The debate on the correct setting for fiscal policy at a time of recession is probably the oldest debate in macroeconomics. One key element in the debate is the trade-off between supporting output growth in the short term, versus the need to control the growth of public debt in the long term.
There are some economists who do not recognise that this trade-off exists at all, because they claim that an increase in the fiscal deficit cannot impact aggregate demand, even in the short run. But this is not a view which I believe to be supported either by empirical research or by economic theory (except on some very restrictive assumptions about Ricardian Equivalence or Says Law).
I recognise that this last statement is very contentious, but it is not my subject today. Instead, I would like to take as given the assumption that a temporary easing in fiscal policy (of the type advocated last week by Martin Wolf for the UK) will increase aggregate demand in the near term. Accepting this, I would then like to ask whether the benefits of an immediate boost to aggregate demand outweigh the costs of higher public debt, and the consequent risks of a fiscal crisis. That is the nub of the issue which is, or should be, exercising policymakers in the real world today.
How should this question be approached? Sometimes, advocates on both sides of the debate assume that the answer is obvious. Supporters of fiscal easing tend to take it as axiomatic that higher public debt will have no effect on inflation or interest rates, and frequently quote the example of Japan in support of their argument.
Meanwhile opponents of fiscal easing (like the UK government) take it as equally axiomatic that a fiscal crisis is to be avoided at all costs, and quote the examples of Greece, Italy and Spain to support their case. They point out that public debt in some developed economies is higher than it has ever been before in peacetime (see the IMF graph below), and conclude that it simply must come down before disaster strikes.
But surely this question involves a much more delicate balance of risk and reward than is often acknowledged. The real question is how much risk a government wants to take of encountering a fiscal crisis over a given period, and how much that risk is changed by increasing the budget deficit to boost GDP. Quantifying these two steps is extremely difficult, but is essential if we want to move beyond the biases and simple assertions which so often pass for public debate on fiscal policy.
Recent publications by the Fiscal Affairs Department at the IMF enable us to make some progress. Last year, in Assessing Fiscal Stress , IMF economists estimated some new indices of fiscal stress, which are driven by 11 different economic variables, including budget deficits, debt ratios, financing needs and demographic trends. The fiscal stress index varies between zero and one, with higher numbers indicating greater concern about fiscal sustainability. The indicator has behaved as follows for the developed economies in recent years:
The most relevant line on the graph is the green one, which shows that the fiscal stress index for the developed economies, weighted by GDP, has more than doubled since the 2008 crisis. The indicator now stands at about 0.43, compared to 0.17 in 2007. The main reason for this, of course, is that budget deficits and refinancing needs have increased greatly during the recession.
We can now use this piece of information to address the questions posed above. The IMF study also provides a chart which translates the level of the fiscal stress index into the probability that a country will enter a fiscal crisis, or remain in fiscal crisis, within one year. Obviously, the higher is the stress indicator shown on the horizontal scale, the greater the probability of fiscal crisis, shown on the vertical scale. When the stress indicator is at its present level of 0.43 for the average developed country, there is a probability of around 0.15 that there will be a fiscal crisis within one year.
Although that may be an acceptable risk in any single year, the risk does cumulate rapidly over five years (cumulatively 0.56) and 10 years (0.80). So the first conclusion is that something needs to be done to correct budget deficits and debt ratios over the medium term. That certainly is the case in the US today.
One rebuttal to this conclusion might be that the IMF study is based on the period from 1995-2010, when global excess savings were not as large as they are now. The increase in savings may reduce the risk of fiscal crisis today compared to the years in which the model was estimated, because there is more private money available to fund a government deficit at low interest rates. Another rebuttal could be that governments might decide that the consequences of a fiscal crisis are not necessarily fatal, and can (with difficulty) be reversed, so it might be worth running this risk anyway.
The second conclusion is somewhat different. If a country has a credible medium-term commitment to reduce the fiscal stress indicator over the medium term, then it does seem that there is scope for a moderate increase in the budget deficit over a period of a couple of years to cushion a temporary period of recession in the economy. For example, a short-term rise of two or three percentage points in the budget deficit would have virtually no effect on the probability of fiscal crisis in any given year (eg. raising the probability of crisis from 0.15 to, say, 0.20), so it might be considered as a risk worth contemplating. That could well be the case in the UK today.
One final point. None of these conclusions is obvious, and they involve making difficult trade-offs where policymakers can reasonably differ. Those who pretend otherwise are grossly over-simplifying a complex and critically important debate.
Copyright The Financial Times Limited 2012.