Martin Wolf: In the long shadow of the Great Recession

It may be hard to avoid crises but it is vital to make them both small and rare

The US and Europe still live with the legacies of the financial crisis of 2007-09 and the subsequent euro-zone crisis. Could better policies have prevented that outcome; and, if so, what might they have been?

A recovery is under way, but only in a limited sense. The change in gross domestic product of crisis-hit countries is now almost universally positive. But GDP remains far below what might have been expected from pre-crisis trends. In most cases, growth has not recovered, mainly because of declines in productivity growth. In the euro zone, GDP was still below pre-crisis levels in the second quarter of 2015. In crisis-hit members, a return to pre-crisis output is still far away. They will suffer lost decades.

From a sample of 23 high-income countries, Prof Laurence Ball of Johns Hopkins University concludes that losses of potential output ranged from zero in Switzerland to more than 30 per cent in Greece, Hungary and Ireland. In aggregate, he concludes, potential output this year was thought to be 8.4 per cent below what its pre-crisis path would have predicted. This damage from the Great Recession is, he notes, much the same as if Germany's economy had disappeared.

A central finding of the work by Prof Ball and, more recently, by Antonio Fatás of Insead and Lawrence Summers of Harvard, is that estimates of potential output track actual output. This suggests that "hysteresis" – the impact of past experience on subsequent performance – is very powerful. Possible causes of hysteresis include: the effect of prolonged joblessness on employability; slowdowns in investment; declines in the capacity of the financial sector to support innovation; and a pervasive loss of animal spirits.


This year Jason Furman, chairman of the US Council of Economic Advisers, brought out the impact of low post-crisis investment: after the crisis, the contribution of investment labour productivity fell to very low levels. This was strikingly so in US, where the estimated impact actually was negative.

The hysteresis hypothesis is not universally accepted. There are at least three other explanations for the enduring post-crisis collapse in output.

First, it is argued that credit booms raised pre-crisis estimates of potential output far above sustainable levels. One objection to this is that the expansion in credit raised asset prices far more than it fuelled actual spending.

Adair Turner, former chairman of the UK's Financial Services Authority, makes this point in his book Between Debt and the Devil . A further objection to this argument is that it confuses the contribution of debt to the structure of demand with its effect on overall supply.

A second explanation for the post-crisis collapse in output is that the impact of new technologies on output is being underestimated. Yet, even if this were true (which is possible), it would not explain the sharp slowdown in the growth of productivity after the financial crisis.

The difficulty of measuring the impact of new technologies also did not suddenly increase in the UK (the country worst affected by a post-crisis slowdown in productivity growth) relative to the US (the home of these new technologies, yet relatively less affected by the productivity slowdown).

A final explanation is that productivity growth slowed before the crisis. In the US, this does seem to be the case. But it is less obviously true elsewhere.

On balance, then, the hysteresis hypothesis retains substantial force. This is why it is so important that we avoid huge crises and respond strongly to any that occur, to minimise their economic impact. Otherwise, the bad cycle might permanently damage the trend. This raises two further questions: could the adverse impact of the crisis have been smaller? And can it still be reversed? The answer to the first must be yes. But it would have required stronger fiscal and monetary responses, and more aggressive restructuring of damaged financial institutions. The euro zone, in particular, should have done far better. Yet, even today, it lacks the will and the institutions it needs.

The answer to whether the losses in output levels and growth rates can be reversed must again be yes. By the early 1960s, GDP per head in the US had regained the level indicated by a continuation of pre-1929 trends. Unfortunately, the fiscal boost from the second World War was a deus ex machina for policy. It cannot be repeated in peacetime. Even so, it might at least be possible to return to pre-crisis trend rates of growth. A mix of aggressive support for demand and contributions to long-term supply – notably via far higher levels of public investment – would hit both objectives at once.

The evidence, then, is that festering recessions have prolonged effects on prosperity. One conclusion is that it is vital to act swiftly to restore demand. Moreover, the evidence is now clear that the big high-income countries enjoyed the policy space needed to act decisively. Whatever many so foolishly said in 2010, they never faced the slightest risk of turning into Greece. The US and, still more, the euro zone should have responded far more aggressively.

Experience indicates something else no less important. It may be hard to avoid crises but it is vital to make them both small and rare. Financial crises lead to deep recessions and prolonged slowdowns, partly because policymakers fear making sufficiently strong responses. For this reason the regulation of finance simply had to be tightened. The question is only whether it has been tightened in the right way. – Copyright The Financial Times Limited 2015