Scientific fraud vs. financial fraud: is there a scientific equivalent of a “market crime”?

A little while back I read Dan Davies’ very good popular book on financial fraud, Lying For Money. Which prompted me to start a series of posts about the interesting analogies (and in some cases, disanaologies) between financial fraud and scientific fraud (see here, here, and here). Today, the latest post in the series: is there a scientific equivalent of a “market crime”?

A market crime is the most sophisticated type of financial crime–indeed, so sophisticated that there’s often disagreement from one jurisdiction to the next as to whether it’s even a crime at all. Insider stock trading, for instance, was banned in the US in 1934, but was legal in the UK until 1980 and legal in New Zealand until 1998. And there’s still no universal agreement as to who the victim of insider trading is. Is it the investor who trades stock with the insider, or the company for whom the insider works? Further, it’s not as if the British and New Zealanders only recently cottoned on to some universal human moral right to be free of insider stock trading. There’s no deep moral principle at stake here. As illustrated by the fact that insider trading isn’t a crime in most other markets besides the stock market. As Davies points out, nobody complains about the fact that Calvin Klein has inside information about the underwear market that its customers don’t have, which Calvin Klein uses to vary the price of underwear among stores and over time so as to maximize its profits.

In contrast to crimes like theft or murder, insider stock trading isn’t a crime because it’s contrary to some deep moral principle. Rather, it’s a crime because it’s contrary to the expectations of small stock investors. Small stock investors expect that they’ll be able to invest on a ‘level playing field’ with professional investors, and that they’ll have some reasonable chance to make a profit. They want to invest in the stock market, but they won’t do so if they feel like they’re being ripped off.* Insider stock trading is a crime only because many of the participants in the stock market want it to be. If they didn’t want it to be, it wouldn’t be a crime, and that would be fine too.

Another example of a market crime (which Davies gave in a talk I saw): performance-enhancing drug use by Olympic athletes. It’s morally bad to try to gain an advantage over your fellow Olympians by taking performance-enhancing drugs. But it’s not against the rules because it’s morally bad–it’s morally bad because it’s against the rules. After all, it’s perfectly within the rules for some athletes to gain an advantage over others in all sorts of other ways! Say, by eating healthier foods, or hiring personal trainers, or training at high altitude, or using better equipment. (Well, until that better equipment is banned.) Performance-enhancing drugs are a market crime–a crime against the expectations of most participants in (and most of the audience for) the Olympics. Those expectations are legitimate even though they have no basis in market-independent morality. Participants in any activity governed by rules are entitled to expect that everyone will play by the same rules. Even if those rules are a matter of arbitrary convention, like the US convention to drive on the right-hand side of the road.

So, are there scientific “market crimes”? Scientific practices that are bad only because they are against the “rules” of science (written or unwritten)? And are there scientific practices that aren’t currently considered “market crimes”, but might come to be treated as such in future, as the rules of the “scientific market” change?

What about self-plagiarism–republishing your own peer-reviewed paper in a different journal? After all, nobody complains about self-plagiarism when, say, a politician gives the same speech to different audiences, or a band plays the same song to different crowds, or a scientist gives the same talk at different conferences. There are plenty of contexts in which repeating yourself is totally fine, even desirable–or at least, not against the rules. Heck, I’ve be reposting a lot of my old blog posts lately, and many readers seem to like it. But peer-reviewed scientific publishing is a context in which there’s widespread agreement that repeating yourself in print is inappropriate–an illegitimate attempt to pad your cv. So I think self-plagiarism of scientific papers is a market crime.

Another example: failure to put make your data available for free download from a public website. I’m old enough to remember a time when there was no expectation that you’d ever share your raw data with anyone else. Now, for various reasons, there is such an expectation, enforced by explicit rules adopted by many journals and funding agencies. The rules of the “scientific market” have changed, so that failure to make your data freely available for download now breaks the rules of the market. It’s a newly-minted market crime.

Can you think of any other examples? Looking forward to your comments.

*Professional investors also want them to invest, because without small investors there’d be much less liquidity in the stock market. National governments want small investors to invest in the stock market too, because it’s considered socially desirable for various reasons.

23 thoughts on “Scientific fraud vs. financial fraud: is there a scientific equivalent of a “market crime”?

  1. Maybe author shenanigans, such as gift and ghost authors would fall into “market crimes?” Thinking of all the celebrities, business captains, and politicians publishing their self-promotional books under their own names. Sometimes there’s a small font “with Actual Writer” statement. Nobody seems too shocked in that setting that these Big Names are hiring uncredited or barely credited professional ghost authors to fill their pages. Not so in scientific writing anymore.

  2. One example would be non-open hiring decisions. In most commercial sectors it’s considered perfectly fine to recruit someone because there’s a gap at the organisation and you can identify a person to fill it, even if they’re a friend or existing contact. In an academic department this would now be considered foul play but only a generation ago it was fairly standard. Even if someone is already on the doorstep, perhaps fulfilling the role in a temporary capacity, you still have to hold an open competition for the position. Please note: I’m putting this forward as a response to the question in the post, not because I dispute the practice!

  3. Hmm, are we talking crime (breaking a law) or violating social conventions? Insider-trading on the stock market is a crime, punishable by jail time, but there is no law that you can’t republish your papers in another journal. If the latter is a crime, then nearly everything about scientific publishing that could be violated would be a market crime. Using present tense in your methods section or publishing statistics with just the p-values and nothing else all violate the conventions of scientific papers. And, in fact, making up the data is not a crime in the same sense that insider-trading is since you can’t be put in jail for it. Obviously, the social punishment of that is much bigger than of not presenting effect sizes with SE in your results, but I’m not sure there is a fundamental difference, just one of scale. Or am I missing something?

    • Well, as you note, fabricating your data isn’t literally a crime, at least not in most jurisdictions. But I wouldn’t say that it merely violates the social conventions of scientists! I don’t think data fabrication is analogous to a “market crime”. Rather, I think data fabrication is analogous to a crime like theft or murder–it’s against the rules because it’s bad, not bad because it’s against the rules. Science can’t function if fabricating data is totally kosher and not subject to any sanction. (And if all scientists thought that science *could* function despite regarding fabricating data as totally kosher, all scientists would be wrong.)

      I agree with you that there are some scientific conventions that it’s not very serious to violate (e.g., using present tense in your methods section). I don’t think of those as analogous to “market crimes”, I guess just because they aren’t very serious.

  4. Slowing down a “competitor” with unwarranted negative reviews.
    Failing to cite appropriate antecedents to better sell the importance of your work.

  5. Something I’m thinking about: is there an important distinction to be drawn between market crimes that violate purely arbitrary market rules, and market crimes that violate market rules for which there’s some sort of rationale?

    For instance, the rule to drive on the right-hand side of the road is a purely arbitrary convention. It’s vital that we all agree to *some convention or other* as to which side of the road to drive on. But there is literally no reason to prefer one convention of the other.

    In contrast, and as the post notes, rules like “no insider stock trading” or “you must share your data” have rationales. Those rationales often are debatable–they’re judgment calls about the pluses and minuses of alternative market rules. “Which market structure would be best” often is a contested question among market participants. But still, there *are* rationales for different market rules, and so debates about which rules would be best on balance are worth having.

    • The distinction is possibly one of practices which are autochthonic, and mostly only affect market participants and those where other parties (e.g. journals, the public) have a stake and exert influence.

      Free/unjustified authorship (or *not* including a contributor) is probably the former. It really doesn’t affect anybody outside the academic system, except in very indirect ways. This allows any rules to be arbitrary.

      Self-plagiarism and data-sharing are the latter.
      Journals (with their own market) take both as quality indicators.
      The same might be said for the public, which has a reasonable expectation of *additional, new* high-quality work (including raw data) for continued support.
      This latter applies to politicians and bands too, where returns decrease with repetition (scale dependent!).

  6. I think that preprinting is a good example of a market crime that’s now being normalized, or undergoing adjudication or whatever. Specifically, the sharing/citation of work that has not undergone formal pre-publication peer review. Some are still strongly opposed, and field-to-field variation is significant. I recently interacted with an editor of a history of science journal to whom the whole idea was apparently bizarre/sketchy that people would cite things without pre-pub PR.

  7. In the EU and UK it is an offence to sell a product to consumers without giving information that might persuade a significant number of people to change their minds. Information that might lead to retraction, for instance. I would argue that it is unlawful not to publish a critical comment. Failure to give information is as bad as giving false information. (I list the regulations below.)
    It is therefore an offence for a journal publisher to sell or offer for sale a paper that the publisher has reason to believe is wrong. If there is a retraction, or an expression of concern, or critical comments have been received, the buyer should be notified.

    The problem is not that the law does not exist, but that it is not enforced.

    I should have thought that the author contracts with the publisher to publish a paper, with the explicit or implicit condition that it is not fraudulent, and the publisher could sue for significant damages.

    I imagine most jurisdictions have similar laws.

    (European Commission. 2005. Directive 2005/29/EC Unfair Commercial Practices Directive.
    Accessed May 27, 2011, from http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?
    uri¼CELEX:32005L0029:EN:NOT
    European Commission. 2005. Directive 2005/29/EC Unfair Commercial Practices Directive.
    Accessed May 27, 2011, from http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?
    uri¼CELEX:32005L0029:EN:NOT
    European Commission. 2011. Directive /83/EU on Consumer Rights. Brussels: The European
    Commission.
    London: The Stationery Office.
    Great Britain. (2008xx). The Consumer Protection from Unfair Trading Regulations. London: The Stationery Office.)

    European Commission. 2011. Directive /83/EU on Consumer Rights. Brussels: The European
    Commission.
    London: The Stationery Office.

  8. How about announcing your paper before acceptance on Twitter?

    This has become a really big issue in ML, as authors use this for blatant self promotion and to circumvent the blinded reviews. Most reviewers are too junior (a massive lack of reviewers in a growing field means not enough seniors). Junior reviewers won’t reject a paper from a big shot.

    Am I the only one who loves self-plagiarism? I have zero issues with this, as long as the papers cite each other. In my field, I find it extremely helpful to read two slightly different descriptions of the same work. One problem is that most ‘good’ journals only accept short papers that are often difficult to read, so having a longer version in another journal gives the best of both world: slightly more stringent review in the better journal and an easy to read long version.

    • “having a longer version in another journal gives the best of both world: slightly more stringent review in the better journal and an easy to read long version.”

      Yes! I’m old enough to remember when it was pretty common in ecology for a Nature or Science paper to be followed up by a longer paper in an ecology journal. The follow-up paper would focus on different aspects of the results than the Nature or Science paper. The advent of online appendices put paid to that. Now instead of a well-written follow-up paper, you have to slog through a badly-written, poorly-organized appendix. I have an old post ranting about this: https://dynamicecology.wordpress.com/2015/10/26/online-supplements-have-ruined-nature-and-science-papers/

      “his has become a really big issue in ML, as authors use this for blatant self promotion and to circumvent the blinded reviews. ”

      I don’t really look at ecology preprints at all, and I’m not on Twitter, so I have no sense at all of how big an issue this is in ecology. Perhaps others who know more than me will comment.

      • You nailed it with your online supplements article.

        Hope your field gets spared twitter. It is just endless self-promotion where PIs post how excited/thrilled/delighted they are that they submitted N papers to Neurips this year. Or congratulate their students for their papers.

        Worse, now entire papers get turned into something like a blog, but posted on Twitter in 50 little fragments. There is even an app that assembles the fragments back into something readable.

        Back in the day, that stuff wouldn’t have been legal.

    • “Most reviewers are too junior (a massive lack of reviewers in a growing field means not enough seniors). Junior reviewers won’t reject a paper from a big shot.”

      This has a a (technically, probably not politically) “easy” solution, remove authors’ names when sending papers to reviewers, This kind of thing has a precedent, and I think it would benefit society greatly if it were adopted more widely:
      https://www.theguardian.com/women-in-leadership/2013/oct/14/blind-auditions-orchestras-gender-bias

  9. Pingback: Scientific fraud vs. financial fraud: is there a scientific equivalent of “control fraud”? | Dynamic Ecology

Leave a reply to Jeremy Fox Cancel reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.