VoxEU Column Macroeconomic policy

Fiscal consolidation: At what speed?

The debate about fiscal consolidation reduces too often to shouting matches about the value of fiscal multipliers, or about the existence of a critical debt-to-GDP ratio. This does not do justice to what is a complex choice, depending on many factors. Our purpose in this article is to review the relevant factors at play and allow for a richer discussion.

In many advanced economies, public debt is very high, and fiscal consolidation must take place. Some factors point to doing more now, others to doing more later. Our purpose in this article is to identify these factors. The right decision, for each country, must depend on a careful weighting of the factors at play.

Less now, more later. Multipliers.

When fiscal multipliers are large, government spending cuts and tax hikes have a large adverse effect on output in the short run, and a small initial effect on the ratio of debt to GDP (Eyraud and Weber, 2013). Indeed, as GDP may initially decline by more than debt, it may lead to an initial increase in the debt-to-GDP ratio, something we have seen in a number of countries in this crisis. (All that is needed, for example, is a multiplier above 1 and the sum of the ratio of revenue-to-GDP and the debt-to-GDP ratio above 100 per cent).

Large multipliers do not necessarily affect the optimal timing of fiscal consolidation, however. If they remain just as large in the future, the adjustment will be as painful later. But, if they are larger now than later, this tilts the adjustment toward doing more later: Less pain now, less pain later. And there are at least three reasons to believe that multipliers are larger now.

First, with nominal interest rates near zero, central banks have less room to cut rates to offset the negative short-term effects of fiscal consolidation on economic activity (Christiano, Eichenbaum, and Rebelo, 2011, for example.) As nominal interest rates normalise in the future, central banks may again be better able to offset adverse effects of fiscal consolidation. In other words, multipliers might then be closer to zero.

Second, a poorly functioning financial system with tighter credit constraints implies that current consumption depends more on current than on future income, and that investment depends more on current than on future profits. Both effects lead to larger multipliers (Eggertsson and Krugman, 2012, and others). As the financial system is healing in a number of economies, this effect may, however, be less important now than it was at the start of the crisis.

Third, a number of empirical studies have found that fiscal multipliers are larger during periods of slack (Auerbach and Gorodnichenko, 2012, and others; for a dissenting view however, see Owyang, Ramey, and Zubairy, 2013).

While this is still a subject of some debate, we read the evidence as showing that fiscal multipliers have indeed been larger than in normal times, especially at the start of the crisis (Blanchard and Leigh, 2013).

Less now, more later. The dangers of low growth.

A second argument in favour of backloading is that, when growth is low, the economy is more vulnerable to “stalling” and slipping into recession (Nailwaik, 2011, Sheets, 2011, and others). Tightening fiscal policy when growth is low is thus riskier than when growth is near normal. (A more accurate statement would probably replace “low growth” by “large output gap” as it is really the level of activity that matters here. But the literature has focused on growth rather than on the gap.)

In this sense, the large fiscal consolidation taking place in the United States this year is not as bad, given relatively strong private demand, as the same fiscal consolidation would be in countries where private demand is very weak.

The empirical evidence on stall speed is actually mixed (Ho and Yetman, 2013). But there are good reasons to think that adverse shocks to growth when growth is already very low can lead to a number of vicious cycles. For example, a decrease in growth from, say, 4 per cent to 2 per cent has only a moderate effect on the proportion of nonperforming loans. A decrease in growth from 0 per cent to –2 per cent has a much larger impact, which in turn leads to weaker banks, forcing them to further tighten credit, thus further decreasing growth.

A related but separate argument relies not on nonlinear effects on output, but on nonlinear effects on welfare. A tax hike that reduces household disposable income is more painful, in terms of welfare losses, if implemented when other developments are already weighing household income down. A further increase in unemployment is more painful when unemployment is already very high.

Less now, more later. Hysteresis.

A third argument in favor of backloading is that fiscal consolidation carries the risk of causing long-term economic damage through hysteresis effects. DeLong and Summers (2012) have argued that a process of “hysteresis” links the short-term cycle to the long-term trend, implying more persistent fiscal policy effects.

Just like large multipliers, however, hysteresis does not necessarily have implications for the timing of fiscal consolidation. The adjustment will be just as painful in the future as it is today. What is needed to tilt the desirable adjustment path is for hysteresis to be stronger today than in the future.

As in the case of multipliers, there are indeed good reasons to think that hysteresis is stronger now. One mechanism goes through unemployment duration. The proportion of long-term unemployed is nonlinear in the unemployment rate. So long as the unemployment rate is relatively low, few people are long-term unemployed. When unemployment becomes high, the proportion of long-term unemployed increases quickly. To the extent that the long-term unemployed are more likely to give up or lose skills, and become unemployable, hysteresis is indeed more of an issue when unemployment is high, as it is indeed today.

More now, less later. Debt overhang and multiple equilibria.

The motivation behind fiscal consolidation is that high debt presents costs and risks, and it is therefore essential to reduce it. Textbooks focus on two types of costs, the displacement of capital by debt, and the distortions implied by the higher taxes needed to service the debt. While both indeed reduce growth, there are at least two even more relevant considerations.

The first one is debt overhang. If bond holders believe that the state may default on its bonds, they will require a risk premium and thus a higher interest rate. The crisis has made investors question the safety of government bonds, and many sovereign bonds now indeed have to pay higher spreads. Not only do these higher sovereign spreads make it harder for the government to sustain debt, they typically lead to higher spreads for private borrowers as well (Zoli, 2013). Also, uncertainty associated with the possibility of default feeds uncertainty about taxation and inflation, and all these effects are likely to reduce growth.

The second one, which is closely related, is the risk of multiple equilibria. When debt levels are high, but not so high that default is certain, there are likely to be two, self-fulfilling, equilibria: “good” and “bad.” The “good” equilibrium is where investors believe that the probability of default is low and ask for a low interest rate. The “bad” equilibrium is where investors believe the probability of default is higher and ask for a higher interest rate to compensate for the risk, making it harder for the government to avoid default, and thus justifying their initial beliefs. The higher the level of debt, the closer the two equilibria, and the more likely that, at some point, the economy suddenly shifts to the bad equilibrium. There is no magic number, but at, say, a 100 per cent debt-to-GDP ratio, a 5 percentage point increase in the interest rate, which leads to an increase in the interest bill of 5 per cent of GDP, may well make the difference between sustainability and eventual default.

Since it is nearly impossible to know what will make investors shift their beliefs, the situation policymakers face here is one of “Knightian uncertainty.” The prudent approach to dealing with such “unknown unknowns,” to use former US defence secretary Donald Rumsfeld’s expression, is to move away from the danger zone. How to do this best, however, is not obvious.

In theory, one way of avoiding these multiple equilibria is to have the central bank stand ready to enforce the good equilibrium, that is, to stand ready to intervene in the market to maintain the low rate. This argument is sometimes given to explain why the United Kingdom has been able to maintain very low sovereign rates, whereas Spain, with slightly less government debt, is paying a substantial premium. The Bank of England could intervene, while the ECB does not have the mandate to do so (the OMT program aims at removing “redenomination risk,” namely at removing multiple equilibria associated with the tail risk of a euro area break-up; this eliminates one source of multiple equilibria, but not the one described earlier). This probably exaggerates the role that central banks can play: Knowing whether the market indeed exhibits the good or the bad equilibrium, and what the interest rate associated with the good equilibrium might be is far from easy to assess, and the central bank may be reluctant to take what could be excessive risk onto its balance sheet.

So, how do countries get out of the danger zone? Even with a large and steady fiscal consolidation, decreasing the debt-to-GDP ratio from, say, 100 per cent to 60 per cent is a slow process, likely to take decades. So the bad news is: Debt will be high for a long time. There is, however, some good news as well: The evidence shows that markets, to assess risk, look at much more than just current debt and deficits. In a word, they care about credibility. The danger zone is not defined by a magic threshold for the debt-to-GDP ratio, but by a much more complex set of characteristics of the fiscal and economic situation.

How best to achieve credibility? A medium-term plan is clearly important. So are fiscal rules, and, where needed, retirement and public health care reforms which reduce the growth rate of spending over time. The question, in our context, is whether frontloading increases credibility.

If one measures credibility by the size of the sovereign spread, the econometric evidence from the crisis is ambiguous. Smaller deficits appear to reduce spreads, but lower growth increases them (Cottarelli and Jaramillo, 2012). Thus, whether or not faster fiscal consolidation decreases spreads depends on whether the effect of a smaller deficit dominates the effect of lower growth. This, in turn, depends on the size of fiscal multipliers. And, for a plausible range of multipliers, the answer can go either way.

The econometric evidence is rough, however, and may not carry the argument. Adjustment fatigue and the limited ability of current governments to bind the hands of future governments are also relevant. Tough decisions may need to be taken before fatigue sets in. One must realise that, in many cases, the fiscal adjustment will have to continue well beyond the tenure of the current government. Still, these arguments support doing more now.

Bottom line

Our discussion shows how deciding on the appropriate speed of fiscal consolidation requires much more than an assessment regarding the size of short-term fiscal multipliers or the debt-to-GDP ratio. While our discussion does not argue in favor of any specific fiscal policy stance in any specific country, we see it as consistent with the call for credible, medium-term, fiscal consolidation plans. We also see it as consistent with the view that, in advanced economies, the pace of fiscal consolidation should depend on, and be adjusted as a function of, the strength of private demand. (Economies that have lost market access obviously do not have such flexibility.) These principles have been at the heart of IMF recommendations since the start of fiscal consolidation (see, for example, Blanchard and Cottarelli, 2010), and we believe they continue to apply.

The views expressed in this article are the sole responsibility of the authors and should not be attributed to the International Monetary Fund, its Executive Board, or its management.

References

Auerbach, Alan, and Yuriy Gorodnichenko (2012), “Measuring the Output Responses to Fiscal Policy,” American Economic Journal – Economic Policy, Vol. 4, pp. 1–27.

Blanchard, Olivier, and Daniel Leigh (2013), “Growth Forecast Errors and Fiscal Multipliers,” IMF Working Paper No. 13/1 (Washington: International Monetary Fund).

Blanchard, Olivier, and Carlo Cottarelli (2010), “Ten Commandments for Fiscal Adjustment in Advanced Economies”, IMFdirect, June 24. 

Christiano, Lawrence, Martin Eichenbaum, and Sergio Rebelo, 2011, “When Is the Government Spending Multiplier Large?” Journal of Political Economy, Vol. 119, pp. 78–121.

Cottarelli, Carlo, and Laura Jaramillo (2012), “Walking Hand in Hand: Fiscal Policy and Growth in Advanced Economies”, IMF Working Paper No. 12/137 (Washington: International Monetary Fund).

DeLong, Bradford, and Lawrence Summers (2012), “Fiscal Policy in a Depressed Economy”, Brookings Papers on Economic Activity, Spring 2012.

Eggertsson, Gauti B, and Paul Krugman (2012), “Debt, Deleveraging, and the Liquidity Trap”, Quarterly Journal of Economics, pp. 1469–513.

Eyraud, Luc, and Anke Weber (2013), “The Challenge of Debt Reduction during Fiscal Consolidation,” IMF Working Paper No. 13/67 (Washington: International Monetary Fund).

Ho, Wai-Hip Alex, and James Yetman (2013), “Do Economies Stall? The International Evidence”, BIS Working Paper No. 407, March.

Nalewaik, Jeremy J (2011), “Forecasting Recessions Using Stall Speed”, Finance and Economics Discussion Series, Federal Reserve Board, 2011-24.

Owyang, Michael, Valerie Ramey, and Sarah Zubairy (2013), “Are Government Spending Multipliers Greater During Periods of Slack? Evidence from 20th Century Historical Data,” The American Economic Review: Papers and Proceedings (forthcoming).

Perotti, Roberto (2011), “The ‘Austerity Myth’: Gain Without Pain?” NBER Working Paper No. 17571 (Cambridge, Massachusetts: National Bureau of Economic Research).

Sheets, Nathan (2011), “Is the US Economy Approaching ‘Stall Speed’?” Citi Empirical and Thematic Perspectives, 11 November.

Zoli, Edda (2013), “Italian Sovereign Spreads: Their Determinants and Pass-through to Bank Funding Costs and Lending Conditions,” IMF Working Paper No. 13/84 (Washington: International Monetary Fund).

19,632 Reads