This Time Is Different: Eight Centuries of Financial Folly

by Carmen M. Reinhart

  • Third, there is evidence to suggest that capital flows to emerging markets are markedly procyclical (that is, they are higher when the economy is booming and lower when the economy is in recession). Procyclical capital inflows may, in turn, reinforce the tendency in these countries for macroeconomic policies to be procyclical as well. The fact that capital inflows collapse in a recession is perhaps the principal reason that emerging markets, in contrast to rich countries, are often forced to tighten both fiscal policy and monetary policy in a recession, exacerbating the downturn.12 Arguably, having limited but stable access to capital markets may be welfare improving relative to the boom-bust pattern we so often observe. So the deeply entrenched idea that the growth trajectory of an emerging market economy will be hampered by limited access to debt markets is no longer as compelling as was once thought.

  The aforementioned academic literature does not actually paint sharp distinctions between different types of capital flows—for instance, debt, equity, and foreign direct investment (FDI)—or between long-term versus short-term debt. Practical policy makers, of course, are justifiably quite concerned with the exact form that cross-border flows take, with FDI generally thought to have properties preferable to those of debt (FDI tends to be less volatile and to spin off indirect benefits such as technology transfer).13 We generally share the view that FDI and equity investment are somewhat less problematic than debt, but one wants to avoid overstating the case. In practice, the three types of capital inflows are often interlinked (e.g., foreign firms will often bring cash into a country in advance of actually making plant acquisitions). Moreover, derivative contracts can blur the three categories. Even the most diligent statistical authority may find it difficult to accurately separate different types of foreign capital inflows (not to mention the fact that, when in doubt, some countries prefer to label a particular investment FDI to make their vulnerabilities seem lower). Given these qualifications, however, we still believe that the governments of advanced countries can do more to discourage excessive dependence on risky nonindexed debt relative to other forms of capital flows.14 Finally, it should be noted that short-term debt—typically identified as the most problematic in terms of precipitating debt crises—facilitates trade in goods and is necessary in some measure to allow private agents to execute hedging strategies. Of course, one can plausibly argue that most of the benefits of having access to capital markets could be enjoyed with relatively modest ratios of debt to GNP.

  All in all, debt intolerance need not be fatal to growth and macroeconomic stability, but it is still a serious impediment. However, the evidence on serial default presented in this book suggests that to overcome debt intolerance, policy makers need to be prepared to keep debt levels low for extended periods of time while undertaking more basic structural reforms to ensure that countries can eventually digest higher debt burdens without experiencing intolerance. This applies not only to external debt but also to the reemerging problem of domestic government debt. Policy makers who face tremendous short-term pressures will still choose to engage in high-risk borrowing, and for the right price, markets will let them. But understanding the basic problem should at least guide a country’s citizens, not to mention international lending institutions and the broader international community, in making their own decisions.

  In our view, developing a better understanding of the problem of serial default on external debt obligations is essential to designing better domestic and international economic policies with regard to crisis prevention and resolution. Although further research is needed, we believe a good case can be made that debt intolerance can be captured systematically by a relatively small number of variables, principally a country’s own history of default and high inflation. Debt-intolerant countries face surprisingly low thresholds for external borrowing, beyond which risks of default or restructuring become significant. With the explosion of domestic borrowing that occurred at the turn of the twenty-first century, on which we present new data in this book, the thresholds for external debt almost certainly fell even from the low levels of a decade earlier, as we shall discuss in chapter 11. Our initial results suggest that the same factors that determine external debt intolerance, not to mention other manifestations of debt intolerance such as domestic dollarization (de facto or de jure substitution of a foreign currency for transactions or indexation of financial instruments), are also likely to impinge heavily on domestic debt intolerance.

  Finally, whereas debt-intolerant countries need badly to find ways to bring their ratios of debt to GNP to safer ground, doing so is not easy. Historically, the cases in which countries have escaped high ratios of external debt to GNP, via either rapid growth or sizable and prolonged repayments, have been very much the exception.15 Most large reductions in external debt among emerging markets have been achieved via restructuring or default. Failure to recognize the difficulty in escaping a situation of high debt intolerance simply through growth and gently falling ratios of debt to GNP is one of the central errors underlying many standard calculations employed both by the private sector and by official analysts during debt crises.

  At the time of this writing, many emerging markets are implementing fiscal stimulus packages that mirror efforts in the advanced economies in order to jump-start their economies. Our analysis suggests that in the “shadow of debt intolerance” such measures must be viewed with caution, for widening deficits leave countries uncomfortably close to debt thresholds that have been associated with severe debt-servicing difficulties. Going forward, after the global financial crisis of the late 2000s subsides, a challenge will be to find ways to channel capital to debt-intolerant countries in nondebt form to prevent the cycle from repeating itself for another century to come.

  - 3 -

  A GLOBAL DATABASE

  ON FINANCIAL CRISES

  WITH A LONG-TERM VIEW

  One would think that with at least 250 sovereign external default episodes during 1800–2009 and at least 68 cases of default on domestic public debt, it would be relatively straightforward to find a comprehensive long-range time series on public sector debt. Yet this is not the case; far from it. Government debt is among the most elusive of economic time series.

  Having defined crises and taken a first pass at analyzing vulnerability to serial default, we now turn to the core of the book, the data set. It is this lode of information that we mine in various ways to explain events. This chapter presents a broad-brush description of the comprehensive database used in this study and evaluates its main sources, strengths, and limitations. Further documentation on the coverage and numerous sources of individual time series by country and by period is provided in appendixes A.1 and A.2. Those are devoted, respectively, to the macroeconomic time series used and the public debt data (which together form the centerpiece of our analysis).

  This chapter is organized as follows. The first section describes the compilation of the family of time series that are brought together from different major and usually well-known sources. These series include prices, modern exchange rates (and earlier metal-based ones), real GDP, and exports. For the recent period, the data are primarily found in standard large-scale databases. For earlier history we relied on individual scholars or groups of scholars.1 Next we describe the data that are more heterogeneous in both their sources and their methodologies. These are series on government finances and individual efforts to construct national accounts—notably nominal and real GDP, particularly before 1900. The remaining two sections are devoted to describing the particulars of building a cross-country, multicentury database on public debt and its characteristics, as well as the various manifestations and measurements of economic crises. Those include domestic and external debt defaults, inflation and banking crises, and currency crashes and debasements. Constructing the database on public domestic and external debt can best be described as having been more akin to archaeology than to economics. The compilation of crisis episodes has encompassed the use of both mechanical rules of t
humb to date a crisis as well as arbitrary judgment calls on the interpretation of historical events as described by the financial press and scholars in the references on which we have drawn, which span more than three centuries.

  Prices, Exchange Rates,

  Currency Debasement, and Real GDP

  Prices

  Our overarching ambition in this analysis is to document the incidence and magnitude of various forms of expropriation or default through the ages. No such study would be complete without taking stock of expropriation through inflation. Following the rise of fiat (paper) currency, inflation became the modern-day version of currency “debasement,” the systematic degradation of metallic coins that was a favored method of monarchs for seizing resources before the development of the printing press. To measure inflation, we generally rely on consumer price indexes or their close relative, cost-of-living indexes. For the modern period, our data sources are primarily the standard databases of the International Monetary Fund: International Financial Statistics (IFS) and World Economic Outlook (WEO). For pre–World War II coverage (usually from early 1900s or late 1800s), Global Financial Data (GFD), several studies by Williamson,2 and the Oxford Latin American Economic History Database (OXLAD) are key sources.3

  For earlier periods in the eight centuries spanned by our analysis, we rely on the meticulous work of a number of economic historians who have constructed such price indexes item by item, most often by city rather than by country, from primary sources. In this regard, the scholars participating in the Global Price and Income History Group project at the University of California–Davis and their counterparts at the Dutch International Institute of Social History have been an invaluable source of prices in Asia and Europe.4 Again, the complete references by author to this body of scholarly work are given in the data appendixes and in the references. For colonial America, Historical Statistics of the United States (HSOUS, recently updated) provides the U.S. data, while Richard Garner’s Economic History Data Desk: Economic History of Latin America, the United States and the New World, 1500–1900, covers key cities in Latin America.5

  On the Methodology Used in Compiling Consumer Price Indexes

  When more than one price index is available for a country, we work with the simple average. This approach is most useful when there are price series for more than one city for the same country, such as in the pre-1800s data. When no such consumer price indexes are available, we turn to wholesale or producer price indexes (as, for example, for China in the 1800s and the United States in the 1720s). Absent any composite index, we fill in the holes in coverage with individual commodity prices. These almost always take the form of wheat prices for Europe and rice prices for Asia. We realize that a single commodity (even if it is the most important one) is a relative price rather than the aggregate we seek, so if for any given year we have at least one consumer (or cost-of-living) price series and the price of wheat (or rice), we do not average the two but give full weight to the composite price index. Finally, from 1980 to the present, the International Monetary Fund’s World Economic Outlook dominates all other sources, because it enforces uniformity.

  Exchange Rates, Modern and Early, and Currency Debasement

  The handmaiden to inflation is, of course, currency depreciation. For the period after World War II, our primary sources for exchange rates are IFS for official rates and Pick’s Currency Yearbooks for market-based rates, as quantified and documented in detail by Reinhart and Rogoff.6 For modern prewar rates GFD, OXLAD, HSOUS, and the League of Nations’ Annual Reports are the primary sources. These are sometimes supplemented with scholarly sources for individual countries, as described in appendix A.1. Less modern are the exchange rates for the late 1600s through the early 1800s for a handful of European currencies, which are taken from John Castaing’s Course of Exchange, which appeared twice a week (on Tuesdays and Fridays) from 1698 through the following century or so.7

  We calculated the earlier “silver-based” exchange rates (trivially) from the time series provided primarily by Robert Allen and Richard Unger, who constructed continuous annual series on the silver content of currencies for several European currencies (for other sources see individual tables in the data appendixes, which list individual authors).8 The earliest series, for Italy and England, begins in the mid-thirteenth century. As described in appendix A.1.4, these series are the foundation for dating and quantifying the “debasement crises”—the precursors of modern devaluations, as cataloged and discussed in chapter 11.

  Real GDP

  To maintain homogeneity inasmuch as it is possible for such a large sample of countries over the course of approximately two hundred years, we employ as a primary source Angus Maddison’s data, spanning 1820–2003 (depending on the country), and the version updated through 2008 by the Groningen Growth and Development Centre’s Total Economy Database (TED).9 GDP is calculated on the basis of purchasing power parity (PPP) in 1990.10 TED includes, among other things, series on levels of real GDP, population, and GDP per capita for up to 125 countries from 1950 to the present. These countries represent about 96 percent of the world’s population. Because the smaller and poorer countries are not in the database, the sample represents an even larger share of world GDP (99 percent). We do not attempt to include in our study aggregate measures of real economic activity prior to 1800.11

  To calculate a country’s share of world GDP continuously over the years, we sometimes found it necessary to interpolate the Maddison data. (By and large, the interpolated GDP data are used only in forming weights and percentages of global GDP. We do not use them for dating or calibrating crises.) For most countries, GDP is reported only for selected benchmark years (e.g., 1820, 1850, 1870). Interpolation took three forms, ranging from the best or preferred practice to the most rudimentary. When we had actual data for real GDP (from either official sources or other scholars) for periods for which the Maddison data are missing and periods for which both series are available, we ran auxiliary regressions of the Maddison GDP series on the available GDP series for that particular country in order to interpolate the missing data. This allowed us to maintain cross-country comparability, enabling us to aggregate GDP by region or worldwide. When no other measures of GDP were available to fill in the gaps, we used the auxiliary regressions to link the Maddison measure of GDP to other indicators of economic activity, such as an output index or, most often, central government revenues—for which we have long-range continuous time series.12 As a last resort, if no potential regressors were available, we used interpolation to connect the dots of the missing Maddison data, assuming a constant annual growth rate in between the reported benchmark years. Although this method of interpolation is, of course, useless from the vantage point of discerning any cyclical pattern, it still provides a reasonable measure of a particular country’s share of world GDP, because this share usually does not change drastically from year to year.

  Exports

  As is well known, export data are subject to chronic misinvoicing problems because exporters aim to evade taxes, capital controls, and currency restrictions.13 Nevertheless, external accounts are most often available for a far longer period and on a far more consistent basis than are GDP accounts. In spite of problems resulting from misinvoicing, external accounts are generally considered more reliable than most other series on macroeconomic activity. The postwar export series used in this study are taken from the International Monetary Fund (IMF), whereas the earlier data come primarily from GFD and OXLAD. Official historical statistics and assorted academic studies listed in appendix A.1 complement the main databases. Trade balances provide a rough measure of the country-specific capital flow cycle, particularly for the earlier periods, from which data on capital account balances are nonexistent. Exports are also used to scale debt, particularly external debt.

  Government Finances and National Accounts

  Public Finances

  Data on government finances are primarily taken from Mitchell for the pre-1963 period and f
rom Kaminsky, Reinhart, and Végh and sources they have cited for the more recent period.14 The Web pages of the central banks and finance ministries of the many countries in our sample provide the most up-to-date data. For many of the countries in our sample, particularly in Africa and Asia, the time series on central government revenues and expenditures date back to the colonial period. Details on individual country coverage are presented in appendix table A.1.7. In nearly all cases, the Mitchell data go back to the 1800s, enabling us to calculate ratios of debt to revenue for many of the earlier crises.

  The European State Finance Database, which brings together data provided by many authors, is an excellent source for the larger European countries for the pre-1800 era, because it offers considerable detail on government revenues and expenditures, not to mention extensive bibliographical references.

  National Accounts

  Besides the standard sources, such as the IMF, the United Nations, and the World Bank, which provide data on national accounts for the post–World War II period (with different starting points depending on the country), we consult other multicountry databases such as OXLAD for earlier periods. As with other time series used in this study, the national account series (usually for the period before World War I) build on the efforts of many scholars around the world, such as Brahmananda for India, Yousef for Egypt, and Baptista for Venezuela.15