An Interview with Chris Sims, 2011 Nobel Laureate

April 2, 2012

Nobel Prize winner Chris Sims explains the economic impact of monetary policy, compares European and U.S. policy models, and shares practical advice for budding economists.

Gary Tapp: Today we are delighted to welcome Dr. Christopher Sims who together with Thomas Sargent was awarded the 2011 Nobel Prize for Economic Science. Dr. Sims is currently the Harold B. Helms Professor of Economics and Banking at Princeton University. Dr. Sims, welcome to the Classroom Economist.

Christopher SimsSims: Thanks. Glad to be here.

Tapp: For high school economics teachers, how would you suggest that they introduce the basic idea of econometric modeling to their students?

Sims: The essence of econometric modeling is that it's statistics applied to economics. What econometric models do and what they're used for could be taught at the high school level. The idea is that there's lots of data available and new data coming in every month, and people who make economic policy need to find some way to make sense of that data and use it to guide their decisions. That means that you have to keep track of it, you need a spreadsheet or something at least, and you need some way to say, well, now that the exchange rate has gone up 10 percent, what does that mean for what we think is going to happen to inflation? That's what econometric models are for, to help policymakers react to the flow of data. I think it's also important... it's not impossible to teach statistics to high school students, and I think high school students should be getting more of an introduction to statistics and probability than is standard, and that should be coordinated if economics is being taught and statistics is being taught in the high school, you can be linking them.

Tapp: So, if I asked you to describe the main contribution of your work to the field of economic modeling and maybe relating back to the traditional model, how would you describe that?

Sims: I think that what the Noble Prize people were singling out was that my work helped sort out the dispute between the monetarists and Keynesians. They, in part by introducing new approaches to statistical modeling in the '60s and early '70s, monetarists were claiming that the main source of business cycle fluctuations was bad monetary policy. The monetary authority was making mistakes, making the growth rate of money vary a lot, and all those variations resulted in recessions and booms, and if only we could force the monetary authority to stop messing with the economy and just keep money growth steady, the business cycle would be greatly reduced or even vanish.

And then the Keynesians were saying that can't be true, but they didn't have statistical models in which they could each put forward their position and ask, well, what did the data say? There were lots of attempts to do that, but with very awkward statistical modeling.

Over the course of about 10 years, things that I did and other people followed up on managed to sort out what the effects of monetary policy changes are and distinguish those from co-movements in money and prices and income that didn't have anything to do with policy. There's now pretty much a consensus on how monetary policy affects the economy, and on what the size of that effect is. The general conclusion is that it accounts for maybe somewhere between zero and 20 or 25 percent of the fluctuations we see, but if you try to trace out historically, you can't blame any recession on monetary policy.

Tapp: One type of question that we get from teachers occasionally is, they see economic forecasts in the media sometimes and their impression is that they're not that accurate. So, the question is, what does the current research say about the forecasting accuracy of models, even the latest generation of models?

Sims: Well, you have to recognize that no model claims that its point forecast is going to be correct; these are probability models. If you see a forecast that says growth will be 2.3 percent next year, the model is not saying and if it's 2.2 percent I'll die, it's saying the most likely value is 2.3 percent, and if you really want to know what the model says you should ask more. How likely is it that it will be 2 percent or less, how likely is it that it will be 3 percent or more? Because a good model, the modern ones, answer those questions, the forecast should be thought of as telling us what the range of uncertainty is, and unfortunately, people often... when a model that does a good job of describing the real range of uncertainty is sometimes dismissed as too inaccurate. If I say, the best I can tell you is that the probability is 80 percent growth will be between 2 percent and 3 percent, that may be perfectly accurate. Somebody else comes along and says it will be 2.5 percent plus or minus 0.1 and sometimes people think, well, that guy's much more accurate than that guy that says 2 to 3 percent. That is, he seems much more sure of himself, but if you look historically, it may be that he is over and over again wrong. People who seem really sure of themselves are often people who over and over again make big mistakes.

Tapp: Some observers have suggested that the pace of financial innovation over the last 10 or 15 years or so exceeded the pace of development of risk management models and that might have contributed some to the crisis of 2008. More broadly, other observers have suggested that macroeconomic models themselves don't have enough of a financial component to them. Would you agree with either the narrow comment about risk management models or the broader model or both of them?

Sims: It's true that financial innovation has been rapid recently. There's a constant tension between regulation and financial innovation. What regulation does is try to control some kinds of financial contracts that create systemic problems, and if there's a need for control it must mean that there's an incentive for somebody to write those contracts and you have to create barriers to doing it. Well, the same incentives that lead to that contract in the beginning are going to lead to people who are going to try to think of ways around any regulation you propose, and that's never going to change. It's a hard problem to see how you avoid new systemic dangers coming up through too slow a speed of response by the regulators. Econometric models, it's true, have not had much of a financial sector in the past and they should have had more. If you looked back historically, you could have seen...we are now seeing when we look at past data, that if we'd had some financial variables in the models, we would have done a better job of tracking what was going on in 2007–2008 and going forward, every policy institution has people working on new models with bigger financial sectors. I'm a little afraid that we're following the usual tendency to fight the last war. We've had a big financial crisis, going forward it may not be financial crisis so much as public finances, debt and deficits, that are the important thing and those have been neglected in these models, too.

Tapp: The current situation in Europe, you have done some work in that area and spoken and written about possible lessons that the current crisis might have for the U.S. Could you just outline what you think some of those lessons might be at this point, given that the situation is still unfolding?

Sims: The sources of the European crisis is that they made some changes in their institutions that took them away from the U.S. model. In the U.S., we have our own currency, an exchange rate between that currency and the rest of the world, and strong fiscal central authority—the federal government that taxes and spends and issues debt. We have state governments also, but they do not have their own currencies, and they have balance budget requirements in their constitutions for the most part.

In Europe, they have a central bank that's like our Fed, but there is no corresponding fiscal authority. There is a European Union that does do a little bit of taxing and spending, but it's very tiny and it does not issue debt backed by a European-wide taxing commitment. It's that situation that's causing them their difficulties. They have all these countries, none of whom can issue their own currency, none of whom can devalue, and they don't have the network of fiscal cushions that we have.

In our country, if there's a housing crisis in Texas and Boston is doing well, which has happened at some points historically, without anyone thinking about it, tax money flows from Boston to Texas because we have an income tax. Incomes are down in Texas, they're up in Massachusetts, and there's a net flow of resources through the federal government that way.

There is no corresponding automatic cushion in Europe. So, these countries that are in economic difficulty like Greece and Spain and Portugal, they find that they don't have the option of borrowing with their own currency, they don't have the option of devaluing to make themselves more competitive, and yet at the same time there is no corresponding fiscal cushion the way there is in the U.S. So, these are all lessons in a way that tell us we ought to be thankful that our system is a lot better set up than theirs.

Tapp: For a high school teacher who has students that they think might be interested in going on further into economics, how would you suggest that they prepare themselves?

Sims: Taking math courses steadily is very important. One of the things that I've found a little bit worrisome in teaching college students is some of the best students have taken advanced placement calculus in high school and then they arrive in college and they stop taking math because they've met the requirements. And then in junior year, they decide they want to take an advanced economics course, and they haven't done any math for two years, and they get very rusty. I think it doesn't matter quite so much exactly what math you're taking, but you should keep at it to keep your skills up in high school and after high school. It's not like there's a certain amount of math you need to learn and then you can quit. Whatever level you're at, you should be keeping your skills sharp and trying to advance in math. So, I think that's the single most important thing. Of course, if all you do is take math courses that can be a creative gap, too. Economics is about the application of math to the real world, and you do have to take courses and be interested in other things as well; history and economics itself if it's taught in your school, and be interested in politics and how decisions are made in your country.

Tapp: I see that your undergraduate major was in mathematics. Can you talk about when you were first exposed to economics in school? What attracted you to the field, and what made you eventually decide to go in that direction for your graduate work?

Sims: I actually didn't have any exposure in school to economics at the time I was going through high school, there were very few high schools that taught any economics. I had an uncle, though, who was an economist who campaigned to get me to become an economist from about the time I was 10 years old. I knew what an economist does from my uncle; also, my grandfather was a labor economist. He was a member of the first of the National Labor Relations Board under President [Franklin D.] Roosevelt. So, there was talk about economic policy issues in family gatherings. My grandfather used to greet me from the time I was about eight years old with, "Well, Chris, what do you think of the present situation of the country?" So I knew about public policy issues and was familiar with discussing them from my family background. I had a pretty good idea of what economists... what kinds of issues economists deal with and why what they do might be interesting and important from a very early age.

I thought he [Sims's uncle] was having the opposite effect of what he intended, but when I got to my senior year and I was considering going on in math, I decided that I probably didn't want to spend my life with pure abstraction, and I knew what economics was all about because of my uncle. I also had an excellent math teacher in high school who went on to become the president of the math teachers association—I can't remember the name of the association [National Council of Teachers of Mathematics]—Stephen Willoughby, and he actually told my mother that I should become an economist, but he never told me [laughs].

Tapp: We've been talking with Dr. Christopher Sims, Nobel Prize winner in economics. Thank you so much for joining us.

Sims: Well, thanks for having me.