In partial defence of fiscal austerity

Recent issues of this Newsletter have included articles somewhat critical of current austerity policies. In this contribution John Fender1, University of Birmingham, puts a partial defence.

Other items related to this theme, published in the Newsletter, include:
LSE blog questions UK budgetary policy (July 2010)
Are we suffering from deficit fetishism? (April 2011)
Debating austerity measures (July 2011)
Surely you're joking, Mr Keynes ( October 2011)
Fiscal stimulus improves solvency in a depressed economy (April 2012)
Understanding the crisis: clarity on measurement, clarity on policy (April 2012)


There has been much debate recently over the UK coalition government’s fiscal consolidation programme. A number of economists, including Robert Neild in the January 2012 Royal Economic Society Newsletter and John Weeks and Dennis Leech in the April 2012 Newsletter, have argued that the pace of fiscal consolidation is too fast and should be slowed down. ‘A Manifesto for Economic Sense’ which was published in the Financial Times of 28th June 2012 is similarly critical of austerity policies. This article is more sympathetic to the fiscal consolidation policy and argues that relaxing the policy is unlikely to have much of a positive effect on economic activity and could well be risky.

In any analysis of the current economic situation, it is important to acknowledge that we do not live in ‘normal’ economic times any more. The last few years have seen:

  • a major world financial crisis;
  • the largest banking crisis in the UK’s history;
  • the greatest worldwide decline in output (in 2008 - 9) since the Great Depression;
  • the largest ever peacetime public sector deficits in the UK;
  • a massive increase in private sector indebtedness in many countries including the UK;
  • a major on-going crisis in the euro-zone.

These events are of enormous importance in understanding what has been happening in the UK economy recently and in formulating appropriate policy.

1. The fiscal policy multiplier in good and bad times.

It is crucial to the arguments of those opposing austerity that the fiscal policy multiplier is reasonably large, or at least can be expected to be large in current economic circumstances. And indeed there are some recent papers which do argue for a large multiplier: for example, Romer and Romer (2010) estimate a multiplier of nearly 3 for tax changes and Christiano et al. (2011) suggest that a government-spending multiplier of 3.7 might be plausible in a constant interest rate environment. However, a crucial question is whether these multipliers are large in times of high (and/or rapidly rising) public sector debt, and there is evidence that in such times fiscal policy multipliers may be much lower and could indeed be negative. This is the message of Perotti (1999), who argues that the effects of fiscal policy in ‘good times’ (i.e. low debt) may be very different from its effects in ‘bad times’ (i.e. high debt). Ilzetzki et al. (2010) present evidence that the fiscal multiplier is zero in countries with debt-GDP ratios above 60 per cent. It has in fact even been suggested that fiscal contraction may be expansionary: Giavazzi and Pagano (1990) argue that the fiscal consolidations undertaken by both Ireland and Denmark in the 1980s were expansionary. There is also evidence (see Reinhart and Rogoff, 2010) that in countries with debt-GDP ratios greater than 90 per cent economic performance deteriorates sharply.

The question that arises, of course, is why low and possibly negative fiscal policy multipliers may occur with high public sector debt. The main mechanism in the literature (see, e.g. Sutherland, 1997) is somewhat as follows. Suppose a country is experiencing a rapidly rising public debt which is unsustainable. Some consolidation is necessary; the only question is when will it be introduced? The longer the delay in introducing the policy, the more painful it will be. The sudden introduction of a fiscal consolidation programme removes the uncertainty about when it will be introduced and means that it is less painful than expected, and for both these reasons it is expansionary. However, there is a complementary explanation for how a fiscal consolidation programme can be expansionary: the programme reduces the budget deficit over a number of years, so at the end of the programme, the stock of public debt is considerably lower than it otherwise would have been. If the debt takes the form mainly of long-term government bonds, and assuming that these bonds are imperfect substitutes for other assets, including shorter term bonds, the lower supply of such bonds means their price will be higher, and hence their yield will be lower. So future long-term interest rates will be lower than they otherwise would have been and with foresight current long-term interest rates will be lower as well. So the policy works by reducing current and expected long-term interest rates; it does this given the time path of expected short-term rates, meaning it reduces the term premium on government bonds. Lower long-term interest rates may increase both consumption and investment spending as suggested by the textbooks. They may well mean higher asset prices, and these may stimulate spending in a variety of ways. Higher share prices may stimulate consumption spending through a wealth effect and investment spending by making it easier for firms to raise equity capital. An increase in asset values may strengthen firms’ balance sheets, and this may encourage bank lending. Perhaps most importantly, it may result in a depreciation of the exchange rate (it raises the price of foreign currency, another asset), and this may stimulate demand by raising exports and shifting domestic spending from imports to domestic goods. So, there are a number of ways in which a fiscal consolidation programme may raise spending through reducing longer term interest rates. For the overall policy to be expansionary, it is necessary that these indirect effects outweigh the direct effects of the policy. Of course, even if they do not completely offset these expansionary forces, they may offset them partially and make the contraction less severe than it otherwise would have been.

A recent IMF study (IMF 2010) has sometimes been cited as evidence that fiscal consolidation is contractionary; the main finding is that a ‘fiscal consolidation equal to 1 percent of GDP typically reduces GDP by about 0.5 percent within two years’ (op. cit., p. 94). But this article does not contend that fiscal contraction is never contractionary. Indeed, the evidence is overwhelming that on average fiscal contraction is contractionary. Rather, the contention is that when there is a ‘problem’ with the public sector debt, the fiscal policy multiplier may well be much smaller than on average and could possibly be negative.

2. Possible consequences of relaxing the fiscal consolidation programme

However, we must ask whether the above argument is applicable to the United Kingdom, which currently has a public debt to GDP ratio of about 65 per cent and a public sector deficit to GDP ratio of about 8 - 9 per cent. We consider the effects of relaxing the fiscal consolidation programme in the light of the above arguments.

(i) Effects of a fully credible relaxation of the consolidation programme.
Suppose we consider a package of a two-year reduction in VAT and some increase in public expenditure (relative to what has been announced) for the same time period. Assume the policy is fully credible (i.e. it is believed that the programme will be implemented as announced). Then the’direct’ effects of the policy should be expansionary But there will be some indirect effects going in the opposite direction; higher future taxes may be expected, and there will be the effects described in the previous section whereby the expectation of higher debt raises longer term interest rates and asset prices. All these effects will tend to reduce spending. But the net effect could still be positive.

(ii) Effects of a less than fully credible relaxation of the consolidation programme (with no possibility of default).
The above discussion assumed that the policy (of temporarily raising spending and reducing taxation) is fully credible. However, this is a dubious assumption. For example, suppose a two-year reduction in VAT is announced. This would involve raising VAT in the year before an election, which politicians may be reluctant to do — perhaps they will claim that the recovery 'is not yet fully consolidated' and postpone restoring VAT to its previous rate. Having relaxed fiscal policy once, the government might be more inclined to do so again. So a greater increase in public sector debt would be expected, and the contractionary forces listed above would be stronger. This would be so even if there is complete confidence that the government will ultimately repay its debts. But the effects might be greater if the policy change erodes this confidence.

(iii) Effects of a less than fully credible relaxation of the consolidation programme (with some chance of default)
It might be argued that there is no chance of any UK government defaulting on its debt. It has not, in over 300 years! But, for there to be a problem, it is not necessary that a default is likely, it is merely sufficient that there be just a small probability that default might take place in certain circumstances, and markets would react negatively. Interest rates on government debt would rise, and the possibility of self-fulfiling expectations arises — interest rates rise, making financing the public debt more expensive, raising the probability of default still further, and so on. Several countries in the euro-zone are familiar with this problem.

A scenario that might lead to a positive expectation of default is as follows: suppose a laxer fiscal policy is pursued over the next two years, and a negative shock then sends the deficit still higher. The deficit is over (say) £100bn. and the debt-GDP ratio is about 80 per cent (and rising) in 2014-15 with a general election pending. Uncertainty about the outcome of the election (with the chance that the resultant government will not be tough on the deficit) could provoke a crisis of confidence. However, although default would occur with unchanged policies, what is much more likely is that the crisis would force a rapid policy change (higher taxes and lower government expenditure) which would restore confidence but which would be much more painful than the current measures.

3. Why is output stagnant in the UK?

The recent low growth in output in the UK has attracted much comment and discussion. We are now in a ‘double-dip’ recession, with output about 4 per cent below its pre-crisis peak. Why has output growth been sluggish? There are several possible explanations. Firstly, and perhaps most importantly, we should mention the aftermath of the credit crunch. There is evidence that recovery from financial crises is often long and painful (e.g. Reinhart and Rogoff, 2010). Many consumers and firms may be over-leveraged, and take considerable time to reduce their debts and start spending at a more rapid pace. Many consumers may find it difficult or expensive to borrow, and hence adopt a more cautious approach, saving up to buy items they might previously have bought on credit. Banks may be unwilling to lend too freely, and may be concerned about future regulatory changes. So, it seems, there will be a period of painful adjustment following a credit crisis which may last many years. However, as households adjust, then one would expect them to spend at a higher rate. The second reason why growth has been sluggish is surely the euro-zone crisis. This has affected exports to the euro zone and lending by banks that hold the debt of the countries that may default, or who are indirectly so exposed. A third plausible reason for sluggish economic growth is high and volatile commodity prices, which may have negative effects on output through both supply and demand side channels.

The fiscal consolidation programme may have been another factor. However, the fact that the UK economy now has record low interest rates (both short term and long term) may be to some extent a product of the fiscal consolidation programme and these may well promote expansion, as might the competitive exchange rate (sterling depreciated by about 25 per cent between mid-2007 and early 2009).

As credit constraints gradually unwind, consumers may spend more freely, although this may take a long time. Confidence that the consolidation programme is working may be another force tending to promote expansion, as may lower commodity prices.

4. History of government debt in the UK

One argument often used by opponents of the fiscal austerity programme is that UK government debt is fairly modest in comparison to its level throughout much of its history. For example, it was well over 200 per cent of GDP after both the Napoleonic Wars and the Second World War, so it might be asked why a debt-GDP ratio of about 65 per cent, which is approximately its current level, be a major concern? (See chart.) This is undoubtedly an important question, to which the following remarks may provide a very brief and inadequate answer.

(a) There is evidence (see Reinhart et al, 2012) that real long-term interest rates have been higher in times of high public sector debt in the UK than in periods of lower debt.

(b) Government debt was clearly on a downward path in both periods (i.e. between 1815 and 1914 and after 1945). For example, the postwar debt-GDP ratio peaked at 238 per cent in 1947; in 1952 it was 162 per cent, in 1957 122 per cent and by 1962 it was just under 100 per cent. The above discussion has suggested that what may be relevant for crowding out is not the current deficit, nor the current debt, but the expected future time path of the debt. Possibly a high but rapidly falling national debt may be less damaging than a low but rapidly increasing national debt.

(c) Reasons for the downward trajectory of national debt in the Victorian period were conservative budgetary policy, and both economic and population growth. In the post 1945 period, inflation and economic growth combined with fairly modest budget deficits were the main factors. However, it is unlikely that either population growth or inflation will reduce debt significantly in the next few years.

(d) Demands upon the state were much lower during these time periods. In much of the nineteenth century, over 90 per cent of central government spending was on debt interest and defence. In the period after 1945, it is true that the welfare state was built, but demands on the welfare state were far more modest than they are now. There are a number of reasons for this. In the early post war period, the proportion of pensioners was far lower, as was the number of university students and the proportion of school children educated beyond the age of 15. Spending on the health service (3 per cent in the early post war era as opposed to about 8 per cent today) was much lower as a percentage of GDP. This list of reasons could surely be lengthened.

(e) Debt interest payments (at about 6 per cent of GDP) were relatively modest in the immediate postwar period in part because much of the debt was issued in 'forced lending' campaigns during the war. Also, as argued by Reinhart and Sbrancia (2011), the government managed to reduce its debt servicing costs by implementing various types of financial repression.

(f) Demographic factors such as an ageing population may mean that the ‘true’ government debt at the moment is much higher than that quoted (because of higher pension costs, extra health expenditures in the future, etc.)


It seems, then, that although debt was extremely high in the periods mentioned, it was clearly affordable, there was no risk of default and it was on a steady downward path. This contrasts with the current situation, where debt is rapidly rising, there are considerable demands on the government budget, and a significant increase in the ratio of taxes to GDP may not be feasible politically. So, the fact that debt has been much higher in the past in the UK than at present is no reason for being sanguine about the current level of debt (or its rate of increase).

5. Qualifications

The Chancellor seeks to eliminate the ‘structural deficit’. However, it might be argued that we do not need to do much more than stabilise the debt-GDP ratio and then set it on a gradually declining path. A deficit of 3 per cent should do this — with public sector debt peaking at about 80 per cent, and nominal GDP growth of about 5 per cent, this should produce a declining debt-GDP ratio. Also, there is nothing in the above analysis to suggest that fiscal consolidation is appropriate for the euro zone. Indeed many of the mechanisms which might bring about an ‘expansionary fiscal contraction’ are neutralised by the fact that the countries are members of a common currency area, and hence unable to adjust their exchange rate or interest rates vis-à-vis the rest of the zone.

It should be clear that this article does not argue that it is never appropriate to use expansionary fiscal policy to combat a downturn in economic activity. Indeed, the expansionary fiscal policies pursued by many countries in response to the 2008-9 financial crisis were almost certainly justified and prevented the recession, bad as it was, from becoming far worse. Rather, the contention is that when the public sector debt is high, and/or rising rapidly, expansionary fiscal policies are far less likely to be successful in boosting economic activity.

A further point is that the macroeconomic consequences of public sector debt may well depend on what the debt is used for. For example, public sector debt used to finance infrastructure projects may have very different effects from that issued to finance social security spending. This is certainly something that needs further research.

6. Conclusion

There is some evidence that the fiscal consolidation programme is working - the deficit-GDP ratio was 11.2 per cent in 2009-10, but fell to about 8.4 per cent in 2011-12, a reduction of 2.8 per cent or a quarter. Figures so far for the current financial year are not encouraging, but figures for both public expenditure and taxation are notoriously volatile. But even though the deficit-GDP ratio may have come down, public sector debt is still rising at a rapid rate.

The basic argument of this article is that a relaxation of the fiscal consolidation programme is unlikely to have a significant expansionary effect and could well be quite risky. Government debt even with current policies seems likely to rise above levels that might be deemed desirable, and pursuing a policy which might allow the ratio to go much higher would be very risky indeed. Relaxation of the programme would be likely to result in higher long-term interest rates, lower share and asset prices and an appreciated exchange rate, as well as expectations that there would eventually have to be higher taxation and/or lower expenditure to get the debt under control, and these are likely to exert a contractionary force on the economy. We have also argued that the fiscal consolidation programme is not responsible, in any major way, for the recent anaemic growth in output. Almost certainly, a number of other factors are responsible for this, in particular the after effects of the financial crisis of 2008-9.

It seems that the economy is in a strange kind of trap, with both conventional monetary policy and fiscal policy unable to stimulate the economy any further, because of the ‘zero lower bound constraint’ and the deficit/debt problem, respectively. However, the Bank of England might be expected to continue to pursue unconventional monetary policies such as Quantitative Easing, which may have some positive effect. The government might pursue policies design to free up bank lending, but there is not much else it can do; it needs to persist with its current programme, and, in due course, as credit constraints are unwound, the economy might be expected to recover.

One would not rule out relaxing the consolidation programme under all circumstances. Faced with a major collapse in the euro zone, for example, some strictly temporary and appropriately targeted government expenditure increases and tax reductions might be implemented. However, it would still be necessary to do this in a way which preserved the credibility of the government’s deficit reduction policy. And more generally, it needs to be emphasised that preserving the credibility of the government’s deficit reduction policy is absolutely key to its success. Credibility is not something that can be achieved easily, and politicians should be very wary of taking any actions that might jeopardise this credibility. This is something that has perhaps not been appreciated sufficiently in recent debates.

Note:

1. Thanks to Tim Besley, Colin Ellis and Neil Rankin for helpful comments on an earlier draft. Needless to say, the views expressed in this article are entirely those of the author, who takes full responsibility for any mistakes it contains.

References:

Ardagna, S. (2009), ‘Financial Markets’ Behavior around Episodes of Large Changes in the Fiscal Stance’, European Economic Review 53, pp. 37 - 55.

Christiano, L., Eichenbaum, M. and Rebelo, S. (2011), ‘When is the Government Spending Multiplier Large?’ Journal of Political Economy 119(1), pp. 78 - 121.

Giavazzi, F. and Pagano, M. (1990), ‘Can Severe Fiscal Contractions Be Expansionary? Tales of Two Small European Countries’, NBER Macroeconomics Annual 1990, pp. 75 - 122.

Ilzetzki, E., Mendoza, E. and Végh, C. (2010), ‘How Big (Small?) are Fiscal Multipliers?’ NBER Working Paper No. 16479.

IMF (2010), ‘Will it Hurt? Macroeconomic Effects of Fiscal Consolidation’, chapter 3, World Economic Outlook, pp. 93 - 124.

Laubach, T. (2009), ‘New Evidence on the Interest Rate Effects of Budget Deficits and Debt’, Journal of the European Economic Association, 7(4), pp. 858 - 885.

Leech, D. (2012), ‘Fiscal Stimulus Improves Solvency in a Depressed Economy’, RES Newsletter no. 157, pp. 11 - 13.

Neild, R. (2012), ‘The National Debt in Perspective’, RES Newsletter no. 156, pp. 20 - 22.

Perotti, R. (1999), ‘Fiscal Policy in Good Times and Bad’, Quarterly Journal of Economics, 114(4), pp. 1399 - 1436.

Reinhart, C. and Rogoff, K. (2010), ‘Growth in a time of Debt’, American Economic Review: Papers & Proceedings 100, pp. 573 - 578.

Reinhart, C. and Sbrancia, M. B. (2011), ‘The Liquidation of Government Debt’, NBER Working Paper No. 16893.

Romer, C. and Romer, D. (2010), ‘The Macroeconomic Effects of Tax Changes: Estimates Based on a New Measure of Fiscal Shocks’, American Economic Review, 100, pp. 763 - 801.

Sutherland, A. (1997), ‘Fiscal Crises and Aggregate Demand: Can High Public Debt Reverse the Effects of Fiscal Policy?’ Journal of Public Economics, 65, pp. 147 - 62.

Weeks, J. (2012), ‘Understanding the Crisis: Clarity on Measurement, Clarity on Policy’, RES Newsletter no. 157, pp. 7 - 10.

From issue no. 159, October 2012, pp.15-17 and 22.

Page Options