Alan Kruger on the Best Recipe for Policy

Alan B. Krueger, presently the Bendheim Professor of Economics and Public Policy at Princeton and formerly chairman of President Barack Obama’s Council of Economic Advisers, recently received the 2017 Daniel Patrick Moynihan Prize from the American Academy of Political and Social Science (AAPSS).

Krueger, who was inducted as an AAPSS Fellow in 2003, is a prolific researcher, writer, and columnist, and he has been a tireless exponent of rigorous investigation of the most challenging social phenomena that need to be addressed in public policy. He chaired President Barack Obama’s Council of Economic Advisers from November 2011 to August 2013. He also served as assistant secretary for economic policy and chief economist of the Department of the Treasury in 2009–10, and as chief economist at the Department of Labor in 1994–95.

“In these political times,” said AAPSS President Ken Prewitt, “the AAPSS doubles-down on Moynihan’s legacy, electing fellows and awarding the Moynihan Prize to those whose engagement contributes in equal measure to better science and more sound policy.”

Social Science Space sat down with Krueger shortly before he delivered his Moynihan lecture on May 18 in Washington, D.C. The interview has been edited for length and clarity.

Could you give some hints or some tips to someone who wants to do true evidence-based economic policy in the current environment? So maybe you could compare and contrast from what we might have known 150 days ago, what we might have expected and what we can do now?

That’s a tough question. The current environment is not the most hospitable for evidence-based research. On the other hand, I think it’s more important now than ever, and I think the environment likely will change in the future. And there tends to be a pendulum.

I can tell you from my experience, when I worked for President Obama, on virtually every issue that we addressed, we started by thinking about what do we know, what’s the evidence. I went through over 60 issues that I worked on, 60 economic policies, from TARP [Troubled Asset Relief Program], to small business lending to long-term unemployment to the strategic petroleum reserve, a wide range of issues. And of those 60 on the vast majority, we started with evidence. When I say vast majority, 95 percent.

Then I said to myself, ‘Well, what was the quality of the evidence?’ And I’d say, ‘Half the time the evidence was quite germane.’

Sometimes the policy was moving so quickly, the economy was collapsing so fast during the financial crisis, you couldn’t say, ‘Well, do we have randomized trials on bank bailout?’ But [even in those cases] we had experience from some other countries which we could grab. And then I said to myself, ‘What fraction of the time is it what I consider the gold standard, where we had an experiment on job training and experiment on disability insurance, and in about 10 percent of the cases — and probably 30 percent of the cases — it would have been feasible in principle to develop methods.

So I think policy is still an area where we can do much better in terms of providing relevant high quality compelling evidence to policymakers.

Do people still want to hear it?

Well, ‘people’ is big definition; it’s a big group. Given the current climate, to be perfectly honest, I’ve talked to some former students who are working in the current administration, and they reached out to talk about evidence and talk about evidence-based policy. So in spite of the fact that the president of the United States seems very proud of his limited attention to research evidence, and shows little respect for firm findings from science, I think it’s still the case that the large set of dedicated civil servants are interested in these issues (and even some of the people he brought along with him).

Do you have any concerns about data collection?

I have many concerns about data collection here. So yes. The first concern is I don’t think our country has done enough to invest in our data, statistical infrastructure, the Bureau of Labor Statistics, the Census Bureau, the Bureau of Economic Analysis, really our crown jewels. If you go back historically, we lead the world. [The United States] had the first GDP accountants, we had the first labor force survey and our methods were copied around the world. That’s one reason why we can compare how we’re doing and study other countries. But the economy has changed so dramatically over the last 30 years, we have not made the investments necessary to keep up with the changes. So one of my concerns is simply that we’re not investing enough given all of these important decisions that depend on accurate data — not only by the government, but also by the private sector, by businesses, by households, by foundations, and so on.

So that’s one concern. The second concern is the fact that some of our core institutions, the Bureau of Labor Statistics is one of them, have become political footballs. And I think that’s very unfortunate. I think democracy requires a certain degree of trust and respect for key institutions, such as the institutions that provide objective, impartial — not perfect, but the best that they can provide — measurements of how the economy is doing.

Apart from say redistricting, why would a policymaker, a legislator, want to defenestrate data collection?

I think there are several reasons. One is benign neglect. There aren’t too many groups looking out for the statistical agencies. The BLS in particular, is in a very difficult situation in terms of its budget. It’s put together with the Labor Department, it’s fighting with HHS for resources, and other than a small group of economists in the government and a small group of statisticians, economists, related social sciences outside of the government, there’s not too many constituents who are fighting for the importance of economic data. So some of it is benign neglect.

Some of it, I think, comes from partisan lenses that are used to interpret how the economy is performing. I’ll tell you a story that when I worked for President Obama, the unemployment rate fell from 8.1 to 7.8 percent in September of 2012. That was announced the first Friday of October 2012, and [former GE CEO] Jack Welch immediately accused the Obama people of pressuring the Bureau of Labor Statistics, manipulating the data, which if he had any idea how the statistics are actually produced was totally beyond the pale.

The only political appointee at the BLS is the commissioner, and she’s presented with the data. It’s really inconceivable and impossible, I think, for federal officials to manipulate the data, particularly the household survey data. The GDP numbers have some subjective judgments and one could imagine there that may be cases where in other countries governments pressuring civil servants to adjust their data. But in the U.S. we have many safeguards against that.

What I worry about in the long term, given the lack of a natural constituency for paying for government statistics, given the tightness of the government budget, and then given the fact that some very prominent, respected individuals have tried to score political points with the government statistical agencies, that will erode support for them even further.

You’re still an active researcher. How do you make different data sets, ugly data sets, procedural datasets, some stuff that’s kind of mostly qualitative and some that’s quantitative, make it all play nice?.

Well, what I try to do my work is draw a collage. That’s what I learned from my teachers. Richard Freeman at Harvard, in particular, was a real master at drawing from various kinds of evidence. And I think if you look back at his record, very few instances where he was wrong, and I think the secret to his success has been his ability to weave together and look for evidence and lots of places and make sure it’s coherent.

I think policy is a little bit like a stew, and it takes lots of ingredients. My experience has been that rarely is there one discovery by somebody in his or her office that leads the policy change. It takes a whole army, it takes a whole village, and there are policies where I could say I had a stamp on this, or even I could say, some of them originated with my idea, but there’s no way I could have gotten Congress to pass it alone. And there’s no way I could have implemented it. And there were changes to it.

It’s a little bit difficult for economists to come to the realization that there are other fields that have views about policy and they’re actually valid in some cases. One of the things I liked about working in the government was working as part of the team where not only economists but lawyers and scientists, in some cases, media people, everybody kind of worked on this stew. And I think that improved the end product.