Fiscal Reform Over Ten Years of Transition

Contributor Notes

Authors’ E-Mail Address:,

This paper analyzes fiscal reforms in transition economies during the decade leading up to 1998. The paper argues that macroeconomic stabilization, price liberalization and privatization—the core reforms visualized by the shock therapy approach—are necessary but not sufficient conditions for a complete transition to a market economy. Further deep changes—such as the creation of new fiscal institutions, changes in incentives and processes, and changes in the role of government—are needed.


This paper analyzes fiscal reforms in transition economies during the decade leading up to 1998. The paper argues that macroeconomic stabilization, price liberalization and privatization—the core reforms visualized by the shock therapy approach—are necessary but not sufficient conditions for a complete transition to a market economy. Further deep changes—such as the creation of new fiscal institutions, changes in incentives and processes, and changes in the role of government—are needed.

I. General Considerations

Over the past decade, we have experienced what is arguably one of the largest economic experiments of this century. Many of the socialist countries that had a centrally planned economy embarked in a process of change aimed at transforming them into market economies. This transformation involved massive changes across a broad spectrum of economic policies, institutions and practices.

With a decade or so of transition behind us, what is our assessment of what has happened so far?2 What have we learned from developments over the past decade? In some countries and in some areas, the transition process has gone very far. In other countries and in other areas, it still has a long way to go. In some countries, one gets the impression that the old system is largely gone, but a new system is far from coming into existence. Thus, in some instances, a kind of institutional vacuum has developed.

For the transition process to be complete, countries will need to develop the institutions that are necessary for private markets to operate successfully and establish a government role that is complementary rather than adversary to a market economy. This paper focuses on the role that public finance has played in the transition process and on the progress made in creating the fiscal institutions that are essential for the proper functioning of a market economy.

A. Initial Conditions

To set the stage for the discussion that follows, some of the essential characteristics of the economic environment that prevailed at the beginning of the transition are presented. At that time, the share of GDP derived from private sector activities was very small in all of the transition countries. It was less than one percent in countries such as Czechoslovakia and Russia and reached a high of almost 20 percent in Poland, which, at the time, was the transition country with the largest private sector. By contrast, the share of GDP derived from the private sector is about 80 percent in the United States. Thus, economic production in the centrally planned economies was overwhelmingly in the public sector because few productive assets could be privately owned and few private activities were allowed. The government was everywhere in the economy.

Prices and genuine economic profits did not play much of a role in the allocation of resources because the use of resources was determined by political decisions made within the planning office. Detailed plans provided quantitative targets for enterprise inputs and outputs, as well as for foreign trade flows. Prices were fixed, primarily to facilitate the establishment and the execution of the plan. In market economies, political decisions have a major influence on the use of public resources, but much less in the remainder of the economy. In the environment that prevailed prior to the onset of transition, it did not make much sense to speak of “public finance” because the existence of public finance presupposes that of private finance. The countries did not need market-type tax systems to raise public revenue because the government could simply appropriate the share of total, and mostly public, production for its own needs. Given that it owned practically everything, the government would decide how it wanted to divide and use the total output. State enterprises were allowed to hold only one account in the monobank, and all transactions (including tax payments) were to go through this single account. Taxes, in this environment, were mostly transfers from one activity to another, particularly in light of the distortions that were introduced by the fixed price and exchange rate regime imposed by the plan. Thus, there was little need for a tax administration, in the sense of those found in market economies. The function of the tax administrators was mostly to ensure that funds were transferred to the government’s account and that they were properly accounted for. More surprising is the fact that there was no budget office, budget law, or treasury in these economies. Budgetary allocations to various functions and programs were made through the plan and the treasury function focussed on its cash-allocation role and was performed by the monobank. In this environment, there could not be well defined or fixed rules of law that could be appealed to by an individual or an enterprise that disagreed with the actions of the government.

Many fiscal activities, which are typically identified with the government in market economies, were not provided by the government, but rather by the state-owned enterprises. These state enterprises provided a range of services to workers and their families, including housing, schooling, vocational training, medical care, and pensions. In addition, enterprises would hire workers because they were required to do so and not necessarily on the basis of need; consequently, there was no official unemployment. As a result, there were no well-developed social safety net programs aimed at protecting unemployed workers.

B. First Phase of Reform

There has been a lot of writing about the economic changes that would have to take place for these countries to become market economies. In the “shock therapy” approach advocated by various economists, including Jeffrey Sachs, during the early phase of the transition, the main elements of the transition were assumed to be privatization, price liberalization, and macroeconomic stabilization. While the advocates of “shock therapy” were not explicit about it, they conveyed the impression that these elements would be sufficient to take these economies from where they were to where, at the time, their policymakers indicated that they wanted to go, namely to become market economies.

Focusing only on the “shock therapy” aspect of the transition gives the impression that much progress has been made (see EBRD, Transition Report, 1998). The private sector share in GDP which was almost insignificant ten years ago, has risen dramatically in many countries (Table 1). In particular, it has reached and, in some cases, exceeded 70 percent in countries like Albania, Czech Republic, Estonia, Hungary, Lithuania, the Russian Federation, and the Slovak Republic. Only in Belarus, Tajikistan, and Turkmenistan has the private sector share in GDP remained below 35 percent. It should be noted, however, that one is looking here at the privatization of ownership, and not necessarily that of management. In many cases despite the advent of privatization, either the management has remained unchanged or the current management has continued to behave as if the enterprises continued to be state-owned.

Table 1

Progress in Transition in Central and Eastern Europe, the Baltic States and the CIS 1/

article image
Source: EBRD Transition Report, 1998.

The index ranks countries as follows: from 1 (little progress) to 4 (much progress).

One intriguing aspect of the privatization experience is the low correlation between what has happened to the state ownership, as discussed in the previous paragraph and shown in Table 1, and the fiscal proceeds from privatization. As mentioned previously, the state owned almost everything before the transition and everything probably meant a multiple of GDP in total value.3 Yet, the fiscal revenue that the government got from the sale of the state assets was minuscule (see Table 2). Privatization proceeds in transition economies as shares of GDP were, in fact, much smaller than in many developing or developed market economies that privatized their public enterprises (with the notable exception of Kazakhstan where privatization centered around the primary sector).

Table 2

Privatization Proceeds

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

There are several reasons for this outcome, but the results are the same. A small group of individuals used their positions often as management insiders, or their connections, to amass enormous wealth.4 This was a fire sale to which only a privileged few were invited. In Russia, for example, it was reported that assets valued at US$50-60 billion were bought for US$1.5 billion. Often the funds used to purchase the assets were obtained from banks controlled by these individuals. As Frye and Shleifer (1998) have commented, when individuals become so rich by essentially raiding the public treasury, would a society be inclined to pass laws that protect private property? As a result, privatization, which must be a fundamental step toward a market economy, becomes in and of itself an obstacle to the enforcement of a fundamental prerequisite for a market economy, namely the protection of private property.

Nomenclature privatization and other similar developments have dramatically changed the income distribution of these countries. Before the transition, these countries had some of the most even income distributions in the world. The first column of Table 3 shows the Gini coefficients for the period 1987-88. In most countries, they were in the low 20s, which was very good by international standards. By the mid 1990s, the Gini coefficients had increased sharply and in Ukraine, Russia and the Kyrgyz Republic had reached values seen only in a small group of developing countries.5 The coefficients have probably increased further more recently. What is worse is that this increase in inequality has occurred not because wealth was created by those who moved up in the income distribution, but because wealth was raided from the government, at times with the implicit connivance of the government.6

Table 3

Changes in Income Inequality During Transition

article image
Sources: United Nations Development Program (1996); Milanovic (1998). Adopted from Kolodko (1998). Note: For most countries, the income category for 1993-95 is disposable income. In 1987-88, it is gross income, since, at that time, personal income taxes were small, as was the difference between net and gross income. Income includes consumption in-kind, except for Hungary and Lithuania in 1993-95.





Countries where income differences had been sharply compressed in the past, found themselves, within a few years, with some of the richest men and women in the world living a life of conspicuous consumption. Some of these new rich individuals also acquired substantial political power. It is easy to guess the reaction of the populations of these countries toward the economic changes that allowed this to happen. It is also easy to understand why the development of a market economy, which is identified with these changes, is blamed for what happened. In such an environment, many of the measures which are necessary to make a market economy vibrant and efficient will be seen by the majority of the population as measures taken to protect the ill-gotten wealth of the new rich. It should therefore be no surprise if such measures do not encounter an easy time in the political process. In conclusion, it should not be surprising if privatization is not seen as a sign of positive progress toward a market economy.7 At some point, the governments of these countries will have to take seriously their role in promoting equity.8 It remains to be seen what form this role will take; though, it is not likely that it will be one which is friendly to a market economy.

There has been much progress in the liberalization of prices and in some other reforms. Table 1 (borrowed from the 1998 EBRD report) provides some useful information. It ranks countries on the basis of an index that goes from 1 (little progress) to 4 (much progress). While some countries show little progress (Belarus, Uzbekistan, Tajikistan), most show some progress and a few display significant progress (Hungary, Poland). Of course, while a movement from 1 toward 4 is welcome and is an indication of progress toward complete price liberalization, it should be noted that in a second-best world the removal of one restriction, while others remain in place, is no guarantee that efficiency has increased. If the remaining price controls are in large and vital sectors such as energy, they may imply large distortions. This point is not generally appreciated.

Figure 1, also taken from the 1998 EBRD Transition Report, is a useful visual rendition of the pattern of shock therapy reforms summarized in Table 1. This would lead one to conclude that much has been accomplished but there is still a long way to go.

Figure 1
Figure 1

Regional patterns of reform

Citation: IMF Working Papers 2000, 113; 10.5089/9781451853698.001.A001

Source: EBRD Transition repot, 1998.

In an interesting paper, Shleifer (1997) pointed out that, while price liberalization eliminates government control over prices, stabilization imposes a harder budget constraint on the government, and privatization deprives the government of the direct control that it previously had over firms. These changes may be sufficient to destroy a centrally planned economy but not to change it into a market economy. For that to occur, many other deeper changes would have to take place. And, for sure, “shock therapy” was silent about the role that the government, and in particular fiscal agencies, should play in the new economic world.9

C. Further Reforms

A process of transition cannot be assumed to be complete at the point when prices have been liberalized, state enterprises have been privatized and the macroeconomy has been broadly stabilized. To be complete, the transition process needs a far deeper transformation of the economy, institutions and processes. Shock therapy deals with the easy or superficial part of the transition. A mature transition requires that most prices be liberalized; that many public enterprises are not only privatized, but are privatized in a process that is seen as transparent and as fair as possible. Furthermore, not only the ownership of enterprises, but also their management must become free of political interference. In this environment, profitability becomes the guiding criterion for most investment decisions. Activities that are deemed to be socially desirable are financed by the government and not by enterprises, be these public or private. The government needs to perform well in its basic or core functions in the economy while it withdraws from, or drastically reduces its role in, many secondary or less basic activities. This is especially important with respect to its regulatory function.

Economies, like traffic, need some regulations. Like those that regulate traffic, the rules that regulate the economy must be few, must be clear, and must leave little scope to interpretation or discretion by either the citizens or the bureaucrats who enforce the rules. While the guiding principle under central planning was that nothing was permitted unless explicitly authorized, the guiding principle in a market economy should be that everything is permitted unless expressly forbidden. If this principle were followed, it would eliminate a substantial portion of unnecessary governmental interference in the economy and would reduce corruption.

To perform their tasks, governments need some well-developed institutions run by competent individuals who are guided by correct incentives. In other words, the objectives of the individuals running the institutions must not diverge from the objectives of the institutions, which must in turn be consistent with the public interest. Obviously, such institutions do not arise magically. They need to be created and continually reformed. In industrial countries, the creation of these institutions took centuries and in some cases, they are still being developed and improved. In transition economies, their creation had to be compressed in a much shorter time span and under much less favorable circumstances. Given that some of the skills needed by these institutions were not easily available domestically, it was inevitable that errors would be made.

To be well-functioning, market economies need governments that are efficient at establishing and enforcing the essential rules of the game (say on competition), at promoting widely shared social objectives, at raising needed revenue to finance public sector activities, at spending the revenue raised productively, at bringing needed corrections to, and controls over, the working of the private sector, at enforcing contracts and protecting property, at producing public goods, and so on. Without the constable at the street corner, the private sector game can become very rough.

When the needed public institutions do not exist or, if they exist, when the existing incentives for those who run these institutions are perverse, the government can easily become an impediment to economic activity because it ends up being used by those who control it for their own private gains. See Nagy (2000). That is what normally happens with corruption (see Tanzi, 1998). Corruption is the privatization of some parts of the government apparatus for the gains of special interest groups or particular individuals. In this case, the achievement of social objectives becomes difficult and some of the actions of the government acquire a predatory nature. Governments cease to be market-friendly and become instead an impediment to the proper working of the market. In this context, and as a commentary to the limitations of the shock therapy approach, it should be noted that governments have many ways, besides price controls, ownership of enterprises, and macroeconomic imbalances, to influence economic activities.

For a process of transition to be successful in the fiscal area, one needs to create necessary and well-working fiscal institutions, and reasonable and affordable expenditure programs, including the provision of basic safety nets against fundamental risks such as becoming unemployed or old or ill. These safety nets and expenditure programs must be similar to some of those that are common in market economies rather than those inherited from the past. The yoke of past commitments must be minimized. This is particularly important in the area of pensions and employment. A market economy cannot guarantee a job for everyone, thus making social provisions for unemployment necessary. The pension commitments by the government must be consistent with the new fiscal reality. It should be also noted that in an environment where real GDP is shrinking, as has been the case for many of these countries, budgeting becomes an extremely difficult activity. It is especially in this environment that the government must focus on essential social needs and reduce its commitments to realistic levels. This will require major reforms.

The final destination of the transition journey must contain spending programs that can be financed from public revenue generated without creating excessive burdens on the private sector.10 Because public expenditures are financed primarily by tax revenue and the level of taxation of a country, ceteris paribus, depends on its economic development and on the sophistication of its tax system and tax administration, this revenue constraint must be kept in mind when determining the programmed level of public spending. Once spending plans are made, they may be difficult to change especially in a downward direction. This is especially relevant for expenditures (such as pensions, health benefits, and public employment) that involve long-term commitments and large number of beneficiaries. Developing countries have tax levels that are generally about half as high as those of the industrial countries. In terms of per capita income and economic structure, most of the economies in transition are, by and large, much closer to developing countries than to the industrial countries. Thus, one would expect that, over time, there would be a tendency for taxes to fall toward the range of 15-25 percent of GDP prevailing in richer developing countries even though, in some of the transition economies, the tax to GDP ratios might still be much higher at the moment. This points to the need for countries to be realistic in terms of the expenditure programs to which the governments commit themselves.11

Finally, because the desirable role of the government emerges not just from economic considerations but also from the interplay of political and economic forces, it is important that, in determining the role of government, the views of the executive branch of government broadly match those of the legislative branch. The political process that generates the legislature and creates the executive branch of government, of course, influences this outcome. If the executive and the legislative branches are far apart on what the government should do, as has been the case in Russia and in some other countries, it is unlikely that an optimal government role or rational policies can emerge until the two sides come closer. In a similar vein, clear rules must establish the fiscal responsibilities of the subnational and the national governments because the role of government cannot be defined only in terms of the national government.12 These rules cannot be too flexible and must be precise enough to indicate when a level of government is overstepping its limits. In several countries, the fiscal arrangements between national and subnational governments are vague and the activities of the local governments have been largely uncontrolled. For some of these countries, in spite of the importance of subnational governments in the fiscal area, a kind of iceberg mentality has developed whereby all the attention has gone to the visible part (the national government) while the role and the activities of the subnational governments have been largely ignored. In Russia, for example, there has been so much attention paid to the difficulties of the national government in raising revenue that a widespread perception developed that Russian taxes are low and that the solution to the Russian fiscal problem would necessarily be an increase in tax revenue rather than a rationalization of total spending or a reallocation of tax revenue between national and subnational governments (See Craig, Norregaard, and Tsibouris, 1997, for a description of the system of intergovernmental fiscal relations in Russia.)13

II. Fiscal Reforms in the Past Decade

The decade of the 1990s has seen major reforms in most transition economies. Some of these have been much more successful than others. As a consequence, as shown in Figure 1, there is a lot of differentiation among countries in terms of the pace of reform. In general, the East European and the Baltic countries have made rapid progress, while other countries have been less successful in establishing fiscal institutions, in controlling fiscal imbalances, and in redefining the role of the state. But even within these broad groups of countries, there is substantial differentiation that cannot be ignored. The discussion which follows provides a broad view of some of these changes, while setting aside some of the details.

A. Tax Policy and Administration

In the pre-transition economies, most tax revenue was obtained from three major sources: the turnover taxes; the taxes on enterprises; and the payroll taxes. Particular characteristics of the centrally planned economies made the collection of these taxes relatively simple. Some of these characteristics were: (i) the knowledge on the part of the authorities of the quantities produced by the state enterprises and of the prices at which the output would be sold (this knowledge was available from the plan); (ii) the role of the monobank in processing payments and the restrictions it imposed on how payments were to be settled; and (iii) the high concentration of economic activities in a few very large enterprises which made controls and tax collection much easier. These were reliable tax bases that generated very high tax revenue levels (at times up to 50 percent of GDP) without the need for full-fledged or sophisticated tax administrations. In the particular case of the USSR, general government revenue amounted to an estimated 41 percent of GDP in 1989 (see JSSE, 1991).

In the pre-transition period, taxes were not collected on the basis of detailed and codified tax laws that precisely defined tax bases and taxed them with parametric tax rates while at the same time respecting taxpayers’ rights. Rather, especially for taxes on enterprises, they were collected largely on the basis of negotiations between the enterprises and government officials. The government was always free to change the rates and, in fact, it often changed them (see Tanzi, 1994).14 In this environment, tax liabilities tended to be soft rather than well-defined and rigid obligations. Thus, negotiations could reduce or increase them. When an enterprise was in difficulty, the taxes were negotiated downward; when the government needed extra revenue, the taxes were negotiated upward. This left a legacy that, unlike death, taxes are not certain.

At the same time, this was an environment where most individuals never had direct contacts with the tax authorities. As most taxes were hidden from the people who finally bore them, most individuals were not even aware that, indirectly, they were paying a large amount of taxes. Thus, a “tax consciousness” or a tax culture never developed. This tradition created a hostile attitude toward the payment of explicit taxes brought about by the transition (see Kornai, 1997). This has made the imposition of a transparent tax system more difficult.

The impact of the transition on the traditional revenue system was radical and damaging. It resulted in a significant decline in tax revenue collections. The process of transition:

  1. Destroyed the plan and thus eliminated the information (good or bad) that the plan had provided on quantities produced and on prices at which the output was sold. The government now had to rely on other sources, including the declarations of the taxpayers, to get this information. As a consequence, the prospect of tax evasion rose;

  2. Eroded what were previously reliable tax bases, as a result of declines in production, in state enterprise profitability, and in real wages;

  3. Increased dramatically the number of producers and thus the number of potential taxpayers, as many private sector activities came into existence. Tax administrations that had been used to dealing with relatively few, friendly enterprises had to deal with hundreds of thousands or even millions of unfriendly taxpayers.15 Some of these administrations were hardly prepared for this change and were slow to adapt. Much of the growth originated from the new small and difficult-to-tax private producers, who required a lot of close attention on the part of the tax authorities because of their high propensity to evade taxes. At the same time, these small producers required protection from unscrupulous tax officials;16

  4. Removed the restrictions on the methods of payments among enterprises and taxpayers in general that had existed when all payments were channeled through accounts with the monobank. Unfortunately, in the new environment, payments in the form of barter and tax arrears have grown and have created major difficulties for the new tax system;

  5. Created production, income and profits in areas which grew proportionately faster relative to other sectors but were lightly taxed, such as agriculture, small services, and exports, or in areas that had not existed in the previous system, such as financial markets.

The changes mentioned above, and others not mentioned, indicate why the old systems could not simply be reformed on the margin. Rather, totally new tax systems, capable of operating in the new environment, were needed. However, new tax systems required not only new tax laws but also new fiscal institutions and new skills. Tax laws could be imported or even copied from other countries.

The new fiscal institutions, however, had to be created from scratch in circumstances that were far from ideal. These new fiscal institutions (such as the tax administrations) needed financial resources (for computers, new staff, etc.) and specialized skills (accountants, lawyers, computer specialists, auditors, etc.). At the same time, they needed clear strategies and objectives, well-defined legal powers vis-à-vis the taxpayers and well-defined legal obligations towards those taxpayers. This would establish clear rules of the game which, like Chinese walls, would separate them from the inevitable pressures—from powerful political figures in the legislative branch, in the executive branch, or in local governments—to reduce the tax liabilities of specific taxpayers or to close their eyes to tax arrears. All these requirements are necessary to establish a market-oriented system guided by the rule of law. However, they are very demanding in terms of time, resources, skills, technical knowledge and political capital. These requirements could be met by only few of the transition economies.17 In many of them, there was a lack of either the financial resources, the specialized skills, a clear understanding of what needed to be done, or the ability/willingness to insulate the day-to-day administrative side of the tax system from political interference.

In many countries, there was an attempt to patch the old institutions to make them behave like new ones. But as the saying goes, it is difficult to teach old dogs new tricks. Often the poorly paid staff, schooled in the old ways, was the main obstacle to change and those who were put in charge of these institutions often had scant knowledge of how tax administrations work in market economies. The incentives for these individuals were to maintain the old system. It would have been far better to create brand new institutions at the outset.

Many governments failed to accept or to understand that, in a market economy, a tax system should be a parametric tool with one overwhelming objective, namely that of raising revenue in as an efficient and equitable way as possible. Rather, the tax system came to be seen as a tool that should do many things including keeping failing enterprises in business, sustaining employment by allowing loss-making enterprises to pay wages instead of taxes, and stimulating economic activity. In some ways, the tax system replaced the plan as the key instrument for economic and social policy. Thus in some of these countries, taxes have continued to be soft or non-parametric especially for large enterprises, and key ministers have continued to spend much time dealing with individual taxpayers’ tax problems rather than in reforming the tax system (see Tanzi, 1998).18 This may have sharply increased the tax burden on the part of the economy that could not benefit from these preferential treatments.

Progress in tax reform has varied across countries in transition. Table 4, borrowed from a study conducted by an IMF staff team led by Ebrill and Havrylyshyn (1999), provides an assessment of progress in tax policy reform for the countries in the Baltics, Russia, and the other countries of the former Soviet Union (BRO) from 1992 through mid-1998. The index ranks the BRO countries from 1 (little progress in reform) to 5 (significant progress in reform). This is an unavoidably subjective ranking which, however, reflects some of the fundamental pillars of tax policy reform, including the adoption of a comprehensive tax code or of completely new laws for the major taxes, the establishment of the VAT, the implementation of a profit tax based on profits in a market economy sense, and the elimination of exemptions and preferences.

Table 4

Progress in Tax Policy Reform1

article image
Source: Staff team led by L. Ebrill and O. Havrylyshyn (1999)

Scale from 1 (very little appropriate market-oriented reform) to 5 (high degree of reform).

One would expect there to be a correlation between progress in tax policy reform and revenue performance. However, it should be noted that many of the reforms in tax policy have occurred relatively recently, thus not yet being fully felt, while certain reforms (such as the elimination of the export tax and the excess wage tax) are revenue reducing in the short run. In addition, many of the tax policy reforms have been hampered by difficulties in tax administration.

As a result of the structural changes during the transition, as well as the difficulties encountered in establishing new modern fiscal institutions in a timely manner, there was a

sharp contraction in tax revenue collections in practically all of the transition economies. Tables 5 and 6 provide data on tax revenue and on overall general government revenue (inclusive of grants) as shares of GDP since 1992 or, in the case of some Eastern European countries, since the late 1980s.

Table 5

Tax Revenue Developments, 1989-98

article image
Source: Data provided by the authorities and IMF staff estimates.
Table 6

Developments in General Government Revenue and Grants, 1989-98

article image
Source: Data provided by the authorities and IMF staff estimates. Also, JSSE (1991).

While the large fall in tax revenues since the late 1980s is evident for many of the countries, the data also show the wide range in tax burdens experienced among the countries. Countries that have lagged in terms of market reforms (e.g., Belarus, Tajikistan, Turkmenistan, and Uzbekistan; see Figure 1) have been able to minimize or delay the decline in tax revenue collections, primarily by continuing to resort to their traditional tax bases, including importantly the still large state enterprise sector. Countries that have been involved in wars or in domestic/border unrest (e.g., Armenia, Azerbaijan, Georgia, and Tajikistan) have seen, understandably, dramatic declines in tax collections, from which they have struggled to recover. These countries which experienced such unrest are now at the low end of the spectrum, with the ratio of taxes to GDP ranging from 12 percent in Tajikistan to about 15 percent in Azerbaijan in 1998. At the high end of the spectrum, one can find Czech Republic, Slovak Republic and Poland with tax burdens exceeding 37 percent of GDP with Hungary and Slovenia having the highest tax rate of 41.2 percent in 1998. All of these countries are among those that have made the most progress in terms of market based reforms (see Figure 1); these higher tax ratios are unlikely to be sustainable over the medium and long run.

The story is broadly similar when one looks at overall revenue collections inclusive of grants (Table 6). Revenue declines can be seen for most of the countries. Countries that have not gone far along the transition path have seen more modest revenue declines than others. At the same time, countries that have been at the forefront of the transition, both in terms of timing and actual reforms, have seen their revenue share in GDP maintained or, in the case of the Baltic countries, increased.

It is also revealing to look at individual components of tax revenue over the transition period. For illustration, tax revenue developments for two countries (Hungary and Russia) are presented in Figures 2 and 3.

Figure 2
Figure 2

Hungary: Tax Revenue Developments, 1991-1998

Citation: IMF Working Papers 2000, 113; 10.5089/9781451853698.001.A001

Figure 3
Figure 3

Russia: Tax Revenue Developments, 1992-98

Citation: IMF Working Papers 2000, 113; 10.5089/9781451853698.001.A001

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

For both countries, personal income taxes have remained roughly unchanged as a share of GDP over the period 1992-98, while payroll taxes have declined very modestly. Taken together, these taxes account for a significant share (somewhat under half) of overall tax receipts. Corporate profit taxes, on the other hand, have declined markedly (particularly in Russia), in large part for the reasons which are mentioned in the previous section. Overall, direct taxes have declined as a share of GDP, not only for these two countries but for most of the transition economies. In contrast to what has been happening in Hungary, revenue collections from the VAT have decreased as a share of GDP in Russia; something which may be in part attributable to the widespread use of barter transactions and to administrative difficulties. In the case of Hungary, Russia and many other of the transition countries, trade taxes have declined in importance, as countries have embarked in trade liberalization, while export duties have been minimized and import tariff rates reduced.

B. Expenditure Policy and Management

Despite the high tax ratios still prevailing in many of these countries, it did not prove possible to completely close fiscal deficits. In many of these countries, and especially in the larger ones, public spending levels have remained very high as shares of GDP. One reason is, of course, that many of these countries have experienced falls in their output which would have required extraordinary cuts in real spending to reduce the spending to GDP ratios, something that would have been very difficult to do given that many of the large expenditure categories, such as wages, and pension benefits, are not easy to compress in the short run. Another reason is that there has not yet been a well thought out policy of shrinking the role of the state. The government remains engaged in far too many activities.19

Data on the share of general government expenditure and net lending in GDP for 1989-98 are presented in Table 7. While expenditures have dropped across the board over the past decade, there has been significant differentiation in individual country experiences. To some extent driven by financing availability as a result of less unfavorable developments on the revenue side, the countries that have been the leaders in the transition process and those countries which have been the laggards have ended up with some of the more modest expenditure cuts. Countries that have experienced the largest revenue drops (particularly, the war-torn countries) have also faced the sharpest expenditure cuts; for example, Tajikistan has experienced expenditure cuts of 50 percent of GDP between 1992 and 1998.20

Table 7.

Developments in General Government Expenditure and Net Lending, 1989-98

article image
Source: Data provided by the authorities and IMF staff estimates. Also, JSSE (1991).

It should be noted that as some of the Eastern European countries have managed to maintain very high expenditure shares, it is not surprising that most of these countries are still experiencing high fiscal deficits (see Table 8). They have not recognized that, as market-oriented economies, their capacity to sustain levels of expenditures typical of rich welfare states will be limited. It would not be prudent to continue aiming for these high expenditure levels. Some saving in spending programs has come about through explicit government policy. For example, price and exchange rate liberalization has resulted in a reduction (in some cases, significant) of government subsidies to the economy.21 In addition, there have been reductions in military outlays (on a cash basis) in many transition countries.22 Despite the significant infrastructure needs of the transition economies, governments chose to give lower priority to capital spending, due in large part to inadequate financing. At the same time, other expenditure categories have placed increased demands on the budget. Interest payments have risen in response to the increase in government debt, the liberalization and emergence of positive interest rates, and the shift from subsidized directed lending to the budget toward more market-based debt instruments.

Table 8.

Overall General Government Balance (Cash Basis)

article image
Source: Data provided by the authorities and IMF staff estimates. Also, JSSE (1991).

Budgetary expenditures on social assistance have increased, as the government had to take over social functions that were previously being provided by enterprises, such as schools and hospitals. Interestingly, the government’s wage bill has remained broadly unchanged as a share of GDP over the transition period, with the partial monetization of non-cash benefits being broadly offset by a rather lackluster reduction in the size of the civil service overall.23 A large part of the expenditure cuts that have been seen in transition economies have not come about from a systematic reassessment of government priorities. Instead, in many cases, the adjustment has come about from sequestration, cash rationing, and non-payment. With overall resources being constrained by declining revenues and by limited access to non-inflationary financing, many countries found it politically infeasible to undertake an expenditure adjustment path arising from parliamentary support for a realistic budget. Instead, budgets were often passed by parliaments, with unrealistic assumptions about revenues and, as a result, budgeted expenditures that largely exceeded financial resources. This resulted in multiple rounds of expenditure reductions over the course of the fiscal year, in part through supplementary budgets, but also through sequestration and accumulation of arrears. In many cases, treasuries (which have been established in all of the transition economies) reverted to month-to-month credit rationing of budget institutions, something that was not particularly effective due to political pressures and resulted in expenditure

arrears. While this rationing process may help preserve the macroeconomic balance, it comes at the cost of corrupting the budgetary process. This raises an important question: if the government does not abide by its legal obligations, how can we expect that others will?

Treasury systems are being developed in all of the transition countries although in some of them, including Russia, their scope is still much too limited.24 Such systems aim to provide a comprehensive and often centralized payment, accounting, and financial management information service for the central government. In some cases, the coverage of the treasury system has been extended to some of the extrabudgetary funds and to lower levels of government. This has required a fundamental reform of existing institutions, processes and structures. For many of the countries, progress has been made on all three major components of a treasury system: (i) improved funding of government; (ii) better accounting of government operations; and (iii) financial management and planning of the government sector. Table 9 presents an index, developed by the IMF staff (Potter and Diamond, 1999), showing relative progress in treasury reform in the BRO countries (except Tajikistan, for which data are not yet available) in the period through end-1998. Inevitably, these ratings are subjective but serve to give some sense of relative progress in: (i) funding control over the payment and receipts process; (ii) centralization of bank accounts; (iii) accounting and reporting; and (iv) financial operations and planning. As can be seen, the Baltic countries and a few others (notably, Kazakhstan and Turkmenistan) have made relatively more progress in treasury reforms compared to the other BRO countries.

Table 9.

Progress in Treasury Reform1

article image
Source: Potter and Diamond (1999)

Index represents the unweighted average of ratings received in subcategories in each of the four areas. Progress is measured on a scale of: 1 denoting no progress, 2 denoting partial progress, 3 corresponding to adequate progress, and 4 denoting high/strong progress.

Most of the pension systems in transition economies have also been placed under strain, with steady declines in tax compliance and the contribution base and an increased use of the benefits to ease the social costs of transition.25 Demographic factors and a low retirement age (both in the pre-transition period and thereafter) have also added to the burden by increasing the dependency ratio of the system in some of the countries. Measures to alleviate the pressure on the pension system have focussed on an increase in the pensionable age, changes in the indexation formula, tightening benefits (e.g., by lengthening the minimum contribution period), and strengthening contribution collections. In many of the transition countries, the pensionable age has been increased, albeit in some cases gradually in order not to penalize individuals on the verge of retirement. In several countries, indexation of pension benefits has been shifted from gross wages to prices—a difficult political decision once the transition process had moved beyond the initial phase and wage growth has started to grow faster than prices. A broadening of the contribution base and the inclusion of either pension contributions or benefits into the tax net were important reform components in several countries, including Bulgaria, Hungary, and Latvia. To the extent that there were financial

imbalances in the short run, despite or in the absence of the aforementioned reforms, countries typically responded by cutting benefits in real terms, increasing contribution rates, and accumulating arrears. More fundamental reforms of the pension system are underway in Hungary, Kazakhstan, and Latvia, by introducing multi-pillar systems with a funded component. Poland is also in the process of introducing a three-pillar pension system starting in 1999; key legislation to this end was recently passed. Legislation resulting in major pension reform is currently being drafted in Lithuania and Slovenia.26

III. Concluding Remarks

The paper started off with a brief discussion of the initial conditions faced by most of the countries that are undergoing an economic transition. In the old system, the government played an overwhelming role. The paper also argues macroeconomic stabilization, price liberalization and privatization—essentially the core reforms visualized by the shock therapy approach—are necessary but not sufficient conditions for a complete transition to a market economy. Further deep changes—such as the creation of many new institutions, changes in incentives, changes in processes, change in the role of government and so on—are needed.

These deep changes are much more difficult and time-consuming, because they involve profound structural reforms and require major changes in attitudes, incentives, and relations.

In the new world of market economies, the role of government must change dramatically; new institutions must be developed in order to be able to cope with the market economy while at the same time fostering it. The emphasis will no longer be on direct controls, but rather by establishing the rules of the game, which would be determined by the tax system, the budget and a few essential regulations. The tax system would be totally reformed to make it efficient and equitable and to provide a reasonable level of taxation. Expenditure policies would be changed to bring them more in line with the reduced public resource.

At the same time, regulations would be fundamentally modified, so as to set the rules of the game, regulate private pensions, enforce competition, etc. Most permits, authorizations, and other ways that lend themselves to an extraction of bribes must disappear. It is a known fact that these regulations promote corruption, which for many of these countries remains very high (Table 10). There is substantial variation in the frequency and extent of bribe payments in the transition economies. However, in many countries, the state directly or indirectly still appears to exert significant intervention in enterprise activities (going beyond the exchange of public goods for taxes). For example, it makes possible for government employees to extract significant bribes from enterprises as shown in Table 11 (EBRD 1999).

Table 10.

Corruption Perception Index (CPI) 1997

article image
Note: 10 implies no corruption; 0 implies maximum corruptionSource: Averages of several indexes. Complied by Dr. Johann Graf Lambsdorff, Gottingen University.
Table 11.

The Frequency and Extent of the Bribe Tax

article image
Source: Business Environment and Enterprise Performance Survey; EBRD (1999).

Given the increase in income inequality (Table 3) and given the experience with privatization in transition economies over the past decade or so, it is not unreasonable to expect that the governments will be asked to play a more positive role in income redistribution. It is not unthinkable to expect that those who made fast money by raiding the public treasuries may be singled out for special attention in future years. Perhaps, the realization on the part of these individuals that this is likely to happen at some future date may provide an incentive for them, in the short run, to take as much capital out of the country as possible. Capital flight has, in fact, been a major problem for some of these countries and especially for Russia.

The policymakers should work hard at harmonizing the vision of the role of the state, and consequent size of government, that seems to prevail in many of the legislatures with one that is feasible, given the macroeconomic conditions and the level of institutional and economic development of these countries. A campaign to educate the public and legislators on what the state is expected to do in a market economy and the limits to what it can do would be useful. However, this will require great credibility for those who take this function.

Large fiscal deficits are often a macroeconomic problem; however, they also become a fundamental problem when they force governments into reneging on their legal contracts by imposing across-the-board sequestration and incurring expenditure arrears. These actions represent a form of corruption of the whole budgetary process and, more generally, of a market economy. When a public employee gives a day’s work and is not paid, or when pensioners do not receive their pensions to which they are legally entitled, there is something fundamentally wrong with the whole political process.27 The transition will not be over until these legal commitments on the part of the government are kept.


  • Aghion, Philippe, and Simon Commander, 1999, “On the Dynamics of Inequality,” Economics of Transition, Volume 7, 2, pp. 275-298.

  • Bokros, Lajos and Jean-Jacques Dethier, editors, 1998, Public Finance Reform During the Transition: The Experience of Hungary (Washington: The World Bank).

    • Search Google Scholar
    • Export Citation
  • Cangiano, Marco, Carlo Cottarelli, and Luis Cubeddu, 1998, “Pension Developments and Reforms in Transition Economies,” IMF Working Paper 98/151 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Cheasty, Adrienne and Jeffrey Davis, 1997, “Fiscal Transition in Countries of the Former Soviet Union: An Interim Assessment,” in MOCT/MOST.

    • Search Google Scholar
    • Export Citation
  • Craig, Jon, John Noregaard, and George Tsibouris, 1997, “Russian Federation,” in Fiscal Federalism in Theory and Practice, edited by Teresa Ter-Minassian (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • de Castello Branco, Marta, 1998, “Pension Reform in the Baltics, Russia, and other Countries of the Former Soviet Union (BRO),” Working Paper 98/11 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Djonkov, Simeon, 1999, “The Restructuring of Insider-Dominated Firms,” Economics of Transition, Volume 7, 2, pp. 467-479.

  • Filotochev, Igor, Mike Wright, and Michael Bleaney, 1999, “Privatization, Inside Control, and Managerial Entrenchment in Russia,” Economics of Transition, Volume 7, 2, pp. 481-504.

    • Search Google Scholar
    • Export Citation
  • Frye, Timothy and Andrei Shleifer, 1998, “The Invisible Hand and the Grabbing Hand,” American Economic Review: Papers and Proceedings, May.

    • Search Google Scholar
    • Export Citation
  • European Bank for Reconstruction and Development, 1998, Transition Report, 1998(London: EBRD).

  • European Bank for Reconstruction and Development, 1999, Transition Report, 1999(London: EBRD)

  • European Bank for Reconstruction and Development, 1999, Transition Report Update, April 1999, (London: EBRD).

  • Havrylyshyn, Oleh, Ivailo Izvorski, and Ron van Rooden, 1998, “Recovery and Growth in Transition Economies 1990-97: A Stylized Regression Analysis,” IMF Working Paper 98/141 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • JSSE, 1991, A Study of the Soviet Economy, (Paris: IMF, World Bank, OECD, and EBRD).

  • Kolodko, Grzegorz W., 1999, “Equity Issues in Policymaking in Transition Economies,” Economic Policy & Equity, Edited by V. Tanzi, K. Chu, and S. Gupta (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Kornai, Janos, 1997, Struggle and Hope, Essays on Stabilization and Reform in a Post-Socialist Economy, Edward Elgar Publishing Limited, (Northampton: Massachusetts) pp. 239-254.

    • Search Google Scholar
    • Export Citation
  • Ledeneva, Alena V., 1998, Russia’s Economy of Favors, (Cambridge University Press).

  • Lopes-Claros, Augusto, and Sergei Alexashenko, 1998, Fiscal Policy Issues During the Transition in Russia, IMF Occasional Paper 155, (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Mokhtari, Manouchelr and Irina Grafova, The Russian Federation at Crossroads: Tax Collection, Transfers, and Corruption,” (Mimeo: no date).

    • Search Google Scholar
    • Export Citation
  • Nagy, Piroska Mohacsi, 2000, The Meltdown of the Russian State, (Cheltenham, United Kingdom: Edward Elgar Publishing Limited).

  • Potter, Barry, and Jack Diamond, 1999, “Evaluation of BRO Treasuries,” mimeo, June.

  • Shleifer, Andrei, 1997, “Government in Transition,” European Economic Review, 41, pp. 385-410.

  • Liam Ebrill and Oleh Havrylyshyn, Staff team led by 1999, “Reforms of Tax Policy and Tax Administration in the CIS Countries and the Baltics” (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Tanzi, Vito 1992, Fiscal Policies in Economies in Transition, (Washington: International Monetary Fund).

  • Tanzi, Vito, 1994, “Reforming Public Finances in Economies in Transition,” in International Tax and Public Finance, pp. 149-163.

  • Tanzi, Vito, 1998, “Essential Fiscal Institutions in Selected Economies in Transition,” Collegium Budapest, Institute for Advanced Study, Discussion Paper No. 53 (November).

    • Search Google Scholar
    • Export Citation
  • Tanzi, Vito, and George Tsibouris, 1999, “Transition and the Changing Role of Government,” IMF Conference on A Decade of Transition: Achievements and Challenges, February 1-3,1999, Washington, D.C.

    • Search Google Scholar
    • Export Citation

Mr. Tanzi is the Director of the Fiscal Affairs Department, and Mr. Tsibouris is a Senior Economist in the European II Department and was previously in the Fiscal Affairs Department. An earlier version of this paper was presented at the Fifth Dubrovnik Conference on Transition Economies, July 23-25, 1999, Dubrovnik, Croatia and will be published in a conference volume. The authors wish to thank Steven Barnett and William Riordan for their assistance in compiling some of the data in this paper and Calvin McDonald and Luc Leruth for comments.


This paper presents developments during the decade leading up to 1998. There have been continued developments and reforms since then, as part of the ongoing transition process, which may not be reflected in this paper.


In fact, given that resources were not used productively, the total wealth to GDP ratio is likely to have been particularly high.


Filotochev, Wright, and Bleaney (1999) provide survey evidence of managerial entrenchment in the post-privatization restructuring of Russian enterprises. Ledeneva (1998) discusses the role that blat (i.e. connections) played in the privatization process. See especially pp. 184-92.


Aghion and Commander (1999) discuss the increase in inequality in selected countries in transition and the channels through which it has manifested itself.


The sharp fall in the income of some groups (e.g., pensioners) has contributed to the deterioration in Gini coefficients.


Djankor (1999) argues that owner/managers’ incentives to restructure privatized firms decreases when they perceive their newly acquired ownership as a windfall gain.


It should be noted that the pursuit of a socially accepted income distribution is considered tobe one of the fundamental roles of government.


A book that gives a feel for what deep reform means is Public Finance Reform during the Transition: The Experience of Hungary, edited by L. Bokros and J. J. Dethier (1998).


For example, some economists have argued that in countries with defined benefit systems, large pension commitments require high payroll taxes. In turn, high payroll taxes discourage employment or force enterprises to go underground thus evading taxes and reducing public revenue. The empirical relevance of this observation is not clear.


There is no question that in some transition economies, such as Hungary, the level of public expenditure is too high. See Bokros and Dethier, 1998.


Obviously the assignment of revenue must bear a clear relationship to the assignment of expenditure. Subnational governments must be given adequate resources to meet their commitments; at the same time, their budgets cannot be soft. There is also a need to reform expenditure because: (a) social expenditures in the past in these economies were largely untargeted and in many instances amounted to nothing more than categorical privileges rather than safety nets to protect against unforeseen circumstances; and (b) in the new emerging market-oriented economies, it is important that the process of reforming social safety nets include a role for individual self-insurance.


At 29 percent of GDP, the level of tax revenue in Russia is not low. General public revenue is even higher. Furthermore, because large sectors of the economy escape the tax net, the fully taxed sector of the economy is likely to experience particularly high tax burdens.


For example, the rates of the turnover taxes were not even published and were often changed.


For example, registered taxpayers for the enterprise profits tax in Russia increased sevenfold in the period between 1990 and 1995.


Tax evasion and corruption on the part of tax officials were phenomena that hardly existed in the previous system.


Even the leaders in the transition such as Hungary and Poland have not yet completed the process of creating new tax systems and tax administrations.


In countries where local governments were important, the personnel of the tax administration has had difficulty in determining whether its allegiance should be to the national government or to the local governments which often provide them with housing, offices, and so forth. See also Mokhtari and Grafova (Mimeo: no date).


For example, in many countries the government is still heavily engaged in the provision of housing and it continues to subsidize energy consumption. The price of energy in the majority of economies in transition remains much lower than in market economies.


Obviously this statement assumes that the GDP estimates are correct.


Budgetary subsidies to the economy in Russia declined from about 14 percent of GDP in 1992 to about 4-5 percent of GDP in 1998; see Lopez-Claros and Alexashenko (1998).


The IMF WEO database indicates that military outlays in transition economies have declined on average from an estimated 6 percent of GDP in 1990 to about 2 percent of GDP in 1997, with most of the decline occurring in the countries of the former Soviet Union.


Kazakhstan and Moldova, unlike the remaining transition economies, have undertaken a significant reduction in their respective civil services.


In Russia for example, the treasury does not cover the power ministries, and the resources generated by the ministries themselves (i.e., own resources); it also does not cover subnational expenditures and off budget accounts.


See Cangiano, Cottarelli, and Cubeddu (1998) for a review of developments in pension systems in 11 transition economies during the 1990s.


When macroeconomic difficulties lead to inflation, and this in turn reduces the real value of what individuals receive, at least the legal obligations have been maintained.