Sunday, March 13, 2011

Must Economists Start Reading Freud?

From reading this review of David Brooks' new book, it seems that he is thinking too hard.   NYU's Nagel writes in his review; "Brooks is right to insist that emotional ties, social interaction and the communal transmission of norms are essential in forming individuals for a decent life, and that habit, perception and instinct form a large part of the individual character. But there is moral and intellectual laziness in his sentimental devaluation of conscious reasoning, which is what we have to rely on when our emotions or our inherited norms give unclear or poorly grounded instructions."

Are people aware of their own strengths and weaknesses and how this list of attributes affects their likelihood of achieving their life goals?  If the answer is no, then they need to hire a "life coach".  Los Angeles has many including Meredith Haberfeld.   If your answer is "yes", then you will be aware that you can make a series of investment choices at any stage of life to improve yourself.

Without reading his book, David Brooks appears to offer a vague, popular version of Jim Heckman's recent work on how a "quality human" is produced.  In Heckman's vision, building up both non-cognitive skills and cognitive skills are both crucial to achieve success in the modern urban capitalist economy.   For an interview with Jim Heckman click here.

David Brooks appears to want to expand how social scientists study human behavior.  While it is exciting to introduce new causal factors, his approach would make hypothesis testing even harder to do as he introduces many new psychological theories based on data that social scientists will have trouble collecting.  To offer a cliche example from "Freudian theory" --- suppose that researchers claim that whether a  mommy smiles at me when I was age 2 plays a crucial role in determining whether I have self control or not at age 45.  Given the challenge in building longitudinal data sets, how will we ever test this claim?   What empirical researchers can do to study "cohort effects" is to see whether those who grew up during the influenza epidemic of 1918 or during the great depression act differently than roughly similar cohorts who missed these "macro" episodes. In this case, we can identify the subset of people at risk for such nasty treatments.

My vote is that we keep economics "simple".  We know who we are, we know what we want, we know what markets will sell us stuff.  There is fundamental uncertainty. We don't know when we will die. We don't know how the business cycle will affect us or what the stock market will do in the near future but we know that we don't know. We know that the world is risky and that climate change is unfolding but we form our best guesses about the future and keep reading to update our uncertainty and plan accordingly.  If we know that we suffer from some of the afflictions identified by behavioral economists, then we engage in various "tricks" and commitment devices to protect ourselves from ourselves.   This vision of economics means that we can formally test hypotheses but we will never have a "perfect" model of explaining and predicting human behavior. David Brooks appears to want to introduce too many parameters to "explain" our choices.

I much prefer the Gary Becker stripped down "household production function" approach.  We recognize that we face a time budget constraint and monetary budget constraint. We have a well stated conception of the good life (quality friends, quality children, comfort) and use our time and market inputs to produce them.  This framework has taken us quite far.