Fed’s Keeper of Secret Teal Book Lifts Veil on New Forecast Era - Bloomberg

Fed’s Keeper of Secret Teal Book Lifts Veil on New Forecast Era - Bloomberg

 

Fed’s Keeper of Secret Teal Book Lifts Veil on New Forecast Era

By 

Craig Torres

March 11, 2021, 10:00 AM GMT

 

 

Stacey Tevlin is the most important person in U.S. economics that you have probably never heard of. She leads a team of 357 people in the Federal Reserve’s Research and Statistics division, which is entrusted with the forecasts for policy makers as they weigh interest rates every six weeks.

 

Her work appears in a document called the Teal Book, which is kept confidential for five years. She doesn’t go on television, give speeches, or even talk to reporters very often. Yet her team’s work has so much influence on the policy debate that one former Fed governor calls the staff forecast “the 13th member” of the Federal Open Market Committee.

 

Bloomberg News spoke to her in February. The interview has been edited for brevity and clarity.

 

1. Where did you grow up, what did your parents do, and how did that shape your world view?

Where I grew up definitely shaped my world view. I grew up outside of Detroit in the 1970s and 1980s. I lived through some of the big recessions that hit Detroit hard. That was a presence as I was growing up. People were losing their jobs and moving out of town. My parents’ jobs weren’t affected by that. My dad was a history teacher and my mom taught English, but we lived in a neighborhood which was mostly auto workers and their families.

 

Tevlin, 52, earned a Ph.D. in economics in 1995 from the Massachusetts Institute of Technology and a B.A. in economics from Northwestern University in 1990. She wrote her dissertation on how successful firms distribute profits. Her adviser was Robert Solow, winner of the Nobel Prize in Economics.

 

When I went to college, I thought I would be an engineer because I really liked math. Then I took Bob Gordon’s macro course in my freshman year and recognized it was about those gyrations we had lived through. That really grabbed me on a personal level, and that’s part of the reason I decided to become an economist.

 

2. How did you forecast the virus and the economy in the days before the vaccine and how do you do it now?

We forecast both the course of the virus and vaccine availability because we have to. They are basic assumptions that underlie our understanding of the economy.

 

A year ago I wouldn’t have had any idea how to do that. We don’t have any epidemiologists on the staff. What we do have are a lot of people who understand complicated models and data really well.

 

A number of people on our staff have invested a significant amount of time learning both the math and the science of pandemics, mitigation strategies, and vaccines so we could build an understanding of how they are likely to evolve and affect the economy.

 

At the very beginning, we developed a whole new set of daily and weekly indicators to track the economy. We started a reading group where we compiled all the Covid-related economic research -- I think it was 200 papers last I checked -- making sure we are on top of the research. We got up to speed pretty quickly on this.

 

3. The Fed is experimenting with novel data sets that are closer to real-time indicators. Which gave you an edge?

Having timely data made a huge difference in our understanding. More than what we had in hand, it was that we had built up knowledge in how you go about acquiring, storing, and transforming these alternative data sets.

 

A lot of the things we wanted to know about changed, so while we did still use the high-frequency data that we had been developing for six years, we also pivoted very quickly to find what we needed.

 

We had a team of entrepreneurial data enthusiasts already in place who were willing to go out and uncover great new data sources before I would have thought to ask for them.

 

4. The FOMC has a new strategy focused on shortfalls from full employment. How has that changed your approach?

When we put together a projection, we need to write down a path for monetary policy that yields a sensible forecast. Using a monetary policy rule is a straightforward and transparent way to do that, even though the FOMC doesn’t use a rule for policy.

 

Last year, we made some changes to our rule to reflect those updates. For instance, we made some tweaks to incorporate and acknowledge what the FOMC said about achieving average inflation of 2% over time, and focusing on shortfalls from maximum employment.

 

They didn’t specify a rule, of course, so we are not going to capture exactly what they are going to do. To help them in their deliberations, we also present different simulations for the economy using different rules.

 

5. The board has few minority economists. Is that something you are trying to change?

Absolutely. Almost every year for the past 10 or so that I can remember we have made some changes in our recruiting process to try and improve how we are doing it -- working on finding ways to overcome implicit bias, making sure we have the best practices.

 

We also spend a lot of time and effort to make sure that we are helping to build a pipeline of new economists. In the last couple of years we have also really been making an effort to address the inclusion piece to make sure once you get diverse people at the board that they feel welcome.

 

I don’t want to make it sound all rosy. We are working hard on diversifying our workforce, but we are not where we would like to be.

 

6. How do you forecast the impact of unprecedented U.S. fiscal stimulus?

We are always having to write down some assumption of where fiscal policy is going to end up. We often have a pretty reasonable guess, but we don’t always.

 

Then it gets harder, because we need to estimate how much of the fiscal outlays actually get spent and when. We base that first on the economic research. But then we also augment those studies to tailor them to the current situation.

 

7. What’s the uncertainty around that?

It’s huge. There is definitely a lot of judgment we are imposing on this. We don’t think people will react exactly as they have in the past. We have a lot of debates on this as a staff.

 

We also run multiple scenarios. We know we don’t have all the right answers but we look at different versions and show different scenarios to policy makers.

 

We don’t know how people are going to behave. There is a little bit of introspection and there is a lot of looking at survey questions as we try and glean things, such as when people are going to be confident to go out and start doing stuff.

 

We can try to build on how they have behaved so far in the pandemic or across countries. We put all that together -- data, surveys, cross-sectional analysis, and model estimates -- and honestly we look at it and ask ourselves if the forecast seems at all reasonable. I wish we had a better crystal ball.

 

Even if we get the health variables right in a forecast, it is still an open question if we have the behavioral response right.

 

9. It’s been a long year. What do you do for a boost to get through the day?

If I can get out of the house, I would go ride my bike. That would be a boost for me. But if I can’t because I am sitting in a meeting, and this is kind of embarrassing to say, I am a big fan of chocolate milk and peanut-butter crackers.

 

10. The pillars of the economy are productivity, population growth, technological change and diffusion. What do they look like post pandemic?

We don’t know. But that is definitely something we are thinking about: What do we want to focus on in the post-pandemic world? It will likely be all of those pillars: What is the labor force going to look like, are people who have left going to stay out permanently, are they going to be drawn back in?

 

Definitely productivity is an important question since we are using technology so differently. How is that going to impact business models? We also have questions about market structure and pricing. There are so many interesting questions, but we don’t know the answers yet.