Board of Governors of the Federal Reserve System

Industrial Production and Capacity Utilization - G.17

Electric Power Explanatory Notes

NOTE: The Federal Reserve discontinued its survey of industrial electric power use with the publication on December 15, 2005, of data for October 2005. The response rate for the voluntary survey had dropped significantly since the early 1990s, with coverage in 2004 nearly 40 percent less than at its peak in 1993. With the publication of the annual revision to the G.17 on November 7, 2005, the monthly indicators for the industrial production indexes for the twenty industries that previously relied on electric power use were changed to production-worker hours for the period 1997 to the present. Although the indexes of electric power use will no longer be updated, the historical indexes will remain available from the historical data page for the G.17. A notice seeking comments on this action was originally published in the Federal Register on September 29, 2005; no comments were received during the public comment period, which ended on November 28, 2005.

Coverage. Electric power data for sales by utilities to industry users and for electric power produced for their own use by cogenerators (manufacturing and mining firms that produce electricity for their own use or to sell to a utility) are generally collected at the 4-digit NAICS and 3-digit SIC level for mining and manufacturing. Aggregates for 3-digit industries, as well as for total mining, durable, nondurable, total manufacturing and total industrial electric power use, are computed. Manufacturing consists of those industries included in the North American Industry Classification System (NAICS), definition of manufacturing plus those industries–logging and newspaper, periodical, book and directory publishing–that have traditionally been considered to be manufacturing and included in the industrial sector. An aggregate showing total industry use excluding nuclear nondefense is shown separately because the value-added proportion for the nondefense nuclear material series (part of NAICS 3251) in total IP is considerably less than its share of total electric power use. In addition, aggregates for utility sales to industrial users and industry generation are computed. While only the major aggregates are shown in the G.17 release, data for the 3- and 4-digit industries are available on the Board's web site.

Source Data. Electric power data are collected from a sample of utilities and cogenerators covering all twelve Federal Reserve Districts. The primary criterion for inclusion of a utility in the panel is whether the utility provides electric power to industrial customers. A comparison of Federal Reserve kilowatt-hour aggregates to estimates from the 1997 Census of Manufactures and recent reporting panel statistics suggests the Federal Reserve data cover about 50 percent of the overall sales to manufacturing in that year. The cogeneration panel covers about 50 percent of cogeneration used directly by manufacturers. In order to provide more complete coverage and correct for any shortcomings of the survey, the series are benchmarked at the 4-digit industry level to the latest available data from the Annual Survey of Manufactures and the Census of Manufactures.

Methodology. The data we receive from utilities and cogenerators are edited for anomalies and summed to the 4-digit NAICS industry levels and above. Where reports are late or unavailable for some reason, responses are estimated.

Seasonal Adjustment. Series are seasonal adjusted at the 4-digit NAICS level, with seasonally-adjusted aggregates typically computed as sums of seasonally adjusted components. The seasonal adjustment procedure (Census X-12 program) is used without trading-day adjustments because the reporting periods of the various utilities are not the same. A leap year adjustment is made where appropriate.


A description of the aggregation methods for industrial production and capacity utilization is included in an article in the Federal Reserve Bulletin , vol. 83 (February 1997), pp. 67-92. The Federal Reserve methodology for constructing industry-level measures of capital is detailed in Capital Stock Estimates for Manufacturing Industries: Methods and Data by Mike Mohr and Charles Gilbert (1996). Industrial Production--1986 Edition contains a more detailed description of the other methods used to compile the industrial production index, plus a history of its development, a glossary of terms, and a bibliography. The major revisions to the IP indexes and capacity utilization since 1990 have been described in the Federal Reserve Bulletin (April 1990, June 1990, June 1993, March 1994, January 1995, January 1996, February 1997 , February 1998, January 1999, March 2000, March 2001, March 2002, and April 2003).

Return to top

Last update: March 25, 2011