Auditing the Producer Price Index: Micro Evidence From Prescription Pharmaceutical Preparations
Ernst R. Berndt, Zvi Griliches, Joshua G. Rosett
NBER Working Paper No. 4009
In this paper we focus on a mystery we uncovered while undertaking a detailed audit of the US Bureau of Labor Statistics producer price index (PPl). We summarize our puzzle as follows. From January 1984 through December 1989.the BLS price Index for SIC 28341 (prescription pharmaceutical preparations) grew at an annual rate of 9.09%. For purposes of comparison, we have obtained monthly price and quantity sales data on all prescription pharmaceutical preparation products sold by four major US pharmaceutical manufacturers, accounting for about 24% of total industry domestic sales in 1989. Using Laspeyres price index construction procedures on these data that mimic BLS methods, we find that over the same time period. the four-company price index increased at only 6.68% per year. Finally. when we employ a Divisia price index procedure with smoothed weights that incorporates new goods immediately, the aggregate price index for these four firms grows at a rate of only 6.03% per year. Why is it that the official BLS price index grows approximately 50% more rapidly (9.09% vs. 6.03%)than the Divisia price index? That mystery is the focal point of our paper.
Published: Journal of Business and Economic Statistics, vol. 11 no. 3, July 1993, pp. 251-264This paper is a revision of the earlier NBER working paper w3490, On the Accuracy of Producer Price Indexes for Pharmaceutical Preparations: An Audit Based on Detailed Firm-Specific Data, Ernst R. Berndt, Zvi Griliches, Joshua G. Rosett