Definition of Government

One of the most basic questions faced by all countries in transition from a centrally planned system to a market economy is to define what exactly the government should be doing and, equally important, what it should not be doing. The “classic” Soviet-era enterprise would, for example, include all kinds of social assets, such as schools and hospitals, which provided services that would in most market economies be provided by the government. As these enterprises were privatized and budget constraints on enterprises began to harden, difficult decisions had to be made on the fate of these assets: would the local government accept responsibility for the hospital or would the enterprise maintain the hospital and simply not pay taxes? The choices made often preserved the murky line between the government and nongovernment sectors. Similarly, many state enterprises that are formally outside the budget continue to undertake quasi-fiscal activities—for example, subsidized energy supplies—that may ultimately generate significant costs that will need to be borne by the budget.

On the other hand, many government agencies have branched out into private sector activities. For example, the National Statistics Committee offers special data for a fee; government agencies rent out real estate to commercial enterprises, possibly in exchange for equity; and some government agencies provide marketing services. While it is not uncommon for public sector institutions in market economies to conduct some commercial operations—for example, a museum renting out space to a concession store—it is the norm that the proceeds (or the bulk of them) flow back to the budget and are taken into account when determining expenditure allocations. This has not been the case in Russia, where budgetary institutions have accumulated equity holdings and funds that are outside the budget and used at their discretion.

A sympathetic view of the off-budget activities is they are undertaken to supplement inadequate funding, and that they were not included in the budget because there was no legal requirement to do so. Even this interpretation raises concerns, in that it implies that actual government expenditures will be a function of expenditure allocations (that presumably reflect the government’s expenditure priorities) plus the agency’s ability to market itself. Less sympathetically, one can view these activities as the exploitation of public sector assets for private gain.

Budgetary procedures have also been complicated by the existence of a large number of earmarked funds that have specific revenue sources tied to them. The number of such funds has been progressively reduced, with the underlying activities either incorporated into the budget or simply eliminated. In the context of the 2001 budget, the federal environmental fund, the federal fund of the ministry of taxes and levies, the federal fund of the federal tax police service, the state fund for fighting crime, and the federal fund for replenishment of mineral resources were all incorporated into the budget. As of end-2001, the only earmarked fund remaining at the federal level was that of the ministry of nuclear power.

Off-Budget Activities

The problem of off-budget activities by budget institutions expanded in the face of expenditure pressures, as the government had difficulty meeting its fiscal targets and resorted to sequestration, and as budgetary institutions realized the opportunities available to them. A strategy for integrating these activities into the budget was already under preparation before the crisis.

• The first step was to get a better handle on the size and range of operations that were being conducted, by capturing the financial flows that passed through the own-resource accounts created for these activities.² An initial inventory of accounts identified by the federal treasury at end-1998 suggested that the balance on these accounts was about 0.6 percent of GDP (Morozov and Sundberg, 2000). Income raised by the ministry of education (for example, from educational services, or leasing premises) amounted to about two-thirds of the federal budget allocation.

• The next step was to introduce a system to consolidate and register in the federal treasury all own-resource accounts. This was achieved through the issuance of a government resolution on August 22, 1998, and a subsequent executive order in June 1999. As might be expected, this met some resistance. However, by end-1999, the Central Bank of Russia (CBR) was able to draw up lists of own-resource accounts, though the comprehensiveness of these lists is difficult to verify owing to complications arising from bank-client confidentiality, which limits the information that the CBR can provide to the federal treasury.

• As the final step, the federal treasury has issued permits to line ministries (and to spending units through them) for opening own-resource accounts. The relevant budgetary agency is required to report on the flows through these accounts. By end-2001, with the exception of the defense sector, most own-resource accounts at the federal level had been brought under the federal treasury, though they were still not captured in the budget.

Quasi-Fiscal Activities

Quasi-fiscal activities take many forms and are extremely difficult to measure, in even the most advanced economies. In transition economies, there are three key areas where quasi-fiscal activities have arisen: (1) the central bank; (2) the banking system; and (3) enterprises, particularly those in the energy sector. In addition to concerns that the existence of quasi-fiscal activities distorts the measurement of the fiscal position, the magnitude of the quasi-fiscal activities in the energy sector is such as to severely distort economic incentives and resource allocation and result in underinvestment in the energy sector.³

Elimination of Noncash Payments

Noncash revenues peaked at about 3 ½ percent of GDP, more than one-fourth of total federal government revenues, in 1996. These noncash payments typically took the form of an offset arrangement in which an enterprise provided services or commodities to government agencies in lieu of cash payment for tax obligations. In addition, there was also frequent resort to arrears clearance schemes in which, for example, the tax obligations of the Unified Energy System of Russia (UES) were reduced by the amount of arrears of budgetary institutions to UES for electricity consumed. In addition to being completely nontransparent and providing extensive scope for graft, the use of these schemes served to undermine budgetary discipline—as it became advantageous for enterprises to accumulate tax arrears as a means of creating demand for their output—and frustrated any attempt to improve expenditure planning.

The significance of the problem created by noncash payments was recognized by the government, and a presidential decree banning the use of such arrangements was issued in early 1998. However, budgetary arrears rose sharply in the immediate aftermath of the crisis, and the government resorted to substantial use of offsets in settling its obligations during the fourth quarter of 1998. In 1999, however, the federal government was successful in resisting the use of noncash operations, aided by the steps taken to strengthen control over expenditure commitments by budgetary institutions (see below). In accordance with the budget code that formally bans the use of such operations at all levels of government, the federal government has also been able to insist on receiving its share of the taxes collected by subnational governments in cash since the beginning of 2000.

Monitoring of Expenditure Commitments and Reduction in Arrears

Before the crisis, expenditure arrears were endemic in several key sectors, such as energy and military procurement, where bills were ultimately settled via noncash operations.⁴ Indeed, the failure of budgetary institutions to make utility payments directly contributed to the culture of nonpayment within the economy and made it difficult, if not impossible, for the energy companies to meet their tax obligations. At the level of the budgetary institution, not paying for utilities was a perfectly rational response to the likelihood of sequestration—why waste the budgetary allocation actually received on the state-controlled utility companies, when there was little or no chance that these utility companies would terminate supplies in response to payment arrears? For the defense sector, part of the problem was likely poor budgeting practices, though it could also reflect a deliberate attempt by elements in the defense sector and traditional defense contractors to enter into commitments beyond budget allocations in the anticipation that the defense lobby would be able to force the ministry of finance to meet these additional obligations.

Resolving these issues required action on a number of fronts. First, increased efforts were made to track expenditure commitments. Beginning in mid-July 1999, a system for recording all commitments at the treasury was introduced for specific spending codes where problems had been encountered in the past—particularly utility payments. Once the commitments are recorded, the information is compiled into a system to track arrears and make the necessary inputs into the budget execution process. Second, to reduce the likelihood of arrears, steps were taken to establish a system based on commitment registration, in which payment orders would only be processed if the contracts had been registered in advance with the treasury. While a system for tracking expenditure commitments has been successfully implemented, the commitment registration system has proven more difficult—notably for communal and housing services where problems had arisen in the past. Delays have stemmed from unresolved legal issues: the implication of the contract registration system is that the federal treasury has the right (through its ability to deny payment) to overrule contracts entered into by budgetary institutions, which appears inconsistent with the Civil Code.

Strengthening the Status of the Treasury and Extending Its Coverage

The first steps toward developing a treasury had occurred in 1994 with the creation of a federal treasury department in the ministry of finance, charged with the execution of the budget and with government accounting and general cash management. However, both key line ministries and the CBR managed to slow its development to preserve the previous system, which provided them with more control over their financial operations. A presidential decree on treasury development approved in 1998 reinforced the position of the treasury by providing a formal statement that the federal treasury was the main controller of budget execution and responsible for cash management. This role was further cemented in the budget code that became effective in January 2000.

While most line ministries had, albeit reluctantly, brought their on-budget operations under the federal treasury, the defense establishment—which as described above was one of the major sources of improper budgeting practices—had stayed outside the system and even continued to use its own accounting system, which was incompatible with the budgetary classification. Recognizing the difficulties for budgetary management that were being created by this sector, the government sought to work with the defense agencies on a system for bringing their operations under the treasury, while providing the necessary safeguards for protecting sensitive information. Two pilot projects were implemented beginning in 1999 in the Privolzhski Military (Samara) region and the Baltic Fleet, in which the treasury essentially processed payment orders. The scope and number of these pilots has been progressively extended with the result that, beginning with the 2002 budget, the vast majority of defense transactions are now covered by the federal treasury.

Implementation of the decree on treasury development has also focused on establishing a treasury single account in the CBR, together with the centralization of all government operations in the federal treasury accounts, to provide an accurate picture of, and control over, budgetary expenditures. In support of these efforts, the treasury has introduced a uniform accounting and reporting system based on a single accounting and budget classification. Progress on the centralization of accounts has, however, been drawn out because of the preponderance of off-budget accounts and deep resistance from some budgetary institutions to reporting on such activities. In terms of achieving the treasury single account, progress in centralizing accounts at the regional level has been slow, with the result that delays in reporting daily cash balances and weaknesses in government cash management persist.

Ability to Resist Pressures for Additional Expenditures

Expenditure allocations in recent budgets have been fully funded, unlike in the years leading up to the crisis, as revenues have exceeded the targets established in the budget law. While this may reflect a trend toward more realistic budgeting, a key factor has been the series of positive macroeconomic shocks that have strengthened budgetary revenues—the larger- and more-rapid-than-anticipated rebound in the real sector following the crisis, as well as higher-than-projected international energy prices. As discussed in Chapter 2, the government’s resistance of the temptation to spend the additional revenues has underpinned the recent macroeconomic stability.

The restraint on expenditures has stemmed from two main factors.

• First, the budget laws in 1999, 2001, and 2002 included very specific language to deal with excess revenues. In 1999, this took the form of an explicit excess-revenue-sharing formula detailing how much of any excess should be saved. In 2001, the budget was based on a high-end forecast of revenues, and the budget included a set of expenditures that would only be made once the revenue targets had been fulfilled. In 2002, there was a formal provision for revenues beyond targets to be channeled into a special financial reserve fund that is to be used to meet debt-service payments in future years—notably the 2003 debt hump.

• Second, the government has been able to marshal strong political support in the Duma. There was a clear consensus that strong action to correct the fiscal imbalances was necessary, including the creation of a reserve to handle the 2003 debt hump, to avoid a repeat of the 1998 crisis.

While these steps have been effective, they do not constitute good budgetary practice and may not be sustainable. The special clauses written into the budgets are distortionary—particularly those in the 2001 budget, which released the contingent expenditures very late in the year—and are subject to renegotiation each year. Also, the political consensus for saving excess revenues may well weaken as elections approach.

Strengthening Expenditure Controls

The principal organ of expenditure control at the federal level is the federal treasury. Similar entities have been slow to develop at the sub-national level and the Development Program for Fiscal Federalism Until 2005 only sees the transition at the subnational level to treasury budget execution being “substantially completed” by 2005. Regions have been given the option of volunteering to have their finances executed through the federal treasury, but the perceived reduction in autonomy has meant few volunteers. Indeed, the only regions that were working via the federal treasury in mid-2002 were those that were formally required to do so—heavily subsidized regions with over 50 percent of revenues coming from intergovernmental transfers, regions undergoing a financial crisis that cannot balance their budgets, and regions that fail to meet their obligations to households, budget recipients, or creditors.⁶

As at the federal level, the widespread use of noncash transactions created enormous difficulties for expenditure management—Lavrov, Litwack, and Sutherland (2001) suggest that the share of noncash payments in subnational tax revenues reached more than 50 percent in 1998, higher than the share in federal taxes. The federal government’s refusal to accept noncash transactions from January 1, 1999, had an ambiguous effect on subnational governments. On the one hand, the federal measure was key to unraveling the chain of noncash payments and increased the monetization of the economy. On the other hand, at least in the short run, pressures on subnational governments to accept noncash payments likely increased, as enterprises had to divert their available cash to meeting their federal obligations. The federal government chose to enforce the provision in the budget code banning all noncash operations at all layers of government by requiring, in 2000, that all revenues transferred to the federal government via tax sharing rules be done so in cash. This had a very sobering effect. Subnational governments faced the stark choice of either insisting on cash payments or accepting noncash payments and then having to divert cash reserves collected from other taxpayers to the federal government. While data on the incidence of noncash transactions at subnational levels are difficult to verify, Lavrov, Litwack, and Sutherland (2001) do show very sharp declines in both 1999 and the first half of 2000. The fact that this is no longer raised as an issue also suggests that the problem has now been largely overcome.

Improved accounting systems and the extension of the treasury have also permitted the creation of more timely and comprehensive data on budgetary execution. Monthly data for each region (at varying levels of disaggregation) on the cash execution of local budgets have been posted on the ministry of finance’s website since 2001. Publicly disseminating such data is a key step toward enhancing accountability at the subnational level.

Transparency and Predictability of Transfers

As might be expected in a country of Russia’s size, there is dramatic variation between the 89 regions. The disparities stem from a combination of the initial resource endowment and the inheritance from the past—in terms of both the extent of industrial development and the accompanying infrastructure. The economic disparities have increased since the beginning of the transition, with wealth (and therefore the tax base) concentrated in the resource-rich regions (notably the oil regions), the cities of Moscow and St. Petersburg, and a few other regions. One result of this is that there are relatively few “surplus” regions that provide net transfers to the federal budget and a large number of net recipients for whom intergovernmental transfers are a key source of income. Even though the overall size of intergovernmental transfers is fairly modest (see Table 5.1), these flows are essential funding for many regions.

Since 1995, the majority of intergovernmental transfers have been distributed via the fund for the financial support of subjects of the federation. The methodology for calculating the size of proposed transfers was largely based on recent budgetary performance—thus, regions showing a strong performance had their transfers reduced, thereby introducing a relatively perverse incentive structure—and the amounts of the transfers were formally specified in the budget law but were frequently adjusted during the Duma debates on the draft budget. The 2000 budget introduced a revised transfer formula, explicitly stated in the budget law, that attempts to compare economic potential with economic needs. While the formula can be challenged on many grounds—for example, economic needs are largely calculated on the basis of Soviet-style norms—it has brought clarity and transparency to what used to be nebulous calculations, particularly since the government has been able to resist arbitrary adjustments to the formula during Duma discussions.

Unfunded Mandates

As noted above, much of the expenditure conducted at the subnational level is done to fulfill federally mandated policies rather than on the basis of expenditure policies established at the regional level. For example, the federal government issues instructions for the provision of free milk to young children or special privileges for invalids such as free public transportation and discounts on housing rents and public utilities, but the region must pay for these services. Total expenditure commitments under the major federal mandates were estimated at 7.7 percent of GDP in 1998, of which less than one-third was financed by the federal budget (OECD, 2000). This low proportion of financing of mandates occurred in spite of a 1993 law that formally eliminated the right of the federal government to delegate unfunded mandates to lower levels of government. The legal system has, nonetheless, tended to interpret the federal mandates, whether or not they are financed and whether or not they were introduced after 1993, as obligations of the subnational governments.

The regional governments have themselves resorted to the use of unfunded mandates to channel resources to groups viewed as vulnerable—such as veterans, invalids, young children, and Chernobyl victims. These mandates are estimated at about 0.5 percent of GDP a year and have, in large part, been financed by decapitalizing state-owned enterprises—particularly the utility companies providing services for housing and heating.

The 2001 budget formally addressed the issue of unfunded mandates for child allowances and disability payments by explicitly moving the financing of these payments to the federal level. The list of federal unfunded mandates has been formally suspended, but not eliminated. The problem of regional mandates remains.

Pension Fund

The pension fund was established in 1991 to implement the state pension law that provides for labor and social pensions. Labor pensions are linked to earnings and are adjusted in an ad hoc fashion, as there is no formal indexation mechanism. Social pensions are paid to invalids, certain military personnel, and various other groups. Labor pensions are formally financed through a payroll tax of 28 percent (all paid by the employer), while social pensions and the costs of administering pension payments are covered by federal budget transfers. As of end-1997, there were over 38 million pensioners receiving payments from the pension fund.

Pension fund expenditures declined sharply in the immediate post-crisis period, reflecting the absence of indexation, as nominal benefits were not increased despite the burst of inflation (see Chapter 2). Revenues soon increased as the economy strengthened and enterprises returned to profitability, enabling the pension fund in 1999 to finally clear its payment arrears, which had plagued its operations since early in the transition. The pension system is undergoing a major overhaul in the context of the pension reforms initiated in 2001 (see Box 5.1).

Medical Insurance Funds

The medical insurance funds comprise both a federal mandatory fund and regional medical insurance funds. The funds are intended to cover health programs for the unemployed and elderly. They are financed from a payroll tax of 3.6 percent supplemented by transfers from local budgets that together amount to about 1 percent of GDP. Resources from the funds are transferred to medical establishments based on various resource indicators, such as the number of beds, rather than need.

Social Insurance Fund

The social insurance fund is designed to provide benefits for work-related illnesses and disability. In 1999, it had revenues of 1 percent of GDP, derived from a payroll tax of 5.4 percent. Since 2000, the finances of the social insurance fund have been supplemented by additional contributions of 0.2 percent to 10.7 percent of payroll, depending on occupational risk.

While there is clearly a need for such a safety net, it is not clear that the existing fund is an efficient way to protect such workers. Much of the resources remain at the enterprise level, with a significant amount spent on sick leave benefits that can only be monitored by the enterprise, which has no incentive to verify claims. The continued absence of controls over expenditures raises clear governance concerns.

Employment Fund

The employment fund was intended to serve as a key component of the safety net for the unemployed, providing financial benefits, as well as retraining opportunities and other job creation activities. These activities were financed by a payroll tax of 1 ½ percent, which, in 1999, amounted to 0.3 percent of GDP.

The employment fund was not particularly effective. In part reflecting the low level of benefits available and the small likelihood of timely payment, only 1.2 million of the estimated 8.7 million unemployed in 1999 bothered to register. In addition, the fund employed about 35,000 people to serve the 1.2 million registered unemployed. The regions were also relatively effective at ensuring that revenues collected in one region were spent in that region, regardless of where the labor markets needs were most pronounced. The fund was abolished effective January 2001 and its functions integrated into the federal budget.

Road Fund

The road fund was responsible for both road maintenance and new road construction. It was made up of a federal road fund (financed by a ½ percent turnover tax) and regional road funds (financed by taxes on gross sales that averaged about 2 percent) that are in some cases on-budget as earmarked funds and in other cases extrabudgetary funds. The road funds also received revenues from a share of taxes on fuels.

Despite the fact that the road funds used the budget classification as the basis for their accounting systems, the different treatment of the funds, with respect to their coverage in the budget, makes it difficult to accurately estimate total expenditures. For 1999, total road fund expenditures were estimated at 2.7 percent of GDP.

There were several problems with the operation of the road funds. First, their operations suffered from a lack of transparency, monitoring, and accountability. In particular, the operations of the regional road funds were suspected of high levels of mismanagement, with contracts awarded to friendly enterprises and a large proportion of expenditures conducted under tax offset operations. Second, the division of expenditures between new road construction and maintenance on existing roads was excessively skewed toward new construction. Third, there was only limited redistribution of resources between the regional road funds, making road expenditures in a region effectively dependent on the revenues generated in that region, although a revised transfer formula was adopted in the 2000 budget to provide for greater redistribution of resources.

Acknowledging the extent of the accountability issues, the authorities considered bringing the road funds under the federal treasury. Despite the fact that the accounting systems were similar, this would not have been a simple task, as it would have significantly increased the number of transactions going through the federal treasury, stretching its technical capacities. In the event, the technical complications and resistance to the move stymied this approach.

A more direct strategy was incorporated in the 2001 and 2002 budgets. The federal road fund was simply cut back and then abolished. The road tax that financed the territorial funds was reduced in stages (from 2.5 percent to 1 percent in 2001 and eliminated in 2003), thereby forcing the move of these expenditures on-budget and eliminating the inefficient turnover tax.

Fiscal Support for Macroeconomic Objectives

One of the implications of Russia’s dependence on the energy sector is that fiscal revenues will be volatile, reflecting both fluctuations in international energy prices, which affect energy sector outcomes, and the second-round effects of these changes on the broader economy. There will be times of relatively high oil prices with high revenues, as in 2000–02, and there will be periods of very low oil prices such as those in 1998, which put great strain on the fiscal accounts. In order to smooth fluctuations in aggregate demand in the economy as a whole, it is desirable for fiscal expenditures to be at least neutral to fluctuations in energy prices and possibly even negatively correlated (countercyclical) where appropriate. Unfortunately, many economies with significant natural resource exports do not achieve this goal, instead ending up with procyclical fiscal policies as expenditures tend to drift upward when revenues are high and get squeezed when revenues (and external financing options) decline.

The government was fairly successful in 1999 and 2000 in resisting pressures to raise expenditures in the face of the higher-than-anticipated revenues. However, expenditures have crept up in 2001 and 2002 and the overall fiscal stance has been loosened. While the loosening has not threatened the sustainability of the fiscal position, it does raise the question of whether the existing fiscal framework is adequate to resist pressures for procyclical expenditure policies or whether it would be helpful to introduce a formal rule/mechanism to try to encourage the adoption of the desired fiscal stance. There are various options for such a rule, ranging from a simple fiscal rule to a formal stabilization fund (as has been introduced in Azerbaijan and Kazakhstan). The challenge is to design a workable and transparent structure that can be implemented: many countries with stabilization funds still end up with procyclical fiscal policy (Davis and others, 2001).

Steps to Improve Budget Execution

Unlike in the precrisis years of repeated sequestrations, budgetary allocations have been fully funded since 1999 and have even been increased in the context of supplementary budgets late in the year. However, budgetary institutions have not fully adjusted to this new reality. Funding is extended by the federal treasury and the resources remain unspent on spending unit accounts. Late in the year, the spending units try to catch up on the implementation of their budgets, resulting in a massive flurry of expenditure in the last few weeks of the year. The macroeconomic impact of this is clear: a sharp spike in liquidity at end-year that is difficult for the CBR to offset, causing a depreciation in the exchange rate and imparting additional (unnecessary) momentum to inflation. This pattern can be seen at end-1999, at end-2000, and at end-2001. More consistent budgetary execution during the year combined with a credible commitment from the federal government that unspent balances by spending units would not be recentralized on January 1 of the following year would help to resolve the issue.

Improving Transparency and Accountability of Government

There is no question that much has been achieved in improving the transparency and accountability of government since the crisis. However, a number of additional measures should be undertaken to further strengthen public sector financial management. Many of these are technical in nature (see Diamond, 2002a, 2002b) but key areas include:

• Expanding the budget documentation to include a comprehensive report on contingent liabilities and a statement of fiscal risk arising from changes in macroeconomic parameters;

• Greater clarity in the role of the government sector, including reviewing quasi-fiscal activity by selected key public financial and nonfinancial enterprises, replacing the existing implicit subsidies with better targeted transparent subsidies, and addressing the commercial activities of state entities through a combination of bringing these operations on-budget with corporatizing and subsequently privatizing the activity;

• Enhancing the credibility of public expenditure management by strengthening external checks on budget performance through developing external audit. In line with international experience, the supreme audit institution—the chamber of accounts—should be free to pursue a systematic, independently determined audit program, with timely access to comprehensive budget execution data;

• Ensuring a consistent approach to debt management through greater clarity of roles for the various agencies involved in this area and some mechanism to ensure the pursuit of a coherent strategy;

• Further strengthening of the central treasury, including by strengthening the quality of accounting and fiscal reporting, to ensure both enhanced control over expenditures and effective monitoring of budget execution; and

• Continuing attempts to develop treasuries at subnational levels, bring off-budget activities into the budgetary sphere, and strengthen expenditure controls of extrabudgetary funds.