Behavioral economists, can people really know what they want?

Economists assume a normative standard of preferences for people. But what if those preferences don't exist?

Photo illustration/OJO Images/Newscom
A woman tries to decide which shoes to buy in a department store. Can people know themselves well enough to understand their true preferences?

At the turn of the new-year the Financial Times published two small articles about why people often do not adhere to their new year’s resolutions. One article was by a philosopher (Julian Baggini) and the other by a psychiatrist (Antonia Macaro). Interestingly, they each seem to focus on whether people really want what they resolve to do or not do. More fundamentally, the authors say, if people understood themselves better they would know more fully what their personal goals are and not have so difficult a time achieving them.

To put things in the words of the ancient philosophical question: If you know the better, will you do the worse?

Perhaps much apparent weakness of will is actually a failure to know oneself well enough to choose appropriate goals. As the philosopher says, “W]eakness of resolution could be a sign of weakness of conviction.” Thus, we are dealing not with a failure of will power but a failure to choose goals that really reflect what, given the agents values, would make life better.

I suggest that this is not an empirical question but a philosophical one. For one reason, there is no independent way of ascertaining when an individual has sufficient self-knowledge to really know the better for himself. So we cannot say: Joe knows what is better (by empirical test A) and yet he does not do or choose it (established by a separate empirical test B). Plato argued that failure to do what is best is failure to know truly what is best.

The error here is thinking that the individual is in a specific kind of disequilibrium in which he knows what he wants to do but cannot (for some reason) do it. Instead, perhaps, the individual is in a constant state of experimenting with his choices and is seeking something better but with no clear idea, ab initio, of what that might be. As both our psychiatrist and philosopher say, the individual will be in a better situation if he reflected more deeply on his values. And yet none of this, it seems to me, means that the outside observer can tell us or him what should be done. Doing and reflecting are part of the same action. It is a process of self-discovery.

This is not a comfortable position for today’s behavioral economists who want to establish normative preferences based on what individuals really want at some deep level. As the economist Frank Knight argued, the only thing we really want at a deep level is better wants. But they never simply arrive in stable form; the process is personal, uncertain and never ending.

The root of the problem may be that economists (behavioral and standard) have allowed a mental construct of preferences – either simply revealed by choice or arrived at by some alternative means – to act as a normative standard. It is true that behavioral economists have argued that preferences are not fixed but sensitive to all sorts of bias-inducing factors like framing, visceral influences on decisionmaking, temptation, and so forth. And yet unless they want to give up subjective value theory entirely, they must privilege some expressions of preference – those made under idealized circumstances of “full” knowledge, perfect cognitive abilities, and no weakness of will. This is their gold standard by which the choices of agents are to be evaluated. But is there anything there? Do such preferences exist?

While this framework may be useful for certain scientific problems of prediction, it does not seem to capture the process of decisionmaking through time. It is rather an attempt to adapt a static conception of choice (fixed preferences, fixed constraints) to internal dynamic problems. People may be dissatisfied with their own behavior and may be seeking to understand what might be better as they choose and act. I do not see any compelling reason to transform a simple construct that is useful for some purposes into a normative standard. If the construct doesn’t work in certain cases to predict behavior, then so much the worse for the construct.

Remarkably, behavioral economists who have spent much time and trouble to show the positive limits of the standard analytical apparatus are saying, “So much the worse for the troublesome individuals.”

This is not the last word on the subject by any means. But it does demonstrate the mental trap into which even smart people can get themselves when they fail to understand that models are constructions useful for some purposes and naive for others.

Add/view comments on this post.


The Christian Science Monitor has assembled a diverse group of the best economy-related bloggers out there. Our guest bloggers are not employed or directed by the Monitor and the views expressed are the bloggers' own, as is responsibility for the content of their blogs. To contact us about a blogger, click here. To add or view a comment on a guest blog, please go to the blogger's own site by clicking on the link above.

of stories this month > Get unlimited stories
You've read  of  free articles. Subscribe to continue.

Unlimited digital access $11/month.

Get unlimited Monitor journalism.