Coming a little more than a week after the Energy Information Administration (EIA) said it is overhauling its 914 report methodology, which will likely lead to a downward revision of U.S. production figures (see NGI, April 5), one energy consultant is warning that the government agency's estimates of natural gas supply are likely off by more than 1 Tcf over the last two years. This under-accounting could leave the United States in a precarious situation once the economy recovers.

According to the Jergens Center for Energy Forecasting (JCEF), when the EIA publishes its natural gas supply and demand data it uses a factor called "balancing item" to account for any difference between its supply total minus its demand total after also adjusting for storage activity. According to public data on www.eia.doe.gov, JCEF said that in 2008 and 2009 this balancing item grew to an "unprecedented" cumulative total of 1 Tcf before recent revisions by the EIA reduced the 2008 supply total by 277 Bcf.

JCEF notes that the "euphoria over shale gas field production" has prompted several analysts and media outlets to claim that the United States is in the midst of a new natural gas supply bubble; a belief at least partially fueled by EIA reports reflecting U.S. supply growth even when drilling slows and well decline rates are rising. JCEF founder Mark Jergens thinks this misconception could put the country into an upside down supply-demand equation down the road.

"The EIA has systematically overestimated the amount of U.S. production and that in actuality the U.S. is 'treading water' since the 2008 production peak and is in danger of falling far short of its supply requirements if the slowly progressing economic recovery translates to higher demand levels," Jergens said recently in a presentation at the Ohio Energy Management Conference.

Jergens noted that state production data, which is available "more quickly" than EIA data, tends to offer much greater detail. "A little common sense can also be applied in this case; the EIA data reflecting 1 Tcf of errors is some combination of an overestimate of supply, an underestimate of demand, or an overstatement of storage," Jergens said. "Furthermore, because the method employed by the EIA to calculate supply is subject to the most extrapolation and because the rig count drop in late 2008 and 2009 was on the magnitude of 55%, it makes more sense that a majority of the balancing error is in the supply totals."

EIA's production estimates were called into question last week after the government agency announced March 29th that it is updating the methodology of its EIA-914 report, which identifies production from major producing states as well as offshore. According to the EIA, there will be downward revisions in January's data when the February report is released on April 30.

The EIA is a statistical unit of the Department of Energy, and it has uncovered a fundamental problem in the way it collects the data from producers across the country. According to Gary Long, acting director of the 914 report, the EIA plans to change its methodology this month, resulting in downward revisions in some areas.

"From some of the preliminary work we have done on the January [production] number, it looks as if we'll likely see a slight revision downward in the Lower 48 total," Long told NGI. "I think we'll see a pretty fair drop in Louisiana. It might be 300-400 MMcf/d...something in that ballpark."

According to the EIA's 914 data released last week for the month of January, Louisiana showed the greatest month-to-month growth in output -- 3.3% to 5.29 Bcf/d.

Long said the EIA started looking at the old methodology about a year ago. "What triggered the review was the data at the end of 2008 and the beginning of 2009 in Texas. There seemed to be a disconnect between December [2008] and January [2009]," he said. "As we looked at it some more and started to get closer to final data from the state, it appeared that we might be too high at the end of 2008 in Texas. At that time we decided we'd review all of our processes, which hadn't been reviewed since we started the 914 in 2005. We brought in some outside folks who could be objective. One thing led to another and here we are.

"Basically, it's a timing issue on two fronts -- frequency and currency. We're updating monthly instead of annually and we're using data that is more current than we have in the past. The changes are both on the sampling side and the estimation side."

Speaking at an energy conference last week jointly sponsored by EIA and Johns Hopkins University School of Advanced International Studies in Washington, DC, Energy Secretary Stephen Chu downplayed the reports that the EIA has been overstating its 914 production data (see related story). "I think there's some questions about the data," he said. "[But] I don't think there's an overstatement of the data. This is something where EIA is continually trying to upgrade its method of data collection."

The revision news was met with mixed reactions from the trading and analyst community. While some market watchers were blaming it for the Thursday April 1 to Monday April 5 rally in natural gas futures values, some analysts were going on the record that the EIA revised production data would not alter the market's fundamentals.

Citi Futures Perspective analyst Tim Evans said news of an EIA methodology shift and a downward revision in January's production data may have contributed to the swing in market sentiment to the upside, but noted that the report is not normally "a focus of attention" for futures traders since it comes out with significant lag time.

"The March 29 report that announced the highlighted change in methodology, for example, only provided detail on U.S. wet gas production for January, a long look back for a market that sometimes devotes only a few minutes of reaction time to the storage data from the prior week," Evans said. "Equity analysts more focused on how much gas is being produced by players in specific shale plays may have greater interest, but for futures analysts we think the data is useful only in providing a post mortem, not a more forward looking analysis."

Responding to another analyst's belief that the news does not alter the fundamentals, Hencorp Futures broker Tom Saal took the sarcastic approach. "Let's see...an undetermined level overestimation followed by an undetermined level of underestimation of the entire U.S. domestic 'supply' of natural gas...one half of the supply/demand balance used by fundamental analysts. [That] does not change the pricing view of fundamental analysts? In my opinion, they must be assuming either a) an insignificant change in supply figures from EIA; or b) supply is not important?"

Stepping beyond production and looking at the supply-demand equation, Jergens noted that errors on demand data do occur, with most resulting from the fact that many utilities submit meter read cycle consumption in place of true calendar consumption. He claims this mismatch in reporting dates can cause inaccurate consumption totals from one month to the next but should inherently catch up and correct itself over time. "Gas storage data should be easier to collect since the number of operators is much smaller than the number of consumers or producers but JCEF is concerned that in the past decade the ratio of base gas to total capacity reported by the EIA has actually declined 2% while the amount of working gas has increased 18%."

Jergens said more accountability is needed by all those involved in the collection, dissemination, and distribution of the data. "The logical first step would be to require the EIA to balance the supply-demand and storage equation within a known small tolerance (say 50 Bcf per month) and diligently work to seek the best estimates possible," he said. "The EIA should not be allowed to project a balancing error in forward forecasts such as the 300-360 Bcf per year error they have built into their 2010 and 2011 forecasts. This is tantamount to planning to fail."

©Copyright 2010 Intelligence Press Inc. All rights reserved. The preceding news report may not be republished or redistributed, in whole or in part, in any form, without prior written consent of Intelligence Press, Inc.