≡ Menu

CPI Bias

In this earlier post, I mentioned work by Jerry Hausman and Ephraim Leibtag, and Mark Bils on why the CPI and inflation are overstated and therefore the standard of living of the average American is understated. Here are two relevant quotes from their work that gives you a feel for the magnitudes involved. The first is from the Hausman/Leibtag paper that discusses how the failure to include Wal-Mart Superstores in the measure of food costs in the CPI biases it upward:

Hausman (2003) discusses four sources of bias in the present calculation of the CPI. A "pure price" index based approach of surveying prices as done by the BLS cannot succeed in solving the problems of bias. We discuss economic and econometric approaches to measuring the first order bias effects from outlet substitution bias. We demonstrate the use of scanner data that permits implementation of techniques that allow the problem to be solved. In contrast, the current BLS procedure does not treat correctly outlet substitution bias and acts as if Wal-Mart does not exist. Yet, Wal-Mart offers identical food items at an average price about 15%-25% lower than traditional supermarkets. The BLS "links out" Wal-Mart’s lower prices. We find that a more appropriate approach to the analysis is to let the choice to shop at Wal-Mart be considered as a "new good" to consumers when Wal-Mart enters a geographic market. This approach leads to a continuously updated expenditure weighted average price calculation. We find a significant difference between our approach and the BLS approach. Our estimates are that the BLS CPI-U food at home inflation is too high by about 0.32 to 0.42 percentage points, which leads to an upward bias in the estimated inflation rate of about 15% per year.

The Bils paper looks at what happens when a manufacturer of a consumer durable (cars, televisions, computers and so on) drops one of the models in their product line and replaces it with another. The BLS has been following the price of the old model and now it has disappeared. They replace it with another model that is "similar." But if the new model is improved in quality (as it often is), the price of that new model is not really comparable and the measured prices will overstate inflation. The BLS does try to control for this effect, but Bils tries to measure how effective those controls are. He finds that in many cases, they are inadequate. Here’s an excerpt from Bils’s intro:

Much of economic growth occurs through growth in quality as new models of consumer goods replace older, sometimes inferior, models.  Moulton and Moses (1997) estimate that Bureau of Labor Statistics (BLS) methods allowed for perhaps as much as 1 percent average quality growth in goods in 1995.  It is often argued, however, that the Bureau of Labor Statistics (BLS) methods miss much of the growth in goods’ quality.  (Hausman, 2003, and Pakes, 2003, are two recent examples.  Shapiro and Wilcox, 1996, review much of the previous evidence.)  The Boskin Commission Report (1996) suggests that the BLS overstates inflation by perhaps one percent per year.  Unmeasured growth in quality of goods is put forth, based on examining a fairly limited set of goods, as the most important component contributing an overstatement of inflation of 0.6 percent per year.   The need to measure quality growth arises from changes in the products available to consumers. To calculate the CPI the BLS tracks the price of a specific product at a specific outlet.  The products followed change for two principle reasons.  At scheduled rotations, roughly every four years, the BLS draws a new sample of stores and products within a geographic area to better reflect current spending.  In addition, a store may stop selling the particular product being priced.  The BLS agent then substitutes another model of that brand or a similar product.  These (forced) substitutions occur on average about every three years for all non-housing CPI items.  They occur much more frequently, nearly once per year, for consumer durables.

And from his conclusion:

For vehicles my results suggest that quality growth has been understated by at least three percent per year.  My results suggest growth for consumer electronics has also been substantially faster than historically measured, by two percent or more per year. For the balance of durables, if price inflation is reasonably well captured by the rate of inflation for matched products, then annual quality growth is about one and one-half percent faster than suggested by methods underlying BLS measurement of the CPI.  Putting it all together, this suggests quality growth that is two-and-half to three percent faster per year than has been measured for these durables, with a resulting growth rate of at least five percent per year even with computers excluded.  My results suggest much faster quality growth for durables than the Boskin Commission Report, which stated BLS methods understated quality growth for durables by about one percent.  This derives from two main differences: (1) The Boskin Report assumed no quality bias for a number of durables; (2) The Report assumed no quality growth for vehicles except from greater durability.

These are not trivial differences. If Hausman and Leibtag, and Bils are right, the CPI we use to deflate wages and earnings is seriously overstated.

Again, I challenge the pessimists, the doom-and-gloomers who accept the wage data as indicating that for 30 years, the average American has made no progress. Either our economic system is broken or are data collection system is broken. There is dramatic evidence that the data collection system is unable to cope with the speed of change in the US economy. Why do you believe instead that it is the economic system that is broken? Where is your evidence?