Journal file / AAS /September 13, 1999
TALK IS CHEAP (or Is It?)
Economics methodology prefers directly observable data such as prices and quantities and does not feel the need to ask people anything. Talk is discounted because people often rationalize their past behavior and don't do what they say they will do. But, is talk really without content and what are the real alternatives?
1. Quantity has a "talk" component. Quantity of what? There has to be a taxonomic term in there somewhere. What the analyst thinks the consumer bought may not be what the consumer thought. How determine this except by asking them? This has many applications such as understanding why consumers pay more for Jello than gelatin. Can't do industrial organization studies of undue product differentiation without talk.
2. If we are comparing two price times quantity outcomes caused by two different institutions, we can not understand their performance unless we describe the institutions. The subjects may not see the same two institutions as the analyst.
3. Data on labor use, expenditures, consumption, & production at the household level is obtained from questionnaires. Sure these contain prices and quantities, but of what? No alternative to talk.
4. We often use proxies for the variables identified by theory. The story that ties the proxy to the theoretical variable is talk.
5. Friedman's permanent income hypothesis makes a distinction between permanent and temporary income and suggests the consumption response. But this distinction is a mental construct. Is the analyst to decide what is permanent or must we talk to people?
6. The Phillips curve which examines the relationship between employment and inflation requires information of inflationary expectations. Expectations are talk.
7. To interpret consumer or producer spending we must know something of how people process information. So the standard methodology is to assume rational maximization of something. Thereby economists use talk, their own assumptions rather than asking people what they are doing. In spite of their protestations, economists use talk, but they prefer their own to that of their subjects.
8. To avoid talk is never to understand the Charlie Russell painting of the trail drive.