This Time Is Different: Eight Centuries of Financial Folly
Page 3
From 1800 until well after World War II, Greece found itself virtually in continual default, and Austria’s record is in some ways even more stunning. Although the development of international capital markets was quite limited prior to 1800, we nevertheless catalog the numerous defaults of France, Portugal, Prussia, Spain, and the early Italian city-states. At the edge of Europe, Egypt, Russia, and Turkey have histories of chronic default as well.
One of the fascinating questions raised in our book is why a relatively small number of countries, such as Australia and New Zealand, Canada, Denmark, Thailand, and the United States, have managed to avoid defaults on central government debt to foreign creditors, whereas far more countries have been characterized by serial default on their external debts.
Asian and African financial crises are far less researched than those of Europe and Latin America. Indeed, the widespread belief that modern sovereign default is a phenomenon confined to Latin America and a few poorer European countries is heavily colored by the paucity of research on other regions. As we shall see, precommunist China repeatedly defaulted on international debts, and modern-day India and Indonesia both defaulted in the 1960s, long before the first postwar round of Latin defaults. Postcolonial Africa has a default record that looks as if it is set to outstrip that of any previously emerging market region. Overall, we find that a systematic quantitative examination of the postcolonial default records of Asia and Africa debunks the notion that most countries have avoided the perils of sovereign default.
The near universality of default becomes abundantly clear in part II, where we begin to use the data set to paint the history of default and financial crises in broad strokes using tables and figures. One point that certainly jumps out from the analysis is that the fairly recent (2003–2008) quiet spell in which governments have generally honored their debt obligations is far from the norm.
The history of domestic public debt (i.e., internally issued government debt) in emerging markets, in particular, has largely been ignored by contemporary scholars and policy makers (even by official data providers such as the International Monetary Fund), who seemed to view its emergence at the beginning of the twenty-first century as a stunning new phenomenon. Yet, as we will show in part III, domestic public debt in emerging markets has been extremely significant during many periods and in fact potentially helps resolve a host of puzzles pertaining to episodes of high inflation and default. We view the difficulties one experiences in finding data on government debt as just one facet of the general low level of transparency with which most governments maintain their books. Think of the implicit guarantees given to the massive mortgage lenders that ultimately added trillions to the effective size of the U.S. national debt in 2008, the trillions of dollars in off–balance sheet transactions engaged in by the Federal Reserve, and the implicit guarantees involved in taking bad assets off bank balance sheets, not to mention unfunded pension and medical liabilities. Lack of transparency is endemic in government debt, but the difficulty of finding basic historical data on central government debt is almost comical.
Part III also offers a first attempt to catalog episodes of overt default on and rescheduling of domestic public debt across more than a century. (Because so much of the history of domestic debt has largely been forgotten by scholars, not surprisingly, so too has its history of default.) This phenomenon appears to be somewhat rarer than external default but is far too common to justify the extreme assumption that governments always honor the nominal face value of domestic debt, an assumption that dominates the economics literature. When overt default on domestic debt does occur, it appears to occur in situations of greater duress than those that lead to pure external default—in terms of both an implosion of output and a marked escalation of inflation.
Part IV broadens our discussion to include crises related to banking, currency, and inflation. Until very recently, the study of banking crises has typically focused either on earlier historical experiences in advanced countries, mainly the banking panics before World War II, or on modern-day experiences in emerging markets. This dichotomy has perhaps been shaped by the belief that for advanced economies, destabilizing, systemic, multicountry financial crises are a relic of the past. Of course, the recent global financial crisis emanating out of the United States and Europe has dashed this misconception, albeit at great social cost.
The fact is that banking crises have long plagued rich and poor countries alike. We reach this conclusion after examining banking crises ranging from Denmark’s financial panic during the Napoleonic Wars to the recent first global financial crisis of the twenty-first century. The incidence of banking crises proves to be remarkably similar in the high- and the middle- to low-income countries. Banking crises almost invariably lead to sharp declines in tax revenues as well as significant increases in government spending (a share of which is presumably dissipative). On average, government debt rises by 86 percent during the three years following a banking crisis. These indirect fiscal consequences are thus an order of magnitude larger than the usual costs of bank bailouts.
Episodes of treacherously high inflation are another recurrent theme. No emerging market country in history has managed to escape bouts of high inflation. Indeed, there is a very strong parallel between our proposition that few countries have avoided serial default on external debt and the proposition that few countries have avoided serial bouts of high inflation. Even the United States has had a checkered history, including in 1779, when the inflation rate approached 200 percent. Early on across the world, as already noted, the main device for defaulting on government obligations was that of debasing the content of the coinage. Modern currency presses are just a technologically advanced and more efficient approach to achieving the same end. As a consequence, a clear inflationary bias throughout history emerges. Starting in the twentieth century, inflation spiked radically higher. Since then, inflation crises have stepped up to a higher plateau. Unsurprisingly, then, the more modern period also has seen a higher incidence of exchange rate crashes and larger median changes in currency values. Perhaps more surprising, and made visible only by a broader historical context, are the early episodes of pronounced exchange rate instability, notably during the Napoleonic Wars.
Just as financial crises have common macroeconomic antecedents in asset prices, economic activity, external indicators, and so on, so do common patterns appear in the sequencing (temporal order) in which crises unfold, the final subject of part IV.
The concluding chapter offers some reflections on crises, policy, and pathways for academic study. What is certainly clear is that again and again, countries, banks, individuals, and firms take on excessive debt in good times without enough awareness of the risks that will follow when the inevitable recession hits. Many players in the global financial system often dig a debt hole far larger than they can reasonably expect to escape from, most famously the United States and its financial system in the late 2000s. Government and government-guaranteed debt (which, due to deposit insurance, often implicitly includes bank debt) is certainly the most problematic, for it can accumulate massively and for long periods without being put in check by markets, especially where regulation prevents them from effectively doing so. Although private debt certainly plays a key role in many crises, government debt is far more often the unifying problem across the wide range of financial crises we examine. As we stated earlier, the fact that basic data on domestic debt are so opaque and difficult to obtain is proof that governments will go to great lengths to hide their books when things are going wrong, just as financial institutions have done in the contemporary financial crisis. We see a major role for international policy-making organizations, such as the International Monetary Fund, in providing government debt accounts that are more transparent than those available today.
Figure P.1. Sovereign external debt, 1800–2008: Percentage of countries in external default or restructuring weighted by their share of world income.
Our immersion in the details
of crises that have arisen over the past eight centuries and in data on them has led us to conclude that the most commonly repeated and most expensive investment advice ever given in the boom just before a financial crisis stems from the perception that “this time is different.” That advice, that the old rules of valuation no longer apply, is usually followed up with vigor. Financial professionals and, all too often, government leaders explain that we are doing things better than before, we are smarter, and we have learned from past mistakes. Each time, society convinces itself that the current boom, unlike the many booms that preceded catastrophic collapses in the past, is built on sound fundamentals, structural reforms, technological innovation, and good policy.
Given the sweeping data on which this book has been built, it is simply not possible to provide textural context to all the hundreds of episodes the data encompass. Nevertheless, the tables and figures speak very powerfully for themselves of the phenomenal recurrent nature of the problem. Take figure P.1, which shows the percentage of countries worldwide, weighted by GDP, that have been in a state of default on their external debt at any time.
The short period of the 2000s, represented by the right-hand tail of the chart, looks sufficiently benign. But was it right for so many policy makers to declare by 2005 that the problem of sovereign default on external debt had gone into deep remission? Unfortunately, even before the ink is dry on this book, the answer will be clear enough. We hope that the weight of evidence in this book will give future policy makers and investors a bit more pause before next they declare, “This time is different.” It almost never is.
ACKNOWLEDGMENTS
A book so long in the making generates many debts of gratitude. Among those who helped is Vincent Reinhart, who consulted on the economic and statistical content and edited and re-edited all the chapters. He also provided the anecdote that led to the book’s title. Vincent worked for the Federal Reserve for almost a quarter century. Back around the time of the collapse of the hedge fund Long-Term Capital Management in 1998, which seemed like a major crisis then but seems less so given recent events, he attended a meeting of the board of governors with market practitioners. A trader with an uncharacteristically long memory explained, “More money has been lost because of four words than at the point of a gun. Those words are ‘This time is different.’”
A special debt of gratitude is owed to Jane Trahan for her extremely helpful and thorough editing of the manuscript, and to our editor at Princeton University Press, Seth Ditchik, for his suggestions and editorial guidance throughout this process. Ethan Ilzetzki, Fernando Im, Vania Stavrakeva, Katherine Waldock, Chenzi Xu, and Jan Zilinsky provided excellent research assistance. We are also grateful to Peter Strupp and his colleagues at Princeton Editorial Associates for skillfully negotiating all the technical details of producing this volume.
PREAMBLE: SOME INITIAL INTUITIONS
ON FINANCIAL FRAGILITY AND THE
FICKLE NATURE OF CONFIDENCE
This book summarizes the long history of financial crises in their many guises across many countries. Before heading into the deep waters of experience, this chapter will attempt to sketch an economic framework to help the reader understand why financial crises tend to be both unpredictable and damaging. As the book unfolds, we will take other opportunities to guide interested readers through the related academic literature when it is absolutely critical to our story. Rest assured that these are only short detours, and those unconcerned with economic theory as an engine of discovery can bypass these byways.
As we shall argue, economic theory proposes plausible reasons that financial markets, particularly ones reliant on leverage (which means that they have thin capital compared to the amount of assets at stake), can be quite fragile and subject to crises of confidence.1 Unfortunately, theory gives little guidance on the exact timing or duration of these crises, which is why we focus so on experience.
Perhaps more than anything else, failure to recognize the precariousness and fickleness of confidence—especially in cases in which large short-term debts need to be rolled over continuously—is the key factor that gives rise to the this-time-is-different syndrome. Highly indebted governments, banks, or corporations can seem to be merrily rolling along for an extended period, when bang!—confidence collapses, lenders disappear, and a crisis hits.
The simplest and most familiar example is bank runs (which we take up in more detail in the chapter on banking crises). We talk about banks for two reasons. First, that is the route along which the academic literature developed. Second, much of our historical data set applies to the borrowing of banks and of governments. (Other large and liquid participants in credit markets are relatively new entrants to the world of finance.) However, our examples are quite illustrative of a broader phenomenon of financial fragility. Many of the same general principles apply to these market actors, whether they be government-sponsored enterprises, investment banks, or money market mutual funds.
Banks traditionally borrow at short term. That is, they borrow in the form of deposits that can be redeemed on relatively short notice. But the loans they make mostly have a far longer maturity and can be difficult to convert into cash on short notice. For example, a bank financing the expansion of a local hardware store might be reasonably confident of repayment in the long run as the store expands its business and revenues. But early in the expansion, the bank may have no easy way to call in the loan. The store owner simply has insufficient revenues, particularly if forced to make payments on principal as well as interest.
A bank with a healthy deposit base and a large portfolio of illiquid loans may well have bright prospects over the long term. However, if for some reason, depositors all try to withdraw their funds at once—say, because of panic based on a false rumor that the bank has lost money gambling on exotic mortgages—trouble ensues. Absent a way to sell its illiquid loan portfolio, the bank might simply not be able to pay off its panicked depositors. Such was the fate of the banks in the classic movies It’s a Wonderful Life and Mary Poppins. Those movies were rooted in reality: many banks have shared this fate, particularly when the government has not fully guaranteed bank deposits.
The most famous recent example of a bank run is the run on the United Kingdom’s Northern Rock bank. Panicked depositors, not satisfied with the British government’s partial insurance scheme, formed long queues in September 2007. The broadening panic eventually forced the British government to take over the bank and more fully back its liabilities.
Other borrowers, not just banks, can suffer from a crisis of confidence. During the financial crisis that started in the United States in 2007, huge financial giants in the “shadow banking” system outside regulated banks suffered similar problems. Although they borrowed mainly from banks and other financial institutions, their vulnerability was the same. As confidence in the investments they had made fell, lenders increasingly refused to roll over their short-term loans, and they were forced to throw assets on the market at fire-sale prices. Distressed sales drove prices down further, leading to further losses and downward-spiraling confidence. Eventually, the U.S. government had to step in to try to prop up the market; the drama is still unfolding, and the price tag for resolution continues to mount.
Governments can be subject to the same dynamics of fickle expectations that can destabilize banks. This is particularly so when a government borrows from external lenders over whom it has relatively little influence. Most government investments directly or indirectly involve the long-run growth potential of the country and its tax base, but these are highly illiquid assets. Suppose, for example, that a country has a public debt burden that seems manageable given its current tax revenues, growth projections, and market interest rates. If the market becomes concerned that a populist fringe candidate is going to win the next election and raise spending so much that the debt will become difficult to manage, investors may suddenly balk at rolling over short-term debt at rates the country can manage. A credit crisis unfolds.
What does economic theory have to say about countries’ vulnerability to financial crises? For concreteness, let us focus for now on governments, the main source of the crises examined in this book. Economic theory tells us that if a government is sufficiently frugal, it is not terribly vulnerable to crises of confidence. A government does not have to worry too much about debt crises if it consistently runs fiscal surpluses (which happens when tax receipts exceed expenditures), maintains relatively low debt levels, mostly borrows at longer-term maturities (say ten years or more), and does not have too many hidden off–balance sheet guarantees.