Sarah Sloat: What is Cognitive Economics? Understanding the World Through New Types of Data

Link to “What is Cognitive Economics? Understanding the World Through New Types of Data” on

I had an interesting discussion with Sarah Sloat a week or two ago about Cognitive Economics, reflected in the article “What is Cognitive Economics? Understanding the World Through New Types of Data” on This is quite an interesting article, but several things became garbled. Let me try to clarify and emphasize a few things:

A. Explaining Nonstandard Behavior

In the passage beginning 

Historically, the first thing that a behavioral economist did was try to document the things people do when their actions look strange from the standpoint of standard economic theory

what I said was that Behavioral Economics has taken as its first task to identify behavior that seems at variance with standard economic theory. Once such behavior is identified I think it makes sense to seek explanations in the following order:

  1. Some deeper explanation using standard economics that was not immediately apparent. 
  2. An explanation based on cognitive limitations or cognitive confusion.
  3. An explanation based on exotic preferences. By “exotic” I mean something like loss aversion or hyperbolic discounting. I don’t count altruism or caring about a wide variety of aspects of well-being (such as happiness, sense of purpose, power, etc.) as exotic preferences.

In the history of thought, many of the early influential behavioral economists reversed the order of 2 and 3.

B. The Goal of Economics

In the passage beginning

Like any scientific discipline, one of the jobs of economics is to understand how the world works

what I said was that beyond the scientific goal of understanding how the world works, economists in their role as policy advisors, have taken on the task of trying to smooth the way for people to get more of what they want.

C. Imperfect Information vs. Imperfect Information Processing

Sarah reports me as saying 

That’s certainly an element. If people don’t know something — what economists call imperfect information — we now have models that are very good with dealing with that imperfect information processing.

I actually said the opposite. We now have very good models for dealing imperfect information, but our tools for dealing with imperfect information processing are still quite rudimentary–primarily because of the infinite regress problem I discuss in my paper “Cognitive Economics.” 50 years ago, economists didn’t realize how much easier it would end up being to model imperfect information than it is to model imperfect information processing. And some of the ways we do have for modeling imperfect information processing model it in a way that, from a formal mathematical point of view treats a problem of imperfect information processing quite imperfectly as if it were a problem of imperfect information. I am thinking for example of the approach of Greg Mankiw and Ricardo Reis, in which what I would describe as a failure of information processing–not going to the effort of using readily available information–is described mathematically as if it were a failure of information availability–not having the information. 

D. Research with Dan Benjamin and Ori Heffetz

Where the article mentions Dan Benjamin, my coauthor Ori Heffetz should be mentioned in the same breath. Much of our joint research is described in posts in my Happiness sub-blog. I have many other cognitive economics coauthors as well–enough that I won’t try to list them all here. 

E. How Cognitive Economics Can Help Save the World

There are two different directions in which I think Cognitive Economics can help to make the world a better place. The first is in helping people deal with their cognitive limitations and in taking account of people’s cognitive limitations in designing public policy–such as in the work of the Consumer Financial Protection Bureau. The second is in using subjective well-being data–data on how people feel and how they think they are doing in various aspects of their lives–to help identify previously unsuspected ways to make people’s lives better. Because a big part of politics is energizing voters to give money and to vote, politics often sets out a very narrow agenda of controversial policies. I have the hope that using subjective well-being data to zero in on what people want will be able to shine a light on changes almost everyone can agree on.