THE BALTIC CHALLENGE AND EURO-AREA ENTRYaei.pitt.edu/12220/1/PC_BalticsZsolt_011209_01.pdf · • The Baltic states’ previous huge current account deficits turned to surpluses by

ISSUE 2009/13 NOVEMBER 2009 THE BALTIC

CHALLENGE ANDEURO-AREA ENTRY

ZSOLT DARVAS

Highlights

• The Baltic states’ previous huge current account deficits turned to surpluses by2009, but competitiveness improvements are still needed.

• The problems of the overvalued exchange rate and the large stock of foreigncurrency loans cannot be solved by any of the options (maintaining the peggedexchange rate, devaluation, introduction of a floating rate) available to the Baltics.

• The best option would be ‘immediate’ euro entry at a suitable exchange ratesupported by appropriate resolution to manage the debt overhang, but there is nolegal basis for this.

• It is in the broader European interest to prevent a collapse in the Baltics, and to helpmedium-term economic growth. Any solution should not be Baltic-specific andshould not incur a risk of moral hazard.

• There are serious concerns about the euro accession criteria. The EU treaty obligesthe Council to lay down the details of the convergence criteria and the excessivedeficit procedure. It is time to fulfil this task and to strengthen the economicrationale of the criteria.

This policy contribution was prepared for the conference ‘The Future of the Baltic Sea Region in Europe’, 27-28 August 2009 in Hamina, Finland, organised to commemorate the 200th anniversary of the Treaty ofHamina by Centrum Balticum and Town of Hamina together with the Finnish Prime Minister’s Office.

The author is grateful for their comments to Torbjörn Becker, Daumantas Lapinskas, Jean Pisani-Ferry,Indhira Santos, André Sapir, Karsten Staehr, György Szapáry, Vilija Tauraitė and Jakob von Weizsäcker, tothe Hamina conference participants and to the seminar participants at the German Marshall Fund in Berlin.The views expressed are the author's and do not necessarily reflect the views of persons mentioned. Maitede Sola, Kristina Morkunaite and Juan Ignacio Aldasoro provided excellent research assistance. Bruegelgratefully acknowledges the support of the German Marshall Fund of the United States to researchunderpinning this policy contribution.

Telephone+32 2 227 4210 [email protected]

www.bruegel.org

BRU EGE LPOLICYCONTRIBUTION

-20

-15

-10

-5

0

5

10

15

-20

-15

-10

-5

0

5

10

15

1996 1998 2000 2002 2004 2006 2008 2010

EstoniaLatviaLithuaniaEuro area

THE BALTIC CHALLENGE AND EURO-AREA ENTRY Zsolt Darvas

02

BR U EGE LPOLICYCONTRIBUTION

THE BALTIC CHALLENGE AND EURO-AREA ENTRYZSOLT DARVAS, NOVEMBER 2009

1. INTRODUCTION

The global economic and financial crisis has hitEstonia, Latvia and Lithuania hard (Figure 1).These countries became increasingly vulnerablebefore the crisis – for example, they had hugecredit, housing and consumption booms and thushigh current-account deficits and external debt. Itwas widely expected even before the crisis thatthese vulnerabilities would have to be correctedat some point, but the crisis amplified themagnitude of the correction. For the 2008-2010period, the three Baltic countries are projected toexperience the sharpest GDP contraction amongthe 182 countries of the world for which forecastsare available.

Figure 1: GDP growth, 1995-2010

Source: IMF, October 2009 World Economic Outlook.

The cornerstone of economic policy in the Balticcountries has been the maintenance of the fixedexchange-rate system. While many observersadvised and predicted devaluation, the threecountries have so far managed to survive undertheir fixed exchange-rate strategy and haveengaged in drastic budget expenditure cuts,including nominal wage cuts. Wages and priceshave also started to fall in the private sector, and

the previously huge current account imbalanceshave been turned into surpluses. While theadjustment in the economy has started, it ishaving a severe social impact, and questionmarks have remained about medium-termeconomic growth.

This paper aims to contribute to the analysis of thepolicy choices of the Baltic countries. To this end,we first discuss some aspects of the pre-crisiseconomic boom in the Baltic countries, describesome scenarios for future growth and consider theissue of the recent current account surpluses. Thisis followed by a discussion of policy options. Asthe best option is euro entry, we revisit the issueof euro-entry criteria, both from economic andlegal perspectives.

2. THE LEGACY OF THE PRE-CRISIS BOOM

In terms of the main macroeconomic conditionsthat prevailed during their pre-crisis economicgrowth phase, most central and eastern European(CEE) countries were different from otheremerging countries in Asia and Latin America.After the dramatic crises of the 1990s and aroundthe turn of the millennium, Asian and LatinAmerican countries fundamentally changed theireconomic models. From being net capitalimporters, they – especially in Asia – becamebalanced, or even substantial capital exporters1.Capital export from poorer to richer countries issometimes referred to as capital moving ‘uphill’.The CEE region was different: capital moved‘downhill’, mostly from rich EU15 countries topoorer CEE countries. The good economic growthprospects and the low level of physical capital, theprospect of eventual EU integration and therelated improvement in the business climate, thegenerally highly-educated labour force and lowlevel of wages, and the low level of domestic creditoffering the potential for substantial credit

1. See, for example, Abiad,Leigh and Mody (2009)

and Darvas andVeugelers (2009).

40

60

80

100

120

140

40 60 80 100 120 140

2007

BG

CZ

CY

EEHULV

LT

MTPLSK

SI

RO

AT

PT

FIIE

GR

ES

BE

DE

FR

ITNL

DK

SEUK

RUPr

ice

leve

l (EA

12=1

00)

GDP per capita at PPS (EA12 = 100)


03


expansion, were the main supply-side factors forcapital flows into CEE countries. These flows tookthe form of foreign direct investment (includingthe buying-up of swathes of the CEE bankingsystems), portfolio investments and loans.

There were demand-side factors as well, whichwere particularly strong in the three Balticcountries. As these countries were on aneconomic-growth path to catch up with the maineuro-area countries, their price levels were alsoset to increase compared to the euro area. Figure2 indicates that in countries with a higher level ofGDP per capita, the price level tends to be higheras well, an observation which has both empiricaland theoretical underpinnings. The relationship isnot clearly one-to-one and there is country-specific heterogeneity, but the relationship isapparent. Figure 3 breaks down this dynamiccorrespondence for each of the Baltic states.Figure 3 (and other figures to be presented later)also reports data for the other six Baltic Seacountries, because for various reasons theyprovide a convenient group of countries withwhich to compare the three Baltic states.

Figure 2: GDP per capita and price levels in the EUand Russia (euro area 12 = 100), 2007

Source: Bruegel’s calculations based on Eurostat and IMFdata. Note: Luxembourg is not shown due to its outlyingGDP/capita figure.

Price level increases compared to the main tradingpartners can occur through either higher inflationor nominal exchange-rate appreciation. Estoniahas had a fixed exchange rate against the euro(and the Deutsche mark before) since it becameindependent, and hence all price-levelconvergence toward euro-area price levels took

* Source for Figure 3:Bruegel’s calculation basedon data from Eurostat, theIMF's October 2009 WorldEconomic Outlook forecast,and ECB exchange-rate sta-tistics. Note: For bettervisual comparability acrosscountries the range of thevertical axis is 50 percent-age points in each panel.The relative price levels for2009 and 2010 were calcu-lated using the IMF WorldEconomic Outlook’s infla-tion forecast for the individ-ual countries and the euroarea and our assumptionson exchange rates againstthe euro. Euro exchangerates for 2009 and 2010are annual averages andwere calculated from actualexchange rates between 2January 2009 and 11 Sep-tember 2009 and anassumed constantexchange rate for the rest of2009 and 2010. Theassumed constantexchange rate is equal tothe average of actualexchange rates between 3August and 11 September2009.

Figure 3: GDP per capita and price level (euro area 12 = 100), 1995-2010*

30

40

50

60

70

30

40

50

60

70

96 98 00 02 04 06 08 10

30

40

50

60

70

30

40

50

60

70

96 98 00 02 04 06 08 10

30

40

50

60

70

30

40

50

60

70

96 98 00 02 04 06 08 10

20

30

40

50

60

70

20

30

40

50

60

70

96 98 00 02 04 06 08 1020

30

40

50

60

70

20

30

40

50

60

70

96 98 00 02 04 06 08 1090

100

110

120

130

140

90

100

110

120

130

140

96 98 00 02 04 06 08 10

90

100

110

120

130

140

90

100

110

120

130

140

96 98 00 02 04 06 08 1090

100

110

120

130

140

90

100

110

120

130

140

96 98 00 02 04 06 08 1090

100

110

120

130

140

90

100

110

120

130

140

96 98 00 02 04 06 08 10

Estonia Latvia Lithuania

Poland Russia Germany

Denmark Finland Sweden

GDP per capita (EA12 = 100) Price level (EA12 = 100)

-25

-20

-15

-10

-5

0

5

10

15

20

-25

-20

-15

-10

-5

0

5

10

15

20

00 01 02 03 04 05 06 07 08 09 10-4

0

4

8

12

16

20

24

-4

0

4

8

12

16

20

24

00 01 02 03 04 05 06 07 08 09 10

Current account/GDP InflationRussiaFinlandSwedenDenmarkGermanyLatviaPolandLithuaniaEstonia

90

100

110

120

130

140

150

160

170

2000 2002 2004 2006 200860

70

80

90

100

110

120

2000 2002 2004 2006 2008

90

100

110

120

130

140

150

160

170

60

70

80

90

100

110

120EstoniaLithuania

LatviaPoland

GermanyFinlandRussia

DenmarkSweden


04


the form of higher inflation. Latvia and Lithuania,on the other hand, had pegs against the SDR (IMFSpecial Drawing Rights) until 2004 and the USdollar until 2002. As a result of the movements ofthe euro against the dollar and SDR, Latvia andLithuania experienced nominal appreciation bothagainst the euro and in nominal effective termsfrom 1999 to 2002 (Figure 4). Nominalappreciation was especially marked in Lithuania(about 50 percent), which contributed toLithuania having the lowest inflation rate amongthe nine Baltic Sea countries in the first few yearsof the 2000s, and even experiencing deflation in2003 (Figure 5). However, once these countriespegged their currencies to the euro, price-levelconvergence with the euro area occurred throughhigher inflation.

The fixed exchange-rate regime enjoyed highcredibility in all three countries because itsurvived both the turbulent period of the firstyears of transition in the first half of the 1990sand the Russian crisis in 19982. The stability ofthe exchange rate created an incentive to borrowin foreign exchange, because the nominal interestrates of euro loans were somewhat below localcurrency loan rates. As inflation started to pick upafter 2004 (Figure 5) people and corporationsrecognised that their wages and incomes, as wellas prices, were rising much faster than thepercentage cost of the loan, ie real interest ratesbecame negative. The negative real rate pushedup demand for loans and amplified the boom.

Furthermore, rapid economic growth fuelled

2. Note however that theeffect of the Russian

crisis on these countriesis dwarfed by the

current crisis (seeFigure 1 and Figure 3).

Figure 4: Nominal effective exchange rates (4 Jan 1999=100), 4 Jan 1999 - 12 Oct 2009

Source: Bruegel’s calculation based on ECB, Datastream and Central Bank of Iceland data.Note: Increases in the index indicate nominal appreciation. The nominal effective exchange rate was calculated against 52trading partners. Weights were derived from average trade flows between 2000 and 2008.

Figure 5: Current account balance (% of GDP) and inflation (%), 2000-2010

Source: IMF - October 2009 World Economic Outlook.


05


expectations that high growth would continue,encouraging people to borrow against their futureincome, as depicted by intertemporal optimisationmodels.

The role of budget policy in amplifying/dampeningthe boom was different in the three countries.According to our previous calculations (Darvas,2009b) budget policy was highly pro-cyclical inLatvia before the crisis, somewhat pro-cyclical inLithuania, but counter-cyclical in Estonia.

Consequently, both supply and demand factorscontributed to the emergence of substantial creditbooms that fostered construction booms, housingbooms, and consumption booms3. Budget policyalso amplified the boom, especially in Latvia, butto a lesser extent in Lithuania. These factorsgradually overheated the economies and, in theyears immediately before the crisis, led to a sharprise in wages4, inflation and current accountdeficits (Figure 5). Although considerable foreigndirect investment (FDI) flowed into the Balticcountries, most of the current account deficit wasfinanced by borrowing from abroad. As aconsequence, massive external liabilities wereaccumulated (Figure 6), of which the bulk wasprivate sector external debt. As the examples ofPoland and a few other new EU countries indicate,such a development was not inherentlyunavoidable. As also shown in Figure 5, Polishexternal liabilities were much lower relative to GDP,and exhibited a more favourable composition,than in the Baltics5.

The risk inherent in private sector net externaldebt did not matter before the crisis but mattersnow for market-risk assessment. Figure 7indicates that the credit default swap (CDS) ongovernment bonds – which is a measure of thecost of insurance against government default – isnow related to countries' net external debt andloan liabilities, which – at least in the three Balticstates – are mainly made up of private sector debtheld in foreign currencies. Should the economicoutlook deteriorate further and/or the exchangerate collapse (eg Baltics), or fall further (egUkraine, Hungary), then even deeper economic

3. Ireland and Spainexperienced similardevelopments after euroentry in 1999 (though to amuch lesser extent than inthe Baltics), due to theirfast economic growth andinflation rate above theeuro-area average, as thelow euro-area interest rateboosted demand andcontributed to the pre-crisishousing and constructionbooms (see Ahearne,Delgado, and vonWeizsäcker, 2008).

4. Migration to westernEurope and the shortage ofqualified workers have alsocontributed to wageincreases.

5. Bulgaria has also had afixed exchange-ratesystem, and hasexperienced very similareconomic developments tothe Baltic countries beforethe crisis, yet Bulgaria hasbeen much less affected bythe crisis. It would beworthwhile studying indetail what lies behind this.

Figure 6: External assets and liabilities (% ofGDP), 2000 and 2007

Estonia

0

20

40

60

80

100

120

140

160

180

200

Assets 2000

Liabilities2000

Assets2007

Liabilities2007

Latvia

0

20

40

60

80

100

120

140

160

180

200

Assets 2000

Liabilities2000

Assets2007

Liabilities2007

Lithuania

0

20

40

60

80

100

120

Assets 2000

Liabilities2000

Assets2007

Liabilities2007

Financial derivativesReservesOther investments

Portfolio debt securitiesPortfolio equity securitiesDirect Investment

Poland

0

10

20

30

40

50

60

70

80

90

100

Assets 2000

Liabilities2000

Assets2007

Liabilities2007







Source: Bruegel based on IMF data: International FinancialStatistics. Note: ‘Other investments’ are mainly loans.

AT BEDK FRIT

NLPTES

SE GBDEGR

HRRU

TR

CZEEHU

LV LT

PLRO

SKSI

BG ATBE

DK FIFR

IT

NLPT

ES

SE

GB

DE

GR

HRRU

TR

CZ

EEHU

LV

LT

PL

RO

SK SI

BG

020

040

060

0

Cred

it de

faul

t dw

ap o

n fiv

e−ye

ar g

ov b

ond

−80 −60 −40 −20 0 20

Net foreign loan and debt liabilities over GDP 2007

EU15 2007 CEE 2007EU15 Sep-Oct 2009 CEE Sep-Oct 2009

A

B

1

2

3

4

Actual output

Potential output

Time

Outp

ut


06


crises could emerge, leading to morebankruptcies, unmanageable bank losses and thecomplete drying-up of foreign capital. Thesefactors may lead to a government default, despitethe low level of government debt (Darvas, 2009b).In countries where foreign banks are prevalent,burden-sharing is an issue. Plotting CDS againstgovernment debt does not indicate a relationship,suggesting that the current level of governmentdebt is not in itself a concern for fiscalsustainability in eastern Europe. The risk inherentin private sector debt matters more.

Figure 7: Cost of insurance against governmentdefault and countries' net external debt

Source: Bruegel’s calculation based on Datastream and IMFdata. Note: Net foreign loan and debt liabilities refer to thewhole economy. CEE refers to new EU member states (exceptCyprus and Malta), plus Croatia, Russia, and Turkey. EU-15refers to the 15 EU countries before 2004 (exceptLuxembourg and Ireland). The Sept/Oct 2009 CDS valueswere calculated from the average between 1 September and12 October 2009.

Deceleration of credit growth already startedbefore the global financial and economic crisis butit greatly amplified the correction and led to verysharp drop in GDP with all the associatedconsequences. The sharp fall in output opened awide gap between GDP per capita and price levels(relative in both cases to the euro area, see Figure3). Further, the exchange rates of some other maintrading partners depreciated, such as the Swedishkrona and the Polish zloty, which helped Swedenand Poland at a time of great contraction in exportdemand6. Indeed, Figure 4 shows that, in nominaleffective terms, the currencies of the three Baltic

countries appreciated by about five to ten percentin late 2008, which incidentally is similar to whathappened in Germany, Denmark and Finland.

3. POST-CRISIS ECONOMIC GROWTH SCENARIOS

Figure 8 provides a schematic picture of actualand potential output before and during the crisis,and offers some scenarios for the future.

The overheated economy, as discussed in theprevious section, has led to faster actual outputgrowth than pre-crisis potential growth, and hencethe actual output level is now greater thanpotential output (thus, point (1) is above point(2) in Figure 8).

Figure 8: Schematic depiction of actual andpotential output scenarios in the Baltic states

Source: Bruegel.

Cerra and Saxena (2008) have demonstrated thatcrises tend to generate a sizeable permanent lossin the level of output compared with the pre-crisistrend when all main country groups (ie advanced,emerging market and developing countries) aretaken into account. Their findings imply that thelevel of potential output in the Baltic countries islikely to have fallen during the recent crisis (frompoint (2) to point (3) in Figure 8).

Actual output should fall markedly between 2008and 2010 in the Baltic countries (from point (1) topoint (4) in Figure 8), while domestic demandshould contract even more sharply. The actual outputfall can be decomposed into three components:

6. The Polish zlotydepreciated by about 45

percent between thesummer of 2008 and

early 2009, but a properevaluation of this

movement implieslooking at what

happened before. Figure4 indicates that thezloty was at record

highs in the summer of2008 and the currentlevel of the exchange

rate at the time ofwriting this contribution

is quite close to theaverage of the past

decade. Figure 3suggests that the zloty

may have beenovervalued before the

crisis (note that the2008 price level wascalculated using the

average exchange rateof the full year, iewithout the huge

depreciation in thefourth quarter, the price

level would have beenmuch higher) and also

that the currentmagnitude of the real

depreciation wasbroadly reasonable,

because the price levelsand GDP per capita

(both relative to theeuro area) converged.


07


• Part of the contraction is a correction ofprevious excess demand (the differencebetween points (1) and (2) in Figure 8);

• Part of the contraction corresponds to the fallin potential output (the difference betweenpoints (2) and (3) in Figure 8);

• But it is also likely that actual output falls belowpotential GDP (the difference between points(3) and (4) in Figure 8), ie the output gapbecomes negative.

While theories of business cycle fluctuations mayneed to be reconsidered in the light of the findingsof Cerra and Saxena (2008), available theoriesoffer a propagation mechanism in which arecession can lead to negative output gaps.Empirical evidence also suggests that recessionstypically result in negative output gaps.Furthermore, financial frictions, such as a suddenstopping of capital inflows and a change in thelending behaviour of banks, are likely to haveconstrained the real economy. This is a furtherreason why the crisis is resulting in a negativeoutput gap in Baltic countries.

An inconvenient implication of the fall in output isthat price levels become too high compared toactual GDP per capita (Figure 3), but probably alsorelative to potential GDP per capita. The price levelbecame too high in the Baltic countries, bothrelative to its historic level, and whenbenchmarked against a global comparison. Inother words, the real exchange rates are highlyovervalued. While this observation holds for allthree countries, Latvia clearly stands out with itsGDP per capita standing at 42 percent of the euro-area average, and its price level at 69 percent ofthe euro-area average (both forecast for 2010, ashighlighted by Figure 3).

The future scenarios depicted in Figure 8 arehighly uncertain. A good outcome would beScenario A, ie a return to the previous potential

growth rate, starting from the reduced level. Underthis scenario, actual output would grow rapidly fora short period until it reaches the potential levelof output, and then would continue to grow at thesame rate as potential output.

Scenario B would be a no-growth scenario. Onecannot exclude the possibility of no potentialgrowth with a highly overvalued exchange rate.Still, actual output may increase in the short runas it converges to the potential level of output byclosing the negative output gap.

Obviously, many other scenarios can beenvisaged. The rate of potential growth for the pre-crisis period is not known (available estimates areso uncertain as to be practically useless) andhence, in the optimistic Scenario A, it is highlyuncertain if future potential growth will be equalto previous potential growth or will behigher/lower. Also, Scenario B may be replaced byscenarios in which potential output grows, butvery slowly, or potential output falls.

4. HOW TO EVALUATE RECENT CURRENT ACCOUNTSURPLUSES

Figure 5 showed that the previous high current-account deficits had already been adjusted by2009 and all three Baltic countries are expectedto have current-account surpluses in 2009 and2010. One may argue that current-accountsurpluses remove the need for further relativeprice adjustment (ie devaluation of the currencyor ‘internal devaluation’, which amounts todomestic wage and price falls). However, thisclaim crucially depends on the fundamentaldrivers behind this adjustment and itssustainability. We list three possible explanations.

1. Financing constraints. One possible explan-ation of the current-account adjustment is thesudden stop of capital inflows, ie the lack of

‘The sharp fall in output opened a wide gap between GDP per capita and price levels (relative

to the euro area) both taking a historical view considering the Baltic countries themselves

and taking a global comparison.’

‘The recent export performance of the Baltic states is at least not worse than in the other

Baltic Sea countries and other EU member states, which is to some extent encouraging.’


08


financing. In this case, domestic agents haveno choice but to cut imports, and theadjustment does not invalidate the need for asignificant relative price adjustment, ie theeconomy may be uncompetitive and anadjustment in relative prices would still berequired for economic recovery and forsustaining the adjustment in the currentaccount. A continued lack of financing of thecurrent account imbalance coupled withinsufficient adjustment in relative prices maynot allow the output gap to turn to zero, with therisk of a scenario where actual output is lowerthan in Scenario B in Figure 8.

2. Negative output gap. A negative output gap7

(eg idle capacity and unemployment) inducesdomestic agents to voluntarily postponeimports due to uncertainty about economicprospects. Again, such a case would not removethe need for relative price adjustment.Whenever the output gap moves toward zero,the current-account imbalance will emergeprovided that there is a source of financing.

3. Disappearance of unsustainable consumptionand investment booms. A third possibleexplanation is that Baltic countries did not havea competitiveness problem, even in the boomyears before the crisis, and the huge currentaccount deficit was merely the consequence ofan unsustainable consumption boom, fuelledalso by intertemporal optimisation (ie agentsexpecting higher future income and borrowingnow against that future income). Unsustainableinvestment booms (concerning investment ineg the real estate sector) may have alsocontributed to the build-up of pre-crisis current-account deficits. With the huge recession,consumption has adjusted partly because ofcloudier future growth expectations. Theunsustainable component of investment hasalso adjusted because of the emergence ofovercapacity in the construction sector, and

property price falls and their future outlook.Under these circumstances the adjustmentachieved so far in the current account is broadlysustainable and a return to more normal growthwithout excessive consumption and investmentbooms will lead to the emergence of ‘reasonable’current account balances.

The three explanations8 have different impli-cations for the need for further adjustment. If thecompetitiveness problem is serious, then asignificant adjustment in relative prices is needed,despite the recently-achieved current-accountsurplus. If competitiveness problems are minor,then relative price adjustment may not be needed,but the government budget would need to beadjusted (public sector wage cuts, expenditurecuts, revenue increases) to keep budgetexpenditures in line with new revenue realities.While all three explanations may have played arole (to different degrees in the three countries),it is difficult to identify one of them as dominant.

The recent export performance of the Baltic statesis at least not worse than in the other Baltic Seacountries and other EU member states, which isto some extent encouraging (Figure 9). Thebroadly similar export performance wasaccompanied by much sharper output falls in theBaltic countries. This implies that the sharp outputcontraction in the Baltic countries was attributablemostly to sectors producing for domestic sales,and export capacity may have not been affectedmore than in other countries. Looking at imports,all three Baltic countries indicate a sharper fallthan the other countries under consideration. Thefall in imports is consistent with all three possibleexplanations of the recent current-accountsurpluses described above.

However, there are at least three reasons thatsuggest that regaining competitiveness is crucialeven if the third explanation given above providesthe main explanation for the previous high current-

7. A negative output gapcan also be of course

the consequence offinancing constraints.

8. A small part of theimprovement in the

current account balanceis due to less

transferring of profit.


09


Figure 9: Volume of export and import of goods and services (2007Q1=100, seasonally adjusted),2005Q1-2009Q2

Figure 10: Unit labour costs in different sectors of the economy, 2000Q1=100, 2000Q1-2009Q1

70

80

90

100

110

120

05 06 07 0870

80

90

100

110

120

05 06 07 0870

80

90

100

110

120

05 06 07 08

50

60

70

80

90

100

110

120

130

05 06 07 0850

60

70

80

90

100

110

120

130

05 06 07 0850

60

70

80

90

100

110

120

130

05 06 07 08

EstoniaLatviaLithuaniaPolandRussiaGermany

EstoniaLatviaLithuaniaPolandRussiaGermany

DenmarkFinlandSwedenIrelandSpainPortugal

DenmarkFinlandSwedenIrelandSpainPortugal

BulgariaCzech Rep.HungaryRomaniaSlovakiaSlovenia

BulgariaCzech Rep.HungaryRomaniaSlovakiaSlovenia

Impo

rtEx

port

Sources: Eurostat and Federal State Statistics Service of the Russian Federation. Note: We used X-12 to seasonally adjustRussian and Bulgarian data, which was available only in unadjusted form.

320280

240

200

160

120

320280

240

200

160

120

2000 2002 2004 2006 2008

450

350300250

200

150

100

450

350300250

200

150

100

2000 2002 2004 2006 2008

240

200

160

120

80

240

200

160

120

802000 2002 2004 2006 2008

130

120

110

100

90

80

70

130

120

110

100

90

80

70

2000 2002 2004 2006 2008

200180

160

140

120

100

80

200180

160

140

120

100

802000 2002 2004 2006 2008

120

115

110

105

100

95

90

85

120

115

110

105

100

95

90

852000 2002 2004 2006 2008

140

130

120

110

100

140

130

120

110

100

2000 2002 2004 2006 2008

160

140

120

100

80

160

140

120

100

802000 2002 2004 2006 2008

140

130

120

110

100

90

80

140

130

120

110

100

90

802000 2002 2004 2006 2008


Poland Ireland Germany

Denmark Finland Sweden

Total economy Construction (F) Financial/business services (J_K)Manufacturing (D) Trade, transport and communication (G_I)

Source: OECD. Note: Data for the Lithuanian construction sector is not available. Data for Ireland ends in 2008Q4.

GDP per capita at PPP

15

20

25

30

35

40

45

50

55

t-3 t-2 t-1 t t+1 t+2 t+3 t+4 t+5

Price Level

35

40

45

50

55

60

65

70

t-3 t-2 t-1 t t+1 t+2 t+3 t+4 t+5


10


account deficits and the recent surpluses (thoughthis is unlikely in our view).

First, the fact that wages also increased muchmore quickly than productivity during the boomyears suggests that the overheated economy mayalso have gradually eroded competitiveness.Among the different sectors of the economy, unitlabour cost (ULC) increases were mostlyconcentrated in the non-tradable sectors inLithuania, as highlighted by Kuodis andRamanauskas (2009), a finding that also appliesto the other two Baltic countries. However, ULC hasalso increased sharply in the manufacturingsector as indicated in Figure 10, especially since2005. The international comparison offered by thefigure highlights that the Baltic countries lostduring the pre-crisis boom a significant degree ofcompetitiveness in manufacturing compared tothe other countries shown9.

Second, the non-tradable sector has developedespecially rapidly during the boom years. Withoutrestoring competitiveness it would be difficult todirect capacities from this sector towards thetradable sector.

Third, Figure 3 has already indicated that a widegap has opened between GDP per capita and pricelevels compared to the euro area. In other words,the actual exchange rate compared to thepurchasing power parity exchange rate, adjustedby the Balassa-Samuelson effect, has becomehighly overvalued. In order to provide a historical,cross-country overview of what used to happenwith the price level after large drops in GDP, wehave looked at all episodes among 182 countriesof the world between 1950 and 2007 in whichGDP per capita has fallen by at least 10 percentwithin a period of two years compared to theweighted average of 22 industrialised countries.We have selected countries with relative GDP percapita levels between 20 and 100 percent in thestarting year and have excluded transitioneconomies and oil-exporting countries. Figure 11indicates that such episodes also tend to befollowed by price level drops. On average, pricelevels fell somewhat sluggishly, but five years

after the start of the depression, the price leveldecline (19 percent) almost equalled the declinein GDP per capita (18 percent); both are relativeto the weighted average of 22 industrialisedcountries.

Figure 11: GDP per capita and price levels afterbig drops in GDP (compared to the weightedaverage 22 industrialised countries)

Source: Bruegel’s calculations using IMF and World Bank data.Note: The data shown refers to 43 episodes in 35 countriesbetween 1950 and 2007 in which GDP per capita at PPPdeclined by at least 10 percent compared to the weightedaverage of 22 industrialised countries with a period of two years(country specific weights were derived from Bayoumi, Lee andJayanthi, 2006). Only countries with GDP per capita between20 and 100 percent (in the starting year) are considered, andtransition countries and major oil-exporting countries wereexcluded. The line with the symbols shows the mean of the 43episodes, while the two dashed lines indicate the interquartilerange (ie the middle 50 percent of the distribution). Time ‘t’ isthe year which precedes the drop in GDP.

5. BALTIC OPTIONS

The cornerstone of the economic policy of Balticcountries has been the maintenance of the fixedexchange rate and the goal of euro entry as soonas possible. There is continued strong political

9. Note that all countriesexcept Poland and

Sweden had anexchange rate fixed to

their main tradingpartners. The Polish and

Swedish currenciesshowed in some caseslarge variations during

the sample periodshown, but did not have

any trend in exchangerate movement (Figure

4). Consequently, acomparison of nominal

ULC data shown onFigure 10 well reflects

changes incompetitiveness among

the nine countries.


11


support for this policy, but it has become thesubject of intense discussion.

The Baltic countries face three main choices, andwe shall argue that none of these choices clearlystands out as the best, as all have advantages anddisadvantages, and pose serious risks. There is afourth option, which would clearly be the best, butit is not within the power of the Baltic countries toadopt. The policy choice of the authorities and thesuccess in implementing their choice willdoubtless have an effect on which of thescenarios shown in Figure 8 will occur. The fouroptions are the following:

• Option 1: Preservation of the current exchangerate level and best endeavours to introduce theeuro as soon as possible;

• Option 2: Devaluation, but maintenance of thefixed exchange rate system;

• Option 3: Introduction of a floating exchangerate;

• Option 4: Immediate euro introduction at asuitable exchange rate, supported byappropriate resolution to manage the debtoverhang.

Option 1: Preservation of the current exchangerate level and best endeavours to introduce theeuro as soon as possible10

The three Baltic countries have continued topursue this option, though with different timehorizons for euro introduction. Policymakers in allthree countries have expressed their aim ofintroducing the euro as soon as possible, butEstonians seem to be the most ambitious with atarget date of 2011, implying that their applicationfor euro-zone membership will be assessed inspring 2010.

There is a clear rationale for choosing this option.With very high foreign currency debt, devaluationwould pose many risks, as we shall discuss underOption 2. The Baltic countries have reasonablyflexible economies and adjustment throughcutting wages and prices could be a solution. Moregenerally, having a fixed exchange rate withoutintroducing the euro implies giving-up monetary

policy without enjoying the benefits of thecommon currency, including the access to ECBfacilities. An independent currency in a very smalland open economy is a source of risk. It is alsosometimes highlighted that keeping the fixedexchange rate, which has survived both theturbulent years after independence in the early1990s and the Russian crisis, is an importantpillar of national pride.

Private sector competitiveness needs to beimproved, but the government also has a majorrole in the adjustment. All three countries haveannounced substantial public sector wage cuts,and other expenditure and revenue measures tocontain the budget deficit.

This strategy implies many risks. The first iswhether sufficient adjustment in wages andprices could be achieved or not. Figure 12 on thenext page shows data on average monthlynominal wages in local currency. In all threecountries public sector wages declined somewhatin the first half of 2009, but to a much lesserextent than announced (about 15, 40, and 25percent in Estonia, Latvia, and Lithuania,respectively). The reason for the discrepancycould be the time required to implement the wagecuts, but it needs to be seen if nominal wages willfall in the public sector, as announced, and ifprivate sector wages and prices will follow. Thegood news for the adjustment is that nominalwages in the private sector have also started todecline (Figure 12); thought the actual magnitudeis still quite small.

The second key risk is caused by the uncertaineconomic environment until a definitive solutionis found. Although all three Baltic countries haveruled out devaluation, markets are not fullyconvinced (Figure 7). As long as investors think acountry may devalue in the near future, they willnot invest, delaying real activity and contributingto the downward spiral.

Third, while the choice is not between devaluationand budget consolidation, because budgetconsolidation would anyway be needed to keep

10. In principle anadditional option would beto keep the peg at thecurrent level but not rushfor euro-area entry; instead,use fiscal policy to dampenthe economic and socialimpacts of the crisis andtarget euro-area entry onlyafter the crisis. While thispolicy would ease the shortrun adjustment needed bygovernment and mayindeed dampen the shortrun social impact of thecrisis, a prolonged period ofuncertainty until euro-areaentry would be detrimentalfor investment, and themany risks discussed forOption 1 would remain.Furthermore, the ability offiscal policy to boostdomestic production islimited in small, openeconomies.


12


expenditures in line with new medium termrevenue realities (Darvas 2009b), theconsolidation requirement is certainly greater ifthe fixed exchange rate is to be maintained.Clearly, budget consolidation at a time of sharprecession strongly amplifies the recession andthe social burden of the crisis. Latvia had to takeespecially drastic measures by cutting publicexpenditures by about 40 percent in 2009 andfurther cuts are required in 2010 in order to fulfillthe obligations of the EU- and IMF-led internationallending programme. Apart from the directconsequences of cutting all expenditure (otherthan interest payments and internationalobligations) by a substantial amount, socialunrest may bring down the government, andconsequently the international lendingprogramme, leading to unforeseeableconsequences.

Fourth, the fall in nominal incomes (both amongthose who continue to be employed and thosewho have recently become unemployed) willincrease the share of non-performing anddefaulting loans. The length of the period ofincreasing bank losses is uncertain but it willprobably be considerable, implying that banks willbe very cautious in their lending for a prolongedperiod, thus lengthening the recession (cf.Scenario B in Figure 8). A lengthened recessionand the uncertainties about its duration will alsolead to postponement of investment decisions, sokeeping the economies in a downward spiral.

Fifth, the lengthened recession will make it more

difficult for governments to meet the euro-areaentry criterion on the budget deficit. Nor can it betaken for granted that the inflation criterion will bemet, though the recession, the overvaluation ofthe exchange rate and the nominal wage cuts willcertainly contain inflationary pressures and willprobably even drive the economy into deflation.

The sixth risk comes from the unpredictablereaction of European institutions to the inflationsustainability criterion. The protocol annexed tothe treaty requires among other things that “...aMember State has a price performance that issustainable...”. Indeed, in the rejection ofLithuania’s 2006 euro-area application thesustainability criterion may have played a role11.Even if the Baltic countries will meet all(backward-looking) criteria, what if theCommission and the ECB do not regard thesituation as sustainable and instead argue thatthe achievement of low inflation was theextraordinary consequence of the deeprecession?

Even if the Baltic countries manage to adopt theeuro in the near future, there will be significantrisks and challenges for the medium and long run.There is no doubt that having the euro (convertedat the current exchange rate) would be better forgrowth than the current situation where there isexchange-rate risk and associated uncertainty inthe business environment. Introduction of theeuro may increase the attractiveness of thecountries and hence could lead to higher inwardinvestment. However, the large private sector debt

11. As we have shown inFigure 4, Lithuania’s

nominal effectiveappreciation between

1999 and 2004 wassizeable, which

contributed to lowinflation in the first half

of the 2000s. Furthernominal appreciation

was not expected.

Figure 12: Average monthly nominal wages in the three Baltic countries (in local currency)18,00016,000

14,000

12,000

10,000

8,000

6,000

4,000

18,00016,000

14,000

12,000

10,000

8,000

6,000

4,000

00 01 02 03 04 05 06 07 08 09

700

600

500

400

300

200

100

700

600

500

400

300

200

10001 02 03 04 05 06 07 08 09

2,800

2,400

2,000

1,600

1,200

800

2,800

2,400

2,000

1,600

1,200

80000 01 02 03 04 05 06 07 08 09

Public sector (raw data) Public sector (seasonally adjusted) Private sector (raw data) Private sector (seasonally adjusted)


Source: Bruegel and Central Statistical Offices of the three countries. Note: We seasonally adjusted the raw data using the X-12method. Quarterly data is available for Estonia and Lithuania (to 2009Q2), while monthly data is available for Latvia (to June 2009).

‘Even if the Baltic countries manage to adopt the euro at the current exchange rate in the near

future, there will be significant risks and challenges for the medium and long run.’


13


and the lower capacity to service the debt willremain even after euro-area entry. If theadjustment in nominal wages and prices provesinadequate and the exchange rate remainsovervalued, then Scenario B of Figure 8 maymaterialise. These outcomes are not justdetrimental in their own right, but may lead toaccelerated migration out of the Baltic countriesto EU15 countries once the recession is over in theEU15, which would further undermine the growthprospects of the Baltic countries (Ahearne,Brücker, Darvas and von Weizsäcker, 2009).

The case of Portugal also provides a warningsignal: following a prosperous economic catching-up period before euro-area entry in 1999, whichwas accompanied by credit, housing andconstruction booms and the build up of largecurrent-account deficits and external debt,Portugal had the slowest growth rate among euro-area countries, and the catching-up processhalted and even went into reverse to some extent(Blanchard, 2006). Spain and Ireland continuedto grow fast after euro-area entry, but are nowfacing the most serious recessions among euro-area members.

Option 2: Devaluation, but maintenance of thefixed exchange rate system12

Whenever restoring competitiveness anddiversion of resources from the non-tradablesector to the tradable sector requires a significantrelative price change, exchange rate devaluationor wage and price cuts could both do the job. Themost important argument in favour of devaluationis that a large correction in relative prices isneeded and it is much easier to do that with anexternal than with an internal devaluation.Deflation is more difficult to achieve and tomanage and also requires more time thandevaluation. As devaluation is quicker, it can giverise to a recovery sooner than the wage and pricecut strategy. There will be bankruptcies both whenthe adjustment is carried out by devaluation and

by cutting wages and prices. Devaluation wouldmostly hurt those who have foreign currencyloans, while wage and price cuts hurt everyone(Becker, 2009). Furthermore, although marketshave stabilised somewhat at the time of writingthis policy contribution, there is a risk that thesocial costs of budgetary cuts and wagereductions will be unbearable, or that markets willanyway enforce devaluation later. A voluntarilyand properly designed devaluation is clearlypreferable to a market- or social unrest-enforceddevaluation, because the latter will most likelyresult in an exchange rate overshoot with furtherdevastating consequences, as many previouscrises have demonstrated.

The key arguments against devaluation are that itwould bring about bankruptcies earlier because ofthe balance sheet effect of foreign exchangeloans, while deflation would provide some time toadjust for all concerned. External financing needsmay not be reduced, because private sectorrollover rates might not improve, as the externaldebt-to-GDP ratio would increase (IMF, 2009a). Afurther risk relates to the reaction of foreign banks(mostly Swedish and other Nordic banks) to alarge number of simultaneous defaults: banksmay decide to withdraw from the Baltic countries,which would further undermine economicrecovery. Furthermore, devaluation would createa precedent because the official fixed rates havenever been devalued since these countries gainedindependence after transition. But devaluationmay induce markets and local people to expectfurther devaluations and to speculate accordingly,including possible runs on banks and conversionof domestic currency deposits into foreigncurrency deposits (Levy-Yeyati, 2009).

Whether a significant change in relative prices isstill needed is also not fully obvious. We haveshown that current accounts have gone intosurplus (Figure 5), and we have discussed in thepreceding section three possible mechanisms

12. A milder form of thisoption is the use of the full+/-15 percent wide ERM2band (instead of the currentpolicy of keeping the rate inthe middle of the band),which would be likely tolead to depreciation to theweak edge of the band.


14


leading to this outcome, one of which does notpoint towards the need for adjustment. Further,the three Baltic countries are probably differentregarding the required size of relative priceadjustment as suggested by eg Figure 3 and 10.Furthermore, for Estonia, IMF (2009b) concludes:“Staff’s assessment is that the real effectiveexchange rate is somewhat overvalued, but thatcompetitiveness is projected to remain adequateas real wage increases are aligned withproductivity. Structural policies to ensure marketflexibility considered to be crucial for facilitatingreal adjustment including a reallocation ofresources from the non-traded to the tradablesector and from low wage to high-value addedactivities.” Such a more-or-less positive assess-ment cannot be found in eg IMF documentsdealing with Latvia.

An additional risk to neighbouring countries is thatif only one (or two) of these countries were to optfor an uncoordinated devaluation, spillover effectsmay necessitate a disorderly devaluation in theother country(ies).

Option 3: Introduction of a floating exchange rateSome CEE countries adopted a floating exchangerate rather successfully. The Czech Republic, forexample, achieved high growth, low inflation,below-euro-area interest rates before the crisis,and foreign currency loans had practically noshare in household loans and made up a very lowshare of loans to corporations. Poland andSlovakia also adopted the floating exchange ratequite successfully and these countries areweathering the storm now much better than theBaltic countries13.

After many years with a fixed exchange rate,Ukraine opted for a floating rate in November 2008in response to the crisis. The exchange rate of thehrivna depreciated by about 40 percent in a fewmonths (helping the relative price adjustment but

hurting borrowers with foreign exchange loans)and was volatile, but it has achieved a remarkabledegree of stability since spring 2009.

With the benefit of hindsight, in the Balticcountries the introduction of a floating exchangerate would have been a good option around 2000,ie some time after the Russian crisis but beforethe huge credit boom started. The floating ratewould have disciplined borrowers and the creditboom would probably have been milder, similarlyto other floating rate CEE countries. However, thereis always a fear that, in very small and very openeconomies, a floating exchange rate may lead toexcessive exchange rate volatility that could bedetrimental to investment and growth. The keyquestion, which at this point has only historicalrelevance, is if the huge boom and bust cycle andall associated consequences caused by the fixedexchange rate or potential exchange rate volatilitywould have been more detrimental to socialwelfare in a longer term perspective.

However, moving to a floating rate at the time ofpanic in a small country with very high externaldebt risks excessive depreciation of the exchangerate, and consequently many of the dangersassociated with devaluation discussed so farwould come into play with greater force14.

The above arguments suggest that the choice isvery hard and there is no clear-cut winner amongthe options. However, there is a further option,which is not at the disposal of the Baltic countriesthemselves, but could be implemented via EU-wide coordination and agreement.

Option 4: ‘Immediate’ euro introduction at asuitable exchange rate supported by appropriateresolution to manage the debt overhang15

This option would combine the benefits of theother options, eliminate most of their risks andspeed up and help sustain the recovery.

13. See Darvas and Szapáry(2008) for a detailed

analysis of theexperiences with

floating exchange ratesin new EU member

states and thecomparison of their

economic performanceto fixed exchange rate

countries.

14. Still, some observersargue for a floating rate

especially because itwould be likely to lead to

an overshooting of theexchange rate andhence may create

attractive investmentopportunities for foreign

investors, therebyhelping the recovery and

medium term growth(see RGE Monitor 2009).

‘The choice is very hard and there is no clear-cut winner among the options facing the Baltic

countries. However, there is a further option: immediate euro introduction at a suitable

exchange rate supported by appropriate resolution to manage the debt overhang.’


15


Why euro introduction? The dilemma around theovervalued exchange rate and the large stock offoreign currency loans cannot be solved properlyby any of the options discussed so far. While euroadoption itself is not a solution to the dilemma, acomprehensive package also including the selec-tion of the suitable exchange rate and resolution ofthe private debt overhang would solve it. It is betterto design a comprehensive package than to dis-cuss the individual elements of it. Morefundamentally, having an independent currency ina very small and open economy is a source of riskand keeping the rate fixed may not be a viableoption under free capital mobility. The credibilityand stability brought by the euro would benefitinvestment and growth. The euro proved to be ashelter during the crisis and crises may occur inthe future. Euro-area members have scope to runcounter-cyclical fiscal policy during a crisis. Eurointroduction at the earliest possible date has beenthe cornerstone of the economic policy of the Balticcountries and, in any case, all new EU entrants areobliged to introduce the euro at some point.

Why ‘immediate’? The Baltic situation, especiallyin Latvia, is getting more and more dramatic.While Baltic governments and central banks aremaking every effort to survive with the peg, thereis a risk of failure. Eventual collapse in one countrymay trigger a collapse in the other two. A Balticcollapse is not in the interest of the EU as whole. Ofcourse, euro introduction cannot be done‘immediately’ and any action has to respect thetreaty of the EU and all other international laws.However, with a unanimous decision of the Councilappropriate provisions can be put in place tostrengthen the economic foundations of the euro-accession criteria, which would increase thecredibility of the euro adoption plan and speed upthe convergence. In the next sections we discussthe economic and legal aspects of this reform.

Why at a ‘suitable’ exchange rate and what doesthis term mean? Introducing the euro at anovervalued exchange rate risks Scenario B ofFigure 8. Defining the appropriate final conversionrate is a difficult task. The evidence we haveprovided so far indicates that the exchange rate is

highly overvalued, especially in Latvia, but also inEstonia and Lithuania. The wage and price fallsthat have already started are helping to restorecompetitiveness, but it is questionable how farthis process can go. We do not take a stand on thelevel of the appropriate conversion rate as itsdetermination should be based on detailedcalculations considering a large number of factors.The issue of the appropriate conversion rateshould be carefully discussed on economicgrounds by European and national authorities.

Under what conditions? Euro-area entryconditions are laid down in the treaty. However,the economic rationale behind the criteria wasdoubtful even before the crisis – but has beenespecially so during the crisis. We shall argue inthe next section that keeping the criteriaunchanged violates in the economic sense theequal treatment principle. The entry criteria shouldbe reviewed and fine tuned within the legalframework of the treaty and only countries thatmeet the reviewed criteria should be allowed tojoin the euro area. Any solution should not beBaltic-specific but should be equally applicable toany EU country. Any solution should not incurmoral hazard, ie should not encourage privatesector actors and governments to count on ‘cheap’future help in crises.

Why and what kind of resolution of private debtoverhang? It is obvious that there will besignificant bank losses even if parity ismaintained or changed, whether or not the euro isintroduced. Neither the people (through indirectchannels including those who do not have loans)and governments of the Baltic countries, nor thegovernments of the home countries of the banksactive in the region, are immune to theconsequences of bank losses. The recentunilateral announcement of the Latviangovernment regarding a possible new bill toretroactively intervene in existing mortgagecontracts16 underlines that the country may notbe able to solve its problem itself and that thesituation may escalate. A potentially escalatingsituation would disad-vantage all partiesconcerned. Consequently, appropriate resolution

15. For Latvia, IMF (2009a)argued for immediate euroadoption possibly at theweak edge of the +/- 15percent exchange rateband, but noted that the EUauthorities have ruled outthis option. Becker (2009)has also argued in favour ofimmediate euro adoption ata devalued exchange rate.Levy-Yeyati (2009) wouldalso favour this option, butacknowledging that it wouldnot get EU-wide support heproposes a containeddevaluation. Nouriel Roubinisuggested depreciating thecurrency, euroise afterdepreciation, restructureprivate foreign currencyliabilities without a formal'default', and boost the IMFplan to limit the financialfallout (see RGE Monitor2009). For the other twoBaltic countries such directsuggestions are rarer.Åslund (2009), on the otherhand, opposes devaluationin Latvia and lists six waysin which Latvia differs fromArgentina and hence why itis unlikely that Latvia’sfixed exchange rate regimewill experience a collapsesimilar to the collapse ofthe Argentine fixedexchange rate regime in2001/2002.

25

30

35

40

45

50

Jan-

99

Jan-

00

Jan-

01

Jan-

02

Jan-

03

Jan-

04

Jan-

05

Jan-

06

Jan-

07

Jan-

08


16


schemes should be put in place (see the thoroughdiscussion of this issue in Mitra, Selowsky andZalduendo, 2009). However, the cross-borderownership of banks re-quires cross-borderparticipation in the design of the resolutionschemes; in the extreme case a multilateralburden-sharing agreement should also not beexcluded, though negotiations for it can beextremely intricate in legal, political and economicterms. The requirement for not incurring moralhazard is especially critical in designing anyresolution scheme.

In the next section we briefly discuss theeconomic rationale behind the review of the euro-area accession criteria, which will be followed byan analysis of the legal options under the treaty.

6. PROPOSAL TO REVIEW THE EURO AREA ENTRYCRITERIA17

Even before the crisis, much discussion took placeabout euro-area entry rules, but a lot more hastaken place in the wake of the crisis. Euro-areaentry criteria were established in the early 1990sbefore the euro area existed and when the EU had12 member states. Intense discussion precededthe drawing up of the rules, and the end result wasa compromise between economics, politics andsimplicity. Now the euro area exists and there are27 members of the EU, but the rules remain thesame. It is easy to show that keeping the samerules in an expanded EU violates in the economicsense the equal treatment principle – newapplicants have to meet tougher criteria thanprevious ones because two of the criteria arebenchmarked on the “three best-performingmember states of the EU in terms of price stability”,which have been interpreted in a special way. Thetreaty does not specify how to determine the “threebest performers” (see the next section); in practicethis been defined as the three EU countries havingthe lowest non-negative inflation rates. Lewis andStaehr (forthcoming) found that according to thisinterpretation the expansion of the EU from 15 to 27members reduces the expected inflation referencevalue by 0.15-0.2 percentage points, and there is aconsiderable probability of a larger reduction.

Another long-known factor is the significant price-level convergence of the new EU member statesdue to their economic catch-up (the so-called‘Balassa-Samuelson effect’), which results inhigher inflation if the exchange rate is to be keptstable. Such inflation is structural and not areflection of unsustainable policies. While thiseffect was present in some current euro-areamember countries, the effect is much stronger inthe new EU member states and hence it is muchharder for them to meet the inflation criterion thanit was for earlier applicants, unless they revaluethe exchange rate as Slovakia did (Figure 13).However, continuous appreciation of the currency,while not against the letter of the treaty, questionsthe usefulness of the exchange-rate criterion, asits rationale must have been to demonstrate thata country can live with exchange-rate stability.

Figure 13: The Slovak koruna against the euroand the ERM-II band, 1999-2008

Source: ECB. Note: A decrease in the exchange rate indicatesappreciation of the koruna against the euro. The red lineindicates the central parity and the two green lines indicatethe edges of the exchange rate band. The width of theexchange rate band was +/-15 percent.

A further long-known factor regarding the inflationcriterion (as well as the interest-rate criterion) isthat it is benchmarked against all EU countries.While this was a perfectly natural idea before theeuro existed, it has become rather questionable.The inflation rates of those EU countries that arenot euro-area members may be affected by largeexchange-rate swings.

Nevertheless, before the crisis and in the first few

16. According to new reportsthe proposed legislation

(which is incidentallystrongly opposed by the

Bank of Latvia) issupposed to have two

key pillars: (1) limitingmortgage lenders’

liability to the value ofthe collateral instead ofthe size of the loan and

(2) requiring that a bankcan repossess a

property in case ofdefault only if it provides

an alternative home tothe debtor with the

debtor's agreement. Inour view retroactiveintroduction of this

legislation is clearlycontrary to international

law and hence cannot beimplemented. Raising

the issue stronglyundermines the trust in

the enforceability ofcontracts and the rule of

law in general, whichmay have devastating

future consequences forinvestments in Latvia.

17. Most economicarguments in this

section first appeared inDarvas (2009a).

0

10

20

30

40

50

60

70

80

90

100

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

%General government balance

General government debt

Inflation

Interest rate

Missing at least one criterion


17


months of it, we shared the view that not allcriteria should be reviewed, restricting ourselvesto advocating a change to the misguidedinterpretation of the inflation criterion mentionedabove (Darvas and Szapáry, 2008; Darvas,2008). Also, four new EU member states have sofar managed to join the euro area. However, inthree of the four cases there was no need forsubstantial price-level convergence and the fourthcase, Slovakia, could only manage it with thesubstantial nominal exchange-rate appreciation(Figure 13) discussed above.

The crisis has prompted a rethink of manypositions, and we ought to rethink the euro-areaentry criteria too, because serious asymmetryexists and serious issues are at stake.

Figure 14: Percent of euro-area member statesmissing the entry criteria, 1995-2010

Source: Bruegel’s calculation based on data from Eurostatand the IMF's October 2009 World Economic Outlook. Note:The percentages are calculated from the actual euro-areamember states in each year (and the first eleven membersbefore 1999): from 1995 to 2000 the 11 countries thatintroduced the euro in 1999; in 2001 Greece is added as the12th country; in 2007 Slovenia is added as the 13th country;in 2008 Cyprus and Malta are added as the 14th-15thcountries; for 2009-2010 Slovakia is added as the 16thcountry. Definition of meeting the general government debtcriterion: a country is considered to meet the criterion if eitherthe debt/GDP ratio is below 60 percent or, if above, thenprojecting the average change in debt/GDP ratio of the latestthree years twenty years ahead will lead to a ratio below 60percent. Three percent is used for the budget deficit criterion.The three EU countries (considering the actual members ofthe EU of the given year) with the lowest (but non-negative)inflation rate were used to determine the inflation andinterest rate criteria.

Asymmetry. Once a country is inside the euroarea, it can do almost anything it likes. TheStability and Growth Pact in principle limits thescope of government action inside the euro area,but not much, as many examples show, both inthe pre-crisis period but especially during thecrisis. Figure 14 shows that 50-60 percent ofeuro-area member countries have violated at leastone entry criterion between 1999 and 200818. Inresponse to the crisis, government deficits anddebt are ballooning in euro-area countries, butcountries wishing to join are subjected toextremely tough and painful measures if they are,in a few years, to be considered eligible.

The countries that have joined the euro area werejudged to have achieved a “high degree ofsustainable convergence”. The large number ofviolations after euro-area entry suggests that thecriteria are inadequate for judging ‘sustainableconvergence’. This fundamentally calls intoquestion both the economic and moralfoundations of the future application of thecurrent entry criteria.

The asymmetry also highlights that the capacityto meet the current entry criteria depends on thebusiness cycle, which is a highly unfavourableproperty. Suppose, for example, that Slovakiaapplies for euro-area membership a year later andhence is evaluated in the spring/summer of 2009instead of the spring/summer of 2008. Slovakiawould have been rejected in 2009 because of thegeneral government balance criterion, thoughfiscal policy has not become ‘irresponsible’between 2008 and 2009. Also, like all regionalfloating currencies, the Polish zloty, the Czechkoruna, the Hungarian forint and the Romanian leuexperienced serious pressures and significantlydepreciated against the euro between thesummer of 2008 and the spring of 2009, and theSlovakian koruna may also have come underpressure at that time without a sure prospect ofeuro-area entry. This in turn may have qualified itfor “severe tensions” status according to the letterof the treaty, putting the evaluation of theexchange-rate criterion at risk, and the tensionsmay have increased government bond yields,

18. The entry criterion forgovernment budgetpositions is the absence ofan excessive deficitprocedure, as we willdiscuss in the next section.Calculations behind thefigure assume a pragmaticdefinition of thegovernment debt criteria(see details in the note tothe figure) and use thethree percent deficitbenchmark. The figure isbased on currentlyavailable data – at the timeof evaluation, real time datawas used; this has beenrevised in some cases.


18


putting the fulfilment of that criterion at risk aswell.

Business cycle dependence is the result of thefact that the two budgetary criteria (deficit anddebt) are phrased in absolute terms (the inflationand interest rate are phrased in relative terms).Business cycle dependence implies that mostcountries can join only in good times19, whichdoes not make much sense: meeting the criteria ingood times obviously does not tell one muchabout long term sustainability, as Figure 14 hasalso shown.

High stakes. One may say that the new applicantsshould have pursued policies similar to those ofthe four newest euro members. However, thestakes are much higher now than just naming andshaming. As we have discussed, the three Balticcountries are in deep trouble and the currentmisery is not only their own fault. A Balticexchange-rate peg failure or a prolonged recessionand a halted economic catch-up would not justcause further pain for the populations of thesecountries. It could undermine trust in the notionof common European values, and lead to a newdivide within Europe (Darvas and Pisani-Ferry,2008). Western European investors in the Balticcountries would also suffer heavy losses. If theBaltic countries do not recover, this will impact theEU’s role as driver of reform in other new memberstates as well as in the neighbourhood countries.

It would be tempting to drop one or the othercriterion20, but that would require changing somekey articles of the treaty. This should be possible,but seems highly unlikely. In fact, the treatyincludes an obligation for the Council to lay downthe details of the convergence criteria referred toin a main article of the treaty, this Council decisionreplacing the relevant protocol to the treaty; thesame obligation exists for the protocol on theexcessive deficit procedure (see details in thenext section). Consequently, it is possible tostrengthen the economic foundations of thenumerical requirements of the current fourconvergence criteria relatively easily, ie with aunanimous decision of the Council. Economic

theory does not provide clear guidelines abouthow to determine the magnitudes, but a fewprinciples can be laid down.

All criteria should be related to the euro-areaaverage for at least four reasons:

• First, all prospective applicant countries arehighly integrated into the euro area (if not, theyshould be), and hence what happens insidematters a lot for those outside. This appliesboth to the inflation rate and also to the generaleconomic outlook, which has an impact onbudget deficits both in the euro area and inapplicant countries.

• Second, it would abolish the peculiar possibilitythat non-euro-area countries or very smallcountries with which the applicant has virtuallyno trade may affect the criteria.

• Third, it would alleviate the asymmetry ofletting the automatic stabilisers run andhelping the economy with discretionarystimulus in euro-area countries while doing thepainful opposite of this in applicant countriesduring a crisis.

• Fourth, as countries in the euro area aredeclared to have achieved “a high degree ofsustainable convergence”, the convergence ofapplicant countries towards the euro-areaaverage seems a natural requirement.

The inflation, interest rate and budget balancecriteria should allow some deviation from theeuro-area average. New EU member states aresmall and open economies characterised by largercyclical swings, and thus need greater scope forcounter-cyclical fiscal policy. Moreover, the needfor public sector investment is greater than in oldEU member states. With regard to inflation, newEU member states have a higher potential growthrate, which implies structural price-levelconvergence, which should be acknowledged.

One option that would allow some deviation fromthe euro-area average is to define the maximumdeviation. For example, the budget balancecriterion could be the average euro-area balanceminus 1.0 percentage points (all measured as a

19. Estonia’s chances alsodepend on the business

cycle but the oppositeway: it could not join in

‘good times’ because ofthe inflationary

pressures discussed inearlier sections. But if

the governmentsqueezes the budget

sufficiently in the midstof a drop in GDP of about

15 percent in 2009 (incontrast to euro-area

governments that usefiscal policy to dampen

their five percent GDPfall), it may have a

realistic chance of apositive assessment in

the spring of 2010,unless the expected

deflation is regarded as‘unsustainable’.

20. For example, Buiter(2005) argues that

“achieving fiscalsustainability prior to

adopting the euro isessential and it is theonly truly necessary

condition for euroadoption. It should also

be a sufficient conditionfor Eurozone

membership.”


19


percentage of GDP) and the inflation criterioncould be the average euro-area inflation rate plus1.5 percentage points. (If the deviation should be1.0 or 1.5 percentage points or another similarnumber should be the subject of discussion.)Another possibility would be to require theapplicant to have better statistics than, say, 25percent of euro-area members.

The requirement for the ratio of government debtto GDP could simply be that this ratio should notexceed the euro-area average (or should be lowerthan in at least 25 percent of euro-areamembers), unless the ratio is diminishingsufficiently and approaching the euro-areaaverage at a satisfactory pace.

In order to ensure that the reformed criteriaprovide a better indication of sustainableconvergence than the original criteria, thecompliance period could be increased from thecurrent one year to the average of the last two orthree years.

Would this change jeopardise the stability andcredibility of the euro area? Certainly not. Therewould still be criteria (but more sensible ones) tokeep applicants on their toes. Furthermore, ingood times the new criteria would be tougher. Forexample, when the budget is balanced in the euroarea, then the new criterion would (rightly) requirea better budget position from the applicants. Themost important threat to the stability of the euroarea is the lack fiscal sustainability – this is a realthreat for many countries currently inside theeuro area, but potential applicants have muchlower government debt-to-GDP ratios and areundoubtedly much better prospects in this regard.And in any case the EU’s surveillance systemneeds a fundamental revision to ensure stabilityand growth in the whole EU. Furthermore,prospective applicants from the new memberstates would make up a very small share of thetotal euro area, and their inclusion would hardly

be noticeable in euro-area aggregates. Theargument that inflation will be higher in the euroarea after admitting countries in which theBalassa-Samuelson effect is persistent, andhence inflation must be lower in old memberstates in order to meet the ECB’s inflation targetfor the euro-area average, is offset by themagnitudes. According to Darvas and Szapáry(2008), the impact of the proposed modificationof the inflation criterion on euro-area averageinflation would be less than an additional 0.05percent per year – a magnitude well within themeasurement error of inflation rates.

Would the revision of the convergence criteriaundermine the trust in EU rules? Clearly not. The‘flexible’ interpretation of the criteria at manyprevious euro-area admissions (see the nextsection) should have already undermined publictrust in the process of euro-area enlargement. Onthe contrary, revising the criteria to be moreeconomically rational and to have less scope fordiscretionary interpretation would evenstrengthen the trust.

Would it be difficult to reach consensus among the27 member states on this particular change? Wethink not. Countries outside the euro area wouldcertainly support it. Countries inside would feelmore comfortable having rules that make moresense. The goal is not to weaken the euro-entrycriteria but to make them more sensible. Thechange should be carefully orchestrated andinitiated by euro-area member states or Europeaninstitutions, not applicant countries.

The suggested change in euro-entry criteria wouldstill require substantial effort from applicants, butit would ease their pain. It would also boostconfidence, helping kick-start the private capitalinflows – rather than inflows of westerntaxpayers’ money – that these countriesdesperately need.

‘The large number of violations of the criteria after euro area entry suggests that the criteria

are inappropriate for judging ‘sustainable convergence.’


20


7. LEGAL OPTIONS UNDER THE CURRENT TREATYFOR INTERPRETING EURO-ENTRY CRITERIA INECONOMIC TERMS

A key question is whether the letter of the treatyand the precedents provided by the previousapplications of euro-area entry criteria allow amore flexible interpretation of rules in order torequire more sensible criteria from future euro-area applicants.

7.1 Full euro-area membership

First of all it is important to recall that for threecountries (Finland, Italy and Slovenia) it is notunambiguous from a legal point of view whetheror not the exchange-rate criterion was met. Thetreaty included and continues to include thefollowing requirement: “the observance of thenormal fluctuation margins provided for by theexchange-rate mechanism of the EuropeanMonetary System, for at least two years” and theprotocol added that “...for at least two years beforethe examination”21 (without devaluing itscurrency on the initiative of the member stateconcerned). The most neutral interpretation of thisregulation is that participation in the exchange-rate mechanism (ERM) is required for at least twoyears before the examination. Neither the treaty,nor its protocol provided any waiver from therequirement for the minimum period of two years.The three countries mentioned did not spend theminimum two-year period prior to the examinationin the exchange-rate mechanism. The 3 May 1998decision of the Council of course recognised this,but decided in any case:

“Italy fulfils the convergence criteria mentioned inthe first, second and fourth indents of Article109j(1); as regards the criterion mentioned in thethird indent of Article 109j(1), the ITL, althoughhaving rejoined the ERM only in November 1996,has displayed sufficient stability in the last twoyears. For these reasons, Italy has achieved a highdegree of sustainable convergence.” (OfficialJournal of the European Communities, 1998, p.32. The article number refers to the treaty in forcein 1999. The first, second and fourth indents cited

refer to the criteria on inflation rate, excessivedeficit procedure and long term interest rate,respectively. The decision for Finland used thesame wording except that Finland joined the ERMin October 1996.)

For the other nine countries different wording wasused, eg for Belgium: “Belgium has achieved ahigh degree of sustainable convergence byreference to all four criteria.” The same wordingwas used for the other eight countries that hadparticipated for at least two years in the ERM.

Consequently, the Council decision of 3 May 1998recognised the absence of the two-year period inthe ERM, but despite this the Council assessed,based on the recommendation of the Commissionand the EMI (European Monetary Institute), thatthe two countries had a stable exchange rate. Onemay say that the decision was in line with thespirit of the treaty, but it is far from being obviouswhether the letter of the treaty was also fullyrespected. This ‘flexible’ interpretation of the treatyshould be noted as providing a precedent22.

On other occasions it was rather questionable ifthe spirit of the treaty was satisfied, though theletter of the treaty was formally respected.Regarding the government budgetary position theletter of the treaty requires the absence of anexcessive deficit procedure (EDP), which isdecided on the basis of the ratios of governmentdeficit and debt to GDP. For the latter, the treatyallows a discretionary decision if “the ratio issufficiently diminishing and approaching thereference value”, ie 60 percent of GDP, “at asatisfactory pace”. On 1 May 1998 the Councilabrogated the EDP against seven countries thatjoined the euro area in 1999, the decision beingsubstantiated by the discretionary optionsallowed by the treaty. However, it was quest-ionable whether, eg the decline of Italy’s generalgovernment debt-to-GDP ratio from 124.9 percentof GDP in 1994 to 121.6 percent by 199723

corresponded to the cited requirement. Similardoubts could be raised for some other countriesregarding the abrogation of the EDP24.

21. All citations of the treatyand the protocol

annexed to it are takenfrom the Lisbon ‘’Treaty

on the Functioning of theEuropean Union.

Consolidated version”(Official Journal C 115 of

9 May 2008). TheMaastricht and the Niceversions have the same

wording, but thenumbering of the

articles is different.

22. A possible alternativeinterpretation is that

ERM participation is notrequired, but only

exchange-rate stability.In this case Bulgaria

may immediately applyfor euro-area

membership as it is nota member of the ERM

but its exchange rate isfixed to the euro without

allowing any volatility.

23. These figures were usedin the 1998

assessment, but havebeen revised somewhat

since then.


21


To sum up, the application of the treaty providesthree precedents where formal adherence to theexchange-rate criterion was suspicious. Therewere some occasions where countries wereallowed to join by formally meeting all the criteria,but these were the results of questionableexercising of the discretionary options allowed bythe treaty with respect to the budgetary criterion.

These precedents are encouraging for theprospects of new applicants, but only if these past‘flexible’ practices are also applied in the future.

Having reviewed some key features of the pastapplication of the treaty, let us now look at theflexibility offered by the letter of the treaty and itsprotocol for future euro-area applicants25.

1. Inflation. The inflation criterion is benchmarkedagainst “the three best performing countries interms of price stability”. Neither the treaty nor itsprotocol define how to determine the best per-formers. In practice the three countries with thelowest, but non-negative, inflation rates wereused. As highlighted by many authors (eg, Buiter,2005; Pisani-Ferry et al, 2008) the applied defini-tion contradicts the ECB’s definition of pricestability, which defines it as close to, but below,two percent. There is nothing in the treaty thatwould hinder the ECB and the European Comm-ission from interpreting the three best performersas the three countries having inflation rates (1)either below or close to two percent, or (2) theclosest to the average inflation rate of the euro area.

2. Excessive deficit26. The room for manoeuvre is‘moderate’ for not opening an EDP or for abrogatinga previously-opened EDP when the “ratio of theplanned or actual budget deficit to gross domesticproduct at market prices exceeds a referencevalue”, which is three percent. The discretionaryoptions provided in the treaty allow the EDP not tobe applied if “either the ratio has declined

substantially and continuously and reached alevel that comes close to the reference value, or,alternatively, the excess over the reference valueis only exceptional and temporary and the ratioremains close to the reference value”27. The words“exceptional”, “temporary” and “close” are notdefined in the treaty, nor in its protocol. EDPs weretypically opened for budget deficits somewhatabove three percent of GDP, but this does not at allmean that “close” must be defined this way. Thecrisis is clearly exceptional and under exceptionalcircumstances, temporariness could last for a fewyears. Consequently, there is some room formanoeuvre, but a proper interpretation ofcloseness at a time of a deep crisis will require anopen attitude similar to past ‘flexible’ practices.

3. Exchange rate. The precedents for theassessment of this criterion must imply that thiscriterion is not really binding. The ‘flexibility’shown so far (eg Italy, Finland, Slovenia andSlovakia) should be extended to future applicants.There is, however, an unsolved issue regarding thiscriterion: the conditions for joining the ERM-II.Without ERM-II membership a country can not jointhe euro area even if it has a stable exchange rateand meets all other criteria28. There should be clearand transparent criteria for joining the ERM-II.

4. Interest rate. In principle, the long-term interest-rate criterion serves as a means to assess thesustainability of the low inflation rate. In practice,however, this criterion reflects the credibility of theeuro-area accession process: when marketsattach a high probability to eventual euro-areaaccession, interest rates will converge, at least tosome extent, regardless of the longer-termsustainability of the inflation rate. The twopercentage point deviation allowed by the treatywill almost surely also be sufficient for futureeuro-area applicants where euro-accessionprospects are credible. Whenever economicreasoning supports future euro introduction,

‘The application of the treaty provides precedents where formal adherence to the exchange-

rate criterion was suspicious. Even when countries were allowed to join by formally meeting

all the criteria, there was questionable exercising of discretionary options.’

24. Some authors, eg Buiter(1995) and De Grauwe(2009), regard these casesas violations of the letter ofthe treaty as well.

25. In addition to the fourcriteria to be discussedbelow, the adequacy ofnational legislation,including the statutes of thenational central bank,integration of markets, thecurrent-account balance,unit labour costs and otherprice indices are alsoexamined for countrieswishing to join the euroarea.

26. The Excessive DeficitProcedure is the crucialcommon component in theStability and Growth Pact(that applies to allmembers of the EU) and theeuro accession criteria.

27. The second criterion ofthe EDP refers to grossgovernment debt and itsdiscretionary option hasalready been cited abovewhen discussing abrogationof the EDP in the case ofItaly in May 1998. However,the precedents of lettingcountries to join with ratioswell above 100 percent ofGDP must imply that thiscriterion will not be bindingfor any prospectiveapplicants having debtratios above 60 percentwhere they are below Italy’sand Belgium’s 120 percentratios at the time of theireuro area admission.

28. For example, Bulgariahas a fixed exchange rate tothe euro and all officialdocuments emphasise theoverriding goal of eurointroduction as soon aspossible. Yet Bulgaria is nota member of the ERM-II andit is difficult to fathom whyfor an outsider given thecurrent lack of transpar-ency of entry conditions.


22


European policymakers can increase thecredibility of applicants’ path toward the euro.

To sum up, there is indeed some room formanoeuvre in the treaty and in its current protocolregarding formal inclusion in the euro area. Butmore importantly, the Council has an obligation toclarify the four convergence criteria and replacethe current protocol:

“The Council shall, acting unanimously on aproposal from the Commission and afterconsulting the European Parliament, the ECB asthe case may be, and the Economic andFinancial Committee, adopt appropriateprovisions to lay down the details of theconvergence criteria referred to in Article140(1) of the said Treaty, which shall thenreplace this Protocol.” (Article 6 of Protocol 13)

The same obligation exists in the treaty to clarifythe details of the excessive-deficit procedure thatwill replace the protocol annexed to the articlediscussing the EDP. In the case of the EDP, some ofthe procedures have been detailed, but even theLisbon version of the treaty includes the referenceto the Council’s authority to adopt appropriateprovisions. It is time to spell out the details of theconvergence criteria and of the EDP, to strengthenthe economic rationale of the convergencecriteria. In particular, the numerical criteria shouldbe benchmarked against the euro-area average asdiscussed in the previous section. As has recentlybeen noted by von Hagen and Pisani-Ferry(2009): “the criteria for joining the euro area wereintroduced in order to ensure that economic logicprevails over political logic, not that legal logicprevails over economic logic” (p. 25). It is stronglyin the European interest to follow this principle.

7.2 Unilateral euroisation

The Council, the Commission and the ECB haverepeatedly ruled out the possibility of unilateral

introduction of the euro (as legal tender) on thebasis that the treaty provides one and only oneway to the euro. Therefore, for political reasons itwould not be wise to implement a unilateral movewhich would earn the clear disapproval ofEuropean policymakers, which is the currentreality. Furthermore, without ECB support, thiswould be very difficult to do in technical andpractical terms. If the unilateral move did not gaincredibility, people could start to withdraw theirdeposits and various other financial assets incash euros. The banking system may not be ableto supply as much euro cash as required, whichmay lead to a breakdown of the financial system.To take a Latin American example, FitchRatings(2009) argues that in the dollarised Ecuador arapid decline in bank deposits or acceleratedcapital flight, combined with limited access toexternal financing, could lead to a crisis driven exitfrom dollarisation.

8. SOME CLOSING THOUGHTS

The three Baltic countries face the deepestrecessions among all countries of the world and itis not just the fault of the politicians and thepeople of these countries that they are in thispredicament. Other eastern European countriesare also suffering disproportionately to their pre-crisis mistakes. The EU has mobilised resourcesto support crisis-hit countries in central andeastern Europe (Darvas, 2009c) according to itsrule-books, but the EU should be more than just arule-book. When everyone is aware that a rule hasdeficiencies, action is needed to modify the rule.

On the one hand, there is broad consensus that animmediate euro introduction accompanied byproper other provisions would serve the bestinterests of the Baltic countries and the EU as awhole, but such a solution is not feasible in thelegal sense and would raise many economic andpolitical issues. On the other hand, it is also clearthat the economic foundations of the current euro-

‘The EU should be more than just a rule-book. When everyone is aware that a rule has

deficiencies, action is needed to modify the rule.’


23


area entry criteria are fundamentally called intoquestion by the fact that it has been a rule ratherthan the exception that euro-area members haveviolated the entry criteria since becoming members,both before the crisis and currently. Adherence tothe current interpretation of the criteria is alsoweakened by the precedent of the ‘flexible’ appli-cation of the treaty at the time of certain previousadmissions to the euro area. The EU’s expansionfrom 15 to 27 mem-bers also made the rules tougherand hence, contrary to common perception, keepingeuro-area entry rules unchanged violates in theeconomic sense the equal treatment principle.

The coincidence of these two consensuses shouldbe used to grasp an opportunity: reform the euro-area entry criteria and demand more meaningful

criteria from all future euro-area applicants. Anysolution should not be Baltic-specific but shouldbe equally applicable to any EU country andshould not incur the risk of moral hazard. At leastthe room for manoeuvre in the protocol of thetreaty should be exploited by requiring from futureapplicants criteria that make more sense. The bestsolution, however, is for the Council to fulfil theobligation placed upon it by the treaty to lay downthe details of the convergence criteria and theexcessive deficit procedure that will replace thecurrent protocols. In the midst of an unexpectedlydeep crisis, two decades after the drawing up ofthe rules and one decade since the launch of theeuro, it is indeed time to reform the convergencecriteria. In designing the reform, the economicrationale of the criteria should be strengthened.

REFERENCES

Abiad, Abdul, Daniel Leigh and Ashoka Mody (2009) ‘Financial integration, capital mobility, and incomeconvergence’, Economic Policy, April 2009, p. 241–305.

Ahearne, Alan, Herbert Brücker, Zsolt Darvas and Jakob von Weizsäcker (2009) ‘Cyclical dimensions oflabour mobility after EU enlargement’, Bruegel Working Paper 2009/02

Ahearne, Alan, Juan Delgado and Jakob von Weizsäcker, 2008, A Tail of Two Countries, Bruegel PolicyBrief 2008/4

Åslund, Anders (2009) ‘How Latvia Can Escape from the Financial Crisis’, Presentation to the AnnualConference of the Bank of Latvia, Riga, October 1, 2009.

Bayoumi, Tamim, Jaewoo Lee and Sarma Jayanthi (2006) ‘New Rates from New Weights’, IMF StaffPapers Vol. 53, No. 2, p. 272-305.

Becker, Torbjörn (2009), ‘Latvia’s options – Internal versus external devaluation’, mimeo, Stockholminstitute of Transition Economics.

Blanchard, Olivier (2006) ‘Adjustment within the Euro Area. The Difficult Case of Portugal’, MITDepartment of Economics Working Paper No. 06-04

Buiter, Willem H. (2005) ‘To purgatory and beyond; When and how should the accession countries fromCentral and Eastern Europe become full members of the EMU?’ In: Breuss, Fritz and Hochreiter, Eduard,(eds.) Challenges for Central Banks in an enlarged EMU. European Community Studies Association ofAustria Publication Series . Springer-Verlag, New York, p. 145-186.

Cerra, Valerie and Sweta Chaman Saxena (2008) ‘Growth Dynamics: The Myth of Economic Recovery’,American Economic Review, Vol. 98(1), 439-57.

Darvas, Zsolt (2009a) ‘Treaty change is needed to make sense of euro-area entry criteria’, VoxEU.org

Darvas, Zsolt (2009b) ‘The Impact of the Crisis on Budget Policy in Central and Eastern Europe’, BruegelWorking Paper 2009/05, to appear in the OECD Journal on Budgeting.


24


Darvas, Zsolt (2009c), ‘The role of the EU in supporting crisis-hit countries in central and easternEurope’, background paper prepared for the Wold Bank’s ’Turmoil at Twenty’ volume, to appear as aBruegel Policy Contribution.

Darvas, Zsolt and György Szapáry (2008) ‘Euro area enlargement and euro adoption strategies’,European Economy – Economic Papers 304, Directorate-General for Economic and Financial Affairs,European Commission, Brussels. To appear as a chapter in: Buti, Marco, Servaas Deroose, Vitor Gasparand João Nogueira Martins (eds.), Euro-The First Decade, Cambridge University Press, forthcoming.

Darvas, Zsolt and Jean Pisani-Ferry (2008) ‘Avoiding a new European divide’, Bruegel Policy Brief2008/10

Darvas, Zsolt and Reinhilde Veugelers (2009) ‘Beyond the crisis: Prospects for transition economies’,mimeo, Bruegel.

De Grauwe, Paul (2009), ‘The politics of the Maastricht convergence criteria’, VoxEU.org

European Commission (2009) ‘Public Finances in EMU’, European Economy 5/2009 (ProvisionalVersion), Directorate-General Economic and Financial Affairs of the European Commission.

FitchRatings (2009) ’Dollarization in Ecuador: Regime Change?’ Sovereigns Ecuador Special Report,May 2009

International Monetary Fund (2009a) ‘Republic of Latvia: Request for Stand-By Arrangement—StaffReport’, IMF Country Report No. 09/3, January, Washington DC.

International Monetary Fund (2009b) ‘Estonia: Article IV Consultation’, March, Washington DC.

Kuodis, Raimondas and Tomas Ramanauskas (2009) ‘From Boom to Bust: Lessons from Lithuania’,Proceedings of OeNB Workshops No. 15 on ‘Recent Developments in the Baltic Countries –What Arethe Lessons for Southeastern Europe?’, p. 102-115, Oesterreichische Nationalbank.

Levy-Yeyati, Eduardo (2009), ‘Is Latvia the new Argentina?’, VoxEU.org

Lewis, John and Karsten Staehr (forthcoming), ‘The Maastricht inflation criterion: What is the effect ofexpansion of the European Union?’ Journal of Common Market Studies.

Mitra, Pradeep, Marcelo Selowsky and Juan Zalduendo (2009) ‘Turmoil at Twenty: Recession, Recovery,and Reform in Central and Eastern Europe and the Former Soviet Union’ World Bank, Washington DC

Official Journal of the European Communities (1998), L 139, Volume 41, 11 May 1998, http://eur-lex.europa.eu

Pisani-Ferry, Jean, Philippe Aghion, Alan Ahearne, Marek Belka, Jürgen von Hagen, Lars Heikensten andAndré Sapir, 2008, Coming of Age: Report on the euro area, Bruegel Blueprint.

RGE Monitor (2009) ‘Is Devaluation in Latvia Inevitable?’ 10 Sep 2009 http://www.rgemonitor.com

Saha, David and Jakob von Weizsäcker (2009) ‘EU stimulus packages’, Bruegel Policy Contribution2009/2

von Hagen, Jürgen and Jean Pisani-Ferry (2009) ’To the Commissioner for Economic and MonetaryAffairs: Priorities for your term of office’. In: Sapir, André (ed) ‘Memos to the New Commission - Europe'sEconomic Priorities 2010-2015’, Bruegel Blueprint

Documents

THE BALTIC CHALLENGE AND EURO-AREA ENTRYaei.pitt.edu/12220/1/PC_BalticsZsolt_011209_01.pdf · • The Baltic states’ previous huge current account deficits turned to surpluses by