This chapter examines changes in transportation rates for contract coal shipped by rail from U.S. producers to investor-owned public electric utilities in the United States between 1988 and 1997. The statistics herein update those presented in EIA's earlier Interim Report⁽¹⁾ by (1) incorporating new data for the years 1994 through 1997, (2) supplementing the basic source data with information and data from other sources, and (3) researching and adding missing data elements in the pre-1994 database to enhance its usefulness. The focus of this chapter--the rail transport of coal--is the primary concern specified under Section 1340 of the Energy Policy Act of 1992.

Railroads constitute the mainstay of U.S. domestic coal distribution, delivering 61.8 percent of total coal distribution in 1997. Eighty-eight percent, or 875.7 million short tons (mst) of total domestic coal distributed, went to electricity generators at utilities (Table 1). This chapter focuses on those public electric utilities that are "investor-owned" because of the availability of representative data for those utilities on coal quality, tonnages, origins and destinations, and shipping rates collected in the Federal Energy Regulatory Commission (FERC) biennial interrogatory known as Form FERC-580.⁽²⁾ (See Appendix A for specifics on Form FERC-580 and EIA's Coal Transportation Rate Database (CTRDB).)

Investor-owned utilities account for almost 80 percent of the coal-fired generation by public electric utilities. By the term "investor-owned utilities," EIA means to distinguish that class from public utilities that are Federal, State, or municipal entities--one of the major criteria used to specify utilities that are not required to submit fuels information or Form 580. Still, not all investor-owned public utilities are subject to Form 580 disclosure. The Form must be submitted only by "jurisdictional" utilities, that is, facilities subject to FERC jurisdiction on the basis of their sale or transmission of electricity across State lines. Further, only data related to coal purchased and delivered under supply contracts of more than 1 year's duration need be reported. Coal contracts of 12 months or less are considered spot market purchases, not subject to Form 580 reporting requirements. For that reason, Form 580 data on coal receipts are identified as "contract coal" data in this report.

Transportation analysts have shown that contract coal prices and rates are a valid indicator of changes in market conditions because contracts since the late 1980's include formulas to account for changes in economic conditions and supply and demand variables.^{(3), (4), (5)} Nonetheless, the absence of spot market data, combined with a growing number of utilities not required to file fuel-related data on Form 580, resulted in full coverage of coal transportation data for only 35 percent of total domestic coal distributed to electric utilities as of 1997 (Appendix A).

In order to raise the level of data coverage, EIA supplemented the Form 580 database. Supplementary data and information for the CTRDB came primarily from the Surface Transportation Board (STB) "Annual Waybill Sample" (coverage is limited to rail shipments) and from the FERC Form 423, "Monthly Report of Cost and Quality of Fuels for Electric Plants." Secondary information was derived from published industry reports and newsletters. The Waybill data and the FERC-423 data together may yield information on coal quality, delivered cost, tonnages, contract coal versus spot, origin and destination, waybill shipping rates, shipping distances, carrier, and coal car ownership. Neither source includes f.o.b. minemouth coal prices or contract details. The Waybill data do not specify the customer. In some cases waybills include coal going to other nearby customers, so the data must be evaluated and edited carefully.

The waybill data apply only to commodities shipped by rail. Also, because it is a sample, waybill data were not available to characterize some "origin-destination pairs." In addition, some itineraries must travel via multiple railroads' trackage systems, so that locomotives from one railroad take over a train of loaded cars from locomotives of another railroad, making it infeasible to trace completely some recurring coal shipments. In all cases in this report, FERC-580 and STB Waybill Sample information designated as confidential is either presented in aggregated form to protect the confidentiality of individual respondents or is withheld.

Overall transportation trends for U.S. coal are presented in the next section, followed by examination of trends in

coal supplied to defined demand regions, coal supplied to electric utilities affected by Phase I of the Clean Air Act Amendments of 1990 (CAAA90), and coal originated in defined supply regions.

Overall Trends in U.S. Rail Coal Transportation,
Sulfur Levels, and Rates

Three major trends define the changes in contract coal transportation by rail during the1988-1997 study period: total tonnage shipped, low-sulfur coal distribution, and high-sulfur coal distribution.⁽⁶⁾ The quantity of contract coal shipped by rail to electric utilities rose from 269.6 to 366.2 million short tons (mst). That is an increase of 36 percent, or a 3.1 percent annual average over the 10-year period (Table 6). As noted in Chapter 2, that rise correlates with increased capacity utilization at the Nation's coal-fired power plants during the period.

Coal Sulfur Levels

As contract coal shipments by rail increased, the portion represented by low-sulfur coal grew most rapidly, from 48 percent in 1988 to 65 percent in 1997. During that time, the share for high-sulfur coal shrank from 18 percent to 8 percent of all contract coal shipped by rail (Table 6). The increases in market share for low-sulfur coal did not begin with the CAAA90, but the rate of increase did double in the 1988-1997 period. Prior to any effects of that legislation, the 48 percent share of rail distribution claimed by low-sulfur coal in 1988 had risen from a 27 percent share in 1979, based principally on requirements of earlier clean air legislation (CTRDB 2000).⁽⁷⁾

Figure 5. Percentage Distribution of Contract Coal Shipped by Rail, by Sulfur Category, 1988-1997

Although not as pronounced as the trend for high-sulfur coal shipments, the amounts of relatively sulfurous "medium-sulfur B" coal shipped decreased also (Figure 5). Distribution remained level for "medium-sulfur A" coals.⁽⁸⁾ These were the highest-sulfur coals that could be burned after January 1, 1995, without treatment or penalties, at power plants affected by Phase I of CAAA90. Figure 5 clearly illustrates the divergence between the distribution levels of low-sulfur coal and those of the other coal categories.

Coal Transportation Distances

Table 7. Average Distance of Contract Coal Rail Shipments by Rail, by Sulfur Category, 1988-1997 (Miles)
Year	All Coal	Low Sulfur	Medium Sulfur A	Medium Sulfur B	High Sulfur
1988	640.2	993.5	439.7	224.8	133.7
1989	653.4	1,004.7	444.2	203.4	121.1
1990	606.7	963.2	438.1	220.0	148.4
1991	623.1	982.1	422.5	208.8	154.4
1992	638.8	994.6	403.8	187.3	151.3
1993	715.5	1,012.0	438.2	191.2	138.7
1994	687.8	980.6	414.1	174.9	172.8
1995	725.9	977.3	422.7	124.6	195.8
1996	743.1	986.2	414.0	233.5	194.3
1997	793.5	1,037.7	419.0	251.0	180.2
Notes: Low Sulfur = less than or equal to 0.6 pounds of sulfur per million Btu; Medium Sulfur A = 0.61 to 1.25 pounds per million Btu; Medium Sulfur B = 1.26 to 1.67 pounds per million Btu; High Sulfur = greater than 1.67 pounds per million Btu. Medium Sulfur A coal meets SO2 emission limits for power plants affected by Phase I of the Clean Air Act Amendments of 1990 (CAAA90). Low-Sulfur coal meets the emission requirements those power plants must attain in Phase II of CAAA90, after January 1, 2000. Statistics based on the Coal Transportation Rate Database (CTRDB) frequently differ from statistics released earlier because between 1995 and 2000 the CTRDB was enhanced with new and supplementary data, including data for years prior to 1995.    Source: Energy Information Administration, Coal Transportation Rate Database.

Figure 6. Average Distance of Contract Coal Shipments by Rail, 1988-1997

Despite the inroads made by Western low-sulfur coals into the Midwest, the Southwest, and some Southeastern States during the 1980's and early 1990's, the actual distances low-sulfur coal is transported have increased very little, if at all. As a result of the greater proportion of total coal receipts that originate in distant low-sulfur supply areas (Figure 5), however, the average distance for U.S. coal distribution overall did increase (Figure 6).

When graphed for individual coal types, distribution distances remain relatively flat from 1988 through 1997. Only high-sulfur coal shows a general upward trend, however slight, which reversed after 1995 (Figure 6). This reversal results from a reduction after Phase I, among power plants located in or near high-sulfur coalfields, in coal purchases from nearby, often in-State, high-sulfur mines. For example, in the generally high-sulfur coal States of Ohio, Illinois, and Indiana, 52.2 mst of in-State contract coal was shipped to power plants in 1994, the final year preceding Phase I. That figure declined to 46.4 mst in 1995 and to 37.5 mst in 1996 before recovering at 42.0 mst in 1997. At the same time, turning to sources for lower-sulfur coal inevitably meant increased shipping distances (CTRDB 2000).

Coal Transportation Rates

Contract coal transportation rates for rail deliveries vary among different pairs of origins and destinations and with factors such as distance, coal tonnage, and length of contract. In this section, averaged data and general trends are described. Variations among U.S. coal demand and supply regions are discussed in the next section, Regional Trends in U.S. Rail Coal Transportation, Sulfur Levels, and Rates.⁽⁹⁾

Dollars per Ton

Figure 7. Average Rate per Ton for Contract Coal Shipments by Rail, by Sulfur Category, 1988-1997

The average inflation-adjusted rate per ton to ship contract coal by rail fell steadily during the study period--a decline of 25.8 percent from 1988 through 1997 (Table 8). The rates for coal in all sulfur categories trended downward, despite a significant reversal in the rates for medium-sulfur B coal in 1996 and 1997 (Figure 7). Clearly, the majority of the contract coal shipped by rail during this period traveled via lower real-dollar rates than in earlier years, and there is no evidence of widespread inflation of shipping rates by the major coal-hauling railroads following enactment of the CAAA90. In fact, the greatest decline in coal rail rates per ton--a 36.0 percent decline in constant dollar terms--was for low-sulfur coal, the very category over which concern may have been greatest.

The rates for high-sulfur coal under contract declined only slightly during the CAAA90 study period. On the other hand, their rail tonnages fell by 57.5 percent from 1988 to 1997, but did not exhibit a decline in 1994, just before the Phase I requirements went into effect. No downturn occurred in 1994 because some power plant operators had committed to the use of high-sulfur coal prior to the beginning of Phase I. These included operators at high-polluting Phase I-affected plants⁽¹⁰⁾ and at plants already in compliance under earlier, tighter emission standards. Whether compliance with the CAAA90 would be through construction of flue gas scrubbers or through buying or trading of emission allowances, those decisions had been implemented gradually, starting prior to 1995. Power plants not affected by Phase I had until January 2000 to plan and initiate any further sulfur dioxide mitigation measures.

Mills per Ton-Mile

The transportation rate per ton-mile is the rate per ton of coal per mile shipped. To obtain significant whole numbers, rail rates per ton-mile are scaled in mills (tenths of a cent) per mile.

Figure 8. Average Rate per Ton-Mile for Contract Coal Shipments by Rail, by Sulfur Category, 1988-1997

Like the average rate per ton, the average rate per ton-mile to ship contract coal by rail declined steadily during the study period. The real-dollar rates for coal in all sulfur categories trended downward (Table 9). The ordering of the rates for coal by sulfur categories shown in Figure 8 is essentially the reverse of those in Figure 7. For example, low-sulfur coal had the highest shipping rate per ton but its rate per ton-mile was the lowest of all. This reversal reflects the fact that low-sulfur coals were located far from most major consumers. Low average rates per ton-mile are found where shipping distances are greater because the fixed costs and loading and unloading costs of carriers are spread over more miles in the net rate calculation. The average rates per ton-mile for high-sulfur coal, on the other hand, were relatively high during the period, while its rates per ton were the lowest on average. These relationships reflect a coal which, while losing market share (Table 6), is concurrently losing customers, especially among traditional customers in the areas where it is mined.

The rail rates per ton-mile were erratic for medium-sulfur B coal--even more than the rates in dollars per ton, and especially from 1993 through 1996 (Figure 8). Rapid changes took place in the rate per ton-mile for medium-sulfur B coal as many customers changed suppliers during the CAAA90 study period. In some cases, as rates per ton were falling, rates per ton-mile rose, as in 1995 and 1996 and less dramatically from 1990 through 1992. In a stable supplier-consumer environment, rising rates may signify higher rail tariffs due to lack of competition. However, the steep rise in the average rate per ton-mile in 1995, which took place when utilities were changing coal suppliers, occurred because average shipping distances had declined at that time. Rail contracts for medium-sulfur B, which had supplied 26.5 mst of coal in 1994, accounted for only 17.1 mst in 1995, while the average shipping distance shrank from 174.9 to 124.6 miles (Tables 6 and 7). Further, contract tonnage for this coal rose from 17.1 mst in 1995 to 18.2 mst in 1996, then declined to 13.7 mst in 1997, indicating that the increase in average distance shipped was coupled with a modest increase in new contracts in 1996, followed by more loss in market share in 1997 (Table 7).

Transportation Cost as a Percentage of Delivered Price

Between 1988 and 1997 a consistent 49 to 52 percent of

Figure 9. Transportation Cost as a Percentage of Delivered Price for Contract Coal Shipments by Rail, by Sulfur Category, 1988-1997

the rail-delivered price of low-sulfur contract coal was spent to transport it. By comparison, transportation costs of the other coal types trended higher, reaching 29 percent in 1997 for the delivered price of medium-sulfur A coals, 26 percent for medium-sulfur B coals, and only 22 percent for high-sulfur coals (Table 10). The stable ratios of transportation costs to delivered price for low-sulfur coal reflect a balance between declining minemouth coal prices and declining western rail transportation rates throughout most of the 1990's (Figure 9). The ratios for the medium- and high-sulfur coals rose because the average minemouth prices of these coals declined. The rail rates per ton declined also, but not as rapidly as coal prices in the unsparingly competitive coal industry.

Railroads serving the PRB also took advantage of inherent economies of scale. Rail rates from the PRB could be held down, on a cost per ton-mile basis, because the flat terrain and space for loading facilities allow efficiencies throughout the haul. The unit trains from the PRB are some of the longest and comprise some of the highest-capacity bulk railcars in the United States, and they can be efficiently loaded and unloaded at uncrowded, modern facilities.⁽¹¹⁾ It was western coal producers and railroads, each competing aggressively to win new markets, who forced coal prices and rail rates downward throughout the country by offering ever lower delivered prices for reliable supplies of low-sulfur coal. In Appalachia, where mining conditions are more challenging, coal producers could not possibly match minemouth prices at PRB and many Rockies mines. Many smaller, less efficient mines closed and the industry offered lower prices by consolidating around fewer, larger, more productive mines with modernized technologies. Eastern railroads lowered their rates also as, even with lowered minemouth prices, the delivered costs were higher than for western coals. It was either lower rail rates or the eastern railroads would have been party to closings of the larger mines and loss of some of their major clients and revenue sources. Table 10 illustrates that both components of competitive coal pricing declined in the three low-sulfur regions--average minemouth price and average transportation rate, with consequent declines in the average delivered price of coal. Similar reductions in cost components and delivered prices followed suit for coal with higher sulfur levels, again, in order to compete and to retain at least a smaller share of coal sales.

Transportation Rates per Million Btu

Coal transportation costs on the basis of the heat content and the sulfur content of the fuel delivered are indicative for many, but not all, electric power producers of the net value of the coal for their purposes. From the customer's perspective, the two most important attributes of any steam coal are: the delivered price of the coal and its value to the customer for use as a fuel. This report is not about delivered prices of coal, even though those data were useful to calculate apparent net transportation rates if rates were otherwise not reported.

The value of a coal to electric power producers currently and in recent years depends primarily on the two coal characteristics that govern its performance and its sulfur dioxide emissions--heat content and sulfur content. Those two coal characteristics are basic. Along with minemouth price, rate per ton, and rate per ton-mile, they affect the bottom-line costs the utility incurs in generating kilowatts. The decisions on heat and sulfur content and other coal specifications have to be made early on, however, so that combustion and emissions technologies can be installed and tested. For that reason, utility fuels buyers negotiate at both the mine level and the transportation level to secure the best buy available for their fuel specifications, including alternate suppliers, alternate fuels in some cases, and alternate modes of delivery.

In most cases, low-sulfur coals offer a better value to power producers. That is, compared with purchasing allowances or investing in flue gas scrubber, the lowest cost option for the greatest number of utilities was found to be switching from high-sulfur to low-sulfur coal.⁽¹²⁾ In some cases, however, a power producer's strategy may include medium- or high-sulfur coal: for example, if scrubbers are already capitalized and being used; if emissions are being offset at other, newer plants; or if because of a plant's age, it is cheaper to purchase the needed emission allowances. In those circumstances, coal purchasers may reckon the value of coals for their operation based more on Btu content, ash content and implied ash disposal options, and factors that affect boiler performance or slagging such as coal volatility, ash fusion temperature, or sodium content.

Figure 10. Average Rate per Million Btu for Contract Coal Shipments by Rail, by Sulfur Category, 1988-1997

Changes in the transportation rates per million Btu and by sulfur content of contract coal delivered to electric utilities are the cost variables in this report that best describe the factors critical to the majority of electricity-generating customers (Table 12). Low-sulfur coal consistently had the highest average transportation rates per million Btu during the study period. As noted earlier, the low-sulfur coals being shipped during the 1980's and 1990's were overwhelmingly low-Btu subbituminous coals from the Powder River Basin. Their low Btu levels, coupled with greater shipping distances than eastern coals, kept transportation rates high on a cents-per-million-Btu basis (Figure 10).

1. Energy Information Administration, Energy Policy Act Transportation Rate Study: Interim Report on Coal Transportation, DOE/EIA-0597, (Washington, DC, October 1995), 136 pp.

2. Energy Information Administration, Energy Policy Act Transportation Rate Study: Availability of Data and Studies, DOE/EIA-0571, (Washington, DC, October 1993), pp. 3-12 and Appendix A.

3. S.M. Dennis, "Using Spatial Equilibrium Models to Analyze Transportation Rates: An Application to Steam Coal in the United States," Transportation Research Forum, Vol. 35 (E) (1997), p. 147.

4. P.L. Joskow, "The Performance of Long-Term Contracts: Further Evidence from Coal Markets," Rand Journal of Economics, Vol. 21(2) (1990), pp. 251-274.

5. J.M. MacDonald, "Transactions Costs and the Governance of Coal Supply and Transportation Agreements," Transportation Research Forum, Vol. 34 (1) (1994), pp. 63-74.

6. High-sulfur coal contains more than 1.67 pounds of sulfur per million Btu of heat input. Low-sulfur coal is defined as containing 0.6 or less pounds of sulfur per million Btu, which meets the Phase II emission limit of 1.2 pounds of sulfur dioxide per million Btu. This category was identified as "compliance coal" in the Interim Report. The term "compliance coal" is widely used because 0.6 pounds of sulfur per million Btu is the upper limit sulfur content that complied with emission limits defined for New Source power plants under the Clean Air Act of 1971. Since publishing the Interim Report, EIA unified its coal classifications, such that the criteria for low-sulfur and compliance coal coincide.

7. "CTRDB 2000" is an acronym/abbreviation used to indicate that the statistics cited were drawn from the primary source data for this report, the Coal Transportation Rate Database, update version of August 10, 2000. The full citation is: Coal Transportation Rate Database, August 10, 2000 (Electronic database, 2000). Energy Information Administration (EIA), Washington, DC. (Distributor: EIA, http://www.eia.doe.gov/cneaf/coal/page/database.html).

8. Medium-sulfur A coal was termed "low-sulfur coal" and medium-sulfur B coal was simply "medium-sulfur coal" in the Interim Report, prior to EIA's unified classification. Medium-sulfur coal statistics were split into two categories to distinguish medium-sulfur A coal which could be burned without further adjustments, after January 1, 1995, from coal that cannot (i.e., medium-sulfur B and, of course, high-sulfur coal).

9. Because the rate data in this report represent regional data aggregations, they do not address alleged inequities in rates to and from isolated locations, or for "captive" shippers (with only one practical coal transportation option), or for small shippers who may not have access to technologically efficient loading equipment or may not qualify for high volume discounts.

10. The CAAA90 listed by name 263 boilers at 261 previously exempted generators that would be required to meet Phase I emission requirements. These were referred to in subsequent Environmental Protection Agency regulations as "Table 1" units, along with 174 additional generating units the utilities brought into Phase I as substitution and compensating units.

11. STB Waybill data indicates averages ranging from 106 to 117 cars in unit trains originating in the Powder River Basin. Union Pacific Railroad reports PRB trains in 1999 routinely hauling 110 to 115 cars, or 135 cars with distributed power (one locomotive positioned within the train of cars). The average carload has increased over recent years as more large-capacity aluminum gondolas are used. The average PRB carload was 112.5 tons in 1997 and 113.5 tons in 1999. (Duane Anderson, Union Pacific Railroad Company, Accounting Group, via letter and personal communication, October 7, 1998 through August 22, 2000.)

12. Energy Information Administration, The Effects of Title IV of the Clean Air Act Amendments of 1990 on Electric Utilities (DOE/EIA-0582 (97)) (Washington, DC, March 1997), pp. 12-13.