شناسایی استهلاک نرخ ارز بر ریسک کشور: مدارک و شواهد از یک آزمایش طبیعی

A natural experiment is used to study exchange rate depreciation and perceived sovereign risk. France suspended coinage of silver in 1876 provoking a significant exogenous depreciation of all silver standard countries versus gold standard currencies like the British pound – the currency in which their debt was payable. The evidence suggests an exchange rate depreciation can significantly increase sovereign risk if a country is exposed to foreign currency debt. We implement a difference-in-differences estimator and find that the average silver country's spread on hard currency debt increased over ten percent relative to non-silver countries.

A currency mismatch occurs when a country's debt is denominated in a foreign currency while its revenue streams are largely in local currency. Currency mismatches make a country vulnerable. A sudden real exchange rate depreciation can abruptly reduce the ability to repay foreign currency debt. This deterioration of a country's “balance sheet” could easily increase default risk.1 Currency mismatches are in fact ubiquitous, and they are deemed by many to create financial fragility by accelerating the onset of a financial crisis or exacerbating the severity of financial crises. Balance sheet problems are at the heart of many explanations for the severity of the 1997 Asian financial crisis and have been analyzed in new micro-founded open-economy models such as that found in Céspedes et al. (2004).2Eichengreen et al. (2005) discuss the impact of ‘original sin’ on various macroeconomic indicators.3 However, few, if any, papers have been able to empirically assess the precise links among currency depreciation, a country's balance sheet, and default risk given the endogeneity of the exchange rate.4 There is typically a link between depreciation episodes and perceived policy problems or poor fundamentals that are hard to measure. Hence econometric studies of the issue make identification of the exchange rate channel difficult. Nevertheless the logic that exchange rates themselves matter independent of policy is quite evident. Fortunately, history provides a natural experiment to isolate the effect on sovereign risk of an exogenous exchange rate depreciation unrelated to fundamentals. We focus on the accelerated depreciation of silver in early 1876 which was connected to the anticipation of France's August 1876 decision to suspend the coinage of silver. France's move in August 1876 accelerated the trend depreciation of silver already underway since 1873. As shown in Fig. 1, this decreased demand for silver (and increased demand for gold) led to an historically abrupt depreciation of silver.5 We argue that France's decision and the depreciation of silver was exogenous for countries with silver-based commodity money systems. The French debate was watched closely by markets throughout early 1876, and French suspension of silver coinage became an increasingly sure thing up to August 1876. Between January 1876 and mid-1876 markets incorporated this information into their expectations perceiving the likelihood of an accelerated and sustained silver depreciation to be greater and greater with each passing week. These factors, combined with the fact that a large portion of countries' liabilities were denominated and payable in gold currency, suggest that the rapid depreciation of silver-based exchange rates vis-à-vis their gold creditors in early 1876 provides a unique historical experiment to study one of the key predictions of the theoretical literature on currency mismatches.We use a new weekly database of sovereign debt prices collected from The Economist, and undertake a before-and-after comparison or event study using a difference-in-differences (DID) regression strategy. This approach eliminates pre-existing differences in risk between silver and non-silver countries and controls for market forces affecting all countries. We measure the impact of the French suspension of silver coinage on sovereign yield spreads on hard currency debt for countries on a silver standard compared to non-silver countries. Our key finding is that the exchange rate is an important determinant of sovereign risk if a country has foreign currency debt. Interest-rate spreads rise by 61 basis points for countries that experience an exogenous depreciation of ten percent depreciation and have a foreign currency debt to export ratio of one. The comparison group is other countries with similar levels of hard currency that did not suffer such a depreciation. The exchange rate shock appears to precipitate a ten percent rise in bond spreads for the average country faced with this shock. Our results are consistent with the idea that exchange rate fluctuations themselves can indeed lead markets to enforce higher default risk premia. This rules out the possibility that the only reason that foreign currency debt is problematic is poor underlying fundamentals. The rest of the paper proceeds as follows. Section 2 reviews different explanations for the French-cum-global demonetization of silver. This is followed by an historical discussion of important events in the market for gold and silver in the 1870s to motivate the event study. We then introduce the new database on sovereign spreads and analyze the effects of the suspension of silver coinage by France on sovereign default risk. Section 4 concludes the paper with a discussion and the implications of our results.

Does a real exchange rate depreciation increase default risk by reducing the perceived ability of a country to repay foreign currency debts? Under normal circumstances, this is a very difficult question to address given the endogenous properties of the exchange rate. Countries with poor fundamentals often resort to inflation finance leading to depreciation thus confounding the impact of policy mismanagement with actual depreciation. Moreover, there is a large debate between those who think original sin is a problem versus those that think debt intolerance (cf. Reinhart et al., 2003) and governance is the problem. Our results show that sharp exchange rate swings can be perceived as problematic for repayment capacity. History offers some insight into the original sin/debt intolerance debate. France's decision to permanently abandon a bimetallic standard and adopt the gold standard in the 1870s led to a large decrease in the demand for silver as the country fully suspended its substantial purchases of the precious metal. In anticipation of the demise of bimetallism, the world gold price of silver fell by a historically unprecedented ten to fifteen percent in the first half of 1876. While the decision to abandon a bimetallic standard and adopt gold may have been endogenous for France, it was an exogenous decision from the viewpoint of countries and policymakers on the periphery. The empirical results show that silver countries experienced a differential rise in yield spreads from the moment that continued sizeable depreciation became more likely. The increase in sovereign risk for countries on the silver standard is most likely a short to medium-term effect, however. This is consistent with a balance sheet/exchange rate channel of financial distress that many observers of the Asian crisis of 1997 said were fundamental. However, risk premia are partially determined by predictions about the future course of policies and events. All else equal, markets must have expected a medium-term improvement in the balance sheet of silver countries. As exports expanded (cf. Nugent, 1973), the ability to service debt and revenue bases would have surely expanded. This could occur either with the expansion of national income or as revenue tariffs increased in step with imports. A stronger export position also could have increased the availability of hard currency that could help pay ongoing liabilities. Many silver countries also eventually adopted the gold standard at the turn of the twentieth century. It is inconceivable however that any of our silver countries could have made the transition to gold within one year of the 1876 shock (see Meissner, 2005). Other silver countries persistently ran loose monetary policies or clung to the commodity standard for a number of years. This could have led to a persistent increase in sovereign risk to the extent the size of the depreciation was unexpected. In the end such expectations would tend to bias downward the true impact on spreads in our regressions. Still, we find a positive and statistically significant impact of depreciation. Overall, we interpret our results as strong evidence that hard currency debts and depreciation can impact sovereign risk even if only in a transitory manner. We have also examined an important public policy question by looking at the relationship between default risk and exchange rate depreciation from one of the most important disruptions to the international monetary system in the last 150 years. We conclude that if LDCs are going to embrace international capital flows denominated in hard currency, then shocks to their exchange rates may make foreign financing significantly more costly.