This from an interview with Allan Meltzer published in the current issue of the Minneapolis Fed’s quarterly:
“I think Chairman Greenspan does not have some overriding view of the world that he tries to impose on the data. Quite the opposite. He looks at the data, and he tries to figure out what is happening in the world and doesn’t have any very precise theory. ... Chairman [William McChesney] Martin and most other officials didn’t have any idea about how what they did today influenced either money or output over any long period of time, and they were never very interested in having such a view. Martin was perfectly happy to operate on his judgment about events and had little confidence in economic analysis. But those are some of the best periods in Fed history.”
Also, this gem:
“[C]redibility is a stock that can be used to surprise markets.”
In other words, telling the truth sets the stage for effectively telling a whopper. The quotation is a good one for explaining the so-called time-inconsistency problem. In fairness to Meltzer, he prefaced this Meltzer-Cukierman observation with the word”Unfortunately,...” And he expressed disappointment about the Fed Chairmen having no theory.
Now, let me see, how does this ponder-the-data approach affect our underestanding of rational expectations? People know, or behave as if they know, what the Fed Chairman can glean, or behave as he can glean, from the data. Something like that.