The Great Depression in Economic Memory

The pro-stimulus camp often refers to the damage wrought by fiscal retrenchment in the US in 1937, four years after Franklin Roosevelt’s election as US president and the launch of the New Deal. According to computations by the economist Paul van den Noord, the net result of the 1937 budget was a fiscal contraction amounting to three percentage points of GDP – certainly not a trivial amount. Economic growth plummeted from 13% in 1936 to 6% in 1937, and GDP shrank 4.5% in 1938, while unemployment rose from 14% to roughly 20%. Although fiscal policy was not the only cause of the double dip, ill-timed retrenchment certainly contributed to it.

So, are we in 1936, and does the budgetary tightening contemplated in many countries risk provoking a similar double-dip recession?

Clearly there are limits to the comparison. For starters, much less time has elapsed since the financial crisis, the recession has been much shallower, and recovery has come faster. Moreover, important developments that occurred between the 1929 stock-market crisis and the 1937 fiscal retrenchment – especially America’s turn to protectionism in 1930 and the monetary turmoil of subsequent years – have no analog today.

Nevertheless, the 1937 episode does seem to illustrate the dangers of attempting to consolidate public finances at a time when the private sector is still too weak for economic recovery to be self-sustaining. (Another case with similar consequences was Japan’s value-added tax increase in 1997, which precipitated a collapse of consumption).

Fiscal hawks also rely on history-based arguments. The economists Carmen Reinhart and Kenneth Rogoff have studied centuries of sovereign-debt crises, and remind us that today’s developed world has a forgotten history of sovereign default. A particularly telling example is the aftermath of the Napoleonic wars of the early nineteenth century, when a string of exhausted states defaulted on their obligations. The 1930’s are relevant here as well, given another series of defaults among European states, not least Germany.

What history tells us here is that defaults are not the privilege of poor, under-governed countries. They are a threat to all, especially in times of high capital mobility, when governments rely too much on foreign lenders’ apparent willingness to provide funds and find themselves in dire straits when capital inflows stop.

Again, there are limits to comparisons: it is especially hard to infer from past episodes the limits to public debt. After all, British public debt exceeded 250% of GDP in the aftermath of World War II, and Britain did not default. But an important insight from history is that unsustainable fiscal policies are more likely to result in defaults when fiscal problems cannot be inflated away. This was the case under the gold-based monetary regimes such as the Gold Standard of the nineteenth century, and it is the case today for countries that have relinquished their monetary autonomy, such as the members of the eurozone.

In normal times, history is left to historians and economic-policy debate relies on models and econometric estimates. But attitudes changed as soon as the crisis erupted in 2007-2008. Indeed, central bankers and ministers were obsessed at the time by the memory of the 1930’s, and they consciously did the opposite of what their predecessors did 80 years ago.

They were right to do so. In extraordinary times, history is, in fact, a better guide than models estimated with data from ordinary times, because it captures variance that standard time-series techniques ignore. If one wants to know how to deal with a banking crisis, the risk of a depression, or the threat of a default, it is natural to examine times when those dangers were around, rather than to rely on models that ignore such dangers or treat them as distant clouds. In times of crisis, the best guides are theory, which captures the essence of a problem, and the lessons of past experience. Everything in between is virtually useless.

The danger with relying on history, however, is that we have no methodology to decide which comparisons are relevant. Loose analogies can easily be regarded at proofs, and a vast array of experiences can be enrolled to support a particular view. Policymakers (whose knowledge of economic history is generally limited) are therefore at risk of being drowned in contradictory historical references.

History can be an essential compass when past experience provides unambiguous headings. But an undisciplined appeal to history risks becoming a confusing way to express opinions. Governance by analogy can easily lead to muddled governance.

Jean Pisani-Ferry is Director of Bruegel, the European think tank.

Copyright: Project Syndicate, 2010.
www.project-syndicate.org