Journal Turn-around Times: The Slower, the Better?
Because I’m a recently defrocked journal editor, a new analysis of journal turn-around times (that is, the elapsed time between submission of a paper and the editor’s first-round decision — accept, reject, or revise and resubmit) recently caught my eye.
In olden days, when one submitted a paper to a social science journal, one could confidently expect to wait anywhere from six months to a year, or even more, for a first-round decision. Times have changed, and social scientists now expect — and editors try to provide — much faster service.
It turns out, though, that in economics journal turn-around times have actually slowed down. Across a wide array of economics journals, turn-around times appear to have lengthened over the last four decades from two months, on average, to between three and six months.
Of course, economists think differently than normal people do, so it stands to reason that they — or at least some among them — have a rather unexpected take on this slowdown.
In a recently published paper (here, gated), economist Ofer Azar notes that at first glance the slowdown in turn-around times in economics journals seems to be “a bad outcome”: “Slower [turnaround] means that new research is disseminated to the academic community less promptly, which is a bad thing.” Nonetheless, Azar considers the slowdown a positive development. Here’s why.
Thinking like the economist that he is, Azar posits the existence of an “optimal” first-response time (FRT for short). It may seem that an FRT of zero would be optimal, or, more realistically, that the closer the FRT is to zero, the better. Not so, Azar contends: A longer FRT “reduces the costs of the refereeing process because a longer FRT reduces the number or submissions of low-quality papers to good journals.” As Azar continues:
With zero FRT, and given the low submission fees in economics [or the absence of submisison fees in political science], the cost for an author of submitting an existing paper to a top journal is so small compard to the potential benefits … that it is worthwhile to do so even when acceptance chances are very low. By submitting the paper, however, the author creates a social cost: referees and editors have to dedicate their scarce time to evaluate the paper. The problem is that the author faces a private submission cost that is much lower than the social cost of submission.
…Increasing the FRT can alleviate this problem since higher FRT increases the submission cost for the author. The FRT delays the publication of the paper and thus creates a cost for untenured authors who want to have publications before their tenure decisions. The FRT also creates a cost for tenured authors because promotion and salary depend on publications. The delay created by the FRT causes the author to think twice before submitting his paper to journals where he has very low acceptance chances and thus decreases the number of submissions of low-quality papers to good journals and reduces the costs of the refereeing system by saving the scarce resources of editors and referees.
There’s more to Azar’s argument than this (and even a formal model!), but that’s the core of it.
Azar’s analysis suggests a perfect excuse for laggardly reviewers to use: Rather than being too lazy to provide prompt reviews, slow reviewers should claim that they’re actually providing an important public service by being so dilatory — though I strongly suspect that they’ll find that account more appealing when they’re on the giving end, as reviewers, than the receiving end, as authors.
Comments
Very interesting. I agree that slow turn-around times discourage article submission. But I'm not sure the implication is efficient pre-screening of articles based on quality.
1. Scholars don't necessarily interpret low acceptance rates as uniform across types of articles. They may view their articles, rightly or wrongly, as having greater or lesser chances of acceptance based on its methodological genre, its subject-matter, etc. Thus, low turn-around times may exacerbate distortion effects of pre-screening for "match." The APSR is a good example of this; I suspect that the fast turn-around time encouraged high-quality diverse submissions under the last regime (but I'd have to see the data).
2. There remains a very high degree of stochasticity in reviews (a pattern demonstrated by multiple studies across the social and natural sciences). I know plenty of stories about articles that rack up more acceptance reviews than rejections, but still need to go to multiple equally competitive journals because of the distribution of those reviews, e.g., accept and reject, 2 accept and 1 reject, and finally 2 accept. Long turn-around times might, therefore, simply lead to the under-placement of articles.
3. Both 1 & 2 suggest the possibility that one category of manuscript most likely to suffer from long turn-around times would be controversial but high-quality articles, i.e., those we'd most like to see in well-read journals.
4. I haven't read the article you link to, but does it deal with the dynamic effects introduced by the possibility that some editors might behave strategically, e.g., that editors of lower-tier journals might seek to increase their volume of submissions and selectivity--and hence enhance the prestige of their journal--by lowering their turn-around times versus existing high status journals?
Posted by: Dan Nexon | January 18, 2008 12:48 PM
Let me add that the article suggests an interesting, but inadequate fix, for an underlying problem.
Increasing reliance on peer-reviewed publications as a metric of academic status and success--particularly with respect to advancement--increases the volume of submissions to peer-reviewed journals. This increase in volume, however, is likely to continually outstrip the size of the pool of those able to provide quality peer reviews. Thus, the effectiveness of screening through peer review will likely decrease over time... which is somewhat perverse, given that this decreased reliability of peer review is an effect of the greater importance placed upon it.
I can't find the citation offhand, but Arthur Stinchcombe first advanced this argument decades ago.
Posted by: Dan Nexon | January 18, 2008 01:01 PM