Recent estimates of natural gas consumption by the EnergyInformation Administration seem to confirm that weather is notalways the determining factor when it comes to energy use. The gasindustry might want to thank God for that. EIA shows consumption isup in the face of much warmer temperatures and much fewer heatingdegree days this winter.

Temperatures in Washington, D.C., for example, have been 22%warmer than normal so far this year. Nationwide, temperatures havebeen 20% warmer than normal and since last November have been about10% warmer than normal.

The National Weather Service shows that since the beginning ofNovember population weighted heating degree days for the entireU.S., which are calculated by measuring differences between themean daily temperatures in 200 cities across the U.S. and 65degrees, have averaged about 10% fewer than normal and 7.6% fewerthan last winter. Since Jan. 1, there have been 20% fewer heatingdegree days than normal (1,438 HDD compared to 1,801 HDDs) and 15%fewer than last year (1,683 HDDs). Every region of the country hashad fewer HDDs than normal during the months of January andFebruary.

CNG’s Energy Index, which is designed to approximate residentialenergy demand, shows energy use so far this year down 25% fromnormal. The index is based on more than temperature and population.It also considers cloud cover and humidity, and incorporates acalculation based on energy usage of a modern energy efficienthome. Once again, every region of the country had below-normalenergy requirements in January and February. The regions with thegreatest departures from normal were the Mid-Atlantic (29.9% belownormal) and Great Lakes (29.1% below normal). The regions with theleast deviation from normal were the Southwest (13.2% below normal)and Pacific (16.3% below normal). Last Tuesday was the first daysince New Year’s Day that the index average was above normal forthe United States as a whole.

“The effects of this classic El Nino are very clear,” saidJoseph H. Petrowski, president of CNG Energy Services. “In parts ofthe country where it is typically below freezing throughout Januaryand February, daytime highs often hit 50 degrees or higher. And, inplaces that are usually warm all winter, the temperatures have beenrather cool. These two trends have resulted in greatly lowering theneed to turn up the furnace or the air conditioner.” At least,that’s what you would think.

Despite temperatures that are much warmer than normal and muchwarmer than last year, the EIA in its February Natural Gas Monthly,which was issued last week, estimated gas consumption so far thiswinter to be up 1.7%, or 1.28 Bcf/d, from the same period lastwinter (November through February).

Estimated total consumption is down 2.3% from winter 1996-96,but gas demand for the industrial sector is even above levelsduring that winter. Industrial gas demand is up 0.7% from lastwinter. Residential gas demand, which is most closely correlatedwith weather changes, has been down about 1.8%, but commercial gasconsumption is up 3% from levels last winter. EIA did not publishestimates of gas demand for electric utility generation.

The Economy as Indicator

Demand is growing. “It seems to be because the economy is good,”said Roy Kass, EIA statistician and director of natural gas surveymanagement. “It’s up in industrial. It’s up in commercial too, butwe really don’t understand that so much. I have no [idea].Commercial users really should be operating things like hotels,shopping centers, things like that. It shouldn’t take as muchenergy to warm those in warm weather as it does in cold weather,”he noted.

Kass noted EIA often makes revisions to its estimates and theconsumption numbers for December, January and February areestimates that could easily change. Kass does admit EIA’s estimatesfor gas consumption in the electric utility sector are “really bad.If we knew [why], we could fix it.”

“If you are aquainted with STEFs, its a short-term energyforecasting model that frankly boggles my mind when I see what’srelated to what. It’s not really for gas; it’s for all energycomponents. My understanding is that for the electricity component,there’s a lot of what the economists call exogenous variables thatwe don’t take account of sufficiently when we run the model becausewe don’t run the whole thing. We only run the gas module. We don’ttake account of, for example, how many nukes are on line. When youcompare the results that come out of the model with the actualsurvey a couple of months hence, the relationship is not accurateenough for us to go public with it.”

Nevertheless, Kass stands by EIA’s gas consumption numbers forthe other sectors. He said EIA surveys 400 gas companies each monththat represent an average of over 90%, but not less than 85%, ofthe gas consumption in each state in the U.S. It later conducts amore comprehensive survey, the results of which are used to furtheradjust EIA’s data.

Growing gas consumption “is the trend,” said Kass. But reportsof increasing demand at a time when weather would strongly indicatethe opposite has led some to question EIA’s estimates. “I do thinktheir numbers are out of whack,” said PaineWebber analyst RonBarone. “40% of gas consumption is weather related. I don’t see howthe other 60% could offset such a big change this winter.”

Others, however agree the impacts of a strong economy and theincreasing number of conversions to gas fired-generation from coaland nuclear power. But Petrowski said yet another factor may bepartly to blame: invisible gas inventories. National surveys areimperfect in many ways but one important short-coming is they donot adequately calculate on-site storage and linepack, he said. “Agas pipeline company has fairly large swings in linepack, theamounts of gas they have on their system.” Petrowski theorizesthere’s a lot more gas in “the system” than normal this winter,resulting in higher consumption calculations without an actualincrease in gas use.

“We like to ballyhoo about how natural gas is grabbing marketshare, and it has. But the fact of the matter is the numbers look alittle higher than I would have thought possible. My own theory isthat invisible inventories of natural gas are much higher thannormal, which would correlate with other things we know in themarket. Interest rates are very low, so your costs of carryinginventories if you’re an industrial are very low. And the absoluteflat price of gas on a historical basis is very low. Those pipelinecompanies and LDCs that I have conversations with indicateeverybody’s system is quite [full]. There’s quite a variability intolerance, in linepack. It could be 10-15% in a normal LDC.

“A lot of this is anecdotal.but I’ve noticed it’s been easygetting gas out of storage this winter but I haven’t got a lot ofwarm responses when I asked to bank some gas for a couple days.People have been reading me the riot act for that. I still thinkconsumption is up on the industrial side. But I don’t believe theactual consumption numbers are as high as they’ve been reported.”There’s a lot of gas in the system and that makes consumption lookhigher than it is.

Rocco Canonica

©Copyright 1998 Intelligence Press, Inc. All rightsreserved. The preceding news report may not be republished orredistributed in whole or in part without prior written consent ofIntelligence Press, Inc.