Hey, Economist! Outgoing New York Fed President Bill Dudley on FOMC Preparation and Thinking Like an Economist
Bill Dudley will soon turn over the keys to the vault—so to speak. But before his tenure ends after nine years as president of the New York Fed, Liberty Street Economics sought to capture his parting reflections on economic research, FOMC preparation, and leadership. Publications editor Trevor Delaney recently caught up with Dudley. This transcript has been lightly edited.
Q: Lawyers sometimes refer to ‘thinking like a lawyer.’ How would you describe what it means to ‘think like an economist?’
The classic joke is that economists are ‘on the one hand…on the other hand.’ Lawyers are advocates for a given point of view and prosecute under a well-defined set of rules of law. I think of lawyers as advocates and economists as almost like a justice balance—where they’re trying to weigh the evidence very carefully to see where the preponderance of the evidence lies.
So economists may be a little bit more open-minded to the facts—not to say that lawyers aren’t, but a lawyer’s job is to do something, to advocate a position, to protect a positon. So they’re starting with a very strong a priori view. I think economists start with a priori views, in the form of a hypothesis, but if the evidence is inconsistent then they start to change the theory and hypothesis, as opposed to arguing that the evidence is obviously not applicable.
Q: You were an economist on Wall Street before joining the Fed. What differences have you noticed between Wall Street economists and the research-oriented economists here on Liberty Street?
A big difference is that Wall Street economists cover a broad range of topics. When I was at Goldman Sachs we had four people covering the U.S. economy—they covered pretty much the gamut of all the policy issues that might have some economic content. Wall Street economists try to synthesize an abundance of information into material that’s useful for the person who’s trying to figure out the world that we live in. A research economist is trying to push the frontier of knowledge outward, so that’s a very different goal.
At the New York Fed, research economists play a hybrid role. They’re doing research that’s trying to push out that knowledge frontier. But they’re also taking all of their knowledge and analytical ability and applying it to real-world policy problems. In my mind, a good research department in a Federal Reserve Bank consists of people who are top-notch in terms of their academic qualifications and ability, and who are interested in real-world policy problems—which creates a tremendous value for the Bank.
Q: During your tenure, has the role of research changed a lot at the New York Fed? …the processes?
I don’t think that it’s changed dramatically. The research department was very good when I came here, it’s very good now. I think probably what’s happened—I don’t know if this is accurate, but it’s a perception on my part—it feels like the research that we do is a little bit more closely aligned with the mission of the Bank than it was ten or fifteen years ago. I don’t think it’s something that was deliberately forced. The work of the Bank has become so interesting in the aftermath of the financial crisis and we have such a wealth of information here compared to other places. I think those factors have caused the research department to want to work on issues that are very consistent with the broader goals of the Bank. The research department is probably providing a little more value to the Bank just because that alignment is a bit closer.
Q: ‘Big data’ is much discussed these days. Has it already had a significant impact on economic research or is that story still evolving?
I think big data is already making a big impact in the world—think about artificial intelligence, machine learning. In the Bank, we’re probably in the early days. I think the biggest data set that we use is the anonymized credit files that we get from Equifax [New York Fed Consumer Credit Panel]. That’s a big data set, but that’s ‘big data’ in a more traditional way, it just happens to be a lot of data. As opposed to big data in the sense that you’re synthesizing a whole bunch of different streams of data and trying to extract information out of it.
The credit analysis has been great though. To be able to take all that information from those credit files and figure out what it means for types and classes of individuals—based on their backgrounds, their locations, their incomes, their educations—has been really valuable. ...MUCH MORE
