HLSS645Wk1

Marine Policy 121 (2020) 104180

Available online 2 September 2020 0308-597X/© 2020 Elsevier Ltd. All rights reserved.

Maritime dependency and economic prosperity: Why access to oceanic trade matters

Jesse M. Lane a,*, Michael Pretes b

a University of North Carolina at Greensboro, USA b University of North Alabama, USA

A R T I C L E I N F O

Keywords: Maritime dependency Maritime trade Landlocked countries Development Trade policy

A B S T R A C T

Maritime trade and access to deep-water territory are important when determining a country’s economic success. Today, as much as 75% of international trade takes place over water due to the fuel efficiency of seaborne freight and worldwide dependency on water as a means of transportation. Issues in economic geography such as being landlocked and maritime trade are directly related to global development patterns. A country’s ability to participate in international trade and transport goods overseas is integral in the modern global economy. Maritime dependency is the ability of a country to participate in maritime trade as determined by their geographic access to international waters and trade dependency. Access to maritime shipping and global participation in maritime trade is key to attracting global capital. This study explores the relationship between five major factors in maritime dependency and economic prosperity. Findings suggest a significant relationship between maritime dependency and gross domestic product (GDP) per capita. Results from this analysis were indexed and an Index of Maritime Dependency was created and mapped to display the geographical distribution of maritime dependency.

1. Introduction

Maritime trade is an essential element in the global economy. As technology has developed at a rapid pace, people have become more interconnected than ever before. This growth in technology has spurred the need for better access to goods and services. Just how important is access to navigable water, and especially oceans, for economic pros- perity? Is there a relationship between the geographic aspects of mari- time dependency and economic prosperity, given that today as much as 75% of worldwide trade is done over water [1]? In order to explore the geospatial relationship between maritime dependency and economic prosperity, this paper begins by developing a set of variables used to determine how exposed a country is to maritime trade. Multiple regression was used to analyze the relationship between maritime de- pendency and economic prosperity.

The importance of maritime trade was recognized early on by coastal states. Coastal access was instrumental in facilitating the growth of such polities as Phoenicia, Greece, Rome, Carthage, and, later, of European states as well as Arab civilization. The rise of Europe to global influence was due largely to maritime access, as ably described by the American Admiral Alfred Thayer Mahan in his classic study The Influence of Sea

Power upon History, 1660–1783. Mahan described the sea as a “great highway” and a “wide common,” of crucial importance because water transportation was cheaper and safer than transport by land [2]; p. 25). Mahan’s concern was mainly with sea power, and far less with the role of trade in times of peace. In this paper we attempt to quantify Mahan’s thinking on geographical position, with a focus on maritime trade.

Maritime trade has become increasingly important to the survival of the economic system. Transportation costs greatly affect the ability of a country to participate in trade. This is one reason why many landlocked countries find it difficult to send and receive foreign goods. As of 2014, domestic freight ships consume 214 British thermal units (Btu) per ton- mile, rail freight consumes 292 Btu per ton-mile, and heavy single-unit trucks consume 21,573 Btu per vehicle-mile [3]. These data show the magnitude of maritime shipping within a globalized world and give rationale for providing quality access to maritime shores. As energy consumption and transportation costs are concerned, maritime shipping is the cheapest form of transportation.

The purpose of this study is to explore the relationship between maritime trade dependency and economic wealth on a global scale. Multiple regression was used to determine if a relationship exists be- tween a country’s deep-water ports, merchandise trade ratio, coastline

* Corresponding author. 237 Graham Building, 1009 Spring Garden Street, Greensboro, NC, 27412, USA. E-mail address: [email protected] (J.M. Lane).

Contents lists available at ScienceDirect

Marine Policy

journal homepage: http://www.elsevier.com/locate/marpol

https://doi.org/10.1016/j.marpol.2020.104180 Received 19 February 2020; Received in revised form 21 April 2020; Accepted 11 August 2020

Marine Policy 121 (2020) 104180

2

length, coast/area ratio, liner shipping connectivity, and Gross Domestic Product per capita. Resulting variables from the best-fit model were indexed together and collectively titled the Maritime Dependency Index (MDI). Maritime dependency is defined as the ability of a given country to participate in maritime trade. The culmination of the above-mentioned factors provides insight on the ability of a country to participate in maritime trade. These factors also represent the interconnection and interdependency of countries around the world. The evaluation of maritime dependency may help provide some clues and insight on future development and growth patterns.

1.1. Landlocked countries

Scholars have studied the impact of geography on economic pros- perity for years and several issues have risen in prominence. Landlocked countries [Fig. 1] face what seems to be an insurmountable obstacle: how to access the coast without bankrupting the private sector. A debate within the literature is to what extent the geographic location of a country affects the overall success of that country’s economy. Some scholars believe that being landlocked is highly disadvantageous for a country [4–6]. According to Arvis [4]; “The high logistics cost and the many developmental problems faced by the landlocked countries of the world can be attributed to their geographical fate. The importance of the transit facilitation agenda to these countries and to the countries of transit stem from these circumstances” (p. 244). The main concern is the ability of a landlocked country to reach inexpensive yet efficient transit systems and trade corridors. Carmignani [6] found that landlockedness had a significantly negative impact on GDP per capita, but argued that the ultimate cause of economic stagnation was institutional quality. While these studies provided valuable information on underdeveloped landlocked countries, they did not provide an explanation as to why some countries such as Switzerland and Luxembourg have been able to circumvent the impediment of being landlocked.

Other authors vary in their explanations of the overall causes of low economic output. Some researchers conclude that the root cause of economic stagnation in landlocked countries stem from relationships with neighboring countries, rather than any inherent landlocked geo- spatial position [7,8]. This explains the economic success of such Eu- ropean landlocked countries as Switzerland and Austria, where neighboring countries are seen as a market, rather than as an obstacle to trade [9,10]. According to Raballand [11]; Switzerland, Austria, and Czech Republic have several options for trading partners, hence have more bargaining power than other landlocked countries. Some scholars have argued that being removed from the interference of main trading powers means that landlocked countries have more freedom to choose their own path [8]. Other scholars argue that there is an inverse rela- tionship between the cost of transportation and the number of items transported [12]. Chowdhury and Erdenebileg’s analysis states that as the number of items transported from a landlocked country increases, the cost of transportation for each item decreases. According to Faye, McArthur, Sachs, and Snow [7]; the economic woes of a landlocked country can be attributed to a combination of its distance from a viable sea port and the high cost of transportation through a neighboring country.

Collier [13] argued that landlocked developing countries find it harder to see substantial growth today because of the control that super giants such as China and India have on their natural resource trade. As he notes, throughout the past few decades, China’s role in resource exploitation in many developing nations, especially in Africa, and the subsequent loans that are given to these developing countries to build infrastructure has given the Chinese government more control over the decision-making abilities of these countries. African borders were drawn by European colonial powers at the Berlin Conference of 1884-85, resulting in many landlocked countries and microstates, which were not based on traditional national boundaries or geographic regions (such as river basins) [14,15]. According to Alden and Alves [16]; these

Fig. 1. Locations of landlocked countries (data source: [41]).

J.M. Lane and M. Pretes

Marine Policy 121 (2020) 104180

3

landlocked African countries have failed to see high rates of economic development due to China’s neocolonial activities on the continent, while Oloruntoba [17] argued that limited development in these coun- tries stems directly from European colonialism.

Landlocked countries often find it difficult to participate in the most basic forms of trade due to high overland travel costs. Authors have explored the effects of logistics cost on exchange traded goods and how this has forced many countries to withdraw from trade they might otherwise benefit [4,5,18]. Others have revealed that improved neighbor relations can help stimulate growth and trade [7,19]. While these studies have provided necessary information about oceanic access, it is important to explore other areas that relate to maritime trade participation, economic geography, and prosperity.

1.2. Maritime trade access

Maritime trade has evolved throughout the twentieth and twenty- first centuries. Currently, there are more than 4500 deep-water ports worldwide [Fig. 2]. What effect do these ports have on an economy? Clott and Wilson [20] determined that ports work as a source of tax money and job security in most port towns and cities. DeSalvo [21] stated that the shutdown of a port would cause all local production associated with that port to cease due to the direct impact of a port on the local economy. Focusing on Chinese cities, Shan Yu and Lee [22] found that increasing cargo flow positively correlates with economic growth in urban centers and large ports have more impact than smaller ports. Interestingly, this study determined that seaports positively impact communities in neighboring port cities as well.

Other authors found similar results in Europe. According to Bottasso Conti, Ferrari, and Tei [23]; ports positively impact local development and have a spillover effect into neighboring economies. Chang, Shin, and Lee [85] found similar results in South Africa, stating that the port sector

significantly impacts the domestic economy and employment across the country. Focusing on the Pearl River Delta, Yangtze River Delta, and the Bohai Rim Economic Zone in China, Chen [24] concluded that the clustering of port infrastructure in these regions has attracted foreign direct investment, created jobs, promoted consumption, and stimulated economic development. Lane [25] found that foreign-trade zones clus- tering near ports in the Southeastern United States had a significant impact on spatial development patterns on the county level. On a global scale, Munim and Schramm [26] concluded that high quality port infrastructure and improved logistical performance had a positive impact on the national economy of 91 countries included within the study.

The classic study of maritime access and sea power is that of the American Admiral Alfred Thayer Mahan [2]. Mahan identified six fac- tors that affect sea power: geographical position, physical conformation, extent of territory, population, national character, and government. By geographical position Mahan was largely referring to coastal access as well as a state’s geographical position with respect to other, especially potentially rival or enemy, states. Physical conformation refers to the nature of the coastline, including the number and quality of harbors. As to extent of territory, Mahan argued that it is not the size of a country, but the ability to defend it. By population and national character Mahan noted that a population must be of sufficient size to populate a navy, and of an outward looking and sea-oriented character. Its government must also be supportive of maritime goals. Though Mahan’s work was largely concerned with sea power and the military role of the oceans, some of his observations, such as those on geographical position, have an application to maritime trade as well. Mahan’s thoughts on sea power and security have been extended and amplified by more recent scholars, including Grygiel [27] and Germond [28].

There is a growing awareness of the impact of geographic distance on market interaction. Tinbergen [29] offered one of the first analyses of

Fig. 2. Deep-water maritime ports around the world (data source: [45]).

J.M. Lane and M. Pretes

Marine Policy 121 (2020) 104180

4

geographic location and international trade when he formulated the gravity model of trade. This model determined that economies of great scale tend to attract other economies and countries that are closer to one another tend to trade with each other. In recent years, other scholars have detailed the importance of understanding geographic proximity of trade patterns in order to make international trade more effective. Hall and Jacobs [30] based their study on the gravity model of trade and suggested that the impacts of proximity are insurmountable when determining trade relations. However, Boschma [31] argued that too little and too much proximity can have a negative impact on innovation and development. This paper also discussed differences amongst geographical, institutional, social, organizational, and cognitive prox- imity as a force for growth. Shaw and Gilly [32] agreed with this assessment, arguing that geographical, organizational, cognitive, and social proximity are important dynamics when explaining economic development and exchange relationships between economic actors.

Understanding the idea of economic gravity, Frankel and Romer [33] found that as distance declined between trading partners, interaction grew. Others determined that distances between countries are inversely proportionate to trade and income [34]. Robinson [35] concluded that maritime port proximity should be maximized in order to increase capital and trade flows. Access to water provides an avenue for in- stitutions to participate in trade. Christiansen, Fagerholt, Nygreen, and Ronen [36] developed a strategic plan for shipping in maritime trade after discovering that seaborne transportation of goods and materials had increased by 67% since 1980. Even in an era with expanding technological innovation, the world still depends on maritime travel for the transportation of commodities over great distances.

Other scholars determined that with all things being equal, lateral trade flows and distance to other markets explained as much as 70% of the variation in median income [37]. According to Overman, Redding, and Venables [38]; “the evidence surveyed here strongly suggests the importance of geography in determining international economic in- teractions, in influencing cross-country income distribution, and in shaping the structure of production across space” (p. 34) Other scholars found that as traded goods move farther from the source, the cost of transporting those goods increases exponentially and impacts trade volumes between the two trading partners [39]. Hoffman and Kumar [40] later determined that new technology has helped to reduce the number of crew members on board ocean liners, thereby reducing the overall cost of maritime transportation.

These scholars have provided profound insight on the study of in- ternational trade. Involvement in international trade is pivotal today in order to remain relevant in the world economy. By participating in trade, lesser developed countries gain access to goods that they would otherwise not be able to produce at home. They also participate in the creation of goods that would otherwise be too expensive to produce in other countries. Because most of these goods are transported in cargo ships and tankers over international waterways [1], the study of mari- time trade is critical when establishing sound policy advice for lesser developed countries.

Considering the importance of international trade, access to mari- time trade, and geographic proximity to markets, this study focuses on the impact of maritime dependency on economic development across the world. Few scholars have compared the number of ports, merchandise trade ratio (dependence on trade), coastline length, coast/ area ratio, and shipping liner connectivity to GDP per capita. The cre- ation of a Maritime Dependency Index from the results in this analysis will allow the academic community to evaluate historical changes to global maritime trade dependency.

2. Materials and methods

There are as many as 195 self-governing institutions and 72 depen- dent territories throughout the world [41]. Data were collected from the Central Intelligence Agency’s (CIA) World Factbook, World Bank,

United Nations Conference on Trade and Development (UNCTAD), and the World Port Source. All data included in this analysis were compiled from 2016 demographics [41–45]. Data were compiled using Microsoft Excel and results were calculated using IBM SPSS Statistics Version 26. Due to a lack of port data from Waters [45] and shipping liner con- nectivity data from Barki and Délèze-Black [42]; 129 countries were included in the analysis [Fig. 3] Unfortunately, this excluded landlocked countries from this analysis. A serious lack in data can present problems in global evaluations, especially when considering maritime trade access to landlocked countries. For this reason, data will be compiled in the future to better evaluate the maritime dependency of landlocked countries.

The most effective way to measure the relationship between one dependent and several independent variables is multiple regression. Multiple regression explores the nature and strength of the relationship between multiple independent variables and a single dependent vari- able. According to Gomez and Jones, [46]; “regression describes the co-variation of two variables measured on an interval/ratio scale. A simple regression means there is only one independent variable” (p. 301). However, in the case of multiple independent variables “an interval/ratio scale can be related to many independent variables simultaneously by expressing it as a linear function of several variables” (p. 304). The resulting equation created by the model can be used to estimate and predict future results.

The factors included in this study indicate how geospatially posi- tioned a country is within the world of maritime trade. By using multiple regression to compare the number of ports, merchandise trade ratio, coastline length, coast/area ratio, and shipping liner connectivity score to gross domestic product (GDP) per capita, this paper will be able to effectively describe the relationship between these factors and economic development. The following formula was used in this analysis:

y= β0 + β1x1 + β2x2 + …βn− 1xi− 1 + βnxi + Σi

where β0 is the intercept, βn is the coefficient estimate slope for variable xi, and Σi is the error term.

Variables included in this analysis provide a description of a coun- try’s ability to participate in maritime trade. Some variables such as coast/area ratio and coastline length are geographic features that pro- mote better access to maritime trade. Other factors such as number of ports and liner shipping connectivity provide evidence of infrastructure and current activity within maritime trade, while merchandise trade ratio reveals a country’s dependence on trade as a percentage of total production.

Coast/area ratio is determined by dividing the total coastline length in meters by the area of landmass in kilometers squared ( m

km2). A country with a high coast/area ratio has easier access to coastal areas, making participation in trade more easily obtainable. For the purpose of this study, coastline length is measured in total kilometers. Analysis is confined to the study of maritime coastline. Larger coastline lengths allow countries to build more ports and increase shipping across greater distances. For that reason, the number of deep-water ports per country were also included in this analysis.

Deep-water ports are considered important infrastructure and directly relate to the ability of a country to participate in maritime trade. Deep-water ports are defined by water depth of at least 30 feet deep at the docking area [47]. This depth allows for the docking of large vessels ocean-going vessels. Small shallow ports were excluded in order to limit the number of ports to those that transport goods between countries. Merchandise trade is calculated by adding imports to exports and dividing by the country’s GDP of the same year (imports+exports

GDP ) [44]. The liner connectivity index is a database including connections of liner services between deep-water ports. This index does not include indirect connections through third parties, instead it only contains direct con- nections between participant countries [48].

These independent variables were compared to GDP purchasing

J.M. Lane and M. Pretes

Marine Policy 121 (2020) 104180

5

power parity (PPP) per capita. The GDP PPP “is the sum of all goods and services produces in the country valued at prices prevailing in the United States” [41]. This allows for an easy and comprehensive comparison of economies across different exchange rates. GDP per capita takes the total GDP in a given year and divides it by the number of citizens of that same year.

The valued performance of the multiple regression model within a given set was evaluated using Akaike Information Criterion (AIC) [49]. At the current scale, the sample sizes were large enough and no null values were present, so no adjustments were made to the AIC score. This data was compiled in order of validity. Coefficient estimates and stan- dard errors were compiled and compared to show the significance of each model. Determination coefficients (R2 values) were also included to demonstrate the amount of variation explained in each model.

After completing the multiple regression model, data from the best- fit model were indexed as a percentage of the maximum value obtained (Index Value = Actual Value

Maximum Value). GDP per capita, number of deep-water ports, merchandise trade ratio, and liner shipping connectivity indices were calculated separately by taking each country’s value and dividing by the maximum value of each category. The final index number was determined by calculating the mean value of all four categorical indices. The Human Development Index was used as a model for the Maritime Dependency Index and similar methods were used to determine the final index scores. This paper used similar rationale as the Human Develop- ment Index for applying non-weighted indices in the final composite index. According to Haq [50]; weights are not recommended in the Human Development Index because there is insufficient rationale for discriminating amongst each variable. Other authors have suggested the use of factor analysis [51] or principal component analysis [52] as al- ternatives to non-weighted indices. However, for the purpose of simplicity and inadequate evidence for applying weights, this study constructed a non-weighted composite index.

3. Results and discussion

3.1. Multiple regression

All independent variables (number of ports, merchandise trade, coast/area ratio, coastline length, and liner shipping connectivity) were included in the original model and were compared with GDP per capita. The best-fit model was determined by comparing AIC score and the model with the lowest value was chosen as the best-fit model. Results from the multiple regression model reveal a significant relationship between the dependent variable and three of the original five indepen- dent variables [Table 1]. These results show a confidence level of at least

Fig. 3. Countries included in the study of maritime dependency (data source: [41–45]).

Table 1 Model averaged parameter estimates of maritime dependency and gross do- mestic product per capita, from 129 countries.

Model Set Kb AIC ΔAIC R2 p-value

GDP Per Capita PORTS + MT 2 2907.4 10.34 0.19 <0.0001 PORTS þ MT þ LSCI 3 2897.06 0 0.27 < 0.0001 PORTS + MT + CAR + LSCI 4 2898.5 1.44 0.27 <0.0001 PORTS + MT + CAR + COAST +

LSCI 5 2900.1 3.04 0.27 <0.0001

Model terms are number of deep-water ports (PORTS), merchandise trade ratio (MT), coast/area ratio (CAR), coastline length (COAST), and liner shipping connectivity index (LSCI). LIN = Linear Relationship. Kb = Number of parameters in model. AIC = Akaike’s Information Criterion Score ​ ’. ΔAIC = The difference between AIC value of the best supported model and the successive models. All models in the 95% confidence are included. Bold row represents the best fit model.

J.M. Lane and M. Pretes

Marine Policy 121 (2020) 104180

6

95% (α = 0.05). The resulting best-fit model reveals a significant rela- tionship between the factors included (p-value < 0.0001) and the r- squared analysis explained 27% of the variation within the model (R2 =

0.27). Slope coefficient estimates for ports, merchandise trade, and liner shipping connectivity were all positive (Ports = 80.63; Merchandise Trade = 155.96; Liner Shipping Connectivity = 129.98) [Table 2]. The following formula represents the best-fit model:

ŷ = 4342.57 + 80.63χ1 + 155.96χ2 + 129.98χ3

where ŷ is the projected values for GDP per capita, χ1represents an in- dividual value for number of deep-water ports, χ2 represents an indi- vidual value for merchandise trade ratio, and χ3 represents an individual value for shipping liner connectivity index. According to this equation, when the number of ports increase by one, GDP per capita increases by $80.63. When the merchandise trade ratio increases by one percentage point, GDP per capita increases by $155.95. Finally, as the liner shipping connectivity index increases by one, GDP per capita increases by $129.98.

While these results show a significant and positive relationship be- tween GDP per capita, number of deep-water ports, merchandise trade ratio, and shipping liner connectivity, regression is just a tool used for describing a relationship and predicting future outcomes. Therefore, these results do not describe a cause-and-effect relationship. Addition- ally, linear regression assumes a constant linear relationship, which is not usually the case regarding income. Easterlin [53] argues that diminishing returns occurs when “a $1000 increase in real income be- comes progressively smaller the higher the initial level of income” (p. 243). A similar effect is apparent when considering maritime dependency.

After analyzing regression results, variables were placed in scatter- plots and quadratic fit lines were used to analyze diminishing returns. The port quadratic scatterplot reveals a plateau effect after the 200 mark. Countries with less than 200 ports have a slope coefficient of 148.91 and countries with greater than 200 ports have a slope coeffi- cient of 49.24. Interestingly, the merchandise trade ratio quadratic plot reveals increasing returns for countries with ratios higher than 150. Countries with a ratio less than 150 have a slope coefficient of 99.35 and countries greater than 150 have a slope coefficient of 166.595. Finally, the shipping liner connectivity quadratic scatterplot shows diminishing returns for countries with a connectivity higher 75. Countries with a connectivity score of less than 75 have a slope coefficient of 273.08 and countries greater than 75 have a slope coefficient of − 232.93. These results provide some clarification to the problems within global linear regression models, which overlook nonlinear relationships that may exist between variables.

3.2. Maritime Dependency Index

Maritime trade accounts for most international trade today and without quality ports to dock and distribute these goods, countries would find it difficult to prosper [86]. Findings from this study reveal that deep-water ports, merchandise trade ratios, and shipping liner connectivity have a significant impact on GDP per capita in countries across the world. Due to the significance of these findings, the dependent variable (GDP per capita) and independent variables (number of deep-water ports, merchandise trade ratio, and shipping liner connec- tivity) were indexed and these values were averaged to create a com- posite Maritime Dependency Index. After indexing the variables, each country was given a rank according to its composite index value. In the initial model presented in this paper, each variable was given equal weight. Because the design of this index is based on methods used to create the Human Development Index and there is insufficient rationale for discriminating amongst variables, weights were not applied to the Maritime Dependency Index.

These results were compiled into a table, joined with a world map vector file, and shaded according to natural breaks using ArcMap GIS software [Fig. 4]. The United States, Hong Kong, and Singapore rank at the top, while Guinea-Bissau, Comoros, and the Democratic Republic of the Congo rank at the bottom of the index [Table 3]. Some countries standout as deviant cases. For example, Indonesia ranks 42nd in Mari- time Dependency, yet it has 154 ports. This is largely due to its GDP per capita of $11,612, merchandise trade ratio of 30.09, and shipping liner connectivity of 29.62. According to Boddin [54]; between 1995 and 2011, Indonesia ranked behind Brazil, China, India, Mexico, and Turkey (other newly industrialized countries) in exports and imports but ranked third amongst these countries in total output, as many of its goods are consumed domestically. This can have a major impact on its merchan- dise trade ratio, as it measures total trade as a percentage of GDP. Dutch colonialism has had a lasting impact on domestic economic growth as well. Yazid [55] concluded that “the social and political chaos in the first decade after independence in 1949 influenced the economic situation in Indonesia” and “most of the basic facilities and economic foundations developed by the Dutch during the colonial period were destroyed during the independence struggle after the war” (p. 83). Infrastructure development has been a major concern for Indonesia while countries such as the United States already have built infrastructure in place.

Other deviant cases such as India have similar reasons for ranking low on the Maritime Dependency Index. India currently has 76 deep- water ports, ranking 18th worldwide, and a shipping liner connectiv- ity score of 58.17, ranking 21st worldwide. However, it ranks 45th on the Maritime Dependency Index. India’s index score was heavily weighted by its GDP per capita of $6,527 in 2016 (ranking 92nd) and merchandise trade ratio of 27.57 (ranking 118th). India’s relatively low GDP per capita is due in part to the legacy of colonialism. Gupta [56] argued that its belated independence from Britain and its concentration in the tertiary sector dominated by foreign firms has kept individual income low compared to its European and North American counterparts. Additionally, in 2016, India’s population was approximately 1.325 billion [57] and 72.43% of its products were consumed domestically. Its large population and land area of approximately 3.29 million square kilometers [41] has had an impact on India’s merchandise trade ratio.

Brazil is another example where the legacy of colonialism, popula- tion, and landmass play a major role in its Maritime Dependency Index rank. Ranking 54th on the Maritime Dependency Index, Brazil has 81 ports and a shipping liner connectivity of score of 38.89. However, in 2016, it had a GDP per capita of $15,128 and a merchandise trade ratio of 18.3. Between 1995 and 2011, Brazil lagged behind other newly industrialized countries such as China, India, and Turkey in exports, imports and total output but still remained a major producer within this group of countries [54]. This is due in part to its lag in economic development in the 19th century during the Industrial Revolution and post-independence. According to Barros [58]; the era after

Table 2 Slope and Standard Error coefficient estimates for maritime dependency and gross domestic product per capita, from 129 countries.

Model terms

PORTS MT CAR COAST LSCI

80.63 (24.78) 155.96 (44.02)

129.98 (66.67)

81.25 (24.84) 160.08 (44.46)

− 1.44 (1.97)

122.66 (67.54)

74.46 (27.20) 159.36 (44.59)

− 1.43 (1.97)

0.06 (0.09)

127.23 (68.11)

107.60 (20.78)

191.21 (40.58)

Where parameter estimates have been omitted in the table, those parameters were not in the model. Models are listed in order of performance (best to worst). Data were collected from 125 countries around the world. Bold parameter estimates are significant at p < 0.05 Model terms are number of deep-water ports (PORTS), merchandise trade ratio (MT), coast/area ratio (CAR), coastline length (COAST), and liner shipping connectivity index (LSCI).

J.M. Lane and M. Pretes

Marine Policy 121 (2020) 104180

7

independence gave countries like the United Kingdom, United States, Canada, Australia and New Zealand a head start over Brazil. This article also concluded that highly educated immigrants from Europe gave the United States, Australia, and New Zealand the human capital needed to develop between the 18th and 19th centuries. This has had a major impact on GDP per capita in Brazil to this day. Additionally, population and landmass play a major role in Brazil’s position on the Maritime Dependency Index. As of 2016, Brazil had a population of approximately 206.2 million [59] and controlled 8.52 million square kilometers of territory [41], both of which may explain why it consumed approxi- mately 81.7% of its products domestically.

Data discussed thus far provide context for a world dependent on maritime trade for global development. The Maritime Dependency Index is an attempt to provide statistical evidence that a significant relationship exists between global development patterns and a country’s ability to participate in maritime trade. In order to better understand the relationship between economic development and maritime dependency, future studies are needed using longitudinal data to further analyze the validity of the Maritime Dependency Index.

4. Conclusions

In this study, multiple regression was used as a tool to analyze impact of five independent variables on GDP per capita. A total of 129 countries were included in the analysis and results reveal that GDP per capita is significantly dependent on the number of ports, merchandise trade ratio, and shipping liner connectivity. The remaining independent variables have been collectively termed maritime dependency. While there may not be a cause and effect relationship between these factors, this study has shown a clear relationship between maritime dependency and prosperity: as maritime dependency increases, GDP per capita increases. The dependent variable and three independent variables included in the best-fit model were indexed as a ratio to the highest value in each category and presented on a scale between 0 and 1. Values from each category were averaged and together and a Maritime Dependency Index

score was tabulated for each country. Results from this index place United States, Hong Kong, Singapore, United Kingdom, and China at the top of the Maritime Dependency Index [Table 3].

Location is an important aspect of any geographic study. In the study of international trade, an emphasis has been placed in the importance of economic policy. While this is an important aspect of international trade and prosperity, location has a profound impact on the ability of a country to participate in the international market. Results from this analysis show just how important deep-water ports, trade as a percent- age of GDP, and shipping liner connectivity are to a given economy. Of these three independent variables, deep-water ports had the most sig- nificant relationship with GDP per capita. According to DeSalvo [21]; “in the absence of the local port, local users of noncomparable imports would have to pay higher prices for imports since they would enter through alternative ports, thereby incurring higher inland trans- portation costs.” In order for countries to import products that cannot easily be produced domestically or export products at a competitive value, ports and access to oceanic waters are necessary functions of a successful institution.

International trade is often seen as a competition, a zero-sum game between competing interests. This overlooks the underlying result of international trade: prosperity. If everyone has access to deep-water ports and maritime trade, the cost of transporting materials throughout the world decreases. If the cost of transportation decreases, then profits increase, and prices decrease. This allows people to have access to goods from around the world and helps to elevate the economic output of countries participating in trade. As various Liberalist scholars have argued, the key here is state preferences, rather than capacities (see, e.g., Ref. [60–62]. Policy choices therefore become more important than limitations imposed by geography and resource endowments.

The next step in the study of maritime dependency is to perform a longitudinal index so that changes in maritime dependency can be tracked throughout the past few decades. The resulting index can be used evaluate the evolution of maritime trade for each country and develop infrastructure improvement plans for developing countries. The

Fig. 4. Choropleth map showing each country’s Maritime Dependency Index score (data source: [41–45]).

J.M. Lane and M. Pretes

Marine Policy 121 (2020) 104180

8

current Maritime Dependency Index is a preliminary model that will need some reconfiguring to provide the best measurements for maritime trade access in the future.

With as much as 75% of international travel and trade still taking place over international waterways, it is easy to see why it is important for a country to be located near deep-water [1]. A more geographic analysis reveals that while a few landlocked countries have benefited from technological advances by participating in banking and money markets abroad, access to navigable oceanic waterways is still necessary for most of the world. This study has shown that as the ability of a country to participate in maritime trade increases, GDP per capita in- creases. Maritime dependency reflects this idea and shows that there is a direct relationship between location and economic success.

5. Limitations

Due to a lack in data between variables, results analyzed in this study are limited by the number of observations and the year at which the study was conducted. For this reason, generalization outside of the 129 countries included in this study is not recommended. While a major portion of this study was dedicated to discussing landlocked countries, few of these countries were included in this analysis due to this lack in data. In order to investigate the validity of the quantitative results from

this study, future analyses are recommended to further evaluate the impact of maritime dependency on economic prosperity.

Authorship contributions

Jesse M. Lane and Michael Pretes conceived of the presented idea. Jesse M. Lane developed the theory and performed the computations. Michael Pretes encouraged Jesse M. Lane to investigate the impact of the variables included in this study on economic development and super- vised the findings of this work. Jesse M. Lane carried out the statistical analysis and evaluated the findings. Michael Pretes revised important aspects of the literature review portion and Jesse M. Lane revised the methods, analysis, and discussion sections of the paper. Jesse M. Lane and Michael Pretes reviewed revisions to ensure validity.

Declaration of competing interest

No declaration of interest.

Acknowledgements

No acknowledgments to declare.

Appendix

Table 3 Maritime Dependency Index and its components.

Rank Maritime Dependency Index

GDP Per Capita Deep-Water Ports

Merchandise Trade Ratio

Liner Shipping Connectivity Index

(value) (PPP $) (total #) (I + E/GDP) (value)

2016 2016 2016 2016 2016

HIGH MARITIME DEPENDENCY 1 United States 0.515 57467 552 19.87 93.63 2 Hong Kong 0.513 58553 2 331.57 100.5 3 Singapore 0.508 87856 2 212.14 119.53 4 United Kingdom 0.427 42609 391 39.51 93.63 5 China 0.383 15535 172 32.91 170.85 6 United Arab Emirates 0.359 72419 9 140.76 73.12 7 Belgium 0.345 46383 8 164.69 86.09 8 Italy 0.343 38161 311 46.60 62.82 9 Japan 0.341 41470 292 25.35 73.9 10 Netherlands 0.339 50898 24 138.51 84.39 11 France 0.334 41466 268 43.57 67 12 Germany 0.323 48730 98 68.82 89.76 13 Qatar 0.308 127523 6 58.58 7.35 14 Malaysia 0.307 27681 25 120.66 102.49 15 Canada 0.296 44025 239 52.75 42.14 16 South Korea 0.291 35751 17 63.89 112.55 17 Denmark 0.287 49496 159 58.94 50.1 18 Spain 0.274 36310 105 48.50 81.44 19 Sweden 0.252 49175 82 54.49 53.18 20 Saudi Arabia 0.241 54431 18 49.22 61.11 21 Vietnam 0.239 6424 15 171.00 61.85 22 Ireland 0.226 68883 49 68.16 11.83 23 Slovenia 0.219 32885 3 141.96 31.5 24 Brunei 0.217 77441 1 69.24 8.86 25 Malta 0.214 37899 2 83.78 51.87 26 Australia 0.208 46790 106 32.15 30.14 27 Poland 0.206 27811 12 84.82 55.8 28 Greece 0.199 26783 103 39.98 47.75 29 Norway 0.197 59302 83 43.70 7.1 30 Portugal 0.188 30624 29 60.22 47.77 31 Thailand 0.188 16917 21 100.63 47.29 32 Iceland 0.187 51399 67 50.65 11.83 33 Turkey 0.184 24244 50 39.50 57.15 34 Russia 0.181 23163 105 36.90 41.07

(continued on next page)

J.M. Lane and M. Pretes

Marine Policy 121 (2020) 104180

9

Table 3 (continued )

Rank Maritime Dependency Index

GDP Per Capita Deep-Water Ports

Merchandise Trade Ratio

Liner Shipping Connectivity Index

(value) (PPP $) (total #) (I + E/GDP) (value)

2016 2016 2016 2016 2016

35 Lithuania 0.175 29966 1 122.52 15.84 36 Mexico 0.172 17862 42 73.69 42.73 37 Finland 0.171 43053 60 49.64 14.85 38 Estonia 0.170 29365 26 120.42 6.6 39 Panama 0.164 23015 13 52.13 50.66 40 Israel 0.164 37901 5 40.60 39.02 41 Croatia 0.159 23596 34 70.46 30.22 42 Indonesia 0.159 11612 154 30.09 29.62 43 Morocco 0.157 7838 15 62.06 59.92 44 Chile 0.153 23960 46 48.34 33.52 MEDIUM MARITIME DEPENDENCY 45 India 0.153 6572 76 27.57 58.17 46 Romania 0.152 23626 18 73.60 28.87 47 Trinidad and Tobago 0.151 31908 11 76.73 17.63 48 Seychelles 0.148 28391 1 105.15 8.46 49 Sri Lanka 0.144 12316 4 36.53 61.91 50 New Zealand 0.144 39059 25 37.74 18.88 51 Cyprus 0.140 32580 4 53.77 22.85 52 Egypt 0.139 11132 28 24.42 58.71 53 Latvia 0.138 26031 6 96.04 8.2 54 Brazil 0.137 15128 81 18.30 38.89 55 Congo, Rep. 0.134 5719 4 108.83 26.45 56 Mauritius 0.132 21088 1 57.64 31.99 57 Argentina 0.132 19934 55 20.78 35.72 58 Namibia 0.130 10585 2 99.82 22.8 59 Colombia 0.128 14158 15 27.01 49.93 60 Ukraine 0.127 8272 18 80.96 28.48 61 Bulgaria 0.126 19199 2 103.01 6.96 62 South Africa 0.125 13225 10 56.41 35.45 63 Lebanon 0.124 13996 2 46.14 41.65 64 Philippines 0.122 7806 68 47.00 27.91 65 Peru 0.121 13022 33 38.13 35.16 66 Uruguay 0.118 21625 10 28.92 33.35 67 Jordan 0.116 9050 3 69.11 30.31 68 Equatorial Guinea 0.113 25535 9 71.13 3.47 69 Bahamas, The 0.110 23173 8 26.98 27.7 70 Cambodia 0.107 3736 2 113.41 8.85 71 Dominican Republic 0.105 15209 14 38.20 27.6 72 Togo 0.104 1491 2 76.95 28.71 73 Ecuador 0.104 11286 9 33.59 35.76 74 Honduras 0.103 4738 6 85.51 17.92 75 Suriname 0.100 14146 2 82.17 6.47 76 Tunisia 0.095 11599 8 78.54 6.6 77 Jamaica 0.095 8835 15 42.46 26.57 78 Nicaragua 0.094 5541 7 89.52 8.85 79 Iraq 0.092 17354 4 48.17 13.57 80 Belize 0.091 8448 1 82.13 8.23 81 Angola 0.091 6499 16 41.79 26.83 82 Saint Kitts and Nevis 0.090 26686 2 42.25 3.53 83 Montenegro 0.090 16854 1 60.49 7.46 84 Georgia 0.090 9997 4 65.43 12.89 85 Algeria 0.090 15075 18 47.54 10.91 86 Marshall Islands 0.089 4072 2 98.72 4.2 87 Fiji 0.089 9561 5 68.93 11.22 88 Guyana 0.089 7819 6 82.44 6.2 89 Costa Rica 0.089 16614 6 43.10 14.51 LOW MARITIME DEPENDENCY 90 Gabon 0.086 18108 6 46.92 8.68 91 Maldives 0.083 13199 1 56.38 9.42 92 Pakistan 0.082 5249 3 24.23 35.45 93 Ghana 0.081 4294 4 57.71 18.73 94 Mozambique 0.081 1217 5 77.72 11.94 95 Antigua and Barbuda 0.078 22414 1 39.83 2.27 96 Barbados 0.076 16816 1 47.23 4.87 97 Solomon Islands 0.075 2236 14 70.29 7.47 98 El Salvador 0.074 8619 3 56.68 9.03 99 Mauritania 0.074 3854 2 70.68 8.41 100 Albania 0.074 11929 4 55.89 4.29 101 Guatemala 0.074 7947 5 39.94 17.47 102 Tonga 0.073 5752 3 70.23 4.78 103 Senegal 0.072 2568 4 55.29 16.27 104 Liberia 0.070 813 4 70.47 9.18

(continued on next page)

J.M. Lane and M. Pretes

Marine Policy 121 (2020) 104180

10

Table 3 (continued )

Rank Maritime Dependency Index

GDP Per Capita Deep-Water Ports

Merchandise Trade Ratio

Liner Shipping Connectivity Index

(value) (PPP $) (total #) (I + E/GDP) (value)

2016 2016 2016 2016 2016

105 Madagascar 0.070 1506 11 50.14 16.44 106 Saint Vincent and the Grenadines 0.069 11606 5 49.66 4.8 107 Micronesia 0.067 3593 2 74.27 1.92 108 Saint Lucia 0.066 11546 4 46.44 4.29 109 Vanuatu 0.065 3081 2 61.40 8.21 110 Haiti 0.063 1784 11 55.07 8.61 111 Dominica 0.062 10975 3 40.68 5.91 112 Grenada 0.062 13928 1 35.99 4.57 113 Nigeria 0.061 5867 12 17.74 21.3 114 Yemen 0.060 2508 8 28.44 20.7 115 Sierra Leone 0.060 1473 3 58.76 7.87 116 Benin 0.059 2168 1 42.00 15.6 117 Myanmar 0.059 5773 5 42.10 9.32 118 Samoa 0.059 6345 1 51.71 4.59 119 Guinea 0.058 1311 2 55.67 8.39 120 Bangladesh 0.053 3581 4 36.04 11.34 121 Tanzania 0.052 2787 13 30.97 11.95 122 Kiribati 0.052 2047 3 52.88 4.42 123 Gambia, The 0.050 1689 1 49.25 6.07 124 Cameroon 0.049 3286 1 26.31 14.9 125 Sudan 0.048 4730 3 11.94 19.54 126 Kenya 0.047 3156 3 28.07 12.31 127 Guinea-Bissau 0.045 1582 2 43.98 5.36 128 Comoros 0.043 1522 3 40.54 5.53 129 Congo, Dem. Rep. 0.034 801 3 31.94 4.62

DEFINITIONS Maritime Dependency Index (MDI): a composite index measuring the access and

openness to maritime trade using the average comparisons of three factors – number of deep-water ports, merchandise trade ratio, liner shipping connectivity, and Gross Domestic Product per capita using purchasing power parity calculations. Deep-water port: a port with cargo water depth of at least 30 feet deep at the docking area that is used to dock, load, and unload cargo from large ocean freight vessels and river barges. Merchandise Trade Ratio: found by adding the total exports and imports of a country and dividing it by the Gross Domestic Product of that same year.

Liner Shipping Connectivity: an index developed by UNCTAD that considers the direct distance between liner service port connections. Gross Domestic Product (GDP) Per Capita: the total production output of a country each year divided by the total population in that same country. GDP Per Capita is measured in US $ using the purchasing power parity. This parity measures income by comparing total output to the going rate for currency exchange. This is a common measurement for measuring average yearly income

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.marpol.2020.104180.

References

[1] H.O. Heiberg, The merchant fleet: a facilitator of world trade, in: R.Z. Lawrence, M. D. Hanouz, S. Doherty (Eds.), Global Enabling Trade Report 2012: Reducing Supply Chain Barriers, World Economic Forum, Geneva, Switzerland, 2012, pp. 85–90.

[2] A.T. Mahan, The Influence of Sea Power upon History, twelfth ed., Little, Brown and Company. (Original work published 1890), Boston, MA, 1949, pp. 1660–1793.

[3] S.C. Davis, S.E. Williams, R.G. Boundy, Transportation Energy Data Book, 37th ed., Oak Ridge National Laboratory, Oak Ridge, TN, 2016. https://tedb.ornl. gov/wp-content/uploads/2020/02/Edition37_Full_Doc.pdf Labratory.

[4] J.F. Arvis, Transit and the special case of landlocked countries, in: L. De Wulf, J. B. Sokol (Eds.), Customs Modernization Handbook, The World Bank, Washington, DC, 2005, pp. 243–264.

[5] J.F. Arvis, G. Raballand, J.-F. Marteau, The Cost of Being Landlocked Logistics, Costs, and Supply Chain Reliability, The World Bank, Washington, DC, 2010.

[6] F. Carmignani, The curse of being landlocked: institutions rather than trade, World Econ. 38 (10) (2015) 1594–1617, https://doi.org/10.1111/twec.12248.

[7] M.L. Faye, J.W. McArthur, J.D. Sachs, T. Snow, The challenges facing landlocked developing countries, J. Hum. Dev. 5 (1) (2004) 31–68, https://doi.org/10.1080/ 14649880310001660201.

[8] T.N. Srinivasan, The costs and benefits of being a small, remote, island, landlocked, or ministate economy, World Bank Res. Obs. 1 (2) (1986) 205–218.

[9] P.J. Katzenstein, Corporatism and Change: Austria, Switzerland, and the Politics of Identity, Cornell University Press, Ithaca, NY, 1984.

[10] A.J. Moore, Quantifying the landlocked trade penalty using structural gravity, J. Quant. Econ. 16 (3) (2018) 769–786.

[11] G. Raballand, Determinants of the negative impact of being landlocked on trade: an empirical investigation through the Central Asian case, Comp. Econ. Stud. 45 (4) (2003) 520–536.

[12] A. Chowdhury, S. Erdenebileg, Geography against Development: A Case for Landlocked Developing Countries, vol. 6, United Nations Office of the High Representative for the Least Developed Countries, New York, NY, 2006.

[13] P. Collier, The Bottom Billion: Why the Poorest Countries Are Failing and what Can Be Done about it, Oxford University Press, New York, NY, 2007.

[14] P. Carmody, The New Scramble for Africa, second ed., Polity Press, Cambridge, UK, 2016.

[15] S. Michalopoulos, E. Papaioannou, The long run effects of the scramble for Africa, Am. Econ. Rev. 106 (7) (2016) 1802–1848.

[16] C. Alden, A.C. Alves, China and Africa’s Natural Resources the Challenges and Implications for Development and Governance (SAIIA Report Occasional Paper No. 41; pp. 1–28), Governance of Africa’s Resources Programme, 2009.

[17] S. Oloruntoba, Africa and the dilemma of transformation: exploring trade for the realisation of the right to development, Afr. Insight 47 (3) (2017) 59–77.

[18] M. Kashiha, J.C. Thill, C.A. Depken II, Shipping route choice across geographies: coastal vs. landlocked countries, Transport. Res. E Logist. Transport. Rev. 91 (2016) 1–14, https://doi.org/10.1016/j.tre.2016.03.012.

[19] B. Lahiri, F.K. Masjidi, Landlocked countries: a way to integrate with coastal economies, J. Econ. Integrat. 27 (4) (2012) 505–519.

[20] C. Clott, G.S. Wilson, Ocean shipping deregulation and maritime ports: lessons learned from airline deregulation, Transport. Law J. 26 (2) (1998) 205–218.

[21] J.S. DeSalvo, Measuring the direct impacts of a port, Transport. J. 33 (4) (1994) 33–42.

[22] J. Shan, M. Yu, C. Lee, An empirical investigation of the seaport’s economic impact: evidence from major ports in China, Transport. Res. E Logist. Transport. Rev. 69 (2014) 41–53, https://doi.org/10.1016/j.tre.2014.05.010.

[23] A. Bottasso, M. Conti, C. Ferrari, A. Tei, Ports and regional development: a spatial analysis on a panel of European regions, Transport. Res. Pol. Pract. 65 (2014) 44–55, https://doi.org/10.1016/j.tra.2014.04.006.

[24] Y. Chen, Survey on influence of maritime port cluster effect on offshore reginal economy based on Grey Correlation Model, J. Coast Res. 94 (sp1) (2019) 707–711.

J.M. Lane and M. Pretes

Marine Policy 121 (2020) 104180

11

[25] J.M. Lane, Foreign-trade zones in the Southeastern United States: do they promote economic development or lead to spatial inequality? SE. Geogr. 60 (2) (2020) 139–156.

[26] Z.H. Munim, H.J. Schramm, The impacts of port infrastructure and logistics performance on economic growth: the mediating role of seaborne trade, J. Shipp. Trade 3 (1) (2018) 1–19, https://doi.org/10.1186/s41072-018-0027-0.

[27] J. Grygiel, Geography and seapower, in: P. Dutton, R.S. Ross, Ø. Tunsjø (Eds.), Twenty-first Century Seapower: Cooperation and Conflict at Sea, Routledge, Milton Park, UK, 2012, pp. 18–41.

[28] B. Germond, The (critical) geopolitics of seapower, in: The Maritime Dimension of European Security: Seapower and the European Union, Palgrave Macmillan, London, UK, 2015, pp. 19–32.

[29] J. Tinbergen, Shaping the World Economy: Suggestions for an International Economic Policy, Twentieth Century Fund, New York, NY, 1962.

[30] P.V. Hall, W. Jacobs, Shifting proximities: the maritime ports sector in an era of global supply chains, Reg. Stud. 44 (9) (2010) 1103–1115, https://doi.org/ 10.1080/00343400903365110.

[31] R. Boschma, Proximity and innovation: a critical assessment, Reg. Stud. 39 (1) (2005) 61–74, https://doi.org/10.1080/0034340052000320887.

[32] A.T. Shaw, Jean-Pierre Gilly, On the analytical dimension of proximity dynamics, Reg. Stud. 34 (2) (2000) 169–180, https://doi.org/10.1080/00343400050006087.

[33] J.A. Frankel, D. Romer, Does trade cause growth? Am. Econ. Rev. 89 (3) (1999) 379–399.

[34] D.A. Irwin, M. Terviö, Does trade raise income? Evidence from the twentieth century, J. Int. Econ. 58 (1) (2002) 1–18, https://doi.org/10.1016/S0022-1996 (01)00164-7.

[35] R. Robinson, Ports as elements in value-driven chain systems: the new paradigm, Marit. Pol. Manag. 29 (3) (2002) 241–255, https://doi.org/10.1080/ 03088830210132623.

[36] M. Christiansen, K. Fagerholt, B. Nygreen, D. Ronen, Maritime transport, in: C. Barnhart, G. Laporte (Eds.), Handbooks in Operations Research and Management Science, vol. 14, Elsevier, Amsterdam, Netherlands, 2007, pp. 189–284.

[37] S. Redding, A.J. Venables, Economic geography and international inequality, J. Int. Econ. 62 (1) (2004) 53–82, https://doi.org/10.1016/j.jinteco.2003.07.001.

[38] H.G. Overman, S. Redding, A.J. Venables, The Economics Geography of Trade Production and Income: A Survey of Empirics (Discussion Paper Series No. 2978; International Trade), Centre for Economic Policy Research, 2003.

[39] N. Limao, A.J. Venables, Infrastructure, geographical disadvantage, transport costs, and trade, World Bank Econ. Rev. 15 (3) (2001) 451–479.

[40] J. Hoffman, S. Kumar, Globalization—the maritime nexus, in: C.T. Grammenos (Ed.), Handbook of Maritime Economics and Business, second ed., Lloyd’s List, London, UK, 2013, pp. 35–64.

[41] The World Factbook, The World Factbook, Central Intelligence Agency, 2017. https://www.cia.gov/library/publications/the-world-factbook/index.html.

[42] D. Barki, L. Délèze-Black (Eds.), Review of Maritime Transport. Geneva, Switzerland: United Nations Conference on Trade and Development, 2017.

[43] The World Bank, GDP Per Capita (Current US$), The World Bank, 2018. htt ps://data.worldbank.org/indicator/NY.GDP.PCAP.CD.

[44] The World Bank, Merchandise Trade (% of GDP), The World Bank, 2018. https ://data.worldbank.org/indicator/TG.VAL.TOTL.GD.ZS.

[45] M.E. Waters, World Ports by Country, World Port Source, 2018. http://www.worl dportsource.com/countries.php.

[46] B. Gomez, J.P. Jones III, Explanatory statistics, in: Research Methods in Geography: A Critical Introduction, Wiley-Blackwell, Malden, MA, 2010, pp. 301–304.

[47] S. Dasgupta, What Are Deep Water Ports? Marine Insights, 2019. https://www.mar ineinsight.com/ports/what-are-deep-water-ports/.

[48] M. Fugazza, J. Hoffman, Liner shipping connectivity as determinant of trade, J. Shipp. Trade 2 (1) (2017) 1–18, https://doi.org/10.1186/s41072-017-0019-5.

[49] M.J. Crawley, Statistics: an Introduction Using R, second ed., John Wiley & Sons, Chichester, UK, 2015.

[50] M. Haq, The birth of the human development index, in: S. Fukuda-Parr, A.K. Shiva Kumar (Eds.), Readings in Human Development: Concepts, Measures and Policies for a Development Paradigm, Oxford University Press, Oxford, UK, 2003, pp. 127–137.

[51] M. Noble, G. Wright, G. Smith, C. Dibben, Measuring multiple deprivation at the small-area level, Environ. Plann. 38 (1) (2006) 169–185.

[52] D. Lai, Principal component analysis on Human Development indicators of China, Soc. Indicat. Res. 61 (3) (2003) 319–330.

[53] R.A. Easterlin, Diminishing marginal utility of income? Caveat emptor, Soc. Indicat. Res. 70 (3) (2005) 243–255.

[54] D. Boddin, The Role of Newly Industrialized Economies in Global Value Chains, International Monetary Fund, 2016, pp. 1–38. WP/16/207; IMF Working Paper.

[55] M.N.M. Yazid, Colonial policy and the impact to the politico-economy stability after independence: the case of Indonesia under the Dutch and Malaysia under the British, Rev. Hist. Polit. Sci. 2 (3) (2014) 69–84.

[56] B. Gupta, Falling behind and catching up: India’s transition from a colonial economy, Econ. Hist. Rev. 72 (3) (2019) 803–827, https://doi.org/10.1111/ ehr.12849.

[57] The World Bank, Population, Total - India, The World Bank, 2020. https://data. worldbank.org/indicator/SP.POP.TOTL?locations=IN.

[58] A.R. Barros, Historical origins of Brazilian relative backwardness, Braz. J. Polit. Econ. 35 (1) (2015) 75–94.

[59] The World Bank, Population, Total - Brazil, The World Bank, 2020. https://data. worldbank.org/indicator/SP.POP.TOTL?locations=IN.

[60] M.W. Doyle, Liberalism and world politics, Am. Polit. Sci. Rev. 80 (4) (1986) 1151–1169.

[61] B. Jahn, Liberal internationalism: from ideology to empirical theory—and back again, Int. Theory 1 (3) (2011) 409–438.

[62] E.B. Shiraev, International Relations, Oxford University Press, New York, NY, 2014.

[85] Young-Tae Chang, Sung-Ho Shin, Paul Lee, Economic impact of port sectors on South African economy: an input–output analysis, Trans. Pol. 35 (2014) 333–340, https://doi.org/10.1016/j.tranpol.2014.04.006.

[86] Jon Helmick, Port and maritime security: a research perspective, J. Transportation Secur. 1 (1) (2008) 15–28, https://doi.org/10.1007/s12198-007-0007-3.

J.M. Lane and M. Pretes