Tuesday, August 4, 2020

I Surpass Keynes: 33 Years Plus One

John Maynard Keynes served as Editor of the Economic Journal--the research journal of Britain's Royal Economic Society--for 33 years plus one issue. He was appointed to the position as a 28 year-old in 1911, and finished with the April 1945 issue. He did have joint or assistant editors for periods of time--Francis Edgeworth (1919-26), D .H. Macgregor ( 1926-34) and Austin Robinson ( 1934-45)--who must have played an especially important role when Keynes was off advising the government. But "for 33 years and one issue Keynes played the central role in the development of the Journal's character and reputation,"  as Donald Moggridge tells the story in "Keynes as Editor" (as chapter 7 in A Century of economics : 100 years of the Royal Economic Society and the Economic Journal, edited by John Denis Hey and Donald Winch, 1990, pp. 143-157). 

The timeline is meaningful to me because with the publication of the Summer 2020 issue of the Journal of Economic Perspectives, I have now been Managing Editor of JEP for 33 years and one issue. For the first and in all probability the last occasion in my professional life, I have surpassed Keynes.  As I crack open a cold Diet Coke from the mini-fridge in my office to celebrate, the Moggridge essay also offers an opportunity to think about how academic journals and the role of their editors has shifted over time. To the modern eye, perhaps the most surprising aspect of a prominent economics journal from the first half of the 20th century is the informality and flexibility of the editorial process, 

Most modern economics research journals have an editor (or group of editors) who send the paper to be evaluated by "referees," who in turn offer recommendations of reject, accept, or revise-and-resubmit. In theory, at least, the process is "double-blind": that is, the authors don't know who the referees are and he referees don't know who the authors are. Modern journals brag if their turnaround process for these referee comments is measured in "only" a few months. But the revisions requested are often nontrivial, and the author has other tasks (like figuring out how to teach online classes), so it may take a few months for the author to respond. At this point, the referees may get another crack at evaluating the paper, which can lead to an additional round of revisions. If and when a paper is accepted, it may then sit for additional months before it actually ends up being published in an issue. 

The editorial process at the EJ under Keynes was much quicker and simpler.  Moggidge writes: 
Ultimately, however, Keynes made the overwhelming majority of editorial decisions himself. He did not regard the job as onerous: indeed, by the 1930s it was one of those activities classed as 'non­work' and done after dinner (Kahn, 1984, 176). As in other areas of his life, he made decisions quickly. To take one example: Lionel Robbins submitted his 'Interpersonal Comparisons of Utility: A Comment' on 31 October 1938; Keynes accepted it on 3 November and the article appeared in the December issue. ...  In fact, in most cases a month was the maximum time between initial submission and final acceptance, and this was exceptional, unless the article was to be subjected to substantial revisions and the whole process then became dependent on the time available to the author. He did not keep a backlog of articles in reserve: at worst an author might see only one issue of the Journal appear between the acceptance and the appearance of his article.
Part of the reason for such quick turnarounds was that Keynes, with relatively few outside reviewers, was the judge of quality. 
As far as contributors other than himself were concerned, Keynes was generally sparing in his use of external referees. He did, of course, use them, particularly in cases where he had doubts about the submission. On several occasions, however, he sent the submission to the referee with a copy of his own draft letter of rejection -- or at the least a strong hint of his own views.
However, there are also numerous cases where Keynes provided extensive comments to authors. Moggridge cites one paper that involved 30 pages of correspondence, and many letters of five pages or more. Keynes also was willing to choose only one part of a much longer paper as deserving of publication, and would sometimes accept a paper but with suggestions that would cut the length by two-thirds. After Keynes retired from the EJ, his successors offered florid praise for the quality of his editorial feedback in the April 1945 issue, where they wrote: 
That editing has been far more than a nominal control of what was to be published. A whole generation of economists would testify to the influence which he exercised with his meticulous, but always constructive, criticism. None but the authors in the secrecy of their own hearts (and perhaps not even they) can know how much of what was ultimately printed was their own, and how much had sprung from the lively ingenuities of his mind. It was to the young and promising that he was particularly lavish with his help, and today (no longer so young) they remember that aid with gratitude.
In part, this may be the kind of fulsome praise lavished upon those as they leave a job, but it probably has some intermixture of honest feeling as well.  

Perhaps the best-known stories about Keynes as an editor have to do with his rather loose rules about publishing his own work. or publishing his comments on the work of others in same issue as the original paper, as well how he phrased some of his negative reactions. Moggridge writes:
Unlike some modern editors, Keynes himself had few inhibitions about placing his own articles in his own journal. While he was editor, his only publications in other English language professional journals, other than seven replies to critics of which only two were over a page in length, number two: his November 1914 invited contribution to the Quarterly Journal of Economics, 'The City of London and the Bank of England, August 1914' (JMK, XI, 278-98), and his 1936 Jevons centenary allocution to the Royal Statistical Society which appeared in that Society's Journal (JMK, X, 109-50). He was not even inhibited from using his position as editor to reply to articles by others in the same issue in which they appeared, going so far as to place his own comments immediately after the piece that had irked him. However, in later years he became slightly more restrained in this, informing one contributor with whom he was in dispute over some points prior to publication that: 'It is a mistake for an editor to quarrel with contributors.' As far as I can tell from the records, he submitted none of his own contributions to an external referee. Nor could he have done so in some cases, for the articles themselves were only finished at the last possible date. 
Perhaps the most famous Keynesian editorial put-down was in writing to a referee about a proposed article by Kalecki: 
I am inclined to return to the opinion that the article is pretentious, misleading, inconclusive and perhaps wrong. I would rather have cheese to a weight equal to the paper it would occupy in 5,000 copies of the Journal.
Or he once wrote to an author: 
[I]t seems to me clear that your article, in its present shape, is half-baked and not fit for publication. I have not been able to spare time to read it carefully enough to know whether there is anything in it at the bottom. But I find it a bit of a rigmarole, of which I fear the reader would make little or nothing. It is neither clear what you are driving at nor where you arrive. And behind all that lies my doubt as to whether the method you are employing is capable of helping much with this particular problem. Also do you not in many cases use symbols where words would do as well and be much clearer? Sorry to be so critical. But I felt I owed it to you to explain why it is I cannot take it for the Journal. 
Or in yet another case, he turned down an article partly for dullness, and partly because Americans should go publish in American journals: 
It strikes me as a competent academic exercise which any competent analytical economist could accomplish if he wanted to. I did not find it interesting or really relevant to anything that matters. I do not see any ground for an American econom­ist to occupy a large number of pages in the Journal with it. Why should he not publish it in America? I did not find myself making any criticisms, but felt, as I have said, that it is just an academic exercise.
Keynes also had the power to solicit articles, rather than just wait for submissions to come in, something he did throughout his editorial career. The Journal of Economic Perspectives, where I work as Managing Editor, is highly unusual among current economics journals in that we solicit the articles that we publish--but then we comment on and edit the articles extensively before publication. 

Overall, how good was Keynes as an editor? Moggridge writes: 
Contemporaries were generally satisfied. Even a critic such as Edwin Cannan could remark in a memorandum of 1 February 1934 to a committee looking for a successor to D.H. Macgregor as Joint Editor:
He may not be exactly the ideal editor, who is, I suppose, one who has no ideas of his own and a great respect for those of other people, but his most hostile critic cannot say that the Society has not prospered during his editorship.
In Keynes own estimation of his editorial performance, he apparently claimed that he "Never rejected what deserved publishing; have published much that wasn't." Moggridge also reports a comment from Keynes to Edwin Cannan on 5 January 1934: 
Of course, I print quite a number of articles which in my personal opinion are not up to much, but I have to compromise as best I can between those which I fancy myself on their merits and those which, on one ground or another, have some sort of claim to appear. I feel much clearer, however, about the de-merit of the articles I reject than I do about the merit of most of those which are included. I have always tried to find space for anything which seemed to have a claim of any substantial sort and the somewhat numerous articles which I reject would, I declare, make a shocking show on any standard, if they were to be assembled. 
As one might expect, what I do as Managing Editor of JEP is in many practical ways a quite different set of tasks from what Keynes did when he edited the EJ. After all, the EJ relied mostly on pickng and choosing among the submissions, while the JEP relies mostly on commenting and editing on solicited papers. But there are also basic differences in process. 

In particular, JEP was started in 1986 with the idea of taking advantage of this newfangled technology called a "floppy disk," where authors could actually send us a Federal Express package with the disk. Next, we would write up comments, which for me also included doing a hands-on edit of the paper for length and clarity. We would then Fed-Ex the floppy disk back to the author, who could revise further, before mailing the floppy disk back to us. We even sent floppy disks directly to the typesetter, rather than paper copies! In the long-ago days before cheap photocopying, there was often literally one copy of a given paper in circulation--with usually a carbon copy saved up by the author. That single copy then needed to be circulated to editor and reviewer, and was perhaps even marked up by hand before going back to the author for revisions. Back in 1986 when starting JEP, we definitely felt we were on the bleeding edge of technology.