Transitional dynamics and the evolution of information transparency: A global analysis

Williams, Andrew; Cheong, Tsun Se; Wojewodzki, Michal Article Transitional dynamics and the evolution of information transparency: A global analysis Estudios de Economía Provided in Cooperation with: Department of Economics, University of Chile Suggested Citation: Williams, Andrew; Cheong, Tsun Se; Wojewodzki, Michal (2022) : Transitional dynamics and the evolution of information transparency: A global analysis, Estudios de Economía, ISSN 0718-5286, Universidad de Chile, Departamento de Economía, Santiago de Chile, Vol. 49, Iss. 1, pp. 31-62 This Version is available at: https://hdl.handle.net/10419/285093 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc-sa/4.0/ Transitional dynamics… / Andrew D. Williams, Tsun S. Cheong, Michal Wojewodzki 31Estudios de Economía. Vol.49 - Nº1, Junio 2022. Págs. 31-62 Transitional dynamics and the evolution of information transparency: a global analysis*1 Dinámica transicional y evolución de la transparencia en la información: un análisis global Andrew David Williams** Tsun Se Cheong*** Michal Wojewodzki**** Abstract The last quarter of the 20th century was a period of sustained economic growth across many countries. Countries’ institutional arrangements have been commonly employed as factors in the convergence studies of economic growth and income levels. However, the issue of whether institutions themselves converge has been under-researched. Using the nonparametric distribution dynamics approach and a sample of 194 countries during the 1980-2010 period, we examine a tendency for countries’ informational transparency (IT) to converge over time. We find that whilst there is some evidence of unconditional convergence across countries, there is stronger evidence for convergence clubs to emerge, at both regional and income levels. Notably, the level of IT of the lowand lowermiddle-income countries and those situated in Africa, and Middle East regions tend to converge towards a level significantly below the global average. We also find a strong relationship between income and IT. Key words: Institutional convergence, information transparency, convergence clubs, distribution dynamics, mobility probability plot. JEL Classification: C4, E02, F55, P48. * We are particularly grateful to three anonymous referees and Prof. Rómulo Chumacero (Editor-in-Chief) for providing valuable comments and suggestions that significantly improved this paper. ** Business School, University of Western Australia, Perth, Australia. E-mail: andrew[email protected] *** Department of Economics and Finance, Hang Seng University of Hong Kong, Hong Kong. E-mail: [email protected] **** [Corresponding author] Department of Economics and Finance, Hang Seng University of Hong Kong, Hong Kong. E-mail: michalwoje[email protected] Received: March, 2021. Accepted: April, 2022. Estudios de Economía, Vol.49 - Nº132 Resumen Varios países presentaron un crecimiento sostenido en los últimos 25 años del siglo recién pasado. Aunque los arreglos institucionales han sido comúnmente utilizados como factores para explicar la convergencia en crecimiento y niveles de ingreso, el estudio de si las instituciones convergen no ha sido suficientemente estudiado. Utilizando el enfoque de dinámica de distribuciones no paramétricas y una muestral de 194 países en el perido 1980-2010, este trabajo examina si existe la tendencia de la transparencia en la información (IT) a converger en el tiempo. Se encuentra que, aunque hay cierta evidencia de convergencia incondicional, también la hay de clubes de convergencia a nivel regional y de ingresos. El nivel de IT de países de ingresos bajos y medios y de aquellos situados en África y el Oriente Medio tienden a converger a niveles significativamente menores que el promedio global. También encontramos una relación estrecha entre IT e ingreso. Palabras clave: Convergencia institucional, transparencia en la información, clubes de convergencia, dinámica de distribuciones, mobilidad. Clasificación JEL: C4, E02, F55, P48. 1. Introduction The issue of convergence has long fascinated economists. Whether that be a convergence of per capita incomes, income inequality, or across a range of different factors, convergence has been a way for economists to think about whether there is a tendency for countries across the world to ‘get closer’ to some common level. With respect to per capita incomes, much of the focus has been on the empirical verification of whether countries are, indeed, converging, or whether we are instead experiencing divergence (e.g., Pritchett, 1997). Although the evidence on unconditional convergence is still mixed, the idea of conditional convergence appears to have some empirical validity. Conditional convergence brings forth the idea that there are convergence clubs, whereby countries with similar initial conditions and circumstances will ultimately converge to a similar level of per capita GDP over time (e.g., Barro, 2015). One of the driving factors behind economic growth is the institutional infrastructure of a country. Therefore, having an acceptable level of institutional capacity in a country not only can maintain the law and order in the society but also offer a required environment for promoting economic growth (e.g., Keefer and Knack, 1997; Hall and Jones, 1999). More specifically, the provision of a transparent and robust institutional environment can attract investment, which, in turn, would lead to an increase in the overall income level of a country, thereby making convergence possible for the developing countries through economic growth. As a result, it calls for a thorough study of the evolution pattern of Transitional dynamics… / Andrew D. Williams, Tsun S. Cheong, Michal Wojewodzki 33 institutional information transparency (IT) of the countries so that pragmatic policy suggestions can be acquired for the formulation of development policies in promoting economic growth. This paper, however, takes a slightly different approach to this issue. Instead of investigating whether the institutional situation of countries plays a role in their convergence of per capita income, we look at the issue of convergence with respect to the informational transparency of the institutional environment itself. At a global level, such institutional convergence might be expected to be observed under periods of substantive globalization. Transactions across borders are not frictionless, being subject to differences in regulations, contract enforcement and so on, across borders. Therefore, to deepen economic integration in the way we have observed in the late 20th century period, one might plausibly expect to see a convergence in institutional information openness occurring concurrently, as countries harmonize institutional arrangements to reduce the transaction costs of cross-border flows. A second consideration relates to the degree of potential spillovers concerning the IT of institutions. If a country is surrounded by countries with stable and transparent institutions, it is certainly plausible that their institutional arrangements would be influenced by this. Conversely, being surrounded by countries with unstable and opaque institutions would make it difficult to maintain one’s institutions robust and effective in the face of this. The question of institutional convergence has received far less attention in the literature (e.g., Beyaert etal., 2019; Pérez-Moreno etal., 2020). The novel contribution of this paper is twofold. Firstly, building on the distribution dynamics methodology developed originally by Quah (1993a), we employ a new analytical framework: the Mobility Probability Plot (MPP) which has not previously been employed in a global context to look at convergence in institutional IT.1 This framework allows us to look not only at convergence, but also at the transition dynamics over time, across both income levels, and regions in a comprehensive worldwide sample of 194 countries. Secondly, we use this information on institutional distributional dynamics to project forward and investigate whether globally we can expect institutional IT convergence (or divergence) in the future. To clarify and motivate empirics, we first look at the existing literature in Section 2. Section 3 discusses the specific institutional measure employed, as well as the methodology of the distributional dynamics. Section 4 discusses our results, while Section 5 offers some thoughts on the implications and limitations of the paper. 1 Although the research output on institutional convergence has been slowly growing in recent years, to the best of our knowledge, Beyaert etal.’s (2019) study on the convergence of euro area’s institutions is the only one to employ the distribution dynamics analysis developed by Quah (1993a). Estudios de Economía, Vol.49 - Nº134 2. Literature Review Institutional capacity is important for economic growth (e.g., Barro, 1991). Deeper institutional factors, such as the existence of a sound and stable legal system, the quality of the bureaucracy, as well as outcome measures such as corruption were all put forward as being plausible reasons why some countries were able to grow faster than others. 2 Relating directly to the idea of institutions and convergence Keefer and Knack (1997) employ datasets from the International Country Risk Guide (ICRG), and Business Environmental Risk Intelligence to isolate the role of the broader institutional landscape of countries in convergence. Other papers subsequently followed, employing different methodologies and institutional datasets (e.g., Chong and Calderon, 2003; Gwartney etal., 2006). Since that time, it has been de rigueur for researchers to employ institutional quality as one of the dominant factors in the convergence literature (e.g., Ahmad and Hall, 2017). However, the issue of whether these institutions themselves converge (either unconditionally or otherwise) has been curiously under-researched. Intuitively, why might we expect institutions to converge over time? Trade and globalization have been put forward as one avenue that could potentially lead to institutional convergence. For example, the movement towards free trade in goods, services and capital may require similar regulations and laws within domestic economies to make them compatible with regulations and laws in the partner countries. It could also be argued that competition, particularly for capital, may lead to a ‘race to the top’ concerning implementing rules that facilitate inflows of FDI (La Porta etal., 2008). In the recent past there has been a concerted effort on the part of some multilateral organizations (such as the World Bank, IMF, and the European Union) to impose certain institutional or governance conditions in order for countries to qualify for loans, or to be eligible for Structural Adjustment Plans. By imposing these conditions, the explicit hope was that countries would improve the institutional quality of their government, thereby promoting something of a virtuous circle which would then allow them to reduce their dependence on aid, or concessional loan facilities (Roland, 2004). Furthermore, there are reasons to think that there might be, at best, conditional institutional convergence. Theoretically, Blackburn etal. (2006) develop a model whereby multiple equilibria corruption clubs emerge. Mukand and Rodrik (2005) are highly skeptical that a ‘one size fits all’ set of institutions would be either useful or indeed workable, at a global level. Roland (2004) discusses the issue of trying to fit a common set of ‘slow-moving’ institutions into an environment where very different ‘fast-moving’ institutions exist. Others (e.g., Rodrik, 2008) caution against the idea that Western-style institutions are 2 It is not the intention of this paper to delve into the entire institutional literature. For further information, please refer to e.g., Rodrik etal. (2004) and Hall and Jones (1999) on the issue of the institutional determinants of income, and Mauro (1995) for the initial uses of these datasets to measure different aspects of institutions. Transitional dynamics… / Andrew D. Williams, Tsun S. Cheong, Michal Wojewodzki 35 universally applicable. De Long and Summers (1993) suggest that regional location might play an important role in what could be termed ‘social capability’. Collectively, these papers suggest that unconditional institutional convergence is unlikely to occur, particularly if no account is made of geographical, historical, and cultural factors. Regarding the empirical evidence on institutional convergence, Ahmad (2008) finds some evidence of corruption convergence clubs emerging. Hall (2016) employs measures of economic freedom to demonstrate convergence. Savoia and Sen (2016), use the ICRG data on the rule of law, corruption, and bureaucratic quality. Their results show some evidence of unconditional β -convergence occurring slowly. Pérez-Moreno etal., (2020) and Schönfelder and Wagner (2019) employ β - and σ -convergence3 to analyze institutional convergence in the euro area countries. Perhaps the closest in spirit to our study is Beyaert etal. (2019) who investigate convergence in the euro area. They employ the unit root tests together with the distribution dynamics approach. Their results suggest a lack of institutional convergence with regards to indicators extracted from the ICRG database. Besides, Beyaert etal. (2019) observe institutional deterioration or even backsliding of the poorer peripheral or new (post-communist) states as compared with the core countries from western and northern Europe. One aspect of institutions that has recently gained some attention is the measurement of the openness or transparency of political, legal, and bureaucratic institutions. For example, Hollyer etal. (2011) and Williams (2009) develop institutional indices derived from how much information is released by governments collated from the World Bank’s World Development Indicators (WDI) database. There are potential economic and political benefits from greater transparency such as lower inflation (Crowe and Meade, 2007), public debt and budget deficits (Alt and Lassen, 2006). Focusing on information transparency to look at the question of institutional convergence has other important benefits. For instance, greater transparency can help reduce information asymmetries, within domestic markets (Gilbert, 2011), as well as promote transactions across borders. In other words, the greater the degree of (informational) openness, the more confidence agents (domestic and foreign) can have when engaging in economic transactions. Furthermore, informational transparency (IT) is likely to also be an important consideration when thinking about the potential spillovers between neighbours within the same geographical region (that is, whether we observe ‘regional clubs’ of convergence). Greater IT may also help impose constraints on the political class, in that their actions can be monitored by society, and consequently help in reducing bureaucratic inefficiency and corruption 3 β -convergence measures whether the countries with lower values of a studied variable (e.g., GDP per capita) are catching up over time with the countries with higher values of this variable. In other words, β -convergence occurs if e.g., a poorer country’s GDP per capita grows faster than that of an initially richer country. The concept of σ -convergence refers to a decrease in a dispersion (variance) of studied variables (e.g., institutional development) across countries. Estudios de Economía, Vol.49 - Nº136 (Brunetti and Weder, 2003). The final reason is a more practical one, in that the measure of IT used in this paper has extensive coverage across time and countries. Governments have (or at least had) a virtual monopoly over the release of information – both its quantity and quality, but also the infrastructure that allows for its dissemination. Information flows can therefore provide important clues as to governmental intent. With improvements in technology over the latter part of the 20th century, particularly concerning the transmission of information, one might imagine that convergence across countries would indeed be possible in this realm. It is to this issue that we now turn our attention. Nevertheless, it is important to note that in what follows we are not ascribing any causal mechanism to this convergence. The crucial point here is to establish the existence (or otherwise) of institutional IT convergence. Once this has been established robustly, as is the intention here, then subsequent research can begin to unpick some of these causal mechanisms. Furthermore, we are not claiming that IT is the only way to observe broad institutional convergence. 3. Data and Methodology 3.1. Information transparency index The measure of IT used in this study comes from the Information Transparency Index developed by Williams (2015). He derives a composite index for IT using existing datasets, for over 190 countries over the 1980-2010 period.4 As such, IT Index focuses on measuring the quantity, quality and infrastructure associated with the release of information by governments such that a higher score represents a higher level of IT in the country (Williams, 2015). As discussed previously, one of the main purported benefits of greater information transformation is that it facilitates greater investment by reducing transaction costs through a reduction of informational asymmetries between contracting parties. This is measured in the above index through both the quantity and quality of information produced. The quantity is important because the information available covers a greater range of economic indicators that might matter to decision-makers when considering an investment in an economy. Therefore, the index includes the quantity of information released by countries, obtained e.g., from the World Bank and IMF databases. 4 Due to the lack of available information for some countries data begins after 1980. More specifically, we use an unbalanced sample, with the number of countries (with scores for the Information Transparency Index) increasing from 153 after 1980 and reaching 191 by 2010 (Williams, 2015). However, over the entire period (1980-2010) in different years, 194 countries have scores for the Information Transparency Index. The list of all 194 countries by region and income groups is shown in TableA1 and A2 in the Appendix. Transitional dynamics… / Andrew D. Williams, Tsun S. Cheong, Michal Wojewodzki 37 However, quantity is not of much use to these agents if it cannot be relied upon, which is why the second sub-index includes the quality of the information. For example, using the World Bank’s Statistical Capacity Indicator, which attempts to measure the quality of information produced by national agencies. The third element of the index relates to the ability of governments to widely distribute that information – hence the inclusion of ‘information flows’ from the index of globalization by KOF Swiss Economic Institute, and the proportional number of radios, as a proxy for the ability of society to receive that information. Collectively, this IT index is a useful candidate with which to look at convergence, even if only as a first step towards future research on this issue, as it broadly captures an institutional aspect that is vital for economic development (information). Figure1 below summarizes these three sub-indices, whilst Table1 provides additional information on all indicators and their sources used in the composition of the IT index. FIGURE 1 COMPOSITION OF INFORMATION TRANSPARENCY INDEX Source: (Williams, 2015). As Tables 2 and 3 demonstrate, the average scores for IT vary considerably across regional groups, as well as income. With respect to regions, Africa (Europe) on average has the lowest (the highest) IT scores over the 30 years of investigation. In terms of income groups, the high-income OECD countries have the highest average IT scores, and low-income countries have the lowest average, thereby suggesting that a bidirectional causality between IT and the capacity of the resources of the country may exist. Estudios de Economía, Vol.49 - Nº138 TABLE 1 SOURCES OF INFORMATION TRANSPARENCY Transparency Sub-Category Indicator (Source) Accessed from Quantity of Information Release of Financial Information Index (IMF’s International Financial Statistics) http://andrewwilliamsecon. wordpress.com/datasets/ Release of Economic and Social Information Index (World Bank’s WDI) http://andrewwilliamsecon. wordpress.com/datasets/ Release of Balance of Payments Information Index (IMF’s Balance of Payments database) http://andrewwilliamsecon. wordpress.com/datasets/ Central Bank Transparency - Economic Transparency (Bank for International Settlements) http://www.central-bankcommunication.net/links/ Institutional Profiles database – Quantity (CEPII’s Institutional Profiles Database) http://www.cepii.fr/institutions/EN/ ipd.asp Statistical Capacity Indicator – Periodicity and timeliness (World Bank) http://go.worldbank.org/ UI0WGV6KW0 Quality of information Banking Disclosure index (World Bank’s Banking Regulation dataset) http://econ.worldbank.org/WBSITE/ EXTERNAL/EXTDEC/EXTRESE ARCH/0,,contentMDK:20345037~p agePK:64214825~piPK:64214943~t heSitePK:469382,00.html Institutional Profiles database – Process (CEPII’s Institutional Profiles Database) http://www.cepii.fr/institutions/EN/ ipd.asp Statistical Capacity Indicator – Source data and Statistical Methodology (World Bank) http://go.worldbank.org/ UI0WGV6KW0 Central Bank Transparency - Procedural Transparency (Bank for International Settlements) http://www.central-bankcommunication.net/links/ Information Infrastructure KOF Index of Globalization (KOF Swiss Economic Institute) http://globalization.kof.ethz.ch/ (Sub-section data on ‘information flows’) Radios per 1,000 population (World Bank’s WDI) WDI (2005) for 1980-2000, Indices of Social Development for 2001-2010 E-government - web measure, infrastructure, participation (United Nations survey) http://unpan3.un.org/egovkb/about/ index.htm Transitional dynamics… / Andrew D. Williams, Tsun S. Cheong, Michal Wojewodzki 45 for values higher than 1. However, for the extremely low-IT countries, the performance of the 2000-2010 decade is slightly worse than that of 1980-1990. This implies that the tendency towards convergence has increased across time, even though it may take a long time to achieve due the high persistence (Figure3). FIGURE 5 MOBILITY PROBABILITY PLOT (MPP) FOR RELATIVE IT OF ALL COUNTRIES ACROSS THREE DIFFERENT PERIODS Note: The horizontal axis represents the value of relative transparency, and the vertical axis represents the MPP. Source: Authors’ calculation. FIGURE 6 ERGODIC DISTRIBUTION OF ALL THE COUNTRIES Note: The horizontal axis represents the value of relative transparency, and the vertical axis represents the proportion. Source: Authors’ calculation. Estudios de Economía, Vol.49 - Nº146 This result is confirmed by the ergodic distribution (Figure6), where the peak lies around the value of one. However, there is another small peak around 1.3, which suggests that convergence clubs may emerge in the future, with many countries attaining the global average level of IT, while some countries attaining an above-average level. 4.2. Distribution dynamics for different regions Although the analysis of the global development in IT is illuminating, it is also of interest to examine the transitional dynamics of different regions in the world. Therefore, the data is divided into seven regions and a stochastic kernel analysis is conducted individually for each of the regions. Figure7 shows the contour maps of the transition probability kernels. The peaks are situated close to the diagonal, which again indicates that persistence is very high for every region. Two clusters of countries can be found in Panel A (Africa). Furthermore, many African countries have below-average relative IT values between 0.6 and 0.8. Panel B shows that the countries in Asia tend to cluster around values between 0.5 and 1. Moreover, we can observe that many European countries (Panel C) have above-average relative IT values of 1.4, whereas in the Middle East (Panel D) many countries cluster around a value of 0.8. Panel E shows that most (some) countries in North and Central America cluster around two aboveaverage IT values of 1.1 (1.4). Similarly, Panel F shows that many of the countries in South America have relative IT values of 1.1 and 1.3. Lastly, countries in Oceania (Panel G) also exhibit similar characteristics as many (some) appear to cluster around the values of 1 (1.6). However, the ‘dumbbell’ shape of the contour map suggests that the disparity between these two groups of countries is substantial. In sum, the stochastic kernels of many regions have twin peaks, thereby indicating that the countries within these regions have very different transitional dynamics. Perhaps unsurprisingly, this would seem to refute the idea that similar institutions are found in areas located geographically nearby. The MPPs by region are shown in Figure8. Given that countries have a higher tendency of moving upwards (downward) if the MPP lies above (below) the horizontal axis, one can expect countries to congregate around a value that is close to the intersection points in years to come. Hence, several intersection points suggest that the ergodic distribution will have multiple peaks, i.e., the emergence of convergence clubs is more likely to occur. Looking at the MPPs of Africa (Panel A) and the Middle East (Panel D) region we can observe that both MPPs intersect the horizontal axis at values smaller than 1, indicating that some of the below-average countries in these regions will move further downwards. This finding is disturbing as it means that convergence to the global average may be difficult for some of these countries. In contrast, Panel B shows the Asian MPP intersecting the horizontal axis at the value of 1, thereby indicating that convergence of Asian countries to the global mean is far more likely. Panel C shows that the MPP of Europe intersects the axis around the values of 1.36, thus it can be expected that the European countries will converge Transitional dynamics… / Andrew D. Williams, Tsun S. Cheong, Michal Wojewodzki 47 FIGURE 7 CONTOUR MAPS OF TRANSITION PROBABILITY KERNEL FOR RELATIVE IT OF DIFFERENT REGIONS Note: The horizontal axis represents the value of relative transparency at time t, and the vertical axis represents the value of relative transparency at time t+1. Source: Authors’ calculation. Estudios de Economía, Vol.49 - Nº148 FIGURE 8 MOBILITY PROBABILITY PLOTS (MPPS) FOR RELATIVE IT OF DIFFERENT REGIONS Note: The horizontal axis represents the value of relative transparency, and the vertical axis represents the MPP. Source: Authors’ calculation. Transitional dynamics… / Andrew D. Williams, Tsun S. Cheong, Michal Wojewodzki 49 to an above-average value in years to come. Panel E shows that the MPP of the North and Central America region is the most volatile along large sections of the horizontal axis. Furthermore, the MPP of the South America region moves down and intersects the horizontal axis at several points (1.11, then at 1.18, and again at 1.29). A similar but more pronounced and dispersed pattern is visible in Oceania (Panel G) where the MPP intersects the axis at 0.97, 1.27, and 1.53. Figure9 presents regional MPPs across three decades. We can observe that with time, many MPPs move higher (lower) for the values below (above) 1. Such a movement in the MPP can lead to faster convergence, as the below- (above) average countries will have a much higher probability of moving upwards (downwards). This tendency, however, is not universally observed. More specifically, Panel G (Oceania) shows that in countries with IT values below 1, the MPP of the 2000-2010 (1980-1990) period is plotted above (below) the other two MPPs. Thus, the transitional dynamics of those below-average countries have deteriorated over time. Figure10 shows the ergodic distributions by regions, which reveals the potential future distribution of relative IT in the long run. It can be observed that Asia will converge to the global average in the future (under the assumption of no changes in transitional dynamics). Africa and the Middle East will, however, converge to a value much lower than the average, whilst Europe, North and Central America, South America, and Oceania will converge to above-average IT values. Moreover, a twin (three) peaks pattern can be observed in ergodic distributions representing Africa, South America, and Oceania (North and Central America) regions. These findings are in line with the conclusions derived from the MPPs (Figure 8). It is worth noting that the results show how a lot of countries situated in the Global North regions enjoy higher levels of IT than their peers in the Global South. For example, the countries in Europe and North America (Africa and the Middle East) would converge to IT values higher (lower) than the global average. Asia would converge to the global average IT, perhaps because this region consists of rich (e.g., Japan and South Korea) and poor (e.g., Afghanistan, and Bangladesh) countries alike. This may indicate that the evolution of IT could be related to the economic performance of a country. It thus calls for an in-depth analysis of the relationship between the levels of income and development in IT in the future. 4.3. Distribution dynamics for different income groups To examine the relationship between income and transitional dynamics of IT, countries have been divided into five income groups, namely: high-income OECD, high-income non-OECD, upper-middle-income, lower-middle-income, and low-income. Figure11 shows the contour maps of these groups. We can observe many clusters within the same income group. In fact, at least two significant peaks can be found in every income group. Another finding is that persistence is high for all the income groups, which suggests that a lot of the countries will have a high probability of remaining at their relative levels of IT. Estudios de Economía, Vol.49 - Nº150 FIGURE 9 MOBILITY PROBABILITY PLOTS (MPPS) FOR RELATIVE IT OF DIFFERENT REGIONS ACROSS DIFFERENT PERIODS Note: The horizontal axis represents the value of relative transparency, and the vertical axis represents the MPP. Source: Authors’ calculation. Transitional dynamics… / Andrew D. Williams, Tsun S. Cheong, Michal Wojewodzki 51 FIGURE 10 ERGODIC DISTRIBUTIONS OF DIFFERENT REGIONS Note: The horizontal axis represents the value of relative transparency, and the vertical axis represents the proportion. Source: Authors’ calculation. Estudios de Economía, Vol.49 - Nº152 FIGURE 11 CONTOUR MAPS OF TRANSITION PROBABILITY KERNEL FOR RELATIVE IT OF DIFFERENT INCOME GROUPS Note: The horizontal axis represents the value of relative transparency at time t, and the vertical axis represents the value of relative transparency at time t+1. Source: Authors’ calculation. Transitional dynamics… / Andrew D. Williams, Tsun S. Cheong, Michal Wojewodzki 53 Figure12 shows that the MPP of the high-income OECD countries (Panel A) intersects the horizontal axis at 1.37 which is the highest amongst all the income groups. Panel B shows that the MPP of the high-income non-OECD countries intersects the axis at several points, however, the values of these intersections are much lower than those in Panel A. In fact, by comparing Panels A and B, it can be observed that the transitional dynamics of the high-income non-OECD FIGURE 12 MOBILITY PROBABILITY PLOTS (MPPS) FOR RELATIVE IT OF DIFFERENT INCOME GROUPS Note: The horizontal axis represents the value of relative transparency, and the vertical axis represents the MPP. Source: Authors’ calculation. Estudios de Economía, Vol.49 - Nº154 countries are far more complicated than those of the OECD countries. Since there are multiple intersection points in Panel B, the high-income non-OECD countries have a high probability of having several peaks in the ergodic distribution. This undoubtedly reflects the significant heterogeneity of this group, given countries ranging from oil-rich Middle East states such as Kuwait and the UAE to Hong Kong, and the Baltic states. It is worth noting that the lower the income, the smaller the values of the MPPs’ intersection points observed in Figure12. For instance, Panel E shows that the MPP of the low-income countries intersects the axis at several points from 0.32 to 0.80, thereby indicating that the below-average transparency countries in the low-income group have a lower tendency of moving upward than their counterparts from the high-income OECD group. Thus, an important finding of this study is that there appears to be a strong relationship between income and upward mobility in IT. Panel E shows that the MPP of the low-income countries lies below the horizontal axis for values greater than 0.83. Furthermore, entities with a relative IT value of 1 have a 27 per cent net probability of moving downwards. This is somewhat concerning, as it means that the below-average countries in the low-income group are more likely to move further downwards in the future distribution. The difficulty of moving upward for the low-income countries suggests the existence of a development trap in IT. Figure13 shows the MPPs of different income groups across the three time periods. In Panel A (high-income OECD countries) the MPP of the 2000-2010 period lies below the MPPs of the earlier (1990-2000 and 1980-1990) periods for the IT values greater than the intersection point, indicating that the countries with high informational transparency have become more susceptible to falling downwards across time. Moreover, also on the left of the intersection with the horizontal axis, the MPP of the 2000-2010 period lies below the MPP of the 19902000 period, thus implying that the upward movement of countries with lower relative IT values decelerated somewhat during the most recent of investigated decades. Furthermore, we can observe a fair degree of resemblance amongst the MPPs of other income groups, which indicates that there is no substantial change to the transitional dynamics over time. Figure14 shows that ergodic distributions of all the income groups are quite dispersed, apart from the high-income OECD countries which have a noticeable high peak. It implies that the OECD countries will achieve similar levels of transparency in the future, while the other countries will exhibit large disparity. It is worth noting that the dramatic difference in shape between the two high-income groups (Panels A and B) suggests that factors other than income may also play a role in the transitional dynamics of countries’ IT. Moreover, it is found that except for Panel A (high-income OECD countries), distributions have multiple peaks, thereby suggesting the emergence of convergence clubs. Transitional dynamics… / Andrew D. Williams, Tsun S. Cheong, Michal Wojewodzki 61 Appendix TABLE A1 THE LIST OF 194 COUNTRIES GROUPED ACCORDING TO GEOGRAPHIC REGIONS Region Country Africa Algeria, Angola, Benin, Botswana, Burkina Faso, Burundi, Cameroon, Cape Verde, Central African Republic, Chad, Comoros, Democratic Republic of Congo, Republic of Congo, Cote d’Ivoire, Djibouti, Egypt, Equatorial Guinea, Eritrea, Ethiopia, Gabon, Gambia, Ghana, Guinea, Guinea-Bissau, Kenya, Lesotho, Liberia, Libya, Madagascar, Malawi, Mali, Mauritania, Mauritius, Morocco, Mozambique, Namibia, Niger, Nigeria, Rwanda, Sao Tome and Principe, Senegal, Seychelles, Sierra Leone, Somalia, South Africa, Sudan, Swaziland, Tanzania, Togo, Tunisia, Uganda, Zambia, Zimbabwe Asia Afghanistan, Bangladesh, Bhutan, Brunei, Cambodia, China, Hong Kong Special Administrative Region, India, Indonesia, Japan, Kazakhstan, Democratic People’s Republic of Korea, Republic of Korea, Kyrgyzstan, Laos, Malaysia, Maldives, Mongolia, Myanmar (Burma), Nepal, Pakistan, Philippines, Singapore, Sri Lanka, Taiwan, Tajikistan, Thailand, Timor-Leste, Turkmenistan, Uzbekistan, Vietnam Europe Albania, Austria, Belarus, Belgium, Bosnia and Herzegovina, Bulgaria, Croatia, Czech Republic (Czechoslovakia), Denmark, Estonia, Finland, France, Germany (Federal Republic of Germany), Greece, Hungary, Iceland, Ireland, Italy, Kosovo, Latvia, Lithuania, Luxembourg, Macedonia, Malta, Moldova, Montenegro, Netherlands, Norway, Poland, Portugal, Romania, Russian Federation (Soviet Union), San Marino, Serbia, Slovak Republic, Slovenia, Spain, Sweden, Switzerland, Ukraine, United Kingdom Middle East Armenia, Azerbaijan, Bahrain, Cyprus, Georgia, Iran, Iraq, Israel, Jordan, Kuwait, Lebanon, Oman, Palestinian Autonomous Areas (also West Bank and Gaza), Qatar, Saudi Arabia, Syria, Turkey, United Arab Emirates, Yemen North and Central America Anguilla (overseas territory of the UK), Antigua and Barbuda, Aruba, Bahamas, Barbados, Belize, Canada, Costa Rica, Cuba, Dominica, Dominican Republic, El Salvador, Grenada, Guatemala, Haiti, Honduras, Jamaica, Mexico, Nicaragua, Panama, St Kitts and Nevis, St Lucia, St Vincent and the Grenadines, Trinidad and Tobago, United States South America Argentina, Bolivia, Brazil, Chile, Colombia, Ecuador, Guyana, Paraguay, Peru, Suriname, Uruguay, Venezuela Oceania Australia, Fiji, Kiribati, Marshall Islands, Micronesia, New Zealand, Palau, Papua New Guinea, Samoa, Solomon Islands, Tonga, Tuvalu, Vanuatu Estudios de Economía, Vol.49 - Nº162 TABLE A2 THE LIST OF 194 COUNTRIES GROUPED ACCORDING TO INCOME GROUPS Income Country High-income OECD Australia, Austria, Belgium, Canada, Chile, Czech Republic (Czechoslovakia), Denmark, Estonia, Finland, France, Germany (Federal Republic of Germany), Greece, Iceland, Ireland, Israel, Italy, Japan, Republic of Korea, Luxembourg, Netherlands, New Zealand, Norway, Poland, Portugal, Slovak Republic, Slovenia, Spain, Sweden, Switzerland, United Kingdom, United States High-income non-OECD Anguilla, Antigua and Barbuda, Aruba, Bahamas, Bahrain, Barbados, Brunei, Croatia, Cyprus, Equatorial Guinea, Hong Kong Special Administrative Region, Kuwait, Latvia, Lithuania, Malta, Oman, Qatar, Russian Federation (Soviet Union), San Marino, Saudi Arabia, Singapore, St Kitts and Nevis, Taiwan, Trinidad and Tobago, United Arab Emirates, Uruguay Upper-middle-income Albania, Algeria, Angola, Argentina, Azerbaijan, Belarus, Belize, Bosnia and Herzegovina, Botswana, Brazil, Bulgaria, China, Colombia, Costa Rica, Cuba, Dominica, Dominican Republic, Ecuador, Fiji, Gabon, Grenada, Hungary, Iran, Iraq, Jamaica, Jordan, Kazakhstan, Lebanon, Libya, Macedonia, Malaysia, Maldives, Marshall Islands, Mauritius, Mexico, Montenegro, Namibia, Palau, Panama, Peru, Romania, Serbia, Seychelles, South Africa, St Lucia, St Vincent and the Grenadines, Suriname, Thailand, Tonga, Tunisia, Turkey, Turkmenistan, Tuvalu, Venezuela Lower-middle-income Armenia, Bhutan, Bolivia, Cameroon, Cape Verde, Republic of Congo, Cote d’Ivoire, Djibouti, Egypt, El Salvador, Georgia, Ghana, Guatemala, Guyana, Honduras, India, Indonesia, Kyrgyzstan, Kiribati, Kosovo, Laos, Lesotho, Mauritania, Micronesia, Moldova, Mongolia, Morocco, Nicaragua, Nigeria, Pakistan, Papua New Guinea, Paraguay, Philippines, Samoa, Sao Tome and Principe, Senegal, Solomon Islands, Sri Lanka, Sudan, Swaziland, Syria, Timor-Leste, Ukraine, Uzbekistan, Vanuatu, Vietnam, Yemen, Zambia Low-income Afghanistan, Bangladesh, Benin, Burkina Faso, Burundi, Cambodia, Central African Republic, Chad, Comoros, Democratic Republic of Congo, Eritrea, Ethiopia, Gambia, Guinea, Guinea-Bissau, Haiti, Kenya, Democratic People’s Republic of Korea, Liberia, Madagascar, Malawi, Mali, Mozambique, Myanmar (Burma), Nepal, Niger, Palestinian Autonomous Areas (also West Bank and Gaza), Rwanda, Sierra Leone, Somalia, Tajikistan, Tanzania, Togo, Uganda, Zimbabwe Note: The list of countries is in accordance to the OECD member countries and World Bank’s Classification as of the end of 2010.