Efficiency is out and political economy is in. But what does that imply about making good policy?
When Barack Obama was asked about carbon pricing in December 2015, at the Paris Climate Accords, he gave an answer that would please economists:
“I have long believed that the most elegant way to drive innovation and to reduce carbon emissions is to put a price on it. This is a classic market failure. If you open up an Econ101 textbook, it will say the market is very good about determining prices and allocating capital towards its most productive use — except there are certain externalities, there are certain things that the market just doesn’t count, it doesn’t price, at least not on its own… If you put a price on it, then the entire market would respond. And the best investments and the smartest technologies would begin scrubbing effectively our entire economy.”
Though a cap-and-trade bill passed the US House of Representatives during Obama’s first year in office, it never became law—and carbon pricing eluded him throughout two terms in office. It would fall to his vice president and successor, Joe Biden, to pass major climate legislation. And Biden got it done by ignoring economists’ preferred approach. Instead, his administration turned to industrial policy.
For the past two months, ProMarket has been publishing a series of articles about industrial policy, motivated in part by the Biden administration’s approach. We’ve asked, among other things: Why now? Why has industrial policy made such a comeback among policymakers and economists in recent years?
There are several answers to that question, but one has stuck with me—especially as a way of explaining the shift in economic policymaking from Obama to Biden. The resurgence of industrial policy around the world is driven in part by “a recognition that we are living in a second-best world of imperfection and imperfect markets,” said Chiara Criscuolo, a researcher at the OECD, during a recent Stigler Center event. “When we live in a second-best world, having an imperfect government intervention might still be welfare-enhancing.”
That comment is the key to understanding the resurgence of industrial policy, at least in the US, and it captures something critical about contemporary debates in economic policy. The Obama years were full of arguments about which policy instruments were optimal—from a public option in health care, to carbon pricing for the climate, to the best way to bail out a bank. Now, in the wake of the Trump presidency and the pandemic, the Biden administration is keenly aware that policies that seem optimal when considered on their own might in practice not be optimal at all. Concerns about efficiency have taken a backseat while concerns over political economy have grown.
The theory of the second-best
The “second-best theorem” that Criscuolo alluded to was formulated in 1956 by Richard Lipsey and Kevin Lancaster. The first fundamental theorem of welfare economics proves that under certain assumptions, competitive markets with perfect information can maximize the size of the economic pie. Lipsey and Lancaster showed that things can get complicated when the world deviates from this theoretical ideal in multiple ways at once.
The second-best theory, as explained in a paper by economists Lori Bennear and Robert Stavins, posits that:
“If there are multiple constraints that prevent the attainment of multiple Pareto optimal conditions, the elimination of only one of the constraints does not necessarily lead to a welfare improvement. Second-best theory suggests that the elimination of one market failure, in a world with many market failures, may not be welfare enhancing.”
In his piece on green industrial policy for ProMarket, John Van Reenen of LSE laid out the many market failures involved in climate policy, from the carbon externality to innovation externalities to free-riding between countries. The implication of second-best theory is that solving just one of these market failures is not necessarily an improvement if you can’t solve the others. And that provides a theoretical case—or maybe more accurately a theoretical reprieve—for the more interventionist, less market-oriented approach that Biden has taken.
The revenge of political economy
Of course, the Obama team was well aware of all these market failures; they knew they were operating in a second-best world. What’s changed is the level of appreciation the Biden policy team has for concerns around political economy. As Bennear and Stavins note, second-best theory applies to constraints imposed by politics, too.
Jared Bernstein, a longtime Biden economic advisor and the president’s current nominee to chair the Council of Economic Advisors, explained this clearly at an event at Brookings in February. The discussion was about the state of the US economy and a theme among the panelists was the need for humility:
“There’s another level of humility that I’d like to suggest and that’s political-economy humility. There are a lot of things that… mainstream economists… see in politics these days—or politics forever—that they don’t like… This stimulus was too large, this ‘Buy America’ thing is against what we’ve learned, and all that. I’m trained in these economics and I understand where that is coming from—I really do. And in many cases I share the critique, and I understand the rent-seek[ing] and all that. But there’s a level of humility that I’d encourage us all to think about which says: We’re trying to do what we believe… is really great, important policy on behalf of strengthening workers. And sometimes that’s not necessarily all kosher economics; sometimes it’s political economics. Sometimes you have to do things that may not be as aesthetically appealing in a neoclassical model to get to the policies that do a lot more good than harm. So I think that’s another level of humility that I would ask us all to embrace.”
Van Reenen gives a version of this argument in his case for green industrial policy, noting the importance of maintaining political coalitions.
Again, the Obama administration were no strangers to political constraint. But the shadows of Trump and populism loom over the US debate around industrial policy. The rising political stakes have lent credence to the view that optimal economic strategy is not synonymous with picking the most efficient individual policy instruments.
Good policy in a second-best world
The theory of the second-best says that in a world of ubiquitous market failures and political constraints, taking one step toward the competitive ideal may not be an improvement. I’ve argued that this sort of thinking opens the door to industrial policy by raising the possibility that policies which might seem distortionary under ideal conditions can make things better in the real world.
But all the second-best theorem says is that it’s possible that eliminating one market failure may not be welfare-enhancing in the presence of others. It doesn’t say that any particular policy actually does more good than harm. And when it comes to industrial policy, there are still plenty of critics. In his piece for ProMarket, Duke’s Michael Munger argues that the existence of market failures isn’t a sufficient justification for industrial policy because there’s little reason to think turning decision-making over to bureaucrats will be an improvement. The key to resolving this debate is to rely less on theory and more on data. Yet it’s striking how little we know about which policies do and don’t work. Some pieces of Biden’s industrial strategy seem justified by both theory and evidence, like investments in scientific research and subsidies for electric vehicle purchases. Others, like the provision that those EVs be made in America, seem dubious.
If the case for some of these policies is not efficiency or innovation but political economy, it’s reasonable to ask for a more-detailed justification on those grounds. If we’re going to rely less on economic theory, let’s at least have more political science in its place.
For instance, is ‘Buy American’ actually critical to maintaining political support for climate policy? The truth is we just don’t know. And much of the public discussion of these political-economy concerns falls prey to what the political scientists Christopher Achens and Larry Bartels call the “folk theory of democracy,” in which an idealized view of politics misconstrues how it actually works. Take the debate among US political strategists and pundits over the idea of “popularism”—meaning roughly that it’s good to propose policies people like. Fair enough, but as Achens and Bartels argue in their book “Democracy for Realists,” elections are usually less about voters rewarding policy proposals that fit their preferences and more about social identity and conflict between groups.
Without a better theory of political economy concerns, the theory of the second-best can become a shield for defending almost any policy. Sure, it runs counter to economic theory, one can say, but you have to consider the ~political economy~.
Still, the fact remains that the Biden administration was able to accomplish through industrial policy what the Obama administration couldn’t get done with a carbon price.
At the end of his remarks on carbon pricing in 2015, Obama added that “[Climate change is] just about the hardest thing for any political system to absorb.” By then he had had seven years to reflect on the political barriers to enacting good policies. “We have to see what works,” he concluded. “When something doesn’t work, we have to change our approach.”
Last year, the US finally passed a climate bill on the scale of the climate challenge. The second-best approach sure looks better than doing nothing.
Articles represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty.