Brykadurev | Building a Genuine Bear Case Before You Commit Capital

There is a well-documented tendency in human reasoning to seek out information that confirms what we already believe. In investment research, this tendency is particularly costly, because the stakes of being wrong are financial rather than merely intellectual. When a private investor builds a thesis around a company or sector, the natural impulse is to gather evidence that supports the case for buying. The result is a body of research that is, in effect, a prosecution without a defence. The bear case — the structured, honest account of why the investment might fail — is not an afterthought or a formality. It is the part of the analysis that forces you to confront the actual shape of the risk you are considering. Without it, you are not making a balanced judgement; you are making a wish with footnotes.

Constructing a genuine bear case requires asking a different kind of question than most research naturally generates. Rather than asking what would have to go right for this to work, you ask what would have to go wrong, and then you ask how plausible each of those things actually is. This means identifying the assumptions that underpin your thesis and testing each one separately. If your bull case depends on a company maintaining its competitive position, the bear case asks what a credible challenge to that position would look like, and whether the conditions for such a challenge already exist. If your thesis rests on a particular macroeconomic environment continuing, the bear case asks what happens to the investment if that environment changes. The discipline here is not to be negative for its own sake, but to be specific. A vague worry that things could go wrong is not a bear case. A detailed account of the mechanism by which things could go wrong, and the conditions under which that mechanism would activate, is.

Scenario analysis is the practical framework through which this kind of thinking becomes usable. Rather than trying to predict a single outcome, scenario analysis asks you to define a small number of plausible futures — a base case, a bull case, and a bear case — and to think through what each would mean for the investment. The value of this approach is not that it tells you which future will occur, because no analytical tool can do that. The value is that it makes your assumptions visible and comparable. When you lay a bear case alongside a bull case, you can begin to ask whether the potential downside in the bear scenario is proportionate to the potential upside in the bull scenario. You can ask whether the bear case is genuinely unlikely or simply uncomfortable to think about. You can ask whether your conviction in the bull case is based on evidence or on the fact that you have spent more time thinking about it. These are not questions that arise naturally from optimistic research. They require the bear case to exist as a real, worked-through document rather than a sentence acknowledging that risks exist.

One of the most useful habits a private investor can develop is to write the bear case before the bull case. This is uncomfortable, because it means beginning your research from a position of scepticism rather than enthusiasm. But the discomfort is informative. If you find it genuinely difficult to construct a credible bear case, that is worth knowing — it may mean the investment is more straightforward than most, or it may mean you have not yet looked hard enough. If the bear case writes itself easily and the scenarios it describes are severe, that too is worth knowing before you have committed capital rather than after. The goal of this kind of structured thinking is not to make you more cautious as a matter of temperament. It is to make your overall judgement more complete. An investor who understands the bear case thoroughly and still chooses to proceed is in a fundamentally different position from one who proceeds without having examined it. The difference is not in the outcome, which neither can control, but in the quality of the decision.