This Time Is Different: Eight Centuries of Financial Folly Read online

  Public Debt and Its Composition

  As we have already emphasized, finding data on domestic public debt is remarkably difficult. Finding data on defaults on domestic debt is, not surprisingly, even more problematic. In this volume we catalog more than seventy instances of outright default on domestic debt dating back to the early 1800s. Yet even this tally is probably a considerable understatement.16

  For the advanced economies, the most comprehensive data come from the Organisation for Economic Co-operation and Development (OECD), which provides time series on general government debt since 1980. However, these data have several important limitations. They include only a handful of emerging markets. For many advanced economies (Greece, Finland, France, and the United Kingdom, to name a few), the data actually begin much later, in the 1990s, so the OECD data on public debt provide only a relatively short time series. Moreover, only total debt is reported, with no particulars provided regarding the composition of debt (domestic versus foreign) or its maturity (long-term versus short-term). Similarly, to consider the IMF’s well-known World Economic Outlook database as extending to public debt requires a stretch of the imagination.17 Data are provided only for the G-7 and only from 1980 onward (out of 180 countries covered in the WEO).

  The most comprehensive data on public debt come from the World Bank’s Global Development Finance (GDF, known previously as the World Debt Tables). It is an improvement on the other databases in that it begins (for most countries) in 1970 and provides extensive detail on the particulars of external debt. Yet GDF also has serious limitations. No advanced economies are included in the database (nor are newly industrialized countries, such as Israel, Korea, or Singapore) to facilitate comparisons. Unlike data from the IMF and the World Bank for exchange rates, prices, government finances, and so on, the database includes no data prior to 1970. Last but certainly not least, these data cover only external debt. In a few countries, such as Côte d’Ivoire or Panama, external debt is a sufficient statistic on government liabilities because the levels of domestic public debt are relatively trivial. As we shall show in chapter 7, however, domestic debt accounts for an important share of total government debt for most countries. The all-country average share oscillated between 40 and 80 percent during 1900–2007.18

  In search of the elusive data on total public debt, we examined the archives of the global institutions’ predecessor, the League of Nations, and found that its Statistical Yearbook: 1926–1944 collected information on, among other things, public domestic and external debt. Although neither the IMF nor the World Bank continued this practice after the war, the newly formed United Nations (UN) inherited the data collected by the League of Nations, and in 1948 its Department of Economic Affairs published a special volume on public debt spanning 1914–1946. From that time onward, the UN continued to collect and publish the domestic and external debt data on an annual basis in the same format used by its prewar predecessor in its Statistical Yearbook. As former colonies became independent nations, the database expanded accordingly. This practice continued until 1983, at which time the domestic and external public debt series were discontinued altogether. In total, these sources yield time series that span 1914–1983 for the most complete cases. They cover advanced and developing economies. For the most part, they also disaggregated domestic debt into long-term and short-term components. To the best of our knowledge, these data are not available electronically in any database; hence, obtaining it required going to the original publications. These data provide the starting point for our public debt series, which have been (where possible) extended to the period prior to 1914 and since 1983.

  For data from the period prior to 1914 (including several countries that were then colonies), we consulted numerous sources, both country-specific statistical and government agencies and individual scholars.19 Appendix A.2 provides details of the sources by country and time period. In cases for which no public debt data are available for the period prior to 1914, we approximated the foreign debt stock by reconstructing debt from individual international debt issues. These debenture (debt not secured by physical collateral or assets) data also provide a proximate measure of gross international capital inflows. Many of the data come from scholars including Miller, Wynne, Lindert and Morton, and Marichal, among others.20 From these data we construct a foreign debt series (but it does not include total debt).21 This exercise allows us to examine standard debt ratios for default episodes of several newly independent nations in Latin America as well as Greece and important defaults such as that of China in 1921 and those of Egypt and Turkey in the 1860s and 1870s. These data are most useful for filling holes in the early debt time series when countries first tap international capital markets. Their usefulness (as measures of debt) is acutely affected by repeated defaults, write-offs, and debt restructurings that introduce disconnects between the amounts of debt issued and the subsequent debt stock.22

  For some countries (or colonies in the earlier period) for which we have only relatively recent data for total public debt but have reliable data going much further back on central government revenues and expenditures, we calculate and cumulate fiscal deficits to provide a rough approximation of the debt stock.23

  To update the data for the time since 1983, we rely mostly on GDF for external debt, with a few valuable recent studies facilitating the update.24 Last but certainly not least are the official government sources themselves, which are increasingly forthcoming in providing domestic debt data, often under the IMF’s 1996 Special Data Dissemination Standard, prominently posted at the IMF’s official Web site.25

  Global Variables

  We label two types of variables “global.” The first are those that are genuinely global in scope, such as world commodity prices. The second type consists of key economic and financial indicators for the world’s financial centers during 1800–2009 that have exerted a true global influence (in modern times, the U.S. Federal Reserve’s target policy interest rate is such an example). For commodity prices, we have time series since the late 1700s from four different core sources (see appendix A.1). The key economic indicators include the current account deficit, real and nominal GDP, and short- and long-term interest rates for the relevant financial center of the time (the United Kingdom prior to World War I and the United States since then).

  Country Coverage

  Table 3.1 lists the sixty-six countries in our sample. We include a large number or African and Asian economies, whereas previous studies of the same era typically included at most a couple of each. Overall, our data set includes thirteen African countries, twelve Asian countries, nineteen European countries, and eighteen Latin American countries, plus North America and Oceania. (Our sample excludes many of the world’s poorest countries, which by and large cannot borrow meaningful amounts from private sector lenders and virtually all of which have effectively defaulted even on heavily subsidized government-to-government loans. This is an interesting subject for another study, but here we are mainly interested in financial flows that, at least in the first instance, had a substantial market element.)26

  As the final column of table 3.1 illustrates, our sample of sixty-six countries indeed accounts for about 90 percent of world GDP. Of course, many of these countries, particularly those in Africa and Asia, have become independent nations only relatively recently (see column 2). These recently independent countries have not been exposed to the risk of default for nearly as long as, say, the Latin American countries, and we have to calibrate our intercountry comparisons accordingly.

  Table 3.1 flags which countries in our sample may be considered “default virgins,” at least in the narrow sense that they have never outright failed to meet their external debt repayment obligations or rescheduled on even one occasion. One conspicuous grouping of countries includes the high-income Anglophone nations, Australia, Canada, New Zealand, and the United States. (The mother country, England, defaulted in earlier eras, as we have already noted.) In addition, none of the Scandinavian coun
tries, Denmark, Finland, Norway, and Sweden, has defaulted, nor has Belgium or the Netherlands. And in Asia, Hong Kong, Korea, Malaysia, Singapore, Taiwan, and Thailand have all avoided external default. Admittedly, two of these countries, Korea and Thailand, managed to avoid default only through massive IMF loan packages during the last debt crisis of the 1990s and otherwise suffered much of the same trauma as a typical defaulting country. Of the default-free Asian countries, only Thailand existed as an independent state before the end of World War II; others have had the potential for default for only a relatively short time. Default or restructuring of domestic public debt would significantly reduce the “default virgin” list, among other things eliminating the United States from the roster of nondefaulters. For example, the abrogation of the gold clause in the United States in 1933, which meant that public debts would be repaid in fiat currency rather than gold, constitutes a restructuring of nearly all the government’s domestic debt. Finally, one country from Africa, Mauritius, has never defaulted or restructured.

  It is notable that the nondefaulters, by and large, are all hugely successful growth stories. This begs the question “Do high growth rates help avert default, or does averting default beget high growth rates?” Certainly we see many examples in world history in which very rapidly growing countries ran into trouble when their growth slowed.

  Of course, governments can achieve de facto partial default on nominal bond debt simply through unanticipated bursts of inflation, as we discuss later, in chapters 11 and 12. Governments have many ways to partially default on debts, and many types of financial crises over the years have taken their character from the government’s choice of financing and default vehicle. The fact that government debt can be a common denominator across disparate types of crises will become even more clear when we take up the links between crises in chapter 16.

  TABLE 3.1

  Countries’ share of world GDP, 1913 and 1990

  - PART II -

  SOVEREIGN EXTERNAL DEBT CRISES

  Most countries in all regions have gone through a prolonged phase as serial defaulters on debt owed to foreigners.

  - 4 -

  A DIGRESSION ON THE THEORETICAL

  UNDERPINNINGS OF DEBT CRISES

  In this book we chronicle hundreds of episodes in which sovereign nations have defaulted on their loans from external creditors. These “debt crises” range from defaults on mid-fourteenth-century loans made by Florentine financiers to England’s Edward III to those on massive loans from (mostly) New York bankers to Latin America during the 1970s. Why do countries seem to run out of money so often? Or do they?

  Former Citibank chairman (1967–1984) Walter Wriston famously said, “Countries don’t go bust.” In hindsight, Wriston’s comment sounded foolish, coming just before the great wave of sovereign defaults in the 1980s. After all, he was the head of a large bank that had deeply invested across Latin America. Yet, in a sense, the Citibank chairman was right. Countries do not go broke in the same sense that a firm or company might. First, countries do not usually go out of business. Second, country default is often the result of a complex cost-benefit calculus involving political and social considerations, not just economic and financial ones. Most country defaults happen long before a nation literally runs out of resources.

  In most instances, with enough pain and suffering, a determined debtor country can usually repay foreign creditors. The question most leaders face is where to draw the line. The decision is not always a completely rational one. Romanian dictator Nikolai Ceauşescu single-mindedly insisted on repaying, in the span of a few years, the debt of $9 billion owed by his poor nation to foreign banks during the 1980s debt crisis. Romanians were forced to live through cold winters with little or no heat, and factories were forced to cut back because of limited electricity.

  Few other modern leaders would have agreed with Ceauşescu’s priorities. The Romanian dictator’s actions are especially puzzling given that the country could presumably have renegotiated its debt burden, as most other developing countries eventually succeeded in doing during the crisis of the 1980s. By the same token, modern convention holds that a debtor country should not have to part with rare national treasures to pay off debts. During Russia’s financial crisis in 1998, no one contemplated for a moment the possibility that Moscow might part with art from the Hermitage museum simply to appease Western creditors.1

  The fact that lenders depend on a sovereign nation’s willingness to repay, not simply its ability to repay, implies that sovereign bankruptcy is a distinctly different animal than corporate bankruptcy. In corporate or individual bankruptcy, creditors have well-defined rights that typically allow them to take over many of the debtor’s assets and put a lien on a share of its future income. In sovereign bankruptcy, creditors may be able to do the same on paper, but in practice their enforcement power is very limited.

  This chapter provides an analytical framework that allows us to think more deeply about the underpinnings of international debt markets. Our goal here is to provide not a comprehesive survey of this extensive literature but a broad overview of issues.2 Readers mainly interested in understanding the historical experience might choose to skip this chapter. In some respects, however, the analysis of this chapter lies at the heart of everything that follows. Why on earth do foreign creditors ever trust countries to repay their debt anyway, especially when they have been burned so regularly in the past? Why would domestic residents in emerging markets ever entrust their money to banks or local currency when they, too, have been burned so often? Why do explosions of global inflation occur sometimes, such as in the early 1990s, when forty-five countries had inflation rates over 20 percent, and not during other periods, such as the early 2000s, when only a couple had such high inflation rates?

  These are not simple questions, and they are the subject of huge debate among economists. We do not come close to providing complete answers; the social, political, and economic problems underpinning default are simply too complex. If future generations of researchers do resolve these issues, perhaps the topic of this book will become moot and the world will finally reach an era in which we can say, “This time really is different.” However, history is littered with instances in which people declared premature victory over such thorny issues.

  We first concentrate on what is perhaps the most fundamental “imperfection” of international capital markets, the lack of a supernational legal framework for enforcing debt contracts across borders. This is an abstract way of saying that if the government of Argentina (a country sporting a famous history of serial default) borrows money from a U.S. bank and then defaults, the bank’s options for direct enforcement of its claims are limited. To sharpen our discussion of the international aspects of the problem, we will temporarily ignore political and economic divisions within the borrowing country and simply treat it as a unified actor. Thus we will ignore domestic public debt (debt borrowed by the government from its own citizens or from local banks).

  It may seem strange to those unfamiliar with economic modeling to group a government and its population together as a unified actor. In altogether too many countries, governments can be kleptocratic and corrupt, with national policies dictated by the political elite rather than by the average citizen. Indeed, political disunity is often a key driver of sovereign defaults and financial crises. The fact that the U.S. subprime crisis became much worse in the run-up to the country’s 2008 election is quite typical. Preelection posturing and postelection uncertainty routinely exacerbate the challenge of developing a coherent and credible policy response. Brazil’s massive 2002 financial crisis was sparked in no small part by investors’ concerns regarding a shift from the centrist government of then-president Fernando Henrique Cardoso to the more populist policies of the opposition leader Luiz Inácio Lula da Silva. The irony, of course, is that the left-leaning winner ultimately proved more conservative in his macroeconomic governance than investors had feared or, perhaps, some of his supporters had hoped.


  Sovereign Lending

  If the reader has any doubt that willingness to pay rather than ability to pay is typically the main determinant of country default, he or she need only peruse our earlier table 2.2. The table shows that more than half of defaults by middle-income countries occur at levels of external debt relative to GDP below 60 percent, when, under normal circumstances, real interest payments of only a few percent of income would be required to maintain a constant level of debt relative to GDP, an ability that is usually viewed as an important indicator of sustainability. Expressed as a percentage of exports or government revenues, of course, payments would typically be several times higher, as we will illustrate later. But even so, a workout would be manageable over time in most cases except during wartime, especially if the country as a whole were clearly and credibly committed to gradually increasing exports over time to a level commensurate with eventual full repayment.

  The centrality of willingness to pay rather than ability to pay is also clear when one looks back several hundred years to international lending during the sixteenth, seventeenth, and eighteen centuries (what we term the early period of default). Back then, the major borrowers were countries such as France and Spain, which commanded great armies of their own. Foreign investors could hardly have expected to collect through force. As Michael Tomz reminds us, during the colonial era of the nineteenth century, superpowers did periodically intervene to enforce debt contracts.3 Britain routinely bullied and even occupied countries that failed to repay foreign debts (for example, it invaded Egypt in 1882 and Istanbul in the wake of Turkey’s 1876 default). Similarly, the United States’ “gunboat diplomacy” in Venezuela, which began in the mid-1890s, was motivated in part by debt repayment concerns. And the U.S. occupation of Haiti beginning in 1915 was rationalized as necessary to collect debt. (Box 5.2 explains how debt problems led the independent nation of Newfoundland to lose its sovereignty.)