The CE and the PCE: a comparison: an analysis of a decline in the ratios of aggregate spending for various categories of expenditures from the BLS Consumer Expenditure Survey and the BEA's Personal Consumption Expenditures over an 11-year period employs a new methodology that takes into account the degree of comparability of those categories

Monthly Labor Review, Sept, 2006 by Thesia I. Garner, George Janini, William Passero, Laura Paszkiewicz, Mark Vendemia

* addressing inquiries about differences in estimates between the CE and other sources,

* assessing the efficacy of the historical CE collection methodology, and

* suggesting possible revisions to improve the quality of CE data.

A summary of points made earlier concerning the methodology and concepts involved in obtaining the CE and PCE estimates is useful to review before examining possible reasons for differences in the estimates. The CE and the PCE each provide a measure of consumer expenditures, but these measures are derived from different types of data. The PCE is defined in terms of sales or the output of production, while the CE is based on purchases. Another important distinction between the two measures is that the PCE includes the expenditures of nonprofit institutions in defining their output. In theory, if(1) all sales and purchases are recorded accurately, (2) expenditures of nonprofit organizations are excluded from the PCE, and (3) the respective populations are adjusted to be the same, the CE and PCE estimates should be similar, if not the same, for the majority of items in the survey. In practice, however, these estimates are disparate.

Three major reasons for differences between CE and PCE estimates are the methodology of the two surveys, their scope (in terms of both whose expenditures are being measured and how expenditures are defined), and the definitions they employ. Aside from including the expenditures of nonprofit institutions, the PCE covers military personnel and others whose expenditures are ignored by the CE. In addition, certain expenditure categories were out of the scope of the PCE in previous comparisons because the BEA used the CE survey as the primary source for the PCE estimates. For example, the BEA used or still uses CE data, directly or through extrapolation, on motor vehicle leasing (cars and trucks), motor vehicle rental, taxis, nursery schools, and childcare. (43) The BEA also used CE estimates for medical and hospitalization insurance premiums in the PCE. Beginning with the 2000 annual revision of the PCE, however, the BEA adopted the Medical Expenditure Panel Survey (MEPS) as the primary data source for the medical care and hospitalization insurance component of the PCE. (44)

Methodological reasons for differences. The methodologies designed to produce CE and PCE estimates are dissimilar and account for some of the difference between the estimates. The BEA starts with a basic initial dollar value for each item. This dollar value consists of the value of manufacturers' shipments of goods or the value of receipts received by service providers. The data are obtained from various economic censuses and surveys. Data from these sources can suffer from reporting errors and, in the case of surveys, sampling errors. Using its expert judgment, the BEA Staff makes adjustments for what it considers to be misreporting errors.

Wholesale and retail trade margins can account for a large proportion of the final purchasers' value of an item assigned to the PCE. The algorithm by which these margins are calculated can be summarized simply as total receipts from sales by wholesalers and retailers, less total costs of acquisition, adjusted by changes in the value of unsold inventories held. Because data limitations do not permit the production of trade margins at the item level, the BEA carries out an iterative series of adjustments and reallocations to obtain a reasonable estimate for wholesale and retail trade margins across items.