reichlin13septfig1.png
VoxEU Column

The financial origin of the euro area fiscal wound

One of the most cited books written in the aftermath of the 2008 financial crisis, which documents the special characteristics of recessions associated with financial crises over time and across countries, is titled 'This Time is Different'. This column argues that the joint behaviour of the public deficit and public debt in the euro area from 2008 to 2013 – characterised by high and persistent public debt despite a severe fiscal consolidation from 2009 – cannot be explained by the unprecedented collapse in output and the historical relationship between macroeconomic, fiscal and financial variables. Rather, it reflects specific characteristics of the crisis years and the size and nature of the public support to the financial sector.

One of the most cited books written in the aftermath of the 2008 financial crisis is titled This Time is Different (Reinhart and Rogoff 2009) and documents the special characteristics of recessions associated with financial crises over time and across countries. In a new paper (Caruso et al. 2019) we ask whether the joint behaviour of public deficit and public debt in the euro area was ‘different’ during the 2008-2013 period, which includes both the Great Recession and the sovereign debt crisis.   

The euro area twin crises – associated with the Great Recession and the euro area sovereign debt crisis – left a legacy of unprecedented high levels of public debts both at the national level and in the aggregate. Projected public debt for the euro area as a whole for 2019 is 85.8% of GDP, decreasing from the all-time peak of 94.4% in 2014 but still around 20 percentage points above the 2007 pre-crisis level. This anomalous high and persistent public debt has been associated with an unprecedented effort of fiscal consolidation which involved a rapid decrease in public deficits starting in mid-2009. 

Figure 1 Euro area government debt/GDP and deficit/GDP

Note: Indices based at 100 in the quarters in which each recession starts.

The charts in Figure 1 plot the paths of debt-to-GDP (left panel) and deficit-to-GDP (right panel) ratios for three euro area recessions starting in 1980, 1991, and 2008.  For each of these episodes, the debt and deficit variables are set equal to 100 at the beginning of the recession. The horizontal axis indicates quarters after that date. 

Following each recession, the deficit-to-GDP ratio increases due to the decline of GDP (the denominator), the decline in tax income, and the effect of fiscal stabilisers on public expenditure. The 2008 recession, however, is of a different order of magnitude. Due to the dramatic decline of GDP, the deficit-to-GDP ratio spikes for the first five quarters and reaches a maximum in the second quarter of 2009, when a massive fiscal consolidation takes place. 

There are many possible explanations for why public debt has not declined faster, such as the unusual magnitude of the initial negative macroeconomic shock that hit the euro area, the response of taxes and automatic stabilisers in a deep recession (which, by design, is stronger than the change in income), but also – potentially – the unique financial nature of the crisis affecting risk premia and public expenditure on rescuing failing financial institutions. 

In our paper, we assess the quantitative importance of these different potential explanations. We do so by counterfactual analysis based on a large vector auto regressive (VAR) model for the euro area. Our model incorporates detailed fiscal variables – spending, taxes, transfers, public investment, and interest payments – macroeconomic and financial indicators, prices and interest rates at different maturities, as well as private debt variables.

The thought experiment we propose is to ask what an observer who had collected data on the previous recessions in the euro area, and who had known with certainty the path of output and prices during the twin crises of 2008 and 2012, could have predicted for public debt and the deficit in particular and for all the other variables included in the model. 

To this end, we estimate the model for the time period spanning from the first quarter of 1981 to the first quarter of 2008 and compute model-based expectations for all variables, conditional on the actual path of output and prices in 2008Q2–2013Q4 and the estimated parameters. This exercise can be interpreted as a test for the statement “this time is different”. A significant difference between the observed path and the median of the simulated path (conditional expectation) would suggest that the exceptional decline of GDP (and realised inflation) alone cannot explain what we have observed, given the historical pattern of business cycle recessions. 

To focus on the fiscal effects of the crisis, we derive a measure of public debt as a cumulated sum of the fiscal deficit and contrast it with realised public debt, which also includes valuation effects and those items related to public intervention in support of the financial system. The latter include government guarantees, which are accounted as debt but not as deficit (the cumulated sum of this component for the period 2008-2011 accounts for just above 6% of GDP; see Table1).

Our results point to the following facts:

1. The jump in the deficit-to-GDP ratio observed in 2009-2010, in the aftermath of the crisis, is statistically significantly different and higher than its counterfactual path. However, it reaches levels which are not significantly different from the simulated path by the end of 2010 thanks to an exceptional fiscal consolidation. This indicates that the anomalous deficit dynamics dissipates by the end of the sample due to the fiscal effort (see the left panel of Figure 2.

Figure 2 Conditional forecast, public debt and public deficit ratios, 2008-2013

Notes: The figures show the realised data (red), the data minus stock-flow adjustment (green) and the counterfactual path (blue). The blue lines are the medians of the forecasts conditional on the path of GDP and inflation, plotted with 68% (dark blue) and 90% (light blue) coverage intervals.

2. The ‘abnormal’ initial increase in deficit is mostly due to the action of fiscal stabilisers in a deep recession, which induced a large increase in expenditure together with a large decline in taxes. Figure 3 shows the fiscal gap that opened in governments’ budgets. Our detailed analysis on budget components shows that while fiscal aggregate shows large responses during the crisis, these are generally at the margin of historical regularities, given the magnitude of the crisis (although the cumulated effect is well off the conditional forecast). Interestingly, the consolidation is achieved via a flattening of expenditures accompanied by an increase in revenues along the historical trend, reversing the effect of the automatic stabilisers in support of income.

Figure 3a The fiscal gap: Euro area government total expenditures and revenues

Figures 3b and 3c Conditional forecast and realised values for government total expenditures and revenues

Notes: The blue lines are the medians of the forecasts conditional on the path of GDP and inflation, plotted with 68% (dark blue) and 90% (light blue) coverage intervals.

3. Importantly, the observed debt is well out of the bands of projected historical regularities. Conversely, the measure of public debt obtained as the sum of the deficit is high relatively to the counterfactual but returns within the bands of historical regularities towards the end of the sample (top panel of Figure 2). This fact points to the unique financial nature of the crisis. Table 1 reports discrepancies between the deficit and the changes in debt – the so-called stock flow adjustments – in the years of the crisis.  These adjustments capture most of the special measures in support of the financial system, which, according to accounting rules, are accounted as debt but not as deficit. Figure 4 reports both the deficit and the first difference of government debt, illustrating the point of the exceptional changes in public debt in 2008 and in 2010.

Figure 4 Euro area government deficit and first differences of government debt

Table 1 Stock-flow adjustments in 2008-2011 (percent of GDP)

Source: Eurostat (2012).

4. As for interest rates, we find that interest rate payments, while below the counterfactual path in the first phase of the crisis, move the upper boundary of the 90% confidence region since 2011 in coincidence with the euro area sovereign crisis. Interestingly, as can be seen in Figure 5, this is not due to an unusually high average long-term interest rate in that period but to an unusually high core-periphery spread, which here we capture as the difference between the Italian and the German ten-year government bond rates. 

Figure 5 Conditional forecast and realised values for interest payments, long-term interest rates, and the Italy-Germany spread

Notes: The blue lines are the medians of the forecasts conditional on the path of GDP and inflation, plotted with 68% (dark blue) and 90% (light blue) coverage intervals.

5. Other interesting results of our analysis point to the fact that the dynamics of macroeconomic variables – such as unemployment, consumption and the current account –   are generally well captured by historical regularities. An important exception is the large and persistent collapse in private investment. The results on consumption and private investment are reported in Figure 6. Other results are in the paper.

Figure 6 Private investment in the euro area

Taken together, these facts suggest that the consolidation effort has been unbalanced due to large transfers towards the financial sector. However, consumption has been relatively resilient while private investment has been hurt more than in previous recessions. This may be an important factor in explaining the change in trend growth post-crisis.

Authors' note: The views expressed in this article are those of the authors and do not necessarily reflect those of Confindustria.

References

Caruso, A, L Reichlin and G Ricco (2019), “Financial and Fiscal Interaction in the Euro Area Crisis: This Time was Different”, European Economic Review 119: 333-355.

Reinhart, C M and K S Rogoff (2009), This Time is Different: Eight Centuries of Financial Folly, Princeton University Press.

2,205 Reads