You are the U.S. President. After the hoopla of the inauguration, you settle into your work focused particularly on implementing your policy agenda, on which you were elected. The success of your campaign breeds triumphalism that energizes your work. Then, within weeks of your inauguration, a financial crisis hits. What should you do?

The Miller Center at University of Virginia focuses on the U.S. presidency and is conducting a series of seminars on the President’s first year, for which it has commissioned a range of essays on potential issues. (See: First Year 2017: Where the Next President Begins.) They asked me to write about the possibility of a financial crisis in the first year.

You can read my essay, “Grab hold of the levers,” here and a shorter version on Fortune’s web site, there. I won’t spoil the message by repeating it in this post. Suffice it to say that historical precedents suggest that financial crises and Presidents don’t mix well. And the difficult mixture happens more often than you’d guess. Eight of the 44 Presidents faced a major financial crisis in the first year of their first term, some at or within days of inauguration. Nearly half of the Presidents faced a major domestic financial crisis anywhere within their first term. And almost 60 percent faced a domestic or international crisis in their first term. In short, history suggests that the risk of a financial crisis is material. The President should prepare. My essay is about what it means to prepare.

I’ve blogged previously about the importance of studying history to the development of future leaders (see this and my annual lists of recommended readings, such as this). History is a powerful tool for framing one’s own experience and for anticipating what the future might hold. Confucius said, “Study the past if you would define the future.” Humans are naturally given to reasoning by historical analogy.

Yet reasoning by historical analogy warrants at least two cautions.

First, the historical experience of oneself or one’s community is usually so impressive as to assure that whatever happened WILL happen again (i.e., with high probability). In financial markets, this tendency to project the past into the future is shown to explain why investors tend to overshoot or undershoot in their pricing of assets—and why bubbles and panics occur. Amos Tversky and Daniel Kahneman described a bias or “heuristic” by which people project an outcome (incorrectly) based on a small sample. People tend to see patterns where none actually exist—this is especially true about patterns of prices in financial markets. This representativeness heuristic draws its power from the individual’s observation and past experience. Another phenomenon, conservatism, reflects the fact that people tend to be slow to change their beliefs in the face of conflicting evidence. Representativeness and conservatism can make a potent brew, by which as Andrei Shleifer wrote, people “[underreact] to individual pieces of information, but [overreact] to conspicuous patterns.” ((See: Andrei Shleifer, Inefficient Markets, Oxford University Press, 2000, page 129.)) Psychological biases are a more or less forgivable error and can be corrected by rigorous assessment of one’s assumptions.

Second, the less forgivable error is to tilt reasoning by historical analogy for some ulterior purpose, such as to justify an unjust war, to exalt a bogus nationality, to validate genocide, or to excuse theft. In her book, Dangerous Games: The Uses and Abuses of History, Margaret MacMillan writes,

Political and other leaders too often get away with misusing or abusing history for their own ends because the rest of us do not know enough to challenge them…Bad history tells only part of complex stories. It claims knowledge that it could not possibly have, as when, for example, it purports to give the unspoken thoughts of its characters…Bad history also makes sweeping generalizations for which there is not adequate evidence and ignores awkward facts that do not fit….Bad history ignores…nuances in favor of tales that belong to morality plays but do not help us to consider the past in all its complexity. The lessons such history teaches are too simple or simply wrong. That is why we need to learn how to evaluate it properly and to treat the claims made in its name with skepticism. ((Pages 36-37.))

These cautions are especially relevant to us in early 2016. Political candidates are wont to make passionate assertions about the past in support of aspired policies. Leaders in government and the financial services industry appeal to history (especially the recession of 1937-38) in support of various proposals to jump-start growth in the global economy right now. Today’s military tensions are compared to geopolitical pivot points in the past.

My advice: take a deep breath. Reasoning by historical analogy is useful in the way that scenario planning is useful: it illuminates the possibilities and dynamic dependencies; it invites critical evaluation of assumptions; and it can promote preparedness. And as long as you respect the two cautions (about biases and intentions) it can be fun.

Thus I offer my essay about how the President should address a financial crisis in the first year, in the spirit of Margaret MacMillan, who wrote:

History, by giving context and examples, helps when it comes to thinking about the present world. It aids in formulating questions, and without good questions it is difficult to begin to think in a coherent way at all. Knowledge of history suggests what sort of information might be needed to answer those questions. Experience teaches how to assess that information….What happened and why? The historian asks. History demands that we treat evidence seriously, especially when that evidence contradicts assumptions we have already made. Are witnesses telling the truth? How do we weigh one version against another? Have we been asking the right or the only questions? Historians go further and ask what a particular event, thought, or attitude signifies. How important is it? The answers in part will depend on what we in the present ask and what we think is important. History does not produce definitive answers for all time. It is a process. ((Page 167.))

The Miller Center’s project, First Year 2017, is exactly that: a valuable process to generate penetrating questions by which all of us can set expectations for the new administration.