A PROTOTYPE OF DIALOGUE WITH ECONOMISTS REVEALING MALPRACTICE OF THEIR ENTIRE PROFESSION

One of the best examples of dialogue with economists regarding their malpractice was illustrated in a conversation in late 1972 between myself and a staff economist of the CEA at the White House.

It seemed to me to be such an important conversation that it was confirmed by letter and I secured permission to paraphrase our conversation and report it without attribution as follows:

• "Burress: Why did the CEA forecast that the saving ratio would not come down in 1972?
• "CEA economist: We were expecting a fairly sharp increase in income and it was a matter of the simple textbook case. That is, when income rose rapidly, you expect saving to rise even more rapidly, pushing the saving ratio higher. It was so high in 1971 it might not go higher. But we had no reason to think that it would come down in 1972.
• "Burress: What is your evidence that, in fact, the textbook case describes the real world? What would be the case for planning policy on the basis of the textbook model?
• "CEA economist: You look to the regression coefficient in econometric models of the consumption function. These show that when income rises rapidly, saving rises even faster and when income falls in a recession savings falls even faster. You can see this in any well-known econometric model.
• "Burress: I am aware that this is found in all the well-known econometric models. But have you looked at specific earlier recovery periods when income rose as rapidly as was expected for 1972? That is, are you aware of what happened to saving as income rose rapidly in 1955, 1959, 1968?
• "CEA economist: No, I am not. The regression coefficients in econometric models provide the best description of the past that is available to economists. I know they show saving rose in recoveries in recent years. But I have not looked at what happened to saving in these specific years and I doubt if any other economist in the CEA has either.
• "Burress: Well, my approach is to look at the specific years. The fact is, the absolute level of saving fell as income rose in 1955, 1959, and 1968. And it looks like that’s going to happen again this year. How can you reconcile these data for individual years with the econometric estimates of the past?
• "CEA economist: You are always going to have some years that depart from your model.
• "Burress: You also mentioned that the econometric models tell you that saving falls in recessions. Are you aware that saving rose, rather than fell as you suggested, in 1958, 1961, and 1970 recessions as well as in the 1967 mini-recession? As I read history you have a pattern here. The econometric model systematically misses even the direction of change of saving in recessions and recoveries since the mid-1950s.
• "CEA economist: If there were such a pattern, it would be picked up in the error term of the econometric model. The econometric models tell us what happened, fundamentally, despite all these 'wiggles' in the data..."

Years of efforts to see the theory of consumer behavior reformulated leaves me convinced that the problem illustrated here is the most serious of all impediments to change. Some may conclude that these comments represented the approach, not of the economics profession, but of this particular CEA economists. But most economists, like the former CEA chairman with whom this discussion took place, will recognize this as a predictable dialogue.

Source: Glenn E. Burress, (testimony), The 1975 Economic Report of the President: Hearings Before the Joint Economic Committee Congress of the United States, March 12, 1975, p.1065.