2004

Reality isn't always what it's reported to be

By Jerry Eyster

Coal Age,  September 2004

During the first half of 2004, market indicators pointed toward an extremely tight coal market in the United States. The Energy Information Administration’s (EIA) weekly coal production estimate showed Appalachian coal output was running below 2003, despite record high spot prices. The Edison Electric Institute (EEI) reported that its Weekly Electric Output data showed total generation was up by 4.8% through the first 25 weeks of 2004 over the same period of 2003. January’s daily coal burn equaled burn rates associated with the summer peak months of July and August. If total generation was growing so strongly and January’s coal burn was so high, it seemed clear that coal use for electricity generation must be up significantly. Traders and market analysts used what data they had available to project an extremely tight coal market for 2004 and beyond. 

The environmental allowance market also responded to this tight market view. With a strong growth in coal use and a limited availability of eastern compliance coal, it appeared that SO2 emissions would increase in 2004, drawing down the allowance bank and straining the liquidity of the short-term allowance market. SO2 allowance prices rose throughout the first half of 2004, reaching an unprecedented peak of about $640 per ton of SO2 in mid-July. 

When EIA’s January generation, fuel use, and stockpile information became available on May 20th, it added fuel to the fire. While the growth rate for coal burn was only 1.4% over January 2003, the burn rate was very high at 3.01 million tons per day. This is equivalent to daily burn rates during the peak months of July and August. The low growth rate was the result of last year’s winter being colder than normal and the coal burn already being high. 

EIA did not release its February electric power data until June 25th. The coal-fired generation data looked very impressive with the coal burn for February 2004 reported as 5.5% above the burn for February 2003. 

At mid year, it appeared as though we had an exceedingly hot coal market. Coal prices were high and still rising. SO2 allowance prices were hitting record highs. Coal stock prices also were hitting new highs. 

All this consensus grew out of consistent data reported by reputable sources. But how good and how timely are the data being used? Let’s look at information that became available around mid year.

Appalachian coal production up, not down 
The Mine Safety and Health Administration’s (MSHA) coal production data for the first quarter of 2004 did not become available until June, and it showed an increase in Appalachian coal production rather than the decline that EIA estimated. MSHA reported Appalachian coal production as 98.3 million tons compared with the 95.2 million tons produced during the first quarter of 2003.  EIA had estimated production based on railcar loadings at 93.5 million tons or 5.1 million tons below actual.

Even when final MSHA data become available, EIA’s coal source and use data did not balance. Over the last five years, the imbalance has run between 3 million tons and 12 million tons (See Table 1). Small changes in preparation plant yields and incorrect reporting by producers selling their raw coal to companies with preparation plants can lead to an understatement of supply. Since coal preparation and the buying of raw coal are concentrated in Appalachia, the imbalances in the EIA data probably originate there. The tonnage imbalances appear small on a national basis, but are much more significant in proportion to the Appalachian region, where the current production shortage is most prominent.

Table 1 - Energy Information Association coal flows (1999 - 2003)
(in millions of tons, Source EIA)

	1999	2000	2001	2002	2003
National production	1,100.4	1,073.6	1,127.7	1,094.3	1,069.5
Domestic consumption	1,038.6	1,084.1	1,060.1	1,066.4	1,090.5
Net exports	49.4	46.0	28.9	22.7	18.0
Change in stocks	24.0	(48.3)	41.9	10.2	(28.1)
Unaccounted for tonnage	(11.6)	(8.2)	(3.2)	(5.0)	(10.9)

During tight markets, steam coal buyers often relax their specifications and accept higher ash coals. Similarly, shifting coal from the steam to the met coal market may create streams of low-quality coal that can be blended into steam coal shipments. It does not take much of a change in preparation plant operations to “create” additional tons of coal that can enter the supply chain without being reported. Appalachian production may be higher than even the MSHA data indicates since “production” at preparation plants may go unreported.

Impact of leap year
While the February coal burn growth sounds impressive, most of the increase is an artifact of February 29, 2004. Roughly 3.6 percentage points of the increase for the month were the result of there being an extra day in February. The average daily burn rate for February (i.e. correcting for the number of days in the month) increased by only 1.8%.  The impact of February 29 declines as the year progresses with the final lift being roughly 0.28% for the year. This means that if the daily burn rate remains constant between 2003 and 2004, total coal use would increase by just less than 3 million tons as a result of Leap Year Day.

EIA data shows little growth 
EIA released March generation data on July 15th and April data on July 30th. Coal use for electricity generation decreased by 0.9% in March and increased by 0.5% in April, according to the EIA. This means that through April the average daily coal burn (correcting for the effect of Leap Year Day) increased by 0.7%. If this rate holds for the remainder of the year, and is combined with the impact of an additional day in February, coal use for electricity generation would increase by only 1% or about 10 million tons for the year. 

The change in electric power generation by fuel source from January through April from 2003 to 2004 is shown in Table 2.  Correcting for Leap Year Day, EIA reports that total generation is up by 1.64%; gasfired generation is up by 4.29%; nuclear generation is up by 3.17%; and coal-fired generation is up by only 0.33%. Coal burn has increased more than coal-fired generation because the average heat content of the coals being used has declined as lignite and sub-bituminous coals increase their share of coal use.

Table 2 - Net generation by source of generation for January - April 2004
Source: EIA Electric Power Monthly

	Monthly generation (billion kWh)			Average daily generation (billion kWh)
Generation source	2003	2004	% change	2003	2004	% change
Coal	633.1	640.5	1.17%	5.28	5.29	0.33%
Natural gas	181.2	190.6	5.16%	1.51	1.58	4.29%
Petroleum gas	37.0	35.9	-3.12%	0.31	0.30	-3.92%
Petroleum coke	4.3	6.1	40.66%	0.04	0.05	39.50%
Nuclear	246.9	256.8	4.03%	2.06	2.12	3.17%
Conventional hydro	88.7	88.5	-0.20%	0.74	0.73	-1.02%
Other	28.8	32.1	11.29%	0.24	0.27	10.37%
Total	1,220.0	1,250.4	2.49%	10.17	10.33	1.64%

EEI revised its generation data significantly downward
EEI announced in late June that it was revising its Weekly Electric Output data series.  While EEI’s report for week 25 indicated that total electric output had increased by 4.8% since the beginning of the year, its report for week 26 indicated that output was up by only 2.9% year to date. On a regional basis, growth in Central Industrial dropped from 4.1% to 2.5%, in mid-Atlantic from 4.9% to 2.3%, and in South Central from 6.5% to 2.4%. Since high growth rates (particularly in industrial demand) are considered good for coal, these significant “declines” in growth came largely out of “expected” coal-fired generation.  This revision brought EEI’s data into line with EIA’s but took much of the fire out of coal market expectations.  The first sign of a change in expectations has been the dramatic decline in SO2 allowance prices. While the SO2 market reached its peak after EEI released its revised data, it takes time for the implications of a data revision as significant as EEI’s to work its way through markets. The SO2 allowance market reached its peak of $640 on July 12th and by early August the price was below $500.

Dire warnings more smoke than fire
Reality proved to be significantly different from what had been reported. While the coal market remains tight and coal pricing should continue to be strong for some time, the dire warnings of impending widespread coal shortages are more smoke than fire. 

Data does not create a tight market. It does confirm or contradict other signals that are being seen in the marketplace. Data may stiffen the willingness of a buyer or seller to hold to a bargaining position. If a coal buyer thinks the market is tight and getting tighter, he is more likely to accept a higher price. Since the need for better data is clear, it is not obvious why coal producers, traders, and users do not push to improve EIA’s weekly production data now that there are fewer mines in production. While Platts COALdat database reports on 1,960 coal operations in 2003, only 1,313 sources produced any coal and 97% of all production came from just 585 sources. A simple weekly e-mail or fax to EIA from fewer than 1,000 entities could provide highly accurate and timely data on weekly coal shipments. Timely data and market intelligence creates a valuable competitive advantage. Reality is reality. However, reality isn’t always what it’s reported to be.

Jerry M. Eyster is a managing consultant with PA Consulting’s Global Energy Practice in Washington, D.C. He has 30 years of experience analyzing coal markets and the impacts of environmental regulations on the coal and electric power industry. He can be reached at jerry.eyster@paconsulting.com or by phone at (202) 442-2543.

  |  Previous  |    |  Next  |