Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
I’m not about to argue that economists are useless. Nor will I argue that they should all be obviously useful. If some want to do research that might seem like “mathturbation” but that is actually pushing forward the frontier of knowledge, good on them. But I do think that there is a gap between the supposedly policy-relevant research supplied by academia and what decision makers actually want. And that it could be smaller.
The fullest complaint is that too often researchers ask the wrong questions, then communicate the answers badly. Some of this isn’t their fault. Academia rewards novelty rather than usefulness. It also can encourage precision (“did the dog tax affect spending on dog food over its first three months?”) rather than breadth (“is taxing pets barking mad?”). And it offers the freedom to think about fixing a single problem with a perfect instrument. Meanwhile, policy is more often tasked with fighting multiple distortions with limited legal tools.
To take an example, economists have spent years trying to estimate the social cost of carbon, arguing that a carbon tax is the best way to fight climate change and criticising industrial policy as misguided. But when the Biden administration started asking how to deploy subsidies, the evidence base was lacking. “Economists don’t know what to do when they just think something is a bad idea,” says Betsey Stevenson of the University of Michigan, adding that on carbon taxes they “should figure out why they haven’t sold the public.”
What to do? The government itself could improve access to timely data. If economists themselves are keen to have more impact, a recent post by Jed Kolko, a former official at the US Department of Commerce, gave examples of work he found handy while in government. New data, like rental prices from Zillow, can aid real-time analysis. Literature reviews save policymakers from wading through oodles of papers. Quantifying the effects of policy changes can be useful too, such as the estimate of labour demand from investments in semiconductor manufacturing that informed workforce policy.
One challenge is that the path to tenure isn’t exactly littered with literature reviews. Academics are not rewarded if their work is cited by a government department or a regulator. But a group including Gopi Shah Goda of Stanford University is working on a metric that captures that kind of reference, which could eventually sit alongside academic citations as a sign of success. “If you can’t measure it, then you can’t even begin to reward it,” she says.
Other initiatives include work by scholars such as that by Eva Vivalt of the University of Toronto to develop reporting standards for published research and make results easier to compare across studies. Gathering more than 600 studies in development economics, she found less than 10 per cent mentioned a policy’s costs. (Policymakers care about prices.)
Researchers might also better appreciate the constraints policymakers face if there were easier routes from academia to government and back again. In Britain, it can be tricky to return to academia after a more practical gig. In the US, Martha Gimbel, who recently worked for the White House’s Council of Economic Advisers, rightly calls it “insane” that in some cases a stint there seemed to hurt academics’ promotion prospects.
My final plea is for economists to write more clearly. I’m not asking for titles like “10 Amazing Consequences of Higher Interest Rates You Won’t Believe Exist.” I am asking for titles that reveal the question or the result. Too many have the format “thing, thing and thing”. Unless one of those things is “sex”, “drugs” or “rock ‘n’ roll”, try again. Less than 15 per cent of National Bureau of Economic Research working paper titles include a question mark. Aim higher.
As for the rest of the paper, please note that impenetrability is a poor signal of quality. Papers with abstracts that have higher readability scores — functions of word and sentence length — get more citations. Admittedly, a comparison of abstracts published in the American Economic Review and two other top journals in sociology and political science suggest that other subjects are worse. But given how many abstracts of NBER working paper abstracts hover around the “difficult” mark, believe me when I say there is room for improvement.
I do not intend to do myself out of a job. Specialisation is important. Comparative advantage is real. Some work is for think-tanks, some is for journalists. But of all people, economists should appreciate that consumers of their research face constraints.