Financial Management

Public Financial Management Systems and Countries' Governance: A Cross-Country Study. 

This content may contain URLs/links that would redirect you to a non-EBSCO site. EBSCO does not endorse the accuracy or accessibility of these sites, nor of the content therein.

Contents

1. Good Governance

2. Public Financial Management in Central Governments

3. Accrual accounting and budgeting

4. Auditing and transparent budgeting process

5. Data and Methodology

6. Variable and data sources

7. Sample and methodology

8. Analysis of Results

9. Discussion and Conclusions

10. Acknowledgements

11. Footnotes

12. References

Full Text

Summary: The objective of this study is to explore whether a relationship exists between public financial management (PFM) systems and expert perceptions of countries' governance in an international cross‐country study. We examine the extent to which variations in accounting, budgeting and auditing practices are associated with governance in a sample of 97 countries that represent different levels of development, analysing the differences between countries classified into factor, efficiency and innovation‐driven economies. Our concept of governance perception includes three dimensions: accountability, government effectiveness and corruption. We find that countries with a higher level of economic development show, on average, more sophisticated PFM systems characterized by the presentation of accrual‐based financial statements, the application of value for money audits and higher budget transparency. When analysing the sub‐samples of countries according to the level of economic development, we find that countries with similar governance perception scores show different patterns of PFM practices, suggesting that there is no one‐size‐fits‐all approach. Copyright © 2015 John Wiley & Sons, Ltd.

governance; accrual accounting; accrual budgeting; audit; budget transparency

Public financial management (PFM) practices have been one of the central issues in the agenda of public management reforms in many countries. The greater demand for transparency, accountability and efficiency forces governments all over the world to adopt new accounting, auditing and budgeting techniques and to improve the scrutiny and timeliness of their reporting. The link between the modernization of PFM and the quality of governance has not been thoroughly explored yet in an international cross‐country setting. For instance, a gap needs to be filled as regards matters of governance and accounting (Broadbent and Guthrie, [ 11 ] ). In a similar vein, the relationship between particular governance perception measures, such as corruption and accounting, has received little academic attention (Alawattage et al., [ 1 ] ).

This study is aimed at contributing to the incipient body of research about the relationship between expert perceptions of governance and several characteristics of PFM. Selecting PFM characteristics, following Pollitt and Bouckaert ([ 47 ] ), we consider three areas: accounting, budgeting and auditing. We focus on the degree of adoption of accrual‐based financial statements and budgeting, the timeliness of the release of the audited financial statements, the scope of value for money (VFM) audit and the transparency of budgeting processes as relevant PFM characteristics, which may influence governance perceptions.

The measurement of the quality of governance has attracted numerous research efforts. A multitude of governance indicators have been developed, reflecting the fact that there is no consensus on the criteria for measuring good governance and there is little agreement on what constitutes high‐quality government (Nanda, [ 40 ] ; Andrews, [ 3 ] ; Fukuyama, [ 18 ] ). Given these difficulties in conceptualization, we focus on three main characteristics of governance perceptions that are likely to be influenced by PFM, namely, voice and accountability, governmental effectiveness and corruption perception.

In our analysis, we also take into account the level of economic development of countries. Countries with a higher level of development are expected to have, on average, both better governance perception measures and a higher degree of sophistication of their PFM. Our first objective is to provide a test for this proposition. Our second objective is to analyse the relationship between selected PFM characteristics and perceptual indicators of good governance, testing the degree of association between them.

Notwithstanding, each country is endowed with its particular institutional setting, influenced by its historical, legal and cultural heritage, which may foster or inhibit the development of particular PFM and influence governance outcomes. We take into account the institutional divergence of countries when interpreting our findings.

Good Governance

Good governance is also known as quality of government; referring to what makes some government institutions look better than others. In this study, rather than searching for an all‐encompassing definition of governance, we focus on those good governance characteristics that are often cited as the intended outcomes of PFM reforms. For instance, the International Public Sector Accounting Standards Board (IPSASB, [ 28 ] : 11) takes the three fundamental principles of corporate governance, openness, integrity and accountability, and redefines them, adapting them to the context of the public sector. The International Federation of Accountants (IFAC, [ 26 ] : 5) also stresses the importance of integrity and openness. One of the ways to achieve them is by implementing good practices in transparency, reporting and auditing to deliver effective accountability.

The World Bank defines governance as the traditions and institutions by which authority in a country is exercised for the common good (Kaufmann, [ 33 ] ) and measures its quality across six dimensions, namely, Voice and Accountability, Governmental Effectiveness, Control of Corruption, Political stability and absence of violence, Regulatory quality and Rule of law. We focus on the first two dimensions as well as on the perception of corruption provided by Transparency International because we argue they may have the closest relationship with PFM. Pragmatically, we define good governance, as that which is deemed free of corruption, is accountable and transparent regarding its policies, and effective in the delivery of public services.

With respect to the measurement of governance, the indicators can be classified as either perception‐based or as facts‐based, with the former being the most frequently used (Buduru and Pal, [ 12 ] ). Perception‐based indicators typically reflect perceptions of governance outcomes but not their causes or the mechanisms involved. The abovementioned World Bank governance indicators are perception‐based and constructed on the data from opinion surveys of citizens, business elites and experts about countries' governance.

Public Financial Management in Central Governments

Accrual accounting and budgeting

Central governments in many countries are increasingly implementing accrual accounting systems. Countries such as New Zealand and Australia pioneered the introduction of accruals into governmental accounting in the 90s, motivated by the limitations of traditional cash‐based systems for tracking the efficiency of spending. The major limitation of cash‐based accounting is that while this system accounts for short‐term cash flows, such as ongoing receipts and payments, it ignores financial flows that do not involve the movement of cash, such as depreciation, write‐off of assets and accrued interest. The latter elements are important if the full costs of governmental services and their effectiveness have to be analysed, for instance in the cases of capital purchase assessment or pension liabilities (Diamond, [ 17 ] ). Governmental entities hold a variety of capital assets, including office buildings, infrastructure, heritage, military equipment and natural resources. Apart from short‐term liabilities, governments owe long‐term obligations such as bonds payable and employee pensions payable. Accrual accounting provides information about capital assets and long‐term solvency (Chan, [ 14 ] ).

The literature identifies two main groups of benefits attributable to the implementation of an accrual accounting system (Jagalla et al., [ 31 ] ). The first group comprises improved internal and external transparency. In several empirical studies (Alt and Lassen, [ 2 ] ; Benito and Bastida, [ 8 ] ), accrual‐based financial statements are considered a measure of fiscal transparency. This is because an accrual accounting system requires a full statement of assets and liabilities as well as revenues and expenses (Diamond, [ 17 ] ).

The second group of benefits is related to decision making. Accrual accounting helps to identify the full costs of activities, enabling improved decision making in resource allocation, enhanced governmental control and better capital investment decisions. For instance, in capital purchase assessment, it is critical to know which assets are owned by the agencies, what costs they generate and at what value they can be replaced. For Warren ([ 53 ] : 9), accrual accounting integrates different systems (tax assessment and debtors, a system for keeping track of purchases made on credit, payroll and employee entitlements, physical asset management and a debt management system) more cohesively than cash accounting, while providing the same important cash information.

Notwithstanding, there is no consensus about the usefulness of accrual accounting in the public sector. Standard setting bodies such as IPSASB have made great efforts in the diffusion of accrual‐based practices. They are also supported by international organizations such as the International Monetary Fund (IMF), the Organization for Economic Cooperation and Development (OECD) and the World Bank. The Asian Development Bank has generally supported, although not necessarily encouraged, developing member countries in their efforts to adopt the accrual basis (Athukorala and Reid, [ 6 ] ). For these institutions, accrual‐based financial reporting represents a benchmark practice that ensures a certain standard of accountability and transparency. Some authors (Deng et al., [ 16 ] ) have also recommended the adoption of accrual accounting in countries known for low levels of fiscal transparency such as China.

PwC ([ 48 ] ) reported the potential benefits of adopting a harmonized accrual accounting system. Navarro‐Galera and Rodríguez Bolívar ([ 41 ] ) have shown, on the basis of a comparison between fair value accounting and historical cost accounting, that IPSAS‐based public sector accounting increases the usefulness of government financial statements for decision making in both developed and developing countries. Kober et al. ([ 35 ] ) conducted a survey among Australian public officials which revealed that generally accepted accounting principles (GAAP)‐based accrual accounting information is perceived as more understandable and useful for financial management than a cash‐based system and government finance statistics accounting. For Torres ([ 52 ] ), the accrual basis can fit—and is fitting—into countries with different levels of development, and accounting can be regarded as a weapon against fraud and waste in governments in developing countries.

Yet, some experts are sceptical about the value of accrual accounting in the public sector, indicating that the evidence about the usefulness of accruals is weak, that the system can produce biased estimates of the costs and that it is complex for lay persons and politicians to understand (Carlin, [ 13 ] ). The accrual‐based system generates even more controversy in some developing countries, as they might not even count with the basic structure of a cash‐based system and where corruption is so high that the choice of one accounting system or another is irrelevant.

Accrual budgeting is defined as the use of accrual accounting records and measures in the budgeting process (Lüder and Jones, [ 37 ] : 35). That is to say, instead of the money received and payments to be made, the accrual budget is structured around revenues earned and liabilities incurred. Some jurisdictions that have operated under dual systems (accrual accounting and cash or modified cash budgeting) have noted that this may hinder the acceptance of the accrual‐based accounting and extensive reconciliations are required between the two systems (IFAC, [ 25 ] : 33; IMF, [ 27 ] : 28–29). Many academics and practitioners (Paulson, [ 44 ] ; Meszarits and Seiwald, [ 39 ] ) have pointed out the scarce usefulness of annual accounts—especially the balance sheet and the income statement—when the budget is prepared under the cash basis. A desire to promote greater transparency and accountability at the national level has prompted a number of countries (Australia, New Zealand and the UK) to align the accounting standards across budgets, statistics and accounts on the accrual basis (IMF, [ 27 ] ). Jesus and Jorge ([ 32 ] ) show that, when translating from government accounting (microeconomic perspective) to National Accounting (macroeconomic perspective), the cash‐accrual adjustments are more diverse and material in relation to the deficit/surplus in countries still having cash‐based budgetary reporting, raising questions concerning the reliability of the deficit/surplus they report.

In practice, however, the rate of adoption of accrual budgeting remains lower than that of accrual accounting (PwC, [ 48 ] ). This reluctance to adopt accrual budgeting can be attributed to its technical complexity, as perceived by politicians that meet to discuss budget appropriations in parliament (Blöndal, [ 10 ] ).

Auditing and transparent budgeting process

A further important dimension of an advanced government PFM system is a well‐developed auditing system, which ensures the veracity of financial reporting and the efficient usage of government funds. By providing unbiased, objective assessments of whether public resources are managed responsibly and effectively to achieve the intended results, auditors help public sector organizations achieve accountability and integrity, improve operations and instil confidence among citizens and stakeholders (Institute of Internal Auditors, [ 23 ] ). The timeliness of financial reporting is considered an essential aspect of good financial management practice (Knack et al., [ 34 ] ).

Financial or compliance audit is concerned with ensuring the legality and regularity of financial reporting. VFM audit, also known as the performance audit, seeks to verify the effectiveness and efficiency of governmental programmes and the use of public funds (INTOSAI, [ 29 ] ). The latter audit form has gained more prominence recently, as governments are under increased pressure to justify their spending between competing welfare programmes (Pollitt, [ 46 ] ). The results of both compliance audits and VFM audits are released publicly, which makes them important accountability devices. Increased accountability and more effective auditing are likely to reduce opportunities for certain forms of corruption and fraud (Hameed, [ 20 ] ).

The public availability of budgetary information and its timely public release is the first requisite for the interested social groups, citizens and opposition politicians to be engaged in the monitoring of public expenditures (Robinson, [ 51 ] ). Perhaps, the best‐known initiative in measuring budgetary transparency belongs to the International Budget partnership (IBP), which publishes the Open Budget Index (OBI) constructed with data gathered by independent budget specialists. According to IBP, three quarters of all the countries surveyed in 2012 fail to meet standards of budget transparency. Although it is very difficult to establish direct causality between transparency and governance outcomes, a greater level of transparency in the public domain is associated with better quality governance (Islam, [ 30 ] ), with reduced corruption (Bastida and Benito, [ 7 ] ) and better socio‐economic and human development indicators (Kaufmann, [ 33 ] ).

In sum, we highlight that the modernization of PFM systems has its origins in the desire to improve the quality of governance by providing greater accountability and transparency for external users and better information sources for improving the decision making of internal and external users. Our contention is that the features of PFM described above are likely to be positively associated with governance perception.

Data and Methodology

Variable and data sources

The source for most of the PFM variables is the OECD International Budget Practices and Procedures Database 2007/2008 , which was updated by the OECD on the basis of in‐depth dialogue with member countries in late 2008 and 2009. The OECD reports quality and reliability analyses of the information gathered. Indeed, these data have been used in other studies, such as Bastida and Benito ([ 7 ] ) and Andrews ([ 3 ] ). We collected the variables (Table [NaN] ) that measure the percentage of the expenditures that are reported on an accrual basis in financial statements and in the budget. We also collected the timeliness of audited financial statements, which measures when the accounts audited by the central audit institution are made publicly available, and the number of VFM audit reports published every year. For 50 of the countries in our sample, we included the OBI drawn up by IBP. This index measures the amount, level of detail and timeliness of budget information, which governments make publicly available.

Variables

Variable	Description	Source
Financial statement basis	On what reporting basis are the financial statements presented to the legislature? Approximate percentage of total on‐budget expenditure in the financial statements on accruals	Organization for Economic Cooperation and Development (OECD) International database of Budget Practices and Procedures 42/2008 . (Question 62)
Budget accounting basis	On what reporting basis is the budget presented to the legislature?	OECD International database of Budget Practices and Procedures 42/2008 . (Question 62)
Approximate percentage of total on‐budget expenditure on accruals
Open Budget Index (OBI)	A measure of budget transparency and accountability.	International Budget Partnership. /
Timeliness of audited financial information publicly available	When are the accounts audited by the central audit institution publicly available? From 0 (no publicly available), 1 (12 or more months after the end of the fiscal year) … to 12 (1 month after the end of the fiscal year).	OECD International database of Budget Practices and Procedures 2007/2008 . (Question 70)
Value for money (VFM) audit	Number of VFM audit reports published every year.	OECD International database of Budget Practices and Procedures 2007/2008 . (Question 67)
Voice and accountability	This measure is composed of the following sub‐indicators: democracy index, vested interests, accountability of public officials, human rights, freedom of association.	The Economist Intelligence Unit. .
.
Government effectiveness	This measure is composed of the following sub‐indicators: quality of bureaucracy or institutional effectiveness, and excessive bureaucracy or red tape.	The Economist Intelligence Unit. .
.
Corruption perception index (CPI)	This variable measures the perceived level of corruption, as determined by expert assessments and opinion surveys. From 0 (highly corrupt) to 10 (highly clean).	Transparency International /.
Stage of development	It takes three values. 1: factor‐driven and: factor to efficiency; 2: efficiency‐driven and efficiency to innovation; 3: innovation‐driven.	Global Competitiveness Report 2008/2009 (p. 22).
.
Gross Domestic Product (GDP) per capita	GDP based on purchasing‐power parity per capita. Current international dollar.	International Monetary Fund, World Economic Outlook Database, April 2012.

Since the OECD budget survey was conducted in 2007/2008, to analyse accounting and governance data for the same period, we collected governance variables for the year 2008. The indicator Voice and Accountability, provided by the Economist Intelligence Unit, is based on expert assessments and includes, among many other indicators, accountability of public officials, democracy index and vested interests. The second dimension is that of Government Effectiveness, which captures perceptions of the competence of the bureaucracy, the quality of public service delivery, the independence of the civil service from political pressures and the credibility of the government's commitment to policies. The third dimension of governance is the Corruption Perception Index (CPI), provided by Transparency International. CPI ranks countries in terms of the degree to which corruption is perceived to exist among public officials and politicians. The CPI includes several indicators, including bribery of public officials, kickbacks in public procurement, embezzlement of public funds or questions about the strength and effectiveness of anti‐corruption efforts, thereby encompassing both the administrative and political aspects of corruption.

Sample and methodology

The sample is made up of 97 countries (Table [NaN] ), which present different levels of economic development. The stage of economic development reached may influence what can be expected as good standards of governance in each country, as well as the specific weight that accounting and budgeting systems play in the national institutional framework (Holmberg et al., [ 22 ] ; Andrews, [ 3 ] ).

Countries included in the sample

The whole sample is made up of 97 countries. Factor‐driven countries, efficiency‐driven countries and innovation‐driven countries are, in total, 87 countries. There are 10 other countries (Congo, Fiji, Guinea, Haiti, Liberia, Papua New Guinea, Rwanda, Sierra Leone, Solomon Islands and Swaziland) included in the whole sample that are not included in the classification of the WEF Competitiveness Report.

Factor‐driven countries (27): Benin, Bolivia, Burkina Faso, Cambodia, Ethiopia, Ghana, Indonesia, Kenya, Kyrgyzstan, Lesotho, Madagascar, Malawi, Mali, Moldova, Mongolia, Mozambique, Nigeria, Philippines, Tajikistan, Uganda, Vietnam, Zambia and Zimbabwe. In this factor‐driven group, we have also included those countries that are in transition from stage 1 to 2: Botswana, Jordan, Morocco and Venezuela.

Efficiency‐driven countries (29): Albania, Argentina, Bosnia and Herzegovina, Brazil, Bulgaria, Costa Rica, Mauritius, Mexico, Namibia, Peru, Rumania, Serbia, South Africa, Suriname, Thailand, Tunisia, Ukraine, Uruguay, Lithuania, Qatar and Taiwan. In this efficiency‐driven group, we have also included those countries that are in transition from stage 2 to 3: Chile, Croatia, Hungary, Latvia, Poland, Russian Federation, Slovak Republic and Turkey.

Innovation‐driven countries (31): Australia, Austria, Belgium, Canada, Cyprus, Czech Republic, Denmark, Finland, France, Germany, Greece, Hong Kong, Iceland, Ireland, Israel, Italy, Japan, South Korea, Luxemburg, Malta, Netherlands, New Zealand, Norway, Portugal, Slovenia, Spain, Sweden, Switzerland, United Arab Emirates, UK and USA.

We, therefore, divided the sample of countries into three groups according to their respective stage of development as revealed by the World Economic Forum (WEF, [ 54 ] , p. 8–9). This report in fact distinguishes five stages of development: factor‐driven, factor to efficiency, efficiency‐driven, efficiency to innovation and innovation‐driven, respectively. However, as the number of countries in the second stage amounted to just four, and only eleven were assessed as belonging to the fourth stage, we considered the countries in stages 1 and 2 and then those in stages 3 and 4 together.

To obtain some initial insights from the data, we conducted descriptive and exploratory analyses of the variables described above. Furthermore, to measure the level of linear association between the variables used in the survey, bivariate Spearman correlation coefficients were determined and a graphical analysis was conducted to highlight within‐group differences.

Analysis of Results

As can be seen in Table [NaN] , for most variables, the average values presented by innovation‐driven countries are higher than those presented by efficiency‐driven economies, which, in turn, are higher than those presented by factor‐driven countries. This is the case for the use of accrual‐based financial statements, the OBI and the application of VFM audits. These results suggest that countries which have a higher level of income also present, on average, a higher application of more sophisticated PFM systems.

Descriptive statistics

		Whole sample (N = 97)	Factor‐driven countries (N = 27)	Efficiency‐driven countries (N = 29)	Innovation‐driven countries (N = 31)
Accrual‐based financial statements	Average	35.48 (N = 87)	20.00 (N = 20)	35.52 (N = 29)	43.68 (N = 31)
(Range 0–100)	SD	47.38	41.03	47.29	49.17
Accrual‐based budget	Average	28.53 (N = 86)	14.29 (N = 21)	32.14 (N = 28)	31.80 (N = 30)
(Range 0–100)	SD	44.94	35.85	47.55	45.99
Accrual‐based financial statementsAccrual‐based budget	Average	2178.30 (N = 88)	1428.57 (N = 21)	2068.97 (N = 29)	2635.16 (N = 31)
(Range 0–10 000)	SD	4131.47	3585.68	4122.50	4424.54
Open Budget Index (OBI)	Average	48.72 (N = 50)	34.82 (N = 17)	55.44 (N = 18)	76.70 (N = 10)
(Range 1–88)	SD	24.70	19.72	12.89	9.77
Timeliness of audited financial information publicly available	Average	5.63 (N = 87)	5.30 (N = 23)	5.19 (N = 26)	6.81 (N = 31)
(Range 0–12)	SD	3.77	4.40	3.38	3.49
Value for money (VFM) audit	Average	13.85 (N = 68)	7.45 (N = 19)	13.11 (N = 19)	20.32 (N = 22)
(Range 0–100)	SD	24.27	22.87	19.49	24.75
Voice and accountability	Average	0.60 (N = 89)	0.42 (N = 25)	0.60 (N = 28)	0.81 (N = 30)
(Range 0–1)	SD	0.21	0.12	0.14	0.13
Government effectiveness	Average	0.43 (N = 90)	0.20 (N = 25)	0.39 (N = 28)	0.71 (N = 31)
(Range 0–1)	SD	0.26	0.15	0.18	0.14
Corruption perception index	Average	4.74 (N = 96)	2.91 (N = 27)	4.35 (N = 29)	7.39 (N = 31)
(Range 1.4–9.3)	SD	2.28	0.92	1.20	1.40

1 †It is an interaction variable consisting of two variables (accrual‐based financial statements and accrual‐based budget) being multiplied in order to test the conjoint effect of them.

As expected, for the whole sample of countries and also within each of the sub‐samples, countries show a higher degree of accrual‐based financial reporting than accrual‐based budgets. The three sub‐samples differ significantly in the degree of preparation of accrual‐based financial statements, albeit only at a very moderate level (p‐value = 0.087). However, efficiency‐driven countries show a higher average application of accrual budgeting than is shown by innovation‐driven countries.

Budget transparency is, as expected, greater in countries that are at higher stage of development, and the Kruskal–Wallis test indicates that the mean differences among the three groups are statistically significant (p‐value = 0.000). According to the IBP report, countries that perform poorly on the OBI are mainly low‐income countries with weaker democratic institutions and a marked dependence on foreign aid (de Renzio and Masud, [ 50 ] ).

In the case of the timeliness of audited financial information, the number of months required by the innovation‐driven countries to make their accounts publicly available is lower (on average) than that required by the efficiency‐driven and factor‐driven countries. However, the differences of means among the three sub‐groups for this variable are not significant according to the Kruskal–Wallis test.

The application of VFM audits varies significantly at the three stages of development and, as expected, the higher the level of a country's development, the greater the number of VFM audits conducted by the central audit institution in each country. The differences in the mean value of this variable at the three economic stages are significant (p‐value = 0.005). Likewise, a considerable increase in the number of VFM audits is recorded when moving from the factor‐driven to the efficiency‐driven group.

In the case of the governance variables, the Kruskal–Wallis test confirms the significance of the differences in the mean values of the three perception measures studied at the three stages of development (p‐value = 0.000). These results are evidences of the link between the countries' governance and the higher per capita income levels. The differences between countries at the three stages of development are more marked in the case of governance variables than they are for the PFM variables.

The Spearman test (Table [NaN] ) provides interesting insights into the correlations between the PFM variables, on the one hand, and the governance measures, on the other.

Spearman correlations by countries' stage of development

	Voice and Accountability	Government Effectiveness	Corruption Perception Index
	Whole sample	Factor	Efficiency	Innovation	Whole sample	Factor	Efficiency	Innovation	Whole sample	Factor	Efficiency	Innovation
Accrual‐based financial statements	0.270	0.187	0.009	0.359	0.205	0.282	−0.267	0.423	0.214	0.022	−0.188	0.523
N = 80	N = 18	N = 28	N = 30	N = 81	N = 18	N = 28	N = 31	N = 86	N = 20	N = 29	N = 31
Accrual‐based budget	0.366	0.439	0.240	0.428	0.223	0.551	−0.053	0.199	0.223	0.248	0.081	0.429
N = 78	N = 19	N = 27	N = 29	N = 79	N = 19	N = 27	N = 30	N = 85	N = 21	N = 28	N = 30
Accrual‐based financial statements Accrual‐based budget	0.342	0.439	0.075	0.451	0.200	0.551	−0.237	0.216	0.236	0.248	−0.056	0.456
N = 80	N = 19	N = 28	N = 30	N = 81	N = 19	N = 28	N = 31	N = 87	N = 21	N = 29	N = 31
Open Budget Index (OBI)	0.770	0.398	0.329	0.226	0.575	0.202	0.056	0.617	0.675	0.261	0.197	0.378
N = 46	N = 16	N = 18	N = 9	N = 47	N = 16	N = 18	N = 10	N = 49	N = 17	N = 18	N = 10
Timeliness of audited financial information publicly available	0.302	0.064	0.451	0.158	0.300	0.188	0.273	0.117	0.297	−0.127	0.249	0.352
N = 81	N = 21	N = 26	N = 30	N = 82	N = 21	N = 26	N = 31	N = 86	N = 23	N = 26	N = 31
Value for money (VFM) audit	0.429	−0.077	−0.069	0.317	0.469	0.354	0.274	0.297	0.403	0.042	0.055	0.348
N = 60	N = 17	N = 18	N = 21	N = 61	N = 17	N = 18	N = 22	N = 67	N = 19	N = 19	N = 22
Gross Domestic Product	0.764	0.182	0.452	0.376	0.813	0.100	0.568	0.306	0.874	0.222	0.514	0.517
N = 89	N = 25	N = 28	N = 30	N = 90	N = 25	N = 28	N = 31	N = 96	N = 27	N = 29	N = 31

· 2 *significant at 5%.

· 3 significant at 1%.

· 4 significant at 0.1%.

· 5 It is an interaction variable consisting of two variables (accrual‐based financial statements and accrual‐based budget) being multiplied in order to test the conjoint effect of them.

For the whole sample of countries, all the PFM variables are positively linked with the three governance perception measures. On the budget transparency side, there is a strong relationship between the OBI and all the governance indicators for the whole sample of countries. All these correlations are above 0.5 and highly significant (p‐value = 0.000). These results are consistent with those obtained in the IBP report ([ 24 ] : 51), which indicated a strong, positive correlation between the OBI and the World Bank's indicator on Voice and Accountability. However, as can be seen in Table [NaN] , the correlations between OBI and governance measures are not significant when they are performed for the sub‐samples of countries. This might be due to the low number of cases involved in each sub‐sample.

The application of VFM audits also shows significant correlations at the 0.1% level with the three governance measures for the whole sample, and the highest correlation is found with government effectiveness (0.469). These results may indicate the fact that effective audit institutions at the central government level play an important role in governance. By ensuring that financial information is faithful and made publicly available in a timely manner, governments signal their accountability and commitment to effective policies. In other words, the positive and significant sign of VFM audit indicates that the higher the level of central government spending that is audited by applying economy, efficiency and effectiveness criteria, the higher the countries' perceived good governance scores. As in the case of the OBI, the correlations of the VFM audits are not significant for the sub‐samples.

Having analysed the descriptive statistics of the variables and the correlations among the whole sample, we now look in greater detail at the most significant associations and the differences among the sub‐samples. Firstly, we also analysed the possible relationships among the PFM variables by sub‐samples, and we find that there are more significant correlations among the innovation‐driven countries than among factor‐driven or efficiency‐driven countries. The data suggest that innovation‐driven countries seem to converge slightly more in the application of more sophisticated PFM practices than the two other sub‐samples. For instance, in innovation‐driven countries (Table [NaN] ), accrual‐based financial statements show a strong correlation with budget transparency (0.864), VFM audit (0.746), and although much more moderately, with timeliness of publicly available audited information (0.379). However, in efficiency‐driven and factor‐driven countries, the bivariate correlations between these same variables are lower, or even negative and nonsignificant.

Spearman correlations within innovation‐driven countries

	1	2	3	4	5	6	7
1. Accrual‐based financial statements	1
2. Accrual‐based budget	0.555	1
3. Accrual‐based financial statements Accrual‐based budget.	0.655	0.919	1
4. Open Budget index (OBI)	0.864	0.607	0.607	1
5. Timeliness of audited financial information publicly available	0.379	0.292	0.273	0.630	1
6. Value for money (VFM) audit	0.746	0.511	0.629	0.829	0.256	1
7. Gross Domestic Product (GDP) per capita	0.309	0.252	0.125	0.358	0.359	0.144	1

· 6 *significant at 5%.

· 7 significant at 1%.

· 8 significant at 0.1%.

· 9 It is an interaction variable consisting of two variables (accrual‐based financial statements and accrual‐based budget) being multiplied in order to test the conjoint effect of them.

In terms of the relationships between PFM variables and governance perception measures, the results show that the accrual basis is associated with governance scores in the two extreme groups of countries (i.e. innovation‐driven and factor‐driven countries). Innovation‐driven economies show the most significant correlations between the application of the accrual basis and governance measures. It seems that the greater the reporting on the accrual basis by central governments in their budgets and their financial statements, the higher the governance scores. The conjoint preparation of accrual‐based budgets and accrual‐based financial statements (interactive variable) is significant at 1% when analysing the innovation‐driven countries or, at a lower magnitude, the whole sample, but not for efficiency‐driven and factor‐driven countries. This may be due to the fact that it is in the innovation‐driven group where we find the highest number of countries that present both accrual‐based financial statements and accrual budgeting. As can be seen in Table [NaN] , the mean of this interactive variable for innovation‐driven countries is the highest.

The highest correlation for innovation‐driven countries is found between the accrual basis of financial statements and the CPI (0.523) at 1%. However, as illustrated in Figure [NaN] , we find different patterns in terms of the joint analysis of accrual‐based financial statements and corruption perception in innovation‐driven countries. This figure exhibits two dichotomous groups of innovation‐driven countries: those who prepare accrual‐based financial statements and those who do not. New Zealand, Sweden, Australia, Canada, Iceland and the UK are the countries that drive a positive relationship between accrual‐based financial statements and the CPI (i.e. the higher the accruals, the lower the perception of corruption). Some of these countries, such as New Zealand (Pallot, [ 43 ] ) and Sweden (Lundqvist, [ 38 ] ), have extensive experience in the preparation of accrual‐based financial statements and score among the best in terms of corruption perception. Experience reflected through the number of years presenting and using accrual‐based financial statements may matter. A ‘learning effect’, that is, preparers and users who learn over time to recognize and value the information generated by the accrual system, may have a positive effect on governance (Kober et al., [ 35 ] ). However, a second group of countries, including Finland, Norway, Denmark, Netherlands and Ireland, among others, also score among the best in corruption perception without having fully implemented accrual‐based financial statements. This reflects that, in general, Scandinavian countries are well known for the quality of their governments, although they do not share the same pattern of PFM characteristics. At the bottom left of Figure [NaN] , we can also identify a third group of countries (Czech Republic, Greece, United Arab Emirates, Malta, South Korea…), which do not have accrual‐based financial statements and get poor scores in corruption perception. As an isolated case, Italy appears to have accrual‐based financial statements but scores poorly in terms of corruption perception. As indicated before, some other PFM variables (such as internal control systems) and non‐PFM variables (political variables and the legal system of the country) may undermine the value of accrual‐based financial statements.

For the efficiency‐driven group, the most significant correlation is found between the timeliness of audited financial information and voice and accountability (0.451). These results are consistent with the fact that the timeliness of public release of the audited statements is better in countries with higher accountability scores.

Within factor‐driven countries, the highest correlation that is significant, although at 5%, is found between accrual budgeting and government effectiveness (0.551). Figure [NaN] shows that Botswana and Philippines stand out as having the highest government effectiveness of this group and, at the same time, they are also the only two factor‐driven countries that present the budget on the accrual basis. The level of income together with the role played by international organizations, reformers and donors are a catalyst for implementing PFM reforms, as in Botswana and Philippines. Although, according to the WEF Competitiveness Reports, these two countries are between stages 1 and 2, that is, factor‐driven countries in our sub‐samples, they are not categorized as low‐income countries by the United Nations (Goldfinch et al., [ 19 ] ). However, these two countries seem to be the exception as Figure [NaN] shows that factor‐driven countries have not implemented accrual budgeting. Instead of accrual budgeting, many countries have medium‐term expenditure frameworks (MTEFs), consisting of setting out projections for both revenues and expenditures over a three to five‐year time horizon. MTEFs aim at guaranteeing the consistency of resource allocation with macroeconomic objectives and increasing the transparency of government operations over a medium‐term policy agenda. Factor‐driven countries, such as Kyrgyzstan Republic, Malawi and Uganda, have adopted MTEFs (de Renzio and Angemi, [ 49 ] ), which may explain why, without having accrual budgeting, these countries try to improve budget transparency with the technical and financial support of donor agencies.

Discussion and Conclusions

Several of the findings reported here are consistent with those made in the previous literature. First, our results show that budget transparency (OBI) is associated with other governance variables (IBP, [ 24 ] ), suggesting that the development of a country is linked to the quality of its budget reports (Bastida and Benito, [ 7 ] ). Second, there is a clear link between a country's level of income and its governance perception scores (Arndt and Oman, [ 5 ] ; Holmberg et al., [ 22 ] ).

We also show that there is a positive relationship between, on the one hand, the level of income and, on the other hand, the application of accrual‐based financial statements, VFM auditing and budget transparency practices. Among the driving forces for the adoption of governmental accounting innovation, Lüder ([ 36 ] ) mentions the socio‐economic status of accounting users. The higher this status, the greater may be the need to know the cost of public services paid by the taxpayers and whether the government is efficiently using the resources collected. This increased pressure on governments to be accountable, coupled with their greater capacity in terms of well‐developed IT systems and appropriately trained public officials, may explain why high income countries are at the forefront of PFM implementation.

The most significant correlations between PFM and governance variables are found for innovation‐driven countries, rather than for the other sub‐groups of countries. This basically indicates that, on average, innovation‐driven countries, which have higher governance perceptions, also display more sophisticated PFM practices. At the same time, these countries show different combinations of PFM practices in order to achieve lower corruption perception. For instance, Norway, Finland and the Netherlands, reported a lower degree of accrual accounting than other countries while they scored among the highest in governance perception. These countries were considering carefully the implementation of accrual accounting via pilot studies in a limited number of agencies.

In factor‐driven countries, the adoption of accrual budgeting is associated with government effectiveness, which leads us to the debate about the convenience of the adoption of accrual budgeting and other PFM reforms in developing countries. The context and resources of each specific country need to be taken into account when deciding whether to implement reforms. For Andrews ([ 4 ] ), the fit or relevance of a best practice depends on the degree of difference between the proposed adoption context and the context in which this practice emerged as best. These reforms need to be sequenced correctly in the public management reform process (Bietenhader and Bergmann, [ 9 ] ). Before accrual accounting is introduced, cash accounting should be robust, external audit should be functioning well and the legislature should be able to call the executive to account. These reforms also require a considerable quantity of resources for the hiring and training of qualified personnel, installing software and contracting consultants (Hepworth, [ 21 ] ). This is particularly relevant for developing countries where priorities are different than in developed countries. At the same time, the implementation of accrual budgeting in developing countries, without being a panacea, can set the scene for profound and durable cultural change in the public sector due to its focus on the delivery of well‐specified outputs at competitive prices and at specified quality levels (Peters, [ 45 ] ).

It can be concluded that countries with a higher level of economic development show, on average, more sophisticated PFM systems characterized by the presentation of accrual‐based financial statements, the application of VFM audits and higher budget transparency. When analysing the sub‐samples of countries according to the level of economic development, we find that countries with similar governance perception scores show different patterns of PFM practices, suggesting that there is no one‐size‐fits‐all approach.

Large sample comparative international studies are designed to capture general tendencies but are limited by the fact that they omit the nuances provided by different institutional and political frameworks, which might influence the governance mode and the way in which PFM systems are enacted. Future studies may explore in‐depth which institutional conditions are especially favourable for the fruitful relationship between governance and PFM systems.