Economic impacts of broadband : Evidence from Turkey

profileal1993
sourcesforbroadbandinTurkey.zip

Country_Report_Turkey.pdf

2013 BSA Global Cloud Computing Scorecard www.bsa.org/cloudscorecard | 1

Country: Turkey Score: 52.38 | rank: 18/24

turkey continues to have some gaps in its coverage of basic cyberlaws. For example, there is currently no data protection law in turkey.

turkey also has rules on Internet content regulation in place that may act as a barrier to cloud services. the enforcement of cybercrime laws remains weak, and there is some concern that the penalties imposed for Internet crimes are not an effective deterrent.

Intellectual property protection in turkey is reasonably up to date, but enforcement is patchy.

turkey is making progress toward integration with the European and international communities, but some domestic preferences are still in place for government procurement opportunities.

the government has an ambitious target of providing fast broadband to 95% of households by 2020. However, turkey faces significant challenges in reaching this goal.

turkey made minor improvements to its infrastructure score in the 2013 Scorecard, but its policy environment remains largely unchanged. Its ranking falls one spot — to 18th.

Q Turkey response explanaTory TexT

DaTa PrIVacy

1. Are there laws or regulations governing the collection, use, or other processing of personal information?

6 A draft Law Concerning Protection of Personal Data has been under development for some time.

2. What is the scope and coverage of privacy law?

Not applicable

3. Is the privacy law compatible with the Privacy Principles in the EU Data Protection Directive?

6 The draft law is based on the EU Directive.

4. Is the privacy law compatible with the Privacy Principles in the APEC Privacy Framework?

6

5. Is an independent private right of action available for breaches of data privacy?

Available Some limited privacy rights are available in Turkey.

Article 20 of the Turkish Constitution on the Confidentiality and Protection of Private Life states that “every person has the right to confidentiality and personal privacy and these rights are untouchable.” This regulation also states that “every person has absolute freedom to decide whether to provide or not his/her personal data and in the latter, he/she shall not be compelled to do so.” A range of exemptions apply. Article 24 of the Turkish Civil Code also states that a person whose civil rights are violated can claim protection.

In practice, these private rights are very difficult to enforce in Turkey.

6. Is there an effective agency (or regulator) tasked with the enforcement of privacy laws?

None The draft Law Concerning Protection of Personal Data includes a proposal to establish a Personal Data Protection Authority.

7. What is the nature of the privacy regulator?

Not applicable

8. Are data controllers free from registration requirements?

4 There are no registration requirements in force in Turkey at this time.

9. Are cross-border transfers free from registration requirements?

4 There are no cross-border data transfer restrictions in place at this time.

10. Is there a breach notification law? 6 Turkey has no relevant data breach notification requirements.

2013 BSA Global Cloud Computing Scorecard www.bsa.org/cloudscorecard | 2

Country: turkey

Q Turkey response explanaTory TexT

SecurITy

1. Is there a law or regulation that gives electronic signatures clear legal weight?

4 The Law of Electronic Signature (No. 5070, dated January 15, 2004) contains the key provisions.

The Ordinance on the Procedures and Principles Pertaining to the Implementation of Electronic Signature Law 2005, contains the procedures and principles for the legal, technical aspects, and implementation of electronic signatures.

2. Are ISPs and content service providers free from mandatory filtering or censoring?

6 Internet censorship is common in Turkey, and sites can be blocked for a number of reasons — many of which appear politically motivated — under the Law on the Internet (No. 5651/2007).

In November 2011, the Information and Communication Technologies Authority (BTK) <www.tk.gov.tr> introduced a system where all Internet users are offered a “choice” of filters (“child” or “family”). However, blocking of specific sites by the government remains common and would still occur where a user opted for the “standard” feed. Internet cafes and kiosks are also subject to mandatory filtering requirements. Around 15,000 sites are now blocked in Turkey.

3. Are there laws or enforceable codes containing general security requirements for digital data hosting and cloud service providers?

None There are no enforceable security requirements for digital data hosting in place in Turkey.

4. Are there laws or enforceable codes containing specific security audit requirements for digital data hosting and cloud service providers?

None There are no security audit requirements in place in Turkey.

5. Are there security laws and regulations requiring specific certifications for technology products?

Comprehensive requirements

(including common criteria)

Turkey is a Certificate Authoring Member (the highest level of membership) of the Common Criteria Recognition Arrangement (CCRA) <www.commoncriteriaportal.org>. The use of certifications for technology products in Turkey is small but growing.

cyBercrIMe

1. Are cybercrime laws in place? 4 The Turkish Penal Code (as amended in 2005) includes numerous computer-related offenses.

The controversial Regulation of Internet Publications and Combating Crimes Committed through such Publications (2007) also covers some cybercrime activities.

2. Are cybercrime laws consistent with the Budapest Convention on Cybercrime?

Turkey signed the Convention on Cybercrime in 2010 but has not yet ratified it.

The full range of offenses in the Cybercrime Convention have not yet been implemented in Turkish domestic law. The key outstanding requirement is that Turkey does not yet have an offense relating to “misuse of devices” as required by Article 6 of the Cybercrime Convention.

3. What access do law enforcement authorities have to encrypted data held or transmitted by data hosting providers, carriers, or other service providers?

Not stated There is no specific law or regulation regarding access to encrypted data in Turkey.

4. How does the law deal with extraterritorial offenses?

Limited coverage

Turkish criminal law includes extraterritorial application to Turkish citizens, no matter where they are located.

InTeLLecTuaL ProPerTy rIGHTS

1. Is the country a member of the TRIPS Agreement?

4 Turkey became a member of the TRIPS Agreement in 1995.

2. Have IP laws been enacted to implement TRIPS?

4 Law No. 5846 of December 5, 1951, on Intellectual and Artistic Works (the Copyright Law) has been regularly updated to comply with international treaties and EU directives.

3. Is the country party to the WIPO Copyright Treaty?

4 The WIPO Copyright Treaty entered into force in Spain in 2008.

4. Have laws implementing the WIPO Copyright Treaty been enacted?

4 Turkey’s Copyright Law is very close to compliant with the WIPO Copyright Treaty. However, some further amendments are under discussion to strengthen criminal enforcement in line with the WIPO Copyright Treaty and other EU instruments.

2013 BSA Global Cloud Computing Scorecard www.bsa.org/cloudscorecard | 3

Country: turkey

Q Turkey response explanaTory TexT

5. Are civil sanctions available for unauthorized making available (posting) of copyright holders’ works on the Internet?

4 Sanctions are available for unauthorized communication, to the public, of any work subject to copyright.

6. Are criminal sanctions available for unauthorized making available (posting) of copyright holders’ works on the Internet?

Undecided It is unclear whether criminal sanctions are available for simply making material available. The takedown process discussed below is more likely to be used than a criminal investigation and prosecution.

7. Are there laws governing ISP liability for content that infringes copyright?

4 Article 4 of the Copyright Law (added as part of the 2004 amendments) sets out a simple regime for ISP liability and cooperation.

8. Is there a basis for ISPs to be held liable for content that infringes copyright found on their sites or systems?

4 Copyright owners who allege an infringement can ask the content provider to stop the infringement. Should the content providers not stop the copyright infringement within three working days, the copyright holder is entitled to ask the public prosecutor to order the relevant ISP to suspend the service to the content provider.

9. What sanctions are available for ISP liability for copyright infringing content found on their site or system?

Civil and criminal

A wide range of civil and criminal sanctions are available, including damages, injunctions, fines, and imprisonment. In practice, enforcement is rare.

10. Must ISPs take down content that infringes copyright, upon notification by the right holder?

4 If infringing material is not removed within three days, ISPs can be ordered to suspend services to the offending subscriber. In practice, online copyright breaches remain common, and the use of takedown notices and suspensions is rare.

11. Are ISPs required to inform subscribers upon receiving a notification that the subscriber is using the ISP’s service to distribute content that infringes copyright?

6 ISPs do not have a direct role in informing subscribers about alleged breaches, although they may be required to issue suspension notices following action by prosecutors.

12. Is there clear legal protection against misappropriation of cloud computing services, including effective enforcement?

Comprehensive protection

Turkey has a moderate level of protection available for cloud services, through a combination of its cybercrime legislation and copyright laws. However, both laws have minor gaps, and Turkey does not yet have data protection legislation in place.

SuPPorT for InDuSTry-LeD STanDarDS & InTernaTIonaL HarMonIZaTIon of ruLeS

1. Are there laws, regulations or policies that establish a standards-setting framework for interoperability and portability of data?

4 Turkey manages standards through the Organization Law (No. 132 of 1960).

2. Is there a regulatory body responsible for standards development for the country?

4 Turkish Standards Institution (Türk Standardlari Enstitüsü ([SE]) <www.tse.org.tr>.

3. Are e-commerce laws in place? The Law of Electronic Signature (No. 5070 of January 15, 2004) contains some limited e-commerce provisions.

4. What international instruments are the e-commerce laws based on?

Other The Law of Electronic Signature is limited; the signature sections are based on the UNCITRAL Model Law on Electronic Signatures.

5. Is the downloading of applications or digital data from foreign cloud service providers free from tariff or other trade barriers?

4 Turkey now follows EU policy on setting tariffs, and there are no relevant tariffs or other barriers in the ICT sector.

6. Are international standards favored over domestic standards?

4 Turkey favors a mix of EU and international standards over domestic standards.

7. Does the government participate in international standards-setting process?

4 Turkey is an active participant in ISO committees and working groups.

2013 BSA Global Cloud Computing Scorecard www.bsa.org/cloudscorecard | 4

Country: turkey

Q Turkey response explanaTory TexT

ProMoTInG free TraDe

1. Are any laws or policies in place that implement technology neutrality in government?

6 There are no specific laws or policies on technology neutrality in Turkey.

2. Are cloud computing services able to operate free from laws or policies that mandate the use of certain products (including, but not limited to, types of software), services, standards, or technologies?

4 There are no relevant mandatory requirements in Turkey.

3. Are cloud computing services able to operate free from laws or policies that establish preferences for certain products (including, but not limited to, types of software), services, standards, or technologies?

4 There are no relevant preferences in place in Turkey, although several government agencies (including defense) have significant open source projects under way.

4. Are cloud computing services able to operate free from laws that discriminate based on the nationality of the vendor, developer, or service provider?

6 Turkey is an observer, but not a member, of the WTO plurilateral Agreement on Government Procurement.

Turkey’s public tender law still provides a price preference of up to 15% for domestic bidders.

IcT reaDIneSS, BroaDBanD DePLoyMenT

1. Is there a national broadband plan? • By 2013, the Broadband Subscriber Penetration Rate to increase to 20%

• By 2013, the proportion of Internet users to increase to 60%

The Information Society Strategy for Turkey (2006-2010) <www2.dpt.gov.tr/konj/DPT_ Tanitim/pdf/Action_Plan.pdf> had the following aspiration for national broadband: • The opportunity of high quality and affordable broadband access to all segments of

the society.

In 2006, the Ninth Development Plan of Turkey (2007-2013) <www2.dpt.gov.tr/konj/ DPT_Tanitim/pdf/Ninth_Development_Plan.pdf> set the following targets: • By 2013 the Broadband Subscriber Penetration Rate increase to 20% (with an annual

growth rate of 28.3%) from 3.5% in 2006 • By 2013 the number of Internet users to increase to 60% (with an annual growth rate

of 17%) from 20% in 2006

While these targets may not be met, Turkey’s Internet usage and broadband growth rates are higher than in most other European economies.

In 2011 the Turkish Prime Minister’s Office undertook a National Broadband Vision Study of Turkey, “From Silk Road to Broadband: Enabling Economic Growth and Competitiveness,” linking development of broadband networks with economic development. Refer to <www.infodev.org/en/Article.779.html> for more information.

2. Are there laws or policies that regulate the establishment of different service levels for data transmission based on the nature of data transmitted?

No regulation and limited

public debate

Issues of net neutrality have not yet been the subject of significant consideration in Turkey.

3. Base Indicators

3.1. Population (2011) 73,639,596 In 2011, the population of Turkey increased by 1.2%.

[International Telecommunication Union (ITU), World Telecommunication/ICT Indicators Database (Dec 2012) <www.itu.int/ITU-D/ict/publications/world/world.html>]

3.2. Urban Population (%) (2011) 72% [United Nations, Department of Economic and Social Affairs, Population Division (2012). World Urbanization Prospects: The 2011 Revision, <esa.un.org/unup/CD-ROM/ Urban-Rural-Population.htm>]

3.3. Number of Households (2011) 16,656,000 In 2011, the number of households in Turkey increased by 3.5%.

[International Telecommunication Union (ITU), World Telecommunication/ICT Indicators Database (Dec 2012) <www.itu.int/ITU-D/ict/publications/world/world.html>]

3.4. Population Density (people per square km) (2010)

95 [World Bank, Data Catalog, Indicators, Population Density (2012) <data.worldbank.org/ indicator/EN.POP.DNST>]

2013 BSA Global Cloud Computing Scorecard www.bsa.org/cloudscorecard | 5

Country: turkey

Q Turkey response explanaTory TexT

3.5. Per Capita GDP (US$ 2011) $10,498 In 2011, the per capita GDP for Turkey increased by 8.5% to US$10,498.

[World Bank, Data Catalog, Indicators: GDP per capita, current US$ (2012) <data. worldbank.org/indicator/NY.GDP.PCAP.CD> and GDP growth, annual % (2012) <data. worldbank.org/indicator/NY.GDP.MKTP.KD.ZG>]

3.6. Public Cloud Services Market Value (2011) (Billions of US$)

0.08 Gartner has calculated the value of the public cloud services market in Turkey in 2011 to be US$0.08 billion. This is a 33% increase from 2010 and ranks Turkey 19 (out of 20 countries) in the forecast. Gartner has projected the five-year compound annual growth rate (CAGR) to 2016 to be 28.3%, and this ranks Turkey 5 (out of 20 countries) for growth in the value of the market for public cloud services to 2016.

[Gartner, Forecast Overview: Public Cloud Services, Worldwide, 2011-2016 (August 2012 Update) <www.gartner.com/id=2126916>]

3.7. Personal Computers (% of households) (2011)

48% In 2011, 48.5% of households in Turkey had personal computers. This is a 9.6% increase since 2010 and ranks Turkey 72 out of 182 countries surveyed. The growth from 2010 is below the five-year CAGR from 2006 to 2011 of 21.5%.

[International Telecommunication Union (ITU), World Telecommunication/ICT Indicators Database (Dec 2012) <www.itu.int/ITU-D/ICTEYE/Indicators/Indicators.aspx>]

Note: In some jurisdictions this is an estimate and subsequent editions of the ITU ICT Indicators Database may update this indicator for prior years.

4. ICT and Network Readiness Indicators

4.1. ITU ICT Development Index (IDI) (2011) (Score is out of 10)

4.38 Turkey’s ITU ICT Development Index (IDI) for 2011 is 4.38 (out of 10), resulting in a rank of 69 (out of 161 economies). The 2011 IDI for Turkey has increased by 5%, and the IDI ranking has declined by three places from a rank of 66 since 2010.

[International Telecommunication Union (ITU), Measuring the Information Society (2012) <www.itu.int/ITU-D/ict/publications/idi/2012>]

Note: In some jurisdictions this is an estimate and subsequent editions of the ITU ICT Indicators Database may adjust this indicator, both for 2011 and prior years.

4.2. World Economic Forum Networked Readiness Index (NRI) (2012) (Score is out of 7)

4.28 Turkey has a Networked Readiness Index (NRI) score of 4.28 (out of 7), resulting in a rank of 59 (out of 142 economies) and a rank of 13 (out of 39) in the upper-middle income grouping of economies. The 2012 NRI for Turkey has increased by 12.9% and improved from a rank of 71 since 2011.

[World Economic Forum, Global Information Technology Report (2012) <www. networkedreadiness.com/gitr>]

4.3. International Connectivity Score (2011) (Score is out of 10)

5.51 Turkey has a Connectivity Score of 5.51 (out of 10), resulting in a rank of 4 (out of 25) in the resource-driven grouping of countries/economies.

[Nokia Siemens, Connectivity Scorecard (2011) <www.connectivityscorecard.org>]

4.4. IT Industry Competitiveness Index (2011) (Score is out of 100)

38.70 Turkey has an IT Industry Competitiveness Index Score of 38.7 (out of 100), resulting in a rank of 41 (out of 66 countries/economies included in the index). The 2011 index score is a 14.8% increase on the 2009 score. Turkey has moved up the ranking by five places since 2009.

[Business Software Alliance (BSA) / Economist Intelligence Unit (EIU), IT Industry Competitiveness Index (2011) <globalindex11.bsa.org>]

5. Internet Users and International Bandwidth

5.1. Internet Users (2011) 31,002,270 [calculated from 8.3.1. and 8.5.2.]

5.2. Internet Users as % of Population (2011)

42% In 2011, 42% of the population in Turkey used the Internet, resulting in a ranking of 89 out of 199 countries surveyed. This is a 5.7% increase since 2010. The growth from 2010 is below the five-year CAGR from 2006 to 2011 of 18.2%.

[International Telecommunication Union (ITU), World Telecommunication/ICT Indicators Database (December 2012) <www.itu.int/ITU-D/ICTEYE/Indicators/Indicators.aspx>]

Note: There may be some variations as to how countries calculate this. Some countries base this upon all or part of the population, such as between 16 and 72 years of age.

Note: In some jurisdictions this is an estimate and subsequent editions of the ITU ICT Indicators Database may adjust this indicator, both for 2011 and prior years.

5.3. International Internet Bandwidth (bits per second per Internet user) (2011)

33,938 Turkey’s International Internet Bandwidth (per Internet user) has increased by 78% since 2010.

[International Telecommunication Union (ITU), Measuring the Information Society (2012) <www.itu.int/ITU-D/ict/publications/idi/2012>]

2013 BSA Global Cloud Computing Scorecard www.bsa.org/cloudscorecard | 6

Country: turkey

Q Turkey response explanaTory TexT

5.4. International Internet Bandwidth (2011) (total gigabits per second [Gbps] per country)

1,052 Turkey has increased its International Internet Bandwidth by 90% since 2010 to 1,052 Gbps and is ranked 19 out of 188 countries surveyed. The growth from 2010 is above the five-year CAGR from 2006 to 2011 of 87%.

[International Telecommunication Union (ITU), World Telecommunication/ICT Indicators Database (Dec 2012) <www.itu.int/ITU-D/ict/publications/world/world.html>]

6. Fixed Broadband

6.1. Fixed Broadband Subscriptions (2011)

7,554,725 Turkey has increased the number of fixed broadband subscribers by 7% since 2010 to, 7,554,725, and is ranked 15 out of 182 countries surveyed. The growth from 2010 is below the five-year CAGR from 2006 to 2011 of 22.2%.

[International Telecommunication Union (ITU), World Telecommunication/ICT Indicators Database (Dec 2012) <www.itu.int/ITU-D/ict/publications/world/world.html>]

Note: In some jurisdictions this is an estimate and subsequent editions of the ITU ICT Indicators Database may adjust this indicator, both for 2011 and for prior years.

6.2. Fixed Broadband Subscriptions as % of Households (2011)

45% [calculated from 8.3.3. and 8.6.1.]

Note: This may be skewed by business usage in some countries (refer to OECD comments about this).

6.3. Fixed Broadband Subscriptions as % of Population (2011)

10% Turkey has increased its fixed broadband subscriptions (as a share of the population) by 5% since 2010, which is below the five-year CAGR from 2006 to 2011 of 20.6%. This ranks Turkey 15 out of 187 countries surveyed.[International Telecommunication Union (ITU), World Telecommunication/ICT Indicators Database (June 2011) <www.itu.int/ ITU-D/ICTEYE/Indicators/Indicators.aspx>]

The OECD figures below present a breakdown on the type of fixed broadband connections in Turkey.

In the OECD, Turkey was ranked 34 (out of 34) for fixed (wired) broadband subscribers as a percentage of population [OECD Broadband Subscribers (Dec 2011) <www.oecd. org/sti/ict/broadband>]

– DSL: 9.3% – Cable: 0.6% – Fiber/LAN: 0.4%

Total: 10.4% (7,576,798 subscriptions). The OECD average total for 2011 was 25.6%.

Turkey’s fixed broadband growth for 2011 was 6.7% (ranked 8 out of 34 for growth), above the OECD average growth of 4.1%.

Note: There may be minor variations in the ITU and OECD subscriber totals due to definition, timing or population baseline differences.

[International Telecommunication Union (ITU), World Telecommunication/ICT Indicators Database (July 2011) <www.itu.int/ITU-D/ICTEYE/Indicators/Indicators.aspx>]

6.4. Fixed Broadband Subscriptions as % of Internet Users (2011)

24% [calculated from 8.5.1 and 8.6.1]

7. Mobile Broadband

7.1. Mobile Cellular Subscriptions (2011)

65,321,745 In 2011, Turkey increased the number of mobile cellular subscriptions by 5.8% and is ranked 19 out of 195 countries surveyed. The number of subscriptions account for 89% of the population.

[International Telecommunication Union (ITU), World Telecommunication/ICT Indicators Database (Dec 2012) <www.itu.int/ITU-D/ict/publications/world/world.html>]

Note: This figure may be inflated due to multiple subscriptions per head of population, but excludes dedicated mobile broadband devices (such as 3G data cards and tablets)/

2013 BSA Global Cloud Computing Scorecard www.bsa.org/cloudscorecard | 7

Country: turkey

Q Turkey response explanaTory TexT

7.2. Active Mobile Broadband Subscriptions per 100 inhabitants (2011)

9 Turkey has increased the number of active mobile broadband subscriptions (as a share of the population) by 340% since 2010. This ranks Turkey 87 out of 144 countries surveyed.

The OECD figures below present a breakdown on the type of mobile broadband connections in Turkey.

For 2011, Turkey’s OECD rank has not changed and was 33 (out of 34) for mobile wireless broadband subscribers as a percentage of population [OECD Broadband Subscribers (Dec 2011) <www.oecd.org/sti/ict/broadband>]

– Satellite: 0% – Terrestrial fixed wireless: 0.2% – Standard mobile broadband subscription: 6.7% (up from 2% in 2010) – Dedicated mobile data subscriptions: 2.1% (up from 0% in 2010)

Total: 8.9% (6,465,606 subscriptions). The OECD average total for 2011 was 54.3%.

Turkey’s wireless broadband growth for 2011 was 345% (ranked 2 out of 34 for growth), well above the OECD average growth of 30.5%.

Note: The mobile broadband subscription types were first reported by OECD in 2010, and ITU data are beginning to have this granularity.

Note: The OECD figures include mobile data subscriptions, which are not as consistently reported in the ITU indicators.

[International Telecommunication Union (ITU), World Telecommunication/ICT Indicators Database (Dec 2012) <www.itu.int/ITU-D/ict/publications/world/world.html>]

Note: This refers to the sum of standard mobile broadband and dedicated mobile broadband subscriptions to the public Internet. It covers actual subscribers, not potential subscribers, even though the latter may have broadband-enabled handsets.

Note: In some jurisdictions this is an estimate and subsequent editions of the ITU ICT Indicators Database may adjust this indicator, both for 2011 and for prior years.

7.3. Number of Active Mobile Broadband Subscriptions (2011)

6,454,801 In 2011, Turkey increased the number of active mobile broadband subscriptions by 346% and is ranked 87 out of 145 countries surveyed.

[International Telecommunication Union (ITU), World Telecommunication/ICT Indicators Database (Dec 2012) <www.itu.int/ITU-D/ict/publications/world/world.html>]

PLAN_26_2_130_146.pdf

Ultimate ICT Network in Turkey For Smart Cities

Akıllı Kentler İçin Türkiye’deki Son Bit (Bilgi ve İletişim Teknolojileri) Ağı

Received: 18.02.2016 Accepted: 15.08.2016

Correspondence: Özge Yalçıner Ercoşkun.

e-mail: [email protected]

Planlama 2016;26(2):130-146 doi: 10.14744/planlama.2016.27247

ARTICLE / ARAŞTIRMA

Özge Yalçıner Ercoşkun

Department of City and Regional Planning, Gazi University Faculty of Architecture, Ankara, Turkey

ABSTRACT A smart city is a place where the traditional networks and services are made more efficient with the use of information and communication technologies (ICT), for the benefit of its citizens. This paper provides an overview of ultimate ICT network built in Turkish cities. It also makes comparison of telecommunication status and infrastructure between Turkish cities by thematic maps using Geographic Information System (GIS) in order to visualize the ultimate ICT network for pav- ing the way for smart cities. The paper begins with smart city concept and focus on best practices in the world. The next part gives some statistics on Turkey’s internet, mobile phone and social media usage. Then using the latest statistics, Turkish cities are compared according to some indicators such as the number of 3G, xDSL, mobile broadband and fiber subscribers, the length of fiber cable infrastructure. After making national analyses, some smart city initiatives and some governmental pilot projects on smart cities are explained to review the cur- rent situation for paving the way for smart cities in Turkey. The chapter concludes with some challenges and future options in city level.

ÖZ Akıllı kent, kentlinin yararına olacak şekilde, bilgi ve iletişim tek- nolojilerinin (BİT) kullanımıyla, klasik ağ ve hizmetlerin daha etkin hale getirildiği yerdir. Bu makale, Türk kentlerinde kurulmuş son teknoloji BİT ağlarının durumunu ortaya koymaktadır. Ayrıca, Türk kentlerinin telekomünikasyon durumu ve altyapısı konusunda bir karşılaştırma yapmakta, Coğrafi Bilgi Sistemleri (CBS) aracılığıyla hazırlanmış tematik haritalarla akıllı kent olma yolundaki son bilgi ve iletişim ağını görüntülemektedir. Makale, akıllı kent kavramı ile başlamakta ve dünyadaki en iyi uygulamalarla devam etmektedir. Dünyadaki akıllı kentler özetlenmekte, IBM’in daha akıllı kentler programından örnekler verilmektedir. Bu örnekler, İskandinav ül- kelerinden, Uzak Doğu’dan ve Amerika’dan bazı kentler üzerine- dir. Sonraki bölüm Türkiye’deki internet, cep telefonu ve sosyal medya kullanımı üzerine birtakım istatistikler vermektedir. Daha sonra en son istatistiklere göre, Türk kentleri bazı göstergelere göre karşılaştırılmaktadır, bu göstergeler, 3G, XDSL, mobil geniş bant, fiber abonelikler ve fiber kablo altyapısı uzunluğuna göre sı- ralanmaktadır. Ulusal analizler yaptıktan sonra, bazı akıllı kent giri- şimleri ve akıllı kentler üzerine pilot projeler anlatılmakta, sonuçta Türkiye’de akıllı kent kavramının önünü açmak için kent düzeyinde zorluklar ve gelecek seçenekler üzerinde durulmaktadır.

Anahtar sözcükler: 3G and mobil geniş bant abonelikleri; fiber kablo altya- pısı; akıllı kentler; Türk kentlerinin CBS ortamında tematik haritaları.

Keywords: 3G and mobile broadband subscriptions; fiber cable infrastruc- ture; smart cities; thematic maps of Turkish cities in GIS.

130

Introduction

“From computers to data to information to communication to democracy” (Saco, 2002, p.xiii). New information and com- munication technology (ICT) platforms have evolved rapidly and transformed vastly during the last 30 years. Some of the concepts used to define the transformation in the Digital Age put forward by Castells (2009), show the cultural change in the globalized world. The popularization of the internet and the wireless systems have caused changes in today’s commu- nication platform. The internet was invented in 1969, but its commercialization expanded in the 1990s. Since then, users grew from 40 million in 1995 to 3.4 billion in 2016 (Castells, 2009, Global Digital Statistics, 2016). After the convergence of internet and wireless systems in the 2000s, the meaning of ‘being connected’ has shifted (Pinzon, 2013); as Castells points “the key feature of wireless communication is not mo- bility but perpetual connectivity” (2009, p.69).

Recent developments in geo-information technologies on wireless platforms to link spatial and relational data provide new and smart insights into patterns of urban life, such as traffic flows within cities, urban mobility and proximity of social relations. Mobile phones allow for large scale moni- toring of people’s movements and physical proximities over time, giving the possibility of understanding cognitive rela- tionships and social ties. Popular social network sites (SNS), like Swarm-Foursquare, Facebook check-in and Quora, ex- tend this real-time and location-based geo-information across highly connected networks of users. Globally, through rapid adoption of smart phone technologies and rise of geo-so- cial networking, the number of active social media users is reached 2.3 billion by 2016 (Global Digital Statistics, 2016).

ICT utilize mobile devices that have changed our lives. They shape how we interact with each other and how we organize friends and family as well as broader networks. They have become the ‘digital fabric of our lives’. The opening ques- tion for the usual telephone conversation has changed from “Whom?” to “Where?” since today we know who we are talking to before answering the phone. Mobile communica- tion coordinates our daily mobility activities, for example, who is going to do the shopping, or who is going to pick up the children that day. These are not planned but talked and decided in real-time. Keeping in touch with social networks has its advantages, but interacting with real life environment is important while living in a smart community (Mobile Tech, 2013).

A smart community is a community where various next- generation technologies and advanced social systems are ef- fectively integrated and utilize in a smart city. In smart cities, digital technologies translate into better public services for

citizens and better use of resources. A smart city is a place where the traditional networks and services are made more efficient with the use of ICT, for the benefit of its inhabitants and businesses.

This paper provides an overview of ultimate ICT network built in Turkish cities. It also makes comparison of telecom- munication status and infrastructure between Turkish cities by thematic maps using GIS in order to prove the ultimate ICT network for paving the way towards smart cities. Some of the concepts used to define smart cities such as eco-tech, livable, green and sustainable cities or smart growth can be considered beyond the scope of this paper since they relate to other spatial components and city planning visions.

The paper begins with smart city concept and continues with Top 10 Smart Cities and Smarter Cities Program in the world. They are explained as best practices from Scandinavia, Far East and America. The next part gives some statistics on Turkey’s internet, mobile phone and social media usage. Then using the latest statistics, Turkish cities are compared accord- ing to some indicators such as the number of 3G, xDSL, mo- bile broadband and fiber subscribers, the length of fiber cable infrastructure. After making national analyses, some smart city initiatives and some national projects on smart cities are explained to review the current situation for smart cities in Turkey. The chapter concludes with some challenges and fu- ture options in city level.

Methodology

The work is based on a literature study which covered pa- pers and reports on indicators, methodologies, rankings and evaluation tools related to ICT and cities. The literature study aimed to define the concept of smart city and to make an overview of smart city examples. National reports, Türk Telekom statistics, GSM operators’ statistics and social media facts are found for the years of 2014 and the first quarter of 2015. Some facts and indicators are extracted from the research of ICT usage of Turkish people made by Turkish Statistics Institute. The city data was blended from publicly available annual province statistics of Information Technolo- gies and Communications Authority (BTK) for Turkey. Then a geo-database was built in ArcGIS environment at city level concerning some indicators such as the number of 3G, xDSL, mobile broadband and fiber subscribers, the length of fiber cable infrastructure of BTK up-to-date statistics. Some the- matic maps were produced to draw quantities using graduat- ed colors to show values. In order to attract attention and to make comparison, five classes were used in the legend of the thematic maps for generating the difference in each indicator towards the ultimate ICT network for smart cities. From gen- eral to specific point of view, some e-municipality efforts in

Özge Yalçıner Ercoşkun 131

Turkish cities were put in a summary table and some ultimate national smart city pilot projects were explained briefly in order to clarify the national vision of being a smart city.

Smart City Concept and Smart Cities in the World

The symbolic importance of labeling and branding cities with ‘cyber’, ‘intelligent’, ‘digital’ or ‘smart’ prefixes is stressed around the world (Graham & Marvin, 1999). Smart cities can be accepted as the demonstration areas of e-government ap- plications. Smart city concept is integrated with the infor- mation based economy. Research and the use of new tech- nologies accounted for the development in science, industry and commerce in a smart city. The impacts of ICTs can be observed in smart cities. The e-governance concept links the administrative institutions to these technologies. Public bod- ies using ICTs in management, commerce and communication across offer e-governance with multi-participation.

The financial services, information technologies and com- munication industry in a smart city bring economic develop- ment. Smart spaces are integrated with infrastructure in a smart city. Local governments should share their decisions, plans and projects with developing their internet networks for effective city management, should present some guide- lines (Odendaal, 2003).

Smart cities own an advanced urban information systems where urban dwellers can use these systems even via their mobile devices. Urban information systems help automate analysis, share information, and encourage teamwork. By visually displaying information, GIS enhances evaluation and performs quickly. Real-time information and sensor updates in GIS support better decision making and improve city man- agement. Creating an urban inventory, thematic maps, and queries also help to identify the gaps and scenarios help in planning measures (Yalçıner, 2002).

Moreover, a city can be defined as smart when investments in human capital, institutional systems, traditional transport and modern ICT communication infrastructure fuel sustainable economic development and a high quality of life, with a wise management of natural resources (Odendaal, 2003). Smart cities are all urban settlements which show strategic effort on the new ICTs in a way to achieve high quality of life, effective- ness and competitiveness on many levels (Angelidou, 2014). For the development of a smart city, instrumenting a city technically and investing in hard infrastructure also provides opportunities for innovation process. However, technology is not enough to guarantee the real smartness of cities, and it does not make urban dwellers themselves to think or act smart (Angelidou, 2015).

On the other hand, some smart city strategies consider spe- cific districts and clusters (Komninos, 2011), such as busi- ness districts, R&D clusters, university technoparks, logistical clusters or neighborhoods. This is a spatial vision that gives character and functions at city level. It addresses some users, who benefit the district or neighborhood where they live in, work in or visit.

Smart city indicators are studied all around world. ITU Acad- emy online notes provide some key performance indicators for the smart cities to understand how ICT performance can be measured (ITU website). When focused on the heading of Urban Sustainability, ICT appears under the connectivity topic. In the connectivity topic, the indicators are: access to internet is measured by the number of broadband and mobile broadband subscriptions; access to telephones is measured by the indicator of the number of mobile 3G subscriptions. Mobile broadband subscriptions are important and their number has a multiplicative effect on GDP, productivity and employment. OECD and the World Economic Forum also use these indicators for the goal of smart cities. So, this paper takes these indicators to visualize the ultimate ICT network built in Turkish cities and to make comparison of telecommu- nication status and infrastructure between them.

As the planet becomes more urban, cities need to get smarter. Large scale urbanization necessitates finding new and smart ways to manage complexity, reduce expenses and improve quality of life. This paper stressed the importance of better use of ICT. But a more comprehensive view of a smart city gives an integrated approach to improving the efficiency of city operations, the quality of life for its dwellers, and growing the local economy. Investments in human capital and modern ICT infrastructures fuel sustainable economic development and a high quality of life, with a wise management of natu- ral resources, through participatory action and engagement (Lombardi et al. 2012).

Canadian urban strategist Dr. Boyd Cohen developed a smart city wheel which allows a common language to develop amongst city stakeholders.

Smart cities can be identified along six main axes: A smart economy, smart mobility, a smart environment, smart people, smart living and smart government (Figure 1). These six axes are linked with traditional theories of urban growth and New Urbanism development and are based on theories of regional competitiveness, transport and ICT economies, human capi- tal, quality of life, and participation of citizens in city gover- nance (Fastcoexist website).

The Smart City Index is widely accepted and created by Co- hen. Classified literature on the categories of a Smart City

PLANLAMA132

Özge Yalçıner Ercoşkun 133

is given in Table 1. Smart Environment can be created by green urban planning. Smart City is powered by “renewable energy,” every function is “carbon free”; it shifts from central to “distributed” small scale power, water and waste systems; and “food production” will be part of the urban green infra- structure. Close-loop cities is emphasized in which the urban eco-cycle is balanced in support of a circular view of urban metabolism. Cities, designed on New Urbanism principles, nurture a special “sense of place” through a renewable and local economy, and they are “walkable” and transit-oriented using alternative fueled vehicles (Ercoskun, 2012). Moreover, ICTs may contribute economic development in a smart city, with smart spaces being integrated with smart infrastructure. The use of the Internet, virtual libraries, online forums, inter- active talks and webinars makes life easier, and facilitates fuel savings as well as flexible and adaptable living and working conditions. Awareness is raised among the city residents in such matters of energy use as smart metering, car pooling and life safety systems (Ercoskun, 2010).

In Smart Government heading (Table 1), Wi-Fi and broadband coverage is one of the important indicators for smart cities (Fastcoexist website). Also under the Smart People heading, the number of internet-connected households and residents with smartphone access is the other indicator of smart cities which are used in this paper.

Smart city solutions have been evolved to more supply-driven rather than demand-driven in the cities (Komninos, 2011). Several conceptual models of smart cities have appeared under their impact of supply and demand. Some ideas for the smart city focus on ICTs which can make cities and their services act

smart. Other smart city paradigms relate technology with in- novation and human capital development to participate in so- ciety and solve problems. Nevertheless, smart city programs are being implemented in hundreds of cities in the world.

The Top 10 Smart Cities around the world are ranked by Cohen who considers the variables about innovation and sus- tainability. These cities are (Fastcoexist website):

1) Vienna: Vienna was the only city that ranked in the top 10 in every category such as: innovation city, regional green city, quality of life and digital governance. Vienna establishes smart-city targets with programs like the Smart Energy Vi- sion 2050 and Roadmap 2020. Carbon reduction, transporta- tion and land-use planning changes make the city number 1 in smart city technologies.

2) Toronto: The highest rated smart city in North America, Toronto is the second in ranking. IBM contributes its smart- ness more. Toronto is also an active in C40 megacities to- wards low-carbon economy.

3) Paris: Paris was highly rated in several categories includ- ing innovation, green cities in Europe, and digital governance. Paris is popular with its automobile and bike sharing pro- grams.

4) New York: New York scored higher than most other cities in the ranking in all of the categories except quality of life. New York partnered with IBM in 2009. IBM helps the city prevent fires.

5) London: London is famous for some of its sustainability innovations such as congestion tax and its robust transit sys- tem. London built a partnership with O2 to launch the largest free Wi-Fi network in Europe.

6) Tokyo: Tokyo scores well in the innovation and digital city categories. Last year, the city announced plans to cre- ate a smart town in the suburbs. This eco-suburb will in- tegrate solar panels and energy efficient appliances all con- nected to a smart grid. Tokyo also promotes smart mobility solutions.

7) Berlin: Berlin has good scores in innovation, greenness and quality of life. In collaboration with the firms, city works on vehicle-to-grid (V2G) technologies for creating a virtual power plant from electric vehicles.

8) Copenhagen: Copenhagen, as the European Green Capital in 2015, takes leadership role on sustainable clean-tech inno- vation. The city has committed to carbon neutrality by 2025 and 40% of its citizens regularly commute by bike.

Figure 1. Smart city wheel.

Table 1. Classified literature on the categories of a smart city

Categories Indicators

Smart environment

Smart mobility

Smart government

Smart living

Description

Green urban planning

Mixed-modal access

Enabling Supply-Demand Side Policy

Safe

Healthy

Cultural&vibrant and happy

Prioritized clean&non- motorized options

Integrated ICT

Green buildings

Green energy

• Pollution control • Biodiversity-green area enhancement • Compact city • Mixed use, mixed income settlements • Walkable-cycleable environment • Strong urban identity • Climate friendly pattern • Urban metabolism for eco-efficiency • Buy local, eat local/self-sufficiency

• More public transport • Integrated fare systems

• Using ICT to deliver energy and information exchange between providers and users

• Crime, terrorism • Smart protection • ICT for emergency

• Life expectancy • Investment on health • Using ICT for healthcare services

• Quality of life • Investment on culture • Equality • Participatory cultural and tourism events

• Bicycle paths-roads • Bicycle sharing systems • EV charging stations

• Optimize logistics for energy-saving • Energy-efficient fuels and engines • Smart card systems for public transit

• Green-certified buildings • Using building automation systems • Adapting/retrofitting for energy-saving

• Using renewables such as solar, wind power, geothermal etc., • Smart metering • Resilient smart grid • Wi-fi connected public lighting

Newman & Jennings (2008), Wheeler (2004), Holmgren (2007), Kazimee (2002), EU (2004), Resilientcity website, Vergunst (2002), Coaffee (2008), Saavendra & Budd (2009), Raven (2010), Godschalk (2003), Hopkins (2008), UNISDR (2010), Newman, Beatley & Boyer (2009), Hodson & Marvin (2009)

Hensher & Button (2003), Tolley, (2003), Ercoskun (2016), Neirotti et al. (2014)

Neirotti et al. (2014), Fastcoexist website, Chourabi et al. (2012), Correia and Wunstel (2011) and Mahizhnan (1999) and Steria (2011)

Neirotti et al. (2014), Fastcoexist website, Accenture (2011), Dirks et al. (2009), Nam and Pardo (2011) and Washburn et al. (2010)

Fastcoexist website Neirotti et al. (2014), Accenture (2011), Atzori et al. (2010), Correia and Wunstel (2011), Dirks et al. (2009); Nam and Pardo (2011), The Climate Group et al. (2011) and Washburn et al. (2010)

Fastcoexist website, Accenture (2011), Dirks et al. (2009), Mahizhnan (1999), Nam and Pardo (2011) and Washburn et al. (2010), Neirotti et al. (2014)

Fastcoexist website, Wheeler, (2004), Kazimee (2002), EU (2004), Resilientcity website, Neirotti et al. (2014)

Atzori et al. (2010), Caragliu et al. (2009), Correia and Wunstel (2011), Dirks et al. (2009), Giffinger et al. (2007), La Greca et al.(2011), Munuzuri et al. (2005), Nam and Pardo (2011), Steria (2011), The Climate Group et al. (2011), Think (2011), Toppeta (2010) Washburn et al. (2010), Neirotti et al. (2014) Ercoskun (2016)

Accenture (2011), Steria (2011), The Climate Group et al. (2011), Think (2011), Washburn et al. (2010), Neirotti et al. (2014)

Chourabi et al. (2012), Correia and Wunstel (2011) , Mahizhnan (1999) Steria (2011), Accenture (2011), Correia and Wunstel (2011), Dirks et al. (2009), Hughes et al. (2013), Nam and Pardo (2011), The Climate Group et al. (2011), Think (2011), Toppeta (2010), Neirotti et al. (2014)

References

PLANLAMA134

9) Hong Kong: Hong Kong is a leader in the use and adoption of smart cards, which are already used by millions of residents for services like public transit, library access, building access, shopping, and car parks.

10) Barcelona: Barcelona is a pioneer in smart city and low- carbon solutions. It was among the first in the world to in- troduce a solar thermal to promote the electric vehicles and charging infrastructure, and the city also launched a project

to develop a living lab for smart-city innovation.

Moreover, IBM Smarter Cities Program gives some other smart city practices. For example, San Francisco, which has equipped with thousands of kilometers of sewer system pipes and treatment facilities that can give alarm against pump fail- ure, broken pipes and overflowing storm drains in real time, with a smart maintenance. Singapore, with its smart traffic system created to lower congestion and carbon emissions

Table 1. Classified literature on the categories of a smart city (Cont.)

Categories Indicators

Smart government

Smart people

Smart economy

Description

Transparency and open data

Inclusive society

Local&global interconnec- tedness

Productivity

Enterpreneurship & Innovation

Embrace creativity

21st century education

ICT & e-Gov

• Increasing reliability and transparency of energy supply systems • Enabling every citizen to access official documents in a simple way

• Raising the number of internet- connected households and residents with smartphone access • Improving the quality of life by stimulating social learning and participation especially for disadvantaged groups

• Integrating the city in national and global markets

• Improving productivity (i.e. output divided input) through automatic routine processes and by powering managers’ decision-making, planning and control activities.

• Fostering the innovation systems and entrepreneurship in the urban ecosystem by local incubators

• Technological platforms harnessing the collective intelligence and creativity of citizens

• Creating more opportunities for students and teachers using ICT tools • Interplay between university and government for e-learning

• Wi-Fi and broadband coverage • ICT-based public participation • E-administration • E-ballots

Neirotti et al. (2014), Fastcoexist website, Chourabi et al. (2012), Correia and Wunstel (2011) and Mahizhnan (1999) and Steria (2011)

Atzori et al. (2010), Bakıcı, Almirall, and Wareham (2013), Caragliu et al. (2009), Chourabi et al. (2012), Correia and Wunstel (2011), Giffinger et al. (2007), Mahizhnan (1999) and Toppeta (2010), Neirotti et al. (2014), Fastcoexist website

Neirotti et al. (2014), Bakıcı et al. (2013), Caragliu et al. (2009), Chourabi et al.(2012), Correia and Wunstel (2011), Giffinger et al. (2007), Mahizhnan (1999) and Toppeta (2010), Fastcoexist website

McKinsey Global Institute (2011), Neirotti et al. (2014), Fastcoexist website

Bakıcı et al. (2013), Caragliu et al. (2009), Chourabi et al. (2012), Correia and Wunstel (2011), Giffinger et al. (2007), Mahizhnan (1999) and Toppeta (2010), Neirotti et al. (2014), Fastcoexist website

Neirotti et al. (2014), Fastcoexist website, Sasaki (2010), Lombardi et al. (2012)

Accenture (2011), Dirks et al. (2009), Mahizhnan (1999), Nam and Pardo (2011) and Washburn et al. (2010), Lombardi et al. (2012), Neirotti et al. (2014), Fastcoexist website

Neirotti et al. (2014), Fastcoexist website, Accenture (2011), Bakıcı et al. (2013), Caragliu et al. (2009), Chourabi et al. (2012), Correia and Wunstel (2011), Dirks et al. (2009), Giffinger et al. (2007), Odendaal (2003), Steria (2011), Think (2011), Toppeta (2010) and Washburn et al. (2010)

References

Özge Yalçıner Ercoşkun 135

by recognizing traffic patterns on a city level. The engineers developed systems for leveraging road pricing, integrated fare management and deep analytics to help predict and miti- gate traffic congestion (IBM, 2011). Technologies providing electronic tolling, advanced driver advisories, parking guid- ance, responsive traffic signaling, intelligent dispatch of cabs and contactless smart card-based fare transactions, have all helped to realize the utility and value of smart transportation systems in managing traffic in Singapore (Hin & Subramaniam, 2012). The City of Rio de Janeiro, which coordinates infor- mation flow from 20 city departments into one operations center for real-time visualization, monitoring and answering of response to incidents across the city. The system uses a smart weather-prediction technology to locate storms and hills vulnerable to landslides. And City of Madrid has an emer- gency coordination center. When authorities receive infor- mation about an emergency, they can recognize it as a unique or a duplicate incident with the improved of 25% response times (IBM, 2011).

Case Study: Turkish Cities

Turkey’s Facts Turkey’s implementation of e-government is shaped by many factors including its size, centralized governance structure, e-government goals, Gross Domestic Product (GDP), popu- lation, etc. Table 2 illustrates some statistics about Turkey be- longing to 2015. The number of mobile phone subscriptions is relatively high where the average annual income per capita is low. Mobile devices, smartphones and internet TVs attract Turkish citizens day-by-day.

TTNet, a subsidiary of Türk Telekom, launched broadband in the early 2000s, providing ADSL services. Mobile broadband over 3G started with three GSM operators: Turkcell, Voda- fone and Avea. Also, broadband services provided fiber and cable-TV networks, in the 2000s. After significant growth in mobile telephony over the last decade, there are more than 70 million subscribers (around 70% of which have 3G), for a penetration rate of 91.5%. According to subscriber numbers, Turkcell, Vodafone, and Avea’s market shares are 49.62%, 28.57%, and 21.87%, respectively. With 8.5 million fixed and 26.4 million mobile subscribers, total broadband subscription in Turkey has reached 35 million. Fixed broad- band subscription is as follows: 79% xDSL, 15% fiber, and 6% cable. Currently, Türk Telekom owns 182,450 km of fi- ber infrastructure, around 123,000 km as backbone network and around 60,000 km that reaches subscribers. Alterna- tive operators own a total of 51,244 km fiber infrastruc- ture reaching subscribers. Ten years after its privatization and the opening of the competitive sector, Türk Telekom is still dominant provider in broadband services (Schkenkkann, 2014).

According to the Vodafone report, with the introduction of the 3G networks in 2009, mobile broadband and smart- phones are rapidly replacing older generation devices in the Turkish market. There are already 5.6 million Smartphone users in 2013 and the number of Facebook users stand at 32,408,540, the 7th largest usage base per country in the world (Mobile Tech, 2013). The Internet, and social media in particular, have become an important part of public life in Turkey. Turkish users spent the second most amount of time online per capita in Europe, after the UK. Facebook is the most popular social network with a penetration equal to 41.59% of the population. Around 90% of the country’s inter- net users utilize Facebook actively. Turkey is ranking the 4th largest in global use of Facebook and 8th largest for Twitter with 31.1% penetration and 11,337,500 active Twitter users in 2014 (Tunç, 2014). Foursquare and Swarm has reached 7.2 million users in Turkey in 2015 (the second country after USA). Users check-in average is about 1.8 million daily. They frequently check-in cafes, restaurants and their homes, they generally prefer Turkish food, fastfood and seafood come next (Foursquare website).

Ultimate ICT Network in Turkey There are 81 provinces in Turkey. Among the 81 provinces, 30 are designated metropolitan municipalities in 2014. The larg- est by far is İstanbul (23rd largest urban area in the world), followed by Ankara (87th) and İzmir (331rd in the world) (Demographia, 2015) (Figure 2).

Turkey welcomed 3G technology in 2009, and the investments spread its benefits in Turkey and get more people involved in this area. A recent report by the Information Technologies and Communications Authority (BTK) for the first quarter of 2015, which reported that there are were nearly 72 million mobile subscribers and 59.4 million 3G subscribers in Turkey in total (TT website). İstanbul, Ankara and İzmir were the top 3 cities in 2014. Bursa, Antalya, Konya was ranked number 4, 5 and 6, respectively. Mersin, Adana and Gaziantep followed as number 7, 8 and 9 in ranking. (Figure 3).

Table 2. Turkey’s facts for e-government (TUIK, 2015)

Indicators-2015 Turkey

Population 77,695,904

Number of internet users 44,700,000

Household possession of mobile computers 28.5%

and tablets (% of population)

Mobile telephone subscriptions 71,888,000

Number of TV with internet (% of population) 20.9%

The average annual income per capita 7714 $

PLANLAMA136

Figure 2. Population map of Turkish Cities-2014.

Figure 3. Number of mobile 3G subscribers in 2014.

Özge Yalçıner Ercoşkun 137

Total broadband subscribers in Turkey approached to 42.9 million at the end of the first quarter (Q1) of 2015. Fixed broadband services were widely provided; the number of fixed broadband subscribers reached 9 million in Turkey (TT website). İstanbul became the undisputed leader in the rank- ings of the number total broadband subscribers. It is followed by Ankara which grabbed second place and İzmir in third. Rounding out the top 8, we see five other municipalities- Bur- sa, Antalya, Konya, Kocaeli and Adana (Figure 4).

XDSL is the main fixed broadband technology in the coun- try with 6.8 million subscribers (TT website). After İstanbul, Ankara and İzmir, Bursa, Kocaeli and Antalya follow them ac- cording to the number of xDSL subscribers in the country in 2014 (Figure 5).

The total number of mobile broadband users was 33.9 mil-

lion corresponding to 43.7% population penetration rate at the end of Q1 of 2015. Mobile broadband users via mobile phones reached to 32.4 million while mobile broadband us- ers via non-phone mobile devices increased to 1.5 million. Compared to OECD average of 78.2% mobile broadband penetration rate, Turkey has an average growth potential in terms of mobile broadband penetration (TT website). After İstanbul, Ankara and İzmir, Bursa, Antalya, Konya, Mersin, Ko- caeli, Adana, Gaziantep and Hatay follow them according to the number of mobile broadband subscribers in the country in 2014 (Figure 6). Number of mobile broadband subscribers via mobile phones in 2014 shows the distribution of same cities (Figure 7).

Additionally, share of fiber subscribers exceeded 1.5 million subscribers and there were also 575,000 cable subscribers at the end of Q1 of 2015. Compared to OECD average of

Figure 4. Number of Broadband Subscribers in 2014.

PLANLAMA138

Figure 5. Number of xDSL Subscribers in 2014.

Figure 6. Number of mobile broadband subscribers in 2014.

Özge Yalçıner Ercoşkun 139

27.4% fixed broadband population penetration rate, Turkey has an average growth potential with its 11.6% penetration rate (TT website). Turk Telekom with its approximately 197 thousand kilometer fiber network has the biggest fiber infra- structure in Turkey. Alternative operators’ fiber length was 53 thousand at the end of Q1 of 2015 (TT website). İstanbul is the leader in the rankings of the length of fiber cable in kilometers. It is followed by Ankara which grabbed second place and İzmir in third. Rounding out the top 9, we see six other municipalities- Bursa, Antalya, Mersin, Adana, Kayseri and Konya (Figure 8).

The fiber network structure is imperative for 4G and 4.5G technologies, and telecommunications companies widely crit- icize Türk Telekom, which dominates Turkish fiber infrastruc- ture ownership. Under an executive order from 2011, the Ministry is supposed to encourage Türk Telekom to reach an agreement with alternative providers for right-of-way conces- sions. Also, some municipalities put barriers to the operators for digging the roads for this infrastructure.

In summary, we can compare the number of mobile tele- phony and broadband subscriptions and the length of fiber cable in top 5 cities. İstanbul is the leader and doubles the percentage of Ankara; Antalya needs more investment as a smart city (Table 3).

Smart City Pilot Projects in Turkey Grounding the national-level ICT policies such as e-Turkey to the urban level is a major challenge that needs to be tackled. Within the frame of e-Turkey, City of Yalova is selected as a pilot city for the initiative of ICT Valley Project in the begin- ning of 2000s. Yalova is selected as Turkish Telecom’s pilot city for the provision of a natural-disaster-resistant internet infrastructure after Great Marmara Earthquake. ICT proj- ects of Yalova were presented as best practices in various national and international conferences, meetings, and plat- forms (Velibeyoğlu & Yiğitcanlar, 2009). After Yalova, Bursa, Kocaeli, Ankara and the other cities were put into agenda of ICT Valley Project. Initiatives supporting e-municipality and

Figure 7. Number of mobile broadband subscribers via mobile phones in 2014.

PLANLAMA140

Figure 8. The length of fiber cable in km in 2014.

Table 3. Ratio of city amount to the country level amount: City amounts of subscriptions/length of fiber are divided by country level equivalent amounts of subscriptions/length of fiber. (Top 5 cities according to population) (BTK, 2014)

Number of mobile telephony Number of broadband The length of fiber-km subscriptions - total subscriptions - total

İstanbul 24,87% 25,94% 11,42%

Ankara 7,43% 8,00% 5,75%

İzmir 5,46% 5,90% 4,86%

Bursa 3,51% 3,74% 3,12%

Antalya 3,13% 3,13% 2,83%

Özge Yalçıner Ercoşkun 141

Table 4. Smart city projects in Turkey

Name of the municipality Status

Yalova ICT Valley Project continues.

Fatih-İstanbul GIS digital inventory has been completed; Smart City Project team has been built.

Kadıköy-İstanbul GIS digital inventory has been completed; system has been integrated with e-municipality applications.

Beyoğlu-İstanbul GIS digital inventory has been completed; system has been integrated with e-municipality applications.

İzmir UIS has been completed; 3D applications are integrated with the system.

Ankara GIS digital inventory has been completed; public transportation system has been built and integrated

with Android and IOS.

Bursa UIS has been completed. Web-based 3D city guide is on use.

e-government, and transition from government to e-gover- nance raised the importance of transparency, communica- tion, public accountability, and participation issues. In this sense, the concept of Urban Information Systems (UIS) be- gan to be popular among the local governments (Velibeyoğlu & Yiğitcanlar, 2009). Some district municipalities of İstanbul such as Beyoğlu and Fatih display 3D street views integrat- ed to Google Earth in their UIS. Fatih Municipality applies augmented reality in their smart city projects. Every single building which has been photographed with mobile phones of 3G-4G technology, the information about the building can be sent to the user. Some smart city projects are listed in Table 4.

In year 2015, two major smart city projects named as Akilli KenTT are launched by Ministry of the Turkish Transport, Maritime Affairs and Communications and Türk Telekom. One of them is city of Karaman in the South of Anatolia. Karaman is one of the fastest growing cities in Turkey. Ac- cording to 2013 data, the young make up a large proportion

of its 180,000 residents, making the city open to many dif- ferent technological and social developments. Türk Telekom chose Karaman, which is also the hometown of the Minis- ter, to launch the country’s first smart city project, creat- ing a model for other cities in the country. Karaman’s Akilli KenTT project differs from other smart city initiatives as all the smart applications operating in the city are managed by a single platform and one operations centre.

Throughout the project, Innova (software developer and in- tegrator company) created overall management of all the sys- tems and applications implemented in the project, including setting up the systems architecture, developing an Internet platform, installing the necessary management and screens, establishing an operation centre, and providing kiosk and digi- tal signage applications. Akilli KenTT applications to be built in Karaman are given and categorized in Table 5.

Minister of the Turkish Transport, Maritime Affairs and Com- munications, Lütfi Elvan, in a conference held in Karaman,

Table 5. Akilli KenTT applications (Innova website, 2015)

Internet and monitoring systems

Transportation and Emergency

Management Platform

Smart Stations for buses and trains

Smart Intersection System

Intelligent Parking System

Traffic Management System (TEDES) Panic Button

Wi-Fi Management System

Security Cam in the Cloud Aerometer in the Cloud

The nerve centre enabling the end-to-end management of system.

Designed to reduce waiting times and to optimize the public transport system. Designed to reduce waiting time at junctions, to lower carbon emissions and fuel consumption, to give priority to emergency response vehicles. Optimized parking services by reducing time spent searching parking lots. Identification of drivers breaking traffic laws and regulations, to improve traffic safety on roads. Calling an ambulance by pressing a single button.

Designed to offer free and secure Internet to all citizens across the city.

Cameras placed in public areas, records can also be viewed by citizens. Designed to measure humidity, temperature and noise levels in the city.

PLANLAMA142

said: “We launch an integrated smart city application for the first time in Turkey, in Karaman. We build a smart city that will bring savings and lead to all public institutions, including the municipal authority, to conduct more effective and effi- cient operations. This project will give us the opportunity to offer better public services at a lower cost” (Innova, 2015). The second Akilli KenTT project was launched in Antalya in May, 2015. The similar applications given Table 4 will be in- stalled in Antalya in near future. Among the most popular solutions - installation of smart traffic lights at intersections will regulate the movement with the busy streets, and orga- nize the movement of vehicles of emergency services. Pas- sengers will be able to track the location of buses, recognize arrival at the desired station, using mobile phone. Pensioners and patients will have regular contact with the doctor and emergency ambulance. Smart Technologies will bring 60% of the decline in traffic accidents, while 25% of the reduction in the waiting time in traffic, and 30-35% of the reduction in carbon emissions.

Many of the improvements will affect the tourism sector positively in Antalya. Wi-Fi zones in the center and on the beaches will be created in these areas: Cumhuriyet Meydanı,

Kapalı Yol, Yavuz Özcan Parkı, Konyaaltı Sahili, Sarısu, Kale Kapısı, Kadınlar Plajı ve Beach Park. Akilli KenTT Project will provide free Wi-Fi zones in 22 km wide which will cover public transport vehicles as well. Such projects have been implemented in London, Dubai and Barcelona before Antalya (Öztuzsuz, 2015).

Conclusion

This paper overviewed smart cities by seeing them as spaces that make better use of ICT. However, smart city concept has a broader approach to improving the efficiency of city op- erations, the quality of life for city dwellers, and growing the local economy. So, many studies can be made on this concept. Table 1 gives the accepted categorization such as environ- ment, mobility, living, government, people and economy. The Top 10 Smart Cities around the world, given in this paper, improve all these categories, national governments make at- tempts in the economy, energy, ICT infrastructure and social dimensions. The municipalities of these Smart Cities make collaborative work with NGOs, private sector and univer- sities to improve Smart Growth plans, smart mobility pro- grams and ICT- controlled energy and infrastructure systems.

Table 5. Akilli KenTT applications (Innova website, 2015) (Cont.)

Social issues

Technical and Telecommunication infrastructure

Smart Home and Office Management System

City Lighting System

Remote Irrigation System

Multimedia Payphones

Information Kiosks

Touchdesk Kiosks

Recycling Machine

Intelligent Waste Collection System

Meter Reading System

City Operations Center

Bulk SMS System

Patient Tracking System

Disabled Amenities

Missing and Unidentified Persons System

Remote security solutions for offices and homes.

Centralized lighting management in public areas to make energy saving.

Remote management to agricultural operations such as irrigation and fertilization to lead energy efficient solutions.

Next generation payphone system with video calls and Internet access to public areas.

Kiosks installed to display public information relating to city matters through a broadcast management system.

Self-service kiosks with a convenient platform to access various public services.

Application to recycle waste instantly allowing remote monitoring of the remaining capacity of waste sites.

Optimized disposal operations for efficient use of resources.

Facilitating centralized readings of electricity and water meters to track illegal usage and other losses of resources.

Designed to enable instant monitoring of all solutions applied on all sites and to give early response.

Used to send warning and reminder messages to all urban dwellers.

Designed to enable the detection of health problems among urban dwellers.

Smart technologies for disabled people in the public spaces.

Designed to find missing people and lost animals by helping to trace back their movements.

Özge Yalçıner Ercoşkun 143

Based on these categories, ICT sector has become an essen- tial part of the Smart Economy, in particular Smart People, since it directly affects the ever-changing cities. Turkey has great potential for growth in telecommunications and the In- ternet sector with its young population and high technology adoption rates. The Internet and social media have become an important part of public life in Turkey. Turkish users spent the second-most amount of time online per capita in Europe, after the UK. More than 90% of the population ages 15-64 have a Facebook account, and more than 70% are on Twitter. Nearly, all of the population owns mobile phones and most of them are the latest technology smart phones. The aggregate attention on digital media is immense. The key explosions are in e-commerce. The e-commerce companies in Turkey already have millions of dollars in monthly revenues.

The BTK has the legal instruments, financial resources, and institutional capabilities to encourage this growth. However, this potential has not sufficiently translated into the expan- sion of the sector. The statistics of Turkey on the number of 3G, xDSL, mobile broadband and fiber subscribers, the length of fiber cable infrastructure is still low or average in OECD countries.

Turkey’s rapid urbanization has transformed the country de- mographically and economically. Urban population has grown from 25% (1950s) to 75% (today). Urban growth has shifted from Turkey’s primate cities to its secondary cities over the last ten years with the impact of the Metropolitan Municipal- ity Law. The thematic maps produced in this paper show that changes in demographic growth patterns and usage of internet broadband, xDSL and mobile 3G technology, Anatolian Tigers such as Konya, Adana, Mersin and Gaziantep are catching up İstanbul, Ankara and İzmir’s statistics. We see Kayseri in the thematic maps in addition to these cities in the rankings of the length of fiber cable in kilometers. As another Anatolian city, Karaman near Konya launched the country’s first smart city project, all the smart applications operating in the city are managed by a single platform and one operation center.

However, Turkish cities have lots to do in Smart Environment and Smart Mobility categories. Green urban planning, green buildings and green energy issues, clean and non-motorized options need to be prioritized and action plans need to set a clear goal to become Smartest City in the world. Turk- ish cities would also make use of the ultimate technology to acquire citizen involvement which offers a range of soft- ware and mobile tools for cities to communicate and engage citizens in a dialog about city projects. Turkish citizens know and contribute very little about city projects. Furthermore, ICT-based safety systems in public spaces save many lives with smart protection against natural and human-made disasters and crime events in Turkish cities. In Smart Mobility category,

the journey to becoming a smart city will stall without a ma- jor commitment to supporting efficient, multi-modal transit. Electric vehicles and the appropriate infrastructure should be placed in many smart-city strategies.

For paving the way towards smart cities, clear strategies should be put with the stakeholders. The first one should be the creation of a vision for the city and citizen engagement. Utilizing ICT and social media can be the way to build ac- tive participation. Each city has its own needs depending on density, topography, existing infrastructure etc. Cities should develop their own smart and sustainable targets and choose indicators on common baselines. Once a city has established clear goals and the indicators to measure their progress, it needs to build plans for longer-term. Starting with a pilot project can test the feasibility of its planned program.

For Smart People category, building a smart community today involves education and trainings at all levels. The subsidies will continue at national level and also the EU is trying to ensure that smart solutions for cities can be explored, implemented and replicated. Many projects for smart cities and communi- ties were funded under Horizon 2020 in the environment which can be an opportunity for the Turkish cities in emerg- ing economies.

PLANLAMA144

REFERENCES

Accenture (2011). Building and managing an intelligent city. Retrieved Feb- ruary 13, 2016 from http://www.accenture.com/SiteCollectionDocu- ments/PDF/Accenture-Building-Managing-Intelligent-City.pdf.

Angelidou, M. (2015). Smart cities: A conjuncture of four forces. Cities, 47, 95-106.

Angelidou, M. (2014). Smart city policies: A spatial approach. Cities, 41, S3- S11.

Atzori, L., Iera, A., & Morabito, G. (2010). The internet of things: A survey. Computer Networks, 54(15), 2787–2805.

Bakıcı, T., Almirall, E., & Wareham, J. (2013). A Smart City initiative: The case of Barcelona. Journal of Knowledge Economy, 4(2), 135–148.

BTK (2014). The Bulletin of the Yearly Statistics at Provincial Level for Turk- ish Electronic Communications Market. Retrieved July 29, 2015 from http://btk.gov.tr/en-US/Pages/Statistics

Caragliu, A., Del Bo, C., & Nijkamp, P. (2009). Smart Cities in Europe. In 3rd Central European conference in regional science – CERS 2009, 7–9 October Kosice, Slovak Republic.

Castells, M. (2009). Communication power: Oxford University Press. Chourabi, H., Nam, T., Walker, S., Gil-Garcia, J. R., Mellouli, S. & Nahon,

K. (2012) Understanding Smart City initiatives: An integrative frame- work. In 45th Hawaii international conference on system sciences (pp. 2289–2297), 4–7 January Maui, HI.

Coaffee, J. (2008). Risk, resilience, and environmentally sustainable cities. En- ergy Policy, 36, 4633-4638.

Correia, L. M., & Wunstel, K. (2011). Smart Cities applications and require- ments. White Paper of the Experts Working Group, Net!Works Europe- an Technology Platform. Retrieved February 13, 2016 from http://www. scribd.com/doc/87944173/White-Paper-Smart-Cities-Applications

Demographia World Urban Areas (2015). 11th Annual Edition, Retrieved August 27, 2015 from http://www.demographia.com/db-worldua.pdf.

Dirks, S., Keeling, M., & Dencik J. (2009). How smart is your city. IBM Insti- tute for Business Value. Retrieved August 27, 2015 from http://www-03. ibm.com/press/attachments/IBV_Smarter_Cities_-_Final.pdf

EU. (2004). Urban design for sustainability, Final Report of the Working Group on Urban Design for Sustainability to the European Union Ex- pert Group on the Urban Environment, Austria

Ercoskun, O.Y. (2016). “Smart Technologies for Sustainable Mobility”. In E.V. Ocalir-Akunal (Ed.), Using Decision Support Systems for Transporta- tion Planning Efficiency (pp.288-310). Hershey, PA: Information Sci- ence Publishing.

Ercoskun, O.Y. (2012). “A Paradigm Shift towards Urban Resilience”. In O.Y.Ercoskun (Ed.), Green and Ecological Technologies for Urban Plan- ning: Creating Smart Cities (pp.1-16). Hershey, PA: Information Science Publishing.

Ercoskun, O.Y. (2010). “Green Urban Planning and Design for Smarter Com- munities”. In A.Hallin & T.K.Gustavsson (Eds.), Organizational Com- munication and Sustainable Development (pp.41-59). Hershey, PA: In- formation Science Publishing.

Fastcoexist website, Retrieved February 13, 2016 from http://www.fastcoex- ist.com/3038818/the-smartest-cities-in-the-world-2015-methodology, http://www.fastcoexist.com/1679127/the-top-10-smart-cities-on-the- planet

Foursquare website, Retrieved June 24, 2015 from http://webrazzi. com/2015/04/16/fousquare-turkiye-kullanici-infografik/.

Godschalk, D.R. (2003). Urban hazard mitigation: Creating resilient cities. Natural Hazards Review ASCE, 4(3): 136-143.

Global Digital Statistics, 2016, Retrieved February 15, 2016 from http:// wearesocial.com/uk/special-reports/digital-in-2016.

Graham, S., & Marvin, S. (1999). Planning cybercities: Integrating telecom- munications into urban planning. Town Planning Review, 70(1), 89-114.

Giffinger, R., Fertner, C., Kramar, H., Kalasek, R., Pichler-Milanovic , N., & Meijers, E. (2007). Smart Cities: Ranking of European medium-sized

cities. Vienna, Austria: Centre of Regional Science (SRF), Vienna Uni- versity of Technology. Retrieved February 15, 2016 from http:// www. smart-cities.eu/download/smart_cities_final_report.pdf

Hensher, D.A., & Button, K.J, (2003). Handbook of Transport and the Envi- ronment, Elsevier Science, USA.

Hin, L. T. W., & Subramaniam, R. (2012). Creating Smart Cities with In- telligent Transportation Solutions: Experiences from Singapore. In Ö.Y. Ercoşkun (Ed.), Green and Ecological Technologies for Urban Planning: Creating Smart Cities (pp.174-190). Hershey, PA: Information Science Publishing.

Hodson, M., & Marvin, S. (2009). Urban ecological security: A new urban paradigm?. Int. Journal of Urban and Regional Research, 33(1): 193-215.

Holmgren, D. (2007). Permaculture: Principles & pathways beyond sustain- ability, Australia: Holmgren Design Services.

Hopkins, R. (2008). The Transition Handbook, Devon, UK: Green Books Hughes, S., Pincetl, S., & Boone, C. (2013). Triple exposure: Regulatory, cli-

matic, and political drivers of water management in Los Angeles. Cities, 32, 51–69.

IBM (2011). Smarter lessons from our smarter cities, Building a Smarter Planet: 9 in a Series, Retrieved August 28, 2015 from https://www.ibm. com/smarterplanet/global/files/us__en_us__smarter_cities__wsj_op_ ad_final.pdf.

Innova website, Türk Telekom and Innova launch Turkey’s first smart city project. Retrieved August 24, 2015 from https://www.innova.com. tr/en/news-detail.asp?haber=781A225F-C412-4132-EAC2-BBA- 001CA111C.

ITU website, Key Performance Indicators (KPIs) and Standards for Smart Sustainable Cities Smart Sustainable Cities Training Programme, Mod- ule SSC-3, Retrieved February 15, 2016 from https://www.itu.int/en/ ITU-D/Regional-Presence/AsiaPacific/Documents/Module%203%20 Smart%20Sustainable%20Cities%20KPIs%20Draft%20H.pdf

Kazimee, B.A. (2002). Sustainable urban design paradigm: Twenty five simple things to do to make an urban neighborhood sustainable. In: C.A. Breb- bia, C. Martin-Duque, L.C. Wasdhwa (Eds.), The Sustainable City II, (pp. 31-41). London: Witpress.

Komninos, N. (2011). Intelligent cities: Variable geometries of spatial intel- ligence. Intelligent Buildings International, 3(3), 172–188.

La Greca, P., Barbarossa, L., Ignaccolo, M., Inturri, G., & Martinico, F. (2011). The density dilemma. A proposal for introducing smart growth princi- ples in a sprawling settlement within Catania Metropolitan Area. Cities, 28(6), 527–535.

Lombardi, P., Giordano, S., Caragliu, A., Del Bo, F. C., Deakin, M., Nijkamp, P., Kourtit, K. , Farouh, H. (2012). “An advanced Triple-Helix network model for smart cities performance”, In Ö.Y. Ercoşkun (Ed.), Green and Ecological Technologies for Urban Planning: Creating Smart Cities (pp.59-73). Hershey, PA: Information Science Publishing.

TUIK (2015). Hanehalkı Bilişim Teknolojileri Kullanım Araştırması, 2015. Retrieved August 20, 2015 from http://www.tuik.gov.tr/PreHaber- Bultenleri.do?id=18660.

TT (Turk Telekom) website, Retrieved August 24, 2015 from http://www. ttinvestorrelations.com/turk-telekom-group/investing-in-turk-telekom/ turkey-telecom-sector.aspx.

Mahizhnan, A. (1999). Smart cities. The Singapore case. Cities, 16(1), 13–18. McKinsey Global Institute (2011). Big Data: The next frontier for innovation,

competition and productivity. Retrieved February 15, 2016 from http:// www.mckinsey.com/insights/ business_technology/big_data_the_next_ frontier_for_innovation.

Mobile Tech. (2013). Mobile technologies – The digital fabric of our lives. Vodafone Institute for Society and Communications, Retrieved August 5, 2014 from http://www.vodafoneinstitut.de/uploads/media/Stud- ies_komplett_FINAL_von_250913.pdf.

Munuzuri, J., Larraneta, J., Onieva, L., & Cortes, P. (2005). Solutions appli- cable by local administrations for urban logistics improvement. Cities, 22(1), 15–28.

Nam, T., & Pardo, T. A. (2011). Conceptualizing Smart City with dimensions

Özge Yalçıner Ercoşkun 145

of technology, people, and institutions. In 12th Annual international con- ference on digital government research, 12–15 June 12–15 College Park, MD.

Neirotti, P., De Marco, A., Cagliano, A. C., Mangano, G., & Scorrano, F. (2014). Current trends in Smart City initiatives: Some stylised facts. Cit- ies, 38, 25-36.

Newman, P., Beatley, T., Boyer, H. (2009). Resilient cities responding to peak oil and climate change. Washington DC: Island Press.

Newman, P., & Jennings, I. (2008). Cities as sustainable ecosystems principles and practices. Washington DC: Island Press.

Odendaal, N. (2003). Information and communication technology and local governance: understanding the difference between cities in developed and emerging economies. Computers, Environment and Urban Systems, 27(6), 585-607.

Öztuzsuz, T. (2015, May 5). Akıllanan Dördüncü Büyük Kent Antalya. Sa- bah Akdeniz Haberleri, Retrieved from http://www.sabah.com.tr/ak- deniz/2015/05/28/akillanan-dorduncu-buyuk-kent-antalya.

Pinzon, L. (2013). Beyond connectivity: The impacts of social media in urban development in Puerto Ayora, Ecuador. (DPU Working Paper No. 160). London: The Bartlett Development Planning Unit.

Raven, J. (2010, May). Cooling the public realm climate-resilient urban de- sign. Paper presented at the meeting of 1st World Congress on Cities and Adaptation to Climate Change Resilient Cities 2010, Bonn, Germany.

Resilientcity website, Retrieved August 8, 2010 from http://www.resilientcity. org/index.cfm?pagePath=RESILIENCE&id=11449.

Saco, D. (2002). Cybering democracy: Public space and the Internet (Vol. 7). University of Minnesota Press.

Saavendra, C., Budd, W. W. (2009). Climate change and environmental plan- ning: Working to build community resilience and adaptive capacity in Washington State, USA. Habitat International, 33, 246-252.

Sasaki, M. (2010). Urban regeneration through cultural creativity and social inclusion: Rethinking creative city theory through a Japanese case study. Cities, 27, S3–S9.

Schenkkann, N. (2014).Turkey as a battleground state for regulating the inter- net, the struggle for Turkey’s internet. A Freedom House Special Report, New York.

Steria-Smart City (2011). Smart Cities will be enabled by Smart IT. Retrieved February 15, 2016 from http:// www.steria.com/uk/fileadmin/assets/ media/STE3899Smart_Cities_brochure_08_APP.PDF

The Climate Group, ARUP, Accenture, & The University of Nottingham (2011). Information marketplaces the new economics of cities. Retrieved June 15, from http://www. theclimategroup.org/_assets/files/informa- tion_marketplaces_05_12_11.pdf

Think (2011). Smart Cities initiative: How to foster a quick transition to- wards local sustainable energy systems. Retrieved June 15, from http:// www.symple.tm.fr/uploaded/pdf/ THINK_smart_cities.pdf

Tolley, R. (Ed.) (2003). Sustainable Transport: Planning for Walking and Cy- cling in Urban Environments, CRC Press, USA.

Toppeta, D. (2010). The Smart City vision: How Innovation and ICT can build smart, liveable, sustainable cities. The Innovation Knowledge Foun- dation. Think!Report, 005/2010.

Tunç, A. (2014). Can pomegranates replace penguins? Social media and the rise of citizen journalism in Turkey, the struggle for Turkey’s internet. A Freedom House Special Report, New York.

Washburn, D., Sindhu, U., Balaouras, S., Dines, R. A., Hayes, N. M., & Nel- son, L. E. (2010). Helping CIOs understand ‘‘Smart City’’ initiatives: De- fining the Smart City, its drivers, and the role of the CIO. Cambridge, MA: Forrester Research, Inc. Retrieved June 15, from http://public.dhe. ibm.com/partnerworld/pub/smb/smarterplanet/forr_help_cios_und_ smart_city_initiatives.pdf

Wheeler, S., (2003). “Planning Sustainable And Livable Cities”, The City Reader, 3rd Edition, (eds.) LeGates, R.T. ve Stout, F., Routledge Urban Reader Series, New York, 487-496.

UNISDR, United Nations International Strategy for Disaster Reduction.

(2010). Making cities resilient: My city is getting ready. 2010-2011 world disaster reduction campaign report. Geneva: UN.

Vergunst, P. J. B. (2002). The potentials and limitations of self-reliance and self-sufficiency at the local level: views from southern Sweden. Local En- vironment, 7(2), 149-161.

Velibeyoğlu, K. & Yiğitcanlar, T.(2009). Information and communication technology for E-regions. In M. Khosrow-Pour (Ed.), Encyclopedia of Information, Science and Technology (pp. 1944-1949). Hershey, PA: In- formation Science Publishing.

Yalciner, O. (2002). Urban Information Systems for Earthquake-Resistant Cities. ArcUser magazine of ESRI, 4, 24-25.

PLANLAMA146

Accelerating-high-speed-broadband-in-Turkey_English.pdf

Foundations, barriers, and policy considerations

Strategy& is part of the PwC network

Accelerating high-speed broadband in Turkey

2 Strategy&

Dubai

Mahmoud Makki Principal +971-4-436-3000 mahmoud.makki @strategyand.ae.pwc.com

Jad El Mir Manager +971-4-436-3000 jad.elmir @strategyand.ae.pwc.com

Istanbul

Bahjat El-Darwiche Partner +90-212-376-5356 bahjat.eldarwiche @strategyand.ae.pwc.com

Mustafa Uçar Principal +90-212-376-5356 mustafa.ucar @strategyand.tr.pwc.com

London

Alastair Macpherson Partner +44-20-7583-5000 alastair.macpherson @strategyand.uk.pwc.com

Contacts

3Strategy&

The Strategy& Digital Prosperity Project brings together leading experts to provide thought leadership at the intersection of technology and economics. The project has developed measures of digitization and digital maturity to better inform policymakers and business leaders on how to use digitization to further economic and social progress.

The members of the project for the Turkey broadband study are: Bahjat El-Darwiche, Alastair Macpherson, Mahmoud Makki, Mustafa Ucar, and Jad El Mir. The following external experts also participated in the project for this report: Andrew Haire, Professor Raul Katz, and Paul Reynolds.

Bahjat El-Darwiche leads the communications, media, and technology practice in the Middle East for Strategy&. He is a partner with PwC Middle East. He provides advice to policymakers and regulators in the areas of sector policy, regulatory management, sector development, socioeconomic impact, and public–private partnerships. He advises telecom and technology players on business development and strategic investments, corporate strategy, digitization, governance, operating models, and restructuring.

Alastair Macpherson is a specialist in regulatory policy, economics, and strategy for the telecoms sector with Strategy&. Based in London, he is a partner with PwC UK, and is the leader of the firm’s International Telecommunications Regulatory Centre of Excellence.

Mahmoud Makki is a member of the communications, media, and technology practice with Strategy&. He is a principal with PwC Middle East, based in Dubai. He specializes in integrated broadband strategy development, commercial turnaround, analytical marketing, and business performance management for telecom operators, as well as over-the-top service providers. He also advises policymakers and regulatory bodies on national broadband plan development and execution.

Mustafa Uçar is a member of the communications, media, and technology practice with Strategy&. He is a principal with PwC Turkey, based in Istanbul. He brings extensive experience in the Turkish telecommunications and ICT markets. With his strategy-to-execution experience, he provides advice to telecom players about business planning, operating model and organization design, integration, and large-scale transformation programs.

Jad El Mir is a member of the communications, media, and technology practice with Strategy&. He is based in Dubai and is a manager with PwC Middle East. He works on strategy, business development, and transformation for telecom operators and technology clients. He focuses on customer-facing strategies, large-scale business transformations, and operating models. He also advises clients on sector strategies and national broadband plans.

About the Strategy& digital prosperity project

4 Strategy&

Andrew Haire is a globally renowned expert in telecommunications policy and regulation, with a 30-year career spanning the U.S., Europe, and Asia. Between 2000 and 2010 he served as the deputy-DG of the Infocomm Development Authority (IDA) in Singapore — the local policy, regulatory, and sector development body of the country — during which time he led the delayering of the telecom-sector structure leading to the structural separation of SingTel.

Professor Raul Katz is adjunct professor in the Division of Economics and Finance at Columbia Business School. He is a globally renowned expert in telecommunications economics and finance, with special focus on national broadband network investments and demand stimulation. He has published five books and more than 20 articles; and he advises multilateral organizations such as the World Bank and the International Telecommunications Union and industry bodies such as GSMA and the World Economic Forum.

Paul Reynolds is a leading international telecom company chief executive with deep experience in separating networks. He has been a leader in two well-known separation cases: first, as CEO of Telecom New Zealand, overseeing the de-merger of the Chorus access network business from Telecom NZ and the creation of two completely separate listed companies; second, as CEO of BT Wholesale, leading BT’s functional separation program with the spin-off of the access network and operations, creating BT Openreach.

This report (“Report”) has been prepared by Strategy& (formerly Booz & Company), part of the PwC network (“Strategy&”), for Türk Telekomunikasyon A.S. (“Türk Telekom”). Although funded by Türk Telekom, this report represents Strategy&’s independent analysis and perspective. The contributing experts reviewed this report to provide an additional independent test of its findings.

This report contains information about mechanisms to accelerate national broadband in Turkey. The report is for informational purposes only and is not advice and should not be treated as such.

In compiling this report, Strategy& has used publicly available information and information in other reports and materials provided by third parties. Strategy& does not assume responsibility for and has relied, without independent verification, on the accuracy and completeness of any such information available at the time of writing the report. Any results from the analysis contained in this report are therefore based on Strategy&’s best professional judgment using the available information; it is possible to reach other reasonable judgments and recommendations.

The reader agrees that Strategy& neither owes nor accepts any duty or responsibility to it, whether in contract or in tort (including but not limited to, negligence and breach of statutory duty), and shall not be liable in respect of any loss, damage or expense of whatsoever nature which is caused by any use (or non-use) the reader may choose to make of this report and its contents, or which is otherwise consequent upon the gaining of access to this report by the reader, including any action or decision taken as a result of such use (or non-use).

Strategy& excludes all representations, warranties, undertakings and guarantees relating to the report. Specifically, Strategy& does not represent, warrant, undertake or guarantee (i) that the information in the report is correct, accurate, complete or non-misleading and (ii) that the use of guidance in the report will lead to any particular outcome or is fit for any particular purpose.

All copyrights and other proprietary rights in this report are reserved and no license, expressed or implied, is granted or intended.

Contributing experts

Important notice from Strategy& (Formerly Booz & Company)

5Strategy&

• Fixed telecom infrastructure investments in Turkey over the last 10 years amount to TL 13.5 billion, of which the incumbent accounts for 81 percent (i.e., TL 10.9 billion from 2005 to 2014).

• There is fixed broadband coverage in Turkey for 98 percent of premises at entry level, 66 percent at high-speed, and 42 percent at ultra-fast. This is at par with, or better than, EU15 averages that stand at 97 percent, 55 percent, and 44 percent, respectively.

• With significant fiber to the cabinet (FTTC) buildup at less than 250 meters from customer premises, 10.7 million premises in Turkey can experience speeds of 70 Mbps upon demand with limited incremental investment.

• Turkey has a fiber coverage gap of 57 percent in suburban and rural areas.

• The Turkish wholesale broadband access market is developed. Essential wholesale broadband access services and regulations are already in place (local loop unbundling or LLU, bitstream, duct sharing, co-location, etc.), and the incumbent’s functional separation is nurturing competition, leading to a Herfindahl-Hirschman Index (HHI) of 51 percent (down from 72 percent in 2010) in the fixed broadband market.

• The country’s entry-level fixed broadband penetration is still at 44 percent, compared to 70 percent in the EU. With high-speed broadband penetration at 13 percent, only about 26 percent of subscribers are high- speed. Ultra-fast penetration stands at merely 0.1 percent.

• Turkey’s digital divide is significant. While the supply gap is only 2 percent (homes that lack service), the demand gap reaches 54 percent (Turkish households that could acquire broadband but do not).

• Turkey is in line with comparable markets in all pillars of the Networked Readiness Index (NRI) except for skills, where it ranks 82nd out of 143 countries.

• Only 56 percent of Turkish households own a computer, compared to 81 percent in relevant benchmark markets.

• Despite lower household incomes compared to the EU, 86 percent of Turkish households can afford high- speed broadband.

• Although 86 percent of the population can afford broadband, when that expense is combined with the expenses of a modem and a notebook computer, the number drops to 69 percent.

• The share of subscribers with higher than 10 Mbps speed increased from 19 percent to 72 percent over the past two years, reaching EU levels much faster than most comparable markets (the same evolution took 4.5 years in the EU).

Key facts and figures

Source: Türk Telekom; operators’ announcements; Information and Communication Technologies Authority (ICTA, BTK in Turkish); World Cellular Information Service (WCIS); World Economic Forum, Global Information Technology Report; Euromonitor; Professor Raul Katz, “The Impact of Taxation on the Digital Economy,” (ITU, 2015); GSMA, “Digital inclusion and mobile sector taxation 2015;” European Commission; European Digital Commission; Turkish Statistical Institute (TÜIK in Turkish); Strategy& analysis

6 Strategy&

Executive summary

Now is the time to accelerate the reach of national broadband in Turkey. Turkey has laid the groundwork for this acceleration. Broadband reach, speed, and affordability have been established, all abetted by a competitive market, a regulatory framework conducive to broadband development, and significant investments.

The barriers that are holding back the Turkish broadband market are: low demand among entry-level users, lagging demand for higher speeds, high taxation on devices, patchy coverage in suburban and rural areas, a challenging wholesale market, and inefficiencies in distribution. Drawing on the examples of successful policy interventions in markets similar to Turkey’s, our research recommends six policy measures that could propel the expansion of Turkey’s broadband reach by overcoming these barriers:

1 Stimulate demand by increasing broadband awareness and digital literacy, emphasizing promotion of high-touch distribution channels, and accelerating the uptake of high-speed broadband.

2 Increase relevant local content and applications, particularly those related to education, government services, and economic productivity.

3 Revise the taxation regime on broadband-related devices and services to reduce the cost of ownership, making broadband more affordable.

4 Infrastructure-based competition can be encouraged through the extension/redefinition of the regulatory holiday and introduction of other regulatory incentives.

5 Extend current subsidies (Universal Service Fund) and bridge coverage gaps in low-density areas.

6 Cooperate with municipalities to aggregate demand and ease right-of- way fees.

At this stage of Turkey’s market development, these six interventions would be more effective, and less disruptive, than alternative supply-restructuring measures such as structural separation.

7Strategy&

Building nationwide high-speed broadband access infrastructure in a timely and efficient manner would propel Turkey into a digital economy. This means providing nationwide access to high-speed Internet and high-value applications, including those that are bandwidth-hungry. This advance would support equitable socioeconomic advancement of Turkey and reinforce the country’s competitive global position, which will reap the related upside on GDP, job creation, competitiveness, and social welfare.

Turkey has the ability to make this change. The acceleration of the high-speed broadband market in Turkey by 25 percentage points is attainable within eight years with the right mechanisms. With this acceleration, 85 percent of households would have broadband by 2023, compared to 60 percent if the current conditions prevail.1 Global studies indicate that even an increase of 10 percent in broadband penetration would result in an increase of 0.3 percent to 1.4 percent in GDP growth.2 Expanding broadband creates jobs, increases worker productivity, and leads to an increase in consumer surplus. In Turkey, increasing fixed broadband uptake by 7 million homes3 could add a cumulative US$130 billion to Turkey’s GDP by 2023, and lead to an additional 0.5 million net jobs4 (see Exhibit 1, page 8).

Other leading markets around the world have benefited from effectively accelerating high-speed national broadband coverage and adoption. Their examples suggest that this change requires balance among demand stimulation, magnitude of investments, and policies and regulations. This type of balance is feasible for Turkey, and can be developed through partnership between government and the private sector.

Given the recent release of Turkey’s 64th government plan and the emphasis it puts on national broadband advancement, the recently granted LTE (Long Term Evolution) spectrum, and the imminent launch of 4.5G services by the country’s telecom operators, as well as the country’s current transition into the second stage of fiber deployment to

Turkey’s broadband opportunity

The acceleration of the high-speed broadband market in Turkey by 25 percentage points is attainable within eight years with the right mechanisms.

8 Strategy&

cover lower-density and less commercially attractive areas, a number of questions arise:

1. Where does Turkey stand in its national broadband coverage and investments? What viable foundations exist and what are the barriers to high-speed broadband advancement?

2. What lessons could be leveraged from relevant national broadband policy interventions around the world? What initiatives worked under what conditions, and what initiatives did not?

3. For the specific case of the Turkish market, what mechanisms could accelerate high-speed broadband investment and deployment, and how could these be efficiently and effectively enacted?

The answers to these questions show how Turkey’s broadband infrastructure could move forward rapidly.

Exhibit 1 Turkey’s fixed broadband penetration forecast

Note: Forecast for the status quo was based on the compound annual growth rate for penetration during the previous three years. The forecast for acceleration initiatives was based on the s-curve trend of comparable and advanced markets.

Source: World Broadband Information Service; Strategy& analysis

Status quo

With acceleration initiatives

0

10

20

30

40

50

60

70

80

90

100

2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

H o

u s e

h o

ld B

ro a

d b

a n

d P

e n

e tr

a ti

o n

( %

)

25 percentage point increase in penetration (7 million additional households)

9Strategy&

Today, around 40 million people in Turkey use the Internet daily. The country’s usage is expanding and evolving. Users are increasingly diverse, from consumers streaming an HD video (which requires a speed of roughly 10 Mbps), to startups conducting video conferencing (25 Mbps), to the government using telemedicine applications in hospitals (50 Mbps). The type of broadband that people use has also changed. Mobile broadband has been the technology getting people online; however, fixed technology remains the most reliable, cost-effective, and capable of delivering high- speed, high-capacity broadband (see “Broadband speed for Turkey”).

Broadband in Turkey today

For this paper, we define entry-level broadband as connections with speeds exceeding 2 Mbps, high-speed broadband as connections with speeds exceeding 24 Mbps, and ultra-fast broadband as connections with speeds exceeding 70 Mbps. These definitions reflect the Turkish market’s requirements for the coming years, and are in line with benchmarks from international institutions (e.g., the International Telecommunication Union or ITU, Broadband Stakeholder Group, and Gartner) and European regulators.

Currently, as Exhibit 2, page 10 shows, fixed broadband coverage in Turkey compares well with those of the European Union. Significant advances in coverage and speeds have pushed Turkey up to EU average levels. Coverage stands at 98 percent of premises for entry-level, 66 percent for high-speed, and 42 percent for ultra-fast. The equivalent numbers for

the EU15 are 97 percent, 55 percent, and 44 percent. The share of fiber coverage from total coverage is evolving rapidly to support high-speed broadband access, reaching 48 percent by 2015 including FTTC (fiber to the cabinet using existing architecture). We calculate that at speeds of 24 Mbps, the Internet needs from a bandwidth perspective of most Turkish households will be satisfied, and customer experience could be at par with international standards.

Under the current network topology, Turkey’s FTTC infrastructure can be used to augment current FTTH/B (broadband to premises) deployment. FTTC buildup at less than 250 meters from customer premises could increase Turkey’s high- speed broadband access infrastructure without the additional expense of FTTH/B coverage and would enable at least 10.7 million premises in Turkey to experience speeds of 70 Mbps upon demand.5

Broadband speed for Turkey

10 Strategy&

Fixed broadband market overview

Over the last five years,6 infrastructure based competition have attracted major investments (exceeding TL 20 billion from operators). The sector accounts for 2.2 percent of Turkey’s GDP, in line with the EU, and employs more than 100,000 people.7

There are reasons to be impressed with Turkey’s progress in fixed broadband to date. Significant advances in coverage and speed have boosted Turkey up to EU average levels over the past five years. Coverage stands at 98 percent of premises for entry-level, 66 percent for high- speed, and 42 percent for ultra-fast. The equivalent numbers for the EU15 are 97 percent, 55 percent, and 44 percent.8 The share of fiber coverage within total coverage has evolved rapidly to support high-speed broadband access, reaching 48 percent by 2015, including FTTC.9

Essential open access services and the accompanying regulations are in place - duct sharing, co-location, LLU, SLU, and bitstream. The current infrastructure-sharing market is developed. Türk Telekom, the fixed broadband incumbent, offers other companies access to ducts and to LLU, bitstream, and sub-loop unbundling (SLU) services. Of more than 1,900 applications received to share infrastructure in the past five years (2011-2015), 75 percent have been approved or are in the process of approval. Applications are generally rejected due to lack of infrastructure in a given route.10

Exhibit 2 Fixed broadband coverage in Turkey by speed

Note: Entry-level is 2 Mbps and above, high-speed is 24 Mbps and above, and ultra-fast is 70 Mbps and above.

BB = broadband

Source: Türk Telekom; World Cellular Information Service; Strategy& analysis

Entry-level

98%

EU 15Turkey

97%

High-speed

EU 15

55%

Turkey

66%

Ultra-fast

42% 44%

Turkey EU 15

(% of premises, 2015)

11Strategy&

Yet, despite these advances, fixed broadband infrastructure presents several challenges in terms of deployment economics, geographical dependency, and speed of deployment. Although the current infrastructure-sharing market is developed, the incumbent incurs access deficit from infra-structure sharing services.11 Access prices should be adjusted to ensure sustainability. Both the fixed broadband and the mobile markets have become more competitive in the past five years. The mobile market’s HHI score (which measures market concentration, with a lower score corresponding to higher competitiveness) has dropped from 40 to 37 between 2010 and 2015. The fixed broadband market has changed more dramatically; its HHI score fell from 72 to 51 between 2011 and 2015. Türk Telekom exhibited the most significant change in the fixed market competitiveness score, with its market share dropping from 84 percent to 69 percent between 2011 and 2015, confirming that the market is on the right trajectory in terms of competition development.

The current market structure with infrastructure based competiton, as well as the supportive policies and regulations, have led to significant high-speed broadband deployment. The regulatory holiday for the incumbent, exempting fiber access from the market analysis process, has incentivized Türk Telekom to increase fiber investments from 0.7 million homes in 2011 to 3.3 million in 2015, and is in effect until October 2016 or until fiber subscribers exceed 25 percent of total fixed broadband subscribers.¹²

Turkey’s regulatory framework is still geared to managing market dominance and imposes more obligations than incentives on the incumbent’s fixed business. While incentives are the regulatory holiday and the absence of virtual unbundled local access (VULA) and IP interconnection remedies, obligations include access to infrastructure, regulation of resale products and fixed calls, co-location and facility sharing, and a single geographic definition.

In short, Turkey has passed the first stage in its high-speed broadband journey: deployment in urban and economically attractive areas. It is now embarking on its second stage: universal coverage. The fiber network in urban and suburban areas has been extended; the challenge is to cover unserved suburban areas and rural areas in a cost-effective way for operators. The coverage gap is currently at 57 percent in suburban and rural areas. But as deployment reaches deeper areas in Turkey, capital expenditures per home rise sharply, and the business case becomes less attractive. Once these issues are addressed, the country will see its digitally related economic growth accelerate.

Turkey has passed the first stage in its high- speed broadband journey: deployment in urban and economically attractive areas. It is now embarking on its second stage: universal coverage.

12 Strategy&

Filling the Gaps

Turkey has two gaps to fill. First, the current topology of high-speed coverage in suburban and rural areas is patchy. Turkey has a fiber coverage gap of 57 percent in suburban and rural areas. Operators have focused primarily on deployment in dense urban, and urban areas, which is economical. Rolling out FTTH in suburban areas in Turkey is estimated to cost twice as much as in dense urban areas; in rural areas, the figure is four times as much.13 Operator-funded infrastructure expansion into those less economically viable geographies will require more guarantees to render the business case appealing.

Second, Turkey has a large demand gap due to slow uptake. The country’s entry-level fixed broadband penetration is still at 44 percent compared to 70 percent for the EU. Fifty-four percent of the Turkish population could be online but is not (see Exhibit 3).14

Exhibit 3 Fixed broadband uptake in Turkey by speed

Note: Entry-level is 2 Mbps and above, high-speed is 24 Mbps and above, and ultra-fast is 70 Mbps and above.

pp = percentage points

Source: WCIS; BTK; Türk Telekom; Strategy& analysis

Entry-level penetration

98%

44%

54 pp

70%

EU 15Turkey

High-speed penetration

EU 15

55%

Turkey

66%

13% 13%

Ultra-fast penetration

42% 44%

11% 0.1%

Turkey EU 15

Coverage Penetration

97%

27 pp

53 pp 42 pp

42 pp 33 pp

(% of households, 2015)

13Strategy&

This digital divide is significant and holding back the country’s potential economic growth. The fundamental issue is not supply; only about 2 percent of Turkish homes lack access to broadband service. However, 54 percent of households have not taken advantage of any level of service that they can already access, leading to a significant demand gap (see Exhibit 4).

Exhibit 4 Turkey’s digital divide

Note: Entry-level is 2 Mbps and above, high-speed is 24 Mbps and above, and ultra-fast is 70 Mbps and above.

Source: BTK; Türk Telekom; Strategy& analysis

Entry-level broadband

High-speed broadband

100%

Total market

0.421.2

14.0

6.8

6.6

Total broadband coverage

Supply gap

20.8

2.7

9.3

Total broadband demand

0.4 (2%)

11.5 (54%)

Digital divide

Supply gap 11.9

Demand gap

Demand Gap = 11.5 million (54% of total households)

High-Speed Demand Gap = 11.2 million (53% of total households)

14 Strategy&

The three principal reasons for the gap include low digital literacy, limited device penetration, and lack of relevant local content. Turkey is in line with comparable markets15 in all pillars of the WEF’s Networked Readiness Index except for skills, where it ranks 80th out of 143 countries.16 In terms of devices for broadband access, only 56 percent of Turkish households own a computer, compared to 81 percent in comparable markets (see Exhibit 5).17 For people in Turkey who are connected to broadband, there is a lack of content relevant to their needs and interests. Turkey has advanced e-health services and a high percentage of its population is on Facebook. However, the country has room for improvement in developing relevant e-government services and retail offerings. Turkey’s content maturity18 is roughly 10 points

Exhibit 5 Skills and device penetration in Turkey

1 Ranking is from best (i.e., 1) to worst (i.e., 143)

Source: Euromonitor; World Economic Forum, Global Information Technology Report (WEF, GITR); WCIS; Strategy& analysis

Desktop & portable computer ownership (% of households) (2010–2014)

2015 NRI ranking1 across pillars

+6.1%

2014

56.0%

2010

44.2%

2014

81.4%

2010

73.7%

2014

80.7%

2010

73.9%

Turkey

EU

Comparable markets

Ranking range of EU and comparable markets

Turkey's rank

T o

ta l

P o

lic y

& re

g u la

ti o

n

B u si

n e ss

& in

n o

va ti o

n

In fr

a st

ru c tu

re

A ff

o rd

a b

ili ty

In d

u st

ry u sa

g e

B u si

n e ss

u sa

g e

G o

ve rn

m e n t

u sa

g e

E c o

n o

m ic

im p

a c t

S o

c ia

l im

p a c t

S ki

lls

2

73

3

108

4

72

3

71

5 1

69

1

61

2

96

3

118

1

85

3

84

122

48 54 37

53

8

80

67

53 55 63

50

+2.5% +2.2%

15Strategy&

lower than that of comparable markets on a 100-point scale. Increasing its content maturity, through the creation and promotion of relevant content, could raise Turkey’s Internet penetration to more than 70 percent of the population.

Turkey’s demand gap is relevant at all speeds. Even homes that are connected to the Internet are largely not opting for the higher speeds. With current high-speed broadband penetration at 13 percent, only about 26 percent of subscribers are on high-speed. Ultra-fast penetration stands at merely 0.1 percent.19 However, the share of subscribers with higher than 10 Mbps speed increased from 19 percent to 72 percent over the past two years, reaching EU levels much faster than most, if not all, comparable markets. The same evolution took 4.5 years in the EU (see Exhibit 6).20

Exhibit 6 Turkey has rapidly attained EU levels of fast broadband subscribers

Source: Türk Telekom Wholesale (December 2015), Superonline Fiber subscribers (Dec 2015) , Cable subscribers (June 2015); EU Commission; European Digital Commission; BTK market reports; Turkish Statistical Institute; Türk Telekom, Strategy& analysis

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

9% 13%

Q3 2008

Q1 2012

10%

0%

13%

23%

8%

16%

5%

49%

56%

64%

39%

Q3 2009

Q3 2010

Q1 2013

14%

4%

Q3 2011

2%

Q1 2011

45%

59%

29%

8%

47%

28%

72% 72%

74%

Q3 2013

66%

19%

70%

Q1 2009

EU

Turkey

F ix

e d

B ro

a d

b a

n d

S u

b s c

ri p

ti o

n s w

it h

h ig

h e

r th

a n

1 0 M

b p

s s

p e

e d

(%

o f

to ta

l)

Around 2 years

Q3 2014

4.5 years

Q4 2015

Q3 2012

Q1 2008

Q1 2010

Q1 2014

Q1 2015

16 Strategy&

Affordability of access offering is not an issue. Turkey’s high-speed broadband prices are lower than other EU nations’ prices, even when purchasing power parity is taken into account. Despite lower household incomes compared to the EU, 86 percent of Turkish households can afford high-speed broadband, and the prices are reasonable compared to other essential household expenses such as water, gas, and electricity.21

Nonetheless, even though high-speed broadband access itself is relatively affordable in Turkey, high taxes on devices put connection out of the reach of many people. Although 86 percent of the population can afford broadband, when that expense is combined with a modem and a notebook computer, the number drops to 69 percent.22

To address slow uptake, many government-led and operator-led demand stimulation initiatives are under way. The government is opening thousands of free Internet centers and education centers and has distributed about 1.3 million devices to students and teachers. Commercial operators have joined the education and school initiatives, and are providing free coverage packages to first-time users. However, operators’ commercialization of fiber deployment across value proposition and sales process needs to improve. Broadband packages tend to be sold on volume, rather than speed, which has led consumers to dismiss speed as unimportant. Fiber prices are significantly reduced through promotions, to prices lower than ADSL (asymmetrical digital subscriber line) in some cases, showing that fiber investment is not properly commercialized. Commercial staff input has not been properly taken into account in demand planning exercises. Coordination between field operations and sales has been suboptimal. In addition, the broadband multi-play market is still in its infancy; most products are sold separately as voice, data, and content services, or commercially camouflaged into a bundle with no single bill across services.

17Strategy&

Turkey’s broadband demand and supply s-curves are not far apart. With supply exceeding demand, it is now time to accelerate Turkey’s demand s-curve to catch up with supply, while smartly pushing supply to fill the remaining gaps with the optimal technology.

To bring about the next stage of universal high-speed broadband coverage, identifying the measures that are appropriate for the Turkey’s market is key. Nations typically advance their broadband capabilities with three goals in mind: increasing the health of the sector, increasing availability of infrastructure, and optimizing return on investments. They accomplish these goals using three acceleration models: supply restructuring, infrastructure investment, and demand stimulation (see Exhibit 7, page 18). The appropriate choice for a given nation depends on the current condition of its broadband market, and on whether the policy objective is to promote socioeconomic development or to correct a market failure. The Turkish market will need well-tailored strategies to advance the second stage of broadband deployment and bridge the digital divide in order to make the Internet more attractive and affordable to the general population.

Supply restructuring. Policymakers typically consider supply restructuring when trying to fix problems of limited broadband supply and an unresponsive incumbent that monopolized the sector. Should policy intervention aim to restructure the broadband supply market, three models have been observed across international case studies: functional or structural separation of the incumbent operator; creation of a co-owned company among industry players and government (NetCo); or creation of a government-owned national broadband company (NBNCo).

Looking to the future

18 Strategy&

Supply restructuring policy interventions that accelerate national broadband deployment typically occur at early stages of the high-speed broadband market development. Singapore, for example, had only 4 percent high-speed broadband coverage and 1 percent penetration when it delayered its broadband industry structure in 2009. In New Zealand, the figures were 4 percent and 0 percent in 2009. Turkey’s current figure of 66 percent high-speed broadband coverage is higher than that of any of the benchmark nations when they restructured their markets, and its current figure of 13 percent penetration trails only that of Sweden when it separated its telecom incumbent.23

Exhibit 7 National broadband acceleration models

ExamplesModels

S o

c io

-e c o

n o

m ic

d e ve

lo p

m e n

t a g

e n

d a

M a rk

e t

fa ilu

re r

e a c ti o

n

Objective

Demand stimulation

Internet education (internet cafes and centers, digital literacy programs, school courses)

End-user subsidies (computers, tickets for SMEs, tax exemptions)

Public benefit applications (e-education, e-gov. services, economic benefit applications)

- National Education Development Fund digitizing and distributing textbooks to schools in Brazil

- UK government’s broadband voucher scheme for SMEs

- Indian government’s Common Services Centers Scheme

Infrastructure investments

Indirect subsidies (tax benefits, regulatory holidays, loan benefits)

Direct subsidies (USF model for rural areas, grants)

Community-based networks (municipality-funded networks, syndicated community networks)

- Tax incentives and low-cost loans in Japan

- Rural Broadband Initiative in New Zealand

- US government grant to operators

- IOptinet in Bristol, Virginia - Stokab in Sweden - Citynet Amsterdam

Supply restructure

Separation (functional separation, structural separation)

Network Company (NetCo) (co-owned network company among industry players and governments)

National Broadband Company (NBNCo) (government-owned network company)

- Skanova in Sweden - Openreach in UK

- NetCo in Singapore - New Chorus in New Zealand

- NBNCo in Australia

Note: SMEs = small- and medium-sized enterprises

Source: Strategy& analysis

19Strategy&

Supply restructuring as a mechanism to accelerate high-speed national broadband infrastructure deployment requires significant investment and execution excellence to realize objectives. Australia’s national broadband network initiative, for example, is far from delivering its promised impact, despite budget overruns and stretched execution time lines. In Singapore, OpenNet and its shareholders (CityNet, NetLink Trust, and SingTel) submitted a consolidated application in 2013 for CityNet (wholly owned by SingTel) to acquire 100 percent of OpenNet. Key regulatory requirements will continue to be upheld in order to maintain competition in the market and benefit the public interest. IDA (Infocomm Development Authority of Singapore) further imposed additional regulatory conditions to safeguard competition, and noted industry and public concerns over OpenNet’s performance issues.24

Infrastructure investment. The mechanism of infrastructure investment has more relevance for Turkey, since it directly addresses the problem of how to reach the final, most hard-to-reach 2 percent of homes with no service, and the 34 percent that lack high-speed service. Government can provide services that are uneconomical for any single company, either through direct subsidy, indirect subsidy and regulatory incentives, or community-based networks.

Coordinated approaches between governments and operators have produced success stories. Such approaches can be indirect for example, Spain’s decision to remove significant market power (SMP) regulation in areas with heavy broadband competition and introduce a speed-based regulatory holiday exempting operators from sharing infrastructure that achieves port speeds greater than 30 Mbps; or direct, such as France’s co-financing agreement with the incumbent to cover less densely populated areas. At the city or state level, the Swedish capital, Stockholm, created Stokab to develop an open and neutral fiber network; and municipalities in the Netherlands aggregate demand from 30 to 40 percent of households in the region for Reggefiber to deploy.25

Supply restructuring as a mechanism to accelerate high- speed national broadband infrastructure deployment requires significant investment and execution excellence to realize objectives.

Spanish regulation has reduced roll-out costs for operators and has incentivized deployment in rural areas. In the most competitive cities, FTTH regulation has been removed, allowing Spanish operators to better monetize super-fast broadband deployment.26

In Stockholm, the Stokab network company exclusively leases dark fiber on equal and transparent terms. By 2015, FTTB/H coverage reached 45 percent and penetration reached 26 percent.27

20 Strategy&

Demand stimulation. The approach that provides the greatest return for the least investment in Turkey is demand stimulation. It addresses the identified problem of the demand gap and provides concrete, immediate techniques to extend the socioeconomic benefits of broadband connection to the entire country.

Many countries have enjoyed successful results with moderate effort and investment. For example, education programs can have a wide reach, such as Tamil Nadu’s partnership with the Ford Foundation in India to launch a rural digital literacy program, connecting over 10,000 people since its inception. To make Internet access more affordable, Malaysia established 2,522 broadband centers, enabling computers to be shared by many citizens, ensuring high usage of each device.

To extend greater public benefits via the Internet, governments can digitize several functions simultaneously, such as healthcare, pensions, etc. Estonia introduced a universal electronic identification card that has become the key to access government services. Internet penetration in Estonia has exceeded 80 percent, and 75 percent of people obtain public administration information online. South Korea provided homeless shelters with free smartphones and wireless Internet services, allowing homeless citizens to search for jobs from any location, as well as reconnect with their families.

21Strategy&

For a decade, nations around the world have been pursuing national broadband acceleration plans. Motives have included increasing the health of the telecom sector, developing the national broadband infrastructure, and optimizing return on investment. We can extract five key lessons.

1 . Address barriers and stage of market development. It is helpful to think of a developing broadband market in terms of axes that show expected growth of supply and expected growth of demand. In very early stages of broadband development, market growth develops well organically (see Exhibit 8, page 22). Later, growth is optimal when governments stimulate demand to drive uptake to commercialize the high supply availability, while investing in infrastructure to push supply to non-economical areas. These patterns of growth can push the market to a mature, ideal state; a disruptive supply restructuring is necessary only to address market failure, when expected demand cannot be met due to lack of supply.

Bridging the gap: Lessons learned from other countries

The U.K. was lagging in fixed broadband penetration in 2005, when incumbent British Telecom had limited competition. In anticipation of mounting regulatory pressure, British Telecom created Openreach, a functionally separate business division, to operate its local access networks. Openreach adopted a policy of product equivalence, requiring nondiscriminatory support of all providers’ retail activities. Wholesale customers could buy services from Openreach or from BT Wholesale.

Although Openreach achieved the target coverage after a halt in investment for several years, customer satisfaction remained low, and setup costs were high. Recently, OfCom, the U.K. regulator, assessed whether BT should be compelled to structurally separate Openreach into an

independent company. The outcome was that Openreach was not forcibly spun off; instead it was compelled to offer duct sharing and pole sharing in order to stimulate infrastructure sharing. It also was compelled to meet tougher service standards for wholesale broadband. In terms of governance, the BT Group will be the holding company of Openreach, with a separate board of directors and financial reporting. BT claimed that it had voluntarily accepted further regulation in order to settle the matter speedily, and said it will proceed with fiber investment plans that keep the U.K. at the forefront of fiber deployment in the G20. In summary, British regulators are avoiding structural separation and adopting other avenues such as healthy regulations and quality of service guarantees as better alternatives.

22 Strategy&

2. Build competitiveness through industry structure. The national benefit from competitiveness, in most cases, stems from broader coverage and greater adoption of high speed. South Korea, Japan, and the U.S., with infrastructure-based competition, are among the world leaders in average download speed and Fiber to the X (FTTX) coverage (see Exhibit 9, page 23). Nations with a common wholesale and a structurally separate network, such as New Zealand, Australia, and South Africa, lag well behind in both categories. The impact of industry structure on increasing average download speed and fiber coverage helps define the optimal national broadband plan.

Exhibit 8 National broadband models selection approach

Source: Strategy& analysis

Low

Expected demand

Typical evolution of supply and demand

Ideal stateDemand stimulation

Infrastructure investments

Supply restructuring

E x p

e c

te d

s u

p p

ly

N a sc

e n

t D

e ve

lo p

e d

A d

va n

c e d

Medium High

Demand stimulation & infrastructure investment

Infrastructure investments

Supply restructuring

Organic market development

Demand stimulation

- Allow the market to develop organically since it is still in its early development stages

- Introduce demand stimulation mechanisms to drive demand uptake to commercialize the high supply availability

- Introduce demand stimulation mechanisms coupled with infrastructure investments to push supply to non-economical areas

- Address market failure of not meeting expected demand due to lack of supply availability, through a disruptive supply restructure

- Inject infrastructure investments to increase supply to non-economical areas where demand is available

Demand stimulation & infrastructure investment

Organic market development

23Strategy&

Exhibit 9 Impact of industry structure on efficiency

1 Although FTTC rollout exceeded 80% of premises in New Zealand, it was superseded by the government co-funded FTTN network after 2011 (which now reached 37% of premises). Chorus was therefore disincentivized from upgrading the FTTC network beyond ADSL2+ technology, so it’s peak speed is only around 20 Mbps

Source: Akamai; Analysys Mason; Strategy& analysis

Common wholesale network

Obligatory wholesale sharing at cost

Infrastructure based competition

Common wholesale network

Obligatory wholesale sharing at cost

Infrastructure based competition

0

10

20

30

40

50

60

70

80

90

100 South Korea

U.S.Sweden

France

Japan

New Zealand 1

Australia

Italy

Spain

Germany

South Africa

UK

0

2

4

6

8

10

12

14

16

18

20

22

Sweden

U.S.

Australia

South Africa

South Korea

Italy

Germany Spain

France

Japan

UK

A ve

ra g

e d

o w

n lo

a d

S p

e e d

( M

b p

s)

F T

T X

C o

ve ra

g e

( %

o f

p re

m is

e s)

New Zealand

3. Be diligent about aligning stakeholders. The incumbent is the natural partner for the governments’ national broadband plan. The capabilities that the incumbent brings in terms of fixed broadband infrastructure and operations should be leveraged in the government’s agenda. As Exhibit 10, page 24 and the Singapore case study indicate, the success of a national broadband plan is highly correlated with fixed incumbent involvement.

24 Strategy&

Singapore was concerned about being left behind amidst the emergence of national networks in South Korea and Japan, and chose to create a separate NetCo to design, build, and operate the passive national fiber infrastructure. IDA, the regulator, championed network open access to incentivize operators to provide competitive access and prices. The government issued a grant to help build a national FTTH network in 2008. The contract was awarded to a structurally separate NetCo, and a separate contract was awarded six month later to an operationally separate operating company for active access services.

OpenNet succeeded in achieving coverage and penetration targets due to

proper regulatory support and risk mitigation. Frequent collaboration between public and private stakeholders led to strong alignment among the government, the regulator, and network operators. The consortium approach spread risk and broadened the experience among several stakeholders. However, OpenNet took three years to launch, and the incumbent, SingTel, neither aligned with nor encouraged NetCo’s objectives.

The Singapore experience illustrates the importance of close collaboration between public and private stakeholders to design and execute national broadband acceleration mechanisms.

Exhibit 10 Correlation between incumbent’s involvement and national broadband plan success

1 Success is defined as the ability to reach aspirational targets through the supply restructuring model

Source: Strategy& analysis

HighNone / Low

Incumbent’s involvement / alignment

Success1

R e

s u

lt

Failure

Success of national broadband plan is highly correlated with incumbent involvement

UKSweden

Italy

Qatar Australia

New Zealand Singapore

25Strategy&

4. Beware the risks of disruptive supply restructuring interventions. Countries that opt for creating NetCos or NBNCos face fundamental risks to commercial performance and operations, as well as highly complex regulatory risks (see Exhibit 11). Australia offers a cautionary tale; the government’s initial cost estimates, made without telco involvement, were subject to controversy, and the NBN’s budget grew from $30 billion to $52 billion. The technology and regulatory framework treated broadband as a “best effort” service, resulting in poorer quality service. By 2015, Australia’s rank in national broadband services fell from 25th to 42nd (according to the WEF’s Networked Readiness Index), mainly due to ongoing delays in the rollout of the project, and the national coverage and speed targets that were initially set were revised downwards.

Exhibit 11 National broadband acceleration risks

Note: SMEs = small- and medium-sized enterprises

Source: Strategy& analysis

Commercial risks

1 Slower than expected broadband uptake

2 Low differentiation capabilities

3 Slow time to market

4 Suboptimal customer experience

5 Alternative technologies cannibalization

6 Weak fixed business competitiveness

7 Complexity in managing OLO demand

1 Limited subsidy amount

2 Absence of key regulatory requirements

3 Limited support for right of way

4 Stringent service requirements from olos

5 Absence of exit / revaluation terms

6 Change in government regulatory position over time

7 Revocation of subsidy

8 Rising regulatory costs

1 Budget overrun

2 Launch delays

3 Inaccurate coverage projections

4 Internal wiring complexity

5 Limited capability of suppliers

6 Suboptimal delivery capabilities

7 Difficulties in capex planning

8 Lack of back-up network

Operational risks Regulatory risks

26 Strategy&

For any national broadband company, commercial risks include slow broadband uptake, slow time to market, suboptimal customer experience, cannibalization of alternative technologies, low differentiation capabilities, weak fixed business competitiveness, and complexity in managing demand from other licensed operators. Operational risks include budget overrun, launch delays, inaccurate coverage projections or capital expenditure forecasts, internal wiring complexity, limited capability of suppliers, suboptimal delivery capabilities and lack of back- up network. Regulatory risks include limited subsidies, absence of key requirements, limited support for right of way, stringent service requirements from OLOs, absence of exit or revaluation terms, change in government regulatory position, and revocation of subsidies.

Even with these risks negotiated and addressed, national broadband companies have had limited impact on broadband penetration. As Exhibit 12, page 27 demonstrates, governments in Australia, New Zealand, and Italy spent significant sums of money for little return. The penatration levels of Qatar and Singapore can be largely explained by geographical factors (City countries with few population and small surface area) that do not apply to Turkey.28

Australia was ranked 25th in provision of national broadband services, according to the WEF’s Networked Readiness Index. Telstra dominated the market, and broadband accessibility was limited due to significant investments required to deploy fiber in Australia. NBNCo was established to design, build, and operate the National Broadband Network in 2009. Telstra finalized an $8.7 billion contract with the government, which saw the company sell off parts of its infra- structure to NBN in 2012. NBN was then given full deployment exclusivity and a large government subsidy.

The model failed. There were execution delays, misaligned budgets and targets, and weak service. The government’s initial cost estimates, without telco involvement, were problematic. Coverage and speed targets were revised downward in 2015 after a new government was elected in 2013. By 2015, Australia had dropped to 42nd position in national broadband services, mainly due to ongoing delays.

The result serves as a caution to policymakers in other nations to consider the time and expense to create a new company and the significant risks involved.

27Strategy&

Exhibit 12 Risk vs. reward matrix

1 UK, Italy and Malaysia are calculated based on FTTC while for other countries calculation is based on FTTP 2 Qatar’s high coverage and penetration results were not achieved by the national broadband network (Q.NBN) but rather the incumbent Ooredoo 3 HSBB coverage at the date of supply restructuring launch 4 Numbers in parenthesis show total government subsidy plan

Source: Analysys Mason; IDATE; Strategy& analysis

0%

20%

40%

60%

80%

100%90%80%70%60%50%40%30%20%10%0

H S

B B

p e

n e

tr a

ti o

n (

2 0

1 5

)

HSBB Coverage (2015)

Australia (1% in 2009)3

(US$20 bn)

UK1

(0% in 2005)3

(US$1.5 bn)

New Zealand (4% in 2011)3

(US$1.5 bn)

Singapore (4% in 2009)3

(0.7 bn USD)

Malaysia1

(12% in 2010)3

(0.6 bn USD)

Sweden (24% in 2008)3

(US$0)

Italy1

(9% in 2008)3

(US$6.5 bn)

Qatar2

(11% in 2011)3

(US$0.55 bn)

Relatively small countries with easy topology

Average coverage pp. increase per year

0%-5%

6%-10%

11%-25%

Bubble size shows government subsidy per capita

Turkey

28 Strategy&

5. Understand the time and complexity a national broadband company requires. Australia and New Zealand took three years to get their NBNs off the ground; Singapore took two years. In any country, stakeholders must resolve questions of scope and coverage, business model, government subsidy, operational and commercial model, and regulation. Due diligence is not quick or easy. In separating IT networks, all product sets must be re-engineered, which can incur costs of $1 billion or more. Regulatory environments require detailed revision to ensure that legacy copper services do not impede uptake of fiber services. Substantial government funding is needed to extend FTTP service into rural areas that do not meet commercial investment criteria, at a per- home cost of up to five times as much as urban areas.29

As Exhibit 13 demonstrates, most supply restructuring cases did not accelerate national broadband deployment and adoption — the average percentage point increase per year is low. This confirms that the time to execute supply restructuring and its complexity can lead to lower impact on the market. In addition, the cases were all at earlier stages of market development than Turkey is at today.

New Zealand’s citizens and government were unhappy with the level of telecommunications industry investment. The digital divide between urban and rural areas was increasing, but the private sector was reluctant to invest widely in FTTP due to New Zealand’s geography and dispersed population. As a solution, in 2008, New Zealand created an Ultra-Fast Broadband program that distributed deployment to public–private partnerships. In 2011, structural separation of the existing businesses into New Telecom (retail) and New Chorus (infrastructure) was finally completed (Telecom is now known as Spark New Zealand). The New Zealand government subsidized 25 percent of the cost of deploying FTTH to 70 percent of the country.

New Zealand’s model failed to achieve strong results due to misalignment of regulatory policy and infrastructure policy, leading to very low copper broadband pricing undermining the take up of high-speed broadband (HSBB), execution delays, large financial risks, and low uptake. Separation of the Chorus network business, first as functional separation, then full structural separation, took three years. There were clear disconnections between retailers’ incentives and Chorus’ objectives to capitalize on scale economy benefits. ISPs were slow to develop fiber plans, leading to low uptake and large financial risk to network operators (particularly Chorus).

These lessons, both positive and negative, from 15 national broadband case studies,30 as well as the current Turkish market foundations and barriers, provide the framework for our six policy recommendations to accelerate Turkish broadband adoption.

Most supply restructuring cases did not accelerate national broadband deployment and adoption.

29Strategy&

We can take significant lessons from the experiences of other countries around the globe over the past decade as they sought to modernize their own broadband networks.

Overall, demand stimulation and infrastructure investment models better address national broadband acceleration objectives when market conditions are comparable to Turkey’s.

Exhibit 13 Progress of supply restructuring cases and Turkey

1 Year of supply restructuring

Source: Analysys Mason; Strategy& analysis

0%

0%

1%1%

17%

0%0% 3%

60%

7%

49%

34%

22%

4% 0%

Australia Singapore New Zealand Italy Sweden UK

20091 2015 2015

56%

24% 8%

86%

45%

100%96%

37%

1% 11%

21 pp 24 pp

XX pp

9%4%4

H S

B B

c o

ve ra

g e

H S

B B

p e n e tr

a ti o

n

Average percentage point increase per year

NBNCo

NetCo

Separation

20111 2015 20111 2015 20091 2015 20081 2015 20081 2015 20051 2015

Qatar Turkey

66%

13%

12 pp 15 pp

9 pp

6 pp 5 pp

5 pp

2 pp

2 pp

1 pp 1 pp

1 pp

1 pp

(% of households)

30 Strategy&

Turkey needs a national broadband plan. Under its own steam, the market has passed the first stage of development, as private companies have built the infrastructure to provide Internet service to almost every home, and high-speed service to most. The country is on its way to being fully connected, and to realizing all the economic and social benefits that will follow.

At its current stage of development, with developed supply but demand below expectations, Turkey does not need a supply restructuring. Restructuring will not address the country’s real barriers; it has proved to have weak market impact, to be risky and complex to execute, and to be costly and slow.

In addition, establishing a national broadband company requires careful analysis of many critical considerations (see Exhibit 14, page 31).

These considerations are formidable and fraught with risks for even the most carefully laid plans, as we have seen in other jurisdictions. Even the most noble of intentions fail under the complexity of such a project.

Turkey’s remaining barriers will not be overcome without deliberate decisions by policymakers. Of the three overarching models — demand stimulation, infrastructure investment, and supply restructure — the balanced approach between demand stimulation and infrastructure investment will most efficiently move the country toward full Internet participation.

We have identified six mechanisms that deliver impact, based on benchmarks and expert input, and taking into consideration Turkey’s market specifics. Three concentrate on demand stimulation and three address deep deployment incentives.

A national broadband plan

31Strategy&

Risks must be evaluated against potential impact. A decision that is complex to execute, or that requires high investment and major regulatory changes, may not achieve its end as quickly as planned. In Turkey’s case, demand stimulation offers

the strongest, fastest impact for the lowest risk, the lowest cost, and the least change in policy. Supply restructuring, even if it reaches its goals, will take a long time to achieve, with associated high costs and complexity and major regulatory changes.

Exhibit 14 Key considerations for national broadband company establishment

Source: Strategy& analysis

Scope & investment

- What is the service / geographic scope of the NBC business and which assets are transferred? - How do you value the current assets to be transferred? How is the financial transfer structured? - What will ensure assets separation with minimum disruption? Who covers separation cost? - What ownership structure would balance the interests of various stakeholders involved? - What government funding is required to ensure sustainability and achieve socio-economic goals?

Operational performance

- How is the company incentivized to innovate and invest in new technology? - How are people incentivized to improve performance given monopoly position? - How is the interface with retail operators managed for planning, SLAs, and QoS? - Who gets the access to the cabinet and how is that operationally controlled? - How are national security risks mitigated and managed?

Regulatory regime

- What new laws and regulations should be enacted with the new monopoly? - What adjustments would be needed on existing regulatory and competition regime? - How will equivalence of services be set, governed, and enforced? - How will internal differences in goals be managed, and under what power? - What institutional adjustments are needed to handle such a regime?

32 Strategy&

1 . Evolve and boost commercialization channels. A hands-on, consultative broadband distribution approach, particularly in low- income areas, can promote digital literacy as it has in other countries. Current operators can improve their distribution to better commercialize their broadband deployment, upgrading people from entry-level to high-speed packages.

In Brazil, the Computers for All program established over 90,000 “LAN houses,” which offered access to 68 percent of the population in the lowest income brackets by 2013. In Sri Lanka, the government established more than 700 communication centers in rural areas, which helped increase computer literacy from 10 percent to 40 percent over a 10-year period. In the U.S., Comcast launched its Internet Essentials plan to offer entry-level broadband and digital literacy training to low-income households for $10 per month, and later increased the speed offered from 1.5 to 5 Mbps while increasing to a higher, but still discounted, price. By 2015, 250,000 households were enrolled in the program. In Italy, Vodafone Italia partnered with a third party specializing in door-to-door sales of non-telco products and revamped its incentive scheme to reward those agents.

To execute this initiative in Turkey, stakeholders could leverage the strong distribution network of fast-moving consumer goods and electronics brands to promote Internet education, and target community entrepreneurs and influencers to increase awareness of broadband services. Strategy& estimates that such a policy in Turkey could increase awareness of broadband in rural areas by 25 percent and increase entry-level penetration by 18 percentage points.

2. Reduce cost of ownership for end users. Distributing subsidized devices and broadband subscriptions to students, cutting taxation on entry-level broadband devices, and subsidizing in-building access facilities installation, will have a sizable positive impact on device ownership. From 2009 to 2012, China ran a PC subsidy program for rural areas where per capita income was $700, offering 13 percent rebate on selected products that cost less than $500. The program was estimated to generate sales of 800,000 computers. In Malaysia, the government offered tax relief of up to $165 on broadband subscriptions, which made access more affordable for 100,000 students. Singapore’s OpenNet contacts home and business owners to pull optical fiber and install fiber terminal points for free, and IDA offers subsidies up to 90 percent of the cost of in-building fiber infrastructure to businesses in Singapore.

Such a program would need participation of several vendors with solid government support, and effective marketing to spread awareness of the subsidy mechanism. In Turkey, if 10 percent of students who do not own computers received subsidized devices, and if the tax contribution of total cost of ownership of broadband were cut half, entry-level penetration

A hands-on, consultative broadband distribution approach, particularly in low-income areas, can promote digital literacy as it has in other countries.

33Strategy&

could rise by 7 percentage points, and high-speed could rise by 2 percentage points, according to Strategy& research.

3. Scale up relevant local applications and content. The applications and content most likely to bring people online that have proven to be successful in other countries are those related to: education; e-government services; and economic benefit. Farmers in India have improved their incomes by up to 15 percent by using broadband to view weather forecasts, commodity prices, and advice on agricultural techniques. The Brazilian Ministry of Education has partnered with Amazon and the National Education Development Fund to digitize and distribute free textbooks to hundreds of thousands of Brazilian educators. In Estonia, a unique electronic identification card and digital signature for each citizen now grants nearly universal access to government services.

Turkey could digitize public services with strong social and welfare impact (e.g., e-education and e-government services) and launch a marketing campaign to make citizens aware of what was available. A supportive ecosystem for entrepreneurs to develop local applications would be necessary. Strategy& estimates that a 10 percent increase in the Turkish content maturity index could raise entry-level penetration by 11 percentage points.

Aside from demand, Turkey can use three separate mechanisms to incentivize deeper deployment into suburban and rural areas. Commercially viable wholesale and infrastructure sharing services should incentivize investment by operators that offer faster broadband speeds. Extension of current subsidies (Universal Service Fund) to high-speed broadband should help bridge the coverage gap in low- density areas. Last but not least, partnerships between operators and municipalities to aggregate demand and ease right-of-way fees should enhance economic viability and accelerate coverage.

4. Promote wholesale and infrastructure sharing services that make commercial sense. Infrastructure-based competition can be encouraged through extension/redefinition of the regulatory holiday. In Spain, the regulator applied a speed-based regulatory holiday above 30 Mbps and removed all SMP regulations in competitive markets, after splitting the country into regional markets. In the Netherlands, the regulator withdrew wholesale access remedies on KPN after redefining the broadband market to include cable.

For such a strategy, regulators should offer acceptable commercial prices to encourage operators to deploy their own networks; offer strong service level agreements in providing backhaul services; and review SMP regulations in regional markets to incentivize

34 Strategy&

infrastructure-based competition. The results, according to Strategy& research, could be a 1.5-fold increase in HSBB coverage for Turkish rural areas, and an increase of 4 percentage points in high-speed penetration and 11 percentage points in high-speed coverage in Turkey.

5. Leverage the Universal Service Fund (USF) to support broadband deployment in pre-defined low-density areas. Turkey should leverage the USF to support broadband deployment in pre-defined low density areas. A targeted list of low-coverage cities should be identified based on expected investment and estimated uptake of broadband services, with clear alignment between operators and policymakers on the required funding and execution time line.

In the U.S., 10 carriers have accepted US$1.5 billion in annual support for rural broadband development from the Connect America Fund, which, together with private investment, will expand broadband to an additional 12 percent of the rural population. Venezuela’s USF subsidized the Puntos de Acceso program, which aims to bring fixed- line and Internet services to people who lack access to ICT services.

To make the same approach work in Turkey, operators and policymakers should be aligned on the conditions for using the USF, and rural areas should be the targets for investment. Strategy& estimates that the initiative could increase high-speed rural penetration by 6 percentage points, and rural coverage by 25 percentage points.

6. Drive municipality-led demand aggregation and incentives. Turkish operators would need to partner with municipalities to support deployment and demand aggregation while offering free access to selected municipality centers. High-revenue-generating municipalities should be targeted first in order to build up momentum. This mechanism would encourage cities to coordinate demand among local institutions. Tactics could also include national competitions on broadband adoption, uniformly lower licensing costs (or right-of-way fees) enforced by municipalities on operators, and an update of building standards and codes to address in-house wiring.

In the Netherlands, as previously mentioned, Reggefiber targets an area for potential fiber deployment only if at least 30 percent of households commit to purchase a connection. Provincial authorities in La Pampa, Argentina, enlisted 21 municipalities, most of them rural communities, to aggregate demand from local governments, hospitals, schools, and libraries for connectivity before building the broadband network.

If the plan is rolled out in rural and suburban areas across all cities, it could increase high-speed penetration by 11 percentage points in rural and suburban areas, and coverage by 20 percentage points.

35Strategy&

A successful national broadband acceleration plan requires collaboration to balance three essential drivers: market competitiveness and sustainability; increasing availability of infrastructure at the right price; and optimizing investment. Policymakers need to define a national broadband plan with clear initiatives and actions including financing demand stimulation and deep deployment mechanisms. Regulators need to facilitate a conducive environment for infrastructure investment and sharing. Operators need to increase and accelerate fiber commercialization investments and support demand stimulation. NGOs need to champion demand stimulation mechanisms and support aggregation guarantees. This collaboration can create consensus on strategy and tactics. Implementing these initiatives requires well- developed institutional capabilities; policies for perpetual execution going beyond the political electoral cycle; and cross-ministerial coordination.

Turkey needs a national broadband plan that sets targets specifically based on its market dynamics and that involves economically sound infrastructure investment to eliminate current barriers. If properly implemented, the six mechanisms described above will fulfill all three drivers and aggressively accelerate Turkey’s high-speed broadband adoption without disruption.

In the first phase, demand stimulation will increase broadband penetration in covered areas, drive upgrade to higher-speed broadband leveraging existing infrastructure, and improve HSBB infrastructure coverage in urban areas. Next, HSBB infrastructure will expand to uncovered suburban areas while demand stimulation increases penetration in newly covered areas. The final phase uses a mix of technologies to bring universal HSBB while continuing to motivate consumers to upgrade to higher speeds. Over the course of the next decade, the plan should raise HSBB supply from 66 percent to 95 percent, and demand should increase even more sharply, from 13 percent to 80 percent (see Exhibit 15, page 36).

Success factors

Turkey needs a national broadband plan that sets targets specifically based on its market dynamics and that involves economically sound infrastructure investment to eliminate current barriers.

36 Strategy&

Turkey does not need a major supply restructuring, thanks to the current state of market development (high supply, low demand), and given the associated complexities, risks, and investments required.

Exhibit 15 High-speed broadband evolution trajectory

Source: Strategy& analysis

H S

B B

p e

n e

tr a

ti o

n /c

o ve

ra g

e

66%

13%

95%

60%

70%

80% 75%

85%

Today 2018 2023 2028

Drive uptake Expand coverage & uptake Achieve universal coverage I II III

∆53%

∆20%

∆15%

∆10%

∆10%

∆10%

∆10%

+9pp

+47pp

1. Demand stimulation to increase broadband penetration in covered areas

2. Upgrade and migration of current BB subscribers to high-speed

3. Infrastructure investment for HSBB in uncovered urban areas

1. Infrastructure investment for high-speed BB in uncovered sub-urban areas

2. Demand stimulation to increase BB penetration in newly covered areas

1. Infrastructure investment for HSBB in rural areas through a mix of technologies (FTTX, LTE)

2. Strong push on upgrading customer to higher speeds

HSBB supply HSBB demand

37Strategy&

Turkey’s existing telecom outlook has the foundation on which to build a vibrant, truly national broadband network that reaches and benefits all people. To leverage that foundation and address current barriers, many measures have been identified that are impactful, affordable, and minimally disruptive. All of these measures have succeeded in other countries whose situations parallel Turkey’s — countries that had a good supply of broadband capacity but worked to build demand through wider distribution channels, reduced costs, and better local content.

International experience suggests that Turkey does not need supply restructuring. Instead, a coordinated effort by all stakeholders — policymakers, regulators, telecom companies, and content providers — can raise the supply of local online content, and increase demand among those who either lack broadband access or have previously seen no need for it. Stakeholders can follow proven policies and regulations to fill in the remaining gaps between supply of high-speed broadband and citizens’ demand, such as aggregating local demand, using mobile technologies as a supplement to fixed lines in remote areas, and doubling down on commercialization efforts to improve monetization of current infrastructure.

Connectivity-aided economic growth offers greater possibilities and prosperity for all sectors of society. If all stakeholders work together on evidence-based strategies to achieve it, while minimizing disruption, Turkey can hasten a highly connected future that benefits everyone.

Conclusion

38 Strategy&

Endnotes

1 Strategy& estimate

2 International Telecommunication Union (ITU)

3 The impact of acceleration measures on the penetration curve is estimated as the difference between the status quo forecast (based on Turkey’s penetration CAGR) and the improvement forecast (based on the s-curve trend of comparable and advanced markets). The source of current numbers is WBIS.

4 Cumulative impact includes years from 2016 to 2023. Strategy& analysis

5 Türk Telekom

6 The years 2005–2014.

7 Tübisad; Invest in Turkey; Strategy& analysis

8 ICTA (BTK); Türk Telekom; World Cellular Information Service (WCIS). EU numbers are for the EU15 nations.

9 Türk Telekom

10 Türk Telekom

11 Türk Telekom

12 Türk Telekom

13 Türk Telekom estimates

14 WCIS; BTK; Türk Telekom; Strategy& analysis

15 Comparable markets are countries with population size between 25 million and 150 million, GDP per capita equal or higher than Turkey’s, and fixed broadband penetration higher than 30 percent. Selected countries are: Australia, Canada, France, Germany, Italy, Japan, Saudi Arabia, Malaysia, Poland, Russia, Spain, South Korea, and U.K.

16 World Economic Forum, Global Information Technology Report.

17 Euromonitor; WCIS

18 Refer to Strategy& Content Maturity Index in WEF GITR 2015

39Strategy&

19 BTK; Türk Telekom

20 EU Commission; European Digital Commission; BTK market reports; Turkish Statistical Institute (TÜIK in Turkish); Türk Telekom

21 The affordability index is computed based on the share of households with a broadband package that is less than 5 percent of the average household income.

22 Strategy& analysis

23 Analysys Mason; Strategy& analysis

24 Infocomm Development Authority (IDA), Singapore

25 Bronwyn E. Howell, “Separation Anxieties: Structural Separation and Technological Diffusion in Nascent Fibre Networks,” SSRN 2014.

26 David Bernal, “Providing broadband services through PPP (Public Private Partnership) models,” ITU 2013; James Allen, Ceri Tinine, Final report for Ofcom: International case studies, Analysys Mason, 2015.

27 FTTH Council; Stokab; ICT Regulation Toolkit; InterConnect Communications; Analysys Mason; Strategy& analysis

28 Analysys Mason; IDATE; Strategy& analysis

29 Strategy& analysis

30 Cases analyzed include Australia, Estonia, France, Italy, Japan, Malaysia, New Zealand, Portugal, Singapore, South Korea, Spain, Sweden, Qatar, U.K., and the U.S.

© 2016 PwC. All rights reserved. PwC refers to the PwC network and/or one or more of its member firms, each of which is a separate legal entity. Please see www.pwc.com/structure for further details. Mentions of Strategy& refer to the global team of practical strategists that is integrated within the PwC network of firms. For more about Strategy&, see www.strategyand.pwc.com. No reproduction is permitted in whole or part without written permission of PwC. Disclaimer: This content is for general purposes only, and should not be used as a substitute for consultation with professional advisors.

www.strategyand.pwc.com

Strategy& is a global team of practical strategists committed to helping you seize essential advantage.

We do that by working alongside you to solve your toughest problems and helping you capture your greatest opportunities.

These are complex and high-stakes undertakings — often game-changing transformations. We bring 100 years of strategy consulting experience and the unrivaled industry and functional capabilities of the PwC network to the task. Whether you’re

charting your corporate strategy, transforming a function or business unit, or building critical capabilities, we’ll help you create the value you’re looking for with speed, confidence, and impact.

We are part of the PwC network of firms in 157 countries with more than 208,000 people committed to delivering quality in assurance, tax, and advisory services. Tell us what matters to you and find out more by visiting us at strategyand.pwc.com/me.

Anil-Koksal.pdf

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

113

The Determinants of Broadband Access and Usage in Turkey: Do Regions Matter?

Emin Köksal

[email protected], Bahçeşehir University,

Istanbul, Turkey

Bülent Anıl

[email protected], Bahçeşehir University, Istanbul, Turkey.

Keywords: Broadband access; Broadband use; Regional disparities; Turkey

Jel Codes: O10, O31, O32

The Determinants of Broadband Access and Usage in Turkey: Do Regions Matter?

Abstract

The aim of the paper is to examine the determinants of broadband internet access in

Turkey, and in particular, whether the probability of adopting broadband access varies

by region. This study also attempts to find out whether there are regional disparities in

engaging in various online activities. The 2012 ICT Household Survey that was

conducted and administered by TURKSTAT is used. Probit model of probability of

adopting broadband access is estimated and results are consistent with the earlier

studies: Education, gender, age and household size matter. Besides this, the

probability of adoption in West and Central Turkey is higher than the Eastern Turkey.

In the second stage of the paper, conditioned on the availability of access, probability

of using various online activities is estimated. The results confirm the existence of the

regional disparities: Being in the West increases the probability of using online

activities that require relatively advanced skills such as searching information and e-

banking, while the probability of using time-required entertaining activities is higher

for the individuals in the East.

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

114

1 Introduction

Broadband internet is widely considered as the crucial element of economic and social

development (OECD, 2013). Thus, many governments are creating policies to increase

broadband penetration and to stimulate regular internet usage. The most notable one

is the Digital Agenda for Europe which defines the targets for the deployment and

penetration of broadband access for EU countries (Commission, 2010). Digital

Agenda’s targets include covering the entire EU with high speed broadband access,

increasing regular internet usage among disadvantaged people, encouraging

populations to buy online, and increasing the use of e-Government. 1

Government policies and regulatory strategies are expected to shape the market

structure and determine the broadband penetration. However, besides these policy

factors, there are various factors which affect broadband penetration and broadening

internet usage. As shown by The Berkman Center for Internet and Society (2010),

variables such as geography, population density, education, employment, and income

are more influential than the policy factors on the broadband performance.

Turkey has experienced both technological change and a considerable deregulation in

the telecommunications industry since 2000. Its obligations to streamline EU’s “acquis

communautaire” have speeded up to establish new institutions (e.g. National

Regulatory Authority) and to adopt new legislations (e.g. Electronic

1 https://ec.europa.eu/digital-agenda/en/our-goals (accessed on November 16, 2013)

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

115

Telecommunications Act, e-Government act, etc.). With the partial privatization of the

state owned incumbent company (Turk Telekom) and with accompanying regulations,

slow but steady competition has begun in the fixed broadband provision market. On

one side, new regulations opened up a service-based competition through allowing

alternative internet service providers to provide access over Turk Telekom’s wired

network. On the other side, divestiture of cable TV network from the incumbent and

emerging fiber deployment by an alternative operator, Turkcell Superonline, has

initiated a facility-based competition in the fixed-broadband market. Moreover, mobile

broadband services have been available since 2009 and have a considerable

competition between three mobile operators; Turkcell, Vodafone, and Avea.

Nevertheless, more than ten years after the deregulation in the telecommunications

industry, Turkey still has the least penetrated broadband market in Europe. 2 Thanks to

Turk Telekom’s countrywide fixed network, fixed-broadband access subscription had

increased rapidly in the early 2000s. However, starting from 2010, the growth rate of

broadband subscription has decreased sharply. The fixed broadband penetration rate

was set to a 10% level, which is considerably lower than EU 25 average (OECD, 2013).

The determinants of broadband access have been widely studied in the literature

(Chaudhuri et al., 2005; Cerno & Amaral, 2006; Moutafides & Economides, 2011;

2 By the end of 2013 broadband penetration in Turkey is 11%. (OECD Broadband Portal,

http://www.oecd.org/internet/broadbandandtelecom/oecdbroadbandportal.htm, accessed on 10 January 2013)

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

116

Hargittai & Shafer, 2006; Hargittai & Hinnant, 2008; Goldfarb & Prince, 2008). 3 Prior

studies conclude that income, education, demographic characteristics and price are

the main determinants in the adoption of broadband access (Chaudhuri et al., 2005,

Cerno & Amaral, 2006, Moutafides & Economides, 2011). Additionally, some studies

have argued that socioeconomic and demographic factors also influence users’ online

activities and their online behaviors (Hargittai & Shafer, 2006; Hargittai & Hinnant,

2008; Goldfarb & Prince, 2008). In their study for the United States, Hargittai & Shafer

(2006) discuss the role of gender and argue that women’s lower self-assessment on

their web-use skills may affect the extent of their online behavior. Moreover, in an

expanded study, Hargittai & Hinnant (2008) state that people with higher levels of

education and resource-rich background pursue capital-enhancing online activities.

Goldfarb & Prince (2008) analyze the issue a bit further and examine the online time

consumption by income and education groups. They find people with higher income

and education level adopt the internet more than other groups but they spend much

less time online. The authors interpret this result as the low-income individuals might

have lower opportunity cost of time, higher marginal benefits from the internet use,

more time or higher value for the internet.

Studies that deal with the regional differences in internet use are very scarce, and to

our knowledge, none of them is conducted in Turkey. A report prepared by Spooner et

3 See, Srinuan & Bohlin (2013) for a recent and detailed review of literature.

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

117

al. (2003) for the Pew Research Center examines the regional disparities of internet

use in the United States. The report states that the penetration rate among regions is

uneven; and education and income are the two main factors behind this variation. The

report also examines the types of internet usage and finds that some basic activities

such as emails are uniformly common in all regions, while activities such as online

shopping, obtaining news and information, searching for health information have

regional variations. The report concludes that each region has its own characteristics

and online activities in the region are in line with these characteristics.

The purpose of this paper is to investigate the determinants of broadband adoption

and usage of households in Turkey. Our main motivation is to examine the influence of

socioeconomic, demographic and geographic factors on broadband access and online

activities. The main contribution of the paper is to analyze the effect of these factors

on online activities emphasizing the regional disparities in Turkey. We classify the

online activities in three categories: “Advanced” includes the activities that require

relatively advanced internet skills such as uploading content, streaming media,

searching information, getting health appointments, making a trip reservation, e-trade

and e-banking; “entertaining” includes the activities that require excessive time use

such as social networking, voice/video calling, online gaming, downloading

and “conventional” includes the basic activities that doesn’t require advanced skills

such as sending/receiving e-mail and getting news. To our knowledge, no other

academic study classifies the services. In addition, this study is the first that analyzes

the determinants of broadband access and usage in Turkey.

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

118

Our empirical results strongly verify that the demographic factors are the main

determinants of broadband adoption in Turkey as well. Our data does not have any

information on income and price; therefore our study cannot control the impact of

these two factors. Our findings support Spooner et al. (2003) and strikingly set out how

online activities change by region consistent with the regional characteristics. Our

findings also support that males are more likely to use the online activities and better

educated people are more likely to use advanced activities. Although our data does

not have the appropriate income and time consumption information, we show that

individuals who have a job and have a higher educational degree are more likely to

adopt broadband access and use more advanced activities.

The closest study to our research was carried out in Thailand, where Srinuan & Bohlin

(2013) investigated the determinants of online activities conditioned on the availability

of broadband at home. Their findings show that the availability of fixed infrastructure

and various demographic variables affect the existence of broadband access at home,

and moreover, each factor has a different impact on specific online activities. Although

we use the same method to estimate the determinants and get comparable results,

our research differs in these distinctive points: We focus on the regional disparities in

broadband use and examine how the use of categorized online activities differs by

region.

The rest of the paper is organized as follows: The next section exhibits the regional

panorama of Turkey. Section 3 introduces the data and the methodology. Section 4

presents the findings. And, Section 5 delivers a discussion of the findings

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

119

2 Regional Panorama of Turkey

Turkey has been divided into three statistical regions to be comparable with EU

countries: NUTS1 (12 regions), NUTS2 (26 regions) and NUTS3 (81 regions). The data

we employ in this study is representative at the NUTS1 level. There are 3 to 9

provinces in each region except Istanbul. TURKSTAT states that the proximity and

population as well as culture are key factors in the selection criteria of regions. 4

Interregional disparities have always been a concern for the Turkish economy. The

regional disparities and its effect on economic growth have been much studied

(Doğruel & Doğruel, 2003; Karaca, 2004; Gezici & Hewings, 2004). The western part of

the Turkey has a higher population density and is always considered as more

“developed” relative to the eastern part (Özarlan et al. 2006). In this section, we

attempt to show the disparities among regions. First, we congregate four NUTS1

regions to construct one region. Our selection criterion is completely geographic 5 .

Then, by using the main indicators, we show whether the regional disparities between

our assembled regions exist. According to our classification, we obtain three

geographical regions as shown in Figure 1:

1. West - Istanbul, West Marmara, Aegean, and East Marmara regions

4 http://tuikapp.tuik.gov.tr/DIESS/SiniflamaSurumDetayAction.do?surumId=164 (accessed on November

29, 2013) 5 When the main indicators are taken into account, NUTS1 regions in each category (West/Central/East)

show similar characteristics. Therefore, we had chosen a completely geographical classification rather than cultural or social grouping.

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

120

2. Central – West Anatolia, Mediterranean, Central Anatolia, and West Black Sea

regions

3. East – East Black Sea, Northeast Anatolia, Middle East Anatolia, and Southeast

Anatolia regions

Figure 1 Regions in Turkey (NUTS1 level & West-Central-East)

Table 1 shows the regional disparities among our regions. The western part of the

Turkey has higher per capita income, a lower unemployment rate and lower poverty

rate and higher educational attainment than the eastern part of Turkey. Similarly,

Central Turkey has better indicators in comparison to Eastern Turkey.

Table 1 Regional indicators

Indicators West Central East

Per capita gross value added in 2011 (US Dollar) i 12 068 9 314 4 948

Population density (person per square kilometer) i 181 76 62

Fixed Telephony penetration – per 100 person ii

23 16,7 8.8

Operator Capacity for Fixed Telephony – per 100 person ii 62.8 47.8 35.6

Fixed-broadband penetration - per 100 person ii 13.6 9.8 4.4

Mobile-broadband penetration - per 100 person ii 20.4 14.2 9.9

Fiber network length (km) ii 85 252 75 022 50 013

Istanbul

West Marmara

East Marmara

Aegean

Mediterranean

West Anatolia

West Black Sea

Central Anatolia

Southeast Anatolia

Middle East Anatolia

Northeast Anatolia

East Black Sea

WEST CENTRAL EAST

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

121

i NUTS2 level Gross Value Added data and Address Based Population Registration System data were taken from TURKSTAT Regional Statistics Database, calculations were made by the authors ii

Data source: ICTA (2013) iii

The distribution of poor among regions. Poor is defined as the person who is under the national relative poverty line. iV

Nonagricultural unemployment rate for the population above age 15

3 Data and methodology

We use the Household Information and Communication Technology (ICT) Usage

Survey, administrated annually by TURKSTAT, in accordance with the regulations of the

EU. The sampling is representative at the national and regional level (NUTS1). Since

there are some modifications in questions over the years, we take the most recent

year available. The 2012 survey consists of 10,605 households with 39,361 individuals.

We only take individuals who are above 15 years old, which reduces our sample to

29,061 individuals. 40% of our sample is from the West, 34% is from Central Turkey

and 26% of our sample is from the Eastern part of Turkey.

The respondents, who state that they had used the internet in the last 3 months, were

asked to answer multiple questions about their internet use, specifically the activity

they performed in the last 3 months. In our analysis, we use various demographic

variables, i.e., age, gender, education; household information, household size, number

of children or the number of children in elementary school; and variables associated

with the broadband usage. We also added the length of fiber network to the first stage

to create an indicator for the investment in that region.

Distribution of Poor iii (2010)

i 23.3 28.9 47.8

Unemployment rate i, iV

11.3 11.0 13.6

Average years of schooling i 5.03 4.55 3.53

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

122

Due to the fact that the income information is unavailable in the ICT survey, we are not

able to control for economic conditions of a household. In order to capture the effect,

we utilize a proxy variable that provides information about an individual’s

employment.

Table 2 presents the information about all variables and descriptive statistics by

region. One notable result from Table 2 is that the indicators in the west are

significantly better than the east, which is consistent with the regional disparity

presented in previous section. For example, households in the west have, on average,

fewer household members with fewer kids. In being consistent with Section 2,

broadband connections are significantly higher in the west.

Table 2 Descriptive Statistics TURKEY WEST CENTRAL EAST

Variable Description Mean (SE) Mean (SE) Mean (SE) Mean (SE)

EDULESSHIGH = 1 if less than high school degree 0.88 (0.32) 0.86 (0.35) 0.88 (0.33) 0.93 (0.25)

EDUMOREHIGH = 1 if more than high school degree 0.12 (0.32) 0.14 (0.35) 0.12 (0.33) 0.07 (0.25)

AGELESS35 = 1 if age less than 35 0.42 (0.49) 0.39 (0.49) 0.39 (0.49) 0.49 (0.50)

AGE3555 = 1 if age between 35 and 55 0.36 (0.48) 0.38 (0.49) 0.38 (0.49) 0.32 (0.47)

AGEMORE55 = 1 if age more than 55 0.22 (0.41) 0.22 (0.42) 0.23 (0.42) 0.20 (0.40)

MALE = 1 if male 0.49 (0.50) 0.50 (0.50) 0.49 (0.50) 0.49 (0.50)

HHSIZE The size of the household 4.71 (2.39) 3.93 (1.67) 4.32 (1.90) 6.14 (2.99)

PUPIL = 1 if kid(s) in elementary school at home 0.49 (0.50) 0.40 (0.49) 0.48 (0.50) 0.62(0.49)

EMPLOYMENT = 1 if individual is working 0.43 (0.49) 0.44 (0.50) 0.43 (0.50) 0.39 (0.49)

FIXED = 1 if individual has fixed telephone at home 0.42 (0.49) 0.53 (0.50) 0.43 (0.50) 0.26 (0.44)

MOBILE = 1 if individual has a mobile phone 0.95 (0.21) 0.95 (0.21) 0.96 (0.20) 0.95 (0.22)

BROADBAND = 1 if individual has broadband access 0.16 (0.37) 0.24 (0.43) 0.16 (0.37) 0.07 (0.26)

EMAIL 0.18 (0.39) 0.23 (0.42) 0.18 (0.38) 0.12 (0.32)

NEWS 0.20 (0.40) 0.26 (0.44) 0.21 (0.41) 0.12 (0.32)

SOCIAL NETWORKING 0.11 (0.32) 0.15 (0.36) 0.10 (0.30) 0.08 (0.27)

VOICE/VIDEO CALL 0.12 (0.32) 0.15 (0.35) 0.11 (0.32) 0.08 (0.28)

ONLINE GAME 0.08 (0.27) 0.09 (0.29) 0.08 (0.27) 0.06 (0.24)

DOWNLOAD 0.13 (0.34) 0.16 (0.37) 0.14 (0.35) 0.09 (0.28)

UPLOAD CONTENT 0.09 (0.29) 0.11 (0.32) 0.11 (0.31) 0.05 (0.22)

STREAMING MEDIA 0.11 (0.31) 0.14 (0.35) 0.11 (0.32) 0.05 (0.22)

SEARCH INFO 0.17 (0.37) 0.23 (0.42) 0.18 (0.38) 0.08 (0.27)

HEALTH APPOINTMENT 0.05 (0.22) 0.08 (0.28) 0.05 (0.21) 0.01 (0.11)

TRIP PLAN 0.05 (0.22) 0.07 (0.26) 0.06 (0.23) 0.02 (0.14)

E-TRADE 0.02 (0.14) 0.02 (0.15) 0.03 (0.16) 0.01 (0.09)

E-BANKING 0.05 (0.21) 0.06 (0.25) 0.05 (0.23) 0.01 (0.11)

Observation 39 361 14 599 13 277 11 485

= 1 if individual is using the activity

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

123

In our empirical estimation, we follow Srinuan & Bohlin (2013) with a slight distinction.

Similarly, we first assume that individuals must have broadband access to use the

internet. Thus, we first estimate the probit model of the availability of broadband

connection on various individual and household level characteristics. Unlike Srinuan &

Bohlin (2013), where availability of fixed broadband connection at home is estimated,

we add mobile connections. It is mostly because of the increasing usage of cell phones

in Turkey (Torlak et al., 2011). In the first stage our estimation is;

(1)

where is a binary variable that takes the value 1 if broadband connection is

available, and 0 otherwise, is the vector of individual and household level

characteristics, is a vector of unknown parameters, and is the cumulative

distribution function of the standard normal distribution.

After the first stage estimation, we adopt Heckman correction for sample selection

(Heckman, 1979). In the second stage, we estimate the probability of activity types

separately, by eliminating some variables that we used in the first stage. The

assumption is that they are affecting the availability of the broadband connection,

however, not the type of activity.

(2)

where is the binary variable that takes the value of 1 if the individual using internet

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

124

for that particular activity does so conditioned on the availability of broadband

connection, is a vector of individual and household characteristics, leaving out the

availability of kids at school age, mobile connection and fiber length and fixed

telephone subscription, and is the error term.

4 Findings

The estimated results to identify the determinants of broadband access and its usage

in Turkey are shown in Table 3. Our first stage results associated with the determinants

of broadband access adoption are shown in the second column of the table.

Broadband access adoption increases with the availability of fixed and mobile phone

subscription. Being younger, employed, and having a degree above high school

increases the probability of broadband adoption. Moreover, living in the west or the

central region increase the adoption rate compared to residing in the east. Likewise

individuals living in urban areas are more likely to adopt broadband. While having

kid(s) at primary school increases the probability of broadband adoption, having larger

families decreases likelihood of the adoption. Fiber network length is insignificant for

the adoption.

Results starting on the third column exhibit the outcome of second stage estimation

for the usage of online activities. Education, age, gender, and household size variables

follow a similar pattern as described in the first stage. Having a degree less than high

school decreases the probability of usage in online activities except streaming,

downloads and online games. Usage of online games increases with having a degree

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

125

less than high school. Being younger increases the usage of online activities but is

insignificant to get news online. Being older generally decreases the probability of the

usage of the applications; however it does not make any difference to getting news

online, searching for information, getting a health appointment, trip planning ore-

trade. Males are more likely to use applications than females. But gender does not

create any difference to searching for information or to getting health appointments.

Household size decreases the usage, but it does not matter for social networking and

online games.

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

126

Table 3 Estimated results for broadband access adoption and usage of online activities

1st stage

BROADBAND E-MAIL NEWS SOCIAL

NETWORKING

VOICE/VIDEO

CALL

ONLINE

GAME DOWNLOAD

UPLOAD

CONTENT

STREAMING

MEDIA

SEARCH

INFO

HEALTH

APPOINTMENT TRIP PLAN E-TRADE E-BANKING

EDULESSHIGH -1.1685*** -0.8162*** -0.5593*** -0.0915* -0.1152** 0.1491*** 0.0420 -0.1608*** -0.0660 -0.2876*** -0.1917*** -0.5283*** -0.3684*** -0.7799***

(0.0293) (0.0580) (0.0567) (0.0493) (0.0482) (0.0512) (0.0486) (0.0505) (0.0481) (0.0500) (0.0554) (0.0577) (0.0743) (0.0593)

AGELESS35 0.6088*** 0.7795*** 0.0480 0.5583*** 0.4044*** 0.5201*** 0.5827*** 0.6300*** 0.4669*** 0.0847** 0.1681*** 0.1355*** 0.1956*** 0.0960**

(0.0217) (0.0400) (0.0386) (0.0360) (0.0355) (0.0374 (0.0350) (0.0373) (0.0360) (0.0355) (0.0394) (0.0396) (0.0514) (0.0418)

AGEMORE55 -1.0002*** -0.3118*** -0.026 -0.3322*** -0.1582** -0.1922** -0.3670*** -0.2636*** -0.2013** -0.1000 -0.0020 -0.0862 -0.0160 -0.2871***

(.03643) (0.0818) (0.0872) (0.0844) (0.0791) (0.0886) (0.0799) (0.0877) (0.0813) (0.0778) (0.0896) (0.0890) (0.1175) (0.0956)

MALE 0.2309*** 0.2991*** 0.2801*** 0.2804*** 0.1547*** 0.4283*** 0.3702*** 0.1465*** 0.0829** 0.0209 -0.0175 0.1108*** 0.2172*** 0.3327***

(0.0219 ) (0.0377) (0.0378) (0.0341) (0.0334) (0.0357) (0.0339) (0.0345) (0.0333) (0.0343) (0.0372) (0.0377) (0.0490) (0.0413)

HHSIZE -0.1176*** -0.0776*** -0.0741*** -0.0125 -0.0474*** -0.0196 -0.0440*** -0.0754*** -0.0693*** -0.0574*** -0.0159 -0.1266*** -0.0645*** -0.1547***

(0.0068) (0.0127) (0.0127) (0.0116) (0.0114) (0.0121) (0.0116) (0.0120) (0.0117) (0.0117) (0.0132) (0.0142) (0.0178) (0.0155)

EMPLOYMENT 0.1162*** 0.1229*** 0.0515 -0.0720** -0.0772** -0.0345 -0.1040*** 0.0685* 0.0462 0.1345*** 0.0681* 0.1815*** 0.2170*** 0.5543***

(0.0228) (0.0379) (0.0382) (0.0350) (0.0341) (0.0365) (0.0349) (0.0355) (0.0342) (0.0351) (0.0385) (0.0391) (0.0512) (0.0436)

WEST 0.3376*** -0.1644*** -0.0135 -0.0580 -0.2753*** -0.2199*** -0.0653 -0.0497 0.1448*** 0.2405*** 0.5719*** 0.0890 0.0677 0.4254***

(0.0474) (0.0489) (0.0493) (0.0490) (0.0533) (0.0845) (0.0504) (0.0632) (0.0505) (0.0489) (0.0779) (0.0683) (0.0349) (0.0350)

CENTRAL 0.1954*** -0.1933*** 0.0643 -0.2515*** -0.2264*** -0.1311*** 0.0591 0.1682*** 0.2660*** 0.3376*** 0.3865*** 0.1777*** 0.3125*** 0.5233***

(0.0320) (0.0541) (0.0540) (0.0494) (0.0480) (0.0507) (0.0493) (0.0507) (0.0498) (0.0497) (0.0635) (0.0590) (0.0771) (0.0686)

URBAN 0.6558*** 0.0401 -0.1492** -0.0661 -0.2903*** 0.1025* 0.0392 -0.0566 -0.0303 0.0466 0.2146*** 0.2518*** 0.1869** 0.0916

(0.0257) (0.0579) (0.0584) (0.0536) (0.0516) (0.0565) (0.0536) (0.0552) (0.0529) (0.0531) (0.0657) (0.0685) (0.0888) (0.0692)

PUPIL 0.2404***

(0.0231)

FIXED 0.9214***

(0.0207)

MOBILE 0.4682***

(0.0604)

FIBERLENGHT 0.0008

(0.0023)

CONSTANT -1.1718*** 1.0055*** 1.5127*** -0.2706*** 0.6002*** -1.038*** -0.2836*** -0.3291*** -0.0983 0.643443*** -1.2312*** -0.3173*** -1.440*** -0.8463***

(0.0852) (0.1097) (0.1088) (0.1025) (0.0972) (0.1071) (0.1017) (0.1054) (0.1004) (0.09948) (0.1239) (0.1200) (0.1632) (0.1299)

rho -0.1336** -0.1848*** -0.0992* -0.2781*** 0.0148 -0.0294 -0.1178** -0.2751*** -0.259912***-0.1698*** -0.2754*** -0.2090*** -0.0524

(0.0570) (0.0568) (0.0534) (0.0526) (0.0555) (0.0526) (0.0550) (0.0526) (0.053928) (0.0606) (0.0610) (0.0803) (0.0677)

Chi square 5.55** 10.80*** 3.43* 27.61*** 0.07 0.31 4.53** 26.83*** 23.52*** 7.62*** 19.25*** 6.34*** 0.59

Observation 26 805 7 349 7 349 7 349 7 349 7 349 7 349 7 349 7 349 7 349 7 349 7 349 7 349 7 349

2nd stage

VARIABLE

The numbers in parenthesis represent the standard deviation. *, ** and *** indicates the significance at the 10%, 5% and 1% levels, respectively.

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

127

On the other hand, employment and regional variables vary with purpose of usage.

Having a job increases the probability of using email, uploading content, searching

information, streaming media, getting heath appointments, e-trade and e-banking; it

decreases the probability of using activities such as voice/video call, social networking

and downloads. Being employed is insignificant in online games, getting news and

streaming media.

Most importantly usage varies with the region. People living in the west and in central

Turkey are more likely to use advanced online applications such as searching for

information, getting health appointment, e-banking, and streaming media. They are

less likely to use some applications such as, email, voice/video call, online games

compared to those in the east. Living in urban areas also increases usage of e-trade,

trip planning, health appointments, online games; and decreases voice/video calls and

getting news online compared to rural areas.

5 Discussion and implications

This paper examines the determinants of the broadband access and broadband usage

in Turkey. Findings of the paper strongly verify the results in the earlier literature; the

regional approach and the categorization of online activities create notable

contributions.

Our main findings related with the determinants of the broadband access are in line

with the earlier studies. Education, age, gender, employment, and availability of fixed

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

128

and mobile telephony are significant determinants. Our regional approach, which is

based on congregating NUTS1 level regions into West-Central-East, indicates

significant differences among the regions. We verify that the interregional disparities

also exist in the probability of broadband adoption. Individuals living in the west and in

central Turkey are more likely to adopt broadband access compared to those in the

east. The availability of fixed networks might be considered as a barrier, but as

described in Table 1, there exists an excess capacity for fixed telephony network. A

possible explanation for the lower probability could be the relatively higher cost in the

east. In Turkey, fixed and mobile operators offer identical prices for their own access

services in all regions. Considering the regional differences in income, adopting

broadband access in the east with those undifferentiated prices becomes relatively

costly. Creating incentives by the national regulatory authority (ICTA) for the operators

to differentiate their prices regionally could be a policy implication of this result.

Furthermore we find that having kid(s) at primary school at home increases the

probability of broadband adoption. In a number of studies it was suggested that

children, teenagers may facilitate internet adoption and influence online behaviors of

family members (Hargittai, 2003; Kiesler et al., 2000; Kraut et al., 1996). However, this

finding has a particular importance for Turkey. The government has just initiated a

countrywide investment project (FATiH project) 6 , which is based on providing ICT

6 See, http://fatihprojesi.meb.gov.tr/tr/english.php (accessed on November 28, 2013).

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

129

equipment to classes, distributing computers to students and teachers in order to

achieve the ICT supported teaching until the end of 2014. Our finding supports the

implementation of such a project in terms of its interactive effect on the broadband

access adoption. Specifically, this program could reduce the regional gap by providing

computers to school-aged children which gives incentive to adopt broadband access at

home.

Findings related with the usage of online activities are also in line with the previous

literature. We confirm that being younger or having a degree above high school

increases the probability of using online activities. We verify that males are more likely

to use the activities than females. Furthermore, we find that having a job decreases

the probability of using entertaining activities such as, social networking, voice/video

call, and downloads.

Most importantly, our categorization of online activities as advanced, entertaining and

conventional activities, and their usage among regions create an empirical contribution

to the literature. We find that conventional activities such as, sending or receiving

emails and getting news, are common in all regions or more likely to be used in the

east compared to the west or the central. Nevertheless, advanced activities such as

uploading content, searching information, streaming media, getting health

appointments, trip planning, e-trade, and e-banking, are more likely to be used in the

central region compared to the east. Most of those mentioned activities (except

uploading content, trip planning and e-trade) are also to be used with higher

probabilities than in the east. Moreover, entertaining online activities such as social

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

130

networking, voice/video calling, and online gaming, are more likely to be used in the

east than in the west and in the central.

In fact, those findings strikingly set out how online activities change across regions

consistent with regional characteristics. One possible explanation of this variation

could be the lower opportunity cost of time in the east, owing to a higher

unemployment rate and lower income, which increases the likelihood of participating

in entertaining / time consuming activities.

Individuals in the West have a noticeably higher probability in engaging in advanced

activities, even after education, gender, age, and household size are controlled. We

suspect the financial literacy as well as experience in such activities plays an important

role in this variation. Unfortunately, we are unable to find the basis of this difference

due to unavailability of necessary information. Still, region specific policies such as

providing training in such activities or encouraging the use of e-government might

overcome the regional disparities.

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

131

References

Cerno, L., & Amaral, T. P. (2006). Demand for Internet access and use in Spain. In In

Governance of Communication Networks (pp. 333-353). HD: Physica-Verlag. Chaudhuri, A., Flamm, K. S., & Horrigan, J. (2005). An analysis of the determinants of internet access. Telecommunications Policy, 29(9), 731-755. Commission, E. (2010, August 26). A Digital Agenda for Europe. Retrieved June 12,

2012,http://eurlex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:52010DC02 45R(01):EN:NOT

Doğruel, F., & Doğruel, A. S. (2003). Türkiye’de Bölgesel Gelir Farklılıkları ve Büyüme. In A. H. Köse, F. Şenses, & E. Yeldan (Eds.), Küresel Düzen: Birikim, Devlet ve Sınıflar (pp. 237-318). İletişim Yayınları.

Flamm, K., & Chaudhuri, A. (2007). An analysis of the determinants of broadband access. Telecommunications Policy, 31(6), 312-326.

Gezici, F., & Hewings, G. J. (2004). Regional convergence and the economic performance of peripheral areas in Turkey. Review of Urban & Regional Development Studies, 16(2), 113-132.

Goldfarb, A., & Prince, J. (2008). Internet adoption and usage patterns are different: Implications for the digital divide. Information Economics and Policy, 20(1), 2- 15.

Hargittai, E. (2003). Informed web surfing: The social context of user sophistication. In S. Jones, & P. Howard (Eds.), Society Online I (pp. 257-274). Sage Publications. Retrieved November 28, 2013, from http://www.webuse.org/pdf/Hargittai- InformedWebSurfing2003.pdf

Hargittai, E., & Hinnant, E. (2008). Digital inequality: Differences in young adults' use of the Internet. Communication Research, 35(5), 602-621.

Hargittai, E., & Shafer, S. (2006). Differences in actual and perceived skills: The role of gender. Social Science Quarterly, 87(2), 432-448.

Heckman, J. (1979). Sample selection bias as a specification error. Econometrica, 47(1), 153-161.

ICTA. (2012). Electronic Communications Market in Turkey Quarterly Market Data (Q4). Retrieved September 14, 2013, from http://eng.btk.gov.tr/dosyalar/2012-4- English_15_03_2013.pdf

ICTA. (2013). The Bulletin of the Yearly Statistics at Provincial Level for Turkish Electronic Communications Market (2007-2012). Retrieved July 15, 2013, from http://eng.btk.gov.tr/kutuphane_ve_veribankasi/yil_istatistikleri/ehsyib.pdf

Karaca, O. (2004). Turkiye’de Bolgeler Arasi Gelir Farkliliklari: Yakinsama Var Mı? Turkiye Ekonomi Kurumu Tartisma Metni, 7.

Kiesler, S., Zdaniuk, B., Lundmark, V., & Kraut, R. (2000). Trouble with the Internet: The dynamics of help at home. Human-Computer Interaction, 5, 323-351.

Kraut, R., Scherlis, W., Mukophadhyay, T., Manning, J., & Kiesler, S. (1996). The HomeNet Field Trial of Residential Internet Services. Communications of the ACM, 39(12), 56-63.

Topics in Middle Eastern and African Economies Vol. 17, Issue No. 1, May 2015

132

Moutafides, G. M., & Economides, A. A. (2011). Demand for broadband access in Greece. Telematics and Informatics, 28(2), 125-141.

OECD. (2013). OECD Communications Outlook 2013. Paris: OECD Publishing. Özaslan, M., Dinçer, B., & Özgür, H. (2006). Regional disparities and territorial

indicators in Turkey: Socio-economic development index. In ERSA conference papers ersa06p858. European Regional Science Association.

Spooner, T., Meredith, P., & Rainie, L. (2003). Internet use by region in the United States. Pew Internet & American Life Project. Retrieved 10 28, 2013, from http://www. pewinternet. org

Srinuan, C., & Bohlin, E. (2013). Analysis of fixed broadband access and use in Thailand: Drivers and barriers. Telecommunications Policy, 37(8), 615-625.

The Berkman Center for Internet & Society. (2010). Next Generation Connectivity: A review of broadband Internet transitions and policy from around the world. Harvard University. Retrieved November 11, 2013, from http://cyber.law.harvard.edu/sites/cyber.law.harvard.edu/files/Berkman_Cent er_Broadband_Final_Report_15Feb2010.pdf

Torlak, Ö., Spillan, J. E., & Harcar, T. (2011). Young Consumers’ Cell Phone Usage in Developing Market: The Case of Turkish Youth Market. Journal of Marketing Development and Competitiveness, 5(3), 47-67.

CAP_Turkeys_Digital_Divides.pdf

Turkey’s Digital Divides By Max Hoffman June 2016

W W W. A M E R I C A N P R O G R E S S . O R G

A P

P H

O TO

/LEFTER IS P

ITA R

A K

IS

Turkey’s Digital Divides By Max Hoffman June 2016

1 Introduction and summary

4 The digital stakes for Turkey

14 Competition and regulation in Turkey

21 How does Turkey compare?

27 Conclusion: Closing Turkey’s digital divide

29 About the author

30 Endnotes

Contents

1 Center for American Progress | Turkey’s Digital Divides

Introduction and summary

Turkey is emblematic of the promise of the digital revolution that is now sweeping through similarly emerging middle-income democracies around the world. Yet its approach to expanding internet penetration is shaped by its own set of political and social conditions. Wider internet access and use could contribute to a more dynamic Turkish economy that is driven by greater online competition and entre- preneurship. Turkey could likewise provide more efficient, responsive government services to more of its citizens by harnessing information and telecommunications technologies. But the efforts to reap these rewards are hindered by wide dispari- ties in internet access and online literacy, as well as by very different customs that divide men and women, the old and the young, and urban and rural citizens.

These divides are evident in the nation’s digital disparities and have roots in the country’s recent political history and social norms. At first glance, Turkey’s rapid but uneven economic development over the past several decades—with all of the accompanying social fissures—is akin to the experiences of other developing nations such as Albania, Chile, or Brazil. And Turkey, like other emerging middle- income democracies, is grappling with the need to privatize the internet and com- munications industries, which are often powerful political players with deep ties to ruling parties and with little interest in fostering serious online competition.

But Turkey’s challenge of providing more internet accessibility at more afford- able prices—a key step to becoming a full-fledged developed democratic nation and a new member of the European Union—faces obstacles that are particular to Turkey’s political economy. The ruling Justice and Development Party—more commonly referred to by its Turkish acronym, AKP—must calculate the political gains and losses of more widespread internet access, particularly for those conser- vative working-class rural voters who are the backbone of its political strength.1

In some ways, Turkey’s digital divide is just fine with the AKP and its de facto leader, President Recep Tayyip Erdoğan, because it gives the government greater control of the information flowing to its most important constituencies. But in other ways, the government recognizes the costs and consequences of Turkey’s

2 Center for American Progress | Turkey’s Digital Divides

digital divide and is actively seeking to bridge the gaps. The question is whether the nation can achieve this transformation and broaden access to the internet and its associated benefits despite the likelihood that this process would loosen the governing authorities’ control over online information that it wants to suppress.

Troublingly, Turkey’s state influence over the internet and telecommunications sec- tors aligns in particular with its efforts to constrain the internet as a space for public access, information, and expression. The country’s extensive internet censorship rules take both legal and extralegal forms, with the government relying sometimes on changes implemented through legitimate political channels and sometimes through the selective or delayed implementation of rules and court decisions. These rules limit the degree to which those with minimal or no access to the internet— many of them supporters of the ruling party—can access information. More encour- agingly, however, some of Turkey’s genuine efforts to boost other aspects of online accessibility and literacy could well lead to a more robust economy and, eventually, a more open and free internet, as is happening among similar economies around the world. This should be the goal of the Turkish government.

This report examines the costs and consequences of this digital divide in Turkey and explores the reasons for the current disparities, the efforts by the government to overcome some of the divisions, and the prospects for success in all of these efforts.

The report begins by detailing the disparities in internet access between Turkish men and women. These disparities are alarming for a high-middle income country that sets its European neighbors as its benchmark. The report details the reasons why Turkey has struggled to develop internet literacy and to deliver content relevant to a wider swath of its population and outlines the ripple effects of these challenges on rural adoption and e-government penetration and effectiveness. It also outlines the consequences of Turkey’s failure to develop a healthy privatized market for the advancement of high-quality, low-cost internet and communications services.

The report closes with some recommendations for the Turkish government and its citizens to consider. In particular, the government should take the following measures:

• Expand e-government services to reduce the need for some in-person interac- tions with the state bureaucracy. This step could entice more individuals who are low-income and less tech-literate to find value in internet usage and improve the efficiency of key government services.

3 Center for American Progress | Turkey’s Digital Divides

• Develop creative ways to diminish the effects of abiding cultural norms that hamper internet access for women. Public internet access points that are located outside of strict familial control in the home could afford women from conservative families the freedom and anonymity to explore and find useful content on the web.

• Study how other developing democracies are handling their own expansions of online business opportunities for small- and medium-sized businesses, as well as how they are handling access to information online while holding true to their recent democratic gains. Albania, Chile, and Brazil offer useful comparisons for the Turkish government.

These are not trivial goals for Turkey to pursue. Extensive investigations by the World Bank and the Organisation for Economic Co-operation and Development, or OECD, into the importance of widespread internet adoption and use among middle-income countries such as Turkey demonstrate the broad economic and political benefits that can accrue to more fully digitized countries.2 For the nation to achieve its larger sociopolitical and socioeconomic goals, closing the digital divide is of paramount importance.

4 Center for American Progress | Turkey’s Digital Divides

The digital stakes for Turkey

The digital transformation that is sweeping developing middle-income nations, including Turkey, presents tremendous economic and political opportunities, yet the widespread and effective deployment of these technologies has not been fully realized. The adoption of a more expansive and free-wheeling internet can open opportunities for greater political inclusion, more dynamic economic growth, increased government efficiency, and expanded innovation. These gains can be realized at the individual, business, and governmental levels.

As the World Bank’s “World Development Report 2016: Digital Dividends” details, the internet offers individuals greater inclusion in economic, political, and social activity through access to information—everything from news to employment opportunities. For businesses, the internet can reduce barriers to market entry and increase efficiency in the workplace. Governments can better gather data, deploy services, increase participation, and understand and address the public voice.3

Leveraging these technological advancements depends on the capable delivery of internet and telecommunications services and their effective usability. This means that while universal access and affordability are prerequisite achievements, they are not sufficient to drive change. As the World Bank’s report concludes, digital and internet literacy, protections of security and user privacy, and healthy internet gover- nance can all affect the internet’s ability to deliver value to a nation’s citizenry.4

Though some progress has been made, middle-income countries continue to grapple with the challenges of access and affordability. A variety of factors con- tribute to these enduring challenges: the underdevelopment of fixed broadband in countries that leapt headlong into the mobile telephony era; the difficult transition from state-owned telecommunications enterprises to true privatization; the online and telecommunications regulatory shortcomings that hamper robust competi- tion, particularly unpredictable tax outcomes and occasionally punitive tax fines; and the difficulty of developing infrastructure to serve far flung populations.5

5 Center for American Progress | Turkey’s Digital Divides

Turkey is grappling with all of these challenges that typify the advancement of internet and telecommunications in the developing world, yet the country exhib- its some stark areas of underperformance compared with its peers—underper- formance that is linked to the nation’s cultural norms and recent political history. A thorough understanding of Turkey’s unique obstacles to developing a robust internet economy and society must begin with a brief history to help place the present hurdles in appropriate context.

While the developed economies of the world saw the internet take off in the 1990s, Turkey remained a poor, developing country. After several years of strong economic growth under the leadership of president and economic and political reformist Turgut Özal, Turkey’s nominal gross domestic product, or GDP, per capita hovered around $4,000 through the late 1990s compared with the United States’ GDP per capita of more than $34,000.6 In the United States—the epicenter in so many ways of the digital revolution—that same period marked the beginning of a sharp increase in the share of Americans who started using the internet. In 1995, just 14 percent of Americans used the internet; by 2001, more than 60 percent were online.7

In Turkey, the internet revolution remained some ways off. The internal political and economic crises of the late 1990s and the 2001 crash of the Turkish economy sparked capital flight, massive inflation, and painful job losses.8 The early 2000s saw political stabilization after the crisis, first under technocratic leadership that spanned the Bülent Ecevit-helmed coalition government in place from 1999 to 2002 and then under the newly formed Justice and Development Party, which won a parliamentary majority in 2002 and remains in power today.

All of this political and economic upheaval, followed by steady steps at eco- nomic and political stabilization, did little to boost internet activity in Turkey. By 2005, with household internet access in the United States nearing 60 percent, only 8.7 percent of Turkish households had access. Similarly, less than 18 percent of Turks used the internet that year compared with more than 70 percent of Americans. In 2005, only 23 percent of Turks had even used a computer.9 Across Western Europe, the share of households with internet access ranged between 40 percent and 80 percent.10 In short, by 2005, Turkey already faced a considerable deficit in advancing internet and telecommunications pen- etration throughout its economy and society.

But then, over the ensuing decade, Turkey experienced a rapid expansion of inter- net adoption and use similar to what the United States had seen a decade earlier. Computer use increased to 55 percent of the population by 2015. Internet usage spiked to 56 percent of the population. Nearly 70 percent of households had some

Major indicators for Turkey

Population

77.7 million

Mobile telephone

subscribers (2014)

71.9 million

Fixed line

subscribers (2014)

8.3 million

Mobile broadband

subscribers (2014)

24.1 million

Broadband

subscribers (2014)

32.5 million Source: Investment Support and Promotion Agency, “Infrastructure and Logistics,” Republic of Turkey, Prime Ministry, available at: http://www. invest.gov.tr/en-US/investmentguide/investors- guide/Pages/InfrastructureAndLogistics.aspx (last accessed June 2016).

6 Center for American Progress | Turkey’s Digital Divides

form of internet connectivity in the same year.11 Turkey benefited from steady economic growth that tracked the overall rise in emerging markets over the same period. Foreign capital inflows fueled growth and rising GDP per capita, and inter- net access became an affordable discretionary expense for many more Turkish citi- zens.12 Yet a deep digital divide developed and remains within the nation’s culture, geography, demography, and political economy.

The digital divide in Turkey today

In part because of the rapid expansion of access over the past decade, Turkey skipped a technological generation. Only one-quarter of Turkish households have desktop computers. Less than 30 percent of households have fixed telephone lines. At the same time, nearly 97 percent of households have mobile phones. Of Turkey’s 32.5 million broadband internet subscribers, just more than one-quarter have fixed lines, while nearly three-quarters have mobile broadband access.13 Turkey’s eco- nomic rise and strides in internet access have leapfrogged the era of fixed connectiv- ity and moved full-scale into the era of mobile telephony. In 2015, nearly 70 percent of households had access to the internet, but individual computer usage stood at 55 percent of Turkey’s population, with internet usage standing at about 56 percent.14

Alas, these clear gains in internet access are not universally shared. Beyond the still-limited penetration of internet access, usage statistics reveal marked disparities across urban-rural, regional, and educational lines, as well as age and gender gaps.

7 Center for American Progress | Turkey’s Digital Divides

Total Male Female

Total 55.9 65.8 46.1

A Istanbul 67.1 74.9 59.1

B West Marmara 59.4 67.1 51.4

C Aegean 56.9 65.9 48.1

D East Marmara 59.2 67.5 51

E West Anatolia 62 70.3 54.1

F Mediterranean 52.6 62.1 43.1

G Central Anatolia 51.5 61.7 40.7

H West Black Sea 50 60.9 39.4

I East Black Sea 52.1 62.3 41.1

J North East Anatolia 40.2 50.2 30.4

K Central East Anatolia 40.2 55.7 24.6

L South East Anatolia 43.1 59.6 28.1

Source: Turkish Statistical Institute, “Information and Communication Technology (ICT ) Usage Survey on Households and Individuals, 2015,” (August 18, 2015), available at: http://www.turkstat.gov.tr/PreHaberBultenleri.do?id=18660.

FIGURE 1

Percentage of internet users by region (2015)

H

G

F

EC

D B

A

I J

K

L

Urban-rural, regional, and educational lines

The urban-rural, regional, and educational divides are clear. While 58 percent of urban individuals use the internet, just less than 29 percent of rural individuals are users.15 While 67 percent of residents in the Istanbul metropolitan area use the internet, only 40 percent in Central Anatolia and 43 percent in Southeast Anatolia use the internet.16 The educational divide is particularly stark. While 6 percent of those with no education and one-quarter of those with primary schooling use the internet, 94 percent of university-educated people and 80 percent of high school- educated people count themselves as internet users.

8 Center for American Progress | Turkey’s Digital Divides

In all likelihood, these three divides exhibit some covariance, namely related to wealth and income. The U.N. Conference on Trade and Development produced a report detailing the high cost of internet service in low-income countries, where internet affordability can be 150 times greater than in a high-income country.17 The effect is certainly intensified in especially low-income segments of middle- income developing countries such as Turkey, as this report examines in more detail in the next section. Educational attainment correlates strongly with income, and urban areas offer greater economic opportunity than rural ones. Urban areas also offer alternative internet access points beyond the home, such as internet cafes, which may be absent from some remote areas.

Age gap

The age gap in internet usage is also pronounced but may be explained by Turkey’s rapid advances in internet access over the past decade. While less than 6 percent of citizens ages 65 to 74 are internet users, 77 percent of 16- to 24-year-olds and 72 percent of 25- to 34-year-olds use the internet. Among seniors, the number of internet users remains durably low. Ten years ago, when Turkey’s eldest citizens were 55 years to 64 years of age, only 2 percent of them used the internet—the same share for senior citizens today.18

TABLE 1A

Percentage of internet users by age

16-24 25-34 35-44 45-54 55-64 65-74

2005 27.8 16.7 9.7 6.3 2.3 5.6

2015 77 71.7 55.4 34 17.2 5.6

TABLE 1B

Percentage of internet users by education

No diploma

Primary school

Secondary school

High school University

2005 0.4 1.7 15.1 32.9 62.6

2015 6.1 25.9 67.2 80.6 94.1

Source: Turkish Statistical Institute, “Information and Communication Technology (ICT ) Usage Survey in Households and Individuals, 2004-2015,” surveys for all years available at: http://www.turkstat.gov.tr/PreHaberBultenleri.do?id=18660

9 Center for American Progress | Turkey’s Digital Divides

In Istanbul’s working class neighborhoods, interviews with senior citizens indicate the recognizable barriers to elder adoption of internet usage: lack of perceived rel- evance and self-doubt at learning something new and seemingly complex. “I don’t know anything about it,” said one 70-year-old retired factory worker. “I was never interested to learn about it.” When asked whether he would be interested in social media as a mode of communication with relatives, he denied needing the internet for that purpose. Similarly, a 57-year-old homemaker expressed doubt about the personal value of internet access before adding, “Even if I were [interested], how would I learn at this age?”19

These conversations reveal the major difficulties of driving technological adop- tion among older people—a challenge that shows up in many countries beyond Turkey. But just as in other middle income countries, Turkey’s 25- to 34-year- olds have wholeheartedly adopted internet use. In 2005, when that same cohort was 16- to 24-years-old, only 28 percent used the internet. Among today’s 35- to 44-year-olds, 55 percent use the internet; only 17 percent were users in 2005, as 25- to 34-year-olds. Nearly 4 out of 5 Internet native Turks—the youngest cohort surveyed—count themselves as users.20 While the age gap has left some Turks behind, the trend lines in Turkey are positive.

Gender disparities

More troubling are the use and access disparities between men and women. While 66 percent of men use the internet nationally, only 46 percent of women do. This marked disparity exists across the urban-rural divide, too. In urban areas, 68 per- cent of men and 48 percent of women are internet users compared with rural areas, where 39 percent of men and 18 percent of women report using the internet.21

10 Center for American Progress | Turkey’s Digital Divides

TABLE 2

Information and communications technology (ICT ) usage

Households and individuals

2005 (%) 2015 (%)

Computer usage (total) 22.9 54.8

Male 30.0 64.0

Female 15.9 45.6

Urban (total) 29.6 59.0*

Male 37.3 69.0*

Female 21.7 49.1*

Rural (total) 11.7 29.5*

Male 17.1 40.1*

Female 6.6 19.3*

Internet usage (total) 17.6 55.9

Male 24.0 65.8

Female 11.1 46.1

Urban (total) 23.1 58.0*

Male 30.3 68.1*

Female 15.6 48.0*

Rural (total) 8.2 28.6*

Male 12.8 39.2*

Female 3.9 18.4*

Households with access to internet

8.7 69.5

* 2013 data was the most recent available.

Source: Turkish Statistical Institute, “Information and Communication Technology (ICT ) Usage Survey in Households and Individuals, 2004- 2015,” surveys for all years available at: http://www.turkstat.gov.tr/PreHaberBultenleri.do?id=18660.

One factor shaping this dynamic is women’s access to education. While 94 percent of female university graduates and 79 percent of female high school graduates use the internet, less than 5 percent of women without a diploma are users.22 Turkey remains a country in which fewer educational opportunities are open to women than to men, and educational attainment strongly correlates with internet use. Illiteracy is symptomatic of this problem: A program to bring internet-use training to women in the southeastern city of Şırnak discovered that women’s illiteracy proved one of the greatest factors hampering the program’s success.23

11 Center for American Progress | Turkey’s Digital Divides

Enduring social norms in particular societal segments may also hinder women’s free access to the internet. Interviews with individuals from working-class, con- servative neighborhoods in Istanbul revealed indications of differential access for young men and women growing up in these communities. One mother insisted that while her sons were allowed to go online, her daughters were not—and that this state of affairs would continue until their marriages. “[My daughter] can do whatever she likes when she gets married. It’s her husband’s problem,” the inter- viewee said. “Until then, no internet, no such things.”24 Other interviewees echoed this view, suggesting that cultural constraints on women constitute one aspect of the gender divide in internet use.

Internet access and use in Turkey

Turkey’s digital divide manifests itself in a variety of ways. The country boasts a big business sector that uses internet access and advanced telecommunications services in the same way that businesses in other developed and major devel- oping member nations of the Organisation for Economic Co-operation and Development do. At the same time, Turkey’s overall internet adoption rates have ranked consistently among the lowest in the OECD.25

Indeed, in the OECD’s “Digital Economy Outlook 2015,” member nations’ efforts to expand broadband networks and service access to their entire populations is a recurring theme due to the economic benefits of high-speed internet and cloud- based computing and data-storage services.26 Turkey’s largest conglomerates have adopted high-speed broadband service and its attendant enterprise service benefits, but Turkey’s government faces a more basic challenge in enabling access and driving adoption of the internet among small- and medium-sized businesses and a wider swath of the population.

One key barrier to deepening internet access and use is the availability of useful content online for new adopters. This may seem surprising given the myriad range of content available, including in Turkish, but it underscores the cultural problems that inhibit the growth of internet use in the country. Consider Uşak, a medium- sized city in Anatolia about equidistant from Antalya and Istanbul. In Usak, Osman Coşkunoğlu—a former member of parliament, engineer, and internet-use expert— held internet training courses for women. Osman taught the participants how to access and use the internet, yet subsequent adoption—that is to say, continuing use—was limited because the women struggled to find meaningful content online.

12 Center for American Progress | Turkey’s Digital Divides

When interviewed, Coşkunoğlu said, “At the root, the crucial aspect is the content: If there was suitable content for all groups, then there would be demand, as well.” Indeed, in most rural areas of the country, many new internet users seek access only to social media.27

TABLE 3A

How do urban and rural people use the internet? (2013)

Purposes Turkey Urban Rural

Sending/receiving e-mails 62.5 63.8 56.1

Participating in social networks 73.2 72.1 78.3

Reading or downloading online news 75.6 76.4 71.8

Subscribe to news services or products to receive them regularly

21.3 21.8 18.7

Seeking health-related information 59.6 62.4 46.2

Looking for information about education, training or course offers

45.9 47.7 37.5

Finding information about goods or services 59.9 62.7 46.3

Downloading software (other than games software) 19.1 20 14.7

Posting opinions on civic/political issues via websites 28.7 28.9 28

Taking part in online consultations or voting to define political issues

12.8 13.6 9.1

Doing an online course (in any subject) 8.4 9 5.5

Consulting wikis to obtain knowledge on any subject 32.6 33.8 26.2

Looking for a job or sending a job application 12.9 13.8 8.4

Participating in professional networks (Linkedln etc.) 4.2 4.7 1.7

Using services related to travel 26.6 27.4 25.5

Selling of goods or services, e.g. via auctions 9.3 10 6

Telephoning / video calls over the Internet 55.1 56.6 47.4

Internet banking 24.8 26.8 15.4

Note: Respondents may choose more than one option, so columns do not equal 100%.

Source: Turkish Statistical Institute, “Information and Communication Technology (ICT ) Usage Survey on Households and Individuals, 2013,” (August 22, 2013), available at: http://www.turkstat.gov.tr/PreHaberBultenleri.do?id=13569.

The Turkish government is actively trying to provide useful content to current and new internet users via access to government services. In Şırnak, women applied for internet training through the Life is Simpler with Internet project, whose moti- vation, in part, is to promote wider use of the online appointment system used by all Turkish state-run public hospitals. The public school system also uses a central- ized online system, called E-Okul, or e-school. Parents are able to follow their children’s academic progress and class notes through the online system, which the government hopes is another potentially motivating factor for wider adoption.28

13 Center for American Progress | Turkey’s Digital Divides

TABLE 3B

How do men and women use the internet? (2015)

Purposes Turkey Male Female

Sending / receiving e-mails 49.5 52.9 44.5

Telephoning / video calls over the Internet 38.1 38.8 37.1

Participating in social networks 80.9 83.1 77.6

Uploading self-created content (text, photos, music, videos, software etc.) to any website to be shared

62.1 63.9 59.6

Reading online news / newspapers / news magazines 70.2 72.3 67.2

Seeking health-related information 66.3 60.1 75.2

Looking for information about education, training or course offers

41.3 37.3 47.1

Consulting wikis to obtain knowledge on any subject 47.2 46.1 48.8

Finding information about goods or services 59.4 61.9 55.7

Downloading software (other than games software) 19.5 22.7 14.8

Posting opinions on civic or political issues 22.3 24.2 19.7

Taking part in on-line consultations or voting to define civic or political issues

12 13.1 10.5

Looking for a job or sending a job application 10.1 10.2 9.9

Participating in professional networks 4.4 4.7 3.9

Using services related to travel 19.7 21.2 17.7

Selling of goods or services (e.g. eBay) 20.8 23.1 17.4

Internet banking 29.3 34.5 21.8

Source: Turkish Statistical Institute, “Information and Communication Technology (ICT ) Usage Survey on Households and Individuals, 2015,” (August 18, 2015), available at: http://www.turkstat.gov.tr/PreHaberBultenleri.do?id=18660.

Elsewhere in government, existing e-government offerings include the prime min- ister’s online complaints system to promote government participation and transpar- ency. In theory, this could help drive internet adoption by people in poor and rural areas—men and women and young and old alike—who want more responsive gov- ernment services. In practice, the system has been used most intensively in Istanbul and other metropolitan areas such as Konya, Adana, and Samsun. Moreover, the plu- rality of the system’s users have been well-educated urban dwellers, with nearly half of the submissions coming from high school and university graduates.29

The Turkish government’s Fatih Project also aims to bring technology to more of society by advancing the adoption of technology in schools. The government program outfits classrooms with smart boards and provides tablets to students. Internationally, these efforts often stem from a belief that technological advance-

14 Center for American Progress | Turkey’s Digital Divides

ment in classrooms will improve learning outcomes, but the Turkish government also seeks to use schools to drive the wider adoption and use of the internet.30

As Coşkunoğlu explains, however, the “second digital divide”—the lack of tech- nological literacy necessary to derive benefit from these technologies—often inhibits poorer people from fully embracing the internet. They are reticent to use these technologies, and as a result, affluent citizens realize most of the gain from e-government and technological innovation initiatives.31 Absent a more intensive effort to make internet use more meaningful, subsidizing technological adoption and dissemination via classrooms may have limited success.

15 Center for American Progress | Turkey’s Digital Divides

Competition and regulation in Turkey

Competition in the internet service provider market is an enduring challenge in Turkey. Law No. 4502 decreed that Türk Telekom would lose its monopoly rights in 2003. Yet by 2008, the telecommunications giant still controlled 91 percent of service and 81 percent of revenue in the telephone services market.32 A 2015 report from the Turkish Information and Communication Technologies Authority, or BTK, reported that Türk Telekom retained a 75 percent market share in the internet service provider, or ISP, market through its subsidiary TTNet.33 Essentially, Türk Telekom sells wholesale service to independent ISPs and simultaneously competes with those providers in the retail market through its TTNet retailing arm, resulting in an unfair playing field for retail competi- tors and reduced pressure on Türk Telekom for high-quality service.34 Though more than 70 other internet service providers compete for market share, Türk Telekom remains the dominant provider.35 Indeed, Türk Telekom was fined 12.4 million Turkish lira, about $9 million, in 2008 for abusing its dominance in the wholesale broadband internet market.36

Part of Türk Telekom’s continuing dominance stems from the lack of rigorous government policing of the devolution of Türk Telekom’s monopoly. In 2005, Türk Telekom was nominally privatized. Yet the company retains close ties to the government. Turkey’s Undersecretariat of Treasury still owns 35 percent of the company;37 the Undersecretary of the Turkish Prime Ministry, Acting Deputy Undersecretary for Transportation of the Ministry of Transport and Communications, the Presidential Secretary-General, and the Chief Advisor to President Erdoğan on economic issues serve on the board of directors.38 In 2014, then-Prime Minister Erdoğan appointed a close adviser to the Türk Telekom board, despite his adviser having no background in telecommunica- tions.39 These intimate relationships between the private sector and the govern- ment hinder Türk Telekom’s privatization efforts.

16 Center for American Progress | Turkey’s Digital Divides

This state of affairs is not unfamiliar to Turkey. Privatization of state-owned enterprises has proceeded sluggishly in other sectors as well. Innovation and competition have sometimes been the casualties of this difficult process. In the energy sector, for example, government monopolies have long dominated explo- ration and production, the refining process, and construction and maintenance of pipelines through Türkiye Petrolleri AO, Tüpraş, and BOTAŞ—the state- owned petroleum exploration and production company, refining company, and pipeline company, respectively.

Given this context and its record thus far, it is fair to question the government’s zeal for genuine privatization and real competition in the internet and telecommu- nications sectors. According to former Member of Parliament Coşkunoğlu—for- merly of the Republican People’s Party, or CHP, the main opposition party—the government gains considerable advantages from the ongoing state of affairs with Türk Telekom through intelligence and surveillance capabilities and through continuing influence over board seats and annuity payments.40 Indeed, the close adviser who President Erdoğan named to Türk Telekom’s board in 2014, eco- nomic advisor Yiğit Bulut, is reported to make more than $160,000 per year from his board appointment.41 Absent sectoral reforms, Turkey will struggle to fulfill its potential for growth in infrastructure and innovation in internet service provision.

FIGURE 2

Internet usage statistics

Information Society Statistics by TUIK 2015

Source: Turkish Statistical Institute, “Use of Information and Communication Technology (ICT) by Enterprises, 2015,” (September 9, 2015), available at: http://www.turkstat.gov.tr/PreHaberBultenleri.do?id=18647.

95.2%

92.5%

65.5%

54.8%

64.0%

45.6%

55.9%

65.8%

46.1%

69.5%

ICT usage in companies

Computer usage

Internet access

Having website

Internet usage (total)

Households with access to the internet

Female

Female

Male

Male

Computer usage (total)

ICT usage in households and individuals

17 Center for American Progress | Turkey’s Digital Divides

Nonetheless, at 92 percent, Turkey has a fairly high rate of internet access for small- and medium-sized enterprises with more than 10 employees.42 Broadband pricing data is difficult to synthesize, as there are many different offerings of vary- ing speed, size, and format, either mobile or fixed. Turkey has a wide range of prices; extremely basic services can be quite affordable, but faster speeds or higher data limits—requirements for business activity—rapidly increase in cost. Overall, fixed broadband access is still quite expensive in Turkey, particularly for the poor. The average of all fixed broadband service offerings—including different speeds and data limits—in Turkey, as reported to the OECD, was the third-highest among OECD countries at more than $83 per month in purchasing power parity terms for residential access. It is worth noting that this data is influenced by the status of the lira.43 For those at the bottom of the income ladder, even entry-level plans can cost 15 percent to 25 percent of the average income, a worse relative position for the nation’s poor than in other countries that were surveyed.44 The cost of fixed broadband internet access in Turkey limits the kind of at-home access individual citizens generally rely on for activities like e-commerce, job searches and applications, and in-depth research.45

By some measures, Turkey’s overall mobile—as opposed to fixed broadband— internet affordability is quite good for individuals when measured against peers: An entry-level mobile broadband plan—500 megabytes, prepaid—costs less than 1 percent of per capita gross national income, which is the Alliance for Affordable Internet’s preferred measure of affordability.46 Yet, again, faster third-generation, or 3G, mobile broadband plans or higher data limits can be expensive. An entry- level plan with 2 gigabytes of data can cost $12 to $17 per month, while plans with 15 gigabytes of data can cost $55 per month including taxes and other fees. With Turkey’s net monthly minimum wage at about $450, for the poor, even mobile internet access is a stretch.47 These cost disparities are a major problem for a coun- try trying to shrink its digital divide—a problem compounded by Turkey’s rising censorship of the internet.

18 Center for American Progress | Turkey’s Digital Divides

FIGURE 3

OECD broadband statistics

Average of broadband pricing offers for residential users in the OECD area, September 2011

Note: prices re�ect average of monthly prices for a range of broadband services in U.S. dollars adjusted for purchasing power parity, as reported to the OECD. Source: Organisation for Economic Co-operation and Development, “OECD Broadband Portal,” available at: http://www.oecd.org/inter- net/broadband/oecdbroadbandportal.htm#prices (last accessed June 2016).

Norway

Slovenia

Turkey

Greece

Australia

Spain

Portugal

Chile

New Zealand

Mexico

Canada

Netherlands

Luxembourg

Italy

United States

Switzerland

Poland

Iceland

Ireland

Czech Republic

Slovak Republic

United Kingdom

Sweden

Belgium

Denmark

Finland

Japan

France

Austria

Hungary

Israel

Germany

Estonia

Korea

$126.12

$101.15

$83.41

$72.41

$72.28

$68.92

$64.99

$64.85

$63.62

$60.60

$60.14

$57.05

$56.99

$55.55

$54.35

$53.07

$52.67

$51.12

$49.44

$49.38

$48.24

$46.02

$46.01

$45.34

42.22

$40.87

$39.81

$39.05

$37.43

$37.11

$35.61

$35.16

$28.05

$24.52

19 Center for American Progress | Turkey’s Digital Divides

FIGURE 4

Fixed broadband monthly subscription price ranges (2014)

All platforms, logarithmic scale, in U.S. dollars adjusted for purchasing power parity

Source: Organisation for Economic Co-operation and Development, “OECD Digital Economy Outlook 2015,” (July 15, 2015), available at: http://www.oecd.org/internet/oecd-digital-economy-outlook-2015-9789264232440-en.htm.

Turkey

Portugal

Estonia

Switzerland

Slovak Republic

Israel

Finland

Austria

Japan

Korea

Denmark

Greece

Slovenia

Czech Republic

Hungary

Mexico

Germany

Poland

Italy

Ireland

UK

Belgium

Chile

Australia

Sweden

USA

France

New Zealand

Netherlands

Canada

Iceland

Norway

Spain

Luxembourg

$259.5

$778.1

$51.0

$105.9

$95.1

$38.8

$86.0

$75.8

$52.0

$34.0

$62.9

$65.8

$1,480.8

$65.9

$67.7

$103.8

$64.3

$128.2

$68.7

$84.7

$59.1

$81.2

$170.6

$105.5

$102.5

$300.0

$53.5

$89.2

$89.5

$173.7

$98.9

$290.1

$71.9

$150.7

$13.6

$14.4

$14.7

$16.3

$17.3

$18.0

$18.8

$20.4

$21.7

$22.2

$22.2

$23.2

$23.3

$23.4

$24.6

$24.8

$24.9

$25.5

$28.1

$28.1

$28.3

$29.9

$30.8

$32.0

$32.3

$32.7

$33.7

$33.9

$35.7

$37.0

$38.5

$38.6

$39.6

$43.8

$10 $100 $1000

Low price ▼

High price ▼

20 Center for American Progress | Turkey’s Digital Divides

Censorship in Turkey

The Turkish government possesses broad powers of internet censorship. In 2007, Turkey passed Law 5651, which enabled the government to block websites and webpages for the purpose of “protecting families and minors.”48 It denoted nine catalogue crimes—all in reference to existing crimes in the Turkish Penal Code—for which sites and pages could be blocked by a unit within the Information and Communication Technologies Authority, or BTK, its Turkish acronym. One of the units at BTK, the Telekom İletişim Başkanlığı—which translates to the Turkish Telecommunications Commission/ Presidency—or TIB, can block websites based abroad without a court order and block sites within Turkey, subject to a subsequent court review.49

In practice, however, the legal checks on censorship are weak to nonexistent. Today, TIB can block entire websites for one piece of harmful content on a single sub-page, and even if a court orders such a ban to be lifted, TIB has on several occasions delayed the unblocking for several days. In the second half of 2015, for example, the government unilaterally blocked access to about 100 websites, arguing that the sites were in technical violation of a variety of often unenforced rules. Some of the websites sued the government; the courts agreed with the plaintiffs, deciding that the blocking of access was politically motivated. The government, however, was slow to respond to the court order and contin- ued to block numerous other websites.50

The 2007 law expanded the censorship tools available to the government in its effort to control access to web content within Turkey and led to a surge in blocked sites. As of the beginning of May 2016, there were more than 110,000 blocked websites, compared with just less than 60,000 in 2004.51 The rise in the number of blocked websites was mostly the result of TIB orders and subsequent court orders upholding the regulator.52 What’s more, several recent amendments to the law have further extended the government’s censorship authority. The April 2015 omnibus law amending Internet Law 5651 included new, broad provisions that allowed sites to be blocked for privacy violations and defamation, increased the speed with which sites could be blocked, and extended the authority to initiate such blocks to government ministers.53 All of these actions have brought internet service providers under more direct government control, enabling the govern- ment to meddle with access to content for political reasons.54

21 Center for American Progress | Turkey’s Digital Divides

Turkey’s internet censorship laws are unprecedented among middle-income democracies. As the report examines in the next section, several of Turkey’s peers around the world have nothing close to this level of internet censorship. No devel- oped democracy—in the European Union or elsewhere—imposes a law similar to that of Turkey. Indeed, Turkey’s efforts to censor objectionable political content on the internet appear more akin to efforts to control information in Russia, which Freedom House classifies as a “Consolidated Authoritarian Regime.”55

Many forms of internet censorship in Turkey, however, can be circumvented with relative ease: Those who are motivated to bypass censorship via virtual private networks, mostly the highly educated and politically engaged, are not the gov- ernment’s primary concern. Rather, the Turkish government seeks to control access to information for those citizens who are sympathetic to the ruling AKP government and inclined to support its policies.56 In this way, the government has managed to mirror its control over television and print news sources on the more unruly web.57 Through control over access, speed, and content, the government ensures that information reaching the mainstream internet-using public harmo- nizes with the government’s interests.

22 Center for American Progress | Turkey’s Digital Divides

How does Turkey compare?

Several countries provide useful points of comparison for Turkey’s challenges with driving the adoption of internet usage across the country and society. This section of the report examines three nations—Albania, Chile, and Brazil—that share different aspects of Turkey’s internet development story in the hopes of offering a varied com- parative perspective. Albania shares similar cultural and religious hurdles, along with a similar privatization trajectory. Chile’s government initially struggled to privatize its telecommunications industry, like Turkey, but has since broken with the Turkish story. And Brazil, a powerful emerging economy and democracy, offers comparable infrastructure, educational challenges, and logistical hurdles. All three countries have taken vastly different approaches to censorship and fair access.

FIGURE 5

Wireless mobile broadband subscriptions

Total per 100 inhabitants, Q2 2010 – Q2 2015, with other OECD members

Source: Organisation for Economic Co-operation and Development, “OECD Broadband Portal,” available at: http://www.oecd.org/internet/broadband/oecdbroadbandportal.htm#prices (last accessed June 2016).

0

30

60

90

120

150

2010-Q2 2011-Q2 2012-Q2 2013-Q2 2014-Q2 2015-Q2 2016-Q2

Turkey

23 Center for American Progress | Turkey’s Digital Divides

All three nations and Turkey are also closely ranked in the World Economic Forum’s extensive database that tracks global equitable growth and development bench- marks. In the most recent update of its “digital infrastructure” category in September 2015, Turkey is only slightly behind Chile and Brazil in most of the benchmarks— households with internet access, fixed broadband internet subscriptions, active mobile broadband subscriptions, and affordability of mobile cellular internet—but leads the two countries in the category of affordability of fixed broadband. Although Albania is ranked as a lower-middle-income country in the database, which means the comparisons with upper-middle-income Brazil, Chile, and Turkey are not directly comparable, Albania ranks near the top of its peer nations in all of these same categories except affordability of mobile cellular access.58 The close similarities in these nation’s digital infrastructures point to possible lessons Turkey could learn.

The report then looks briefly at the Broadband Initiatives passed as part of the American Recovery and Reinvestment Act of 2009 in the United States, which in sur- prising ways is seeking to address some of the challenges also faced by Turkey. Indeed, the World Economic Forum’s database of digital infrastructure puts the United States well behind in most categories compared with its peers among advanced econo- mies—except in the category of fixed broadband costs, in which it is the most highly ranked—an indication of the lagging development of most digital infrastructure in rural America.59 The report also looks at the age gap in users’ access to the internet in the United States, which offers additional perspective on Turkey’s similar problem.

Albania

Though far smaller in area and population than Turkey, Albania’s economic and cultural profile is more similar to that of Turkey. After the fall of the Soviet Union, Albania—a majority Muslim nation—ranked among the poorest coun- tries in Europe. Like Turkey, it registered high economic growth rates in the 1990s and 2000s as capital inflows buoyed the economy. And as the economy grew, the Albanian government made internet and telecommunications devel- opment a top priority. The government rapidly expanded e-government ser- vices—moving 100 percent of public procurement online, in addition to the civil registry, licensing, and business registration.60 Albania also improved its national employment service’s job vacancy portal.61

24 Center for American Progress | Turkey’s Digital Divides

Alongside these usefulness improvements, Albania expanded access points for its citizens. At each of Albania’s 510 post offices, free internet service is provided, and the country plans to further expand coverage across the coun- try—presumably through other public buildings.62 Albania also invested in e-classroom initiatives. Beyond merely acquiring and distributing hardware and wiring schools for broadband, Albania has also adopted curricula that are purpose-built for the wired classroom.63

Yet Albania also shares Turkey’s history of uneven and incomplete privatization of its internet and telecommunications industries. The first mobile operator in Albania, Albanian Mobile Communications, was founded in 1996 as a state-run entity and was privatized with the sale of 85 percent of its stock in 2000. Several firms have since entered the market through a flurry of foreign direct investments—includ- ing through one of the privatized companies, ALBtelecom, which the government sold—primarily to a Turkish consortium.64 But competition remains low.

Overall internet usage also remains low. Roughly 60 percent of the population used the internet last year. As in Turkey, Albania mostly skipped the fixed broad- band stage and jumped straight to mobile. About 35 percent of access occurred through mobile phones compared with 5.8 percent through fixed broadband.65

But Albania—despite similarities in its history and some shared cultural norms with Turkey—has taken a fundamentally different approach to internet censor- ship: There is next to no censorship or restriction of internet content. The U.S. Department of State reported in 2015 that the Albanian government “did not restrict or disrupt access to the internet or censor online content, and there were no credible reports the government monitored private online communications without appropriate legal authority.”66

Chile

Chile’s experience with fostering greater internet adoption and usage demon- strates that some of Turkey’s shortcomings are common to other middle-income countries that have enjoyed rapid economic growth. Chile today faces a digital divide between urban and rural residents similar to the one evident in Turkey: 65 percent of urban households in Chile have internet access, but only 40 percent of rural homes have access. Just as in Turkey, internet use came rapidly to Chile. Usage rates have surged in the past 20 years: from zero percent in 1990 to 16.6 percent in 2000, 45 percent in 2010, and 72.4 percent in 2014. From 1997 to 2001, the number of internet users in Chile grew from 200,000 to 2.5 million.67

25 Center for American Progress | Turkey’s Digital Divides

Chile has undertaken some efforts to broaden and deepen internet adoption among its citizenry. The government has created “infocentros”—public spaces with internet connections oriented toward people without private access. Like Turkey, Chile also provides students with computers, but this program has faced effectiveness chal- lenges because some of the population is not technologically literate.68

A primary focus of Chile’s internet development programs has been its busi- ness sector, especially entrepreneurs and small- and medium-sized enterprises. Recognizing that increased interaction with internet and advanced telecom- munications technologies can improve these businesses’ competitiveness and growth prospects, Chile prioritized these efforts under its successive digital action plans from 2008 through 2014.69

Tellingly, Chile’s efforts to deregulate its internet and telecommunications markets and to foster competition have been more effective and complete than those in Turkey. Starting in 1982 with the General Law on Telecommunications, Chile liberalized its internal telecommunication markets to drive private investment in order to connect the entire nation. Legal ambiguities continued to give regulatory agencies the authority to make rulings that shaped the market until 1997, when Chile adopted a deregulation agenda that was considered one of the most innova- tive in the world and created intense market competition.70

Significantly, Chile has paired this open, competitive environment with a degree of national consciousness supporting the idea of fair internet usage: Chile was the first country in the world to adopt national net neutrality laws.71 And, most importantly, the U.S. State Department reports no major restriction or censor- ship of the internet in Chile.72

Brazil

Brazil’s digital divide is more pronounced than in Turkey, Albania, and Chile, likely due to the greater number of people, larger rural areas, and challenging geography. In Brazil, 48 percent of urban households have access to the internet, but only 15 percent of rural Brazilian households do.73

Brazil also experienced rapid internet adoption after a slow initial uptake, though its adoption was slower than Turkey, Albania, and Chile. Brazil’s usage rate stood at 2.8 percent in 2000 but quickly grew to 28.2 percent in 2006 and 57.7 percent in 2014.74

26 Center for American Progress | Turkey’s Digital Divides

This growth pattern is most likely due to lowered costs of access and the process of deregulation that started in 1997, which has improved competition—though not to the same degree as in Chile.

Broadband use in Brazil has accelerated rapidly, mainly due to mobile use, as in Turkey. The 1.7 million mobile broadband subscribers in 2008 grew to 20.6 million by the end of 2010. Fixed broadband growth has been slower, increasing from 10 million to 13.8 million subscribers over the same period. PricewaterhouseCoopers points to low levels of competition as a main factor in the low penetration of fixed broadband in Brazil.75

Brazil, like Chile but unlike Turkey, has also made working toward internet access for all a value-based priority. In April 2014, the Brazilian government enacted the Marco Civil da Internet, a civil bill of rights for the internet that made internet use a right for all Brazilian citizens.76 The bill built upon Brazil’s National Broadband Plan, which was intended to expand internet usage by some 40 million users.77

United States

The United States has faced some of the same challenges experienced by the four middle-income countries discussed above and at about the same time. There is a clear lag in rural-to-urban usage of the internet in the United States, though the gap has narrowed sharply in recent years, partly through the Broadband Initiatives passed as part of the American Recovery and Reinvestment Act of 2009. The Recovery Act allocated a $4 billion grant for a Broadband Technology Opportunities Program aimed at closing the digital divide in the United States by deploying broadband internet infrastructure and expanding public internet access.78 The age gap in internet use in the United States has also narrowed.79

In 2000, 42 percent of people in rural areas connected to the internet, compared with 53 percent in urban areas. By 2015, 78 percent of rural residents were con- nected online, compared with 85 percent of urban residents. Government and private-sector efforts to expand rural broadband use have helped to reduce costs and have offered more ways to connect across the board, contributing to these high overall usage statistics, but these efforts have made a particular difference in rural areas, which had previously lagged far behind cities in access and use.80

27 Center for American Progress | Turkey’s Digital Divides

The United States has also come a long way toward overcoming the age gap that is so visible in Turkey. Americans ages 18 to 29 still use the internet at far higher rates than older Americans, but seniors have seen the greatest rate of change since 2000. In 2000, 70 percent of the younger cohort used the internet. By 2015, 96 percent of young Americans used the internet. For those ages 65 and older, just 14 percent used the internet in 2000, increasing to 58 percent by 2015.81

This comparative analysis of the United States and three of Turkey’s middle- income peers demonstrates that the Turkish government is not an outlier in its efforts to bridge its own digital divides. Indeed, many of the same problems faced by these other countries also offer up possible solutions for Turkey to consider. To do so, however, will require the ruling AKP party to directly confront those aspects of Turkey’s digital divide that are unique to the nation, particularly censor- ship and the gender divide.

28 Center for American Progress | Turkey’s Digital Divides

Conclusion: Closing Turkey’s digital divide

The economic and sociopolitical stakes for Turkey to close its digital divide are potentially enormous. Expanding internet access to lower-income Turks and especially to women would connect these citizens to their government and to the wider economy—allowing for more efficient and frequent interactions with gov- ernment agencies and businesses while helping to develop the online literacy, con- nections, and markets that are essential to a dynamic modern economy. Enforcing the deregulation of the internet service provider marketplace is similarly critical to the country moving up the economic value chain in manufacturing and services and exports, a challenge at the heart of Turkey’s long-term economic plan. Better internet access for all Turks will improve access to jobs—through more efficient job searches and wider commercial opportunities—which would help to decrease high levels of unemployment, particularly in rural areas.

None of these developments will be easy to achieve without concerted govern- ment efforts. But the hurdles are not insurmountable. Extending e-government services to obviate the need for some in-person interactions with state bureau- cracy could entice some low-income, less tech-literate individuals to find value in internet usage. Extended relevant services could then be paired with more systematic training—beyond what nongovernmental organizations can pro- vide—to drive higher internet adoption. To enable these sorts of interactions among low-income citizens, the Turkish state will have to expand public access points to the internet.

Yet Turkey has quite a long way to go in developing its e-government offerings. In 2013, about one-third of users described government website technical failure issues, 15 percent noted unclear or outdated information, and more than 10 per- cent cited lack of support in using e-government services. Nearly half of all users experienced at least one problem.82 But these are solvable problems. Turkey’s Ministry of Development, for example, has included provisions for poor residents to receive free internet service in its 2018 strategy document.

29 Center for American Progress | Turkey’s Digital Divides

The Turkish government should also consider creative ways to diminish the effects of abiding social norms that hamper internet access for women. Public internet access points outside of the home could provide women from conserva- tive families with opportunities to search for useful content on the web, free from strict family supervision and control in the home. If the government sees the value in more egalitarian access, it could devise other strategies to drive adoption, in particular by increasing the affordability of mobile cellular internet access and by raising the number of active mobile broadband subscribers—both of which could be achieved by increasing competition in the telecommunications sector.

It is possible, though, that the government sees no serious reasons to be concerned about social, communal, and familial restrictions to women’s access. Indeed, the government might see social and political value in not intervening on such a charged issue—just one of several reasons why the current AKP government might not expend political capital to expand internet access: to maintain its control over information for large swathes of the population. Yet this would be shortsighted of the government, as is evident when looking at the ways in which other developing democracies have worked to close their own digital divides and reaped the benefits.

The ways in which Albania, Chile, and Brazil have dealt with their digital divides offer some constructive examples for Turkey to consider. First and foremost, none of these nations attempt to block citizens’ access to information on the internet. This is the most serious issue Turkey must tackle, particularly for its least advan- taged rural citizens and for women. Broadening competition in internet and telecommunications services in Turkey is a second major problem that—in many ways and to lesser degrees—is a problem shared with Albania, Chile, and Brazil, but there are specific lessons to be learned by examining Chile’s more successful efforts in privatization and less fettered regulations.

Is Turkey’s ruling party committed to overcoming these internet censorship and competitiveness hurdles? The evidence to date is not encouraging, yet the impor- tance of bridging the digital divide is recognized by the government, seen in its e-government services push and its e-education efforts in classrooms across the country. If Turkey takes further steps toward greater privatization and deregulation of the internet and communications industries—steps accompanied by less sweep- ing censorship rules and regulations—then the nation would have a better chance at a more competitive economy and a more effective and responsive government.

30 Center for American Progress | Turkey’s Digital Divides

About the author

Max Hoffman is the Associate Director for the National Security and International Policy team at the Center for American Progress, where his research focuses on Turkey and the Kurdish regions. He has organized and undertaken repeated research trips to Turkey and the wider region.

Acknowledgments

We are grateful to United Minds for Progress for their support and continuing efforts to expand educational opportunity and access to information. We also appreciate the input and expertise of Pinar Ersoy, Ed Paisley, Dov Friedman, and Jacob Greenberg.

31 Center for American Progress | Turkey’s Digital Divides

Endnotes

1 Michael Werz, Max Hoffman, Mark Bhaskar, “Previewing Turkey’s General Election,” Center for American Progress, (June 2, 2015), available at: https:// www.americanprogress.org/issues/security/re- port/2015/06/02/114206/previewing-turkeys-general- election/

2 See, for example: Organisation for Economic Co- Operation and Development (OECD), “OECD Digital Economy Outlook 2015,” (December, 2015), available at: http://www.oecd.org/sti/oecd-digital-economy- outlook-2015-9789264232440-en.htm

3 World Bank, “World Development Report 2016: Digital Dividends,” (Washington: World Bank, 2016), available at http://www.worldbank.org/en/publication/wdr2016.

4 Ibid.

5 World Bank, “World Development Report 2016: Digital Dividends,” (Washington: World Bank, 2016), available at http://www.worldbank.org/en/publication/wdr2016.

6 World Bank, “GDP per capita (current US$),” available at http://data.worldbank.org/indicator/NY.GDP.PCAP. CD/countries/US-TR?page=3&display=default (last accessed May 2016).

7 Kathryn Zickuhr and Aaron Smith, “Digital differences,” (Washington: Pew Research Center, 2012), available at http://www.pewinternet.org/2012/04/13/digital- differences/.

8 See, for example, Rabobank, “The Turkish 2000-01 banking crisis,” (September 4, 2013), available at: https://economics.rabobank.com/publications/2013/ september/the-turkish-2000-01-banking-crisis/

9 Ibid; Thom File, Computer and Internet Use in the United States: Population Characteristics (U.S. Bureau of the Census, 2013), available at https://www.census. gov/prod/2013pubs/p20-569.pdf; Turkish Statistical Institute, “Information and Communication Technology (ICT) Usage Survey on Households and Individuals,” 2004–2015, surveys for all years available at: http:// www.turkstat.gov.tr/PreHaberBultenleri.do?id=18660 (last accessed June 2016). Urban-rural estimation for the year 2014 was not provided. The study was con- ducted in the 16 to 74 age group.

10 Organisation for Economic Co-operation and Develop- ment, “Information and communication technology (ICT): Internet access,” available at https://data.oecd. org/ict/Internet-access.htm#indicator-chart (last ac- cessed March 2016).

11 Turkish Statistical Institute, “Information and Commu- nication Technology (ICT) Usage Survey on Households and Individuals,” 2004–2015, surveys for all years avail- able at: http://www.turkstat.gov.tr/PreHaberBultenleri. do?id=18660 (last accessed June 2016).

12 Dani Rodrik, “The Turkish Economy in Comparative Per- spective” (Boston: Harvard University, 2015), available at http://drodrik.scholar.harvard.edu/files/dani-rodrik/ files/turkish_economy_in_comparative_perspective. pdf.

13 Turkish Statistical Institute, “Information and Commu- nication Technology (ICT) Usage Survey on Households and Individuals, 2015,” (August 18, 2015), available at: http://www.turkstat.gov.tr/PreHaberBultenleri. do?id=18660 (last accessed June, 2016).

14 Ibid.

15 Ibid.

16 Turkish Statistical Institute, “Information and Commu- nication Technology (ICT) Usage Survey on Households and Individuals,” 2004–2015, surveys for all years avail- able at: http://www.turkstat.gov.tr/PreHaberBultenleri. do?id=18660 (last accessed June 2016).

17 Ali Acılar, “Exploring the Aspects of Digital Divide in a Developing Country,” Issues in Informing in Science and Information Technology 8 (2011), available at http://iisit. org/Vol8/IISITv8p231-244Acilar248.pdf.

18 Turkish Statistical Institute, “Information and Commu- nication Technology (ICT) Usage Survey on Households and Individuals, 2015,” (August 18, 2015), available at: http://www.turkstat.gov.tr/PreHaberBultenleri. do?id=18660 (last accessed June, 2016).

19 Interviews in Soganlik and Ugur Mumcu neighbor- hoods in Istanbul, August, 2015.

20 Turkish Statistical Institute, “Information and Commu- nication Technology (ICT) Usage Survey on Households and Individuals, 2015,” (August 18, 2015), available at: http://www.turkstat.gov.tr/PreHaberBultenleri. do?id=18660 (last accessed June, 2016).

21 Ibid.

22 Turkish Statistical Institute, “Information and Commu- nication Technology (ICT) Usage Survey on Households and Individuals,” 2004–2015, surveys for all years avail- able at: http://www.turkstat.gov.tr/PreHaberBultenleri. do?id=18660 (last accessed June 2016).

23 Mehmet Bilir, telephone interview, August 19, 2015.

24 Interview with Havva (last name withheld at authors’ discretion), Istanbul, August 2015.

25 Organisation for Economic Co-operation and Develop- ment, “Information and communication technology (ICT): Internet access,” available at https://data.oecd. org/ict/Internet-access.htm#indicator-chart (last ac- cessed March 2016).

26 Organisation for Economic Co-operation and Develop- ment, “OECD Digital Economy Outlook 2015” (2015), http://dx.doi.org/10.1787/9789264232440-en.

27 Osman Coşkunoğlu, interview in Istanbul, August 27, 2015.

28 Mehmet Bilir, telephone interview, August 19, 2015.

29 Prime Minister’s Communications Portal (Başbakanlık İletişim Merkezi – Sayilarla Bimer), Report of Applica- tions to Complaint System, available at: http://www. bimer.gov.tr/Forms/Docs/SayilarlaBimer.pdf (last accessed June, 2016).

30 The official Fatih Project website (in English and Turk- ish) are down for maintenance, but the outlines of the program are discussed here: Michael Trucano, “Observ- ing Turkey’s ambitious FATIH initiative to provide all students with tablets and connect all classrooms,” The World Bank, (December 18, 2013), available at: http:// blogs.worldbank.org/edutech/observing-turkeys- ambitious-fatih-initiative-provide-all-students-tablets- and-connect-all-classrooms

32 Center for American Progress | Turkey’s Digital Divides

31 Osman Coşkunoğlu, interview in Istanbul, August 27, 2015.

32 Ibid.

33 Information and Communication Technologies Author- ity, “Electronic Communications Market in Turkey” (2015), available at http://www.btk.gov.tr/File/?pa th=ROOT%2F1%2FDocuments%2FPages%2FMark et_Data%2F2015-Q1-En.pdf.

34 Osman Coşkunoğlu, interview in Istanbul, August 27, 2015.

35 Ibid.

36 Osman Coşkunoğlu, “Infrastructure and Independence: Why Turkey’s Telecommunications Sector Is Not Keep- ing Pace with Demand” (Washington: Freedom House, 2014), available at https://freedomhouse.org/report/ struggle-turkeys-internet/infrastructure-and-indepen- dence-why-turkeys-telecomms-sector-not-keeping- pace-demand.

37 Türk Telekom, “Fact Sheet: Financial & Operational Information,” (last accessed June, 2016), available at: http://www.ttinvestorrelations.com/financial- operational-information/fact-sheet.aspx.

38 Türk Telekom, “Board of Directors,” (last accessed June, 2016), available at: http://www.ttinvestorrelations.com/ corporate-governance/board-of-directors.aspx.

39 Nate Schenkkan, Osman Coşkunoğlu, Aslı Tunç, “The Struggle for Turkey’s Internet,” Freedom House, (August, 2014), available at: https://freedomhouse.org/sites/ default/files/The%20Struggle%20for%20Turkey’s%20 Internet.pdf

40 Osman Coşkunoğlu, interview in Istanbul, August 27, 2015.

41 Cihan, “Record salary for PM aide at Türk Telekom board,” (May 7, 2014), available at: https://www.cihan. com.tr/en/record-salary-for-pm-aide-at-turk-telekom- board-1425324.htm.

42 Turkish Statistical Institute, “Information and Commu- nication Technology (ICT) Usage Survey on Households and Individuals,” 2004–2015, surveys for all years avail- able at: http://www.turkstat.gov.tr/PreHaberBultenleri. do?id=18660 (last accessed June 2016).

43 OECD, “Broadband statistics: Average of broadband pricing offers for residential users in the OECD area, September 2011,” available at: http://www.oecd.org/ internet/broadband/oecdbroadbandportal.htm.

44 Alliance for Affordable Internet, “The 2015-16 Afford- ability Report,” (2016), available at http://a4ai.org/ affordability-report/report/2015/.

45 OECD, “Broadband statistics: List of offers of Broadband pricing for residential users in the OECD area, Septem- ber 2011,” (last accessed May 2016), available at: http:// www.oecd.org/internet/broadband/oecdbroadband- portal.htm.

46 Alliance for Affordable Internet, “The 2015-16 Afford- ability Report.” (2016), available at http://a4ai.org/ affordability-report/report/2015/.

47 Barın Kayaoğlu, “Turkey’s new 4G mobile network comes with many dark clouds,” Al Monitor, (April 14, 2016), available at: http://www.al-monitor.com/pulse/ originals/2016/04/turkey-4g-mobile-network-what- cost.html#ixzz4BD5rZvbO.

48 Mustafa Akgül and Melih Kırıldoğ, “Internet Censorship in Turkey,” Internet Policy Review 4 (2) (2015).

49 Ibid.

50 Bureau of Democracy, Human Rights and Labor, Turkey 2015 Human Rights Report, U.S. Depart- ment of State, (2016), available at http://www. state.gov/j/drl/rls/hrrpt/humanrightsreport/index. htm?year=2015&dlid=252909; Freedom House, “Freedom on the Net 2015: Turkey,” (2015), available at https://freedomhouse.org/report/freedom-net/2015/ turkey.

51 Engelliweb, available at https://engelliweb.com/ (last accessed May 1, 2016).

52 Bureau of Democracy, Human Rights and Labor, Turkey 2015 Human Rights Report; Freedom House, “Freedom on the Net 2015: Turkey.”

53 See, for example: Bureau of Democracy, Human Rights and Labor, Turkey 2015 Human Rights Report.

54 Ibid.

55 Freedom House, “Nations in Transit 2015: Russia” (2015), available at https://freedomhouse.org/report/nations- transit/2015/russia.

56 Osman Coşkunoğlu, interview in Istanbul, August 27, 2015.

57 See, for example: Max Hoffman and Michael Werz, “Freedom of the Press and Expression in Turkey,” Center for American Progress, (May 14, 2013), available at: https://www.americanprogress.org/issues/security/ report/2013/05/14/63159/freedom-of-the-press-and- expression-in-turkey/.

58 Richard Samans, Jennifer Blanke, Gemma Corrigan, Margareta Drzeniek Hanouz, “The Inclusive Growth and Development Report 2015,” (Geneva: World Economic Forum, 2015), available at http://reports.weforum.org/ inclusive-growth-report-2015/.

59 Ibid.

60 Regional Cooperation Council, “Interview with Genc Pollo, Minister for Innovation, Information and Com- munication Technology (ICT), Albania,” Our South East Europe Newsletter 15 (2011), available at http://www. rcc.int/interviews/0/42/interview-with-genc-pollo-min- ister-for-innovation-information-and-communication- technology-ict-albania.

61 Sofie Maddens-Toscano, “Strategies for the Promotion of Broadband Services and Infrastructure: A Case Study on Albania,” (Geneva: International Telecommunication Union, 2012), available at http://www.itu.int/ITU-D/ treg/publications/BBD_MDG_Albania_Final.pdf.

62 Endri Mataj, “Digital Albania,” US-Albania Investment Forum, (September 19, 2011), available at:

http://www.developingmarkets.com/sites/default/files/ session-3-endri-mataj.pdf.

63 Ibid.

64 Maddens-Toscano, “Strategies for the Promotion of Broadband Services and Infrastructure.”

65 Bureau of Democracy, Human Rights and Labor, Albania 2015 Human Rights Report, U.S. Depart- ment of State, (2015), available at http://www. state.gov/j/drl/rls/hrrpt/humanrightsreport/index. htm?year=2015&dlid=252815.

33 Center for American Progress | Turkey’s Digital Divides

66 Ibid.

67 World Bank, “World Development Indicators: Chile,” available at http://databank.worldbank.org/data/ reports.aspx?source=2&country=CHL&series=&period (last accessed April 2016).

68 Juan M. Gallego and Luis H. Gutierrez, “ICTs in Latin America and the Caribbean: Stylized Facts, Programs and Policies,” Inter-American Development Bank, (Buenos Aires: Diálogo Regional sobre Sociedad de la Información July, 2015), available at: http://dirsi.net/ web/files/files/Gallego%20y%20Guti%C3%A9rrez%20 KSF_ICT_DIRSI_Discussion%20Paper_10%20JUL%20 %2015.pdf.

69 Ibid.

70 Pablo T. Spiller and Carlo G. Cardilli, “The Frontier of Telecommunications Deregulation: Small Countries Leading the Pack,” The Journal of Economic Perspectives 11 (4) (1997): 127–138, available at http://www.jstor. org/stable/2138467.

71 See, for example, El Mundo, “Chile publica su ley que ga- rantiza la neutralidad de la Red (Chile announces its law guaranteeing net neutrality),” August 27, 2010, avail- able at http://www.elmundo.es/elmundo/2010/08/27/ navegante/1282907501.html; Carin Zissis and Rachel Glickhouse, “Net Neutrality Lessons from Latin America,” U.S. News & World Report, (May 9, 2014), available at http://www.usnews.com/opinion/blogs/ world-report/2014/05/09/the-fcc-can-learn-some-net- neutrality-lessons-from-latin-america.

72 Bureau of Democracy, Human Rights and Labor, Chile 2015 Human Rights Report, U.S. Department of State, (2015), available at http://www.state. gov/j/drl/rls/hrrpt/humanrightsreport/index. htm?year=2015&dlid=252999.

73 Juan M. Gallego and Luis H. Gutierrez, “ICTs in Latin America and the Caribbean: Stylized Facts, Programs and Policies,” Inter-American Development Bank, (Buenos Aires: Diálogo Regional sobre Sociedad de la Información July, 2015), available at: http://dirsi.net/ web/files/files/Gallego%20y%20Guti%C3%A9rrez%20 KSF_ICT_DIRSI_Discussion%20Paper_10%20JUL%20 %2015.pdf.

74 World Bank, “Internet users (per 100 people): Brazil,” available at http://data.worldbank.org/indicator/IT.NET. USER.P2/countries/1W-BR?display=graph (last accessed April 2016).

75 PricewaterhouseCoopers, Communications Review: Exploring telecom markets in Latin America 16 (2) (2011), available at https://www.pwc.com/ve/es/publicacio- nes/assets/communications-review-latin-america- vol16-no2.pdf.

76 See, for example, Ronaldo Lemos, “Brazil’s Internet Law, the Marco Civil, One Year Later,” Council on Foreign Relations, (June 1, 2015), available at http://blogs.cfr. org/cyber/2015/06/01/brazils-internet-law-the-marco- civil-one-year-later/.

77 International Telecommunication Union, “Measuring the Information Society,” (2013), available at http:// www.itu.int/en/ITU-D/Statistics/Documents/publica- tions/mis2013/MIS2013_without_Annex_4.pdf.

78 See, for example: Federal Communications Commis- sion, “Recovery Act Broadband Initiatives,” (accessed May, 2016), available at: https://www.fcc.gov/general/ recovery-act-broadband-initiatives. BusinessUSA. gov, “Broadband Technology Opportunities Program,” (accessed May, 2016), available at: http://business.usa. gov/program/broadband-technology-opportunities- program-btop.

79 Andrew Perrin and Maeve Duggan, “Americans’ Internet Access: 2000-2015,” (Washington: Pew Research Center, 2015), available at http://www.pewinternet. org/2015/06/26/americans-internet-access-2000-2015/.

80 Andrew Perrin and Maeve Duggan, “Americans’ Internet Access: 2000-2015,” (Washington: Pew Research Center, 2015), available at http://www.pewinternet. org/2015/06/26/americans-internet-access-2000-2015/.

81 Ibid.

82 Organisation for Economic Co-operation and Develop- ment, “OECD Digital Economy Outlook 2015.”

1333 H STREE T, NW, 10TH FLOOR, WASHINGTON, DC 20005 • TEL: 202-682-1611 • FAX: 202-682-1867 • W W W.AMERICANPROGRESS.ORG

Our Mission

The Center for American Progress is an independent, nonpartisan policy institute that is dedicated to improving the lives of all Americans, through bold, progressive ideas, as well as strong leadership and concerted action. Our aim is not just to change the conversation, but to change the country.

Our Values

As progressives, we believe America should be a land of boundless opportunity, where people can climb the ladder of economic mobility. We believe we owe it to future generations to protect the planet and promote peace and shared global prosperity.

And we believe an effective government can earn the trust of the American people, champion the common good over narrow self-interest, and harness the strength of our diversity.

Our Approach

We develop new policy ideas, challenge the media to cover the issues that truly matter, and shape the national debate. With policy teams in major issue areas, American Progress can think creatively at the cross-section of traditional boundaries to develop ideas for policymakers that lead to real change. By employing an extensive communications and outreach effort that we adapt to a rapidly changing media landscape, we move our ideas aggressively in the national policy debate.

Infrastructure_and_Independence_-_Why_Turkeys_Telecomms_Sector_is_Not_Keeping_Pace_with_Demand.pdf

As other sections of this report highlight, Internet use in Turkey and the role of the Internet in the country’s economy have grown significantly in recent years. Yet the development of Turkey’s Internet infrastruc- ture—especially next-generation networks requiring broadband that are necessary for the new phase of In- ternet development—has not kept pace. This section explains why the regulatory apparatus in Turkey has failed to stimulate infrastructure development despite meaningful reforms.

INFRASTRUCTURE The last two decades have seen significant changes for Turkey’s telecommunications sector, which falls under the jurisdiction of the Ministry of Transporta- tion.13 In 1994, Law No. 4000 created a state-owned, fixed-line operator company, Türk Telekom (to be privatized in 2005), and granted it exclusive rights for all telecommunications infrastructure, with the excep- tion of mobile service.

Also in 1994, two private companies, Turkcell and Telsim, launched mobile telephony services and were granted licenses in 1998. Later, Avea, a subsidiary of Türk Telekom, became the third mobile operator.

TTNet, a subsidiary of Türk Telekom, launched broad- band in the early 2000s, providing ADSL services over their public switched telephone network (PSTN). Mobile broadband over 3G started towards the end of

Infrastructure and Independence: Why Turkey’s Telecommunications Sector Is Not Keeping Pace with Demand Osman Coşkunoğlu

the decade, with three GSM operators: Turkcell, Voda- fone (which had bought Telsim), and Avea. Likewise, broadband services started over fiber and cable-TV networks, as well as through satellites, at the end of the 2000s.

The penetration rate of fixed-line telephony has been decreasing steadily over the last decade as consum- ers switch to mobile service. With more than 13 million subscribers, the current penetration rate of fixed-line telephony is 17.34 percent. Considering that the average household size in Turkey is 3.69 persons, it can be assumed that the fixed-line market is satu- rated. Around 86 percent of fixed-line users subscribe to the incumbent Türk Telekom, and the remaining 14 percent to alternative operators.

After significant growth in mobile telephony over the last decade, there are more than 70 million subscrib- ers (around 70 percent of which have 3G), for a pen- etration rate of 91.5 percent. If the 0-to-9 age group is excluded, the penetration rate of mobile telephony exceeds 100 percent. According to subscriber num- bers, Turkcell, Vodafone, and Avea’s market shares are 49.62 percent, 28.57 percent, and 21.87 percent, respectively. According to revenues, these market shares are, respectively, 46.45 percent, 33.38 percent, and 20.17 percent.

With 8.5 million fixed and 26.4 million mobile sub- scribers, total broadband subscription in Turkey has reached 35 million (broadband subscription through satellite is negligible). Fixed broadband subscription is as follows: 79 percent xDSL (67 percent of which is owned by Türk Telekom subsidiary TTNet and 12 percent by other ISPs), 15 percent fiber, and 6 percent cable. Currently, Türk Telekom owns 182,450 km of fiber infrastructure, around 123,000 km as backbone

In 1994, Türk Telekom was created as a state- owned, fixed-line operator company with exclusive rights for all telecommunications infrastructure except for mobile service.

SPECIAL REPORT The Struggle for Turkey’s Internet

8

network and around 60,000 km that reaches sub- scribers. Alternative operators own a total of 51,244 km fiber infrastructure reaching subscribers.14 In other words, nine years after its privatization and the opening of the sector to competition, Türk Telekom remains the overwhelmingly dominant provider in fixed-line and broadband services.

Scoreboard Turkey lags behind all other OECD countries in fixed broadband penetration with 11 percent, less than half of the OECD average of 26.7 percent. In mobile broadband, Turkey is near the bottom but ahead of three member countries (Chile, Hungary, and Mexico), with 26.44 million subscribers corresponding to 34.5 percent penetration, which is still only about half the OECD average.15

Two annual reports provide global comparisons. The Global Information Technology Report of the World Economic Forum ranks countries according to the Network Readiness Index (NRI). In the report pub- lished in 2014,16 Turkey ranks 51st out of 148 coun- tries on the NRI. One sub-index in the rating is “Readi- ness,” where Turkey ranks 48th. Under this sub-index, Turkey ranks 42nd on the pillar of “Infrastructure and Content.” Turkey ranks 44th on the “Environment” sub-index, but 55th on the “Political and Regulatory Environment” pillar within it.

The International Telecommunication Union’s Measur- ing the Information Society ranks countries accord- ing to the ICT Development Index (IDI). In the latest report published in 2013,17 Turkey ranks 69th out of 157 countries according to the IDI. On the sub-index “Access” that captures ICT readiness and includes five infrastructure and access indicators (fixed-telephone subscriptions, mobile-cellular telephone subscrip- tions, international Internet bandwidth per Internet user, percentage of households with a computer, and percentage of households with Internet access), Turkey ranks 72nd.

REGULATORY MECHANISMS Why have infrastructure development and broad- band penetration lagged behind expectations? Since 1998, Turkey has increasingly adopted market-based mechanisms for telecommunications and Internet infrastructure development. As such, the two most important functions of the regulatory agency are to achieve liberalization of the sector and to ensure fair competition while protecting consumers. Yet effective

liberalization, particularly in the broadband sector, has not been achieved. A review of the legislative changes shows an improvement in the regulatory framework that is not matched by an improvement in implemen- tation of liberalization.

Background (1998–2008) The first legislative step towards the liberalization of Turkey’s telecommunications sector was the enact- ment of Law No. 4502 in 2000. With this law, the Ministry of Transportation established its regulatory oversight of the communications sector through the Telecommunications Authority (Telekomünikasyon Kurumu, TK), with powers to regulate prices, intercon- nection, and access.

Even before this law, however, liberalization in telecommunications had started when two mobile operators, Turkcell and Telsim, were granted licenses in 1998, rendering them independent of Türk Tele- kom’s monopoly. This launched genuine competition in the mobile telephony segment of the sector. In 2003, Avea, a subsidiary of Türk Telekom, entered the mobile market, and in 2005 Vodafone bought Telsim. Although the mobile segment had liberalized, Turkcell dominated the market due to the first-mover advan- tage until the TK took two measures in 2007: institut- ing retail and wholesale tariff controls and mobile number portability. Fair competition in the mobile market was also reinforced by later reducing mobile call termination rates.

Liberalization in fixed-line telephony started only with the adoption of Law No. 4502. According to the law, the monopoly rights of Türk Telekom would be termi- nated at the end of 2003. However, by 2008, the share of Türk Telekom in overall telephony services was still 91 percent by traffic and 81 percent by revenue. The degree of competition in fixed broadband services was even worse. By 2008, there were more than 70 ISPs in the market, yet the share of Türk Telekom in

Nine years after its privatization and the opening of the sector to competition, Türk Telekom remains the overwhelmingly dominant provider in fixed-line and broadband services.

Türk Telekom 4.3M (95%)

Türk Telekom 86%

Türk Telekom 60,000 KM (54%)

Other ISPs .2M (5%)

Other Operators 14%

Other ISPs 51,244 KM (46%)

FIXED-LINE USER SUBSCRIPTIONS

FIXED RETAIL BROADBAND CONNECTIONS (BY 2008)

FIBER INFRASTRUCTURE REACHING SUBSCRIBERS

www.freedomhouse.org

Freedom House

9

fixed retail broadband connections was about 95 per- cent (4.3 million out of a total of 4.5 million).

Despite some shortcomings in Law No. 4502, it was a good start. However, implementation of the regula- tory framework and the resulting market outcomes were less than satisfactory. The implementation was uneven and conspicuously favored Türk Telekom. Although the Ministry and the TK vigorously enforced competition in the mobile telephony segment, they were ineffective in the fixed telephony and broadband segments, where Türk Telekom continued to dominate.

As the European Commission noted in 2008 in its progress report on Turkey’s accession, “Liberalisation of local telephony is still pending and undermines competition in the fixed and broadband markets. The regulatory body is well staffed and is self-financed. However, it lacks independence—notably in the authorisation process—and the decision making process of the Telecommunications Authority is not transparent.”18

The European Competitive Telecommunications As- sociation (ECTA) also noted the Telecommunications Authority’s uneven and ineffective implementation. The ECTA’s Regulatory Scorecard report contains an overall score, based on qualitative and quantita- tive analyses, for the effectiveness of the regulatory environment in each of the EU member and some candidate countries. In ECTA’s 2008 report, Turkey scored the worst of all countries surveyed.19 Turkey scored particularly poorly in the “Entry Enabler” and

“Economic Market Conditions” dimensions. Even though Turkey’s score was best in the “Institutional Framework” dimension, within that dimension it performed worst in the “National Regulatory Agency (NRA) Independence” and “NRA Enforcement Powers” sub-dimensions.

It is also revealing to note that, when the Telecom- munications Authority persistently remained ineffec- tive in countering the dominance of Türk Telekom in fixed-line telephony and broadband segments, the Competition Authority—which had had an impec- cable record of acting independently at that point in time—had to interfere. In a landmark decision in November 2008, the Competition Authority imposed a fine of 12.4 million TL (about $9 million) on Türk Telekom for abusing its dominance in the wholesale broadband Internet market.20 That it required another authority’s interference to enforce competition speaks

to concerns that the Telecommunications Author- ity was insufficiently independent in encouraging competition.

Electronic Communications Law (2008) The Electronic Communications Law (Law No. 5809) of November 2008 was a landmark piece of legislation. Although not perfect, the ECL constituted a signifi- cant improvement over Law No. 4502. Article 5 of the ECL redefined the Ministry of Transportation’s role and duties in the communications sector:21 determining strategy and policy for electronic communications services that use scarce resources (such as num- bering, satellite positions, and frequencies). It also authorized the Ministry to determine the principles and policies for the promotion of competition in the electronic communications industry, and to take supportive measures. This is in line with the general approach adopted in the liberalization of network industries, whereby the Ministry takes on responsibil- ity for general formulation of policy and strategy for the industry, and an independent regulatory authority is given responsibility for the formulation and imple- mentation of necessary regulations. As we shall see, however, the question of independence of the regula- tory authority has been a constant issue for effective implementation.

The ECL empowered the regulatory agency and changed its name to the Information and Communi- cation Technologies Authority (Bilgi Teknolojileri ve İletişim Kurumu, or BTK).

More specifically, Article 6 authorizes the BTK to do the following:22

• To undertake regulations so as to install and protect competition and to prevent activities that hamper or distort competition, and to impose remedies on operators with significant market power (SMP) and on other operators if necessary;

• To undertake market analysis and determine opera- tors with SMP;

• To maintain Board decisions, along with justifica- tions and procedures, on matters of interest to operators and consumers open to public;

• To approve, as necessary, tariffs and reference ac- cess offers.

Two virtues of the law deserve to be emphasized. First, there is an explicit reference to market analysis to determine operators with SMP (even if the market

TURKEY'S STATS

FIXED-BROADBAND PENETRATION

TURKEY 11% OECD AVERAGE 26.7%

MOBILE-BROADBAND PENETRATION

TURKEY 34.5% OECD AVERAGE 68.4%

SPECIAL REPORT The Struggle for Turkey’s Internet

10

analysis procedures lack a firm basis in the law).23 Second, the requirement that the BTK has to publish its decisions with justifications is of particular impor- tance for transparency and accountability.24

Following the adoption of the ECL, the BTK released or renewed the necessary secondary legislation, including ordinances on the following: Authoriza- tions, Rights of Way, Access and Interconnection, Determination of Operators with SMP and Obligations that can be Imposed on Them, Tariffs, and Number Portability.

The European Commission’s views on the legisla- tion were mildly positive:25 “[ECL] is a significant step towards aligning Turkey’s regulatory framework with the EU acquis, notably as regards the authorisation rules and the tasks of the regulator… The amendments to the law have the potential to create the conditions for competition on the fixed telephony market.” The EC also expressed some reservations:26 “However, the Electronic Communications Law is not in line with the acquis, in particular the provisions on universal service obligations, and as regards the scope of authorisation rules, which does not apply to existing concession agreements.”

Although the legal gaps between the EU and the Turkish regulatory frameworks were not eliminated, they were narrowed down. The foremost implication of this is that the lack of progress in the development of competition was then due to weak implementation and enforcement rather than the lack of legal instruments.

Implementation (2008–Present) As shown in the “Infrastructure” section above, despite regulatory liberalization, the incumbent Türk Telekom’s domination in the fixed-line telephony and broadband segments has continued, and the progress in competition as well as infrastructure development remains slow.

Some limited but favorable developments have oc-

curred in the last six years. The following is compiled from the European Commission’s annual Turkey Prog- ress Report from 2010–13:

• The BTK is publishing three-year strategic plans (so far, 2010–12 and 2013–16) and quarterly market reports. It also publishes justifications of decisions taken, which is a positive development for the transparency and accountability of its activities.

• A communiqué was published in February 2012 to improve the service quality of Internet service providers. However, the quality of ISPs has not improved significantly.

• Fixed number portability and wholesale line rental became operational in early 2012.

• Mobile termination rates (MTR) were further reduced. Fixed termination rates, however, remain higher than the EU average, and calls of interna- tional origin are exempted from the MTR regulation, resulting in distortion of the market.

In addition to the foregoing, in 2010 Türk Telekom was finally required to provide naked DSL services. Provision of naked DSL means that Türk Telekom can no longer bundle voice and data services together and that subscribers can subscribe to DSL services alone, without having to pay for voice services as well. Despite pressure from public and alternative opera- tors, for years the BTK had not asked Türk Telekom to provide naked DSL, which would mean revenue loss for the incumbent and would increase the com- petitiveness of alternative ISPs. The unavailability of naked DSL was slowing down consumer switching from fixed to mobile telephony and retarding demand for VoIP services.

Yet again it was only through a Competition Authority ruling that Türk Telekom was forced to provide naked DSL services.27 What should have been the BTK’s role as the independent agency regulating telecommuni- cations infrastructure development was again fulfilled by the Competition Authority.

Around the same time that the ECL was enacted in Turkey, a new technological concept had emerged in telecommunications at the global level: New (or Next) Generation Networks (NGN).28 Being packet- based, NGN uses Internet Protocol (IP) and enables the deployment of access-independent services over converged fixed and mobile networks. As such, broad- band plays the core role in NGNs and has become the most crucial technology in telecommunications.

The lack of progress in the development of competition is due to weak implementation and enforcement rather than the lack of legal instruments.

www.freedomhouse.org

Freedom House

11

Faced with this new challenge, in an October 2011 decision the BTK excluded fiber from market analysis for the next five years or until the percentage of fiber- based subscriptions reaches 25 percent of all fixed broadband subscriptions. Türk Telekom’s reference offer, as approved by the BTK, includes the terms and conditions of facility-sharing, and Türk Telekom com- mitted to provide wholesale and bitstream services over its fiber network on a nondiscriminatory basis.29 The objective of exclusion was to incentivize Türk Telekom to invest in fiber development that ISPs could then use through facility-sharing agreements.

Excluding fiber from market analysis has not served its purposes, however. Türk Telekom’s investments in broadband in general and in fiber in particular have not increased in recent years. In fact, according to an analysis by the Competitive Telco Operators Associa- tion (TELKODER)—members of which are all the tele- communications and Internet firms in Turkey, except for Türk Telekom—the investment/revenue share of Türk Telekom between 2010 and 2013 is 17.2 percent, while the combined alternative operator revenue is 38.8 percent.30

There are a number of other implementation problems regarding the growth of broadband in general and fiber infrastructure in particular. The most important is that alternative operators face difficulties in acquir- ing rights-of-way to deploy mobile/wireless and fiber networks. Under an executive order from 2011, the Ministry is supposed to encourage Türk Telekom to reach an agreement with alternative providers for right-of-way concessions. Yet, due to lack of respon- siveness on the part of Türk Telekom, in the last eight months alternative operators have been unable to install a single yard of fiber.31

CONCLUSION Turkey has great potential for growth in telecom- munications and the Internet sector given its young population and high technology adoption rates. The BTK has the legal instruments, financial resources, and institutional capabilities to encourage this growth. However, this potential has not sufficiently translated into the expansion of the sector. In fact, according to the BTK’s own annual figures, in dollar terms there has

been no significant growth in the telecommunications and Internet sector from 2008 to 2013.32

With the exception of the incumbent Türk Telekom, there is consensus in the sector that the root cause of this insufficient growth is the lack of progress in fostering competition. The underlying reasons for this lack of competition are the BTK’s scant degree of inde- pendence from the Ministry and the continued close ties of Türk Telekom with the current administration.

In regard to the BTK, the Ministry of Transportation remains responsible for overall government telecom- munications policy, proposes names for BTK board members to the cabinet, and exercises approval power over changes in license fees proposed by the BTK. It also oversees Türksat, the government opera- tor of Turkey’s three communications satellites and the holder of Türk Telekom’s former cable company assets. The Cabinet of Ministers appoints BTK board members without any consultation with the sector.

The prime minister’s attitude towards independent regulatory agencies was shown early in his premier- ship in 2004, when he expressed frustration that “We give instructions to the boards of regulatory agencies, they say ‘yes, sir’ but then they don’t do what we say.”33 After 10 years of one party in power, fewer and fewer people are not doing what they’re told. The last 10 years have demonstrated the consequences of this approach, with decreasing independence evident among regulatory agencies not only in telecommu- nications but in other areas as well, such as media regulation and public procurement.

As for Türk Telekom, despite privatization in 2005, the company remains very close to the government. The Turkish Treasury owns 30 percent of the company, and the Undersecretary of the Ministry of Transportation is the vice-chair of the board. In March 2014, the chief advisor to Prime Minister Erdoğan (a former newspa- per columnist with no background in telecommunica- tions) was appointed to the board as an independent member.34 These close relationships ensure that Türk Telekom continues to receive preferential treatment regardless of the legal obligations that should prevent favoritism.

In order for Turkey to fulfill its potential for growth Internet infrastructure and telecommunications ser- vices, the first and foremost step must be to eliminate political intrusion into the sector.

In the last eight months alternative operators have been unable to install a single yard of fiber.

SPECIAL REPORT The Struggle for Turkey’s Internet

12

InfodevDocuments_1132.pdf

i

ii

iii

© 2011 The International Bank for Reconstruction and Development / The World Bank

1818 H Street NW

Washington DC 20433

Telephone: 202-473-1000

Internet: www.worldbank.org

E-mail: [email protected]

All rights reserved

The findings, interpretations and conclusions expressed herein are entirely those of the author(s) and do not

necessarily reflect the view of infoDev, the Donors of infoDev, the International Bank for Reconstruction and

Development/The World Bank and its affiliated organizations, the Board of Executive Directors of the World Bank or

the governments they represent. The World Bank cannot guarantee the accuracy of the data included in this work.

The boundaries, colors, denominations, and other information shown on any map in this work do not imply on the

part of the World Bank any judgment of the legal status of any territory or the endorsement or acceptance of such

boundaries.

Rights and Permissions

The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without

permission may be a violation of applicable law. The International Bank for Reconstruction and Development/The

World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work

promptly.

For permission to photocopy or reprint any part of this work, please send a request with complete information to

infoDev Communications & Publications Department; 2121 Pennsylvania Avenue, NW; Mailstop F 5P-503, Washington,

D.C. 20433, USA; telephone 202-458-4070; Internet: www.infodev.org; E-mail: [email protected].

All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher,

The World Bank, 1818 H Street NW, Washington, D.C. 20433, USA; fax: 202-522-2422; e-mail: [email protected].

Cover design by infoDev

To cite this publication: Telli, Cagatay, 2011. Broadband in Turkey: Compare To What? Washington, D.C: infoDev / World Bank.

Available at http://www.broadband-toolkit.org.

iv

Table of Contents

Executive Summary……………………………………………………………………………… 7

1. Turkish ICT Sector: A General Assessment…………………………………………………… 8

1.1 Background……………………………………………………………………………… 8

1.2 ICT Sector………………………………………………………………………………. 10

2. Turkey’s Broadband Ecosystem: A General Assessment……………………………………… 13

2.1 Synopsis of Near History: Evolution of Markets and Public Reform……………………… 13

2.2 Institution Setup: Current Policy, Legal and Regulatory Framework……………………… 14

2.3 Infrastructure Development……………………………………………………………… 17

2.3.1 Broadband penetration……………………………………………………………. 18

2.3.2 Broadband technologies…………………………………………………………… 21

2.3.2.1 Backbone network…………………………………………………………… 21

2.3.2.2 Local access technologies…………………………………………………… 21

2.3.3 Prices……………………………………………………………………………… 26

3. Applications…………………………………………………………………………………… 30

3.1 Government……………………………………………………………………………… 30

3.2 Electronic Commerce…………………………………………………………………… 34

3.3 Other Applications: The Rise of Social Media…………………………………………… 36

4. Usage and Diffusion…………………………………………………………………………… 39

4.1 People…………………………………………………………………………………… 39

4.2 Business…………………………………………………………………………………… 41

5. Conclusions…………………………………………………………………………………… 44

5.1 The Likely Impacts of High Bandwidth Networks on Turkish Economy………………… 44

5.2 Factors Contributing to Turkey’s Broadband Development……………………………… 45

5.2.1 Lessons learned……………………………………………………………………. 45

6. References……………………………………………………………………………………… 48

6.1 Online Resources………………………………………………………………………. 49

v

List of Figures, Tables and Boxes

Figure 1-1: Geographic position of Turkey……………………………………………………… 8

Figure 1-2: Global Competitiveness Index, Turkey, 2010………………………………………… 8

Figure 1-3: Fastest growing economies in ICT spending, 2003-2009……………………………… 10

Figure 1-4: The composition of ICT spending in Turkish economy, 2009……………………… 11

Figure 1-5: Evolution of ICT foreign trade in Turkish economy, US$ million, 2004-2009……… 12

Figure 2-1: Fixed broadband penetration gap with selected countries, 2010……………………… 18

Figure 2-2: Fixed broadband (per 100 inhabitants), net increase, June 2009-2010………………… 19

Figure 2-3: Selected penetration indicators, percent……………………………………………… 19

Figure 2-4: Technological configuration of broadband infrastructure as percent of shares, 2010… 22

Figure 2-5: Breakdown of fixed broadband subscribers by access speeds, 2010………………… 22

Figure 2-6: Shares of technology in total broadband subscriptions, percent, 2010………………… 25

Figure 2-7: Percentage of fibre connections in total broadband, June 2010……………………… 25

Figure 2-8: Broadband average monthly subscription price, Oct 2009, USD PPP………………… 27

Figure 3-1: Number of public sector owned websites…………………………………………… 31

Figure 3-2: The maturity level of selected public services, percent, 2007………………………… 31

Figure 3-3: Ownership of websites, by employment size of enterprises, 2009…………………… 34

Figure 3-4: E-commerce activities of enterprises by sector of activity, 2007……………………… 34

Figure 3-5: The benefits reported through e-commerce, 2008…………………………………… 35

Figure 3-6: The Turkish social media landscape, 2010…………………………………………… 38

Figure 4-1: Household broadband penetration, as percent……………………………………… 39

Figure 4-2: Internet usage by place of connection, male and female, 2009……………………… 40

Figure 4-3: Broadband usage indicators by enterprises, as percent……………………………… 42

Figure 4-4: Internet usage in enterprises by purpose, 2007-2009………………………………… 42

Figure 5-1: Baseline and broadband enhanced growth scenarios, real GDP 2009=100…………… 45

Figure 5-2: Turkey connectivity performance by scorecard component, 2010…………………… 46

Table 2-1: Major institutions and establishments in ICT sector………………………………… 15

Table 2-2: Broadband indicators, Turkey, 2010………………………………………………… 20

Table 2-3: Subscription indicators by technology……………………………………………… 21

Table 2-4: Prices of selected broadband service packages in Turkey, March 2011………………… 28

Table 3-1: Major e-government applications in Turkey…………………………………………… 32

Table 3-2: Top cities on Facebook……………………………………………………………… 37

Table 4-1: Internet activities of individuals who have accessed the Internet in the last 3 months… 41

Box 2-1: Recent strategies and initiatives related to e-government and broadband ecosystem…… 17

Box 2-2: Leveling the field: The battle of broadband in national market………………………… 23

vi

Acknowledgments

The author of this report is Çagatay Telli from the State Planning Organization (SPO) of Turkey, with guidance from Michael Minges, who directed the case studies for the Broadband Strategies Toolkit. This case study was commissioned by Tim Kelly (infoDev) and Carlo Rossotto (ICT Sector Unit of the World Bank Group). Thanks are due to Samhir Vasdev for editing and preparing the document for publication.

A number of people contributed to this study by participating in expert meetings and interviews and providing up to date information on the national broadband ecosystem, its relative strengths and challenges faced for future growth. Most notably Ertugrul Karacuha (Vice Chairman, Information and Communication Technologies Authority), Ahmet Hasanbeseoglu (Regional Director, Cisco), Ugur Terzioglu (Manager, Cisco), Ahmet Kaplan (Vice President, TURKSAT), Pasa Yasar (President, ENVITEL), Ramazan Altınok (Head, E-Government Group, Prime Minister’s Office), Emin Sadık Aydın (Head of Information Society Division of SPO), and Furkan Civelek (Planning Expert, Information Society Division of SPO) commented on the draft report.

This case study is one of an initial series of seven that will contribute to the Broadband Strategies Toolkit, an online resource for policy-makers and regulators, especially in developing countries (see www.broadband-toolkit.org). The Toolkit and its case studies are generously funded by the Korean Trust Fund (KTF) on Information and Communications for Development (IC4D). The KTF is a partnership between the government of the Republic of Korea and the World Bank Group whose purpose is to advance the ICT4D agenda to contribute to growth and reduce poverty in developing countries. The KTF, as well as Philippe Dongier, Sector Manager (TWICT) and Valerie D‘Costa, Program Manager (infoDev), are owed the author‘s gratitude for their support and guidance on this project.

7

Executive Summary

Turkey has throughout history been prominent as a center of commerce because of its land connections to the continents of Europe, Asia and Africa and the sea surrounding it on three sides An Organization for Economic Cooperation OECD member, it has been awaiting European Union (EU) membership since 1987. As an upper-middle-income economy, Turkey suffers from comparison with these mainly high-income groupings. Its fixed broadband penetration stood at 9.4 subscriptions per 100 inhabitants in June 2010 compared to the OECD average of 24.2 and 34% of Turkish homes had a broadband connection compared to the EU average of 61% in 2010.

These statistics disguise the fact that, compared to other countries in its income group, Turkey is doing relatively well. E-government initiatives have been a major driving force for development of the broadband ecosystem. This has triggered demand by enterprises in the ICT sector and motivated citizens to increase Internet usage. Ensuring a shared vision among political leaders and technocrats has also been an important factor in pushing e-government programs. Political leaders saw e-government as a central instrument that would support public reforms and larger changes in the political system. A central organizational structure was formulated to develop strategies and put public money into the pipeline for a set of strategically important projects with high value and high transaction.

The high tempo growth of Turkish economy in the last decade is another supportive factor. The various market-oriented reforms have been implemented complemented with a proactive

foreign policy resulting in large sums of overseas capital flowing into the country.

Communications, software and hardware segments of ICT industries have expanded rapidly. This has included significant investment in upgrading mobile networks to broadband. Broadband mobile networks were only launched in 2009, yet by the end of 2010 around a quarter of the population were capable of accessing high-speed wireless services.

The Turkish population has also reacted to social networking in a major way. The country is the fourth largest Facebook market in the world. Turkey’s own social media content is growing and Turkish web sites are becoming more popular and increasingly diversified.

Nevertheless the country continues to face economic and social barriers to effectively absorb broadband technologies on a large scale and better utilize them for leveraging overall economic competitiveness. Fixed broadband competition is limited and dominated by ADSL technology. ICT skill gaps among small and medium enterprises and the less educated need to be adequately addressed with participation of private initiatives. The lack of a suitable national accounting framework for more detailed analysis hinders international benchmarking in ICTs and innovation.

If Turkey can overcome these barriers, the results could be considerable. According to the National Broadband Vision study, broadband could boost economic growth by 0.8-1.7 percentage points per year. This economic momentum enabled by an enhanced broadband ecosystem would create 180,000-380,000 new jobs each year.

8

1 Turkish ICT Sector: A General Assessment

1.1 Background

Geographically, the Republic of Turkey is located at a point connecting the continents of Europe, Asia and Africa. Because of its geographical location the mainland of Anatolia possesses an extraordinary geo-strategic value. In the past, it was an important expansion post to many great political powers including Hatti, Hittite, Ancient Greek, Urartu, Phrygia, Ionia, Persia, Roman, Byzantine, Arab, Seljuk and Ottoman. It has also been prominent as a center of commerce because of its land connections to three continents and the sea surrounding it on three sides (Figure 1-1).

Modern Turkey was founded in 1923 by Mustafa Kemal following the collapse of the 600-year-old Ottoman Empire. After a period of one-party rule, since the 1950 election, Turkey has experienced a democratic political system based on multi-party politics. Turkey became a European Union (EU) candidate country in 1999 and, in line with requirements, went on to

introduce substantial political and economic reforms. After intense bargaining, EU membership negotiations were launched in October 2005.

Turkey has two levels of government and several types of administrations: National administration

Figure 1-1: Geographic position of Turkey

Figure 1-2: Global Competitiveness Index, Turkey, 2010

(Source: Global Competitiveness Index, 2010)

9

with ministries and agencies at the central government level and field offices in provinces and districts. Local governments include 3,225 municipalities in areas with dense populations and 81 special provincial administrations with jurisdiction beyond municipal boundaries. In the last few years Turkey developed another organizational structure to initiate a more diversified and flexible decision making system to support endogenous growth of its economic regions by introducing Development Agencies. Currently 26 Development Agencies have become fully operational.

With an increasingly urban population of around 75 million1 and a GDP per capita of around US$10,000 in purchasing power parity, the Turkish economy is largely developed. Benefiting from both a capable and young labor force2 and diverse natural resources the country is among the world's leading producers of agricultural products; textiles; motor vehicles, ships and other transportation equipment; construction materials; consumer electronics and home appliances. Turkey has a rapidly growing private sector not only in industrial activities but also in competitive services such as banking, transport, tourism and communications.

Following the financial crisis in 2001 major economic reforms were launched and successfully fulfilled. The monetary, fiscal and financial reforms implemented in the 2000s made Turkey’s macroeconomic framework very robust. The economy rebounded rapidly and economic restructuring was transformed into strong and sustained growth. This profile of macroeconomic path was confirmed by impressive figures: real GDP growth leading the OECD and Euro area, inflation rates coming down to single digit levels sharply, strong fiscal adjustment and large inflows of capital rushing into the country.

Internalization and underlying dynamism of the economy can be best reflected by the fact that Turkey has lately become a major destination for

1 More than 70 percent of the Turkish population is concentrated in urban areas and Turkey ranks among the leading countries in terms of urbanization rate. Source: Turkish Statistical Institute, http://www.turkstat.gov.tr/PreTablo.do?tb_id=39&ust_id =11 and OECD (2010) Factbook: Economic, Environmental and Social Statistics, Paris. 2 The median age of the country is 28.8 indicating a concentrated structure of young population in the total. More than 60 percent of total population is below 35 age. Source: Turkish Statistics Yearbook (2009), Address Based Population Registration System Population Census Results.

Foreign Direct Investment (FDI) flows.3 According to the figures the Turkish economy started to absorb growing amounts of foreign capital and it ranks before many similar developing economies such as Chile, Greece, Korea, and Israel. The average level of FDI inflows was around just US$ 100 million throughout the 1990s. Today with an annual average of around US$ 15-20 billion, Turkey is the 15th most attractive destination for FDI in the world. 4

From a traditional agricultural and low value added industrial configuration Turkey has transformed into a dynamic and sophisticated economy. While services account the larger share in national value added with 43%, industry claims 11%, and trade, transport and communication sectors each have 8% shares. Agriculture accounts for only 5% of total production.

After its recent aggressive growth records, Turkey’s GDP doubled reaching above USD 650 billion in 5 years. Following the recent economic downturn Turkish economy proved to be resilient and showed a strong rebound. According to estimates GDP growth in 2010 should be very strong, and be probably among the highest in the OECD and EU areas at an annual rate between 7-8%. 5

A number of challenges loom in the horizon. In a world of highly connected economic and political interdependences and growing uncertainty, raising overall productivity and competitiveness is the key to Turkey’s economic success in the long run. For sustained economic performance, major elements of the policy mix are maintaining the predictability of macroeconomic policies, reducing capital and employment costs, supporting employment levels, and financing external imbalances with resilient resources such as FDI.

3 For more information on general investment climate in Turkey also see The World Bank (2010), Turkey Investment Climate Assessment: From Crisis to Private Sector Growth. This report draws on the analysis of firm- level survey data collected during April 2008-January 2009, supplemented by other sources, to provide a comprehensive and up-to-date description of the investment climate facing Turkish firms of all size classes, including the impact of government regulations. 4 UNCTAD (2008-2010), World Investment Prospects Survey. 5 According to the January edition of the Global Economic Prospects 2011 report prepared by the World Bank, Turkey’s gross domestic product (GDP) growth is expected to register 8.1 percent for 2010.

10

The Turkish economy needs microeconomic reforms to upgrade its productive potential. Information and Communications Technologies (ICT) have shown positive and significant effects on economic growth and are recognized to be among most critical general purpose technologies which have great transformative potentials. In the current economic environment, policies and investment that promote and accelerate ICT adoption can have an important role to play and improve the country’s competitiveness.

According to Global Competitiveness Index, (2010), Turkish economy is positioned between innovation and efficiency driven stages of development (Figure 1-2). It is an open market economy in structural transition facing fierce competition in traditional sectors like textile, agriculture, light machinery and automobiles from other developing countries like China and India. Raising productivity and supporting innovation in these sectors will be of critical importance to keep its competitiveness intact. Therefore in order to make a successful leap from an efficiency-drive to an innovation-driven economy, the Turkish economy needs to upgrade traditional sectors and harness the transformative potential of ICTs.

1.2 ICT Sector

In the Turkish case, the lack of a suitable national accounting framework for more detailed analysis hinders international benchmarking in most emerging policy areas notably ICT and innovation. There still remains a persistent information gap on this issue following many reforms in the national statistical system. The Turkish national accounts do not provide information about the size of ICT sector in terms of value added or jobs. Nevertheless utilizing other indicators, it is estimated that Turkey’s ICT sector accounts for a lower share of economic production activities compared to advanced industrial economies but has a very robust growth potential.

Market size and ICT spending are among the available indicators for comparison. Turkey was placed 13th in the list of highest spending countries within the OECD in 2007 with spending of USD 36 billion or 6.8% when compared to GDP. Various institutions put forward similar estimates concerning the Turkish ICT sector. However estimates vary and a commonly agreed figure is still unavailable. In the absence of official statistics analysts frequently rely on business intelligence data provided by different consulting companies.

0,0

5,0

10,0

15,0

20,0

25,0

30,0

35,0

40,0

45,0

C A

G R

o f

IC T

S p

e n

d in

g

Figure 1-3: Fastest growing economies in ICT spending, 2003-2009 (Source: OECD database, Economist

Intelligence Unit and EITO. The aggregate for Turkey is calculated for the period 2005-2009)

11

Even the State Planning Organization (SPO) uses these estimates in its official documents and reports. According to the SPO the size of the ICT market in Turkey was around US$ 20-25 billion in 2009. 6 Turkey ranks among the highest growth countries in ICT spending (Figure 1-3).

The Turkish ICT sector has a great potential for growth. The ICT market experienced double- digit growth over the last ten years since the financial crisis the country experienced in 2001, but the share of software and services is significantly behind Western markets, indicating substantial growth potential. 7

The Turkish ICT market is dominated by communications, constituting approximately 70 percent of the total and the whole IT market, which constitutes the hardware and software categories, comprising the other 30 percent.8

The computer hardware segment is the second largest part of the ICT sector with a 25% share following communications. The penetration rate of personal computers (PCs) has grown rapidly and reached levels of 253 per 1,000 people in 2009 from 92 per 1,000 in 2006. This figure is

6 SPO (2010), Information Society Statistics, Ankara (Turkish). 7 Turkey Invest’s ICT sector information at http://www.invest.gov.tr/en-US/sectors/Pages/ICT.aspx 8 SPO (2010) Information Society Statistics, Ankara.

higher in European markets, being 519 per 1,000 people in Italy, 725 in Germany and 766 in France. The temporary value-added tax (VAT) reduction on consumer durables, introduced in March 2009 in the Turkish market to counter the impact of the financial crisis, augmented the sales of PCs and laptops in 2009.9

Leading multinational enterprises, such as IBM, Hewlett-Packard, Dell, Siemens, Cisco, and NCR, have considerable shares in domestic IT market. These firms typically operate through their local subsidiaries, which assemble PCs and other IT hardware components imported from overseas. Sales are realized both domestically and for export to the EU, Eastern and Central Europe, and the Middle East.

Computer services and computer software have a much lesser role in the Turkish economy because of lower business use of ICTs and the rapid growth of mobile and other consumer communication services. In contrast more

9 Economist Intelligence Unit (2010). The reduction in value added tax (VAT) on consumer durables, introduced in March 2009 to combat the impact of the financial crisis, boosted PC and laptop sales. The number of PCs per 100 population is expected to rise to 38 per 100 population as of 2014.

25%

2%

3%

70%

Computer hardware Computer software Computer services Communications

Figure 1-4: The composition of ICT spending in Turkish economy, 2009 (Source: SPO 2010,

Information Society Statistics, Ankara; based on data provided by IDC.

12

sophisticated economies have a somewhat larger share of computer services and software. 10

While being relatively less important and at its early stage of progress, the software industry is among the fastest developing and most promising sectors in the country. Over the last two decades, the sector has gained great momentum in terms of capacity, production and usage. According to some analysts, the software industry accounts for a much larger share in the total IT market than previously thought. It is estimated that national software market reached US$ 1.6 billion in 2009 with double-digit growth over the last years.11

Turkey has rapidly developed its international linkages in the ICT sector. Export and import volumes in the ICT sector have reached to US$ 3.27 billion and US$ 9.05 billion respectively as of 2009 (Figure 1-5). In terms of international trade balance, the Turkish economy is positioned to be a net importer of ICT goods and services due to mainly being a large market for consumer

10 For further reading on market segments of ICT spending please see OECD (2008), Information Technology Outlook, Paris, pp. 56. 11 Source: Interpromedya (2009), ICT Market Research. Please see www.interpromedya.com.tr

electronics and computer hardware. In 2009, while the Turkish economy exported US$ 472 million computer hardware products it imported US$ 4,444 million of the same goods leading to a US$ 3,972 million trade imbalance.

The ICT sector plays a significant role in terms of employment creation opportunities and generation of R&D demand in Turkish economy. According to the latest available data, direct employment by the ICT sector was 160,644 in 2006. 12 Including employed people with ICT skills by other industries, a broader definition of “ICT related employment” indicates the sector provides much larger opportunities for job creation: the percentage share of ICT related employment in total employment was above 10 percent in 2006. A considerable and steady increase in this figure has been observed over the recent years and it reached to 10.8 percent of total employment as of 2009.

12 Source: TURKSTAT Annual Industry and Service Statistics Survey

-12.000

-10.000

-8.000

-6.000

-4.000

-2.000

0

2.000

4.000

2004 2005 2006 2007 2008 2009

Imports Exports

Figure 1-5: Evolution of ICT foreign trade in Turkish economy, US$ million, 2004-2009

(Source: Undersecretariat of Foreign Trade –Export Promotion Center, IGEME)

13

2 Turkey’s Broadband Ecosystem: A General Assessment

This section develops a country snapshot of the national broadband ecosystem in a historical, institutional and economic context. Different aspects of national broadband adoption will be covered such as infrastructure, services, market environment, applications and usage.

2.1 Synopsis of Recent Evolution of Markets and Public Sector Reform

Reform and structural change in the overall ICT and broadband sector started early with the most important segment of telecommunications. The growth and evolution of the Turkish telecom industry in the last 30 years had five separate phases:

1. Infrastructure pickup (1980 to 1984): Network infrastructure buildup was a priority and public investments played a major role in this period. As a result, the number of access lines grew by 14 percent on average; however, despite the acceleration in telecom investments and growth in the subscriber base, long waiting lines remained during this time.

2. Fast wire line growth (1985 to 1994): The late 1980s was a period of ongoing intense infrastructure investment and fast growth. During this period, the government’s annual telecommunications investments averaged US$ 656 million and access lines grew by an average of 20 percent annually. Also, the PTT (“Posta Telefon Telegraf”) was split into postal services and Turk Telekom. Two GSM 900 licenses were granted in 1993 to Turkcell and Telsim, with revenue sharing agreements with Turk Telekom.

3. Wireless revolution (1995 to 1999): The late 1990s were characterized by explosive growth in wireless phone subscriptions. In 2000, annual subscription growth exceeded 100 percent, and wireless phone penetration reached 25 percent. Average annual investment in the wireless subsector was more than US$ 1 billion. During this period, the government continued

investments in wire line at US$ 560 million per year and wire line penetration reached 28 percent. There were several unsuccessful attempts to privatize Turk Telekom.

4. Preparation for liberalization (2000 to 2004): This period is characterized by maturity in wire line and decelerated growth and increased competition in wireless. Two GSM 1800 MHz licenses were auctioned in May 2000, at a substantially higher fee (close to US$ 3 billion including VAT, compared to US$ 500 million for the initial licenses bought in 1999) to end the revenue- sharing agreements. The Telecommunications Law in 2000 established an independent regulator, the Information and Communication Technologies Authority (ICTA) and predetermined full market liberalization starting from January 2004.

5. The period of post liberalization and broadband revolution (2004-...): The government maintained its full support for liberalization and privatization of the telecommunications sector. The ownership of Telsim, a privately owned telecom operator, was transferred to the government after its owners were convicted of fraud. The operator was afterwards privatized in an international tender won by Vodafone in December 2005. 3G mobile licenses were awarded to all three operators, Turkcell, Vodafone and Avea, in 2008 and services began in 2009. This period is also characterized with increasing competition, declining fixed line penetration and falling voice revenues. Broadband, both fixed and mobile, have become a major source of revenue and a general technological platform for overall communication services.

As a result of structural policies in order to liberalize the market competition flourished, new entrants emerged as strong operators and foreign capital flew into the country.13

13 The incumbent operator of the telecommunication sector is Turk Telekom. The legal monopoly of Turk Telekom in

14

Nonetheless a distinction should be made between fixed and mobile segments of communications. As to the national market, the fixed and mobile subsectors have radically dissimilar configurations in terms of productivity, liberalization, and ownership. Mobile is more competitive and productive, with a more advanced regulatory structure. On the other hand, in the fixed subsector, productivity is relatively low.14

As observed in many other developed and saturated markets, fixed-line penetration is in continuous decline since its peak period of 2001- 2004. Consequently, the penetration rate is down to 23.3 percent in 2009, from around 28 percent in 2004. The penetration is low compared to EU countries, for example 37 percent in France, 43 percent in Germany, 46 percent in Greece and 31 percent in Hungary.15 The Turkish mobile telecommunications segment has achieved a considerable growth with a share of 60 percent within total sector revenue in 2008 and is expected to continue growing. Mobile number portability (MNP), launched in November 2008, as well as 3G mobile services introduced in July 2009, have accelerated the competition between the three market players.

Analysts forecast mobile virtual network operators (MVNO) are also expected to start operating in the market along with the three mobile operators. Mobile-phone subscribers are expected to grow at an annual rate of 5.5 percent between 2010 and 2014, following a fall of 0.5 percent in 2009. This will increase the mobile- phone penetration rate in Turkey to about 113 percent in 2014, similar to most EU countries, where penetration rates are generally around 100-120 percent.

the field of fixed line telephone services ended in 2004 and the sector was opened to full competition. Since then, many operators entered the sector to operate in the field of fixed line telephone services. 55% of the public shares in Turk Telekom were privatized in 2005. As regards to mobile subsector services started as early as 1994 and currently three operators, namely Turkcell, Vodafone, and Avea, are operating in the field of mobile communication. 14 McKinsey Global Institute (2003), Turkey: Making the Productivity and Growth Breakthrough 15 According to the analysts the outlook for fixed-line telephone penetration does not look promising. Fixed-line penetration is expected to decline to 19 telephone main lines per 100 people by 2014.

2.2 Institutional Setup: Current Policy, Legal and Regulatory Framework

Table 2-1 exhibits major institutions and establishments that are involved in the ICT and broadband sector. The State Planning Organization (SPO) is the leading institution regarding government policies and strategies for the overall ICT sector. It prepares pivotal national strategies and programs such as development plans, economic programs and sector strategies and action plans. SPO is also involved in resource allocation to e-government projects and applications and has access to private sector organizations through various mechanisms. The Office of the Prime Minister has lately become a significant coordinating and a consulting body in terms of e-government initiatives. In regards to ICT related research and determining the national agenda for general science and technology policies the Scientific and Technological Research Council (TUBITAK) is the main body.

The Ministry of Transport and Communication (MoTC) is responsible for sector oversight in the provision of public services including information and communications to the end users. The MoTC is also the top official body on issues of safety, quality, standardization and balanced development of the communications infrastructure. The Information and Communications Technologies Authority (ICTA), founded in 2000, has the overall regulatory responsibility over communications sector.

A number of non-governmental organizations are also involved in the general direction and structure of the ICT sector. The most influential establishments include the Turkish Informatics Association (TBD), Informatics Sector Association (TUBIDER) and Turkish Informatics Industry Association (TUBISAD).

Turkey’s public sector has a tradition of passing legislation rather than using secondary regulations to interpret basic laws. In an attempt to liberalize communications, to regulate provision, diffusion and usage of information and enable e-transformation different legislation was enacted throughout the 2000s. As to the legal framework in effect (or pending) the following documents constitute the fundamental architecture for the nation:

15

Institutions Abbreviations Functions Web pages

Office of the Prime Minister NA Office of the Prime Minister coordinates numerous e-government projects, gives advisory services to

implementing agencies, encourages the adoption of ICT by public institutions. NA

Ministry of Transport and Communication MoTC MoTC is to provide the production and the control of quality, balanced, safe, environmental friendly, fair

and economic transport, information and communication services for all users.

www.ubak.gov.tr

The State Planning Organization SPO

SPO provides general vision and strategies including ICT sector, integrates these specific visions into

broader economic programs and coordinates the implementation process. SPO also approves public

investment projects including ICT and e-government applications. www.spo.gov.tr

Information and Communication Technologies Authority BTK BTK prepares plans in telecommunication sector according to Wireless, Telephone and Telgraph Law .

Then BTK presents the plans to MoTC. BTK also regulates and audits tellecommunication market. www.tk.gov.tr

International Satellite and Cable Operator TURKSAT Turksat A.S. is the only satellite operator company in Turkey. Turksat manages and operates three satellites

(Turksat 1C, Turksat 2A, Turksat 3A) and provides all types of satellite communications. www.turksat.com.tr

The Scientific and Technological Research Council of Turkey TUBITAK

The Scientific and Technological Research Council of Turkey (TÜBİTAK) is the leading agency for

management, funding and conduct of research in Turkey. It was established in 1963 with a mission to

advance science and technology, conduct research and support Turkish researchers. www.tubitak.gov.tr

Turkish Informatics Association TBD TBD is a non governmental organization w hich was established in 1971 to expand Informatics Culture by

members. www.tbd.org.tr

Informatics Sector Association TUBIDER

TUBIDER IT Sector Association was founded in November in 1999 so as to protect rights and interests of

the IT companies operating in the informatic sector and to ensure vocational regulations are implemented.

TUBIDER continues its operations having reached to 800 registered members and more than 1500

applicants. www.tubider.org.tr

Turkish Informatics Industry Association TUBISAD

Established in 1979, TUBISAD is the largest non-governmental organization of the Turkish private ICT

sector, including industries and services, with a representation base of 95 % through its direct

membership. TUBISAD has a group of members comprising of almost 180 leading ICT companies of which

are Software Developers, Hardware Manufacturers, Hardware and Software Distributors,

Telecommunication Companies, System Integrators, Local Subsidiaries of IT and Communication

multinational companies and/ or Consultants. www.tubisad.org.tr

General Directorate of Postage and Telegraph Organization PTT Postage and telegraph services are operated by General Directorate of PTT. www.ptt.gov.tr

Table 2-1: Major institutions and establishments in ICT sector

16

 In the field of Information Society: Right to Information Law (2003), Inclusion of cybercrime in the Criminal Code (2004), Electronic Signature Law (2004), Amendments in the Intellectual Property Rights Law, Universal Service Law (2005), Law No. 5651 on regulating broadcast in Internet and combating crimes committed through such broadcast (2007), Electronic Communications Law (2008), and secondary legislation on telecommunications.

 In the field of IT and e-government: Public Procurement Law (2008), Census Services Law, Law No.5942 amendments in the Traffic Law No.2918, Circulars regarding the payment of taxes online, Draft Data Protection and Privacy Law (pending in the Parliament), e-Commerce studies regarding the harmonization of 2000/31 EC, and Draft law covering partial amendments for various e-government services.16

Box 2-1 gives a quick snapshot of major national strategies and initiatives targeting e- transformation in the public sector and broadband adoption at large. Evolution of public policies with regard to Internet based technologies and e-transformation can be analyzed under four distinct periods:

 Data processing and computerization (1970-1990): During this period the focus of public policy was mainly on automating back-office functions, such as processing of the census and taxes.

 Building basic ICT capabilities (1990- 2000): In the 1990s, there emerged an effort to provide basic ICT facilities and capabilities from a modernization perspective of the government sector. The “Turkey: Informatics and Economic Modernization” project of the World Bank in 1993 was an initiator in this sense. In the second half of 1990s the National Informatics Infrastructure Main Plan of the Ministry of Transportation was an early example

16 For a more detailed discussion please see OECD (2007) E Government Studies: Turkey, Paris.

of the first organized action in the public sector. However the succession of governments; high inflation and recession; and political and economic instability prevented much implementation from taking place.

 The era of e-government (2000-2010): Political stability with Development and Justice Party rule following 2002 and negotiation process with the EU helped government agencies to devise more integrated and organized frameworks for action which included active participation of private sector and non- governmental organizations. The E- Transformation Project and the Information Society Action Plan are good examples of this period. The overall goal in these documents was defined as promoting Information Society polices to increase Turkey’s competitiveness. A further goal was to move from labor-intensive production to a higher-value-added production and from providing a source of low-cost labor to a highly educated workforce in a knowledge based economy. However going beyond rhetoric, the real focus was on implementing e-government applications and demand aggregation policies rather than promoting e- commerce and the digital economy.

 Beyond e-government (2010-…): With increasing diffusion of ICTs into business applications and government operations, there are now new signals calling for a shift in public policies. Growing businesses are in clear need of accelerated ICT adoption to leverage their competitive power. The high level of investment in e-government applications is clearly just one component of modernizing the public sector. To better harness the transformative potential of ICTs government agencies are now starting to look beyond e-government. The National Vision for Broadband Strategy and updating studies of the Information Society Strategy provide good illustrations of this process.

17

Turkey faces a new window of opportunity with regard to upgrading competition and productivity by means of better ICT adoption. Promoting and creating a viable broadband ecosystem is the key to success. The country can either let the market determine the pace of development or promote a more rapid development of broadband infrastructure, as has been done in advanced countries. Defining, monitoring, and promoting the broadband ecosystem with effective means is an important area where more focus should be devoted.

2.3 Infrastructure Development

One of the fundamental aspects of broadband ecosystems is broadband infrastructure. As broadband technologies and applications

flourish rapidly, providing good infrastructural capacity has been a key policy concern among government officials. Evaluating infrastructure developments require analysts to monitor a number of indicators at the same time since any single dimension will not be sufficient to provide a complete picture.

Broadband is typically defined as a “high-speed communications network” that offers internet connectivity with download speeds of at least 256 kbit/s. 17 This definition based on the minimum level of speed is a matter of debate though it is clear that broadband technologies

17 http://www.oecd.org/document/46/0,3343,en_ 2649_34225_39575598_1_1_1_1,00.html

 “Turkey: Informatics and Economic Modernization” report was prepared by the World

Bank in 1993.

 The Ministry of Transportation published “the National Informatics Infrastructure Main Plan”

(TUENA) in 1998. The plan aimed at developing policy actions and strategies for enabling

transition to an Information Society. The Internet Advisory Council and The E-Commerce

Coordination Council were established in the same year.

 Turkey joined the eEurope+ Initiative, together with other EU candidate countries, in June

2001. Soon after, Turkey started the eTurkey Initiative, which includes the same goals as

eEurope+.

 A new government of the Justice and Development Party (AKP) took office on 18

November 2002. The Urgent Action Plan (UAP) was developed. This plan included the

foundation of the e-Transformation Turkey Project. The Project initiated in February 2003

with Prime Ministry Circular. The State Planning Organization (SPO) was assigned to co-

ordinate the e-Transformation Turkey Project. To realize the stated objectives and to

ensure the success of e-Tr, a new co-ordination unit, the Information Society Division, was

established in SPO in March 2003.

 Within the framework of e-TR two action plans were executed: Short Term Action Plan

(STAP) with 73 actions (2003-2004) (completion rate: 47%), 2005 Action Plan with 50 actions

(2005) (completion rate: 40%)

 Information Society Strategy (2006-2010) and Action Plan were developed by SPO and

took effect by a High Planning Council decision on 11 July 2006. A new institutional

structure was formed by introducing e-Transformation Turkey Executive Board, e-

Transformation Leaders and an Advisory Board.

 Prime Minister’s Office prepared the countries first National Broadband Vision in

cooperation with relevant stakeholders and participation with private sector. The National

Broadband Vision under three versions (Turkish, English and Arabic) is to be published

before end March 2011.

 SPO commenced studies to update its Information Society Strategy and Action Plan to

finalize before end of 2011, and decided to arrange a separate chapter on broadband

ecosystem.

Box 2-1: Recent strategies and initiatives related to e-government and broadband ecosystem

(Source: Interviews with Dr. Ramazan Altınok, Dr. Ertugrul Karacuha, Emin Sadık Aydın, Furkan

Civelek, Ahmet Hasanbeseoglu and Ugur Terzioglu).

18

are significantly different from dial-up lines in terms of functioning. No distinct definition for broadband exists in Turkey.

Three basic criteria are incorporated to provide an integrated picture of Turkish broadband development: Penetration, technologies and prices. A benchmarking technique is utilized in order to obtain a better picture of the relative performance of national broadband capacity both across countries as well as over time. In isolation, there is a tendency to view national growth rates as high, since, as with any new technology, growth is often in the double digits.

2.3.1 Broadband penetration

Overall performance in making broadband an essential part of daily life has been uneven and subject to significant variations among nations. The divide in performance of countries is best reflected by the concept of broadband penetration gap. The broadband gap is defined as the discrepancy between benchmark countries and that of host country in terms of adoption of broadband technologies generally measured in household or population penetration. In the case of Turkey, there is a significant fixed broadband gap as compared to other OECD members (Figure 2-1).

A number of factors are important in determining penetration rates but as a general observation countries with relatively high fixed broadband penetration rates tend to have relatively higher per capita GDP and well developed communications infrastructure.18

18 OECD (2008), Broadband Growth and Policies in OECD Countries, Paris.

Studies indicate that, though the relationship is not necessarily casual, per capita income is correlated to fixed broadband penetration with a significant correlation coefficient of 0.70. Since the Turkish economy has depicted a strong rebound in the post crisis environment and medium to long-term outlook is promising, rising per capita income levels should establish a major driver force for broadband infrastructure development.

Turkey faces a challenge in order to catch up with OECD members in broadband technologies. Despite growth in broadband infrastructure, when compared to advanced countries, Turkey has recorded somewhat poor performance and has not been able to close the broadband gap. As of the second quarter of 2010, in terms of fixed broadband technologies the leading countries reached penetration levels above 35 percent while Turkey managed only a 9.4 percent subscription based penetration rate.

Turkey’s fixed broadband growth rate is lagging the OECD area. During 2010 Turkish fixed broadband growth was 0.83 percent compared to the OECD average of 1.72 percent (Figure 2-2. When compared to major developing economies, Turkish broadband growth is considerably lower; for instance India is expanding its broadband capacity with a growth of 62 per cent in 2010.

Nevertheless the fixed broadband penetration rate in Turkey still exceeds some European

0

5

10

15

20

25

30

35 France

Germany

United Kingdom Canada

United States Japan

Turkey

Italy

Broadband gap among selected countries and Turkey through, June

S

Figure 2-1: Fixed broadband penetration gap with selected countries, 2010

Source: OECD broadband portal and author’s own calculations)

19

countries such as Poland, Italy, Bulgaria and Romania, and is very close to the rates in Portugal, Hungary, Spain and Estonia. On the other hand, the personal computer (PC) penetration level in Turkey in 2009 was only about 25.3 percent, compared to 77 percent in the UK. Since broadband usage depends on PC penetration, increasing PC usage and ownership

in Turkey are expected to create opportunities for the broadband market.19

Turkey’s relatively well-developed mobile network has a penetration level around 85 percent and provides another means for broadband access (Figure 2-3). Wireless broadband has shown a significant expansion

19 Economist Intelligence Unit, 2010.

-3

-2

-1

0

1

2

3

4

5

OECD net increase

Source : OECD

Figure 2-2: Fixed broadband (per 100 inhabitants), net increase, June (2009-2010) (Source: OECD)

0

10

20

30

40

50

60

70

80

90

100

2002 2003 2004 2005 2006 2007 2008 2009

Fixed Line Mobile Fixed Broadband

Figure 2-3: Selected penetration indicators, per 100 inhabitants (source: author)

20

following the decision of Turkish government to issue 3G licenses during November 2008, making the market an appealing area for future investment. Since then 3G mobile Internet subscriptions using data cards have grown at a rapid pace and as of end 2010 reached 1.5 million subscriptions (total 3G subscriptions reached 19.3 million as of December 2010 according to ICTA). The annual rate of growth in this segment of broadband was 265 percent between 2009 and 2010. As a result mobile broadband is growing much faster than fixed broadband subscriptions (27 percent annual growth as of 2010). Total fixed and mobile broadband through data card subscriptions reached 8.5 million by the end of 2010.

A saturated voice market with declining revenues has motivated operators to rely on mobile broadband network investments for optimizing

profits.20 There has been a considerable growth in the investments to mobile subsector and more importantly telecommunication operators have changed their strategy significantly by allocating increasingly far more resources to mobile investments. Building up of 3G communication infrastructures constitutes a determining factor for this shift in corporate strategies.

Experiences in many countries show that governments should leave the choice of technology and infrastructure expansion as much as possible to market forces while promoting a level playing field for different technologies. Turkey is implementing a policy of

20 The revenues from broadband access services provided from the fixed network are rapidly increasing. Parallel with the decrease in the number of fixed line telephone subscribers, the revenues from these services decreased by 17.5% between 2006 and 2009 and the revenues of mobile telephone services increased by 24.4% during the same period.

Item Subscriptions

Penetration

(per 100

people)

Comment

Total broadband 8,516,898 11.6 Fixed and mobile through data

cards

Fixed broadband 7,068,878 9.6 ICTA. ADSL, Cable modem, FTTP

3G subscriptions 19,400,000 26.3

ICTA. Technical ability to access

mobile broadband without

consideration of whether

actively using to access Internet

Mobile broadband

data cards 1,448,020 2.0 ICTA

% of

households

Households with

computer Turkstat

- Desktop 33.8 Turkstat

- Portable 16.8 Turkstat

Households with

access to the Internet 41.6 Turkstat

Households with

broadband access to

the Internet

34 Eurostat

- of which DSL 30.5 Turkstat

Table 2-2: Broadband indicators, Turkey, 2010

(Source: Adapted from ICTA, Turkstat, Eurostat) (note: Subscription penetration calculated based on

population of 73,722,988 reported by Turkstat at December 31 2010)

21

increased competition through the privatization of its incumbent telecom operator (55% was sold to the private sector in August 2005) and implementing rules to increase competition such as providing competitors access to telephone lines.21 Yet this policy was not transformed into increased competition and a high rate of growth in broadband infrastructure. Alternative operators accounting for only 7 percent, the level of competition in fixed broadband access is still very limited.22

As the National Broadband Vision (2010-2023) highlights, if Turkey wants to develop its broadband ecosystem rapidly and close the broadband penetration gap with advanced economies the government should follow an integrated and strategic approach in terms of broadband policy. 23 Waiting for growth in household demand for high-speed Internet access to encourage the development of broadband will take time. The public sector should consider taking a more pronounced role in stimulating the development of broadband infrastructure and using this capacity for modernizing the public sector.

2.3.2 Broadband technologies

Penetration data does not deliver enough information about the quality improvements in the broadband infrastructure. While the growth rate is slowing down there may still be

21 In March 2010, Oger Telecom declared its interest in acquiring an additional share of the Government's stake in TT. The Turkish Government plans to divest a further 15- 20 per cent of its stake via an initial public offering. The remaining 15 per cent stake in the operator is already listed on the Turkish stock exchange. 22 Source: SPO (2010) Information Society Statistics. 23 Office of the Prime Minister (2010) National Broadband Vision, Forthcoming.

considerable internal changes as users upgrade their connections with faster speeds. This section helps to illuminate Turkish broadband development from a technological perspective.

2.3.2.1 Backbone network

Until recently there have been two main technologies or platforms over which broadband services have been delivered to consumers. The first is Digital Subscriber Line (DSL) which entails upgrading the legacy public switched telephone network (PSTN). The second platform consists of the cable-modem technology, which entails upgrading the cable-tv network.

2.3.2.2 Local access technologies

There has been an impressive shift away from dial-up Internet connections to broadband. In 2005, dial-up connections still accounted for 40 per cent of fixed Internet connections but had already fallen to just 10 percent by 2010.

In Turkey, xDSL is the most widely used technology accounting for 94 per cent of broadband subscribers.24 While 3.9 percent of the subscribers use cable networks and 2 percent of subscribers have fiber technologies for broadband access the number of xDSL subscribers has touched to 6.6 million by end of 2010. When compared to EU countries Turkish broadband is relatively far more concentrated on conventional technologies like xDSL.

24 ICTA (2009), Electronic Communication Sector in Turkey, Market Statistics Report, February 2010, p. 20

Table 2-3: Subscription indicators by technology (Source: ICTA)

22

According to the recent data there has been a rapid shift away from lower speed connections.25 The great majority of ADSL subscribers (64

25 Source: Information and Communication Technologies Authority (2010), Communications Market Report, Ankara (In Turkish).

percent) have an 8 Mbps connection. The share of the subscribers who have 1 Mbps connections has decreased dramatically from 42% to 26% in just six months of 2010.

,94

,4 ,2 ,1

,79

,15

,2 ,4

,0

,10

,20

,30

,40

,50

,60

,70

,80

,90

,100

xDSL Cable Fiber Other

Turkey EU

Figure 2-4: Technological configuration of broadband infrastructure, as percentage

shares, 2010 (Source: author)

2%

26%

2%

2%

64%

4%

<=512 Kbit/s 1 Mbit/s 2 Mbit/s 4 Mbit/s 8 Mbit/s > 8 Mbit/s

Figure 2-5: Breakdown of fixed broadband subscribers by access speeds, 2010

23

The incumbent Turk Telekom (TT) provides landline, mobile and Internet services. It launched

ADSL services in 2003, and currently offers Internet services in selected regions through its

subsidiary TTnet. In 2005 in a drive for privatization, the government sold a 55 per cent stake of

TT to the Saudi-owned Oger Group. In March 2010, Oger Telecom declared its interest in

acquiring an additional share of the Government's stake in TT. The Turkish Government plans

to divest a further 15-20 per cent of its stake via an initial public offering. The remaining 15 per

cent stake in the operator is already listed on the Turkish stock exchange.

One of the key benefits of TT’s privatization is the fact that it has lost the special status of a

state-owned company and as a result is obliged to follow the same regulations as any other

Turkish telecoms operator. TT is likely to remain aggressive within the telecoms marketplace

and in the regulatory process as it tries to maintain its leading position. Thus, the low level of

competition in the broadband internet is partly due to the strategic behavior of Turk Telekom

to prevent the entry of alternative Internet service providers. In most countries incumbent

operators such as Turk Telekom have been under the regulatory obligation of allowing new

entrants to use the existing network to provide their own DSL service In a landmark decision

taken in November 2008, the Competition Authority has imposed a fine of 12.4 million TL

(about €6.2 million) on Turk Telekom for abusing its dominance in the wholesale broadband

Internet market through price squeeze in the retail Internet market. In its decision, the

Competition Board stated that Turk Telekom and its Internet subsidiary TTNet endured

operating without profits for long periods of time and implemented campaigns that would

not cover losses in reasonable amounts of time and that these strategies were executed in

order to monopolize the sector. In a recent decision the Competition Authority stated that

Türk Telekom should provide naked DSL services. Provision of naked DSL means that Turk

Telekom can no longer bundle voice and data services together and that the subscriber can

subscribe to DSL services alone, without having to pay for voice services as well.

In recent years Turksat with its UYDUNET cable services has increased its market share by

winning some 274 thousand broadband customers as of 2010. However since the cable TV

infrastructure is operational mostly in urban areas and lacking somewhat more effective

business models the company has not made a major breakthrough in the domestic market.

Another striking development was the introduction of 3G services which had a significant

impact on the broadband sector. The three mobile operators, Turkcell, Avea and Vodafone,

are trying to find new sources of revenue in anticipation of the difficulties associated with

increasing subscriber numbers in the saturating market. All three won 3G licences in

December 2008 and launched 3G services at the end of July 2009. Avea allocated TRY 1

billion to be spent mostly on 3G network development during 2009. Vodafone followed suit

with a TRY 1.3 billion investment allocation. At the end of December 2010, there were

1,448,020 mobile broadband subscribers, showing a 265 per cent growth over the last 12

months. At the end of 2010, the number of customers on 3G networks reached 19.4 million

With the emergence of alternative network technologies TT is now under pressure for

adopting a more business oriented strategy rather than enjoying its domination in the fixed

market. TT introduced its new Internet speed package days before 2011 which enables

speeds of up to 32 Mbps with VDSL2 and up to 8 Mbps through ADSL. TT shall begin offering

packages with speeds of up to 50 Mbps and 100 Mbps with VDSL2 technology, and shall

increase the speed up to 16 Mbps with ADSL2+.

TT will need to work hard to retain its market dominance however, as mobile operator Turkcell

is investing in considerable fibre infrastructure through its subsidiary Superonline. Vodafone

Turkey also bought a major alternative operator in late 2009. Another point is that EU

candidacy is prompting Turkey’s telecoms regulator to strengthen the regulation of the sector

and curtail TT’s monopolistic powers.

Box 2-2: Leveling the field: The battle of broadband in national market

(Source: Atiyas Izak (2010) Regulation and Competition in the Telecommunications

Industry: An Update; and interviews with Dr. Ramazan Altınok, Dr. Ertugrul Karacuha, Emin

Sadık Aydın, Furkan Civelek, Ahmet Hasanbeseoglu and Ugur Terzioglu)

24

Turk Telecom is leading provider of DSL and owns all of fixed telephone infrastructure (16 million connections in 2010). As of 2009 the share of alternative operators reached to 6.3%. Compared to the previous year, alternative operators’ subscriber number increased by 29%. Currently two types of Local Loop Unbundling (LLU) namely ‘full access’ and ‘shared access’ have been implemented in Turkey. 163 central offices have been opened to access of alternative operators by the end of 2009 under the LLU regulations of the Authority. Currently alternative operators have technical access to almost 7 million PSTN and 2.6 million ADSL subscribers via the above mentioned 163 central offices. Besides LLU, operators can offer broadband Internet access services to their customers using Turk Telekom’s other wholesale broadband access services such as xDSL simple resale and xDSL bit-stream access (BSA) which enable ISPs to access the fixed broadband network at different levels. Some promotional campaigns including LLU in the above mentioned wholesale access services have been approved by the Authority upon Turk Telekom’s proposal in 2009.

Co-location areas and problems encountered at central offices opened to access have been observed within the central office surveys carried out in June. Moreover the outdoor DSLAM implementations of the incumbent operator are also observed.

In accordance with the relevant legislation and regulations, a Reference xDSL Resale Offer has been offered to the Authority by Turk Telekom at the end of June 2009. The public consultation process regarding to the draft offer has been completed and approval procedure of the reference offer is ongoing. Besides, some new re-sale tariffs proposed by Turk Telekom have been included in the existing offer upon approval by the Authority. In this context, ADSL 2 Mbit/s with 4 GB quota, ADSL 8 Mbit/s unlimited, MEB ADSL 2 Mbit/s, 4 Mbit/s unlimited, ADSL up to 8 Mbit/s unlimited, with 4 GB and 6 GB quota, ADSL 512 Kbit/s with 512 MB quota and ADSL2+ 16 Mbit/s unlimited resale tariffs have been approved in 2009 and included in Turk Telekom Reference xDSL Resale Offer.

Mobile broadband offers a competitive alternative to fixed broadband. Turkcell, Avea and Vodafone, the three operators in mobile segment had authorization from the government to provide 3G services as of December 2008,

and became operational by July 2009. Since then mobile broadband subscriptions recorded an explosive expansion. With some 1.45 million subscriptions (December 2010) and a growth rate of 265 percent, mobile broadband holds great promise for Turkey, particularly given the limited competition on the fixed market. Mobile broadband provided over the UMTS platform with HSPA extensions came relatively late to Turkey. The 3G market is benefiting from the deployment of HSPA technology and by the introduction of new smartphones. Growth has been dramatic and mobile broadband accounted for 17% of all broadband subscriptions by the end of 2010.

Though the level of competition in fixed broadband is limited compared to more advanced economies, with a broader definition of broadband including the mobile segment, Turk Telekom as the incumbent operator is losing its market share rapidly. TT’s broadband market domination has been somewhat challenged by mobile operators’ (Turkcell, Avea and Vodafone) success in targeting mobile broadband customers and partly due to increases in cable and fiber subscriptions. TTNet’s market share in the broadly defined broadband market was down from 93% to 71% since early 2009. By contrast the market share of mobile broadband operators grew explosively up to 17% following the opening of 3G services in the country. It is estimated that mobile broadband subscribers will occupy an increasingly prominent place in Turkey’s broadband sector. These are broadband customers who use devices such as netbooks, smartphones and USB sticks to connect to the Internet wirelessly via a high-speed (3G/HSPA) network. The share of cable network operator Turksat has also augmented from 1.3 % to 3.2 % during the last two years and fiber connections expanded their share from 1% to 1.8 % since early 2010 (Figure 2-6).

Turksat provides both satellite and cable television services. It has around 1.2 million cable television subscribers.26 Turksat has upgraded its networks and made cable broadband services available to the majority of homes in urban areas. Nonetheless the cable network is in its takeoff phase and needs an effective management model for further expansion. The Turksat “Uydunet” service provides its subscribers with broadband Internet access through its cable infrastructure at speeds

26 http://www.turksat.com.tr/english/v2/cable-services

25

Figure 2-6: Shares of technology in total broadband subscriptions, percent, 2010

xDSL (TTNet) xDSL Other Cable (Turksat) Mobile Fiber Other

0% 10% 20% 30% 40% 50% 60%

Ireland

Germany

Finland

Switzerland

Turkey

Poland

Italy

Netherlands

Portugal

United States

Iceland

Hungary

Czech Republic

OECD

Denmark

Norway

Sweden

Slovak Republic

Korea

Japan

Figure 2-7: Percentage of fibre connections in total broadband, June 2010 (Source: OECD)

26

ranging from 1 Mbps to 20 Mbps. One challenge for expanding cable broadband access is that most TV subscribers in Turkey have a preference for satellite delivery (some 10 million satellite dishes in mid-June 2010).27

Fiber networks are ideal because the capacity is much higher than traditional copper lines and relatively easy to expand once the fiber is in place simply by adding additional lasers to a line. When compared to advanced Asian and European economies Turkey lags in fiber broadband penetration (Figure 2-7). However the recent expansion in fiber subscriptions is very promising and it indicates the demand for high bandwidth network access could possibly be ready in the country. During 2010, the share of fiber subscriptions in the total broadband market reached 1.8 percent.

In terms of technology deployment, though there have been outstanding market developments that reflect the overall potential in demand and supply sides, the majority of Turkish broadband infrastructure is based on traditional forms of connectivity devices which do not allow high bandwidth data services such as advanced e-health and multichannel television. Turkey shifted from a dial up network structure to a broadband dominated one at a very rapid pace. Turkey’s fiber Internet backbone provides for high-speed connections, but this has not translated to a high number of broadband users.

In the long run the leading policy challenge for the government would be facilitating the development of high speed broadband networks most notably fiber. High-speed broadband has strategic importance and should be promoted. Much as traffic expands to fill roadway capacity after the building of superhighways, having more broadband capacity will likely facilitate the delivery of government e-services, provide a foundation for e-government in whole sectors of government, and stimulate the development of e-business. Besides the recent rapid shift into high bandwidth subscriptions and explosive growth in mobile and fiber segments indicate that the future demand for broadband ecosystems will be strong. Turkish universities have already developed their own broadband network, as has the National Adjudication Network Project (UYAP). The e-health initiative

27 http://www.eutelsat.com/products/broadcast- audience.html

is proposing to build its own network for exchanging data.

The lack of a complete and strategic approach would impair this great opportunity. There are signs of a better understanding emerging in government circles. The National Broadband Vision has urgently called on the public and private sectors for the development of a holistic government approach to broadband. SPO is getting ready to update the nation’s Information Society Strategy and disclosed publicly that it shall prepare a separate chapter on broadband. A more organized approach should prepare for harder policy designs to broadband challenges and ensure competition in the national broadband market since it is the best option and the main driver of continuous improvements.

The Turkish government should also ensure a more effective regulative approach for broadening competition in the broadband market. The development of competition in broadband has been extremely slow, primarily because the Ministry of Transport and the Information and Telecommunication Technologies Authority have been slow in adopting, implementing and enforcing the necessary secondary legislation. Analysts argue that Turk Telekom’s influence on the Ministry and the Information and Telecommunication Technologies Authority’s lack of full independence from the Ministry are to a large extent responsible for this state of affairs.

2.3.3 Prices

Prices are a crucial indicator in evaluating the state of broadband development in any given country. Analysis of prices, the level of service, and the range of choice available to users, can help inform questions about the take-up and use of broadband services as well as in assessing how competitive broadband access is in individual markets. While there is still not enough data to systematically examine price and performance changes, the trend in prices is very much downward; and the trend in performance is upward. Competition is the main driver of improvements in price and performance.

International comparison suggests that prices of broadband services in Turkey are among the highest in OECD and EU economic areas. The average broadband monthly price per advertised Mbit/s in Turkey is about 40 percent higher than the OECD average. When prices are corrected for purchasing power parity, prices in Turkey are the highest (). Turkey is also lagging behind in

27

high-speed connectivity prices. Prices in Turkey are relatively cheaper for low speed connections and relatively more expensive for high-speed connections.

Presenting the latest situation as of March 2011, Table 2-2 confirms that high prices and affordability are among main concerns in Turkish broadband market. Turk Telekom offers DSL services with different speed options. The baseline ADSL offer at 1 Mbps costs US$ 13.2 per month with a 1 GB cap on data transfer. Higher speed services are much more expensive than EU countries. The highest speed 100 Mbps service is priced at US$ 94.3 per month with unlimited data option and a fiber 32 Mbps service costs US$ 27.2 per month with a 10 GB Cap. Turk Telekom’s ADSL2+ 16 Mbps services are priced at US$ 37.3 and USD 20.9 with unlimited and 5 GB Cap respectively. Turk Telekom also offers ADSL technology to consumers with the highest speed option 8

Mbps going from US$ 69 per month with unlimited data transfer and decreasing prices for more limited speeds and data cap options.

Among independent ISPs offering broadband Internet services, Superonline provides DSL and fiber Internet access with different speed options. The baseline starts with a 1 Mbps speed service at USD 28.7 with unlimited data transfer, and 8 Mbps speed option goes from USD 24.7 and USD 18.3 with 6 GB and 4 GB Cap respectively. Among fiber connections the highest option is 100 Mbps costing USD 125.9 with unlimited data transfer.

As the largest cable TV service provider, Turksat runs its Uydunet to offer broadband services at different speed options. The baseline is the 1 Mbps speed service costing USD 18.3 per month with unlimited data transfer. However higher speed services are priced relatively higher than the average ADSL services. The 10 Mbps option costs USD 62.6 and the 20 Mbps service

135.36

91.29

79.53

62.76

59.45

57.06

56.71

54.24

51.64

50.11

49.25

48.34

46.85

41.01

40.63

40.20

39.85

39.75

39.39

38.87

38.50

35.18

33.25

32.70

32.50

32.46

31.33

30.80

29.91

27.72

0 20 40 60 80 100 120 140 160

Turkey

Luxembourg

Portugal

Norway

Australia

New Zealand

Mexico

Canada

Czech Republic

Austria

United States

Spain

Slovak Republic

Netherlands

Iceland

Denmark

Poland

Belgium

Hungary

Germany

Korea

Japan

Switzerland

France

Sweden

Finland

Ireland

United Kingdom

Italy

Greece

Average Subscription Price, USD, PPP

Figure 2-2: Broadband average monthly subscription price, Oct. 2009, USD PPP (Source: OECD)

28

Table 2-4: Prices of selected broadband service packages in Turkey, March 2011

(Direct source inquiry to author)

costs USD 125.9 per month without any cap on data transfers. Prices per Mbps offered are significantly higher than ADSL operators.

The three mobile operators, Turkcell, Vodafone and Avea are also offering mobile Internet services based on HSDPA technology and there are quite differentiated service packages for consumers. Table 2-4 presents a selection of these mobile packages most notably the baseline options. All three operators offer 1 GB cap mobile data options at around USD 18-19 as of March 2011.

Since the competition in fixed broadband market has been relatively limited in Turkey, the general downward trend in prices seems to be somewhat partial. The effect of taxes on broadband service pricing is another significant variable in assessing the affordability and performance of the market. As part of the need to augment the tax base during the economic crises of 2000-2001, Turkey raised taxes on telecommunication services. While these measures have increased tax revenues, they had a negative effect on e-services take-up. In 2008 the government considered ways to increase the affordability and thereby the use of e-services by reassessing telecommunication taxes but little has been achieved.

The tax rate applied for broadband products and services include the 18% Value Added Tax (VAT) and an additional 5% Special Communication Tax (SCT) calculated over the

price of the related service including VAT. Compared to the other communication services, Internet services are taxed relatively lower: SPC rates on mobile and fixed line communication services are 25% and 15% respectively. Though taxes on Internet services are relatively low the fiscal burden still accounts for a major impact on a middle-income household.

One can expect that when Internet access requires a lower share of monthly per capita income, usage rates increase, everything else being equal. Internet access cost – measured as the Internet subscription cost as share of average monthly income – relatively high. According to analysts, middle-income consumers are generally ready for broadband packages costing less than 3 % of their per capita income.28 In Turkey the Internet access cost of an average broadband package has come down significantly in recent years but still exceeds 3% threshold with a rate around 5%.

National growth performance has created opportunities for broadband expansion and service affordability. The high growth in per capita income in recent years affected positively the broadband affordability by increasing the denominator. It is clear that rising incomes have supported the market and affordability of

28 Source: Interview with Ahmet Hasanbeseoglu from Cisco.

Operator Technology Speed (Mbps)

Price per month

(TRY)

Price per month

(US$)

Price per Mbps

(US$) Note

TTNet (TTelekom) ADSL 1 20,9 13,2 13,2 1 GB Cap

TTNet (TTelekom) ADSL <=8 29 18,3 2,3 4 GB Cap

TTNet (TTelekom) ADSL <=8 32 20,2 2,5 6 GB Cap

TTNet (TTelekom) ADSL 8 109 69,0 8,6 Unlimited

TTNet (TTelekom) ADSL2+ <=16 33 20,9 1,3 5 GB Cap

TTNet (TTelekom) ADSL2+ <=16 59 37,3 2,3 Unlimited

TTNet (TTelekom) Fiber <=32 43 27,2 0,9 10 GB Cap

TTNet (TTelekom) Fiber <=100 149 94,3 0,9 Unlimited

Superonline ADSL 1 45,3 28,7 28,7 Unlimited

Superonline ADSL <=8 29 18,3 2,3 4 GB Cap

Superonline ADSL <=8 39 24,7 3,1 6 GB Cap

Superonline Fiber <=50 59 37,3 0,7 12 GB Cap

Superonline Fiber <=100 199 125,9 1,3 Unlimited

Uydunet (Turksat) Cable 1 29 18,3 18,3 Unlimited

Uydunet (Turksat) Cable <=10 99 62,6 6,3 Unlimited

Uydunet (Turksat) Cable <=20 199 125,9 6,3 Unlimited

Turkcell HSDPA 7,2 29 18,3 2,5 1 GB Cap

Vodafone HSDPA 7,2 30 19,0 2,6 1 GB Cap

Avea HSDPA 7,2 29 18,3 2,5 1 GB Cap

29

broadband services in a country where competition was somewhat imperfect.

Internet use will likely increase significantly if the cost of accessing the Internet can be reduced. The government should prompt action to reduce taxes on both telecommunication and broadband services. On the supply side the Information and Communication Technologies Authority should continue to actively stimulate competition in the telecommunications sector to

promote faster and more affordable Internet access in line with EU initiatives in this area.

Mobile phones can provide an alternative, less expensive channel for delivery of e-services to citizens and businesses. Given the high number of mobile subscribers in Turkey, the government should consider this channel to provide information and services. In this regard a simple, flat and more affordable tax rate applied to communication services could be a good and less distortive option.

30

3 Applications Government policies should ensure not only development of broadband infrastructure but also have to ensure that broadband technologies are adopted and used increasingly by the country’s businesses, households and public sector. The previous sections of the report focus on penetration, price, and speed of high bandwidth Internet networks in Turkey and highlight how the country tried to promote competition and encourage investment in order to increase connectivity. This part presents the content and applications created by private, civil and public agents over broadband platforms.

3.1 Government

The period of 2000-2010 can also be called the era of e-government in Turkey. The negotiation process with the EU and the rule of a single party government helped public agencies to design and manage more concerted efforts on e- government applications and infrastructure. Since then Turkey has made strong progress in implementing e-government. The number of services integrated with the e-Government gateway reached 139 (6 fold increase in one year) and the number of visitors increased over half million people (50 fold augmentation in one year) in 2010. While the central government allocated US$ 267 million for 203 e-government projects in 2002, the public investment budget rose to US$ 590 million for 244 projects in the pipeline in 2010. The number of web sites to provide access for public services grew explosively during this period and as of 2010 the total number was above 20 thousand (Figure 3-1).

The share of services provided through electronic channels in total public services reached 66% and the e-services user satisfaction rate was recorded as 95% in 2009. In a benchmarking study conducted by the EU in 200729, the maturity level of 20 basic public services in Turkey scored very much near the EU average and even overtaking slightly the EU score in the business services segment (Figure 3-2).

Table 3-1 presents leading e-government applications implemented in Turkey. These

29 European Commission. 2007. The User Challenge Benchmarking The Supply Of Online Public Services. http://ec.europa.eu/information_society/eeurope/i2010/d ocs/benchmarking/egov_benchmark_2007.pdf

projects have been enablers of e-government and e-transformation in the Turkish public sector. Many of the projects are high volume and high value ones with important transactions on a daily basis such as MERNIS and UYAP. Turkey has also demonstrated good progress in terms of electronic tax returns. The Internet Tax Office has become operational and started to accept declarations and payments online. The Custom Modernization Project, one of the long waited initiatives, finally completed in this period and became effective in 2009. E-government applications also include some public portals to provide up to date information to targeted audiences or to the broader public. The Turkish Foreign Investment Portal, KOBINET and Tourism Portal are examples in this regard.

In the last ten years Turkey has made strong progress to develop e-government applications and infrastructure. Some main findings and policy lessons could be drawn from this transformation process to inform similar countries:

 Strong political leadership: One of the reasons for the rapid development of e- government in Turkey is its strong support within the high-level political leadership. E-government leadership is provided by the e-Transformation Turkey Executive Board, chaired by the Deputy Prime Minister. The new government’s modernization agenda considers e-government as a major tool for change.

 Effective organizational management: In terms of planning, financing and monitoring functions The Executive Board is aided by the State Planning Organisation (SPO), which is responsible for scrutinizing all public investments. From a broader perspective the SPO has provided necessary technical and organizational services to develop, implement and monitor a coherent strategy for e- transformation in the public sector organizations.

 A strategic approach to e-government: Together with line ministries and private sector agencies these organizations

31

0

5.000

10.000

15.000

20.000

25.000

2003 2004 2005 2006 2007 2008 2009 2010

gov.tr Other Public Inst. Total Public

Figure 3-1: Number of public sector owned websites

Figure 3-2: The maturity level of selected public services, percent, 2007 (Source: European

Commission 2007. The Use Challenge Benchmarking The Supply Of Online Public Services)

0

10

20

30

40

50

60

70

80

90

100

Turkey EU27

20 Services 12 services for citizens 8 services for business

32

Project Explanation Web pages

E-Government Gateway

Launched in December 2008, one stop shop, integrated, multi-channel (internet, mobile, call center) service delivery

platform for identification and payments. Mobile portal launched in October 2009 currently 139 public services are

integrated.

www.turkiye.gov.tr

Central Census Management System (MERNIS) Unique ID number for all citizens (over 130 million records www.nvi.gov.tr

Identity Sharing System (ISS)

ID data sharing platform through web services. More than 2,500 agencies, both public and private, signed 19 protocols to

use the system. PM Circular dictates public agencies do not require citizens to submit ID records and address information

and obliges them to acquire this information through ISS. www.nvi.gov.tr

Address Record System Citizens address records are integrated with MERNIS through ID numbers. www.nvi.gov.tr

National Judiciary Informatics System (UYAP)

Aims to ensure fast, reliable and robust judicial system. Comprises all of the courts, public prosecutors, prisons, other

judicial institutions and other government departments. Serves to some 40,000 officials and judges, 66,000 lawyers and for

all citizens. Litigate online case, reach and examine case information. Informs citizens on their cases through SMS. Ministry

of Justice announces 2.7 billion € savings with

applications of UYAP. www.uyap.gov.tr

Internet Tax Office Online tax declarations and payments. Online payment of motor vehicle taxes and traffic fines. 99 % of tax returns are

declared online. OECD study shows unit cost of tax revenue collection decreased from 2 $ to 0.35 $. www.gib.gov.tr

Police Network and Information System (POLNET)

Covers 684 locations. Internal applications (26 main applications, 51 search programs and 26 local applications) and some

online applications: passport application, driving license (application and information on penalty points), denounce on

public order and security, search for lost persons. www.egm.gov.tr

Informatics Sector Association

TUBIDER IT Sector Association was founded in November in 1999 so as to protect rights and interests of the IT companies

operating in the informatics sector and to ensure vocational regulations are implemented. TUBIDER continues its operations

having reached to 800 registered members and more than 1500 applicants. www.tubider.org.tr

Accounting Office Automation Project (Say2000i) Network for communication and data transfer for over 6,000 users in 1,660 auditing branches and 39,500

budget offices throughout the country. All financial transactions are processed online and monitored real-time. www.muhasebat.gov.tr

Customs Modernization Project

Facilitates legal trade and prevents illegal trade by using ICT. Customs formalities are carried out in a computerized media in

real-time. 100% of customs transactions are processed electronically. Around 6.3 million import and export declarations

were processed by using this system in 2009. www.gumruk.gov.tr

Social Security e-Declaration Project Social security premiums of employees are declared online. People pay the premiums electronically, verify payments made

and outstanding balance. More than 90% of firms process declarations online.

www.sgk.gov.tr

Turkish Foreign Investment Portal

Guide for foreign investors on: starting a business, cost of doing business, taxes and incentives, special investment zones,

demography and labor force, business environment and business legislation, infrastructure, economic outlook, Investment

Support and Promotion Agency assists investors in

Arabic, Chinese, English, French, German, Italian, Japanese, Korean, Russian, or Spanish as well as Turkish. www.invest.gov.tr

KOBINET for SMEs

Provides information about business environment (marketing, management, taxes, human resources, internet and

technology, finance, insurance and labor regulations). Information about firms, online business opportunities, 22,915

members. Provides user ID, password, e-mail address, web pages in 7 languages as free of charge. www.kobinet.org.tr

Tourism Portal Provides information on history, heritage, life-culture, travel guide, where to go and business. www.goturkey.com

Electronic Signature and Certification Authority

Electronic Signature Law is effective since 2004. Electronic signature has the same legal effect as that of a

handwritten signature. Public Certification Authority (PCA) for public agencies. 3 private electronic certificate service

providers www.kamusm.gov.tr

Mobile Electronic Signature Introduced by a Turkish GSM operator. Initial applications in banking, customs, justice and some local administrations. www.btk.gov.tr

Table 3-1: Major e-government applications in Turkey (Source: Author’s compilation)

33

managed to develop a shared vision and a focused strategy including quantitative targets for outputs and outcomes in order to make e-government happen.

 Prioritizing projects: Instead of creating many e-government services Turkey with a clear focus, has managed to prioritize projects that have high volume and high transactions: e-procurement, making social security and health payments, and collecting customs, tax and social security payments etc. This selective list also included projects to establish e-government infrastructure and ICT network platforms to enable provision of e-services.

On the supply side the main challenge for Turkey now seems to be improving Internet access and broadband development. Turkey’s basic communications infrastructure is the telephone line network and the country clearly needs alternative technologies such as fiber networks for next generation broadband services. Internet access costs remain high causing the penetration rates to go flat. Public policy should assume a more pronounced role in terms of ensuring broadband infrastructure development.

Another supply side challenge is increasing the capacity of line ministries and local governments to develop and implement e-government within their respective sectors and in their subordinate agencies. A strong case is needed to make for co-operating rather than developing solutions in isolation. As e-government applications proliferate and the number of projects in the pipeline augment Turkey needs to transform its centralized organizational setup in e-government to a more horizontal and participative one.

An early study conducted by OECD (2001)30 defined lack of ICT skills as a hidden threat to e- government and e-transformation. Turkey’s e- government infrastructure and applications are relatively more advanced than her ICT skill base in public offices. To transform the government by means of e-government Turkey should focus on building appropriate skills and ensure people are willing to work in the public sector.

On the demand side, the results of the Household and Enterprise Use of Information

30 OECD (2001) The Hidden Threat to E-Government, Avoiding large government IT failures, PUMA Policy Brief No. 8.

Technologies Surveys carried out by TURKSTAT provides information regarding access to and use of online public services by citizens and enterprises. According to these results, enterprises use public web pages more than the household sector. As of 2008 while 68% of the surveyed enterprises responded that they use public web pages to get information only 26% of households replied the same question positively. During their visit to the public web pages, 91% of households spent time by getting information, 30% by downloading forms and 34% sending information and uploading forms in 2009. With a 92% response share, getting information is among the leading motivation for enterprises visiting public sector web portals. 83% of enterprises get involved in downloading forms whereas 70% responded positively that they filled and sent some forms through web sites. More interestingly 7% of enterprises submitted proposals for public tenders via public web portals.

Individuals who do not visit public web pages responded that the preference to carry out transactions face to face in public offices seems to be the chief reason for not using e- government services. 19% of the group replied the unavailability of online services has prevented them to use public web sites frequently. Likewise 74% of the enterprises not preferring to use public web sites state the major reason for this behavior to be the preference to meet face to face. 44.2% give the reason as the unavailability of the required services online whereas another 36.1% report delays of feedback. Surveys indicate that enterprises need more sophisticated services and the demand for e-government services becomes more developed in the case of business transactions with the public.

Another interesting finding is that a great majority of households and enterprises which do not prefer to use e-government applications report that they are motivated to meet face to face rather than making transactions online. This may be regarded as an indication of a cultural impact on the e-government applications since Turkish people are generally known to be socializers. However since broadband technologies have also transformed the social space and socialization process itself, the Turkish society seems to be in a period of continuing adoption to ICTs and e-government applications.

34

Hence the chief policy challenge on the demand side is to provide and maintain good skill creation mechanisms and programs for increased ICT adoption by households and enterprises. Education policy should play a significant role through formal education and lifelong learning programs.

3.2 Electronic commerce

The abilities of enterprises to acquire, generate, and manage knowledge have become a main driver of economic growth. To do this, the adoption of ICT by business and its widespread and efficient use in enterprises has crucial importance. Electronic commerce (e-commerce)

58.7 55.2

76.7

84.8

0

10

20

30

40

50

60

70

80

90

100

10+ 10-49 50-249 250+

Figure 3-3: Ownership of web sites, by employment size of enterprises, 2009

(Source: State Planning Organization Information Society Statistics, 2010)

16,4

19,7 17,4

21,6

18,6

10,4 13,0

11,7

4,7

10,1

28,8

10,1

3,8

10,1

0

5

10

15

20

25

30

Motion picture and video activities,

radio and television

Real estate, renting and business

services

Transport, storage and communication

Hotels, motels, camping sites

Wholesale and retail trade, repair of motor vehicles, motorcycles and

personel and household goods

Construction Manufacturing Industry

p e rc e n t

Placing order

Receiving Order

Figure 3-4 E-commerce activities of enterprises by sector of activity, 2007

35

applications have been at forefront of this transformation process.

Within the group of enterprises employing 250 or more employees, 85% have established their own web sites in 2009 (Figure 3-3). In the group of enterprises with 50-249 employees, web site ownership ratio decreases to 76 % and in the group of 10-49 employees size, it further declines to 55%. The adoption of e-commerce is positively correlated to financial and managerial capacity of business units and Turkish enterprises have a growing interest in electronic commerce.

According to survey data by TURKSTAT, 77.9 % of enterprises have provided product catalogs and price lists through web sites. It is the leading form of e-commerce activity exercised by enterprises. In 2008 44.8 % of enterprises reported to engage in marketing of their products and services through internet and 28.4 % responded of giving after sales support services. The share of enterprises conducting online purchase order, reservation and registration transactions is 16.3 % as of 2008. As e-commerce activities get more sophisticated ability of adoption by enterprises is observed to decrease significantly (Figure 3-4).

E-commerce growth and adoption among Turkish enterprises is uneven and subject to significant variation among different sectors of economic activity (Figure 3-4). The leading

economic sector to adopt e-commerce has been tourism services. 21.6 % of hotels, motels and camping sites reported to place orders while 28.8 % of the same group reported to receive orders via internet. Real estate, motion picture and video activities, wholesale and retail trade and transport and communication sectors are among leading sectors in terms of enhanced e- commerce applications. More industry oriented activities such as construction and manufacturing, by contrast reported somewhat poorer adoption rates of e-commerce than service based sectors.

Banking and financial sector is an interesting example where e-commerce systems have been developed and embedded into business models rapidly. Sayar and Wolfe (2007) found that Turkish banks offer a wider range of services from their internet branches compared to British banks, despite the fact that the UK has a more favorable environment for internet banking in terms of the level of sophistication of its banking sector and technological infrastructure.31

While in most of the service-based industries the main motivation is to augment sales and increased market penetration, in the Turkish banking sector the main incentive is lower costs rather than increased sales. Analysts calculate

31 Ceren Sayar, Simon Wolfe, (2007) "Internet banking market performance: Turkey versus the UK", International Journal of Bank Marketing, Vol. 25 Iss: 3, pp.122 - 141

0,0

10,0

20,0

30,0

40,0

50,0

60,0

70,0

80,0

Other Increased turnover Lower transaction costs

Penetration into new markets, increased

sales potential

percent

Figure 3-5: The benefits reported through e-commerce, 2008

36

that an Internet transaction costs a Turkish bank around 5% of the cost of the same transaction in a traditional bank branch.32 Most of the larger commercial banks in Turkey offer Internet- based banking services. Other companies offer online services. The most active are airlines and retailers of books and electrical goods, largest supermarket chains also offer online services.

The data on benefits reported through e- commerce activities can also provide a good example in terms of assessing the relative degree of e-commerce adoption by Turkish enterprises (Figure 3-5). A great majority (70 %) of surveyed firms have responded that penetration into new markets and increased sales potential were the foremost benefits after engaging in e-commerce activities. Lower transaction costs and increased turnover with a 60 % response share each are among positive effects perceived by these firms. This clearly shows that enterprises are getting aware of the benefits offered by e-commerce applications.

There are also a number of challenges that limit increased adoption of e-commerce. According to the Survey data in 2008 enterprises reported to face following problems: Customers reluctance to use e-commerce applications (56.4 %), difficulty to adopt products and services into internet based commerce systems (55%), security issues during payment (48%), uncertainties in legal infrastructure of e-commerce (45.5%), technical issues and problems (42.4%), difficulties to reorganize business models (42.2%), and negative experiences on electronic sales (14.2%).

In other parts of the Survey, enterprises gave increased attention on security issues related to e-commerce applications. More than 17 % reported to experience some security problems during their electronic activities in Internet. There are also some indications that the share of enterprises having some problem on security issues is on the rise. Enterprises started to introduce necessary measures to overcome these challenges on two fronts:

 Having installed secure servers (94.9% of the group reported to have at least one server) and software (37.9 % reported to have installed firewalls).

32 Source: Economist Intelligence Unit, news at http://www.ebusinessforum.com/index.asp?layout=rich_st ory&doc_id=11173&title=Turkey%3A+Overview+of+e- commerce&channelid=4&categoryid=31

 Having employed information specialists (48.6% of the group with 250 and more employees reported to employ information specialists) or even creating their own divisions for this purpose.

E-commerce is growing rapidly in both demand and supply sides. However the degree of adoption differs among enterprises significantly and the size of firms plays a major role in enhancing e-commerce applications. Larger firms with expanded financial and managerial capacity are in favorable condition to develop e- commerce solutions and online services. Smaller enterprises with fewer resources are finding difficult to get involved in these next generation commerce activities.

Government policies should address more rapid and balanced development of e-commerce applications in the enterprises sector. KOSGEB, the agency in charge of assisting small and medium-sized enterprises (SMEs), should increase its efforts and design additional programs to address this challenge. In the past KOSGEB established 55 Internet cafes with instructors to assist SMEs in using the Internet. But this is not sufficient; much more could be accomplished by, for example, working with universities to provide courses in using the Internet, as well as providing counseling and training services in e-commerce to SMEs.

3.3 Other Applications: The Rise of Social Media

The development of social media has been rapid in Turkey. While the country is lagging behind in terms of broadband penetration than more advanced economies, Turkey’s youthful population structure makes it a very vibrant market for most social media applications and services. In Turkey, where 70 percent of the population is under 35, online communications is very appealing to young people.

With more than 26 million users, Turkey is one of the 15 largest Internet populations in the world and roughly the size of the Spanish, Italian or Canadian markets. Turkey’s mobile penetration is larger than Internet penetration, which means that people increasingly access social network applications from mobile phones.

37

Turkey has the fourth highest Facebook population in the world.33 According to one source, Istanbul has been ranked as the city with the second largest number of Facebook accounts in the world with over 9.6 million people. Following Facebook, the most popular social networking services are Mynet Eksenim, Netlog and MySpace.34 Twitter and personal blogs have also gained in popularity.

Turkey’s own social media content is growing day by day. According to a recent study conducted by a consulting firm, Turkish web sites are getting popular and increasingly diversified. Figure 3- presents a snapshot of the social media landscape in Turkey categorized around different activities (e.g., photo, video, document and music sharing, news services, e-

33 http://www.ictdata.org/2011/03/monaco-tops-in- facebook-penetration.html 34 comScore, Inc. 2009. “Instant Messaging Accounts for More Than 25 Percent of Total Time Spent Online in Turkey, Followed by Social Networking at 10 Percent”, July 16. http://www.comscore.com/Press_Events/ Press_Releases/2009/7/Turkey_Instant_Messaging_and_S ocial_Networking

commerce, business networking, social networking, social bookmarking, blogging and micro blogging, customer tracking, etc.). The figure demonstrates that social media are increasingly popular in society and Turkish content creation is at its emerging stage except for news services where there is a natural tendency to utilize the mother language. In the electronic commerce services segment the leading companies are domestic with uniquely designed and developed web applications.

Istanbul recently hosted the World Social Media Forum providing a meeting place for sharing global experiences and knowledge in this regard.35 Turkish experience with social media applications presents a useful example for other countries with similar economic and cultural backgrounds.

35 http://www.worldsocialforum.info/news/european- social-forum-turkey-make-them-pay-for-their-crisis

Table 3-2: Top cities on Facebook (Source: Socialbakers.com)

38

2 Figure 3-6: The Turkish social media landscape, 2010

(Source: Ticktock Boom at www.ttboom.com/index.html)

39

4 Usage and Diffusion

Since the Internet can be viewed as a public good necessary for economic and social development, governments should take part in fostering nationwide internet access. A competitive private sector needs to develop the capability to use e-commerce. E-commerce will develop slowly if there are few Internet users, and users will not develop if charges for connections are high, coverage low, and the Internet provides negligible content and services. This section draws analysis on Internet usage and diffusion by people and enterprises. Discussion herein progresses around issues of how individuals and enterprises in Turkey use the Internet and differences in usage as to gender, age, and where people live – the so- called digital divide.

4.1 People

Household broadband access is relatively low in Turkey compared to European Union members. In 2010 more than three out of five households living in EU countries had a broadband connection while in Turkey the ratio was 34% (Figure 4-1). This relatively low share of broadband access represents a barrier to the take-up of online government services as well as the development of e-commerce.

Significant differences exist within Turkey:

between urban and rural areas, men and women, young people and individuals aged 55-74, and individuals with higher education and people without higher education. In 2009 urban areas have 32.5% broadband household penetration whereas rural areas have only 11%. However it is observed that the rural broadband connections are growing at a superior rate than that of urban areas which means the digital divide is getting smaller (another factor is because of the restrained growth of broadband infrastructure nationwide and particularly in largely populated urban centers).

The Internet usage proportion is the highest in the 16 – 24 age group (59.4 %) according to the results of the survey applied to the 16 – 74 age groups. The usage of Internet by women (46%) significantly lags behind the usage of men (74.1%) in 16 – 24 age group and all age groups. Consequently, it is important for the government to seek to increase the proportion of individuals using the Internet through better geographical coverage and by reducing the cost of accessing and using the internet.

The level of education and Internet usage are highly correlated. Internet usage of individuals increases in parallel with their education levels. Internet usage by women with college,

0

10

20

30

40

50

60

2004 2005 2007 2008 2009

EU (27) Turkey

Figure 4-1: Household broadband penetration, as percent (Source: EUROSTAT)

40

university, and higher education (85.8%) is quite close to the usage levels of men with the same educational background (89%). Those using the Internet most, according to employment status, are students (88.2%), employers (66.1%), waged/salaried employees (56.8%), and the unemployed (41.6%). 36

A great majority of users (68.2 % for females and 51.6 % for males) report using the Internet at home. The proportion of home usage by females is significantly surpassing the usage by males. The workplace usage is among the top forms of Internet connection activities among households and more interestingly Internet cafes continue to offer a wide mean for internet in particular for male groups (31.8%). Individuals also report that they are using internet during social activities and visits. (Figure 4-2).

Though home access have greatly increased recently Turkish people still account for a different structure of access type by place. Internet cafe use is much higher and public libraries are not used at all. This finding shows that Internet access from home or workplace is not always the most revealing measures of actual individual internet access for comparison between countries.

36 The data come from the Survey results by TURKSAT and figures are for 2009.

Survey data indicate basic Internet activities such as sending email (73%), reading online news (59%), surfing the web and making social connections (64%) are among the leading forms of Internet usage. Usage rates for more sophisticated Internet activities are still relatively low: Internet banking (17%), looking for a job (10%), online education (6%) and selling goods or services (4%).

Analyses show that Turkey faces a number of challenges in Internet usage. First and foremost improving Internet access and broadband development should be a priority on the supply side. Turkey needs to promote a high degree of broadband coverage using appropriate technologies. Initiatives to reduce the digital divide in Turkey should also include expanding the number of computers and computer classes in primary and secondary schools, and providing increased access to the Internet through community use of school computers. Turkey has demonstrated an ambitious case in trying to close the divide in digital literacy between urban and rural areas by means of education policy. The Ministry of National Education (MNE) and local governments have taken part in this collective effort to foster education in digital skills. The number of students per computer is 30.8 in primary schools and 25.1 in secondary schools nationwide. 27,999 information technology (IT) laboratories were established.

0,0

10,0

20,0

30,0

40,0

50,0

60,0

70,0

80,0

Other During social events or visits

Internet cafes Education or training

institutions

Workplace Home

percent

Female Male

Figure 4-2: Internet usage by place of connection, male and female, 2009

41

1,850 Public Internet Access Points (PIAP), to provide ICT access and ICT competency to citizens, have been completed by the MNE.

Since the early take up of information society and e-transformation initiatives Turkey has launched a motivated program for teaching computer and information literacy in primary and secondary schools, but more is needed. Turkey should provide means to increase computer and information literacy for people who are not in school. Online distance learning programs and new incentives and programs for Internet cafes to provide a social training base for households living in rural areas may constitute the essence of an effective policy mix. The government should also find means to further motivate citizens to use the Internet by demonstrating and informing citizens about the benefits and advantages of accessing information and government services online.

4.2 Business

The business sector has shown somewhat more success in ICT adoption than households. The Turkish business sector has become heavily computerized in the last ten years. Computer usage has become a standard nationwide even in micro sized business units. In 2009 computer penetration rates are as follows: In enterprises with 250 or more employees 99.3%, in enterprises with 50-249 employees 97.7%, in enterprises 10-49 employees 89.5% and in enterprises with 10 or more employees 90.7%.

Similarly Internet access by the business sector has grown explosively and reached a level almost identical to that of computer penetration rates. In January 2010, 90.9% of enterprises with at least 10 persons employed used the Internet. This rate was 98.4% of enterprises with 250 or more employees, while enterprises with 50-249 employees was 96.9% and enterprises with 10-49

Turkey

Urban

Rural

Sending/receiving e-mails 72.8 73.9 66.9

Posting messages to chat sites, blogs, news groups or

online discussion forum, use of instant messaging

64.2 65.0 60.4

Reading or downloading online news/

newspapers/news magazines

58.8 58.8 58.9

Finding information about goods or services 55.7 57.7 45.6

Playing or downloading games, images, films or music 51.2 51.1 51.5

Seeking health-related information (e.g. injury,

disease, nutrition, etc.)

47.3 48.6 40.8

Telephoning over the Internet/ video calls (via web

cam) over the Internet

47.1 48.3 41.0

Listening the web radios or watching web television 41.1 42.1 36.1

Consulting the Internet with the purpose of learning 37.1 38.1 32.1

Uploading self-created content (text, images, photos,

videos, music, etc.) to any websites to be shared

30.3 31.3 25.1

Looking for information about education, training or

course offers

26.4 27.0 23.5

Using services related to travel and accommodation 22.6 24.9 11.1

Internet banking 16.8 17.8 11.3

Downloading software (other than games software) 14.8 15.6 10.5

Looking for a job or sending a job application 10.2 10.5 8.9

Doing an online course (in any subject) 6.3 6.7 4.5

Selling of goods or services, e.g. via auctions 4.1 4.3 3.1

Table 4-1: Internet activities of individuals who have accessed the Internet in the last 3 months,

by private purposes, %, 2010

(Source: TURKSTAT, Results of the ICT Usage in Households and by individuals, 2010)

42

employees was 89.7%. 88.8% of enterprises had Internet access in January 2009.

The business sector has demonstrated a major achievement in adopting and utilizing ICTs and broadband. Turkish enterprises have high rate of broadband usage and as compared to EU 27 countries Turkey slightly surpasses the EU averages (Figure 4-3). In terms of connection type the most frequently used access technology in enterprises was DSL with a 94.6% proportion in 2009. However 19.5% of enterprises reported

to have access via traditional dial up connections also. Mobile technologies accounted for 13.5 % and cable technology had a proportion of 10.2%.

Figure 4-4. portrays how firms are integrating internet usage into internal business process. Two points can be drawn from data: The trends are similar between years and a major change is not observed. Secondly, business units are utilizing online connections most for getting financial services (76.3%) and assessing relevant market trends (78%). Diffusion of internet based

0,0

10,0

20,0

30,0

40,0

50,0

60,0

70,0

80,0

90,0

2007 2008 2009

Turkey EU 27

Figure 4-3. Broadband usage indicators by enterprises, as percent

77,5 77,6 76,3

32,6 33,4 31,6

75,9 77,0 78,0

2007 2008 2009

as percent Financial services Education and training Market trends

Figure 4-4: Internet usage in enterprises by purpose, 2007-2009

43

applications into education and training services accounts for lower but still a significant proportion (31.6%) in 2009.

Lack of sufficient number of candidates with the required qualifications and job experience (39.4%) stood out as the main difficulty faced by enterprises when employing information specialists in 2007. While 68.3% of enterprises reported that the salary expectation of specialists were high, a greater proportion of 74.4% stated that the work experience and competency of specialists as being insufficient. Similarly even a greater proportion of 84.5% reported there was very limited number of candidates with required qualifications in 2007. 37 According to the Survey, enterprises, with a proportion of 8.9 % in 2007, had started to run internal training programs to address the skill gap in ICT technologies.

Though the broadband penetration has been somewhat limited compared to more advanced countries, Turkish businesses have adopted a high level of Internet usage, motivated to a large degree by the desire to use government e- services and penetration into new markets. One exception to this fact is the banking sector where the main motivation is cutting costs rather than demand aggregation.

37 Source: SPO (2010), Information Society Statistics, Ankara.

However, this has not resulted in a fully competent business sector with a high level of broadband usage and applications within integral business processes. A large share of small and medium-sized enterprises (SMEs) selling in local markets with few resources find it difficult to learn and develop new marketing channels. These business units are slow to adapt broadband technologies. Industrial structure is increasingly fragmented and innovative collaboration frameworks and programs are needed. This is where the public policy should engage in effective mechanisms and incentive structures to enable a favorable environment for technological take up.

Given that broadband technologies have a very strong stimulus for economic growth, the government might consider means to stimulate business usage through infrastructure development, as well as providing content and e- services over the Internet through various mechanisms. Requiring firms to interact electronically with e-government services should increase private sector use of the Internet. Additionally, increased broadband availability will allow faster development of Internet usage and serve as an important factor in attracting foreign domestic investment in the Turkish economy. Stronger efforts to foster human resource development and to address the ICT skill gap are needed also.

44

5 Conclusions

5.1 The Likely Impacts of High Bandwidth Networks on the Turkish Economy

According to the National Broadband Vision Study of Turkey,38 through fostering broadband development the Turkish economy could gain US$ 4.9-10 billion extra value added each year thereby boosting its economic growth by 0.8-1.7 percentage points. This economic momentum enabled by an enhanced broadband ecosystem would bring 180,000-380,000 new jobs and provide new income opportunities. This implies that employment creation in new broadband- related activities would eventually compensate for job loss due to process optimization and structural displacements.

Through help of a toolkit39 the Broadband National Vision Study reports to find a significant growth impact for the Turkish economy fueled by broadband enhancing public policies. In the baseline growth scenario Turkish economy would take off by an annual rate of 5.0 per cent as depicted in Figure 5-1, and it would reach the US$ 1,216 billion GDP level by the end of 2023.

Under the broadband enhancing growth scenario in which a set of active government programs to boost broadband impact are introduced, the economic growth forecast is much faster, with an additional 2.3 percentage points, reaching to 7.3% on annual basis. Government policies targeting enhanced broadband infrastructure and ecosystem will boost economic growth significantly. By 2023 the Turkish economy would have a GDP at

38 Turkish Prime Minister’s Office (2011), From Silk Road to Broadband: Enabling Economic Growth and Competitiveness, Ankara. 39 Broadband Dynamic Value Assessment (BDVA) is a toolkit that fills a niche between the macroeconomic models that link broadband to aggregate economic growth and models that take a wholly microeconomic approach, measuring specific benefits of broadband by sector or type of usage. The BDVA focuses on broadband implementation objectives of public policy that link broadband penetration rates to “pillars”, which in turn, drive GDP growth. The pillars are essentially proxies for broadband implementation objectives. In this manner, the model user can identify which of these pillars have the greatest impact on GDP growth and use this information to formulate a national policy mix for achieving the greatest benefits from wider broadband usage

USD 1,652 billion rather than USD 1,216 billion of the baseline scenario. This is a GDP level 36% greater than the baseline. It may accurately be called the broadband effect.

The broadband effect encompasses two distinctive categories: Industry benefits (measured as USD 289 billion during 2010-2023) comprising of the overall broadband productivity impacts within and beyond the ICT industry; and pillar effects (measured as USD 147 billion during 2010-2023) which might directly be associated with broadband enhancing policies of the government. Since industry benefits are nearly two times larger than pillar benefits for Turkish economy, broadband enhancing policies should trigger a multiplier effect within the economy and their indirect impacts could potentially outpace the direct effects.

The study also analyzes detailed impacts of enhanced broadband polices on different economic sectors, the so called pillar effects. Pillar effects can be decomposed into different sectors as follows:

- 7 per cent of total pillar effects will come from revitalized small/medium business;

- 6 per cent contribution will arrive from improved government services;

- 4 per cent to come from the benefits of broadband-enabled education;

16 per cent to come from a more attractive business environment;

- 67 per cent to come from other sectors, as well as multi-factor productivity and the benefits of collaboration across the economy.

Analyses indicate forecast benefits from increased ICT adoption by the SME sector will be large and e-government applications would provide a significant amount of contribution through cutting costs and improving productivity. Another important policy area that should be focused is education. The main benefit seems to be improvements in the general business

45

environment. Under the broadband enhanced scenario policies to foster the business environment, 16 % of total pillar effects will come from improvements in the business framework alone. The largest one is the multiplier effect. Broadband adoption will bring benefits diffusing beyond the ICT sector through channels of multifactor productivity.

5.2 Factors Contributing to Turkey’s Broadband Development

E-government initiatives have been a major driving force for development of the broadband ecosystem. Turkey has implemented a highly motivated e government program that created the necessary demand for enterprises in the ICT sector and motivated citizens for increased Internet usage. In this regard Turkey has demonstrated an interesting example that countries with relatively lower level of Internet penetration and usage ratios can exploit e- government programs as a means to ignite development of the broadband ecosystem.

Ensuring a shared vision among political leaders and technocrats has also been an important factor in pushing e-government programs. Since the coming of the new national government into

power in 2002, political leaders saw e- government as a central instrument that would support public reforms and larger changes in the political system. A central organizational structure was formulated to develop strategies and put public money into the pipeline for a set of strategically important projects with high value and a high transaction volume.

The EU candidacy of Turkey and the membership negotiation process has provided a good opportunity for the country to reform its ICT and broadband legal and regulative frameworks. This process helped bureaucrats and top-level managers get closer to European institutions, policies and initiatives, and market developments in these sectors.

The high tempo growth of Turkish economy in the last decade is another factor supporting in the Internet revolution. The new national government has implemented ambitious market oriented reforms complemented with a proactive foreign policy which resulted in large sums of overseas capital flowing into the country. Communications, software and hardware segments of ICT industries have expanded rapidly.

5.2.1 Lessons learned

75

100

125

150

175

200

225

250

2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

Base Line

Broadband Enhanced

5,0 %

7,3 %

Figure 5-1: Base line and broadband enhanced growth scenarios, Real GDP2009=100 (

46

The Connectivity Scorecard is a composite index to assess the relative development level of ICTs (Figure 5-2).40 Turkey was found to have a robust consumer infrastructure due to its high mobile and PSTN penetration and also the recent progress in broadband networks. It also scores strong on consumer usage and skills metrics such as frequent Internet usage and uptake of voice services. According to the Scorecard the country’s business infrastructure is above average, with high penetration of secure Internet servers, and substantial business investments in ICT. Turkey also impresses on government-related metrics such as government spending on ICT and provision of government services online.

However the Scorecard points out that Turkey ranks somewhat poorly in business usage and

40 The Connectivity Scorecard is the first index to examine the quality and quantity of ICT usage and infrastructure and to link it to a country’s social and economic prosperity. 25 ‘Resource and Efficiency Driven’ and 25 ‘Innovation Driven’ economies are studied (as defined by the World Economic Forum (WEF)) in this first phase. The first step taken is to divide the economy into 3 ‘pillars’, business, consumer and government and assign weights to these pillars. The greatest weighting is given to the business pillar since it is a key contributor to productivity growth. For each component of the scorecard countries are benchmarked against the best-in-class in their tier. Low scores reflect gaps in a country’s infrastructure, usage or both. For more information see “Connectivity Scorecard” at: http://www.connectivityscorecard.org/methodology/

skills, where its score is considerably lower than one might expect given the state of its business infrastructure. Turkey has made considerable progress in developing a robust ICT infrastructure but its weaker usage and skills scores indicate that the country is yet to fully realize its benefits. Therefore, it needs to place greater effort on ensuring the adoption and diffusion of ICTs to leverage competitive power.

Turkey faces a strategic opportunity in terms of broadband infrastructure development. The majority of the national broadband network is based on slower speed connections which do not optimally support advanced applications of next generation connectivity such as e-education, e-health, etc. There are signals suggesting the current broadband network has become somewhat overloaded.

To enhance a rapid take up of high bandwidth broadband networks the government should play a more effective and active role. First and foremost is ensuring effective regulation. The public stake in the incumbent operator should not prevent authorities from enforcing rules for fostering competition in the broadband market. Secondly, Turkey should design a complete and integrated broadband strategy for coordinating various individual pieces in the same direction.

To enable increased ICT and broadband adoption by businesses the government should exercise greater efforts to design effective

0

0,2

0,4

0,6

0,8

1

Consumer (Infrastructure)

Consumer (Usage&Skills)

Business (Infrastructure)

Business (Usage&Skills)

Government (Infrastructure)

Government (Usage&Skills)

Turkey

Best Performing Country

Figure 5-2: Turkey connectivity performance by scorecard component, 2010

(Source: www.connectivityscorecard.org/countries/turkey)

47

programs and incentives. In particular skill gaps should be identified and adequately addressed with participation of private initiatives.

The lack of a suitable national accounting framework for more detailed analysis hinders international benchmarking in most emerging

policy areas notably ICTs and innovation. Turkey should take measures to address this problem and ensure reliable and timely indicators are developed and made available to the public.

48

6 References

Atiyas, İzak. 2010. Regulation and competition in the Turkish telecommunications industry: an update. In The Political Economy of Regulation in Turkey. New York: Springer. http://research.sabanciuniv.edu/14658/.

Information and Communication Technologies Authority. 2011. Annual Report 2010. Ankara: ICTA. http://www.btk.gov.tr/eng/basineng.html.

McKinsey & Company. 2009. Mobile broadband for the masses: Regulatory levers to make it happen. http://www.mckinsey.com/en/Client_Service/Telecommunications/Latest_thinking/Mobile_broadband _for_the_masses.aspx.

McKinsey Global Institute. 2003. Turkey: Making the productivity and growth breakthrough. http://www.mckinsey.com/mgi/publications/turkey/index.asp.

Organisation for Economic Co-operation and Development. 2001. “The Hidden Threat to E-Government: Avoiding Large Government IT Failures.” Public Management Policy Brief, March. http://www.oecd.org/document/36/0,3746,en_2649_34225_1916324_1_1_1_1,00.html.

———. 2007. e-Government Studies: Turkey 2007. Paris: OECD. http://www.oecdbookshop.org/oecd/display.asp?sf1=identifiers&st1=9789264028449.

———. 2008. Broadband Growth and Policies in OECD Countries. Paris: OECD. http://www.oecd.org/document/1/0,3746,en_2649_34225_40931201_1_1_1_1,00.html.

———. 2009. Communications Outlook 2009. Paris: OECD. http://www.oecd.org/document/44/0,3746,en_2649_33757_43435308_1_1_1_1,00.html.

———. 2010a. Information Technology Outlook 2010. Paris: OECD. http://www.oecd.org/document/20/0,3343,en_2649_33757_41892820_1_1_1_1,00.html.

———. 2010b. Factbook 2010: Economic, Environmental and Social Statistics. Paris: OECD. http://www.oecdbookshop.org/oecd/display.asp?K=5KMLHLH8X1ZN&lang=EN&sort=sort_date%2 Fd&stem=true&sf1=Title&st1=factbook&sf3=SubjectCode&st4=not+E4+or+E5+or+P5&sf4=SubVer sionCode&ds=factbook%3B+All+Subjects%3B+&m=2&dc=22&plang=en.

Point Topic. 2011. World Broadband Statistics: Short report Q4 2010. http://point-topic.com/dslanalysis.php.

Prime Minister’s Office. 2011. From Silk Road to Broadband: Enabling Economic Growth and Competitiveness.

Sayar, Ceren, and Simon Wolfe. 2007. “Internet banking market performance: Turkey versus the UK.” International Journal of Bank Marketing 25 (3): 122-141. doi:10.1108/02652320710739841. http://www.emeraldinsight.com/10.1108/02652320710739841.

State Planning Organization. 2010. Information Society Statistics 2010.

The World Bank. 2009. Information and Communications for Development 2009: Extending Reach and Increasing Impact. http://go.worldbank.org/NATLOH7HV0.

———. 2010. Turkey - Investment climate assessment : from crisis to private sector led growth. http://go.worldbank.org/QLE6A8NI60.

———. 2011. Prospects for the Global Economy. http://go.worldbank.org/RWLJPPA0D0.

Turkish Statistical Institute. 2009. Annual Industry and Service Statistics. http://www.turkstat.gov.tr/jsp/satis/common/kitaplar_tem.jsp?pageNum=4.

UNCTAD. 2008. World Investment Prospects Survey 2008-2010. http://www.unctad.org/Templates/Page.asp?intItemID=5171&lang=1.

World Economic Forum. 2010. The Global Competitiveness Report 2010-2011. http://www.weforum.org/reports.

49

5.1 Online Resources

www.carelink.se/en/mission/sjunet

http://www.dtc.umn.edu/mints/home.php

http://econ.worldbank.org/

www.oecd.org

The Central Bank of the Republic of Turkey at www.tcmb.gov.tr

The Undersecretariat of the Treasury at www.treasury.gov.tr

Invest in Turkey at http://www.invest.gov.tr

http://www.invest.gov.tr/en-US/sectors/Pages/ICT.aspx

Export Promotion Center (IGEME) at www.igeme.org.tr

http://www.turkstat.gov.tr/PreHaberBultenleri.do?id=8450.

http://www.turksat.com.tr/english/v2/cable-services

http://www.eutelsat.com/products/broadcast-audience.html

Economist Intelligence Unit, news at

http://www.ebusinessforum.com/index.asp?layout=rich_story&doc_id=11173&title=Turkey%3A+Over view+of+e-commerce&channelid=4&categoryid=31

Ticktock Boom at http://www.ttboom.com/index.html

http://www.connectivityscorecard.org/methodology/

http://www.socialmedia-forum.com/

50

About infoDev

infoDev is a global development financing program among

international development agencies, coordinated and served by an

expert Secretariat housed at the World Bank Group, one of its key

donors and founders. It acts as a neutral convener of dialogue—and

as a coordinator of joint action among bilateral and multilateral

donors—supporting global sharing of information on ICT for

development (ICT4D), and helping to reduce duplication of efforts

and investments. infoDev also forms partnerships with public and

private sector organizations who are innovators in the field of ICT4D.

infoDev is housed in the Financial and Private Sector Development

(FPD) Vice Presidency of the World Bank Group.

For additional information about this study or more general information

on infoDev, please visit www.infodev.org/publications.

AxonVodafoneTurkey.pdf

º

Turkey’s digital future: Vision for the Broadband Plan

Independent report prepared by Axon Consulting

January 2017

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 ii

Contents

Executive summary ............................................................. 1

Introduction ....................................................................... 4

Turkey’s digital future: Vision for the Broadband Plan in

Turkey ...................................................................................... 7

3.1. Broadband as a national strategic priority ...................................................7

3.2. Choice of a future-proof technology for Turkey .......................................... 10

3.3. Turkey’s fixed broadband opportunity: a market with high growth potential .. 13

3.4. Understanding Turkey’s Broadband Gap .................................................... 18

3.5. Crafting a Broadband Plan for Turkey ....................................................... 24

3.5.1. The Open Access Network (OAN) scheme ............................................... 24

3.5.2. Investment needs of the open access network scenarios in Turkey ............ 26

3.5.3. Business case analysis of the OAN scenarios ........................................... 28

3.5.4. Role of the government to the expansion of FTTH/B networks in Turkey ..... 36

Conclusions ...................................................................... 41

Annex A: Open access broadband networks and government

support in the international experience ....................................... 47

Annex B: Principles of the business case model for Turkey ............. 51

Annex C: Sources of reference for the figures of coverage, take-up

and prices ............................................................................... 57

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 1

Executive summary

In recent years, international policy makers have been observing the power of

broadband networks as enablers of economic and social development, hence

promoting their deployment. Public initiatives focused on the development of fixed

broadband markets move through successive phases: Deployment, Adoption and

Integration. Initially, policy makers tend to focus on the national footprint for

broadband network availability. In this first phase, monitoring focuses on broadband

network deployment to make services as widely available as possible, using standard

indicators of coverage, capacity, take-up and price. The current state of the fixed

broadband market in Turkey, as explained in this Report, would be classified under

this phase.

In the Deployment phase, several technologies exist today that may be considered

for delivering broadband connectivity to households (e.g. ADSL 2+, 3G, FTTH/B,

WLAN/WIMAX). Among these technologies, FTTH/B is technically the most advanced

due to a number of unique characteristics, proving it to be a future-proof technology.

Our study1 starts with the presumption – and the vision – that expanding FTTH/B

coverage to a significantly larger part of the population than its current state should

be the goal of Turkey’s Broadband Plan. Coupled with a properly designed investment

undertaking of a one-time nature, the whole initiative would avail an infrastructure

to the nation that will be usable for many decades.

The current consumption profile of fixed broadband users in Turkey indicate

demonstrated growth over the recent years. We can observe in general that the

Turkish consumers will not be staying behind worldwide usage trends in the years to

come. Growth of FTTH/B connections in servicing the growing demand, has however

demonstrated a relatively slow growth over the recent years. Turkey stands at a

right point in time to consider and realize the vision of expanding FTTH/B coverage

to a significantly large part of the population. The challenge of this vision lies in the

fact that wide scale FTTH/B expansion projects require significant investments and

consequently a careful analysis of the business case for the medium to long-term

recovery of those investments.

1 This study was undertaken by Axon Partners Group Consulting S.L. (‘Axon Consulting’) to provide an independent report (‘Report’) regarding the expected Broadband Plan in Turkey. It has been commissioned by Vodafone Telekomunikasyon A.S (‘Vodafone’).

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 2

Expansion of FTTH/B networks in Turkey would imply taking forward the household

coverage ratio of this technology from its level of 23% in Q4 2015. The initiative

would not only bring the coverage levels at par the levels observed in leading

countries, but also would attempt to take it forward to ensure maximum adoption

and take-up by the consumers.

We recommend that the deployment of an FTTH/B network through an open access

network (OAN) scheme will enable the objective of a network with a sufficiently

large footprint in Turkey. An open access network would cater for the needs of the

Turkish market by:

 Avoiding duplication of resources through multiple service providers

investing in FTTH/B access networks. It would instead enable focusing on the

timely expansion of service to reach the widest level of FTTH/B coverage

through optimized roll-out and investment plans (hence meeting the supply

objective).

 Creating the environment for effective and non-discriminatory access to

network elements by multiple retail providers, promoting an environment of

affordable prices and volume gains by retail policies that enhance take-up

of FTTH/B services (hence meeting the demand objective).

Our economic analysis on the expansion of FTTH/B coverage in Turkey through an

OAN concludes the following:

Conclusion 1: Under the current circumstances, there is a risk that Turkey

will miss the opportunity associated to the high growth potential of the

Turkish fixed broadband market, with up to 15,16 million households being

left without access to FTTH/B networks

According to our study, under the current circumstances, i.e. if no action is taken,

only between 5 and 6,50 million households in Turkey will be covered over the next

five years by FTTH/B technology. This is an improvement – albeit a small one - from

the levels of coverage currently achieved (5 million households).

Our study has shown, however, that the expansion of FTTH/B in Turkey can go up to

11,26 million households by market-driven mechanisms under the right regulatory

and market conditions. A first conclusion from our study is therefore that keeping the

current course in terms of broadband policy would imply a very high cost for Turkey.

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 3

Conclusion 2: Promoting a friendly environment for commercially-driven

network roll-out would open up the investment opportunity and could lead

to an additional 535 million USD in additional investment and 4,76 million

additional households with FTTH/B coverage

The investments required to build a FTTH/B network in Turkey would be reduced by

between 31% to 45% in that case that existing civil infrastructure could be reused.

More importantly, introduction of agile and efficient civil infrastructure reutilization

measures would result in an increase (not a decrease) of investments by private

agents in FTTH/B networks. In fact, investments by private operators in FTTH/B

networks would increase by up to 535 million USD with the presence of civil

infrastructure reuse, leading to an increase in the number of households covered of

up to 4,76 million.

Conclusion 3: Moderate financial and regulatory support from the

government to an Open Access Network could help boost expansion and

provision of FTTH/B networks to all Turkish households living in cities of

over 100.000 inhabitants

Under the right conditions, the government could consider some sort of intervention

to help expand the current footprint of FTTH/B networks through an OAN. We observe

from our economic analysis that this would enable the ambitious yet achievable target

to provide coverage, over a 5-year period, to all households in Turkey located in

population centres with 100.000 thousand or more inhabitants (i.e. 76% coverage of

all households in Turkey).

The above conclusions, point out that, provided the right support, the OAN scheme

can enable the pooling of resources and the deployment of an efficient and shared

FTTH/B network in Turkey that reaches to 76% of all households. This would put

Turkey as a clear leader in the region in terms of FTTH/B coverage and at par with

some of the most advanced countries in the world.

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 4

Introduction

Vodafone Telekomunikasyon A.S (‘Vodafone’) commissioned Axon Partners Group

Consulting S.L. (‘Axon Consulting’) to provide an independent report (‘Report’)

regarding the expected Broadband Plan in Turkey.

This report, titled “Turkey’s digital future: Vision for the Broadband Plan”, presents

Axon Consulting’s independent vision and perspective on this subject.

About Axon Consulting

Axon Consulting (previously known as ‘SVP Advisors’), established in 2006 in Madrid

(Spain), is an international consulting firm that provides consulting services to an

international client base in the broad technology sector, with a strong focus on the

Telecommunications, Internet and Media industries.

Axon Consulting has earned a solid reputation as a world-class leader in developing

and applying state-of-the-art analytical frameworks for practical problem solving.

Axon Consulting’s approach combines the disciplines of economics, engineering,

marketing, econometrics, accounting, finance, and computer science with which we

achieve the strategic thinking required by our clients.

Axon Consulting is the consulting arm of Axon Partners Group, which resulted from

a merger between SVP Advisors, an international management consulting firm for

the electronic communication industries, and Axon Capital, an investment firm

specializing in investments in early-stage technology firms in the telecoms, IT and

media sectors.

Within the TMT industry, Axon Consulting offers relevant expertise in the fields of

strategy & innovation, regulation & public policy, and profitability & operations. This

includes supporting institutional bodies in developing techno-economic and pricing

models; setting regulatory policy and competition monitoring; and advising telecom

operators on business, commercial and regulatory strategy.

The exhibit below illustrates Axon Consulting’s principal areas of expertise.

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 5

Exhibit 2.1: Axon Consulting’s areas of expertise [Source: Axon Consulting]

Since its foundation, Axon Consulting has worked with top-class clients, including

telecoms operators such as Telecom Italia, Telefónica, Vodafone, Saudi Telecom

Group, Turkcell, ETB, UNE and Etisalat; media and internet companies such as

Babelgum and VirginPlay; telecoms regulators including OUR (Jamaica), SUTEL

(Costa Rica), IFT (Mexico), CRC (Colombia), AFTIC (Argentina), OCECPR (Cyprus),

TRA (UAE), ANAC (Cape Verde), ARTP (Senegal), EETT (Greece), CNMC (Spain),

AGCOM (Italy), ERST (Denmark), CITC (Saudi Arabia) and TRA (Oman); international

organizations such as the Inter-American Development Bank and the European

Commission; and financial institutions, including Banco Santander and Citigroup.

In total, Axon Consulting has worked on or carried out consulting assignments on

approximately 50 countries, as illustrated below:

Exhibit 2.2: Axon Consulting’s international footprint [Source: Axon Consulting]

Projects Undertaken

Axon Offices

Miami

Madrid (HQ)

Delhi

Bogotá

Istanbul

Mexico DF

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 6

Notice from Axon Consulting

In compiling this Report, Axon Consulting has used publicly available information

published either by the Information and Communication Technologies Authority

(‘ICTA’) or by third parties. Certain figures pertaining to the year 2016 (including but

not limited to the number of fixed broadband subscribers, fixed and mobile broadband

consumption) have been forecasted based on actual data available from ICTA for the

first three quarters of 2016.

Axon Consulting does not assume any responsibility on the accuracy, robustness and

completeness of such information available at the time of writing this Report.

No part of this Report may be copied or made available in any way to third parties

without Axon Consulting prior written consent.

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 7

Turkey’s digital future: Vision for

the Broadband Plan in Turkey

3.1. Broadband as a national strategic priority

Broadband access to Internet is an essential pillar of a successful

economy. The widespread availability and use of broadband services

bring economic and social benefits and are a major driver of global

growth and innovation.

In recent years, policy makers have been observing the power of

broadband networks as enablers of economic and social

development, hence promoting their deployment. Conveniently,

available broadband empowers people, nurtures technology and

service innovation, and triggers positive change in businesses as

well as in the society as a whole.

According to a study carried out by ITU2 in 2012, every 10%

increase in broadband coverage in a country market results in a

GDP3 increase between 0,25% and 1,38%, with the exact impact

depending on the country’s dynamics. A similar report4 covering 33

OECD countries by Ericsson concludes that doubling the broadband

speed in an economy leads to a GDP increase of 0,3%, equivalent

to 126 USD billion considering the total GDP of the OECD countries

in year 2011, the year covered by the report.

Despite the undisputed positive impacts of broadband coverage and

speed, the economics of broadband networks typically imply that a

compromise needs to be taken between the two. The priority placed

on each of these aspects determines the general policy direction in

countries and therefore needs to be carefully considered in the light

of national strategies.

The economics of broadband networks have typically led to the

above factors being viewed by policy makers in a mutually exclusive

context – i.e. either to focus on policy initiatives to achieve high

2 Report ‘Impact of broadband on the economy’, 2012. 3 Gross Domestic Product. 4 Report ‘Socioeconomic effects of broadband speed’, 2011.

“The widespread

availability and

use of broadband

services bring

economic and

social benefits

and are a major

driver of global

growth and

innovation”

“Every 10%

increase in

broadband

coverage in a

country market

results in a GDP

increase between

0,25% and

1,38%”

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 8

levels of ‘social coverage’ (affordable and universalized service

across the whole population), or to prioritize quality and

performance. None of these two approaches has traditionally paid

the needed attention to the ‘efficiency’ of the services, i.e. making

a high performance service widely available and at affordable prices.

However, current public initiatives are beginning to consider the

efficiency target for their broadband plans as an enabler to boost

not only the availability of the services across the whole population,

but also the affordability of such services. This phenomenon is

illustrated in the Exhibit 3.1 below:

Exhibit 3.1: Economics of broadband networks [Source: Axon Consulting]

Currently government initiatives focus on delivering high-speed and

quality broadband services to all premises in the country, while also

ensuring the affordability of such services. As such, specifying

technical characteristics alone turns out to be insufficient, if there

do not also exist requirements on affordability.

The exhibit below illustrates some example targets of National

Broadband Plans (NBP) from international practice in terms of their

focus on performance and availability.

Coverage

S p

e e d

HIGH- PERFORMANCE

NETWORKS

SOCIAL COVERAGE NETWORKS

HIGH- EFFICIENT NETWORKS

“Currently

government

initiatives focus

on delivering

high-speed and

quality

broadband

services to all

premises in the

country, while

also ensuring the

affordability of

such services”

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 9

Exhibit 3.2: National Broadband Plan objectives in international examples

[Source: Axon Consulting based on information published by Cullen

International] Note: Medium coverage: Below 50% / High coverage: 50% or above.

Medium speed: Below 70 Mbps / High speed: 70 Mbps or above.

Research conducted by the Broadband Commission suggests that

the introduction or adoption of public initiatives oriented to the

attainment of high-efficient networks is associated with 2,5% higher

fixed broadband penetration.

This result is consistent with a broadband line of action focusing

efforts across the industry in coordination with policy-makers,

emphasizing broadband as a national priority, and signalling

national commitment to the roll-out of broadband networks.

As illustrated in the exhibit below, public initiatives focused on the

development of fixed broadband markets move through successive

phases: Deployment, Adoption and Integration.

Exhibit 3.3: Phases of the application of governmental initiatives [Source:

Axon Consulting based on reports published by the ITU]

Medium High

Coverage

M e d

iu m

H ig

h

S p

e e d

AUT DNK FIN

FRA ITA KOR

LUX SLO SWE

AUS BEL CZE

EST POR

GRE HUN POL

SPA GBR

 Broadband access and capability building for effective use

 Stimulation of digital literacy and community access programs

PHASE 2: ADOPTION

 Broadband integration in economy and society

 Integration of multiple ICT services (e-health, e-governance, e- commerce) in a wide range of sectors

PHASE 3: INTEGRATION

 Broadband network deployment (availability)

 Broadband network uptake (affordability)

Focus

PHASE 1: DEPLOYMENT

Phase

“Research

conducted by the

Broadband

Commission

suggests that the

introduction or

adoption of public

initiatives

oriented to the

attainment of

high-efficient

networks is

associated with

2,5% higher fixed

broadband

penetration”

“Public initiatives

focused on the

development of

fixed broadband

markets move

through

successive

phases:

Deployment,

Adoption and

Integration”

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 10

Initially, governments tend to focus on the national footprint for

broadband network availability. In this first phase, monitoring

focuses on broadband network deployment to make services as

widely available as possible, using standard indicators of coverage,

capacity, take-up and price. The current state of the fixed

broadband market in Turkey, as explained later, would be classified

under this phase.

In their second phase, governmental initiatives include usage or

adoption factors such as digital literacy programmes or community

access projects. Basic telecom indicators remain important, but the

focus expands to include subscription rates, network resilience and

quality of the network.

Initiatives in the third phase tend to focus on evaluating the social,

economic and institutional uses of broadband underpinning the

wider use of ICT services in a range of sectors (including health,

government, education, commerce, public information and the

media). Indicators of speed, quality and reliability become more

important in this phase.

3.2. Choice of a future-proof technology for Turkey

Several technologies exist today that may be considered for

delivering broadband connectivity to households. These may

typically be viewed under two perspectives: connectivity (‘Basic’ and

‘Next generation’) and access medium (‘Fixed’ and ‘Wireless’), as

summarized in the table below:

Exhibit 3.4: Classification of broadband access technologies [Source:

Report ‘The broadband State aid rules explained’]

ADSL 2+

DOCSIS 2.0

FTTH/B

FTTC/VDSL

DOCSIS 3.0/HFC

3G (UMTS)

SATELLITE

Fixed Wireless

Access medium

N e x t -

g e n

e r a t io

n B

a s ic

C o

n n

e c t iv

it y

WLAN/WIMAX

LTE/LTE-A

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 11

Consistent with the categorization of fixed broadband in Exhibit 3.4,

the European Commission (EC) considers FTTH/B, Cable DOCSIS

3.0, VDSL and other superfast broadband technologies as Next

Generation Access (NGA) technologies.

Among these NGA technologies, FTTH/B is technically the most

advanced due to a number of unique characteristics:

 Capability to provide very high bandwidths, up to 1Gbps,

thereby able to accommodate future redefinitions of the

“broadband access” in terms of download speed.

 Lowest possible latency compared to other alternatives.

 Flexibility to upgrade bandwidth capacity over time with

minimal investments.

 Long physical lifetime of at least 40 years.

 Relatively lower energy consumption.

The above characteristics prove FTTH/B to be a future-proof

technology, assigning a truly ‘next generation’ perspective to the

significant initiative of bringing broadband to a large portion of the

population. This perspective is achieved not only through the

capability to provide high bandwidths, but also through the low

latency characteristic of this technology, which becomes

increasingly important with applications that require real time

interaction with remote servers (e.g. online storage, tele presence,

virtual reality).

In Exhibit 3.5 below we provide an overview of the connectivity

requirements of an illustrative group of consumer applications and

the corresponding technologies that can meet those requirements.

Advanced applications such as augmented reality, 4K resolution

video streaming, telemedicine, Internet-of-Things based

communications such as smart homes, emergency and response

systems, smart power grids all have requirements both for high

speed and high quality connections. FTTH is the only technology

poised to meet both of these requirements, being capable of

providing very fast typical speeds (e.g. >1 Gbps) and high

connection quality (via ultra-low latencies of less than 1.5ms)

simultaneously.

“Among the NGA

technologies,

FTTH/B is

technically the

most advanced

due to a number

of unique

characteristics”

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 12

Exhibit 3.5: Quality and speed requirements of different applications and

corresponding technologies [Source: Axon Consulting based on recent

research by the European Parliament]

Recent studies carried out by the FTTH Council Europe5 also confirm

that FTTH/B is the only technology with enough bandwidth to handle

projected consumer demands in the close future from a reliable and

cost-effective perspective.

The preferential characteristics of FTTH/B are recognized also in

Turkey, observable through:

 The deployments of FTTH/B networks in the recent years by

the incumbent operator as well as an alternative operator,

albeit focused in the densely populated, urban zones and

5 The FTTH Council Europe is an industry organisation whose mission is to accelerate the availability of fibre-based, ultra-high-speed access networks for the benefit of consumers and businesses. Created in 2004 by five founding members (Alcatel- Lucent, Cisco, Corning, Emtelle and OFS), the FTTH Council Europe now has more than 150 members and a contracted team working to promote the benefits of fibre access across the continent.

Requirements for transmission speed

High

Low

Low FTTHFTTC-VDSLADSL2+

High

Telemedicine

Online video games

VoIP

Email

Web browsing

Symmetrical file-sharing via cloud-computing

R e q

u ir

e m

e n

ts f o

r

t e c h

n ic

a l q u

a li t y

4K resolution streaming

Augmented Reality

“Recent studies

carried out by the

FTTH Council

Europe also

confirm that

FTTH/B is the

only technology

with enough

bandwidth to

handle projected

consumer

demands in the

close future from

a reliable and

cost-effective

perspective”

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 13

green field projects (further discussed in the coming

sections).

 Turkey’s 2015-2018 Information Society Strategy and Action

Plan released by the Ministry of Development, which points

out FTTH/B as the only long-term solution to the constantly

increasing need for bandwidth and consumption.

A number of countries across the world have been leading the

deployment and usage of FTTH/B technologies since quite a few

years. Among these Bulgaria, Denmark, Finland, Japan, Norway,

Portugal, Romania, Russia, South Korea, Spain and Sweden can be

cited as countries with sizable geographic areas. The common factor

in all these countries is the fact that deployment of FTTH/B has been

the main national target in enhancing fixed broadband offered to

consumers.

Our study starts with the presumption – and the vision – that

expanding FTTH/B coverage to a significantly larger part of the

population than its current state should be the goal of Turkey’s

Broadband Plan. Coupled with a properly designed investment

undertaking of a one-time nature, the whole initiative would avail

an infrastructure to the nation that will be usable for many decades.

3.3. Turkey’s fixed broadband opportunity: a market

with high growth potential

Turkey’s fixed broadband market has been growing in the recent

years in terms of both market size and demand. As can be observed

in the exhibit below, as of the end of 2016, the number of fixed

broadband connections in Turkey are forecasted to have reached

10,4 million, while the total annual data exchanged over fixed

broadband connections are forecasted to have been 8,4 TB6 million7.

6 Terabytes. 7 The forecasted figures for total fixed broadband connections and annual data exchange have been based on the actual data available from ICTA for the first three quarters of 2016

“Our study

departs from the

presumption that

expanding

FTTH/B coverage

to a significantly

larger part of the

population than

its current state

should be the

goal of Turkey’s

Broadband Plan”

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 14

Exhibit 3.6: Evolution of fixed broadband connections and total data

exchanged over fixed broadband in Turkey in the period between 2011

and 2016 (forecasted) [Source: ICTA]

As of Q4 2015, the total fixed broadband connections corresponded

to 83% of the 11,5 million fixed line connections in the market8. At

the same time, the total fixed broadband connections implied a 45%

take-up of fixed broadband connections within the households that

are covered by broadband services9.

The annual data exchange through fixed connections increased from

1,6 TB million in 2011 to the forecasted 8,4 TB million for 2016,

implying a CAGR10 of 38,6%, which compares significantly high in

relation to the 6,5% CAGR of fixed broadband subscriptions in the

same period. The above observation implies that per connection

demand for fixed broadband is increasing in Turkey.

In fact, as can be seen in the exhibit below, the average monthly

data consumption per fixed broadband connection has been growing

over the 2011-2016 period.

8 Source: ICTA, Q4 2015 market indicators report. 9 As per data from the Turkish Statistical Institute, there were 21,7 million households in Turkey in 2015. As per recent announcements of Turk Telekom, 98% of the households in Turkey are covered with fixed broadband. Hence, the estimated take up percentage of 45% is obtained through dividing 9,5 million fixed broadband connections by 98% of 21,7 million households. 10 Compound Annual Growth Rate.

7.6 7.9

8.4 8.9

9.5

10.4

0

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

10

11

2011 2012 2013 2014 2015 2016F

M il li o n s o

f T b y te

s e

x c h a n g e d

M il li o n s o

f c o n n e c ti o n s

Fixed Broadband Connections

Fixed Broadband Connections in Turkey (in millions)

Fixed data consumption in Turkey (millions of Tbytes)

CAGR: +6,5%

“The average

monthly data

consumption per

fixed broadband

connection has

been growing

over the 2011-

2016 period”

“Turkey’s fixed

broadband

market has been

growing in the

recent years both

in terms of

market size and

demand, implying

a 45% take up of

fixed broadband

services among

households

covered”

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 15

Exhibit 3.7: Evolution of fixed data consumption per connection and annual

increase in the period between 2011 and 2016 (forecasted) [Source: ICTA]

Note: Data consumption per connection has been calculated by dividing the total data

consumption and the average broadband connections along the year.

The average monthly fixed data per connection has increased at a

CAGR of 30,5% between 2011 and 2016. It appears from above that

the growth trend observed above will continue into the future.

The observed phenomena of increased usage in the fixed broadband

market results relevant also against the data usage trends in the

mobile market. The following exhibit provides a snapshot of the

monthly average data consumption in 2011 and the forecast for

2016 and the CAGR between these years in both markets:

18.2

23.0

30.3

39.8

54.9

68.8

0

10

20

30

40

50

60

70

2011 2012 2013 2014 2015 2016F

Fixed data consumption per connection in Turkey (Gbyte/connection-

month)

CAGR: +30,5%

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 16

Exhibit 3.8: Evolution of data consumption per connection between the

years 2011 and 2016 forecast in the fixed and mobile markets [Source:

ICTA] Note: Data consumption per connection has been calculated by dividing the

total data consumption and the average broadband connections along the year.

We note the following from the above exhibit:

 The growth in average consumption of fixed broadband data

(CAGR of 30,5%) has been substantially higher than the

growth in the average consumption of mobile data (CAGR of

23,5%).

 The average consumption of fixed broadband data was about

30 times (18,2 GB fixed data divided by 0,6 GB mobile data)

the average consumption of mobile data in 2011, while the

same multiple has increased to 38 times in 2016 (68,8 GB

fixed data divided by 1,8 GB mobile data).

We observe from the above that historically the appetite of Turkish

consumers for data consumption has been much higher for

consumption through fixed services compared to consumption

through mobile services. This could be attributed to the overall

higher speeds and usage quotas of fixed connections11.

11 Fixed broadband plans in Turkey usually allow unlimited data usage or usage subject monthly fair usage quotas on volumes of data downloaded. Such volume

18.2

68.8

0.6

1.8

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

0

10

20

30

40

50

60

70

Year 2011 Year 2016F Year 2011 Year 2016F

Fixed data consumption per connection in Turkey

Mobile data consumption per connection in Turkey

M o b il e , G

B / m

o n th

/ c o n n e c ti o n

F ix

e d , G

B / m

o n th

/ c o n n e c ti o n

CAGR:

+30,5%

CAGR:

+23,5%

“The appetite of

Turkish

consumers for

data consumption

has been much

higher for

consumption

through fixed

services

compared to

consumption

through mobile

services”

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 17

The LTE-A12 services recently launched in Q2 2016 may bridge the

gap between the fixed and mobile broadband data consumption,

although it is also true that wireless technologies tend to be

supplementary technologies compared to fixed technologies for

fixed broadband provision.

We observe that the usage habits of the Internet have been

changing dramatically over the recent years. The Internet has

evolved from a means for exchange of electronic messages and

downloading documents to a channel for a wide suite of applications

and services for consumers and businesses including, video media

streaming of increasing quality (with multiple simultaneous

streaming devices in a household), video conferencing, cloud-based

services and peer-to-peer file sharing, among others. Data available

from a recent report by Ericsson13 indicates that the worldwide

average monthly data consumption per fixed broadband connection

is expected to reach 180 GB/month by the year 2021. Attaining this

level of monthly data consumption by the same year in Turkey would

imply that the average consumption of 68,8 GB/month forecasted

in 2016 needs to increase by a CAGR of 21,2%, as presented in the

exhibit below.

based on quotas are significantly higher than the typical monthly allowances of mobile data plans. 12 Abbreviation of Long Term Evolution–Advanced. This terminology is currently referred to as the 4.5G technology in the Turkish market. 13 Report ‘Ericsson Mobility’, 2015.

“The LTE-A

services may

bridge the gap

between the fixed

and mobile

broadband data

consumption,

although wireless

technologies tend

to be

supplementary

technologies

compared to fixed

technologies for

fixed broadband

provision”

“Such a growth

seems plausible

given the

demonstrated

growth in

consumption over

the recent years,

as well as the

expected changes

in the usage

habits”

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 18

Exhibit 3.9: Projections of data consumption per connection between the

years 2016 and 2021 [Source: Ericsson]

Such a growth seems plausible given the demonstrated growth in

consumption over the recent years, as well as the expected changes

in the usage habits. We can observe in general that the Turkish

consumers will not be staying behind worldwide usage trends in the

years to come.

3.4. Understanding Turkey’s Broadband Gap

Subscribers in the Turkish fixed broadband market are served via a

number of different access technologies that include xDSL, FTTH/B,

cable and satellite. The exhibit below presents the share of these

technologies among the fixed broadband connections in Turkey for

the period between 2013 and 2016:

18.2

68.8

180.0

-

20

40

60

80

100

120

140

160

180

200

Year 2011 Year 2016F Year 2021 (Projection)

Fixed data consumption per connection in Turkey

G b y te

/ m

o n th

- c o n n e c ti o n

CAGR:

+30,5%

Projected CAGR:

+21,2%

“xDSL is still by

far the largest

access

technology in

2016, forecasted

to be accounting

for 73,9% of the

total fixed

broadband

connections as of

Q4 2016”

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 19

Exhibit 3.10: Breakdown of fixed broadband connections by technology in

Turkey [Source: ICTA]

From the exhibit above, we observe that xDSL is still by far the

largest access technology in 2016, accounting for around 74% of

the total fixed broadband connections. We also observe a moderate

decrease of this percentage from its level of 78,7% in 2013, mostly

in favour of the FTTH/B and cable access technologies.

In terms of the advertised connection speeds of the connections, we

see – as demonstrated in the next exhibit – that there has been a

migration towards higher speed connections in the period between

2013 and 201614:

14 The 2016 figures have been estimated based on the actual data available from ICTA for the first three quarters of the year.

78.7% 76.2% 74.9% 73.9%

5.8% 6.3% 6.6% 6.8%

14.1% 16.4% 17.5% 18.0%

1.4% 1.1% 1.0% 1.2%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2013 2014 2015 2016F

Breakdown of fixed broadband connections by technology in

Turkey

xDSL Cable FTTH/B Other

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 20

Exhibit 3.11: Breakdown of fixed broadband connections by speed in

Turkey [Source: ICTA]

We note that the observed switch from <10 Mbps speeds to 10-30

Mbps speeds has come through a technology upgrade of the existing

copper networks, making VDSL technology available to a larger part

of the households. In other words, the technology upgrade has still

kept xDSL as the major access technology, as viewed also on the

previous Exhibit 3.10.

Looking into the future, the expected growth of fixed broadband

demand in the coming years (as observed in the previous section)

implies that new and more profound upgrades or changes of access

technology may be timely and appropriate for Turkey.

The shifts in the habits of consumers, who more and more expect

to use high performance communication services, multiple

streaming music and video instances, and cloud services in many

aspects of daily life (e.g. public e-government services, online

storage), point out to the need for high speed, low latency networks

with high availability rates.

The above observations justify that setting the objective of the

national broadband plan to focus on deploying FTTH/B is both timely

and appropriate. Before moving on with our proposed vision for

implementing this objective, we find it useful to provide in the

coming paragraphs a snapshot of the current FTTH/B deployment,

81.0%

52.9%

31.1% 23.4%

18.3%

45.7%

66.4% 72.3%

0.7% 1.4% 2.5% 4.3%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2013 2014 2015 2016F

Breakdown of fixed broadband connections by speed in Turkey

(%)

< 10 Mbps 10-30 Mbps > 30 Mbps

“The expected

growth of fixed

broadband

demand in the

coming years

imply that new

and more

profound

upgrades or

changes of

access

technology may

be timely and

appropriate for

Turkey”

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 21

prices of FTTH/B services and the competition in the fixed broadband

market.

Current FTTH/B coverage and take-up

The levels of FTTH/B coverage15 and take-up16 in Turkey (23% and

33% by the end of 2015, respectively) compare significantly low

against the average of the leading FTTH/B countries17 (66% and

41%, respectively).

Exhibit 3.12: FTTH/B coverage15 and take-up16 gaps in Turkey against the

average of the leading FTTH/B countries [Source: European Commission,

FTTH Council and the OECD]

Until recently, the FTTH/B deployment in Turkey has been under a

regulatory holiday period18 introduced by the ICTA at the end of

2011. Under these observations, it is not clear whether the

regulatory holiday period has been effective in triggering the

investments needed for ensuring coverage levels in line with the

leading FTTH/B countries.

15 Coverage: households passed over total households in the country. 16 Take-up: households connected over households passed. 17 We refer to Bulgaria, Denmark, Finland, Japan, Norway, Portugal, Romania, Russia, South Korea, Spain and Sweden as the leading FTTH/B countries as mentioned under Section 3.2. 18 The regulatory holiday period ensures that neither the newly deployed fibre networks of the incumbent operator nor those of alternative service providers would be subject to market analysis, and hence obligations for wholesale network access arising from a potential designation of significant market power (SMP).

66%

23%

41%

33%

0%

10%

20%

30%

40%

50%

60%

70%

Leading FTTH/B

countries

Turkey Leading FTTH/B

countries

Turkey

FTTH/B Coverage FTTH/B Take-up

FTTH/B coverage and take-up in in Turkey against the average of the

leading FTTH/B countries

Coverage gap

43pp

Take-up gap

8pp

“The current

levels of FTTH/B

coverage and

take-up in Turkey

compare

significantly low

against the

average of the

leading FTTH/B

countries”

“It is not clear

whether the

regulatory holiday

period introduced

by the ICTA at the

end of 2011 has

been effective in

triggering the

investments

needed for

ensuring coverage

levels in line with

European

countries”

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 22

Price levels of FTTH/B services

The exhibit below compares the average purchasing power parity

(PPP) adjusted monthly price of a FTTH/B subscription with a

download speed of 100 Mbps in Turkey with the average observed

in the leading FTTH/B countries in 2015.

Exhibit 3.13: Price of reference for a FTTH/B connection in Turkey against

the average of the leading FTTH/B countries with comparable offers

[Source: OECD and information gathered from websites of service

providers in the respective countries adjusted for PPP]

The observed gap in PPP terms of 67,6% could be attributable to

the lower levels of coverage and take-up compared to the leading

FTTH/B countries (see Exhibit 3.12).

Competition in the fixed broadband market

In the Turkish fixed broadband market (i.e. broadband connections

over xDSL, cable and FTTH/B), the incumbent operator Turk

Telekom is the main player. As can be observed in the exhibit below,

the significantly high subscribers’ market share of Turk Telekom in

the market has been in a declining trend over the 2013–2016 period

(from 80% by the end of 2013 to a forecasted level of 69% as of

the end of 2016).

41.9

70.2

0

10

20

30

40

50

60

70

80

Leading FTTH/B countries Turkey

M o n th

ly U

S D

P P P p

e r

F T T H

/ B c

o n n e c ti o n

PPP adjusted monthly price of FTTH/B connection offering 100 Mbps

download speed, leading FTTH/B countries average vs. Turkey in 2015

+67,6%

“The average

price of a FTTH/B

subscription in

Turkey is 67,6%

higher than

comparable

existing offers in

the average of

the leading

FTTH/B

countries”

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 23

Exhibit 3.14: Market share in the fixed broadband market [Source: ICTA]

Despite a decline in Turk Telekom’s market share, it is still high in

comparison with international standards, especially when compared

with the average of the leading FTTH/B countries where incumbent

operators’ average market share of fixed broadband connections

stands at 40,5%, as presented in the exhibit below.

Exhibit 3.15: Incumbent operators’ market share in the fixed broadband

market in Turkey against the average of the leading FTTH/B countries

[Source: ICTA, EU Digital Agenda Scoreboard and other public sources]

80.0% 76.5% 71.8% 68.9%

11.3% 14.9% 17.6%

19.1%

8.8% 8.6% 10.6% 12.0%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2013 2014 2015 2016F

Market share of fixed broadband connections

Turk Telecom Superonline Others

40.5%

68.9%

0%

10%

20%

30%

40%

50%

60%

70%

80%

Leading FTTH/B countries Turkey

Incumbent operators' market share in the fixed broadband market

in leading FTTH/B countries and Turkey

Gap

28pp

“Despite a

decline in Turk

Telekom’s market

share, it is still

high in

comparison with

international

standards,

especially when

compared with

the leading

FTTH/B

countries”

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 24

3.5. Crafting a Broadband Plan for Turkey

In formulating a Broadband Plan, each country needs to consider its

conditions and priorities, taking into account the overall national

priorities, the socio-economic climate and geography, as well as

levels of broadband awareness among key stakeholders (such as

government agencies, business and community leaders and the

public).

Concerning the Turkish fixed broadband market, we can summarize

our observations in the previous sub-sections as follows:

 A market with high potential growth in terms of demand and

adoption of new services over fixed broadband.

 A market that appears to be impacted from low affordability

of high download speeds which in reality may be achieved

through new technologies and effective competition, and

instead basing its supply principally on legacy copper

networks in the foreseeable future.

These observations justify that Turkey stands at a right point in time

to consider and realize the vision of expanding FTTH/B coverage to

a significantly large part of the population. The challenge of this

vision lies in the fact that wide scale FTTH/B expansion projects

require significant investments and consequently a careful analysis

of the business case for the medium to long-term recovery of those

investments.

Expansion of FTTH/B networks in Turkey would imply taking forward

the household coverage ratio of this technology from its level of 23%

in Q4 2015 (as presented in the previous Exhibit 3.12). The initiative

would not only bring the coverage levels at par with the levels

observed in the leading FTTH/P countries, but also would attempt to

take it forward to ensure maximum adoption and take-up by the

consumers.

3.5.1. The Open Access Network (OAN) scheme

We discuss in this section the so-called open access network

(OAN) scheme which we believe suits well the objective of

deploying an FTTH/B network with a sufficiently large footprint in

Turkey.

“An open access

network (OAN)

scheme suits well

the objective of

deploying an

FTTH/B network

with a sufficiently

large footprint in

Turkey”

“Turkey stands at

a right point in

time to consider

and realize the

vision of

expanding

FTTH/B coverage

to a significantly

large part of the

population”

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 25

In the international examples of open access networks, a separately

founded entity in the market undertakes the deployment of a

common passive wholesale FTTH/B access network. This passive

network lets third parties – i.e. licensed retail service providers –

use the deployed network assets to serve end users in an effective

manner. In other words, in an OAN scheme, competing service

providers reach end user premises through a common passive

network/infrastructure offered at the wholesale level. The service

providers attach their active equipment to the passive network to

complete the full service to the end user. The end users may

choose among multiple service providers who offer ICT services.

The revenue source of an open access network is typically monthly

wholesale access fees collected from retail service providers for each

retail connection activated by end users. It is a common practice

that this access fee is determined on a cost-oriented, affordable and

non-discriminatory basis to ensure fair and equal passive access to

all players in the market.

An open access network with the above characteristics would cater

for the needs of the Turkish market by:

 Avoiding duplication of resources through multiple

service providers investing in FTTH/B passive access

networks. It would instead enable focusing on the timely

expansion of service to reach the widest level of FTTH/B

coverage through optimized roll-out and investment plans

(hence meeting the supply objective).

 Creating the environment for effective and non-discriminatory

access to network elements by multiple retail providers,

promoting an environment of affordable prices and volume

gains by retail policies that enhance take-up of FTTH/B

services (hence meeting the demand objective).

We also note that an open access network enables the utilization of

any government support and facilitation in the deployment of

FTTH/B networks, since it facilitates the pooling of resources and

capital, and efficient expansion. We discuss in Annex A, selected

examples of open access networks from the international

experience, providing also a perspective on the different levels

government support that was provided to such networks.

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 26

Bearing in mind the suitability of an open access network for Turkey,

we present in the next section an analysis of the likely investment

cases for such a network.

3.5.2. Investment needs of the open access

network scenarios in Turkey

We present below the results of an investment and business case

analysis we have carried out to assess the economic fundamentals

of a shared FTTH/B network in Turkey. The calculations behind the

presented results are based on a business case model19 developed

for this exercise.

We consider two main scenarios in the business case model:

 Scenario A (with reutilization of civil infrastructure):

Roll-out of the FTTH/B network with reutilization of the

existing civil infrastructure (in which ducts are rented through

a wholesale price) by the open access network entity. Such

entity in turn provides wholesale services to retail service

providers to make use of the deployed infrastructure.

 Scenario B (no reutilization of civil infrastructure): Roll-

out of the FTTH/B network with own deployment of the civil

infrastructure (i.e. no reutilization of the existing

infrastructure) by the open access entity. Such entity in turn

provides wholesale services to third parties to make use of

the deployed infrastructure.

We note that under the above Scenario A, we have not assessed the

actual feasibility of full reutilization of existing telecommunications

ducts in Turkey. However, we are of the opinion that some degree

of reutilization should be achievable, given the international

experience, and the possibility that ducts constructed not

necessarily only for telecommunications networks, but also for other

utilities (such as electricity and gas), may also turn out as at

disposition under the right conditions.

19 Further details on the basics and principles of the business case model are presented in Annex B.

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 27

We also note that under the above Scenario B, the assumed

deployment cost of ducts20 represent the cost of construction only.

Any fees that may be charged by various state entities (such as the

municipalities) in the form of rights of way or other descriptions are

not included. Therefore, the investment and business case results

of Scenario B presented further below should be interpreted as

outcomes in the absence of such charges.

Under both scenarios, we evaluate the investments needed to bring

the FTTH/B household coverage from its existing level of 23% (see

Exhibit 3.12) to higher levels. While explained in further detail in the

Annex B of this Report, we summarize below the principal household

coverage percentage levels referred in the presentation of our

results:

 23% FTTH/B coverage: the level of coverage as of Q4 2015

attained by own infrastructure of retail service providers.

 76% FTTH/B coverage: The coverage level signifying full

coverage of the households located in urban areas of Turkey

(i.e. cities with more than 100.000 inhabitants).

 90% FTTH/B coverage: The coverage level signifying full

coverage of the households located in urban and suburban

population centres in Turkey (i.e. all cities and towns with

more than 30.000 inhabitants).

 100% FTTH/B coverage: The coverage level signifying full

coverage of the households located in urban, suburban and

rural population centres in Turkey.

The exhibit below summarizes the resulting key performance

metrics under both scenarios in terms of FTTH/B coverage and

households passed.

20 Details of cost assumptions are provided in Annex B.

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 28

Alternative FTTH/B

coverage Million HH

passed

No action (Status Quo) – Current Situation 23% 5,00

Coverage of urban areas 76% 16,46

Coverage of urban and suburban population centres 90% 19,49

Coverage of urban, suburban, rural population centres 100% 21,66

Exhibit 3.16: Summary of performance metrics under different coverage

targets [Source: Axon Consulting]

The estimated investments needed by the open access network

entity for each scenario illustrated above are presented below.

Alternative

Estimated investments by the OAN (USD million)

Diff. Scenario A

(reutilization of civil infrastr.)

Scenario B (no reutilization of civil infrastr.)

No action (Status Quo) - -

Coverage of urban areas (>100.000 inhabitants)

1.740 3.167 x1,8

Coverage of urban and suburban population centres (>30.000 inhabitants)

2.757 4.561 x1,7

Coverage of urban, suburban, rural population centres

4.607 6.681 x1,5

Exhibit 3.17: Summary of investments needed for different household

coverage percentages under Scenario A and Scenario B [Source: Axon

Consulting] Note: Figures in USD millions.

As seen above, under Scenario B, with no reutilization of civil

infrastructure, depending on target coverage rates, the OAN entity

would need to undertake investments that are between 1,5 and 1,8

times more than Scenario A, where instead existing civil

infrastructure is reutilized. In other words, the possibility of

reutilizing existing civil infrastructure has a significant impact on the

investment case.

3.5.3. Business case analysis of the OAN scenarios

In this sub-section, we analyse the investment figures presented

above against the projected revenue streams foreseen for each

Scenario. We present in the remainder of this section an analysis of

Scenario A and B under the following assumptions concerning the

revenue model of the OAN:

“Under Scenario

B, with no

reutilization of

civil

infrastructure,

the OAN entity

would need to

undertake

investments that

are between 1,5

and 1,8 times

more than

Scenario A,

where instead

existing civil

infrastructure is

reutilized”

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 29

Input Assumption

value

Take-up under Sensitivity Case 1 33%

Take-up under Sensitivity Case 2 45%

Exhibit 3.18: Characteristics of the scenarios considered in the business

case analysis [Source: Axon Consulting] Note: Take-up figures are presented in

terms of households connected over households passed.

Among the assumptions, the wholesale access fee of the eventual

open access network is foreseen at a fee level that approximates

the currently regulated wholesale prices of traditional xDSL services

in the market. Hence, by setting a similar level, we aim to fulfil the

objective that the final service provided to the end users will be

priced at affordable levels. We note here, however, that the foreseen

wholesale price for this exercise has been chosen for purely

illustrative purposes and that once the OAN goes ahead, a detailed

analysis would need to be undertaken to determine the right

wholesale prices that are cost-oriented, and that ensure affordability

and non-discrimination.

Concerning the take-up of FTTH/B lines in the covered zones, the

business case analysis considers two sensitivity cases:

 Sensitivity Case 1 assumes that the retail service providers

will convert households covered to connections at the same

rate currently observed in the market (33%, as presented in

Exhibit 3.12).

 Sensitivity Case 2 assumes that this rate will reach the level

of 45%, i.e. the take-up of fixed broadband in Turkey as of

the end of 2015 (as outlined in sub-section 3.3).

We present in the following paragraphs the outcomes of the 4

scenario/sensitivity case combinations that result from the above.

Each scenario/case combination is presented in terms of the

evolution of:

 Estimated revenues

 Estimated expenditures (comprising operating expenses,

depreciation and cost of capital)

 Estimated economic margin (difference between revenues

and expenditures)

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 30

The above items are presented in net present value (NPV) terms

over the projections for the next 5 years and along a continuous

scale that represents different FTTH/B coverage levels starting from

23% up to 100%.

Scenario A – Sensitivity Case 1

The economics implied by Scenario A - Sensitivity Case 1

(Reutilization of civil infrastructure and take-up of 33%) are

presented in the exhibit below:

Exhibit 3.19: Economics of an open access network under Scenario A and

Sensitivity Case 1 in a 5-year period [Source: Axon Consulting]

We observe from above that in all coverage expansion points above

23%, Scenario A – Case 1 does not produce positive economic

margins for the open access network.

Scenario A – Sensitivity Case 2

The economics behind the Scenario A - Sensitivity Case 2

(Reutilization of civil infrastructure and take-up of 45%) are

presented in the exhibit below:

-4.000

-3.000

-2.000

-1.000

0

1.000

2.000

3.000

4.000

5.000

23% 31% 39% 47% 55% 63% 71% 79% 87% 95%

U S D

m il li o n

Scenario A - Sentivity Case 1

Estimated net present value of expenditures for a 5-year period

Estimated net present value of revenues for a 5-year period

Estimated net present value of margin for a 5-year period

URBAN

Current FTTH/B

coverage: 23%

SUBURBAN RURAL

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 31

Exhibit 3.20: Economics of an open access network under Scenario A and

Sensitivity Case 2 in a 5-year period [Source: Axon Consulting]

We observe above that under Scenario A – Case 2, the improved

take-up percentage assumption enables the open access network to

have positive and efficient economic margins for coverage levels up

to 52%. After this level of coverage, further expansion would imply

deteriorating margins compared to 52% coverage level.

In other words, the above economics suggest that the open access

network would be able to increase the FTTH/B coverage in the

country up to 52% (implying 11,26 million households covered) in

an economically efficient manner.

Scenario B – Sensitivity Case 1

The economics behind the Scenario B - Sensitivity Case 1 (No

reutilization of civil infrastructure and take-up of 33%) are

presented in the exhibit below:

-3.000

-2.000

-1.000

0

1.000

2.000

3.000

4.000

5.000

23% 31% 39% 47% 55% 63% 71% 79% 87% 95%

U S D

m il li o n

Scenario A - Sentivity Case 2

Estimated net present value of expenditures for a 5-year period

Estimated net present value of revenues for a 5-year period

Estimated net present value of margin for a 5-year period

URBAN

Current FTTH/B

coverage: 23%

SUBURBAN RURAL

Economic FTTH/B

coverage: 52%

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 32

Exhibit 3.21: Economics of an open access network under Scenario B and

Sensitivity Case 1 in a 5-year period [Source: Axon Consulting]

We observe from above that in all coverage expansion points above

23%, Scenario B – Case 1 does not produce positive economic

margins for the open access network.

Scenario B – Sensitivity Case 2

The economics behind the Scenario B - Sensitivity Case 2 (No

reutilization of civil infrastructure and take-up of 45%) are

presented in the exhibit below:

-5.000

-4.000

-3.000

-2.000

-1.000

0

1.000

2.000

3.000

4.000

5.000

6.000

23% 31% 39% 47% 55% 63% 71% 79% 87% 95%

U S D

m il li o n

Scenario B - Sentivity Case 1

Estimated net present value of expenditures for a 5-year period

Estimated net present value of revenues for a 5-year period

Estimated net present value of margin for a 5-year period

URBAN

Current FTTH/B

coverage: 23%

SUBURBAN RURAL

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 33

Exhibit 3.22: Economics of an open access network under Scenario B and

Sensitivity Case 2 in a 5-year period [Source: Axon Consulting]

We observe above that under Scenario B – Case 2, the improved

take-up percentage assumption enables the open access network to

have positive and efficient economic margins for coverage levels up

to 30%. After this level of coverage, further expansion would imply

deteriorating margins compared to 30% coverage level. These

economic outcomes suggest that the OAN would be able to increase

the FTTH/B coverage in the country up to 30% (implying 6,50

million households covered) in an economically efficient manner.

Economic implications of the 4 ‘scenario–case’ combinations

As mentioned in the previous paragraphs, we note that in some of

the ‘scenario–case’ combinations the OAN may avoid negative

economic margins by limiting the FTTH/B deployment at

economically viable levels.

The table below summarizes these coverage levels, and the relevant

economic indicators:

-4.000

-3.000

-2.000

-1.000

0

1.000

2.000

3.000

4.000

5.000

6.000

23% 31% 39% 47% 55% 63% 71% 79% 87% 95%U S D

m il li o n

Scenario B - Sentivity Case 2

Estimated net present value of expenditures for a 5-year period

Estimated net present value of revenues for a 5-year period

Estimated net present value of margin for a 5-year period

URBAN

Current FTTH/B

coverage: 23%

SUBURBAN RURAL

Economic FTTH/B

coverage: 30%

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 34

Scenario-Case

Economic FTTH/B

coverage of the

country

HH reached

with FTTH/B

(millions)

Estimated investments by the OAN

(USD millions)

NPV of economic

margin in a 5-year

period (USD millions)

Scenario A (with reutilization of civil infrastructure)

Sensitivity Case 1 (33% take-up) 23% 5,00 - -

Sensitivity Case 2 (45% take-up) 52% 11,26 948 202

Scenario B (without reutilization of civil infrastructure)

Sensitivity Case 1 (33% take-up) 23% 5,00 - -

Sensitivity Case 2 (45% take-up) 30% 6,50 413 21

Exhibit 3.23: Coverage and relevant economic indicators at economically

viable levels of FTTH/B coverage [Source: Axon Consulting]

We observe from above that the Sensitivity Case 1, i.e. preserving

the current FTTH/B take-up level of 33%, does not yield positive

economic margins for the OAN under any scenario. On the other

hand, under an increased take-up case, it appears that the OAN

would be able to undertake economic investments up to 52%

coverage (implying 11,26 million households covered) in the case of

reutilization of existing infrastructure, and 30% coverage (implying

6,50 million households covered) in the case of digging own ducts.

Summary and assessment of the scenario outcomes

The exhibit below provides a comparative summary view of the

outcomes of the different ‘scenario-case’ combinations presented

above. Particularly the exhibit presents:

 The total foreseen households reached in Turkey (including

the households covered with own infrastructure of retail

service providers as of Q4 2015).

 The foreseen network investments by the OAN entity.

 The foreseen NPV of the impact on economic profit21 in a 5-

year period of the household coverage percentage under each

case.

21 Difference between the NPV of the estimated margin for a certain level of FTTH/B coverage and the NPV of the estimated margin obtained under the economical FTTH/B coverage.

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 35

HH reached with

FTTH/B (millions)

Estimated investments by the OAN

NPV of variation on economic profit in a 5-

year period

Sensitivity Case 1

Sensitivity Case 2

Scenario A (with reutilization)

Economic coverage under sensitivity case 1 5,00 - - -

Economic coverage under sensitivity case 2 11,26 948 - -

Coverage of urban areas 16,46 1.740 -801 -81

Coverage of urban and suburban areas 19,49 2.757 -1.547 -653

Coverage of urban, suburban, rural areas 21,66 4.607 -3.043 -2.033

Scenario B (without reutilization)

Economic coverage under sensitivity case 1 5,00 - - -

Economic coverage under sensitivity case 2 6,50 413 - -

Coverage of urban areas 16,46 3.167 -1.391 -489

Coverage of urban and suburban areas 19,49 4.561 -2.293 -1.217

Coverage of urban, suburban, rural areas 21,66 6.681 -3.900 -2.709

Exhibit 3.24: Summary of performance metrics under Scenario A and

Scenario B [Source: Axon Consulting] Note: Figures in USD millions.

We can observe the following from above summary:

 The reutilization of civil infrastructure would lead the OAN to

undertake higher investments and reach more households.

Specifically, we observe that under a scenario with no

reutilization of civil infrastructure, the OAN would undertake

an investment amount of 413 million USD and increase

coverage of FTTH/B infrastructure from 5 million households

to 6,50 million households. The same figures would be 948

million USD and 11,26 million households under the scenario

of reutilization of the civil infrastructure.

 The reutilization of civil infrastructure would also enable the

OAN to reach to a larger coverage. For instance, while under

Scenario B (no reutilization), the OAN would need to invest

3,16 billion USD to cover all the households in urban areas

(population centres with more than 100.000 inhabitants), by

investing a substantially lower amount of 2,76 billion USD

under Scenario A (reutilization), the OAN would be able to

cover both urban and suburban areas (population centres

with more than 30.000 inhabitants).

 Under the reutilization scenario, the NPV of variation on

economic profit of the OAN to cover all urban areas varies

between -81 million USD and -801 million USD, depending on

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 36

the take-up rate of FTTH/B services. The magnitude of this

range suggests that with adequate levels of government

support, the coverage of all urban areas under the OAN

scheme is a viable and potentially realizable outcome.

All the above observations indicate that with adequate and timely

support, the OAN scheme can realistically cover all population

centres over 100.000 habitants in Turkey. Implying 76% of the

households, this would be an excellent result, bringing a state-of-

the-art and future-proof technology to the majority of the country,

and potentially contributing to the competitiveness of Turkey in the

international arena.

Government support to enable such a desirable outcome would

primarily need to focus on:

 Measures for the promotion of a friendly regulatory

environment in which it is enabled the reutilization of existing

civil infrastructure, e.g. the facilitation or removal of rights of

way or other fees currently charged by various state entities

such as the municipalities.

 Moderate financial support followed by a careful assessment

of the business performance of the OAN and potentially

providing grants of a relatively minor magnitude based on the

take-up of the FTTH/B services.

3.5.4. Role of the government to the expansion of

FTTH/B networks in Turkey

Given the social and economic implications of an FTTH/B network

with wide coverage, a key question that needs answering is the role

of the government and the extent of any eventual government

support in such an initiative. Government support to the deployment

of next generation broadband networks is a widely encountered

phenomenon in the international experience.

Government support may be in the form of facilitations (such as

rights of way), support in-kind, as well as outright financial aids. The

European Commission, for example, has a clear set of guidelines

“Given the social

and economic

implications of an

FTTH/B network

with wide

coverage, a key

question that

needs answering

is the role of the

government”

“Government

support to

broadband

deployment

initiatives would

be justified if

certain social and

economic

principles were

met”

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 37

targeted at the rapid deployment of broadband networks22 in

member countries. These guidelines define proper mechanisms for

the Commission to receive, evaluate and approve European aid

requests from member states or their jurisdictions (e.g. individual

regions, cities) for the deployment of next generation broadband

networks.

The guiding principles of the EC guidelines as well as the underlying

motivations of interventions encountered in other international

markets point out that government support to broadband

deployment initiatives would be justified if certain social and

economic principles were met. Such principles may be summarized

as in the exhibit below:

Exhibit 3.25: Key principles pursued by government interventions [Source:

Axon Consulting based on international practice]

We briefly explain below each of the above principles:

Principle 1: Delivering a material change in broadband

service level

A crucial point for justifying government intervention is whether the

relevant initiative is able to ensure a ‘step change’ in terms of

broadband availability. Such a step change would imply significant

new investments regarding infrastructure and significant new

capabilities to the market in terms of service characteristics.

22 ‘EU Guidelines for the application of State aid rules in relation to the rapid deployment of broadband networks’, 2013.

KEY

PRINCIPLES OF

A STATE AID

AND

INTERVENTION

Minimize the amount

of state fiscal subsidy where possible

Delivering a material

change in broadband service level

1 4

Maximize total

economic benefit

Intervene only where

the market does not deliver on the

broadband objectives

3 6

Promote competition

through fair wholesale access

Stimulate private

investment2 5

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 38

Principle 2. Stimulate private investment

Government intervention should be designed in a way to incentivize

the private sector to undertake the major part or the entirety of the

investment needed to reach the national broadband objectives.

Measures such as the facilitation or removal of rights of way or other

fees currently charged by various state entities to the operators

could, for example, become effective forms of intervention in this

direction.

Principle 3: Intervene only where the market does not deliver

on the broadband objectives

The government intervention should only be limited to areas where

the market either has failed to deliver service or does not deliver

service in line with the policy objectives.

Principle 4: Minimize the amount of state fiscal subsidy

where possible

In order to maximize value for money in the use of public funds,

government intervention should take place in the form of financial

aids only if such aids are absolutely necessary for reaching the

objectives of the intervention.

Principle 5: Promote competition through fair wholesale

access

Effective competition brings about a wide range of benefits. In

recognition of this, government intervention should aim at

promoting competition. In other words, all retail service providers

should be able to offer service under reasonable and equivalent

terms and conditions.

Principle 6: Maximize total economic benefit

Government support should take place with the recognition that

wide availability latest broadband technologies will yield numerous

benefits to the society and the wider economy.

The following exhibit provides an assessment of which conditions an

FTTH/B network in Turkey and its undertaking would qualify for

government support in the light of the above principles:

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 39

Principle Assessment of the Turkish case and the proposed

FTTH/B deployment Principle justified?

Principle 1: Delivering a material change in broadband service level

In those regions that are only covered by basic broadband technologies (e.g. ADSL and ADSL+), the widespread deployment of FTTH/B is expected to bridge the connectivity gap and provide a significant ‘step change’.

In regions where there are next generation technologies (e.g. FTTC/VDSL), deployment of FTTH/B will still bring material improvements in terms of competition, quality service and future proofing of the needs of the consumers.

Yes

Principle 2. Stimulate private investment

The national broadband initiative crafted for Turkey should be based primarily on investment by private parties, who may be existing players in the local market or new players looking to enter the market.

Yes, with the right

investment structure

Principle 3: Intervene only where the market does not deliver on the broadband objectives

The proposed deployment is directed at bringing FTTH/B to areas that are currently not covered with this technology.

Yes

Principle 4: Minimize the amount of state fiscal subsidy where possible

FTTH/B deployment should have the necessary transparency to permit timely and continuous assessment of network and business performance and determine the right amount of government support needed, if at all, through changing market conditions.

Yes, through a

sufficiently transparent undertaking

Principle 5: Promote competition through fair wholesale access

The deployed FTTH/B network should be available to all retail service providers at equal and suitable conditions.

Yes, through ensuring equal and fair access

Principle 6: Maximize total economic benefit

The proposed undertaking should make possible the existence of multiple competitors at the retail level, leading in time to product innovations. This, for example, could come through a pooling of resources at the wholesale level.

Yes, under a

resource pooling

approach at the

wholesale level

Exhibit 3.26: Justification of the key principles of government intervention in Turkey [Source:

Axon Consulting]

It can be observed from the above assessment that an initiative to

deploy an FTTH/B network would need to be complemented by the

following elements in order to qualify for an eventual government

support:

 An ownership structure consisting of private investors who

undertake the primary investment and the risk of the

initiative.

 A transparent governance structure, permitting continuous

monitoring of its technical and business performance and

timely adjustments to the support that the undertaking

receives from the government.

 Service provision at the wholesale level, through regulated

access prices and services equally available to all licensed

retail service providers.

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 40

 Pooling of individual funds and promoting sharing of

infrastructure in order to maximize the economic benefits that

would be passed to the consumers.

The proposed OAN scheme for deploying FTTH/B infrastructure in

Turkey would qualify for government support as deemed

appropriate since it would:

 Have a private ownership structure composed of one or more

parties who commit to provide the main funding of the

common passive wholesale network and undertake the

business risk.

 Be a separate undertaking offering access at the wholesale

level only, and ensuring a setup with a governance structure

of maximum transparency.

 Provide wholesale service through regulated access prices.

 Enable the pooling of individual investors interested in

network deployment in Turkey, and ensure that funds are

spent economically.

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 41

Conclusions

We present below our conclusions and relevant recommendations from our previous

analyses:

Conclusion 1: Under the current circumstances, there is a risk that Turkey

will miss the opportunity associated to the high growth potential of the

Turkish fixed broadband market, with up to 15,16 million households being

left without access to FTTH/B networks

The Turkish fixed broadband market is growing both in terms of consumption per

connection and in terms of the number of connections. Consumers are increasingly

adapting to the services available over fast broadband speeds, and are on track to

match the expected worldwide consumption trends.

This environment of growing demand represents an excellent background for the

extension of future-proof FTTH/B technologies to serve a large percentage of the

population in Turkey.

According to our study, however, under the current circumstances, we expect that

only up to an additional 1,50 million households in Turkey will be covered over the

next five years by FTTH/B technology, bringing the total number to 6,50 million

households. This is an improvement – albeit a small one - from the levels of coverage

currently achieved (5 million households).

This implies that up to 15,16 million households in Turkey would remain without

access to FTTH/B technologies and would have to wait for the next technology-change

cycle to obtain such access.

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 42

Exhibit 4.1: Proportion of total households with access to FTTH/B under current

circumstances [Source: Axon Consulting]

This gap in FTTH/B coverage does not necessarily have to materialize. As our study

has shown, the potential for the expansion of FTTH/B in Turkey can go significantly

above the indicated number of 6,50 million households. This number could potentially

increase to 11,26 million households in Turkey by market-driven mechanisms under

the right regulatory and market conditions.

A first conclusion from our study is therefore that keeping the current course in terms

of broadband policy would imply a very high cost for Turkey. The cost would come in

the form of foregone investments, reduction of ultrafast coverage, choice and quality

of service for consumers and, most importantly, slower transition of the country

towards a digital economy.

Conclusion 2: Promoting a friendly environment for commercially-driven

network roll-out would open up the investment opportunity and could lead

to an additional 535 million USD in private investment and 4,76 million

additional households with FTTH/B coverage

Our business case analysis presented in the previous sections indicates that a large

part of the cost for the OAN would comprise the cost of deploying new civil

infrastructure. The investments required to build a FTTH/B network in Turkey would

be reduced by between 31% to 45% in case that civil infrastructure could be reused.

23% (5,00 MM HH)

30% (6,50 MM HH)

77% (16,66 MM HH)

70% (15,16 MM HH)

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Current situation Expected situation in a 5-year period

under the current circumstances

% of households with no access to FTTH/B

% of households with access to FTTH/B

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 43

More importantly, our study shows that the introduction of agile and efficient civil

infrastructure reutilization measures would result in an increase (not a decrease) of

investments by private agents in FTTH/B networks. Our model suggests that

investments by private operators in FTTH/B networks would further increase by up

to 535 million USD with the presence of civil infrastructure reuse, leading to an

additional up to 4,76 million increase in the number of households covered.

Metric

Without reutilisation of

civil infrastructure (status quo)

With reutilisation of

civil infrastructure

Impact of civil infrastructure reutilisation

Total households covered (in million)

6,50 11,26 +4,76

Additional households covered with respect to current coverage (in million)

1,50 6,26 +4,76

Additional private investments (in USD million)

413 948 +535

Private investments per additional household (USD/HH)

275 151 -124

Exhibit 4.2: Impact of civil infrastructure reuse in terms of covered households and private

investment (under the take-up sensitivity case 2) [Source: Axon Consulting]

The findings of our study are fully aligned with the international experience.

Government and regulatory bodies around the world have been aware for some time

already of the importance of enforcing adequate mechanisms to facilitate an efficient

use of existing civil infrastructure for new FTTH/B networks.

Measures typically adopted by governments and regulatory agencies around the

world to facilitate such reuse include the following:

 Imposition of obligations to operators to provide access to existing ducts, poles

and other civil infrastructure, either on the basis of market dominance (such

as in United Kingdom, France, Spain and many other EU Member States), or

as a general obligation applicable to all service providers (such as in Portugal).

 Reuse of civil infrastructure from other industries and/or public utilities, such

as electricity, water or transportation.

 Facilitation of permits, and reuse of civil infrastructure from public authorities.

 Obligations on any new constructions (including water, energy, transport,

sewage networks, new buildings) to leave suitable space for fibre deployment.

 Support the development of a common platform that would enable through

time an instantaneous and online referencing of available ducts and other

channels in a certain area.

In this regard, it is worth highlighting that even though there currently exists in

Turkey a regulatory initiative to promote the sharing of the ducts of the incumbent

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 44

fixed-line operator, the reduced take-up of the service suggests that such mechanism

is not working well in practice. As for reuse of other civil infrastructure, the extent to

which such measures are present in Turkey is at best very limited, or inexistent. In

fact, according to the statistics from the local regulators and Vodafone Turkey, as of

2015 there were less than 1.000 km of shared ducts in Turkey, as opposed to 28.760

km in France, 10.000 km in Spain and 7.523 km in Portugal.

We understand that although the history of duct sharing regulation dates back to

2012, the process since then has been a rather slow one, allowing alternative service

providers to file duct-sharing applications only since April 2015.

Conclusion 3: Moderate financial and regulatory support from the

government to an Open Access Network could help boost expansion and

provision of FTTH/B networks to all Turkish households living in cities of

over 100.000 inhabitants

Our study suggests that the right conditions apply in Turkey for the government to

consider some sort of intervention to help expand the current footprint of FTTH/B

networks. We have seen that all six principles that, according to international best

practice should guide any government intervention in broadband markets, apply

largely in Turkey.

As per our analysis, the intervention scheme that best fits the particular

circumstances in Turkey would be those of an ‘Open Access Network’, built, owned

and (mostly) financed by the private sector, which would operate under certain

obligations, including the provision of cost-oriented, affordable and non-

discriminatory wholesale access to all retail operators.

Such a scheme would diminish the financial commitments required from the

government and would eliminate any risk for the government associated with the

operations of such network (or networks) – other than general supervision of its

operations and determination of the appropriate conditions of access. The presence

in Turkey of several competing telecommunication operators with strong financial

and operational capacities, which have signalled a determination to operate in this

market, would also facilitate the implementation of such a scheme.

Based on our analysis of the economics currently associated to the deployment of

FTTH/B networks in Turkey, we believe that an ambitious yet achievable target for

such a scheme would be to provide coverage, over a 5-year period, to all households

in Turkey located in population centres with 100.000 thousand or more inhabitants.

Such target would amount to 76% of all households in Turkey having FTTH/B

coverage. If achieved, this would put the country as a clear leader in the region in

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 45

terms of FTTH/B coverage and at par with some of the most advanced countries in

the world.

Expressed in alternative terms, the suggested intervention would bring FTTH/B

coverage to between extra 5,20 million to 11,46 million Turkish households which

would otherwise not have coverage.

Exhibit 4.3: Impact of the recommended intervention in terms of households covered with

FTTH/B [Source: Axon Consulting]

According to our estimations, the suggested scheme would have public funding

requirements in the range of 81 million USD (in the most optimistic scenario of high

take-up by users) to 489 million USD. While these are sizable amounts, they are not

misaligned with previous broadband policy interventions by the government to bring

broadband coverage to rural areas. These estimations assume that the measures for

reuse of civil work described in our second recommendation are in place and are

available for the builder of the Open Access Network. This public funding would in

turn mobilize estimated additional private investments of 792 million USD, which

would contribute to increased employment and economic activity.

5,00 5,00

11,26

16,46

0

2

4

6

8

10

12

14

16

18

Sentivity Case 1 Sentivity Case 2

Current situation Expected situation in a 5-year period

under favourable civil infrastructure

sharing mechanisms

Expected situation

in a 5-year period

under

recommended

intervention

M il li o n s o

f h o u s e h o ld

s

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 46

Metric

Expected situation in a 5-year period under favourable

civil infrastructure mechanisms

Expected situation in a 5-year period under potential

incentives

Impact

Total households covered (in million)

11,26 16,46 +5,20

Additional households covered with respect to current coverage (in million)

6,26 11,46 +5,20

Public funding requirements (in USD million)

- 81 to 489 +81 to 489

Additional mobilization of private investment (in USD million)

948 1.740 +792

Exhibit 4.4: Impact of the recommended scheme on public funding and private investments

[Source: Axon Consulting]

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 47

Annex A: Open access broadband

networks and government support in the

international experience

As a consequence of the strategic importance placed on next-generation broadband

by international governments, there is a significant body of literature available that

examines the open access networks’ ownership models that can be and have been

used to deliver significant investment in next-generation networks and services in

other countries.

The typical ownership models observed internationally are briefly summarized below.

Exhibit A.1: Summary of typical ownership models observed internationally [Source: Axon

Consulting based on international practice]

 Option 1: Private sector build, finance, own and operate with

obligations. Government contracts with a private market player (or several

market players) who will finance, design, build, own and operate the fixed

broadband infrastructure. Capital subsidy will be paid to the private market

players through grants, which are paid during deployment and through the

operational life of the contract. The private market players retain ownership of

the network at the end of the expiry contract.

 Option 2: Private sector finance, build and operate with asset reversion.

Government contracts with a private market player (or several market players)

who will finance, design, build and operate the broadband infrastructure. The

private market players will derive economic benefit from the infrastructure for

the duration of the contract at which point ownership will revert to the public

sector. On a similar basis to the previous option, capital subsidy will be paid to

the private market players during deployment and through the operational life

of the contract.

 Option 3: Corporate joint venture. The government and a private market

player (or several market players) form a joint venture that will design, build

Option 2: Private sector finance, build and operate with asset reversion

Option 1: Private sector build, finance,

own and operate with obligations

Option 3: Corporate

joint venture

Option 4: Public sector finance and own with

private sector design, build and operate

Option 5: Public sector build, finance,

own and operate

Involvement of the government

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 48

and operate the wholesale network. Both parties own equity in the entity and

split the risk and rewards of ownership. Equity is invested by the government

and the private market players over the deployment period in proportion to their

shareholdings.

 Option 4: Public sector finance and own with private sector design, build

and operate. Government funds and owns the wholesale network and contracts

with a private market player (or several market players) for the network’s

design, build and operation. The partner derives the economic benefit from the

network and bears the commercial risk of wholesale operations until contract

expiry.

 Option 5: Public sector build, finance, own and operate. Government

designs, builds, operates and owns the wholesale network. It will derive all

benefits associated with development and operation of the wholesale network

and assume all risk. There is no private sector involvement in the project beyond

the contracting of network design and build.

A wide number of countries have implemented the aforementioned ownership

models in their intervention strategy. The exhibit below summarizes some examples.

Ownership models International

examples

Option 1: Private sector build, finance, own and operate with obligations

Ireland, United Kingdom

Option 2: Private sector finance, build and operate with asset reversion

-

Option 3: Corporate joint venture New Zealand, Singapore

Option 4: Public sector finance and own with private sector design, build and operate

France

Option 5: Public sector build, finance, own and operate

Australia

Exhibit A.2: International examples of ownership models [Source: Axon Consulting]

Australia, due to lack of private sector participation in the process, is the only country

that has opted for Option 5 by taking a government-owned open access approach

by establishing the National Broadband Network Company (NBNCo) in 2009, that is

responsible for the design and the build of the infrastructure. Australia’s NBNCo

model places all of the risk on the public sector and has run into public funding issues

several years after its launch.

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 49

In contrast, New Zealand and Singapore have instead opted for creating network

companies through Option 3, which is forming a joint venture that is co-owned

among several private entities and the government. As mentioned before, this

method has the added benefit of distributing the risk among all of the co-investors

in the open access network while allowing government support in the initial

deployment phases where there are high cost barriers. For example, in New Zealand

the government is responsible for the roll-out down the street while the private

sector is responsible for funding the fibre to the premises upon customer request.

Therefore, the government bears the initial risks of the immature market. As end-

user fibre take-up increases, the private sector will gradually repay the government’s

investment, which can in turn be invested again towards infrastructure roll-out.

Through this model, fibre roll-out in New Zealand has been progressing very well

with the initial targets increases from the conclusions of revision conducted in

September 2014.

As for Singapore, NetCo is the entity formed for the fibre network which has an open

access structure. It is a consortium of several private companies with a 36 percent

funding from the government, while the rest (more than 1.4 USD billion) is expected

to come from the private sector. As a high-income island state, Singapore might not

be a template for national service provisioning, however it raises an interesting point.

Even in an affluent city-state where operational circumstances (high proportion of

multi-dwelling blocks) are favourable, significant government funding is needed. This

suggests that government funding may be needed in most countries.

Sweden, which has the highest penetration of FTTH/B in Europe among the OECD

countries, has taken a diversified approach and created open access networks on

municipal levels.

For the case of France, France Telecom was selected through a competitive tender

process to undertake the design, build and operation of the network as part of a 10-

year contract in order to provide access networks on municipal levels (e.g. Auvergne

region).

In addition to this, recently the French regulator has mandated network sharing for

in-building wiring on national level. Under this model, one operator signs a contract

with the building owner and becomes the main operator within the building. This

operator is in charge of constructing and maintaining the networks and offering

passive access to alternative operators, either through a dedicated or shared fibre

line.

For the cases of Ireland and the United Kingdom, Option 1 has been chosen as the

appropriate approach in creating an open access network, where the government

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 50

has contracted private sector companies to finance, design, build, own and operate

the fixed broadband infrastructure.

It is important to note that the analysis of the international experience demonstrates

that the government intervention has had positive outcomes in the fixed broadband

market across a wide number of countries. However, a variety of ownership models

are used across the world, illustrating that the choice of ownership model employed

needs to be tailored to the situation and that there is no ‘one-size-fits-all’ solution.

A decision on the appropriate ownership model therefore needs to consider all of

ownership options on their respective merits.

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 51

Annex B: Principles of the business case

model for Turkey

In order to perform a numerical assessment of relevant economic and performance

metrics for the configuration of an open access network in Turkey, a business case

model has been developed.

An overview of the high-level structure of this business case model is presented in

the exhibit below.

Exhibit B.1: Overview of the structure of the business case model [Source: Axon Consulting]

In the following paragraphs, a description of the inputs and assumptions considered

in the business case model is provided.

Economic inputs

The exhibit below summarizes the economic inputs considered in the business case

model, together with the source of reference.

Economic inputs Demographic inputs

Network elements characteristics Network dimensioning assumptions

Households connected by FTTH/B

Network dimensioning

Estimated CapEx + OpEx

Wholesale price

Estimated margin

Estimated revenues

Inputs/Assumptions

Calculations performed

OUTPUTS

Households passed by FTTH/B

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 52

Input Value Source of reference

Weighted Average Cost

of Capital (WACC) 12%

Axon Consulting based on

international references

Wholesale price 76,80

USD/year Vodafone

Exhibit B.2: Summary of economic inputs used in the business case model [Source: Axon

Consulting]

Demographical inputs

A characterization of the Turkish territory has been performed in order to typify the

households according to the following classification: urban, suburban and rural. The

exhibit below illustrates an overview of the methodological approach followed for

classifying the existing households in the country.

Exhibit B.3: Overview of the approach followed for classifying the Turkish territory [Source:

Axon Consulting]

As illustrated above, the first step within the methodological approach gathers the

inhabitants of Turkey into a total of 2012 population centres. The source of reference

for this information is GeoNames, a geographical database available and accessible

through various web services under a Creative Commons attribution license, which

integrates geographical data such as location, elevation and population.

Inhabitants per population centre [#2012 POPULATION CENTRES]

URBAN GEOTYPE (Pop>100k)

SUBURBAN GEOTYPE

(30k<Pop<100k)

RURAL GEOTYPE (Pop<30k)

Number of inhabitants per geotype at a REGIONAL LEVEL [#82 PROVINCES]

Inhabitants/HH for each region

Number of HH per geotype at a REGIONAL LEVEL

[#82 PROVINCES]

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 53

Each population centre is identified with the correspondent province and classified

according to the following thresholds:

 Urban: Population centres of more than 100.000 inhabitants.

 Suburban: Population centres of more than 30.000 inhabitants and less than

100.000 inhabitants.

 Rural: Population centres of less than 30.000 inhabitants.

The result of the previous step is the number of inhabitants for each region and

classified into urban, suburban and rural areas. According to this classification, the

next step comprises the application of the average inhabitants per household for each

of the 82 regions. This information is extracted from the Turkish Statistical Institute

for the year 2015.

The outcome of this calculation is the number of households for each of the 82 regions

and clustered into urban, suburban and rural areas. According to this approach, the

characterization of the Turkish territory would be as follows:

Geotype % of

households

Urban 76%

Suburban 14%

Rural 10%

TOTAL 100%

Exhibit B.4: Characterization of the territory [Source: Axon Consulting]

The resulting disaggregation is aligned with the urban population rate published by

the World Bank for the year 2015 (74%).

Network elements characteristics

When estimating the expenditures (CapEx23 and OpEx24) in the business case model,

a number of network elements of reference have been considered, as illustrated in

the exhibit below.

23 Capital expenditures. 24 Operational expenditures.

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 54

Exhibit B.5: Overview of the network elements considered in the business case model

[Source: Axon Consulting]

The exhibit below summarizes the characteristics of these network elements in terms

of unitary CapEx, unitary OpEx and useful life.

Element Unit Unitary CapEx (USD)

Unitary OpEx (USD)

Useful life (Years)

Source of reference

MSAN per

equipment 31.116 768 5

International benchmark

ODF per

equipment 29.968 877 5

International benchmark

Cabinet per

equipment 2.342 - 5

International benchmark

Fibre cable per km 123 13 24 International benchmark

BTU per

equipment 31 - 5

International benchmark

Line card per

equipment 629 7

International benchmark

Ducts

Scenario A25 (rental)

per km - 279 1

Regulation on right of way

regarding fixed and mobile

communications infrastructure

Scenario B25 (deployment)

per km 22.638 63 30 Vodafone

Exhibit B.6: Summary of the network characteristics inputs used in the business case model

[Source: Axon Consulting]

25 The descriptions of Scenario A and B are detailed in the section ‘Crafting a Broadband Plan for Turkey’.

BTU

Civil infrastructure

Civil infrastructure

Cabinet MSAN ODF

Line cards

Fibre deployment (Secondary NW)

Fibre deployment (Primary NW)

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 55

We note that the deployment cost of ducts (22.638 USD/km) presented above

represents the cost of construction only. Any fees that may be charged by various

state entities (such as the municipalities) in the form of rights of way or other

descriptions are not included.

For the quantification of the net present value of the network expenditures, the above

CapEx and OpEx have been annualized along the useful life of the element. The

annualized figures for a 5-year period (i.e. 5 times the annualized figure) have been

calculated as a net present value utilizing the WACC value presented in Exhibit B.2.

The unitary net present value figures for each of the network elements are presented

in the exhibit below.

Element

Unitary NPV

expenditure

(USD)

MSAN 33.885

ODF 33.128

Cabinet 2.342

Fibre cable 104

BTU 31

Line card 497

Ducts Scenario A (rental) 1.004

Scenario B (deployment) 10.359

Exhibit B.7: Summary of the net present value of the expenditures for each of the network

elements [Source: Axon Consulting]

Network dimensioning assumptions

When estimating the number of network elements needed for the FTTH/B expansion,

a number of assumptions have been considered in the business case model. These

assumptions are summarized in the exhibits below.

Network Geotype

Average distance

of the fibre/civil

infrastructure

(Kms.)

Primary network

Urban 0,3

Suburban 0,9

Rural 1,8

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 56

Network Geotype

Average distance

of the fibre/civil

infrastructure

(Kms.)

Secondary network

Urban 0,6

Suburban 1,5

Rural 5,0

Exhibit B.8: Summary of the assumptions for the deployment of fibre cables and civil

infrastructure [Source: Axon Consulting]

Element Value

Maximum number of customers per cabinet 2.000

Maximum number of cabinets per node 5

Max number of lines per FTTx line card 256

Exhibit B.9: Summary of the net present value of the expenditures for each of the network

elements [Source: Axon Consulting]

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 57

Annex C: Sources of reference for the

figures of coverage, take-up and prices

Sources of reference for the figures of COVERAGE

Country Year of

reference Source of reference

Bulgaria 2015 FTTH Council of Europe

Denmark 2015 European Commission

Finland 2015 European Commission

Japan 2014 Japan Telecoms Market Report 2015

Norway 2015 European Commission

Portugal 2015 European Commission

Romania 2015 European Commission

Russia 2015 FTTH Council of Europe

South Korea 2014 South Korea Telecoms Market Report 2015

Spain 2015 European Commission

Sweden 2015 European Commission

Exhibit C.1: Sources of reference for the figures of coverage

Sources of reference for the figures of TAKE-UP

Country Year of

reference Source of reference

Bulgaria 2015 FTTH Council of Europe

Denmark 2015 OECD

Finland 2015 FTTH Council of Europe

Japan 2014 Statistical Handbook of Japan 2015

Norway 2015 FTTH Council of Europe

Portugal 2015 OECD

Romania 2015 FTTH Council of Europe

Russia 2015 FTTH Council of Europe

South Korea 2015 South Korea Telecoms Market Report 2015

Spain 2015 OECD

Sweden 2015 FTTH Council of Europe

Exhibit C.2: Sources of reference for the figures of take-up

Turkey’s digital future: Vision for the Broadband Plan

Axon Partners Group Consulting © - 2017 58

Sources of reference for the figures of PRICES

Country Year of

reference Source of reference

Bulgaria 2016 Operator's website

Denmark 2015 OECD

Finland 2015 OECD

Japan 2015 OECD

Norway 2015 OECD

Portugal 2015 OECD

Romania 2016 Operator's website

Russia 2016 Operator's website

South Korea 2016 Operator's website

Spain 2015 OECD

Sweden 2015 OECD

Exhibit C.3: Sources of reference for the figures of prices