Fiscal Transition in Countries of the Former Soviet Union
An Interim Assessment1/
Author: Adrienne Cheasty and Jeffrey M. Davis
Publication Date:
01 Jun 1996
ISBN:
9781451848120
Language:
English
Keywords:
WP; government expenditure priority; FSU government; government intervention; subsidy entitlement; General government deficit; government capital formation; intervention in the economy; government arrears; FSU governments record
Download PDF (2 MB)

This paper reviews the uneven record of fiscal performance in the states of the former Soviet Union since independence. Deficits have come down rapidly, an important contribution to stabilization. Given, however, the unexpectedly severe revenue decline and limited financing, the stabilization was reached by sharp cuts in expenditure. The cuts were abrupt and not focused on transition goals, and the instruments used vitiated normal budgetary processes. Hence, benchmarks of fiscal success other than stabilization are elusive. Government intervention and subsidies remain important, social spending is inefficient, and there is little evidence in the budget of restructuring.

Abstract

This paper reviews the uneven record of fiscal performance in the states of the former Soviet Union since independence. Deficits have come down rapidly, an important contribution to stabilization. Given, however, the unexpectedly severe revenue decline and limited financing, the stabilization was reached by sharp cuts in expenditure. The cuts were abrupt and not focused on transition goals, and the instruments used vitiated normal budgetary processes. Hence, benchmarks of fiscal success other than stabilization are elusive. Government intervention and subsidies remain important, social spending is inefficient, and there is little evidence in the budget of restructuring.

I. Introduction

After the dissolution of the Soviet Union, budget deficits surged in most of its successor states. By now, they have been reduced more or less globally. This paper examines fiscal stabilization in the Baltics, Russia, and other countries of the former Soviet Union (FSU) in the last four years, contrasting the path envisaged at the outset of their transition with actual developments. 1/ Successive sections review the pressures which have complicated deficit reduction, and arguments for caution in cutting deficits during the transition; the factors underlying the deficit reduction that took place despite the pressures; as well as the way in which the deficit cuts have been achieved, and the costs of inappropriate means of stabilization.

In 1992, the unweighted average deficit in the region of the FSU rose to 13 percent of GDP, compared with 8 1/2 percent in the Soviet Union in 1989 and broad balance until the mid-1980s (Table 1). 2/ The deficit remained minimal only in the Baltics, which had begun to reform prior to 1992, and in Azerbaijan and Turkmenistan, which profited from energy price liberalization. Factors contributing to the large deficits varied across countries, but included: (i) governments and electorates which ill-understood financing constraints and could not resist the temptation to resort to the inflation tax; (ii) costly regional conflicts; (iii) the loss of up to 20 percent of GDP in inter-U.S.S.R. transfers; and (iv) the initiation of price liberalization without a concomitant dismantling of government procurement and subsidy entitlements--implying a jump in subsidy bills.

Table 1.

FSU: The Evolution of the General Government Deficit, 1992–95

article image
Source: Data provided by the authorities and IMF staff estimates.

The regional contraction in the deficit since then has been broad-based and substantial. The unweighted average deficit dropped to 12 percent of GDP in 1993 and to an estimated 4 1/2 percent two years later. By 1995, 9 of the 15 states had a deficit of below 5 percent of GDP. The sharp fiscal consolidation has been one of the more striking achievements of the region. Nevertheless, there are concerns that deficits have often been reduced through arbitrary cuts in government spending, nonpayment of bills, and shifting of expenditures off-budget. Such methods not only vitiate the role of the budget in restructuring the economy, but raise questions about the sustainability of fiscal consolidation to date. 1/

II. Constraints on Deficit Reduction

The decline in FSU fiscal deficits is particularly noteworthy because a number of factors have been more adverse than expected. Shocks have been more severe, and some expected remedies have proved disappointing. Moreover, there were several influential arguments that questioned the appropriateness of pursuing stabilization by targeting the measured fiscal deficit at the outset of the transition. This section explains why FSU deficits might have been expected to remain large.

1. Unexpectedly adverse factors

The path of fiscal adjustment has differed from what was predicted and recommended by experts partly because several elements of the transition process turned out to be unexpectedly unfavorable. These include: a more severe revenue decline than anticipated; the unexpectedly rapid erosion of interstate financing; disappointing returns to the state from privatization; the tenacity of subsidies in the budget; and a smaller than hoped-for peace dividend. 1/

a. Revenue decline

Some revenue decline had been anticipated, but its depth--an average loss of 11 percent of GDP between 1991 and 1995--was unexpected (Table 2). In essence, the decline implies an additional “hidden” 11 percent of GDP adjustment underlying the apparent cut in the deficit figures.

Table 2.

FSU: The Evolution of General Government Revenue, 1991–95

article image
Source: Data provided by the authorities and IMF staff estimates.

Figures for 1991 and 1992 are adjusted to take account of a revision in the Kyrgyz national accounts from 1993 on.

Figures for 1994 and 1995 are adjusted to take account of a revision in the Moldovan national accounts from 1994 on.

Within this range, individual country experience has varied significantly. To generalize, slow reformers (Belarus, Tajikistan, Ukraine and Uzbekistan) remained relatively protected from the fall, but all began to show evidence of impending loss by 1994–95. The decline in some fast reformers (the Baltics and Moldova) was showing tentative signs of bottoming out at the end of 1995. Unsurprisingly, war-torn countries have lost control over their tax collection and have fared worst. Turkmenistan also fared badly, since its revenues depended so heavily on the energy payments from its neighbors which fell into arrears.

The causes and consequences of the decline have been discussed elsewhere, and will be summarized only briefly here. 1/ Revenue fell because of a combination of systemic shocks, the concentration of the output loss in traditional tax bases, revenue-losing tax reform, inadequate institutional response, and failure to tap the full revenue potential of the energy sector.

The systemic shocks ranged from regional wars to the sharp terms of trade shock. As discussed below, the cessation of interstate financing aggravated the disruption in production and income and the needed adjustment in the majority of states. Other shocks with fiscal consequences included the decentralization of foreign trade--which eliminated previous large rents and price differential taxes--and the overhaul of the Soviet pension system--which quadrupled contribution rates and eroded enterprise profitability and nonearmarked revenues.

Traditional tax bases shrank more sharply than other components of GDP. Industrial production, state enterprise profitability, the real wage, and retail distribution through state stores all shrank by more than GDP in the opening years of the transition. These were the main bases for income taxes on enterprises and individuals, VAT and excises. 1/ Growth areas since 1992 have been exports, agriculture, small services and the rest of the informal private sector. All of these have been relatively lightly taxed, in part because they have been difficult to include in the traditional administration. Tax collection efforts in the FSU continue to center on a withholding tax for wages, and the culling of the enterprise income tax and VAT directly from enterprises’ bank accounts.

Many of the tax reforms that have taken place since 1992 have been revenue-losing. Some of these losses were desirable in the sense that they removed distortions that would have impeded the development of the market. The most expensive reforms include the dismantling of the trade revenue system mentioned above, a narrowing of the enterprise income tax base to a more economic definition of profits, and the lowering of VAT rates from the 28 percent rate at which it was introduced to a median 20 percent by 1995.

These problems have been exacerbated by deficiencies in governance. The most tangible cost of weak government has been a proliferation of exemptions which have eroded the base for major revenue earners, notably VAT and excises. 2/ Tax administration has been complicated not only by the escalation in the number and diversity of potential taxpayers, but also by declining enforcement powers and by eroding real wages and large vacancy rates in the tax authority. At the same time, shifting tax legislation and uneasy relations between central and sub-national governments have contributed to honest confusion, opportunities for evasion, and withholding of tax payments at different stages of collection.

One reason the severity of the revenue decline was underestimated was the enormous potential for revenue from the energy sector, once prices were liberalized. The fuel and energy sector accounted for 11 percent of industrial value-added in the USSR, despite the fact that, at the beginning of 1992, oil prices were administratively set at less than 5 percent of world levels (McPherson, 1996). For energy exporting countries, revenue was expected from income taxes, royalties, natural resource taxes, and dividends on the state’s share in energy companies; for both exporters and importers, excises and other taxes on energy use were considered appropriate, given the need to rationalize energy consumption in the region. But in fact, the fiscal contribution from energy has been much below expectations for several reasons: continuing repression of domestic energy prices and large scale extra-budgetary subsidization; governments’ more general reluctance to add taxes in the short run, when the relative price change had been so great; decentralization, which has made rich regions more autonomous and reduced interregional income redistribution; and the success of powerful lobbies in delaying tax measures from being legislated.

b. Loss of interstate financing

A related adverse influence on budgetary performance was the unexpectedly rapid erosion of interstate trade financing. In 1992, Russia provided nearly 13 percent of GDP in explicit financing to other FSU states through the monetary mechanism of the ruble area, and a further estimated 13 1/2 percent in trade subsidies (Wolf, et al., 1994). By 1993, as part of the dismantling of the ruble area, credits (including cash) had been cut to 2 percent of GDP and trade subsidies are estimated to have dropped to around 1/5 their 1992 level. The loss of the cross-border subsidy--and so suddenly--is estimated to have had a negative impact of an average 13 percent of GDP on countries other than Russia (Citrin, 1995). The shock was concentrated in the traditional sectors which benefitted most from the subsidies and which were the areas where tax collection was strongest. Moreover, the replacement of what had been low- or no-cost financing with market-related loans contributed to the debt burden and expenditure of some countries.

c. Privatization

Privatization has fallen short of expectations in several respects: it has not generated the budgetary financing, revenue, or expenditure savings that might have been anticipated.

It had generally been recognized that actual receipts from privatization would amount to only a fraction of their potential--taking into account transaction costs, difficulties in identifying the right market price, the serious liquidity constraints facing buyers, the limitations put on foreign participation, and the need to include a large redistributional bonus to compensate the public for the sale. 1/ Nonetheless, given the large share of national wealth in the state’s hands, privatization receipts could have generated a significant flow of funds to the budget to help cover fiscal imbalances during the transition. The potential is illustrated in Table 3. Under certain simplifying but conservative assumptions, nine FSU countries could, in principle, have generated an average of 13 percent of GDP annually for the budget between 1989 and mid-1995 from their privatized assets. 1/ Collection of even part of those amounts would have helped finance the restructuring costs of enterprises and the hump in the social safety net needed to support those adversely affected by the transition process.

Table 3.

FSU: Budgetary Privatization Receipts

article image
Source: IMF (1996).

The calculation of potential revenues from privatization represents the annual average percentages of GDP that could have been earned between 1989 and mid-1995 by sales of state assets, assuming a price-earnings ratio of 3, and that the USSR private sector share of GDP in 1989 (13 percent) applied evenly across all states at that time.

The implied rate of return is calculated as actual cumulative revenues over potential cumulative revenues, 1991–95. No data are available on privatization receipts in 1989–90; it is assumed these were minimal.

In reality, as shown in Table 3, budgetary privatization receipts have been minimal. Cumulatively through 1995, they have not amounted to much more than 1 percent of GDP in any country other than Kazakstan. The implied average rate of return to government from privatization so far is less than 3 1/2 percent of the value of the divested assets. The insignificant return reflects the widespread use of voucher (rather than “cash”) privatization, larger-than-recommended free share-outs to insiders (workers and management), constraints on foreign participation, the uncertainties created by the volatile political and economic climate, and the decapitalization of balance sheets caused by inflation, interenterprise arrears, and inadequate accounting and regulatory practices.

Privatization had always been seen as implying some tax revenue loss. Enterprises could no longer be called upon to pay ad hoc surcharges or extraordinary tax rates. Moreover, the process was expected to entail some downsizing of the average firm size--creating difficulties for the tax administration because of the increase in the number of potential taxpayers and because small enterprises are harder and less cost-effective to tax. However, it was hoped that the loss would be compensated by increases in the efficiency of production from the privatization, part of which would come from the enterprises’ newfound ability to charge market prices for their output. In practice, the net revenue yield from privatized sectors has tended to decline sharply. Factors contributing to this decline have included: important tax breaks granted to investors (particularly foreign investors with sophisticated legal backing) as part of the privatization/restructuring package; 1/ unforeseen cost increases that reduced enterprises’ tax bases relative to when they were in the public domain (for instance, setting-up and retooling costs or a shift in accounting to economic depreciation); and, perhaps most often, the disappearance of the profitable activities of the privatized enterprises into the informal sector.

Privatization was seen as a short cut to limit the burden of enterprise restructuring costs to the budget. But to protect employment and because of a dearth of buyers for unprofitable enterprises, the trend, following Germany’s Treuhandanstalt and the World Bank’s “enterprise hospital” model, has instead turned towards restructuring enterprises before privatization, often using budgetary funds. Hence, the privatization process has tended to imply the offloading of profitmakers and the direct assumption of responsibility by government for bailing-out unprofitable concerns (including periodic repayment of interenterprise arrears). 2/

d. Subsidies

Subsidies still weigh heavily on FSU budgets, largely because price liberalization has remained incomplete in most countries, notably in the energy sector and in agriculture. In 1989, measured subsidies in the USSR amounted to 13 1/2 percent of GDP. Their elimination through upfront price liberalization was seen as the main automatic stabilizer for the post-Soviet budget (though the need for some offsetting increase in social safety net transfers was always recognized). But at end-1994, production support still amounted to 11–13 percent of GDP in Kazakstan, Russia, Ukraine, and Uzbekistan, and consumer subsidies, where measurable, added a further 2–3 percent of GDP to the subsidy bill (see Chapter IV, below). Moreover, the budgeted amounts underestimate total government support for production and consumption, because the failure of enterprises to pay tax bills and of consumers to pay utility bills ultimately translated into a higher fiscal deficit.

e. The peace dividend

The peace dividend has been significantly smaller than had been hoped, particularly in its impact on the public finances. It was expected to have two parts: savings in defense expenditure in the post-Soviet bloc, and western aid related to the demilitarization. 1/ The macroeconomic impact of cuts in defense expenditure, though these have been substantial as a share of total military outlays, has been limited by the fact that they started from a level repressed by low administered prices, and because of some offsetting costs. These include the various regional conflicts, conversion of defense industries (though see Chapter IV), demobilization and military housing (seen as politically imperative), and the proliferation of military structures--15 instead of one. Moreover, the potential for decommissioning the post-Soviet armed forces (and unleashing their labor into the civilian economy) has been limited by fears of consequent social and political unrest. Aid from the West was not, as had been mooted, linked to defense savings there, nor to the scale of military conversion in the Soviet bloc. 2/ External support has not been proportionate to western defense cuts, perhaps in part because the money was needed at home to shore up domestic demand as defense-related activities were scaled down.

2. Arguments against emphasizing deficit cuts

Besides the fact that the adverse economic climate made deficit cutting difficult, the decline in FSU deficits is noteworthy because reduction of the measured deficit was not universally seen as an appropriate objective for governments in the early years of transition. This section outlines the problems that were seen as creating difficulties for judging the appropriate level of the fiscal deficit.

a. Difficulties in defining government

Transition experts argued that the coverage of any deficit measure would inevitably be arbitrary in FSU countries, because of the indistinct--often indefinable--division between government functions and production in a planned system (Tanzi, 1993). In a sense, “any” deficit target could be met by redefining fiscal boundaries.

b. The need to reduce the size of government

Further, a repeated repositioning of fiscal boundaries was seen as a central element of the transition strategy--since a primary goal is to move activities out of government into the private sector. It would be unlikely for this restructuring to be budget-neutral. Hence, deficits, however measured, might be expected to fluctuate widely as piecemeal functions were moved off-budget. The macroeconomic impact of deficit movements under large scale restructuring could be quite different from changes in the fiscal stance of an equivalent size in industrial countries, and might in many cases amount only to changes in statistical classification. Hence the movements should not necessarily be constrained or offset.

c. Monetization of implicit subsidies

Not only might deficits move significantly and unpredictably, but arguably they should be expected--and permitted--to widen at the outset of the transition. Under planning, much of government’s redistribution had been effected by regulation rather than through the budget--for instance the repression of food prices, the effect of which on farmers was partly offset by repressed interest rates on agricultural borrowing, and the cross-subsidization of household energy consumption by industry. As price liberalization took place, off-budget subsidization had to become monetized; and as finances tightened, enterprises and banks lobbied to transfer their quasi-fiscal activities to government. 1/ While some of the subsidization could be dismantled immediately, in some cases the social cost needed to be cushioned. The transfer of previously implicit costs to the budget was seen as an important interim step in deregulation (Holzman, 1991). Price liberalization without using the budget as a buffer could exacerbate enterprises’ difficulties and slow their reform, and could entail unacceptable social costs. Moreover, budgetizing the costs of administered systems was seen as a way of helping governments prioritize and of monitoring the dismantling of intervention.

d. Financial demands of restructuring

More generally, it was also recognized that restructuring would entail inevitable fiscal costs--including capital injections into enterprises and banks, the overhaul of the tax administration and civil service and their computerization, the reform of the legislative and accounting systems, compensation to displaced workers, and new investment, particularly in infrastructure. These activities could only to a limited extent be carried out by the private sector and their delay would have wider repercussions for the pace of transition. 1/ Hence, it could be optimal for the fiscal deficit to remain sizeable.

These arguments make the point that targeting a given deficit, and in particular, effecting stabilization by cutting deficits, may not have been appropriate at the outset of the FSU transition. In a situation where the distinction between government and the rest of the economy is not known or is not intrinsic to the stabilization problem, it may be difficult to judge the appropriate fiscal deficit, and so monetary or exchange rate policy may be more operationally suitable for achieving stabilization than fiscal policy. The targeting of a given credit expansion--which could be allocated to government or the rest of the economy depending on the fluctuating assignment of functions--would seem to achieve adjustment with less distortion than would targeting a steady downward trend of the deficit; likewise, exchange rate based stabilization would be relatively neutral to a shifting role of government in the economy.

In reality, as is evident from Table 1, the predicted erraticism in the deficit on account of changing coverage has not been observed. The downward movement of FSU deficits has been more or less uninterrupted except in a few extreme cases; and it has been accompanied by fairly smooth downward trends in revenue and expenditure. On the revenue side, governments’ commercial profit-generating functions disappeared early in the process, privatized of their own accord (for instance, foreign trade and the management of depreciation funds).

As regards the expenditure side, the next section of the paper argues that the inelasticity of the financing available to governments has severely limited the scope for budgetizing new spending functions. Little of the restructuring referred to above can be shown to have taken place through the budget (see Chapter IV). The scarcity of financing appears to have caused it to be postponed (or, to some extent, be conducted haphazardly and nontransparently outside the control of fiscal policy makers).

It should also be noted, however, that some revisionism has begun. With respect to the budgetization of regulatory taxes and subsidies, there are growing doubts about the advisability of governments’ assuming new responsibilities. The incorporation of some functions into the budget could raise their likely total costs, by reducing prior incentives to enterprises to rationalize, and might be an unnecessary detour that could postpone their elimination (Cheasty, 1995). Moreover, incorporating an activity into the budget could ratify its legitimacy as a public function, and governments could face political pressures to maintain it beyond those imposed on enterprises. For these reasons, FSU governments have proven reluctant to expand the coverage of the budget.

As regards possibilities for relying on monetary rather than fiscal policy: besides the fact that financing remained low not only for governments but for the FSU countries as a whole, it further became clear that the division of credit between government and the private sector was not irrelevant. Some credit needed to be left to nongovernment, to finance monetized domestic privatization, and allow the private sector to invest.

III. Why Deficits Have Fallen

In light of the upward pressures, on the one hand, and the arguments for ignoring or accommodating the deficit, on the other, a major reason deficits have come down must be traced to difficulties in financing a higher imbalance. These difficulties too have been greater than anticipated, and made the pursuit of stabilization imperative.

Some of the financing shortfalls have already been discussed. Privatization receipts did not materialize, nor did the hoped-for peace dividend. Other sources of financing--from abroad, from the issue of government paper, from the inflation tax, and even from arrears--have also had limitations.

1. Lack of foreign financing

It had been assumed at the outset of the transition that new governments would find it relatively easy to borrow from abroad because of low initial levels of external debt--which should, presumably, make them creditworthy borrowers and give them cheap access to credit. It was also seen as advisable to maximize foreign borrowing in the very short run, as a way of quenching inflationary pressures by financing demand through a net inflow of foreign resources. 1/

But in fact, positive net foreign financing of the budget has been small, for several reasons: (i) though inflows have been higher than often intimated in the press, they have been sporadic, slow to arrive, tied to sometimes low-priority expenditures and projects, and unevenly spread; (ii) importantly, much of FSU sovereign borrowing has been channelled nontransparently outside the budget; (iii) access to capital markets proved unaffordable for almost all countries, so foreign financing had to come, in the main, from official sources; and (iv) even though the official share of foreign financing has been higher than expected, debt service costs have grown rapidly. In some countries they have already become sufficiently large to offset the budgetary contribution of the inflows.

Table 4 shows the evolution of net foreign budgetary financing. Outliers in 1992 and 1993 make the average financing figures uninformative. In 1994 and 1995, the average was around 2 percent of GDP. Eliminating Armenia, the most significant recipient of foreign aid, from the average gives perhaps a more realistic picture of the foreign support recently available to FSU budgets--not more than 1 1/2 percent a year in net terms.

Table 4.

FSU: Foreign Financing for the General Government Deficit, 1992–95

article image
Source: Data provided by the authorities and IMF staff estimates.

The 1992 data are particularly speculative.

The receipt of foreign capital and its contribution to the deficit has varied a lot from country to country and over the period. In 1993, only the Baltic States were in a position to cover more than half of their (relatively small) fiscal imbalances with foreign funding. In 1994–95, as more countries entered into higher-tranche programs with the IMF, 6–7 of them were able to finance the equivalent of more than half their deficits from abroad. 1/ Armenia has benefitted most from foreign support because of its abundant diaspora, which has both provided direct financing and created important interest groups to influence the international lending community. Despite important western aid, the net impact on Russian budget financing has turned negative (−0.2 percent of GDP) in 1995, partly because of Russia’s vast size and partly because of the statistical impact of the appreciating ruble. Slow reformers, such as Belarus, Tajikistan and Ukraine have clearly been penalized on the financing front. Ukraine has begun to receive significant inflows in connection with its recent reforms, but these have been more than offset by a heavy energy-based debt service bill to Russia and Turkmenistan. Turkmenistan has been a net creditor to the rest of the region throughout the period, to some extent involuntarily because its energy importing partners (notably Ukraine) have run up sizeable arrears.

Countries are also becoming conscious of the accumulation of external debt, and the constraints it imposes on fiscal policy. They began the transition with essentially zero debt (apart from Russia), but the large deficits of 1992 and 1993, the terms of trade shock, high real interest rates, extreme exchange rate depreciations, and mainly negative growth have quickly created a sizeable debt burden for some (Brau, 1995). In 1993, for instance, when the average FSU deficit was 12 percent of GDP, the external debt of the region grew ten-fold (IBRD, 1994–95). Many states were slow to curb independent borrowing by state enterprises, and later found that the growth in their public debt exceeded their measured deficit. By now, some of them have already been classified as heavily-indebted (Georgia, the Kyrgyz Republic and Tajikistan). As a group, they are catching up quickly with other debtor countries.

2. Limitations on domestic financing

a. The absence of domestic securities markets

It had been predicted that demand by individuals and enterprises for domestic financial assets would grow rapidly after 1992, in line with an anticipated rise in measured private saving and the liberalization of enterprise finances. For instance, JSSE (1991) argued that the public would need quasi-indexed new vehicles for saving, to insulate itself from inflation, and to provide for old age in an environment in which state support was no longer certain. 2/ Hence, it was thought that a market for government paper would be relatively easy to develop, and would provide a significant noninflationary source of financing for the budget.

However, nonbank financing has remained marginal. Table 5 shows that there has been none in most countries. 1/ Until 1994, all of domestic financing outside the banking sector consisted of extraordinary items such as gold sales. Russia began to issue government paper in 1994 and Ukraine in 1995. However, Russia’s first issues were offered at below market rates, and used to cover tax liabilities. Likewise, a 1994 Kazak issue was the securitized deferral of government obligations on account of interenterprise arrears, and represented to some extent involuntary financing.

Table 5.

FSU: Domestic Financing of the Budget, 1992–95

article image
Source: Data provided by the authorities and IMF staff estimates.

So far, both government demand for nonbank financing and domestic supply have remained muted. On the demand side, governments which are still receiving credits at negative real rates from captive central banks have been reluctant to seek financing at the high interest rates that would be necessary to attract credit markets. Since FSU central banks do not consistently transfer profits (as defined according to some economic measure) to government, the budget can make a net gain by borrowing at zero interest from the central bank. On the supply side, dollars and consumer durables have until recently been perceived as a more trustworthy inflation hedge than domestically-denominated government bonds. Moreover, real wages have remained low, and the share of consumption in wages has grown, so the fount of private savings to be tapped remains limited.

b. The erosion of the inflation tax

Lack of alternative financing sources has led governments to rely heavily on the inflation tax. Bank financing of the budget amounted to 11 percent of GDP in 1992--more than 80 percent of the whole (unweighted average) FSU deficit (see Table 5). By 1995 it had dropped to an average of less than 2 percent of GDP, 40 percent of the FSU deficit and slightly lower than average seignorage for all developing countries in 1990–95. 2/ But six countries, including Russia, continued to finance more than two-thirds of their deficit through the banking system in 1995.

The decline in banking system finance was to some extent imposed by broader external factors. Real money demand had fallen drastically throughout the FSU in the preceding high-inflation years, and hence the scope for financing through the inflation tax had been significantly eroded. Table 6 shows that real money balances dropped on average by 90 percent between 1991 and 1994 and were almost wiped out in Armenia, Georgia, Kazakstan and Uzbekistan. Bank credit was able to remain the main source of deficit financing only because of the sharp retrenchment in the overall size of the fiscal gap. In other words, a major factor underlying the decline in FSU deficits has had to have been governments’ inability to finance higher imbalances as the inflation tax base eroded and alternative sources of funds failed to materialize.

Table 6.

FSU: Decline in the Real Domestic Money Stock, 1991–94

article image
Source: Citrin, et al. (1995, page 90).

c. Disorderly financing through government arrears

The forced adjustment was difficult to accept for FSU electorates. Governments’ frequent response was to continue to promise unrealistically generous budgets, while recognizing the inevitability of resorting to cash rationing when financing was exhausted. This “brute-force” method of expenditure control is discussed in the next section. The application of various types of sequestration and cash rationing led to sizeable government expenditure arrears. To the extent that arrears exist, the decline in FSU deficits is less impressive than suggested by the figures, and “financing” has been higher by the amount of the arrears.

Few consistent figures for arrears are available. In general, expenditure arrears are difficult to measure and prone to differing interpretations; FSU governments record spending only on a cash basis. 1/ Documented figures for spending arrears (which are likely to be partial) include: arrears of 6.3 percent of GDP in Armenia and 8.7 percent of GDP in Moldova in 1994; and in 1995, of 2 percent of GDP in Azerbaijan, 10.3 percent of GDP in Tajikistan, and 0.9 percent of GDP in Ukraine. These numbers suggest that the decline in recorded cash deficits significantly overstates the contraction of the policy gap between spending and revenues.

Arrears, too, tend to be an unsustainable method of financing. In Russia, popular discontent against the nonpayment of wages and benefits is said to have been the main motivation for the vote which unseated the reform government and returned the Communists to power in Parliament in December 1995. 2/

IV. Methods of Deficit Reduction and their Costs

The containment of the deficit has been surprisingly effective, given the pressures on the budget. Since revenues have proved so disappointing, it has been achieved entirely through expenditure control.

Expenditure in FSU countries has dropped from an average 45 percent of GDP in 1992 to 29 percent in 1995 (Table 7). All countries have made cuts, but experience has varied, perhaps less predictably than on the revenue side. Fast reformers like the Baltics are coupled with slow reformers like Belarus and Uzbekistan in evincing relatively small expenditure cuts--if any. 3/ The Caucasus and Tajikistan have had to adjust by extreme amounts, with cuts of 46 percentage points in Tajikistan, 39 percent in Georgia, 35 percent in Armenia, and 21 percent in Azerbaijan.

Table 7.

FSU: The Evolution of General Government Expenditure, 1992–95

article image
Source: Data provided by the authorities and IMF staff estimates.

Figures for 1991 and 1992 are adjusted to take account of a revision in the Kyrgyz national accounts from 1993 on.

But, in many instances, rather than taking stock of their spending priorities and selecting programs for elimination or down-sizing, governments have found it politically easier to compress spending by under-budgeting, across-the-board freezes, sequestration, and cash rationing. Western budget practices for cutting expenditure--the elimination of low-priority programs, reduction in civil service numbers, increases in cost-recovery, or scaling-back of entitlements--have been absent or played down in the policy strategy presented in budget documents. 4/ As illustrated in the previous section, the unorthodox restrictions have had the nearly inevitable consequence of generating arrears in wages, social benefits and pensions, utility bills and payments to suppliers. There is increasing evidence that some of the cutbacks are unsustainable--in the sense that expenditures have been postponed or sequestered rather than eliminated--and that the way they have been carried out is eroding political and social stability. Moreover, the disorderly nature of the cuts has caused the composition of spending to diverge from what was seen as appropriate for the transition. This section discusses the unorthodox tools used to achieve the deficit cuts in FSU countries over the last five years, and their implications.

1. Unorthodox means of deficit control

Several methods of compressing expenditure without structurally reducing the budget have been applied.

a. Under-budgeting

Under-budgeting has taken the form primarily of constructing the budget around an inflation prediction lower than was forecast to occur. For instance, in 1995, the Russian budget assumed an inflation rate which was one-third lower than the inflation target built into monetary policy. This implies that all programs, wages, pensions and other benefits were under-funded compared with what they were expected to cost. Full payment of some (as provided for by the sequestration mechanism--see below) meant that resources were inadequate to pay the others, although they possessed approved budget lines and were therefore in the process of implementation. Another facet of under-budgeting observed across the FSU has been a tendency to over-estimate revenues, to permit higher expenditure allocations for a given deficit. Failure of the revenue to materialize has then required expenditure to be contained through cash rationing. 1/

The problems caused by under-budgeting have been compounded by layers of conflicting legislation. Frequently, labor and social protection laws grant entitlements to benefits higher than are provided for in the budget law--for instance, by using different indexation mechanisms. The budget law in most if not all countries overrides the labor laws, by a clause stating that all entitlements will be fulfilled only if the government has the means to pay. The existence of conflicting laws explains the confusion about how large arrears actually are: the numbers can differ significantly depending on whether the social laws or the budget law are used as the benchmark. Understandably, workers and pensioners tend to evaluate government performance with reference to the more favorable laws. Since the social laws to a large extent embody the negotiations between unions and government, failure to observe them creates a transparent breach of the social contract. The breach affects individuals usually by reducing their income--arguably the most direct means of increasing the distrust of the public in its government.

If revenue is lower than projected by the budget, countries apply various means of sequestration and cash rationing. These are similar forms of delaying payment. The Russian sequestration system prioritizes types of payments ex ante, which are then paid in order, as resources arrive. The Ukrainian cash rationing system allows for more discretion (but therefore less prior predictability), by having high-ranking officials choose among payments on a day-to-day basis, as requests are submitted. If the budget and the macroeconomic framework were realistic, sequestration and cash rationing could lead to only within-year arrears: mild forms of payment postponement and creative cash management are applied by many governments. The problem becomes a social threat only when resources do not catch up with spending commitments by the end of the fiscal year, so that nonpayments become an observed default by government. Besides any political problems they may create, such defaults erode the credit rating of the government.

2. The costs of expenditure compression

In the short run, the budgeting methods described above are attractive to both the administration and politicians. Under-budgeting and sequestration allow governments to postpone unpopularity. Parliaments do not have to choose between programs, but can be seen to vote in favor of alt of them. Cash rationing allows governments to stay within stabilization program deficit limits despite an inconsistent budget. Only in the longer term, through the type of electoral backlash seen in Russia, do the consequences become evident.

a. Political and social costs

Russia is a well-publicized example of the political implications of expenditure control by sequestration. General government expenditure has been cut from 60 percent of GDP in 1992 to 32 percent in 1995, a remarkable contraction of 28 percent of GDP. 1/ Sequestration continues to be the main policy instrument to keep spending in line with revenues and available financing. Because of the severity of Russia’s resource shortfall, sequestered expenditures have been large. At the central government level only (which accounts for something less than half of total outlays), they amounted to an estimated 3 percent of GDP in 1995, 2/ Besides sequestered expenditures (which are not defined as arrears), identified central government spending, arrears on wages were reported to amount to 3/4 percent of GDP at end-1995, and central government arrears to the Pension Fund were one-third of the total payment for 1995 as mandated by pension laws. These numbers are not surprising considering that expenditure was cut by 12 percent of GDP in 1995 (Table 7). Media observers considered the shortfalls to be the most tangible reason for the resurgence of the Communist vote in 1995.

b. Macroeconomic spill-overs

The problems created by under-budgeting and sequestration also have adverse macroeconomic effects. Two tendencies have been observed in the fiscal and monetary statistics of several countries--both attempts by affected parties to protect themselves from unpredictable government payments.

The first is that regions begin to withhold their revenues from the central government, to cushion themselves against nonreceipt of promised central transfers. This leads to a contentious and uneven de facto decentralization of government; and, because the withholding tends to be competitive, it is likely to generate a downward spiral of central government control. The redistributive function of central government loses its power, and since this is an important cohesive force, political union-begins to erode. Arguably, this is already being observed in Russia, where the share of revenue controlled by the central government has dropped from 73 percent in 1992 to an estimated 40 percent in 1995.

Second, government spending agencies build up precautionary balances to give themselves more control over their outlays. Agencies which are paid for low-priority items (because they request it on a “slow day” at the cash rationing center) hoard the funds rather than using them as allocated, in case higher-priority payment obligations are returned without funding later. 1/ The hoarding leads to a loss of control by the Ministry of Finance over spending ministries, and also an erosion of the central authorities’ ability to control monetary policy. The last happens because the precautionary balances manifest themselves as a build-up of government deposits, but at a decentralized agency level. 2/ The deposits can be run down by agencies if the central authorities try to tighten monetary policy by restricting their access to credit--thus rendering monetary policy impotent. A common feature of FSU countries since 1992 has been a tight first half-year (as funds were hoarded), followed by a profligate second half (as agencies began to run down their balances before the end of the budget year).

There is no easy solution to the evident erosion of government control. The only remedy is a return to orthodox budgeting. This would require two steps: the design and ratification by parliaments of a transparent, comprehensive, and realistic budget; and the elimination of sequestration and cash rationing, so that subnational governments, regions, and voters regain trust that expenditures, if budgeted, will be financed.

3. Delays in realigning expenditure priorities

Besides the problems outlined above, expenditure compression has tended to delay the expenditure restructuring towards market-economy priorities that will be a benchmark for the success of the transition. The shift to market would be expected to imply a significant reduction in government intervention in the economy, and a change in the composition of expenditure towards provision of public goods and outlays connected with restructuring. An adequate social safety net would also be needed. As discussed below, progress in these areas has been unimpressive so far.

Table 8 compares expenditure in the closing days of the USSR with outlays in the sample of its successor states for which a comparable breakdown of data could be constructed for 1994. 1/ Kazakstan, Moldova and Russia are generally considered relatively fast reformers, while Tajikistan, Ukraine and Uzbekistan have been slow to reform. Some broad conclusions may be drawn.

Table 8.

The Composition of General Government Expenditure in 1994 A Comparison of the USSR and Some of its Successor States

article image
Source: JSSE (1991) for the USSR; IBRD (1995b) for Russia; IMF for others.

Includes expenditure from social funds. In countries where some of these are now managed outside the budget, total expenditure may differ from that shown in Table 7.

Includes expenditure “on the national economy”, budgetary loans to state enterprises (other than in the context of a restructuring program), and other coverage of state enterprise losses by government. Where applicable, numbers are adjusted to exclude road funds and the military conversion program.

The decline in the size of government compared with the USSR is significant only in Kazakstan and Moldova. In the slow reformers, and in Russia, government expenditure remains close to Soviet levels--though the 1995 data for Russia in Table 7 suggest that a decline has now begun.

In the same vein, apart from in Moldova, the decline of budgetary resources devoted to government intervention in the economy has been minor. In the USSR, support for the national economy was the largest spending item, amounting to 14 percent of GDP in 1989. This category includes production subsidies (price supports, direct government procurement, coverage of operating losses and generalized sectoral spending (such as on irrigation and nuclear power development)), and therefore constitutes the main difference between government spending in planned economies and industrial countries.

Despite the widescale price liberalization since 1992, the countries in the sample have made little progress in reducing their budget bill for production support. 2/ In Russia, besides the 12 1/2 percent of GDP spent on direct support for the economy in 1994, enterprises received a further 3–4 percent in tax exemptions (IBRD, 1995a, 1995b). In Ukraine, direct support of 13 1/2 percent of GDP was largely devoted to subsidizing producer prices for agricultural goods through direct government procurement (7 percent of GDP) and to support for the coal sector (3 3/4 percent of GDP). 1/

Social spending has become very expensive. It pre-empts a higher share of GDP than in the USSR in all countries other than Kazakstan, and even there it accounts for a significantly higher share of total expenditure. It has always been recognized that social safety net costs would rise in the transition. However, there are several suboptimal aspects to the high spending.

- Spending on education and health 2/ has been unexpectedly buoyant, growing faster than direct social protection in countries other than Russia and Moldova. In most countries, employment has increased in these sectors and relative wages have remained stable or risen (Horton, 1996). This is contrary to expectations, which were that wages--and in consequence, employment--in health and education would fall relative to wages in the rest of the economy as the market sector expanded. Some decline was seen as appropriate, because in many instances teacher-student ratios and other service indices are higher than in industrial countries. But in at least some FSU states, these wages are linked by law to industrial wages, and, partly because of the political importance of teachers and medics, the link has proved difficult to break. Moreover, jobs in these sectors are often linked to housing privileges, etc., and to some extent represent disguised social protection.

- Unemployment benefits have remained unimportant. High unemployment (as long as it was temporary) was seen as a necessary element of the transition because labor mobility was seen as key to rapid sectoral restructuring. Hence, unemployment benefits that protected living standards were considered an important instrument for drawing workers away from declining sectors and giving them search time to find jobs in the private sector. This strategy has not worked. Open unemployment remains lower than predicted; in Russia, for instance, the registered unemployed amount to only 3.3 percent of the labor force. In consequence, the outlays of employment funds (only part of which are unemployment benefits), have remained a minor component of the social safety net, an average of around 1 percent of GDP in 1994.

Several reasons have been identified for the surprisingly slow growth of unemployment. To summarize: (i) enterprises have been willing to keep unproductive workers on their books for social reasons; (ii) workers have been reluctant to leave, because the wage package has turned out to be only part of their total compensation, with tied housing and social services also significant; and (iii) unemployment benefits, in reality, have remained low, difficult to qualify for, and undependable. In particular, in most countries they have been tied to the minimum wage, which has been adjusted much less frequently than inflation.

- More generally, cash transfers have remained a small share of social expenditure. The goal of social safety net designers was to replace subsidies by a consolidated cash transfer to individuals affected by various relative price shifts; this could then gradually be phased out as wage levels adjusted. While some cash transfers have been put in place, the model envisaged by the designers ran into unforeseen difficulties. For one thing, since various types of social benefits are delivered by different agencies (the pension fund, trade unions, the workplace), effective means-testing has proven impossible and little effective consolidation of benefits has taken place. For another, governments have been reluctant to switch from some subsidies to cash transfers, because they bear only part of the costs of the subsidies, with producers bearing a further share. If producers were to eliminate their subsidies pari passu with government, the demand for cash compensation from the budget could be higher than the current budgetary cost of the subsidy.

- Partly for the reasons discussed, the share of consumer subsidies in the social safety net remains significant in countries where information is available. 1/ These are primarily related to costs of heating and housing (“communal services”). In Ukraine, these amounted to 3.3 percent of GDP in 1994, and in Moldova and Uzbekistan to around 1 percent of GDP. In Russia, heating subsidies are covered by the energy sector rather than the budget (and are not included in Table 8), but are estimated to be higher than in Ukraine. Since increases in cost recovery have been difficult, and since energy price liberalization is still continuing in most countries, these subsidies have tended to rise since 1994. In Ukraine, subsidies were budgeted at 5.3 percent of GDP for 1995. 2/

- Besides weighing on the budget, the persistence of housing and household energy subsidies has had adverse consequences at a macroeconomic level: (i) households have been reluctant to lose access to subsidies, which has slowed housing privatization and the development of a housing market--in turn hindering labor mobility and sectoral adjustment; and (ii) utility companies pay part of communal services subsidies, through under-pricing and cross-subsidization of their services, and payment arrears from households and government. The underpayments have decapitalized them, and they themselves have run up arrears--notably external arrears with Russia and Turkmenistan. External energy arrears, which governments largely see as beyond their control (and which seem to the casual observer to be far removed from the problem of housing charges), have become a serious obstacle to the normalization of FSU trade and financing relations.

The most striking adjustment, in all countries, has taken place through cutbacks in defense spending, which has been nearly halved in Russia compared with Soviet outlays, and has fallen further, to around 2 percent of GDP or below in most other countries. 1/ 2/ The downsizing of defense spending is even more impressive than the net changes suggest, because it has taken place despite factors which would have tended to put upward pressure on military outlays since 1992. These include: (i) expensive regional conflicts; (ii) the expected disproportionate impact of price liberalization in the defense sector, because prices there were significantly more repressed than in other parts of the economy; 3/ (iii) the dismantling of strategic stockpiles (“state reserves”), which removed an important source of commercial revenue from the military; 1/ and (iv) the military conversion effort. In effect, government spending on conversion and weapons dismantling, expected to be a central restructuring burden for transition governments, has been minimal in the countries where data are available. For instance, in Russia it was 0.1 percent of GDP in 1994. In Ukraine, it amounted to 0.4 percent of GDP in 1994 and was cut to 0.2 percent of GDP in the 1995 budget. The reasons for the lack of budgetary attention to conversion are not clear. Some FSU governments see the problem as one of shortfalls in foreign financing for conversion compared with what they expected at the outset of the transition. The spending that did take place was in many cases seen as wasteful, given budgetary exigencies. It is also likely that much of the conversion has been self-financing, particularly in industries being privatized.

Finally, spending on investment and restructuring has been low. Government capital formation is difficult to measure given the dispersed sectoral allocation of investment responsibilities within government and possible overlap with the intervention discussed above. Nonetheless, it has clearly declined markedly. Soviet budgeted capital spending was 7 percent of GDP in 1989. 2/ In 1994, comparably classified capital outlays in Kazakstan were only 0.2 percent of GDP, in Moldova and Ukraine around 2 percent of GDP, and in Uzbekistan around 4 percent of GDP. Only in Tajikistan did government investment in 1994 exceed Soviet levels. Table 9 provides evidence of the breadth of the decline from a broader range of countries. Capital spending has been one of the main victims of ad hoc budgeting. This is partly because of the significant savings to be had from delaying new projects, but also partly because many projects are reimbursed only upon completion--giving government an easy target for delaying payment. Arguably, some cutbacks in capital spending are appropriate, at a time of budget tightening, unmet social needs, and continuing volatility in relative prices--implying that good investments might be hard to identify. On the other hand, in light of the diagnosis that an obsolete capital stock was a main cause of the Soviet Union’s demise (Fischer and Easterly, 1993), and that inadequate linkages were another (Arrow and Phelps, 1990), significant inputs of new infrastructural investment were seen as playing a central role in the recovery of output. It is clear that these have been deferred.

Table 9.

FSU: Expenditure on Capital and Restructuring by General Government, 1994 1/

article image
Source: Data provided by the authorities and IMF staff estimates.

The coverage of these figures may differ from country to country.

V. Conclusions

This paper has examined the progress of fiscal control and restructuring in FSU countries since the dissolution of the Soviet Union. The record is deeply uneven. Control appears to have been captured almost everywhere, in the sense that deficits--which ballooned after price liberalization--have come down rapidly. The tightening was somewhat exogenously enforced, because financing sources quickly became inelastic. It was achieved by sharp and often inappropriate cuts in public expenditure, which became necessary to achieve stabilization in the face of an unanticipatedly steep revenue decline. Because the cuts in public spending have been done abruptly and without a focus on transition goals, and because the instruments used to ensure them vitiated the normal budgetary process, benchmarks of fiscal success other than stabilization have been elusive. Government intervention in the economy remains important and subsidies have not been eliminated. Social spending is not efficient, and there is little evidence in the budget of restructuring. Most worryingly, the unorthodox budget control mechanisms appear to have begun to elicit some social and political backlash. It is clear that the need to realign government expenditure priorities and establish orderly budgeting procedures is now the most pressing fiscal requirement in the region of the former Soviet Union.

Same Series

Other IMF Content

Other Publishers

Cover IMF Working Papers
Fiscal Transition in Countries of the Former Soviet Union: An Interim Assessment
Author: Mr. Jeffrey M. Davis
Volume/Issue:
Volume 1996: Issue 061
Publisher:
International Monetary Fund
ISBN:
9781451848120
ISSN:
1018-5941
Pages:
36
DOI:
https://doi.org/10.5089/9781451848120.001