In my opinion there are four broad objectives for the emerging quantitative work on currency crises. The goals of this research program can be summarized in a series of questions:

1. What are the determinants of crises?

2. Can one predict crises or construct “early warning systems”?

3. Is there joint causality across countries – that is, contagion? What are the channels?

4. Is there joint causality between banking and currency crises?

In this fine chapter, Glick and Hutchison focus on the last issue, the problem of joint causality. This is an important issue, and the stakes are high for their work. A finding of joint causality between banking and currency crises has strong implications for our understanding of the causes of both, and, more importantly, for policy actions to prevent future crises. They also ask an important question which has thus far not been directly addressed in the literature, namely, “does country aggregation matter?” The extant literature disaggregates by time and sometimes the degree of capital mobility, but most papers use either OECD or emerging market data, but not both. Glick and Hutchison emerge from their extensive empirical analysis with two key conclusions. First, there is in fact joint causality; more specifically, banking crises tend to cause currency crises. Second, emerging markets are different. Both of their findings are plausible and sensible. This adds to the appeal of their chapter, though it makes the job of the discussant a bit more demanding.

One important methodological issue rears its head at the outset. Can one do an investigation of the fourth issue without taking a strong stand on the first three issues?