How bad are twins? output costs of currency and banking crises

We investigate the output effects of severe banking and currency crises in emerging markets, focusing on whether twin crises (simultaneous occurrence of currency and banking crises) exist as a unique phenomenon and whether they entail especially large losses. Recent literature, mostly relating to the East Asian crisis, emphasizes the interplay and reinforcement between currency and banking crises, presumably making twin crises particularly damaging to the real economy. Using a panel data set over the 197597 period and covering 24 emerging-market economies, we find that twin crises do not contribute any additional (marginal) negative impact on output growth. That is, twin crises do not adversely impact output over and above the independent effects associated with a currency and banking crisis taken together. We find that currency (banking) crises are very damaging, reducing output by about 58 (810) percent over a two-four year period. The cumulative output loss of both types of crises occurring at the same time is therefore very large, around 1318 percent, and should alarm policymakers. However, twin crises are bad only in that they entail output losses associated with both currency and banking crises, not because there are additional feedback or interactive effects further damaging the economy. This result is robust to alternative model specifications, lag structures and using IV and GMM estimation procedures that correct bias associated with simultaneity and estimation of dynamic panel models with country-specific effects.