Friday, April 28, 2006

Fundamental Attribution Error

In the psych class I've been auditing, Steven Pinker yesterday talked about a phenomenon called the fundamental attribution error. When evaluating others, people have a tendency to overestimate the importance of personal characteristics and underestimate the role of situation.

Example:
Why am I smiling? Because it is a sunny day.
Why is he smiling? Because he is a cheerful person.

I wonder if this common error can help explain some unfortunate impulses in economic policy.

Example:
Why did I raise my price? Because demand increased more than supply.
Why did the gasoline station raise its price? Because oil companies are greedy price gougers.

Similarly, in judging policymakers, perhaps we give too much credit to those who were simply lucky and too much blame to those who were unlucky. In this paper, I did not refer to the fundamental attribution error, but I was following its logic:

If you were to poll monetary historians, most of them would tell you that Alan Greenspan is a hero among central bankers and that Arthur Burns is a goat. Just as Greenspan gave us low and stable inflation, together with robust and stable growth, Burns gave us high and rising inflation, together with anemic and volatile growth. The standard assessment of these two men is easy to understand.

Yet, in looking back at these polar two experiences, I wonder whether we exaggerate the role of policy decisions and understate of role of luck. One reason is that the bad inflation performance of the 1970s and the good inflation performance of the 1990s were not limited to the United States. Most developed countries had about the same experience. If there was a policy failure in the 1970s and success in the 1990s, the blame and credit go to the world community of central bankers, not to the single person leading the Federal Reserve.

I suspect, however, that the difference cannot be fully explained by policy at all. These two eras saw very different exogenous supply shocks. The relative price of food and energy was extraordinarily volatile during the 1970s and extraordinarily tame during the 1990s. The standard deviation of this relative price differs in these two decades by a factor of almost three. (Table 1.3, Mankiw 2002) Moreover, the 1970s witnessed an unexpected slowdown in productivity growth and an increase in the natural rate of unemployment, whereas the 1990s witnessed an unexpected acceleration in productivity growth and a decline in the natural rate of unemployment. The favorable supply-side developments of the 1990s were not caused by monetary policy, but they did make the job of monetary policymakers a lot easier. Luck plays a large role in how history judges central bankers.

The same could be said of Presidents. I have long thought that Bill Clinton gets too much credit for the booming economy of the 1990s, and Jimmy Carter gets too much blame for the lousy economy of the 1970s. Now I have a term for it: the fundamental attribution error.

Update: Arnold got here first.