APPENDIX A.2
PUBLIC DEBT
This appendix covers the government debt series used, while appendix A.1 is devoted to the database on macroeconomic time series.
Abbreviations of Frequently Used Sources and Terms
Additional sources are listed in the tables that follow.
CLYPS
Cowan, Levy-Yeyati, Panizza, Sturzenegger (2006)
ESFDB
European State Finance Data Base
GDF
World Bank, Global Development Finance
GFD
Global Financial Data
IFS
International Monetary Fund, International Financial Statistics (various issues)
Lcu
local currency units
LM
Lindert and Morton (1989)
LofN
League of Nations, Statistical Yearbook (various years) MAR Marichal (1989)
MIT
Mitchell (2003a, 2003b)
RR
Reinhart and Rogoff (year as noted)
UN
United Nations, Statistical Yearbook (various years)
WEO
International Monetary Fund, World Economic Outlook (various issues)
TABLE A.2.1
Public debentures: External government bond issues
TABLE A.2.2
Total (domestic plus external) public debt
TABLE A.2.3
External public debt
TABLE A.2.4
Domestic public debt
APPENDIX A.3
DATES OF BANKING CRISES
TABLE A.3.1
Banking crisis dates and capital mobility, 1800–2008
APPENDIX A.4
HISTORICAL SUMMARIES
OF BANKING CRISES
TABLE A.4.1
Banking crises: Historical summaries, 1800–2008
NOTES
Preface
1. Notably those of Winkler (1928), Wynne (1951), and Marichal (1989).
2. More recently, there was Ferguson’s (2008) excellent and equally engaging history of the foundations of currency and finance. See also MacDonald (2006).
Preamble: Some Initial Intuitions on Financial Fragility
and the Fickle Nature of Confidence
1. See Shleifer and Vishny (1992) and Fostel and Geanakoplos (2008) for interesting technical analyses of how changing fortunes of optimists and pessimists can drive leverage cycles.
2. Classic articles on multiple equilibria and financial fragility include those of Diamond and Dybvig (1983) and Allen and Gale (2007) on bank runs, Calvo (1988) on public debt, and Obstfeld (1996) on exchange rates. See also Obstfeld and Rogoff (1996), chapters 6 and 9.
3. See Buchanan and Wagner (1977).
4. Krugman (1979).
5. See, for instance, North and Weingast (1988) and also Ferguson (2008).
6. See Bernanke (1983) and Bernanke and Gertler (1990), for example.
7. We term the recent global crisis the “Second Great Contraction” in analogy with Friedman and Schwartz’s (1963) depiction of the 1930’s Great Depression as “The Great Contraction.” Contraction provides an apt description of the wholesale collapse of credit markets and asset prices that has marked the depth of these traumatic events, along with, of course, contracting employment and output.
Chapter 1 Varieties of Crises and Their Dates
1. See Reinhart and Rogoff (2004).
2. Frankel and Rose (1996).
3. Ibid.; Kaminsky and Reinhart (1999).
4. See Kaminsky and Reinhart (1999) for the construction of thresholds to date equity price crashes and Reinhart and Rogoff (2008b) for a depiction of the behavior of real estate prices on the eve of banking crises in industrialized economies.
5. See Kaminsky and Reinhart (1999), Caprio and Klingebiel (2003), Caprio et al. (2005), and Jácome (2008). For the period before World War II, Willis (1926), Kindleberger (1989), and Bordo et al. (2001) provide multicountry coverage on banking crises.
6. See Camprubri (1957) for Peru, Cheng (2003) and McElderry (1976) for China, and Noel (2002) for Mexico.
7. This is not meant to be an exhaustive list of the scholars who have worked on historical sovereign defaults.
8. Notably, that supplied by Lindert and Morton (1989), Suter (1992), Purcell and Kaufman (1993), and MacDonald (2006). Of course, required reading in this field includes Winkler (1933) and Wynne (1951). Important further readings include Eichengreen (1991a, 1991b, 1992), and Eichengreen and Lindert (1989).
9. At present, Honduras has remained in default since 1981.
10. Apparently an old saw in the marketplace is “More money has been lost because of four words than at the point of a gun. Those words are ‘This time is different.’ ”
11. For example, during the mid-1990s Thailand claimed not to have a dollar peg but rather a peg to an (unspecified) basket of currencies. Investors could clearly see, however, that the basket did not contain much besides the dollar; the exchange rate of baht to dollars fluctuated only within narrow bands.
12. Central banks typically lose money in any unsuccessful intervention to prop up the currency because they are selling hard currency (e.g., dollars) in exchange for the local currency (e.g., baht). When the exchange rate for the local currency collapses, the intervening central bank suffers a capital loss.
Chapter 2 Debt Intolerance
1. Later, in chapter 8, we use new historical data on domestic public debt in emerging markets and find that it is a significant factor in some cases. However, introducing this consideration does not fundamentally change the remarkable phenomenon of serial default examined here.
2. The figures for Japan’s level of debt relative to GDP are from International Monetary Fund, World Economic Outlook, October 2008.
3. Following the World Bank for some purposes, we divide developing countries according to their level of per capita income into two broad groups: middle-income countries (those with a GNP per capita in 2005 higher than US$755) and low-income countries. Most (but not all) emerging market economies with substantial access to private external financing are middle-income countries. Similarly, most (though not all) of the low-income countries do not have access to private capital markets and rely primarily on official sources of external funding.
4. Note that many of these default episodes lasted several years, as discussed in chapter 8.
5. Note that tables 2.1 and 2.2 measure gross total external debt because debtor governments have little capacity to tax or otherwise confiscate private citizens’ assets held abroad. For example, when Argentina defaulted on US$95 billion of external debt in 2001, its citizens held foreign assets abroad estimated by some commentators at about US$120–150 billion. This phenomenon is not uncommon and was the norm in the debt crises of the 1980s.
6. Using an altogether different approach, an International Monetary Fund (2002) study on debt sustainability came up with external debt thresholds for developing countries (excluding the highly indebted poorest countries) that were in the neighborhood of 31 to 39 percent, depending on whether official financing was included or not. The results, which we will present later, suggest that country-specific thresholds for debt-intolerant countries should probably be even lower.
7. For particulars about the survey, see the September 2002 issue of Institutional Investor and their Web site. Though not critical to our following analysis, we interpret the ratings reported in each semiannual survey as capturing the risk of near-term default within one to two years.
8. One can use secondary market prices of external commercial bank debt, which are available for the time since the mid-1980s, to provide a measure of expected repayment for a number of emerging market countries. However, the Brady debt restructurings of the 1990s converted much of this bank debt to bond debt, so from 1992 onward the secondary market prices would have to be replaced by the Emerging Market Bond Index (EMBI) spread, which remains the most commonly used measure of risk at
present. These market-based indicators introduce a serious sample selection bias: almost all the countries in the EMBI, and all the countries for which there is secondary debt price data for the 1980s, had a history of adverse credit events, leaving the control group of nondefaulters approximately the null set.
9. See the debt glossary (box 1.1) for a brief explanation of the various concepts of debt used in this study.
10. This exercise updates the work of Reinhart, Rogoff, and Savastano (2003a), who used thresholds based on a smaller sample of countries over 1979–2002.
11. Prasad, Rogoff, Wei, and Kose (2003) found that during the 1990s, economies that were de facto relatively financially open experienced, on average, a rise in consumption volatility relative to output volatility, contrary to the premise that the integration of capital markets spreads country-specific output risk. Prasad et al. also argue that the cross-country empirical evidence on the effects of capital market integration on growth shows only weak positive effects at best, and arguably none.
12. See Kaminsky, Reinhart, and Végh (2004) on this issue.
13. Of course, it was not always so. Prior to the 1980s, many governments viewed allowing foreign direct investment (FDI) as equivalent to mortgaging their futures, and hence preferred debt finance. And where FDI was more dominant (e.g., in oil and natural resources investment in the 1950s and 1960s), many countries eventually ended up seizing foreigners’ operations later on, again leading to considerable trauma. Thus FDI should not be regarded as a panacea for poor growth performance.
14. Rogoff (1999) and Bulow and Rogoff (1990) argue that the legal systems of creditor countries should be amended so they no longer tilt capital flows toward debt.
15. The issues of debt reduction and debt reversal are taken up in box 5.3.
Chapter 3 A Global Database on
Financial Crises with a Long-Term View
1. Detailed citations are in our references and data appendix.
2. See Williamson’s “regional” papers (1999, 2000a, 2000b). These regional papers provided time series for numerous developing countries for the mid-1800s to before World War II.
3. For OXLAD, see http://oxlad.qeh.ox.ac.uk/. See also Williamson (1999, 2000a, 2000b).
4. See http://gpih.ucdavis.edu/ and http://www.iisg.nl/hpw/. Although our analysis of inflation crises begins in 1500, many of the price series begin much earlier.
5. HSOUS is cited in the references as Carter et al. (2006); Garner’s Economic History Data Desk is available at http://home.comcast.net/~richardgarner04/.
6. Reinhart and Rogoff (2004).
7. See Richard Bonney’s European State Finance Database (ESFDB), available at http://www.le.ac.uk/hi/bon/ESFDB/frameset.html.
8. Allen and Unger’s time series, European Commodity Prices 1260–1914, is available at http://www2.history.ubc.ca/unger/htm_files/new_grain.htm. Sevket Pamuk has constructed comparable series for Turkey through World War I (see http://www.ata.boun.edu.tr/sevket%20pamuk.htm).
9. See Maddison (2004). The TED is available at http://www.ggdc.net/.
10. PPP is calculated using Geary-Khamis weights. The Geary-Khamis dollar, also known as the international dollar, is a hypothetical unit of currency that has the same purchasing power that the U.S. dollar had in the United States at a given point in time. The year 1990 is used as a benchmark year for comparisons that run through time. The Geary-Khamis dollar shows how much a local currency unit is worth within the country’s borders. It is used to make comparisons both between countries and over time.
11. There are exceptions. For instance, Rodney Edvinsson’s careful estimates for Sweden from 1720 to 2000 or HSOUS for the United States beginning in 1790 offers a basis from which to examine earlier economic cycles and their relation to crises.
12. It is well known that revenues are intimately linked to the economic cycle.
13. See, for example, calculations in the background material to Reinhart and Rogoff (2004), available on the authors’ Web pages.
14. See Mitchell (2003a, 2003b) and Kaminsky, Reinhart, and Végh (2004).
15. See Brahmananda (2001), Yousef (2002), and Baptista (2006).
16. These numbers are a lower bound because they do not include the many sovereign defaults prior to 1800 and, as regards domestic defaults, we have only begun to skim the surface; see Reinhart and Rogoff (2008c).
17. This description comes from the IMF’s Web site, http://www.imf.org/external/data.htm: “Download time series data for GDP growth, inflation, unemployment, payments balances, exports, imports, external debt, capital flows, commodity prices, more.”
18. For some countries, such as the Netherlands, Singapore, and the United States, practically all public debt is domestic.
19. For Australia, Ghana, India, Korea, and South Africa, among others, we have put together debt data for much of the colonial period.
20. See Miller (1926), Wynne (1951), Lindert and Morton (1989), and Marichal (1989).
21. Flandreau and Zumer (2004) are an important data source for Europe, 1880–1913.
22. Even under these circumstances, they continue to be a useful measure of gross capital inflows; the earlier sample included relatively little private external borrowing or bank lending.
23. Indonesia prior to 1972 is a good example of a country where this exercise was particularly useful.
24. Jeanne and Guscina (2006) compiled detailed data on the composition of domestic and external debt for nineteen important emerging markets for 1980–2005; Cowan et al. (2006) performed a similar exercise for all the developing countries of the Western Hemisphere for 1980–2004. See Reinhart, Rogoff, and Savastano (2003a) for an early attempt to measure domestic public debt for emerging markets.
25. http://www.imf.org/.
26. The fact that some of the world’s poorest countries often fail to fully repay their debts to official lenders cannot be construed as a financial crisis in the conventional sense, because the official lenders often continue to provide aid nevertheless. See Bulow and Rogoff (2005) for a discussion of the related issue of whether multilateral development banks would be better structured as outright aid agencies.
Chapter 4 A Digression on the
Theoretical Underpinnings of Debt Crises
1. During the late 1920s, Stalin’s collectivization of farms led to mass starvation, and Russia needed money to import grain. As a result, in 1930 and 1931 the country sold some of its art treasures to foreigners, including the British oil magnate Calouste Gulbenkian and the American banker Andrew Mellon. But Stalin surely did not contemplate using any of the proceeds to repay old Tsarist debts.
2. See Persson and Tabellini (1990) and Obstfeld and Rogoff (1996) for literature surveys.
3. Tomz (2007).
4. See Eaton and Gersovitz (1981).
5. If countries simply borrow ever greater amounts without ever repaying, levels of debt relative to income must eventually explode provided the world (risk-adjusted) real interest rate exceeds the country’s long-term real growth rate, which generally appears to be the case both in practice and under reasonable theoretical restrictions.
6. More generally, the game theoretic approach to reputation detailed by Eaton and Gersovitz typically admits to a huge variety of equilibria (outcomes), all of which can be rationalized by the same reputation mechanism.
7. See Bulow and Rogoff (1989b).
8. See Bulow and Rogoff (1989a).
9. Bulow and Rogoff (1989b) present a simple example based on a tariff war; Cole and Kehoe (1996) place this argument in a more general setting.
10. Borensztein et al. (1998) examine the empirics of the relationship of FDI and economic growth.
11. See Diamond and Rajan (2001).
12. See Jeanne (2009).
13. See Sachs (1984).
14. See, for example, Obstfeld and Rogoff (1996, chapter 6).
15. See Bulow and Rogoff (1988a, 1989a).
16. The citizens of a creditor country (outside of banks) may reali
ze gains from trade just as the citizens of the debtor country do. Or, in the case of an international lending agency such as the IMF, creditors and debtors may be able to induce payments based on the IMF’s fear that a default will lead to contagion to other borrowers.
17. See Jayachandran and Kremer (2006).
18. See, for example, Broner and Ventura (2007).
19. See Barro (1974).
20. See North and Weingast (1988).
21. See Kotlikoff, Persson, and Svensson (1988).
22. See Tabellini (1991).
23. For example, see Barro and Gordon (1983).
Chapter 5 Cycles of Sovereign Default on External Debt
1. MacDonald (2006); Ferguson (2008).
2. Cipolla (1982).
3. North and Weingast (1988); Weingast (1997).
4. Carlos et al. (2005).
5. Kindleberger (1989) is among the few scholars who emphasize that the 1950s still has to be viewed as a financial crisis era.
6. This comparison weights defaulting countries by share of world income. On an unweighted basis (so that, for example, the poorest countries in Africa and South Asia receive the same weight as Brazil or the United States), the period from the late 1960s until 1982 saw an even lower percentage of independent countries in default.
7. Kindleberger (1989) emphasizes the prevalence of default after World War II, though he does not provide quantification.
8. Note that in figure 5.2 the debt crises of the 1980s do not loom as large as the previous cycle of defaults, for only middle- and low-income countries faced default in the 1980s while, in addition to emerging market economies, several advanced economies defaulted during the Great Depression and several more defaulted during World War II.
This Time Is Different: Eight Centuries of Financial Folly Page 28