Why the level of public debt matters after all

By: Susanne Mundschenk

 

Economists tend to agree more easily on the economic effects of monetary policy than fiscal policy. While it is generally agreed that an increase in interest rates will lead to a slowdown in inflation and growth eventually, it depends very much on one’s underlying model whether government consumption raises or lowers private consumption and real wages. In fact, both results can be obtained in empirical estimates. The result depends on how the causal relationship between fiscal policy and the economy (output, inflation, interest rate) is identified. In a recent NBER paper entitled « Debt and the effects of fiscal policy » Carlo Favero and Francesco Giavazzi  argue that these models share one common weakness, namely that they omit the response of taxes, government spending and interest rate service on the level of public debt. Once the public debt is included in the analysis, it is no longer surprising that consumers respond differently to government consumption. The consumers’ response depends on whether they expect the government to meet its intertemporal budget constraint by adjusting the tax rates and government spending in the near future.

 

Using US data, the authors provide some explanations for two puzzles from the empirical literature. The first is that there is no hard evidence from the standard literature that fiscal shocks affect long term interest rates. In the paper’s setting, whether or not a fiscal shock affects interest rates depends on the level of public debt. If a growing stock of outstanding government debt raises fears of future inflation or debt default, fiscal shocks are more likely to raise risk premia and therefore long term interest rates. The second puzzle from the standard literature is that the effect of fiscal policy shocks on the economy seemed to have changed over time. While government expenditure had a significantly positive effect on US output in the 1960's and 1970's, the effect became insignificant in the 1980's and 1990's. Favero and Giavazzi find that the US authorities responded differently to the accumulation of public debt in these two periods. Since the early 1980's fiscal policy instruments were adjusted over time to stabilise the debt to GDP ratio.

 

The different reaction to a fiscal policy shock in these two periods can be linked to the dynamics of the debt-to-GDP ratio. In the 1960's and 1970's the economic growth rate was always higher than the cost of debt financing. Under these circumstances the government could grow out of its public debt without further efforts. Tax income was always sufficient to service the outstanding public debt. These golden years came to an end in the 1980's when the economic growth rate fell below the average cost of debt servicing. Extra efforts from the government were needed to stabilise an otherwise exploding debt ratio.

 

 

 

This paper is about the US. What is the relevance for Europe? The importance of the level of public debt has been explicitly acknowledged by the revised Stability and Growth Pact. Countries with higher debt levels are now required to undertake stronger fiscal efforts than those countries with lower public debt levels. What about the debt dynamics?  A simple look on the Italian data in the two charts confirms that the two periods of debt dynamics can also be identified for Italy. Since the 1980’s economic growth is lower than the real interest rate (upper chart). The debt service increased dramatically, but so did the primary surplus, albeit at a slower pace (lower chart). This suggests that Italy also reacted to the change in the debt dynamics and increased its primary surplus to stabilise public debt. Unfortunately, the chart only shows data until 2001. Today, Italy still has one of the highest debt levels in the euro area. It benefits from a lower debt service cost since the start of EMU, but its primary surplus deteriorated from 2001 until 2005 due to rising primary expenditures. We still expect the economic growth rate to be much lower than the real interest rate, implying a potentially unstable debt path. This suggests that to test the model of Favero and Giavazzi in Europe, it may be useful to identify a third period marking the start of monetary union, when the fiscal response to the debt dynamics changed again.

 

 

mundschenk(at)eurointelligence.com

---