POLICY RESEARCH WORKING PAPER 2472 Uzbekistan and Kazakhstan Despite a common parentage for most of the twentieth century, Uzbekistan and A Tale of Two Transition Paths Kazakhstan followed seemingly different paths in Asad Alam the transition to a market Arup Banerj'i economy. Uzbekistan adopted a cautious, gradual approach to market reform, while Kazakhstan followed a more aggressive strategy. But has Kazakhstan done better economically? The World Bank Europe and Central Asia Region Poverty Reduction and Economic Management Sector Unit H November 2000 I POLICY RESEARCH WORKING PAPER 2472 Summary findings Uzbekistan adopted a cautious, gradual approach to several competing and complementary hypotheses about market reform. Kazakhstan followed a more aggressive why the paths the two Central Asian countries took may strategy. But while Kazakhstan may have achieved a have led to different economic outcomes. One possibility better policy environment, its economic performance has is that missing pieces in reform-especially deficiencies not been better than Uzbekistan's. in the competitive environment-in combination with a Alam and Banerji examine the interplay between rapidly diminishing role for the state may have limited policies, institutions, and initial conditions to examine the gains from the policy reforms in Kazakhstan. This paper-a product of the Poverty Reduction and Economic Management Sector Unit, Europe and Central Asia Region-is part of a larger effort in the region to understand differences in policies, institutions, and performance in transition economies. Copies of the paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contact Lorie Henson, room H4-347, telephone 202-458-4026, fax 202-522-2751, email address lhenson@worldbank.org. Policy Research Working Papers are also posted on the Web at www.worldbank.org/research/ workingpapers. The authors may be contacted at aalam@worldbank.org or abanerji@worldbank.org. November 2000. (25 pages) The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the senies is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Produced by the Policy Research Dissemination Center UZBEKISTANAND KAZAKHSTAN: A TALE OF TWO TRANSITION PATHS? by Asad Alam and Arup Banerji* * The authors are senior economists at the World Bank. The authors are grateful to Pradeep Mitra, M. Ataman Aksoy, Lev Freinkman, and K. Tanju Yurukoglu for comments on an earlier draft. Research assistance for a first draft was provided by Rohit Malhotra. UZBEKISTAN AND KAZAKHSTAN: A TALE OF TWO TRANSITIONPATHS? Two roads diverged in a wood, and I-- Took the one less traveled by, And that has made all the difference. --Robert Frost "The Road Not Taken" (1916) I. INTRODUCTION Uzbekistan and Kazakhstan-the two most populated countries in Central Asia- provide an interesting contrast in terms of their transition strategies and economic outcomes. Kazakhstan has adopted an aggressive strategy of liberalization, while Uzbekistan has been much more conservative, preferring to transform its economy using its own form of gradualism and slow sequencing of reforms. Kazakhstan clearly Figure 1: EBRD Transition Indicators Compared, 1999 Large-Scale Privatization Securities Mkts. & NBFI KAZ Small-Scale Privatization Banking Reform & /'^¢,,i Banteres Raef Lib. Enterprise Restructuring Interest Rate Lib., Competition Policy Price Liberalization Trade and Forex Liberalization Source: EBRD 1999 Transition Reoort dominates Uzbekistan in the pace and extent of its policy reforms, particularly in areas such as trade liberalization, small privatization, price liberalization, and interest rate liberalization (see Figure 1). This judgement fits comfortably with the verdict of other Page 2 observers. ' However, in two institutional areas critical for transition success-enterprise restructuring and governance, and competition policy-the institutional framework is equally deficient in both countries. But how have the two countries fared in terms of their economic outcomes? Certainly, if GDP data for the two countries are to be believed, Uzbekistan has outperformed Kazakhstan in terms of aggregate output growth since independence. Both countries saw their production drop sharply in 1992, due to the severe terms of trade shock and economic dislocation that accompanied the breakup of the Soviet Union. The extent of the decline was much smaller in Uzbekistan and the subsequent rate of recovery Figure 2: Uzbekistan and Kazakhstan: 1999 Macroeconomic Performance Compared Real G DP Index (% of 1991) 95 5. U BEKISTAN Foreign Exchange Reserves Average CPI Inflation (%) ($ bn) Current Account Deficit (% Total Revenues (% of of GDP) GDP) KAZAK STA Fiscal Deficit (% of GDP) Source: World Bank also higher. Consequently, by 1999, Uzbekistan's GDP had recovered to 96 percent of its 1991 level as compared with 63 percent for Kazakhstan.2 In fact, over the five years 1995-99, Uzbekistan saw a cumulative 10 percent real economic growth at a time when the Kazakhstan economy shrank by a cumulative 9 percent. Even on the basis of other macroeconomic performance indicators, it is not clear that Kazakhstan has performed any better. While inflation has been lower and foreign exchange reserves higher, revenue performance has been poorer and the twin deficits on the fiscal and current accounts were For instance, the World Bank's Country Policy and Institutional Assessment (CPIA) for 1999 and the Heritage Foundation and Wall Street Journal's 1998 Indices of Economic Freedom. 2 The cumulative decline in GDP in Uzbekistan has been the lowest of the former Soviet Union countries even if alternative estimates based on electricity consumption are used. See Zettlemeyer ( 1999) for these estimates. Page 3 larger in Kazakhstan (see Figure 2). Clearly, prima facie, the evidence suggests that despite better economic policies, Kazakhstan has not outperformed Uzbekistan in terms of macroeconomic outcomes. This paper attempts to explain this seeming paradox by putting forth a set of hypotheses, and then matching them with available facts. These hypotheses separately, or, perhaps more likely, together, seem to explain much of the difference. While it is difficult to come up with definitive conclusions given the ongoing economic and political transformation in these countries, it is possible to lay out a set of issues and identify areas for further investigation. In this context, it is relevant to note that previous empirical research on the CIS countries, including those of Central Asia, have focussed either on individual country studies or cross-country econometric analyses.3 This note takes a comparative case study approach, seeking to specifically analyze the dichotomies in policies, institutions, and performance between these two neighbors, in an effort to understand why the outcomes were very different from those expected ex ante. After this introduction, the paper begins by examining the recent socioeconomic history of the two countries in Section II. Thereafter, it goes on to lay out six hypotheses, examining each in turn. II. THE SOCIO-ECONOMIC STORY SO FAR ... Despite more aggressive inflation management-or perhaps because of it-output collapse was more dramatic in Kazakhstan and more prolonged. The balance between monetary and fiscal adjustment in their stabilization policies has been quite different, with Kazakhstan clearly demonstrating a tighter monetary stance. The quality of fiscal adjustment has been poor in both countries, but Uzbekistan has had a much larger emphasis on social and physical investment, and has managed to avoid any large buildup of payments arrears. The biggest payoff to Kazakhstan from its more market-friendly economic policies has perhaps been the large inflows of foreign direct investment that it has been able to attract. While the narrow enclaves of the extractive sector into which these inflows have gone limits the potential of significant multiplier effects in the Page 4 economy, these foreign flows have helped strengthen Kazakhstan's balance of payments even as Uzbekistan struggles with foreign exchange restrictions, declining exports, import controls, and severe import compression. Notwithstanding these differences, the prospects for sustained, robust and equitable growth in both of these countries remain dubious-though for separate reasons. Uzbekistan's prospects are undermined by the underlying macroeconomic vulnerabilities in its policies. Kazakhstan's macroeconomic policy stance, while more flexible and market-responsive than its neighbor's, is both buoyed and weakened by its reliance on exports of commodities. Output Collapse. Both the depth and the length of the post-independence recession has been more severe in Kazakhstan. In 1999, Kazakhstan's GDP was still only 63 percent of its 1991 levels as compared with 96 percent in Uzbekistan (see Figure 3). Uzbekistan has also had 4 years of continuous growth, while Kazakhstan grew only in 1997 and again in Figure 3: Real GDP Index (1991=100) [ KAZAKHSTAN U-ZBEKISTAN 100 - 95 85- 80- 75 ~ 70- 65 __ 60 - ,,T 1991 1992 1993 1994 1995 1996 1997 1998 1999 Source: World Bank 1999. In 2000, Kazakhstan does expect robust growth, but that has been highly influenced by the large rebound in the price of Kazakhstan's commodity exports, especially petroleum. 3 See, for instance, Selowsky and martin (1996), de Melo et al (1997), Fisher et al (1998). Page 5 Stabilization. Even as output has fallen more in Kazakhstan, inflation has decelerated faster. Kazakhstan has been very successful in bringing inflation down to single-digit levels, while Uzbekistan is still struggling to keep it under 30 percent (see Figure 4). In their stabilization policies, the two countries differed sharply in the balance Figure 4: Inflation, 1992-99 * K azakhstan -U zbekistan I 0000 Su 1000f 0 0- 10- 1992 1993 1994 1995 1996 1997 1998 1999 Source: World Bank between monetary and fiscal adjustment. In Kazakhstan, the policy bias was clearly in favor of a tighter monetary stance given the low revenue base (see Table 1). This may have been responsible for the faster deceleration of inflation in Kazakhstan. Compared to very high levels of monetary growth during the early years (1992-94) when it averaged between 500-700 percent in both countries, the averages fell sharply in 1997-99 to 34 percent in Uzbekistan and just 14 percent in Kazakhstan. The monetary adjustment in Kazakhstan, therefore, was noticeably sharper. In contrast, fiscal adjustment in Kazakhstan was not as acute as revenue constraints and the need to maintain basic public expenditures impinged upon fiscal policy choices. At the end of 1999, Kazakhstan had a fiscal deficit of about 4 percent compared with 2.2 percent in Uzbekistan. Quality of Fiscal Adjustment. But what was the quality of this fiscal adjustment? Uzbekistan may have had larger overall fiscal contraction, but was that appropriate in terms of maintaining social welfare and enabling future growth? The data suggests that Page 6 Uzbekistan may indeed have managed the fiscal adjustment better, notwithstanding the large unknown stocks of contingent liabilities in both countries emanating from the banking and enterprise sectors. There are four areas of clear difference. Table 1: Monetary and Fiscal Adjustment, 1992-99 M2 Growth Fiscal deficit 1992 1999 Av. 97-99 1992 1999 Av. 97-99 Uzbekistan 468 32 34 -18.5 -2.2 -2.6 Kazakhstan n.a. 3 14 -6.9 -4.1 -5.2 Source: World Bank The first set of evidence relates to the large build-up of pension and payments arrears in Kazakhstan until 1999. Despite the smaller fiscal contraction, Kazakhstan had by 1998 built up budgetary arrears equivalent to 5 percent of GDP, which are now being gradually reduced as the commodity boom brings in unexpectedly high revenues. In Uzbekistan, there is no evidence to suggest a systemic build up of budgetary arrears, though delays in wage and pension payments are reported.4 Second, govermnent investment has suffered heavily in Kazakhstan, falling to below 2 percent of GDP by 1999. A large part of even that investment is connected directly or indirectly to the construction of Kazakhstan's new capital, Astana. In contrast, public investments in Uzbekistan were 7 percent of GDP. Investments in productive physical assets-roads, schools, and other infrastructure-have received special attention in the Uzbekistan government's public investment program, though there are some questionable investments in the industrial sector. Third, social sector investments have been severely eroded in Kazakhstan, while they have been protected (at least in relative terms) in Uzbekistan. In 1998, Kazakhstan invested 3.4 and 2.6 percent of GDP on education and health respectively, while Uzbekistan invested 7.2 and 3.3 percent of GDP respectively.5 4 The official position of the Government of Uzbekistan is that there are no budgetary arrears. However, evidence from the Voices of the Poor consultations conducted for the World Development Report 2000 reveals that the poor regarded non-payment and delays in wage payments as one of their most serious problems. In addition, a Cabinet of Minister's Resolution of November 19, 1999, instructed all economic entities, including budgetary organizations, to clear all wage arrears before December 1, 1999, and made the heads of such entities and of the commercial banks servicing those entities personally accountable for the prompt payment of salaries. A key policy practice in Uzbekistan which may have precluded the growth of arrears on wages and pensions is the restriction on cash payments by the banking system for wages and pensions only. See World Bank (1999) chapter 7 for data on the evolution of health and education expenditures in Uzbekistan. Page 7 Finally, the pattern of budgetary financing in Kazakhstan raises questions both over the sustainability of the current fiscal stance and over the appropriate use of budgetary resources, as the large government consumption was essentially financed through capital sales. Revenues from privatization were on average 4-5 times higher as a share of GDP than Uzbekistan (see Table 2). In fact, Kazakhstan's fiscal proceeds from privatization are the largest for any of the countries in transition, spurred mostly by sales of large oil concessions in the latter half of the 1990s. These higher privatization proceeds enabled Kazakhstan to finance its large budgetary deficits. But to the extent that fiscal needs drove privatization, the quality of privatization may have suffered. Moreover, to the extent that capital sales have financed the current government consumption, such an approach raises not only questions about the best use of privatization proceeds but also about its intergenerational equity. TABLE 2: PRIVATIZATION PROCEEDS, 1993-98 (% of GDP) 1993 1994 1995 1996 1997 1998 Kazakhstan 4.5 1.7 3.1 3.8 3.1 4.3 Uzbekistan 0.2 0.7 0.8 0.8 0.5 1.0 ECA Transition Average 0.6 0.5 0.5 0.6 0.9 0.7 Source: Tanzi, V. (1999) External Vulnerability. The generally better policy environment in Kazakhstan has enabled it to develop a stronger balance of payments position, despite larger current account deficits. Yet both economies have external vulnerabilities, albeit for different reasons. Traditional ways of looking at external vulnerabilities-such as months-of- import coverage and external debt-to-GDP ratios-are misleading for Uzbekistan. This is because imports in Uzbekistan are severely compressed through administrative means and therefore give an upward bias to the import coverage figures. Similarly, the foreign exchange is administratively set at a grossly over-valued rate.6 As a result, the GDP figure in US dollars is biased upwards and the ratio of external debt-to-GDP is biased downwards. 6 A 50 percent devaluation of the official exchange rate in May 2000 and various other liberalization measures have helped to reduce the curb market premium on foreign exchange from a peak of 500 percent in end- 1999 to about 200 percent by end-June 2000. Page 8 Liquidity indicators for the two countries as at the end of 1999 were reasonable with debt-service to exports of 18% for Uzbekistan and 21 percent for Kazakhstan (see Figure 5) and foreign exchange reserves were adequate to provide coverage against short- term debt flows and debt-service.7 However, in both countries, the ratio in debt-service to exports has been increasing, driven of part by new short and medium term borrowings. In Kazakhstan, recent forays into the Eurobond market have left a legacy of high-priced borrowings of relatively shorter maturities. A 1999 Eurobond placement was at 825 basis points over US five-year treasuries, while an April 2000 issue was, at a premium of 500 basis points, still expensive debt, especially given the uncertainties of the commodity- driven fiscal cycle in Kazakhstan. Figure 5: External Liquidity Indicators, 1999 *Uzbekistan *Kazakhstan 100 17.8 20.7 1 0 2. 2.4 1 - '_ _ _ _ _ Reserves to Short- Reserves to Debt Debt Service to Term Debt (%) Service (%) Exports (%) Source: World Bank, IMF Quality of Foreign Direct Investment (FDI). In 1998, FDI per capita in Kazakhstan was $74 compared with $7 in Uzbekistan, while cumulative FDI inflows over 1989-98 were $5,661 million in Kazakhstan compared with $533 million in Uzbekistan.8 However, more is not necessarily better. Most of the FDI in Kazakhstan has been into the extractive industries, especially the oil and gas sector, with low multiplier effects in Empirical work by Rodrik and Velasco (1999) has shown that reserves to short-tenn debt ratios of less than unity are strongly associated with future financial crises EBRD (1999), Table 3.1.6, pg. 79. Page 9 the economy. Consequently, potential growth impacts were not commensurate with the large flows. In relative terms, Uzbekistan saw more FDI flows into sectors with larger multipliers-automobiles, electronics, textiles, chemicals, mining, and agro processing. At the same time, this does not mean that despite their short-term economic gains, these foreign investments in Uzbekistan are viable over a more medium- to longer term horizon. Much of the foreign investment in Uzbekistan is directed by the Government into sectors that the Government feels are 'strategic' for the future and are consistent with its vision of an industrialized nation, but not necessarily with the country's comparative advantage. A strategy of picking 'winners' entails high risks and is typically subject to large economic losses over time. Moreover, the distorted policy environment in Uzbekistan, particularly as it relates to the incentive bias against exports, is already undermining the profitability and growth prospects of these new industries. Figure 6: Trends in life expectancy at birth (years) 75 - 70- 0 65 - 60- 55 1991 1992 1993 1994 1995 1996 1997 1998 1--Kaz. Male -O--Kaz. Overall ---Uzb. Overall Source: World Bank Social indicators. Both Uzbekistan and Kazakhstan have maintained the very similar basic social indicators that inherited from the Soviet Union (see Table 4, below). But Kazakhstan has slipped dramatically in terms of life expectancy at birth, driven mostly by a rapid fall in male life expectancy. This has fallen to 59 years as of 1998, compared to 66 years in Uzbekistan (Figure 6). At the same time, educational participation has also been comparatively poorer in Kazakhstan (Table 3). Page 10 These overall trends is against the backdrop of large differences between the two countries in their public expenditures on health and education. As noted earlier, Kazakhstan's spending on health and education is about half that in Uzbekistan. Health Table 3: Changes in School Enrollment Patterns Primary enrollment as Secondary enrollment as Tertiary enrollment as % % of relevant age group % of relevant age group of relevant age group 1980 1996 1980 1996 1980 1996 Kazakhstan 85 98 93 87 34 32 Uzbekistan 81 78 106 94 29 36 Source: World Bank and education outcomes, of course, depend on both public and private expenditures on the provisions of these services, the efficiency of these expenditures, as well as on a host of other factors-such as sanitation, supply of drinking water, and behavioral issues. Differences in these, difficult to document, may have offset the stark differences in public expenditures in health and education. Nonetheless, the deliberate cuts in social spending resorted to in Kazakhstan in response to the fiscal pressures is of serious concern, as it risks under-investment in the human capital stock of the country and thereby undermines the future growth potential. There are other indicators of human development are not so evenly shared by the two countries. Some of these indicators reflect a higher level of social breakdown in Kazakhstan. For instance, male suicide rate in Kazakhstan at 38 per 100,000 males is more than four times Uzbekistan's 9 per 100,000 and almost double the European Union's average of 20 per 100,000 males.9 Moreover, evidence suggests stronger gender bias in this, as the male suicide rate in Kazakhstan is more than 4 times that for females compared with a ratio of 3 in Uzbekistan. In terms of their social safety nets, the two countries have adopted drastically different approaches. In Uzbekistan, policymakers have revitalized and strengthened the traditional mahalla system (which provides decentralized benefit-targeting using local communities) as the primary vehicle for providing social assistance to the most 9 tUNDP (1 999). Page 1 I vulnerable groups in the society. Reforms of its pensions system are only just beginning. By contrast, Kazakhstan is very advanced with pension reforms and is at the frontier of countries, with private pension funds accounting for 45 percent of all contributions. However, social assistance payments from the Government are generally low, poorly targeted, and often the residual expenditures in stretched local government budgets. III. TWO PATHS, TWO DESTINATIONS? As the discussion in the previous section shows, Uzbekistan and Kazakhstan are now at very different points of evolution in their economies, with the relatively poor policy performer, by some measures, having better economic outcomes at this stage of the game. How can this disjunction between economic policy and performance be explained? Has one path to transition been clearly dominant over the other? To examine these issues, we offer six hypotheses: 1. The initial conditions between the two countries were different. In other words, comparing the efficacy of the reform experience of the two countries as a natural experiment-just because they were both once part of the Soviet Union-is not justified. 2. A special set of initial conditions affected the choice of policies and their relative performance. In particular, the presence of mineral resources (especially oil) in Kazakhstan affected both the pace and outcomes of reforms. 3. Gradualism with regard to liberalization was the appropriate response in CIS countries. This hypothesis would point to the relative performance of the two countries as evidence that the recipe for reform espoused by the Washington IFIs was too simplistic-or just plain wrong. Additionally, in Kazakhstan, the "shock therapy" reforms in next door Russia and close economic and social ties with her may have influenced, even constrained, policy choices. 4. The relative greater growth in Uzbekistan is an artifact of the seriously distorted prices there. This hypothesis would argue that economic actors in Kazakhstan face a much more realistic set of relative prices than do their Uzbek counterparts. With the same set of relative prices, the Uzbek economic performance may be as poor as, or worse than, that in Kazakhstan. Page 12 5. There is not that much 'effective ' diference between the two countries ' reforms. This argument is focused on Kazakhstan, and would argue that despite the apparently greater progress in reforms there, the overall set of reforms there have been incomplete and thus unable to engineer an economic rebound.'0 A more judgmental version of the hypothesis would contend that the resultant combination of progress in some areas, and not in others, has created a dysfunction that has lowered growth more. 6. The data is wrong. While this hypothesis seems the most dismissive, it has to be weighed against the reality of extremely poor data collection by many of the post- Soviet states, especially when it came to gauging market activity. In the rest of this paper, we examine these hypotheses in turn, to see if we can arrive at an understanding of the complex set of circumstances and interactions that produced Kazakhstan's and Uzbekistan's development outcomes. Hypothesis 1: Is History Bunk? Are the initial conditions faced by the two nations what helped to determine their subsequent development paths? Looking at the initial conditions both countries faced, it would be easy to think that the two countries would have a common development experience. However, on balance, it appears that some favorable initial conditions in Uzbekistan may have contributed to its better economic performance. Key similarities between the two countries were in terms of their inheritance of a common legacy of colonialism, their economic structure, common macroeconomic imbalances, and some similar social indicators (see Table 4). Politically, both Uzbekistan and Kazakhstan gained independence in 1991 after a common legacy of about 55 years of Tsarist domination and 71 years of Soviet rule, and had their new post-independence administration led by the same leaders as during the latter part of the Soviet regime- Presidents Karimov of Uzbekistan and Nazarbayev of Kazakhstan. They continue to share fairly autocratic states, though Uzbekistan's early years were marked by more 10 This is the "pitfalls of partial reforms" argument of Schleifer et al (1992). Page 13 authoritarian policies than Kazakhstan's." Both inherited similar economic structures with the share of output of agriculture, industry, and services quite evenly distributed. Both started with the same levels of repressed inflation and large black market exchange rate premia (common to all FSU countries). Both were (and are) landlocked countries though Uzbekistan is the only 'double' landlocked country in the world. Their social indicators-in terms of life expectancy (though not infant mortality), literacy rates, and school enrollment rates-were almost identical. Yet, in many other ways, they were very different. The key differences were in terms of their demographic profile, their economic and social links with Russia, the structure of agriculture and industry, and their level of income and natural resource endowments. Uzbekistan was, of course, more populous. It also had a much younger profile of the population, with 60 percent of the population under the age of 16 compared with 40 percent for Kazakhstan. But Uzbekistan also had a more ethnically homogeneous population with 70 percent of the population of Uzbek ethnicity. In contrast, Kazakhstan had a much greater diversity of ethnicity with about 40 percent of the population Kazakh.12 The large Russian population and the stronger economic ties to Russia in terms of trade and financial flows also meant that events in Russia had a larger impact on the economic evolution in Kazakhstan. The historical differences are also strong- Uzbekistan has a thousand year history of settled civilization, culture and learning, while the Kazakh people were mostly nomadic till the early 20th century. Even though agriculture and industry contributed almost equally to output in both countries, the structure of production was quite different and may have had implications for growth and the future course of economic reforms. Uzbekistan was primarily a cotton producing economy with cotton contributing to 60 percent of the agricultural output and 30 percent of export earnings. The production of cotton also meant forward linkages in the economy in terms of ginning, and textiles production. This, in combination with Freedom House's Freedom in the World 1992-93 rated Uzbekistan's political freedoms and civil rights as 6 on a 7-point scale, similar to Cambodia and Bosnia-Herzegovena. Kazakhstan's were 5 for both-the latter on par with Egypt, Kuwait, and Ghana. 12 Various empirical analyses from around the world have documented the adverse effect of ethnic heterogeneity, in the face of weak institutions, on income, growth, and economic policies. See for Page 14 TABLE 4: UZBEKISTAN AND KAZAKHSTAN: INITIAL CONDITIONS COMPARED (1990) UZB KAZ SIMILAR Political Structure Number of Years of Soviet Central Planning 71 71 Change in Leadership upon Independence No No Structure of Economy Share of Agriculture in Output (%) 31 29 Share of Industry in Output (%) 33 34 Share of Services in Output (%) 36 37 Macroeconomic indicators Repressed Inflation 25.7 25.7 Black Market Exchange Rate (% diff. over official) 1,828 1,828 External Debt (% of GDP, 1991) 0 0 Social Indicators Life Expectancy at birth (1981-90) a/ 68.1 68.2 Secondary School Enrolment Rate (%) 98 99 DIFFERENT Demographic Structure Population (million) 23 15 Urban population (% of total) a/ 40.6 57.6 Share of Population under 16 60 40 Population Growth rate 2.5 1.2 Percentage of dominant ethnic group (1993) 70 43 Structure of Economy Share of Extractive Industry in Industrial Output 12 65 Share of manufacturing and Food in Industrial Output 70 33 Natural Resource Endowments and Production Moderate High Proved recoverable oil reserves (mil. Met tons) c/ 41 449 World Rank in Production of lead, 1992 c/ n/a 5 th World Rank in Production of cadmium, 1992 c/ n/a 7th World Rank in Production of zinc, 1992 c/ n/a 8th World Rank in Production of copper, 1992 c/ n/a gth Macroeconomic Indicators GNP per capita (PPP, US$1989) 2,740 5,130 GDP (billion US$) 23.7 40.3 GDP Growth Rate, 1981-89 a/ 3.4 2.0 Exports (% of GDP) 29 74 Social Indicators Infant Mortality per 1,000 live births (1981-90) a/ 42.7 29.2 Poverty (%, headcount index) b/ 24 5 Poverty (% below $2 a day) (Uzb.: 1993, Kaz. 1996) 26.5 15.3 Source: De Melo et al (1997); World Bank; a/ from WDR 1996; b/ Milanovic 1998, Table 5.1, pp.68-9; for instance, Easterly and Levine (1997) for this as an explanation of Africa's poor growth performance. c/ calculated from World Resources Institute (1994) Page 15 agriculture-based food industry meant that more than 50 percent of the industrial output was from manufacturing. By contrast, the industrial sector in Kazakhstan was skewed towards the extractive industries-fuel, energy, metallurgy, and chemicals-which together accounted for about 65 percent of the industrial output; manufacturing and food comprised only one-third of industrial output. While the structure of the economy may have been associated with Uzbekistan's relatively lower income levels and higher poverty and mortality rates at independence, there is also evidence to suggest that the exceptional mildness of Uzbekistan's transitional recession and subsequent growth can, in part, be accounted for by its low degree of initial industrialization and its cotton production.'3 One way to interpret this result is to say that Uzbekistan appears to have been effective at preventive the collapse of its relatively small industrial sector by combining rigid state control with subsidies that were in large part financed by cotton exports. Whether this is sustainable or not is, of course, a different issue. Hypothesis 2: Midas' Curse? Another critical difference in initial conditions that may have affected the choice of reform strategy-and thus the particular outcomes in Kazakhstan versus Uzbekistan- is the presence of enormous natural resources in Kazakhstan. It is claimed that in Kazakhstan, there are extractable amounts of every metal in the Mendeleev periodic table, not to mention some of the richest oilfields in the region. As Table 4 demonstrates, Kazakhstan, even at independence, was one of the largest exporters of a range of metals, and was just beginning to realize the potential of unexploited oil reserves along its Caspian shore. The need to attract foreign capital to exploit the mineral resources may have led to a faster pace of reforms and a greater degree of outward-orientation in Kazakhstan. But for the same reason, outcomes may have suffered as it is by now well-established that the exploitation of mineral wealth can amplify the problems of transition (and other) economies.14 In Uzbekistan, on the other hand, an alternative set of initial conditions, including heavy dependence on energy and grain imports and a lower level of over- 3 Zettlemeyer (1999); a third factor is the achievement of near self-sufficiency in energy. 14 See, for instance, Auty, R. (1998). Page 16 industrialization, led the Government to adopt autarkic policies in support of energy and grain self-sufficiency and import-substitution led industrialization. The implication of this difference in natural resource endowment was, thus, in the choice of industry. Kazakhstan's economy was concentrated in sectors linked to mineral and petroleum products-including steel, zinc processing, copper smelting, aluminurn processing, petrochemicals and oil refining. Light industry, though present, was secondary in importance. Uzbekistan had a more diversified economy. Its own resource endowments of cotton and gold may have also played a key role in maintaining foreign exchange inflows, and growth, but they served as a launching pad for a wider array of manufacturing activities than has been true in Kazakhstan. In the post-independence era, therefore, Kazakhstan was much more exposed to commodity price shocks than was Uzbekistan-since its major raw material exports and processed exports were all co- cyclical. The 1998-1999 drop in worldwide prices for commodities, therefore, hit Kazakhstan very hard (and may have depressed its growth rates below its trend). Empirical work has confirmed that the growth effects of manufacturing are much smaller when manufacturing industry is resource-based rather than more diversified. 5 Most important for Kazakhstan is the presence of oil, which has been extracted there for the past 100 years. The collapse of the Soviet Union roughly coincided with an increased interest in the potentially vast oilfields of the Caspian, and the eventual exploration and development of the Tengiz oilfield, one of the richest in the world. But the geographic isolation of Kazakhstan has made it difficult to ship large volumes of oil out, until new pipelines to the Black Sea are built. The perceived need to quickly attract foreign investors-with foreign capital and technology for the economic development of this oil-meant that the policymakers in Kazakhstan had to adopt market-oriented policies that provided the best bet for a more rapid infusion of foreign investments into the extractive sector. The haste to attract investors led Kazakhstan to offer substantial (perhaps excessive) concessions to foreign investors, including, for example, a 25 percent depreciation allowance on all investment as well as generous tax breaks on corporate taxes. This has led to a large under-taxation of the minerals sector in Kazakhstan, a consequent fiscal crisis in the late 1990s, and abysmally low public investment (below 2 percent of GDP in 1998-99). This has Page 17 compromised the medium-term growth prospects of Kazakhstan's non-oil sector. On the other hand, Uzbekistan-as part of its import-substitution industrialization strategy-has been more selective in opening sectors of its economy to foreign investors. Thus, widespread and rapid liberalization of the economy has not been a necessary driving force in its development strategy. Hypothesis 3: The Tortoise And The Hare The fundamental differences in the reform program adopted by the two countries lie both in the approach to the pace of market-oriented policy reforms and the sequencing of institutional and market reforms. Uzbekistan clearly adopted a more gradualist approach, with the idea that the unfettered market may not be compatible with the Government's aims of socioeconomic development. Thus, this line of thinking argued that the failures of transition lie in a misunderstanding of the foundations of a market economy-particularly in its informational requirements-as well as in a misunderstanding of the basics of an institutional reform process.16 Alternative arguments stress that it takes time to build a new world, adjustment costs can be high and politically and socially destabilizing, and that the pace of new job creation is likely to be slow.'7 In Uzbekistan, gradualism received official support primarily on grounds of national and historical factors including ethnic diversity, the younger age profile of the population, low living standards relative to the rest of the former Soviet Union, the need to maintain social cohesion and stability, and the deeply-entrenched public psychology of dependency on the state.'8 This implied that the state be given a principal role during transition not only as the developer and implementer of reforms-and also as the collective entrepreneur, production regulator, and investor in priority sectors. Many of these justifications, therefore, are linked to the earlier discussion of initial conditions.. As a result, many of even the "first generation" policy reforms, such as price liberalization, trade and foreign exchange liberalization, and interest rate liberalization were slower to develop in Uzbekistan (as discussed earlier). In Kazakhstan, on the other 15 See, for instance, Wood and Berge (1997). 16 See Stiglitz, J. (1999). 17 A rich literature enlightens this debate. See, for instance, Nuti and Portes (1993), Dewatripont and Roland (1992) Aghion and Blanchard (1994). 18 Karimov, I. (1993). Page 18 hand, a more aggressive liberalization program was initiated and led by a group of young, Western-influenced reformers, who adopted some reforms which were radically dependent on the free market. One such step was the adoption of a drastic pension reform program in 1998, which phased out the entire pay-as-you-go system and replaced it with a funded system based on individual contributions. At the same time, the Uzbeks viewed the creation of market institutions as a precursor to policy reforms, while the Kazakhs allowed the institutions to develop endogenously with the market. This difference in choice of sequencing was deliberate and reflected the Uzbek view both of gradual step-by-step reforms and the role of the state as the developer and implementer of reforms. Thus, in terms of the size of the state (proxied by total expenditures and total revenues) Kazakhstan is just over half the size of Uzbekistan (Figure 7). The desire to find market solutions, driven in large part by the ideology, led to a sizeable downsizing of the state in Kazakhstan. Interestingly, however, both Uzbekistan and Kazakhstan appeared to have progressed to similar levels on Figure 7: Size of the State Kaz Total Expenditures - Kaz Total Revenues Uzb Total Expenditures "*Uzb Total Revenues 40 35 A 30 7Z 20 15 1992 1993 1994 1995 1996 1997 1998 Source: World Bank institutional reforms pertaining to enterprise restructuring, corporate governance, and competition policy, all central to the transition process.'9 19 Some argue that the faster pace of policy reforms in Kazakhstan may have been conditioned by geography, the closer economic links with Russia, and the liberal views on the role of the state in the post-communist era. Given the higher level of industrialization in Kazakhstan and the larger Russian population, economic and social ties were closer. Thus events in Russia, including the Page 19 Hypothesis 4: The Distorting Prism? One of the major areas where Uzbekistan's reforms have lagged behind is in price liberalization, especially in the foreign exchange market. This limited progress in price reforrns is a key reason why Uzbekistan's macroeconomic performance has actually been better in some areas. There are major questions, however, about the sustainability of this performance. After good progress in the early years in Uzbekistan, the process of foreign exchange liberalization was reversed in late 1996, when a system of multiple exchange rates was formally introduced. Growing restrictions on access to foreign exchange and smaller sales of foreign exchange by the central bank led to a growing parallel market premium over the official rate, which by end-1999 had reached a maximum of over 500 percent.20 By contrast, the exchange rate regime in Kazakhstan-lightly managed until August 1998, heavily managed until April 1999, and moderately managed since-has still been generally responsive to market pressures, with an unified exchange rate since 1994 and no black market premium. But the multiple exchange rate system in Uzbekistan introduced an insidious system of taxation and subsidization which severely distorts the prices faced by producers and consumers; as a corollary, economic agents faced smaller changes in relative prices than their compatriots in Kazakhstan. Combined with soft budget constraints, the consequent fall in industrial output-which usually results from the restructuring of potentially viable and the liquidation of non-viable enterprises-has been much smaller in Uzbekistan. By contrast, Kazakhstan's producers have had to be more exposed to the competitive pressures engendered by the relatively more level playing field. This has been particularly true for exporters in light industries, who have been heavily penalized by the price advantage offered to their Russian competitors by the steady appreciation of the tenge against the ruble until April 1999. With an industrial legacy more oriented towards the Russian market than that in Uzbekistan, the Kazakhstan industrial sector has evolution of "shock therapy" economic reforms there, could have been critical in shaping economic policies in Kazakhstan. 20 The foreign exchange devaluation of 50 percent in May 2000 and some other liberalization measures have reduced this premium to 200 percent by July 2000. Page 20 been less able to muster internal financial and managerial resources to retool and refocus their economic activities in the face of the initial shock of the breakdown of the Soviet trading system. Hypothesis 5: Missing The Pieces? The clear impression of most observers, who have noted that Uzbekistan has lagged behind Kazakhstan in liberalizing prices, is that Uzbekistan's reforms been less comprehensive than Kazakhstan's. But is liberalization a sufficient predictor of reforms? The question is, therefore, whether the formal liberalization of prices has led to greatly improved channels of transmission of accurate price signals. There is some evidence that even as both countries may have generally comparable levels of institutional reforms, their implementation may have been less effective in Kazakhstan. Figure 8: Obstacles to Business | - - -Uzbekistan i Kazakhstan Taxes and Regulations Infrastructure Inflation Functioning of the Judiciary Financing Organized Crime/Mafia Policy Instability Street Crime/Theft/Disorder Exchange Rate Corruption Source: World Bank, Business Environment and Enterprise Performance Survey Thus, if we accept this reasoning, even as the policy reforms may have created some conditions for an environment conducive to growth in Kazakhstan, the more rapid liberalization has depressed economic growth relative to Uzbekistan (as in the previous hypothesis) while the lack of better institutional incentives has constrained production and income generating activities. Page 21 This is confirmed in a recent business environment survey which reveals that managers of firms surveyed perceive the impediments to business to be much more acute in Kazakhstan (see Figure g)721 This may be a central factor behind the relative lack of economic rebound in Kazakhstan.22 Particularly revealing are the perceptions of businessmen in areas such as inflation, exchange rates and policy stability-the less controlled and thus inherently less predictable market-friendly environment in Kazakhstan is seen as posing relatively more obstacles to them than the controlled policy environment in Uzbekistan. As important, of course, is that the more laissez faire environment is also accompanied by weaker market institutions such as a strong judiciary and low levels of crime and petty harassment. Nowhere is the interplay between institutional and policy changes as revealing as in the area of tax reform. A relatively low level of trade taxes but a very corrupt customs service in Kazakhstan has meant that there is an inherent bias by economic agents towards imports and away from domestic production. At the same time, a relatively narrow tax base coupled with a poor administrative and legal environment has meant that a small group of businesses (including start-ups) have been frequently harassed by inspectors. As a result, the business environment in Kazakhstan is one often fraught with unexpected transactions costs, especially for those who are not "insiders." Thus, partial reforms-and less than fully synchronized policy and institutional reforms-may have undermined potentially very successful economic outcomes in Kazakhstan. A related issue is the progress with respect to industrial enterprise restructuring. Here, despite clear differences in privatization paths followed by the two countries, effective outcomes are not that different. While both countries moved swiftly at the start of the transition to privatize their housing units and small enterprises engaged in retail trade, it has been in the privatization of other small, medium, and large enterprises that the two countries differed. The pace of privatization was much faster in Kazakhstan, at least as suggested by the number of enterprises privatized. This reflected the differing philosophies to transition in the two countries as discussed above. Moreover, private 21 It is possible that the general responses from Uzbek firms were muted in their evaluation of the severity of constraints due to the larger proportion of state-owned firms and the political situation in Uzbekistan. 22 Moreover, given the lack of implicit subsidies from price distortions and relatively few explicit subsidies from the Government, the business environment may actually be worse in Kazakhstan. Page 22 participation in strategic sectors of the economy-such as power, telecomrnmunications, banking-is much more advanced in Kazakhstan. In addition, the initial approach to privatization of medium and large enterprises in Kazakhstan followed the Russian model of voucher privatization (except for oil fields), in an effort to quickly transfer ownership to as diversified a population group as possible. This has diffused ownership, and often allowed the old, less innovative managers to effectively retain control without accountability to the diverse shareholders. In Uzbekistan, the approach was guided by the need to transfer ownership to real owners capable of using the property and ensuring its effective utilization. Towards that end, a scheme of privatization investment funds was developed-which, while providing for widespread private ownership, also attempted to create independent financial investment entities that would improve corporate governance and promote capital market development. However, the Uzbek experience has so far been disappointing and may be explainable by the lack of an enabling macroeconomic environment and the frequent changes to the legal framework (including the partial re-nationalization in 1997).2 IV. CONCLUSION (HYPOTHESIS 6): DAMN STATISTICS? The discussion of the five alternative hypotheses above range from some that have policy implications (hypotheses 3, 4 and 5) and some that do not (1 and 2). Part of the agenda in future work is thus to test the roughly sketched discussions in this note in greater detail, and to work out the relevant policy package if, indeed, there is one. But the efforts to find the answers, at least in the short run, will be bedeviled by the fact that in both these countries, data is generally unreliable, and often unavailable. In fact, the entire debate about relative performances of the two countries may be clouded by the fact that the statistics are very unreliable, with the state statistical agencies unable to accurately gauge the extent of market activity. The methodology in vogue for estimating the national accounts suffers from various deficiencies. One particular weakness is in adequately accounting for the sizeable informal sector; as a result, relative growth rate stories may be meaningless. Estimates of the informal sector were 25 percent 23 See World Bank (1999) for a discussion of the quality of the privatization program in Uzbekistan. Page 23 in Kazakhstan in 1998, and 10 percent in Uzbekistan in 1997 but there are no estimates of its trend growth since independence.24 So, to turn full circle, is indeed the performance in Uzbekistan better than that in Kazakhstan? By many measures, yes-and by others, not. While Kazakhstan's policy performance is definitely superior, economic performance is not so. Various hypotheses discussed above may each have some merit in explaining this disjunction between policy and performance, but a key reason is the missing pieces in the reform-in particular the greater deficiencies in the competitive environment-which in combination with a diminishing (or weaker state) may have led to such mixed outcomes in Kazakhstan. Thus, while this paper has laid out some intersecting threads of reasons to explain the dichotomous paths taken by these two Central Asian countries in transition, a fuller understanding of their paths will have to rely on further, focused research on these hypotheses. This would then contribute to a better understanding of the different roads to growth that would lead to better outcomes-not just for the economies of Kazakhstan and Uzbekistan and their people, but for all developing countries. 24 EBRD (1997). The size of the informal sector is likely to have grown since then, if the rapid growth of the curb market exchange rate premium is any indicator. Page 24 Bibliography Aghion and Blanchard, 1994, "On the Speed of Transition in Central Europe" NBER Working Paper Series No. 4736. Auty, R., 1998, "Mineral Wealth and the Economic Transition: Kazakhstan", Resources Policy, Vol. 24, No. 4, pp. 241-249, Pergamon, U.K. De Melo, J., C. Denizer, A. Gelb and S. Tenev, 1997, "Circumstance and Choice: the Role of Initial Conditions and Policies in Transition Economies", World Bank Working Paper No. 1866. Dewatripont, M and G. Roland, 1992, "Economic Reform and Dynanic Political Constraints" Review of Economic Studies, October, 59 (4), pp.703-30. EBRD, 1997, Transition Report 1997, London. EBRD, 1999, Transition Report 1999, London. Easterly, W., and R. Levine, 1997, "Africa's Growth Tragedy: Policies and Ethnic Divisions", Quarterly Journal of Economics, CXII (4), 1203-1250. Fischer, S. R. Sahay, and C. Vegh, 1998, "From Transition to Market: Evidence and Growth Prospects" IMF Working Paper No. WP/98/52, IMF. IMF, 1999, Review of Social Issues and Policies in IMF-Supported Programs. Karimov, I., 1993, Building the Future: Uzbekistan-Its Own Model for Transition to a Market Economy, Uzbekistan Publishers, Tashkent. Milanovic, B., 1998, Income, Inequality and Poverty during the Transition from Planned to Market Economy, World Bank, Washington, D.C. Nuti, D. and R. Portes, 1993, "Central Europe: The Way Forward" in R. Portes ed. Economic Transformation in Central Europe, London: CEPR. Rodrik, D. and A. Velasco, 1999, "Short-Term Capital Flows", Paper presented at the Annual Bank Conference on Development Economics, April, Washington, D.C. Selowsky, M. and R. Martin, 1996, "Policy Performance and Output Growth in Transition Economies", Papers and Proceedings of the American Economic Association, May. Schleifer, A., K. Murphy, and R. Vishny, 1992, "The Transition to a Market Economy: Pitfalls of Partial Reform", Quarterly Journal of Economics, August. Stiglitz, J., 1999, "Whither Reform? Ten Years of the Transition", Paper presented at the Annual Bank Conference on Development Economics, Washington D.C. Page 25 Tanzi, V. and G. Tsibouris, 1999, "Fiscal Reform over Ten Years of Transition", Paper presented at the Fifth Dubrovnik Conference on Transition Economies, July 23-25. UNDP, 1999, Human Development Report, New York. World Bank, 1999, Uzbekistan: Social and Structural Policy Review, Grey Cover Report No. 19626. World Resources Institute, 1994, World Resources 1994-95, Oxford University Press, London and New York. Wood, A. and K. Berge. 1997, "Exporting Manufactures: Human Resources, Natural Resources, and Trade Policy," Journal of Development Studies Vol. 34, 35-59. Zettlemeyer, J., 1999, "The Uzbek Growth Puzzle", IMF Staff Papers Vol. 46, No. 2, June. Policy Research Working Paper Series Contact Title Author Date for paper WPS2455 The Effects on Growth of Commodity Jan Dehn September 2000 P. Varangis Price Uncertainty and Shocks 33852 WPS2456 Geography and Development J. Vernon Henderson September 2000 R. Yazigi Zmarak Shalizi 37176 Anthony J. Venables WPS2457 Urban and Regional Dynamics in Uwe Deichmann September 2000 R. Yazigi Poland Vernon Henderson 37176 WPS2458 Choosing Rural Road Investments Dominique van de WalIe October 2000 H. Sladovich to Help Reduce Poverty 37698 WPS2459 Short-Lived Shocks with Long-Lived Michael Lokshin October 2000 P. Sader Impacts? Household Income Martin Ravallion 33902 Dynamics in a Transition Economy WPS2460 Labor Redundancy, Retraining, and Antonio Estache October 2000 G. Chenet-Smith Outplacement during Privatization: Jose Antonio Schmitt 36370 The Experience of Brazil's Federal de Azevedo Railway Evelyn Sydenstricker WPS2461 Vertical Price Control and Parallel Keith E. Maskus October 2000 L. Tabada Imports: Theory and Evidence Yongmin Chen 36896 WPS2462 Foreign Entry in Turkey's Banking Cevdet Denizer October 2000 I. Partola Sector, 1980-97 35759 WPS2463 Personal Pension Plans and Stock Max Alier October 2000 A. Yaptenco Market Volatility Dimitri Vittas 31823 WPS2464 The Decumulation (Payout) Phase of Estelle James October 2000 A. Yaptenco Defined Contribution Pillars: Policy Dimitri Vittas 31823 Issues in the Provision of Annuities and Other Benefits WPS2465 Reforming Tax Expenditure Programs Carlos B. Cavalcanti October 2000 A. Correa in Poland Zhicheng Li 38949 WPS2466 El Nino or El Peso? Crisis, Poverty, Gaurav Datt October 2000 T. Mailei And Income Distribution in the Hans Hoogeveen 87347 Philippines WPS2467 Does Financial Liberalization Relax Luc Laeven October 2000 R. Vo Financing Constraints on Firms? 33722 Policy Research Working Paper Series Contact Title Author Date for paper WPS2468 Pricing, Subsidies, and the Poor: Ian Walker November 2000 S. Delgado Demand for Improved Water Services Fidel Ordonez 37840 in Central America Pedro Serrano Jonathan Halpern WPS2469 Risk Shifting and Long-Term Mansoor Dailami November 2000 W. Nedrow Contracts: Evidence from the Robert Hauswald 31585 Ras Gas Project WPS2470 Are Larger Countries Really More Stephen Knack November 2000 P. Sintim-Aboagye Corrupt? Omar Azfar 38526 WPS2471 Validating Operational Food Luc J. Christiaensen November 2000 L. Christiaensen Insecurity Indicators against a Richard N. Boisvert 81463 Dynamic Benchmark: Evidence John Hoddinott from Mali