Lionel Robbins defined economics as the study of the allocation of scarce resources among competing uses. For an economy at full employment, where the opportunity cost of government spending is the private spending it displaces, this remains a good characterisation. But what if there is a deficiency of aggregate demand, so the nation’s resources are being underused – as evidenced by significant involuntary unemployment? In this case, in so far as it employs unused capacity, expenditure by the government will add to the resources available for investment and consumption. This is conventionally measured by the “multiplier” – the ratio of extra output available per unit of government expenditure. If the multiplier is equal to one, for example, there is no opportunity cost to public spending, in effect: the resources would otherwise be unused.
In his 1933 essay “The Means to Prosperity”, John Maynard Keynes argued that “a very conservative figure . . . makes the multiplier to be at least two” . Increased tax take and reduced benefit payments would reduce the net budgetary cost of a fiscal expansion considerably. But this calculation was explicitly contingent on “present circumstances”. In conditions of high employment, as J Bradford DeLong and Lawrence Summers note in their March 2012 Brookings paper, “a policy relevant multiplier close to zero is in fact likely to be a better approximation for thinking about discretionary policy”. There would be no output gain from increased spending, only a rise in prices.
The contrast between Robbins’s perspective and that of Keynes is echoed in the fierce debate in Europe on how far and how fast to proceed with fiscal consolidation in present circumstances. The message we can take from the earlier debate is that consolidation plans should take account of potential positive externalities. In particular, to “consolidate” in a slump is misguided: one should be exploiting the positive externality of public spending to expand demand. This, in a nutshell, is what the chief economist at the IMF, Olivier Blanchard, argues in a recent paper written with a colleague, Daniel Leigh.
In fact, DeLong and Summers made an even stronger case for government spending in a slump. Not only would this generate positive benefits in terms of current output, it could also avoid negative “hysteresis” effects on future supply. The argument is that factors unused in the present ipso facto become less usable in the future.
Like Keynes, DeLong and Summers were concerned with the blight of unused resources and how to mitigate this. Nothing in their analysis, however, would countermand a policy of additional fiscal consolidation if that were judged appropriate, with spending reductions to reduce the ratio of debt to GDP, for example, as and when the economy regained full employment.
This brings us to the latest contribution to the austerity debate from Ryan Bourne and Tim Knox, both of the Centre for Policy Studies. They argue that DeLong and Summers must be wrong because a stimulus would imply a “free lunch”, and Milton Friedman said there is no such thing! They go on to question the evidence for hysteresis and to warn of the danger that the costs of borrowing may be higher than the two Americans assume. How damning are these arguments?
The crucial issue now, as in the 1930s, is where things stand at the time of asking; and what can be expected by private agents acting in their own interest. If aggregate demand lies significantly below supply, there is a prima facie case for a government either to spend more in its own right or to subsidise private expenditure. How then do things stand?
Under George Osborne’s three-year stewardship, the British economy has hardly grown: GDP today is still 2.6 percentage points lower than its pre-recession peak, the slowest rate of recovery since the 19th century. Firms are piling up cash; overleveraged households are holding back spending; the economy is virtually at a standstill. Monetary policy easing may have mitigated a slump – but it has not triggered recovery. So the conditions for positive externalities from fiscal expansion policy are satisfied.
What of the argument that the costs of debt service are higher than DeLong and Summers allowed for? The yields on indexlinked borrowing in the UK are in fact less than zero – far below the rates that DeLong and Summers assume in their analysis.
Bourne and Knox dismiss hysteresis effects, citing the work of an American economist who found no evidence that “temporary increases in government spending have permanently increased output in the past”. We are given no information about the data leading to this conclusion: but unless it is restricted to periods when there was undoubted spare capacity, it will miss the point.
It is not only in the UK but all across Europe that untimely fiscal consolidation has cast its pall of austerity and despair – on the youth above all. So why not, as Blanchard and Leigh put it, do your consolidation later when the costs are low? The usual counterargument is one of credibility: that one cannot trust a government that spends more today and says it will spend less later. This is not an argument that Bourne and Knox deploy but it is worth taking seriously. What it suggests, however, is not prompt consolidation in a slump, but medium-term planning for consolidation on recovery.
The two CPS economists promote inappropriate, pro-cyclical fiscal policy action now in the self-defeating pursuit of the long-term objective of debt reduction. The policy we recommend – minimising output losses now because current consolidation is counterproductive and a buoyant economy will better support the aim of long-term debt reduction – is both more rational and more humane.
Marcus Miller and Robert Skidelsky are professors of economics at the University of Warwick