آیا PEG ها مداوم واقعا بیشتر در معرض بحران ارز هستند؟

This paper evaluates the treatment effect of consistent pegs (i.e., a policy in which countries actually adopt announced pegged regimes) on the occurrence of currency crises to examine whether consistent pegs are indeed more prone to currency crises than other regimes. Using matching estimators as a control for the self-selection problem of regime adoption, we find that countries with consistent pegs have a significantly lower probability of currency crises than countries with other exchange rate policies. More interestingly, we find that countries with consistent pegs have a significantly lower probability of currency crises than those with a “fear of announcing a peg” policy (i.e., a policy in which countries actually adopt pegged regimes but do not claim to have pegged regimes). The results stand up to a wide variety of robustness checks.

It is well known that several countries' actual (de facto) exchange rate regimes are inconsistent with their official (de jure) exchange rate regimes ( Calvo and Reinhart, 2002, Reinhart and Rogoff, 2004 and Levy-Yeyati and Sturzenegger, 2005). For example, Calvo and Reinhart (2002) suggest that many countries that claim to have floating regimes in reality actively manage their exchange rates. According to Alesina and Wagner (2006), because a large depreciation (or devaluation) of an exchange rate makes market participants recognize that such countries are vulnerable in terms of monetary and exchange rate regimes, many countries try to actively manage exchange rates to avoid such situations, even if they claim to have floating regimes. Therefore, many countries strategically (or are forced to) follow regimes that differ from their announced regimes. It is often claimed that announcing the adoption of pegged regimes increases the risk of currency crises because official pegs may become targets of speculative attacks (e.g., Levy-Yeyati and Sturzenegger, 2005 and Genberg and Swoboda, 2005). However, countries with consistent pegs (i.e., a policy whereby countries claim to have pegged regimes and actually adopt the announced pegged regimes) may avoid speculative attacks because they can enhance the credibility of their currencies by following through on their commitments to adopt pegged regimes. In a world with increasingly integrated capital markets, are consistent pegs really prone to speculative attacks and currency crises, as is commonly assumed? The purpose of this paper is to answer this question. According to Levy-Yeyati and Sturzenegger (2005), to avoid speculative attacks, many countries adopt pegged regimes but do not claim to have pegged regimes when they wish to stabilize their currencies. Alesina and Wagner (2006) call this behavior “fear of announcing a peg” (FOAP).1 Is this policy effective? Are there significant differences in the probability of currency crises between the policy of adopting regimes different from the announced regimes and that of actually adopting the announced regimes?2 These questions must be addressed to evaluate actual exchange rate policies. However, empirical literature investigating the links between actual regimes, announced regimes, and the occurrence of currency crises is limited.3 This paper seeks to address this research gap. The choice of optimal exchange rate regimes is one of the most important topics in international economics that has been studied and debated over many decades.4 In particular, much interest has been focused on the question of the optimal regime for currency crisis prevention following the major currency crises in the 1990s and early 2000s (e.g., the European Monetary System (EMS) crisis in 1992–1993, the Mexican crisis in 1994–1995, the Asian crisis in 1997–1998, and the Argentine crisis in 2001–2002). Recently, several empirical studies have examined the links between the choice of exchange rate regimes and the occurrence of currency crises, using various dataset and methods (e.g., Ghosh et al., 2003, Bubula and Ötker-Robe, 2003, Husain et al., 2005, Haile and Pozo, 2006, Coulibaly, 2009 and Esaka, 2010). For example, Ghosh et al. (2003) estimate the occurrence of currency crises under alternative exchange rate regimes in IMF member countries from 1972 to 1999, using the data based on the IMF de jure classification, and they find that the probability of crises is the highest for floating regimes. Haile and Pozo (2006) investigate whether exchange rate regimes affect the incidence of currency crises in 18 developed countries from 1974 to 1998 by estimating probit models based on the data of the IMF classification and the de facto Levy-Yeyati and Sturzenegger (2005) classification. They find that, while the probability of currency crises is significantly higher for pegged regimes than for other regimes when the IMF classification is used, there is no significant link between exchange rate regimes and currency crises when the Levy-Yeyati and Sturzenegger (2005) classification is used. Esaka (2010) examines whether de facto exchange rate regimes affect the occurrence of currency crises in 84 countries from 1980 to 2001 by estimating probit models using the data based on the de facto Reinhart and Rogoff (2004) classification. The study by Esaka (2010) finds that pegged regimes significantly decrease the likelihood of currency crises compared with floating regimes. Using the combined data of exchange rate regimes and the existence of capital controls, it also finds that pegged regimes with capital account liberalization have a significantly lower likelihood of currency crises compared with other regimes. However, as noted above, previous studies provide a mixed view of the effects of exchange rate regimes on the occurrence of currency crises. Therefore, it is useful to determine which types of exchange rate regimes are more susceptible to speculative attacks and currency crises and which exchange rate regimes can avoid currency crises. As Levy-Yeyati and Sturzenegger (2005) point out, if many countries strategically follow actual regimes that differ from their announced regimes to avoid speculative attacks, an examination of whether deviations of actual exchange rate regimes from announced regimes affect the occurrence of currency crises will provide new and useful information. Accordingly, this paper empirically evaluates the effect of consistent pegs on the occurrence of currency crises to examine whether countries with consistent pegs have significantly higher or lower probabilities of experiencing currency crises than countries with other exchange rate policies. In doing so, we investigate whether deviations of actual exchange rate regimes from announced regimes affect the occurrence of currency crises. We use the relationship between the de jure IMF classification and the de facto Reinhart and Rogoff (2004) classification as an indicator of discrepancy or consistency between announced and actual regimes. To properly estimate the effect of exchange rate regimes on the incidence of currency crises, we must carefully control for self-selection problems with respect to regime adoption. Previous studies (cited above) do not explicitly address this issue. In the estimations of these studies, self-selection bias can arise because a country's exchange rate regime choice is non-random (i.e., there are systematic differences between countries that do and do not adopt a specific regime). Thus, previous studies may provide an inaccurate picture of the effect of exchange rate regimes on the occurrence of currency crises. In this paper, we employ the bias-corrected matching estimator of Abadie and Imbens (2006) to address the self-selection problem of regime adoption. Abadie and Imbens (2006) show that a simple matching estimator will be biased in finite samples because matching is not exact when the matching variables are continuous. To remove this bias, they propose a bias-corrected matching estimator, which adjusts for differences in covariate values within the matches. To our knowledge, no other paper investigates the effect of consistent pegs on the occurrence of currency crises using matching methods. The central concept of matching methods is matching each treated unit with control units that have similar observed characteristics and then comparing the outcomes of the treated units and those of the control units. The advantage of matching methods is that they can formally control for the non-random selection problem and avoid specification of a functional form because they are nonparametric techniques (Imbens, 2004). Matching methods thus avoid selection bias and provide unbiased estimates of treatment effects (Imbens and Wooldridge, 2009). Using matching estimators, we estimate the average treatment effect of consistent pegs on the likelihood of currency crises. We find evidence that consistent pegs significantly decrease the likelihood of currency crises compared with other exchange rate policies. This result is robust to a wide variety of matching methods and different definitions of currency crises. The paper is organized as follows. Section 2 presents our research hypotheses and empirical methodology for matching methods. Section 3 presents data on currency crises and de jure and de facto exchange rate regimes. Section 4 presents estimations of the average treatment effect of consistent pegs on the risk of currency crises, using matching methods. Finally, Section 5 presents our conclusions

In this paper, we formally evaluate the treatment effect of CP on the occurrence of currency crises in 84 countries from 1980 to 1998. We carefully address the self-selection problem of regime choice that previous studies have not explicitly addressed. To address the self-selection problem, we employ the bias-corrected matching estimators. Using matching estimators, we estimate the average treatment effect of CP on the incidence of currency crises. The results of matching methods provide interesting policy implications for exchange rate polices. After controlling for self-selection bias, we find that consistent pegs significantly decrease the probability of currency crises compared with other exchange rate policies. Therefore, we statistically confirm that deviations of actual exchange rate regimes from announced regimes significantly affect the occurrence of currency crises. We can reasonably conclude that countries that consistently maintain announced pegged regimes can substantially avoid speculative attacks and currency crises because such policies enhance the credibility of their currencies and cause speculators to perceive a positive signal regarding governments' willingness to defend the exchange rate. Moreover, and more interestingly, we find that consistent pegs significantly decrease the probability of speculative attacks (both successful and unsuccessful ones) compared with a policy of FOAP. That is to say, in countries that adopt de facto pegs, when two countries have similar observed characteristics (e.g., similar economic conditions), and when one country adopts CP and the other does not (i.e., adopts the FOAP policy), the country with CP is substantially less prone to speculative attacks. Therefore, we can surmise that, although the strategy of “announcing a peg” may become target of speculative attacks, it is substantially effective in avoiding speculative attacks because this policy can cause speculators to recognize a positive signal of policymakers' willingness to defend the exchange rate peg. Finally, our results suggest that countries that actually adopt pegged regimes should announce their adoption of currency pegs if they truly wish to avoid speculative attacks.