blog




  • Essay / CPI Bias - 1035

    The accuracy of the CPI (consumer price index) as a measure of the rate of increase in the cost of living has long been the subject of controversy scrutiny, with many studies showing that it overestimates the numbers - although by different amounts. extents. One of them, “Measurement error in the consumer price index: where are we?” (2003) by David Lebow and Jeremy Rudd, provides a comprehensive analysis of the five causes of CPI bias and a new set of estimates for them. They claim to be more accurate than previous estimates thanks to procedural improvements at the BLS (Bureau of Labor Statistics), new research, and alternative judgment on existing information. This essay will briefly explain the report's argument, giving more detail on the two main sources of bias; higher level substitution and new or improved goods entering a market. It will then explain the practical implications of this for government and the economy given the wide range of uses of the CPI, highlighting the need for policymakers to pay more attention to this issue. In their report, Lebow and Rudd conclude that the US CPI overestimates increases the cost of living by 0.87%, within a range of (0.3%-1.4%). This figure can be broken down into five categories of bias: new outlets, weighting, lower level substitution, higher level substitution and quality change or new items. The first three have a significantly smaller effect on the bias, accounting for only 0.2% of the 0.87% estimate. The new outlet bias, estimated at 0.05%, occurs because the CPI uses only quality changes to explain price differences between old and new outlets. This does not take into account changes in purchasing habits which often result in price changes. Weighting bias, which had taken place in the middle of the article......to make informed decisions. They should also encourage research in these areas and improvement of the way the CPI is collected, for example using the PCE instead of the CEX, and calculated, for example using a geometric mean aggregation formula at place of Laspeyres. : where are we? provided a detailed analysis of the causes of the bias in the US Consumer Price Index and the uncertainty behind previous estimates and their own estimates. Of these, higher level substitution and quality or quantity changes have proven to be both the most important and the most controversial. With the bias figure calculated at a plausible 0.87%, the practical implications of their argument for the wider economy have become clear and policymakers need to track the bias estimates and inform their choices by recognizing them...