1. Taxation

Under socialist central planning, the major sources of government revenue are turnover taxes, various tax and nontax levies on enterprise earnings, and wage taxes, including social security contributions. ^4/ With the possible exception of wage taxes, the tax system is neither transparent nor parametric (i.e., the tax liability does not bear a predetermined functional relationship to a well-defined legal base). In particular, enterprise taxation is exposed to highly discretionary and uneven enforcement, including the requisition of profits.

The turnover tax, levied on goods and services usually at the distribution or production stage, is calculated as a residual wedge between the retail price and the cost of production, normally with provision for wholesale and retail profit margins. Given the planners’ reluctance to change prices, these product-specific implicit tax rates change commensurately with changes in costs of production; less frequently, tax rates are increased to raise revenue. It is not unusual to find thousands of variable tax rates in such a system. However, in recent years, in an effort to introduce a general turnover tax and as part of a price reform, some countries have reduced the number of rates while converting them into explicit fixed rates. ^1/

The reduction in the number of turnover tax rates is regarded as an indispensable step toward a value-added tax (VAT), which was introduced in Hungary in January 1988 and in the former GDR in July 1990 (under the terms of the state treaty concluded with the FRG in May), ^2/ and announced in Czechoslovakia and Poland for adoption by 1992 and in Yugoslavia by 1994. Adoption of the VAT, to be levied at very few rates, is intimately connected with price liberalization. Under freely determined output and input prices, the tax is neutral with respect to factor use and trade flows and should assist in promoting allocative efficiency.

Following a two-year preparation, Hungary replaced the turnover tax with the VAT at 25 percent standard and 15 percent reduced rates, while retaining selected excises. A large number of basic commodities (especially food products) and most services, accounting for roughly one half of the potential, tax base, is effectively exempt from the tax. ^3/ Unlike Hungary’s relatively high standard rate and narrow base, the former GDR has fully adopted the FRG’s VAT base and rate structure (14 percent standard and 7 percent reduced rates) in line with the average EC rates and EC Commission’s proposal for VAT harmonization.

In the past, foreign trade taxes operated in a manner similar to turnover taxes. Product-specific import and export taxes or subsidies (partly in the form of differentiated foreign exchange coefficients, tantamount to a multiple exchange rate system) were applied largely as wedges between controlled domestic prices and import or export prices, respectively. During the 1980s, this system had been partially replaced in some countries with fixed-rate customs duties on imports from the convertible currency area. By the end of 1990, with the shift to convertible currency transactions among CMEA member countries, the traditional system of price equalization taxes was to be abandoned practically in all Central and East European countries. The rationale for fixed-rate trade taxes—with as few rates as possible—is, of course, buttressed by the introduction of a flexible exchange rate and domestic price liberalization. Also, the reform of trade taxation is to proceed pari passu with the phase-out of state trading and quantitative restrictions. This has been virtually completed in Hungary, Poland, and Yugoslavia, while the former GDR’s trade regime was integrated with that of the FRG.

In a number of countries enterprise taxation and/or transfer of profits are still enforced largely through a bargaining process—^1/ in part due to a lack of proper accounting practices—as socialized enterprises strive to maximize the deductibility of bonuses to managers and of contributions to the enterprise’s social fund (for expenditure on employee facilities, etc.) and to its development or investment funds (for certain investment outlays). In January 1989, these taxes, often levied at differentiated rates across different forms of establishments, along with various other taxes, were replaced in both Hungary and Poland by a uniform company income tax, currently at a 40 percent statutory rate (lowered from 50 percent in January 1990 in Hungary), and in Czechoslovakia at a 55 percent rate to be fully phased in by 1992. In June 1990, the USSR adopted a 45 percent basic company income tax rate, effective January 1991. For the most part, calculation of the tax base has remained unchanged: very large number of conservative straight-line depreciation rates, virtually no allowance for loss carryover (in Hungary a meager two-year period is provided), and nondeductibility of certain interest costs. Thus, generally, effective tax rates in these countries remain high—especially on equity-financed investment—relative to rates in market-oriented European countries that otherwise have a broadly comparable investment environment. ^2/

However, perhaps reflecting a subconscious planning instinct, the authorities have singled out a number of specific activities (the “winners” as against “losers”) for reduced company income taxation. In particular, tax holidays ranging from two to five years, tax-free reinvestment of earnings, and/or reduced statutory tax rates are granted to foreign investors. Since these incentives are normally not available to residents, they discriminate heavily against domestic entrepreneurs. ^3/ Internally, the main beneficiaries of preferential treatment are agriculture, food processing, and export-related activities. Also, ad hoc tax reliefs were provided to enterprises facing financial difficulties, but many of these have now reportedly been eliminated. As of January 1990, Poland abolished tax preferences accorded to specific activities, including exports—consistent with the introduction of a realistic exchange rate policy.

Potentially the most contentious element of tax reform in these countries involves personal income taxation. The point of departure is typically a mixture of schedular taxes: proportional and/or mildly progressive wage taxes withheld at source, and taxes on gross profits or income from self-employment in the secondary economy, some of them levied at highly progressive rates, with top marginal rates of up to 95 percent in certain countries. However, for the most part, under central planning, the income tax burden is borne implicitly at a highly progressive rate structure by labor in the form of centrally set wage levels that deviate significantly from the marginal product of labor—the higher the excess of the marginal product over the wage, the higher the implicit tax rate. Analogously, the interest rate on savings deposits, the only recognized private income from capital, is implicitly taxed to the extent it is subject to a ceiling below the rate (compounded by shortages and uncertainty) of time preference between present and future consumption. Thus, the switch to a global progressive personal income tax can be viewed as an explicit, neutral (across most income sources), and potentially fairer revenue-raising instrument, to complement wage liberalization—in the same way that unified commodity taxation is a logical complement to price liberalization.

In January 1988, Hungary introduced a nearly comprehensive personal income; tax, albeit still retaining a number of schedular features. The initial rate structure exhibited a sharply progressive segment above a relatively high exempt threshold, with a 60 percent top marginal rate. While progressivity was mitigated and the top marginal rate cut to 50 percent by January 1990, the proliferation of tax preferences proceeded unchecked. Besides a number of deductions for various types of income, agricultural income is subject to tax only above a high exemption level. Interest income is subject to a 20 percent final withholding tax. At the time of introduction of the personal income tax, wages and salaries subject to withholding were grossed up by the new tax so as to leave after-tax incomes unchanged. Effective July 1990, the USSR enacted a personal income tax with somewhat differentiated structures by type of income, though in all instances the top marginal rate is 60 percent. The former GDR has broadly adopted the FRC’s income tax system. Neither the company income tax nor the personal income tax legislation in any of these countries provides indexing of the tax base for inflation—an issue that is likely to command attention in the future.

A number of criticisms have been leveled in Hungary at the personal income tax on technical grounds: definition of the legal base by source of income rather than at the household level, sharp rate progressivity, and imperfect wage gross-up at introduction, among others. Apart from these technical flaws, most of which can be (or have been already) corrected, personal income taxation is regarded by some observers as entailing an invasion of privacy. ^1/ This view, which does not extend to other forms of taxation (including withholding taxation or implicit taxation through wage controls), may have to be taken into account in the design and implementation of global personal income taxation in neighboring countries that share a similar cultural background.

More generally, the tax reform undertaken thus far shows considerable progress toward transparency for both the taxpayer and the authorities, greater uniformity, neutrality, and objectivity. However, the opportunity for streamlining the system has not been fully realized. The proliferation of tax preferences (including virtual exemption of the agricultural sector and foreign direct investment) under all three major tax categories has prevented adoption of relatively low marginal rates; hence, allocative distortions and supply-side disincentives remain. An important lesson is that tax reform should be oriented simultaneously to tax rate reduction and base expansion, following the lead of present tax reforms in industrial countries. ^2/ This approach is particularly relevant for economies in transition which, lacking effective tax administration and given a traditional reliance on withholding taxation through state-owned enterprises that operated under a soft budget constraint, are likely to suffer large-scale evasion of company income tax and social security contributions (typically set at relatively high rates) by newly privatized enterprises.

In the absence of sufficiently developed demand management tools, including an effective incomes policy, and upon relaxation of planning, most countries felt compelled to rely on microeconomic intervention through tax measures, regulation or ad hoc controls—largely on the basis of second-best arguments. Thus, to contain the expansionary drive of socialized enterprises, the authorities had relied extensively on taxes on investment and/or depreciation (replaced with the company income tax in Czechoslovakia, Hungary, and Poland). ^1/ Moreover, in the USSR, as an antimonopoly device, profits in excess of a standard rate of return are to be subject to marginal tax rates of up to 90 percent, instead of the basic rate.

Similarly, given the state-owned enterprises’ propensity to award wage increases—particularly when permitted by some degree of wage liberalization—a steeply progressive, in principle prohibitive, tax was imposed on wage increases above either (a) a predetermined average percentage increment consistent with the inflation target, or (b) the actual increase in the enterprise’s value added. Originally adopted in Hungary in the 1970s, Poland in the mid-1980s, and the USSR in late 1989, and subject to occasional modification, taxes on excess wage increases continue to be applied. In Poland, the present adjustment program includes such a tax in the form of (a), with marginal rates of up to 500 percent, that has successfully contained wage pressures. Whereas a major drawback of approach (a) is its likely restraining effect on productivity, ^2/ alternative (b) tends to be ineffective given the close relationship between the wage bill and value added. In either case, the tax is justified as a temporary device until the introduction of financial discipline on enterprises or of an effective incomes policy. In Yugoslavia, under an incomes policy adopted as part of the adjustment program launched in December 1989, the wage level was to be maintained unchanged in nominal terms.

2. Subsidies

In centrally planned economies, a large share of government expenditure consists of consumer and producer price subsidies and transfers to socialized enterprises. In addition, a number of implicit subsidies, especially in the form of preferential interest rates are provided through the banking system to households and enterprises.

Basic commodities (bread, sugar, milk, meat, pharmaceuticals, children’s clothing, heating fuel) and services (rental housing, mass transport) are usually sold at below-cost priced to satisfy merit wants. The difference between unit costs of production and retail prices determines the magnitude of these product-specific price subsidies, the rates of which are, of course, subject to frequent change. In other words, they operate as negative turnover taxes.

Artificially depressed prices often create widespread shortages and associated hoarding, queuing, and black market transactions for these commodities and certain services (e.g., informal transfer of housing rental rights). Although price subsidies are intended to benefit everybody, in fact their availability can be quite regressive, as incomes in the informal economy far exceed ordinary wage and salary levels. Poland was the first country that moved decisively to eliminate, with few exceptions, price controls and subsidies, between August 1989 and January 1990. Similarly, since mid-1990, subsidies have been abolished in the former GDR, except where they are in line with EC regulations. Following a more gradual approach, in Hungary and Yugoslavia about two thirds and three fourths of retail prices, respectively, have been freely determined since late 1989. ^1/ In July 1990, Czechoslovakia raised on average by 25 percent retail prices on a very large number of commodities and provided a fixed income transfer to households; although budget neutral, the combination of these measures permitted a welcome adjustment in relative prices. There are, of course, justified exceptions to the elimination of price subsidies. Specifically, the retention of subsidies on passenger mass transport is warranted on grounds that, as a public service, they confer substantial external benefits, including alleviation of urban traffic congestion and environmental pollution.

Socialized enterprises have been major recipients of subsidies specified either with respect to an activity or product, or simply as an ad hoc grant. Product-specific subsidies (mostly on material inputs, such as coal) were provided typically to promote, for example, heavy industry. Also, in the past, subsidies were extended to permit socialized enterprises to honor CMEA trade contracts. Over the last decade, in some highly indebted countries (e.g., Poland), enterprises engaged in production of convertible currency exports received budgetary subsidies, besides tax reliefs and preferential foreign exchange allocation. At the beginning of 1990, Hungary started to phase out budgetary support of CMEA-related activities, while Poland halted subsidies and tax preferences for convertible currency exports as the exchange rate became a major instrument of external adjustment. Besides CMEA-related grants, discretionary transfers (as well as easy credit) have been made available to socialized enterprises experiencing financial difficulties ^1/ following relaxation of planning and increased enterprise independence, in the absence of market discipline. However, under partial liberalization and decentralization, not all enterprises face a soft budget constraint. ^2/ Instead, profitable enterprises tend to bear a relatively high tax burden, while loss-making enterprises enjoy a host of subsidies and tax preferences.

Among the implicit interest rate subsidies, a potentially important subsidy has been extended through long-term mortgages (30 or 40 years) at low fixed nominal interest rates (3 percent or less) in Czechoslovakia, Hungary, and Poland. As inflation accelerates and nominal interest rates are allowed to increase to positive real values, as part of the financial reform, the subsidy element increases. For the sake of transparency, in Hungary since 1989 and in Czechoslovakia as of 1990, the state budget explicitly reimburses the banking system for the subsidy element. But, apart from this presentational improvement, the authorities have not been able to discontinue the subsidy element on outstanding housing loans, as this can be interpreted as a breach of contract. ^3/ Early phase-out of this subsidy—while allowing limited grand-fathering on existing mortgages—is important because of its negative effect on the budget, the concomitant inhibiting effect on a flexible interest rate policy, and the distortion it creates in the housing market.

The removal of price subsidies can make the most immediate and significant contribution to fiscal consolidation, with the most visible supply-side impact in an economy of shortage. While arguably abolition of price subsidies does not necessarily have adverse equity implications in a dynamic context, many low-income households are likely to suffer in the short run. Therefore, price subsidies cannot be trimmed in isolation but must be accompanied by a social safety net—discussed in the following section.

3. Social security

A fundamental tenet in socialist countries—usually enshrined in the constitution—is the right to work and the right to social benefits, including access to free health care and education. The sources of financing these benefits are payroll taxes and general revenues, usually channeled through the state budget; in turn, the benefits are distributed and administered through the workplace. In principle, assuming that the economy is on a high steady-state growth path and that labor productivity and demographic forces remain stable, these social goals can be realized in such an internally consistent conceptual framework, since the absence of poverty and unemployment obviates social assistance and unemployment compensation schemes.

However, most of these countries have experienced decelerating economic growth, decline in labor productivity, and unfavorable demographic trends (i.e., resulting in an increased old-age dependency ratio), while attempting to relax eligibility requirements for benefits and adherence to universal access. In various degrees, these trends led to a rise in the payroll tax burden, which so far has been relatively easy to enforce, ^1/ and were accompanied by a substantial deterioration in health care benefits and erosion in the real value of cash benefits, made available to deserving as well as undeserving beneficiaries. In addition, the slowdown in economic activity resulted in substantial concealed unemployment and growing social needs. Work force redundancies not only were hidden in socialized enterprises, but also imposed an excessive burden on certain benefits, such as sick pay and disability or early retirement pensions, serving in part as substitutes for unemployment compensation.

Under these circumstances, price liberalization and enterprise restructuring are likely to exacerbate open unemployment and push an increasing number of households below the poverty level. ^2/ This requires major efforts to establish a social safety net with an appropriate administrative apparatus to ensure cost effectiveness. Such a safety net would include: social assistance consisting of cash or in-kind benefits, subsidized rents and mortgages, etc., targeted to low-income households; automatic indexation of old-age pensions for inflation; unemployment compensation, subject to the usual eligibility requirements, for income support during short-term unemployment; ^3/ and manpower training and retraining programs, possibly accompanied by wage subsidies to private employers, to alleviate structural unemployment. Training should be particularly helpful to assist industrial restructuring, especially as regards the labor force engaged in the production of nontradables or of exports for the CMEA market, endowed with obsolete skills and a relatively weak work ethic. Also, some training could be provided in combination with temporary employment in public works—warranted by much needed investment in infrastructure.

The task ahead in this area can be expected to be monumental. For the reasons discussed earlier, the authorities for the most part are ill-equipped to create a cost-effective social safety net. The administration of social welfare payments to households below the poverty level, left largely to local jurisdictions, is likely to be quite uneven, especially in countries (e.g., Yugoslavia) with a high degree of regional autonomy. Training programs, if available, are too specific and inadequate for the tasks in demand (especially services and exports to convertible currency markets). So far, Hungary in 1989, followed by Bulgaria, the former GDR, and Poland in 1990, have introduced nationwide unemployment compensation schemes that provide benefits to a fraction of the admittedly modest number of unemployed, except in the former GDR and Poland which have allowed for a rapid rise in open unemployment.

In addition to establishing new social security schemes, European countries in transition will need to overhaul their existing social security programs especially by: tightening eligibility for old-age and disability pensions and sick pay provisions; introducing automatic indexation of old-age, disability and survivors’ pensions; enhancing the cost-consciousness of both patients and medical care providers; rationalizing health care benefits and substituting open remuneration for the current gratuity payments; and allowing for private pensions, private insurance, and private medical practice. In May 1990, the USSR enacted a pension reform law which provides a considerable expansion of benefits, including indexation, but without tightening eligibility requirements. In July, the FRG’s pension system was extended to the former GDR. Only modest progress has been made on the health care front—limited mostly to Hungary—with important potential benefits in terms of budgetary savings, improved resource allocation, and above all better quality of benefits at an affordable cost to the needy.

4. Public investment

Capital formation to promote economic growth occupies a prominent place in the public sector of centrally planned economies. Traditionally, public investment, through both the state budget and socialized enterprises, was undertaken according to the central plan, in areas that would maximize net material output, especially in heavy industry—regardless of consumer preferences, resource requirements, or environmental impact. Also, investment in defense and in domestic and joint CMEA prestige projects was favored. ^1/ Typically, investment projects were large-scale and took a long time to complete because of resource shortages coupled with considerable waste and pilferage. As the budget constraint tightened—often on the strength of external credit limitations—many of these projects were abandoned, trimmed, or simply not started, and capital expenditure declined. However, in the 1980s, high rates of capital formation and high incremental capital-output ratios, by international standards, were still visible in most European centrally planned economies. The low efficiency of investment expenditures was reflected not only in global statistical indicators but equally in individual projects and in the performance of socialized enterprises. Further, the neglect of investment in social infrastructure, housing, consumer durables, and services became progressively obvious.

By the late 1980s, most countries began to curtail public investment projects and capital transfers to socialized enterprises. Nevertheless, none of these countries was ready yet to initiate a systematic and comprehensive review of public investment at various levels of government and to apply standard social rate of return criteria that account explicitly for external economies or diseconomies. ^1/ Of course, such a review would be limited to government-sponsored projects through the budget or extrabudgetary funds, whereas industrial investment by socialized enterprises must be predicated on private rate of return and risk considerations. In any event, investment undertaken by government agencies or socialized enterprises should be prevented from crowding out—in physical or financial terms—incipient private capital formation. A distinct risk that lies in the period ahead in some countries in transition is a possible overreaction against all forms of public investment, including investment in the production of essential public goods and services, which cannot be expected to be provided by the fledgling private sector but are highly complementary to private goods and services.

5. Public enterprises

In socialist economies, nonfinancial state-owned or socialized enterprises account for virtually the entire productive capacity in the economy and are, in many ways, indistinguishable from the general government. The state, on the one hand, is the legal owner of these enterprises through the founding or branch ministries and, on the other, it has economic authority over them by levying taxes, granting subsidies or setting regulations; in either case, the enterprises’ goals are subservient to the central plan. Key management decisions are exercised by branch ministries.

First in Yugoslavia, and more recently in Hungary and Poland (and to a much lesser extent in the USSR), the locus of decision-making has shifted increasingly to the enterprises, in various institutional forms (e.g., workers’ councils in Poland and Yugoslavia, and enterprise councils in Hungary), while the role of branch ministries was substantially reduced or eliminated. Increased enterprise self-management has not remedied the lack of a well-defined objective function (in essence, a mix of social, growth and profit maximizing goals) and the continued prevalence of a soft budget constraint for loss-makers and a relatively hard one for profitable ones. ^1/ Also, the monopoly and/or monopsony position of large enterprises that do not face the risk of bankruptcy—as neither the antimonopoly nor the bankruptcy laws enacted in the 1980s in Hungary, Poland, and Yugoslavia were effectively applied—has not been conducive to allocative and managerial efficiency.

As the initial enthusiasm over privatization in 1989 in both Hungary and Poland was displaced by controversy and disillusionment, it has become evident that the sale of socialized enterprises to domestic or foreign investors will take considerably longer and will be more limited in scope than envisaged earlier. Nevertheless, considerable efforts are now under way in Czechoslovakia, Hungary, Poland, and the former GDR in privatizing state-owned enterprises in a speedy, orderly, and transparent manner. ^2/

In any event, it is necessary to redefine the goal of the remaining socialized enterprises along commercial lines and to expose them fully to a uniformly hard budget constraint and to the risk of bankruptcy. Difficult as it may be, against the above background of several decades, the government would have to exercise unambiguously its ownership right and hold management accountable for the enterprise’s profitability, and at the same time, as authority, maintain an arm’s-length relationship with the enterprise, rather than engage in bargaining over taxes, subsidies, or bankruptcy action. Although recently a number of steps—in some cases, through legislative action—have been made in this direction in Czechoslovakia, Hungary, Poland, and Yugoslavia, a thorny outstanding issue is the future role of the workers’ councils or enterprise councils in management and the extent to which professional managers—a scarce resource in these countries—will be effectively in charge of enterprise operations.

Hungary, Poland, and Yugoslavia have announced the implementation of stepped-up antimonopoly campaigns (including the breakup of large enterprises) and the initiation of bankruptcy action by commercial banks, which in turn should be subject to strengthened banking supervision and to rules on prudential practices. Further, the authorities declared their intention to Launch an orderly process of rehabilitation of enterprises facing financial difficulties. Against an unimpressive track record in this area, however, it is not yet clear how successful these efforts will be in the immediate future.

More generally, socialized enterprises should be encouraged to strengthen their financial structure. It is widely recognized that this involves shifting reliance from outside financing (easy credits and subsidies) to retained earnings, and undertaking investment projects on the basis of risk and return criteria. By now, the earlier practice of transferring all after-tax earnings to the budget or to extrabudgetary funds has been abandoned in most countries. Thus, the question remains as to how the government should exercise its claim as owner and dispose of socialized enterprise earnings. In both Hungary and Poland, the simplest solution was adopted, namely, to specify an ex ante dividend payout requirement. In Poland, since 1989, each socialized enterprise has been obliged to pay a dividend in fixed proportion to the enterprise’s equity, equivalent to the interest rate on refinancing credits, set by the National Bank of Poland; but, as in the case of the company income tax, preferential treatment was accorded to enterprises engaged in certain activities. In Hungary, as of January 1990, socialized enterprises have been liable for an 18 percent dividend-earnings ratio. Albeit serving as imperfect proxies for market-determined yields on equity, in the absence of preferential treatment these rules are objective and simple. Further, along with enterprise income tax reform and phase-out of subsidies, they contribute to transparency in the relationships between the government (both as authority and as owner) and non-financial socialized enterprises.

Traditionally, official financial institutions (including the state-owned commercial banks, following the banking reform in Hungary and Poland) have been regarded as a component of the state budget and subservient to the central plan. As such, these institutions served fully as fiscal agents, collecting implicit taxes (beyond those associated with seignorage) and allocating implicit subsidies. This arrangement violated most criteria that underlie effective fiscal and monetary control and efficient financial intermediation. To be sure, some progress has been made in certain countries (especially in Czechoslovakia, Hungary, and Yugoslavia) toward transparency by shedding the quasi-fiscal functions of the banking system and setting up separate commercial banks. However, these banks have not yet been fully exposed to bankruptcy risk ^1/ or to required accounting for nonperforming loans, implying a concealed subsidy element. In the meantime, legislation is under consideration in several countries to endow the central bank with some independence from the government in the conduct of monetary policy.

6. Government debt

In centrally planned economies, budget deficits are financed typically with central bank credit subject to neither an interest charge nor an amortization schedule. Along with generous bank lending to public enterprises, credit to the government contributes to the accumulation of an inflationary overhang—in the presence of price controls and import barriers. Occasional attempts at placing government paper with the nonbank public met with little success owing to inadequate interest yields; hence, enterprises and their employees were obliged to hold such paper or counterpart deposits, as done repeatedly, for example, during the last decade in the USSR.

With an increasingly active interest rate policy, recently some countries (notably Hungary) began to experiment with limited periodic auctions of government obligations to the nonbank public at market terms. Clearly, the development of financial markets should be assisted with the introduction of negotiable instruments with a wide range of maturities and yields to suit a variety of portfolio needs. As part of this process, it is necessary to establish an external and internal debt management capacity. Ultimately, however, the authorities should prevent the buildup of public debt to near unsustainable levels in relation to GDP, which—as illustrated by a number of Western countries, including industrial ones—can be very difficult to reverse or even to contain especially in the absence of sufficient economic growth.

In the near term, economies in transition face the task of cleaning up the fledging commercial banks’ portfolio of nonperforming loans inherited from the central bank, following the banking reform. This task emerges with the application of prudential bank regulations and the bankruptcy law and is compounded with the advent of possibly high market-clearing nominal interest rates. A similar problem is encountered with regard to outstanding bank liabilities (including mortgages) extended at low fixed nominal interest rates and long maturities. ^1/

In view of the limited scope for absorbing rapidly such losses—mitigated by a gradual phase-out of the interest subsidy on outstanding loans—through the buildup of bad debt reserves by banks or through wider intermediation margins, shifted mainly onto bank lending rates, the government may have to assume the bulk of this burden. This solution may entail financial transfers from the budget (or from an extrabudgetary fund created for this purpose, such as in Hungary’s Housing Fund) for the amount of the subsidy, or takeover of the nonperforming debt in exchange for government paper, with a resulting interest charge shown in the budget (or the extrabudgetary fund). ^1/ On a parallel track, the government may be compelled to take over the external debt service obligations (e.g., on convertible currency liabilities of Polish enterprises) or the nonperforming external credits (e.g., CMEA export credits extended by the Czechoslovak state bank) of public enterprises.

As part of the solution of the bad debt problem, the proceeds from the privatization of state-owned enterprises should be earmarked primarily to offset the stock of government liabilities, rather than to finance the growth of ordinary budgetary outlays. All told, the cleanup of nonperforming liabilities, which may take a number of years, is essential for the unencumbered conduct of monetary policy and the competitiveness and viability of the initial commercial banks (i.e., spinoffs of the central bank) relative to newly established domestic and foreign-owned banks that can start up with a high grade portfolio.

7. Fiscal decentralization

There are incipient forces to decentralize fiscal functions in some Central and East European countries that are likely to undermine the conduct of fiscal policy for stabilization purposes. Fiscal decentralization is manifest in the proliferation of extrabudgetary funds within the central government and in increased assignment of fiscal resources to regional and local governments.

Over the past decade, a number of government functions (e.g., housing, education, export promotion, foreign debt service, public investment, etc.) has been transferred from the state budget to extrabudgetary funds, financed from general revenue, earmarked taxes, or special financing from monetary institutions. This trend can be observed in varying degrees in Bulgaria, Czechoslovakia, Hungary, Poland, and Yugoslavia. ^2/ Admittedly, the creation of special-purpose funds may seem useful to protect the integrity of a particular government program from competing spending needs, or to enhance the visibility of a program—thereby strengthening the popularity of the program and of the government. However, relatively few programs (e.g., social security schemes) genuinely warrant a separate status insofar as they involve coneingent government liabilities. ^1/ Otherwise, there is a very real risk of relinquishing budgetary scrutiny with the concomitant erosion of control over the fiscal stance.

In most of these countries, including the USSR, there is a strong undercurrent of popular discontent—rooted in cultural, ethnic, and regional attributes—with the legacy of a highly centralized structure of public administration. Modeled after the concept of local councils (soviets), first implanted in the USSR, republican, provincial, county, or municipal jurisdictions have been subservient economically to the central planning agency and ultimately to party authority within each country. This explains much of the centrifugal movement toward autonomy, responsive to regional or local demands and challenging the central authority—even when this authority may now be constituted by elected officials and responsible to a multi-party parliament. While the fiscal implications of this movement are not yet entirely clear, local and regional authorities have claimed an increasing share of general government revenue and assets over which they can dispose freely.

Regional claims on fiscal resources are perhaps most pronounced and deeply entrenched in Yugoslavia, where the federal administration controls only around one third of general government revenue—primarily from indirect taxes—and has inadequate data on republican and local government finances. Moreover, a significant portion of the federal government share consists of earmarked tax revenue. Although more recent, the movement toward fiscal federalism has become almost equally intense in the USSR, where consideration was given to shifting most revenue and spending functions to the republics, which in turn would make transfers to the union budget for certain common expenditures on the basis of GNP (or GNP per capita) in each republic; ^2/ at any rate, this issue appears to be far from settled. ^3/ In response to a more moderate drive toward local autonomy, Bulgaria, Czechoslovakia, Hungary, and Poland are also considering changes in revenue-sharing arrangements—often involving considerable bargaining—between levels of government, increasingly in favor of the local jurisdictions. ^4/

Clearly, there is a case for some degree of fiscal federalism, based on the fact that a number of government functions can be most effectively attended to at the regional or local levels. Certain public services (education, social assistance, etc.) and infrastructure investment projects (hospitals, roads, etc.) are at least in part the responsibility of local governments. Therefore, it is necessary to provide local governments with sufficient resources—either through tax earmarking and/or intergovernmental budgetary transfers—to meet these responsibilities. However, these resources should be provided, or shared with other levels of government, on the basis of objective and transparent rules, ^1/ to be accompanied by appropriate budgeting, accounting, and expenditure monitoring at the local level. The absence of such rules, of safeguards limiting the access to foreign and domestic borrowing by local governments, and of overall intergovernmental fiscal coordination, can be detrimental to public debt management and to the use of fiscal policy for macroeconomic purposes. This is a particular source of concern for a country that needs a major and sustained stabilization effort, yet lacks sufficiently developed macroeconomic instruments. ^2/

1. Restructuring government finances

The attraction of Central and East European countries to the instilutional framework of West European countries is evident in many areas, including the public sector. Repeated statements by leaders and policymakers from economies in transition attest to a strong desire to move closer to, or to join, the European Community (EC) and other West European-based institutions. It is partly in this context that the economies in transition intend to substitute private ownership and supply-side incentives for direct state ownership and interference in the enterprise sector; to create an adequate social safety net, replacing price subsidies; to develop financial markets; to engage in external liberalization; to clean up of the environment; and to rein in the military establishment. These fundamental policy objectives have important implications for the level and composition of government operations. Hence, an examination of the differences in size and structure of government finances in these countries and in the EC should provide a broad gauge of the extent of the changes necessary in economies in transition to approximate the fiscal structure of Western Europe in general and of EC member countries in particular. Admittedly, such comparisons are too crude to capture differences in the qualitative and administrative aspects of each country’s revenue system and budgetary practices that prevail in former socialist countries—even after the initial reform steps taken in some of these countries. Also, they ignore state interference, through regulation and planning, which characterized until recently these countries’ enterprise sector.

Above all, comparisons for the second half of the 1980s (Tables 1 and 2) indicate that the size of general government operations in most Central and East European countries exceeded by a wide margin that in EC member countries. In Bulgaria, Czechoslovakia, and Hungary general government operations accounted for about 60 percent of GDP or more, ^1/ as against less than one half of GDP in the Community. As an exception, Yugoslavia’s general government expenditure has risen to barely 40 percent of gross social product (GSP), which is a relatively low share even if corrected for a downward bias. ^2/

Table 1.

Selected European Countries: Summary of General Government Operations, 1985

(In percent of GDP/GNP)

Sources: International Monetary Fund (1990), national authorities, and author’s estimates.

^1/Data shown in percent of an adjusted measure of gross social product.

^2/Weighted average of EC member countries, with weights given by GDP/GNP converted at market exchange rate, excluding Luxembourg and Greece. Data for Italy and Portugal refer to central government.

^3/Including entrepreneurial income from state-owned enterprises and financial institutions. In addition, for the USSR, including transfers to extrabudgetary centralized funds, and for Yugoslavia, mandatory interenterprise transfers.

^4/Product-specific turnover taxes, excises, sales tax, and value-added tax.

^5/Including other tax and nontax revenue and grants.

^6/Consist of social security benefits and other targeted transfers, and may include transfers to certain nonprofit institutions.

^7/Including product-specific price subsidies, explicit interest rate subsidies, and debt service on behalf of enterprises and institutions. In addition, for the USSR, inducing transfers from extrabudgetary centralized funds, and for Yugoslavia, mandatory interenterprise transfers.

^8/Including transfers abroad as well as unidentified transfers.

Table 1.

Selected European Countries: Summary of General Government Operations, 1985

(In percent of GDP/GNP)

		Bulgaria	Czechoslovakia	Hungary	Poland	Romania	USSR	Yugoslavia 1/	European Community 2/
Revenue		54	56	60	49	49	48	35	44
	Enterprise income taxes and transfers 3/	19	23	16	12	15	16	7	5
	Individual income taxes	4	7	1	1	—	4	1	9
	Social security contributions	10	6	13	9	6	3	16	13
	Domestic taxes on goods and services 4/	13	17	17	14	20	13	7	11
	Revenue from international trade	2	1	3	3	—	8	3	—
	Taxes on property	—	—	3	2	—	—	—	2
	Other revenue 5/	6	2	8	8	8	4	1	4
Current expenditure		48	51	53	42	28	41	34	44
	Expenditure on goods and services	25	23	19	15	19	24	9	17
	Interest payments	1	—	1	—	1	—	—	5
	Transfers to households 6/	10	13	13	11	7	8	12	16
	Subsidies and transfers to enterprises 7/	13	15	21	16	1	9	13	3
	Other current expenditure 8/	—	—	—	—	—	—	—	3
Capital expenditure		8	7	8	7	18	9	—	3
	Investment	8	5	7	—	—	⋯	—	2
	Transfers	—	2	1	7	18	⋯	—	1
Net lending		—	—	—	—	—	—	—	1
Overall balance (surplus/deficit)		-1	-1	-1	-1	3	-2	1	-4

Sources: International Monetary Fund (1990), national authorities, and author’s estimates.

^1/Data shown in percent of an adjusted measure of gross social product.

^2/Weighted average of EC member countries, with weights given by GDP/GNP converted at market exchange rate, excluding Luxembourg and Greece. Data for Italy and Portugal refer to central government.

^3/Including entrepreneurial income from state-owned enterprises and financial institutions. In addition, for the USSR, including transfers to extrabudgetary centralized funds, and for Yugoslavia, mandatory interenterprise transfers.

^4/Product-specific turnover taxes, excises, sales tax, and value-added tax.

^5/Including other tax and nontax revenue and grants.

^6/Consist of social security benefits and other targeted transfers, and may include transfers to certain nonprofit institutions.

^7/Including product-specific price subsidies, explicit interest rate subsidies, and debt service on behalf of enterprises and institutions. In addition, for the USSR, inducing transfers from extrabudgetary centralized funds, and for Yugoslavia, mandatory interenterprise transfers.

^8/Including transfers abroad as well as unidentified transfers.

View Table

Table 1.

Selected European Countries: Summary of General Government Operations, 1985

(In percent of GDP/GNP)

		Bulgaria	Czechoslovakia	Hungary	Poland	Romania	USSR	Yugoslavia 1/	European Community 2/
Revenue		54	56	60	49	49	48	35	44
	Enterprise income taxes and transfers 3/	19	23	16	12	15	16	7	5
	Individual income taxes	4	7	1	1	—	4	1	9
	Social security contributions	10	6	13	9	6	3	16	13
	Domestic taxes on goods and services 4/	13	17	17	14	20	13	7	11
	Revenue from international trade	2	1	3	3	—	8	3	—
	Taxes on property	—	—	3	2	—	—	—	2
	Other revenue 5/	6	2	8	8	8	4	1	4
Current expenditure		48	51	53	42	28	41	34	44
	Expenditure on goods and services	25	23	19	15	19	24	9	17
	Interest payments	1	—	1	—	1	—	—	5
	Transfers to households 6/	10	13	13	11	7	8	12	16
	Subsidies and transfers to enterprises 7/	13	15	21	16	1	9	13	3
	Other current expenditure 8/	—	—	—	—	—	—	—	3
Capital expenditure		8	7	8	7	18	9	—	3
	Investment	8	5	7	—	—	⋯	—	2
	Transfers	—	2	1	7	18	⋯	—	1
Net lending		—	—	—	—	—	—	—	1
Overall balance (surplus/deficit)		-1	-1	-1	-1	3	-2	1	-4

Sources: International Monetary Fund (1990), national authorities, and author’s estimates.

^1/Data shown in percent of an adjusted measure of gross social product.

^2/Weighted average of EC member countries, with weights given by GDP/GNP converted at market exchange rate, excluding Luxembourg and Greece. Data for Italy and Portugal refer to central government.

^3/Including entrepreneurial income from state-owned enterprises and financial institutions. In addition, for the USSR, including transfers to extrabudgetary centralized funds, and for Yugoslavia, mandatory interenterprise transfers.

^4/Product-specific turnover taxes, excises, sales tax, and value-added tax.

^5/Including other tax and nontax revenue and grants.

^6/Consist of social security benefits and other targeted transfers, and may include transfers to certain nonprofit institutions.

^7/Including product-specific price subsidies, explicit interest rate subsidies, and debt service on behalf of enterprises and institutions. In addition, for the USSR, inducing transfers from extrabudgetary centralized funds, and for Yugoslavia, mandatory interenterprise transfers.

^8/Including transfers abroad as well as unidentified transfers.

Table 2.

Selected European Countries: Summary of General Government Operations, 1989

(In percent of GDP/GNP)

Sources: International Monetary Fund (1990), national authorities, and author’s estimates

^1/For Poland, Belgium, Portugal, and the United Kingdom, data refer to 1988; for Ireland and Spain, data refer to 1987.

^2/Preliminary data shown in percent of an adjusted measure of gross social product.

^3/Weighted average of EC member countries, with weights given by GDP/GNP converted at market exchange rate, excluding Luxembourg and Greece. Data for Italy and Portugal refer to central government.

^4/Including entrepreneurial income from state-owned enterprises and financial institutions. In addition, for the USSR, including transfers to extrabudgetary centralized funds, and for Yugoslavia, mandatory interenterprise transfers.

^5/Product-specific turnover taxes, excises, sales tax, and value-added tax.

^6/Including other tax and nontax revenue and grants.

^7/Consist of social security benefits and other targeted transfers, and may include transfers to certain nonprofit institutions.

^8/Including product-specific price subsidies, explicit interest rate subsidies, and debt service on behalf of enterprises and institutions. In addition, for the USSR, including transfers from extrabudgetary centralized funds, and for Yugoslavia, mandatory interenterprise transfers.

^9/Including transfers abroad as well as unidentified transfers.

Table 2.

Selected European Countries: Summary of General Government Operations, 1989

(In percent of GDP/GNP)

		Bulgaria	Czechoslovakia	Hungary	Poland 1/	Romania	USSR	Yugoslavia 2/	European Community 1/ 3/
Revenue		57	60	61	47	53	46	41	44
	Enterprise income taxes and transfers 4/	23	17	14	14	15	18	8	4
	Individual income taxes	4	7	6	1	—	4	5	9
	Social security contributions	9	15	16	7	7	4	18	13
	Domestic taxes on goods and services 5/	11	18	18	13	19	12	6	11
	Revenue from international trade	5	1	3	2	—	6	2	—
	Taxes on property	—	—	—	2	—	—	—	2
	Other revenue 6/	5	2	5	7	12	2	2	5
Current expenditure		52	55	57	40	27	48	41	42
	Expenditure on goods and services	24	24	22	13	15	21	9	16
	Interest payments	3	—	3	—	—	1	—	5
	Transfers to households 7/	10	13	16	10	11	8	19	15
	Subsidies and transfers to enterprises 8/	15	19	16	17	1	18	13	3
	Other current expenditure 9/	—	—	—	—	—	—	—	3
Capital expenditure		5	6	7	7	18	8	—	4
	Investment	5	4	5	1	—	⋯	—	3
	Transfers	—	2	1	6	17	⋯	—	1
Net lending		3	8	—	1	—	—	—	—
Overall balance (surplus/deficit)		-3	-9	-2	-1	8	-10	—	-3

Sources: International Monetary Fund (1990), national authorities, and author’s estimates

^1/For Poland, Belgium, Portugal, and the United Kingdom, data refer to 1988; for Ireland and Spain, data refer to 1987.

^2/Preliminary data shown in percent of an adjusted measure of gross social product.

^3/Weighted average of EC member countries, with weights given by GDP/GNP converted at market exchange rate, excluding Luxembourg and Greece. Data for Italy and Portugal refer to central government.

^4/Including entrepreneurial income from state-owned enterprises and financial institutions. In addition, for the USSR, including transfers to extrabudgetary centralized funds, and for Yugoslavia, mandatory interenterprise transfers.

^5/Product-specific turnover taxes, excises, sales tax, and value-added tax.

^6/Including other tax and nontax revenue and grants.

^7/Consist of social security benefits and other targeted transfers, and may include transfers to certain nonprofit institutions.

^8/Including product-specific price subsidies, explicit interest rate subsidies, and debt service on behalf of enterprises and institutions. In addition, for the USSR, including transfers from extrabudgetary centralized funds, and for Yugoslavia, mandatory interenterprise transfers.

^9/Including transfers abroad as well as unidentified transfers.

View Table

Table 2.

Selected European Countries: Summary of General Government Operations, 1989

(In percent of GDP/GNP)

		Bulgaria	Czechoslovakia	Hungary	Poland 1/	Romania	USSR	Yugoslavia 2/	European Community 1/ 3/
Revenue		57	60	61	47	53	46	41	44
	Enterprise income taxes and transfers 4/	23	17	14	14	15	18	8	4
	Individual income taxes	4	7	6	1	—	4	5	9
	Social security contributions	9	15	16	7	7	4	18	13
	Domestic taxes on goods and services 5/	11	18	18	13	19	12	6	11
	Revenue from international trade	5	1	3	2	—	6	2	—
	Taxes on property	—	—	—	2	—	—	—	2
	Other revenue 6/	5	2	5	7	12	2	2	5
Current expenditure		52	55	57	40	27	48	41	42
	Expenditure on goods and services	24	24	22	13	15	21	9	16
	Interest payments	3	—	3	—	—	1	—	5
	Transfers to households 7/	10	13	16	10	11	8	19	15
	Subsidies and transfers to enterprises 8/	15	19	16	17	1	18	13	3
	Other current expenditure 9/	—	—	—	—	—	—	—	3
Capital expenditure		5	6	7	7	18	8	—	4
	Investment	5	4	5	1	—	⋯	—	3
	Transfers	—	2	1	6	17	⋯	—	1
Net lending		3	8	—	1	—	—	—	—
Overall balance (surplus/deficit)		-3	-9	-2	-1	8	-10	—	-3

Sources: International Monetary Fund (1990), national authorities, and author’s estimates

^1/For Poland, Belgium, Portugal, and the United Kingdom, data refer to 1988; for Ireland and Spain, data refer to 1987.

^2/Preliminary data shown in percent of an adjusted measure of gross social product.

^3/Weighted average of EC member countries, with weights given by GDP/GNP converted at market exchange rate, excluding Luxembourg and Greece. Data for Italy and Portugal refer to central government.

^4/Including entrepreneurial income from state-owned enterprises and financial institutions. In addition, for the USSR, including transfers to extrabudgetary centralized funds, and for Yugoslavia, mandatory interenterprise transfers.

^5/Product-specific turnover taxes, excises, sales tax, and value-added tax.

^6/Including other tax and nontax revenue and grants.

^7/Consist of social security benefits and other targeted transfers, and may include transfers to certain nonprofit institutions.

^8/Including product-specific price subsidies, explicit interest rate subsidies, and debt service on behalf of enterprises and institutions. In addition, for the USSR, including transfers from extrabudgetary centralized funds, and for Yugoslavia, mandatory interenterprise transfers.

^9/Including transfers abroad as well as unidentified transfers.

As to the composition of government operations (Charts 1 and 2), the comparisons reflect a heavy reliance of Central and East European governments on enterprise income transfers and turnover tax revenue for redistribution through price subsidies and current or capital transfers back to enterprises. After the enterprise income tax reform, both Hungary and Poland continued to draw substantial revenue from such taxes, in contrast to EC member countries where the corporate income tax yield is relatively insignificant. The case of Yugoslavia stands apart from other socialist countries in that labor-managed enterprises retain most, profits and are subject to a lower formal income tax liability.

CHART 1

Selected European Countries: General Government Revenue, 1985 and 1989

(In percent of total revenue)

View Expanded

Sources: International Monetary Fund (1990) and national authorities.

CHART 2

Selected European Countries: General Government Expenditure, 1985 and 1989

(In percent of total expenditure)

View Expanded

Sources: International Monetary Fund (1990) and national authorities.

In Central and East European countries, the explicit tax burden on households’ income or property for the most part has been either very low or nonexistent as compared with EC member countries. The modest individual income tax revenue shown for Bulgaria, Czechoslovakia, and the USSR represents the yield from wage taxes; for 1989, Hungary shows the receipts from the newly introduced progressive individual income tax. While, on average, social security contributions by both employers and employees constitute a major component of general government revenue of EC member countries, the revenue from such contributions, levied almost solely on employers, varies considerably across economies in transition, representing a considerable portion of government revenue in Hungary and Yugoslavia.

By EC standards, the share of outlays on goods and services was large—though with a relatively small wage bill component as wage rates were compressed at low levels—in most former socialist countries. The magnitude of outlays on material inputs can be ascribed to inefficiency and waste and in some instances (notably the USSR) to substantial spending on defense and internal security. Meanwhile, unlike many EC member countries, Central and East European governments made insignificant interest payments—except in some cases on external liabilities—given low or zero interest rates charged on domestic public debt held mostly by central banks.

A major question that emerges from these comparisons involves the extent to which economies in transition can expect to approximate the fiscal structure of the EC. Whereas over the medium run strides can be made toward reducing the size of the public sector through stepped-up privatization, the restructuring and contraction of government operations are likely to take more time. While there is ample scope for phasing out price subsidies and transfers to loss-making public enterprises, for trimming the government payroll, and in some countries for curbing military outlays, the resultant savings are likely to be offset by new demands on government expenditure associated with a comprehensive reform effort. These spending requirements arise in connection with the restructuring of public enterprises (often preceding privatization), establishment of a targeted social safety net, investment in infrastructure, environmental cleanup, market wages for civil servants, and market interest rates on public debt. The possible scale of such expenditures is nowhere illustrated more dramatically than in the transformation of the former GDR—to conform closely with the institutional structure of the FRG—while being exposed to market forces prevailing in the EC.

Thus, conceivably, the size of government outlays may increase beyond the initial stage of the transformation. Moreover, at the same time, tax revenue is likely to stagnate or decline due to supply-side rate cuts, proliferation of tax preferences, and inadequate administrative capacity to contain the erosion in tax collections as the economy becomes increasingly privatized. In sum, the combination of upward pressures on government expenditure and a potential fall in revenue poses a serious risk of widening fiscal imbalances during the transition. ^1/

2. Speed, sequencing, and consistency

A major dilemma facing Central and East European countries is the speed at which economic transformation from command to market economy should take place. ^1/ In other words, does the preferred strategy involve a big bang or gradualism? In general, the speed of the adjustment is dictated by the economy’s openness and the extent of external and internal disequilibria. At one extreme is the former GDR, where it has been practically unavoidable to proceed with transformation at a break-neck speed in the context of German unification. Because of the extent of the internal disequilibrium, reflected in high and accelerating inflation fueled by a wage-price-exchange rate spiral, both Poland and Yugoslavia have been compelled to adjust very rapidly. Near the other extreme, the apparently milder imbalance experienced by Czechoslovakia allowed some room to debate the appropriate pace of adjustment.

The speed of implementing fiscal reform measures is, of course, to be viewed within such a broad context. Yet even if a big bang strategy is adopted, fiscal reform cannot be achieved overnight, not only on account of technical and institutional constraints but also because of attitudes and practices conditioned over several decades in a command economy. As an example of the first set of constraints, tax reform is contingent on the administrative capacity—in terms of manpower, computer facilities, accounting standards—of a given country. To illustrate the second point, while reform of the legal and institutional arrangements between the government and state-owned nonfinancial enterprises or commercial banks can be implemented rather quickly, the reorientation of actual operations of these enterprises—and the attitudes that ultimately shape them—along commercial lines will surely take place over a longer period of time. ^2/

The sequencing of policies is particularly crucial when a country faces the double objective of stabilization-cum-liberalization. The standard prescription—that a front-loaded stabilization must be accompanied by structural measures aimed at liberalizing domestic commodity markets, followed by domestic financial liberalization and foreign trade liberalization, and finally by removal of all capital controls—^1/ is broadly applicable to European economies in transition as well. However, in many respects, sequencing in these countries is far more complex and diverse. Again, the case of the former GDR stands out insofar as that adjustment must be pursued on all fronts at once, by restructuring the domestic sector and by opening up both the external current and capital accounts simultaneously. In other words, the former GDR has to achieve the transformation from having been perhaps the most disciplined centrally planned economy to full-fledged membership in the European Monetary System (EMS) within a period of less than a year, a feat that no other major country has attempted or achieved in recent memory. In a sense, this transformation can serve almost as a laboratory experiment, illustrating the pressure points that arise without sequencing but which may be relieved with budgetary transfers from the FRG. As for other economies in transition, in general it would be desirable to initiate or announce most structural and institutional market-oriented reforms at the outset, even if the pace of implementation were sequenced. At the very beginning of the transition, however, it is essential within each country to arrive at the broadest possible consensus—if necessary through a formal constitutional accord among regional governments—about the basic rules that will govern intergovernmental economic relations, and, in particular, the division of fiscal resources and responsibilities among various levels of government, subject to binding limits on their budget deficits and their financing.

In the case of countries suffering either major internal and external disequilibria (Poland and the USSR), or primarily an internal disequilibrium (Yugoslavia), or mainly an external imbalance and large indebtedness (Bulgaria and Hungary), it is essential to combine demand management and exchange rate adjustment with structural measures that reinforce, and are consistent with, the stabilization objective. Accordingly, in the fiscal area, it is particularly important to phase out quickly most product-specific price subsidies and enterprise transfers that cannot be justified in terms of external benefits, and to fix and unify turnover tax rates (or at least reduce their number) in tandem with substantial price liberalization. Meanwhile, it is necessary to design and implement early in the transition a social safety net, consisting of unemployment compensation and targeted social assistance in the form of cash payments or benefits in kind. In addition, it is important to abolish immediately interest rate subsidies on mortgages and to phase out over a couple of years such subsidies on outstanding mortgages.

The combination of these front-loaded measures should be very useful in several respects. Elimination of most subsidies (especially those pegged on prices and interest rates) would have an immediate corrective effect on the budget deficit, and thus exercise—along with a commensurate cutback in credit to the public sector—a restraining effect on domestic demand, prices, and the external imbalance. At the same time, the accompanying price liberalization would assist in improving the allocation of resources and alleviating commodity shortages, and, together with interest rate flexibility, it would help drain excess liquidity and stimulate financial intermediation.

Much of the budgetary saving from the elimination of price subsidies and transfers would be offset, probably with some lag, by the cost of targeted subsidies, indexation of pensions, unemployment compensation, manpower training, and wage subsidies. Nevertheless, unemployment benefits and training are key conditions for the broad acceptability of the adjustment program and the authorities’ willingness to encourage industrial restructuring and dismissal of redundant workers. The latter would also help contain the upward wage drift that is likely to accompany wage liberalization. And, if the social safety net were to impose a heavier-than-anticipated burden on the budget (i.e., if not adequately compensated for by subsidy cuts or if revenue declined), the authorities should stand ready to implement contingency measures, such as reduction in government employment and increases in excise rates.

At the second stage of the adjustment, which may start as early as the end of the first year of the transition, wage liberalization should proceed in earnest, possibly in combination with an effective incomes policy or a tax on excess wage increases. This tax—preferably based on the increase in the enterprise’s wage bill, in reference to an ex ante inflation target—would be imposed temporarily, that is, as long as it takes to cool wage inflation and to develop an acceptable and orderly collective bargaining mechanism or a binding agreement among social partners. Also in this second stage, the authorities should begin preparations (subject to administrative capacity constraints) to replace the turnover tax with a broad-based value-added tax that would serve as the main revenue engine for the budget. Income taxation of enterprises and individuals should be unified and streamlined as soon as feasible thereafter. Moderate statutory rates (with a relatively mild progressivity on personal incomes) should be imposed on a broad definition of taxable income, allowing only a very small number of special-purpose exemptions (e.g., nonprofit institutions) and deductions. This effort would need to be assisted by a major improvement in tax administration over the medium term. ^1/ Low effective income tax rates would be made possible by administrative improvements and a strengthened fiscal position on account of the abolition of most subsidies, especially in countries that, upon having cut a high rate of inflation (e.g., Poland and Yugoslavia), may experience an endogenous surge in tax revenue (and to a lesser extent, a fall in government expenditure), reversing the earlier trend of lengthened tax collection lags. ^2/

At a third stage, the complete removal of quantitative trade restrictions would permit rationalization of the import tariff structure, aimed at raising some revenue and at setting up a time-limited moderate effective rate of protection. In no way should tariffs, non-tariff barriers, and export subsidies be used as a substitute for a realistic exchange rate policy—to be adopted from the beginning of the transition.

During the transition and extending well into the medium term, the authorities should pursue the restructuring of public enterprises as well as their privatization in a speedy yet transparent manner. The revenue from such asset sales should be earmarked to write off outstanding government liabilities rather than regarded as ordinary revenue for the state budget. The restructuring and more competitive behavior of enterprises, assisted by the dismantling of quantitative trade restrictions, would further dampen price pressures. Meanwhile, it is unlikely that sizable foreign direct investment inflows would materialize—despite generous tax holidays and other investment incentives—at the initial stages of the transition period. Direct investment from abroad is usually attracted by sustained favorable macroeconomic performance (combination of steady growth, decelerating inflation, gains in market shares, etc.); a stable, transparent, and credible tax and regulatory environment; clearly defined ownership rights; an adequate infrastructure (working communication facilities, etc.); and a sufficiently trained but inexpensive labor force. A major incentive, of course, would be conferred at the final stage of the transition by the phase-out of exchange and payments restrictions (i.e., the establishment of a fully convertible currency).

Other fiscal reform areas where continuous, steady progress is necessary through various stages during the transition, involve public investment, government debt, and social security. The conduct of a review of public investment projects and implementation of an appropriate selection process is a largely technical exercise, under the constraints of external credits and domestic macroeconomic considerations. Development of a financial market for government securities requires appropriate institutional and technical groundwork. By contrast, reform of the social security system might be subject to a protracted political debate. Even on purely technical grounds, some social security reform measures (e.g., an increase in the retirement age) are likely to be implemented over the medium to long term, dictated by various contractual, equity, and cultural considerations, as well as demographic and labor market pressures. All along, it is essential that the sequencing of reform measures in these areas be consistent with the maintenance of macroeconomic stability.