1 This is the reason for an exception such as Friedman and Schwartz's (1963) monumental monetary history. This is the first formal application of the core dataset described in Reinhart and Rogoff (2009), and. the scope of the data set has also been significantly expanded. We use representative country stories to elaborate and complement some of the patterns seen in the global aggregates.
11 Table 1 in the appendix provides a brief glossary of the main categories of debt that we examine in this paper. In this case, a large increase in the number of bankruptcies or bad loans could be used to mark the beginning of a crisis. For many early episodes, it is difficult to determine how long the crisis lasted.
The Big Picture and Country Histories
Recovery in the debt ratio usually precedes the rise in default rates, as the regressions for the world aggregates shown at the bottom of Figure 3 confirm. The pattern between debt and default is consistent with the lines already discussed in the context of the world aggregate shown in Figure 3. The "characteristic" rise in debt on the eve of a debt crisis, banking crisis or both is quite evident in most of the episodes on the timeline of Brazil and Greece's last two.
The fact remains that, as Bordo and Eichengreen (2001) point out, the number of years between default episodes is much lower in the more recent period. Before World War II, serial banking crises were the norm in advanced economies; when the larger emerging markets developed a financial sector in the late 19th century, these economies joined the "serial banking" crisis club. It should also be noted that as financial markets have developed in the smaller, poorer economies, the frequency of banking crises has increased.16.
The regressions (shown at the bottom of Figure 9) confirm what the visual inspection of the time series plotted in the figure suggests. Increases in external debt systematically help predict increases in the proportion of countries defaulting and the comparable share of emerging markets in systemic banking crises. A large number of episodes documented in the Chartbook show this "prototype" pattern.
In light of the previous discussion of the time profile of external debt before, during and after the debt and banking crisis, it is not surprising that capital flows show the boom-bonanza phase in the years leading up to the crisis and the Dornbusch-Calvo -type sudden stop syndrome just before or during the year of the crisis (even in crises of a previous century and in advanced economies). Notes: Only the first year of banking crises (black lines) and defaults (light line) are shown in the top panel of the figure. Other similar examples populate country history in Reinhart and Rogoff (2010a), including the build-up of household debt almost across the board in OECD countries in the years immediately preceding the onset of the 2007–08 global crisis.
A similar pattern occurs in the run-up to government defaults (which, in this particular exercise, immediately follow banking crises). The small table in Figure 10, which shows Iceland's external debt, also shows a marked increase in the share of short-term debt as the crisis approaches, from about 17 to 49 percent. Several examples of the debt crisis that engulfed Latin America in the early 1980s and lasted for a decade are documented in various figures in the Chartbook.
Theoretical Underpinnings of “This Time is Different” Syndrome Our results beg the question of how to explain the remarkable universality of
The inset to Figure 5 on the eve of hyperinflation in the late 1980s in Brazil is another entry in this long list. Theoretical foundations of the "It's different this time" syndrome Our results raise the question of how to explain the remarkable universality. Many equilibrium models and related refinements appear to offer an explanation for one central feature of the “It's Different This Time” syndrome: it is usually much easier to determine when an economy is vulnerable to a financial crisis than to estimate the probability or timing of a collapse.
Multiple equilibria in financial markets, especially in debt markets, are quite general and therefore consistent with the proximate. The build-up of short-term debt seen on the eve of financial crises (perhaps to save on interest rate costs as debt rises) is certainly consistent with the multiple equilibrium story. With that kind of "This time it's different" mentality, you wouldn't recognize that the economy has its back to the proverbial cliff until it's too late.
In the absence of an underlying sunspot model, it is difficult to estimate the level of risk in different economies. Recent quantitative analyzes of sovereign defaults, including, for example, Aguiar and Gopinath (2006) suggest that high sovereign discount rates are a key element of any convincing explanation of the debt-default cycle. A notable example is the way in which governments regularly guarantee the debt of quasi-governmental agencies that may be taking on significant risk, particularly as was the case with mortgage giants Fannie Mae and Freddie Mac in the United States.
At the time of this writing, one only has to read the debacle in the financial press about Greece's hidden debts easily facilitated by its underwriter Goldman Sachs. Even taking all these promising strands of the political economy and financial crisis literature together, one suspects that there are still large gaps in our understanding of the arrogance and ignorance that underlies most financial crises—repeating a reading of Winkler (1933) is highly recommended. Ignorance, of course, stems from the belief that financial crises happen to other people at other times in other countries.
For example, Kahenman, Slovic, and Tversky (1982) provide examples of overconfidence in the sense of underestimating the variability of future shocks.
Debt, Banking Crises and Default: Cross-country Evidence
As noted earlier, the predominant pattern emerging from the country's history seems to suggest that banking crises come before the debt crisis. The causality tests used here mimic the spirit of the standard vector autoregression (VAR) setup. Both variables (banking and debt crisis dummies) are treated as potentially endogenous, which may or may not be explained by their own delayed values and the delayed values of the second variable.
We include as additional (exogenous) regressors the financial crisis model for the global financial center and allow the intercept to vary depending on whether the country is an advanced or an emerging market. BCt-1 at t-3 and DCt-1 at t-3 are the three-year moving averages of the two crisis variables. The main results, regardless of which sample period or estimation strategy is chosen, is that systemic banking crises in financial centers help explain domestic banking crises and domestic banking crises help explain sovereign defaults.
For the longer sample, the total public debt (domestic plus external) is PD_Yt; for the period after 1970, we also consider external (public and private debt) for the subgroup of emerging markets. 23 We estimated the same model for the entire period of the crisis - instead of just for the first year. The main result was that the lagged dependent variables were significant; this is not surprising given the fact that both the banking and debt crises are largely multi-year phenomena.
Banking crises in financial centers are still important in the domestic banking crisis equation, as Table 5 highlights. In fact, based on a close examination of country history, which links banking crises to rising private debt. Turning to the debt crisis equation, domestic banking crises continue to be a significant predictor of debt crises, while financial center crises have no direct independent effect (there is apparently an indirect relationship through a systematic relationship with domestic banking crises).
Increases in public debt have the expected significant positive effect on the likelihood of default, although the relationship appears to be somewhat weaker for the 1947-2009 runner-up.
Concluding observations
Caprio, Gerard, and Daniela Klingebiel, Luc Laeven, and Guillermo Noguera, "Banking Crisis Database." In Patrick Honohan and Luc Laeven (eds.), Systemic Financial Crises, Cambridge: Cambridge University Press, 2005. Friedman, Milton and Anna Jacobson Schwartz, A Monetary History of the United States Princeton: Princeton University Press, 1963). Reinhart, "The Twin Crises: The Causes of Banking and Balance of Payments Problems", American Economic Review, Vol.
Kletzer, Kenneth M., "Asimmetries of Information and LDC Borrowing with Sovereign Risk," The Economic Journal, Vol. Morton, "How Sovereign Debt Has Worked," in Jeffrey Sachs, red., Developing Country Debt and Economic Performance, Vol. Morris, Stephen en Hyun Song Shin, "Rethinking Multiple Equilibria in Macreconomic Modelling," in Kenneth Rogoff en Ben S Bernanke (reds.) NBER Macroeconomics Annual 2000 (Cambridge, Nasionale Buro vir Ekonomiese Navorsing).
Creditors often determine all the terms of the debt contracts, which are normally subject to the jurisdiction of the foreign creditors or to international law (for multilateral credits). Total government debt (total public debt): total debt obligations of a government with both domestic and foreign creditors. Domestic public debt: all debt obligations of a government issued under - and subject to - national jurisdiction, regardless of the nationality of the creditor or the currency denomination of the debt (so it includes domestic public debt in foreign currency, as defined below).
The terms of the debt contracts can be determined by the market or set unilaterally by the government. Foreign currency domestic government debt: debt obligations of a government issued under national jurisdiction that are nevertheless denominated in (or pegged to) a currency other than the country's national currency. Hidden debt includes contingent liabilities from the government, these could be (i) explicit guarantees (in which case they are not completely hidden).
Although we have not yet encountered any time series on public debt that quantify such guarantees, more recent measures of government guarantees are now published under the International Monetary Fund's Standard Data Dissemination System (SDDS) framework; (ii) implicit guarantees which may extend to all types of private sector debt. iii) Liabilities of the central bank (see above). iv) Off-balance sheet liabilities arising from transactions in derivative markets. v) Last but not least, any liability of the government not included in official debt statistics (so official statistics will underwrite true public sector debt) not already included in (i)-(iv) above.
