Does The Cpi Inflate Inflation?

Robert Kuttner's column "Why it's too soon to monkey with the CPI" (Economic Viewpoint, Nov. 27) leaves the impression that I agree with the views of Dean Baker [of the Economic Policy Institute], which I do not. Since Kuttner did not speak to me directly, he must be referring to my June 6 testimony before the U.S. Senate Finance Committee, where I did, indeed, express reservations about the empirical underpinnings of the whole debate. But I also indicated my belief that, on balance, the consumer price index did overstate inflation in recent years by a significant amount.

Since the time of the testimony, the Advisory Commission, of which I am a member, has met several times, reexamined the evidence, and concluded unanimously in its interim report that the recent rate of overstatement of price growth by the CPI was on the order of 1.5% per year, with a range of reasonable uncertainty from 1% to 2% or higher. There was more evidence on this than I had realized at the time of my testimony, most of it actually discovered and provided by the Bureau of Labor Statistics itself.

The commission is pursuing this topic further, not only to improve the quality of this particular "number" but also to investigate workable suggestions of how the BLS could improve the CPI and make it into a better measure of the cost of living, a goal that we share with them.

Zvi Griliches

Professor of Economics

Harvard University

Cambridge, Mass.

Before it's here, it's on the Bloomberg Terminal.