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

  We do not want to overemphasize the rationality of lenders. In fact, there are many cases in which the very small risk premiums charged sovereign nations are hardly commensurate with the risks involved. High-risk borrowers, of course, not only have to face interest rate risk premiums on their borrowing but often bear significant deadweight costs if debt problems amplify recessions in the event of default. For borrowers the this-time-is-different mentality may be even more costly than for creditors, but again we will need to revisit this issue in a broader calculus of default.

  Odious Debt

  Another deep philosophical issue, in principle relevant to thinking about international lending, surrounds the notion of “odious debt.” In the Middle Ages, a child could be sent to debtors’ prison if his parents died in debt. In principle, this allowed the parent to borrow more (because the punishment for failure to repay was so great), but today the social norms in most countries would view this transfer of debt as thoroughly unacceptable. But of course nations do borrow inter-temporally, and the children of one generation may well have to pay off the debts of their parents. At the end of World War II, the gross domestic debt of the United States reached more than 100 percent of GDP, and it took several decades to bring it down to a more normal 50 percent of GDP.

  The doctrine of odious debt basically states that when lenders give money to a government that is conspicuously kleptomaniacal and corrupt, subsequent governments should not be forced to honor it. Jayachandran and Kremer argue that one can modify standard reputation models of debt to admit a convention of not honoring odious debt, and that this can be welfare improving.17 However, there is quite a bit of controversy about whether odious debt can be clearly delineated in practice. Everyone might agree that if the leaders of a country engaged in genocide were to borrow to finance their military, the lenders should recognize the debt as odious and at risk of default in the event of a regime change. However, one can imagine global bureaucrats arguing over, say, whether debt issued by the United States is odious debt, in which case, of course, the concept would not provide sufficient discrimination to be useful in practice. The practical guidelines regarding odious debt must be sufficiently narrowly construed so as to be implementable. In practice, though, weaker versions of odious debt do, perhaps, have some relevance. The circumstances under which a debt burden is accumulated can affect a debtor’s view of “fairness,” and therefore its willingness to pay. On occasion, the international community may also be willing to treat debtors more gently in these circumstances (at the very least by giving them greater access to subsidized bridge loans).

  Domestic Public Debt

  If the theory of external sovereign debt is complex, the theory of domestic public debt is even more so. For the purposes of this discussion, we will assume that domestic public debt is denominated in domestic currency, adjudicated within the issuing country, and held by domestic residents. Of these three strictures, the only one that is really absolute in our definition in chapter 1 is the assumption that the debt is adjudicated by domestic authorities. Beginning, perhaps, with Argentina’s U.K. pound–denominated “internal” bonds of the late nineteenth century, there have been a number of historical examples in which domestic debt has been indexed to foreign currency (mostly famously the tesobono debt issued by Mexico in the early 1990s and the precedents noted in box 7.1), and in recent years that phenomenon has become more prevalent. As more emerging markets have moved to liberalize their capital markets, it has become increasingly common for foreign residents to hold domestic public debt. The nuance that both foreign and domestic residents may hold a certain type of debt can be relevant, but we will set this nuance aside to simplify our discussion.18

  Domestic debt is debt a country owes to itself. In Robert Barro’s famous Ricardian model of debt, domestic public debt does not matter at all, for citizens simply increase their savings when debt goes up to offset future taxes.19 Barro’s analysis, however, presumes that debt will always be honored, even if savings patterns are not homogeneous and debt repayments (as opposed to repudiations) favor some groups at the expense of others. This presumption begs the question as to why political outcomes do not periodically lead countries to default on domestic debt, and assumes away the question as to why anyone lends to governments in the first place. If old people hold most of a country’s debt, for example, why don’t young voters periodically rise up and vote to renege on the debt, starting anew with a lower tax for the young at the cost of less wealth for the elderly?

  One of the more startling findings in part III of this book, on domestic debt, is that such outright defaults occur far more often than one might imagine, albeit not quite as often as defaults on sovereign external debt. Governments can also default on domestic public debt through high and unanticipated inflation, as the United States and many European countries famously did in the 1970s.

  What, then, anchors domestic public debt? Why are domestic bondholders paid anything at all? North and Weingast argue that a government’s ability to establish political institutions that sustain large amounts of debt repayment constitutes an enormous strategic advantage by allowing a country to marshal vast resources, especially in wartime.20 They argue that one of the most important outcomes of England’s “glorious revolution” of the late 1600s was precisely a framework to promote the honoring of debt contracts, thereby conferring on England a distinct advantage over rival France. France, as we shall see, was at the height of its serial default era during this period. The Crown’s ability to issue debt gave England the huge advantage of being able to marshal the resources needed to conduct warfare in an era in which combat was already becoming extremely capital intensive.

  In democracies, Kotlikoff, Persson, and Svensson suggest that domestic debt markets might be a convention that can be sustained through reputation, much as in the Eaton and Gersovitz model of sovereign external debt.21 Tabellini, in a related article, suggests that debt might be sustainable if young voters care sufficiently about older voters.22 All of these theories, and others for the case in which the government is a monarchy rather than a democracy, are built around the assumption that debt markets are self-sustaining conventions in which the costs and benefits narrowly match up to ensure continuous functioning. Yet, as we have discussed, the incentives for repayment of any kind of government debt probably involve broader issues than just the necessity of smoothing out tax receipts and consumption. Just as failure to honor sovereign debt might conceivably trigger broader responses in international relations outside the debt arena, so might domestic default trigger a breakdown in the social compact that extends beyond being able to borrow in the future. For one thing, in many economies government debt is not simply a means for governments to smooth tax receipts but a store of value that helps maintain the liquidity of credit markets. Governments may periodically default on their debts, but in most countries the record of private firms is even worse.

  Financial repression can also be used as a tool to expand domestic debt markets. In China and India today, most citizens are extremely limited as to the range of financial assets they are allowed to hold, with very low-interest bank accounts and cash essentially the only choices. With cash and jewelry at high risk of loss and theft and very few options for accumulating wealth to pay for retirement, healthcare, and children’s education, citizens still put large sums in banks despite the artificially suppressed returns. In India, banks end up lending large amounts of their assets directly to the government, which thereby enjoys a far lower interest rate than it probably would in a liberalized capital market. In China, the money goes via directed lending to state-owned enterprises and infrastructure projects, again at far lower interest rates than would otherwise obtain. This kind of financial repression is far from new and was particularly prevalent in both advanced and emerging market economies during the height of international capital controls from World War II through the 1980s.

  Under conditions of financial repression, governments can, of course, potential
ly obtain very large amounts of resources by exploiting to the fullest their monopoly over savings vehicles. However, as we will show later, domestically issued debt has flourished in many emerging markets even when financial repression has been quite limited, for example, during the decades before World War II.

  We will defer further discussion of domestic debt until we look at the issue empirically in chapters 7–9. There we will also show that there is an important interaction between sovereign debt and domestic debt. Again, as in the case of sovereign external debt, the issue of multiple equilibria often arises in models of domestic debt.23

  Conclusions

  In this chapter we have given a brief overview of the key concepts governing sovereign debt and default, as well as other varieties of crises including currency and banking crises. This chapter, while admittedly abstract, has addressed fundamental questions about international financial crises. We will return to some of these themes later in the book as our expansive new data set helps to cast light on some of the more difficult questions.

  In many regards, the theoretical work on the underpinnings of international lending and capital markets raises the question of why defaults are not more frequent. Even Venezuela, the modern-day sovereign default champion, with ten episodes since it achieved independence in 1830, still averages eighteen years between new defaults. If crises recurred almost continuously, the this-time-is-different mentality would seldom manifest itself: every time would be the same, borrowers and lenders would remain constantly on edge, and debt markets would never develop to any significant degree, certainly not to the extent that spectacular crashes are possible. But of course, economic theory tells us that even a relatively fragile economy can roll along for a very long time before its confidence bubble bursts, sometimes allowing it to dig a very deep hole of debt before that happens.

  - 5 -

  CYCLES OF SOVEREIGN DEFAULT

  ON EXTERNAL DEBT

  Policy makers should not have been overly cheered by the absence of major external sovereign defaults from 2003 to 2009 after the wave of defaults in the preceding two decades. Serial default remains the norm, with international waves of defaults typically separated by many years, if not decades.

  Recurring Patterns

  We open our tour of the panorama of financial crises by discussing sovereign default on external debt, which, as we have just been analyzing theoretically, occurs when a government defaults on debt owed to foreigners. (Some background on the historical emergence of sovereign debt markets is provided in box 5.1.)

  Figure 5.1 plots the percentage of all independent countries in a state of default or restructuring during any given year between 1800 and 2008 (for which our data set is most complete). For the world as a whole (or at least those countries with more than 90 percent of global GDP, which are represented by our data set), the relatively short period of few defaults before the late 2000s can be seen as typical of the lull that follows large global financial crises. Aside from such lulls, there are long periods when a high percentage of all countries are in a state of default or restructuring. Indeed, figure 5.1 reveals five pronounced peaks or default cycles.

  The first such peak was during the Napoleonic Wars. The second ran from the 1820s through the late 1840s, when at times nearly half the countries in the world were in default (including all of Latin America). The third began in the early 1870s and lasted for two decades. The fourth began in the Great Depression of the 1930s and extended through the early 1950s, when again nearly half of all countries stood in default.5 The final default cycle in the figure encompasses the debt crises of the 1980s and 1990s in the emerging markets.

  BOX 5.1

  The development of international sovereign debt markets in England and Spain

  Modern debt institutions as we now understand them evolved gradually. This was particularly the case with domestic borrowing, in which the relationship between taxes, repayments, and power was historically often blurred. Loans were typically highly nontransparent, with ill-specified interest rates and repayment schedules and often no specific dates on which principal repayments would be made. A king’s promise to “repay” could often be removed as easily as the lender’s head. Borrowing was frequently strongly coercive in nature. Early history is replete with examples of whole families who were slaughtered simply to seize their lands and other wealth. In thirteenth-century France, the Templars (of Crusades fame) were systematically exiled by the French kings, who seized their wealth.

  In medieval times, the church enforced usury laws that were intended to prevent Christians from lending to each other at interest. Of course, non-Christians, especially Jews, were allowed to lend, but this gave sovereigns access to only a very small pool of their nation’s total funds. In order to gain access to larger wealth pools, borrowers (sometimes with the help of theologians) had to think of ways to try to circumvent church law. During this period, international lending markets were sometimes helped by the device of having a borrower repay in a stronger, more stable currency than was specified in the original loan, perhaps repaying in currency that was not being as aggressively debased. Of course, such devices are tantamount to paying interest, yet they were often viewed as acceptable.

  By far the most sophisticated early financial markets appeared in the Italian city-states of Genoa, Florence, and Venice in the late thirteenth century. (See, for example, the excellent discussions of MacDonald or Ferguson.)1 Early loans took the guise of “repayable taxes,” but soon the system evolved to the point at which sovereign loans were sufficiently transparent that a secondary market developed.

  As historian Carlo Cipolla has emphasized, the first true international debt crisis had its roots in loans made by Italian merchants to England starting in the late thirteenth century.2 In that era, it was Italy that was the developed financial center and England the developing nation rich in natural resources, especially sheep’s wool. As we have already discussed, a sequence of Italian loans helped finance various stages of a long series of wars between England and France. When Edward III of England defaulted in 1340 after a series of military failures, the news reached Florence quickly. Because the major banks had lent heavily to Edward, a bank run hit Florence’s economy. The whole affair played out in slow motion by modern standards, but one major Italian lender, the Peruzzi Bank, went bankrupt in 1343, and another, the Bardi Bank, did in 1346. Thus England, like so many emerging markets in later eras, went through the trauma of sovereign external default (and more than once) before it eventually “graduated” to the status of nondefaulter. Before its graduation, England was to experience several more episodes of government debt restructurings; however, these more recent credit events involved only domestic debt—as we will document.

  Indeed, England did not truly cast off its status as a serial defaulter until the Glorious Revolution in 1688, which led to a substantial strengthening of Parliament’s power. As North and Weingast argued in their seminal work, this provided, for the first time, a self-renewing institution that stood behind British debt. Weingast further argued that the Bank of England, by providing a bureaucratic “delegated monitor” to oversee the government’s debt service, provided the key instrument through which Parliament expressed its power.3 Certainly a number of other factors helped support Britain’s success, including the government’s practice of using short-term debt to finance wars, then converting the debt to longer-term debt after each war’s conclusion. Short-term financing of wars makes sense, of course, because uncertainty over the war’s conclusion forces the government to pay a premium, which it will not want to lock in. The issuance of long-term debt also facilitated an active secondary market that helped make English debt liquid, a point underscored by Carlos et al.4 Finally, it cannot be overemphasized that one of the main factors underlying England’s relatively pristine repayment record is the country’s remarkable success in its many wars. As we have already seen with regard to the early British monarchs, nothing causes debt failure to the extent that war
failure does. We will return to the issue of graduation toward the end of this book.

  Prior to 1800, few nations other than England had achieved the capacity to build up significant international debts and then default on them. To achieve large-scale serial default requires a sufficient store of wealth to keep convincing each new generation of creditors that the earnings needed to repay the debt will eventually be available (that this time it will be different) and that the country is sufficiently stable to ensure that it will be around to make the payments. After 1800, thanks to rapid global income growth in the wake of the Industrial Revolution as well as to Britain’s capacity for spinning off excess savings, many countries began to fulfill the wealth criteria. Prior to 1800, aside from the early Italian cities, plus Portugal and Prussia on one occasion each, only France and Spain commanded the resources and stability to engage in big-time international defaults. And default they did, Spain six times by our count and France eight, as we illustrate in this chapter.

  Spain’s first string of defaults, in 1557, 1560, 1575, and 1596 under Philip II (1556–1598), have been extensively studied and debated by economic historians, as have the later and far uglier episodes that occurred under Philip II’s successors in 1607, 1627, and 1647. The Spanish experience illustrates a number of issues that have continually recurred in later cases of serial default. Spain is also extremely important historically as the last country to threaten the domination of Europe until Napoleon.

  Prior to the sixteenth century, Spain was sufficiently diffuse and its regions’ finances sufficiently tenuous that large-scale international borrowing was not feasible. The discovery of the New World changed all that. Spectacular lodes of silver were found in Mexico and Peru, with truly massive amounts beginning to arrive in Europe by the 1540s. The huge increase in revenues greatly enhanced the power of the king, who was no longer so reliant on domestic tax revenues, which required the cooperation of Parliament. At the same time, the influx of precious metals, especially silver, had a huge inflationary impact on prices in Europe.