LINKING MACROECONOMIC POLICY TO POVERTY REDUCTION IN UZBEKISTAN

Transcription

1 LINKING MACROECONOMIC POLICY TO POVERTY REDUCTION IN UZBEKISTAN TASHKENT 2005

2 TEAM OF AUTHORS TEAM LEADERS: GALINA SAIDOVA, First Deputy Minister of Economy of the Republic of Uzbekistan GIOVANNI ANDREA CORNIA, University of Florence, Italy COORDINATORS: TERRY MC KINLEY, Policy Adviser of Bureau for Development Policy, UNDP UKTAM ABDURAKHMANOV, Research Coordinator, Center for Economic Research AUTHORS: MAX SPOOR, Institute for Social Studies, The Netherlands DAVID KOTZ, Amherst University, USA ZAVIKHULLA NASRITDINKHODJAEV, Chief of sub-division, Information-analytical Department of Cabinet of Ministers of the Republic of Uzbekistan AKTAM KHAITOV, Deputy Chairman, Committee for economic insolvency YULDUZ ABDUGANIEVA, Chief specialist, Information-analytical Department of Cabinet of Ministers of the Republic of Uzbekistan ARTUR TUKHTAROV, Leading specialist, Information-analytical Department of Cabinet of Ministers of the Republic of Uzbekistan PARTICIPATED IN THE PREPARATION OF PUBLICATION: ALISHER YAKUBOV, Assistant Research Coordinator, Center for Economic Research MUZAFFAR KASIMOV, Research Associate, Center for Economic Research JAKHONGIR ALIMBABAEV, National Expert, Center for Economic Research

3 CONTENTS CHAPTER 1 The Record of Economic Transition in Uzbekistan 4 CHAPTER 2 Sources and Features of the Uzbek Growth Puzzle 19 CHAPTER 3 Trends in Poverty, Inequality and Human Deprivation 29 CHAPTER 4 An Overall Strategy for Pro-poor Growth 50 CHAPTER 5 Agriculture, Rural Development and Poverty 65 CHAPTER 6 Employment Promotion and Poverty Reduction 76 CHAPTER 7 Taxation, Public Expenditure and Poverty Reduction 87 CHAPTER 8 Institutional Development for Growth with Equity in Uzbekistan 99

4 CHAPTER 1 THE RECORD OF ECONOMIC TRANSITION IN UZBEKISTAN 1. INTRODUCTION When Uzbekistan emerged as an independent state from the demise of the Soviet Union in September 1991, the international financial institutions offered advice to the new government about how to conduct the transition to a market economy. The advice was similar to that offered to other transition countries, stressing moving quickly to liberalize prices and markets both domestically and internationally, privatise enterprises and free them from state control, and seek macroeconomic stabilization through tight fiscal and monetary policies. It was believed that such policies would set the transition countries, after a relatively brief period of difficulties, on a path toward an efficient, rapidly growing market economy with rising living standards for the population. Uzbekistan s reluctance to follow the conventional policy recommendations on transition became apparent very early. Ignoring the warnings from abroad, the government of Uzbekistan followed a policy of gradual transformation of the economy. While housing and small enterprises were privatised quickly and significant price liberalisation took place within a few years, the government maintained strong control over the allocation of resources in the economy and over its interface with the international economy. Privatisation of large enterprises has been slow and partial, with the government often retaining an active role in large enterprises even after official privatisation. Outside observers expected poor results from Uzbekistan s unorthodox policies. An example was the following warning offered in 1993: In the absence of comprehensive reform, the few scattered and partial reforms that the Government is implementing would have no major impact. This muddle through scenario would imply that none of the sectors is restructured significantly. This failure to adjust would lead to a supply-led contraction of the level of economic activity, reduction in savings and investment, and unnecessary hardship on the population in the medium term represented by a decline in consumption per capita of no less than 30 percent by 1997 (World Bank, 1993: xx). Contrary to expectations, Uzbekistan s growth performance during , despite its policy of gradual and partial transformation, has been the best of the former Soviet republics. Uzbekistan s initial contraction was the mildest of the former Soviet republics, it ended relatively early, and since 1996 GDP has grown at an average rate of 4 percent per year. Rather than falling by 30% by 1997, consumption per capita had slightly surpassed (by 0.5%) the 1993 level in 1997 (Ministry of economy). Uzbekistan s economic performance since 1992 has earned the nickname the Uzbek growth puzzle, which even appeared in the title of an IMF Staff Paper in 1999 (Zettelmeyer, 1999). By 2001 Uzbekistan s GDP was 3% above the 1989 level, making that country the first former Soviet republic to surpass its 1989 GDP level (see table 1.1). By contrast, in 2001 the fourteen other former Soviet republics had GPD s ranging from 35% (Moldova) to 88% (Belarus) of their respective 1989 levels. Table 1.1. Real GDP in 2001 in the Fifteen States Formerly Part of the Soviet Union (percentage of 1989 level) Armenia 69 Azerbaijan 57 Belarus 88 Estonia 87 Georgia 36 Kazakhstan 78 Kyrgyzstan 69 Latvia 69 Lithuania 69 Moldova 35 Russia 62 Tajikistan 52 Turkmenistan 84 Ukraine 44 Uzbekistan 103 Commonwealth of Independent States Average 72 Source: EBRD (2002: 17). A country of some 25 million people in Central Asia, Uzbekistan inherited from the Soviet period an economy centred around cotton. Two great rivers, the Amu Darya and the Syr Darya, had provided water for agriculture in that region for thousands of years, and in the Soviet period extensive irrigation works were constructed for cotton growing. Besides cotton, Uzbekistan also grew a variety of fruits and vegetables and, in addition to agriculture, had a significant industrial base. In 1991 agriculture represented 37.2 percent of GDP by value added while industry contributed 26.2 percent (Ministry of economy). Industry included agro-food (13 percent), light industry (39 percent), and heavy industry (41 percent) (World Bank, 1993: xi). There were a number of strengths in Uzbekistan s economy at the start of independence. It had a readily marketable export in cotton. It had the 14th largest natural gas reserves in the world, substantial gold deposits, and also silver, copper, lead, zinc, tungsten, coal, and uranium (EIU 2002: 16). It had some oil reserves, largely undeveloped. It had a well-educated and healthy population. 4

5 Linking macroeconomic policy to poverty reduction in Uzbekistan However, the obstacles it faced seemed enormous. Uzbekistan had received a net subsidy from the Soviet Union s central budget. In the last few years of the Soviet Union, the subsidy rose rapidly from 7 9 percent of Uzbekistan s GDP during to 21% in 1991 (IMF, 1992: 4). This subsidy was suddenly cut off with independence. Uzbekistan s industry had been part of the industrial complex of the Soviet Union, and the breakup severed its traditional supplier and customer relations. Its economy had depended on cheap Soviet oil, and, despite its large agricultural sector, it depended on food from elsewhere. Uzbekistan is a double-landlocked country, whose goods must cross two countries to reach a port, and it faced uneasy relations with several of its neighbours. Table 1.2 shows four key indicators of macroeconomic performance since Like all of the former Soviet republics, Uzbekistan underwent a large economic contraction after the demise of the Soviet Union. GDP fell by a cumulative 18.4 percent during Then it rose by 26.2 percent from 1996 through The very rapid inflation of the early 1990s peaked in 1994, then fell in steps before settling at about 25% per year after The government fiscal deficit shot up to 18.3 percent of GDP in 1992, then declined to single digits and, after 1996, fell to 3 percent of GDP or less. The current account balance followed a more complex path, first falling rapidly from a very large deficit in 1992 to a small surplus in 1994, then heading back into the red through 1997, after which it has been close to balance. This report is not concerned only with impersonal economic indicators but with the impact of economic developments on human welfare. In 1992 the real income of Uzbekistan s population fell to 68.5 percent of its 1991 level. By 2000 it had climbed to 19.1 percent above the 1991 level (see table 1.23 below). Since the population grew by 18.2 percent during that period, real income per capita just surpassed its pre-independence level by However, income became more unequally distributed over that period. According to official statistics, the Gini coefficient for income rose sharply, from 0.26 in 1991 to 0.35 in 2000, an increase of 50 percent, and the decile coefficient rose by 45 percent. Because of rapidly rising inequality, it is likely that only a minority of the population has surpassed its preindependence level of real income. (Chapter 3 examines inequality and poverty over this period in detail.) While the official unemployment rate, which counts only a small fraction of the unemployed, rose little over the period, estimates of actual unemployment in recent years have ranged from 6.3 percent of the labour force to 26.5 percent (a UNDP estimate reported in EIU, 2002: 13). Thus, while Uzbekistan s economic growth performance has been relatively good compared to other transition countries, it appears likely that a majority of the population still awaits an improvement in material conditions beyond the level they had experienced prior to independence. Since 1992 Uzbekistan s economy contracted for four years and then expanded for six. In section 2 we will examine key policy decisions, external events affecting the economy, and a few major performance indicators for the contraction phase of In section 3 we will do the same for the expansion phase of Section 4 offers a more detailed account of Uzbekistan s economic performance over the whole ten-year period. Section 5 sums up the strengths and weaknesses of the country s economic performance. The aim of this chapter is to present a clear picture of Uzbekistan s policies and economic performance. We leave to chapter 2 an analysis of the connections between policies and performance, along with the lessons for future policy aimed at sustainable economic growth and poverty reduction. Table 1.2. Indicators of Uzbekistan s Macroeconomic Performance, GDP growth rate Consumer price inflation rate Fiscal balance Current account balance Percentage change in real GDP from the previous year. 2 Percentage change in the consumer price index from the previous year. 3 Surplus (+) or deficit ( ) of the general government budget as a percentage of GDP. 4 Surplus (+) or deficit (-) on current account as a percentage of GDP. 5 Estimate. Source: GDP and CPI data are from Ministry of economy; fiscal balance and current account data are from EBRD (2002: 19 20). 5

6 Chapter 1. The record of economic transition in Uzbekistan Table 1.3. Money and Interest Rates Stage 1 Stage Broad money growth rate GDP deflator growth rate Refinancing rate Bank loan rate Producer price inflation rate Percentage change in M3 from the previous year. 2 Percentage change in the GDP price deflator from the previous year. 3 Percentage per year. 4 Percentage change in the producer price index from the previous year. Source: Growth rates in the GDP deflator and producer price index were calculated from data from Ministry of economy; the M3 growth rate, the refinancing rate, and the bank loan rate are from EBRD (2002: 91). Table 1.4. GDP Growth Rates for Five Central Asian Republics and the CIS Average, Stage 1 Stage Kazakhstan Kyrgyzstan Tajikistan Turkmenistan Uzbekistan CIS Weighted averages. Source: EBRD (2002: 17). Data for Uzbekistan differ in some years from these in table 1.2, which were from a different source. Table 1.5. CPI Inflation for Five Central Asian Republics and the Commonwealth of Independent States, Stage 1 Stage Kazakhstan Kyrgyzstan Tajikistan Turkmenistan Uzbekistan CIS Median value of CPI inflation. Source: EBRD (2002: 18). The data for Uzbekistan differ in some years from those in table 1.2, which were from different source. Table 1.6. General Government Budget Balance for Five Central Asian Republics and the Commonwealth of Independent States, Stage 1 Stage Kazakhstan Kyrgyzstan Tajikistan Turkmenistan Uzbekistan CIS Unweighted averages. Source: EBRD (2002: 19). 6

7 Linking macroeconomic policy to poverty reduction in Uzbekistan 2. STAGE 1: CONTRACTION, As the transition began in 1992, Uzbekistan chose a different course from the Washington Consensus approach. Instead of immediately lifting price controls, both price controls and rationing were kept for many goods at first and only gradually abandoned. Rather than freeing enterprises from state control, a system of state orders for industry and agriculture was kept in place. The state actively financed investment and provided subsidies to enterprises. Unlike other transition countries, most of which were slashing state spending, Uzbekistan maintained state spending at a relatively high level (it fell no lower than 35.3 percent of GDP during stage 1 see section 4.3 below). As was noted above, privatisation was very gradual and partial. In this stage, only small enterprises, most of them in trade and consumer services, were privatised, along with housing. Foreign trade was kept under tight control. Economic policy was initially focussed on limiting as much as possible the decline in the nation s consumption and its production capacity, as well as maintaining order. The policy initially sought to maximize imports and limit exports, to the extent possible given financing constraints. 1 However, believing that continuing reliance on imported oil and food would threaten the country s independence, the government also sought to develop its oil resources and grain production as rapidly as possible. For much of this stage, the Uzbekistan authorities were unable to pursue an independent monetary policy, because the country remained in the ruble zone at first. Only in 1993 did the country leave the ruble zone, introducing a tempowrary currency, the soum-coupon, in November of that year. On 1 July 1994 a permanent currency, the soum, was issued. Table 1.3 gives monetary and interest rate data for , which indicate a mixed picture of the government s monetary stance in the last two years of stage 1 (data were not available for ). While the growth rate of broad money in 1994 and 1995 was high, the GDP price deflator grew significantly more rapidly than the money supply in those two years. This suggests tight monetary policy, as the real value of the money supply was reduced and the ratio of money to GDP fell. Inflation was able to continue due to a rapidly rising velocity of money, suggesting a powerful inertial inflationary process. The rate of money change was in itself exerting downward pressure on inflation, which did decline sharply in 1995, the last year of the stage of contracting GDP. On the other hand, monetary policy was not tight enough to produce positive real interest rates in stage 1. While high in nominal terms, real interest rates were negative during 1994 and 1995 (the two interest rates shown in table 1.3 were below the rate of producer price inflation). Table 1.2 above showed Uzbekistan s fiscal deficit as a percentage of GDP. While the government ran a budget deficit every year during , the size of the deficit shrank rapidly and steadily as a share of GDP. Hence, while it could not be said that fiscal policy was tight in the conventional sense, the policy moved steadily in a tightening direction from 1992 to Table 1.4 shows the GDP growth rates for the five Central Asian states and for the average of the CIS. 3 In 1992 Uzbekistan s GDP performance was not outstanding. Its GDP decline of 11.1 percent was in the middle of the group. At that time the World Bank s warning, cited above, seemed quite plausible. However, in Uzbekistan s GDP decline slowed sharply and each year was substantially below that of its Central Asian neighbours. Uzbekistan s decline was also well below the CIS average. As Table 1.5 shows, Uzbekistan s inflation performance was relatively good in 1992, second only to Turkmenistan (which also had not liberalized quickly) within the CIS as a whole. As Uzbekistan gradually freed prices, inflation picked up and, by 1994 it was far in excess of inflation in Kyrgystan or Tajiskistan and barely below the CIS median inflation rate. In 1995 inflation slowed, but now it was above the CIS median rate. Uzbekistan s fiscal deficit performance placed it about in the middle its Central Asian neighbours during , and somewhat better than the CIS average in , as table 1.6 shows. Despite better growth performance, the government s determination to maintain spending, in the face of declining output and tax revenues, brought only mediocre performance on this indicator in stage 1. 1 Domestic consumption of oil and oil products declined by 15 percent in , which contributed to the achievement of oil selfsufficiency. However, even had domestic consumption not declined, domestic production would have provided 92.5 percent of domestic consumption by 1995, up from 48.1 percent of domestic consumption in Table1.7. Current Account Balance for Five central Asian Republics and the Commonwealth of Independent States, Stage 1 Stage Kazakhstan Kyrgyzstan Tajikistan Turkmenistan Uzbekistan CIS Unweighted averages. Source: EBRD (2002: 20). 7

8 Chapter 1. The record of economic transition in Uzbekistan The current account balance of Uzbekistan improved dramatically from 1992 to 1994, when it turned positive, and it was essentially in balance in 1995, as table 1.7 shows. In it was one of only three CIS countries with a current account in surplus or very close to balance, the other two being the well-endowed raw material exporters Turkmenistan and Russia. The program to achieve oil self-sufficiency reached its goal in three years, by developing additional oil resources which had lain undeveloped in the Soviet period. This was accomplished without any foreign direct investment. Table 1.8 shows Uzbekistan s energy balance during (no data were available for 1992). In the Soviet period Uzbekistan had imported about 60 percent of its petroleum (Griffin, 1996: 32). From domestic oil extraction increased by 92 percent, while oil imports fell by 95 percent and net imports of oil and oil products became negative. The program to increase grain production also was successful. From 1.91 million tons in 1991, grain production rose to 3.22 million tons in 1995, an increase of 69 percent (Ministry of economy). Today Uzbekistan produces about 90 percent of the foodstuffs consumed in the country. Near the end of the contraction stage, the crisis conditions of 1992 had largely abated. Oil self-sufficiency had been achieved and domestic grain production was growing rapidly. While the budget deficit was still large and inflation was still high, the current account was under control and output was no longer falling significantly. At this point, Uzbekistan appeared to contemplate a shift in policy toward the Washington Consensus approach. An IMF stabilization program was accepted in 1994 which provided a stand-by arrangement loan. In 1995 some trade liberalisation began, in line with the agreement, and current account convertibility was promised by the end of 1996 (EIU, 2002: 22, and 2000: 21). The process of price liberalisation was essentially completed by early Large loans were obtained from the World Bank, the EBRD, and the Asian Development Bank. If Uzbekistan had continued along this path, no Uzbek growth puzzle would have arisen. However, the turn toward the Washington Consensus approach was short-lived. 3. STAGE 2: EXPANSION, In 1996 both internal and external conditions in cotton put a severe squeeze on the economy. Cotton production, which had varied only moderately between a high of 4.23 million tons and a low of 3.94 million tons during , suddenly fell to 3.35 million tons in 1996, a decline of 15 percent from the previous year (Ministry of economy). At the same time, the world market price of cotton fell by 15% (IMF, 2000: 43). Cotton exports had represented 48 to 56 percent of Uzbekistan s foreign exchange earnings in , and this caused an economic crisis. The government responded by making a large loan to the cotton sector and reversing course on economic policy. In January 1997 the government formally introduced a multiple exchange rate system along with tighter controls on trade (IMF 2000: 7). The newly tightened trade and currency regimes were intended to support a strategy of import substitution, aimed at diversifying the economy and reducing its dependence on cotton exports over time. Under this industrial policy, certain sectors were given access to foreign exchange at a favourable rate to purchase imported inputs, while sectors not targeted for assistance had to compete for limited supplies of foreign exchange at less favourable rates. Favoured sectors were both protected from foreign competition, using both tariff and non-tariff barriers, and were subsidized by resource transfers from other sectors of the economy, particularly from the cotton sector. The government also directed cheap credit through the Central Bank of Uzbekistan to certain sectors via the state-controlled commercial banks. The IMF responded to the shift away from liberalisation by suspending its stand-by loan facility. Large-scale lending by the EBRD and World Bank were also suspended (EIU, 2002: 22). A joint IMF Uzbekistan working group to discuss economic policy failed to reach agreement. Relations with the IMF were basically put on hold, a situation that lasted until late There were some moves toward liberalisation and privatisation after In late 1996 a privatisation program aimed at medium and large enterprises was begun, based on investment funds as the new shareholders. In 1998 a third stage of privatisation began with World Bank support, calling for case by case privatisation of large state enterprises, with foreign participation encouraged (IMF, 2000: 14). In the agricultural sector, state orders in cotton declined from 75 percent of output in 1994 to 30 percent in 1997, while state orders were ended in all other crops except grain after The government began to encourage foreign direct investment in certain sectors, which led to investments in autos, electronics, textiles, chemicals, mining, and agro-processing (Alam and Banerji: 9). Table 1.8. Energy Balance of Uzbekistan (thousand tons) Domestic crude oil extraction 3,944 5,517 7,586 7,621 7,891 8,104 Oil and oil product imports 4,762 3, Domestic consumption of oil products 8,201 7,368 6,961 6,547 6,520 6,934 Net oil and oil product imports 4,258 2, , Source: IMF (2000: 51). 8

9 Linking macroeconomic policy to poverty reduction in Uzbekistan Large foreign investors included Daewoo, BAT, Newmont Mining, Mercedes, and Lonrho. Official statistics show that by 2000 the nonstate sector of the economy accounted for 72.6 percent of GDP and of 76 percent of employment (Ministry of economy). However, the government has retained significant control over large enterprises, whether still public or officially privatised, and also over cotton and grain production. A 1998 decree allowed the government to block corporate strategic decisions even by majority foreign shareholders in certain cases (EIU, 2002: 22). By the end of the 2001, Uzbekistan, while having a private market oriented sector, was still following in the main a state-guided development strategy. In 1996 the cotton crisis described above had a big effect on both monetary and fiscal policy. Monetary policy clearly became loose that year, as table 1.3 above shows. For the first time the rate of money growth exceeded price inflation and the ratio of money to GDP increased. The previously huge gap between nominal interest rates and inflation rates fell sharply, although real interest rates remained negative. Fiscal policy turned sharply expansionary that year, as the budget deficit shot up from 4.1 percent to 7.3 percent of GDP. In the following years monetary policy returned to a relatively tight stance, based on the measure we have been employing. The growth rate of the money supply was slower than the rise in the GDP price deflator every year during (see table 1.3), and the real money supply gradually trended downward. The money supply reached a low of 12.4 percent of GDP in In 2001, monetary policy eased as money growth rose above the inflation rate. During the entire period, real interest rates remained negative, except for 1998 when the real bank refinancing rate was positive. During fiscal policy was mildly expansionary, as the budget deficit stayed between 2.4 percent and 3.0 percent of GDP. After 1999 fiscal policy moved in a tighter direction as the deficit fell to the one percent of GDP range. Uzbekistan s foreign exchange earnings potential encountered serious difficulties during the expansion phase, from three different sources. First, its largest hard currency earner, cotton, had serious production problems. Following the sharp drop in cotton output in 1996 mentioned above, output recovered in 1997, then had a series of ups and down, with bad years in 1998 and Second, world cotton prices hit a peak in 1995, declining steadily thereafter, to 43 percent of the 1995 level in 2001, based on the Twelve Markets Cotton Price Index. Third, the Russian financial crisis of 1998 dealt a hard blow to Uzbekistan s exports, as the ruble lost three fourths of its value, causing the real exchange rate of the soum against the ruble to appreciate by a factor of three from December 1997 to December 1998 (see section 4.5 below). Russia had been the destination of 18 percent of Uzbekistan s total exports in 1997 (IMF 2000: 74 75). 2 Despite these obstacles, Uzbekistan maintained steady economic growth starting in While Uzbekistan grew faster than the CIS average during this subperiod, its rate of growth was not impressive compared to its Central Asian neighbours, all of which grew faster than Uzbekistan, as Table 1.9 shows. 3 However, the meaning of this comparison is difficult to interpret since, during , the other Central Asian republics were recovering from much larger GDP declines than Uzbekistan had experienced. Table 1.9. Average GDP Growth Rates for Five Central Asian Republics and the CIS, Kazakhstan 4.2 Kyrgyzstan 5.5 Tajikistan 4.0 Turkmenistan 4.8 Uzbekistan 3.5 CIS These figures are compound annual average GDP growth rates. 2 Weighted average. Source: Calculated from data from EBRD (2002: 17). Uzbekistan s inflation rate stabilized after 1997 in the percent range (see table 1.5 above). While stable inflation at such a rate is not inconsistent with relatively good performance of real economic variables, it is a high rate compared to that of the other Central Asian states. Only Tajiskistan has had higher inflation over that period, while the others were at or near the single digit level by the end of the period. The CIS median inflation rate was below that of Uzbekistan in every year during We saw above that, after the difficult year of 1996, Uzbekistan s fiscal deficit was not large, ranging from 2.4 percent to 3.0 percent in and falling to about one percent in This is a very respectable performance. In light of the serious blows to Uzbekistan s foreign exchange earning capacity during stage 2 cited above, and in view of the tendency for the balance of payments to worsen in an expanding economy, Uzbekistan s current account balance record in stage 2 is impressive. During the current account balance averaged negative one-quarter of one percent of GDP (from table 1.7). 4. A CLOSER LOOK AT UZBEKISTAN S ECONOMIC PERFORMANCE DURING GDP Growth Data There is some debate about the reliability of Uzbekistan s official data for GDP growth, based on a number of methodological concerns (IMF, 2000: 8, EIU, 2002: 25). The 2 The Daewoo joint venture auto factory was trying to sell to the Russian market, but the ruble depreciation made this difficult. On the other hand, Russian machinery became cheaper for Uzbekistan to purchase. 3 For comparability across countries, table 1.9, as well as table 1.4, uses GDP growth rates from the EBRD, which differ for some years from the official growth data in table 1.2. The GDP growth rate average during is 4.0 percent rather than 3.5 percent using official data. The only large difference is for 1997 see section 4.1 for a discussion of this matter. 9

10 Chapter 1. The record of economic transition in Uzbekistan EBRD and the IMF have published GDP growth rates for Uzbekistan, the former covering (with an estimate for 2001) and the latter covering (EBRD, 2002: 91; IMF, 2000: 38). The differences between the published IMF and EBRD series and the official government series are not great. Of the ten years, the three growth series are identical for four years, are within 0.2 percentage points of each other for five years, with a sizeable difference only for 1997, when official statistics list a 5.2 percent growth rate while the IMF and EBRD cite a 2.5 percent growth rate. We have used the official GDP growth rate data in most cases, because the differences among the series are small and the official series is the only one with final figures for all ten years. Some have called attention to the disparity between GDP growth rates and electricity usage growth rates. Official statistics show electric power output declining during , when GDP was growing (Ministry of economy). However, whether this is inconsistent with reported GDP growth rates is a complex matter. The structure of Uzbekistan s economy changed significantly over the period. The share of trade in GDP rose from 6.0 percent in 1995 to 11.5 percent in 2001, while the share of industry fell from 19.7 percent to 16.2 percent over that period (table 1.18 below). An IMF staff economist found that, using output estimates based on electricity consumption through 1997, Uzbekistan s growth performance still ranked as the best of the former Soviet republics (Zettelmeyer, 1999: 275). The EBRD series on GDP growth for transition countries since 1989 shows Uzbekistan s cumulative growth so far ahead of that of the other former Soviet republics (table 1.1) that it seems unlikely that Uzbekistan s first place ranking in GDP growth could be solely a product of inaccurate growth data. Like all of the other transition countries, Uzbekistan has a sizeable informal sector that is, a sector that operates more or less illegally. A knowledgeable official estimated the informal sector at percent of GDP in An estimate of this sector is included in the official GDP growth data. The difficulty of accurately estimating the informal sector may impart a downward bias to official GDP growth data, as for other transition countries Growth, Employment, and Population Despite declining GDP in , official statistics show a slight increase in the number of employed workers during that period, at a rate of 0.6 percent per year. While a compositional shift toward more labor-intensive sectors could hypothetically account for this, it seems likely that underemployment rose over this period. During the period of GDP expansion from , official statistics show employment growing at the rate of 1.3 percent per year, which seems plausible given the 4.0 percent per year GDP growth rate over that period (Ministry of economy). However, employment growth fell short of the growth rate of the able-bodied adult population, which increased at the rate of 2.9 percent per year during The total population grew at the rate of 1.7 percent per year during , which is well below the 3.9 percent GDP growth rate over that period. Hence, during that period GDP per capita grew at about 2.2 percent per year. 4 GNP per capita in US dollars, based on purchasing power parity, rose from $ 2440 in 1995 to $ 2982 in 2000, an increase of 4.1 percent per year (Economic Trends: Uzbekistan, 2001: 9) Fiscal Performance Uzbekistan s fiscal balance performance has as one of its underpinnings an ability to effectively collect taxes that is shared by few other former Soviet republics. Table 1.10 shows the evolution of total government expenditure and revenue during Both were close to half as large as GDP in 1991, the last year of the Soviet Union. In 1992 revenues fell to 31.4 percent of GDP, but there was no significant further decline thereafter. From revenues remained in the range percent. While there is a slight downward trend in revenues, they ended up in 2001 only 0.7 percentage points below the 1992 ratio to GDP. Other former Soviet republics experienced declining tax revenues, forcing a choice between sharp slashes in public spending or rapidly rising deficits. Because in Uzbekistan the authorities were able to continue to collect taxes effectively, they could reduce their initially high fiscal deficit with gradual and relatively modest reductions in public spending, compared to what happened elsewhere. Public spending fell from nearly half as large as GDP in 1992 to 35.3 percent in It rose to 41.6 percent in 1996 when the cotton crisis hit, then fell the next year to 32.5 percent of GDP, remaining within 3 percentage points of that for the rest of the period. The share of public spending in GDP at the end of the period was slightly greater than the similar figure for the US of 30.6 percent in Investment and Saving Table 1.11 provides data on investment and saving during the period of expansion. Investment averaged 20.0 percent of GDP during while saving averaged 19.7 percent of GDP. That is, domestic saving was essentially adequate to finance investment. However, given the aver- 4 The trend in population growth has been downward, reaching about 1.5 percent per year by the end of the period. Table Government Expenditure, Revenue, and Budget Balance (percentage of GDP) Government Expenditure Government Revenue Budget Balance Source: Calculated from data from Ministry of economy and EBRD (2002: 92). 10

11 Linking macroeconomic policy to poverty reduction in Uzbekistan age budget deficit over that period of 2.6 percent of GDP, there was a small shortfall of domestic saving to cover both investment and the government budget deficit. During the last three years, the budget deficit averaged only 1.4 percent of GDP, while investment was 19.0 percent of GDP and saving was 18.9 percent of GDP. These data suggest that only a very modest capital inflow should have been sufficient to finance the actual level of investment during the growth stage, and particularly during the last three years. However, the question of whether the actual investment rate has been adequate remains. Table 1.12 shows the sources of investment funds during The government budget provided between 23 and 29 percent. Enterprise funds supplied a declining proportion, falling from 46.6 percent to 27.5 percent over the period. Despite the near adequacy of domestic savings for financing investment and the budget deficit, the data show a significant and rising share of investment financed by foreign capital inflow, which rose from 16.1 percent to 29.3 percent of investment sources over the period. This means that an outflow of capital nearly as large as the inflow was taking place. 5 Table 1.13 shows the distribution of investment among sectors during Industry claimed over half of investment, led by the fuel and energy complex. Transport and communication received one-fourth of investment, while under ten percent went to agriculture. The high share of investment in transport and communication 5 For example, in 1998 the sum of foreign direct investment plus foreign loan drawings was $ 1,055 million, which equaled 47 percent of Uzbekistan s total domestic investment that year. However, foreign loan repayments that year totaled $ 1,051 million (IMF, 2000: 71). A substantial part of the capital inflow evidently found its way into the financing of domestic investment it financed 19.7 percent of domestic investment but in the same year a nearly equal value of capital flowed out as flowed in. Table Investment and Saving as a Percentage of GDP, Year Investment Saving Source: Ministry of Economy. Table Sources of Investment Funds as a Percentage of Total Investment Budget funds Funds of enterprises Bank credits Funds of off-budget foundations Funds of population Foreign investments and credits Other sources Source: Ministry of Economy. Table Sectoral Composition of Investment, Investment by Sector Percentage of Total All sectors Industry 54.9 Fuel and energy complex 19.0 Machine building and metal working 8.3 Chemical and petrochemical industries 10.1 Other industry 17.5 Agriculture 9.4 Construction 0.9 Transport and communication 25.0 Other sectors Other sectors include trade, public catering and services. Sources: Ministry of Economy. 11

12 Chapter 1. The record of economic transition in Uzbekistan indicates a commitment to improving the infrastructure. Large investments were made during those years in new rail lines and air transport Foreign Trade and Investment We noted above Uzbekistan s surprising achievement of current account balance during Table 1.14 shows how this was achieved. As Uzbekistan s exports dropped precipitously after 1997 in response to the Russian financial crisis and falling cotton output and prices, imports fell by even more, producing a trade surplus every year during Thus, import compression allowed Uzbekistan to weather its export problems without incurring a large current account deficit. Uzbekistan has been a moderately trade-dependent nation, with exports and imports each varying in the range of 25 to 35 percent of GDP since Prior to 1998 Uzbekistan s merchandis exports did relatively well, growing by 28 percent during This occurred despite a 135 percent appreciation in the real exchange rate of the soum against the dollar from December 1993 to December 1997 and a 10 percent appreciation of the real exchange rate against the ruble (table 1.15). After the Russian financial crisis, Uzbekistan s exports fell to a level approximately equal to that of the early 1990s. Uzbekistan s export performance, which is measured in US dollars, as is usual for balance of payments data, is understated by that measure because of the rapid rise in the value of the US dollar during the 1990s relative to that of other major currencies. A broad trade-weighted index of the value of the US dollar, relative to other currencies, rose by 52 percent from 1993 to The last line of table 1.14 shows exports deflated by that index of the value of the dollar. This gives a better measure of the import-purchasing power provided by Uzbekistan s exports. While the year-to-year pattern in this series is similar to that of exports in US dollars, the long-term trend is different, in that the value of exports in 2001 based on the alternative measure is 46 percent greater than exports in Uzbekistan has attracted little foreign direct investment on aggregate. In its best year, 1997, FDI totalled $ 167 million, which was only 4.5 percent of export earnings and 1.6 percent of GDP. The cumulative FDI inflow from 1989 to 2001 was only $ 30 per capita, compared to the CIS average of $ 196 per capita during that period. It had the second lowest FDI per capita of the CIS countries, with only Tajiskistan having a lower rate (EBRD, 2002: 24). Uzbekistan renounced its share of the former Soviet Union s debts and assets in November of However, after that foreign debt was steadily accumulated through 1999, rising from $ 1107 million in 1994 to $ 4310 million in 1999, with no further appreciable increase since 1999 (table 1.14). Most of this was a result of borrowing from international agencies. As a percentage of exports, foreign debt rose from 35.9 percent in 1994 to just over 150 percent of exports in The debt service ratio, which measures the percentage of export revenue required to make payments of interest and principal coming due, was under 20 percent through In the last two years of the period it rose sharply to 30.4 percent. This would appear to indicate a looming repayment problem, but this actually reflects short-term borrowings coming due in 2000 and The debt service ratio is expected to decline after 2002 (in the absence of new heavy borrowing), and a World Bank official knowledgeable about Uzbekistan s debt situation remarked that its current debt burden is sustainable based on its recent economic trajectory. The nominal value of the Uzbekistan soum against the dollar has steadily depreciated over time, as shown in table This is an inevitable accompaniment to the high rate of domestic inflation. Each encourages the other, as domestic inflation puts downward pressure on the international value of the soum, while depreciation of the soum against other currencies puts inflationary pressure on its domestic value. The market (illegal) exchange rate, in soums per dollar, has been higher than the controlled official rate, and the spread between the two grew during Estimates of the real exchange rate of the soum, against the US dollar and the Russian ruble, are also shown in table This shows a sharp appreciation against the dollar in 1995, followed by very gradual further appreciation through 1999, then a decline in The real exchange rate of the soum against the ruble was relatively stable until the Russian financial crisis tripled the rate in Since the end of 1998, the real depreciation of the soum against the dollar, together with the appreciation of the ruble against the dollar during that period, depreciated the soum s real exchange rate against the ruble significantly. A major aim of the Uzbekistan authorities was to wean the economy from its dependence on cotton exports. Table 1.16 shows that cotton, which represented almost half of export earnings in 1992, initially increased further until 1994, then fell steadily to under one fourth of export earnings in Exports of fruits and vegetables have increased significantly over the period (CER, 1999: 12). The two categories of exports in table 1.16 that had rising shares are services, which might reflect rising tourism and the earnings of the new national airline, and the other category. 6 The change in import composition was consistent with the government s import substitution policy goals. Energy and foodstuffs dropped sharply over the period as a share of imports, replaced by domestic production. Machines and equipment rose from 9.2 percent of imports in 1992 to 47.9 percent in 1995, thereafter remaining between 35 percent and 47 percent of imports. The Uzbekistan authorities also sought to diversify their trade beyond their traditional CIS trading partners. Table 1.17 shows that, in 1992, 61.9 percent of exports and 55.5 percent of imports were in their traditional trading sphere. After the ruble zone broke up, Uzbekistan s exports rapidly shifted away from the CIS, whose share fell to 22.9 percent in After 1996 the CIS export share recovered, thereafter ranging from about 25 to 35 percent of exports. The import experience was similar, with the CIS import share dropping rapidly during to 28.0 percent. In imports too there followed a recovery to percent of imports by the end of the period. 6 Tourism grew by a factor of 2.7 from , while foreign visitors grew 5.4 times (CER, 1999: 12). 12

13 Linking macroeconomic policy to poverty reduction in Uzbekistan Table International Transactions of Uzbekistan (millions of US dollars, unless otherwise noted) Merchandise exports 2, , , , , , , , ,755.0 Merchandise imports 3, , , , , , , , ,479.0 Trade balance Current account balance Foreign direct investment External debt 1, , , , , , , ,392.0 External debt/exports Debt service/exports Real Exports 2 2, , , , , , , , , Expressed as a percentage. 2 Real exports are the dollar value of merchandise exports adjusted by a broad index of the nominal trade-weighted value of the US dollar. Source: IMF (2000: 71), EBRD (2002: 24, 91), EIU (1999: 46), Economic Report of the President (2002: 446). Table Exchange Rates of the Soum Official exchange rate (soum/$) Market exchange rate (soum/$) ,408.6 Market to official ER ratio Real index of $/soum Real index of rubles/soum Average nominal exchange rate. 2 Estimated real exchange rate index at the end of each year, except for Estimated real exchange rate index at the end of September, Source: Economic Trends: Uzbekistan (2001: 45, 47). Real exchange rate indexes were estimated from a graph. Table Composition of Exports and Imports by Product (percentage of total) All Exports Ginned cotton Chemical products and plastics Energy resources Ferrous and nonferrous metals Machines and equipment Foodstuffs Services Other All Imports Chemical products and plastics Energy resources Ferrous and nonferrous metals Machines and equipment Foodstuffs Services Other Source: Ministry of economy. 13

14 Chapter 1. The record of economic transition in Uzbekistan Table Trade with the Commonwealth of Independent States (CIS) and Others, (percentage of total exports or imports) Exports CIS/Total Others/Total Imports CIS/Total Others/Total Source: Calculated from data from Ministry of economy Diversification of the Economy The economy has undergone a substantial shift in its output and employment structure. Tables 1.18 and 1.19 show the sectoral composition by value added (which measures the share of GDP) and by employment respectively. The share of agriculture in GDP declined significantly in the middle of the period but then rose again, ending up only a few percentage points lower as a share of GDP in 2001 than it had been in Measured by employment, agriculture s share did not fall appreciably until the last few years of the period, falling from 41.9 percent to 33.5 percent of total employment over the decade. The difference in the shift out of agriculture by these two measures suggests an increase in value added per worker in agriculture over the period, which may have been due to a change in the product mix. Table 1.20 shows the output change of several major agricultural products from 1991 to Cotton declined by about 30 percent, and large increases were registered in grain, potatoes, and fruit. Thus, there has been some success in diversifying away from cotton in agriculture. Outside of agriculture, the major shifts measured by value added (table 1.18) were large declines in the shares of industry and construction while the shares of transportation and communication, sales and catering, and other services rose substantially. A rise in the share of services was desirable, but since Uzbekistan was not a highly industrialized part of the former Soviet Union, it may be that the drop in the share of industry was greater than would be desirable. The employment shares of the non-agricultural sectors generally changed in the same direction as the value added shares, with the exception of transportation and communication. Industry s share of employment fell much less than its share of GDP, which may partly reflect increased overstaffing. The relative expansion of the sales sector was much smaller by employment than by value added, which probably reflects the relatively high incomes earned in that sector. Table 1.19 also shows employment shares in two service occupations. Public health employment rose as a share of the total, which would seem to be a desirable shift. The devotion of significant resources to education, culture, art, and science was a positive feature of the former Soviet Union, both for economic progress and human welfare. While the share of those occupations in total employment fell by 0.8 percentage points during , it is a good sign that the fall was not sharper Industrial Performance According to official data, industrial production declined only in the first year of the period, 1992, and increased every year thereafter. Table 1.21 shows the annual rate of change of industrial output and its main components during According to these data, total industrial production fell by only 2.0 percent from 1992 to 1995, while in the growth phase it rose by 30.7 percent. Based on these data, industrial output increased by 28.2 percent over the whole period. 7 A number of outside (and inside) observers have expressed doubt about the reliability of the official data on industrial output, due to methodological problems. However, alternative series from the EBRD and Economic Trends: Uzbekistan, which cover the years 1994 to 2000, actually have higher estimates for most years (EBRD, 2002: 91; Economic Trends: Uzbekistan, 2001: 9). 8 Nevertheless, in light of the widespread criticisms of these particular data, and the seeming conflict with available data for individual physical output series, it appears that the data in table 1.21 should be treated with caution. Table 1.22 shows industrial capacity utilization rates for a range of industrial products. Most of the products have utilization rates of less than 60 percent. The exceptions are mainly in raw material processing. Some of the industrial products have very low utilization rates, below 40 percent. Although utilization rate data are not among the most reliable of data for any country, these numbers are so extreme that it would appear reasonable to conclude that the economy has substantial excess productive capacity in the industrial sector, although the quality of the unused plant and machinery may be low Real Income According to official statistics, the real income of the population fell by nearly one-third in 1992, and by 1995 it was still just over half of the 1991 level (table 1.23). Thereafter it rose, reaching percent of the 1991 level by The sources of household income shifted dramatically over the period, reflecting changes in the structure of the economic system, as table 1.24 shows. Wages fell from 7 Changes in industrial production and GDP are not comparable because a different implicit price deflator is used in the two cases. 8 The Economist Intelligence Unit has published estimates of industrial output growth for 1994 through 2000 which are somewhat lower for two of the years, identical for two years, and higher for three years (EIU, 2002: 51). 9 Capacity utilization is inherently difficult to measure accurately, and it is even more difficult to measure in an economy undergoing major structural change. The greatest problem is when equipment should be considered obsolete and removed from the capacity measure. 10 It may be plausible that real income had risen 19 percent above its 1991 level by 2000, although GDP was still 1.4 percent below its 1991 level that year. The consumer price index rose significantly more slowly than the GDP price deflator over the period. 14