January 2025 Europe and Central Asia Hindering or enabling? The role of states in MIC to HIC transitions in ECA P180441 Governance © 2025 The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved. This work is a product of The World Bank. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. The boundaries, colors, denominations, links/footnotes and other information shown in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. The citation of works authored by others does not mean the World Bank endorses the views expressed by those authors or the content of their works. Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Attribution—Please cite the work as follows: “World Bank. 2025. Hindering or enabling? The role of states in MIC to HIC transitions in ECA. © World Bank.” Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Hindering or enabling? The role of states in MIC to HIC transitions in ECA Table of Contents 1. What an effective enabling state can do: the role of the state in South Korea’s and Finland’s transitions to high income, high innovation countries Error! Bookmark not defined. 2. The role of states in economic development in ECA ........................................................... 4 States in pre-1990 ECA – creating foundations as well as obstacles for growth ........................ 4 Different speeds of growth and innovation in post-1990 ECA ................................................... 5 How has institutional performance differed between HIC convergers and others?.................. 7 ECA’s Large fiscal footprints and the need for efficient delivery of enabling services ............. 11 Avoiding capture: ensuring that markets and opportunities remain competitive ................... 14 Pursuing institutional and policy reforms in good and in bad times ........................................ 16 3. Evolving the role of the state across different functions ................................................... 19 The role of states in innovation in ECA ..................................................................................... 20 4. Conclusions and policy implications................................................................................. 23 i Acknowledgements and Disclaimer This note was developed as a background paper to the ECA WDR 2024 companion report. FCDO generously contributed to the development of the note. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. The paper was prepared by Verena Fritz (Lead Governance Specialist) and Ruggero Gambacurta-Scopello (Young Professional) in collaboration with Eric Arias (Economist), Jonathan Hanson (Consultant/Lecturer in Statistics for Public Policy, University of Michigan), Rachel Sigman (Consultant/Associate Professor, University of Denver), and Lorenzo Montrone (Consultant). Dan Boyce (former Practice Manager for Governance ECA East) and Rachel Ort (Senior Governance Specialist) provided inputs to earlier versions and the initial design of the note. The team thanks Asad Alam, Regional Director for Prosperity, ECA; Ivailo Izvorski, Chief Economist for ECA; Michael Lokshin, Lead Economist, ECA; Chiara Bronchi, Practice Manager for Public Institutions, Data and Analytics; Adrian Fozzard, Practice Manager for Governance in ECA East, and Fabian Seiderer, Practice Manager in ECA West for their helpful comments and guidance in developing the note. Convergers Upper MICs (‘potential Lower MICs convergers’) (GNI per capita ≤ 4900 in 2012) (GNI per capita of ≥ 4900 in 2012) Croatia Belarus Albania Czechia Bosnia Armenia Estonia Kazakhstan Azerbaijan Hungary Montenegro Georgia Latvia Russian Federation Kosovo Lithuania Serbia Kyrgyz Republic Poland Türkiye Moldova Romania Turkmenistan North Macedonia Slovakia Tajikistan Slovenia Ukraine Bulgaria Uzbekistan Late convergers (selected figures) Romania and Bulgaria ii 1. What an effective enabling state can do: the role of the state in South Korea’s and Finland’s transitions to high income, high innovation countries South Korea transitioned from low, to middle, and then to high income country in just over 35 years - approximately the time that has passed since the fall of the Berlin wall to today. In 1953, South Korea was a country devastated by the Korean War, with a GDP per capita of only US$158 in 1960 – lower than Haiti, Ethiopia or Yemen at that time. Following several years of slow growth under postwar import substitution policies (Kim 1991), the country started a systemic transformations in the late 1950s and early 1960s with re-building an education system and land reform. From 1962 onwards, policy efforts focused on ‘export is first’ 1, underpinned by effective incentives for firms. The South Korean government used several levers to pursue the goals of industrialization and exports: (i) it set up institutions and coordination mechanisms, and developed 5-year plans that provided guidance on policies and industrialization priorities; (ii) it channeled public investments towards infrastructure to facilitate industrialization and exports; and (iii) it subsidized loans to private companies with industrial and export targets and making further access conditional on meeting agreed-upon targets – as a key incentive mechanism. Important motivators were the perceived need to have sufficient defense capability vis-a-vis the north. The country’s real GNP tripled in every decade from 1962 -1991, driven by successive rounds of structural and industrial transformation – from investments, to infusion, to innovation (WDR 2024) - in which the state played a significant role. While the state played a strong role in South Korea’s development, the state itself underwent a push for reforms in terms of promoting effective as well as impartial institutions. At the outset, institutional quality in South Korea was only marginally better than MIC averages. This was followed by major improvements during the country’s upper middle-income stage. A rapid shift towards near-OECD institutional quality occurred between 1980 and 2000, as South Korea transitioned from MIC to high-income status – when government rapidly brought corruption levels down and decisively moved to improve the public administration (Figures 1a and b). Governments invested in education and attracted high-caliber graduates to the public sector (Soh, Koh, and Aridi 2023)2. An increasingly capable public sector leveraged foreign aid to support industrialists in adopting modern technologies and reinvesting profits. While risks of capture are inherent to industrial policy – and can turn such policies into a fiscal burden fueling rent-seeking -- continuous public sector improvements and high- level commitment to success allowed maintaining a focus on performance rather than incumbency protection (Ilyna et al. 2024). By the early 1990s, the economy had developed a dynamic manufacturing sector. Figures 1a and 1b: Corruption and Public Administration Quality (MIC averages, OECD, and South Korea 1960-2022)3 Source: V-DEM; note: average of MICs from across all regions 1 Following policy approaches being pursued in Japan since the mid-1950s that were beginning to show success. 2 As the authors note: “Given the central role of the government in the “developmental state” model, talent recruitment and management of the civil service were critical”; civil service entry exams became rigorous and highly selective from the 1960s onwards. 3 V-DEM data is based on expert surveys: https://v-dem.net/about/v-dem-project/methodology/. The composition of country groupings evolved over time. Data includes countries before and after their potential separation. For instance, data from Yugoslavia is replaced by the data of the countries created after the dissolution. 1 Improving institutions enabled South Korea to implement consecutive rounds of industrial policy while limiting capture and wasteful efforts (WDR 1997). Key governance arrangements were structured around the Economic Planning Board and relied on a clear political leadership and commitment and strong cooperation with the private sector. While South Korea had substantial capture and clientelism especially at early stages (Kang 2009), a key to success was a unifying motivation and the ability to improve institutions over time. Land reforms and continued investments in education first reduced and then contained inequality, while a social safety net was gradually developed since the 1970s (Soh, Koh, and Aridi 2023). Improved institutions also enabled an effective response to the 1997 Asian financial crisis. South Korea reached high income status by 1995 but experienced a slide backwards in 1997-98 due to the crisis. With well- calibrated government actions – notably reforms of the financial sector and shifting access to preferential credits from large firms to smaller companies (Soh, Koh, and Aridi 2023), the economy recovered from the severe shock, progressing back to high income status in 2001 and continuing its upwards trend. The reforms allowed for greater market competition and larger foreign ownership, enhancing productivity growth in non-chaebol firms (Aghion, Guriev and Jo 2020). Since reaching high income statues, South Korea has pursued becoming and remaining a highly innovative economy. While slower than in prior decades, growth in the 2000s was still double that of advanced economies, with GDP per capita reaching US$34,998 in 2022 (Soh, Koh, and Aridi 2023). South Korea has been ranked as one of the top twenty countries on the Global Innovation Index since the 2010s, advancing to a top 10 country in the 2020s.4 Compared to the previous emphasis on large firms and industries, the government became more focused on promoting small and medium enterprises and technology entrepreneurs. In 2017, South Korea established a Ministry of SMEs and Startups, managing various startup support schemes. Venture capital investments reached US$6.4 billion in 2021, a 78% year-on-year growth5, with the number of new jobs created by startups surpassing those created by the four largest conglomerates combined (Yoon 2022). Achievement of high levels of human capital development has been complemented by an expanded social safety net and a more integrated approach to education and training. Finland – with a GNI per capita of 1370USD in 1962 – made a rapid transition from middle to high-income status over a 30-year period between the end of WWII and the late 1970s. At the end of WWII, Finland’s economy remained highly dependent on primary and low value-added products (forestry, agriculture, and pulp and paper production); moreover, Finland was obliged to pay reparations to the USSR and faced a shrunken territory. To pay for reparations, Finland sought to upgrade its domestic machinery and shipbuilding industries (Hjerppe 2008). By 1969, industries such as wood, pulp and paper, and metals (including shipbuilding), transport, and electrical equipment accounted for more than a third of employment. However, income per capita still trailed behind nearby countries such as Sweden, prompting many Finns to emigrate to find better jobs (Andre 2019). Finland’s successful transition was enabled by three key factors: (i) early breakthrough into new heavy industries (shipbuilding), (ii) already good institutional quality that was maintained over time, and (iii) government-led investments in education and research pursued in close collaboration with the private sector. As Soviet demand for Finnish ships grew in the 1950s and 60s, so too did the industry’s technical and design abilities, along with its innovation potential. The state focused on incremental measures – particularly in the areas of education, research, and innovation – to create a solid foundation for increasing competitiveness and technological sophistication (Hirvonen 2004). One center of gravity of these arrangements was a close and productive relationship between the Finnish state and Nokia as the dominant corporate entity (Lindén 2021). Different from South Korea, Finland had accountable and capable institutions already in the 1960s – reducing risks of capture or poor execution of state-led efforts – and maintained these over time. In the 1960s, Finland’s institutions had a level of performance already well ahead of its income status; which then ‘caught up’ by the 1980s. In the 1970s and 1980s, Finland’s economy converged with more advanced industrialized countries such as Germany and Sweden, with the income per capita gap shrinking to less than 10%. 4 https://www.wipo.int/en/web/global-innovation-index. 5 https://www.weforum.org/agenda/2022/01/startups-in-south-korea-are-thriving-this-is-why/ 2 While Finland’s success with developing a world -class general education system is well known, a second key element for success were government efforts at fostering research and innovation – in close collaboration with the private sector. In 1967, Finland created Sitra, the Finnish Innovation Fund, which serves as both economic think tank and investment company. The foundation’s objective was to support project that increase economic efficiency, improve education and research, and study future development scenarios. In the two decades following its creation, government officials worked closely with both Sitra and private sector leaders, especially Nokia. The government planned the creation of new universities and established ground rules for cooperation between universities and companies (Lindén 2021). An opaque system of commissioned research was replaced with a more open, transparent system. Additionally, the establishment of Tekes – the Funding Agency for Technology and Innovations – in 1983 facilitated closer cooperation between business and universities utilizing both private and public money (OECD 2017). Effective state-led innovation governance structures also helped Finland to weather and recover from major economic downturns in the 1990s and 2000s. With the collapse of the Soviet Union and the end of the Cold War, demand for Finnish products declined, sending Finland into a deep recession. Working through agencies such as Sitra and Tekes, the Finnish state adopted educational policy and tax reductions that helped Nokia transition to mobile phone production and spurred broader ICT sector development. In the 1990s, Tekes funded around 26% of Nokia’s R&D spending, enabling them to develop the mobile phone products that propelled them to compete in the international ICT sector for some time. As Nokia’s success with GSM products waned and it began shedding jobs, the Finnish government responded with programs to incubate small businesses. Despite the declining economic situation, Tekes had its annual budget for funding new and innovative busines ses increased to around €550 million. The Finnish Ministry of Employ ment and Economy launched Vigo, a start- up accelerator in 2009 and several new Finnish startups attracted significant volumes of venture capital (Halme et al. 2018).6 6 See https://www.wired.com/story/finland-and-nokia/# for additional detail. 3 2. The role of states in economic development in ECA This note focuses on the role of the state in the pursuit of middle- to high-income transitions in ECA. ECA MIC to HIC transitions account for almost one third of the countries (11 out of 34) that newly achieved high-income status since 1990 (WDR 2024) – suggesting that states in ECA convergers must have gotten something right. 7 Furthermore, comparing within the region – between HIC convergers and remaining MICs, as well as between ‘early’ and ‘late convergers’ – allows looking for patterns of relative similarity and differences. This note explores differences in the role of the state across these groups, while also relating these back to the experience of South Korea and Finland. In some ways, the challenges and opportunities for the role of the state in ECA have been unique relative to other middle-income regions, while in others they are similar to those faced by economies elsewhere . In terms of exceptional conditions, on the one hand, most ECA countries experienced an excessive role of the state during communism, followed by deep institutional disruption and volatility in the 1990s. On the other hand, EU accession has provided a combination of funding and external stabilization as well as a strong pull towards institutional reforms at least until reaching membership. Yet ECA countries have also struggled with challenges common across middle-income countries aspiring to achieve high incomes: how to enhance economic productivity, how to weather and rebound after periodic crises and how to avoid zig-zagging between gains and losses in per capita income levels and maintain an upward dynamic. The note explores the following questions: 1. How different are HIC convergers from remaining MICs in ECA in institutional effectiveness, integrity, and in constraining state capture? 2. Are the region’s states’ mostly large fiscal footprints a constraint, and how do converging and non - converging countries compare? 3. How have states in ECA pursued reforms during ‘good’ and ‘bad’ times? 4. How has the role of the states been recalibrated across different functions to meet evolving policy challenges and objectives – including efforts at becoming more knowledge-intensive economies? In a Schumpeterian perspective – central to the WDR 2024 - states should ‘get out of the way’, but also create the right conditions for innovation, provide public goods, and protect competition . Basic implications from Schumpeter work are that states should simplify business regulations and enable access to capital for firms with more proactive elements such as offering - fair and criteria based - tax incentives aimed at fostering innovation and investing in education and infrastructure. More recent economists sharing a Schumpeterian view have emphasized the need to balance between competition and incentives for innovation; and the critical role of the state in protecting market entry for new competitors. This matters especially for sectors and markets close to the technological frontier to enable continuous innovation (Aghion et al. 2012). States in pre-1990 ECA – creating foundations as well as obstacles for growth During the communist period, states in the ECA region took on a dominant rather than enabling role in economic and social development. First, states functioned as ‘owners’ of all major means of production. Second, states operated a wide range of restrictions: market forces were largely not allowed to operate and the state sought to set comprehensive plans for economic development, operationalized through targets for factories and farms (Dabrowski 2023), with some variation across countries.8 De facto, all communist ECA countries had ‘shadow’ or grey markets to varying degrees, and many of these became more extensive in the final years of the communist period (Grossman 1989; Ivanov and Aisin 2023; Gerhani and Cichowski 2023). While pre-1990 states in ECA suppressed entrepreneurialism, they laid some important growth foundations, especially by creating basic infrastructure and investing in education . Urbanization and infrastructure connectivity increased across the region – from railroads to airports to mass transit systems (see also Carlin, Schaffer and Seabright 2013). In many of today’s convergers, literacy rates were already high during the pre - 7A further 3 convergers – Greece, Malta, a Portugal – are part of the wider region, but not considered in this note. 8Different communist countries varied to the extent to which some private ownership and initiative remained formally permitted – generally more so in Central Eastern Europe, e.g. Hungary (small businesses) and Poland (private land ownership). 4 communist period (Darden and Grzymla-Busse 2006), while elsewhere, primary and secondary education became universal during the communist period. During the 1960s and 70s, most ECA countries matched or exceeded South Korea’s levels of education. Tertiary education and research capacity varied more widely: from having at least a few strong HE and research institutions per country in Central Eastern Europe, comparatively strong higher education (HE) systems in Soviet Belarus, Russia, Ukraine and the three Baltic states, to weaker systems in Southeastern Europe and in Central Asia.9 Especially in the Soviet Union, research was heavily focused on military objectives (Chan 2015). Post-1990, ECA had some important pre-conditions for knowledge-based growth, but also had to grapple with profound transition disruptions. Comparatively strong levels of general education and lower income inequality relative to other global regions provided foundations for growth and for the development of effective and equitable institutions. However, post-communist transitions were disruptive: they required a complete reconfiguration of the role of the state, they triggered deep economic and fiscal contractions, and they involved a reconfiguration of state geographical boundaries for a majority of ECA countries. As rules and organizational roles profoundly changed, institutional capacity initially dropped as processes had to be redefined and stakeholders within public administrations as well as businesses and citizens had to establish, agree and learn new ways of operating. Rapid divestment of previously state-owned property on a large scale was prone to capture and rent-seeking – although much better managed in some countries than others (Piatkowski 2018). In parallel to fundamentally changing the role of states, ECA saw many new states emerge from previous larger entities (22 new states as well as one merger vs 6 continuing). 10 Newly constituting states needed to create many institutional capacities – to various degree building on previous regional-level institutions or temporally distant precursors. The depth and duration of economic and fiscal crises impacted institutional deterioration as well as the pace of recovery. Transition disruption – varying significantly across the region – worsened the delivery of public goods, including law and order, and led to poor maintenance of public infrastructure and services especially in rural areas, impacting existing growth foundations, notably infrastructure and education (World Bank 1997 ). In Georgia, the tax to GDP ratio dropped from 34% in 1991 to 5.4% by 1995, and in Bulgaria to 11.2% in the same year– figures typically associated with low income or even failed states. 11 The Russian Federation, Armenia and Azerbaijan, Moldova, as well as Tajikistan, Turkmenistan, and Lithuania also saw sharp revenue decreases (IMF 1996). Higher education and research suffered searing budget cuts in countries experiencing deep and extended fiscal contractions.12 The combination of these shocks and trends led to increased perception of institutions as corrupt and self-serving (Fritz 2003). In Central Eastern Europe, recessions and fiscal deterioration were present, but less deep and lasting for a shorter period. The transition period left institutional scarring, especially in countries with deep and prolonged fiscal deterioration, as informal practices and corruption became widespread.13 Different speeds of growth and innovation in post-1990 ECA Between the mid-1990s and 2012, nine ECA countries moved from middle to high income status; followed by two ‘late’ convergers: Romania in 2019, and Bulgaria in 2024 . As Figure 2 shows, the initial ECA convergers, rather than getting stuck in a middle-income trap achieved rapid increases in per capita incomes between 1993 and the start of the global financial crisis (GFC), with average growth rates of 4.5 per cent. While impacted by the global financial crisis, their growth trajectory eventually resumed from the 2010s. By contrast, both lower and upper middle income ECA countries experienced a stretch of almost 15 years until the mid-2000s before starting to experience some growth accelerations, followed by another slowdown in the 2010s, before resuming stronger growth in recent years. The two late convergers (Romania and Bulgaria) followed very similar 9 Even in countries with stronger HE systems, the share of age cohorts attending remained lower (up to 20%) compared to Western Europe and other high-income countries. 10 States emerging from the former Soviet Union and former Yugoslavia, the splitting up of Czechoslovakia, and German reunification. 11 WDI data. 12 In the Russian Federation in particular, predominantly military focused research went through a steep fall in funding and output during the transition. As one measure, patents per million population sharply decreased (from 265.9 in 1992 to 102.1 in 1997). 13 While South Korea also experienced a transition in the 1980s – from an authoritarian to a competitive political system in 1987 – the deep and difficult transition of the role of the state in the economy during the middle-income stage is specific to former communist ECA countries. 5 trajectories to ‘potential convergers’/upper MICs in ECA for most of the time, but then pulled ahead from the mid-2010s onwards. Figure 2: which states succeeded in the MIC to HIC transitions in ECA and how fast GNI per capita, Atlas method (current US$) 25000 20000 15000 10000 5000 0 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 average lower MICs average upper MICs average early convergers average late convergers LMICs global UMICs global High income threshold 2022 Source: WDI Relative to global high-income averages, the income gains made in ECA are substantial for early converging countries, but much weaker for the rest of the region. Early convergers in ECA moved from 37.3 to 46.5 per cent of the average global high-income levels. However, all other groups only made small to marginal gains – potential convergers from 20.6 to 23.9 per cent, and lower ECA MICs moved from 9.7 to 10.7 per cent of per capita incomes of high-income countries over the 33-year period from 1990 to 2023. Compared to all UMICs, potential and late convergers in ECA experienced less cumulative gains in per capita incomes. The region’s poorer countries saw greater growth compared to LMICs but fell significantly behind the trends of UMICs globally. Regarding knowledge intensity, ECA countries saw a substantial amount of infusion in the 30 years since 1990. Technology infusion - as the second phase of moving towards high income levels as set out by the WDR 2024, ECA generally benefitted from significant FDI, which scaled up first and most strongly in subsequent converger countries. Large volumes of FDI flowed to subsequent convergers from the 1990s (Lankes and Stern 1999; Marin 2010). In upper MICs and other ECA countries, FDI arrived later and remained more limited: FDI to Russia and Ukraine only took off from the mid-2000s and did not reach the same scale in per capita terms; Azerbaijan and Kazakhstan primarily attracted FDI into resource extraction rather than a broader range and more knowledge- intensive industries and services. While FDI is not the only mechanism for technology infusion, it was especially important for ECA convergers and involved a transfer of technology as well as management and financing practices that were new to the region. In terms of reaching high level innovation – i.e. emulating South Korea’s or Finland’s level of success – only a few countries in ECA so far show significant potential. While the Global Innovation Index has ranked South Korea as a top ten in recent years, in ECA, only two countries – Estonia and Czechia - have so far made it into the top 30 (on the overall rank and ‘knowledge and technology output’) (Figure 3), followed b y a large group of HIC convergers and upper MICs that achieve a middling rank (30-65 out of 133 economies included). Globally, ECA countries have been far surpassed by South Korea since the 1990s and the People’s Republic of China since the 2010s. More recent years have seen efforts to strengthen R&D and to rebuild on the region’s comparativ e strength in ICT; while Türkiye made significant progress from an initially low base. 6 Figure 3: Global Innovation Index ratings Score 2022: Knowledge and technology outputs index 70 60 50 40 30 20 10 0 Denmark Bangladesh PR China Cyprus Canada Belarus Morocco Slovak Republic Bulgaria Romania Serbia Viet Nam Sri Lanka Montenegro Georgia Ethiopia Poland Thailand Latvia Lithuania Moldova Tunisia Kyrgyz Republic France Norway Ukraine Chile Brazil North Macedonia Uruguay Kuwait Argentina Jordan Indonesia Uzbekistan Botswana Cameroon Germany Japan Slovenia Luxembourg Portugal Australia Malaysia Iran, Islamic Rep. Hong Kong SAR, China Kenya Philippines Mexico Qatar Bahrain Egypt, Arab Rep. Panama New Zealand United Arab Emirates Peru United States Finland Netherlands Israel Singapore Ireland Czechia Austria Iceland Greece Costa Rica Colombia Mauritius Hungary South Africa Saudi Arabia Pakistan Tajikistan Mongolia Jamaica Albania Senegal United Kingdom Italy Belgium Estonia Malta India Croatia Russian Federation Spain Bosnia and Herzegovina Armenia Guatemala Switzerland Sweden Korea, Rep. Turkiye Kazakhstan Trinidad and Tobago Dominican Republic Oman Source: Global Innovation Index 2022 (inverted); orange: Finland and South Korea; dark: ECA countries. As outlined above, this note explores what accounts for the different speeds of recovery, growth, and knowledge intensity in the ECA region during the post-1990 period, and in particular the role played by public institutions. While causality is difficult to attribute for country level changes, the note maps out what institutional features show greater and less difference, and traces patterns of change across countries as well as over time. In particular, it focuses on four aspects: (i) comparing various institutional dimensions across countries and time, (ii) differences in the size of the state’s footprint (fiscal, staffing, and state ownership), (iii) the ability of countries and their states to pursue reforms during growth and crisis periods, and (iv) how the state’s footprint has evolved for specific functions. It also touches on how the evolution of institutional performance has intersected with the opportunities offered by EU accession. How has institutional performance differed between HIC convergers and others? Wealthier countries have better institutions than poor ones (WDR 2017) – but strong institutions are rarely present from the outset in middle income countries seeking to move towards high income levels. Rather, developing good institutions is a process that needs to accompany growth and economic sophistication – as was the case in South Korea. To regain momentum towards increasing income levels, ECA countries had to reconfigure and rebuild institutions, newly establish competition as a driver of growth, and recover and repair growth foundations. While today’s HIC convergers in ECA follow the global pattern of having markedly better institutional performance than those that are still middle income countries (see also Raballand and Zovighian 2024), this note explores how they got there. Furthermore, exploring which institutional features improved across ECA versus mostly in HIC convergers is useful to better understand what institutional improvements are more critical to enabling transitions to the high-income level. Across ECA, property rights were established across most country sub-groups, while the speed and timing of improvements have varied. The Baltic states and Central Eastern European countries rapidly closed the gap with Western Europe (Baltics) or narrowed it substantially (Central Eastern Europe). Both groups reached a level of 0.8 (on a scale of 0 to 1) or higher from the early 1990s 14, largely removing property rights from being a direct bottleneck to economic development. In the Balkans and the Caucasus, this level was reached from the 2000s, and the 2010s respectively, only Central Asia has continued to lag significantly behind. 14 Source: V-Dem Property Rights Index. The index is based on a survey questions. 7 In contrast, for the ‘rule of law’ the gap between convergers and potential converging countries has remained wide (Figure 4). The rule of law and legal institutions are crucial for improving the functioning of other institutions, enabling wider access to credit, and delivering justice in society (WDR 2017 - Governance and the Law). Even for convergers, the average performance remains significantly below that for Western Europe, but approximates the Rule of Law quality in South Korea. The assumed importance for investors and the empirically observed trends in converging and non-converging ECA countries suggests that at least a ‘good enough’ legal predictability and contract enforcement is important to accelerate domestic and foreign investments. Early convergers (e.g. Slovenia, Czechia) had notably stronger rule of law performance than later convergers. Among the latter, Romania (reaching high income in 2019) exceeded Bulgaria (high income since 2024) over a period of time, although both remain at lower level of performance than earlier convergers. Figure 4: Rule of Law performance 1996 to 2022 Source: WGI. The data is based on different representative sources. Rule of law captures perceptions of the extent to which agents have confidence in and abide by the rules of society, and in particular the quality of contract enforcement, property rights, the police, and the courts, as well as the likelihood of crime and violence. The displayed data is the average rank. Overall government effectiveness (i.e. the overall ability of implementing policies, regardless of regime type or economic policy approach - Huntington 1968) has been much higher in early convergers compared to MICs, and the latter only made limited gains over time (Figure 5). The difference between early convergers and other ECA countries was already wide by the mid-1990s. Late convergers (Bulgaria and Romania) show a low- intermediate performance throughout and a deterioration in most recent years driven by declines in Bulgaria. For the group of ‘potential convergers’, government effectiveness improved until 2014 and then started stagnating, followed by deterioration in most recent years. ECA’s low middle income countries performed lowest for most of the past 30 years, but with only a small gap to the upper MIC group. After initial stagnation in the 1990s and 2000s, that group’s government effectiveness gradually improved after 2008 and in most recent years has exceeded the (declining) upper MIC group. Figure 5: GDP per capita and government effectiveness 8 The timing and trends of institutional quality across countries and groups in ECA indicate that domestic political commitment plays a primary role – and only where this is given can EU accession provide an additional pull. Early convergers - such as Hungary, Poland and the Baltic states - applied for EU membership in 1994-5 and became member states in 2004. At the point of starting accession processes, their general government effectiveness and rule of law quality was already quite robust, indicating greater inherent institutional readiness to start the evolution towards high income economies. Late convergers appear to have benefitted more from the pull provided by the accession process – but their institutional progress still remained more limited. Among ECA’s current MICs, 9 out of 19 have become EU accession candidates. Some started this process over a decade or longer ago, while three (Georgia, Moldova, and Ukraine) were recently added. The institutional progress in those that have been candidate countries for a while, such as Türkiye and several Western Balkan countries, further suggests that prospects for EU accession on their own – including significant support for improving public administrations, public procurement, and so on 15 – have limited impact. Institutional improvements need to be underpinned by domestic political commitment and associated conditions, which accession processes can then reinforce. A comparison of current MICs to converging countries five years prior to reaching high income status indicates further which institutional dimensions diverge more or less across the groups (Figure 6) . The greatest overlap exists for property rights and for access to state business opportunities. Government effectiveness shows as further apart. The dimensions for which the groups differ the most are (i) ‘transparent laws with predictable enforcement’, (ii) the rule of law index, (iii) rigorous and impartial public administration, and (iv) public sector corruption, as well as (v) judicial constraints on the executive. In all of these five dimensions, very few current MICs overlap with the performance that convergers achieved in the years leading up to high income status. In addition, on average, the higher income MICs perform somewhat worse on these measures compared to lower income MICs.16 Figure 6: Current middle-income countries compared to HIC convergers five years prior to reaching high income levels17 Differences in overall institutional quality translated in how quickly and fully ECA’s transition countries were able to shift the role of the state from dominating the economy to enabling private businesses as key drivers of growth. Performance improved more rapidly and strongly in converging countries, where governance of incumbent state enterprises changed especially quickly, ensuring greater financial discipline and reducing SOEs’ crowding or shutting out new competitors. Competition policies also improved more strongly among the subsequent converging countries, albeit more gradually (Figure 7b). Potential convergers/upper MICs continued 15 Including through extensive TA via OECD SIGMA (https://www.sigmaweb.org/) and regular assessments through progress reports and recommendations (https://neighbourhood-enlargement.ec.europa.eu/enlargement-policy/strategy-and-reports_en), as well as financial support and incentives. 16 Several of the wealthier middle-income countries in ECA derive a substantial share of incomes from natural resource wealth. Globally, higher NRM shares in economies are associated with lower quality of governance, as well as with greater economic volatility, given shifting market conditions. 17 Sources: V-Dem indicators, which the exception of the Government Effectiveness - Worldwide Governance Indicators. The scales of these variables have been standardized such that the data are centered at their worldwide means. A unit on the scale is thus a standard deviation of country scores in the original distribution. 9 to perform substantially worse and made a renewed push towards further policy progress only from the mid- 2000s. Among lower income MICs in ECA, progress stalled and flatlined from the mid-1990s until the mid-2000s, before resuming improvements, and starting to overtake upper MICs from the late 2000s. 18 Figure 7a and 7b: Governance and Enterprise Restructuring and Competition Policy Source: EBRD, based on assessments by the EBRD’s Office of the Chief Economist. Ratings range from 1 (soft budget constraints (lax credit Ratings range from 1 (No competition legislation and and subsidy policies weakening financial discipline at the institutions) to 4+ (Standards and performance typical of enterprise level); few other reforms to promote advanced industrial economies: effective enforcement of corporate governance) - to 4+ (standards and competition policy; unrestricted entry to most markets). performance typical of advanced industrial economies: Link to the methodology: https://2021.tr- effective corporate control exercised through domestic ebrd.com/structural-reform/. financial institutions and markets, fostering market- driven restructuring). Early HIC convergers experienced a virtuous cycle between better institutions and greater private investments, which in turn enabled further improving public sector performance. Among subsequent convergers, public institutions were good enough to enable an infusion of new technologies and management techniques from the mid-1990s. The ability to attract large-scale FDI – apart from bringing an infusion of new technologies – also helped to reduce fiscal stresses and enabled earlier re-building of public services, relatively free from front-line corruption (Figure 8). By contrast, corruption increased and remained elevated in many MICs or reduced significantly later in late convergers (Bulgaria and Romania).19 Figure 8: Public Sector Corruption Index with pre-and post-1990 periods for selected countries Source: V-Dem 18The EBRD’s Transition indicators were discontinued after 2014. 19‘Front-line corruption’ – i.e. requests for bribes at the point of service delivery. See also Global Corruption Barometer 2003-2017 editions (www.transparency.org). 10 The overall business environment in upper MICs improved between the 1990s and mid-2000s, but has mostly stagnated since then – with late convergers pulling ahead 5-10 years prior to reaching high income levels. While business environments do not correspond 1:1 to institutional quality, the direct and indirect links are significant. Other factors shaping the business environment, such as access to finance, logistics, and others are themselves be held back (e.g. financing) when key institutional dimensions such as the rule of law are performing poorly. Figure 9 shows the rating for the overall business environment, with the blue line being the average for convergers 5 years prior to reaching high income status20, the grey line being the late convergers (Bulgaria and Romania), and the orange line for other ECA MICs for which the data is available. Potential convergers have gradually improved their business environment, but this has been a slow process. Bulgaria and Romania crossed the ‘HIC-5y average’ just around the time they respectively reached high income status. Figure 9: Overall business environment (1993-2023) 7.5 6.5 5.5 4.5 3.5 2005 1993 1995 1997 1999 2001 2003 2007 2009 2011 2013 2015 2017 2019 2021 2023 Converger HIC-5 average Upper MICs Late Convergers (Romania and Bulgaria) Source: EIU data ‘overall business environment’. Based on the assessment of The Economist Intelligence Unit; note: limited da ta available for the lower MIC group. As data is not available prior to the 1990s, we cannot compare back to Finland and South Korea during their middle to high income transition years. By 1995, Finland scored 7.7 and South Korea 6.6, i.e. above and similar to ECA convergers five years prior to their respective years of reaching high income. Overall, most strikingly, ECA’s early convergers had significantly better institutions from an early stage, while for others, gains have been slow and uneven – leaving a large potential for change. While early convergers had to transform the role of the state, they benefited from early relatively strong government effectiveness and general rule of law – somewhat similar to Finland in the 1960s – in turn enabling early progress on creating enabling economic conditions for private businesses. Early and significant inflows of FDI helped with sustaining a positive momentum of bolstering economic recovery and fiscal revenues and avoiding intense and long constraints on fiscal space experienced by many ECA MICs. Late convergers – Romania and Bulgaria – shared institutional features closer to current upper MICs in ECA and to South Korea during its early middle-income stage – but eventually improved key institutional dimensions as well as their business environment. Their experience suggests that initial HIC transitions are possible with institutional performance still well below high income averages – and that even limited additional further gains may pay off. Furthermore, in recent years, gains in institutional quality have been more widespread at the lower end of the income spectrum than among countries already closer to the high-income threshold – helping to accelerate growth among the group of countries that has seen the least degree of economic convergence over the past 30 years. ECA’s Large fiscal footprints and the need for efficient delivery of enabling services One notable feature of states in ECA are large fiscal footprints. Revenue to GDP ratios are especially high in ECA HICs, but are also substantial in ECA MICs compared to middle income countries in other regions (Figure 10a). Both groups have followed similar patterns of more expansionary and consolidation periods in fiscal 20 i.e. using the average of the rating for each converger, 5-years prior to reaching the HIC threshold. 11 management over time, in response to external shocks, in particular the global financial crisis and COVID 19 (Figure 10b). Figure 10a and 10b: Government expenditure and Size of Government in MICSs by Region Source: IMF/GFS; the comparison across regions includes only middle-income countries. Aging societies – associated with substantial social protection and health expenditures - contribute to large fiscal footprints in many ECA countries. In South Korea, the fiscal size of the state was significantly smaller prior to its MIC to HIC transition and has remained smaller than in either ECA group – but challenges of funding an aging society set in only more recently. Most countries in both ECA HICs and MICs have sizeable social transfer expenditures (14.6% of GDP on average for HICs and 8.9% for MICs [2021 IMF/GFS data 21]); while spending on health (average: 6.8% of GDP in HICs and 3.7% in MICs) and education (4.9% in HICs, 4.6% in MICs) is more variable among MICs (ranging from 1.5% spending on health in Azerbaijan, to 5.6% of GDP spent in Belarus and Moldova). Moreover, spending on economic affairs ranges widely among MICs – from 2.5% in Armenia and 3.5% in Kazakhstan, to 6.6% in Azerbaijan and 7.7% in Tajikistan, but tends to be more sizeable across most HICs (ranging from 4.9% in Estonia to 9.2% in Hungary). Defense spending is higher in MICs than HICs (2.1% versus 1.5% on average), but also highly variable across countries in this group, with the highest levels in Armenia and Azerbaijan. Spending on administrative services varies widely among MICs – from 1.5% of GDP in Kazakhstan to 8.3% in the Russian Federation. Among HICs, spending on administrative services is the lowest in the Baltic states and Bulgaria (below 4% of GDP), and especially high in Hungary (8%). The comparatively large fiscal footprints of most converging countries - the smallest at the point of reaching HIC status were those in Estonia (33.4% of GDP - 2006), Romania (33.3% - 2019), and the Slovak Republic (36.4% - 2007) - indicate that fiscal size as such is not a constraint to the high-income transition. At a basic level, the recovery of revenue to GDP ratios has been positive. However, focusing spending on growth enabling policies and spending efficiently remain important concerns – and in turn depend on institutional quality. Taxes are a substantial cost to businesses and citizens, but a similar tax burden has different implications depending on the quality of institutions, and the resulting efficiency and effectiveness with which public services are delivered. In growth-enabling states, companies may face substantial direct and indirect taxation, but also benefit from many public services – high-quality transport infrastructure, a large pool of well-trained workers, various innovation programs, reliable and effective dispute settlement mechanisms, effective security provision and so on. In contrast, when institutional quality is low(er), a similar revenue ratio will be more burdensome: tax collection tends to be more unfairly distributed on the one hand, and on the other, a significant share of revenue collected ends up wasted on corruption and mismanagement in public procurement22, various payments to politically important interest groups, as well as inefficient processes in public administrations and service delivery. 21 2019 data for the Russian Federation; no breakdown by function available for most Western Balkan countries except Albania. 22 While difficult to measure directly, one indication is perceived corruption, which remains a top-level concern among most ECA MICs as well as in the late convergers (Bulgaria and Romania) (IPSOS 2021 regional survey). 12 Both HICs and MICs have taken some measures to improve spending efficiency and effectiveness . ECA remains the best performing region for aggregate average CPIA and PEFA scores, although further progress stalled in most countries since the 2000s - following rapid earlier progress. More recently, almost all ECA governments have made progress on digitalizing core systems (Figure 11), with several ECA HICs exceeding the performance of established high-income countries. Beyond direct efficiency gains, digitization has also supported reductions in (front-line) corruption (World Bank 2012). Many ECA countries have also reduced the share of public to total employment over time. In most ECA HICs, this is now in the range of 20-22 per cent of total employment but varies more widely among MICs. Figure 11: Progress on Digitizing core Government Systems Source: GovTech Maturity Index. The Core Government Systems Index (CGSI) is based on 15 indicators which capture core elements of the whole-of-government approach. More needs to be done to enhance spending efficiency, especially in ECA MICs that combine large fiscal footprints with limited institutional effectiveness and integrity. There are several important inefficiencies: in countries with continuing substantial capture and corruption problems, competitiveness in public procurement remains lower or decreased (Figure 12), and issues persist in ensuring better management of public sector companies (SOEs), such as non-meritocratic appointments and low productivity, increasing risks of recurrent bailouts.23 Also, especially for ECA MICs, PERs/PFRs highlight opportunities for better organizing and managing the large service delivery sectors – general education and health - to optimize care and education provided relative to resources deployed. While overall PEFA scores are strong for the region, efficiency/value for money related indicators, such as ‘performance information for service delivery’ remain weak (i.e. D and C rated). Logistics performance remains low in some MICs, indicating transport infrastructure as a major constraint that public sectors have insufficiently addressed. Among upper middle-income countries, Serbia stands out as having a high revenue to GDP ratio of 46.3%24, but a low quality of trade and transport related infrastructure (Logistics Performance Index infrastructure score of 2.4 25 in 2022), while Poland (3.5), Estonia (3.5) or Slovenia (3.6) achieve better infrastructure with a similar or lower revenue burden. Generally, Public Investment Management Assessments for ECA countries indicate significant weaknesses in planning, selecting, and especially implementing public investments (IMF 2020). 23 As documented in various recent ECA Public Finance Reviews (e.g. North Macedonia 2024 – inefficiencies in SOEs; Bulgaria 2023 on inefficiencies in public procurement due to low competitiveness); European Commission 2016. 24 GFS, 2022, including social contributions. 25 LPI ranges from 1 to 5, scoring countries on a range of dimensions. See: https://lpi.worldbank.org/about. 13 Figure 12: Access to State Business Opportunities trends 2000-2020 Avoiding capture: ensuring that markets and opportunities remain competitive As Schumpeter and later scholars have emphasized, the essence of creative destruction is to ensure that incumbents do not shift from profit- into rent-seeking, to preserve competitive markets and the drive to innovate. As the WDR 2024 (chapter 5) stresses, while elites have an important role to play in the middle- to high-come transition, constraining elites not to capture opportunities to the exclusion of others is a key challenge. While risks of state capture tend to be higher in middle- than in high income countries given that institutions fostering (political and economic) competitiveness tend to be less robust and entrenched – as shown earlier, capture can also occur at high income levels when a narrow elite group manages to weaken checks and balances. As the experiences of South Korea and Finland show, disciplining against rent-seeking and preservation can happen even when strong incumbents become established (the chaebols and Nokia respectively), provided institutions remain sufficiently strong and independent of incumbent interests. State capture enables greater rent- rather than profit seeking among connected firms while constraining others - but it has been hard to measure in ways that allows cross-country comparisons (David-Barrett, Kaufmann and Ceballos 2023). State capture constraints innovation-based growth, given that it leads to a channeling of resources and opportunities – e.g. for importing goods, winning public tenders and concessions, or enjoying tax exemptions – to connected firms, while holding back other companies. Such as distribution of opportunities between incumbents and non-incumbents can still allow some infusion of technology from abroad, but the resulting rent-seeking increasingly becomes a binding constraint as countries move towards high income thresholds (Pritchett, Sen and Werker 2017). It is also more prone to social upheaval due to discontent of unconnected businesses and citizens bearing the cost of incumbent protection – as well as a greater tendency of governments in captured environments to resort to repression and control. A new index allows tracing the evolution state capture across countries over time . Convergers performed far better than both upper and lower ECA MICs since the start of the series in the mid-1990s. However, there has also been some uptick among convergers since the mid-2010s. Furthermore, while initially state capture was higher in lower than in upper MICs, the gap started closing from the late 2000s, and reversed since the late 2010s, as more lower than upper MICs sought to constrain capture tendencies (Figure 13). One notable example is Moldova which initiated more serious reforms following a billion dollar fraud scheme (over 10 per cent of the country’s GDP at the time) committed in the early 2010s, in which a former Prime Minister, as well as the country’s largest oligarch were directly implicated (Longhurst 2023). Moldova’s recent governments have sought to systematically reduce and constrain state capture – making significant while still fragile gains. 14 Figure 13: State capture index, 1996-2022 Competitiveness among firms and the meritocratic selection of management can offer one way of observing potential state capture. Low competitiveness among firms will often be rooted in public sector actions, such as favoritism in licensing, taxing, public procurement, financing, or regulatory enforcement; while wide-spread non-meritocratic management indicates weak corporate governance rules and/or enforcement. The combination of both is most constraining – since it indicates that firms favored through regulatory action are not incentivized to show high performance, as might be the case in a country pursuing some form of well- calibrated industrial policy. Data indicates that both dimensions are low in North Macedonia, Bosnia and Herzegovina, as well as in Kazakhstan and Kyrgyzstan (Figure 14). In some HIC convergers, these dimensions are also still limited: Croatia, Romania, and Hungary perform below the 40 th percentile at least on one measure or both; and performance remains below the 50th percentile across both dimensions for most other MICs. Figure 14: Competitiveness among firms and meritocratic selection of management (WEF) 26 90 80 meritocratic selection for 70 management 60 50 partially 40 competitive & 30 meritocratic 20 10 high capture/low competition 0 0 20 40 60 80 Competitiveness among firms SOEs often form part of (partial) state capture, while the capture of specific institutions such as tax and customs administrations occurs in more severe cases. For instance, Szarzec et al., 2020 show that the average number of changes in the composition of management boards is higher in SOEs than in private enterprises, and higher shortly after election periods. The Corporate Governance of State-Owned Enterprises in Europe and Central Asia A Survey27 also shows variable vulnerability to political interference. For instance, SOEs in Belarus, 26 WEF/INSEAD – extent of market dominance - https://www.insead.edu/system/files/2023-11/gtci-2023-report.pdf (p. 213, and 221 – professional management); ratings are based on an executive survey. While this is a subjective method, it is one of the few ways of identifying ‘de facto’ rather than de jure performance. 27 https://documents1.worldbank.org/curated/en/210521608727226154/pdf/Corporate-Governance-of-State-Owned-Enterprises-in- Europe-and-Central-Asia-A-Survey.pdf 15 Kyrgyzstan Republic, Tajikistan and Uzbekistan are not required to establish an Audit Committee. Poor SOE management associated with extensive capture has also been a challenge in MICs in other regions (Alam, Mokate and Plangemann 2016). In several MICs in ECA, powerful individuals or networks have captured public institutions more extensively (Leitner and Meissner 2017). Greater state capture negatively affects state legitimacy, i.e. citizens' acceptance of authority and trust in governmental institutions. State legitimacy (‘trust in the state’) strongly influences a state's ability to implement policies, maintain social order, and foster economic development. Generally, state legitimacy in ECA remains well below OECD/EU values. While direct measurement of state legitimacy for ECA countries is limited, available statistics for the period 2006 to 2022 show a slight increase in state legitimacy for converger countries, following trends for other EU countries. In upper MICs, state legitimacy has stagnated over most of the past 15 years, at a comparatively low level (Figure 15). For lower MICs, the data suggests a gradual improvement from a low base from 2018 onwards. The most recent Life in Transition Survey (LiTS, Round IV) shows widely dispersed levels of trust in governments and presidents (as representatives of the state), with especially low levels in the Western Balkans, Moldova and the ‘late convergers’, Bulgaria and Romania. Compared to the 2016 LiTS III round, trust most strongly improved in several of the lower-income MICs – notably Armenia, Moldova, and Kosovo – which undertook reforms to reduce state capture and corruption. Figure 15: State Legitimacy trends – 2005 to 2022 Source: Fragile States Index28. Because state legitimacy conveys important advantages such as greater readiness to pay taxes and comply with regulations, higher state capture have a tendency to ‘shore up’ legitimacy with popular but costly and regressive subsidies or emphasizing nationalism (David-Barret 2021). Energy subsidies range from 2–19% of GDP in ECA countries (according to the IEA 2021 energy consumption subsidy database) – imposing a huge cost in the countries at the higher end of the range (World Bank 2024), including most of the Central Asian countries and Azerbaijan (IEA 2022). While less directly measurable, an emphasis on nationalism has increased in parallel to decreasing equity in business opportunities in several countries in the region, including some high income convergers. Pursuing institutional and policy reforms in good and in bad times As middle-income countries pursue the transition to high incomes, they will experience periods of buoyant international growth as well as shocks and recessions. Countries that have progressed faster from middle to high income levels are those which have been able to sustain growth and to minimize backsliding over time (Pritchett at al. 2017). South Korea had to grapple with the Asian crisis in 1997 but managed to re-accelerate growth after a few years, while Finland experienced challenges in the early 1990s and after the global financial crisis, followed by renewed growth spurts after some time. Achieving sustained growth requires continuing 28 The State Legitimacy Indicator measures how well a government maintains public trust, openness, and political integrity. It looks at factors like election fairness, corruption, and basic state functions such as tax collection. Additionally, it considers social unrest, protests, insurgencies, and political violence to gauge public perception of the government’s authority and stability. These elements reflect the state's effectiveness and how legitimate its power is in the eyes of its people. 16 efforts at institutional improvements, enabling the use of such institutions to develop and implement effective policies in response to challenges. Using periods of widespread growth to improve institutions and policies helps countries to be prepared for shocks and to rebound more strongly, while certain painful reforms can be more readily implemented during crises. During ‘good times’, reforms are facilitated by expanding fiscal space, some of which can be used to invest in improvements such as public sector digitization, greater bureaucratic capacity, or reforms to service delivery systems. Periods of economic growth and increasing revenue also enable adopting more social buffers for those negatively impacted by reforms. Furthermore, they allow for more focused attention relative to periods of economic shocks and recessions which often push governments into ‘fire -fighting’ mode. However, benign periods can also reduce the urgency for change and for ensuring that states operate efficiently, and it can be harder to win sufficient support for changes that hurt established interests. Some governments have effectively used crises to enact difficult reforms – notably towards further liberalization in specific sectors or markets, efficiencies in service delivery, or to remove subsidies (Ruch 2020). Especially the early converging countries in ECA used periods of stable economic growth for institutional and policy reforms – benefitting from the timing of EU accession prospects . For early convergers, the late 1990s and early 2000s were a period of intense policy and institutional reforms – coinciding with strong global and regional growth and an EU accession process which amplified motivation and provided a template for the combination of market liberalization and institutional strengthening. 29 Romania and Bulgaria – the ‘late convergers’ – started from substantially lower initial levels of institutional quality (Figure 5 above) and received a greenlighting as candidate countries five years after the initial group.30 They undertook some reforms prior to gaining accession in 2007, but stalled at a substantially lower level before being impacted by the global financial crisis. Following the GFC shock, Romania resumed greater reform efforts during a period of renewed substantial growth (>4% annual average for 2017-22). The contrast is even greater with today’s MICs, which pursued far less institutional reforms during the high growth period of the 2000s (see e.g. Figure 6 and 7 above), with some resuming reform efforts in the late 2010s. Countries that had completed more reforms prior to the global financial crisis were generally better able to respond when the crisis hit in 2008-09 and were able to overcome steep recessions (Brownbridge and Canagarajah 2021). Facing sharp contractions, several ECA countries had to undertake large-scale fiscal adjustments to avoid ballooning deficits (Figure 20). For example, in response to the crisis, Latvia further streamlined its public administration, closing half of its 75 state agencies at the time, started implementing an already prepared masterplan to make its network of hospitals more fiscally sustainable and to reduce the number of schools in the face of demographic changes (Aslund 2012). Several countries also undertook important reforms of labor markets to enable more flexible work in an effort to address unemployment – which increased due to the overall economic contraction as well as layoffs and hiring restrictions in public sectors. In contrast, countries that were less well prepared had to rely on simpler austerity measures, cutting expenditures across the board without targeted reforms. The post-crisis pace of reforms in the region slowed rather than accelerated (EBRD 2010). As a result, growth rates only partially recovered between 2010-2019 (Figure 17), while the Baltic states in particular saw a more rapid recovery and a new period of strong growth from 2014 onwards. 29 ECB (2010): Before the [global financial] crisis, the CESEE region was experiencing an economic boom with rapid GDP and credit growth, which in turn was driven by large capital inflows and benefited from strong global growth and easy global liquidity conditions. In addition, strong economic growth in the region was supported by positive expectations of EU convergence and euro adoption. 30 In late 1999 compared to 1994 for the early convergers. 17 Figure 16: Net Lending/Borrowing as % of GDP Figure 17: Change in growth rates before and after the global financial crisis Source: World Bank national accounts data, and OECD National Accounts data files. Many ECA countries adopted fiscal rules as an instrument to promote greater macro-fiscal stability in the 2010s. This includes all EU MSs (covered by EU fiscal rules), but also many ECA MICs. Exceptions include Türkiye, three countries in Central Asia, and Belarus, Ukraine, and Moldova, and three of the six countries in the Western Balkans.31 However, as it is challenging to ensure that fiscal rules are binding, this has been subsequently followed by the establishment of Fiscal Councils – so far mostly as an additional institutional mechanism in HIC convergers/EU MSs, but also in Serbia and Georgia.32 South Korea also established a Fiscal Stability Board in 2003 – following its experience of the Asian Financial Crisis – but did not adopt a binding fiscal rule. 33 Some countries used renewed periods of growth to restart greater reform efforts, but more need to follow . As Figure 9 indicates, ECA MICs initiated some reforms to their business environment once they emerged from the aftermath of the global financial crisis from 2014 onwards. Notable reform efforts emerged in Armenia and Moldova as well as Ukraine to reduce capture and corruption. Following the end of the COVID19 crisis and the aftermath of high interest rates, more countries should seek a renewed push towards reforms – especially as the risk of future shocks due to climate change and challenging geopolitics remains high. 31 IMF: Fiscal Rule dataset. 32 https://www.imf.org/en/Data/Fiscal/fiscal-council-dataset. 33 https://www.imf.org/en/Data/Fiscal/fiscal-council-dataset. 18 3. Evolving the role of the state across different functions The role of states varies and evolves across different policy domains . States' involvement varies from limited to extensive across sectors like economic management, education, energy, healthcare, and transport. States need to adapt their roles, selecting between direct delivery, regulation, and enabling private delivery. Cross- cutting institutional quality shapes how well states can perform the chosen limited or more extensive roles. Touching on challenges in areas such as competition policy, delivering quality education and healthcare services, and managing energy sector transitions, this section stresses the importance of clarifying and evolving the role of states - alongside improving governance overall - to enhance efficiency, effectiveness, and equity in service delivery. The specific role of states varies across policy domains, while being interconnected with overall institutional quality and performance. As Schumpeterian economic thinking has highlighted, the role of states is a combination of ‘getting out of the way’, setting a framework for private enterprise and investments, and taking action to provide certain goods and services. Thus, in some policy areas, states are likely to retain sizable direct delivery involvement; in others, the role has been or has yet to be recalibrated away from comprehensive direct delivery towards effective regulation, as well as towards public private collaboration, or towards more minimal involvement. Across these different role of states - changing and adjusting at different points in time - cross cutting institutional quality ‘shows up’ and impacts the ability to evolve as well as the quality of governance within specific functions. In economic management, states got ‘out of the way’ across most of ECA at early stages of the transition, but ensuring a competitive environment continues to vary substantially. Across the region, ‘getting out of the way’ by abolishing price and other controls happened early in the transition. However, establishing a framework to protect and enable competition also requires at least some pro-active role, good enough rules, and institutional capacities and commitment to implement and enforce policies. Key institutions involved include small and specialized but influential agencies such as competition (or anti-trust) authorities, as well as the judiciary – especially courts involved with commercial disputes, and various regulatory functions. As noted above, in many ECA MICs, judiciaries and regulatory enforcement are often not yet sufficiently geared towards ensuring level playing fields. In the energy sector, the role of states is currently facing a major re-configuration – and this is one of the challenges with accelerating energy transitions in the region, especially in ECA MICs with weak energy sector governance. Across many countries in ECA, states (partially) own legacy energy sector assets, as well exercising regulation and providing subsidies where those exist. Energy transitions entail opening more space for private investments, as well as significantly changing the regulatory role of the state from focusing on ‘keeping energy cheap’ towards orchestrating complex transitions that require large scale private investments and complementing public regulations and investments. Making this shift is demanding for the institutions involved, including energy regulators, Ministries of Energy, as well as center of government institutions coordinating across different economic and social interests of power producers and consumers. Overall government effectiveness and efficiency weaknesses set out in section 3 impact the government’s management of the sector in many ECA MICs, with limited financial transparency and high political influence on managerial appointments of energy sector SOEs (IMF 2017; SELDI 2022). On the one hand, states in ECA have gradually opened energy sectors to RE investments and adopted enabling legislation, but on the other hand, hold back such investments by providing subsidies for fossil power and governance challenges impacting investor certainty and bankability. Improving how states function and adjusting the role of the state matters also across other policy domains, notably health care and transport. For example, health care sectors have undergone various shocks and shifts, evolving towards mixed public-private systems, but with many inconsistencies and leaving a high burden on citizens. Even where delivery mainly remains public, out of pocket payments have often reached levels of 40 or even 80 per cent of total spending (e.g. Serbia, Armenia) as insurance schemes have not yet sufficiently expanded. In transport infrastructure development and management, the role of states has gradually shifted towards greater private involvement especially in highway development. For each of these areas, a longer-term model of the role of the state and ensuring clear funding and risk-sharing models are still ‘work in progress. Across these different functions, clarifying the extent of the state and how it contributes to effective, efficient and equitable delivery is critical, as well as ensuring (greater) integrity. 19 The role of states in innovation in ECA States have long played a substantial role in education, including higher education, in funding research and development, and in incentivizing business R&D as key ingredients for economies to become able to move into high value added products. Governments have established universities and research institutes. The other main source of funding are companies – typically medium to larger in size, and various forms of venture capitalism – which has vastly expanded in scale over recent decades. In developing skills, and to a lesser degree in research and development, states retain large roles across the region - but how the state exercises its role needs to evolve. General and to a significant extent also post- secondary education remain predominantly public across ECA. General education is one of the largest segments of public sector employment, and as such is linked with the quality of public sector HR systems and practices. While support for research and development remains more limited compared to HICs in nominal terms, direct and indirect government funding plays an important role especially in upper-middle MICs in ECA.34 However, the quality and innovation focus of ECA’s public education systems varies widely – ranging from an excellent school and university system in Estonia to still very weak in several lower MICs. In tertiary education, states in several MICs seek to exercise excessive control, while neglecting a strong enough role in effective quality assurance, especially of new private universities. Given public-private complementarities, translating innovation into firm and overall economic growth - is likely to be greatest in countries where public and private efforts combine. As Mazzucato and Rodrik (2023) as well as Aghion and Schumpeter earlier have emphasized, states can play especially important roles in funding and stimulating foundational research that led to breakthrough discoveries. Private R&D is often especially good at covering the stretch from basic research findings to practical application, developing user friendliness, and commercialization. Strong complementarities exist both in Finland and in South Korea which rank 1 st and 8th globally on the number of patent applications per million people. 35 Pre-1990, more highly developed parts of ECA had built up significant state-run research capacity within the constraints of an overall command and control systems. 36 This capacity combined both civilian and military R&D (with the latter being most closely linked to state funding and direction across countries, and linked to the core role of the state of organizing external defense). Within the civilian sector, innovation was typically weaker, given the lack of competition as an incentive, a primary focus on meeting planning targets based on existing technologies within companies, as well as political constraints on the free flow of ideas. Türkiye is the notable exception within the region with stronger private traditions, and several highly performing private universities that were established in the 1990s.37 During the initial transition crisis, much of the existing state-run research, as well as higher education capacity, deteriorated in the context of general institutional and fiscal weakness. A lack of resources and direction impacted public R&D, while some capabilities transitioned into the newly emerging private sector. In higher education, bribery and in some countries a rapid proliferation of new for-profit academies and universities eroded quality. R&D funding contracted, and many scientists emigrated (Vizi 1993). With economic recoveries, higher education systems re-stabilized, but effective quality assurance and academic excellence remain challenges even 30 years on. Higher education improved more rapidly in early convergers such as Slovenia or Czechia than in later ones such as Romania and in current MICs (World Bank 2010). A majority of ECA countries joined the Bologna Accord and the European Higher Education Area (EHEA) to promote comparability in the standards and quality of higher-education qualifications.38 As part of the transition, furthermore, governments in ECA had to make a shift from a link between (public) higher education 34 Based on Eurostat data, available for the Western Balkans, Türkiye, and Russia, but not for other ECA MICs. The share has remained especially high in Russia where public funding accounts for over 60 per cent of total R&D spending. It remains also high in Serbia and other Western Balkan countries. 35 WIPO data 2022. 36 See e.g. Loren Graham (1993). 37 Cankaya and Istanbul Bilgi Universities. 38 Signatories to the Bologna Accord include all ECA countries which joined the EU, plus: Türkiye (2001), Albania, BiH, North Macedonia, Russia, Serbia (all 2003), Armenia, Azerbaijan, Georgia, Moldova and Ukraine (all 2005), Montenegro (2007), Kazakhstan (2010), and Belarus (2015). 20 systems and public companies, to different and more complex interfaces between public and private initiatives (see e.g. World Bank 2013 on R&D in the Western Balkans). However, the quality of higher education and research often remains uneven and fell behind relative to emerging strong academic institutions in Asia – with only two universities in the region reaching the top 100-200 range in international rankings.39 To date, state funding for R&D, whether directly or via tax incentives, has remained lower even among high income convergers, compared to Finland, South Korea, or EU member averages . Romania – as a late converger, performs especially low40, while Bulgaria is slightly ahead, despite overall slower convergence (Figure 18). Among the ECA MIC countries for which data is available, Bulgaria, Serbia and the Russian Federation have relatively higher public R&D spending. Figure 18: R&D expenditure as % of GDP Patent applications per million population range widely among ECA MICs and HICs (Figure 19a and b), while some countries in the region do well on the ratio of ‘unicorn’ companies 41 per population. For patents, Russia and Türkiye are clear front runners among the MIC group, in absolute terms (14 th and 21st globally) as well as per million population.42 Poland and Czechia have the highest number of patents among the high income convergers. Compared to Western Europe, R&D spending and patents are still significantly lower, even among regional front-runners such as Poland. On ‘unicorn density’ (i.e. unicorns per population), Estonia is a front - runner, followed by Lithuania, Latvia and Poland as other strong performers in the region. Among EU member states, Estonia also has the highest venture capital investments in per capita terms by far. Figure 19a and b: R&D spending and patent applications (including and excluding Western Europe) 39 Charles University in Czechia, and Lomonosov University in the Russian Federation. 40 See also: Rusu (2022) https://blogs.worldbank.org/en/europeandcentralasia/how-commercialization-helping-unlock-innovation- potential-research. 41 I.e. privately held companies with valuations of over USD1bn. 42 By comparison, South Korea ranks 4th, and Finland 19th in the total number of patents (WIPO data). https://www.wipo.int/edocs/statistics-country-profile/en/_list/l1.pdf. 21 The employment level of tertiary graduates is another indicator of greater knowledge intensity . A greater absorption of graduates into the private sector can signal that private sector innovation and higher end outputs are intensifying and that the private sector is overall becoming more dynamic. Across most countries in the region for which data are available, employment levels of tertiary graduates have trended upwards – but they remain substantially higher in ECA convergers compared to countries that have remained at middle income levels (Figure 20).43 Among middle income countries with comparable data, Türkiye stands out as the main declining performer in recent years, while most other MICs have seen improving trends, albeit all still below the averages seen in convergers. Figure 20: Employment ratios of tertiary graduates 90 80 70 60 50 2014 2015 2016 2017 2018 2019 2020 Converger average MIC average Turkiye Source: Eurostat. MICs included: Bulgaria, Montenegro, North Macedonia, Serbia, Türkiye The performance of ECA’s innovation systems shows that it remains challenging to shift from infusion to cutting edge innovation even after reaching high-income levels, alongside some remarkable successes. On the one hand, several countries in the region reached high income status over a decade ago, but operate mainly based on strong technological infusion, such as parts of car and other manufacturing moving production from Western Europe into the region (Pellenyi 2020). On the other hand, Estonia, and to a lesser extent other Baltic states, Czechia and Poland show a real take off of innovation often focused on digitalization, and based on comparatively strong general public education systems, and - especially in Estonia - deliberate efforts at creating favorable regulation and a welcoming eco-system of public and private initiatives (World Bank 2016). 43 Overall employment rates include employment in public and private sectors. 22 4. Conclusions and policy implications Building and continuously adapting states to play enabling roles for economic growth and innovation is a challenge and an opportunity across ECA. States, and the specific institutions they are composed of, play critical roles in providing a framework for markets and for innovation. Some states in ECA already play effective enabling roles and are doing so in ways that (mostly) use fiscal resources efficiently. Some are in the process of undertaking a renewed push for better institutions to create a more supportive and open environment for businesses and people to achieve their potential. Many still grapple with various combinations of state capture, inefficient allocation and management of financial and human resources, and with establishing institutions suited to effectively supporting knowledge economies. Challenges manifest across functional areas – economic management, education and innovation, managing energy sectors and low carbon transitions, as well as others not covered in this note. Considering the experience across different ECA countries over the 35 years since 1990, as well as Finland’s and South Korea’s earlier transition experience, five key issues stand out: • First, improving institutional quality beyond the levels already reached by most upper middle-income countries matters (even) for making the MIC to HIC transition. Early converging ECA countries all had key elements of institutional quality from the early 1990s that were well ahead of other upper middle- income countries globally, as well as of other countries in ECA. Earlier, Finland in the 1960s had institutional qualities far above MIC averages. Late convergers started pulling ahead of other upper MICs in terms of institutional quality at least to some extent in the run-up to reaching high income levels. South Korea’s experience highlights that the ‘take off’ in terms of institutional development can start from a low base, and still enable development from low- to middle income status, but also shows a real acceleration during the middle-income years. • Second, ‘upscaling’ institutions requires social and political consensus, with a mandate and political will, which EU accession processes can help to reinforce, but not generate where they are absent. Most of the early convergers started with already higher institutional quality at the time of applying for EU membership. In contrast, the two late convergers and Türkiye started from similar income levels as well as similar levels of (lower) institutional quality in the 1990s. EU anchoring helped to ‘pull’ Romania and Bulgaria (to a lesser extent) towards institutional quality sufficient to reach high income levels by 2019 and 2024 respectively, while Türkiye saw institutional backsliding. More recently, institutional improvements across different dimensions have started to happen in several of the lower MICs in ECA, which is encouraging for their prospects of higher rates of growth over the coming years. However, sustaining the social and political consensus needed to underpin continuous gains will remain a challenge – as indicated by recent experiences in Georgia and Moldova. Reforms should aim at building and strengthening the consensus for better institutions and an enabling business environment. • Third, large states – by fiscal, staffing or SOE footprints – have not been a hindrance for MIC to HIC transitions, but only as long as institutional quality ensures that these resources are used well enough. Compared to South Korea, states in ECA were fiscally larger when reaching high income status, in part due to aging societies and associated high social transfers; public sector wage bills in ECA convergers also remain slightly above high-income averages. However, institutional quality was comparatively high in most HIC convergers already during their MIC years. By contrast, some of the region’s current upper MICs combine (very) large fiscal, personnel, and SOE footprints with comparatively high corruption, limited rule of law, and limited equity of accessing state business opportunities. This imposes a high tax burden on businesses and households without commensurate service delivery, including large fiscal risks and often mediocre or poor-quality delivery by SOEs. • Fourth, state capture continues to pose a risk in several MICs as well as in some HICs in ECA and can stifle competitive and dynamic markets at different stages - at a scale beyond simple economic incumbency. With state capture, the regulatory, financial, and other powers of the state are used to provide greater opportunities to companies associated with political leaders or core elites compared to others, in a way that seeks to cement the position of incumbents. In contrast to industrial policy – also involving the use of regulatory powers - there is no or only a weak link to encouraging technological 23 upgrading and competitiveness. The objective is to entrench an incumbent individual or group both politically and economically. State capture is pernicious because it is systemic rather than limited to selected economic incumbents and actively uses the power of the state to curtail open competition. To sustain growth, ECA countries need to maintain or shift towards providing a regulatory and financial level playing field for all companies both de jure and de facto. • Fifth, to regain higher growth performance, ECA states need to manage risks and pursue reforms both in good and in bad times. Reforms such as streamlining internal organization or improving regulatory governance are easier to carry out when fiscal resources are growing (‘good times’). However, many ECA MICs have at best pursued partial reforms during such windows of opportunity. To periods of crisis, they have reacted with austerity measures, but limited deeper reforms that strengthen performance, efficiency, and resilience to future shocks. • Sixth, the role of the state needs to be calibrated across different functions, responding to policy challenges and objectives arising at different points in time. In education (and health) service delivery, well performing ECA states continue to combine direct public sector delivery with appropriate regulation and quality assurance of private providers and enable private-public interfaces and collaboration. In energy sectors, the role of the state needs to undergo a significant transition from still widespread ownership of generation and distribution capacity to enabling a much greater private share and guiding a complex transition. Reluctance to adjust the role of the state holds back progress across a number of countries and sectors in the region. • Finally, on the relationship between developing knowledge intensive economies and institutional quality, ECA offers three main insights: first, the region’s most innovative economies also have some of the highest levels of institutional performance. South Korea as well as Finland and other highly knowledge-intensive countries have continued to evolve how the state enables research and development and innovation in the economy to maintain competitiveness in an evolving world – enabled by overall effective and non-corrupt institutions. Second, some of ECA upper MICs – notably Türkiye and Russia - have developed comparatively strong knowledge generation and innovation capacity as measured in terms of the number of patents per million population. Yet, turning research outputs into broader knowledge-based growth requires better institutions and an enabling business environment – as found in regional innovation-based economy leaders such as Estonia, Slovenia, Latvia and Lithuania, Czechia, and Poland. Third, technological infusion remains an important opportunity, especially for the ECA’s lower MICs that are pursuing improvements to their overall institutional and business environment. All of these points also outline the agenda of reforms which ECA countries, and current MICs pursuing transitions to high income status in particular: upgrading institutions and constraining capture to enter a virtuous cycle towards dynamic growth, capable of generating increasingly knowledge intensive and higher quality jobs. 24