Redemption or Abstinence? Original Sin, Currency Mismatches and Counter-Cyclical Policies in the New Millennium

The emerging market crises of the 1990s focused the attention of economists on issues of debt composition and particularly currency denomination. Since in bad times the real value of the domestic currency tends to weaken, servicing foreign currency debt becomes more difficult exactly when the capacity to pay is diminished. This makes for riskier debt, less room for counter-cyclical fiscal policies and a monetary policy geared towards currency not output stability.

A debate emerged in the late 1990s and early years of the past decade regarding the causes of the prevalence of foreign currency foreign debt in emerging markets. Some saw it as a consequence of moral hazard. We worked at the time on the original sin hypothesis which was introduced by Hausmann (1999) and Eichengreen and Hausmann (1999) to describe a situation in which “the domestic currency is not used to borrow abroad or to borrow long-term even domestically” and argued that original sin is not a mere consequence of bad policies or institutions (Eichengreen, Hausmann, and Panizza, 2005a, b).

Several observers now argue that original sin is no longer a problem –if ever it was– because many developing countries expiated their sins and achieved redemption. The standard claim is that many emerging market countries can now issue external debt in their own currency and that they often do so by making greater use of the domestic debt market, where original sin is not as pervasive as in the external debt market and where foreigners are playing an increasingly active role. And so, we find anecdotal statements to the effect that “governments and corporations increasingly are borrowing in their own currencies” (David Wessel, The Wall Street Journal, May 3, 2007). Or that: “The principal emerging markets sales desk pitch of recent years has been the expiation of the “original sin” of governments’ borrowing in foreign currencies.” (John Dizard, Financial Times, October 21 2008). Summing it all up, we find Martin Wolf’s (2006, p.37) statement: “I don’t believe in original sin.” 

In fact, redemption from original sin was argued as one of the reasons for the impressive performance of emerging markets since the 2008 crisis and for their ability to conduct counter-cyclical macroeconomic policies for the first time. 

In Hausmann and Panizza (2010), we update our measures of original sin to 2008 and look at the redemption hypothesis by using data on the international and domestic bond markets for a large sample of developing, transition, and emerging market countries.  Our main findings can be summarized as follows.

First, we do find a reduction in original sin, but this reduction is small and concerns a limited number of countries. Figure 1 plots the evolution of the outstanding stock of international bonds issued by developing countries. The columns show that this market grew from approximately USD200 billion in 1993 to just above USD1 trillion in 2007 (the market shrunk by approximately 2 percent in 2008). The value of international bonds denominated in the developing countries’ currencies went from nil in 1993 to a peak of USD193 billion in 2007 (it was USD185 billion at the end of 2008). The solid and dotted lines plot the evolution of the weighted averages of the two indexes of original sin originally defined by Eichengreen, Hausmann, and Panizza (2005a). OSIN1 barely moved and went from 1 in 1993 to 0.96 in 2008. OSIN3, instead, went from 1 to 0.815, indicating that developing country can only hedge 18.5 percent of their foreign currency exposure.[1]  Out of a sample of 65 developing countries, for which we have data, only 9 ever managed to issue at least 15 percent of their debt in own currency (and only 7 countries had an index of OSIN1 lower than 0.85 in 2008) and only 18 would have ever been able to swap at least 25 percent of their international debt securities (11 countries in 2008).

Second, we find that several countries are making greater use of the domestic bond market and that this market is becoming less “sinful” in the sense that the proportion of long-term, fixed-rate debt is increasing relative to FX-, CPI- or interest-indexed debt. However, we do not find any evidence that foreign investors are now more willing to take currency risk by increasing their exposure to domestic currency bonds traded in local markets. In particular, we use data collected by the US Treasury and show that less than 10 percent of the international bonds issued by developing countries and held by US investors are denominated in the currency of the issuing country. We conclude that the domestic bond market is still not a venue through which developing countries can borrow from foreigners in local currency. If a country faced the need to borrow abroad, it would still need to do so mostly in foreign currency, and hence still suffers from original sin. This is especially true if one considers that our findings are heavily influenced by the stock of debt issued during the period 2003-2007. These were years characterized by extremely low risk aversion and by the presence of investors desperate for yield and willing to buy almost anything that promised a decent return.

Third, we document that there has been an important retreat from reliance on foreign debt as expressed in major declines in gross and net foreign debt to GDP ratios. We show that countries have reduced their currency mismatches not because they are borrowing less in foreign currency but because they are borrowing less abroad. We compute an aggregate measure of currency mismatches and decompose it into three elements: total external foreign-currency borrowing; the attenuation brought about by redemption from original sin, and the attenuation brought about by self insurance via reserve accumulation. Figure 2 shows that the average mismatch went from about 56 percent in 2000 to 27 percent in 2008. This 29 percentage point reduction was due to a 20 percentage point reduction in external debt, a 7 percentage point increase in international reserves, and a 2 percentage point reduction in original sin. Redemption from original sin explains less than 10 percent of the decrease in aggregate currency mismatch that took place since the turn of the century.  Therefore, while we find that original sin is still with us, meaning that the great majority of countries that need external funds would have to borrow in foreign currency, we show that fewer countries are willing to put themselves in a position of having to borrow abroad.

Finally, we document that developing countries exhibited a more anti-cyclical response of monetary policy to the crisis in 2008-2009 relative to the past. However, we show that these countries are now able to conduct countercyclical policies because they have lower gross and net levels of external debt and not because they have been redeemed from original sin. By running a fixed-effects panel regression of a simple Taylor rule, we find that the coefficient of the response to variations in output has moved from negative to positive and it has done so in a manner that is proportional to the reduction in the mismatch. However, as noted above, the change in the mismatch was driven by abstinence and not by redemption. Our results support the view that it is abstinence from foreign debt rather than redemption from original sin that explains the newfound ability to adopt counter-cyclical policies.

Summing up, original sin has declined but only marginally and in a few selected countries. Original sin continues to make financial globalization unattractive and developing countries have opted for abstinence rather than sin. Redemption has been the exception, not the rule. One conjecture is that, given original sin, abstinence is a better strategy than indebtedness: foreign currency debt is too risky to be sensible, given the mismatches it generates and since countries are unable to borrow without creating currency mismatches they have opted to forgo net borrowing and have extracted some of the benefits, such as the ability to adopt counter-cyclical policies.  The promised paradise of financial globalization will need to wait for redemption from original sin.

Figure 1: Bonds issued by developing countries

 

image001_512_68.gif

 Figure 2: Composition of aggregated mismatches (simple average)

 

image002_512_73.gif

References

Eichengreen, Barry and Ricardo Hausmann (1999) “Exchange Rates and Financial Fragility,” Paper presented at the symposium New Challenges for Monetary Policy, August 26–28, Jackson Hole, WY.

Eichengreen, Barry, Ricardo Hausmann, and Ugo Panizza (2005a) “The Pain of Original Sin,” in B. Eichengreen and R. Hausmann (eds.) Other People’s Money, Chicago University Press

Eichengreen, Barry, Ricardo Hausmann, and Ugo Panizza (2005b) “The Mystery of Original Sin” in B. Eichengreen and R. Hausmann (eds.) Other People’s Money, Chicago University Press

Hausmann, Ricardo (1999) “Should there be five currencies or 105?”, Foreign Policy (Fall)

Hausmann, Ricardo and Ugo Panizza (2010) “Redemption or abstinence? Original sin, currency mismatches and counter-cyclical policies in the new millennium”, Center for International Development at Harvard University, Working Paper No. 194, January.

Wolf, Martin (2006) “FIXING GLOBAL FINANCE 3. Task Unfinished: making global finance work,” The Bernard Schwartz Forum on Constructive Capitalism, March 30th 2006. Available at: www.sais-jhu.edu/mediastream/powerpoint/wolf033006.ppt

 


[1] Moreover, the 18.5 percent figure is an overestimation of the ability to hedge because some countries may have a level of outstanding debt (the debt that needs to be hedged), which is lower than the amount of debt denominated in their own currency.  


Opinions and comments on RGE EconoMonitors do not necessarily reflect the views of Roubini Global Economics, LLC, which encourages a free-ranging debate among its own analysts and our EconoMonitor community. RGE takes no responsibility for verifying the accuracy of any opinions expressed by outside contributors. We encourage cross-linking but must insist that no forwarding, reprinting, republication or any other redistribution of RGE content is permissible without expressed consent of RGE.