Ralph Bradley

Feasible methods to estimate disease based price indexes

There is a consensus that statistical agencies should report medical data by disease rather than by service. This study computes price indexes that are necessary to deflate nominal disease expenditures and to decompose their growth into price, treated prevalence and output per patient growth. Unlike previous studies, it uses methods that can be implemented by the Bureau of Labor Statistics (BLS). For the calendar years 2005-2010, I find that these feasible disease based indexes are approximately 1% lower on an annual basis than indexes computed by current methods at BLS. This gives evidence that traditional medical price indexes have not accounted for the more efficient use of medical inputs in treating most diseases.

Analytical Bias Reduction for Small Samples in the U.S. Consumer Price Index

The U.S. Bureau of Labor Statistics (BLS) faces sample size constraints when computing its Consumer Price Index (CPI–U). The samples are not sufficiently large for the index to equal a true “fixed basket” price index. This study adjusts for this small-sample bias by estimating the second order of a stochastic expansion of the index. Unlike increasing sample size, this adjustment is inexpensive because it uses the same data used to compute the CPI–U. From the beginning of 1999 to the end of 2003, we estimate that 63% of the difference between the BLS superlative index (C–CPI–U) and the CPI–U is the result of finite-sample bias, and the other 37% is commodity substitution bias.