Risk Policy

Risk Policy

Reviewed by Faisal Zaidi

A risk policy is Daniel Kahneman's name, in Thinking, Fast and Slow, for a broad frame applied as a routine. A person with a risk policy does not weigh each risky choice as a new and separate problem. The choice is placed among the others of its kind and settled by a rule made in advance. Kahneman's examples are always taking the highest possible deductible when buying insurance and never buying extended warranties.

What it is

The book's chapter on risk policies sets out the problem first. People who are prone to narrow framing construct a preference every time they face a risky choice. The book's answer is to decide once about the whole class of choices, and then apply the decision whenever a relevant problem comes up.

Kahneman pairs the idea with the outside view from his chapters on planning. The outside view shifts attention from the specifics of the case to the statistics of outcomes in similar cases. A risk policy does the same for gambles: it embeds a particular risky choice in a set of similar choices. Both are broad frames, one for forecasting and one for risk.

The term appears in the book as Kahneman's own usage. This vault has not traced it to an earlier source, so the page does not credit it further back than that.

In effect

The book's starting case is Paul Samuelson's coin toss. A friend would not take a single bet to lose $100 or win $200, but would take a hundred of them. The hundred together have an expected return of $5,000, a 1 in 2,300 chance of losing any money and a 1 in 62,000 chance of losing more than $1,000, according to the figures the book credits to Matthew Rabin and Richard Thaler. A person with a policy for such bets does not have to wait for the aggregate to be offered.

Kahneman also gives a short mantra for the broad frame, that you win some and lose some.

A related experiment, which the book describes without a result reported on this page, asked students to treat gambles either as the only decision of their kind or as the work of a trader who makes such decisions all the time.

What it does not say

It does not say the mantra works everywhere. The book gives it three limits. The gambles must be genuinely independent of each other, so it does not apply to several investments in the same industry that would all go bad together. A possible loss must not cause worry about total wealth. And it should not be applied to long shots, where the probability of winning is very small.

It is not an experimental finding. The risk policy is a recommendation from a trade book, built on experimental findings about narrow framing, and its advantages are argued in the book rather than measured. The page treats it as a prescription.

It does not mean never deciding case by case. A policy is a default for a class of similar choices, and the limits above mark where a case may not belong to the class.

Sources

  1. Kahneman, D. (2011). Thinking, Fast and Slow. New York: Farrar, Straus and Giroux, as recorded in the vault's reading digest of the book.
  2. Kahneman, D., & Lovallo, D. (1993). "Timid choices and bold forecasts: A cognitive perspective on risk taking." Management Science, 39(1), 17 to 31. doi:10.1287/mnsc.39.1.17.