Thursday, June 07, 2007
Bernie Black: When To Post On SSRN
The recent discussion of what was, by all accounts, a less than spectacular article that was publicly distributed (through BE Press) prompts me to write again on my thoughts about when to post a working paper on SSRN. Here are some factors that matter to me. They don't all point in the same direction.
1. I want what I write to be read, built upon, cited, change the way people think about the world. None of that can happen until I post it. Some articles are more time sensitive than others, and will simply be less important if I delay posting too long.
2. I don't want to be preempted. Once I post, I have intellectual ownership of an idea. I realize that someone else, most likely someone else who is already working in the same area, might see what I write, build on it, not acknowledge me, perhaps even steal my idea and try to beat me into print. But usually the larger risk is that someone else will preempt me before I post, rather than steal from me after I post. Or so I want to believe -- and I think this is usually right.
3. I assume that I get one shot at any particular reader -- anyone who reads a working paper will likely not reread the final. So this draft should be reasonably complete, enough so that I'm happy if this is the version that gets read.
4. I assume that I will be judged on the quality of the work that I publicly distribute, and properly so. The draft should be spell-checked, proofread, ideally read first by a few colleagues, perhaps given at a workshop or two if the opportunity to do so was available. It's a DRAFT, but it ought to be a GOOD draft.
5. If I know a paper will be attacked, I am more cautious, and post a later, more polished draft. This has been a factor in a series of papers on medical malpractice I've written with David Hyman, Charlie Silver, Bill Sage, and Kathy Zeiler. Our data, and our analysis of that data, has political implications. It has been and will be attacked by people who have strong political priors, and by other people who care only about results.
6. For empirical work, if I have a new idea, but it relies on publicly available datasets, I am more cautious than if my idea relies on a hand-collected dataset, due to preemption risk.
7. When a revised, better draft is available, I post it, as a revision to the old draft. At least from then forward, the old draft disappears (but it still sits in the SSRN archives, together with the posting date, in case I later need to prove intellectual ownwership of an idea).
8. When possible (almost always with student-run law reviews, if you insist, and often with peer-reviewed journals), I obtain permission to post the final published version, and then do so, as a revision to the working paper. I add the citation to SSRN, including the magic words "as published in" [citation]. That's my way of telling readers this is the final version. When this is not possible, I add the citation (without said magic words) to SSRN, and post the latest draft I have. In substance, this is the final version, even though it will look different.
In the end, I usually post what I think of as an advanced working paper, perhaps ready to be submitted to journals, perhaps not quite there yet, but still solid. Earlier for some papers, later for others, based on a rough sense of the factors above, and perhaps other implicit factors I'm not thinking of right now. With coauthors, I tend more often to be pushing for posting, rather than pushing for delay.
None of this speaks directly to the idea of intentionally posting a short 10-15 page "idea only" draft, as Larry Solum has suggested, where it is clear that you are posting early and rough, to get an idea out there and claim ownership, taking the preemption risk that someone else who has a similar idea will say, "darn it, that was my idea, I had better write it up soon and better than that yo-yo." And yes, you'll be cited, but you had better be on track to get a fuller version out there too.
Finally, I don't usually flog my own work, but here are a couple of recent examples of the judgments we make:
A. A medical malpractice paper, just posted, and simultaneously submitted to a journal. Posted relatively late because of the political concerns mentioned above. Bernard Black, David Hyman, Charles Silver & William Sage, The Costs of Litigating Medical Malpractice and Other Personal Injury Cases: Evidence from Texas, 1988-2004 (working paper, June 2007) (http://ssrn.com/abstract=979163)
B. A law and finance paper, Vladimir Atanasov, Bernard Black, Conrad Ciccotello & Stanley Gyoshev, How Does Law Affect Finance? An Examination of Financial Tunneling in an Emerging Market (working paper March 2007) (http://ssrn.com/abstract=902766). This one has been circulating for a while. I wasn't even a coauthor on the first posted version. It's been revised several times, has gotten stronger over time, and is now under submission at a journal (having already been rejected at two other journals, but heck, it's a finance paper about Bulgaria, and a lot of finance people think almost nothing outside the US is interesting). I like this paper a lot, I want the idea out there, and it relies on data that we uniquely own so no one can directly compete with us.
[managing director, SSRN]
TrackBack URL for this entry:
Listed below are links to weblogs that reference Bernie Black: When To Post On SSRN: