It’s what they do, not what they say: How to infer the stabilization objectives of a central bank
I first visited the Graduate Institute of International Studies in 1982. Alexander Swoboda and Hans Genberg had been awarded a Swiss National Science Foundation grant to study the effect on Switzerland of foreign economic shocks. Alexander and Hans planned to undertake a statistical analysis using Vector Autoregressions and needed a project overseer. They had originally recruited Richard Meese but, late in the Spring, Meese discovered that he could not come. It thus happened that I received a “cold call” one April evening asking me to join the research team. My immediate reaction was that it was too late to ask for a leave. Teaching assignments had been set and the market for visitors concluded. But, as I told Hans, the opportunity was too good for me to assume that a leave could not be arranged. And so, expecting the worst, I made an appointment to see my department chair. Sometimes the gods smile. My department chair agreed that the opportunity was too good to pass up. And so, family in tow, I arrived in Geneva the following August to begin a personal and academic adventure that has provided rich opportunities and a lot of fun from then until today. The project went well. Alexander, Hans, and I were able to publish its findings (Genberg, Salemi and Swoboda, 1983 and 1987). Good things led to others and within a couple of years we began to study Swiss monetary policy. In particular, we were interested in whether a formal analysis would confirm conventional wisdom that the Swiss National Bank (SNB) was an “inflation hawk.” The claim was not obviously true since estimates of the SNB reaction function showed that it responded to many different variables including the exchange rate. The SNB project was my first attempt to use inverse control and began a research program that I have been working on ever since. It is that program, which owes its beginning to Alexander, Hans, and HEI, that is the subject of this chapter. Ironically, as I will explain later, the program has cycled back to once again consider Swiss monetary policy. I first offer a non-technical explanation of inverse control. I describe decisions that a researcher must make to undertake an inverse control analysis. I explain how several coauthors and I have used inverse control to study central bank policy in a variety of settings. Finally, I connect inverse control to some of Alexander’s published ideas.