Notes

1 Some of the terms in the previous coal pricing equations (underground and surface) were specified with relatively large exponents. These functional forms indicated somewhat spurious variations in coal prices for some regions and mine types for relatively moderate changes in capacity utilization and factor input costs.

2 Originally, only data on average daily capacity were reported for 1979 through 1986 in EIA’s Coal Production reports. More recently, annual capacity utilization data for 1984, 1985, and 1986 have been published in issues of EIA’s Coal Industry Annual.

3 Data for coal mines in the Northwest supply region (Alaska and Washington) were not included in the regression model. The average mine price of coal for those States is withheld from EIA publications to avoid disclosure of individual company data.

4 A statistical methodology for estimating missing data (based on the EM algorithm) has been used successfully by EIA for estimating suppressed cells in data published from the EIA-846, “Manufacturing Energy Consumption Survey.” In the future, the same procedure may be used to estimate coal prices for cells suppressed in tables in EIA’s Coal Production and Coal Industry Annual reports, so that estimates of confidential coal data can be incorporated into the database for the coal pricing regression model. For a description of the EM algorithm, see R.J.A. Little and D.B. Rubin, Statistical Analysis With Missing Data (New York, NY: John Wiley and Sons, 1987), Section 7.2.

5 Throughout this chapter, tons refers to short tons (2,000 pounds).

6 The free on board mine price is the price paid for coal at the mining operation site. It excludes freight or shipping and insurance costs.

7 U.S. Census Bureau, 1992 Census of Mineral Industries, web site www.census.gov (accessed April 9, 1998).

8 The sample size for the PPI for mining machinery and equipment (Series ID: PCU3532) for the February 1998 reporting cycle was 63 establishments. Personal communication (e-mail) from Chris Anfang, U.S. Department of Labor, Bureau of Labor Statistics, Washington, DC (April 2, 1998).

9 D.W. Carlton and J.M. Perloff, Modern Industrial Organization (London, UK: Scott, Foresman and Company, 1990), Appendix 3A.

10 Dr. Kevin Forbes, Science Applications International Corporation, formulated and estimated the Two-Stage Least Squares model of the U.S. coal market used for AEO98. This section draws upon material provided in the report by Science Applications International Corporation, An Econometric Model of Coal Supply: Final Report, prepared for the Energy Information Administration (December 20, 1996).

11 G.S. Maddala, Introduction to Econometrics: Second Edition (New York, NY: Macmillan Publishing Company, 1992), Chapter 9.

12 The regression coefficient for the labor productivity term for Central Appalachian underground mines is equal to the overall coefficient for labor productivity plus the labor productivity coefficient for underground mines plus the AEO98 adjustment (-0.728 = -0.953 + 0.051 + 0.174).

13 The regression coefficient for the labor productivity term for Powder River Basin surface mines is equal to the overall coefficient for labor productivity plus the regional productivity coefficient for the Powder River Basin plus the AEO98 adjustment (-0.996 = -0.953 - 0.217 + 0.174).

14 U.S. Census Bureau, 1992 Census of Mineral Industries, web site www.census.gov (accessed April 9, 1998).

Back To
Coal Pricing Methodology for the Annual Energy Outlook 1998