Chapter

3. Financial Integration, Growth, and Imbalances

Author(s):
International Monetary Fund. European Dept.
Published Date:
May 2011
Share
  • ShareShare
Show Summary Details

In the run-up to the crisis, financial integration in Europe boosted investment and reduced saving in countries that previously had high interest rates. As capital inflows increasingly went into the nontradable sector and contributed to credit and housing booms, countries in the euro area periphery and countries in emerging Europe with fixed exchange rates built up large current account imbalances, with ultimately unsustainable trajectories of net external asset positions. Financial markets did not pay sufficient attention to these vulnerabilities, and policies did too little to address market failures. When capital flows slowed, the boom ended, and sharp recessions ensued. The absence of EU-wide institutions to deal with banking crises and the incomplete integration of capital markets compounded the crisis. To overcome the crisis decisively, the most critical factor in the longer run is restoring growth in the crisis-affected countries. To prevent new crises, more vigilance is needed, better institutions to deal with financial sector problems must be developed, and more, rather than less financial and economic integration is needed.

The establishment of the Economic and Monetary Union (EMU) in 1999 marked an important step toward financial integration in Europe. In 1999, 11 member states of the European Union (EU) adopted the euro as their common currency, and six more countries followed in the subsequent years.1 Several countries in central and eastern Europe (CEE)—notably the Baltic countries and Bulgaria—pegged their currencies unilaterally to the euro, thus tying their monetary policies to that of the European Central Bank (ECB).

Rising current account imbalances accompanied financial integration, and countries with high current account deficits were particularly hard hit by the 2008–09 financial crisis. In the euro area periphery, strong cross-border capital flows in the run-up to the crisis fueled credit, asset price, and domestic demand booms, which led to a surge in imports, rapid expansion of the nontradable sector, deterioration of competitiveness, and widening current account deficits. This pattern was also observed in countries with exchange rates pegged to the euro in anticipation of early euro area entry. When capital flows slowed, the domestic demand booms ended, and sharp recessions ensued.

The legacy of the boom years and subsequent crisis is likely to depress growth in the euro area periphery and countries with exchange rates pegged to the euro for some time. The debt overhang in the private sector and sharply deteriorated public finances will subdue domestic demand, while the erosion of competiveness during the boom is bound to depress exports. In the absence of labor mobility and exchange rate flexibility, and with limited wage and price adjustment capacity, turning these dynamics around is proving to be quite challenging.

This chapter first discusses the contribution of financial integration to rising current account imbalances before the crisis. It shows how financial integration led to sharp compression of interest rate differentials, which boosted investment and reduced saving in countries that previously had high interest rates, and how strong credit and housing booms led to massive current account deficits.

The chapter then reviews why the unwinding of the imbalances led to such a severe crisis. It will show that the widening of current account deficits ultimately was the result of an unsustainable and risky growth pattern, which went on for too long as markets paid insufficient attention to rising risks, and policies did too little to address these market failures.

The aggravation of the crisis by the absence of EU-wide institutions to deal with banking crises and by the incomplete integration of capital markets is also discussed. Although the EU has fostered financial integration by relaxing constraints, harmonizing various aspects of the financial system, and adopting a common currency, this process has been allowed to outpace development of the institutions necessary to support the single financial market. In the absence of EU-wide institutions, the approach to dealing with banking sector problems remained national throughout the crisis. Banking and sovereign debt problems have thus exacerbated each other, leading to vicious circles in the periphery.

Finally, the chapter discusses that while the crisis has led to a sharp adjustment of earlier current account imbalances, just dealing with imbalances is not enough. The most critical factor in the longer term is restoring GDP growth in the crisis-affected countries, with stronger roles for the tradable sector and exports, and less reliance on the nontradable sector, capital flows, and domestic demand. Ultimately, growth and convergence will need to be backed by productivity increases. To foster efficiency increases and prevent the reemergence of imbalances, better policies are needed at the national level, while better governance at the EU level would give teeth to such policies.

The chapter focuses on the original EMU members, Greece, the Baltic countries, and Bulgaria. These countries shared the euro, or had a hard peg to the euro, for at least five years before the start of the global crisis in 2007.2 To highlight the role of financial integration, as opposed to trade integration, developments in four central European countries (the Czech Republic, Hungary, Poland, and the Slovak Republic), which during the run-up to the global crisis all had flexible exchange rates, are compared with developments in the focus countries.

Imbalances and Crisis

The elimination of exchange rate risk in the wake of monetary integration led to rapid reductions in risk premiums and to interest rate convergence, particularly for money market and government bond rates (Figure 3.1). Progress in fiscal consolidation, aimed at meeting the Maastricht debt and deficit criteria, further reduced risk premiums. In the first half of the 1990s, the governments of Finland, Italy, Portugal, and Spain still paid spreads of 400 basis points or more over German bund rates, reflecting a history of frequent devaluations. In the run-up to the euro introduction, during 1995 to 1998, these spreads disappeared almost entirely, bringing immediate and tangible benefits in the form of lower funding costs for the public sector, corporations, and households. The process repeated itself for countries adopting the euro at later stages, as well as for the hard peg countries in emerging Europe that had tied their currencies to the euro. Interest rate convergence tended to be far less pronounced in the central European countries that kept flexible exchange rates, and for countries that were not members of the EU.3

Figure 3.1Selected EU Countries: Convergence of Long-Term Government Bond Rates, 1990–2010

(Percent)

Source: IMF, International Financial Statistics.

Monetary integration also encouraged larger cross-border financial exposures (Figure 3.2). In the banking sector, the share of interbank loans to banks within the euro area increased from 15 percent in the late 1990s to 25 percent in the late 2000s, and to 20 percent from 10 percent for interbank loans to banks in the EU but outside the euro area. In addition, home bias for investment funds’ allocations of equity and debt securities declined significantly.4

Figure 3.2Selected EU Countries: Indicators of Financial Integration, 1998–2008

(Percent)

Source: European Central Bank.

Note: MFI stands for monetary financial institutions.

Not all financial markets became equally integrated: by 2007, debt markets had become most integrated, while cross-border flows in foreign direct investment (FDI) and equity portfolio investment remained more limited. External debt liabilities were generally several times larger than FDI and external equity investment (Figure 3.3). Within debt markets, a number of countries’ cross-border holdings were largely confined to sovereign debt (Figure 3.4).

Figure 3.3Selected EU Countries: International Investment Position, Liabilities1

(Percent of GDP)

Sources: IMF, Balance of Payments Statistics database; and IMF, World Economic Outlook database.

1 For Ireland, 2001 instead of 1999 data have been used reflecting lack of data availability.

215th and 85th percentile.

Figure 3.4Selected EU Countries: Components of Debt Securities Liabilities1

(Percent of GDP)

Sources: IMF, Balance of Payments Statistics database; and IMF, World Economic Outlook database.

1For Ireland, 2001 instead of 1999 data have been used reflecting lack of data availability.

Cross-border mergers and acquisitions remained small compared with domestic mergers and acquisitions (Figure 3.5), and remained much lower than mergers and acquisitions across regions in the United States (Umber, Grote, and Frey, 2010). High concentration of corporate control (Becht and Mayer, 2000), regulatory differences (particularly in the application of takeover regulations), and predictability of national regulatory agencies seem to have been the most important barriers to cross-border equity flows, including in the banking sector in the EU (Koehler, 2007; and Campa and Moschieri, 2008).

Figure 3.5Selected EU Countries: Cross-Border Mergers and Acquisitions for Selected Target Countries, 2001–07

(Percent of completed deals)1

Sources: Thomson One Banker; and IMF staff calculations.

1Completed deals in which the target company and the acquirer company are based in the Europe 15 area (Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, Netherlands, Portugal, Spain, Sweden, and the United Kingdom). Deals include only those that imply a change of control of the acquired company (20 percent of ownership), but exclude those in which the acquirer already had control before acquisition.

The Widening of External Imbalances

The decline in interest rates boosted investment and reduced saving in countries where interest rates had previously been high. The impact was particularly pronounced in relatively poor countries because the expected rapid income growth there made borrowing more attractive (Figures 3.6 and 3.7, and Table 3.1).

Figure 3.6Euro Area: Current Account Balances in 2007 and Starting Positions in the 1990s

Sources: IMF, World Economic Outlook database; IMF, International Financial Statistics; and IMF staff calculations.

Figure 3.7Euro Area: Saving-Investment Balances, 2000–09

(Percent of GDP)

Sources: Eurostat; and IMF staff calculations.

Note: Surplus (deficit) countries are countries with a current account surplus (deficit) in 2007. Surplus countries include Austria, Belgium, Finland, Germany, and the Netherlands. Deficit countries include France, Ireland, Italy, Portugal, and Spain.

Table 3.1Selected EU Countries: Sectoral Saving-Investment Balances, 2000–07(Percent of GDP, unless otherwise indicated)
HouseholdsNonfinancial corp.Financial corp.General governmentTotal
SINetSINetSINetSINetSINet
2007
Surplus countries10.26.53.812.611.11.51.00.30.72.81.71.026.619.67.0
Netherlands6.97.5-0.517.09.37.71.50.41.23.43.30.128.820.48.4
Germany11.56.25.311.410.50.80.70.10.62.41.41.026.018.37.6
Finland3.97.7-3.714.912.52.50.70.30.47.62.55.127.122.94.2
Austria10.25.35.012.016.0-4.02.30.71.62.61.11.427.223.24.0
Belgium10.06.73.313.613.70.01.30.80.61.71.60.126.722.83.9
Deficit countries19.07.7-1.06.812.8-2.91.50.50.62.83.2-0.420.024.2-4.2
Italy10.26.93.36.312.2-5.91.30.40.92.32.30.020.121.9-1.8
France10.27.03.17.911.1-3.20.80.80.01.13.3-2.220.022.2-2.2
Ireland3.312.2-8.89.39.8-0.44.90.64.34.14.7-0.521.727.2-5.5
Spain6.99.7-2.85.217.2-12.02.10.12.06.94.02.821.031.0-10.0
Portugal4.95.8-0.95.613.6-8.02.61.01.6-0.52.4-2.912.722.9-10.2
Greece2-1.19.4-10.510.77.23.52.00.31.7-2.52.8-5.39.019.7-10.7
Hard peg-6.34.8-11.116.026.0-10.01.30.11.16.05.40.616.936.3-19.4
Lithuania-3.23.9-7.113.921.5-7.61.50.11.33.65.3-1.715.830.9-15.1
Estonia-0.98.5-9.415.526.1-10.6-0.2-0.1-0.17.65.12.522.039.6-17.7
Latvia-3.05.0-8.012.429.1-16.73.00.03.05.76.3-0.618.140.4-22.3
Bulgaria-14.93.5-18.421.028.0-7.00.70.40.37.64.92.714.336.8-22.5
2000
Surplus countries9.67.22.58.912.2-3.31.50.70.82.42.00.422.422.00.4
Netherlands6.97.0-0.113.810.43.43.11.41.74.53.11.428.422.06.4
Germany10.57.52.96.912.0-5.11.20.50.71.61.8-0.220.221.8-1.6
Finland4.46.6-2.314.711.63.10.10.2-0.19.32.46.828.520.97.6
Austria8.95.73.111.316.0-4.72.11.01.11.31.5-0.223.624.3-0.7
Belgium10.65.94.812.513.8-1.31.00.90.12.62.00.626.722.54.2
Deficit countries19.26.52.78.811.9-3.11.20.60.62.02.9-0.921.221.9-0.7
Italy9.96.53.48.611.3-2.80.80.60.31.32.3-1.020.620.7-0.1
France9.85.74.08.310.8-2.51.40.80.52.13.1-1.021.620.51.1
Ireland34.97.4-2.59.48.90.62.20.41.83.94.3-0.320.520.9-0.4
Spain7.57.40.110.315.3-5.01.50.41.13.03.2-0.122.326.3-4.0
Portugal7.58.8-1.28.015.4-7.41.70.71.00.53.7-3.117.828.5-10.7
Greece2.411.5-9.18.37.80.60.70.40.3-0.23.6-3.811.323.3-12.0
Hard peg-1.12.2-3.313.417.6-4.21.80.71.13.12.60.517.223.0-5.8
Lithuania4.53.70.95.612.5-6.91.30.41.01.52.4-0.813.018.9-5.9
Estonia2.33.4-1.016.021.2-5.11.50.31.23.53.8-0.323.428.7-5.3
Latvia1.51.40.214.419.9-5.52.71.11.60.21.4-1.118.923.7-4.8
Bulgaria4-9.30.8-10.118.819.2-0.41.60.80.86.12.93.217.323.8-6.5
Change (2007 over 2000)

(percentage points of GDP)
Surplus countries0.6-0.71.33.7-1.04.8-0.5-0.4-0.10.4-0.20.64.2-2.46.6
Netherlands0.00.4-0.43.2-1.14.3-1.6-1.1-0.5-1.10.2-1.30.4-1.62.0
Germany1.0-1.32.34.5-1.45.9-0.5-0.4-0.20.8-0.31.25.8-3.59.3
Finland-0.41.0-1.50.20.9-0.60.50.10.5-1.70.0-1.8-1.42.0-3.4
Austria1.4-0.41.80.70.00.70.2-0.30.51.3-0.41.63.6-1.14.7
Belgium-0.60.9-1.51.2-0.11.30.4-0.10.5-0.8-0.3-0.50.10.3-0.3
Deficit countries1-0.21.3-3.7-2.00.90.20.2-0.10.10.80.30.5-1.12.3-3.5
Italy0.30.4-0.1-2.20.9-3.20.5-0.20.61.00.01.0-0.51.2-1.7
France0.41.3-0.9-0.40.3-0.7-0.6-0.1-0.5-1.00.2-1.2-1.61.7-3.3
Ireland-1.64.7-6.3-0.10.9-1.02.70.22.40.20.4-0.21.26.3-5.1
Spain-0.62.3-2.9-5.11.8-6.90.6-0.30.93.80.93.0-1.34.7-6.0
Portugal-2.6-3.00.4-2.4-1.7-0.60.90.30.6-1.0-1.20.2-5.0-5.60.6
Greece-3.5-2.1-1.42.4-0.52.91.2-0.11.3-2.4-0.8-1.5-2.3-3.61.3
Hard peg-5.32.6-7.92.58.4-5.8-0.5-0.50.02.92.80.1-0.313.3-13.6
Lithuania-7.70.2-8.08.39.1-0.80.1-0.20.42.12.9-0.82.812.0-9.2
Estonia-3.35.1-8.4-0.64.9-5.5-1.7-0.4-1.34.11.32.8-1.411.0-12.4
Latvia-4.53.6-8.1-2.09.2-11.30.2-1.11.45.55.00.5-0.816.7-17.5
Bulgaria-5.62.7-8.32.28.8-6.6-1.0-0.4-0.51.52.0-0.5-3.013.0-16.0
Sources: Eurostat; IMF, World Economic Outlook database; and IMF staff calculations.Note: S denotes gross saving; I denotes gross investment; Net = S minus I. Country group averages are weighted with 2000 and 2007 nominal GDP weights.
  • Household balances deteriorated as household saving declined and residential real estate investment increased. Countries with housing price booms saw the largest deterioration in the saving-investment balance.

  • Corporate balances worsened as corporate saving declined, probably as a result of the increase in unit labor costs (see next section). In the hard peg countries, corporate investment booms also contributed significantly to the current account deterioration.

By contrast, in countries where income By levels and interest rate levels had already converged, financial integration did not further compress interest rates, and domestic demand remained much more muted. In fact, private sector balances improved, although partly for reasons not directly related to financial integration. Corporate saving increased because wage moderation led to an increase in the share of profits in national income. Germany had entered the euro area with an impaired competitive position, reflecting in part the overhang from German reunification, which took many years of internal devaluation to correct. An additional factor hurting its competitiveness was the convergence of interest rates in the wake of the euro’s introduction, which negated Germany’s comparative advantage of low funding costs. Corporate investment was weak, reflecting both weak domestic demand and outsourcing to suppliers in emerging Europe.

Current account balances in the EU started to widen as a result, peaking in 2007 (Figures 3.8, 3.9, and 3.10). Current account balances deteriorated sharply in Ireland, Spain, and the hard peg countries. By 2007, current account deficits in Greece, Portugal, Spain, and the hard peg countries exceeded 10 percent of GDP. By contrast, current account balances improved sharply in Austria, Germany, and the Netherlands.

Figure 3.8Selected EU Countries: Current Account Balances, 1999–2009

(Percent of GDP)

Sources: IMF, World Economic Outlook database; and IMF staff calculations.

Figure 3.9External Balances

(Percent of GDP)

Sources: IMF, World Economic Outlook database; and IMF staff calculations.

Note: Surplus (deficit) countries are countries with a current account surplus (deficit) in 2007. Surplus countries include Austria, Belgium, Finland, Germany, and the Netherlands. Deficit countries include France, Greece, Ireland, Italy, Portugal, and Spain.

Figure 3.10Euro Area 11: Current Account Imbalance Indicator, 1990–20091

(Percent of GDP)

Sources: IMF, World Economic Outlook database; and IMF staff calculations.

1 Sum of the absolute values of current account balances in euro area divided by aggregate GDP.

The convergence in interest rates not only fueled private sector spending, it also boosted government primary spending. The interest rate compression provided a substantial windfall for countries like Italy and Greece, allowing these countries to reduce their overall deficits while increasing primary spending (Figure 3.11). The boom in private sector domestic demand fueled a surge in government tax revenue, which created further space to boost primary expenditure (Figure 3.12). As a result, countries in the periphery saw a sharp increase in primary expenditure between 2000 and 2007—even though this was not visible in headline fiscal balances at the time.

Figure 3.11Euro Area 11: Changes in General Government Expenditure, 1995–2007

(Percentage points of GDP)

Sources: IMF, World Economic Outlook database; and IMF staff calculations.

Figure 3.12Selected EU Countries: Real Domestic Demand, Government Revenue, and Expenditure Growth, 2000–07

(Percent)

Sources: IMF, World Economic Outlook database; and IMF staff calculations.

In general, the change in fiscal balances played only a modest part in increasing external imbalances. The improvement in headline balances was the real problem, because it disguised deterioration in the underlying fiscal situation. Interest savings and what turned out to be the temporary revenues had been used to boost primary expenditure.

Shocks external to the EU and euro area also contributed to rising imbalances. The integration of China into the world economy benefited Germany, which exported high-end machinery, but hurt southern Europe (Chen, Milesi-Ferretti, and Tressel, forthcoming). The integration of CEE countries into Europe boosted German firms’ productivity as they set up production platforms in the region, but competed with other countries’ exports (Marin, 2010). Similarly, the appreciation of the euro between 1999 and 2008 had a bigger impact on some countries in the periphery, for which trade with countries outside the euro area is very important, than on countries in the core, for which trade with other euro area countries is more important.5

The Crisis

The credit booms in Europe came to a sudden end in September 2008 after Lehman Brothers filed for bankruptcy. All types of capital flows reversed in the fall of 2008, but cross-border banking flows experienced the most severe retrenchment (Milesi-Ferretti and Tille, 2011). As risk aversion among investors rose sharply and equity markets plunged, many advanced-country banks, when confronted with liquidity and capital shortages, sharply curtailed new lending or even deleveraged.

In a change of strategy, they advised their subsidiaries and branches in emerging Europe that new credit would henceforth need to be financed solely by increases in local deposits (Figure 3.13).6

Figure 3.13Europe: Bank Exposure, 2003–10

Sources: Bank for International Settlements; IMF, World Economic Outlook database; and IMF staff calculations.

Note: EA4 comprises Greece, Ireland, Portugal, and Spain.

In the euro area periphery, Ireland and Spain suffered wrenching private sector adjustments. Before the crisis, Ireland and Spain had both experienced investment booms, which now collapsed. In Portugal and Greece, whose investment paths had been stable or declining, the impact of the crisis was initially less severe. Subsequent stress in the public sector affected them more (Figure 3.14, Table 3.2).

Figure 3.14EA4: Saving-Investment Balance, 1999–2009

(Percent of GDP)

Sources: IMF, World Economic Outlook database; and IMF staff calculations.

Table 3.2Selected EU Countries: Current Account Balances and Real Domestic Demand, 2003–10
Precrisis VulnerabilitiesAdjustments During Crisis
Current Account Balance,

2007 (percent of GDP)
Real Domestic Demand

Growth, 2003–07 (percent)
Change in Current

Account Balance, 2007–10

(percent of GDP)
Change in Real Domestic

Demand Growth, 2007–10

(percent)
Euro area countries with high current account deficits
Greece-14.415.83.9-10.0
Ireland-5.327.14.6-22.9
Portugal-10.17.30.2-1.0
Spain-10.020.75.5-7.6
Hard peg
Bulgaria-30.243.929.5-11.3
Estonia-17.242.520.8-29.3
Latvia-22.362.825.9-35.4
Lithuania-14.653.116.4-20.5
Sources: IMF, World Economic Outlook database; and IMF staff calculations.

The hard peg countries in emerging Europe, where imbalances had been most pronounced, were also hit hard.7 Domestic demand plunged, further exacerbated by a collapse in housing prices, and GDP contracted sharply, leading to a steep rise in unemployment. The CEE countries that during the boom years had flexible exchange rate regimes suffered a less severe recession. These countries had not experienced the same size credit boom and had much lower current account deficits, and went through much smaller forced adjustments (IMF, 2010b).

Concerns about the public sector soon exacerbated the crisis (Figure 3.15). The decline in domestic demand contributed to a severe drop in government revenue, and fiscal balances deteriorated sharply. Risk premiums in the periphery surged, even in countries that had entered the downturn with low deficits and debt (Figure 3.16). A vicious circle emerged: while the sovereign debt problems worsened as a result of the fiscal costs of banking problems, the concerns about the public sector exacerbated the problems for the private sector and financing costs for both went up in tandem (Figure 3.17).

Figure 3.15Selected EU Countries: Change in Fiscal Balance, 2007–10

(Percentage points of GDP)

Sources: IMF, World Economic Outlook database; and IMF staff calculations.

Figure 3.16Selected Countries: Five-Year CDS Spreads, January 2007–April 2011

(Basis points)

Source: Bloomberg, L.P.

Figure 3.17Selected EU Countries: Change in Sovereign and Bank Credit Default Swap Spreads, January 2010–March 2011

Sources: Bloomberg L.P.; and IMF staff calculations.

Note: Data points not labeled are for Austria, Denmark, Norway, and Switzerland.

Why Was the Crisis So Severe?

Why did the widening of current account imbalances culminate in such a severe crisis? In itself, the widening of external imbalances was not surprising. Capital can be expected to flow from richer to poorer countries, where the marginal productivity of capital is higher. Indeed, current account positions in the late 2000s were closely linked to both per capita income and interest rate differentials before monetary integration (Figure 3.6).8 Nor are higher current account deficits always a concern. To the extent that inflows are used to expand production, especially export capacity, current account deficits should be temporary and matched by the increased future capacity of the recipient countries to service their debts. As exports subsequently grow, imbalances should decline.

However, these large capital inflows fueled an unbalanced and unsustainable growth pattern:

  • Growth in the current account deficit countries was increasingly driven by sectors such as construction and financial intermediation with small tradable components, while growth in the surplus countries was tilted toward the tradable sector, especially industry (Table 3.3). As a result, the surplus countries saw sizable increases in their export-to-GDP ratios (Figure 3.18), while countries with small current account deficits experienced much smaller increases, and countries with very high current account deficits became even more closed in the 2000s.

  • Strong growth in the nontradable sector, in turn, contributed to rising wages, which put profitability in the tradable sector under pressure (Figures 3.19, 3.20, 3.21) and made the current account deficit countries less attractive for FDI.

  • Investment in the nontradable sector received a further boost from its relatively closed nature and the failure to prevent bubbles. With competition rather limited, investors in the nontradable sector enjoyed rents unavailable in the tradable sector, which was fully exposed to the harsh winds of global markets. Profitability in the nontradable sector jumped as asset price bubbles developed unchecked in key subsectors such as construction (Figures 3.22 and 3.23).

  • With incentives stacked toward the nontradable sector, investment took off and foreign capital flowed in. Beginning in 2002–03, the share of FDI in the current account deficit countries declined while bank and portfolio inflows surged (Figure 3.24). As the capital stock in the nontradable sector grew, marginal productivity of capital declined over time (Table 3.4).

  • This pattern of growth led to a steady widening of current account balances, which resulted in large changes in net external asset positions (Figure 3.25)—an ultimately unsustainable trend.

  • As current account deficits widened, countries became increasingly dependent on continuing capital inflows, and a sudden stop of capital inflows could cause a large-scale financial disruption, with a severe impact on growth.

Table 3.3Selected EU Countries: Growth in Value Added and Contribution by Sector, 2000–07(Percent)
Surplus CountriesDeficit CountriesHard PegCentral Europe
Total13.014.863.234.0
Agriculture0.0-0.1-0.70.6
Industry4.31.814.114.4
Construction-0.61.36.81.1
Trade, transport, and communication3.33.223.69.4
Financial intermediation, and real estate4.55.714.66.3
Other services1.62.84.82.6
Sources: Eurostat; and IMF staff calculations.

Figure 3.18Selected EU Countries: Exports of Goods, 1995–2009

(Percent of GDP)

Sources: IMF, World Economic Outlook database; and IMF staff calculations.

Figure 3.19Euro Area: REER (ULC Manufacturing Based), 1995–2009

(Index, 1999 = 100)

Sources: European Commission; and IMF staff calculations.

Note: Aggregation is weighted by GDP.

Figure 3.20Selected EU Countries: Appreciation of REER (ULC Manufacturing Based), 2000–071

(Percent)

Sources: European Commission; and IMF staff calculations.

1Real effective exchange rates are relative to the rest of EU27 and based on quarterly data.

2BLEU stands for Belgium-Luxembourg Economic Union.

Figure 3.21Selected EU Countries: Appreciation of REER (ULC Manufacturing Based) Versus Change in Corporate Saving to GDP, 2000–07

Sources: European Commission; Eurostat; and IMF staff calculations.

Figure 3.22Selected EU Countries: Nominal House Prices, 2000:Q1–2010:Q3

(Index, 2000:Q1 = 100)

Sources: Bank for International Settlements; national authorities; and IMF staff calculations.

1For Finland (panel 1) and Greece (panel 2), 2006:Q1 = 100.

Figure 3.23Selected EU Countries: Profitability by Sector, 2000–071

(Index, 2000 = 100)

Sources: EU KLEM database; and IMF staff calculations.

1Profitability is computed as value added deflator minus wage rate plus gross value added per hours worked (or an inverse of labor productivity).

2 Data for Portugal are for 2000–06.

Figure 3.24Selected EU Countries: International Investment Position, 2000–09

(Billions of U.S. dollars)

Sources: IMF, International Financial Statistics; and IMF staff calculations.

Note: Surplus (deficit) countries are countries with a current account surplus (deficit) in 2007. Surplus countries include Austria, Belgium, Finland, Germany, and the Netherlands. Low-deficit countries are countries with a 2007 current account deficit less than 5 percent of GDP and include France and Italy. High-deficit countries are Greece, Ireland, Portugal, and Spain.

Figure 3.25Selected EU Countries: International Investment Position, 2000–07

(Percent of GDP)

Sources: IMF, International Financial Statistics; and IMF, World Economic Outlook database.

Note: Net International Investment Position is defined as the difference between total external assets and total external liabilities.

12001 instead of 2000.

Markets Underestimated Risks

Before the recent global crisis, markets tended to underestimate these risks. This pattern was also observed in countries with exchange rates pegged to the euro in anticipation of early euro area entry. Risk premiums remained low, so current account imbalances became wider and more persistent, thus contributing to large changes in net external asset positions. Markets showed little concern until mid-2007, ignoring solvency risk and covering large financing needs at low interest rates, which boosted self-feeding bubbles in nontraded sectors. By failing to demand higher risk premiums, markets failed to rein in this unhealthy development until it was too late for a soft landing. This situation is not unique to the European debt crisis but follows a pattern familiar from other crisis cases, including the Latin American debt crisis of the early 1980s, the Asian crisis of the late 1990s, and the U.S. subprime crisis of the late 2000s.

One reason markets underestimated risks may have been the expectation that euro area banks and governments would be bailed out. Markets found it difficult to imagine that the euro area countries would not come to the rescue of members in trouble—particularly given the interconnectedness of their banking systems. The regulatory environment also considered all sovereign bonds to be safe assets.

Yet the underestimation of risk was not confined to Europe. The belief was widely held that macroeconomic volatility had declined and the central problem of depression prevention, for all practical purposes, had been solved (the “Great Moderation”). Large capital flows seemed to carry few risks when crisis was seen as a remote possibility.

Economic Policies Failed to Address Imbalances

Economic policies failed to sufficiently correct market failures and check overoptimistic expectations.

  • With interest rates set at the euro area average and markets not differentiating between countries, the credit boom became difficult to stop once it had started. Indeed, as the economy heated up and inflation and wages started to rise, high inflation led to negative real interest rates, further boosting demand for credit (Figure 3.26). In emerging Europe, and as discussed in the October 2010 Regional Economic Outlook, the new EU member states that had fixed their currencies to the euro at an early stage (Bulgaria, Estonia, Latvia, and Lithuania) all experienced credit and domestic demand booms and very high current account deficits, while countries that retained greater monetary independence generally experienced less pronounced imbalances (Box 3.1).

  • However, other policy instruments to curb domestic demand and excessive expansion of the nontradable sector were used insufficiently. Tighter fiscal policy could have dampened domestic demand.9 Although fiscal balances in the periphery were, on average, not in worse shape than in the core (with the exception of Portugal and Greece, Figure 3.27), this situation primarily reflected savings on interest payments and revenues that turned out to be temporary.

  • Stronger macroprudential regulation would have required financial institutions to build up larger capital buffers and provisions in good times, which would have raised the cost of capital and slowed the expansion of financial sector balance sheets, or at least made banks less vulnerable in the subsequent bust.10

Figure 3.26Selected EU Countries: Inflation, June 2008

(Year-over-year, percent)

Source: Eurostat.

Figure 3.27Selected EU Countries: Fiscal Balance, 2007

(Percent of GDP)

Source: IMF, International Financial Statistics.

While credit excesses, housing booms, competitiveness losses, and lack of fiscal discipline all played a role, their contributions were country specific. Greece suffered most from the lack of fiscal discipline. Ireland and Spain experienced a lack of political fortitude to use fiscal policy and macroprudential tools to manage credit and housing cycles.11

Why Is Resolution of the Crisis So Protracted?

The boom-bust cycle in the euro area periphery was not unique. During the boom years, countries throughout the world experienced credit and asset price escalations. Stark differences across regions also occurred in other monetary unions. In the United States, Arizona, California, Florida, and Nevada went through severe boom-bust cycles, while other states were much less affected.

The crisis in Europe was prolonged by the incomplete integration of the financial system and the absence of a centralized mechanism to deal with it. The result was a vicious circle in which sovereign debt and banking sector problems aggravated each other.

Financial Integration Is Incomplete

Financial integration in the EU and the euro area is still incomplete and uneven. Some elements of the financial system are highly integrated: capital flows cross borders with little impediment, and banks transact freely in the money market. But home bias in portfolio allocations remains, securitization is very much a national affair (for example, no uniform mortgage contract exists), and cross-border retail banking is virtually nonexistent—cross-border provision of retail financial services amounts to less than 1 percent of total loans. Apart from some regional clusters, cross-border mergers and acquisitions are still limited in most European banking markets, with foreign acquisitions accounting for only 20 percent of banking activity, compared with 40 percent in other sectors.12For the euro area as a whole, foreign bank participation in domestic markets amounted to 27 percent in 2009 (Figure 3.28).

Figure 3.28Selected EU Countries: Foreign Bank Presence

(Percent of assets)

Source: ECB Report on EU banking structures, consolidated banking data.

Note: Foreign bank presence is the total assets of foreign banks’ branches and subsidiaries as a percent of the total assets of the banking sector.

1Data from 2005.

Perceptions of a lack of cross-border synergies played a role, but differences in business and regulatory environments seem to be important factors explaining the incomplete financial integration in the euro area. The absence of a unified legal framework is a clear obstacle but so is the desire to maintain national champions, build niche activities, take advantage of regulatory and supervisory arbitrage, and retain full national accountability for explicit and implicit guarantees of the banking system. It is not surprising, then, that the efficiency benefits conferred by heightened competition in a single market have been limited at best and have recently suffered a setback (Box 3.2).

Table 3.4Euro Area: Output-Capital Ratios, 2000–07(Percent)
Average1Marginal2
200020072000–07
Italy0.440.390.16
Spain0.440.370.23
Portugal0.580.520.24
France0.460.420.26
Netherlands0.410.400.32
Germany0.410.400.32
Belgium0.390.380.32
Greece0.430.400.34
Austria0.380.370.35
Ireland0.650.620.56
Finland0.420.450.67
Sources: Organization for Economic Cooperation and Development, Economic Outlook database; and IMF staff calculations.

As a result, banking flows to the euro area periphery during the boom years largely took the form of debt rather than equity, which exposed banks in the periphery to rollover risk. By contrast, in emerging Europe, with fewer implicit barriers to bank expansion, most of the bank flows went from parent banks to their local subsidiaries. Although nominally in the form of debt, in practice the flows were more like equity—parent banks could not suddenly withdraw funding from their subsidiaries because such an action would lead to liquidity shortages in the subsidiary and likely intervention by supervisors. The more stable funding structure benefited banks in emerging Europe during the crisis, and may have been one of the reasons that banking crises in the region were largely avoided.

Institutions Supporting the Single Financial Market Are Still Being Built

The EU has fostered financial integration by relaxing constraints (for example, the single passport for cross-border banking) and harmonizing various aspects of the financial system, and by adopting a common currency in the euro area, but this process has been allowed to run ahead of the institutions necessary to support the single financial market. No effective instruments were put in place to detect and handle cross-border risks or mitigate the buildup of imbalances financed by cross-border financial flows.

As a result, when the crisis hit, reactions were mostly national, with no regard for the cross-border implications of such policies (for example, guarantees of liabilities, bans on short selling, and proposals for bail-ins). Arrangements to deal with cross-border banks broke down over burden-sharing issues. Throughout the ongoing resolution of the crisis, the approach to banking sector problems has remained national.

EU policymakers have responded to the crisis by setting up new institutions. The European Systemic Risk Board (ESRB) will look into macroprudential risks to spot credit developments before they give lead to unsustainable imbalances, while the new European Supervisory Authorities should strengthen and harmonize regulation and supervision.13

But dealing effectively with interconnected financial institutions operating in a single financial market requires a pan-EU, or at least a euro area-wide, approach, a step beyond the currently planned coordination. There is little alternative to harmonization of regulations and supervisory practices at the EU level and an approach to crisis management and resolution that employs a pan-EU backstop. Only then will location no longer matter for financial institutions’ costs of and access to funding, and only then will the financial system be decoupled from the sovereign, a link that in the current crisis has proved to be detrimental in both directions.

Box 3.1Why Did the Currency Board Countries in Emerging Europe Have Such Large Imbalances?

Selected EU Countries: Composition of Growth, 2000–07

Percentage points)

Sources: IMF, World Economic Outlook database; and IMF staff calculations.

As discussed in Chapter 3 of the October 2010 Regional Economic Outlook: Europe (IMF, 2010b) the large capital flows that went to emerging Europe in the years before the global economic crisis did not affect all countries equally. In some countries, current account deficits widened to over 15 percent of GDP and inflation reached double digits, while in others imbalances remained more contained.

One important factor in these differences was the exchange rate regime. The largest imbalances occurred in new EU member states with hard currency pegs to the euro (Bulgaria, Estonia, Latvia, and Lithuania). In most new member states with more flexible exchange rate regimes, imbalances were milder (the Czech Republic, Hungary, Poland, and the Slovak Republic). Imbalances were also less in countries with fixed exchange rates that were not yet EU members (Bosnia and FYR Macedonia). Capital flows into these countries were lower, probably partly because of the memory and legacy of various conflicts in the region, and partly because they were not yet in the EU).

Why did imbalances in new member states with hard pegs become so much larger than in new member states with more flexible exchange rate regimes?

  • From a demand perspective, a boom in domestic demand drove GDP growth in countries with fixed exchange rates (see figure). These countries saw sharp increases in their domestic demand to GDP ratios, while the exports to GDP ratios increased much less. In countries with floating exchange rates, the growth pattern was more balanced.

  • From a supply perspective, GDP growth was driven to a large extent by the nontradable sector. Countries with more flexible exchange rates had much stronger growth in manufacturing. These differences reflect differences in capital inflows. In the Baltics and Bulgaria, capital flows largely went to the nontradable sector, whereas in the countries with more flexible exchange rates, a much larger share went to the tradable sector.

When faced with large capital inflows, the fixed exchange rate regime countries had few instruments at their disposal to stop the credit boom. Indeed, they experienced a vicious circle. As the economy heated up and wage growth and inflation increased, real interest rates dropped, further boosting demand for credit. Sharp wage increases led to a deterioration of the tradable sector’s competitiveness, which led to a reduction in the growth of exports. Countries with more flexible exchange rates did not experience the same vicious circle—they could keep real interest rates higher by letting the nominal exchange rate appreciate, which mitigated the credit boom.

Note: The main author of this box is Jeta Menkulasi.

More Complete Financial Integration Would Have Helped Resolve the Crisis

How would this deeper integration fortified by proper institutions have helped in resolving the crisis?

  • First, banks would have suffered less spillover from sovereign debt trouble. Deposit guarantees or any other implicit guarantees would not have depended on the sovereign signature, which would have kept funding costs for banks in euro area periphery countries in check.

  • Second, it would have given the EU more options for dealing with weaknesses in the banking system. If domestic markets had been more open to foreign banks, national public sector policies for supporting and recapitalizing banks—which has perpetuated the adverse feedback between banks and sovereign—would not have been the only viable measures.14

  • Third, it would have facilitated consolidation of the financial sector typical in the aftermath of a crisis and helped avoid the tendency to ring-fence national systems. Consolidation is now occurring slowly, if at all, and often within borders. In many cases, restructuring has led to refocusing on the domestic market and sales of foreign operations, thus reducing financial integration. Similarly, national authorities, mindful of their taxpayers, are ring-fencing operations under their purview, thus diminishing the benefits of the single market.

  • Finally, had a pan-EU supervisory regime been in place, excessive exposures or expansions of banking systems well beyond the fiscal capacity of the sovereign may have been spotted and ill-considered unilateral policy moves avoided.

Experience in the United States illustrates that more centralized crisis resolution mechanisms and more integrated capital markets can make a difference (Box 3.3). The United States also went through a severe boom-bust cycle, with large differences across regions. Although the U.S. fiscal deficit is now higher than that of the euro area countries, and some individual states have come under significant fiscal pressures, the U.S. capital market has not disintegrated along state lines.

In sum, increased financial integration will improve risk sharing and should help stabilize national demand, contributing to the resolution of imbalances and helping prevent their reemergence. But it will need to be accompanied by robust institutional arrangements and backstops at the EU or euro area level. Finally, financial integration will remain imperfect for some time, so fiscal and structural policies will also need to be implemented to facilitate adjustment to shocks under a common monetary policy.

Box 3.2Recent Developments in Bank Competition in Europe

Globalization and financial integration have changed the business environment in which banks operate. The impact has been profound in Europe, supported by the introduction of the euro, significant liberalization, and harmonization of financial sector regulation as policymakers pursue the goal of a single market. A fully integrated financial market promotes innovation and competition and ensures efficient provision of financial goods and services. At the same time, the global financial crisis has made clear that harmonized and well-designed regulation and prudent supervision of the banking sector are indispensible for preventing excessive risk taking. The crisis took a heavy toll on European banks and required the introduction of bank support schemes to avoid disorderly bank failures that could have pushed the financial sector into collapse. With advice from the ECB and the European Commission, national authorities implemented support schemes for their domestic banks. The Commission had to balance the need for state aid on the one hand with safeguarding competition and avoiding market segmentation along national borders on the other. Whether this has been achieved is a key question for policymakers. A competitive and sufficiently integrated banking sector should facilitate adjustment and the needed recapitalization, and cross-border restructuring through mergers and acquisitions is an essential element.

The level of bank competition can be measured in various ways. Historically, inferences about the level of competition relied on structural characteristics such as concentration ratios for the largest banks, or overall profitability. However, high capital returns may simply reflect high efficiency instead of monopolistic rents. Moreover, structural features could be less relevant in gauging market competition in very open, globally integrated, and highly contestable markets. Unrestricted entry and exit of firms, including those from abroad, may force banks to act competitively even if their numbers are small. Indeed, Claessens and Laeven (2004) show for a large cross-country sample that being open to new entry is the most important competitive pressure; they find no evidence that banking system concentration is negatively associated with competitiveness.

The H-statistic (H) developed by Panzar and Rosse (1987) is a widely used measure for competitive behavior that relies on market contestability and potential competitors. It focuses on the degree to which changes in the costs of inputs lead to subsequent changes in revenues. Under a monopoly, an increase in input prices will increase marginal costs and reduce equilibrium output and total revenues (H < 0). Under perfect competition, an increase in input prices will be fully passed on to the output price (H = 1). Positive values less than 1 imply some degree of monopolistic competition. Estimation results for the periods 1995–2000, 2001–07, and 2008–09 show the following:1

  • In the period shortly after the introduction of the euro (2001–07), competitive bank behavior broadly converged across euro area member countries to a lower overall level and closer to the U.S. level.

  • In the aftermath of the financial crisis, several European countries and the euro area as a whole saw a further small but statistically significant deterioration in their levels of banking competition. However, the overall deterioration in competition in Europe appears no worse than the deterioration in the United States.

The finding that large and financially integrated countries or regions tend to exhibit less competitive behavior than smaller regions exhibit is in line with other studies, including Bikker and Spierdijk (2008), who also find some deterioration in competitive behavior over time for Europe’s banks. They argue that banks in large and integrated financial markets are pushed by rising capital market competition and tend to shift from traditional intermediation to more sophisticated and complex products associated with less price competition. However, the further deterioration in competitive pricing observed in the aftermath of the crisis is unlikely to yet reflect structural market changes and evolving bank business models. Instead, significant bank losses and the subsequent need for recapitalization may have temporarily limited competitive pressures. Finally, little evidence yet suggests that national support schemes in Europe have further hindered competition. Nevertheless, the European Commission and other regulatory agencies should continue to insist on bold restructuring and balance sheet repair in instances in which state aid has been granted with a view toward maintaining a level playing field for healthy and competitive banks with sound business models.

Bank Competition Measured by H-Statistic over Time
1995–20002001–20072008–2009Change
Before and After EMUBefore and After Crisis
Country/regionH-StatisticStandard

Error
H-StatisticStandard

Error
H-StatisticStandard

Error
ΔHStandard

Error
ΔHStandard

Error
(1)(2)(3)(4)(5)(6)(7)=(3)–(1)(8)(9)=(5)–(3)(10)
Austria0.580.040.600.030.710.100.020.040.100.11
Finland0.800.100.550.060.650.05-0.25**0.120.100.08
France0.640.010.580.020.630.05-0.05***0.020.040.05
Germany0.430.010.450.010.360.020.020.02-0.09***0.02
Greece0.820.080.520.060.390.10-0.3***0.10-0.130.11
Ireland1.020.160.750.070.590.13-0.270.17-0.170.14
Italy0.880.010.590.010.500.03-0.29***0.02-0.09***0.03
Netherlands0.900.160.410.060.610.08-0.49***0.160.20**0.10
Portugal0.710.040.680.050.850.17-0.030.070.170.18
Spain0.700.030.800.030.510.050.09**0.04-0.29***0.06
United Kingdom0.510.040.650.030.620.050.14***0.04-0.030.05
United States0.310.010.430.000.270.010.12***0.01-0.16***0.01
Euro area0.700.010.520.010.440.01-0.18***0.01-0.07***0.01
Source: IMF staff calculations. For more detail, see Sun (forthcoming).Note: *** and ** indicate significance at the 1 and 5 percent level, respectively.
Note: The main author of this box is Thomas Harjes.1 Postcrisis estimates only provide preliminary evidence in view of the limited number of observations and the fact that structural changes in the aftermath of the crisis may have distorted the long-term market equilibrium in some countries, which could invalidate the H-statistic.

Box 3.3Crisis Resolution and Financial Integration in the United States

Like Europe, the United States experienced large regional differences in the housing boom-bust. California, Florida, Nevada, and Arizona went through the most severe cycles. In these states, housing prices more than doubled from 2000 to their peak, and then declined sharply—in Nevada by more than 50 percent. In other states, the amplitude of the cycle was not as large. These differences in the housing boom-bust were associated with stark differences in the severity of the recession (first three figures). Between late 2006 and late 2010, the unemployment rate rose by 10 percentage point in Nevada, and only ½ percentage point in North Dakota.1 Differences in unemployment rate levels were similarly stark—ranging from 4 percent in North Dakota to 14½ percent in Nevada.

However, unlike in Europe, financial markets in the United States remained fully integrated. Some states experienced sometimes severe fiscal pressures,2 but there were no concerns that these pressures would affect the financial sector or reduce market access for the private sector, and there was no differentiation in private sector financial market access by state.

United States: Housing Boom-Bust, 2000–10

Sources: Bureau of Economic Analysis; Federal Housing Finance Agency; and IMF staff calculations.

United States and Selected EU Countries: Boom and Bust, 2003–09

Sources: Bureau of Economic Analysis; Bureau of Labor Statistics; Federal Housing Finance Agency; IMF, World Economic Outlook database; and IMF staff calculations.

Note: Annual house prices are based on averages of seasonally adjusted quarterly indices. Unemployment rates are seasonally adjusted.

This marked difference between the United States and Europe is likely the result of the different financial sector crisis resolution mechanisms, the better integration of capital markets, and the larger role of the federal government. With much smaller state debts, and with bail-out responsibility assigned to the federal level, concerns about the public finances of individual states do not spill over to the private sector.

  • In Europe, individual countries were responsible for rescuing insolvent banks, while in the United States, financial bailouts took place at the federal level.3 Thus, in Ireland bank support weighed on public finances, while in the United States the bailouts of AIG and investment banks were done at the federal level rather than by the state where the bank was headquartered.

  • In the United States, failing banks often are taken over quickly by other banks—including from other states—while in Europe such takeovers of problem banks have been uncommon.

  • In the United States, state debt is relatively low—in 2008, federal debt held by the public was about 2½ times as large as debt issued by states.4 Constitutional and statutory provisions in many states limit debt issuance, and most states are governed by balanced budget provisions (last figure, top panel).5

  • In the United States, the borrowing flexibility and safe haven status of the federal government provides a backstop for the individual states during bad times. A sizable part of the federal stimulus went to states, helping to prevent sharp spending cuts at the state level. More broadly, unlike in Europe, the much larger role of the central fiscal authority in the United States may have a stabilizing effect on individual states. Spending by the federal government is about as large as spending of individual states (last figure, lower panel).

United States and Selected EU Countries: Unemployment Rates, 2006–10

Sources: Bureau of Labor Statistics; Eurostat; and IMF staff calculations.

Note: Unemployment rates are seasonally adjusted.

United States and Selected EU Countries: Debt and Expenditure, 2008

(Percent of GDP)

Sources: Bureau of Economic Analysis; dXdata; IMF, World Economic Outlook database; and IMF staff calculations.

1 Fiscal year basis. Debt held by the public (IMF, 2010d).

Note: The main author of this box is Lone Christiansen.1 North Dakota also benefited from an oil industry boom.2 In California, the recession’s negative effect on revenues together with the strict balanced budget provision led to the issuance of IOUs in lieu of some payments in the summer of 2009. The budget approval process was further complicated by the requirement of a two-thirds super majority of votes to pass the budget bill.3 The federal government stepped in to save Bank of America, Wells Fargo, U.S. Bancorp, PNC Financial, and a host of others. In September 2008, the Treasury was given authority to use $700 billion under the Troubled Asset Relief Program to purchase assets, make equity investments and loans, and provide asset guarantees in a range of financial institutions and markets. At the height of the crisis, the Treasury also guaranteed more than $3 trillion in assets to prevent runs on money market mutual funds. The Federal Deposit Insurance Corporation extended the coverage of insured deposits to $250,000, provided an unlimited guarantee of transactions deposits, and guaranteed new bank debt issues. The Federal Reserve provided a range of liquidity support to depository institutions, securities dealers, select foreign central banks, and key markets, and conducted unconventional large-scale asset purchases to support the housing sector and the economy. See IMF (2010c).4 By 2010, outstanding federal debt had increased to 60 percent of GDP, suggesting that the ratio increased even further. No data on 2010 individual state debt are available yet.5 Some of these provisions are implicit rather than explicit: “some balanced budget requirements are based on interpretations of state constitutions and statutes rather than on an explicit statement that the state must have a balanced budget” (National Conference of State Legislatures, 2010).

Box 3.4How Far Have Current Account Imbalances Adjusted?

To gauge how far rebalancing has come since the onset of the financial crisis, the macroeconomic balance approach is used to compare actual current account balances with “equilibrium” current accounts, which are determined as a function of fundamental factors.1 This notion of “equilibrium” measures levels of the current account that are consistent with underlying macroeconomic conditions and does not explicitly address the sustainability of current account positions.

A simple empirical model was used, based on the Consultative Group on Exchange Rate Issues methodology.2 In this model, current accounts as ratios to GDP are determined as a function of the initial net foreign asset position, the fiscal balance, the per capita growth rate, the level of per capita GDP, the oil trade balance, and key demographic features (see Annex for technical details). This approach confirms that current accounts moved away from their equilibrium values in the period immediately before the global crisis (first figure, first and second panels).

During 2001–04, most actual current account balances were not significantly different from their estimated equilibrium values. Reflecting comparatively older populations (and, in some cases, natural resources and importance as financial centers) a number of countries, including Austria, Belgium, Denmark, Germany, and the Netherlands were in surplus, whereas countries in the southern periphery of the euro area (Greece, Portugal, and Spain) were showing a modest deficit, and new EU member states (notably the Baltics, Hungary, Romania, and the Slovak Republic), characterized by much lower GDP levels, displayed somewhat larger deficits. This juxtaposition is consistent with convergence theory, according to which capital should flow downhill to equilibrate its relative return (Abiad, Leigh, and Mody, 2007).

Subsequently, during 2005–08, current accounts moved away from their estimated equilibrium values in a significant number of cases. Deficits in the periphery of the euro area and new member states became larger than could be explained by equilibrium models. In the periphery of the euro area, the gaps between the deficits and their estimated equilibrium intervals ranged from about 1 percent of GDP in Portugal to nearly 5 percent of GDP in Spain. Among the new member states, the gaps ranged from about 1 percent of GDP in Estonia and Lithuania to more than 9 percent of GDP in Bulgaria. Overoptimistic expectations of rapid income convergence together with underestimations of risk and policies that supported the allocation of resources to the nontradable sector, underpinned these developments (see main text). Conversely, in a few countries (notably Germany) surpluses began to exceed equilibrium values. While these diverging current account developments were not mirror images inside the euro area or the EU, they were financed largely by intra-euro area or intra-EU capital flows (Chen, Milesi-Ferretti, and Tressel, forthcoming).

Current accounts began moving back toward equilibrium as a consequence of the global crisis. The return of risk aversion and risk differentiation sharply curbed cross-border capital flows. The resulting contraction in private and public domestic demand led to an acute reduction of current account deficits. Current account balances in surplus countries also narrowed as exports declined and domestic demand held up better than in the deficit countries.

Estimated Ranges for Current Account Norms and Actual Current Accounts

(Percent of GDP)

Sources: IMF, World Economic Outlook database; and IMF staff calculation.

Fiscal Adjustment and Current Account Balances

(Percent of GDP)

Sources: IMF, World Economic Outlook database; and IMF staff calculation.

Looking ahead using IMF World Economic Outlook projections of current accounts for 2009–12, with a few notable exceptions (Greece and Portugal), deficits are set to decline to levels compatible with equilibrium values for the current account (first figure, third panel).3 In Greece, policies are in place to further improve the external accounts under the EU- and IMF-supported adjustment programs. For some countries, including Belgium, France, Italy, and Spain, a modest further improvement in the external accounts would be desirable. Strikingly, a number of countries in eastern Europe are set to run current account balances well above what would be consistent with fundamentals. The gaps between the deficits and their estimated equilibrium intervals are projected to range between about 2 percent of GDP in Bulgaria to 10 percent of GDP in Latvia. This possibly reflects a combination of private sector balance sheet repair currently under way, and still-heightened risk aversion toward the region, both of which may temporarily depress domestic demand. A subsequent recovery of domestic demand would likely bring current account balances back in line with equilibrium values. Similarly, a number of surplus countries are set to continue to run surpluses that exceed values justified by fundamentals.

The ongoing fiscal adjustment could have sizable effects on current account balances in a number of countries. Fiscal deficits rose sharply in the aftermath of the global financial crisis and are currently being brought down to preserve sustainability of the public finances across the EU. In some countries (for example, Greece), projected fiscal deficits remain higher than targets stipulated in their most recent Stability and Convergence Programs (SCPs), while in other countries (for example, Germany) the situation is now reversed (second figure, first panel). A simulation based on the model mentioned above shows that additional adjustment to align projected fiscal balances to current SCP targets could be enough to bring Italy’s, France’s, and Belgium’s current account deficits within the normal range (second figure, second panel). In Greece, continued fiscal consolidation would further reduce, although not eliminate, remaining imbalances. At the same time, additional consolidation could further increase the current account balance in Ireland.

Note: The main author of this box is Irina Tytell.1 Two other approaches to determining sustainable current account balances exist. One is a reverse “early warning” type exercise that links the probability of a crisis to the degree of imbalance and derives “safe levels” of current account imbalances. This approach is based on historical observations of the association of imbalances with subsequent crises. Another approach is based on evaluating the current account balance that stabilizes the net foreign asset position of the country in question (see Lee and others, 2008).2 See Lee and others (2008); and also Jaumotte and Sodsriwiboon (2010); Barnes, Lawson, and Radziwill (2010); and Decressin and Stavrev (2009).3 Note that for the purpose of computing the equilibrium values for this period it was assumed that countries follow sustainable fiscal policies. For the EU this means adherence to targets of the Stability and Growth Program updates. In addition, to abstract from the potential influence of output gaps on the estimates of the equilibrium current accounts, the European Commission’s projections for potential growth were used rather than those for actual growth.

Restoring Growth and Preventing Future Excessive Imbalances

The crisis led to a sharp adjustment of earlier current account imbalances, and current account balances in many countries are now close to “equilibrium” (Box 3.4). Current account deficits in Spain and Ireland declined significantly by 2010, while current account deficits in the Baltics and Bulgaria disappeared or turned into surpluses. Looking ahead, deficits are set to be reduced to levels compatible with equilibrium values of the current account, with a few notable exceptions (Greece and Portugal). In Greece, policies are in place to further improve the external accounts under the EU/IMF-supported adjustment programs.

To overcome the crisis decisively, adjusting imbalances is not enough: the most critical factor in the longer term is restoring GDP growth in the crisis-affected countries to secure lasting prosperity. Strong GDP growth would also make it easier to put public finances on a sustainable footing and improve the health of banking systems.

A sustainable recovery in crisis-affected countries will require a growth pattern different from that of the precrisis years, with a stronger role for the tradable sector and exports, and a less prominent role for the nontradable sector, capital flows, and domestic demand. The previous growth pattern, which relied on large capital inflows, led to pronounced boom-bust cycles. These cycles increased growth volatility but had no appreciable effect on average growth while leaving a legacy of high external indebtedness and impaired competitiveness (Figure 3.29). Surplus countries could focus on removing obstacles to the expansion of their nontradable sectors.

Figure 3.29Selected EU Countries: Capital Inflows and GDP per Capita Growth, 2003–10

Sources: The Conference Board Total Economy database, January 2011; IMF, World Economic Outlook database; and IMF staff calculations.

Figure 3.30Selected EU Countries: Growth of Real Exports of Goods and Services, Average 2010–11

(Percent)

Sources: IMF, World Economic Outlook database; and IMF staff calculations.

After a difficult 2009, such a shift seems to be taking hold in the Baltic countries and Bulgaria. Exports grew rapidly in 2010 and so far in 2011, and competitiveness indicators improved markedly, although the reallocation of resources to the tradable sector is by no means complete (Figures 3.30 and 3.31).15 In the euro area periphery, Ireland’s competitiveness has improved considerably, while Spain has experienced a strong increase in exports. In Greece and Portugal, export growth is still weak, even though Greece has seen an improvement in competitiveness.

Figure 3.31REER (ULC Manufacturing Based)

(Peak = 100)

Sources: European Commission, Price and Cost Competitiveness Indicators; IMF, Information Notice System; and IMF staff calculations.

Ultimately, growth and convergence will depend on productivity increases. On this front, some countries have performed poorly over the past decade, despite ample access to foreign capital. In Portugal, for example, labor productivity grew by much less than its starting position pointed it toward. Apparently it failed to put the large capital inflows it enjoyed to their best productive uses (Figure 3.32).

Figure 3.32Selected EU Countries: Labor Productivity

Sources: The Conference Board Total Economy Database, January 2011; and IMF staff calculations.

Better Policies Are Needed at the National Level

Sustainable convergence in the EU will require better policies at the national level—policies that promote increases in efficiency and productivity and that prevent boom-bust cycles. To a considerable extent, the current crisis arose from a failure to use national policies to manage local conditions. Therefore, policymakers need to rely more strongly on national fiscal policy, structural reform, and macroprudential tools to manage national developments in demand, credit, prices, and wages relative to developments in the EU as a whole (Box 3.5).

Better Governance at the EU Level Would Help Enforce Strong Policies

Reinforced surveillance of imbalances at the EU level would encourage appropriate and timely national responses and help shield the region from asymmetric developments. Previous work by European Commission staff in the context of the Stability and Growth Pact (SGP), the Lisbon Strategy, and the review of competitiveness developments pointed in the right direction. However, related political discussions (within the Eurogroup, the Economic and Financial Affairs Council, or the European Council) were neither systematic nor binding.

The proposed new surveillance framework for imbalances—the Excessive Imbalances Procedure (EIP)—aims at filling this gap and provides a strong platform for discussing imbalances at the EU level and for handling them at the national level. The EIP will mimic the SGP in that it will have both a preventive and a corrective arm. To be most effective, the procedure should be designed such that it avoids deadlocks in all intermediate steps between diagnostics and sanctions and ensures prompt action to address severe vulnerabilities. The mechanism would encompass all EU27 countries, but more demanding enforcement rules should apply to euro area member states.

Box 3.5How Can the Reemergence of Excessive Imbalances Be Prevented?

Policies can help prevent a recurrence of imbalances. They could do so, for example, by improving fiscal institutions, helping to channel capital flows to productive uses, and promoting strong private sector balance sheets. In addition, structural reforms could improve domestic price and wage adjustment and stimulate healthy private saving and investment behavior.

Identifying effective policy measures for reducing or preventing excessive current account imbalances is not an easy task. First, many policies may have multiple effects that yield ambiguous overall implications for imbalances, thus requiring a case-by-case analysis to determine their suitability. Second, many policies are difficult to quantify, so that existing indicators often suffer from measurement errors and small-sample biases. For these reasons, the recommendations below are merely suggestive. Actual policies need to be carefully tailored to specific circumstances in every country, and the estimated effects on imbalances should be treated with caution.1

Fiscal

Although fiscal adjustment can help reduce excessive imbalances in the near term, over the longer term the quality of budgetary institutions underpins fiscal discipline. Sound public finances, in turn, can help to avoid excessive current account imbalances. The quality of fiscal institutions in a number of EU countries that developed excessive deficits ahead of the global financial crisis—the Baltics, Bulgaria, Greece, Ireland, Portugal, and Romania—ranks relatively low (figure, first panel).2 Raising the quality of their fiscal institutions to the average level in the EU could have reduced excessive deficits in 2005–08 by about 1 percent of GDP in Greece, Ireland, and Portugal, and by up to ½ percent of GDP in the Baltics, Bulgaria, and Romania.

Imbalances and Policies, 2005–08

Sources: IMF, World Economic Outlook database; and IMF staff calculations.

Financial

In the wake of the global financial crisis, effective financial and macroprudential regulation was identified as a key tool for preventing credit and housing bubbles that could cause unsustainable consumption and investment booms. Distortions in mortgage markets are especially damaging in countries where home ownership rates are high. Countries with high loan-to-value ratios (LTVs) on mortgage loans and high home ownership rates include Bulgaria, Estonia, Ireland, Lithuania, and Spain, all of which ran excessive current account deficits ahead of the global financial crisis (figure, second panel).2 This suggests that measures to keep LTVs in such countries at more reasonable levels could help prevent excessive deficits in the future. Bringing the combination of maximum LTVs on mortgage loans and home ownership rates to the EU average could have reduced excessive deficits in 2005–08 by 1–1½ percent of GDP in Bulgaria, Estonia, Ireland, Lithuania, and Spain.

Structural

Although the link between structural policies in labor and product markets and growth is well established, the evidence on the role of these policies in relation to current account imbalances remains relatively limited (see, for example, Berger and Nitsch, 2010; and Kerdrain, Koske, and Wanner, 2010). The lack of evidence may reflect potentially opposing effects of these policies on imbalances and resulting ambiguities in their overall effects, which make some of the recommendations in this area particularly tenuous.4

A high tax wedge on labor raises costs and lowers corporate profitability, thereby reducing investment and pushing the current account balance toward surplus. It could also increase unemployment, raising the need for precautionary saving on the part of households, which would also push the current account into surplus. Reducing the average tax wedge could help lessen excessive surpluses in Austria, Germany, and Sweden (figure, third panel). Bringing it to the EU average could have reduced excessive surpluses in 2005–08 by about ½ percent of GDP in these countries.

The link between the strictness of employment protection legislation (EPL) and the extent of product market regulation (PMR) to current account imbalances is less clear. Although stricter EPL gives workers more job security, it also leads to longer unemployment spells and reduces productivity, with ambiguous overall effects on precautionary saving and the current account. Similarly, more extensive PMR tends to lower productivity, hurting the working population that has a high propensity to save, but it also tends to raise costs and lower corporate profitability, thereby reducing investment, so that the overall effect on the current account is unclear. On the whole, there is no strong evidence suggesting that reducing EPL or PMR would lessen or prevent excessive imbalances in the EU.5

In addition, certain labor market practices not directly considered here—for example, collective bargaining systems that lead to excessive wage demands—have likely contributed to the widening of the imbalances. Structural policies that improve wage flexibility could facilitate current account adjustment and help to prevent a resurgence of imbalances in the future. Similarly, further liberalization in nontradable sectors would help in surplus countries by stimulating domestic demand and in deficit countries by moderating prices and improving competitiveness in tradable sectors.

Note: The main author of this box is Irina Tytell.1 The effects of policies are estimated using regressions of estimated current account adjustment needs on policy variables. See Annex for more detail.2 The quality of fiscal institutions is measured using an index produced by the European Commission (http://www.ec.europa.eu/economy_finance/db_indicators/fiscal_governance/framework/index_en.htm).3 Maximum LTVs refer to the highest observed LTVs in the mid- to late 2000s. Home ownership rates reflect census data from the early to mid-2000s. See Crowe and others (2011; forthcoming); and IMF (2011).4 Indicators of structural policies in labor and product markets come from the Organization for Economic Cooperation and Development (OECD) and Kerdrain, Koske, and Wanner (2010).5 In the literature, the evidence on the effects of EPL and PMR on current accounts is mixed. For example, Kerdrain, Koske, and Wanner (2010) do not find any significant effects of EPL on saving rates, a somewhat surprising positive effect on investment rates (possibly Reflecting increased substitution of capital for labor), and a negative effect on current accounts in OECD countries; they find no lasting effects of PMR on investment rates and no significant effects on current accounts. In contrast, Berger and Nitsch (2010) find that European countries with higher EPL and PMR exhibit systematically lower trade surpluses or higher trade deficits than do their peers with more flexible labor and product markets.

Several elements of the proposed EIP could be strengthened to make the process more effective:

  • The surveillance process should be kept to the shortest possible time frame. The current proposal to align, under the European semester,16 the preventive arm of the EIP with the surveillance conducted under the SGP, Europe 2020, and the European Systemic Risk Board is a welcome step and should facilitate the identification of relevant linkages between imbalances, fiscal policy, growth-enhancing reforms, and macrofinancial stability. However, weeks rather than months should suffice to identify countries with potential problems, if, as foreseen, the initial screening is based on a mechanistic application of a heat map complemented, as appropriate, with past assessments of individual countries. By nature, the corrective arm will be decoupled from the European semester, but the elapsed time from diagnosis to reform implementation should not be more than a year.

  • The initiation of the preventive arm should be as automatic as possible. The subsequent in-depth analysis should involve a larger element of judgment.

  • The scoreboard should err on the side of caution and leave little margin for countries to escape in-depth analysis, especially because the intent of the alert system is to identify potential imbalances in a timely manner. Ideally, the number of indicators should be small, but broad in scope, and should be based on definitions or accounting relationships to remain immune from influence peddling. A balance between stock and flow indicators would be useful.

  • Policy adjustment recommendations should best be subject to reverse majority rules—under which recommendations are adopted unless a majority opposes—as in the proposals for a modified SGP.

  • Finally, these policy actions need to be fully integrated with fiscal policy recommendations under the SGP, and with risk warnings and recommendations of the ESRB.

Conclusions

In the run-up to the crisis, countries in the euro area periphery, and countries in emerging Europe with fixed exchange rates, built up large current account imbalances. Although the causes of the imbalances varied across countries, the pattern was similar: strong capital flows into the nontradable sector drove up nontradable prices and wages and eroded competitiveness. Because capital flows boosted demand rather than supply, and imports rather than exports, they contributed to large and ultimately unsustainable changes in net external asset positions.

Financial integration played a critical role in permitting the financing of imbalances, but lack of institutions and a “national” approach to crisis resolution prevented the private sector and markets from playing a larger role. With banking problems addressed at the national level, banking and sovereign problems in euro area periphery countries exacerbated each other; and as financing costs in the private sector increasingly came to depend on the national origin of the borrower, financial integration reversed.

To overcome the crisis, the crucial factor in the longer run is restoring GDP growth in the crisis-affected countries. Ultimately, growth and convergence will need to be backed by productivity increases, which some countries have struggled with over the past decade, despite ample access to foreign capital. To foster efficiency increases, better national-level policies are needed, while better governance at the EU level would help enforce such policies. Finally, greater vigilance is needed, both nationally and across borders, and better institutions to deal with financial sector problems, and more, rather than less, financial and economic integration.

Note: The main authors of this chapter are Lone Christiansen, Yuko Kinoshita, Jeta Menkulasi, Esther Perez, Irina Tytell, Nico Valckx, and Johannes Wiegand.

Austria, Belgium, Finland, France, Germany, Ireland, Italy, Luxembourg, the Netherlands, Portugal, and Spain. They were followed by Greece (2002), Slovenia (2007), Cyprus, Malta (both 2008), the Slovak Republic (2009), and Estonia (2011).

Cyprus, Malta, Slovenia, and the Slovak Republic, which entered the euro area between 2007 and 2009, did not meet this criterion. Luxembourg is excluded because of its small size and role as a financial center.

The exception is the Czech Republic.

By contrast, retail lending remained largely within national borders because banking groups typically expanded into other European countries by establishing branches or subsidiaries rather than through direct cross-border lending.

The bulk of the appreciation of the CPI-based real effective exchange rate in Greece and Portugal was due to the nominal appreciation of the euro (Chen, Milesi-Ferretti, and Tressel, forthcoming).

The effect was compounded by the freezing of the international syndicated loans market, as well as a halt in the growth of direct cross-border loans.

Although the adjustment in the Baltics had already started in the fall of 2007, when Swedish banks became concerned about their exposures and wanted to engineer a gradual slowdown, the real shock to the region came in September 2008, after Lehman Brothers filed for bankruptcy.

Together, per capita income and interest rate differentials in the early 1990s explain about 85 percent of the variation in current account balances in the late 2000s.

It would also have created fiscal cushions that could have been used during the subsequent downturn.

During the boom years, many countries in emerging Europe strengthened prudential regulations to slow credit growth. Although these measures had limited effectiveness in slowing credit, they resulted in the creation of large capital and liquidity buffers, which protected the banks during the crisis of 2008–09.

During the boom years, Spanish banks were required to build supplemental reserves to cover future loan losses (“dynamic provisioning”). While this helped to build buffers, it was less successful in slowing credit growth.

The situation is different in many countries of emerging Europe, where foreign banks tend to dominate the banking sector, reflecting a conscious decision by the governments to divest to Western banks because domestic and state-owned banks repeatedly caused crises and slowed the transition process.

In the international context, various initiatives are under way to make the banking sector safer, and progress is being made in setting up better cross-border tools for crisis management and resolution.

Achieving these changes would require rapid progress on the single regulatory rulebook for banks and on harmonizing supervisory practices to eliminate multiple reporting requirements. Standardization of products, the establishment of a pan-European approach to securitization, and more uniform consumer protection regimes are also needed.

See also Chapter 2.

The European semester is a six-month period every year during which the member states’ budgetary and structural policies will be reviewed to detect any inconsistencies and emerging imbalances. The aim is to reinforce coordination while major budgetary decisions are still under preparation. See http://www.consilium.europa.eu/showFocus.aspx?lang=EN&focusID=504.

    Other Resources Citing This Publication