Journal Issue

Euro Area Policies: Selected Issues

International Monetary Fund
Published Date:
July 2007
  • ShareShare
Show Summary Details

II. The Euro and the New Member States29

A. Introduction

26. The new member states of European Union are expected to gear their policies toward fulfilling preconditions for joining Economic and Monetary Union (EMU) and adopt the euro when they do. The European Commission (EC) and the European Central Bank (ECB) are regularly monitoring progress in this area. The entry preconditions are embedded in the Maastricht Treaty and require countries to achieve a high degree of sustainable nominal convergence before they can participate in EMU.30 The Maastricht criteria have for the past fifteen years served as the cornerstone of a gradual approach to expanding EMU, aiming to ensure its credibility and sustainability. The criteria helped create a shared culture of stability among the “old” members (OMS) and are now expected to play the same policy-anchoring role for the new member states (NMS).31

27. This paper examines the macroeconomic policy challenges the NMS face as they prepare for joining monetary union.32 These challenges largely stem from the convergence of incomes and prices in the NMS to euro-area levels, capital inflows and financial deepening, and the resulting inflation and exchange rate developments. Indeed, foreign investors generally perceive the exchange rate risk in the NMS as low and are willing to on-lend capital to the domestic sectors in the NMS in euros and other European currencies.

28. Against this backdrop, the paper explores the policies the NMS would need to pursue to enter the euro area smoothly and prepare for a good performance inmonetary union. Using a dynamic stochastic general equilibrium model, the paper quantifies the degree of policy adjustment the NMS would need to undertake to meet the entry preconditions. The paper also explores structural features of NMS economies, notably their flexibility to adjust to shocks in monetary union.

29. The paper is organized as follows. Section II.B provides background information on the NMS. Section II.C summarizes the Maastricht criteria and Section II.D the policy challenges the NMS face in meeting them. Section II.E presents estimates of the policy adjustment needed. Section II.F summarizes the findings and concludes.

B. Background

30. The NMS are catching-up economies that are in the process of converging to the euro area in real and nominal terms. Per capita incomes in the NMS are generally lower than in the OMS and are growing fast. Price levels are also lower in the NMS and are converging to the euro-area levels, implying that inflation tends to be higher in the NMS and/or their nominal exchange rates tend to appreciate vis-à-vis the euro. Convergence in nominal interest rates, in part driven by declining risk premia for the NMS, is also proceeding at a rapid pace. Together, these phenomena tend to come with larger current account deficits in the NMS than typical in the OMS.

GDP PPP per Capita, 2006

(in U.S. dollars)

Source: IMF, World Economic Outlook.

GDP PPP per Capita, 2006

(annual percentage change)

31. The economic and financial linkages between the NMS and the OMS are strengthening. In general, the degree of business cycle synchronization between the NMS and the euro area is lower on average than between the OMS and the euro area. However, business cycle correlations between various NMS and the euro area now exceed those for Greece and Portugal. Production structures in the NMS are characterized by a higher share of agriculture and a lower share of services, but are gradually converging to those in the euro area.33 Inflation correlations and variance shares explained by common euro-area shocks are lower than for the OMS, but the transmission of common euro-area shocks to the NMS does not differ significantly from those to the OMS.34 About two thirds of NMS trade is with the euro area. The degree of integration of the NMS’ equity markets has increased in recent years, especially for the larger NMS such as the Czech Republic, Hungary, and Poland.35 Local bond prices in the Czech Republic and Poland exhibit fairly high comovement vis-à-vis Germany. With foreign banks, mostly from the euro area, accounting for a significant share of assets in the NMS, the banking systems of the OMS and the NMS are closely integrated.

32. Most NMS have announced their plans to adopt the euro in the coming years. This would necessitate changes to their current exchange rate regimes, which range from currency boards to freely floating exchange rates. Membership in monetary union is expected to bring long-term benefits in the form of further integration of NMS and OMS markets for goods, services, labor and capital and faster real and nominal convergence (Box II.1). When the NMS join the euro area, they will lose monetary policy independence and the nominal exchange rate will no longer be able to act as a shock absorber. Staff analysis suggests that this may not entail major losses.36 Moreover, the elimination of exchange rate risks should at least partly compensate for these costs. The euro area is also expected to benefit from NMS euro adoption through further market integration and improvements in production efficiency.

Exchange Rate Regimes in the NMS and Euro Adoption Plans
Progress in Euro Adoption
IMF Classification 1/ERM IIEMU 2/
Czech RepublicFloatHas not joined yetNo official target date has been set. On current budget plans, 2012 would be the earliest feasible date.
HungaryIntermediateHas not joined yetNo official target date has been set. The Convergence Program of December 1, 2006 aims at meeting the Maastricht criteria in 2009.
PolandFloatHas not joined yetNo official target date.
Slovak RepublicFixedJoined on November 28, 2005Target date is set for January 1, 2009.
EstoniaFixedJoined on June 28, 2004The government is committed to adopting the euro at the earliest possible date, which it now estimates to be 2011, based on current inflation forecasts.
According to the information released by the Ministry of Finance, in 2007 the
LatviaFixedJoined on May 2, 2005Government would discuss a new target for the changeover to the euro, tentatively in 2011-2013.
No official target date has been set. According to the government, Lithuania
LithuaniaFixedJoined on June 28, 2004will aim to join the euro area as soon as possible and the more favorable period for the country to join the euro area starts in 2010.
BulgariaFixedHas not joined yetTarget date is set for January 1, 2010.
RomaniaFloatHas not joined yetTarget date is set for January 1, 2014.
Sources: IMF Annual Report on Exchange Arrangements and Exchange Restrictions and International Financial Statistics; European Commission; European Central Bank; National central banks.

“Fixed” includes currency boards, conventional pegs, and narrow bands. “Intermediate” includes tightly managed floats and broad bands. “Float” includes managed and independent floats.

Latest information available from European Commission and national authorities.

Sources: IMF Annual Report on Exchange Arrangements and Exchange Restrictions and International Financial Statistics; European Commission; European Central Bank; National central banks.

“Fixed” includes currency boards, conventional pegs, and narrow bands. “Intermediate” includes tightly managed floats and broad bands. “Float” includes managed and independent floats.

Latest information available from European Commission and national authorities.

33. The latest plans represent a delay in euro adoption compared to the original schedules announced in 2004, shortly after the NMS’ accession to the EU. The main reason for the delay was a failure of most NMS to satisfy convergence criteria. Growing skepticism about benefits from euro adoption and reform fatigue also contributed to a weakening of political support for euro adoption in some NMS. The latest Eurobarometer survey (May 2007) indicates that the majority of population in most NMS still supports euro adoption, but the margins are small in the Czech Republic and Estonia. Most citizens in Latvia and Lithuania are against euro adoption. The proportion of people in the Baltic countries expecting net benefits from euro adoption fell in May 2007 compared to September 2006.37

C. Maastricht Criteria for EMU Membership

34. The Maastricht Treaty leaves the timing of EMU entry open. The NMS are expected to join the euro area if and when they satisfy the entry preconditions. However, unlike Denmark and the United Kingdom, the NMS do not have an indefinite opt-out regarding participation in EMU.38 The fulfillment of the Maastricht criteria is assessed by the EU Council on the basis of the reports prepared by the EC and the ECB at least once every two years or at the request of a member state wishing to adopt the euro.39 Notwithstanding the NMS’ commitment to join the euro area, there are no legal limits on how long they can stay outside the euro area, and there are no sanctions for not satisfying the Maastricht criteria. Satisfying them before entering EMU is required. Adopting the euro otherwise, for example, through “euroisation” (i.e., using the euro as a legal tender) would be inconsistent with the spirit of the Maastricht Treaty.

35. The Maastricht criteria require prospective members to achieve sustainable nominal convergence before entering monetary union. This involves the following:

  • Pricestability. The average annual rate of inflation should not exceed by more than 1½ percentage points that of the three best performing EU countries in terms of price stability. The notion of “best performance” is not defined in the Treaty. The 2004 Convergence Reports took “best performance” to mean “the lowest non-negative inflation.”40 Sustainability is also not defined in the Treaty. In practice, the EC examines whether inflation was at or below the Maastricht reference value in the recent past, bearing in mind one-off and temporary effects, and assesses whether such convergence is likely to be sustained over the year following the evaluation. The ECB uses a broadly similar approach. In this context, the EC and the ECB also assess the recent trends and outlook for unit labor costs and the current account balance to gauge if these point to overheating pressures.

  • Exchangeratestability. Countries are required to keep their exchange rates within the “normal” fluctuation margins provided for by the exchange rate mechanism of the European Monetary System (ERM-II) without severe tensions for at least two years before the examination and not to devalue their currency during this period.41 When assessing tensions in the exchange rate, the EC and the ECB examine how far the exchange rate is from the central parity, the size of the short-term interest rate differential and the size of foreign exchange interventions. Specific limits on the size of interventions or revaluations that would be consistent with the fulfillment of the exchange rate stability criterion have not been specified.

  • Convergenceoflong-terminterestrates. The average over the latest 12 months of the nominal long-term interest rate should not exceed by more than two percentage points the average of the three best performing EU member countries in terms of price stability.

  • Fiscalsustainability. The fiscal deficits should not exceed 3 percent of GDP, and gross government debt should not exceed 60 percent of GDP.42 The examination of the fiscal deficit and debt under the excessive deficit procedure outlined in the Maastricht Treaty is designed to take into account whether the fiscal deficit ratio “has declined substantially and continuously and reached a level that comes close to the reference value” or that “the excess over the reference value is only exceptional and temporary and the ratio remains close to the reference value.” The government debt ratio is allowed to be “sufficiently diminishing and approaching the reference value at a satisfactory pace.”

36. When assessing progress in fulfilling the Maastricht criteria, the euro-area authorities guard against countries’ satisfying the criteria in an “opportunistic” manner. Accordingly, they would consider the roles of unusually favorable external conditions (e.g., declining oil prices), rapid nominal exchange rate appreciation (where applicable), and changes in administrative prices or indirect taxes in lowering inflation. The reason is that the effects of temporary factors or “quick-fix” measures would unwind after entry into EMU.43

37. Although most NMS already comply with at least some of the Maastricht criteria, achieving all of them is proving difficult. As of December 2006, inflation was below the Maastricht reference value in only two out of the nine NMS: the Czech Republic and Poland. Low inflation in these two countries largely reflected the strength of their domestic currencies (Figure II.1). Fiscal deficits exceeded the 3 percent limit in the CEECs, and were particularly large in Hungary. However, the Baltics, Bulgaria, and Romania all had small deficits or surpluses. Government debt was below 60 percent in all NMS, except Hungary. Long-term interest rates were within the Maastricht reference limits in all NMS, except for Hungary where fiscal sustainability problems loom large.44

Figure 1.NMS and Euro Area Members: Maastricht Criteria

Sources: Eurostat, national statistical offices, IMF staff estimates.

D. Policy Issues in the Context of Euro Adoption

Achieving Nominal Convergence

38. Achieving nominal convergence among countries with different per capita incomes can be challenging. Income convergence tends to be accompanied by convergence in price levels. This takes place either through inflation running above or the nominal exchange rate appreciating relative to that in higher income partner countries. Both imply an appreciation of the real effective exchange rate of the lower income countries. The best-known mechanism through which price convergence takes place is the Balassa-Samuelson effect, whereby productivity growth in the traded goods sector leads to an overall increase in prices and puts upward pressure on CPI-based real exchange rates.45

NMS: GDP per Capita and Prices, 1995-2006

(In percent of EU-25 levels)

Source: World Economic Outlook, IMF staff estimates.

Note: Countries shown on the chart include Bulgaria (BG), the Czech Republic (CZ), Estonia (EE), Hungary (HU), Latvia (LV), Lithuania (LT), Slovenia (SI), Poland (PL), Slovak Republic (SK), Romania (RO).

39. While Balassa-Samuelson related challenges could be large for the NMS, empirical studies suggest that this is not necessarily the case. Productivity differentials seem to explain from ⅓ to 2⅓ percentage points of annual inflation differentials in the transition economies vis-à-vis the euro area, with most estimates clustered around 1-2 percent. In some NMS, no evidence of the Balassa-Samuelson effect is found currently, for example, because productivity in the non-tradable goods sector is growing faster than in the tradable goods sector, owing to significant inflows of foreign direct investment (FDI) in the non-tradable goods sector. In others, however, the Balassa-Samuelson effect is likely to be operating, putting pressure on prices or the exchange rate.

Empirical Estimates of the Balassa-Samuelson Effect, 1995-2005(In percent per year)
Latest Estimates UsingLatest Estimates UsingAverage of EstimatesAverage of
Total Services 1/Market-Based Services 1/Available in the Literature 2/Columns (1)-(3)
Czech Republic(1)
Slovak Republic(4)
Average of Rows (1)-(7)
Sources: Égert (2007), Égert, Halpern and McDonald (2006).

Égert (2007).

Based on the literature survey by Égert, Halpern and McDonald (2006). The time period varies across the studies surveyed, but generally precedes 2001.

Sources: Égert (2007), Égert, Halpern and McDonald (2006).

Égert (2007).

Based on the literature survey by Égert, Halpern and McDonald (2006). The time period varies across the studies surveyed, but generally precedes 2001.

40. Several considerations suggest that the equilibrium rates of nominal convergence may be higher than suggested by the Balassa-Samuelson estimates (Figures II.2-3). These relate to the rapid structural transformation of the NMS economies that followed their opening up to the world economy and long-term improvements in living standards. For instance, the quality of tradable and non-tradable goods (for example, health care, education, telecommunication and financial services) is likely to gradually improve, commanding higher prices. Also, in tandem with rising incomes, consumption preferences in the NMS are likely to shift away from food and other basic goods to services. Similarly, EU membership may prompt lasting real appreciations, for example, via the transfers from the OMS to the NMS for the purpose of upgrading infrastructure.46

Figure 2.NMS: Real Effective Exchange Rates, 2000-06

Source: Eurostat.

Figure 3.NMS: Export Market Share to the EU15, 2000-06

Source: IMF, Direction of Trade Statistics.

41. By the same token, more transient or disequilibrium phenomena are also likely to be embedded in rates of real appreciation. These include, for instance, the one-off, level-adjustment associated with the elimination of the undervaluation gap in the NMS currencies observers believe existed at the beginning of transition.47 Another set is the strongly favorable impulses associated with EU membership and the confluence of rapid financial integration and unusually benign global financial conditions. Crucially, EU membership entails a whole range of highly beneficial legal and institutional reforms. Their implementation is likely to have fostered an initial, stock-adjustment-type economic boom, notably heavy investment in the NMS economies by foreign investors.48 Expectations of continued rapid economic and monetary integration, including euro adoption might have also contributed to these booms. Information asymmetries in financial markets might in some cases have lead to irrational exuberance and exchange rate overshooting in the course of such stock- or level-type adjustments.

42. Accordingly, there is significant uncertainty as to what drives differentials between NMS and OMS price and exchange rate developments. These could be transient or more permanent forces. The transient forces could be fundamentals, such as the adoption of the EU acquis, or disequilibrium trends such as investor exuberance and policy shortcomings. No comprehensive estimates of trend real appreciation associated with the factors that have been discussed above are available.49

43. The uncertainty about the size of the “true” real appreciations associated with real convergence has contributed to controversies over the Maastricht criteria. The inflation criterion allows a 1½ percentage point margin over inflation in the three EU member countries with the lowest nonnegative inflation, while the exchange rate stability criterion allows up to 15 percent nominal appreciation during the two years countries need to spend in ERM-II and there is scope to adjust the central parity. This raises at least two considerations:

  • Some have questioned whether the room under the Maastricht inflation criterion is sufficient to fully accommodate equilibrium real appreciation trends. While Balassa-Samuelson effects are typically estimated to be small, this might not be the case for all NMS and other factors might push equilibrium inflation beyond the 1½ percent margin over the best EU performers.50

  • Others consider that current NMS inflation and exchange rate developments are appreciably transient in nature, driven by both transient “equilibrium” stock-adjustment (such as the acquis communitaire) and “disequilibrium” investor exuberance and the NMS country authorities’ policy complacency. These observers do not consider the Maastricht inflation criterion inordinately tight and argue that the “disequilibrium” developments could come back to haunt the NMS in the case of premature euro adoption.

In this setting, the key challenge for NMS policymakers, including with respect to euro adoption, is dealing with the current economic booms that have raised vulnerabilities. At the same time, work on strengthening the NMS economies’ resiliency to shocks in monetary union needs to continue.

Inflation and Exchanges Rates in the Context of an Accession-related and Global Upswing

44. Against the backdrop of EU membership and interest rate declines, domestic credit and demand booms have been unfolding in the NMS. Country risk premia for the NMS seem to have declined earlier than they did in the catching up OMS prior to their joining EMU. Financial markets are viewing the NMS favorably, pricing their sovereign assets some 50-100 basis points below the levels that would be expected based on standard policy fundamentals.51 The NMS’ success in macroeconomic stabilization and structural reforms, EU accession, the policy-anchoring role provided by the euro adoption plans, all seem to have contributed to rapid interest rate convergence. Global factors, such as low interest rates, ample liquidity and a widening of the investor base for emerging markets have also played a role.52 As a result, capital flowed to the NMS, in the form of direct investment (especially in the CEECs), bank loans (especially in the Baltics), and portfolio investment. Last year, for example, capital inflows reached about 6 percent of GDP in the Baltics, 12 percent of GDP in Bulgaria and Romania, and about 3 percent of GDP in the CEECs.53 As in the catching up OMS, convergence-driven booms in the NMS are associated with rapid credit and domestic demand growth, appreciating real exchange rates and inflationary pressures.

Sources: Bloomberg, IMF staff estimates.

1/ t corresponds to the time of EMU entry for Greece, Ireland, Portugal and Spain and 2011 for the NMS.

NMS: Net Capital Flows, 1999-2006

(In percent of GDP)

Source: IMF, World Economic Outlook.

Sources: Eurostat; IFS, National Statistical Offices; and IMF staff estimates.

45. With large capital inflows also come some vulnerabilities. Rapid credit expansions are raising concerns about overheating, widening external imbalances, and rising balance sheet risks in some NMS. Domestic borrowers have been contracting loans in euros and other European currencies (mostly the Swiss franc), leading to a build up of currency mismatches in the private sector balance sheets. For NMS banks, many of which are foreign owned, lending in euros and other major European currencies facilitates or reduces the need for hedging. As a result, the share of foreign currency lending in the NMS exceeds that in Western Europe and in Latin American and East Asian emerging markets. Currency mismatches make the private sector vulnerable to an exchange rate depreciation, and through credit risk, the NMS banking sector might also be affected. Via the potential for spillovers, vulnerabilities extend beyond the NMS (Boxes II.2-Box 3).

Financial Market Developments in the NMS Compared to other Emerging Markets

(January 2006-February 2007)

Source: Bloomberg, Luengnaruemitchai and Schadler (2007).

Foreign Currency-Denominated Loans, 2005

(in percent of total outstanding loans)

Source: National authorities, Fund staff estimates.

Note: Regional averages for East Asia and Latin America cover emerging market countries.

46. Cooling off an economy during a surge in capital inflows and rapid credit growth is a challenge. The NMS with fixed exchange rate regimes have to rely on fiscal policy for managing demand pressures. Even in countries with floating exchange rates, monetary tightening would be unlikely to slow credit growth, as higher interest rates would tend to encourage further capital inflows and borrowing in foreign currency. Fiscal tightening could help rein in inflation, but it might need to be of considerable magnitude to have the necessary impact. Prudential measures might slow credit expansion, but their effectiveness is likely to decline over time, if the underlying incentives for credit expansion remain strong, for example, because of a pent-up demand for credit in formerly financially repressed economies or because foreign banks move business abroad. Capital controls might help slow a surge in capital inflows, but they are illegal in the EU in other than crisis situations.54

Fostering Resiliency to Shocks in Monetary Union

47. Successful performance in the monetary union depends on a variety of features, which NMS economies display to varying degrees. These features relate to trend productivity growth, adjustment to shocks, integration with other union members (intra-and inter-temporal risk sharing), and scope for using fiscal policy (intertemporal risk sharing). Productivity growth in the NMS has been running significantly above the levels experienced by the OMS. Ascertaining the extent of the sustainability of this strong productivity growth, part of which may be related to the current economic booms in the NMS, is difficult, but significant catching up potential in the NMS remains. Other macroeconomic indicators in the NMS compare favorably, notably with those OMS that are widely considered to have faced difficulties (Figures II.4-5).

Figure 4.Portugal and the Baltics: Selected Macroeconomic Indicators 1/

Source: AMECO; OECD; and WEO.

1/ For Portugal, t corresponds to 1999, when the country joined EMU. For the Baltics, t is 2006.

Figure 5.Portugal and the CEECs Selected Macroeconomic Indicators 1/

Source: AMECO; OECD; and WEO.

1/ For Portugal, t corresponds to 1999, when the country joined EMU. For the CEECs, t is 2006.

Total Factor Productivity Growth

(2002-2006 average; in percent)

Sources: Eurostat, Ameco, national statistical offices, IMF staff estimates.

48. Considerable labor market flexibility in the NMS should facilitate restoring competitiveness in the event of adverse shocks but product markets are more rigid. Employment protection legislation is less restrictive in the NMS than OMS, minimum wages are lower,55 collective bargaining structures in the NMS are less centralized, and unemployment benefits are less generous.56 Measures of labor market flexibility provide comfort that the NMS are relatively well poised to adjust to asymmetric shocks in monetary union. However, downward flexibility in wages has largely been untested in the NMS and this provides grounds for caution. In addition, there is variation in labor market flexibility across the NMS, with wages being more responsive to productivity and unemployment in the Baltics than in the CEECs.57 Further improvements of product and labor market flexibility, including through deregulation, welfare reforms, and measures to facilitate geographical labor mobility, remain a priority for many NMS.58

NMS, Euro Area and Other OECD Countries: Labor Market Regulations, 2001 1/

Source: OECD.

1/ Sorted by 2001 values. The scale of indicators is 1-10 from most to least regulated. The value for the euro area is calculated as a simple average.

NMS and Euro Area: Measures of Labor Market Rigidities 1/

Sources: OECD, Boeri and Garibaldi (2006), IMF staff calculations.

1/ Data refer to the latest year available. Minimum wage is in percent of average wage. Unemployment benefits are in percent of averages wages. Marginal effective tax rates refer to 100 percent of average wages for single persons without children. Country averages are calculated based on data availability.

NMS, Euro Area and Other OECD Countries: Business Regulations, 2001 1/

Source: OECD.

1/ Sorted by 2001 values. The scale of indicators is 1-10 from most to least regulated; Euro area (simple average).

49. Significant financial integration of the NMS with the OMS should facilitate risk sharing. Foreign banks, mostly from euro-area countries, account for a significant portion of banking system assets in the NMS, greater on average than in Western Europe and in emerging markets in other regions of the world. Foreign presence is significant in other financial sectors, albeit to a lesser degree. The variety of financial products and services available to NMS consumers is thus similar to that in the OMS. The NMS have also been major recipients of cross-border capital flows, both FDI and portfolio investment, which are likely to have contributed to a decline in consumption volatility in recent years. However, judging by the de facto indicators of the degree of financial development and integration,59 NMS financial systems remain less developed and less integrated into the global financial system than those in Western Europe and other emerging markets. Continued financial deepening, notably development of market-based finance, should further enhance the ability of the NMS to adjust to shocks in EMU.

Share of Foreign-Owned Banks

(In percent of total assets)

Source: European Central Bank, national central banks.

1/ Data refer to 2004 or 2005, depending on availability.

External Assets and Liabilities

(in percent of GDP)

Source: Lane and Milesi-Ferretti (2006).

50. The NMS also seem to have fiscal flexibility, which they can draw on for stabilizing their economies in response to asymmetric shocks in monetary union. Although the automatic stabilizing properties of the NMS budgets overall seem weaker than in the euro area, this is not appreciably so. Some budget categories in the NMS, especially in the Baltics, have higher cyclical elasticities than those in the OMS. A broad measure of budget flexibility—variation in different categories of expenditures—is higher in the NMS than in the euro area, largely owing to greater variation in discretionary spending.60 The Baltic countries score particularly well on this measure. Nonetheless, there is scope for further improvement of budget flexibility in the NMS, particularly in nondiscretionary expenditures.

Budget Elasticities for a One Percentage Point Change in the Output Gap 1/

Income Tax
Social Security


Income Tax

Czech Republic1.190.801.391.00-0.020.39
Slovak Republic0.700.701.321.00-0.060.37
Euro area1.480.741.431.00-0.110.48
Source: Girouard and André (2005), Convergence Programs; and IMF staff estimates.

Estimates for the CEECs and the euro area are unweighted averages. Expenditures are current primary expenditures. The last column is the semi-elasticity, which measures the change in the budget balance, as a percent of GDP for a 1 percent change in GDP, based on 2003 weights. Budget elasticities for Latvia are averages for 2002-06, except that for the corporate income tax (CIT), which is estimated for 2001-05. The very high elasticity for CIT reflects sharp revenue increases in spite of tax cuts in 2003 and 2004.

Source: Girouard and André (2005), Convergence Programs; and IMF staff estimates.

Estimates for the CEECs and the euro area are unweighted averages. Expenditures are current primary expenditures. The last column is the semi-elasticity, which measures the change in the budget balance, as a percent of GDP for a 1 percent change in GDP, based on 2003 weights. Budget elasticities for Latvia are averages for 2002-06, except that for the corporate income tax (CIT), which is estimated for 2001-05. The very high elasticity for CIT reflects sharp revenue increases in spite of tax cuts in 2003 and 2004.

Variation in Key Expenditure Categories, 2000-2005

(Coefficient of variation, in percent)

Sources: National Statistical Offices; Eurostat; and IMF staff estimates.

1/ The coefficient of variation is defined as the ratio of the standard deviation to the mean. The Baltics in this chart cover the Estonia, Latvia, and Lithuania. The CEECs comprise the Czech Republic, Hungary, Poland, and the Slovak Republic.

E. Quantifying the Macroeconomic Policy Adjustment Needed to Prepare for Participation in Monetary Union

51. A dynamic stochastic general equilibrium model (DSGE) is used to quantify the effects of monetary and fiscal policies in the NMS.61 The Global Integrated Monetary and Fiscal Model (GIMF) developed by Kumhof and Laxton (2007) brings in the non-Ricardian features of overlapping generations (OLG) models by Blanchard (1985) and Weil (1989) into an open economy monetary business cycle model.62 The combination of finite planning horizons and liquidity-constrained consumers imply that fiscal policy has strong short-term and long-term effects. For example, the OLG structure for an open-economy means that permanent changes in government debt will have long-term effects on the net foreign liability position of a country. Consequently, GIMF predictions differ from the predictions based on standard open economy macroeconomic models with an infinitely-lived representative agent, rational expectations and non-distortionary taxation, which assume that Ricardian equivalence holds.63 The non-Ricardian features of the model are described in Box II.4.

52. A two-country annual version of GIMF is calibrated to representative NMS and the euro area. The results should not be taken as the precise estimates of the impulse responses in any particular NMS; the focus rather needs to be on the qualitative conclusions. Two versions of the model are considered: a NMS with a floating exchange rate regime and another with a fixed exchange rate regime. In the latter case, the “monetary” reaction function aims at stabilizing the exchange rate. To mitigate against uncertainty inherent in calibration-based modeling, extensive sensitivity analysis is undertaken. The approach to calibration is pragmatic and eclectic.64 The calibration of medium- and long-term responses relies on microeconomic studies, where available. Some parameters reflect the key characteristics of the NMS and the OMS, for example, the relative size of countries, their levels of trade, and their capital-output ratios.

Long-Run Values and Parameters Used in Calibration(In percent, unless specified otherwise)
Steady State ValuesNMSOMSStructural ParametersNMSOMS
Long-Run World Real Interest Rate3.003.00Elasticity of Substitution between Good Varieties:
Steady State Inflation2.002.00-Nontradables Manufacturing6.006.00
Long-Run NMS GDP as a Share of World GDP0.5099.50-Tradables Manufacturing6.006.00
Share of Liquidity-Constrained Agents in the Population0.400.25-Wholesale21.0021.00
Steady State Ratios-Consumer Goods Retail21.0021.00
Imports of Intermediate Goods/GDP20.000.10-Importers of Intermediate Goods41.0041.00
Exports of Intermediate Goods/GDP20.000.10-Importers of Final Goods41.0041.00
Imports of Final Goods/GDP20.000.10-Labor Market6.006.00
Government Consumption Spending/GDP15.0015.00-Home and Foreign Final Goods Imports1.505.00
Government Productive Investment Spending/GDP3.003.00-Home and Foreign Intermediate Goods Imports0.505.00
Private Investment/GDP16.0016.00-Capital and Labor in the Nontradables Sector0.750.75
•;Nontradables Output/GDP50.0050.00-Capital and Labor in the Tradables Sector0.750.75
Labor Income Share (total)60.0060.00Price Stickiness Parameters:
Labor Income Share in Nontradables Sector60.0064.00-Nontradables Manufacturing10.0010.00
Government Debt/GDP25.0025.00-Tradables Manufacturing10.0010.00
Structural Parameters-Wholesale10.0010.00
Intertemporal Elasticity of Substitution0.250.25-Importers of Intermediate Goods (if there is pricing to market)0.0010.00
Habit Persistence0.400.40-Importers of Final Goods (if there is pricing to market)0.0010.00
Depreciation Rate of Private Capital0.100.10-Labor Market10.0010.00
Depreciation Rate of Public Capital0.040.04
Source: IMF staff estimates.
Source: IMF staff estimates.

Modeling Fiscal and Monetary Shocks

53. First, a permanent 1 percent decrease in the fiscal deficit implemented through a balanced package of expenditure and tax measures is considered. Specifically, it is assumed that the package includes cuts in government consumption (with the weight of 0.33), an increase in labor taxes (0.33) and a decrease in labor transfers (0.34).65 Public investment, which is assumed to be a productive complement to private investment in GIMF, is exempt from cuts. As part of sensitivity analysis, alternative compositions of adjustment (exclusively through consumption and/or investment expenditures or exclusively through taxes and transfers) are considered as well.

54. The long-term effects of permanent fiscal consolidation in this non-Ricardian, intertemporal optimization model are independent of the exchange rate regime. When fiscal deficits permanently decline, the government’s debt position improves. The net foreign liabilities (NFL) position also improves in the long run, because consumers with finite life horizons do not increase consumption one-for-one with a decline in the tax burden on future generations.66 The real interest rate falls to equilibrate the savings-investment balance, while investment rises to match a higher level of savings in the new equilibrium. The real exchange rate appreciates in the long run, because a lower stock of debt obligations requires a lower current account surplus to finance debt obligations.67 In contrast to the above, the short- and medium-term effects of fiscal policies differ in economies with fixed and flexible exchange rate regimes, largely reflecting differences in how these economies adjust to shocks.

55. For the NMS with flexible exchange rates, fiscal consolidation reduces domestic demand and inflation in the short run but over time inflation returns to the central bank target. The bank’s reaction function entails cuts in nominal interest rates in response to fiscal tightening. This lowers the impact of fiscal consolidation on domestic demand and inflation, albeit with a lag. The decline in nominal interest rates exceeds the decline in inflation and therefore the real interest rate falls, crowding in investment. The nominal exchange rate depreciates in line with the interest rate parity, further easing monetary conditions. Since the degree of nominal depreciation exceeds the decline in inflation, the real exchange rate also depreciates. The current account improves, which over time improves the NFL position. In the medium term, the central bank brings inflation back to target. How long inflation remains below the target depends on the profile of fiscal adjustment: a consolidation program with a constant annual decrease in the fiscal deficit will have a more lasting effect on inflation than a one-off temporary or permanent reduction in the fiscal deficit.68

NMS: Effects of a One-Percent Permanent Decrease in Fiscal Deficit

(In percent or percentage point deviation from the baseline)

Source: IMF staff estimates.

56. For the NMS with fixed exchange rates, a permanent decrease in the fiscal deficit has a larger impact on inflation in the short run, but a perverse impact over the medium term. Cuts in government expenditures and transfers and a tax hike crowd out domestic demand and reduce inflation in the short run. Inflation and real GDP fall by more than under the flexible exchange rate regime, because monetary conditions cannot be eased quickly through interest rate cuts and nominal exchange rate depreciation. Lower inflation pushes the real interest rate up, which crowds out investment. Private consumption declines owing to higher taxes. Lower domestic demand depresses imports, while exports strengthen as the real exchange rate depreciates. Eventually, the excess of savings over investment puts downward pressure on the real interest rate. As the real interest rate declines, domestic demand and inflation pick up in the medium term. Such a reversal of the short-term effect of fiscal consolidation on inflation is independent of the pattern of fiscal adjustment (whether it is one-off temporary, one-off permanent, or persistent permanent adjustment, i.e., a given decrease in the fiscal deficit per year). Greater wage and price flexibility and lower nominal rigidities imply faster price adjustment, reducing output volatility, but inflation volatility becomes more pronounced.

NMS with Fixed Exchange Rate Regimes: Effects of a One-Percent Permanent Decrease in Fiscal Deficit

(In percent or percentage point deviation from the baseline)

Source: IMF staff estimates.

57. The effects of a monetary policy shock in GIMF are predictable. The real interest rate rises and the exchange rate appreciates in response to an interest rate shock. Private investment and consumption weaken, leading to excess supply and disinflationary pressures. These effects unwind over time, and monetary policy neutrality holds in the long run.

NMS with Flexible Exchange Rates: Effects of a One Percent Interest Rate Increase

(In percent or percentage point deviation from the baseline)

Source: IMF staff estimates.

Lowering Inflation

58. When setting policy targets to fulfill the Maastricht criteria, the NMS need to build in a safety margin to accommodate unanticipated shocks. As emerging market economies, the NMS are susceptible to risk premium shocks to a larger extent than the OMS. The impact of oil price shocks on inflation in the NMS is likely to be stronger than in the OMS, because of the higher share of energy products in the consumer price baskets in the NMS. Monetary policy can offset the inflationary impact of the above shocks, but it would do so with a lag. Based on the historical volatility of oil prices, a moderately conservative strategy for satisfying the inflation criterion would call for setting the inflation target about 0.3 percentage points below the Maastricht reference value, which in the December 2006 Convergence Report was 2.8 percent. Such a margin would accommodate differences in the estimated impact on the NMS and the euro area of about 70 percent increase in oil prices as well as exchange rate shocks of up to 2¼ percent.

Safety Margins for Accomodating Oil Price and Exchange Rate Shocks
Coefficient 1/Shock 2/Safety Margin 3/
Oil price increase0.00468.60.3
Exchange rate depreciation0.102.30.2
Source: IMF staff estimates.

Difference between the estimated pass-through coefficients for a 1 percent increase in oil prices in the euro area and the NMS.

Two standard-deviation oil price shock, based on the historical data for 1999-2006. A depreciation shock consistent with the exchange rate stability criterion.

The impact of the shocks in question on inflation.

Source: IMF staff estimates.

Difference between the estimated pass-through coefficients for a 1 percent increase in oil prices in the euro area and the NMS.

Two standard-deviation oil price shock, based on the historical data for 1999-2006. A depreciation shock consistent with the exchange rate stability criterion.

The impact of the shocks in question on inflation.

59. Quantifying the degree of required policy adjustment and potential output losses is not straightforward. One question is which sacrifice ratios to use to assess output losses associated with disinflation. For the NMS with floating exchange rates (for example, the CEECs),69 the available estimates of the long-run sacrifice ratios for a 100 basis point permanent reduction in inflation among the NMS range from -½ percent of GDP to -4 percent of GDP.70 The sacrifice ratios tend to be higher in countries with lower policy credibility and weaker monetary transmission mechanisms. For the NMS with fixed exchange rate regimes (for example, the Baltics), long-run estimates of the sacrifice ratio are not available. Instead, GIMF simulations provide estimates of the short-run output losses from a decrease in fiscal deficits and show that the resulting disinflation is sustainable over the one-year horizon.71 These short-run losses might be offset by the long-run gains from fiscal consolidation, to the extent that adjustment allows crowding in private investment and consumption by lowering public debt obligations and distortionary labor and consumption taxes and does not involve cuts in productive public investment compementary to private investment. Given the strong fiscal positions in the Baltics, the long-run gains associated with further fiscal consolidation are likely to be limited, we focus our attention on the short-run losses.

60. Another question is which levels of inflation to use. Since the objective of the analysis is simply to illustrate potential output losses, we abstract from forecasting inflation for the NMS and other EU countries and instead use a backward-looking approach to estimate potential output losses associated with disinflation. We use the reference inflation rate reported in the December 2006 Convergence Reports, adjusted for the safety margin (paragraph 58), as a reference value for the Maastricht criterion; the adjusted reference value comes to 2½ percent. For the starting value of inflation, we consider two approaches, aiming to estimate a range of potential losses, given uncertainties about the equilibrium rates of real exchange rate appreciation in the NMS (paragraph 40):

  • The first approach uses the actual values for inflation corresponding to December 2006. Output losses relating to disinflation would then range from 5½-6½ percent of GDP in Hungary and Latvia to 2-3 percent of GDP in Estonia and Lithuania to 1½ percent in the Slovak Republic.72 Significant fiscal retrenchment would be needed in the NMS with fixed exchange rate regimes to achieve such disinflation, for example, by about 5 percent of GDP in Latvia and 2⅓ percent of GDP in Estonia. This would bring budget surpluses in these countries to 5¼ percent of GDP and 6 percent of GDP, respectively. However, the adequacy of the starting point—inflation in 2006–is questionable. As discussed above, various temporary, NMS-specific factors might have boosted inflation in 2006, notably demand shocks related, for example, to interest rate convergence (falling risk premia), EU accession in 2004, and the prospects for EU transfers.

  • Accordingly, an alternative but still simple approach is to argue that only inflation reduction to below the longer-run equilibrium trend is costly for purposes of meeting the Maastricht criteria. Assuming this longer-run equilibrium trend is captured in all NMS only by the Balassa-Samuelson estimates that add 1½ percent to average OMS inflation, which is running just under 2 percent, this would refer to any inflation reduction below 3½ percent. Relative to the cautious 2½ percent reference value discussed above, the losses would thus relate to a 1 percentage point temporary reduction in inflation. Assuming the NMS manage to achieve the policy credibility and structural flexibility of the best performers among them, the related output loss would be in the range of ½–1½ percent of GDP.

These two groups of estimates provide some rough idea about upper and lower bounds of possible output losses. Which group of estimates is more relevant for a given country depends on country-specific circumstances, particularly the degree to which inflation and exchange rate developments in the country in question are sustainable and the extent to which the country authorities would be willing and able to strengthen their policy frameworks, including monetary policy credibility, the effectiveness of the monetary transmission mechanism and the structural flexibility of the economy.

Output Effects Associated with Disinflation 1/
Output Losses from a 1
Desired Degree ofPercentage PointOutput Losses Associated
Current InflationDisinflation 2/Decrease in Inflation 3/with Desired Disinflation
(In percent)(In percent)(In percent of GDP)(In percent of GDP)
Czech Republic2.60.1-0.50.0
Slovak Republic4.41.9-0.8-1.5
Sources: ECB (2006), Bulř and Hurník (2006), IMF staff estimates.

The table illustrates the output losses associated with disinflation. Actual inflation data and the reference values for the Maastricht criterion are taken from the December 2006 Convergence Report of the ECB.

Actual inflation and the Maastricht reference value (as of December 2006) minus the ¼ percent safety margin.

Data for the CEECs are estimates of full-horizon sacrifice ratios from Bulř and Hurník (2006). Data for the Baltics are 4-quarter output losses from a decrease in the fiscal deficit needed to achieve a 1 percentage point decrease in inflation, based on GIMF. Disinflation of 1 percentage point is sustainable for 4 quarters.

Sources: ECB (2006), Bulř and Hurník (2006), IMF staff estimates.

The table illustrates the output losses associated with disinflation. Actual inflation data and the reference values for the Maastricht criterion are taken from the December 2006 Convergence Report of the ECB.

Actual inflation and the Maastricht reference value (as of December 2006) minus the ¼ percent safety margin.

Data for the CEECs are estimates of full-horizon sacrifice ratios from Bulř and Hurník (2006). Data for the Baltics are 4-quarter output losses from a decrease in the fiscal deficit needed to achieve a 1 percentage point decrease in inflation, based on GIMF. Disinflation of 1 percentage point is sustainable for 4 quarters.

Maintaining Exchange Rate Stability

61. The potential for inconsistency between inflation and exchange rate objectives in ERM-II makes meeting the exchange rate stability criterion challenging for the NMS with floating exchange rate regimes.73 Inflation targets in such NMS would have to be set relatively low (the Maastricht reference value, as of December 2006, minus the safety margin renders 2½ percent). Any equilibrium real appreciation would likely have to take place through the nominal exchange rate. Assuming that equilibrium real appreciation is limited to the Balassa-Samuelson effects (estimated at 1½ percent on average), the 3 percent cumulative appreciation during the two years that the NMS would spend in the ERM-II can be easily accommodated within the 15 percent appreciation part of the ERM-II band. However, the remaining room for appreciation (6 percent per annum) might not be sufficient if other equilibrium factors are at play. Interest rates can be reduced in this case, but this might put the fulfillment of the inflation criterion at risk. Simulations suggest that a 1 percentage point increase in interest rates can lower inflation by ½ percent, with the exchange rate appreciating by 1½ percent. Fiscal consolidation would help resolve the tensions between the inflation criterion and the exchange rate stability criterion. A 1 percent of GDP permanent decrease in the fiscal deficit would reduce inflation by about ⅓ percentage points, according to GIMF simulations, and the exchange rate would depreciate by about 2 percent on impact—an effect that would gradually unwind. The potential tensions involved in simultaneously satisfying the exchange rate and inflation criteria argue for minimizing the time the NMS spend in the ERM-II.

Reining in Fiscal Deficits

62. The NMS also need to build in safety margins in their fiscal targets relative to Maastricht criteria, given the historical volatility of their GDP and budget elasticities. The variability of output is greater in the NMS—emerging market economies—than in the OMS. A cyclical slowdown or an adverse external demand shock, for example, can widen budget deficits, putting the fulfillment of the fiscal criterion at risk. To mitigate this risk, the NMS would need to set fiscal targets below the Maastricht and SGP limit of 3 percent of GDP. Such a buffer would serve the NMS well in the euro area too, as it would create room for the operation of automatic stabilizers or discretionary policy in the event of a cyclical downturn. Going below the Maastricht limit might be necessary to insure against the risk that a negative shock shortly before euro adoption would raise the deficit above the limit and require a procyclical tightening of fiscal policies to bring the deficit below the limit. Based on the historical volatility of GDP and budget elasticities, the necessary margins are estimated at about 1-2 percent of GDP. Clearly the Maastricht 3 percent fiscal deficit limit therefore does not appear to be a prudent target for the NMS.74 Going beyond the Maastricht requirements may also help contain inflation.

63. As for the effects of a decrease in the fiscal deficit on output, there is a trade-off between short-run and long-run effects. In the short run, fiscal consolidation reduces GDP, while in the long run the effects are positive owing to the improvement in the government’s debt position, and the long-run effect tends to outweigh the short-run effect in present value terms. The extent of the fiscal adjustment needed to reach prudent fiscal targets (i.e., the Maastricht 3 percent of GDP ceiling minus the safety margin) varies across countries. Using fiscal positions described in the December 2006 Convergence Report as an example, the magnitude of the needed fiscal adjustment would be 9 percent of GDP in Hungary and about 2 percent in the Czech Republic, Poland and the Slovak Republic. The Baltic countries are already meeting their prudent medium-term targets.

Output Effects Associated with Achieving Prudent Fiscal Targets(In percent of GDP)
PrudentDifference fromShort-Run OutputMedium-Term Output
FiscalFiscalthe PrudentLosses from FiscalGains from Fiscal
Balance 1/Balance 2/BalanceConsolidation 3/Consolidation 3/
Czech Republic-3.5-1.2-2.3-1.20.8
Poland 4/-3.9-1.9-2.0-1.00.7
Slovak Republic-3.4-1.6-1.8-0.90.6
Source: ECB (2006), Schadler and others (2005), and IMF staff estimates.

Data refer to 2006.

Based on GIMF simulations, a 1 percent decrease in the fiscal deficit results in a ½ percent decrease in the level of GDP in 1 year and a ⅓ percent increase in the level of GDP over the 5 year horizon.

The 3 percent of GDP Maastricht and SGP headline deficit ceiling minus the prudent fiscal buffer estimated in Schadler and others (2005).

With the second pillar pension pillar classified outside government.

Source: ECB (2006), Schadler and others (2005), and IMF staff estimates.

Data refer to 2006.

Based on GIMF simulations, a 1 percent decrease in the fiscal deficit results in a ½ percent decrease in the level of GDP in 1 year and a ⅓ percent increase in the level of GDP over the 5 year horizon.

The 3 percent of GDP Maastricht and SGP headline deficit ceiling minus the prudent fiscal buffer estimated in Schadler and others (2005).

With the second pillar pension pillar classified outside government.

F. Conclusion

64. The NMS face considerable macroeconomic policy challenges as they prepare for participation in monetary union. These challenges relate to their lower per capita incomes and financial depth than those in the OMS and the ensuing pace of convergence. As a result, the NMS are experiencing higher output and productivity growth than the OMS and rapid credit growth, supported by large capital inflows. This has led to significant appreciation of the NMS real effective exchange rates relative to those of their OMS partner countries, via appreciation of nominal exchange rates or higher inflation. Some part of these real appreciations is clearly of an equilbrium character. One element is the so-called Balassa-Samuelson effect. Estimates of this effect vary, but average around 1½ percent per year. Other long-lasting equilibrium mechanisms may also be at play, for example, EU transfers, which are projected to be sizable in the coming years, are also likely to contribute to real appreciation, and improvements in the quality of services in tandem with rising demand for them might also command higher prices. Other mechanisms prompting real appreciations may be less lasting, including those associated with the adoption of the EU acquis and the confluence of rapid financial integration, unusually benign global financial conditions, and structural change in the investor base for emerging market countries. These may have fostered stock-adjustment-type economic booms that will fade appreciably sooner than longer-lasting forces. These booms may well be fully justified, given the economic potential of the NMS but they may also go further than fundamentals warrant. Distinguishing benign appreciation trends in exchange rates and prices from overheating trends is thus difficult in the NMS.

65. Against this backdrop, the degree of macroeconomic policy adjustment the NMS would need to undertake to join monetary union is generally uncertain and depends on country-specific circumstances. The analysis in this paper is based on a dynamic stochastic general equilibrium model that allows for a joint analysis of monetary and fiscal policies. The crucial issue is the extent to which today’s high real appreciation rates reflect longer-rather than shorter-run forces. If they are largely driven by the former, simulations show that appreciable short-run output losses might be needed for the purpose of meeting the Maastricht criteria. To the extent that recent real appreciation trends largely reflect the latter forces, transient in nature, or irrational exuberance, short-run output losses may be more manageable, because real appreciation rates are likely to slow over time or require policy responses regardless of the Maastricht criteria. In either case, fiscal adjustment is not only needed but would also be beneficial, especially in the NMS that are still far away from their “prudent” medium-term fiscal targets. On current estimates of output volatility and budget elasticities in the NMS, such prudent targets are significantly below the Maastricht fiscal deficit limit, and would strengthen performance in the medium to long term.

66. A key challenge for the NMS and euro-area authorities is to ensure that the NMS enter in positions that allow them to perform well in monetary union. The long-standing Maastricht criteria play a useful role as simple and transparent anchors for macroeconomic policies on the way to monetary union. A pursuit of consistent macroeconomic policies, further improvements in the structural flexibility of the economy, and financial deepening are essential for the NMS to excel in EMU. Sound prudential policies, particularly supervision of foreign currency and real estate lending, are also needed to maintain the quality of credit and bank strength. Clear communication of the NMS’ euro-adoption prospects would help households, businesses, and financial markets make correct decisions and perhaps facilitate the unwinding of the currency mismatches accumulated on earlier, more optimistic expectations of euro adoption.

67. The benefits of euro adoption for both the NMS and the OMS are likely to be considerable. The NMS would benefit from the enhanced policy credibility that comes with the economic governance framework of EMU members as well as a further reduction in exchange rate-related risk. And both the NMS and the OMS would benefit from further expansion in trade and investment that comes with a common currency. This expansion in trade would inject greater competition, efficiency and growth in the EU and, via trade in intermediate goods, also help the OMS to optimize their production structures and improve their competitiveness in an increasingly competitive global economy.

Box II.1.Economic Effects of the NMS’ Joining the Euro Area

Long-Run Welfare Gains

Euro area membership is likely to boost growth by about 1 percent per year, speeding up convergence in living standards between the NMS and the OMS.1/ Gains to the euro area are likely to be smaller, given differences in the relative size of the euro area and the NMS economies—the NMS as a group account for 6 percent of the euro area’s GDP and 25 percent of population. Nonetheless, even for the euro area the gains are likely to be noticeable. They are likely to occur through trade, investment, and migration.

Trade. Empirical studies agree that a currency union is likely to boost trade over the long run, although the magnitude of this effect is believed to be smaller now than a few years ago.2/ The elimination of currency risk and currency transaction costs between the NMS and the euro area is likely to augment the already achieved gains from trade resulting from the elimination of trade barriers between the two regions, resulting in considerable trade creation. Judging by the boost to euro area exports and trade balances that the NMS’ accession to the EU provided, gains from trade with the NMS will be significant.

NMS are driving exports of euro-area countries.

Source: IMF, Direction of Trade Statistics and Fund staff calculations. Growth rates refer to exports of goods denominated in euros.

Investment. Lower currency risk is also likely to encourage further FDI in the NMS, promoting vertical and horizontal integration between firms in the two regions. FDI flows will facilitate technological and quality upgrading in the NMS. The euro area will share in these benefits, as outsourcing to lower-wage NMS is likely to improve the competitiveness of euro-area firms.

Migration. Increased migration flows between the NMS and the euro area are also likely to bring in efficiency gains, although the role of euro adoption in this is likely to be very limited. Recent studies for Ireland and the United Kingdom (major recipients of migrant flows from the NMS in recent years) suggest that these flows can boost GDP in euro-area countries by ½-1½ percent.3/

Implications for Macroeconomic Management

With euro adoption, the nominal exchange rate will no longer serve as a shock-absorber in the NMS; adjustment to asymmetric shocks would need to take place through prices and wages. To the extent that the latter are rigid, the volatility of output and inflation in the NMS after euro adoption is likely to increase. However, this cost of joining monetary union needs to be weighed against the benefit resulting from lower exchange rate risk. This would be important in the NMS that are susceptible to risk premium shocks (for example, owing to lower than euro-area policy credibility) or where significant currency mismatches have accumulated in recent years in the private sector balance sheets.4/ Another consideration is that increased trade and investment flows with the euro area are likely to speed up convergence of business cycles and production structures, reducing the incidence of asymmetric shocks.

1/ Schadler and others (2005).2/ Rose (2000), Faruqee (2004), and Baldwin (2006).3/ Barrell, Fitzgerald, and Riley (2007).4/ Karam and others (2007).

Box II.2.Financial Instability in the NMS: Costs and Spillovers1/

Empirical studies suggest that the cost of a financial crisis if it were to originate in a NMS are likely to be significant, slowing real convergence. De Gregorio and Lee (2004) estimate that twin currency and banking crises in emerging market countries on average result in the output loss of 16 percent of GDP, cumulative over 5 years of an average duration of the crisis. Hutchison and Noy (2005) put the average duration of crisis at about 3 years and output losses at 10 percent of GDP. The loss could be significantly higher, if the decline in output is permanent (Boyd and others, 2004).

A downturn in one NMS may affect other NMS and, to a lesser extent, the euro area through trade links. Exports to the NMS accounted for about 7 percent of total euro-area exports in 2005. For the NMS, the euro area is the most important trading partner. However, the importance of intra-NMS trade increased in recent years. For example, 32 percent of Latvia’s total imports came from the NMS in 2006 and 35 percent from the euro area. The respective shares were 22 percent and 39 percent in 2000.

A widening of the risk premium for one NMS may prompt investors to reassess sovereign risk of other NMS. For example, CDS spreads for Hungary and Poland are highly correlated in periods of a sharp increase in spreads. Investors may be also forced to sell assets of other NMS to finance margin calls or cash outflows from the country where the shock originated. The potential for capital outflows is significant: according to the IMF’s Coordinated Portfolio Investment Survey, the outstanding stock of portfolio investment was $17.4 billion in the Czech Republic, $8.8 billion in Poland, $3.0 billion in Estonia, and $3.4 billion in the Slovak Republic in 2005.

Source: IMF’s Coordinated Portfolio Investment Survey, BIS, and IMF staff estimates.

Cross-border financial linkages are another important channel for spillovers between the NMS and the euro area. Foreign banks, from the euro area and Sweden, account for about 80 percent of total banking sector assets in the NMS. Foreign banks have played a prominent role in the rapid expansion of foreign currency-denominated credit in the NMS—through lending to NMS subsidiaries, leasing companies, and corporates. The stock of outstanding cross-border claims on the NMS is large.

Most of these claims are owned by a few banks from a few countries—common lenders to the NMS. The banks that are most active in lending to the NMS are from Austria, Belgium, France, Germany, Italy, the Netherlands and Sweden. Although assets in the NMS represent a small fraction of these banks’ total balance sheets, the share in profits is higher. For example, for Austrian banks, activities in the CEEC countries accounted for about 16 percent of consolidated total assets in 2005, while profits from these activities reached 35 percent of pre-tax profits.2/ Changes in earnings outlooks of the banks active in the NMS are likely to affect equity prices for these banks, and through interbank linkages this might have implications for other European banks.3/ The fact that some banks are common lenders to several NMS points to the scope for transmission of shocks across the NMS.

1/ Prepared by Natalia Tamirisa, with inputs from Gavin Gray and Andy Jobst (MCM).2/ Hilbers (2007).3/ Gropp, Lo Duca, and Vesala (2006) find evidence of significant cross-border contagion among European banks.

Box II.3.Cross-Border Bank Contagion between the NMS and the Euro Area1/

A measure of joint asymptotic tail dependence that underlies bivariate extreme value distribution (EVD) functions is used to assess the potential for financial contagion between European banks active in the NMS and a broader aggregate of European banks. Conventional Pearson’s linear correlation coefficients characterize variable dependence as a central tendency and are thus diluted in the presence of skewed distributions. We consider an alternative measure of bivariate dependence—the Chi(q) measure plot of quantile-based dependence between two i.i.d. random vectors—defined as χ(q)=2log(pr(FX(X)<q,FY(Y)<q))/log(q), where FX and FY are the marginal distribution functions, quantile q Є [0,1].2/ The statistic measures dependence at extreme values of the distribution.

The analysis suggests that markets are likely to believe that problems in the European banks active in the NMS will get transmitted to other European financial firms. We find a precipitous increase in quantile-based dependence between equity prices of European financial firms and the banks exposed to either the Baltics or central and eastern Europe at low percentiles of individual equity return distributions. This indicates joint asymptotic tail behavior in line with extreme value dependence during periods of large declines in equity prices. Although large equity price movements are likely to reflect common shocks to the European banking sector rather than shocks specific to the NMS, high extreme-value dependence between the NMS-active banks and other European banks could point to little investor differentiation among these two groups of banks in times of stress. The high share of index-based funds investing in European banking stocks might explain this result.3/

Extreme-Value Dependence between Equity Prices of Banks Active in the Baltics and Central and Eastern Europe and European Financial Firms Overall

Source: Bloomberg, Bankscope, annual reports of individual banks, and IMF staff estimates.

Note: The chart shows the Chi square measure plot on daily log returns of MSCI Equity Index for European Financials and a composite equity price index of selected European banks with the largest asset exposures to the Baltics and the CEECs (from 11/1/2000-4/13/2007; 1,684 obs.). The confidence interval assumes independent observations at a statistical power of 95 percent. The plot is estimated on 1,000 intervals using the delta method, which may lead to poor interval estimates around q=0 and q=1.

1/ Prepared by Andy Jobst (MCM), with inputs from Gavin Gray and Natalia Tamirisa.2/ Coles, Heffernan, and Tawn (1999).3/ Sáez, Fratzscher and Thimann (2007) also find that shocks to equity prices of banks in Emerging Europe (the Czech Republic, Poland, Russia, and Turkey) are transmitted to the European financial sector.

Box II.4.Non-Ricardian Features in GIMF

The non-Ricardian properties of GIMF result from the following assumptions. First, the OLG consumers have finite planning horizons and are therefore myopic with respect to future tax liabilities. Second, the life-cycle income patterns of consumers feature declining life-cycle productivity and labor income profiles. Third, some consumers are liquidity-constrained because they lack access to financial markets and as a result they change their consumption one-for-one with their after-tax labor income. This feature is particularly relevant for the NMS economies with underdeveloped financial markets. Lastly, labor and consumption taxes are distortionary, because labor effort and consumption depend on relative prices, which in turn depend on tax wedges.

Several other elements of the model enrich the dynamic response to fiscal policy shocks. Preferences and technologies are defined broadly enough, highlighting the role of the intertemporal elasticity of substitution in the propagation of fiscal shocks. A multi-sector production structure includes traded and nontraded goods manufacturers, import agents, distributors, retailers and trade unions. Labor supply decisions and capital formation are endogenous. The latter serves as an additional channel through which government debt can crowd out economic activity. The model also allows for productive government investment in infrastructure, which makes it particularly relevant for the analysis of fiscal policy effects in the catching up NMS with considerable infrastructure needs.

Nominal and real rigidities in the model are typical of monetary business cyclical models. Nominal rigidities include producer currency pricing for manufactured goods and final goods, local currency pricing for imported goods, and sticky wages. Real rigidities include habit persistence in consumption, adjustment costs in investment, imports and retail trade, as well as international trade frictions.

The non-Ricardian features and rigidities in the model allow for monetary and fiscal policies to be examined jointly. Monetary policy is based on an interest rate rule, similar to the rules suggested by Orphanides (2003), which aims to stabilize inflation and output growth, while smoothing interest rates. The main difference, reflecting the non-Ricardian features of the model, is that the steady-state real interest rate is not constant but depends on fiscal policy. Fiscal policy aims to stabilize government debt-to-GDP ratio by adjusting labor and capital tax rates or by reducing government spending for consumption and investment purposes. The policy is implemented by a fiscal rule where the primary surplus adjusts to deviations of the government balance ratio from a desired value. Interactions between monetary and fiscal policy determines the short-run dynamics of the model, while the medium- and long-run dynamics depend on the fiscal rule. For a more detailed description of the model, see Kumhof and Laxton (2007).


    AhearneA. and J.Pisani-Ferry2006The Euro: Only for the Agile,Bruegel Policy BriefIssue 2006/1February.

    AfonsoA.C.Nickel andP.Rother2006Fiscal Consolidations in the Central and Eastern European Countries,Review of World Economic Studies 142(3) 402–421.

    AngeloniI.M.Flad andF. P.Mongelli2005Economic and Monetary Integration of the New Member States: Helping to Chart the Route,ECB Occasional Paper No. 36 (Frankfurt: European Central Bank).

    BackeP.C.ThimannO.ArratibelO.Calvo-GonzalezA.Mehl andC.Nerlich2004The Acceding Countries’Strategies Towards ERMII and the Adoption of the Euro:An European Central Bank).

    BalassaB.1964The Purchasing Power Doctrine: A Reappraisal,Journal of Political Economy72585596.

    BaldwinR.2006The Euro’s Trade Effect,Working Paper No. 594(March) (Frankfurt: European Central Bank).

    BarrellR.J.Fitzgerald andR.Riley2007EU Enlargement and Migration: Assessing the Macroeconomic Impacts,mimeo.

    Bini-SmaghiL. andP.Del Giovane1996Convergence of Inflation: A Necessary Prerequisite for EMU?Open Economies Review7 pp. 117126.

    BlanchardO.2006Adjustment within the Euro. The Difficult Case of Portugal,mimeo.

    BlanchardO. andF.Giavazzi2002Current Account Deficits in the Euro Area: the End of the Feldstein-Horioka Puzzle,Brookings Papers on Economic Activity2 147186.

    BoeriT. andP.Garibaldi2006Are Labour Markets in the New Member States Sufficiently Flexible for EMU?Journal of Banking and Finance30 pp. 13931407.

    BoydJ.S.Y.Kwak andB.SmithThe Real Output Losses Associated with Modern Banking Crises,Journal of Money Credit and Banking Volume 37 Number 6December2005 pp. 977999.

    BöwerU. andC.Guillemineau2006Determinants of Business Cycle Synchronization Across Euro Area Countries,ECB Working Paper No. 587 (Frankfurt: European Central Bank).

    BruhaJ. andJ.Podpiera2007Transition Economy Convergence in a Two-Country Model: Implications for Monetary Integration,ECB Working Paper No. 740 (Frankfurt: European Central Bank).

    Brzoza-BrzezinaMichał2005Lending Booms in the New EU Member States: Will Euro Adoption Matter,ECB Working Paper No. 543 (Frankfurt: European Central Bank).

    BuiterW. andC.Grafe2002Anchor, Float or Abandon Ship: Exchange Rate Regimes for Accession Countries,CEPR Discussion Paper No. 3184 (London: Center for Economic Policy Research).

    BuiterW.G.Corsetti andN.Roubini1993Excessive Deficits: Sense and Nonsense of the Treaty of Maastricht,Economic Policy 85890.

    BuiterW. andA.Siebert2006Eurozone Entry of New EU Member States from Central Europe: Should They? Could They?Development & Transition UNDP-LSE Newsletter 4June pp. 16 -19.

    BulířA. andJ.Hurník2006The Maastricht Inflation Criterion: How Unpleasant Is Purgatory?Working Paper No. 06/154 (Washington, D.C.: International Monetary Fund).

    CappielloL.B.GerardA.Kadareja andS.Manganelli2006Financial Integration of New EU Member States,Working Paper No. 683 (Frankfurt: European Central Bank).

    ChoueiriN.F.Ohnsorge andR.van Elkan2007Inflation Differentials in the EU: A Common (Factors) Approach with Implications for EU8 Euro Adoption Prospects,mimeo (Washington, D.C.: International Monetary Fund).

    CoffinetJ.J.Matheron andC.Poilly2007Une Évaluation Structurelle Du Ratio De Sacrifice Dans La Zone Euro,Notes D’Études Et De RechercheBanque de France.

    ColesS.G.HeffernanJ. andJ. A.Tawn1999Dependence Measures for Extreme Value Analyzes,Extremes Vol. 2 pp. 339365.

    CoricelliF andBJazbec(2001):Real Exchange Rate Dynamics in Transition Economies,CEPR Discussion Paper No. 2869 (London: Center for Economic Policy Research).

    CrowleyP.1996EMU, Maastricht, and the 1996 Intergovernmental Conference,Contemporary Economic Policy Vol. XIV pp. 4155.

    De BroeckM. andT.Sløk(2001):Interpreting Real Exchange Rate Movements in Transition Countries,IMF Working Paper 01/56 (Washington, D.C.: International Monetary Fund).

    De Grauwe2005Economics of Monetary Union” (Oxford: Oxford University Press).

    De GregorioJ. andJ.-W.Lee2004Growth and Adjustment in East Asia and Latin America,Economía Volume 5Number 1Fall 2004 pp. 69134.

    Delors Report1989Report of the Committee for the Study of Economic and Monetary Union (Luxembourg: Office for Official Publications of the European Communities).

    DornbuschR.1990Two-Track EMU, Now!,in Pohl et al. Britain and EMU (London: Centre for Economic Performance.”

    ÉgertB.L.Halpern andR.McDonald (2006) Equilibrium Exchange Rates in Transition Economies: Taking Stock of the Issues,Journal of Economic Surveys 20(2) pp. 257324.

    EichengreenB.1992Should the Maastricht Treaty Be Saved?Princeton Studies in International Finance No. 74Princeton University.

    EichengreenB.1993European Monetary Unification,Journal of Economic Literature Vol. XXXI (September) pp. 13211357.

    EickmeierS. andJ.Breitung2005How Synchronized Are Central and East European Economies with the Euro Area? Evidence from a Structural Factor Model,Discussion Paper Series 1: Economic Studies No. 20 (Frankfurt: Deutsche Bundesbank).

    ÉgertB.2002Investigating the Balassa-Samuelson Hypothesis in Transition: Do We Understand What We See? A Panel StudyEconomics of Transition 10273309.

    ÉgertB.2007Price Level Convergence and Inflation Rates in Europe: East-West Divide,mimeo.

    European Central Bank2003Inflation Differentials in the Euro Area: Potential Causes andPolicy Implications (Frankfurt: European Central Bank).

    European Central Bank2006Convergence Report,December (Frankfurt: European Central Bank).

    European Commission2005The Framework for Fiscal Policy in EMU: What Future After Five Years of Experience,European Commission Economic Paper No. 223March.

    FabrizioS.D.Igan andA.Mody2007The Dynamics of Product Quality and International Competitiveness,IMF Working Paper 07/97 (Washington: International Monetary Fund).

    FaganG. andV.Gaspar2007Adjusting to the Euro,ECB Working Paper No. 716 (Frankfurt: European Central Bank).

    FaruqeeH.2004Measuring the Trade Effects of EMU,Working Paper No. 04/154 (Washington, DC: International Monetary Fund).

    FiláčekJ.R.Horváth andM.Skorepa2006Monetary Policy Before Euro Adoption: Challenges for EU New Member States,Working Paper No. 853NovemberThe William Davidson Institute at the University of Michigan.

    FischerC.2002Real Currency Appreciation in Accession Countries: Balassa-Samuelson and Investment Demand,Deutsche Bundesbank Discussion Paper No. 19/02 (Frankfurt: Deutsche Bundesbank).

    FrankelJ. andA.Rose1998The Endogeneity of the Optimum Currency Area Criteria,Economic Journal 108July.

    FrankelJ. andA.Rose2000Estimating the Effect of Currency Unions on Trade and Output,CEPR Discussion Paper 263 (London: Centre for Economic Policy Research).

    GiovanniniA.R.Cooper andR.Hall1990European Monetary Reform: Progress and Prospects,Brookings Papers on Economic Activity Vol. 1990 No. 2 pp. 217291.

    GirouardN. andC.André2005Measuring Cyclically-Adjusted Budget Balances for OECD Countries,OECD Economics Department Working Papers No. 434 (OECD Publishing).

    GroppR.M.Lo Duca andJ.Vesala2006Cross-Border Bank Contagion in Europe,Working Paper No. 662 (Frankfurt: European Central Bank).

    GrosD.2004The Maastricht Criteria after Enlargement: Old Rules for New Members?” a paper prepared for the XVI Mongragone International Economic Seminar on “Rules,International Economy and Growth”CEIS—University of Rome “Tor Vergata,”June2324.

    HalpernL. andC.Wyplosz (2001):Economic Transformation and Real Exchange Rates in the 2000s: the Balassa-Samuelson Connection,Economic Survey of Europe No 1 (Geneva: United Nations Economic Commission for Europe).

    HilbersP.2007Austria: Managing Eastward Expansion of Its Banks,IMF SurveyJune 182007 Vol. 36 No. 10 pp. 1567.

    HutchisonM. andI.Noy2005How Bad Are Twins? Output Costs of Currency and Banking Crises,Journal of Money Credit and BankingOhio State University Press, Vol. 37(4)pages 72552August.

    International Monetary Fund2004Austria—Financial System Stability Assessment,IMF Country Report No. 04/238 (Washington: International Monetary Fund).

    International Monetary Fund2006Republic of Poland: 2006 Article IV Consultation—StaffReport; Staff Supplement; Public Information Notice on the Executive BoardDiscussion; and Statement by the Executive Director for the Republic of Poland,IMF Country Report No. 06/391 (Washington: International Monetary Fund).

    KaramP.D.LaxtonD.Rose andN.Tamirisa2007Macroeconomic Benefits and Costs of Euro Adoption,a paper prepared for the conference on “Open Economy Models for Policy Evaluation”Washington, DC,April1821.

    KaufmanH.1996European Economic and Monetary Union: Prospects and Pitfalls, Is EMU Premature?Paper presented at the Third World-ECSA Conference in BrusselsSeptember1920.

    KenenP.1992EMU After Maastricht,Group of ThirtyWashington, DC.

    KumhofM. andD.Laxton2007A Party Without a Hangover? On the Effects of U.S. Deficits,unpublished (Washington, DC: International Monetary Fund).

    LamfalussyA.1989Macro-Coordination of Fiscal Policies in an Economic and Monetary Union in Europe,in Collection of Papers Submitted to the Committee for the Study of Economic and Monetary Union (Luxembourg: Office for Official Publications of the European Communities).

    LanePhilipR. andGian Maria Milesi-Ferretti2006The External Wealth of Nations Mark II: Revised and Extended Estimates of Foreign Assets and Liabilities, 1970- 2004,IMF Working Paper No. 06/6 (also forthcoming in the Journal of International Economics).

    LaxtonD. andP.Pesenti2003Monetary Policy Rules for Small, Open, Emerging Economies,Journal of Monetary Economics Vol. 50pp.11091146.

    LipschitzL.T.Lane andA.Mourmouras2002Capital Flows to Transition Economies: Master or Servant?IMF Working Paper 02/11 (Washington: International Monetary Fund).

    LuengnaruemitchaiP. andS.Schadler2007Do Economists and Financial Markets’ Perspectives on the New Members of the EU Differ?Working Paper No. WP/07/65 (Washington, DC: International Monetary Fund).

    MihaljekD. andM.Klau2003The Balassa-Samuelson Effect in Central Europe: A Disaggregated Analysis,BIS Working Papers No. 143.

    McKinnonR.1963Optimum Currency Areas,American Economic Review 5371725.

    MendelR.1961A Theory of Optimal Currency Areas,American Economic Review Vol. 51 pp. 657665.

    Organization for Economic Cooperation and Development2007Economic Survey of Euro Area” (Paris: Organization for Economic Cooperation and Development).

    RoseA.2000One Money, One Market? The Effect of Common Currencies on International Trade,Economic Policy Vol. 30 pp. 745.

    SáezL.M.Fratzscher andC.Thimann2007The Transmission of Emerging Market Shocks to Global Equity Markets,ECB Working Paper No. 724 (Frankfurt: European Central Bank).

    SamuelsonP.1964Theoretical Notes on Trade Problems,Review of Economics andStatistics 46145154.

    SchadlerS.P.DrummondL.KaisZ.Murgasova andR.van Elkan2005Adopting the Euro in Central Europe: Challenges of the Next Step in European Integration,Occasional Paper 234 (Washington, D.C.: International Monetary Fund).

    SzapáryG.2001Maastricht and the Choice of Exchange Rate Regime in Transition Countries during the Run-up to EMU,ENEPRI Working Paper No. 6 (Brussels: European Network of Economic Policy Research Institutes/Center for European Policy Studies).

    TamirisaN. andD.Igan2006Credit Growth and Bank Soundness in the New Member States,in Czech Republic Republic of Estonia Hungary Republic of Latvia Republic of Lithuania Republic of Poland Slovak Republic and Republic of Slovenia—Country Report 06/414 (Washington, DC: International Monetary Fund).

    Von HagenJ. andI.Traistaru-Siedschlag2006Macroeconomic Adjustment in the New EU Member States,SUERF Studies SUERF - The European Money and Finance Forumnumber 2006/4edited byMorten BallingFebruary.

Prepared by Natalia Tamirisa with Douglas Laxton (RES), Andy Jobst (MCM), Gavin Gray, Thomas Harjes, and Emil Stavrev (all EUR).

The criteria are set out in Article 121 of the Treaty establishing the European Community and further detailed in a Protocol attached to the Treaty. The Convergence Reports prepared by the ECB and the EC describe how the criteria are applied in specific country cases.

For the purposes of this paper, the OMS comprise Austria, Belgium, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal, and Spain. Within this group, Greece, Ireland, Portugal, and Spain are considered catching-up economies. The NMS are defined as the central and eastern European (CEEC) countries in the 2004 wave of EU enlargement (the Czech Republic, Hungary, Poland, and the Slovak Republic), the Baltic countries in the same wave of enlargement (Estonia, Latvia, and Lithuania) and the two south-east European countries in the 2007 wave (Bulgaria and Romania). Slovenia, which adopted the euro in January 2007, and two island economies, Cyprus and Malta, which are soon to follow, are considered to be the former members of the NMS group. Per capita incomes in these three countries are higher than those in other NMS and/or economic growth is lower, and hence the analysis presented in this paper applies to them to a much lesser extent than to other NMS.

For extensive IMF staff analysis of euro adoption issues, see Schadler and others (2005).

Angeloni, Flad and Mongelli (2005). Frankel and Rose (2000) suggest, however, that a currency union can foster endogenous convergence, making countries less susceptible and more adaptable to asymmetric shocks.

Eickmeier and Breitung (2005).

Cappiello and others (2006).

Schadler and others (2005).

Lithuania’s application was rejected on the grounds that the country did not meet the inflation criterion (by 0.1 percentage points) and that, crucially, convergence in inflation rates was considered not sustainable.

Sweden does not have an opt-out but is not participating because of domestic political considerations.

The euro-area authorities indicated that the assessments are guided by the principle of equal treatment of the NMS vis-à-vis the OMS. (The informal Ecofin document adopted on April 5, 2003, as cited in Backé, Thimann and others, 2004).

The euro-area authorities faced a question in 2004 as to whether countries with negative inflation (Lithuania at that time) should be included in the calculation of the reference value for the inflation criterion. See Filáček, Horváth and Skorepa (2006) for a detailed review of how the Maastricht criteria were interpreted in past Convergence Reports.

Requirements concerning participation in the ERM-II and the exchange rate stability criterion jointly imply that the exchange rate is allowed to appreciate within the 15 percent of the ERM-II band but it cannot depreciate by more than 2¼ percent from the central parity (see Schadler and others, 2005, for more details). The ECB examination of exchange rate stability against the euro focuses on the exchange rate being close to the ERMII central rate, while also taking into account factors that may have led to exchange rate appreciation.

The fiscal criteria are supplemented by the definitions of the excessive deficits and debt, the “no-bail-out” clause (which makes countries responsible for servicing their own debt) and a ban on direct central bank financing and access to favorable financing of public deficits.

Szapary (2001) coined the term “the ‘weighing-in’ effect” to describe this strategy, comparing it to that of a boxer who refrains from eating and drinking before a weigh-in only to binge afterwards.

ECB (2006).

Balassa (1964) and Samuelson (1964).

For example, using empirical estimates from the aid literature, IMF staff put real appreciation that would result from the projected 2 percent of GDP increase in EU transfers to Poland during 2006-2011 at about 2 percent per year and 13 percent cumulatively over a five year period (see IMF, 2006; p. 10).

Lipschitz, Lane, and Mourmouras (2002). The IMF’s CGER exercise also assumes that the NMS currencies covered by the exercise were significantly undervalued at the beginning of transition.

Significant FDI the NMS received since the beginning of transition is likely to have contributed to improvements in the quality of the NMS tradable goods, triggering shifts in foreign consumer preferences toward the NMS goods and helping eliminate the undervaluation gap. No direct estimates of such quality-adjustment effects are available, but, in an indirect take on the issue, Fabrizio, Igan and Mody (2005) estimate that quality upgrading helps eliminate the negative impact of real appreciation on the market share in developing countries by almost a half: the impact of a 10 percent appreciation can be offset by a 22 percent improvement in quality (measured in terms of unit value ratios).

Empirical studies on the Balassa-Samuelson effects do not control for the afore-mentioned additional factors that might result in trend real appreciation. Furthermore, these studies are based on strong assumptions. They assume that the law of one price holds for tradable goods, which is not confirmed by the data—goods prices are 10 percent to 40 percent lower in the NMS than in the euro area (Égert, 2007), possibly owing to differences in taxes, competition practices, and pricing-to-market behavior. Price differentials for market and non-market services are even larger.

For example, during 1999-2006 four catching up euro-area countries posted an average inflation rate in the 3-3½ percent range; in 2006, these countries posted an average inflation rate of 3.2 percent. The Maastricht reference value for the purpose of the December 2006 Convergence Report was 2.8 percent. As Choueri, Ohnsorge, and van Elkan (2007) show, the reference value for the Maastricht inflation criterion might be unduly depressed because low inflation in the “three best performing EU economies in terms of price stability” might reflect idiosyncratic factors, for example, changes in regulated prices and indirect taxes or recession.

Luengnaruemitchai and Schadler (2007).

Brzoza-Brzezina (2005) estimates that post-euro adoption credit booms are likely to be smaller in the CEECs, because the degree of interest rate convergence the NMS have already achieved is greater than what the catching up OMS were able to achieve during the comparable period.

Discussions with market participants suggest that carry trades based on using the Swiss franc (and, to a lesser extent, the Czech koruna) as the funding currencies for investment in other NMS currencies supported capital flows to the NMS.

All restrictions on capital movements and payments, both between Member States and between Member States and third countries, are prohibited, although there are exceptions for crisis situations and related to taxation, prudential supervision, public policy considerations, money laundering, and financial sanctions agreed under the Common Foreign and Security Policy.

Although union density is similar in the NMS and the euro area, it does not imply the same degree of labor market rigidity. In contrast to the euro area, where the influence of trade unions goes beyond their membership, non-unionized workers in the NMS typically are not covered by wages negotiated under collective agreements.

Boeri and Garibaldi (2006). In addition, the authors argue that the view held in some quarters that the NMS have rigid labor markets is based on the fact that the job content of growth is low. The paper shows that the latter fact reflects productivity-enhancing job destruction in an environment of traditional labor hoarding.

For more details on specific reform measures, see staff reports for individual NMS, available at

See Lane and Milesi-Ferretti (2006) on the latter.

We leave aside here the long-standing debate on the appropriateness and effectiveness of discretionary fiscal policy in monetary union. See EC (2005).

Structural change, short data series and measurement problems in the NMS make the econometric estimation of policy effects, especially on the fiscal side, unreliable.

The empirical literature provides mixed evidence on Ricardian equivalence, in part owing to difficulties in controlling for macroeconomic interactions and feedback effects in an econometric setting.

For example, Obstfeld and Rogoff (1996), Caselli (2001), and Corsetti and Pesenti (2001).

We draw on the existing DSGE models for the euro area and the NMS, DSGE models for the U.S., and selected forecasting models of the NMS central banks. See, for example, Kumhof and Laxton (2007), Laxton and Pesenti (2003), and Karam and others (2007).

The latter might include, for example, public sector wages or social payments.

In an intertemporal optimization model where the Ricardian equivalence does hold, a permanent decrease in the fiscal deficit does not have any effect on the current account balance and the net foreign liability (NFL) position in the long run. When the short- and medium-term effects of fiscal policy unwind, the NFL ratio converges to its assumed long-run value. By focusing on the intertemporal and stock-flow implications of fiscal policy the GIMF goes well beyond the Mundell-Fleming model, which restricts the analysis to flows.

These effects are negligible for a small open economy.

The effects of fiscal consolidation obtained through simulations are broadly consistent with the estimates based on panel regression analysis of fiscal consolidation in the CEECs (Afonso, Nickel, and Rother, 2006).

The analysis below focuses on the CEECs and the Baltics because reliable estimates of the sacrifice ratios for Bulgaria and Romania are not available.

Bulíř and Hurník (2006). This compares to the sacrifice ratios ranging from -½ percent of GDP to -3½ percent of GDP in EU-15 countries (Bulíř and Hurník, 2006) and -1¼ percent of GDP in the eurozone (Coffinet, Matheron, and Poilly, 2007).

This would appear sufficient, given the interpretation of the sustainability principle in practice (paragraph 35).

In the Czech Republic and Poland, strengthening currencies helped bring inflation below the Maastricht reference values.

Satisfying the exchange rate stability criterion is trivial for the NMS with fixed exchange rate regimes.

Schadler and others (2005).

Other Resources Citing This Publication