BOJ-IMES special trialogue
Research at Central Banks

Wakatabe Next, I would like to ask you about how research at central banks should look like. In other words, how research at central banks can best contribute to promote the banks' mandates, the design of monetary and financial policy?

Orphanides From time to time, we have fads in academia that are sometimes based on pure academic research motives. They are very important but basic research is not always ready to be applied before it is properly filtered. The rational expectations revolution [of Robert Lucas] mentioned by Carl is a very good example of this.

This revolution took a couple of decades to be filtered slowly into policymaking. It's important for central banks to have research staff so that they can distill what is most important from academic research and translate it to what can be used for policy purposes.

But at the same time, the policymakers must be very careful not to fall into the trap of fads and be overly influenced by academics where the theories and the understanding are not tested.

Walsh Let me emphasize what Athanasios has just said, that is, a critical role of economists within central banks is translating academic research and presenting it to policymakers.

More particularly, their essential role is distilling what is useful from academic research and that is likely to offer real insights into policy questions. That's really an important role.

If it's lacking, then the policymakers within the central bank are not getting all the potential insights academic research can provide. This is true even for policymakers who have strong academic records and experience themselves, because it's tough to keep up on everything if you are a busy policymaker. So, you still need staff economists to look at what are the emerging trends in academia and to highlight those that really offer insights that policymakers need to think about.

Wakatabe I also think that translation or acting as intermediary between the academic community and the policymaking side is an extremely important role of central bank economists. So, I'm glad that we can agree upon this.

Orphanides True, but let me add one more thing. I used to work as an economist at the Fed and eventually moved to the academia. In my experience, watching things from inside and outside central banks is very useful.

From my past experience, very often the gaps in theoretical models become apparent first inside policy institutions and only later with a lag outside. So, researchers inside the central bank, if they are given sufficient time for research activity to develop those ideas, have an advantage in being at the forefront of policy research.

Actually, I myself had the luxury of it at the Fed. One of the most cited papers I have on monetary policy is about the Taylor rule.[4]

The idea of this paper was born during an internal briefing at the Fed. At the time, people in academic circles thought that Fed policy could be explained with the Taylor rule. Instead, using data actually available when policy decisions were made, I was showing that the Taylor rule was not really useful for explaining policy. Top Fed officials were very interested in these findings, which prompted additional research. Eventually it became a paper.[5]

This may be a good example of the importance of interaction between academics and central bank research community as well as how central bank economists can best contribute to policymaking and research.

<Next page: Academics and central bank economists have different incentives>

Wakatabe That is very intriguing, thank you. So, we agree that we do need central bank economists who are aware of the fads and the pitfalls of new academic research.

My next question is what would be the key considerations to foster good central bank economists who have a balanced understanding of new developments in academic research and their limitations? And also how should the interaction between central banks and the academic community look like for better policymaking?

Orphanides I would like to first point out that policymakers need to be cognizant that the incentives in academia and elsewhere are very different. Researchers at a university want to do research that will be valued by their peers and be published in academic journals. And this kind of research will not necessarily be practical for policy purposes.

On the other hand, it's not right to adopt the same quantitative criteria of publication in the academia for researchers inside policy institutions.

In my view, the most valuable research inside the policy institution is the one that has an impact on policy, and a lot of that research in a good policy institution is only known inside the institution.

Another big difference is the degree of simplification of a model. When researchers in academia focus on a new idea, they are trained to simplify the models they use. In the process of simplification, the academics may throw out elements that are absolutely essential for policy analysis inside the central bank.

Walsh Following up on that, it's true that academics tend to simplify away things except the core issue one wants to address to really get strong insights and implications on that particular issue.

Given this, it's very important to have somebody who can say to the policymakers, "Yes, there is this literature and that result. But to get that, these are the important assumptions they make," because those assumptions don't usually hold in practice.

Speaking of the incentives for academics, it's hard to publish an article on, let's say, the optimal monetary response to some event that nobody thinks is ever going to happen. For people to be interested in your research, you need to pick topics that people are interested in. And so it's going to be inevitable that the research community is not going to focus on what turns out to be the new event that surprises everyone.

For example, academics would not pick up unexpected issues such as the global financial crisis, COVID-19, or recent surge of inflation, which caught policymakers by surprise as well.

If you think about the period leading up to the global financial crisis, the standard academic models had essentially no financial sector. There were a few exceptions, but not many.

We had foundational work by people like George Akerlof who won the Nobel Prize on asymmetric information and how it can lead to fragility, particularly in financial markets. And in 2022, Ben Bernanke, Doug Diamond and Phil Dybvig won it for their work on financial frictions. [6] This work had important impacts on financial regulatory policy, but not as much on monetary policy until the global financial crises, because it hadn't really addressed what the consequences of these frictions for monetary policy might be.

Orphanides Yes, Carl's example is a good one for our point. The standard macroeconomic models before the global financial crisis were not very useful for understanding the crisis, because for simplicity, some features that proved important during the crisis had been eliminated.

So, economists inside the central bank needed to resurrect classical studies that take into account practical aspects relevant for the central bank, and integrate them with the new studies and methodologies to make them practicable.

In addition, in my opinion, it's very, very important for successful policy to integrate practice and theory from different fields.

For example, it's very important to have people who understand markets, because central banks operate in the markets. And it's very important to have both practitioners and academics to do this right.

Walsh Since the global financial crisis, we have seen the huge shift of resources within the academic and the central bank communities to start thinking much more deeply about financial frictions and their influence on the economy and financial markets.

These contributions are helpful for policymakers going forward in distilling which frictions create problems, and how we should address them through regulatory policy and monetary policy. I think this is another area where having that conversation between academic and central bank researchers is important.

Wakatabe Thank you very much for your fruitful discussions on the interactions between academic and policy communities. Occasionally, scholars make a distinction between "professional or disciplinary knowledge," knowledge by and for academics, and "practical knowledge," knowledge created and exercised by practitioners.[7] In central banking world, both are more closely related to each other.

<Continue to Series 2: "Art versus Science" in the Conduct of Monetary Policy (Coming Soon) >