One country’s fiscal nudge is another country’s fiscal dud. In an important piece research, Ethan Ilzetzki and colleagues from the University of Maryland shine a light on the key interaction between monetary policy and fiscal stimulus.
When the 2008 financial crisis hit the world, the world hit back. Policymakers around the globe responded rapidly and aggressively to the most significant financial crisis the world had experienced in close to a century. Central banks slashed interest rates, which fell to close to zero in industrialised countries. They unleashed a barrage of unconventional monetary policies – “quantitative easing” in the United Kingdom, “credit easing” in the US, “swap” transactions between the United States Federal Reserve and the European Central Bank. In the rush to fix things, there was a tendency to look for big, broad solutions, for one-size-fits-all strategies that would give the impression that our leaders, on whose watch the chaos had struck, were at least now in charge of events.
Of course, it was never going to be easy, if only because a crisis born of almost unimaginable complexity was never going to submit to broad-brush countermeasures. In the wake of the first wave of the crisis (for the crisis is still very much with us), Enrique Mendoza and Carlos Vegh of the University of Maryland and I have examined the varying effects of fiscal stimulus programmes using data from 44 countries.
The idea behind fiscal stimulus is to boost demand and nudge economies out of recession. One of the first acts of the newly elected Obama administration in February 2009 was to pass a $787 billion stimulus package through the US Congress. Most recently, in December, President Obama passed, with bi-partisan support, a bill prolonging both the previous Bush administration’s tax cuts and his own administration’s extension of unemployment benefits. Obama justified these measures as an additional attempt to stimulate economic activity. In the developing world, governments made similar attempts; the Chilean government, for example, increased its expenditure by close to 3 per cent of GDP.
Prior to the 2010 election, the UK pursued similar policies, but more recently, the Coalition Government has put more of an emphasis on fiscal austerity, with some ministries facing budget cuts in excess of 25 per cent over the next several years. Austerity has also been the response favoured by eurozone governments. Most recently, the Irish government, facing a loss in market confidence for its sovereign debt, took measures to increase public savings by more than 3 per cent of GDP.
Policy advice by economists on the relative merits of economic stimulus and austerity has been divided. In a January 2009 piece in the Wall Street Journal, the Harvard economist Robert Barro argued that the peacetime “fiscal multiplier” – the dollar increase in GDP caused by a one dollar increase in government spending – is essentially zero. At the other extreme, Christina Romer, when she was chair of President Obama’s Council of Economic Advisers, used multipliers as high as 1.6 in estimating the job gains generated by the 2009 fiscal stimulus package. In the UK, in duelling letters to The Times and the Financial Times, professors from LSE and other universities were similarly divided on whether immediate action was required to reduce fiscal deficits or whether fiscal consolidation would deepen the UK’s recession.
Why so much disagreement? One reason is that the evidence of the effects of fiscal policy on economic activity is limited and often conflicting. One of the biggest hurdles to obtaining precise and consistent estimates of fiscal multipliers has been the limited availability of data. Most studies in this area have relied on evidence from a small number of countries, typically the US. What evidence does exist from other sources shows a very wide range of effects across time and countries. As a result, we sought better, broader and deeper data. Rather than adding yet another estimate of the effects of fiscal stimulus to an already confusing list of conflicting evidence we used international data to ask where and when fiscal stimulus is likely to be effective.
Until recently, most countries reported fiscal data on a quarterly basis, but they didn’t collect data quarterly; instead, they used statistical methods to estimate quarterly patterns. Fortunately, improvements in data collection encouraged by the International Monetary Fund and Eurostat have now made such data available. We worked with the statistical agencies and finance ministries of a number of developing and high-income countries in order to ensure that the data used in their sample of 44 countries was reliably collected on a quarterly basis.
What we found was that the impact of government expenditure shocks depends crucially on key country characteristics, such as the level of development, the exchange-rate regime, the country’s openness to trade, and public indebtedness. Specifically:
- the output effect of an increase in government consumption is larger in industrial than in developing countries;
- the fiscal multiplier is relatively large in economies operating under fixed exchange rates but zero in economies operating under flexible exchange rates;
- fiscal multipliers in open economies are lower than in closed economies; and
- fiscal multipliers in high-debt countries are zero.
A key finding has to do with the role of a country’s exchange rate regime in determining the fiscal multiplier effect of stimulus programmes. The long-run fiscal multiplier is large (approximately 1.5) in countries with fixed exchange rates; in contrast, in countries with flexible exchange arrangements, the long-run multiplier is essentially zero. A similar result was found when we compared countries where trade comprises only a small part of overall economic activity (long-run fiscal multiplier: 1.4) and those highly exposed to international trade (long-run fiscal multiplier: approximately zero).
These findings may help to explain the significant differences in the effects of fiscal policy across countries and time periods that had been found in earlier studies. For example, in a 2004 study, the Italian economist Roberto Perotti estimated that the government expenditure multiplier in the UK declined from 0.1 in 1960-79 to minus 1.2 in 1980-2001. The significant decline in the expansionary power of government purchases in the UK may be explained by the fact that the pound was pegged to the US dollar until the early 1970s, but allowed to float freely thereafter. Moreover, international trade has played an increasing role in the UK’s economic activity, with the ratio of trade to GDP almost doubling since 1960.
We also found a link between monetary policy and fiscal stimulus. Central banks committed to maintaining a stable exchange rate tend to lower interest rates (by an average of 125 basis points) in response to every 1 per cent of GDP rise in government consumption during the two years following a fiscal stimulus. Conversely, central banks with other aims (such as an inflation target) increase interest rates (by an average of 60 basis points); they do so presumably to counteract the inflationary pressures caused by the fiscal expansion, but with the side effect of potentially negating the effects of stimulus altogether.
These findings have important implications for policymakers. The interaction between fiscal and monetary policy is a crucial determinant of the effects of fiscal stimulus. For example, it is vital to consider the reaction of the Bank of England in assessing the potential economic fallout from the UK government’s current austerity measures. If the Bank of England is capable and willing to respond to the Treasury’s actions by maintaining low interest rates and continuing unconventional expansionary monetary policies, our estimates imply that they may be able to significantly mitigate the costs of fiscal austerity. Ireland, because it is part of the eurozone and cannot pursue an independent monetary policy, is in a much tougher bind. The cost of austerity measures pursued there could lower GDP by as much as 4 per cent in the upcoming five years.
This article first appeared in the Spring 2011 issue of LSE Research magazine, with additional reporting by Stryker McGuire.
Please read our comments policy before commenting.