IV Public Expenditure Reform

Sergei Alexashenko, and Augusto López-Claros
Published Date:
April 1998
  • ShareShare
Show Summary Details

As part of its gradual move toward establishing a market economy based on the rule of law, Russia has scaled down significantly the size and scope of the public sector and has redefined the state’s role as primary producer, allocator, and distributor in the economy. The period 1992–96 has witnessed a sustained reduction in the expenditure to GDP ratio that—for the consolidated government—fell from some 67 percent in 1992 to 38 percent in 1994 and to some 32 percent in 1996.27 The reduction in federal expenditures has been especially pronounced, from 56 percent of GDP in 1992 to 24 percent in 1994 and to some 18½ percent in 1995–96, thus accounting for all of the total contraction (Table 7). On the whole, by 1996, expenditures at the federal level were dominated by debt service, defense and security, and various “protected” items with relatively limited discretion, while expenditures at the regional level were mainly accounted for by housing and other communal services, health, and education.

Table 7.Government Expenditure(In percent of GDP)
Expenditures at federal level155.927. 824. 317. 619. 8
National economy26.
Public health0.
Culture, arts, and mass media0.
Social protection0.
Law enforcement1.
Intergovernmental transfers1.
Interest payments0.
Net lending33.
Unbudgeted import subsidies11.92.3
Transfers to CIS states8.52.0
Central bank directed credits415.55.02.3
Working capital transfers3.3
Expenditures at regional level13.017.017.514.514.7
National economy55.
Public health2.
Culture, arts, and mass media0.
Social protection60.
Law enforcement0.
Net lending0.
Consolidated expenditures767.142.138.330.632.2
Memorandum item:
GDP (in trillions of rubles)181726111,6302,256
Sources: Ministry of Finance; World Bank; and IMF staff estimates.

Including unbudgeted import subsidies, central bank directed credits and working capital transfers, but excluding transfers to other CIS states.

Includes subsidies to industry, energy, construction, agriculture, fishing, compensation for “price differences,” capital investments, maintenance, among others.

Credits for investment and industrial reconversion and other budgetary loans.

To agriculture, fuel and energy, industry, the Northern Territories, among others.

Includes subsidies for housing and utilities.

Includes children’s allowances.

Net of intergovernmental transfers.

Sources: Ministry of Finance; World Bank; and IMF staff estimates.

Including unbudgeted import subsidies, central bank directed credits and working capital transfers, but excluding transfers to other CIS states.

Includes subsidies to industry, energy, construction, agriculture, fishing, compensation for “price differences,” capital investments, maintenance, among others.

Credits for investment and industrial reconversion and other budgetary loans.

To agriculture, fuel and energy, industry, the Northern Territories, among others.

Includes subsidies for housing and utilities.

Includes children’s allowances.

Net of intergovernmental transfers.

Coverage and Classification Issues

Analysis of the evolution of government operations is complicated by a number of changes in the structure of the budget that have significantly affected the coverage of expenditures. Official budget execution data for 1992 shows consolidated expenditures of some 33 percent of GDP and a deficit of just under 4 percent of GDP, figures that would appear at first sight to suggest both the relatively small size of the public sector in Russia in relation to that of other countries in Central and Eastern Europe and the restrained stance of fiscal policy. In reality, it reflects the fact that in the early stages of the transition many large operations were simply not included in the budget. During 1993–95 a number of quasi-fiscal and credit operations previously off-budget were gradually incorporated into the budget.

Until mid-1993, the official budget did not fully account for foreign exchange revenues and expenditures. On the expenditure side, in particular, outlays on foreign currency debt (including to residents), centralized imports, and various foreign exchange allocations to ministries and other agencies were incorporated in the budget only insofar as they were financed by ruble allocations. Other expenditures, such as those financed by sales of gold and precious metals or foreign credits, were also not covered in the budget. Although attempts were made to include these expenditures (the inclusion of the equivalent of 12 percentage points of GDP in import subsidies in the expenditure data for 1992 is a good example of this), the coverage remained incomplete mainly because of the lack of systematic and comprehensive accounting of such operations. For instance, because of inadequate information, large balances on escrow accounts abroad to pay for centralized imports and financed by a share of the foreign exchange surrendered by exporters and certain export taxes were not included in more comprehensive versions of the government’s operations.

A range of quasi-fiscal or net lending operations, some guaranteed by the Ministry of Finance, others in the form of directed credits by the central bank to the enterprise sector and which had not been included in the budget in 1992–93, gradually began to be incorporated in subsequent years as the scope of the underlying operations was reduced significantly. Such examples include (1) credits for the indexation of working capital (of the order of 3 percent of GDP in 1992); (2) credits to the Commonwealth of Independent States (CIS), mainly in the form of central bank correspondent accounts and technical credits amounting to over 8 percent of GDP in 1992; (3) credits to agriculture, to support planting and sowing in the spring and harvesting later in the year, a share of which began to be included in the budget as loans in mid-1994; (4) credits to the Northern Territories, incorporated in 1995; and (5) credits to support electric power generation in the Far East and purchases of fuel in the winter, among many others.

Various extrabudgetary funds were created at the federal and regional levels, particularly during 1991–92. In addition to the social funds (see Section V), various industrial and sectoral funds were also established, financed by a share of production costs and with the aim of furthering “investment and research and development.” While no information is available on the number of these funds at the regional and local levels, it is estimated that at the federal level there were at least 50 in operation, collecting about 2–3 percent of GDP in revenue, all of it deductible from the profit tax base.

Apart from the issues of the appropriate coverage of budgetary expenditures, certain aspects of the classification of expenditures further complicate analysis of trends and sharply limit the usefulness of budget expenditure data. Budgetary expenditures are classified according to functional rather than economic criteria. The main categories of expenditure still reflect the main sectors of the economy as envisaged in the earlier national plans—such as sociocultural activities, defense, science, and national economy—with each category including both current and capital outlays (not separately identified) and sometimes consolidating large resources in a highly aggregated fashion. Although a new system of budgetary classification was introduced with the 1995 budget, which lists expenditures by government function, economic characteristics, and spending units, no economic classification for the 1995 outturn is yet available.28


As shown in Table 8, subsidies (budgeted and un-budgeted) during 1992 amounted to about 26 percent of GDP and consisted mainly of subsidies to agriculture, the coal industry, military conversion projects, interest payments, and imports, as well as some other amounts provided through local budgets. The most important of these were import subsidies associated with purchases of commodities through the budget under the centralized imports scheme and those linked to tied foreign credits involving the resale by the government of commodities (grain, spare parts, medicines, and processed food) financed by external loans at prices that implied a large subsidy element. The latter type were not included in the budget; the combination of the two was equivalent to 15 percent of GDP in 1992, with the bulk (12 percent of GDP) corresponding to the unbudgeted type. The rate of effective subsidization for various commodities (given by the difference between the market price implied by the prevailing exchange rate and the ruble counterpart actually collected from the receiving enterprise) varied over time but was generally in the range of 90–100 percent. An additional 11 percent of GDP was provided through various consumer and producer subsidies. Consumer subsidies took the form of price subsidies for a number of food items, medicines, heating, and rent and transportation, with the latter provided through local budgets. Price liberalization notwithstanding, local governments continued to monitor closely the prices of essential food items and provided direct consumer subsidies, as needed and as dictated by the availability of resources and local prerogatives. This practice, to a greater or lesser degree, remained in force for the next several years in a large number of regions. De facto then, during 1992, there was a significant shift in expenditure responsibilities to the local level for the subsidization of essential items. In the aggregate, regional housing and utilities subsidies amounted to 3–4 percent of GDP in 1995–96.

Table 8.Subsidies to the Economy in 1992(In billions of rubles)
1. Nonimport subsidies1,384
(In percent of GDP)(7.7)
Coal industry180
Of which:
Farmers’ Fund54
Producer subsidies141
Military conversion130
Local budgets585
Interest subsidies1151
Other subsidies30
2. Import-related subsidies2,721
(In percent of GDP)15.0
3. Other subsidies650
4. Total subsidies (1 +2+3)4,755
(In percent of GDP)(26.3)
Coal subsidy1.0
Military conversion0.7
Total interest subsidy20.6
From local budgets3.2
Total import subsidies15.0
Other subsidies3.2
GDP (in trillions of rubles)18.1
Sources: Ministry of Finance; and IMF staff estimates.

To Rosselkhosbank and Northern Territories.

Through the Farmers’ Fund and commercial banks and excluding quasi-fiscal operations associated with interest payments on directed credits.

Sources: Ministry of Finance; and IMF staff estimates.

To Rosselkhosbank and Northern Territories.

Through the Farmers’ Fund and commercial banks and excluding quasi-fiscal operations associated with interest payments on directed credits.

Producer subsidies were mainly provided to agriculture and the coal industry. Subsidies to agriculture took the form of allocations for the improvement of agricultural land, for livestock production, to compensate for “high” energy costs, for housing construction, and to offset the “high” cost of borrowing, the latter typically provided so as to reduce the effective interest rate to a fraction of the central bank refinance rate. Those to the coal industry consisted of wage subsidies and direct subsidies to finance social expenditures. In addition, industrial and agricultural enterprises received interest rate subsidies on centralbank-directed credits amounting to some 3 percent of GDP, to cover the difference between the central bank refinance rate and the interest paid by enterprises to commercial banks. In 1992, directed credits amounted to some 19 percent of GDP, both from the central bank and the government, the latter through working capital injections equivalent to some 3½ percent of GDP. Since the bulk of these directed credits (and additional amounts disbursed in 1993–94) are not expected to be repaid, it may be more appropriate to think of them as grants, given to finance the provision of social benefits and services, to continue in an indirect way to provide some degree of consumer subsidization, and to finance capital flight.29

Much progress was made during the transition period in eliminating the bulk of such subsidies through the emergence of a more transparent system of resource allocation. As price liberalization led to a more rational structure of signals and incentives in the economy, the extent of the prevailing distortions became glaringly evident. By 1993 and in the context of a unified exchange rate, import subsidies had been sharply curtailed, amounting to no more than 2½ percent of GDP for the year as a whole; indeed, the reduction in the fiscal deficit that year is mainly accounted for by a drop of nearly 10 percentage points of GDP in unbudgeted subsidies. Progress in eliminating other subsidies was considerably slower but, against the magnitude of the underlying distortions, made some headway. The move to a market-based system of interest rate determination allowed for a more transparent accounting of the interest cost associated with subsidizing activities in the enterprise sector, particularly in agriculture. While interest rate subsidies continued to be provided through the period under review, these were increasingly limited in scale and aimed at a relatively small number of activities, mainly in the agricultural sector. For instance, the 1995 budget identified four major areas as recipients of budgetary loans at a fraction of the central bank refinance rate: investment programs and defense industry conversion (one-fourth of the central bank refinance rate); supplies to the Northern Territories (one-third); purchases of agricultural products for the Federal Food Funds (one-third); and the Federal Fund for Fuel Procurement (one-third). It should be noted, however, that although the bulk of new lending to agriculture was at market rates, a large share of the loans was rolled over and little interest was actually collected.30

The difficulties in compressing expenditure significantly beyond the levels implied by the elimination of subsidies and other obviously inefficient expenditure items have been apparent in other economies in transition; indeed, in some of these countries (such as Poland and Hungary) the total expenditure to GDP ratios were higher in 1992, two years after the onset of the transition, than in 1990. It is interesting to compare the evolution of selected components of expenditure (especially current expenditure) in several of the transition economies in Eastern Europe with those in Russia. As Figure 6 shows, spending on wages and salaries and interest payments rose significantly relative to pretransition levels. The increase over the four-year period beginning one year before the implementation of the authorities’ most comprehensive economic reform program and ending two years later shows an average combined rise in these two components of expenditure of some 4–5 percentage points of GDP. The rise mainly reflects liberalization of interest rate policy and the subsequent emergence of positive real interest rates, together with growing borrowing from domestic and foreign financial markets, and wage policies intended to prevent a massive shift of qualified personnel to the rapidly growing private sector, as well as the monetization of in-kind benefits.31 In Russia, as in other countries, interest payments in relation to GDP also rose in response to the liberalization of interest rates and the move away from subsidized directed central bank credits toward market-related debt instruments (see the section Budget Financing below).

Figure 6.Selected Countries: Government Expenditure by Components1

(In percent of GDP)

Source: IMF staff estimates.

1 Starting year in chart corresponds to first year of “transition.”

2 Social expenditure in Russia defined as total expenditures of the four main funds (Pension, Social Insurance, Employment, and Medical) as well as those administered through the Ministry of Labor and Social Development.

Producer subsidies fell sharply in all transition countries and, in general, the larger the drop the more pronounced was the corresponding fall in industrial output, since the bulk of these subsidies was allocated to industrial enterprises. As noted earlier, cutbacks in producer subsidies adversely affected the financial position of enterprises and contributed to less ambitious investment plans and production, layoffs, and tax and payments arrears. Purchases of goods and services and capital spending fell across all of these countries. The declines were especially pronounced in Bulgaria and Romania (not shown in Figure 6; in Romania capital expenditure fell from 18 percent of GDP in 1989 to 6 percent in 1992).

Expenditures in the form of income transfers (consumer subsidies plus social expenditures) went up in all transition countries, with the fall in subsidies being more than offset by the rise in expenditures; the increases in social expenditures in Poland in 1990–92 were particularly high (9 percentage points of GDP), mainly in the form of increases in pensions and unemployment benefits. These increases were intended to offset declines in real wages and in income from consumer subsidies following price liberalization. In 1992–96, transition countries made limited progress in improving the efficiency of social spending. In contrast, in Russia, as noted earlier, consumer subsidies were reduced significantly and social expenditures fell simultaneously. In relation to GDP, social expenditures carried out by the social funds (pensions, unemployment compensation, and social benefits provided through the Social Insurance Fund, among others) fell by some 3 percentage points of GDP between 1992 and 1996.32

Scope for Additional Expenditure Compression

One argument often made in Russia about the limited scope for a further reduction in the size and the functions of the public sector is that, once the most apparent inefficiencies and distortions are eliminated, particularly subsidies, areas will remain where significant spending may be necessary. A number of observations can be made in this regard. First, while it is true that many of the institutions of the centralized economy have been eliminated, such as planning and price offices, branch ministries, and so on, new ones have been created or will need to be created to oversee previously nonexisting activities, such as privatization, bank regulation and supervision, and tax policing. Others will need to be strengthened considerably to be able to deal effectively with expanded functions and responsibilities in the context of an emerging market economy. A key example would be the State Tax Service, given the large increase in the number of taxpayers. Likewise, while some of the functions of state security agencies have been phased out in the context of building up democratic institutions in Russia and greater awareness of the importance of basic human rights, there are important needs in the fight against a rising tide of crime, which is a fundamental concern of the population. There will be growing needs in the judicial area as well, as efforts continue to establish the rule of law and, as a result, more cases are settled through the courts. It may not be feasible nor desirable, therefore, to compress further expenditures that directly finance essential functions of the state in a modern economy (see Table 9).33 Furthermore, pressures for expenditures in other areas will also continue. Some of the more important of these are examined below.

Table 9.Selected Countries: General Government Expenditures(In percent of GDP)
United Kingdom44.040.743.143.543.143.241.9
United States32.933.434.433.933.033.233.3
European Union countries49.048.850.451.950.850.249.8
OECD countries38.939.340.541.140.340.340.3
Source: Organization for Economic Cooperation and Development, Economic Outlook, June 1997.
Source: Organization for Economic Cooperation and Development, Economic Outlook, June 1997.

Interest payments. In a context of high real interest rates and still relatively short maturities, payment pressures will continue, given the accumulation of public debt associated with budget deficits and the increasing use of market-based debt instruments to finance budgetary shortfalls. There are two aspects to this. The first is the level of interest payments (including on external debt), which have grown rapidly in recent years to more than 5 percent of GDP in 1996. The other pertains to the need to nurture growing confidence in Russia’s financial system against a backdrop of disruptions such as confiscatory “monetary reforms” (for example, under the Pavlov government, 1991), bankruptcies (Vneshekonombank, affecting over $10 billion of foreign currency deposits of residents, 1991), and old/new ruble swaps (1993), all of which undermined public trust and contributed to the high premiums being demanded for the holding of treasury bills and other debt instruments. Yet another aspect of this is the budgetary cost of bank recapitalization through government bond issues. Since a share of the loan portfolio of Russia’s banks may be nonperforming, some systemwide solution may be warranted that could entail significant yearly interest costs.

Wages. Wages will come under pressure as various hidden subsidies at the enterprise level continue to be eliminated (housing, some forms of social protection, and so on) and also to narrow the gap with the private sector and, more generally, with Russia’s main trade partners. The average monthly wage in the public sector in 1994–95 was equivalent to some 50–75 percent of the average wage for the economy.34

Capital spending. To stem the further deterioration of physical infrastructure and of the health and education systems, both of which have come under heavy strain during the transition, capital spending will rise. Neither of these categories of expenditure can be neglected, given the development lessons of the past two decades about the future economic costs of inattention to them. It is by now well accepted that the economic returns to investment in education and health “are often extremely high” and that the economy’s future growth potential will be undermined by the prolonged disregard of these two sectors.35 In addition, experience in other countries has shown that a deteriorated transport and communications infrastructure can be an important deterrent to foreign direct investment. Significant environmental cleanup will also be necessary, given the unfortunate legacy of disregard for Russia’s habitat. This is particularly the case as regards the safe disposal of nuclear waste (of which Russia has a large world share) and the deactivation of a number of nuclear power facilities, many of which are associated with the military-industrial complex. There are justifiable concerns that inattention to these issues poses considerable risks for the environment, in some cases on a regional scale. There is also broad consensus that such cleanup is extremely costly and must be carried out over a long-term horizon.

Industrial reconversion. Some form of industrial reconversion will continue, particularly in the military. Industrial activity in Russia has contracted sharply in the past several years (a cumulative drop in industrial production of 53 percent over 1991–96 and affecting every sector from electric power generation—21 percent—to light manufactures—87 percent), owing to a broad range of factors. These include excess capacity in the military-industrial complex that had evolved in the past on the basis of certain assumptions about the likely evolution of the international political climate, assumptions that proved to be wrong; inefficient pricing policies; overmanning; delays in the modernization of the capital stock (an especially important consideration in the energy sector); excessive growth of labor costs in the pretransition period and an unusually elastic budget constraint that did not prepare the enterprise sector for the more competitive environment brought about by price and trade liberalization. A process aimed at restoring enterprises’ balance sheets in those enterprises in which government participation is likely to remain high (for example, within the military-industrial complex), is likely to entail significant financial costs.

Beyond this, there remains the problem of the Closed Administrative Territorial Units, a list of 40 cities officially recognized by the federal budget as being centers for “specialized military production.” With an average population of some 200,000 inhabitants, these cities are fully supported by the federal budget, mainly in the form of subsistence level wages and subsidies. Because they are typically far from other urban centers and because of the exclusively military nature of their production, it is unlikely that these cities will be economically viable. They will thus continue to be a drain on the budget and, to the extent that they represent unused resources, a drain on the economy as well. The development of a medium-term strategy to close many of them, to finance the migration of their inhabitants to other regions, and to restructure and modernize some of the production facilities will also entail financial costs. In addition, the government will also have to meet some of the costs associated with the reconstruction of Chechnya, particularly in the area of infrastructure.

Social assistance. The financing of social assistance in general (see Section V for further discussion) will continue. This will include the financing of unemployment compensation and the selective transfer to the budget of some of the social functions presently performed by the enterprise sector, a process that, although costly, is itself an important component of the transition to a market economy. Indeed, Tanzi (1993b, p. 5) has argued that an undue focus on the budget deficit as a measure of macro-economic performance “that ignores this transfer might be met by the government delaying the transfer of these social functions from the enterprises to the budget.” This would undermine rather than enhance the transition. All the above are likely to exert upward pressure on budgetary expenditures over the medium term.

Tanzi’s point warrants further comment. He notes that focusing on the conventionally measured budget deficit (as opposed to a comprehensive and economically meaningful measure of the fiscal deficit that might include such items as social expenditures carried out by the enterprise sector and that, in market economies, are typically the responsibility of the state, or cheap loans and other quasi-fiscal operations) may create perverse incentives by inducing the government to adopt policies that go against other key elements of the transition. Some examples with respect to Russia would be (1) the unwillingness of the government to improve the level and coverage of unemployment compensation that has resulted in continued labor hoarding at the enterprise level, notwithstanding the large output drops; (2) subsidized credit by the central bank that during 1992–94 greatly reduced the cost of debt service by the government; and (3) the shifting of expenditures out of the budget to extrabudgetary accounts.

Budget Process and Institutional Reform

As in other countries in transition, a redefinition of the role of the public sector may be called for, from one largely focused on control and direction to one of supporting the private sector through the further freeing of market forces and the establishment of a simple regulatory framework based on transparent rules. The Russian economy needs to be freed from excessive intervention, arbitrary decisions, and the inconsistent application of rules and policies, all of which are likely to hinder business activity and slow the pace of private sector investment.36 The adjustment process in Russia has sometimes been undermined by weaknesses in the government’s administrative capacities, underscoring the need for ambitious institutional reforms. A case can be made that the effectiveness of adjustment policies will depend in no small measure on the extent to which they are supported by policies aimed at improving the institutional setup on which sustained implementation ultimately rests. The freeing up of prices, for instance, will encourage a supply response but its magnitude is likely to be larger in the context of adequate infrastructure for transport and credit institutions that allocate resources relatively efficiently. Adequate legal underpinnings and a framework for public accountability are also key elements of institution-building reforms. In all these processes, the role of the government is critical and must be geared to the efficient management of economic policy, facilitating the transition to better policies and the design and implementation of structural and institutional reforms.

Because of their central role in the implementation of fiscal policy, reforms are urgently needed in the formulation and execution of the budget. A review of the experience with the elaboration of the budget since 1990 shows that no two years have been alike. Budgetary procedures have changed significantly each year linked to changes in the economic and political environment. This section discusses budgetary procedures for the 1995 and 1996 budgets, as these contain the key elements that are likely to remain in place over the medium term. A number of serious weaknesses are also identified and possible solutions suggested.

The Ministry of Finance plays the central coordinating role in the budgetary process. Work on the following year’s budget begins with the issuing of a government resolution (postanovlenie) on the preparation of the budget, which contains various deadlines for the submission of information necessary for the elaboration of the budget to the Ministry of Finance by all the various spending ministries and agencies (henceforth referred to as “spending units”). The Ministry of Economy is asked to prepare a basic macroeconomic scenario, which forms the basis of the budget calculations. This mainly involves the setting of an inflation target for the year for which the budget is being prepared together with the quarterly inflation profile as well as “wage coefficients” that are to be applied to estimate the wage bill through the year on a quarter-by-quarter basis. Following the issuing of this resolution, there is usually a brief discussion between the staffs of the Ministry of Finance and the Ministry of Economy on the adequacy of the targets underlying the Ministry of Economy’s scenario, and shortly thereafter these estimates are sent to all spending units expected to submit spending requests to the Ministry of Finance.37 Major progress has been made in recent years in starting the budgetary exercise early, to prevent situations such as those that prevailed until 1994, when the final budget for that year was approved only in June, which meant that spending units did not know the final level of appropriations until that time and thus could not adequately plan or commit resources.

The Ministry of Finance asks spending units for forecasts of their individual budgetary execution through the end of the year as well as a list of their demands for the following year. This exercise allows for drawing up an “unchanged policies” base for the current year from which to make forecasts for the year for which the budget is being prepared. At the time the spending units prepare their individual current-year forecasts, they may also include any extraordinary or unanticipated expenditure requests arising, for instance, from recent spending decisions made by the executive that they feel they are obliged to cover, legislation that may have been approved recently that could have an impact upon their individual units’ spending behavior during the following period (for example, unforeseen adjustments in the previous year’s wage coefficients) or, more generally, new spending they regard as essential for the implementation of the functions under their jurisdiction. Upon receipt of this information, the Ministry of Finance will instruct the spending units what expenditure items to exclude from the projected base and will provide information on the areas that are more likely to be financed; the aim of this is to work with as realistic a base for the current year as possible. This will typically involve cuts even in areas that may be part of the officially approved budget for the current year, given the shortfalls in revenue that have been a permanent feature of budgetary implementation in recent years. The figures provided by the spending units to the Ministry of Finance will typically consist of base-year forecasts multiplied by an appropriate array of price and wage coefficients. They will also include a listing of all new expenditures, separately identified and justified, particularly for those involving new decisions. While this work at a disaggregated level goes on, the Ministry of Finance prepares a detailed revenue forecast, identifies additional sources of financing, and sets limits on the budget deficit.

In 1993–95 the initial set of demands submitted by the over 100 spending units typically were some-where between two to three times higher than the sum of revenues plus identified financing. This is the point at which the Ministry of Finance’s most difficult task begins, the ultimate aim of which is to match spending authorizations within the budget with the sources of revenue and additional financing, consistent with the targeted deficit. With the sharp deceleration of inflation in 1995/96 and the emergence of medium-term fiscal deficit targets, it is thought that the initial gap between resources requested and resources ultimately approved could be narrowed somewhat.

There are typically three stages to the above negotiation. To more easily identify the nature of the discussions, an illustrative example may be useful. In the first stage, the Department of the Social Sphere in the Ministry of Finance will meet with representatives from the Financial Department of, say, the Ministry of Culture. This ministry’s requests will be discussed and a compromise will be attempted trying to reconcile the needs of the ministry with the overall expenditure and deficit targets worked out by the Ministry of Finance. Full compromises are seldom reached at this stage, although the gaps are usually narrowed. These discussions will then be followed by a round of meetings between the Budgetary Department of the Ministry of Finance and other divisions with primary responsibility for different areas of the economy within the ministry—social sphere, industrial production, agriculture, and so on—at which a second round of cuts will be made to bring the aggregate requests emanating from the first stage down to a level that is more consistent with the Ministry of Finance’s targets. Finally, at a third stage, the Minister of Culture himself or some high-ranking official from that ministry will come to the Ministry of Finance and will meet with either the Minister or, more typically, a Deputy Minister of Finance, to make a case against some of the cuts that may have been proposed by the Ministry of Finance but that the Ministry of Culture is simply not ready to accept. At this last stage, the Ministry of Finance will typically provide the Ministry of Culture with the upper limits of expenditure consistent with the deficit target and the available sources of financing, although many of the spending units will reserve for themselves the right to appeal these decisions to the next level of decision making (see below).

In recent years, these three stages have typically taken somewhere between three to four weeks, although it is expected that in the future the process will be longer. In the high inflation environment of 1993–95, when the budget process was initially delayed and the budget itself was revised several times in the course of the year, the Ministry of Finance was under great pressure to present a budget to the government, a situation that did not permit a more considered view of spending priorities (see below).

Following this last round of negotiations and during a two- to three-week period, the Ministry of Finance will consolidate all spending units’ allocations into an aggregate budget. It will draft explanatory notes, prepare tables, and submit a first draft of the budget to the government. This is typically done by the Minister of Finance and some of his deputies meeting with the Prime Minister to report on the results of the exercise and to identify for him the salient features of the budget and spending priorities, indicating also those areas where the most significant “cuts” were made and what some of the possible risks could be in cutting down the spending units’ initial expenditure requests to more reasonable levels. At this stage, the Prime Minister may raise questions, including requesting additional information on the relative share of spending allocated to each unit, the increase with respect to the previous year, the extent of the cuts made and areas where these may have been made, and so on.

When the budget is presented to the Prime Minister, there will typically be an unallocated reserve equivalent to 3–5 percent of total revenue. At this stage, the Prime Minister will allocate this reserve fully to those units that he may feel were perhaps unduly cut or where the ministry itself may have felt a plausible case could be made for some additional spending. These discussions with the Prime Minister will typically take no more than about a week. Immediately thereafter the Prime Minister will present the draft budget to the government at a meeting of the Cabinet. At this meeting, many ministers will raise objections and typically argue that the interests of their respective sectors are being adversely affected by the large cuts they are being asked to sustain with respect to their initial demands. Ahead of this meeting considerable time will have been spent by each agency preparing papers and explaining (sometimes in great detail) the well-justified nature of their demands. The experience in the last two budgetary exercises has been that, notwithstanding the entreaties made by ministers at the Cabinet meeting, no fundamental changes were made to the proposals submitted by the Ministry of Finance. The essential problem has been that while all would like to have more resources to spend, the new sources of additional financing are not always clear. Given government reluctance to move more aggressively on the revenue front (as noted earlier, the government itself has often taken initiatives that have undermined revenue collection) and since, at this point, the size of the deficit is typically not negotiable, there is little, if any, room to accommodate such expenditure requests.

One of the main weaknesses at this stage of the budgetary process is that the Ministry of Finance is given the responsibility of negotiating with the spending units without having a clear understanding of the nature of their operations. Negotiations on expenditure cuts are often conducted in a vacuum of relevant data that might allow better informed choices. To go back to the example of the Ministry of Culture, the following situation might occur. The Ministry of Culture was able in previous budgetary exercises to obtain financing for the purchase of art works and other inputs for a number of important museums. These purchases were carried out, as budgeted, but, say, no such expenditures are necessary in the budget for 1997. However, it may well be that the Ministry of Culture submits a request for additional acquisitions so as not to lose the budgetary allocation given in earlier exercises, although, in fact, no such acquisitions are expected in the forthcoming fiscal year, or not to the extent requested. Because the negotiation is made with the Ministry of Finance at a very aggregate level, the aim of which is to try to compress expenditures down to a level consistent with the overall macroeconomic aggregates, little attention is paid to the individual structure of expenditure within each of the spending units. Consequently, the Ministry of Finance often finds itself in the position of not knowing whether the cuts it eventually enforces on the Ministry of Culture are leaving that ministry underfinanced or overfinanced, whether they are leading to an extremely difficult and inefficient situation—in terms of the overall interests of the economy—or whether because of earlier “creative accounting” on the part of an individual spending unit they are, in fact, leaving it in a relatively comfortable position and are, therefore, imposing costs on other spending units where no such accounting took place and where perhaps the proposals were more transparent at the outset.

The absence of long time series on selected expenditure items for the spending units, the inflationary environment in which the budgetary exercises have thus far taken place—which made two- or threefold increases in nominal expenditures seem “reasonable,” shortages of qualified staff at the Ministry of Finance who might follow up, over time, spending behavior in some of the key spending units and make ex post comparisons between “approved” and “actual” expenditures, all contribute to make the exercise extremely inefficient, potentially unfair, and liable to abuse. Incremental budgeting in a high-inflation transition economy poses certain risks that demand special vigilance on the part of the authorities. High inflation may give an aura of legitimacy to requests for higher nominal spending at a time when the transition itself may require the government to withdraw from certain traditional areas of financing. Identifying those areas, deciding how best to allocate scarce resources, and deciding where cuts will best serve the interests of economic efficiency require considerable administrative capacity that needs to be developed as a key priority of economic management.

Following the Cabinet meeting the government sends the budget to the Duma. The Duma will typically analyze the budget in considerable detail over two to four weeks. At its first hearing, it may either approve or reject the budget as a whole. Approval of the budget on the first reading implies acceptance of the underlying macroeconomic assumptions and other basic assumptions that may have been included in the preparation of the budget—for instance, “no monetary financing in 1995.” If it is rejected, it is typically sent back to the government with various comments and requests.

Or alternatively, the Duma, as was the case for the 1995 and subsequent budgets, may decide to appoint an interagency Conciliatory Commission to try to bring consensus to the large body of different opinions that will have formed on the appropriateness of the budgetary stance proposed by the Ministry of Finance and the government. Typically these are the most detailed discussions on the budget and will involve a number of different committees within the Duma, the Ministry of Finance, the central bank, and various spending units. Following approval of the budget at the first reading, the second reading will involve approval of the revenue, expenditure, and deficit aggregate figures. At the third reading, individual items of revenue and expenditure are discussed, with the emphasis on the latter, often involving significant additional reallocation within the various expenditure chapters. The fourth and final reading is a general discussion at which the Duma also approves such aspects of the budget as external financing proposals and fiscal federalism issues.

It is important to emphasize that at the time of the second reading when the aggregate parameters of the budget are discussed and approved there is not a great deal of scope for boosting revenue given that in the first reading the Duma has already approved the macroeconomic assumptions underlying the budget and, in particular, the price and wage forecasts used in the entire exercise. This has not prevented the Duma, however, from inflating revenues, often significantly, in every budget exercise during the past several years, increases that are then allocated to various spending categories. A variant to this (for example, the 1996 and 1997 budgets) involves the Ministry of Finance presenting overly optimistic revenue estimates at the outset that, ex post, give the government considerable sequestration discretion when, inevitably, revenues fall short (see below).

One of the weakest points in this process is the lack of an effective linkage between such additional revenue “requests” on the one hand and the necessary underlying tax legislation on the other, often with no attempt being made by the Duma to formalize the linkage that exists between the two. For instance, the Duma will “find” alternative sources of revenue (usually motivated by pressures for additional spending) but will not follow through with the necessary legislation that will make it possible to collect that additional revenue. This, therefore, creates a “reality gap” that leads to the overestimation of revenues by as much as 10–20 percent. A case in point: the Duma identifies Rub X trillion of additional revenue by proposing the elimination of certain import duty exemptions for a particular group of organizations. It allocates the additional revenue that is generated by the “elimination” of the exemptions to, say, defense expenditure, but then once the budget is approved it does not follow through with the appropriate legislative acts that might make it actually possible to eliminate the exemptions and collect the revenue. In the meantime, however, the higher level of expenditure has been approved and an inconsistency is thus created between the reasonably carefully worked out proposals of the Ministry of Finance and the figures that are eventually approved and become part of the budget law. The government has often accepted this inconsistent approach as part of the strategy to “buy support” for the budget, for instance, from the agrarian lobby or other interest groups. However, this typically leads to a situation where the sum of actual revenues and financing turn out to be less than the levels anticipated in the budget, which creates problems for budgetary implementation. A similar situation arises as a result of the indexation of wages, including the minimum wage, and pensions. The principle is basically the same: a number of decisions are taken in the course of the year that were not fully anticipated in the budget eventually approved and that again create difficulties for implementation.

An additional source of “noise” in the system once the budget has been approved is the number of decrees or resolutions of the government and the executive that are issued practically on a daily basis and that typically involve additional spending. In some cases, the Ministry of Finance may claim that these decisions are, in fact, already incorporated in the budget. Again, an illustrative example might be that the Ministry of Culture has a budget approved for the construction of a new museum. A senior member of the government visits a particular town and promises the population and the local authorities to build a new museum for Rub X billion. The Ministry of Finance may subsequently claim that the museum promised was, in fact, already budgeted for and included in the original request of the Ministry of Culture, an observation unlikely to be appreciated by this ministry since they may have had an entirely different place in mind on which, even, work may already have commenced. The Ministry of Finance, however, may not always be able to do this because the promises made (or something resembling them) are nowhere to be found in the original budget proposals. In this case, the Ministry of Finance finds itself in the position of having to cut elsewhere. Obviously this creates imbalances; in the absence of additional revenues, departures with respect to “approved” spending paths are inevitable and the possibility of major inequities in the spending process is increased.

The opposite situation may be created when actual inflation is higher than that incorporated in the budget, leading to additional revenues that can facilitate the implementation of such promises, assuming that the additional revenues thus generated more than offset the adverse impact of overly optimistic baseline forecasts. It is estimated that “new promises” requiring spending cuts elsewhere after the budget has been approved amount to 5–10 percent of total expenditures.

When faced with lower-than-anticipated levels of financing (including lower revenues), the government is forced to cut expenditure accordingly, and for this purpose it has developed a number of guiding principles. Capital spending tends to be cut first. Current spending is typically cut across the board, although certain components, such as debt service, wages, stipends, and some items of a social nature are considered “protected” and thus, in theory, not subject to such cuts. In practice, however, these guidelines are not always adhered to. It may not be feasible to cut in the short term certain capital projects, and thus, for instance, wage arrears may build up. A considerable level of arbitrariness in the spending process is thus introduced. The spending authorizations included in the budget rapidly become no more than a loose framework providing spending units with an upper indicative limit for resources likely (or unlikely) to be received. Considerable uncertainty in their operations is thus created, and much time is then spent on deciding how to reallocate priorities within their (effectively) reduced level of spending and on lobbying the government for maintaining a certain flow of financing. Indeed, the larger the financing shortfall the more the activities of the spending units are turned into emergency cash management and lobbying operations.

On occasion, as it happened in August 1995, a particular spending unit (in this case, the Ministry of Defense) may unilaterally decide to carry out spending not authorized in the budget: for example, a 25 percent increase in the wages of the military amounting to some Rub 7 trillion (0.8 percent of GDP) for the last five months of the year, which is then validated by the government.38 As in the earlier example, given the budget’s overall deficit limits, such decisions force drastic cuts elsewhere, contributing to the emergence of situations where the budget at times is able to finance little beyond the payment of wages and interest on the public debt.

Sequestration inevitably contributes to the growth of arrears. Beyond those associated with the nonpayment of otherwise “authorized” spending, it creates an environment in which not meeting commitments is regarded by economic agents as tolerable behavior, including, of course, the payment of taxes to the budget. It greatly reduces the credibility of the budget as an instrument of fiscal policy and of the government as the chief architect of that policy. Fiscal policy, the elements of which, in most cases, governments tend to revise once a year on the occasion of the preparation of the budget, is then reduced to weekly and/or daily cash management mainly involving “who gets how much.” Such an approach is enormously inefficient and detracts the authorities’ attention from the kind of policy concerns and strategic planning and thinking that is essential in Russia, given the serious problems that remain, as noted in earlier sections. Since the above inefficiencies are precipitated, by and large, by revenue shortfalls, there is no alternative, over the medium term, to considerably improving revenue performance and tax administration. Alternatively, faced with tight deficit ceilings and financing shortfalls, the government may authorize spending units to borrow short term from commercial banks under government guarantees (not recorded in the budget), as begun to be made in the latter part of 1995 and continued in earnest in the course of 1996.

The experience of the past several years has shown that the structure of expenditures after the budget was implemented was radically different from that anticipated in the approved budget (Table 10). Spending units are obliged to present their expenditure plans to the Ministry of Finance following a system of economic classification but the distribution of expenditure is, in effect, purely indicative and no attempt is made later to compare the initial budget proposals with the figures actually executed (Figure 7). Since spending units will have typically received less than the budgeted amounts, the Ministry of Finance does not feel it has the authority to go back and verify discrepancies or departures with respect to initial plans. Indeed, in the course of the year, as resources become available, the Ministry of Finance will credit the accounts of the spending units, but there is no formal requirement that the resources received should be spent in a way that reflects the original proposals presented by the government and ultimately approved by the Duma (for example, x percent to wages and salaries, z percent to capital investment). If it becomes evident that resources received have been grossly misallocated, the Ministry of Finance may “inform” the government, but typically, no further action is taken. Furthermore, there are certain spending units (for example, the Ministry of Defense) for which the informational requirements associated with the uses of the resources allocated to them are considerably less stringent. In light of the above, it is perhaps not surprising that no time-series data exist, for instance, on the share of budgetary spending allocated to wages and salaries.39 Indeed, no official budget execution data has yet been issued for 1995 by economic classification.

Table 10.Budgeted and Actual Federal Expenditures, 1994(In percent of GOP)
Total federal expenditure31.022.0
National economy7.23.0
Education, culture, arts, and public health2.11.5
Law enforcement2.01.7
Intergovernmental transfers4.33.5
Source: Ministry of Finance.
Source: Ministry of Finance.

Figure 7.Federal Spending: Budget Versus Actual, 1994

(Share of total spending)

Source: Ministry of Finance.

In addition to the impact of “new promises” and the unanticipated cuts they force in other areas, and the effects of expenditure sequestration, many of the important wage decisions are often made after the budget has been approved and are not always consistent with the wage coefficients included in the budget at the time of its preparation. Because of the large weight of wages in total spending (probably around 16–20 percent), this inevitably forces sharp changes in the structure of budgetary spending. A t the same time, shortfalls in foreign financing may also induce a restructuring of the budget as the budget is implemented. For this reason, because there are sizable differences between the structure of expenditures as approved and as actually implemented, the Duma has argued for shifting the elaboration of the budget to a quarterly basis in an attempt to have more control over spending; this has not been supported by the government. Reflecting the above discrepancy, the actual budget execution for the previous year has never been discussed in a session of the Duma. There seems to be agreement that, as long as the deficiencies are not corrected, such debate would not be especially useful and might, by highlighting the underlying weaknesses, actually further undermine the credibility of the budget and the budget process.

Budget Financing

Treasury bill auctions were started by the Ministry of Finance in May 1993. The authorities’ motivation appears to have been twofold: to tap an important source of resources for financing the fiscal deficit and to widen the range of market-based instruments with which the central bank could conduct monetary policy. Given the large deficits registered in the early part of the transition, the recourse to central bank credit with the associated repercussions for the price level, and the limited scope for securing additional foreign financing, the authorities felt that treasury bill issues would fill an important void as both instruments of monetary control and public debt management. While, in net terms, financing of the budget that year through issues of treasury bills was small (less than Rub 200 billion, equivalent to 0.1 percent of GDP) and consisted almost exclusively of three-month maturities, gross issues were increased significantly in 1994 and thereafter and began to cover an expanding share of the federal deficit (Table 11). For the 1995 budget the government adopted a two-pronged strategy—while projecting a sharp reduction in the deficit, it moved to eliminate direct central bank credits altogether, theretofore granted at a yearly nominal interest rate of 10 percent.

Table 11.Financing of the Federal Deficit(in trillions of rubles)
Financing of the deficit2.011.269.788.5186.5
Net foreign financing-0.1-0.10.1-3.114.5
Domestic financing2.111.369.691.7172.0
Banking system1.911.261.079.6152.5
Monetary authorities2.411.253.524.148.4
Banking system-0.57.555.6104.1
Sources: Ministry of Finance; and IMF staff calculations.
Sources: Ministry of Finance; and IMF staff calculations.

A move from central bank financing to treasury bill financing, whether through the banking system or the nonbank public or both, transfers the costs of the deficit from one source—the central bank—to another, the budget.40 Central bank financing in Russia had led to the rapid growth of the monetary base and created a need for extensive sterilization—to stem the inflationary impact—at a time when, other than through interventions in the foreign exchange market, the availability of such instruments was strictly limited. As has been observed in other countries, the high reserve requirements usually associated with sizable sterilization operations (equivalent to taxation levied through the financial system) tend to lead to a large spread between borrowing and lending rates as banks attempt to preserve profitability through high lending rates on that portion of their assets not subject to reserve requirements, and/or relatively low rates on deposits.41 In Russia the treasury bill market appeared and then grew in the context of a highly inflationary environment and at a time when confidence in the ability of the government to fulfill its obligations was not high, given certain past developments involving callbacks of banknotes, the blocking of foreign exchange deposits at the Vneshekonombank, and delays in the reimbursement of commodity bonds, all of which had undermined government credibility. These factors, together with widespread perceptions that legal protection of debt holders is quite weak, helped create a situation characterized by relatively short maturities and high nominal and real interest rates. In addition, exchange rate stability during much of 1995–97 turned high ruble real interest rates into high dollar real interest rates as well. Weak budgetary revenues and tight foreign financing thus contributed to growing domestic borrowing and continued upward pressures on interest rates.

Treasury bill financing is not necessarily a particularly severe constraint in the conduct of fiscal policy if the deficits are small or declining with respect to GDP and, furthermore, has none of the negative impact that central bank financing has on the allocation of financial resources. However, if the budget deficit is expected to be relatively high over the medium term and the economy is characterized by high inflation, agents will show a marked preference for short-term instruments and demand an interest rate premium. In conditions of high inflation the fiscal deficit then becomes a function of both the stock of public debt and the level of the nominal interest rate and, through it, the inflation rate. In Russia at the end of 1996 domestic debt was equivalent to about 12 percent of GDP and total debt was around 35 percent of GDP.

Because it is often difficult to reduce current expenditure, the authorities may, in an effort to keep the deficit within agreed levels, find it tempting to reduce public investment expenditure or outlays for human capital, both of which are likely to have adverse implications for future growth. A distortion may then be introduced in the pattern of public investment expenditure as projects are no longer judged on the strength of their merits but rather on the extent to which they might contribute, in the short run, to a widening of the deficit. The sensitivity of the interest component of public expenditure to fluctuations in interest rates may also impose additional constraints on the exercise of monetary policy and may lead to attempts to push interest rates on government paper down merely on the grounds that not to do so would prove a threat to the budget. This higher degree of intervention in the financial markets could come at a time when attempts are being made to move toward freer, less-regulated financial markets. This “forced” reduction of interest rates may in turn result in net reductions in the holdings of treasury bills by the nonbank sector as the relative rates of return on alternative assets go up.42

As Russia’s financial market is developed and a process of accumulation of interest-bearing government debt in the hands of the nonbank sector deepens,43 it will be necessary to ensure that an additional source of uncertainty is not introduced in the evolution of the monetary aggregates and, by implication, in the implementation of monetary policy. While in Russia the bulk of treasury bills remain within the banking system—and thus at least in theory subject to whatever reserve requirements the monetary authorities might consider consistent with the achievement of the monetary targets—the emergence of a large market for bills in the hands of the nonbank sector in a context of still relatively high inflation could serve as a potential source of instability for the financial system.

However, if it is assumed that the government will follow over the medium term a reasonably cautious, consistent, and responsible course as regards financial policies—and as part of which there will be no manipulation of interest rates—then these instruments should prove to be an important source of noninflationary finance. Furthermore, the emergence of the treasury bill as an additional policy instrument for the conduct of monetary policy should be regarded as a welcome development. Purchases and sales of government securities in the market to influence liquidity in the system should enhance the central bank’s capacities in the area of monetary control.

27These ratios exclude the operations of the extrabudgetary funds that, in relation to GDP, fell from some 12 percent of GDP in 1992 to about 9 percent in 1996, mainly because of a contraction in the expenditures of the Pension Fund. and credit operations previously off-budget were gradually incorporated into the budget.
28For a fuller discussion of these issues, see the section on Budget Process and Institutional Reform below.
29Directed credits from the central bank in 1992 amounted to roughly $14 billion.
30For instance, Government Resolution No. 126 of February 23, 1994, on measures “to assist the agricultural complex,” instructs the Credit Policy Commission and the central bank to allocate during the first half of 1994 “no less than Rub 5 trillion” (0.8 percent of GDP) in directed credits to agricultural enterprises for the purchase of various inputs for the spring sowing season. The resolution also instructs the central bank to grant these credits with a maturity of “up to three years” and explicitly defers payment of principal and interest until September 1, at which point it is expected that farmers would have the resources to begin to pay their loans. The Ministry of Finance was also instructed to continue to finance in 1994 the interest rate differential on preferential credits granted to certain categories of farmers during 1992 and 1993; the bulk of these credits was granted at 28 percent.
31It is noteworthy that, notwithstanding this rise, wages and salaries in transition economies remain well below levels in Western industrial economies.
32For a more detailed discussion of social issues, see Section V.
33The table provides a listing of general government expendture levels in OECD member countries.
34Average real wages in the public sector (federal level) fell by 40 percent during 1995.
36There is much international evidence, for instance, that a strong disincentive to foreign investment is the presence of an in-consistent and complex set of signals, subject to a high degree of uncertainty. Investment decisions involve issues of long-range planning; from the investor’s perspective a well-identified, simple, and stable set of rules—even if somewhat restrictive—may be preferable to one perceived to be opaque and subject to unpredictable changes.
37Occasionally, at the time that these estimates are sent, other sup-porting materials may also be forwarded to all the spending units. For the 1995 budget, for instance, the Ministry of Finance sent an explanatory note on the new system of budgetary classification.
38Indeed, this increase was not incorporated as part of the 1996 budget either, but was fully financed by the Ministry of Finance.
39It is estimated that 16 percent of total federal expenditure in 1996 was allocated to wages and salaries.
40The case is sometimes made that in either case the costs are borne by the budget although through different mechanisms. Unremunerated or below-market credit to the government will reduce—or totally eliminate—the transfer of central bank profits, frequently an important source of nontax revenue.
41This interest rate wedge has well-known adverse allocational repercussions. It introduces distortionary effects on savings and investment and thus represents a form of taxation of the economy’s productive sectors. In the extreme case of unsterilized central bank financing, the result will be inflation, which is also a form of taxation. In Russia during the first half of 1995, the difference between the monthly interbank lending rate and the deposit rate for legal entities amounted to more than 20 percentage points.
42A key issue associated with the question of the desirability of developing a market for public debt is whether the instruments chosen will encourage additional savings or whether they will simply lead to asset substitution. If the intent of the authorities is to encourage savings—while at the same time remaining faithful to a policy of diminishing intervention in the financial system—some form of indexation of treasury bills could be considered. A lengthening of maturities may alleviate pressures on the financial system but will not reduce the burden on the budget.
43One-year government obligations (OGSZs) began to be sold to the nonbank public in the fall of 1995, The notes were introduced as bearer obligations carrying four coupons for quarterly interest payments. More recently, two-year obligations carrying four coupons for semiannual interest payments were introduced. As of end-1997 some Rub 13 trillion had been issued in denominations of Rub 100,000 and Rub 500,000. The potential market is seen to be large.

    Other Resources Citing This Publication