Eight Centuries of Financial Folly
By Carmen M. Reinhart and Kenneth S. Rogoff
Princeton University Press, 2009, 496 pp., $35
One of the fallouts of the global financial crisis, especially after the collapse of Lehman Brothers in September 2008, has been the questioning of the credibility of much of what passes for economic science, whether delivered by the mathematical adepts of sophisticated financial modeling in the business world or by academia. This kind of economics promised a rational world of ever-increasing prosperity and stability. It was rather dismissive of economic history, which was cast off as an obsolete discipline based on old-fashioned, low-technology techniques. The only value economists saw in history was as a mine for microeconomic data that could be fed into complex mathematical models. By contrast, economic history (which is often rather gloomy) now looks rather more interesting and important. Carmen Reinhart and Kenneth Rogoff have delivered a powerful and eloquent statement explaining why.
Reinhart and Rogoff’s story is about the dismal cycle of recurring economic and financial crises. That cycle depends on historical ignorance. Financial innovators, as well as academics and politicians, want to tell stories about why the old boom-and-bust cycle is dead. They invariably think that “this time”, their own time, is different, because the present is completely different compared to every past experience. The most memorable political exponent of this thesis is the currently rather vulnerable British Prime Minister Gordon Brown, who as Chancellor of the Exchequer repeatedly claimed to have created a new economic framework that would end “the violence of the repeated boom-and-bust cycles of the past.”
The most important lesson of the book is already revealed in the title, “This time is different.” As Reinhart and Rogoff put it, the damaging syndrome of blindness to history holds that
financial crises are things that happen to other people in other countries in other times; crises do not happen to us, here and now. . . . The old rules of valuation no longer apply. The current boom, unlike the many booms that preceded catastrophic collapses in the past (even in our country) is built on sound fundamentals, structural reforms, technological innovation, and good policy. Or so the story goes.
Examples of the myopia that leads to an inability to learn from past patterns of behavior include the boom of the 1920s that preceded the Great Depression, the pumping of debt into Latin America before August 1982, the “Asian miracle” of the 1990s, the dot-com boom, but also, obviously (though it wasn’t obvious in 2005 or 2006), the miracle of rapid and widely dispersed global growth in the 2000s.
In some measure, responsibility for the recent crises goes to an obsession with very short-run time series, taken from runs that often extended for less than five years. On this basis, calculations of variation could be used to judge risk, and to show that, assuming normal statistical behavior, purchasers of assets bore no exposure to the possibility of default. If they felt they did, they might insure against default at extraordinarily cheap rates (since the default predictions for credit-default swaps were also very low). The combination of sophisticated statistical techniques and a very limited range of comparative data led to the assumption of a largely risk-free world in which high rates of return did not correspond to any possibility of loss.
Looking back over time to 12th-century China and medieval and early modern Europe, as Reinhart and Rogoff do, gives a much broader database and leads to a much more skeptical view about the sustainability of debt levels and the gullibility of investors. The beginning of their analysis concerns the often-repeated phenomenon of states being unable to pay their creditors and defaulting on their debt. A second part of the book deals ostensibly with a rather different topic: banking crises. As with sovereign default, history is replete with examples of frauds and miscalculations followed by dramatic failures. But in 2007 that history appeared to be remote, and widespread banking failures in industrial countries quite unthinkable. An historical survey of severe banking crises shows that they cause major economic downturns and that it takes a very long time to recover from such shocks to the credit system. That, indeed, was a lesson that Ben Bernanke presented in a 1983 article in American Economic Review for the experience of the Great Depression.
As it happens, the two topics of the books are linked because the damage done by bank collapses is so great that it can only be dealt with—as in the Great Depression or again after the collapse of Lehman Brothers—by some form of public-sector involvement. Such state engagement may be a direct state takeover of banking, or recapitalization schemes that were voluntary (as in Germany or Italy) or compulsory (as in France and the United States). The result is that major banking crises lead to surges in public-sector indebtedness that are hard to deal with. Only a very small subset of the cases, mostly a Scandinavian subset, managed to confine the cost of bailouts for collapsed banking systems to 2 to 4 percent of GDP. In most cases, the figure is well over 10 percent and can be as high as 50 percent.
The long-term consequences of public sector indebtedness are so unpleasant that we do not like to think of them. Instead, we often comfort ourselves with some vague promise to introduce stability at a future stage, through balanced budget promises or even legal requirements. Reinhart and Rogoff look at cases of what they call debt reversal, when the share of debt falls significantly. In most of these cases, the debt reversal is a consequence of a default. But defaults do long-term damage to expectations about the security of investments and assets, and make for damagingly higher levels of uncertainty.
Two comparisons of societies that were highly divided socially and politically are instructive here. France after 1815, with the Bourbon monarchy restored, did not default on the debt it inherited from the Revolution and Napoleon. There was an immediate cost in terms of higher debt service, but the long-term consequences were growth and stability in the 19th century. Russia defaulted in August 1998 in the aftermath of the Asian debt crisis and then apparently grew at an extravagant pace, but debt repudiation did a great deal to undermine the security of and trust in property relations, since owners of assets will continually fear that governments will be tempted to repeat an apparently successful maneuver.
Many countries in Reinhart and Rogoff’s overview of cases of debt crisis have no alternative to debt repudiation as a way of reducing their debt level, because the borrowing is in a foreign currency or metallic money. Medieval and early modern rulers often defaulted by adjusting downwards the relationship of their currency to its metallic content, but the United States took a similar course as recently as 1933 in repudiating gold clauses in debt contracts. An alternative and less-dramatic course for some countries is to default on the sly, through inflation. Monetary inflation was the great drama not just of the 20th century, with the vivid and well-known examples of post-World War I Germany or post-World War II Hungary, but of every previous age of social and political upheaval—and that includes most of human history. Inflation may not be as dramatic as sovereign default, but its consequences can be even more so.
Reinhart and Rogoff have done an extraordinary job in putting together statistics on government debt—a task that economic historians should have done long ago but shied away from because of the difficulties of defining “government”, which is often complex and multi-layered. On inflation and the gains from inflation, they cannot provide the last word, as they acknowledge: It would require more information on the maturity structure of debt and on interest payments than the dataset provided. But they have posed a challenge to historians. We would be wise to take it up.