Lately, the big concern roiling financial markets has been fear of Greek default. The risks seem obvious: Greek government debt is at levels that have historically signaled deep trouble for middle-income nations, and debt is still rising rapidly thanks to a large deficit. Meanwhile, Greece is suffering a severe recession in large part because costs have gotten far out of line with the rest of Europe. And one more thing: Greece has a long history of default—in fact, the nation has been in arrears on its debt for half its modern history.
Yet as recently as last September, nobody seemed worried. Credit default swaps on Greek debt—insurance against a possible default—were fairly cheap; Greece was able to borrow at only modestly higher interest rates than that paragon of fiscal rectitude, Germany. Why were investors so complacent? The answer was that almost everyone believed that historical precedents were irrelevant. Greece was now part of Europe, and even more important, since 2001 part of the eurozone—sharing a currency with its more affluent neighbors. And that changed everything. Except that it didn’t.
The Greek crisis came after the publication of This Time Is Different: Eight Centuries of Financial Folly, by Harvard’s Kenneth Rogoff and the University of Maryland’s Carmen Reinhart, but it was a dramatic illustration of the point they make with their sarcastic title: the more things change in the financial world, the more they stay the same. The Greek debt crisis of 2010 bears a strong resemblance to the Mexican debt crisis of 1827; inflation in Zimbabwe is just the latest episode in a history of currency debasement that goes back to ancient Greek city-states; and last but not least, the US subprime crisis of 2008 followed the script of scores of banking crises past, going back at least as far as eighteenth-century Scotland.
From an economist’s point of view, there are two striking aspects of This Time Is Different. The first is the sheer range of evidence brought to bear. Reading Reinhart and Rogoff is a reminder of how often economists take the easy road—how much they tend to focus their efforts on times and places for which numbers are readily available, which basically means the recent history of the United States and a few other wealthy nations. When it comes to crises, that means acting like the proverbial drunk who searches for his keys under the lamppost, even though that’s not where he dropped them, because the light is better there: the quarter-century or so preceding the current crisis was an era of relative calm, at least among advanced economies, so to understand what’s happening to us one must reach further back and farther afield. This Time Is Different ventures into the back alleys of economic data, accepting imperfect or fragmentary numbers as the price of looking at a wide range of experience.
The second distinguishing feature is the absence of fancy theorizing. It’s not that the authors have anything against elaborate mathematical modeling. Professor Rogoff’s influential 1996 book Foundations of International Macroeconomics, coauthored with Maurice Obstfeld, contains literally hundreds of fairly abstruse equations. But This Time Is Different takes a Sergeant Friday, just-the-facts-ma’am approach: before we start theorizing, let’s take a hard look at what history tells us. One side benefit of this approach is that the current book manages to be both extremely useful to professional economists and accessible to the intelligent lay reader.
The Reinhart-Rogoff approach has already paid off handsomely in making sense of current events. In 2007, at a time when the wise men of both Wall Street and Washington were still proclaiming the problems of subprime “contained,” Reinhart and Rogoff circulated a working paper—now largely subsumed into Chapter 13 of This Time Is Different—that compared the US housing bubble with previous episodes in other countries, and concluded that America’s profile resembled those of countries that had suffered severe financial crises. And sure enough, we had one too. Later, when many business forecasters were arguing that the deep recession would be followed by a rapid, “V-shaped” recovery, they circulated another working paper, largely subsumed into Chapter 14, describing the historical aftermath of financial crises, which suggested that we would face a prolonged period of high unemployment—and so we have.
So what is the message of This Time Is Different? In a nutshell, it is that too much debt is always dangerous. It’s dangerous when a government borrows heavily from foreigners—but it’s equally dangerous when a government borrows heavily from its own citizens. It’s dangerous, too, when the private sector borrows heavily, whether from foreigners or from itself—for banks are basically institutions that borrow from their depositors, then make loans to others, and banking crises are among the most devastating shocks an economy can face.
Yet people—both investors and policymakers—tend to rationalize away these dangers. After any prolonged period of financial calm, they either forget history or invent reasons to believe that historical experience is irrelevant. Encouraged by these rationalizations, people run up ever more debt—and in so doing set the stage for eventual crisis. (One odd omission by Reinhart and Rogoff, by the way, is their failure to mention the late Hyman Minsky, a heterodox economic thinker who made a similar argument and is now experiencing a renaissance in influence.)
Debt-driven crisis can take a variety of forms. There are sovereign debt crises, in which investors lose faith in the ability and/or willingness of governments to fulfill their financial obligations. There are inflationary crises, which happen when governments turn to the printing press either to pay their bills or to inflate away the real value of their debts. There are banking crises, in which people lose that trust in private-sector promises that is essential to a fully functioning market economy. And all of these afflictions are often associated with currency crises, in which speculation leads to a sharp fall in a currency’s value in terms of other currencies.
What we’re in the middle of right now is what Reinhart and Rogoff call the “second great contraction”—a giant banking crisis afflicting both sides of the Atlantic, with effects that have spilled over to the entire world. The first great contraction was, of course, the Great Depression. In the past, banking crises have often led to sovereign debt crises as well, since banking collapses depress the economy, reducing government revenue, at the same time that they often require large outlays to rescue the financial system. Greece may be only the first of many stories of troubled governments; most obviously, Spain, Portugal, and Italy are all in some danger.
It’s worth noting, as an aside, that the Reinhart-Rogoff interpretation of the Great Depression is, implicitly, a critique of other interpretations—most notably, Milton Friedman’s famous claim that the Depression was fundamentally a failure of monetary policy, which could easily have been avoided if only the Fed had prevented a fall in the money supply. Although This Time Is Different doesn’t provide an extensive discussion of events leading up to the Depression, it’s easy to confirm from other sources that the late 1920s looked very much like the prologue to other severe financial crises: irrational exuberance in the stock market, a surge in household debt, and an ever more overextended banking system. There was even a real estate bubble in Florida, memorialized by the Marx Brothers in The Cocoanuts: “You can have any kind of a home you want. You can even get stucco. Oh, how you can get stucco.” That’s not to deny that better policy could have alleviated the pain, a question we’ll return to later. But the Depression looks much more like the product of excessive private-sector debt than like the government failure of monetarist legend.
So now we’ve experienced a severe financial crisis, fundamentally similar to those of the past. What does history tell us to expect next? That’s the subject of Reinhart and Rogoff’s Chapter 14, “The Aftermath of Financial Crises.” This chapter can usefully be read in tandem with two studies by the International Monetary Fund that take a similar approach, published as chapters in the April 2009 and October 2009 editions of the semiannual World Economic Outlook. All three studies offer a grim prognosis: the aftermath of financial crises tends to be nasty, brutish, and long. That is, financial crises are typically followed by deep recessions, and these recessions are followed by slow, disappointing recoveries.
Consider, for example, the case of Sweden, which experienced a severe banking crisis in 1991, following a major housing bubble. Sweden’s government has been widely praised for its response to the crisis: it stabilized markets by guaranteeing bank debt, and restored faith in the system by temporarily nationalizing and then recapitalizing the weakest banks. Despite these measures, however, Swedish unemployment soared from 3 percent to almost 10 percent; it didn’t start coming down until 1995, and progress was slow and fitful for several more years.
It’s true that there have been some “phoenix-like” recoveries from financial crises, to use a term introduced by Columbia University economist Guillermo Calvo. But such recoveries, like South Korea’s bounce-back from the 1997–1998 Asian crisis, have invariably been associated with large depreciations of the afflicted nation’s currencies—the Korean currency, the won, for example, lost more than half its value against the dollar—followed by huge export booms, presumably due to the way a weak currency made that nation’s exports more competitive. Nothing like that can be expected for America now. For one thing, the dollar actually rose in the face of the crisis, as investors sought the safest haven they could find. Beyond that, this is a global crisis, and we can’t all export our way out of it—not unless we can find another planet to trade with.
How long does the pain last? According to the second of those IMF studies, the answer, to a first approximation, is “forever”: financial crises appear to depress not just short-term performance but also long-term growth, so that even a decade after the crisis real GDP is substantially lower than it would otherwise have been.
Reading these studies, we find ourselves wondering what Obama administration economists were thinking when they circulated their now-infamous prediction that the US unemployment rate would peak at 8 percent in the third quarter of 2009. If that had happened, it would have been an exceptional performance, in that both the rise in unemployment and its duration would have been much less than is normal in these cases. In fact, of course, things have turned out considerably worse than the administration’s prediction, and are running fairly close to the historical norm. As Rogoff told one of us in conversation, the United States is experiencing a “garden-variety severe financial crisis.”
History says that the next few years will be difficult. But can anything be done to improve the situation? Unfortunately, This Time Is Different says little on this score.
In part, that may reflect the limits of a history-based, theory-shy approach. In important ways the Reinhart-Rogoff approach resembles that of Wesley Mitchell, who founded the National Bureau of Economic Research in 1920. Under Mitchell’s direction, the NBER focused on quantitative studies of business cycles, tracking just what happens during booms and busts; to this day the organization is responsible for officially dating the beginnings and ends of recessions. Valuable work—but by itself it offered little guidance to policy: it could tell you what usually happens but not how to change the outcome. It wasn’t until John Maynard Keynes offered a theoretical explanation of how it is that economies come to be persistently depressed—an explanation that was informed by historical experience but went far beyond a simple description of past patterns—that economists could offer useful advice to policymakers about how to fight a slump.
That said, history can offer some evidence on the extent to which Keynesian policies work as advertised. As we’ve noted, Reinhart-Rogoff don’t address that question, but others have. Thus the IMF, squinting hard at a relatively limited run of experience (it looks only at advanced countries since 1960), finds evidence that boosting government spending in the face of a financial crisis shortens the slump that follows—but also finds (weak) evidence that such policies might backfire when governments already have a high level of debt, a point we’ll come back to. Interestingly, the IMF also finds that monetary policies, usually the recession-fighting tool of choice, don’t appear effective in the wake of financial crises, perhaps because funds don’t flow easily through a stricken banking system.
There has been even more suggestive work from the economic historians Barry Eichengreen of Berkeley and Kevin O’Rourke of Trinity College in Dublin, who have coauthored two hugely influential papers exploiting the similarities between the current slump and the Great Depression. In the first of these papers, they showed that from a global point of view the first year of this slump was every bit as bad as the Depression: world industrial production fell as steeply, world financial markets were if anything more disrupted, and so on. All this suggests that the shock to the system was just as big this time around.
In successive updates, however, they have shown current events increasingly diverging from the historical record, with the world experiencing a recovery that may be disappointing, but is far better than the continuing downward spiral between 1929 and 1933. The obvious difference is policy: rather than emulating the grim austerity of policymakers three generations ago, who slashed spending in an effort to balance budgets and raised interest rates in an effort to preserve the gold standard, today’s leaders have been willing to run deficits and pump funds into the economy. The result, arguably, has been a much smaller disaster.
An even better test comes from comparing experiences during the 1930s. At the time, nobody was following Keynesian policies in any deliberate way—contrary to legend, the New Deal was deeply cautious about deficit spending until the coming of World War II. There were, however, a number of countries that sharply increased military spending well in advance of the war, in effect delivering Keynesian stimulus as an accidental byproduct. Did these countries exit the Depression sooner than their less aggressive counterparts? Yes, they did. For example, the surge in military spending associated with Italy’s invasion of Abyssinia was followed by rapid growth in the Italian economy and a return to full employment.
Since conditions in the 1930s resembled those now in important ways—as Eichengreen and his coauthors put it, now as then we live “in an environment of near-zero interest rates, dysfunctional banking systems and heightened risk aversion”—this seems to suggest that the right course of action now is to spend freely on stimulus and pay for it later.* But doing so would mean running large budget deficits and adding to debt levels that are already historically high in many countries. How dangerous is doing that?
Much of This Time Is Different is devoted to sovereign debt crises, in which governments lose the confidence of lenders, are unable to service their debt, and respond by defaulting, engaging in inflation, or both. Implicitly, then, the book warns against taking it for granted that nations can get away with deficit spending. On the other hand, advanced nations have historically been able to go remarkably deeply into debt without creating a crisis. Britain’s debt, for example, was larger than its gross domestic product for four decades, from World War I until the 1950s, yet the country’s credit remained good. Japan has run large budget deficits for almost twenty years, yet it can still borrow long-term at very low rates.
So should we be comforted or worried by the historical record? One reason to worry is that advanced countries today may not be as creditworthy as they once were. Reinhart and Rogoff write of the “debt intolerance” of nations suffering from “weak institutional structures and a problematic political system”; might not that description be applied to America today?
In work that postdates This Time Is Different, Reinhart and Rogoff have also argued that there are hidden costs to debt. In a recent working paper they show that even among advanced countries that have not had debt crises, economic growth has historically been lower when government debt exceeds 90 percent of GDP—a threshold the United States might cross in a few years. This result has been widely cited by deficit hawks.
A closer look at the data suggests, however, that in this case correlation may not imply causation. In the case of the United States, for example, the years of high debt were all in the immediate post–World War II period. During that period US real GDP did, in fact, fall—but not because of debt. Instead, GDP was falling thanks to postwar demobilization, as Rosie the Riveter became a suburban housewife. In the case of Japan, the high-debt years all followed the financial crisis of the early 1990s, from which Japan has never fully recovered, so that debt is arguably a consequence of slow growth rather than the other way around.
The truth is that the historical record on the consequences of government debt is sufficiently ambiguous to admit of different interpretations. We read the evidence as supporting a policy of stimulate now, pay later: spend strongly to promote employment in the crisis, but take measures to curb spending and raise revenue once the crisis has passed. Others will see it differently. The main thing to notice, perhaps, is that there is no safe path: debt has long-term risks, but so does failing to engineer a solid recovery. The IMF’s research suggests that the long-term cost of financial crises is less when countries respond with strong stimulus policies, which means that failing to do so risks damage not just this year but for years to come.
Clearly, the best way to deal with debt crises is not to have them. Is there anything in the historical record indicating how we can do that?
Reinhart and Rogoff don’t address this question directly, but Chapter 16 of This Time Is Different, which provides an overview of the ups and downs of crises over the course of the twentieth century, is suggestive. What the data show is a dramatic drop in the frequency of crises of all kinds after World War II, then an irregularly rising trend after about 1980, with a series of regional crises in Latin America, Europe, and Asia, finally culminating in the global crisis of 2008–2009.
What changed after World War II, and what changed it back? The obvious answer is regulation. By the late 1940s, most important economies had tightly regulated banking systems, preventing a recurrence of old-fashioned banking crises. At the same time, widespread limitations on the international movement of capital made it difficult for nations to run up the kinds of large international debts that had previously led to frequent defaults. (These restrictions took various forms, including limits on purchases of foreign securities and limits on the purchase of foreign currency for investment purposes; even advanced nations like France and Italy retained these restrictions into the 1980s.) Basically, it was a constrained world that may have limited initiative, but also left little room for large-scale irresponsibility.
As memories of the 1930s faded, however, these constraints began to be lifted. Private international lending revived in the 1970s, making possible first the Latin American debt crisis of the 1980s, then the Asian crisis of the 1990s. Bank regulation was weakened, enabling the mid-1980s savings and loan debacle in the United States, the Swedish bank crisis of the early 1990s, and so on. By the early twenty-first century, the rapid growth of “shadow banks”—institutions like Lehman Brothers that didn’t accept deposits, and so were not covered by conventional banking regulations, but that in economic terms were carrying out banking functions—had recreated a financial system that was as vulnerable to panic and crisis as the banking system of 1930.
As all this happened, proponents of looser regulation extolled the virtues of a more open system. Indeed, there were real advantages to laxer control: without question, some people, businesses, and governments that would not have had access to credit got it, and some used that credit well. Others, however, ran up dangerous levels of debt. And the old cycle of debt, crisis, and default returned.
Why didn’t more people see this coming? One answer, of course, lies in Reinhart and Rogoff’s title. There were superficial differences between debt now and debt three generations ago: more elaborate financial instruments, seemingly more sophisticated techniques of assessment, an apparent wider spreading of risks (which turned out to have been an illusion). So financial executives, policymakers, and many economists convinced themselves that the old rules didn’t apply.
We should not forget, too, that some people were making a lot of money from the explosive growth both of debt and of the financial industry, and money talks. The world’s two great financial centers, in New York and London, wielded vast influence over their respective governments, regardless of party. The Clinton administration in the US and the Labour government in Britain succumbed alike to the siren song of financial innovation—and were spurred in part by the competition between the two great centers, because politicians were all too easily convinced that having a large financial industry was a wonderful thing. Only when the crisis struck did it become clear that the growth of Wall Street and the City actually exposed their home nations to special risks, and that nations that missed out on the glamour of high finance, like Canada, also missed out on the worst of the crisis.
Now that the multiple bubbles have burst, there’s obviously a strong case for a return to much stricter regulation. It’s by no means clear, however, whether this will actually happen. For one thing, the ideology used to justify the dismantling of regulation has proved remarkably resilient. It’s now an article of faith on the right, impervious to contrary evidence, that the crisis was caused not by private-sector excesses but by liberal politicians who forced banks to make loans to the undeserving poor. Less partisan leaders nonetheless fret over the possibility that regulation might crimp financial innovation, even though it’s very hard to find examples of such innovation that were clearly beneficial (ATMs don’t count).
Equally important, the financial industry’s political power has not gone away. Banks have waged a fierce campaign against what many expected to be an easily passed reform proposal, the creation of a new agency to protect financial consumers. Despite the steady drumbeat of scandalous revelations—most recently, the discovery that Goldman Sachs helped Greece cook its books, while Lehman cooked its own books—top financial executives continue to have ready access to the corridors of power. And as many have noted, President Obama’s chief economic and financial officials are men closely associated with Clinton-era deregulation and financial triumphalism; they may have revised their views but the continuity remains striking.
In that sense, this time really is different: while the first great global financial crisis was followed by major reforms, it’s not clear that anything comparable will happen after the second. And history tells us what will happen if those reforms don’t take place. There will be a resurgence of financial folly, which always flourishes given a chance. And the consequence of that folly will be more and quite possibly worse crises in the years to come.
—April 15, 2010
See Miguel Almunia, A. S. Bénétrix, B. Eichengreen, K.H. O'Rourke, and G. Rua, "The Effectiveness of Fiscal and Monetary Stimulus in Depressions," VoxEll.org, November 18, 2009.↩
See Miguel Almunia, A. S. Bénétrix, B. Eichengreen, K.H. O’Rourke, and G. Rua, “The Effectiveness of Fiscal and Monetary Stimulus in Depressions,” VoxEll.org, November 18, 2009.↩