Price and Expenditure Measures of Petroleum Products: A Comparison: Price Changes in Petroleum Products Are Closely Correlated among the Three Price Programs of the Bureau of Labor Statistics; Changes in the BLS Petroleum-Product CPI Also Correlate with Changes in Consumer Spending on Those Products, as Measured by the Consumer Expenditure Survey

Duly, AbL., Harris, Jeffrey A., Khatchadourian, Ara M., Ulics, Rozi T., Wolter, Melissa C., Monthly Labor Review

Political events in oil-producing countries, hurricanes in the Gulf of Mexico, and increasing global demand for petroleum products have all contributed to sharp increases in recent years in prices of crude oil and of petroleum products derived from crude oil. From January 2000 to July 2006, the average price for a gallon of unleaded regular gasoline in the United States increased 130.5 percent. (1) Personal consumption expenditures on gasoline rose from $175.7 billion to $287.3 billion from 2000 to 2005. (2) Rising prices and increasing expenditures are a concern for consumers, business leaders, and Federal policymakers. As a result, reliable information on the prices and costs of crude oil and petroleum products is more vital than ever in making decisions at all levels of the economy.

The Bureau of Labor Statistics (BLS, the Bureau) conducts price and consumer expenditure surveys that measure both changes in prices of, and expenditures for, petroleum products throughout the various levels of the economy. This article introduces the programs that carry out these surveys, describes the petroleum data compiled by those programs, explains the methodology underlying the various crude-oil and gasoline surveys, and provides historical comparisons of price data across the BLS programs.

Price and expenditure programs

The Bureau conducts three price surveys and a survey of consumer expenditures: the International Price Program (IPP), which measures import and export prices; the Producer Price Index (PPI), which measures prices received by domestic producers; the Consumer Price Index (CPI), which measures consumer prices paid out of pocket; and the Consumer Expenditure Survey (CE), which measures out-of-pocket consumer expenditures. Chart 1 illustrates the high correlation among the three price indexes. (Two different PPI's are shown.) Each program has a different scope, measurement goal, and methodology for collecting and compiling data related to crude oil and petroleum products, and the differences among the programs must be understood in order to properly interpret and compare the movements among the respective indexes. A description of each program's measures of petroleum product prices and expenditures follows. The appendix presents an exhibit summarizing the program methodologies.

[ILLUSTRATION 1 OMITTED]

Measurement of traded-goods inflation: the IPP's import and export price indexes. The IPP produces indexes for import and export goods and services. These indexes measure price changes on the basis of the actual transaction prices of specifically defined items and services coming into and leaving the country. Published IPP indexes for crude oil and refined petroleum products reflect three classification systems: the Harmonized Tariff System of the United States, the Bureau of Economic Analysis End Use classification system, and the North American Industrial Classification System (NAICS.) These indexes are used. among other things, to deflate various foreign trade statistics produced by the U.S. Census Bureau and the Bureau of Economic Analysis.

Although the me calculates indexes for all petroleum products, only some of these products meet the program's dollar-value-of-trade threshold for public release. The IPP publishes indexes solely for products with an annual dollar value of trade for exports of at least $3.0 billion or a dollar value of trade for imports of at least $3.0 billion. Because U.S. crude-oil exports no longer exceed $3.0 billion per year, the IPP does not publish export indexes for crude oil. By contrast, crude-oil imports have a trade dollar value that greatly exceeds the IPP threshold. Indeed, on the basis of a dollar value of more than $131 billion in 2004, imported crude oil accounts for 9.35 percent of the overall IPP import index, with all other imported petroleum products together accounting for 3.45 percent of the weight of the import index. Among the import and export price indexes the IPP produces for other petroleum products are indexes for gasoline and distillates.

Unlike most of the product indexes in the IPP, which reflect price data collected directly from importers and exporters, the import crude-oil index is based primarily on monthly transaction data obtained from the Department of Energy's Energy Information Administration (EIA) Form-856 survey. (3) This survey collects data on prices and quantities of virtually all crude oil imported into the United States. Because of its broad scope, EIA-856 data are more comprehensive than data that can be obtained from a sample of importers. EIA-856 data also encompass transactions for the entire calendar month, so that, in contrast to price indexes for goods and services, for which the reference period is the first week of each month, the IPP crude-oil index is composed of transaction data received over the entire month.

One limitation of using the EIA-856 survey, however, is that most of the transaction data are not available from the Department of Energy in time for the first published estimate of the crude-oil import index. Instead, the complete set of EIA-856 data for a given month is available on a delayed basis. For example, most of the June 2006 prices were not available until July or August 2006. To correct for this limitation, the me employs two different estimation procedures to calculate its crude-oil import index: a procedure for calculating the preliminary and first revised index and a procedure for calculating two subsequent revisions of the index.

To calculate the preliminary and first revised index when the EIA-856 data are incomplete, the IPP uses a regression model based on the incomplete EIA-856 data to estimate the percent change in the price for crude oil. This regression model's estimates are much closer to the final estimate of the monthly percent change in the crude-oil import index than are the estimates derived from merely aggregating the incomplete EIA-856 data.

The two subsequent revisions are calculated directly from the EIA-856 data with the use of a modified Laspeyres index formula based upon fixed item and index weights calculated each year by the IPP. (4) The weights are lagged 2 years. For example, in 2006, the weights were calculated from 2004 trade values. In January 2007, the IPP will start using 2005 weights. At the item level, the IPP calculates the weight for each type of imported crude oil, or …

Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: Price and Expenditure Measures of Petroleum Products: A Comparison: Price Changes in Petroleum Products Are Closely Correlated among the Three Price Programs of the Bureau of Labor Statistics; Changes in the BLS Petroleum-Product CPI Also Correlate with Changes in Consumer Spending on Those Products, as Measured by the Consumer Expenditure Survey. Contributors: Duly, AbL. - Author, Harris, Jeffrey A. - Author, Khatchadourian, Ara M. - Author, Ulics, Rozi T. - Author, Wolter, Melissa C. - Author. Journal title: Monthly Labor Review. Volume: 129. Issue: 12 Publication date: December 2006. Page number: 56+. © 1999 U.S. Bureau of Labor Statistics. COPYRIGHT 2006 Gale Group.

This material is protected by copyright and, with the exception of fair use, may not be further copied, distributed or transmitted in any form or by any means.