Facing the Challenges

The basic medium-term policy agenda in these five countries is set by the common goal of promoting strong and sustainable growth, with fiscal policy having to reconcile three broad objectives:

• realization of a strong fiscal position that will (i) address country-specific concerns of external vulnerability; (ii) facilitate and sustain disinflation; (iii) ensure room for maneuver in conducting effective stabilization policies in the face of unexpected shocks; and (iv) prepare for longer-term challenges arising from population aging. Over time, of course, a goal will be to meet the requirements of the EU’s Convergence Programs, and later the Stability and Growth Pact—but countries’ Pre-Accession Economic Programs are not subject to the excessive deficit procedure, which will not apply for many years (Box 7.1);

• a reduction in the high tax burden on labor income, in order to boost private-sector-led growth and employment, while adjusting indirect taxes and tariffs to prepare for the single EU market; and

• an increase in some areas of government spending associated with reforms to (i) complete the transition agenda; (ii) modernize the economy through investment in areas such as infrastructure and the environment (also meeting EU requirements); and (iii) improve pension and health care systems in ways that will contain longer-term costs.

Specifics of reform differ significantly, however—as do macroeconomic circumstances in each country. First, on reforms, the Czech and Slovak Republics still face a significant burden resulting from bank and enterprise restructuring; and in Poland, with steel and coal sector restructuring ongoing, the largest challenge lies in agriculture—with uncertain fiscal implications. Second, though demographic pressures loom in all of the five, magnitudes differ considerably, with Slovenia facing by far the largest shock: Hungary, Poland, and Slovenia, have adopted reforms to strengthen the sustainability of their pension systems, though these imply additional short- and medium-term fiscal pressures. Third, spending related to adoption and implementation of the acquis communautaire is bound to differ—though comparisons are made difficult by the uncertainty at present about scale and phasing. Fourth, macroeconomic situations are diverse and subject to fluctuations. Using 1999 as an illustrative reference point, inflation rates ranged from 2 percent in the Czech Republic (4 percent in 2000) to more than 10 percent (in both years) in the Slovak Republic. In terms of external positions, the Czech Republic was an external creditor (if FDI is excluded), while Hungary’s net foreign debt ratio exceeded 20 percent of GDP.⁹⁶ Poland had a current account deficit of 7½ percent of GDP in 1999 (although the 2000 outcome was close to 6 percent).⁹⁷ This compares with a deficit in 1999 of 3 percent of GDP in the Czech Republic (rising to some 4¾ percent of GDP in 2000). Finally, the different challenges for fiscal policy are also shaped by the monetary and exchange rate regimes, although the regimes in the five countries are similar. Four of them pursue formal or informal inflation targeting (albeit, in Hungary, within a wide exchange rate band), while Slovenia combines monetary targeting with a managed float.

Even in terms of the fiscal starting position, similarities are less than suggested by a simple comparison of general government deficits. With the exception of Slovenia, deficits in 1999 were almost identical, at slightly above 3½ percent of GDP (Figure 7.1).⁹⁸ However, this comparison is misleading: these official “headline” measures⁹⁹ are not fully consistent across countries, and deviate to differing degrees from the definition in the European System of Accounts (ESA-95).¹⁰⁰ Also, they take no account of quasi-fiscal deficits in the public enterprise and banking sectors, which have been largely eliminated in Hungary, for example, but not in the Czech and Slovak Republics. Moreover, as a result of past fiscal trends—and, in the case of Poland, the impact of official debt restructuring—the interest bill, and thus the primary balance, differs substantially across countries.¹⁰¹ The Czech Republic was running a primary deficit of 2½ percent of GDP in 1999 (1¼ percent of GDP after recent data revision)—which more than doubled in 2000, driven by banking sector problems¹⁰²—while Hungary was showing a primary surplus of almost 4 percent of GDP. Finally, the size and structure of revenues and expenditures differs quite substantially within the group. In all five countries, the relative size of the government—measured by either 1999 revenue or expenditure shares in GDP—has fallen below the EU-average (Figure 7.2).¹⁰³ However, both ratios were considerably lower in the Czech Republic than, for example, in Hungary or the Slovak Republic—which also have the highest tax wedge (social security contributions and personal income taxes as a share of gross salaries). On the expenditure side, Poland spends a particularly large share on transfers to households, while public consumption as a share of GDP is highest in the Slovak Republic. Thus, the similarities in the accession group are smaller than a first glance might suggest, indicating scope for differences also in medium-term fiscal strategies.

View Full Size

Government Balances in the CEC5 and the European Union, 1993-99

(Percent of GDP)

Sources: IMF, World Economic Outlook; and IMF staff estimates.¹Excluding privatization receipts (in Hungary, of central government only). Data for Hungary are on a consolidated GFS basis from 1998 on, only.²Defined as total revenue minus noninterest expenditure.³The CEC5 aggregate is derived as the unweighted average for the five central European countries under study; the EU aggregate is the unweighted average in the European Union, excluding Luxemburg.

• Download Figure
• Download figure as PowerPoint slide

Medium-Term Fiscal Positions: Choosing an Appropriate Target

The governments of all five countries under study are aiming to strengthen their fiscal positions over the medium term in order to sustain confidence and foster domestic savings (Box 7.2). Goals differ considerably, partly reflecting differences in macroeconomic circumstances, fiscal starting conditions, and remaining reform agendas discussed above. As countries review their objectives in the context of the accession process, a crucial question they face is what benchmarks to use, and what considerations to weigh, in setting the goal for the overall medium-term fiscal balance. The ultimate yardstick is arguably the lasting impact of fiscal policy on economic growth: while appropriate tax and expenditure policies can significantly strengthen economic performance, a prudent overall fiscal position is crucial to ensure that growth is sustainable. And in countries where investment and growth depend strongly on capital inflows, sustainability is to a high degree a question of containing external vulnerabilities.

View Full Size

Revenue and Expenditure Trends in the CEC5 and the European Union, 1993-99

(Percent of GDP)

Sources: IMF, World Economic Outlook; and IMF staff estimates.¹Excluding privatization receipts (in Hungary, of central government). Data for Hungary are on a consolidated GFS basis from 1998 on, only.²For Czech Republic, includes grants to transformation agencies. For Hungary, see footnote 1.³The CEC5 aggregate is derived as the unweighted average for the five central European countries under study; the EU aggregate is the unweighted average in the European Union, excluding Luxemburg.

• Download Figure
• Download figure as PowerPoint slide

The broad tax and expenditure priorities identified by these five countries also reflect a concern to foster medium-term growth. Most countries under study (the Czech Republic being an exception) intend to meet their medium-term fiscal balance objectives through cuts (sometimes sizeable) in expenditure ratios, while also planning to lower tax rates on labor income. While the relationship between fiscal policy and economic growth is far from clear-cut, combinations of tax cuts on labor income with spending restraint focused on durable adjustments in the public wage bill and transfers are likely to encourage growth over the medium term. Moreover, such policies should help counter any short-term contractionary effects of fiscal consolidation (Box 7.3).¹⁰⁴ In general, for growth to benefit over the medium term, it is crucial that spending targets be achieved by lasting reductions in current programs, and not by a suppression of investment and other growth-enhancing expenditure. Moreover, if fiscal consolidation is perceived as credible and lasting—reinforced by a realistic and transparent medium-term fiscal strategy—the positive effects on growth are likely to emerge faster. Meeting targets for the fiscal balance, while achieving revenue objectives and safeguarding priority expenditure, is the true challenge—and will require difficult policy choices.

To ensure that growth is sustainable, fiscal policy over the medium term needs to safeguard public debt and external sustainability. High fiscal deficits and their translation into rising public and external debt ratios and large current account deficits have been found to increase a country’s vulnerability to crisis.¹⁰⁵ Moreover, if a crisis occurs, its impact on output is typically most serious in emerging market economies characterized by large and potentially volatile capital inflows;¹⁰⁶ and it is more likely to occur when these inflows are debt creating.¹⁰⁷ In this regard, fiscal consolidation can play a key role in buttressing sustainable economic growth: it allows an expansion of private investment, without jeopardizing external viability. Appendix I demonstrates the linkages in these economies between fiscal policy and public and external debt dynamics and the current account, using the situation in 1999 as its illustrative reference point. Based on a number of simplifying assumptions, it shows how a quantification of these linkages can provide a valuable safety check for fiscal policy.

A first and fundamental issue is whether public debt sustainability is a binding constraint. The quantitative analysis in Appendix I suggests that public debt sustainability does not pose a constraint to existing fiscal plans—though an important caveat is needed. It assumes that entitlement reform will be the main route to address the fiscal impact of population aging. Otherwise, stronger fiscal positions could be required to ensure debt sustainability. Indeed, Hungary, Poland, and Slovenia have moved ahead with reforms to strengthen the sustainability of their pension systems—an area where some existing EU members are lagging in reform.

For some of the countries, concerns of keeping external debt and current account deficits within prudent ranges appear to be relevant when setting medium-term fiscal balance objectives. The quantitative analysis in Appendix I illustrates this relationship, by determining fiscal positions that are consistent with prudent external debt objectives. The link occurs through the current account deficit, the share of nondebt financing, and stylized assumptions about private saving and investment behavior. Based on the economic situation in 1999, the analysis derives illustrative targets for fiscal adjustment, taking into account the relatively high external indebtedness (in Hungary), a considerable current account deficit (in Poland), and the potential for a deterioration in the private sector saving-investment balance—or alternatively, somewhat smaller FDI inflows (in the Czech and Slovak Republic). For Slovenia, none of the factors were found to warrant fiscal tightening, but a goal of approximate budget balance appeared well-grounded in light of a particularly severe demographic shock as well as the need to prepare for the impact of the final stages of external liberalization.

In practice, it is crucial to look beyond headline numbers for deficits and debt. First, the nature and composition of external financing is key in assessing external vulnerability; it is not only the debt characteristics but also the maturity profile of capital inflows that matters, with a large build-up of short-term debt, particularly in relation to the stock of reserves, raising the probability of crisis.¹⁰⁸ Secondly, the exchange rate system plays a role. All countries have given up soft pegs, thus reducing their vulnerability to misalignments and speculative attacks. Nevertheless, in small open economies that rely heavily on foreign financing, the exchange rate is too important a price to ignore in the conduct of domestic economic policy. With monetary policy focused on price stability, fiscal policy may need to play a role in limiting external imbalances and preserve competitiveness. Third, the domestic counterpart of external imbalances also needs to be assessed in judging sustainability. Imbalances are typically of much greater concern if driven by excessive consumption. But—a more complex issue—the efficiency with which foreign inflows are being utilized for investment in the private sector is also key: even FDI inflows may take advantage of distortions in domestic markets or rely on implicit guarantees.

The above considerations point to the importance of a sound financial sector when appraising sustainability and the extent of vulnerability to external risks. Especially if capital inflows are liberalized without adequate supervision and regulation in place, the potential for excessive external borrowing and on-lending—often encouraged by implicit or explicit government guarantees—can rise rapidly. This places emphasis on reforms to strengthen banking systems which directly diminish financial vulnerability and—while often adding to explicit public deficits and debt—may not result in a deterioration of the external current account.¹⁰⁹ In this area, the Fund’s enhanced surveillance over the financial sector should help shed light on strengths and vulnerabilities—and four of the five economies have participated in IMF World Bank Financial Sector Assessment Programs (FSAPs), with an FSAP for the Slovak Republic being conducted in the first half of 2002. In sum, judging external sustainability entails an assessment of these qualitative aspects of the domestic setting—an assessment that is perhaps easier to make in the smaller and most open economies, and those where banking systems have been most fundamentally restructured.¹¹⁰

After setting appropriate targets, the pace of medium-term fiscal consolidation may need to be adjusted in the event of temporary output shocks and cyclical pressures. All five countries will continue to face shocks—emanating, for example, from energy prices, emerging market contagion, or the late stages of transition. Exposure to shocks varies across countries and, in many instances, is likely to diminish: transition-related shocks will recede over time, and the status of these countries as leading EU accession candidates should serve to mitigate their exposure to contagion from emerging markets. But when shocks occur, fiscal stabilizers can play a potentially useful role in softening output fluctuations, thus warranting possible adjustments in the pace of consolidation (indeed, the global economic slowdown in 2001 serves as a case in point). Of course, the effectiveness of fiscal policy is constrained in relatively open economies, and flexible exchange rates in all five countries help reduce the burden on fiscal policy to stabilize output in the event of a negative demand shock. Irrespective of the exchange rate system, however, fiscal policy will need to play a role when excessive domestic demand exerts pressure on resources, whether mainly on domestic prices or directly on the current account.

In the end, decisions on medium-term targets for the fiscal balance will reflect a varied and complex set of issues. A definitive judgment will clearly need to take account of the various factors discussed above, as well as longer-term issues, such as demographic trends. Moreover, with strong and sustainable growth as the ultimate objective, the decision about medium-term fiscal adjustment must consider the trade-offs between revenue and expenditure objectives, and the target for the fiscal balance. As it turns out, IMF surveillance, which takes account of country-specific circumstances, has tended to favor fiscal goals for these countries ranging between moderate deficits and a position of overall balance implying, in most cases, recommendations for further consolidation.¹¹¹

The IMF’s assessments of medium-term fiscal requirements are now typically made public and updated at least annually in the context of Article IV consultations. They can be found on the country pages of the IMF’s website (http://www.imf.org). A key point of these assessments is that even against a common background of transition, recommended medium-term fiscal balances are not identical. This reflects many factors, including different starting positions, different economic prospects (for example, for the behavior of saving-investment balances), and also differences in definitions—which hamper comparisons across countries. In view of the usual host of uncertainties on the macroeconomic front, and also because of measurement problems—which should lessen as data quality improves over time—projections will undoubtedly be revised moving forward. Fiscal balance targets may need to be reconsidered as a result. It is also fair to say that the debate between the IMF and country authorities has not focused exclusively on issues of deficit size. It has focused deeply on the need to eliminate quasi-fiscal deficits and to move forward with public expenditure reforms—and thus to reorient the public finances in support of growth, taking into account the potential tensions between fiscal adjustment, on the one hand, and growth-supporting revenue and expenditure reforms on the other. Revenue and expenditure reforms are the subject of the next two sections.

Revenue Reforms on the Way to EU Accession

A reduction in the tax burden on labor income should be a priority in all five countries. When social security contribution rates are combined with personal income taxes, the nominal tax wedge exceeds 40 percent of gross salaries in all five countries, and reaches about 60 percent in Hungary and the Slovak Republic.¹¹² These tax wedges on labor are high even by the standards of existing EU members—where the average nominal tax wedge is below 40 percent. Reducing the wedge between gross and net salaries should boost both labor demand and incentives to work, thus fostering higher growth and a lasting reduction in unemployment rates.¹¹³

A key challenge for the five accession candidates is to ensure that a reduction in the tax rates on labor income is well coordinated and sequenced with a broadening of tax bases and improved tax administration, to avoid excessive revenue losses.¹¹⁴ The authorities could broaden tax bases if they eliminated various tax credits, reduced exemptions, and strengthened tax administration so as to improve collection rates. The revenue effect of lower tax rates on labor could be offset, at least in part, by higher employment and a shift of activity from the gray economy to the formal sector.

Moving closer to principles applied in more advanced economies could help address weaknesses in tax administration. The effectiveness of tax administration differs across the five countries. Depending on the circumstances, there may be scope to improve the management and organization of tax offices and to enhance coordination between tax collection agencies. The focus should be on greater specialization in the functions of the staff in tax offices to support a system based on self-assessment (for example, processing returns, collecting tax arrears, and carrying out audits). In addition, headquarters operations may require strengthening so that tax departments are better equipped to implement critical reform measures. The maintenance of consistent data on taxpayers, and a regular exchange of information between revenue agencies are also important elements of an effective administration. Efficiency gains obtained through specialization would free resources from routine undertakings and could be redirected toward audit and enforcement activities. To this end, improvements in staff training and introduction of modern systems (including a unique tax identification number) would be crucial. Finally, given the high level of tax arrears in some countries, tax administration reforms need to include strengthening collection enforcement procedures, granting the necessary legal powers to the tax administration to combat tax evasion, and reforming the framework of penalties and fines in order to discourage noncompliance. Efforts to conform with EU accession requirements—on which work has already begun—should generate further improvements in tax administration.¹¹⁵

A comprehensive reform of the personal income tax may also strengthen revenue collection and would be a useful element of tax harmonization and improved competition. Revenue generation through personal income taxes is hampered by the number of deductions and exemptions, and generally high marginal tax rates. In some of the countries, individuals also seem to avoid the personal income tax precisely to avoid paying large social security contributions, contributing to the fact that a significant portion of economic activity is not captured within the tax net.¹¹⁶ A reform strategy could include reductions in rates, the number of tax brackets, and the extent of exemptions and deductions. Moreover, in the case of countries with low single-digit inflation, the reduction in the marginal tax rates could be combined with the taxation of interest income.

Some changes in corporate income tax rates and tax incentives may be warranted. Following the recent reforms to lower corporate income tax rates in Poland and the Slovak Republic, the scope for further rate reductions is quite limited.¹¹⁷ In addition to lost revenues, any decision to lower corporate income tax rates also needs to take account of the potentially undesirable effects resulting from growing divergences in taxes on capital and labor.¹¹⁸ A careful evaluation of the effectiveness of other investment incentives may also be useful, with a view to reducing certain tax holidays.¹¹⁹ Finally, emphasis should be given to broadening the tax base by removing special tax incentives, particularly in the Czech Republic, where the difference between the standard and the effective rate is largest. A neutral tax base (that is, a similar tax burden for different business activities) contributes to a level playing field, as does a simple depreciation system.

Medium-term revenue projections must take into account the effects of changes in indirect tax rates and tariffs implied by EU accession and compliance with WTO agreements, noting the range of uncertainty. The process of integrating the central European economies into the single market—free of restrictions on the movement of goods, services, and factors of production between countries—will imply a number of reforms. First, VAT systems could move closer to the ones of EU countries, which would generally entail an increase in the lower rate (especially in the Czech Republic and Slovenia), accompanied by a gradual decline in the standard rate (particularly in Hungary).¹²⁰ In addition, better harmonization with EU standards would entail moving a number of goods, services, and activities that are now taxed at the lower rate to the standard rate (for example, construction activities in the Slovak Republic). Second, excise taxes will have to be aligned with those of the EU, involving increases in some rates (tobacco in Hungary and the Slovak Republic, for example). Finally, accession countries will need to continue implementing tariff cuts agreed under the Uruguay Round of the WTO, although the revenue impact of such cuts would be limited, given the small share of customs revenue in total revenue. The overall revenue effects of these measures are difficult to evaluate: in many cases the direction and magnitude would depend on substitution effects between higher and lower taxed goods, as well as offsetting rate and quantity adjustments. Moreover, scope remains, again, to broaden revenue bases.

In all, these reforms likely imply a reduction in revenue-to-GDP ratios in most of the countries. Except for the Czech Republic and Slovenia, this is generally consistent with official plans. To facilitate a reduction in the high tax rates on labor, without compromising fiscal targets, it is important to coordinate them with other measures that enhance revenues, notably the realization of revenue gains from better tax administration and broader tax bases. In any event, whether significant tax rate reductions can be achieved without compromising fiscal balance objectives will also depend crucially on success in containing public spending.

Need and Scope for Expenditure Reforms

Various tensions are evident in the effort to reduce spending. This is particularly the case in light of spending pressures arising, albeit to differing degrees, from the need to (i) complete the process of economic restructuring; (ii) initiate reforms to lessen the long-term fiscal impact of population aging, implying additional medium-term fiscal costs; (iii) comply with EU regulations; and (iv) meet other country-specific spending obligations, such as membership in NATO. While these expenditures can be stretched out to some extent (for example, by negotiating longer transition periods for compliance with certain aspects of the EU’s acquis), spending restraint in other areas is still inevitable to secure the medium-term fiscal targets.

Assessing the Fiscal Implications of Reforms

A number of areas are of particular relevance when developing estimates of the costs of reforms:

In the Czech and Slovak Republics, there remains a comparatively large unfinished agenda to complete the restructuring of the enterprise sector and absorb the fiscal impact of bank restructuring. The direct cost, particularly of enterprise reforms, to the countries’ budgets are difficult to project, as an undetermined share is expected to be borne by the private sector or offset by privatization receipts. Notwithstanding the positive longer-term effects of restructuring, indirect costs—in terms of higher unemployment benefits and other transfers—are also likely to add to the direct costs over the short and medium term. The fiscal cost of bank restructuring in the Slovak Republic is estimated at about 1 percentage point of GDP a year in interest on recapitalization bonds (whose total estimated value is equivalent to 12 percent of 2000 GDP), while in the Czech Republic, future cumulative bank restructuring costs are estimated at 14 percent of 2000 GDP (excluding costs related to Investični a Poštovni Banka (IPB) which could exceed 5 percent of GDP).¹²¹ In the Slovak Republic, the ongoing restructuring of the railways and electricity sectors could put additional pressures on fiscal policy. In Poland, the restructuring of the coal and steel sectors is still ongoing and reforms of the large agricultural sector have barely begun. The total cost of the coal and steel sector restructuring, covering closures, investment requirements, and severance packages—including the part borne by the affected companies—has been estimated at close to 3 percent of 1999 GDP.¹²² The effective fiscal impact of agricultural reforms is more uncertain, depending, among other things, on prospects for creating nonfarm jobs in rural sectors.

As in most of Europe, all five countries will experience sizable demographic pressures over the coming decades, with a diminishing share of workers to support a growing number of pensioners (Figure 7.3). The associated fiscal challenges, in terms of long-term pressures on pension and health expenditures, will be highest in Slovenia. Here, the old-age (or demographic) dependency ratio—namely, population of 65 and older in percent of population of 15–64—is projected to double to nearly 40 percent by 2030, and the deficit of the pension system (currently 4 percent of GDP) is expected to triple by 2010, in the absence of reforms. In the other four countries, the increase in the dependency ratios between 2000 and 2030 is projected to be around 10 percentage points—albeit from different bases, with the highest current rate of more than 20 percent in Hungary.¹²³

View Full Size

Old-Age Dependency Ratio in the CEC5 and Europe, 1990-2030

(Population aged 65 and above in percent of working-age population of age 15-64)

Source: World Bank, World Development Indicators, 2000.

• Download Figure
• Download figure as PowerPoint slide

In response to the demographic outlook, Hungary, Poland, and Slovenia have implemented comprehensive pension reforms over the past years, culminating in the introduction of a partly funded multipillar system. Pension reform is also needed in the Czech and Slovak Republics, likely implying pressures on the fiscal accounts over the coming years. Although these reforms are expected to generate sizable savings over the longer term,¹²⁴ the short- and medium-term effects on the government balances—even if supported by supplementary cost-saving measures¹²⁵—are likely dominated by the revenue loss associated with transfers to the pension funds (provided these funds are not part of the general government). It should be noted that the immediate revenue loss associated with the reclassification of public to private pensions does not imply an underlying weakening in the fiscal stance from the perspective of national savings.¹²⁶ Some fiscal tightening may well be appropriate, however, to cover various cost associated with the pension reform, including administrative expenses for setting up private funds as well as higher borrowing cost to finance the public sector deficit.¹²⁷

The demographic outlook also affects health care expenditure, and needed reforms to the health systems to mitigate long-term costs could raise expenditure somewhat in the earlier years.¹²⁸ While fiscal pressures could be mitigated by a variety of measures,¹²⁹ they still create incentives for delaying reforms in the absence of a longer-term policy orientation and an explicit strategy for the coming years.

Compliance with EU legislation (that is, adoption of the acquis communautaire) is associated with additional spending for legal approximation and institution building, development of the transport infrastructure, and compliance with environmental standards.¹³⁰ When assessing the fiscal implications of these measures, it should be noted that part of the additional expenditure is offset by transfers from the EU (Box 7.4).¹³¹ While many of these expenditures are already occurring, additional outlays, particularly for environmental investments, could be sizable. Estimates over the medium term are uncertain, depending, among other things, on how the acquis is met, the length of the transition periods granted to achieve compliance, and the portion of the costs borne by the private sector. While official projections are not available, earlier tentative estimates by the World Bank suggested annual government spending on environmental investments in the range of 2½-3¾ percent of 1997 GDP in the Czech Republic and 1¾-4½ percent of 1997 GDP in Hungary.¹³² In Poland, IMF staff estimates, based on projections by the World Bank and conservative assumptions for transition periods, suggested that spending on environmental investment (including operations and maintenance costs) would need to increase by 1¼-4½ percentage points of GDP between 1999 and 2005. Taking into account existing outlays, this would imply average spending of 1¼-3½ percent of GDP annually over 2000–05, to comply with EU environmental regulations.¹³³ In Slovenia, total annual accession costs are estimated at 2½-4 percent of GDP, largely related to a strengthening of the country’s administrative capacity. As accession discussions continue over the next year or so, it will be important that countries develop firmer estimates of the related expenditure. A quantification of the fiscal implications associated with implementing the reform agenda will also be necessary to come to realistic agreements with the EU on the length of possible transition periods.

Box 7.4.

EU Pre-Accession Assistance

A number of EU funds are available to support the accession process.¹ To be eligible for them, candidate countries need to provide cofinancing and have in place the necessary administrative infrastructure.

The Phare program is the main instrument for financial and technical cooperation with the countries of central and eastern Europe. It was set up in 1989 to support economic and political transition and by 1996 it had been extended to include 13 partner countries from the region. Starting with an allocation of €4.2 billion for the 1990–94 period, the Phare budget was increased to €6.7 billion for the 1995–99 period. It is envisaged that for the period 2000–06 the program will provide €1.56 billion of annual support. In Agenda 2000, the European Commission proposed to focus the Phare program on preparing the candidate countries for EU membership by concentrating its support on two crucial priorities in the adoption of the acquis communautaire: institution building and investment support. Institution building pertains to adapting and strengthening democratic institutions, public administration, and organizations that have a responsibility in implementing and enforcing Community legislation. It is envisaged that approximately 30 percent of Phare funds will be used to meet institution building needs, in accordance with the conclusions of the Luxembourg European Council, in particular through the so-called twinning mechanism.

The Instrument for Structural Policies for Pre-Accession (ISPA) has been set up as part of Agenda 2000 and is dedicated to supporting environment and transportation projects so as to help align infrastructure standards with those of the EU. It is envisaged that over the period 2000–06, €1.04 billion a year (at 1999 prices) will be made available for such purposes. The main priorities of ISPA in preparing the applicant countries for accession will be: (i) familiarizing them with the policies and procedures of the Union; (ii) helping them catch up with EU environmental standards; and (iii) expanding and linking with the trans-European transport networks.

The Special Accession Programme for Agriculture and Rural Development (SAPARD) aims at providing structural adjustment support for agriculture and rural development. It will also support measures to enhance efficiency and competitiveness in farming and the food industry and create employment and sustainable economic development in rural areas. It is envisaged that the program will have an annual budget of €520 million (at 1999 prices) until 2006.

¹source: The European commission website: http//europa.eu.int/comm./enlargement.

Finally, further spending obligations—though significantly more modest—are occurring in the Czech Republic, Hungary, and Poland from the need for modernizing the military infrastructure under NATO membership. The Slovak Republic and Slovenia are planning for similar expenditures over the next few years as part of their preparation for possible NATO membership. Although very difficult to project, NATO-related expenditures could raise defense outlays over the coming years by up to ½ percentage points of GDP annually. On the other hand, there may be scope for sizable efficiency gains, if part of the defense-related services are provided by the private sector.

Adding up the fiscal implications of reforms is more than an arithmetic exercise—it requires a careful consideration of structural reform priorities and their phasing. The desire—politically or economically motivated—to press ahead with reform measures creates tensions with the objectives of reducing deficits and tax rates. For example, if deficit and tax reductions were regarded as the more relevant objectives, and pressures were strong to speed up compliance with EU standards, there would be a risk that important projects (such as infrastructure investments) would be “crowded out” and difficult reforms with mostly longer-term benefits (for example, pension reforms) would be delayed. On the other hand, with a more flexible attitude toward the timing of tax cuts and longer transition periods for some EU-related outlays, other reforms could possibly be advanced. At all events, these trade-offs need to be carefully considered, and the development of a medium-term fiscal framework would facilitate that.

Reconciling the fiscal objectives in a sustainable manner will ultimately require adjustments in “managed” spending. Most countries anticipate a reduction in revenue ratios—largely motivated by the desire to reduce the heavy tax burden on labor income (see Box 7.2), In addition, all countries will need to absorb spending associated with the reform agenda outlined above. Thus, a successful fiscal strategy, which resolves the various tensions in an efficient forward-looking manner, calls for strict limits on the growth of the remaining “managed” expenditures—and, in some cases, most likely, a reconsideration of revenue targets, if balance objectives are adhered to.

Scope for Savings

To identify areas where spending is high and savings are feasible, it is useful to compare the expenditure structures in the five accession candidates with those in existing EU member countries. Such a comparison, illustrated in Figure 7.4, suggests that the share of expenditure on public consumption (defined as the sum of the wage bill and expenditure on goods and services) is largest in the Slovak Republic and Slovenia. Poland, on the other hand, spends a relatively large share on transfers to households, while Hungary and the Czech Republic may have room to reduce subsidies to enterprises. In all five countries, a significant portion of current expenditure is assigned to social security and welfare functions, with both Poland and Slovenia spending an even greater share of GDP for social purposes than the average EU member (notwithstanding higher income levels in the EU).

View Full Size

Indicators of Public Spending in the CEC5 and the European Union, 1998

(Percent of GDP)

Sources: National authorities; IMF, Government finance Statistics; and IMF staff estimates.¹ Data on subsidies in the Czech Republic include bank restructuring costs.² Refers to consolidated central government only. EU average is derived on the basis of the latest data available (ranging from 1995 to 1998), and excludes Belgium, France, Italy, and Portugal, for which no consistent data is available.

• Download Figure
• Download figure as PowerPoint slide

When comparing spending ratios, it is useful to take into account the generally positive correlation of government spending with income levels. This correlation suggests that lower spending ratios in the five countries would be more commensurate with their income levels. Within the European Union, there is a distinctly positive relationship between the current primary spending ratio (as well as its main components) and the per capita income level (Figure 7.5).¹³⁴ On the basis of the trends derived for the European sample—and even more so in comparison with countries of similar income levels in other regions—all accession candidates have primary current spending ratios above the levels commensurate to their per capita incomes. The main cause for elevated spending ratios in the Czech Republic, Hungary, and particularly Poland, is a high level of transfers to households.¹³⁵ In Hungary, fairly large enterprise subsidies also contribute to an elevated level of expenditure. The Slovak Republic spends a significantly larger share on public consumption than its income level would suggest. Finally, Slovenia’s spending ratios exceed comparative EU levels for both transfers and consumption, though not by much.

View Full Size

Per Capita Income and Primary Current Spending in Selected Countries, 1998¹

(Percent of GDP)

Sources: European Commission; World Bank Development Indicators; and IMF staff estimates.¹Regression lines are fitted to the observations on EU countries only, as the basis for comparison. Comparison of spending ratios is hampered by different coverage across countries. For example, data for Korea and Tunisia refer to consolidated central government only, where as data for Chile include public enterprises.

• Download Figure
• Download figure as PowerPoint slide

Box 7.5.

Planning Spending Restraint—An Illustration for Poland

• The decision where savings should be realized, and the pace of change, is a political one, which will reflect each country’s starting position and priorities. Once such priorities have been determined, an important aspect of the decision-making process is the design of specific reforms to achieve the desired result and an assessment of their impact. What follows is a very simple illustration for the case of Poland (based on the situation in 1999) of how savings could be allocated to three main spending categories, in order to bring about a targeted reduction in the ratio of primary current spending to GDP.¹

• The starting point is the formulation of baseline projections for total government spending in the absence of reforms. Given the government’s target for the overall fiscal balance—which, at the time, was the elimination of the (economic) deficit over the medium term—the baseline would imply an increase in the revenue ratio of 3½ percentage points of GDP between 1999 and 2005. To avoid this, and instead provide scope for a reduction in tax rates—while safeguarding the fiscal balance objective—the government would therefore have to achieve expenditure saving in excess of 3½ percentage points of GDP (assuming no major improvement in revenue collection).

• Savings in public consumption can be achieved by reductions in the wage bill, as a share of GDP, through wage moderation, a reduction in public sector employment, or a mix of the two. Baseline projections assume unchanged employment in the public sector and an increase in real wages in line with total labor productivity in the economy. If average wages in the public sector were rather kept constant, in real terms, public consumption could be reduced by 1½ percentage points of GDP over the medium term. Without wage moderation, the same savings in the total wage bill would require a reduction in public sector employment by some 4¼ percent annually.

• In the baseline, subsidies to enterprises are assumed to grow in line with GDP. Savings of at least 1 percentage point of GDP could be achieved after 2003, when the restructuring of the coal and steel sectors are expected to be completed.

• Transfers to households are particularly high in Poland, and could be reduced by further reforms to curb pension outlays, accounting for approximately three-fourth of total transfers in 1998. While the recent comprehensive reform of the pension system promises significant long-term fiscal gains, savings over the medium term would need to be achieved by measures that increase the low effective retirement age (stricter enforcement of eligibility rules for disability pensions, and greater financial disincentives for early retirement, for example). A gradual increase of the effective retirement age by 2½ years could generate fiscal savings of about 1–3 percentage points of GDP over the medium term, depending on the extent to which the additional supply of labor can be absorbed by the market.

• The above-mentioned measures combined could thus generate fiscal savings in the range of 3½-5½ percentage points of GDP, which under the projections would be sufficient, at a minimum, to achieve the balance objective without an increase in the revenue ratio.

¹Based on the analysis presented in IMF (2000d).

It is crucial to recognize that restraint in managed expenditure can only be sustained over the medium term if it is achieved through structural reforms. While it is generally feasible to achieve short-term savings in public consumption, for example, by containing annual wage increases, this is unlikely to be sustainable or even desirable given that public sector wages in some areas have been squeezed considerably in the past. To avoid a general deterioration in skills in the public sector, future measures need to focus on civil service reform—reducing staff in areas where productivity is low, while allowing greater flexibility to attract qualified personnel. Finally, there is scope to improve the efficiency of public services, including the strengthening of incentives for public sector employees (for example, through performance-based remuneration).

In the area of social transfers, structural reforms need to target improvements in incentive systems. To this end, it is important to undertake substantive reforms that alter basic features (such as eligibility and indexation rules) and improve compliance with existing criteria. At the same time, however, it must be recognized that spending on education and health care constitutes investment in human capital, which is important for growth-oriented strategies, and thus needs to be carefully considered.¹³⁶ Careful consideration should also be given to a basic social safety net, which would include training and education programs, to ensure that those who are displaced during the transition process do not fall into poverty. Finally, as economies grow stronger, they will become able to deliver progressively more generous support, though they will of course have to make sure that these benefits do not overstretch budgets, or create inappropriate disincentives to work or saving. Ultimately, the decision about expenditure reforms is a political one. To the extent that the existing expenditure structures already reflects differences in political priorities, a combination of measures affecting the entire spectrum of primary current spending may well prove the most effective approach. Box 7.5 in the previous two pages illustrates such a possible expenditure-restraint scenario, using Poland as an example.

Reconciling Fiscal Objectives

The design of fiscal policy is complicated by several conflicting factors. First, in some countries, significant restructuring costs still have to be absorbed by public finances, and the cost of reforms in public services, such as health care, could be substantial. Countries will also have expenditures associated with the adoption of the acquis communautaire. Second, although taxes on labor are typically heavy and need to be reduced, there is uncertainty about how fast the total tax burden on the economy can realistically be cut, while keeping within targets for the fiscal balance. Third, if private saving does not increase in parallel with private investment, public saving may have to increase to relieve pressure on external current accounts. Fourth, reconciling the fiscal objectives will require difficult adjustment in managed spending, namely, public consumption, subsidies, and social transfers.

Bringing together the arguments made above, six steps can be identified for designing a sound fiscal framework:

• determining appropriate goals of fiscal policy, taking into account debt dynamics and external vulnerability;

• developing a medium-term fiscal deficit path, consistent with the considerations above, based on prudent assumptions about growth and the private saving-investment balance;

• formulating a medium-term strategy for tax reforms;

• identifying spending that is precommitted or that cannot be curtailed in the near future (such as interest on debt);

• building in the estimated costs of reforms in the real economy, the financial sector, and the government; taking into account the extent to which they add to the public debt or affect the saving-investment balance; and assessing the costs of infrastructure improvements and a phased approach to environmental policy; and

• setting limits on managed public spending that reconcile the objectives above.

A first attempt in setting up a fiscal framework—on the basis of the six steps—may lead to unrealistically tight limits on managed spending and call for further iterations to converge to the deficit target. This would involve weighing the difficulties of certain cuts in managed spending against the costs of adjusting other components of the framework, such as postponing certain infrastructure projects, spreading out some accession-related investments, or revising the phasing of tax cuts. Box 7.6 illustrates possible applications of this approach on the basis of two contrasting country examples, featuring Hungary and the Slovak Republic in 1999/2000. At that time, limits for managed spending in Hungary seemed tight but feasible. In the Slovak Republic, on the other hand, a first iteration suggested that reconciling the various fiscal objectives would require significant adjustments in revenue objectives in further iterations, as well as more fundamental reforms of public spending programs.

Medium-Term Fiscal Management in Practice

Most of the accession candidates have formulated medium-term fiscal policy goals and guidelines in the context of multiyear projections, but have so far shunned the adoption of more formal medium-term fiscal frameworks. While the distinction between the two is not clear-cut, a key element of the formal medium-term approach, as understood here, is its explicit link to the annual budget process and the political commitment behind the medium-term objectives (Box 7.7 on page 180 provides a definition of medium-term fiscal frameworks, as used here). While many of the advantages of full-fledged medium-term fiscal frameworks occur at the microeconomic level (requiring, however, a number of institutional improvements to reap the benefits, in terms of better efficiency and accountability),¹³⁷ the pros and cons at the macroeconomic level are closely related to the relative benefits of commitment versus flexibility. Drawing on the experience in advanced economies, the following discussion focuses on how to strike a balance between commitment and flexibility that would be appropriate for the accession candidates, in terms of generating many of the benefits of medium-term fiscal frameworks without unduly limiting the room for maneuver.¹³⁸

Adopting Medium-Term Fiscal Frameworks: Commitment Versus Flexibility

The appropriate balance between commitment and flexibility in medium-term fiscal frameworks depends on country-specific circumstances, including the political feasibility of making credible commitments and the degree of uncertainty in the medium-term outlook. One of the fundamental arguments in favor of medium-term policy commitments is that they reduce the scope for political haggling with at least two desirable implications: (i) policy becomes more reliable, facilitating efficient planning at all levels; and (ii) medium-term objectives are more likely to be achieved, which in turn strengthens confidence. On the other hand, commitments eliminate flexibility which could either result in inferior policies in the face of unexpected developments or—if political pressures become too strong—failure to maintain the commitment, undermining its credibility and essential purpose. For this reason, firm commitments should be limited to areas where implementation is realistic. In other areas, more flexible commitments in the form of policy targets can be suitable substitutes, in particular if large uncertainties necessitate greater flexibility.

Experience in other countries may provide useful insights for designing effective medium-term fiscal frameworks in the accession countries. OECD economies have increasingly resorted to medium-term frameworks to guide fiscal policy. About half of all OECD countries are conducting their annual budget negotiations within a medium-term fiscal framework, including Australia, Austria, Canada, Denmark, Finland, Germany, the Netherlands, Sweden, and the United Kingdom. Not all of these frameworks would meet the formal criteria outlined in Box 7.7. Some countries, such as Finland, the Netherlands, and the United Kingdom, have successfully adopted such a formal approach, incorporating, however, different degrees of rigidity as well as other country-specific variations (see Table 7.1 for an overview of the main elements of fiscal frameworks in these three countries and a brief discussion of the pros and cons of alternative arrangements). The experience in these and other advanced economies suggests some tentative lessons for the design of medium-term fiscal frameworks under different country circumstances.

Table 7.1.

Medium-Term Fiscal Frameworks in Finland, the Netherlands, and the United Kingdom

Table 7.1.

Medium-Term Fiscal Frameworks in Finland, the Netherlands, and the United Kingdom

	Finland	The Netherlands	The United Kingdom	Pros and Cons
Horizon	Four years, on annually rolling basis	Four years, set at outset of coalition period	Three years on a two-year rolling basis (that is, with successive Frameworks overlapping by one year) for noncyclical expenditures (see coverage); cyclical or volatile spending is managed on an annual basis	Longer horizon and fixed ceilings provide more certainly and discipline to managers, but reduce flexibility in case of changing circumstances
Objective	In past, mainly geared to deficit reduction; in Future also to provide room for tax cuts	To achieve a parallel reduction in the deficit and tax burden	Over the cycle, to balance the current budget and maintain investment (and hence borrowing) at a sustainable level, while meeting the Government’s priorities	Prominence of deficit goal should reflect considerations of public debt sustainability (factoring in demographics), plus room for stabilizers; goal for taxes should reflect, among other things, extent tax wedge distorts labor market and thus long-run fiscal outlook
Coverage of medium-term spending targets	Total central government expenditures and subceiling for individual ministries, including transfers to local governments and social security funds; excludes local governments’ self-financed expenditure	Expenditures of central government (excluding infrastructure Fund) and social security Funds; subceilings on central government, health, and social security	Departmental Expenditure Limits (DELs), covering most noncyclical primary expenditures (some half of total expenditures); includes central government support for local government but not local authority’s self-financed expenditure	Comprehensive coverage is more transparent and effective but may constrain automatic stabilizers unduly; inclusion of local governments requires their autonomy to be limited, with risks of adverse incentive/accountability effects
Statutory nature of medium-term spending targets	Budget year binding; outer years agreed by cabinet as nonbinding norm to ministries; published but not submitted to parliament	Budget year binding; outer years politically but not legally binding; submitted ho parliament for information; contingency reserve to deal with changing circumstances	Budget year binding; outer years not legally binding but overall DELs announced in parliament; contingency reserve to deal with changing circumstances	Politically binding targets favor discipline and credibility; legally binding limits might constrain flexibility unduly
Nominal or real targets	Medium-term at constant prices of budget year; for new budget translated to current prices based on specific price and cost deflators	Medium-term at constant prices; for one-year budget, translated to current prices based on projected GDP deflator	All targets set in current prices	Medium-term based on constant prices facilitates real resource planning; one-year budget must be nominal as inflation anchor; use of category-specific deflators may risk discouraging desired adjustment to relative price changes
Macroeconomic assumptions	Underlying assumptions reflect central forecast	Growth forecast cautious; rule splits “growth dividend” between additional deficit and tax cuts	Growth forecast at conservative end of potential output range	Reasonably prudent assumptions an growth, inflation, and interest rates essential; but, if unduly so, may over time create credibility problems and “second-guessing”
Adjustments in target levels for outer years	Ceilings For outer years can be adjusted; automatic adjustment for wage and salary increases	No adjustment in real expenditure; only “automatic” translation of targets into current prices based on aggregate inflation forecast	Medium-term DELs in current price are adjusted only if new inflation forecasts differ significantly from original projections	s Adjustments provide flexibility in case of unanticipated exogenous developments or desired policy changes but reduce imposed discipline, transparency, and credibility of the system
Autonomy and incentives at lower levels	Ministries and agencies bound by annual budget; some appropriations transferable, avoiding “forced” consumption at end of fiscal year	Ministries and agencies bound by annual budget; part of funds can be carried over to next year; no penalties have been imposed for repeated healthcare overruns	Individual departments bound by DELs; within these limits departments can recycle any efficiency savings they make; unspent DEL funds can be carried over to next year; overruns are avoided by strict DELs	Carry-over discourages excessive spending at end of fiscal year; tolerance of overruns lowers credibility but is difficult to avoid, if coverage includes cyclical items or other expenditures that are difficult to control
Output objectives and quality control	Agencies establish result and efficiency targets, with ex post evaluation by ministries	No formalized output objectives	Publication of Public Service Agreements alongside spending plans, stating departments’ goals and setting specific quantified policy and efficiency targets	Experience with performance objectives positive; but tensions between output objectives and spending restraint possible, if output objectives introduced before budgetary discipline is inculcated

View Table

Table 7.1.

Medium-Term Fiscal Frameworks in Finland, the Netherlands, and the United Kingdom

	Finland	The Netherlands	The United Kingdom	Pros and Cons
Horizon	Four years, on annually rolling basis	Four years, set at outset of coalition period	Three years on a two-year rolling basis (that is, with successive Frameworks overlapping by one year) for noncyclical expenditures (see coverage); cyclical or volatile spending is managed on an annual basis	Longer horizon and fixed ceilings provide more certainly and discipline to managers, but reduce flexibility in case of changing circumstances
Objective	In past, mainly geared to deficit reduction; in Future also to provide room for tax cuts	To achieve a parallel reduction in the deficit and tax burden	Over the cycle, to balance the current budget and maintain investment (and hence borrowing) at a sustainable level, while meeting the Government’s priorities	Prominence of deficit goal should reflect considerations of public debt sustainability (factoring in demographics), plus room for stabilizers; goal for taxes should reflect, among other things, extent tax wedge distorts labor market and thus long-run fiscal outlook
Coverage of medium-term spending targets	Total central government expenditures and subceiling for individual ministries, including transfers to local governments and social security funds; excludes local governments’ self-financed expenditure	Expenditures of central government (excluding infrastructure Fund) and social security Funds; subceilings on central government, health, and social security	Departmental Expenditure Limits (DELs), covering most noncyclical primary expenditures (some half of total expenditures); includes central government support for local government but not local authority’s self-financed expenditure	Comprehensive coverage is more transparent and effective but may constrain automatic stabilizers unduly; inclusion of local governments requires their autonomy to be limited, with risks of adverse incentive/accountability effects
Statutory nature of medium-term spending targets	Budget year binding; outer years agreed by cabinet as nonbinding norm to ministries; published but not submitted to parliament	Budget year binding; outer years politically but not legally binding; submitted ho parliament for information; contingency reserve to deal with changing circumstances	Budget year binding; outer years not legally binding but overall DELs announced in parliament; contingency reserve to deal with changing circumstances	Politically binding targets favor discipline and credibility; legally binding limits might constrain flexibility unduly
Nominal or real targets	Medium-term at constant prices of budget year; for new budget translated to current prices based on specific price and cost deflators	Medium-term at constant prices; for one-year budget, translated to current prices based on projected GDP deflator	All targets set in current prices	Medium-term based on constant prices facilitates real resource planning; one-year budget must be nominal as inflation anchor; use of category-specific deflators may risk discouraging desired adjustment to relative price changes
Macroeconomic assumptions	Underlying assumptions reflect central forecast	Growth forecast cautious; rule splits “growth dividend” between additional deficit and tax cuts	Growth forecast at conservative end of potential output range	Reasonably prudent assumptions an growth, inflation, and interest rates essential; but, if unduly so, may over time create credibility problems and “second-guessing”
Adjustments in target levels for outer years	Ceilings For outer years can be adjusted; automatic adjustment for wage and salary increases	No adjustment in real expenditure; only “automatic” translation of targets into current prices based on aggregate inflation forecast	Medium-term DELs in current price are adjusted only if new inflation forecasts differ significantly from original projections	s Adjustments provide flexibility in case of unanticipated exogenous developments or desired policy changes but reduce imposed discipline, transparency, and credibility of the system
Autonomy and incentives at lower levels	Ministries and agencies bound by annual budget; some appropriations transferable, avoiding “forced” consumption at end of fiscal year	Ministries and agencies bound by annual budget; part of funds can be carried over to next year; no penalties have been imposed for repeated healthcare overruns	Individual departments bound by DELs; within these limits departments can recycle any efficiency savings they make; unspent DEL funds can be carried over to next year; overruns are avoided by strict DELs	Carry-over discourages excessive spending at end of fiscal year; tolerance of overruns lowers credibility but is difficult to avoid, if coverage includes cyclical items or other expenditures that are difficult to control
Output objectives and quality control	Agencies establish result and efficiency targets, with ex post evaluation by ministries	No formalized output objectives	Publication of Public Service Agreements alongside spending plans, stating departments’ goals and setting specific quantified policy and efficiency targets	Experience with performance objectives positive; but tensions between output objectives and spending restraint possible, if output objectives introduced before budgetary discipline is inculcated

Box 7.7.

Defining Elements of Medium-Term Fiscal Frameworks

Formal medium-term fiscal frameworks, as defined here, are essentially plans for fiscal activity. They cover revenue and expenditure for the current budget and the subsequent period, typically two to three years. While these plans are naturally based on various assumptions and projections (for example, GDP growth and inflation), they also incorporate policy commitments, similar to annual budgets, which can vary in the degree to which they are politically binding. A key defining feature of a medium-term fiscal framework, in contrast to simple medium-term projections, is its formal status within the annual budget process. The two need to be fully reconciled, and the forward estimates are effectively agreed as notional budgets for subsequent years, and form the starting point of the following year’s budget negotiations. While the specific features may differ across countries, the following are some basic characteristics of formal medium-term fiscal frameworks:

• A macroeconomic framework with projections of key variables relevant for fiscal revenue and expenditure developments, such as real GDP growth, inflation, and unemployment.

• A statement of policy objectives, including the path for the fiscal deficit, tax measures included in revenue projections, and expenditure targets, consistent with the revenue and deficit paths.

• Top-down translation of overall expenditure targets into spending norms for individual ministries, and bottom-up forward estimates of expenditures by the individual spending ministries. These estimates should include a costing of existing policy commitments and identification of proposed new commitments.

• A transparent decision-making process for ensuring consistency between the top-down and bottom-up estimates, typically resolved at the cabinet (or equivalent) level.

• Publication of the framework (and possible approval by parliament), highlighting both macroeconomic objectives and monitorable commitments on sector spending and ouputs.

While paying particular attention to the uncertainties dominating the fiscal outlook in the five accession countries, one can summarize the key lessons as follows:

• Focusing primarily on a time horizon of three years in addition to the budget year is most common. To assess better the implications of individual spending programs and tax measures, it may be useful to extend the projection period (for example, to five years as in the case of PEPs). Of course, the feasible duration for political commitments may be shorter, giving projections in outer years a more informative purpose.

• The choice between fixed or rolling frameworks is a major factor in determining the degree of flexibility. Rolling frameworks—which are updated annually and are extended to include one additional year of projections—entail a lower degree of precommitment and are more appropriate in circumstances of high uncertainty.

• The underlying macroeconomic projections should be realistic, possibly erring on the side of caution rather than optimism, to provide a safety margin, particularly when fiscal adjustment is vital. A conservative approach, which errs on the side of fiscal overperformance, is also justified, when the political cost and credibility loss—or, for that matter, the economic consequences—are larger in the case of underperformance. Moreover, the fiscal measures to correct underperformance, namely, additional expenditure cuts or tax increases, are clearly more difficult to implement than the tax reductions or higher spending, which could be considered in the case of overperformance. Obviously, the timing of the latter needs to take account of the cyclical conditions so that fiscal overperformance does not result in pressures to add stimulus (through additional tax cuts or expenditure increases) when the economy is at risk of overheating.

• Both the policy statement regarding the fiscal objectives and the underlying assumptions should be transparent. This includes clear indications about the circumstances that would justify deviations from individual targets. Transparency in this context is essential to distinguish policy commitments from projections and avoid a loss of credibility in the case of simple projection errors. Again, a higher level of uncertainty, which raises the likelihood and magnitude of projection errors, gives this feature more importance.

• Policy commitment is typically expressed in terms of expenditure ceilings. This permits the operation of automatic stabilizers in both directions on the revenue side, while securing the consolidation effort on the spending side.

Coverage should ideally extend to all levels of general government. However, commitments, in terms of spending ceilings, may have to be limited to central government expenditures, depending on the degree of autonomy of local governments (see below). Cyclical spending components are sometimes excluded, to permit expenditure stabilizers to operate in both directions and avoid slippage in noncyclical outlays in times of economic upturns. Similarly, particularly uncertain and volatile spending items could be excluded from the ceilings to avoid having to compensate for their unexpected movements by ad hoc adjustments in other expenditures. The risk, however, is that excluded expenditures may undermine the medium-term objectives. A possible alternative may be to combine tight subceilings for managed expenditures with some upper bound for overall spending.

Approaches to inflation differ greatly. Real planning provides greater resource certainty, as inflation shocks are accommodated. The use of specific deflators for different spending categories, however, is problematic as it discourages desired adjustments to relative price changes.

These lessons, translated to the specific circumstances of the accession candidates, suggest four basic recommendations for implementing formal medium-term fiscal frameworks.

First, given the wide range of uncertainties in the external environment and the cost implications of reforms, medium-term frameworks should be formulated on a rolling basis (that is, updated every year to include an additional year of projections), in order to permit adjustments to shocks. The time horizon should be sufficient to signal pressures ahead, and indeed, a five-year period will be mandatory under the PEPs—though the firmness of the political commitment would be stronger in the earlier years and influenced by the election cycle. The medium-term framework should be complemented by longer-term projections for demographically sensitive revenue and expenditure items (such as pensions) to indicate potential reform needs early on.

Second, medium-term fiscal objectives should be based on conservative macroeconomic projections and prudent assumptions about private sector behavior. As argued above, this rule is justified by the ease of implementing subsequent tax cuts, in the case of fiscal overperformance, relative to the political difficulty of introducing additional spending restraint (or undesirable tax increases) when fiscal deficit targets are in danger to be exceeded. Some bias in favor of fiscal overperformance is also consistent with the general objective of reducing deficits. However, mechanisms will be needed to help determine how policy should respond when faced with fiscal overperformance, taking into account cyclical conditions and external pressures, at the time.

Third, medium-term fiscal frameworks should specify the government’s strategy for tax reforms, but maintain (and clearly state) some flexibility. This includes both allowing stabilizers to operate and maintaining discretion about the precise phasing in of envisaged tax cuts. The latter provides a safety margin to protect priority spending (on infrastructure, for example) in the event of negative surprises.

Fourth, firm political commitments should be made in terms of ceilings for managed expenditure, covering the bulk of primary current spending that is targeted for adjustment over the medium term. Setting the ceilings is an iterative process that reconciles deficit and revenue objectives with quantified spending needs in priority areas and projections of spending on uncertain transition- and accession-related items. Critically, the ceilings must catalyze agreements on reform measures to achieve them, and firm political commitments, by increasing the stake of meeting the ceilings, help ensure that they are realistic and credible. Nonetheless, preservation of fiscal objectives may call for subsequent adjustments in managed spending—possible in a rolling framework—or a reconsideration of the other components of the medium-term plan, such as the timing of tax cuts or of certain transition- and accession-related spending, if warranted.

Additional Risk Factors: Privatization Receipts and Fiscal Decentralization

While privatization inflows—still expected to be sizable in the Czech and Slovak Republics and Poland—can be instrumental in reducing the public debt and future interest payments, they also risk weakening the fiscal reform momentum, if spent unwisely. Projections of average annual inflows over the period 2001–05 range from fairly small amounts (in percent of GDP) in Hungary (where privatization is mostly complete) and Slovenia, to 1 percent of GDP in Poland, and some 2½ percent of GDP in the Czech and Slovak Republics, but both the magnitude and timing are highly uncertain. In addition, all countries are expected to generate receipts from the sale of mobile phone (UMTS) licenses. While these inflows, if used to retire public debt, can help offset part of the cost of reforms by lowering future interest expenditures,¹³⁹ they also entail risks: specifically, vast additional receipts are likely to generate pressures for higher spending or, at least, weaken public support for expenditure restraint. Thus, privatization receipts and other “windfall gains” (such as the ones associated with the sale of mobile phone licenses) might well result in higher medium-term deficits (excluding these receipts) in the absence of a clear and credible agreement to repay public debt. By reducing future interest payments, a debt-reduction strategy would spread the benefits of privatization over a longer period, without weakening the reform momentum. The freed resources could then be used to advance reforms, tax cuts, or the envisaged deficit reduction.

An additional and more lasting complication for the achievement of medium-term fiscal objectives stems from the devolution of fiscal authority to local governments. In three of the five accession candidates, namely, the Czech Republic, Hungary, and Poland, local government activity accounts for at least one-fifth of overall government spending. Decentralization, while partly a historical country-specific phenomenon and partly driven by the accession process (with the need for regional levels of government to qualify for EU cohesion funds), also reflects deliberate decisions of governments to allow greater responsiveness of spending decisions to the preferences of citizens in different constituencies.¹⁴⁰ On the other hand, a large fragmentation of fiscal authority poses problems of its own. Thus, it is important to put in place mechanisms to ensure budgetary discipline.¹⁴¹

A possible prudent approach would suggest maintaining fairly tight budget constraints on local governments while allowing sufficient room for spending decisions within these constraints. This could be achieved by assigning some limited taxing powers to local governments; linking earmarked transfers from the central government to objective criteria; and setting effective limits on local government borrowing subject to their payment capacity. Besides striking an appropriate balance between control and accountability, the main challenge is to ensure that the mechanisms supporting fiscal discipline do not undermine other objectives—including the provision of sufficient resources to deliver the desired level and quality of public services; and the maintenance of broad equity across regions and municipalities. Reforms to strengthen the efficiency of local governments are warranted and likely to require larger effective autonomy at the subnational level over time, both in terms of generating own revenues and gaining greater access to financial markets. This will call for mechanisms to enhance fiscal coordination, create additional incentives for local governments to pursue policies consistent with overall medium-term objectives, and improve the local government’s management capacity. The latter will be particularly challenging in light of EU accession, as many EU-related public investments will have to be carried out at the subnational level.

These additional considerations will need to be factored in when designing fiscal frameworks. This complicates the process, not least because of the effort required to get commitments upfront. But by making these commitments, the likelihood of fiscal policy success is significantly increased.