Research Highlights Podcast
May 7, 2024
The pace of economics publishing
Daniel Hamermesh discusses publishing in economics and how to speed it up.
Source: NuPenDekDee
Timely publication of research in peer-reviewed journals is critical for economists seeking tenure and important for audiences looking for high-quality, trustworthy studies. But in recent decades, there has been an increasing concern that the pace of publishing in economics is too slow.
In a paper in the Journal of Economic Literature, authors Aboozar Hadavand, Daniel S. Hamermesh, and Wesley W. Wilson analyzed the publication lag in top economics journals and compared it to other fields. They found that economics publishing takes nearly twice as long as comparable fields in the other social sciences.
However, Hamermesh says that some innovative journals, such as AER: Insights, are taking steps to shorten the time between submission and publication. He recently spoke with Tyler Smith about the pace of publishing in economics, how to fix it, and some advice for young economists trying to publish their work.
The edited highlights of that conversation are below, and the full interview can be heard using the podcast player.
Tyler Smith: With all the resources economists have for getting work out, such as posting on websites and social media, why do you think a timely publication process for academic journals is still important for the economics profession?
Daniel Hamermesh: There are immense numbers of people trying to get their ideas out in the public. If you're one of 25,000 economists worldwide who are trying to do research and get it disseminated, the number of people who are going to see it is minuscule if you’re trying to put it out on your own. It's not going to hit much of an audience. So I think that's the real reason why being timely is important. I think the other reason is that it just provides a way of giving a stamp of approval to the research. Unless it's in a peer-reviewed journal, many journalists will not publicize your research. In the sciences, that is the sine qua non. In economics, it’s less so, but if it takes two or three years to publish, we're not getting the ideas out in a timely fashion.
Smith: If we can speed up the publication process in economics, who do you think stands to benefit the most?
Hamermesh: I think the biggest beneficiaries would be junior economists. I know economists are often in a university with other social scientists or with hard scientists, and they get their stuff published within three to six months or even a year after submission. When some poor assistant professor in economics comes up for tenure after six or seven years, she or he has very little published, and we look stupid and we suffer compared to those people in other fields. I've seen this happen repeatedly in the several years I've served on a promotion and tenure committee at the University of Texas, so those are the major beneficiaries. The second beneficiary, I think, would be the general public who gets this stamp of approval, this implicit bona fides earlier, and thus the results may impinge on the public more quickly.
Smith: How does the pace of economic publications compare to other fields, whether that's the hard sciences or social sciences?
Hamermesh: There are several comparisons to be made. First of all, compared to the hard sciences, from the time I submit a paper to the time it appears in print, it takes 6 to 8 months for the average paper to get published. A more relevant comparison is the other social sciences that are quite similar, such as psychology and sociology. And there the evidence says it takes us twice as long from the time the paper is submitted, to the time it's accepted and appears in print. We shoot ourselves in the foot vis-à-vis the other social sciences. Finally, one might ask—and this is much harder to get evidence on—have we gotten faster or slower over time? I've been trying to publish for 57 years now in academic journals. My first real paper was published in early 1969, and I remember the first paper in one of the so-called top-five journals was submitted in October of 1969. It was sent back to me in January; they asked for minor revisions, and it appeared in print in August of 1970—ten months after submission. These days, if I were to do it at a top journal in two and a half years, from submission to print, or even a year and a half from submission to acceptance, that would be very good. It has gotten slower, and contrary to all the efforts we've been making at these very top journals, it has not improved in the last decade at all.
At the extreme, I mean the slowest 75 percent, we're talking four years from submission to a final publication. And this assumes the paper is accepted eventually at the first place to which one sends it. In most cases, one sends it to some place with acceptance rates that are 5 percent or less. If you have to go elsewhere, that makes the process even slower.
Smith: How long did it take you to publish this article we're talking about today?
Hamermesh: This is the beauty of this paper. The paper was accepted after a couple of revisions in June of 2022. It appeared in print in March of 2024, 21 months later. Part of that was my fault. I had to make sure the data were 100 percent replicable. That took 9 of the 21 months. But even aside from that, it was still another year. So I think the paper illustrates the problem it's talking about.
Smith: What are some of the most important reasons for economics publishing being slower than these other fields you talked about?
Hamermesh: There are three groups who might be at fault, if you will, in causing the slowness: the referees whom the editor asked to look over the paper; the editor, him or herself, who spends time cogitating about it; and the authors. I can't separate the referees’ from the editor's slowness. If you break it down between editors/referees and authors, the authors are the major cause. It's their time from the time they get the paper back with suggested revisions to the time they send it back to the journal that’s the slowest. Given that, however, the editors will simply transmit so often the referees comments and say take them into account. And many referees view themselves as unmentioned coauthors and feel that anything they say the author has to take into account. And the editors, rather than editing, will simply send the thing on and say take care of it. People write responses to referees and editors which in many cases are longer than the paper, which seems nuts. So I point the finger at the editors who aren't editing, and too often they’re channeling referee comments. To my mind, the referee's job is not to be a coauthor, but simply to tell the editor if there are mistakes. And that's not true anymore.
I tell young people that before they even send the paper off anywhere, they should have a refereeing tree in mind. If it doesn't get to the top branch where you're sending it, have in mind immediately where you want to send it next.
Daniel Hamermesh
Smith: Some journals like Economic Inquiry have this policy where you can submit it, but without resubmissions. Do you think this is the best way to fix the problem? Are there any other ways to speed the process up?
Hamermesh: The problem with that is who chooses to submit without resubmission. And we looked at that. We had a lot of data from Economic Inquiry. It happens that people who choose to submit tend to be more secure in their jobs, or to want a quick decision because they're facing a tenure decision soon. It's not a random sample, but that certainly does go a lot faster. I think, similarly, the AEA has established AER: Insights, which has a policy essentially of accepting or rejecting with some minor revisions. That seems to get things done an awful lot faster. That's one way out, and I think that's excellent. It puts the discretion on the editor's shoulders, which is where it should be. The editor is not just a transmitter of referee comments, she or he is the person who should be making a decision.
Another solution is to have journals accept referee reports from a paper refereed elsewhere. The AEJ: Macro, AEJ: Applied, AEJ: Micro, and AEJ: Policy all do that, and the papers that are submitted with prior referee reports do get published faster. A lot of other journals are now following the same procedure. I'm not sure it's a great idea, but the acceptance rates on those who send them and those who don't appear to be identical, just as they are in the Economic Inquiry.
Smith: If you could go back early in your career, what would you want to know about the publication process in economics and getting through it?
Hamermesh: I tell young people that before they even send the paper off anywhere, they should have a refereeing tree in mind. If it doesn't get to the top branch where you're sending it, have in mind immediately where you want to send it next. But unless there is something wrong with the paper, obviously, when you get it back from the first place at which you're rejected, don't spend time doing every little thing. And certainly don't spend time responding to every referee's trivial comment, because the referees in the next place are going to have totally orthogonal, totally different trivial comments. I think a refereeing tree is just absolutely crucial in saving yourself time.
♦
“Publishing Economics: How Slow? Why Slow? Is Slow Productive? How to Fix Slow?” appears in the March 2024 issue of the Journal of Economic Literature. Music in the audio is by Podington Bear.