Papers, Payola, and Principles

Alex Tabarrok links to a quirky proposal from EJMR:

I have a new paper that I consider my best work. For a variety of reasons, the marginal return professionally for this paper is very small for me. But I think it has an excellent shot a top
journal, I would estimate 1/3 at the AER (I have published there before). So I am offering it for sale. Here are the details:

1. This paper is not yet posted on my website. It has not been circulated and I have not yet presented it.

2. The paper is applied micro although I will sell it to anyone.

3. Email bids to Use a fake account and make sure to send no revealing information.

4. Your bid is for an AER or QJE. If it ends in Restud, you pay 65%. If it ends in the Journal of Labor Economics, Journal of Public Economics, or EJ, you pay 35%. Other journals are negotiable. You can choose the submission path as long as it starts with one of the top journals.

5. I will contact the winning bid (or highest real bid) to arrange an in person meeting in Philly at the meetings. We will never leave a paper trail.

6. Half of payment is due with a revise and resubmit. I will also make the needed changes. The final half is due with final acceptance.

7. Spare me any discussion of the ethics here. I am dead serious and I will not be commenting further on this thread.

I want to respond only to point (7). At first, I thought the ethics were disgusting. Very similar, indeed, to fraudulent data in the natural sciences. In a systematic sense, I think I was right. The impetus for both purchasing quality papers and cheating on empirics is to elevate one’s stature in a profession that is deeply-wedded to journal output. “Publish or perish”, they say.

But in and of itself, this is better approximated by the more interesting and “grey” debate behind payola. As an up-and-coming musician, I can pay a prominent radio jockey to play my record during their normal hours. The ethical “questionability” comes because, traditionally, the audience (and, for that matter, radio station) isn’t informed of this transaction; they believe they are listening to something “picked” by the RJ himself.

There are a few important criteria for success. For one, the audience can’t suspect the RJ is engaging in such practices; why would we voluntarily listen to an advertisement? Further, the music itself cannot be too bad. That is, audience members must be convinced the record in question would be part of a natural playlist designed by the RJ.

The ethics here are not as clear. There is a principal-agent problem between the RJ and his station (where his immediate and sustained profit might, in the long-run, ruin general credibility of the host). But this can be solved if the station is made aware of any such transactions. This could even allow them to pay the RJ at a lower wage rate, increasing overall welfare. Something feels “wrong”, but we are always free to switch to another radio station that we believe plays better music.

The analogy is a little tenuous, but informative. In place of the music are the paper purchaser’s academic credentials. That is, a five year-old will not credibly convince anyone that he’s penned a prodigious paper; in the same way my singing will not be aired on primetime radio for any sum of money. The audience is academic economists. The buyer himself (separate from his credentials) is the musician, and the authentic author the RJ. The act of being published in a famous journal is tantamount to one’s music being “picked” by a popular RJ. (They are both effectively credentials).

The transaction is almost certainly socially efficient (there are externalities of the ethical kind which I will discuss shortly). The marginal benefit of fame and fortune from a “top pub” for the genuine author is clearly dominated by that of a potential buyer, who himself will have to be smart enough to pull it off. So no one totally undeserving is going to get a free paper. That is, if you think devious academics are deserving to begin with, but that’s another story that… I’m about to discuss.

The obvious externality isn’t – as some on MarginalRevolution threads have suggested – the “honest” grad student who won’t be able to get his paper published because someone deviously bought the paper. The journal editors select the best papers on the quality of the work (though apparently “prestigious” authors get a leg-up, but clearly Larry Summers won’t be purchasing a paper from hormonal idiots on EJMR).

The external cost, rather, is the potential academic and professional success that is derived from this paper. From two talented candidates for tenure-track at a top college, other things equal, the position will be allocated to the one with a fancy paper in AER. But in a perfectly competitive and ideal academic job market, the non-talented but “credentialed” economist will immediately be excluded in favor the better person. Therefore, this is an indictment of the system itself – which places undue regard on publications and citations – rather than the microcosmic act of trading papers for cash.

But the system can also save itself. Let’s assume “publish or perish” is a fair constraint. Therefore, if a relatively untalented grad student lands a too good job because he purchased this paper, he will become all but irrelevant if he does not continue to publish. If this is true, and he is actually untalented, he will be forced to write more papers.

Let’s now stipulate that there are a group of very smart people who derive very little marginal benefit from publishing an economics paper. Maybe they’re brilliant, but not interested in the subject. Maybe they’re working in other fields. Maybe they’re Paul Krugman and have decided the New York Times is more influential than AER. These highly potential minds, then, are not devoting themselves to good research because incentives are misaligned. However, if there emerges a demand for high-quality papers – because there is a captive demand among the untalented who entered academia on false credentials – then these people will reenter active scholarship because the updated marginal benefit now exceeds all costs.

Therefore, there is reason to believe research output would increase.

A final externality is that eventually, if this system becomes sufficiently pervasive, credentials will loose all value. That’s not a good thing, but also not different from payola. But there’s also good reason to believe life cycle factors will prevent this from “taking over” the profession. It is unlikely that older economists will derive much value from buying papers because the discounted future value from the paper is small (they’re well into their careers). Further, if they have not done something brilliant thus far, they will be under scrutiny from peers who are surprised (Dr. Yitang Zhang at your service).

The union of the sets of people who (a) care about getting published in AER (b) are young (c) can afford a lot of money (d) do not consider it unethical (e) want a career relevant to economics and (f) are sufficiently smart to credibly convince the community that it is their work is small. The externality is non-linear in nature. That is it will “suddenly” become a problem if everyone’s doing it.

But by definition everyone is not. And therefore I believe paper payola is economically, socially, and intellectually beneficial. Flame me.

P.S.:  I retrospectively add (from my comment on MR):

I would add this is a self-correcting system in the classical sense. If the erosion of credential value (as inevitably happens if this paper sale becomes larger and more pervasive) is large, the value of buying a paper will collapse. And if that collapses, people will regain faith in the value of credentials. And the rest is history (repeated).

In fact, this brings curious contradiction to the application of “you make the system, the system makes you” type criticism to purchasing papers. The more you purchase papers, the more you disincentivize others from doing the same. Negative feedback cleanup!

P.P.S.: I’m not covering my ass. Of the six mentioned sets, I believe I fall in only 1.5. I will leave it as an exercise for the reader to figure which.

  1. ‘… credentials will loose all value. That’s not a good thing….’

    I’m not too sure about that. It might be all for the better if the ‘journal system’ went the way of the dinosaur. And, with blogs having established themselves, it might be happening.

    And, it isn’t like the AER hasn’t published fraudulent papers in the past under the traditional peer review system.

    • That’s true. I imagine in the short-run there are some negative effects, though. For example, if I write what I think is a brilliant paper, do not want to sell it and place a high value on fame/credit, I am cheated by the system.

      In the long-run it’s all self-correcting.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s