The World Economic Outlook published by the International Monetary Fund last week has created new excitement around the debate on the effectiveness of fiscal retrenchment. The evidence that governments have consistently been underestimating the size of the fiscal multiplier – which measures the impact of deficit reduction measures on economic growth – has been used to suggest that austerity policies are doomed to fail.
That is the wrong conclusion to draw from the IMF study.
The fiscal multiplier quantifies the impact of a given change in taxes or public expenditures on gross domestic product (GDP). Those who interpreted the IMF study as evidence that austerity is self-defeating mistakenly believe the multiplier measures the impact of a change in the budget deficit on the long-term rate of growth of the economy. In fact the multiplier measures the one-off impact of budget changes on output when the adjustment is made.
Let us consider a simple example of a country with a budget deficit of 1 per cent of GDP, a debt to GDP ratio of 100 per cent and a trend growth rate of 1 per cent per year. A multiplier lower than one implies that when the deficit is reduced from 1 per cent to zero, GDP continues to rise – although temporarily at a lower rate than the trend – and the debt falls as a percentage of GDP.
If the multiplier is larger than one, the reduction of the deficit by 1 percent of GDP initially leads to a reduction in the level of GDP and to a rise in the debt-to GDP ratio. But if the deficit is kept at zero in the following years, the debt ratio would start falling in the second year, as growth returned to its trend rate.
This example is consistent with what has been observed in several euro area countries over the past couple of years: the sharp reduction of the deficit from very high levels has initially led to a contraction of GDP and an increase in the debt ratio.
But a multiplier larger than one does not mean that a cut in the deficit is not effective in reducing the debt burden in the long-term. It only implies that it takes more time and more pain to achieve a reduction of the debt.
What policy implications can be drawn from this finding? Should countries delay the adjustment or spread it over a longer period of time?
Front loading austerity measures produces an initial sharp reduction in output, and a short-term increase in debt over GDP. But soon after the economy should recover and the growth rate should pick up. A more gradual adjustment implies a prolonged period of low or stagnant growth, with debt ratios not falling for several years. It is not clear which scenario is preferable.
It partly depends on the reasons for the multiplier being larger than one. Some commentators have suggested that it is due to the fact that interest rates are close to the zero lower bound, which prevents monetary policy from compensating for the restrictive effect of the fiscal adjustment. In cases of severe fiscal adjustment, the economy might benefit from negative interest rates, but this is not generally an option for central banks. In such a case, however, the best way to reduce the debt burden would paradoxically be to engineer a fiscal expansion, even though that seems to contradict the experience of the eurozone in the run up to the financial crisis.
But another reason for the multiplier being higher than one could be that the potential growth rate of advanced economies has fallen sharply during the financial crisis, partly as a result of the deleveraging that is taking place in the private sector. If this is the case, growth would remain lower even in the absence of fiscal adjustment and the size of the multiplier would not fall over time.
Finally, the study does not show any non-linearity in the relationship between fiscal adjustment and level of output. That is the fiscal multiplier does not increase in the case of larger contractions. That would suggest there is no advantage in postponing adjustment.
The optimal timing ultimately depends on the patience of voters to endure a prolonged adjustment effort and the patience of financial markets to continue financing a government’s deficits.
The more patient the voters, the more the adjustment effort can be spread over time. Recent experience in the euro area suggests that voters may lose their patience if austerity drags on for too long and if they don’t see the light at the end of the tunnel.
The patience of financial markets depends on several factors, such as the size of the adjustment effort, the level of the debt, the underlying pace of economic activity and the willingness of the central bank to finance the Government. If markets become impatient, credit risks start to rise, market access is impaired and the overall adjustment costs increase in a dangerous feedback loop. In this case, market impatience, rather than the effect of excessive frontloading, leads to a higher fiscal multiplier.
The only way to ensure that a gradual fiscal retrenchment is sustainable over time and credible in the eyes of the markets and the voters is to implement it in the context of a multi-year adjustment program, possibly under the auspices of the IMF. The policy implication is that countries wishing to spread out fiscal adjustment should probably seek the assistance of the IMF.
That, rather than a rejection of austerity, may be the subtle message of the IMF study.
The writer is a former member of the executive board of the European Central Bank and currently a visiting scholar at Harvard’s Weatherhead Centre for International Affairs