What are the consequences of betting on one's own research?

Question

(Details fictionalized.) My research involves a new way of forecasting the weather. Unlike other forecasting models, the model I have developed says that the weather in a certain region will be amenable to farming next year. Before I post the preprint of my paper, I am considering "betting" on my model by investing in some agricultural stocks in the region in question.

Would placing a bet on my research have a negative effect on my ability to publish down the line?

What other ethical and legal issues should a researcher consider before making a bet like this?

Are there historical examples of scholars making bets like this, and how did it play out?

Additional details:

• I recognize that being financially staked in the correctness of my research yields a conflict of interest, and I would report this when posting the preprint and seeking publication.
• This question concerns the consequences of such a bet for me as a researcher, not whether placing such a bet is a wise financial decision. In this hypothetical scenario, I would not invest more money than I am willing to lose. However, if the size of the bet changes the ethics, then you may note this in your answer.

---

Edit: I greatly appreciate the interest this question has attracted, but I haven't selected an answer yet because most of the answers so far (with a few exceptions, which I have gratefully +1ed) have focused on the question of whether there is a conflict of interest rather than on the question in bold, which is what the consequences of the conflict of interest would have for my ability to publish.

The ethical debate can be summarized as follows: "There is nothing wrong with betting on yourself being right; if anything it lends you credibility" vs. "Stock prices are themselves indicators of beliefs rather than morality, so the situation described still creates a conflict of interest." As commenters have pointed out, this debate ultimately turns on the details of the circumstance, the nature of the financial products, and so on—details I have not provided. However, for the purposes of this SE question, the fact that this bet creates a conflict of interest is given. That is, assume that if my research is widely accepted, I stand to make a (very small) profit, regardless of whether the particular predictions pan out. How would, for example, a paper referee react to such a conflict of interest?

I concede that this may be a more banal question, but it also keeps us more safely within the margins of academia.SE's "no opinion questions" rule. Possible answers could include

• Everyone pursuing an academic career is, to one extent or another, "betting on their research," and throwing a stock purchase into the mix doesn't rise above the threshold of suspicion.
• In principle, monetary instruments can create conflicts of interest, but there are ways to clarify to the referees (how?) that you are betting a small amount of money, and doing so in a way that is maximally correlated to the correctness of the results rather than people's expectations, which could be swayed by the impact of the research. As long as you make this clarification, then it's no big deal.
• Even if there is not an ethical issue in the philosophical sense, any sort of monetary bet creates the appearance of a conflict of interest, and for that reason it will be harder to publish this paper in a high-quality journal.

In my own field (which, as some of you have figured out, has nothing to do with the climate or farming), I have seen a paper published in a peer-reviewed journal that argued for the efficacy of a certain sampling methodology, and then included on the final page a disclosure saying that the study was funded by a company whose primary revenue source is performing the very same sampling methodology. In this case, it was very easy to draw an arrow from the funding source to the research outcome, but the paper was transparent and well argued. In a situation like this, does the paper ultimately get published because it's good enough to overlook the conflict of interest, or because the conflict of interest was commonplace enough not to warrant additional attention?

Answer 1

I think some (but not all) of the answers here are over-interpreting your use of the word "bet" to focus on some more trivial cases: the sorts of bets one might have among friendly rivals of different sports teams, for example. Petty cash, bottles of liquor, risking gastrointestinal distress by drinking something one wouldn't normally.

However, the question as asked is not about these trivial bets but about having a financial stake in the matter. This changes things into a potential conflict-of-interest scenario. The key in this scenario is disclosure: you'll need to reveal your position (possibly even if it is a future planned position, though that might spare you from some legal if not ethical rules).

Yes, these conflicts may impact whether a journal wants to publish your paper as well as how others view it. Of course, if you turn out to be correct, then it will be hard for others to see the research too negatively. On the other hand, if you turn out wrong, you might look bad beyond the cost to your pocket: it may appear, whether intended or not, that the paper was just a ploy to influence the investment patterns of others to your benefit. In other words, consider that you may be making more than just your financial bet.

These issues come up in the pharmaceutical literature all the time, though more often in the form of direct payments (unless the authors are actually employees of the company). I don't think there is anything inherently wrong in papers being funded by a company with an interest in the results (and I think there is scientific benefit to having those papers released publicly rather than being used only for internal decision making, or selective release of only positive results), but when some product has a bunch of mixed literature from the broader scientific community and only rosy publications from those with a financial stake, well, it doesn't look that great: for the company, product, or authors.

However, peer review should always be focused on a specific submission, not the past. Dabbling a bit in the sort of thing you discuss doesn't seem like it would be harmful long-term; building a reputation as a charlatan is another matter.

Answer 2

What you’re describing is actually not that unusual. There is a sort of tradition among physicists and mathematicians to make bets on whether a research prediction they are making will turn out to be true. These bets are usually made with colleagues who hold a conflicting opinion. The stakes are usually fairly low (a bottle of wine, meal at an expensive restaurant, etc) but sometimes involve large amounts of money, for example $10,000 in a very recent example.

In terms of the ethics, I don’t see a problem (and neither does anybody else as far as I can tell) with betting that your prediction will turn out to be true — after all, even those of us who don’t make explicit bets are effectively invested in the outcome of our predictions. If the predictions work out we gain reputation, prestige etc, which can lead to job promotions, awards, and other material rewards. But the incentives are aligned correctly: we only get the rewards if we make a correct prediction, and the scientific community actually wants researchers to try as hard as they can to make correct predictions. So despite what some people here are suggesting, this situation is not a conflict of interest (at least in a generic example, and in your particular example; perhaps someone can come up with a special case where there would indeed be a conflict).

Now, on the other hand, if you were involved in some weird scheme to bet against your prediction turning out to be true, then that would be a genuine conflict; this is analogous to various scandals that occurred in professional sports where players made bets against their own teams. From some googling I did it seems that the major professional sports leagues explicitly forbid their players from betting against their own team, and with good reason. So don’t do the scientific analogue of this practice.

Answer 3

From my perspective, the conflict of interest arises if you go public with your research before the time period in question passes. In this case, if the public believes that your prediction is correct, that information may influence the market of the underlying agricultural stocks. You might then sell the now inflated stocks to make a profit. If the information you make public is materially false, that might be considered market manipulation which is against multiple federal regulations in the US (insert disclaimer about not a lawyer).

If you (very correctly) disclose this conflict, an editor might take this into consideration when making a decision. Conversely, research conducted by commercial entities with direct financial conflicts of interests are published in high impact journals all the time. Disclosing the conflict adequately is the key.

However, if the investment was made historically, and the time period predicted is in the past, I am not sure there is a conflict.

Also note that many institutions (including the National Institutes of Health) have guidelines on reporting ownership of particular securities.

Answer 4

Would placing a bet on my research have a negative effect on my ability to publish down the line?

You are required to declare conflicts of interest, but it's improbable that a bet would negatively affect your ability to publish. It would make the others examine your paper more closely, but if your work is sound it should still be publishable.

Are there historical examples of scholars making bets like this, and how did it play out?

Sure, see Timothy Springer, a Harvard biochemist who became a billionaire investing in biotechnology stocks. His lab is still publishing a lot of papers, as well.

Answer 5

Are there historical examples of scholars making bets like this, and how did it play out?

Although not entirely analogous to your situation, there are some well-known cases where academics have made bets on outcomes that relate to their research work and academic hypotheses. Usually this happens when two academic antagonists get pissed off with each other's contrary pontifications, until one finally says to the other, "Care to put your money where your mouth is?"

A famous example of this is the Simon-Ehrlich wager between the economist Julian Simon and the biologist Paul Ehrlich. Prior to the bet, Ehrlich had long argued that growing human population would lead to resource scarcity and other global catastrophes, whereas Simon had argued that resources would generally become more plentiful and cheaper. Simon challenged Ehrlich to choose any raw material and they would observe the inflation-adjusted price on an agreed date more than a year away. The pair ultimately agreed to bet on the inflation-adjusted prices of copper, chromium, nickel, tin, and tungsten, at an end date approximately a decade from the initial bet. All five materials declined in real price over that time and so Simon won the best convincingly.

Answer 6

Talk to a Hedge Fund

Why make a small bet? If you really have a meaningfully better way of predicting the future (weather) then there are companies that will be willing to make very large bets, although they'll probably prefer to call them investments. This is what many of these firms do day in day out. There need be nothing inherently unethical about it as long as you disclose your interests where appropriate in the publication process, but then you may decide you don't want to publish... which raises a different set of ethical issues. If all goes well you'll be able to ponder them from your Yacht.

Answer 7

You've done your research. You're not placing a bet on it. You're placing a bet on some stock or commodity performance.

In research, the ethics comes down to conflicts -- reported or unreported. If you design a drug, say, and have IP rights to that drug, readers have reason to suspect that your financial interest creates a situation where you might highlight good findings and suppress poor findings. Thus, you would need to declare your conflict and manage it. In this case, you may be blinded from your own findings, or an independent monitor of your data may need to do any analysis before you do. Readers would be made aware of the conflict and mitigating strategy.

Here, it doesn't feel like you would be creating a conflict. You're not trying to generate cherry-picked data for your model. You're trying to make money. If you, however, were going to write about your model, and you wanted to publish your experience to make your model more valuable for licensing purposes, you may well be creating a conflict.

You might discuss this with your chair (I'm not a lawyer), and see if you need to update any of your conflict reporting forms.

Answer 8

I think one additional point of view, not taken in the previous answers, is to start from a definition of a conflict-of-interest. There are of course many definitions, but this one seems to appear in slightly different forms in Elsevier's journals:

When an investigator, author, editor, or reviewer has a financial/personal interest or belief that could affect his/her objectivity, or inappropriately influence his/her actions, a potential conflict of interest exists.

Now, first, does this betting influence your objectivity? Most probably not. There is no incentive for you to be wrong, and the betting takes place after you have performed rigorous research.

Second, do you have a personal interest that affects your actions? The answer to this questions seems to be yes. However, are these actions inappropriate? For more context, inappropriate would here usually mean having some kind of dual commitment:

Conflict of interest exists when an author (or the author’s institution), reviewer, or editor has financial or personal relationships that inappropriately influence (bias) his or her actions (such relationships are also known as "dual commitments", "competing interests", or "competing loyalties").

This should clear up the issue. Whatever happens during peer-review / post-publication, you must first be loyal to the research. For example, you must not insist on not making changes the peer-reviewers or editors ask, if the reason is that it means you are suffering monetary losses.

Thus, I would say this does influence the publishing process. More often that not manuscripts are heavily edited after being first submitted for publication; the editor and reviewers will be unsure whether you can be trusted in making requested changes. Declaring the conflict-of-interest of course does give you +1 but it is still easy to go below zero.

Answer 9

If your paper demonstrates a way to predict company bankruptcies for example, and you short companies which your technique predicts to go bankrupt prior to your paper being published, then yes I think that would be classified as unethical. Specifically if you were intending to close the short on any subsequent dip in prices (as other people presumably flocked to short based on the new technique), rather than waiting for a bankruptcy.