Published by EH.NET (April 2010)
Carmen M. Reinhart and Kenneth S. Rogoff, This Time is Different: Eight Centuries of Financial Folly. Princeton, NJ: Princeton University Press, 2009. xlv + 463 pp. $35 (hardcover), ISBN: 978-0-691-14216-6.
Reviewed for EH.NET by Richard Sylla, Department of Economics, Stern School of Business, New York University.
For financial economists and historians this is a welcome, timely, and seminal book. It is welcome because it fills a gap in the literature of financial crises. Earlier work is of two kinds, episodic and analytic in a non-quantitative way. The episodic approach ? literary and sometimes flowery accounts of major historical financial crises ? began with McKay (1852) and continued down to Chancellor (1999), and also includes many studies of individual crises. The analytic, non-quantitative approach, describing and analyzing the typical pattern of a financial crisis and illustrating it with historical examples, is best exemplified by Kindleberger (1978, with several later updated editions). Missing until Reinhart and Rogoff filled the gap was a thorough study of the quantitative aspects of various categories of financial crises ? inflation crises, currency crashes and debasements, external sovereign debt defaults, domestic debt defaults, and banking crises. Their quantitative analysis is based on a huge database covering sixty-six countries that recently accounted for some ninety percent of world economic product. Most of the database is drawn from the past century or two, but some of the inflation, debasement, and external sovereign debt default evidence goes back six to eight hundred years.
The book is timely because it provides an extended account ? about a quarter of the text ? of the financial crisis of 2007-2009, showing that it has many of the characteristics of previous crises. This account can be read separately from the rest of the book, for in it the authors repeat the main quantitative findings from their treatment of all the previous crises they identify. Reinhart and Rogoff demonstrate that, despite the Great Moderation, the sophistication of modern finance, and the skill of today?s central bankers and other economic policy makers, this time was not different. That lesson, obliquely conveyed in the book?s title, and the lucky timing of its appearance have given it a widespread popularity in the media ? a popularity rather unusual for a volume full of tables, charts, and a hundred pages of appendices describing the data. Some may find it too scholarly and a somewhat dry study, and I would surmise that once many of its purchasers put it down, they won?t be able to pick it up again. I had the opposite reaction. I couldn?t put it down until I had gone all the way through it, and then I immediately ordered it as an assigned text for my Spring 2010 MBA course, ?The Development of Financial Institutions and Markets.? My students are finding it useful and engaging.
The book should prove seminal because the valuable database on which it is based will attract the interest of quantitative analysts as ants are drawn to honey, and because, as Reinhart and Rogoff more or less admit, they are making a first pass at the data, raising as many unanswered questions as answers. Their findings are mostly central tendencies, i.e., means and medians. Not much information is presented on the dispersions of the data around these means, an obvious avenue to explore in future research. Nor is there much detail on any particular crisis ? this, of course, is the bread and butter of the episodic approach to crisis studies. Moreover, there is little theory in the book, so theorists will be attracted to its findings in order to explain why the quantitative facts are what they are, and to explain the cause-effect relationships that facts have with one another.
What are some of the key findings? A selection might include:
1. From 1340, when England defaulted and ruined prominent Italian bankers, to 1799, there were nineteen instances of external debt default. From 1800 to 1945, there were 127 episodes of external debt default with a median of six years from default to resolution by means of restructuring, repayment, or debt forgiveness. From 1946 to 2008, there were 169 such episodes with a median of three years from default to resolution. Defaults often follow spikes in capital inflows.
2. Although domestic (as contrasted with external) debt defaults are little studied, domestic debt averages two-thirds of total public debt. There are at least 70 instances of domestic default since 1800. Domestic defaults involve forced conversions to lower interest rates, reductions of principal, suspension of debt-service payments, and, of course, inflations that erode the real value of domestic debts. Output declines an average of eight percent in the three years before default, and inflation averages 170 percent in the year of default and remains above 100 percent in subsequent years. Governments are far from transparent in accounting for their domestic debts, hiding many of them in off-budget guarantees ? think, for example, of the unfunded Social Security and Medicare liabilities of the U.S. government.
3. Although a few countries have avoided debt defaults altogether, and others appear to have graduated from repeated defaults long ago to the avoidance of default for decades or centuries, no country has been able to escape from having banking crises. World financial centers have had numerous banking crises since 1800: the UK 12, the U.S. 13, and France 15. Banking crises lead to decreases in tax revenues and increases in public spending, so on average real government debt increases an average of 86 percent during the three years following a banking crisis. Costs of bailing out financial institutions are but a part of this increase. Banking crises are more usual in periods of high international capital mobility. Real house prices and equity prices tend to boom just before a crisis, and then fall for several years starting with the crisis year. On average real house prices decline 35 percent over six years, and equity prices decline 56 percent over three and one half years.
4. Moneys around the world have been debased rather continuously for six or seven centuries. Episodes featuring a stable value of money for any lengthy period of time are few.
5. By 2006-2007, the U.S. and a few other countries exhibited several leading indicators of imminent financial crises: massive borrowing from other countries, asset price inflation (especially in real estate), rising household leverage, soaring profits of highly levered financial firms, and a slowing of output growth. A number of academic observers warned of a crisis about to happen; others put forward ?not to worry? analyses and soothing prognostications, as did leading policy makers. So opinions differed. But it could hardly be said that since the crisis was inherently unpredictable, the best course of action for policy makers was to wait until it happened and then act to contain the damage.
Reinhart and Rogoff, however, say little about what policy makers should do when the tell-tale signs of an imminent crisis are flashing. It?s a difficult problem. Were policy makers to act in a way that prevented a crisis from happening, they might well be accused of overreacting to a problem that did not exist, and of needlessly reducing economic growth ? not to mention stanching the rise of house and equity prices, and putting the kibosh on the bonuses of bankers ? for no obvious and good reason. What policy maker or politician would want to run that risk? Must crises continue to happen because it is in no one?s ex ante interest to prevent them from happening?
Taking away the punch bowl just when the party is getting good is a thankless task. It may, however, be a needed task if we are to mitigate or avoid the crisis carnage so amply documented in This Time Is Different. By documenting quantitatively the warning signs and the negative consequences of financial crises, Reinhart and Rogoff have made a substantial contribution. Their work and the further work it will undoubtedly inspire could give some momentum to reforms that would reduce both the incidence of crises and the damage that results from them. Were that to happen, maybe the next time will be different.
Edward Chancellor (1999), Devil Take the Hindmost: A History of Financial Speculation. New York: Farrar Straus Giroux.
Charles P. Kindleberger (1978), Manias, Panics, and Crashes: A History of Financial Crises. New York: Basic Books.
Charles McKay (1852), Memoirs of Extraordinary Popular Delusions and the Madness of Crowds. Reprint ed., Boston: L. C. Page, 1932.
Richard Sylla is Henry Kaufman Professor of the History of Financial Institutions and Markets, and Professor of Economics, Stern School of Business, New York University, and Research Associate, NBER. With Douglas Irwin, he is co-editor and a contributor to Founding Choices: American Economic Policy in the 1790s, forthcoming in 2010 from the University of Chicago Press.