Magazine April 6, 2020, Issue

Can Well-Designed Experiments Make the Case for Government Intervention?

Esther Duflo and Abhijit V. Banerjee following their Nobel lectures at Stockholm University in Sweden, December 8, 2019 (TT News Agency/Christine Olsson/Reuters)
Good Economics for Hard Times, by Abhijit V. Banerjee and Esther Duflo (PublicAffairs, 432 pp., $30)

As a political independent, I had the privilege of reading Abhijit Banerjee and Esther Duflo’s Good Economics for Hard Times without assuming a defensive crouch. Their sometimes surprising, sometimes inconclusive attempts to measure the effects of popular policy ideas will frustrate anyone looking for a five-point plan. But for someone mistrustful of easy answers, it was a treat to see so many buzzword solutions subjected to experimental testing.

The authors, a married couple, are both economics professors at MIT. While it seems fair to conclude from their proposals that they are liberal, they cannot reasonably be called bleeding hearts. They won the 2019 Nobel Prize in Economics, “for their experimental approach to alleviating global poverty,” in recognition of their decades-long campaign to test the outcomes of new policy proposals in the same way as new drugs, through the use of randomized controlled trials (RCTs). These studies are the gold standard of proof in the field of medicine, and they sound a lot like the scientific method you learned in school: Take a group of equivalent subjects, act on some, leave the rest alone, and watch whether their destinies diverge or not.

This approach’s most famous success so far might be in directing foreign aid to fight malaria. When economists weren’t sure whether to give away insecticidal bed nets for free or sell them at a low price (to promote their use for safe sleeping rather than fishing, and to avoid crippling any local markets), a series of RCTs in Africa showed that free nets were used just as wisely as bought nets and that the free nets reached more people than the bought nets did. A consensus was reached, the naysayers caved, and a range of organizations began giving away bed nets, preventing hundreds of millions of deaths over the last 20 years. For Banerjee and Duflo, “good economics” is informed by well-controlled experiments like these. “Bad economics” is driven by intuition and anecdotes.

That evidence-based mindset can appeal to conservatives. During the Nixon administration, it was Donald Rumsfeld who pushed for some of the first economic experiments of this kind in the U.S. to test a proposed negative income tax (NIT) instead of rolling it out directly. In a 1987 radio address on welfare reform, Ronald Reagan called for “experimental changes, through carefully tested and evaluated demonstrations.” National Review writers have cited Banerjee and Duflo approvingly in the past, with Jim Manzi, for instance, expressing the hope that their work would make social science “more accurate and more hospitable to limited government and political liberty.”

These hopes may be disappointed. Although the authors offer frequent caveats — such as “Economists often get things wrong. We will no doubt do so many times in this book” — they end up calling for “a much-expanded role for government” as “the only possible way out” of our present predicaments. That assurance is interesting, because Banerjee and Duflo’s only previous book for a general audience, Poor Economics — which won the Financial Times and Goldman Sachs Business Book of the Year Award in 2011 — did not take a strong position on the proper roles of the public and private sectors. It presented the results of 15 years’ worth of RCTs in the developing world in a way that would be helpful for anyone with money to throw at a humanitarian crisis, no matter where they worked. In their new book, Banerjee and Duflo focus on America, which is not their primary research domain, and on the kinds of policy choices open only to governments, such as setting tax rates and immigration quotas. This brings them into bloodier oceans of debate, with fewer of their own bespoke studies and more extrapolation from the best available parallels in other countries or at other points in history.

Will conservatives like the results? The answer may depend on whether their wariness toward welfare spending comes more from the premise that government should not do such things (i.e., “Socialism is theft”) or from the conclusion that it hasn’t appeared to do them very well (i.e., “Socialism always fails”). Some readers will reject the authors’ assumption that governments should have goals such as “restoring dignity” or “reducing anxiety.” Others might take heart from Banerjee and Duflo’s criticism of “old-fashioned government handouts” and their call for “a wholesale rethinking of the social policy apparatus.” The evidence they present shows that “nothing terrible happens when one gives cash to the poor” — cash transfers don’t lead to family breakdown, less work, or more spending on alcohol or tobacco — but they don’t lead to lasting rises in household income, either. Banerjee and Duflo are also deeply concerned about the effects that handouts have on people’s self-respect. One of the more humorous experiments in the book was conducted on customers at an H&R Block in California who were eligible for SNAP benefits but had not registered for them. They became much more receptive to the program when a brochure billed it to them as a “Golden State Advantage Card,” allowing them to “get more at the grocery shop,” in place of the traditional language about welfare and food stamps.

What exactly should we be doing instead? The authors’ big-picture goal is to “break the impasse” between Right and Left with “careful program design,” persuading people that welfare is worth raising taxes for, if it has proven results. One of their most confident recommendations, repeated several times, is that government should help workers move out of the areas hurt worst by free trade. They propose to do this by massively expanding the Trade Adjustment Assistance program (TAA) on the same scale as the G.I. Bill, citing a 2018 study that found that displaced workers who retrained under its auspices earned $50,000 more over the next ten years than their fellows who did not. They also hold out hope that expanding TAA would reduce disability payments, which 9.9 percent of workers who lost jobs to trade have now claimed, according to data in the American Economic Review. Perhaps the most unusual suggestion in the book, which the authors themselves describe as “somewhat radical,” is paying firms hurt by trade to retain their employees over the age of 55 who can’t or won’t move, even if the subsidy payments must exceed the employees’ salaries. “It will be very difficult to help people in these places,” the authors admit, suggesting that the key point is to prevent more people from becoming stuck in the same way.

They are leery of a federal jobs guarantee in the style of the Green New Deal, questioning the utility of the jobs, which would come from private companies bidding on government contracts, and drawing comparisons to India’s National Rural Employment Guarantee Act (NREGA) of 2005, which has so far failed to provide jobs for everybody who wants one. However, they are game to offset the effects of both trade and automation with government funding for “labor-intensive public services” such as elder care and child care, which would likely lead to more jobs in these areas, without any fixed quotas. To illustrate the value that more workers in these fields could bring, they point to a 2011 RCT from Tennessee, which found that kindergartners in smaller classes not only scored higher on standardized tests but were more likely to attend college, get married, own homes, live in better neighborhoods, and save for retirement.

Still, the heart of Good Economics for Hard Times is in the method, not the message. The best-guess recommendations the authors land on at the end of each chapter don’t appear to match up with the platform of any particular politician, but that doesn’t seem to concern them. Their goal is to ground both sides of our national debate in hard evidence, and the process of coming up with ways to do that is pretty interesting in itself. Some chapters are baldly inconclusive. The book might feel encyclopedic at times, with reams of examples from different places and eras, but it could not be re-shelved in the philosophy section. Eager to distance themselves from the previous generation of economists who argued from first principles, Banerjee and Duflo say nothing that smells like special pleading. It would be hard to take umbrage with such studied humility. The authors admit, “We clearly don’t have all the solutions, and suspect no one else does either.” Even so, the prospect of a path towards consensus solutions through iterated experiments is enough to make for a compelling read.

This article appears as “Putting Government Intervention to the Test” in the April 6, 2020, print edition of National Review.

Something to Consider

If you enjoyed this article, we have a proposition for you: Join NRPLUS. Members get all of our content (including the magazine), no paywalls or content meters, an advertising-minimal experience, and unique access to our writers and editors (conference calls, social-media groups, etc.). And importantly, NRPLUS members help keep NR going. Consider it?

If you enjoyed this article, and were stimulated by its contents, we have a proposition for you: Join NRPLUS.

LEARN MORE
Laura Ball is a science journalist, a former Thiel Fellow, and a research assistant at Mila, the Quebec Artificial Intelligence Institute.

In This Issue

Articles

Features

Books, Arts & Manners

Sections

Letters

Letters

A reader responds to Madeleine Kearns's article, ‘Pornography Is a Public-Health Problem.’

Most Popular

Politics & Policy

The Suicide of the Cities

I noted on Twitter last night that rioting, coming right after the virus, is a catastrophe for the cities. Regardless of whether Trump or Biden is elected in November, it's easy to envision the following happening: Americans will flee the cities as they did in the post-1968 era. Thirty years of great progress for ... Read More
Politics & Policy

The Suicide of the Cities

I noted on Twitter last night that rioting, coming right after the virus, is a catastrophe for the cities. Regardless of whether Trump or Biden is elected in November, it's easy to envision the following happening: Americans will flee the cities as they did in the post-1968 era. Thirty years of great progress for ... Read More
Law & the Courts

‘Dominating’ the Streets

Since the revolution in policing that began in the early 1990s, we have had a generation of peace and prosperity. Without the rule of law -- i.e., without order, without the presumption that the laws will be enforced -- that kind of societal flourishing is not possible. We are seeing now what happens when the ... Read More
Law & the Courts

‘Dominating’ the Streets

Since the revolution in policing that began in the early 1990s, we have had a generation of peace and prosperity. Without the rule of law -- i.e., without order, without the presumption that the laws will be enforced -- that kind of societal flourishing is not possible. We are seeing now what happens when the ... Read More