By Martin Wolf

Might fiscal expansion be a free lunch? This is the question addressed in a thought-provoking paper “Fiscal Policy in a Depressed Economy”, March 2012, by Brad DeLong and Larry Summers, the most important conclusion of which is obvious, but largely ignored: the impact of fiscal expansion depends on the context. *

In normal times, with resources close to being fully utilised, the multiplier will end up very close to zero; in unusual times, such as the present, it could be large enough and the economic benefits of such expansion significant enough to pay for itself.

In a liquidity trap fiscal retrenchment is penny wise, pound foolish.

Indeed, relying on monetary policy alone is the foolish policy: if it worked, which it probably will not, it does so largely by expanding stretched private balance sheets even further.

As the authors note: “This paper examines the impact of fiscal policy in the context of a protracted period of high unemployment and output short of potential like that suffered by the United States and many other countries in recent years. We argue that, while the conventional wisdom rejecting discretionary fiscal policy is appropriate in normal times, discretionary fiscal policy where there is room to pursue it has a major role.”

There are three reasons for this.

1. First, the absence of supply constraints means that the multiplier is likely to be large.

It is likely to be made even bigger by the fact that fiscal expansion may well raise expected inflation and so lower the real rate of interest, when the nominal rate is close to zero.

2. Second, even moderate hysteresis effects of such fiscal expansion, via increases in the likely level of future output, have big effects on the future debt burden.

3. Finally, today’s ultra-low real interest rates at both the short and long end of the curve, suggest that monetary policy is relatively ineffective, on its own.

The argument is set out in a simple example. “Imagine a demand-constrained economy where the fiscal multiplier is 1.5, and the real interest rate on long-term government debt is 1 per cent. Finally, assume that a $1 increase in GDP increased tax revenues and reduces spending by $0.33. Assume that the government is able to undertake a transitory increase in government spending, and then service the resulting debt in perpetuity, without any impact on risk-premia.

“Then the impact effect of an incremental $1.00 of spending is to raise the debt stock by $0.50. The annual debt service needed on this $0.50 to keep the real debt constant is $0.005. If reducing the size of the current downturn in production by $1.50 avoids a 1 per cent as large fall in future potential output – avoids a fall in future potential output of $0.0015 – then the incremental $1.00 of spending now augments future tax-period revenues by $0.005. And the fiscal expansion is self-financing.”

This is a very powerful result.

It rests on the three features of the present situation: high multipliers; low real interest rates; and the plausibility of hysteresis effects.

A table in the paper (Table 2.2) shows that at anything close to current real interest rates fiscal expansion is certain to pay for itself even with zero multiplier and hysteresis effects: it is a “no-brainer”.

Why is this? It is because the long-term real interest rate paid by the government is below even the most pessimistic view of the future growth rate of the economy. As I have argued on previous occasions, the US (and UK) bond markets are screaming: borrow.

Of course, that is not an argument for infinite borrowing, since that would certainly raise the real interest rate substantially!

Yet, more surprisingly, the expansion would continue to pay for itself even if the real interest rate were to rise far above the prospective growth rate, provided there were significantly positive multiplier and hysteresis effects.

Let us take an example: suppose the multiplier were one and the hysteresis effect were 0.1 – that is to say, the permanent loss of output were to be one tenth of the loss of output today. Then the real interest rate at which the government could obtain positive effects on its finances from additional stimulus would be as high as 7.4 per cent.

Thus, state the authors, “in a depressed economy with a moderate multiplier, small hysteresis effects, and interest rates in the historical range, temporary fiscal expansion does not materially affect the overall long-run budget picture.” Investors should not worry about it. Indeed, they should worry far more about the fiscal impact of prolonged recessions.

Are such numbers implausible? The answer is: not at all.

Multipliers above one are quite plausible in a depressed economy, though not in normal circumstances. This is particularly true when real interest rates are more likely to fall, than rise, as a result of expansion.

Similarly, we know that recessions cause long-term economic costs. They lower investment dramatically: in the US, the investment rate fell by about 4 per cent of gross domestic product in the wake of the crisis. Businesses are unwilling to invest, not because of some mystical loss of confidence, but because there is no demand.

Again, we know that high unemployment has a permanent impact on workers, both young and old. The US, in particular, seems to have slipped into European levels of separation from the labour force: that is to say, the unemployment rate is quite low, given the sharp fall in the rate of employment. Workers have given up. This is a social catastrophe in a country in which work is effectively the only form of welfare for people of working age.

Indeed, we can see hysteresis effects at work in the way in which forecasters, including official forecasters, mark down potential output in line with actual output: a self-fulfilling prophecy if ever there was one. This procedure has been particularly marked in the UK, where the Office for Budget Responsibility has more or less eliminated the notion that the UK is in a recession. Yet such effects are not God-given; they are man-chosen. They are the product of fundamentally misguided policies.

This is an important paper. It challenges complacent “do-nothingism” of policymakers, let alone the “austerians” who dominate policy almost everywhere. Policy-makers have allowed a huge financial crisis to impose a permanent blight on economies, with devastating social effects. It makes one wonder why the Obama administration, in which prof Summers was an influential adviser, did not do more, or at least argue for more, as many outsiders argued.

The private sector needs to deleverage.

The government can help by holding up the economy. It should do so. People who reject free lunches are fools.