Econ Journal Watch:

The latest issue of Econ Journal Watch is out and I wanted to call readers' attention to a couple of articles in particular.

William McEachern of the University of Connecticut has an article presenting evidence on the campaign contributions of "American Economic Association Members, Committee Members, Officers, Editors, Referees, Authors, and Acknowledgees." Subject to obvious caveats about the nature of the data set, McEachern finds that in the AEA generally the Democratic to Republican contribution ratio is 5.1 to 1 and that on average contributions to Democrats were approximately 20 percent larger than to Republicans. He also finds that about 10% of the editors of the American Economic Review contributed to candidates (9 out of 84 Americans), and of those, all of them contributed to Democrats. McEachern similarly studies the editors, referees, etc., of the Journal of Economic Literature, and Journal of Economic Perspectives. He concludes by asking whether it is possible that the ideological orientations of editors and referees have the potential for influencing their opinions as to the quality and relevance of various articles.

Dan Klein follows-up with an an essay on the possible implications of McEachern's research (and that of others) for the professional practice of economics. Klein makes the provocative argument that the editorial leadership of the AER's various journals may be reflected in the pattern of articles published there. These journals are among the most prestigious in the profession, so there is some import from this. Klein suggests that the ideological predispositions of the editors is reflected in the type of articles that are accepted and published (his argument that McEachern's data reflects itself in the articles published is largely anecdotal). Put crudely here (and with more nuance there), Klein suggests that "liberal" economists who serve as editors of the AER are less likely to publish articles that are critical of interventionist economic policies.

Note that itis not necessary for editors to be consciously-biased in order for ideology to influence publication processes. Ideological bias may mainifest itself as simply as the degree of skepticism applied to a given argument or the extent to which a reader believes that the author has "considered" competing explanations for the observed phenonmenon, for instance. Even more diffusely, what a given editor believes to be an "interesting" or "original" contribution will be colored to some extent by ideological views.

Note more generally that Klein's argument is anchored in the McCloskeyite notion of the economic enterprise as a sort of conversation among economists, rather than "truth" per se. In this McCloskeyite view, the preexisting ideological perceptions that economists bring to the table would be relevant to understanding this process in a way that it would not be in the standard Popperian sort of analysis of scientific economics. Obviously this assumption itself in Klein's paper is likely to prove controversial.

Both articles present the question whether subconscious ideological beliefs influence the publication process of professional economics journals. Klein also praises (p. 185) the role of blogs in creating a more spirited discussion about economics than is present in most scholarly journals. If that is so, and it does seem plausible to me, Klein argues that there is room for some degree of disclosure in the publication process to determine whether there may be ideological bias that may color editors' and reviewers' perceptions of an article.

There is an interesting analogy to Klein's--it is commonly expected that if someone has a financial interest in a matter or their research was financially supported that a disclosure is appropriate. Given that a person's ideological bias may also influence their independence, it is an open question whether it might also be appropriate to make similar sorts of disclosures about ideology. I'm not sure, but if Klein is right that ideological bias matters in professional publications (and he's the only one I have seen that actually looks at that question), the logic of the argument suggests that such disclosure might be appropriate.

These articles are sure to prove controversial, both in the findings of the data and the implications that the authors draw from them. By the way, Econ Journal Watch is a fascinating publication (most of the January 2006 issue is actually focused on completely different issues) in that it is an effort to create the sort of robust debate that is actually somewhat lacking in most academic journals today. In some sense it can be thought of as a hybrid between academic journals (in their seriousness and depth) and blogs (in their effort to create a real dialogue of ideas, rather than just a series of monologues).

If Klein is right, there does seem to be implications of this for law reviews. My impression is that most law professors work under the assumption that ideology is almost certainly present in the selection process of student edited law reviews, given the nature of the process. It is an observation frequently remarked upon in passing, but to which little thought is given as to the possible implications. It almost certainly influences where articles get accepted (or even read) and in my experience it can impact the editing process quite dramatically. I have no original ideas about whether this actually turns out to be a substantial problem in practice and if so what might be done about it. But it seems like an important question that has heretofore been largely ignored.

Anyway, while I don't necessarily agree with the entirety of the argument (just because I haven't thought it all through in depth) I thought the argument sufficiently interesting that I wanted to call to readers' attention this conversation in Econ Journal Watch.

Commenterlein (mail):
The right way to read McEachern's results combined with the prior evidence cited in the paper seems to be that the majority of academic economists lean democrat, and that this disparity is reflected in the leanings of AEA officers and the editoriol boards of the AEA journals.

It is not clear to me how one would construct a "bias" argument from these findings. It would be more disturbing, imho, if rank-and-file AEA members leaned one way and the AEA leadership leaned differently.
1.27.2006 3:51pm
alkali (mail) (www):
The discussion of "bias" and "ideological orientation" in the foregoing warrants some analysis. Consider the following discussion by Michael Bérubé, posted to his blog just today:
[M]ost critics of universities don’t seem to distinguish between unconscious liberal bias and conscious, articulate liberal convictions. They take the language of “bias” from critiques of the so-called liberal media, where it is applied to outlets like the New York Times and CBS News that, in the view of some conservatives, lend a leftish slant to the news both deliberately and unwittingly. But the language of “bias” is not very well suited to the work of, say, a researcher who has spent decades investigating American drug policy or conflicts in the Middle East and who has come to conclusions that amount to more or less “liberal” critiques of current policies. Such conclusions are not “bias”; rather, they are legitimate, well-founded beliefs, and of course they should be presented—ideally, along with legitimate competing beliefs—in college classrooms.
In particular, I don't know what it means to say that "a person's ideological bias may also influence their independence." Suppose, for example, we were considering the case of a biology professor reviewing an essay arguing in favor of intelligent design. In what sense does the biology professor have an "ideological bias"?
1.27.2006 4:26pm
As an economist, I have always been struck by how unideological the field is in comparison to, say, law or political science. Even ideas with ideological resonance (e.g., schemes for social security reform) are typically evaluated relatively dispassionately. Good economics (and good applied statistics) plays by certain rules, and attempts to cheat are usually relatively obvious.

With regard to subtext in this post (which I assume reflects McEachern's point), I don't think either my liberal or conservative colleagues feel that journal decisions are related to political ideology (although other kinds of biases are often alleged).

In particular, well-placed conservative economists with stellar publication records are not that hard to find. Feldstein, Poterba, Hoxby, Mankiw, Hubbard, do I have to go on?

And finally, on whether "liberal" economists favor interventionist solutions -- in almost no field are practitioners more cognisant of the potential failures (inefficiencies or perverse, unintended consequences) of government action. There is a strong reluctance among all economists to intervene in even poorly functioning markets.
1.28.2006 4:48pm