Latvia's Macroeconomic Options in the Medium Term Fiscal and Monetary Challenges of EU Membership Célestin Monga1 Abstract Latvia's experience over the past decade shows that economic growth and real convergence can no longer be assumed to be exogenously driven processes determined by given technological improvements and relatively higher factor returns. Instead, it is an endogenously driven process led by many variables, including policy variables. European membership clearly brings enormous economic benefits to Latvia. However, it also brings important challenges on the macroeconomic front, especially prior to the adoption of the euro. Substantial progress has been made toward fiscal consolidation since the mid-1990s but several risks remain. Given the facts that domestic demand remains buoyant and that Latvia will have to accommodate the expenditure commitments associated with NATO and EU membership while simultaneously aiming to fulfill the medium-term goal of a balanced budget, a more prudent fiscal policy should be the main short-term policy objective. The authorities should therefore reassess their current medium-term budget framework and strengthen their fiscal rules so that off-budget spending is eliminated and controls over spending ministries and local government finances are reinforced. They should also rethink any further tax reductions until a clear strategy is designed for compensating for projected losses in government revenues. Monetary policy can contribute to sustainable growth and job creation in Latvia in the medium and long term by providing an environment for price stability. It will have to be complemented by further microeconomic reforms and prudent wage developments aligned with productivity growth. World Bank Policy Research Working Paper 3307, May 2004 The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Policy Research Working Papers are available online at http://econ.worldbank.org. 1Célestin Monga was Senior Country Economist for the Baltic countries (Estonia, Latvia, and Lithuania) when this research was initiated. He is with the Macroeconomics Group in the Europe and Central Asia Region of the World Bank. The author would like to thank Olivier Blanchard, Bernard Funck, Inguna Sudraba (formerly Ministry of Finance of Latvia), and Helmuts Ancans (Bank of Latvia) for their comments on a previous draft. Andrejs Jakobsons and Lodovico Pizzati provided excellent research assistance. 1 Contents 1 Introduction..........................................................................................................3 2. Latvia's Stabilization Program: An Interim Balance Sheet................................................4 2.1 Rationale and Main Features ..............................................................................4 2.2 Outcomes of Stabilization .................................................................................6 3. Looking Ahead: The Quest for Convergence.................................................................11 3.1 Long Term Sources of Growth: Lessons from History................................................11 3.1.1 Determinants of Growth.........................................................................11 3.1.2 Lessons for Latvia.................................................................................12 3.1.3 Growth Scenario in the Long Term...........................................................14 3.2 Why Labor Market Issues are Key to Latvia's Convergence with the EU.........................14 4. Macroeconomic Options with EU Membership..............................................................15 4.1 A Fiscal Strategy for the Medium Term................................................................17 4.1.1 Fiscal Developments in the Last Decade.....................................................17 4.1.2 Fiscal Policy and Unemployment: How Latvia Dealt with the Business Cycle........19 4.1.3 Medium-Term Outlook and Expenditure Strategy.........................................20 4.1.4 The Need to Rethink Fiscal Rules in Latvia................................................23 4.1.5 Strengthening the MTEF with New Rules and Targets....................................24 4.1.6 Practical Issues in Designing Expenditures Rules for Latvia ............................25 4.2 Monetary Policy, Growth, and Employment..........................................................27 4.2.1 Benefits of Price Stability for Growth and Job Creation..................................28 4.2.2 Latvia's Monetary Policy and Convergence................................................30 5. Conclusion...........................................................................................................31 Text Boxes Box 1: Latvia's Macroeconomic Recipe, 1992-2002............................................................................11 Box 2: Growth Accounting in CEEC ..............................................................................................13 Tables Table 1: Indicators of Latvia's Macroeconomic "Big-Bang".....................................................................4 Table 2: Projected TFP Growth Rates for the Baltic Countries.................................................................14 Table 3: Sustainable primary fiscal balance......................................................................................21 Table 4: Tax Burden and Effective Tax Rates (2002 or latest year available)................................................22 Table 5: Latvia and the Maastricht Criteria.........................................................................................30 Table 6: Episodes of Fiscal Adjustment and Stimulus in Transition Economies, 1992-2001..............................36 Figures Figure 1: GDP Growth, Productivity, and Unemployment in Latvia............................................................7 Figure 2: Financing of the Fiscal Balance (in millions of Lats)...................................................................8 Figure 3: Balance of Payments, 1998-2002 (percent of GDP)...................................................................8 Figure 4: Savings-Investment Balance, 1995-2001 (percent of GDP)..........................................................9 Figure 5: Government Revenue and Expenditure (as percent of GDP)........................................................17 Figure 6: Composition of Expenditure, 1992-2002................................................................................18 Figure 7: Fiscal Balance and GDP Growth (as percent of GDP)................................................................19 Figure 8: Composition of Government Revenue, 1992-2002, as percent of GDP ...........................................20 Figure 9: The Fiscal Cost of EU-Accession (as percent of GDP)..............................................................22 2 1 Introduction Latvia is part of the first wave of a group of 10 Central and Eastern European countries (CEECs) to become members of the European Union (EU) on May 1, 2004. For a country that regained its independence only in 1991, the transition from a communist regime into a market economy has been difficult but remarkable. Latvia has recorded an average real GDP growth rate of almost 5 percent over the past 7 years. This is well above the EU average and that recorded in many transition economies. In contrast with the pattern observed in some other transition countries, this post-contraction growth has been evenly paced; boom-and- bust experiences have been avoided, and the expansion has been well balanced on both the demand and supply sides. While output in 2002 was still lower than in 1992, real per capita GDP now stands more than 50 percent above its 1995 level. Recent good performance on the output front contrasts with the experience during the first years of independence and transition from communism. In fact, the story of Latvia's transition is one of two phases. First, the initial transition, from 1991 to 1995, was characterized by a sharp decrease in output,2 a sharp decrease in employment, and a sharp increase in unemployment. Second, the recovery, starting in 1996, has been marked by an increase in output, substantial productivity growth, and a decrease in unemployment, largely explained by a decrease in participation. The impressive growth performance has improved labor market conditions but unemployment remains high (11 percent in 2003 from a peak of almost 21 percent in 1995). Moreover, unemployment and underemployment are largely concentrated among certain vulnerable groups and in some economically disadvantaged areas. The country still faces the challenge of creating enough employment to keep an important fraction of the labor force at work. The severity of the initial output shock may have provided some benefits. A disastrous early 1990s situation quickly helped focus the national economic and political debate on the deeper and longer-run issue of structural reforms. The need for and acceptance of reforms was facilitated by several factors: first, the political consensus on the need to join the EU as soon as possible--a vision shared by most people in the three Baltic countries (Latvia, Estonia, and Lithuania); and second, the fears created by the uncertainties of the first years of transition (output collapse, high inflation, and rising unemployment in the context of political fragmentation). Section 2 of the paper briefly reviews Latvia's macroeconomic performance during its first decade of independence and presents the main features and outcomes of a rather successful stabilization program. Section 3 outlines the challenges for long-term convergence with EU countries, focusing on the role of the labor market. Section 4 suggests a macroeconomic strategy after EU accession, focusing on the role of fiscal and monetary policies. Section 5 offers some concluding remarks. 2Ĺslund (2001) suggests that the fall in real output during the first years of transition may have been overstated, though it remains large. Measures of GDP both prior to transition and in the initial years suffer from important methodological problems: valuation of output produced under non-market conditions is difficult. The problem is compounded with the rapid development of informal sector activities (not well captured in official statistics) after independence. 3 2. Latvia's Stabilization Program: An Interim Balance Sheet Like all strong reformers from the former Soviet Union, Latvia quickly embraced the two- pronged macroeconomic strategy of stabilization and liberalization (see Mitra and Selowsky 2002 and World Bank 1996). Stabilization involved designing macroeconomic policies aimed at reducing inflation and containing domestic and external imbalances. Liberalization meant freeing prices, trade, and privatizing the large state enterprise apparatus. 2.1 Rationale and Main Features The stabilization effort in Latvia drew from the lessons learned in historical waves of stabilization including Europe in the 1920s, and Latin America and Africa in the 1980s. Shortly after independence, policymakers in Latvia designed a very ambitious yet feasible stabilization package. Although the issue was not to repair the damage suffered by an existing market economy but rather to start one from scratch, they applied the same logic to conceiving the economic strategy that the country then followed rigorously over 10 years. This was achieved despite the fact that Latvia averaged one new government per year during this period. In addition to the traditional stabilization package, some features reflecting the specificities of the country's political and social context were added to the strategy-- sometimes in defiance of policy recommendations from international financial institutions. At the origins of the process were the collapse in trade and output observed in Latvia (and in most transition economies) in the early 1990s following the disintegration of the Soviet Union. In 1992, the first year after independence, gross domestic product fell by 32 percent (see Table 1). At the same time, the price of imported goods--especially energy--increased dramatically to international levels, causing a decline in the terms of trade of 53 percent during that year. The following year, GDP continued to fall by over 11 percent and stood at about half its 1990 level. Table 1: Indicators of Latvia's Macroeconomic "Big-Bang" Indicator 1991 1992 1993 GDP growth -12.6 -32.1 -11.4 Inflation (consumer price index, in percent) 262 958 35 Real wage change -16.3 -15.7 0.5 Unemployment 0.1 2.3 5.8 Terms of trade change 1.0 -53 8 Fiscal balance/GDP 6.4 -0.8 0.6 Current account balance/GDP n.a 1.8 6.8 Source: World Bank and Latvian authorities Inflation was facilitated by two fiscal factors. First, the shock directly affected government revenues. These decreased sharply from 37.4 percent of GDP in 1991 to 28.2 percent of GDP in 1992. While the Latvian authorities managed to cut public expenditure, the fiscal balance went from a surplus of 6.4 percent of GDP in 1991 to a deficit of 4.0 percent in 1994 (a swing of over 10 percentage points). Second, under heavy political pressure to reject the economic contraction, the first government after independence tried to reconcile income claims from various social groups and accommodated wage and price increases through 4 money expansion. This triggered a wage-price spiral.3 Because fiscal and monetary policies were too lax in 1992, the interaction between the budget deficit and inflation eventually resulted in accelerating inflation. This led private agents in Latvia to economize on money balances. In turn, the change in money-holding behavior required a higher rate of money growth to generate the same amount of real revenue. This led to even higher inflation. Inflation reached 958 percent in 1992. At high inflation rates and with its collection lags and the fall in output, Latvia's tax system yielded less revenue. This aggravated the deficit and the pressure on money growth. Attempts by the authorities shortly after independence to protect some social groups by increasing subsidies on publicly supplied goods (e.g. heating and housing) further worsened the price structure and put extra pressure on the budget. By mid-1992, economic disruptions and inflation reached very high levels. The economic uncertainty triggered political uncertainty and huge social fears, and engendered the support necessary to implement a stabilization program. Latvia's stabilization package was launched in July 1992 and was supported by a stand-by arrangement with the IMF. It was largely conventional in its main features, but the very strong political commitment by the authorities made it different and successful. It included the following macroeconomic measures: · Fiscal restraint to bring down the government deficit and thereby restrain aggregate demand pressures; · Monetary restraint aimed at reducing the growth of absorption and inflation. This was done in a very decisive way by de-linking the currency system from the ruble area and introducing a new currency, the lat; · Exchange rate action to adopt a nominal anchor and to ensure a real exchange rate that improves international competitiveness and also creates incentives for expanding the production of tradable goods (the lat was pegged at a fixed rate to the SDR); · Interest rate policy aimed at keeping real interest rates positive to ensure the credibility of the fixed exchange rate system, but not so high as to run the risk of aggravating the economic contraction; · External financing to ensure sustainability of the current account, attract foreign direct investment, and limit foreign borrowing; · Structural reforms to make the Latvian economy flexible and efficient. This involved the liberalization of producer prices (except in a few areas such as heating and rent) and trade. Simultaneously, privatization, financial sector reform, and tax reform were undertaken. 3 This process is typical of many countries faced with a huge adverse shock, a big change in political power, or a shift to democracy that lead to inconsistent claims to the distribution of income. For Eastern European countries, see Blanchard et al. 1992. 5 EU accession has always been one of the driving forces behind the country's adjustment and reform efforts. As pointed out by two senior government officials, because Latvia must meet specific conditions and implement specific policies to get into the "club," "EU accession also served as a primary goal, helping to break political deadlocks and counterbalance the destructive activities of lobbies."4 Furthermore, economic policy programs and lending programs with the IMF and the World Bank, as well as the goal of joining the World Trade Organization--which Latvia did in February 1999--also served as external anchors, as they "unquestionably influenced policymaking because the conditions associated with membership provided targets that helped policymakers reach consensus on launching and sustaining politically difficult reforms designed to liberalize the economy and trade."5 2.2 Outcomes of Stabilization By all accounts, Latvia's economic strategy has been relatively successful and the country has now completed its transition from communism to a market economy. Real GDP has grown on average by nearly 6 percent annually over the period 1995-2003, following falls in the early years of transition. Growth has been broad-based, with especially strong gains in forestry, business services, financial intermediation and manufacturing. The key factors in building the growth momentum were the acceleration of domestic private and public demand. Domestic consumer and investment demand were supported by growing real incomes and tax revenues, and reinforced by improving consumer and investor confidence and projections of rising real incomes in the medium term. Low capital-intensive and fast-maturing foreign direct investment (FDI) contributed to an increase in productive capacity and the enhanced competitiveness of Latvia's manufacturing base. The economic expansion was easily financed with relatively inexpensive domestic credit provided by foreign direct portfolio investment inflows and a competitive financial system. Growing demand for Latvia's exports in both the traditional Commonwealth of Independent States (CIS) and new EU markets, as well as sound trade policy, also contributed to growth momentum. During its first decade as an independent country, Latvia needed to retain established economic links to its Eastern neighbors as well as to develop trade with new Western partners. In 1997 and 1998, exports equaled 51 percent of GDP, but fell to 44 percent of GDP after the sharp depreciation of the ruble in 1998 and default on Russia's sovereign debt. The real effective exchange rate appreciated by about 25 percent compared to its pre-crisis level, and the trade balance deteriorated to 19 percent in 1998 (from 15 percent in 1997). The Russian crisis sparked a process of adjustment in Latvia's external trade. Although the ratio of exports to GDP did not change significantly, the composition of trade flows changed substantially. A shift toward Western European markets continues to occur. Prior to the crisis, EU and CIS markets accounted for 49 percent and 30 percent of the country's exports, respectively. In 2003, more than 60 percent of Latvia's exports were directed toward EU countries and only 10 percent to the CIS. Despite the appreciation of the real effective exchange rate in 1997-2001 and the recent slowdown in the world economy, Latvia has 4 Zile and Steinbuka (2001). 5 Idem. 6 managed to maintain strong export growth. This reflects improved productivity in key sectors and the resilience and sound fundamentals of the economy. But strong economic growth and increased productivity have not yet been reflected in a structural improvement in labor market conditions. Employment has declined sharply during the first decade of independence, broadly similar to the experience in transition economies in general, reflecting in part a larger initial output loss. Starting from virtually zero at independence in 1991, unemployment in Latvia rose to a peak of 20.7 percent in 1996 (ILO definition). Figure 1 shows only a gradual decrease over recent years, in spite of sustained GDP and productivity growth. Figure 1: GDP Growth, Productivity, and Unemployment in Latvia 25 12000 20 10000 15 8000 10 6000 5 4000 0 2000 1995 1996 1997 1998 1999 2000 2001 2002 2003* -5 0 GDP growth, % (LHS) Unemployment rate, % (LHS) P roductivity, LVL (RHS) Monetary policy has been consistently tight as the Central Bank concentrates on maintaining price stability in the country. Since its inception in 1994, Latvia's fixed exchange rate regime, tied to the SDR, has served the country well. After reaching 109 percent in 1993, CPI inflation has been at single digit levels since 1997. In 2003 the CPI grew by 2.9 percent. Interest rates have come down significantly since the 1998 Russian crisis that affected the banking system. The annualized yield on a 12-month Treasury bill, which was 9.8 percent at the beginning of January 1999, was 3.5 percent in December 2003. Fiscal policy has been conservative from the outset of the stabilization program with a few notable exceptions caused by domestic or external shocks. Owing to its generally good fiscal discipline, Latvia has low levels of public debt, and its credit ratings have improved, giving the country easier and less expensive access to international capital markets. However, the fiscal situation in 1999 and even in 2000 showed the consequences of the Russian crisis. From a surplus of 0.1 percent of GDP in 1997, the fiscal deficit rose to 3.9 percent of GDP in 1999 and remained relatively high in 2000 at 3.3 percent. This was above the target agreed with the IMF. In 2001, the situation improved somewhat and the fiscal deficit decreased to 1.4 percent of GDP. Still the budget for 2002 was for an increase in the fiscal deficit to 2.7 percent, significantly above the target of 1.5 percent. More worrisome is the fact that the 7 2003 budget envisaged an even higher deficit (3.2 percent of GDP) notwithstanding the high growth rates of recent years. Fortunately, Latvia has been able to finance its fiscal deficit over the past decade mostly with privatization receipts (1997-98), through the banking system (1999-2000), and with some foreign financing (1999 and 2001), without either putting pressure on government debt or crowding out the private sector. Figure 2: Financing of the Fiscal Balance (in millions of Lats) 150 100 50 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 -50 -100 -150 -200 Privatization Banking system Other domestic Foreign financing Fiscal balance Following several years of small external current account deficits that were entirely financed by foreign direct investment, Latvia's current account deficit (including official transfers) rose from 6.1 percent of GDP in 1997 to 9.8 percent in 1999, with only half financed by FDI. This deterioration was largely initiated by the Russian crisis. There was some improvement in 2000 when the current account deficit was brought to just under 7 percent of GDP despite appreciation of the lat against the euro and the rise in oil prices. However, it widened again in 2001, reaching 9.6 percent of GDP. Figure 3: Balance of Payments, 1998-2002 (percent of GDP) While the external position Balance of Payments, 1998-2002 improved slightly in 2002 with the (Percent of GDP) deficit falling to 7.8 percent of GDP, 16 12 this reflected the impact of improved 8 4 statistical coverage that helped raise 0 private transfers. The fragile -4 -8 external position remains the main -12 -16 risk to the Latvian economy. -20 -24 -28 Like many transition economies, Mar Jun Sep Dec Mar Jun Sep Dec Mar Jun Sep Dec Mar Jun Sep Dec Mar Jun Sep Dec 1998 1999 2000 2001 2002 Est. Latvia has recorded relatively Service Account Trade deficit Income and transfers modest saving rates (19.5 percent of Current Account Source: Bank of Latvia. GDP on average over the past five years) even as investment needs during transition have been high (gross domestic investment averaged 28 percent of GDP in 1998-2003). Such conditions suggest that running a sustainable current account deficit, both to smooth consumption over time and to finance much-needed capital projects to catch up with EU countries, may be necessary from a welfare viewpoint. However, sizable investments will be needed to sustain economic growth in the medium term, as well as a potential catch-up in private sector consumption. These factors may compromise improvements in the external current account deficit, as imports will continue to grow 8 strongly and FDI-related profit remittances are expected to rise. On the other hand, a continued strengthening of Latvian exports to the EU, together with diversification in the export base under broad-based FDI, is projected to materialize during the period 2004-2006. As FDI is projected to remain buoyant-- Figure 4: Savings-Investment Balance, 1995-2001 (percent of GDP) because of the forthcoming 35 0 privatization of the 30 -2 remaining large public Gross Domestic enterprises like Ventspils 25 Investment -4 Nafta and Lattelekom 20 and the sale of new Gross National -6 Savings telecommunications 15 licenses--the external -8 Foreign Savings account is expected to be 10 (current account largely covered by non- deficit) 5 -10 debt-creating flows, thereby containing - -12 Latvia's already 199519961997199819992000200120022003* favorable debt indicators. Despite a wider current account deficit and lower FDI coverage in 2002 than in previous years, Latvia's debt indicators--including those for short-term debt--remain moderate. Public sector external debt represented 11.4 percent of GDP in 2002 and total external debt (excluding currency and deposits) represented 53 percent of GDP. While the most comprehensive measures of short-term debt, including foreign liabilities of the banking system, suggest that Latvia's short-term debt is high (47.8 percent of GDP), these liabilities are largely matched by high-quality, liquid foreign assets.6 In addition, the Latvian banking sector is sound and well supervised. Latvia has made great strides in building the institutions needed for developing a market economy through its aggressive structural reforms. For example, prices, trade, and other markets have been liberalized, and the necessary legal reform, institutional development, and social safety net improvements are being implemented. The remaining structural reform program is directed at attracting foreign investment and sustaining growth. The focus of the agenda is to implement measures identified as catalysts to improve the business climate, to foster competition in the energy and telecommunications sectors, to complete the privatization of large-scale public enterprises, and to combat corruption.7 The restructuring and privatization process for state enterprises has been completed with the notable exceptions of several large companies--Latvenergo (power), Lattelekom (communications), and Ventspils Nafta (oil pipeline company). Recent privatization transactions included the sale of 6 According to a recent IMF study, over two-thirds of Latvian short-term debt consists of nonresident deposits in commercial banks. These deposits are largely redeposited in foreign banks or placed in other secure liquid foreign assets in OECD countries chosen to match the currencies and maturities of the deposits. When assessing vulnerability, it is reasonable therefore to exclude nonresident deposits from short-term debt or to net the liabilities against their corresponding assets. 7 See FIAS (2003). 9 all remaining shares in Latvijas Gaze in February 2002 and the sale of LASCO through an offering of 32 percent of its shares for vouchers and 51 percent of its shares for cash on the Riga Stock Exchange at a price approved by the Board of the Latvian Privatization Agency. (The remaining 17 percent will go to employees, the State Pension Fund, and the privatization reserve.) Parliament voted in 2001 to put Latvenergo on the list of strategic companies not to be privatized. The authorities went along with the decision in order to prevent a public referendum that might have stymied privatization for many years to come, and possibly halted efforts to restructure the company. Taking into account lessons learned in similar situations in transition countries, the authorities and the Bank agreed that the appropriate sequencing was for Latvenergo first to be broken up. This was done so that the government wound not be privatizing a monopoly. With smaller companies, real competition over the grid could be organized, and a regulatory framework consistent with EU requirements could be put in place. In designing its electricity market, the government has adopted EU Directive 96/92/EC concerning common rules for the internal market in electricity and is adopting measures to further liberalize the market. With respect to Lattelekom, its strategic investor Sonera has taken the government to arbitration court to resolve the issue of compensation for shortening the company's exclusivity period from 2013 to 2003. The authorities have committed themselves vis-ŕ-vis the WTO and the EU to shorten this period. On Ventspils Nafta, the government has received the privatization plan and the company's financial valuation from the international investment bank hired to prepare Ventspils Nafta's privatization. They have reaffirmed their commitment to privatizing their shares in the company in an open, competitive manner and at a fair price in 2003. In the area of private sector development, an ambitious effort was launched in 1999 with the Action Plan to Improve the Business Environment, which aims at simplifying tax administration, streamlining the administrative procedures of customs clearance, and providing transparency and better protection of investors' interests. This plan has recently been updated. A new version approved in October 2002 envisages measures to improve tax policy and administration, combat corruption, improve public procurement procedures, and foster confidence in the rule of law. 10 Box 1: Latvia's Macroeconomic Recipe, 1992-2002 Like the two other Baltic countries (Estonia and Lithuania), Latvia will probably be studied by future economic historians as an interesting case of a rapid, successful transition from a communist regime to a market economy. The following (non-exhaustive) list of key policy decisions served the country well during its first decade of independence: A vision backed up by a national consensus on long-term political goals and strong external anchors. In spite of Latvia's wide political spectrum, most political leaders agreed at the outset of the transition on the country's main goals. The prospects of integration to the EU, the goal of joining the WTO, and program arrangements with the World Bank and the IMF have also generated over the years strong incentives for the adoption of sound policies. A conservative fiscal stance throughout most of the decade. Latvia's fiscal deficit only exceeded 3 percent in 1994 (owing to increases in net lending), in 1995 (owing to the banking crisis), and in 1999 and 2000 (owing to the Russian crisis). By and large, the authorities have always opted for a tight fiscal stance as a way to fight inflation and limit external imbalances. A prudent monetary policy with a clear exit strategy. To combat inflation, support business confidence and enhance stability in financial markets, the BOL has pegged the lat to the SDR since 1994--the structure of the SDR basket largely corresponds to the currency composition of Latvian foreign trade and debt. The BOL has also demonstrated commitment to defend the external value of the Lats by raising interest rates whenever it has come under pressure. The widening of the fiscal deficit over the past two years makes the goal of maintaining a peg and keeping inflation in check more difficult. The main challenge ahead is to maintain the Lats' external competitiveness, prepare to repeg it to the euro (following the Lithuanian strategy with the litas), and ultimately determine an appropriate central parity for it in the ERM II. Financial sector reforms. Faced with a major banking crisis in the mid-1990s, the authorities responded by immediately implementing tighter prudential regulations, raising the minimum capital requirement for banks, enhancing monitoring through more frequent on-site inspections and the requirements of external audits. Structural reforms. Latvia liberalized its economy quickly, freeing prices at the beginning of its transition. The economy was opened to the world, allowing not only goods and services to flow freely over its borders but also capital, implying full currency convertibility for most current and capital account transactions. The Latvian authorities quickly dismantled the restrictive and inefficient trade regime inherited from the Soviet Union, established trade agreements with the EFTA members and the EU, and applied for WTO membership. Privatization was given an early boost with the sale of all small enterprises, mainly via auctions and open tenders to the highest bidder for cash. The sale and restructuring of some large utilities like Latvenergo (power), Lattelekom (communications), and Ventspils Nafta (oil pipeline company) was delayed to build political consensus. Finally, a series of laws were adopted by Parliament to enhance the transparency and efficiency of public sector operations. 3. Looking Ahead: The Quest for Convergence Latvia has achieved enough progress in macroeconomic stabilization and structural reforms to gain EU membership. Yet, convergence to EU income levels will take time. Supply factors will determine economic growth in the long term, whereas demand factors will dominate movements in economic activity in the short run. This section outlines the long- term prospects for Latvia and the importance of labor market reforms in the convergence process before and after EU accession. 3.1 Long-Term Sources of Growth: Lessons from History 3.1.1. Determinants of Growth The prospects for the real sector in Latvia will determine the appropriate policy framework before and after EU accession. Therefore, it is important to identify the key factors likely to 11 shape the pace and sources of growth in the medium and long term. Historical evidence from European and Asian growth (Japan and also the so-called East Asian Tigers) in 1950-1975 shows that the sources of growth can vary enormously: while total factor productivity (TFP) dominated the growth in Western Europe and Japan, increased factor inputs seem to be the predominant factor in East Asia. Four factors usually explain diverging patterns of growth across nations:8 (i) differences in initial income levels: Western countries had a much higher per capita GDP relative to the United States in 1950 than did the East Asian countries in 1960; (ii) demographic trends: East Asia experienced a rapid rise in the labor force, which was not the case in Western Europe where the demographic transition was completed prior to the postwar golden age; (iii) differences in skill: the quality of the labor force (measured by average years of education) was much higher in Western Europe than in East Asia; and (iv) differences in market development: Western Europe had already established a well-functioning market system in 1950 whereas Asian countries relied on large integrated enterprises, with the state playing a major role in economic policy. 3.12 Lessons for Latvia From this historical experience, it appears that Latvia (like many transition economies) resembles Western Europe at the outset of its golden age in all the four policy areas identified above: · Latvia's income level relative to the United States (at current international prices) was about 25 percent in 2002, below the range of Western European countries in 1950 (from 32 in Italy to 59 percent in the United Kingdom). This initial income level means that the scope for growth from increased capital is likely to be as important as it was for East Asian countries. · Latvia's labor force is projected to decrease over the next decade. This demographic trend implies that labor may not be available to make a significant contribution to growth as it did in East Asia. · The average number of years of education in Latvia is relatively high, close to the levels prevailing at the outset of Western Europe's growth spurt (12 years).9 This means that the labor force may be better able to assimilate and use new technologies and make productivity gains than was the case for East Asian labor at the outset of the convergence process. · Finally, Latvia has already achieved significant progress in establishing many of the types of market institutions that were essential to the economic development of 8See Doyle et al. (2001). 9Source: 2001 World development Indicators CD-ROM, World Bank. According to Doyle et al. (2001), average years of schooling in Western Europe in 1950 varied between 4.9 in Italy to 9.4 in the United Kingdom. It should be noted that the quality and nature of schooling in transition countries during the socialist period, even for highly educated workers, has been subject to debate. 12 Western Europe in the 1950s. Though the relative merits of different institutional frameworks in supporting economic development remain subject to debate, empirical studies have shown that the systems in East Asia have been less effective in promoting efficiency and innovation. Thus, the main lesson from the international evidence is that TFP, rather than mobilization of factor inputs, has the greatest potential to spur growth in Latvia in the medium and long term. Box 2: Growth Accounting in CEEC Boldrin and Canova (2003) attempt to quantify the extent to which the next wave of EU accession countries (CEEC) are economically behind the EU15. They review trends in GDP per capita and labor productivity, savings and investment patterns, and carry out an ambitious growth accounting exercise. They find that despite the fact that the quest for convergence with the EU began to bear fruit in the mid-1990s, "the average CEEC is still where it was ten years ago relative to the average EU15 country." They document that, with the current investment rates in physical capital, it will take a considerable amount of time for CEECs to reach the capital-output ratio existing in the EU. The key drive for faster results will be the dynamics of technological acquisition in these countries. From the viewpoint of a simple growth accounting exercise, productivity gains, as measured by the growth rate of total factor productivity (TFP), seem to be the main determinant of long-term growth. Sources of TFP include foreign direct investment, shrinkage of the public sector or the agricultural sector, and institutional change. In the case of Latvia, it appears that the TFP contribution accounted for over 80 percent of output growth during the period 1991-2002, while labor contribution was negative. Time Period y/y k/k n/n TFP FDI/FDI Bulgaria 1991-2001 -0.04 0.006 -0.04 -0.006 0.001 Czech 1993-2000 0.02 0.02 -0.002 0.002 0.003 Estonia 1992-2001 0.02 0.015 -0.015 0.02 0.004 Hungary 1992-2001 0.023 0.014 -0.004 0.013 0.003 Latvia 1991-2000 0.077 0.008 -0.013 0.082 0.002 Lithuania 1991-2000 0.049 0.018 -0.014 0.044 0.002 Poland 1991-2001 0.032 0.012 -0.006 0.04 0.001 Romania 1991-2001 -0.004 0.01 0 -0.015 0.001 Slovak 1993-2000 0.03 0.022 -0.01 0.017 0.003 Slovenia 1992-2000 0.05 0.017 -0.01 0.042 0.001 CEEC average 0.025 0.014 -0.011 0.022 0.002 Notes: y/y = output growth; k/k = capital growth; n/n = employment growth; TFP = TFP change; FDI/FDI = FDI growth; AGR n/n = contribution of agricultural employment. The increments in capital stock are computed summing up investment over the period and subtracting yearly depreciation. Data for depreciation is available only for the three Baltic States: the depreciation rate is estimated to be around 40-50percent of the investment rate. We apply a value of 45percent to the other seven countries. The increments on the labor input are computed using bodies and not hours which are not available in many countries. Data for Hungary indicate that the difference is small. The domestic increment in the capital stock is calculated as residual. The same depreciation rate is applied to domestic capital and FDI, therefore biasing downward the contribution of FDI to growth. The contribution of agriculture to growth is calculated multiplying n/n by the decrement in the population employed in agriculture. Source: Boldrin and Canova (2003). 13 3.13 Growth Scenario in the Long Term Empirical research has highlighted the role of the income gap and the level of education in the convergence process. Using a methodology developed by Benhabib and Spiegel (1994) and already used by Doyle et al. (2001) for a panel of transition economies, we can estimate Latvia's TFP growth rate for the next decade. At the steady state, Latvia's capital output ratio will remain constant. As a result, per capita GDP growth at the steady state will be TFP growth over the labor share plus changes in the quality of the labor force. Projected TFP growth is presented in Table 2 below. Table 2: Projected TFP Growth Rates for the Baltic Countries Average years of Gap with US Projected TFP growth Labor force growth schooling 1999-2010 average Lithuania 12 4.6 4.2 0.2 Latvia 12 4.6 4.5 -0.4 Estonia 13 3.2 3.4 -0.2 Poland 9.5 3.8 3.0 0.2 Sources: Calculations using the equation from Benhabib and Spiegel (1994), which is: [.0007*Schooling*Gap + .0014*Gap]. Gap is the ratio of the US per capita GDP to that of the country concerned in 2000 at current international prices from the World Bank Development Indicators. Labor force growth is average annual growth rate over 1999-2010 calculated from World Bank Development Indicators. Data for Poland is from Doyle et al. (2001). These projections indicate that Latvia has the potential for rapid growth but convergence with EU income levels will probably take time to materialize. An important caveat to bear in mind, however, is that the projections are based on an assumption of a constant capital-output ratio. The realism of this assumption is questionable given uncertainties about the viability of capital stock inherited by Latvia from the former Soviet Union. The pace of convergence with Western European countries would clearly be influenced by any substantial change in the capital-output ratio. This has major policy implications for the medium term, as the Latvian authorities will need to ensure adequate investment ratios, stimulate domestic and foreign savings to fund the necessary investment, and implement measures to create the legal and business environment that will be most conducive to TFP-intensive growth. 3.2 Why Labor Market Issues are Key to Latvia's Convergence with the EU The potential benefits for Latvia when it adopts the euro are well known: the single currency will reduce transaction costs with agents in EMU countries, and hence raise the static efficiency of the economy. It will eliminate exchange rate risks as well as the cost of hedging and currency exchange. The resulting improvement in transparency will increase the intensity of competition in goods and factor markets, thus raising the dynamic efficiency of the economy. This in turn should foster innovation, trade, investment, and growth, thereby improving employment prospects. The downside will be the heightened labor market risks for Latvia. In theory, the exchange rate constitutes a useful tool for macroeconomic adjustment in the case of asymmetric shocks. Exogenous shocks that hit countries asymmetrically require a response of real exchange rates between the countries concerned in order to adjust relative wages between them. This is usually done through adjustment of nominal exchange rates or adjustment of nominal goods and factor prices in one country relative to another--so that higher unemployment is mitigated. This is not an option in Latvia where monetary policy is similar to a currency board arrangement. Nor will it be an option when Latvia formally gives up any 14 national monetary policy to adopt the euro. Therefore, a rise in unemployment in Latvia resulting from the loss of exchange rate flexibility can only be avoided if there is little probability that the country is hit by asymmetric shocks. Whether or not Latvians will be able to respond to local labor market shocks via migration to Western Europe will be determined by EU policies on immigration. Latvia's historic patterns of susceptibility to shocks may not persist when the country joins the euro zone. Indeed, the future probability of asymmetric shocks hitting Latvia will depend upon its underlying economic structure. The critical question will then be how participation in the euro system will affect this structure and what impact it will have on the synchronization of business cycles. There are two schools of thought on this issue. Some argue that an increasing synchronization of business cycles within the EMU is to be expected because trade linkages between European regions are likely to increase further over time. Thus, any shock in a given country is likely to be passed on quickly to all countries through trade linkages. It is even argued that tighter international trade linkages between EMU countries will make their economic structures and business cycles more similar, and asymmetric shocks less likely. On the other side, some researchers argue that the EMU will create opportunities for scale economies and therefore, a more specialized group of countries and regions (with different production structures). This would aggravate the likelihood of asymmetric shocks.10 Regardless of what happens to the Latvian economy, clearly labor market flexibility (aggregate and relative wage flexibility, working time flexibility, spatial and job mobility) will be crucial for absorbing shocks. Latvia has made progress in recent years in improving aggregate wage flexibility. However, it is not enough to restore and maintain labor market equilibrium in an era of increased openness and competition. Aggregate wage flexibility must be complemented by a high degree of relative wage flexibility, regionally and sectorally. Preliminary evidence seems to suggest that Latvia's relative wage flexibility is still too low. In some rural areas, a relatively high level of unemployment compensation may be acting as a disincentive to work. Moreover, working time flexibility (which may compensate for insufficient wage flexibility and help raise productivity) is still too low in Latvia. Low spatial and job mobility also seems to explain the persistence of high unemployment in Latvia. Although several important institutional changes have been made since 1995, Latvia has not consistently implemented the comprehensive mix of reforms that would enhance flexibility. As long as these rigidities impede the transition process, shocks are likely to threaten the country's remarkable macroeconomic performance and aggravate its structural unemployment. 4. Macroeconomic Options with EU Membership While EU membership will give Latvia new economic opportunities, it will also entail new, tougher macroeconomic challenges requiring deeper fiscal and monetary policy adjustments. On the fiscal front, given the prevailing framework in the EU and the jurisprudence of the Stability and Growth Pact (SGP), the main issue for Latvian authorities will be to decide between setting rules versus using discretion. This challenge will be all the more important as Latvia faces the need to substantially increase public spending as part of EU and NATO 10 For the two sides of the argument, see Frankel and Rose (1998) and Krugman (1993). 15 requirements.11 In this chapter, the design of fiscal policy will be discussed from the perspective of growth and job creation. The question specifically is whether a more appropriate fiscal framework for public spending is desirable from that perspective. Lessons from theory and history show that the design of fiscal institutions can have potentially large effects, for good or ill, on economic performance.12 This is particularly true in the case of Latvia where, because of the fixed exchange rate regime, fiscal policy is the main macroeconomic tool at the government's disposal to limit the current account deficit. Latvia's already high tax burden severely constrains rate increases to improve government revenue. In fact, like all Baltic countries, Latvia has been trying to reduce direct taxes to attract more foreign investment, while at the same time harmonizing its tax system with EU requirements and norms. In this chapter, we argue that, in order to sustain economic growth and maintain a business environment favorable to job creation, Latvia should maintain its medium-term objective of a balanced budget and place more emphasis on fiscal policy rules (especially spending rules) within the boundaries set by the Stability and Growth Pact. Latvia's current fiscal framework is geared toward achieving a balanced budget in the medium term. En route, annual deficits are supposed to follow a declining path as articulated in the country's 2002-06 National Development Program. However, fiscal discipline eroded in 2002; the government deficit reached 2.7 percent of GDP (1 point over the target agreed with the IMF), and the 2003 budget initially projected a deficit of 3 percent of GDP (the actual outcome was about 2 percent). On the monetary front, things seem to be under control. From a purely legal and institutional viewpoint, Latvia does not have to make any major change in its exchange rate regime prior to EU membership. However, after EU accession and as the prospects of adopting the euro increase, monetary and exchange rate policies in Latvia are expected to face new pressures for mutation. Like all acceding countries, Latvia will be expected to participate in the exchange rate mechanism (ERM II) arrangement that will limit fluctuations of non-EMU (European Economic and Monetary Union) EU countries' national currencies relative to the euro. The Latvian authorities have already indicated that they intend to join ERM II in 2004-2005 and to repeg the Lats to the euro at the same time. But while EU accession itself does not imply any major changes to existing monetary policy, Latvian authorities should take into account the fact that their exchange rate policy will become a "matter of common concern," starting on May 2004. The role that monetary policy could play in reducing unemployment in Latvia appears to be relatively modest, given the central bank's emphasis on exchange rate stability. The current stable exchange rate imposes a constraint on domestic monetary policy (the so-called nominal anchor), which is perceived by the private sector and foreign investors as a useful safeguard against unsound policies. This makes it even more desirable to implement labor market reforms that will reduce wage pressure and further reduce inflation. 11According to an IMF study, acquis-related expenditure will be sizable, as the annual costs of meeting environmental requirements alone could reach 2 percent of GDP. As a NATO member, Latvia is expected to raise its military expenditure to 2 percent of GDP over the medium term. "While transfers from the EU to finance such spending could exceed 1 percent of GDP during the pre-accession phase and reach up to 4 percent of GDP after accession, this would still imply a significant contribution from national budgets." (Mueller et al. 2002:4). 12Masson (2000: 112). See also Tanzi and Zee (1997). 16 This section examines Latvia's fiscal performance over the first decade of independence and discusses whether (and if so, how) fiscal policy could be used to stimulate growth and employment. It also discusses what role, if any, monetary policy could have in the medium term--especially during ERM II--in the quest for growth, employment, and convergence. Its key conclusion is that Latvia should resist the temptation to stimulate growth and job creation through a shift toward expansionary fiscal and monetary policies. Such "active" policies would endanger macroeconomic stability, growth, and medium- and long-term job prospects. 4.1 A Fiscal Strategy for the Medium Term Fiscal policy discussions typically focus on three questions: (i) whether a well-designed tax system and good spending programs can foster long-term growth and employment, (ii) whether expansionary fiscal policy can play a stimulating role in the economy, and (iii) how high fiscal deficits and government debt can become a source of inflation and balance-of- payments problems.13 This section reviews Latvia's fiscal policy mostly from the perspective of questions (ii) and (iii). We start with an analysis of fiscal developments over the first decade of independence and their relationship to the business cycle. We then suggest a fiscal framework for strengthening credibility after EU accession. 4.1.1 Fiscal Developments in the Last Decade Latvia has made substantial progress over recent years in adopting a fiscal stance consistent with its strategic goal of EU membership. During the first years of transition, public finance performance was uneven, with the fiscal Figure 5: Government Revenue and Expenditure (as percent of deficit nearing 4 percent of GDP in GDP) 1994-1995. Latvia's fiscal discipline has 5 0 generally been good since then, 4 5 contributing to the country's 4 0 3 5 macroeconomic stability. Since 1996, 3 0 the fiscal deficit has been below the R e v e n u e 2 5 E x p e n d i t u r e Maastricht Treaty ceiling of 3 percent of 2 0 GDP, except for during 1999 when it 1 5 1 0 reflected the consequences of the 5 Russian crisis. 0 Fiscal adjustment was based especially on expenditure restraint, which is historically more durable,14 especially when reductions occur in primary current expenditures as has been the case in Latvia during the second half of the past decade. Figure 2 indicates that the adjustment effort on transfers to households has been modest, while wages and salaries have been kept under tight control throughout the decade. Overall, the ratio of expenditures to GDP has been reduced by 2 points in the last 6 years (from 40.7 percent in 1997 to 38.8 in 2003). 13See Hemming et al. (2002). 14See Alesina and Perotti (1997). See also Mueller et al. (2002) for an analysis of recent fiscal performance in the Baltic countries. 17 Figure 6: Composition of Expenditure, 1992-2002 100% 90% 80% 70% 60% Capital expenditure Other current expenditure 50% Interest Transferts to households 40% Wages and salaries 30% 20% 10% 0% 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 This trend has recently begun to be reversed, with the 2003 budget envisaging a government deficit of nearly 3 percent of GDP. And while public debt remain moderate by international standards (14.6 percent of GDP at the end of 2002), its level has more than doubled between 1998 and 2003, making Latvia the most indebted country among the three Baltic states. These are sources of concern to Latvian policymakers, especially as the rapid aging of the population will increase spending on pensions and health care while reducing the labor force and therefore the tax base. Reducing public spending and public debt is politically difficult as it always implies making some social groups worse off. The process can be particularly difficult in a politically fragmented country like Latvia, where major elections have brought to power an average of almost one new government per year since independence. To circumvent these difficulties, several countries have attempted to limit discretionary fiscal policy, which is considered ineffective in stimulating economic growth and employment in a sustainable manner without creating inflation. They have done so by adopting various types of formal fiscal rules, from full-fledged balanced-budget rules to multiyear budget frameworks.15 On the one hand, a clear and binding commitment to a medium-term objective can help policymakers withstand pressure for increasing spending or delaying fiscal adjustment. On the other hand, budget rules that are too rigid may constrain the fiscal authorities' ability to run countercyclical fiscal policy. Moreover, even in countries where fiscal rules exist in the book, there is always some scope for creative accounting, which influences the effectiveness of these rules. 15Some authors have suggested setting up an independent institution (central bank type) to manage fiscal policy. See the review of the literature by Hemming et al. (2002). 18 4.1.2 Fiscal Policy and Unemployment: How Latvia Dealt with the Business Cycle As a tool to stabilize aggregate demand shocks, fiscal policy has an important role in a country with an exchange rate peg regime like Latvia. This role will become even more important after EU accession, as monetary policy within the euro area is designed to address developments in the entire union. While rigid fiscal rules may cause fiscal policy to be procyclical, Latvian authorities could design rules that allow for some cyclical stabilization and reduce demand shocks that create frictional unemployment. Figure 7: Fiscal Balance and GDP Growth (as percent of GDP) 10 8 6 4 2 0 -2 -4 -6 -8 -10 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003* Fiscal Balance Real GDP A quantitative analysis of the use of fiscal policy to manage the business cycle would normally rely on comparing indicators of the cyclical situation (typically the so-called output gap) and indicators of the fiscal policy stance (usually the change in the overall fiscal balance and the change in the structural primary balance, which is often used as a proxy for the discretionary elements of fiscal policy (the "fiscal impulse").16 Thus, a positive impulse reflects an improvement in the structural balance that is a discretionary fiscal contraction. If fiscal policy is procyclical, periods of negative output gap should be accompanied by fiscal contraction. Likewise, if fiscal policy is a countercyclical, a negative output gap should be accompanied by a deteriorating fiscal balance reflecting either discretionary measures by the government or the operation of automatic stabilizers. A rigorous estimation of Latvia's structural budget balance would involve three steps: estimating the output gap derived from the country's potential output; measuring the cyclical component of public revenues and expenditures; and deducting cyclical revenues and expenditures from their actual levels.17 While it is difficult in the current circumstances to systematically carry out such an analysis (because of the rapid transformation of the economy 16Different methodologies can lead to different estimates of the fiscal impulse. For an analysis of these issues in the context of a sample of EU countries, see IMF (2001). 17See Hagemann (1999) for a detailed discussion of these three basic steps. 19 over the past decade), it appears from basic empirical observation that fiscal policy in Latvia has rarely been countercyclical over the past decade. There are at least two reasons why the fiscal impulse in Latvia often has been neutral: (i) in the very early years of independence, monetary policy was available to smooth the business cycle, especially when capital controls were still in place (1991-94); and (ii) in Latvia (like in most transition economies), output gaps are very difficult to gauge and policymakers did not attempt to fine-tune fiscal policy, which affects the economy with delay anyway. But other factors explain why fiscal policy in Latvia was not only neutral but even procyclical. First, fiscal policy tended to be expansionary during upturns because policymakers were overly optimistic and failed to recognize the temporary nature of the fiscal improvements during upswings. Also, given the rather fragmented political market, it has always been difficult for newly elected government officials to withstand pressures from spending ministries and interest groups when resources are available. Second, under the stabilization programs adopted by the authorities in the mid-1990s and more systematically in the framework of Stand-By Arrangements concluded with the IMF, the authorities adopted a tighter fiscal stance. As a result, the deficit was considered economically and politically unsustainable-- especially in the late phases of downturns-- thus forcing a discretionary contraction. Figure 8: Composition of Government Revenue, 1992-2002, as percent of GDP 100% 80% Other 60% Taxes on goods and services Income and 40% social tax Corporate tax 20% 0% 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 4.1.3 Medium-Term Outlook and Expenditure Strategy Under the Maastricht Treaty and the Stability and Growth Pact, EU countries must keep the general government deficit within 3 percent of GDP except for temporary reasons, and the gross general government debt must be below 60 percent of GDP. Like all accession countries, Latvia has met these requirements. But EU membership will imply the obligation to comply with EU regulations and resolutions that have strengthened the Maastricht Treaty, requiring union members to maintain a fiscal position "close to balance or in surplus" in the medium term. Thanks to its prudent fiscal policy throughout most of the past decade, Latvia still enjoys a low level of public debt (14.6 percent of GDP as of end-2002). Its large current account deficit has so far been financed mostly with foreign direct investment. These two factors contribute to limit the usefulness of theoretical fiscal sustainability exercises (see Mueller et 20 al. 2002). Moreover, fiscal sustainability does not simply entail the government's ability to finance itself. It also requires fiscal and monetary policies to be consistent with the expected growth, inflation and interest rates. Sustainability does not necessarily require the government to be able to pay off its debt in the long-run. It implies that real debt is increased only at a rate less than the real interest rate paid on it. In other words, the government is accountable for the net real interest rate (real interest rate, r, minus the real growth rate, µ) paid on the debt to GDP ratio, b0 . This can be financed either with a primary surplus g ­ , or with seigniorage revenue, which is represented by the inflation tax paid on the money demand to GDP ratio, L(r+) which is a decreasing function of nominal interest rate, (r+). This sustainability condition is represented as: (g ­ ) + (+µ) · L(r + ) = ( r ­ µ ) · b0 Is Latvia's primary fiscal balance in line with this long-run sustainability condition? Assuming a long-run growth rate of 4 percent of GDP and expected average inflation rate of 3 percent, it would appear that Latvia can sustain a long-run deficit of approximately 0.5 percent of GDP. Table 3 below shows the level of sustainable fiscal balance under different growth and interest rate assumptions. Table 3: Sustainable primary fiscal balance18 Interest Rate Real growth rate 4 5 6 7 8 3 -0.43 -0.27 -0.11 0.04 0.20 4 -0.67 -0.51 -0.35 -0.19 -0.03 4.5 -0.79 -0.63 -0.47 -0.31 -0.15 5 -0.91 -0.75 -0.59 -0.43 -0.27 As a member of the EU, Latvia now has to set its fiscal policy according to its national framework but within the boundaries of the Maastricht Treaty and the SGP. Every year, it has to present a 4-year fiscal plan to the ECOFIN Council, which will issue an opinion on the plan's consistency with EU principles of sound public finance. Usually, 4-year plans presented by EU countries contain only indicative fiscal targets and do not lead to sanctions, even if they deviate from the "close to balance or in surplus" path. Thus, different countries have chosen to pursue different routes in designing their medium-term fiscal framework. Fiscal policy in Latvia has generally been conservative from the outset of the transition process with a few notable exceptions caused by internal or external shocks. Due to generally good fiscal discipline, Latvia has low levels of public debt (though the World Bank has recently reclassified Latvia from "less indebted" to "moderately indebted"), and its credit 18It should be noted that the sustainable long-run deficit calculated in Table 3 is an approximation based on the following assumptions: We use the 2002 figure of 14.6 percent debt to GDP ratio and a GDP growth rate of 4 percent; net interest payments are calculated using the long-run real interest rate of the 20-year average on the US Treasury Bond rate (5.5percent). While the net interest payments on debt account for most of the sustainability burden on the government budget, seigniorage revenue does not contribute greatly to alleviating interest payments on debt. The inflation tax is the inflation rate (set at a long-run value of 3 percent) times the money demand to GDP ratio. Since money demand is assumed to be a decreasing function of the nominal interest rate, L(·) is approximated accordingly. Therefore, the impact of an increase in interest rates is twofold. It increases debt payments and decreases seigniorage revenue. The money demand function is approximated using the following form: A·e- (r+) where A=0.109; =1.99 has been estimated using available data. 21 ratings have improved, giving Latvia easier and less expensive access to international capital markets. Table 4: Tax Burden and Effective Tax Rates (2002 or latest year available) Effective Tax Rate (in percent of tax base) Tax receipts Non-wage Personal income Capital Consumption (percent of GDP) labor costs Latvia n.a 25 percent n.a 18 percent VAT; excise 29.6 tax Germany 31.8 17.8 15.9 17.9 43.1 France 32.1 15.2 22.6 24.5 46.4 United States 11.6 13.9 22.7 9.3 29.9 For the medium term, Latvia is committed to further fiscal consolidation accompanied by some relief in the tax burden, which is still high relative to several western countries (Table 4). The main question at this stage is what strategy and what specific policy instruments the authorities should adopt to reach their goal of a balanced budget. Lessons from the previous wave of EU accession countries of the 1970s and 1980s (notably Greece, Figure 9: The Fiscal Cost of EU-Accession (as percent of GDP) Ireland, Spain, and Portugal) suggest that Latvia will face some major fiscal 6 challenges in the medium term as it 5 reconciles several conflicting policy choices: Specifically, the country will 4 have to accommodate additional public Grants from EU 3 expenditure (acquis-related spending EU-related expenditure and an increase in its military budget as 2 part of NATO membership) while complying with tax harmonization 1 requirements, pursuing its lower direct 0 tax rates policy, and making efficient 2002 2003 2004 2005 2006 use of transfers from the EU that will generate pressure on the size of government. In order to carry out these new responsibilities without diverging from the path of a balanced budget, further rationalization of public expenditure and reduction of non-priority spending will be required. Projections of the likely financial flows to and from the EU budget carried out by the Mueller et al. (2002) indicate that under the baseline scenario, EU-related expenditure could average 5.4 percent of GDP in 2004-06 (see Figure 9). Latvia would need a reduction in discretionary spending by about 2 percentage points in the medium term, which could be effected by "moderate wage increases, coupled with civil service reforms, improved targeting of social benefits, and effective cuts of expenditure on goods and services." Savings will need to be generated in other parts of the government to increase funding for priority programs. More fundamentally, the key challenges for the government are first to develop and use a (proper) methodology to make appropriate choices among policy priorities and, second, to allocate its resources to those priority areas in a manner conducive to sustainable, cost-effective results. As an integral part of Latvia's fiscal policy for the medium term, expenditure policies could make an important contribution to growth and convergence. A recent World Bank report on 22 Central and Eastern European countries outlines the main elements of a strategy that would benefit Latvia:19 (i) more public resources should be left in the hands of the private sector where they help foster factor productivity growth; (ii) subsidies and transfers should be curtailed--Latvia still spends more than twice as much relative to GDP (4 percent) as the average EU country on enterprise subsidies (1.5 percent); and (iii) some social transfers, especially in the education sector (see chapter 11), should also be reduced to enhance quality of service and efficiency in the use of public resources. 4.1.4 The Need to Rethink Fiscal Rules in Latvia In spite of Latvia's overall good track record in managing public finances, many challenges remain. During pre-electoral contexts, politicians tended to discount the future more heavily than private sector agents. In 2001-02, differences of opinions among ruling coalition partners often generated disagreements over the budget process and slowed down fiscal consolidation. Also, the very nature of Latvia's fragmented political system and the collective form of decisionmaking on budget matters has tended to create the "common pool problem," that is, high spending or policy distortions reflecting the aggregation of uncoordinated decisions made at the level of each ministerial department (as parties in the ruling coalition sometimes ask for expenditures programs to satisfy their constituencies without taking into account their implications for the budget). In spite of Latvia's overall good track record in managing public finances, many challenges remain. During pre-electoral contexts, politicians tended to discount the future more heavily than private sector agents. In 2001-02, differences of opinions among ruling coalition partners often generated disagreements over the budget process and slowed down fiscal consolidation. Also, the very nature of Latvia's fragmented political system and the collective form of decisionmaking on budget matters has tended to create the "common pool problem," that is, high spending or policy distortions reflecting the aggregation of uncoordinated decisions made at the level of each ministerial department (as parties in the ruling coalition sometimes ask for expenditures programs to satisfy their constituencies without taking into account their implications for the budget). As noted in the 2001 Public Expenditure Review, the lack of policy priorities has often led to an increase in unfunded mandates and ineffectiveness in the budget formulation process. Among the circumstances that tend to lead to over-commitments and unpredictability of resources, thus limiting incentives and program performance are the absence of a forum where alternative policy proposals compete, poor policy formulation in some sectors, a budget process based on incremental budgeting, and ad hoc commitments to unfunded policy proposals during budget preparation or execution. Placing policies at the center of budget decisions is further hampered by structural rigidities in the budget (limited fungibility), notably earmarking in special budgets (about 42 percent of Latvia's revenues are earmarked), the existence of own revenue, and the proliferation of autonomous agencies (208 agencies in 2000). Similarly, fairly rigid rules on financing and allocations limit the scope for true policy decisions. 19See Funck (2002). For more specific policy recommendations on expenditure management, see World Bank 2002. 23 The Latvian authorities have long recognized the need to limit discretion in fiscal policy. The government decided in 2000 to move toward a Medium-Term Expenditure Framework (MTEF). A reform program was adopted that includes several important elements of an MTEF, such as a medium-term fiscal policy statement, a budget priorities statement, and the development of guidelines for line ministries to design their own strategic plans. While these were important steps, Latvia would benefit from additional fiscal rules that could constrain fiscal policy. Indeed, beyond the substantial public finance challenges ahead, the level of public expenditure is still high, and the implications of the aging population for future social spending have yet to be fully quantified. The rules needed would obviously operate within the broader constraints of the Stability and Growth Pact. In Latvia's case, the key issues to be addressed after EU accession will be the choice between expenditure and deficit rules, the implementation mechanisms (enforcement), and the relationship between the central government and the various levels of local government.20 4.1.5 Strengthening the MTEF with New Rules and Targets While the current plan for an MTEF is appropriate for Latvia, the country would benefit from strengthening public finance management with more stringent rules and targets. Recent difficulties in containing distortions in the budget process point to the need to address the remaining deficit bias either through some type of deficit rule or through an expenditure ceiling that would constrain spending ministries, agencies, and local governments. Latvia could anchor its medium-term fiscal policy on a numerical rule relating to the central government fiscal balance, the level of aggregate public expenditure, or the public debt. The authorities have already expressed their goal to balance the budget over the medium term--a rule as simple as the SGP rule on a maximum budget deficit of 3 percent of GDP. Such a decision has the advantage of being straightforward and easy for the private sector and general public to understand and monitor. But recent experience shows that it does not adequately address the bias toward excessive spending, as the government can raise taxes to finance higher expenditures. Furthermore, in Latvia's case, fiscal policy would remain procyclical: an adverse shock reducing public revenues and raising expenditure would inevitably result in discretionary tightening to maintain the budget balance. In principle, Latvia could address some of the difficulties of its medium-term balance rule by targeting the budget structural balance.21 This supposes having some good quantitative indicators and parameters to measure business cycles and their implications on government revenues and expenditures. But given the enormous methodological problems of such an approach in the Latvian context, the scope for uncertainty over the numbers could be large. Even if an independent and credible institution is in charge of estimating the cyclically adjusted deficit, enforcement and monitoring problems will remain. Indeed, it would be 20See the World Bank 2003 report on Latvia's territorial reform. 21Following the IMF (WEO), the structural budget position is defined as the actual budget deficit (or surplus) less the effects of cyclical deviations of output from potential output. Because of the margin of uncertainty that attaches to estimates of cyclical gaps and to tax and expenditure elasticities with respect to national income, indicators of structural budget positions can only be interpreted as broad orders of magnitude. Moreover, it is important to note that changes in structural budget balances are not necessarily attributable to policy changes but may reflect the built-in momentum of existing expenditure programs. 24 difficult to explain to the private sector how the rule is actually measured, and how much confidence can be placed in the methodology in case of changing economic circumstances. A good model of fiscal rule for Latvia could be to set upper bounds on expenditure increase. This framework would complement the current MTEF and would directly address the issue of policy distortions that currently lead to excessive spending. Furthermore, it would not automatically lead to a procyclical fiscal stance, because stabilizers on the revenue side of the budget would remain free to operate. In fact, there could be a legal requirement in the budget bill requiring that the fiscal stabilizers be allowed to operate (meaning that shortfalls or surpluses in realistic revenue projections be used to increase or reduce the balance). This rule would also limit the tendency of spending ministries to increase their programs during upturns, and it would be easy to explain to the private sector if it is simply designed. It has been successfully adopted over recent years by several countries like Sweden, Finland, and the Netherlands. An important shortcoming of this type of expenditure rule for Latvia is that it would not necessarily prevent a bias toward excessive fiscal deficit through a large tax cut or the systematic over-optimistic projection of government revenues. However, the country's recent budget management experience suggests that the likelihood of this happening is rather small. In addition, within the SGP, Latvia already has another safeguard against excessive deficit: its medium-term zero budget balance target, which would complement an expenditure rule, in spite of the fact that the medium-term target is a weaker binding commitment.22 Furthermore, empirical studies indicate that fiscal consolidation based on expenditure reduction is usually more sustainable. 4.1.6 Practical Issues in Designing Expenditures Rules for Latvia · Choosing a Set of Macroeconomic Projections. Latvia should improve its MTEF process to make it a tool to tighten government expenditure programs, assist decisionmakers to gauge what is affordable in aggregate over the medium term, and to reconcile this with spending policies and their costs over the same period. In this context, an important decision is to adopt a more realistic scenario for future macroeconomic developments. The so-called cautious scenario adopted by the authorities for the 2002 budget delivered some "favorable" surprises ex post as cyclically sensitive spending turned out lower than initially projected, which allowed the government to deal with overruns in non-cyclical expenditures categories, thus facilitating negotiations within the coalition cabinet during an election year. It also helped address unexpected spending pressures without deteriorating the budget deficit. However, it also obfuscated the true goal of Latvia's MTEF: making the spending targets explicit for economic agents. A realistic budget scenario could be complemented with a small contingency fund to be used as a margin for discretionary spending. It is not in Latvia's interest to present to the Saeima a 22Another possibility would be to adopt an explicit target for the debt-to-GDP ratio, seen as a more appropriate benchmark for changes in fiscal policy than the budget balance. Switzerland has adopted an interesting variant of this method whereby revenue forecast errors are cumulated into a "notional" debt stock that must be reduced to zero over time. But given Latvia's political history and its tradition of one new government on average since independence, such an approach may not be appropriate. 25 budget bill that does not reflect the government's most accurate projections for revenue and expenditure. Such a practice may encourage constant second-guessing by spending ministries, and build momentum and pressure for a supplementary budget throughout the fiscal year, especially in a politically fragmented environment like Latvia. Also, there is often the temptation to use additional spending room from an overly cautious budget scenario to engage in procyclical spending when economic conditions are favorable. A limited contingency fund would provide both transparency and the opportunity for adjusting spending during the fiscal year. · Taking Expected Inflation into Account. A second important step in designing the expenditure rules for the medium term is to decide whether to use real or nominal spending ceilings. Although Latvia's very prudent monetary policy has brought inflation under control since the decision to peg the lat to the Special Drawing Right (SDR) in 1994, the country's budget rules should focus on the evolution of real spending instead of nominal spending. While a nominal rule makes the framework easier to design (as changes in the inflation outlook would not impose changes in spending ceilings), a nominal target would imply lower real public expenditures if for some reason inflation becomes a matter of concern. Therefore, it would be better for Latvia to opt for real spending rules. Using the general consumer price index (rather than sector-specific deflators, which are much more difficult to forecast accurately), the spending ceiling could be converted every year into a nominal ceiling. If inflation arises and is the result of buoyant domestic demand, cyclical stabilization could be made through reduction of real spending. And if the country is hit by a permanent adverse supply shock, setting a lower lever of real public spending would help absorb the shock. To ensure that the ceilings on the most important entitlement programs are easily set and respected, important spending categories like pensions could be indexed to expected inflation. · Choosing the Right Spending Target. In deciding which expenditure target to choose for its strengthened fiscal framework, Latvia should opt for an aggregate category (i.e., total government expenditure as a percentage of GDP) that has macroeconomic significance and is easy to monitor over the years. Indeed, the temptation is always there to choose a narrow expenditure category. But experience from many countries shows that a narrow target is almost always circumvented with the introduction of new spending programs. There is, however, one important caveat to the option of choosing an aggregate spending target: given Latvia's history of containing expenditure growth by limiting capital expenditures23, it would be reasonable to exclude them from the target and focus on current government expenditures. And to ensure that the distinction between current and capital expenditures is transparent, the criteria for the classification should be clearly laid out. Also, because interest payments are not under the direct control of the government during the budget process, they may be excluded from the target. To improve the ability of the new fiscal rule to adequately deal with fluctuations of the business cycle, expenditures that are very cyclically sensitive, like unemployment compensation, should be excluded. Unemployment spending is a rather small component of 23In 2002, for instance, the IMF reported that the Latvian authorities could only meet one of the end- December performance criteria on fiscal policy by delaying some public expenditures. 26 government spending in Latvia (unemployment benefits have averaged 0.5 percent of GDP per year over the period 1998-2001, compared to an average of 1.5-2.4 percent of GDP in countries like Germany, France, or Spain)24. Therefore, a case could be made to include them in the spending target. However, recent history has shown that in case of recession the additional burden on the government budget may be substantial. For instance, the fiscal outcome for 1999 was affected adversely by the impact of slower economic growth on revenue, as well as the higher pension and unemployment benefits and wage increases granted in late 1998. As a result, the consolidated general government deficit reached 3.9 percent of GDP in 1999. To prevent such cyclical factors from negatively affecting the monitoring of the fiscal rule, it would be preferable to exclude unemployment benefits from the target. By contrast, the rule should cover not only discretionary expenditures but also entitlements-- especially the growing cost of pension. In principle, including the burden of pension in the spending target may complicate enforcement of the new fiscal rule and make it easy for some policymakers to develop creative accounting (i.e., disguising other expenditures as entitlements). Latvia should nevertheless consider that option because the ongoing pension reform program will eventually alleviate the burden of entitlements if implemented as scheduled. In the late 1990s, the Latvian authorities adopted two measures that severely affected the financial sustainability of the pension system: (i) a one-off pension increase permanently raised all pensions by approximately 10 percent and (ii) a regressive and unfair pension recalculation more than doubled pensions for a select group of pensioners, those who continued to work. Fortunately, beginning in 1999-2000, the Saeima adopted legislation to reform the pension system, starting with the establishment of a second tier. In 2001, further pension reform resulted in the creation of a three-tier pension system that is one of the most advanced among the transition countries. It includes a privately funded and managed pillar. Mechanisms for better targeting of social assistance are being implemented. The government's reform program is geared toward completing the pension reform agenda by 2005 and achieving an equitable and financially sound system that is less likely to contribute to macro-imbalances when the economy is affected by an adverse shock. Hence Latvian authorities need to review their medium-term fiscal strategy and accelerate the pace of the medium-term budget planning framework, which is critical to the necessary prioritization of spending. 4.2 Monetary Policy, Growth, and Employment At an early stage of the transition process, Latvia introduced its own currency, thus enabling the Bank of Latvia to pursue an independent monetary policy that aims to maintain price and exchange rate stability and control the amount of banks' reserves so as to limit excessive lending. The central bank's exchange rate policy is similar to that of a currency board, and the monetary base is backed by gold and foreign currency reserves. Latvia has been using the exchange-rate-based stabilization program since 1993. In mid-February 1994, the Bank of 24Unemployment insurance was introduced in 1997. Benefits can be relatively high during the first 3 months of unemployment, after which both the rate of income replacement and the benefit ceiling are substantially reduced (the maximum benefit period is 9 months). 27 Latvia pegged the lat to the SDR basket of currencies at the rate of 1 SDR equals 0.7997 LVL. Latvia was not required to make any important change in its exchange rate regime prior to EU membership. However, after EU accession and as the prospects of adopting the euro increase, monetary and exchange rate policies in Latvia are expected to face new pressures for mutation. Like all acceding countries, Latvia will be expected to participate in the exchange rate mechanism (ERM II) arrangement, which will limit fluctuations of non-EMU EU countries' national currencies relative to the euro. The Latvian authorities have already indicated that they intend to join ERM II in by 2005 and repeg the lat to the euro at the same time. But while EU accession itself does not imply any major changes to the existing monetary policy, Latvian authorities should take into account the fact that their exchange rate policy will become a "matter of common concern" starting in May 2004. Latvia is well positioned to take the "fast-track" approach, that is, to join ERM II as soon as possible and stay at this stage only for the minimum required period before adopting the euro. Then, the country will have to adopt the European Committee for Banking Standards operational framework. While the authorities have made clear that they do not intend to radically change their existing fixed exchange rate policy, there has been some public debate as to what alternative monetary policy options Latvia (and other EU acceding countries) could consider before adopting the euro in the medium term.25 4.2.1 Benefits of Price Stability for Growth and Job Creation Stability as the Priority. The main task of the Bank of Latvia (BOL) currently is to maintain price stability, thus contributing to the process of restructuring the national economy. This task will become increasingly difficult as Latvia (a small country with an open economy) undergoes extensive institutional and structural changes to adjust to all aspects of the acquis communautaire. As Latvia organizes its transition from its current exchange rate regime (almost identical to that of a currency board) into ERM II and the adoption of the euro, there is, in principle, the possibility of relaxing monetary policy, that is, abandoning the peg and allowing the lat to fluctuate within a wider margin compared with the euro. Some researchers have suggested that such an approach to monetary policy could help foster growth and job creation. The Latvian authorities have indicated that the country's monetary policy will remain the same after EU accession. Indeed, in order to sustain growth and employment in the medium and long run, the core task of the Bank of Latvia should remain to safeguard the purchasing power of the lats, that is, to maintain monetary stability in Latvia prior to the adoption of the euro. That goal will require that the lats has a stable internal value (price stability) and, given the growing importance of trade relations with EU countries, that there are favorable conditions for a stable external value (exchange rate stability). The Eurosystem's definition of price stability as an annual rate of increase in consumer prices of less than 2 percent to be maintained over the medium term has been criticized for being either too specific and too stringent or too vague and less effective as an anchor for inflation 25See, for instance, the proceedings of the seminar on "Monetary Policy Strategies for Accession Countries" organized in Budapest, Hungary, February 28. 28 expectations than a point inflation target26. The Bank of Latvia has no specific quantitative ceiling for price stability: its goal is to achieve low and stable inflation in the medium term, focusing on a CPI that excludes administered, energy, and unprocessed food prices. Clearly, there may be benefits for Latvia from adopting a point target for medium-term inflation. But these should be weighed against the potential credibility cost of switching to a new monetary policy objective. Furthermore, since Latvia will not adopt the euro in the immediate future, there is not strong rationale for the BOL setting a quantitative inflation target similar to the European Central Bank (ECB) just after becoming the member of EU. Moreover, internal stability of the lat was achieved over the past decade: successful financial policies have been reflected in the continuous fall of inflation from 958.7 percent in 1992 to 23.1 percent in 1995, 7.0 percent in 1997, and 2.9 percent in 2003. Over the past decade, Latvia has also benefited from the peg to the special drawing rights (SDR)--a monetary policy aiming for a stable external value of the lat against the euro and the dollar as the key to achieving price stability27. In a small open economy like Latvia, the existence of close trade relations with EU countries was indeed a further justification for such an indirect approach to fighting inflation: the stability of the lat helped remove artificial barriers to international competition and thus promoted trade with important partners. While the strategy was clearly adequate, success was also due the good policy mix, especially a prudent fiscal policy, labor market reforms responsible for moderate wage increases, and other structural reforms to create efficient markets. Potential Economic and Social Costs of Inflation. In principle, ERM II provides the possibility of more flexibility for monetary policy, as it involves stable but adjustable central rates to the euro for participating currencies, with fluctuation bands of plus or minus 15 percent around the central rate. Because its fixed exchange rate regime and the peg to the SDR was determined to be incompatible with ERM II, Latvia will be required to either repeg the lat to the euro, or adopt a managed float within ERM II bands at a central rate agreed with the ECB28. In theory, this second option would open the possibility of a more active monetary policy, with potential short-term gains for economic growth and job creation. But Latvia should resist the temptation of using monetary policy as a tool for short-term stimulus of the business cycle, and stick to its current goal and strategy. The shift to a more active monetary policy in the interim period prior to the adoption of the euro could create an inflationary environment with heavy economic and social costs. First, even in the case of anticipated inflation (when the public fully expects price increases), the burden on businesses would be heavy. Indeed, after enjoying a rather long period of stability with the exchange rate peg, businesses (and the public at large) would have to anticipate the effects of increased prices in their decisions. The efficiency losses stemming from attempts to constantly revise price lists or economize on non-interest-bearing money holdings or from misallocating resources due to adjusting to a non-indexed tax system would be large and detrimental to growth and job creation. 26See Begg et al. (2002) for the first type of criticism, and Svensson (2002) for the second type. 27The BOL's monetary policy also aims at maintaining exchange rate stability and controlling the amount of banks' reserves so as to limit excessive lending. The monetary base is backed by gold and foreign currency reserves. 28See ECOFIN (2001). 29 Second, the cost of uncertainty and unanticipated inflation would be significant. Latvia's current environment of stable prices offers a very reliable basis for corporate decisionmaking. In considering investment opportunities and hiring and firing decisions for the medium term, it is important for firms to continue to discriminate between relative price adjustments and general changes in the price level. This process would be particularly difficult in a context of high inflation. During its participation in ERM II, Latvia would benefit from maintaining its current monetary policy strategy, which creates a non-inflationary environment and allows cross-border relative price comparisons to be made without the distorting influence of exchange rate fluctuations. Keeping stable prices would allow firms to enter into medium- and long-term investment contracts easily (instead of looking for short-term, quick return contracts), which is likely to foster job creation. Finally, inflation--which functions as a tax--would lead to negative redistributive consequences, especially during ERM II. If wage increases become the main determinant of inflation, income distribution will change to benefit workers and harm profits (firms). Wage increases in excess of labor productivity growth would put pressure on prices. And Latvian firms facing international competition will have to cut their profit margins to adjust to higher unit labor costs in order to keep their market shares. 4.2.2 Latvia's Monetary Policy and Convergence A More Ambitious Approach to Real Convergence. The Latvian authorities have announced their intention to join ERM II as soon as possible and to accelerate the process of adopting the euro. In the run-up to the adoption of a single currency, Latvia could use monetary policy to foster both nominal and real convergence. This will imply maintaining the price stability achieved over recent years, preserving gains from stabilization (especially by implementing the prudent budget policies outlined in Section I), and completing the remaining structural reforms agenda to enhance the country's medium-term growth potential. Latvia has already achieved a significant degree of nominal convergence, as the inflation rate and short-term and long-term interest rates have fallen over the past decade, reflecting lower inflation expectations and the prospects of EU membership (Table 5). Table 5: Latvia and the Maastricht Criteria Variable Maastricht Convergence Criteria Latvia, end 2003 Inflation should not exceed by more than 1.5 percent the rate 3.6 (end-of-period) Inflation in the three member states showing the lowest level of price increases 2.9 (annual average) Fiscal deficit Annual budget deficit lower than 3 percent of GDP ~1.9percent of GDP (103 million Public LVL) Finance Debt National debt lower than 60 percent of GDP ~15percent (end-of-year) Within 2 percent of the rates in the three countries with Long-term interest rate lowest rates of inflation 5.125percent (fixed income 10- year T-Bonds) A country should have respected the normal fluctuation The Lats is still pegged to SDR (in Exchange rate margins of the ERM for at least 2 years without severe principle meets this criterion) tensions and without devaluing its currency against any other member's currency on its own initiative 30 However, progress on real income convergence--as measured by the most common indicator, i.e. per capita income--has been limited. Latvia's per capita income stands at about 33 percent of the EU average in purchasing power standards. Given the fact that this picture is not likely to change in the short run and that there are already significant differences in income levels both across and within EU countries29, policymakers in Latvia may want to consider a more ambitious approach to convergence, such as ensuring that the country quickly adopts international best practices and standards in terms of institutions and economic structures. Such an approach ("real structural convergence" as Padoa-Schioppa30 puts it) would refocus the national debate after EU accession on the institutional framework and business environment that Latvia will need to eventually catch up and compete on an equal footing with other EU countries. Inflation Risks in Latvia During ERM II. In spite of its remarkable performance in achieving nominal convergence with the euro area over recent years, Latvia will face serious challenges in fighting inflationary pressures during the run-up to the adoption of the euro. First, the impact of higher productivity growth and the so-called Balassa-Samuelson effect31 will lead to a gradual increase of prices toward the level of richer EU countries. Also, the consequences of further price liberalization32 and deregulation imposed by EU membership will be an additional source of inflationary pressures, though this will be a normal adjustment process for the real economy. Second, if the recent trend in fiscal policy in not quickly reversed (the fiscal balance went from a surplus of 0.3 percent of GDP in 1997 to an estimated deficit of 2 percent of GDP in 2003), there is the possibility that macroeconomic factors combined with inappropriate wage policies after EU accession could create inflation. Thus it is essential to pursue and complete Latvia's structural reform agenda and to make all stakeholders involved in the wage bargaining system aware of the perils of claiming wage levels prevalent in rich EU countries too quickly. The BOL's Medium-Term Challenge: Avoiding Stringent Rules. In ensuring that monetary policy helps create an environment conducive to growth and employment over the next few years, the BOL will face uncertainty. In that context, it will be crucial for the central bank to maintain credibility and transparency in its goals and actions. In principle, Latvian monetary authorities could choose to reach that objective by committing to a set of monetary policy rules, that is, establishing a clear mechanism as to how monetary policy instruments should be used as a function of a specific list of variables (i.e., constant money growth, interest rate rules, etc.).33 29For instance, the per capita income level of the city of Brussels is twice the EU average, and that of Lombardia is twice that of Sicily. 30See Padoa-Schioppa (2002). 31The Balassa-Samuelson effect refers to the fact that in fast-growing economies like Latvia productivity growth in the tradable sector is higher than in the non-tradable sector, which tends to lead to the appreciation of the real exchange rate. 32It is estimated that administered prices still account for around one-fifth of the consumer price index in most transition countries, despite substantial progress on price liberalization over the past decade. In Latvia, prices in some key sectors like housing rents are still fixed at levels below cost recovery. Therefore, liberalization as required for EU membership is likely to create upward pressures on the CPI. 31 Such an approach--choosing to set stringent rules--would be a difficult path for a country like Latvia to follow, as its policymakers face a very limited knowledge of the specific aspects of the structure of the economy. Therefore, it would be difficult to analyze the functioning of the channels of monetary policy transmission, and the precise lags at which monetary policy decisions are transmitted to the economy. Furthermore, attempting to observe future inflation at a given horizon (as required by the inflation targeting approach) would be impaired by the difficulty of observing current and future shocks and understanding their underlying nature (for example, whether they are transitory or not, and whether they will affect demand or supply). A Monetary Strategy (Not Rules) Better for Employment. Given the rapidly evolving structure of the Latvian economy and the uncertain external environment ahead, a mechanical use of monetary policy rules would not be the best option for the BOL in pursuing price stability, which is a key pre-requisite to sustained growth and employment. However, because the lags of monetary transmission are also uncertain, the central bank should not solely rely on discretionary actions, as these may exacerbate economic fluctuations. In order to stabilize expectations and increase the likelihood of success for its actions during ERM II, the BOL should commit itself to a monetary policy strategy34 of the type followed by the ECB. This is particularly appropriate for Latvia as the structure of the economy has been changing rapidly over the past decade, making it one of the reasons why it is so difficult to estimate the NAIRU (non-accelerating inflation rate of unemployment). Monetary policy in Latvia in the run-up to the adoption of the euro will be effective in fostering welfare only if complemented by a stable macroeconomic environment (especially a prudent fiscal policy) and the pursuit of microeconomic and regulatory reforms. 5. Conclusion Latvia's macroeconomic progress since independence has been substantial. Growth has been among the highest in Europe, and labor productivity has increased. This has helped to maintain competitiveness despite some appreciation of the real exchange rate of the lat, and the private sector has attracted strong FDI flows. These flows have financed the current account deficit and kept debt indicators at moderate levels. However, there are some causes for continued concern about Latvia's progress. Per capita GDP (PPP based) remains at about one-third of the EU average and unemployment is still much higher than the EU average. Fiscal complacency is also a risk. The deterioration of the external accounts and the relaxation of fiscal policy are the most urgent macro challenges on the agenda. The recent lapses in fiscal policy (in spite of strong economic growth) may prove overly procyclical and lead to inflationary pressures, further deterioration of the current account, and 33Such "simple rules" are often criticized because the information used for monetary decisions is too narrow. Some authors have suggested using "optimizing rules" like inflation targeting, which allows the central bank to act on a larger set of information. 34I define monetary strategy here in the sense suggested by Issing (2003:3), that is, "a systematic framework for the analysis of information and a set of procedures designed to achieve the central bank's main objective. As such, it differs from the strict formulation of a rule because it is contingent on all relevant information [emphasis supplied], and it does not have a simple analytical representation." 32 overvaluation of the exchange rate. Given the facts that domestic demand remains buoyant and that Latvia will have to accommodate the expenditure commitments associated with NATO and EU membership while simultaneously aiming to fulfill the medium-term goal of a balanced budget, a more prudent fiscal policy should be the main short-term policy objective. The authorities should therefore reassess their current medium-term budget framework and strengthen their fiscal rules so that off-budget spending is eliminated and controls over spending ministries and local government finances are reinforced. They should also rethink any further tax reductions until a clear strategy is designed for compensating for projected losses in government revenues. Monetary policy can contribute to sustainable growth and job creation in Latvia in the medium and long term by providing an environment for price stability. However, it would be a mistake to consider that the BOL assumes sole responsibility for establishing price stability for the Lats and stimulating job creation in the years to come. Monetary policy will have to be complemented by sound fiscal policy and prudent wage developments aligned with productivity growth. Indeed, given the recent deterioration of the fiscal stance and the challenges of increased public spending associated with EU and NATO accession, if the Latvian authorities do not aggressively pursue their goal of a balanced budget over the medium term, the BOL may be forced to take measures in the fight against inflation that will entail some short-term loss in output and employment. This would obviously have adverse effects on public opinion but in the long run, the foundations for balanced growth and employment would be protected. Because the effects of monetary policy on the economy occur with lags that are difficult to identify precisely (especially given the huge structural transformation taking place in Latvia), it would be impossible to try to use it for fine-tuning purposes. Monetary policy objectives can only be pursued in the medium term. Moreover, issues related to growth and employment creation cannot be addressed only by sound monetary policy. This fact is illustrated by Latvia's current unemployment problem: As its causes are mostly structural, its solution requires structural reforms, not expansionary monetary policy. An important area for further research is a systematic assessment of the output gap in Latvia (the degree of slack in the economy). Given the fast-changing structural transformation of the economy, such a task will be a challenging one. But it would help the authorities make a more rational judgment on the appropriate fiscal stance, the underlying external position, and the causes and solutions to unemployment. 33 Reference Alesina, Alberto F. and R. Perotti (1997), "Fiscal Adjustment in OECD: Composition and Macroeconomic Effects," IMF Staff Papers, 44, June, 210-248. Ĺslund, A. (2001), "The Myth of Output Collapse after Communism," Carnegie Endowment Working paper, no. 18. Begg et al. (2002), "Surviving the Slowdown," Monitoring the European Central Bank, 4, CEPR. Benhabib, Jess, and Mark M. Spiegel (1994), "Role of Human Capital in Economic Development: Evidence from Aggregate Cross-Country Data," Journal of Monetary Economics, vol. 34, pp. 143-173 Blanchard, Olivier J., Rudiger Dornbusch, Paul Krugman, Richard Layard, and Lawrence Summers (1992), Reform in Eastern Europe, Cambridge, MA. MIT Press. Boldrin, Michele and Fabio Canova (2003), "Regional Policies and EU Enlargement," in Bernard Funck and Lodovico Pizzati (eds.), European Integration, Regional Policy, and Growth, World Bank, Washington D.C., pp. 33-93. Doyle, Peter, Louis Kuijs, et Gurong Jiang (2001), Real Convergence to EU Income Levels: Central Europe from 1990 to the Long-Term, working paper 01/146, Washington D.C., IMF. ECOFIN (2001), Reportby the ECOFIN to the European Council in Nice, http://europa.eu.int/comm/economy_finance/publications/european_economy/2001/c2001_01 _en.pdf. FIAS (2003), Developing Knowledge Intensive Sectors, Technology Transfers, and FDI, Washington D.C., World Bank. Frankel, Jeffrey A. and Andrew K. Rose (1998), "The Endogeneity of the Optimum Currency Area Criteria," The Economic Journal, vol. 108, n0 449. pp. 1009-1025. Funck, Bernard, ed. (2002), Expenditure Policies Towards EU Accession, Technical paper no. 533, World Bank, Washington D.C. Hagemann, Robert (1999), The Structural Budget Balance: The IMF Methodology, working paper, IMF, Washington D.C. Hemming, Richard, Michael Kell, and Selma Mahfouz (2002), The Effectiveness of Fiscal Policy in Stimulating Economic Activity--A Review of the Literature, working paper, Washington D.C., IMF. IMF (2001), Selected Euro-Area Countries: Rules-Based Fiscal Policy and Job-Rich Growth in France, Germany, Italy, and Spain, report no. 01/203, Washington D.C. IMF, Staff Reports for Article IV Consultations, Washington D.C., various years, 34 http://www.imf.org/external/country/LVA/index.htm Issing, Otmar (2003), Considerations on Monetary Policy Strategies for Accession Countries, Paper presented at the seminar on "Monetary Policy Strategies for Accession Countries" organized in Budapest, Hungary, February 28. Krugman, Paul (1993), "Lessons of Massachusetts for EMU," in: Francisco Torres and Francesco Giavazzi (eds.), Adjustment and Growth in the European Union, New York, Cambridge University Press. Martin, Carmela, Francisco J. Velasquez, and Bernard Funck (2001), European Integration and Income Convergence: Lessons for Central and Eastern European Countries, technical paper no. 514, Washington D.C., World Bank. Masson, Paul (2000), "Fiscal Policy and Growth in the Context of European Integration," in: Jan Smets and Michel Dombrecht (eds.), How to Promote Economic Growth in the Euro Area, Cheltenham, UK and Northampton, MA, USA, Edward Elgar, pp. 112-140. Mitra, Pradeep, and Marcelo Selowski (2002), Transition--The first ten years : analysis and lessons for Eastern Europe and the former Soviet Union, Washington D.C., World Bank. Mueller, Johannes et al. (2001), The Baltics: Medium-Term Fiscal Issues Related to EU and Nato Accession, Washington D.C., IMF. Padoa-Schioppa, Tommaso (2002), Trajectories towards the Euro and the Role of ERMII, Speech at the 2002 East-West Conference, Vienna, November 4. Svensson, Lars E.O. (2002), "A Reform of the Eurosystem's Monetary Policy Strategy is Increasingly Urgent,"Briefing Paper for the European Parliament's Committee on Economic and Financial Affairs, http://www.princeton.edu/~svensson Tanzi, Vito and H. Zee (1997), "Fiscal Policy and Long-Run Growth," IMF Staff Papers, 44, pp. 179-209. World Bank (1996), World Development Report 1996: From Transition to Market, Washington D.C.. World Bank (2002), Republic of Latvia: Public Expenditure Review, report no. 22393-LV, Washington D.C. World Bank (2003), Latvia Beyond Territorial Reform, report no. No. 25466-LV, Washington D.C.. Zile, Robert and Inna Steinbuka (2001), "Latvia on the Way to the European Union," Finance and Development, vol. 38., no. 2, June. 35 ANNEX Table 6: Episodes of Fiscal Adjustment and Stimulus in Transition Economies, 1992-2001 Adjustment Stimulus Country Adjustment Stimulus 2/ (Expenditure Based) 3/ (Expenditures Based) 3/ CEE countries Albania 93-95; 97-98 - 93-95; 97-98 - Bulgaria 2/ 94 93; 97 94 - Croatia - 95; 99 - - Czech Republic 93 94 93 - Hungary 2/ 94-96 93 94-96 - Poland - - - - Romania 93; 99 - 93 - Slovak Republic 93-94; 99 97 93-94; 99 - Slovenia - - - - Baltics Estonia 94; 97 95; 98-99 97 95; 98-99 Latvia 93; 96-97 94; 99 - 94; 99 Lithuania 93 - 93 - FSU Armenia 94-95; 97 99 94-95; 97 99 Azerbaijan 95-96 98 95-96 - Belarus - 93 - 93 Georgia 2/ 95 - 95 - Kazakhstan 95; 97 94; 96; 98 95; 97 - Kyrgyz Republic 93-94; 96 95 96 - Moldova 93; 95; 98-99 - 93; 98-99 - Russian Federation 93; 95 94 93; 95 - Turkmenistan 96 98 96 98 Ukraine 95-96; 98 - 95-96; 98 - Uzbekistan 94; 97 96 94; 97 96 Mongolia 4/ 94-95 93; 98 94-95 93; 98 Total episodes 33 23 27 10 Source: Excerpts from Purfield (2003) 1/ Episodes successful in sustaining the reduction in the primary balance after the end of the adjustment are highlighted in bold. Expansionary stimulus episodes are also highlighted in bold. 2/ The sample excludes the 1996 Georgia, 1998 Hungary, and 1999 Bulgaria episodes because the widening of the deficit reflects structural breaks in the data. In Georgia and Bulgaria, off-budget accounts were incorporated; and in Hungary the widening deficit reflected pension-system reform. 3/ An expenditure-based adjustment (stimulus) is one where at least 60 percent of the improvement in the deficit is derived from cuts in primary expenditure and net lending. 4/ Mongolia is included in the group of Former Soviet Union (FSU) countries because as a former satellite of the Soviet Union, it shares many of the characteristics of other FSU transition countries. WB157883 L:\Latvia\Macro\latvia-wp-0504-final.doc May 11, 2004 9:48 AM 36