Before 2007, systemic financial crises were mainly the fate of emerging economies. EU countries had been relatively immune, the most notable exceptions being Finland and Sweden during the 1990s and Spain at the end of the 1970s. For instance, Laeven and Valencia (2008) report 122 systemic financial crises episodes between 1970 and 2007. Of these, only 22 crises occurred in today's EU27 (that is, including the countries that acceded to the EU in 2004) and the OECD.
These crises have usually been very costly. The same authors estimate that net fiscal outlays to rehabilitate the banking system averaged 13% of GDP and increases in the public debt ratios averaged 20% of GDP. Owing to their relative immunity from crises in the past, EU countries were surprised by the deterioration in public finances following the global crisis. This was despite developments in the run-up to the crisis that suggested some countries, especially those that had experienced the fastest credit growth, were particularly vulnerable to the economic downturn to come. Rising government deficits, low economic growth, and support to the financial sector are leaving a legacy of rapidly growing government debt ratios, which EU countries are currently tackling at different speeds according to their fiscal consolidation programmes. A phasing out of the stimulus measures and cyclical recovery, including a rebound in tax revenue from the crisis-related lows, will be insufficient to prevent government debt ratios from rising to even higher levels before the end of the next decade.
By historical standards, although the rise in debt in most EU countries comes on top of comparatively high starting levels – reflecting the increase recorded in the 1980s which was only partially stemmed subsequently (see Figure 1) – the projected sharp increase in government debt ratios is nothing out of the ordinary in a financial crisis. While it is indisputable that the global financial crisis has led to a sharp deterioration of EU countries' public finances, opinions remain divided over the most appropriate consolidation strategy to follow. Particular attention has been focused on the timing of fiscal consolidation in relation to the path of economic recovery, including:
- the trade-off between consolidation and stabilisation;
- fiscal consolidation in the context of a distressed banking system where the credit channel is hampered and without which economic recovery can hardly take place, and
- the absence of exchange-rate adjustment in the Eurozone which could make it more difficult for countries with competitiveness problems to achieve successful fiscal consolidation.
In a recent paper (Barrios et al. 2010) we set out to investigate these questions by considering European countries together with other OECD economies where financial crises had also happened during the period 1970-2008.1 We conduct an econometric analysis of the determinants of successful fiscal consolidations, where the criterion used to define fiscal consolidations as "successful" depends on the size of the reduction in the level of public debt in the medium run. Making use of the data provided by Laeven and Valencia (2008) and Reinhart and Rogoff (2008, 2009) as a control for the occurrence of systemic financial crises episodes, we pay close attention to the specific features of the situation of most EU economies at the onset of the current financial crisis – namely the high starting debt level and/or fast rising interest rate paid on government bonds – to determine whether sharp or gradual fiscal consolidations are more appropriate.
Figure 1. Evolution of debt to GDP ratio during major debt increase episodes2
Sources: Commission services
Our results (based on probit estimations) suggest that repairing the financial sector significantly increases the chances of successful fiscal consolidations, especially so if sharp consolidations are called for. In such cases, fixing the banking sector increases the probability of achieving successful fiscal consolidations by approximately 30% (where success is measured by a reduction in the debt to GDP ratio by at least five percentage points three years after a fiscal consolidation episode has been initiated). Our findings also suggest that fiscal consolidations need to be tailored to country-specific situations such that a proper coordination of national fiscal plans is vital in order to minimise the adverse effects of a "rush to the exit". While the starting debt level (and related-risk of rising debt servicing burden) play a significant role, we find that these elements become less important economically once account is taken of the strong incentives to undertake fiscal consolidations that countries with high debt already face, which may in part also explain their success or failure.
We show that countries facing high debt levels and high interest rates or those at risk of low GDP growth in the coming years would have a better chance of achieving successful fiscal consolidations if these are sharp and sustained. Other countries where such constraints are less binding meanwhile, would be better off by undertaking more gradual fiscal consolidations. These results are presented in Figure 2 below showing the estimated probabilities of gradual vs. cold shower fiscal consolidations depending on the starting level of public debt and the snowball effect of public debt.3
Figure 2. The estimated probability of success of gradual and cold shower fiscal consolidation depending on the snowball effect and the level of debt
Note: Figures based on tow-stage probit estimations as reported in Table 5 of Barrios et al. (2010)
Furthermore, we find that public-expenditure-cuts-based consolidations tend to be more effective in part because these send convincing signals regarding the political will of the fiscal retrenchment as well as its medium-run viability.
Finally, we also find evidence suggesting that while exchange-rate depreciations may help to complement fiscal consolidation effort, these offer no guarantee for achieving successful fiscal consolidations. A number of possible factors can explain this result. First, the contribution of depreciation depends on the degree of exchange-rate pass-through and the degree of trade-openness, which are to a larger extent out of governments' control. Furthermore, past experiences of successful fiscal consolidations conducted following large exchange-rate devaluations were made possible through the anchoring of inflation expectations by pegging national currencies to an inflation-proof currency, a role which was typically played by the Deutsche Mark in the EU before the advent of the euro. Overall, ongoing discussions on the difficulty to undertake fiscal consolidations in the absence of exchange-rate devaluation in the Eurozone tend to disregard the fact that past successful devaluations have often come together with wide-ranging macroeconomic reform packages, which were arguably the most important factor of adjustment and did not depend on devaluation in order to be implemented.
This article reflects the views of the authors writing in their personal capacity.
Barrios,S., S Langedijk, and LR `Pench (2010), "EU fiscal consolidation after the financial crisis. Lessons from past experiences", European Economy Economic Paper n°418, Directorate General for Economic and Financial Affairs, Brussels.
Laeven, L and F Valencia (2008), “Systemic banking crises: a new database”, IMF Working Paper 08/224.
Reinhart, CM and KS Rogoff (2008), “Banking crises: an equal opportunity menace”, NBER Working Paper 14587.
Reinhart, CM and KS Rogoff (2009), This time is different: eight centuries of financial folly, Princeton University Press.
1 All EU27 countries are included in our study and in addition Australia, Canada, Japan, Mexico, Norway, Switzerland, Turkey, and the US.
2 Note: Debt increase episodes are identified as corresponding to a minimum of 20% increase in a maximum of five years. The year t0 corresponds to the first year marking the debt increase episode which in the current crisis corresponds to 2007. The last year in the current debt increase episode is 2011 (data taken from the Commission Autumn 2009 forecast) and the year t-5 is 2002 and is set in order to cover a period of 10 years. For the other debt increase episodes the last year t+4 is defined as the one where the debt increase over five year (on a moving average basis) reached its maximum value. The years t0 and t-5 are then determined recursively to cover a time span of 10 years as for the current debt increase episode. The (unweighted) average value of the debt to GDP ratio for the following groups of countries (with time periods covered indicated in parentheses) are considered: EU, past large debt increases: Belgium (1974-1983), Denmark (1974-1984), France (1986-1995), Greece (1978-1987), Ireland (1975-1984), Italy (1975-1984 and 1985-1994) Malta (1990-1999), the Netherlands (1976-1985), Portugal (1975-1984), Spain (1976-1985 and 1987-1996, Sweden (1973-1982). Non-EU, OECD: Japan (1970-1979), Canada (1976-1985 and 1984-1993) and Iceland (1986-1993). Finland, Sweden 1990s financial crisis: Finland (1985-1994) Sweden (1985-1994)
3 A gradual consolidation is measured by an improvement in the cyclically adjusted primary balance of at least 1.5 percentage points over a three-year period while a cold shower implies that such an adjustment takes place over one year only. The snowball effect is also sometimes termed the debt-stabilising primary balance and is defined according to the following expression: Debt/GDP(t-1) * (i - y/ (1+ y)), where i is the interest rate and y is the nominal GDP growth in year t. The three levels of debt chosen to group countries in three categories (i.e. high, medium, low) are set according to the level of debt of EU countries forecasted for 2011 (Source: European Commission Spring 2010 forecast).