Monday, April 5, 2010

Natural-Gas Data Overstated

Energy Department Will Make Revisions to Its Monthly Report on Output

By CAROLYN CUI

The Energy Department is preparing to make sweeping revisions to its U.S. natural-gas production data after finding it has been overstating output, raising new questions about the government's collection of energy information.

The monthly gas-production data, known as the 914 report, is used by the industry and analysts as a guide for everything from making capital investments to predicting future natural-gas prices and stock recommendations.

But the Energy Information Administration, the statistical unit of the Energy Department, has uncovered a fundamental problem in the way it collects the data from producers across the country—it surveys only large producers and extrapolates its findings across the industry. That means it doesn't reflect swings in production from hundreds of smaller producers.

The EIA plans to change its methodology this month, resulting in "significant" downward revisions in some areas, according to Gary Long, the acting director of the 914 form, who led the review.

The Wall Street Journal last month reported that the EIA also has key deficiencies in its collection of market-moving oil-inventory data that has caused swings in its survey.

The EIA has been overtaken by advances in technology, oil-shale finds and changes in the industry and has been less able to account for smaller companies that increase or decrease production. Some commodities analysts and gas producers, such as EOG Resources Inc., have long suspected that the EIA has overstated domestic natural-gas output—a factor they argue has helped push prices to seven-year lows in 2009.

"The model we have now overestimates" production, Mr. Long said in an interview. He said the review was prompted by the EIA noticing aberrations in some states. "We saw some numbers we didn't like in Texas; we thought they were a little too high," Mr. Long said.

Mr. Long said the EIA plans to change its methodology, though he didn't give details. The changes could lead to a downward revision of the nation's gas production. While overall there mightn't be a big change, Mr. Long said, some states will see "significant" revisions in production.

The EIA data showed that gas supply rose 4% in 2009, despite a 60% decline in onshore gas rigs. The conflicting numbers have perplexed analysts.

Analysts also point to the discrepancy between supply (how much gas is produced or imported) and demand (the amount that is stored or used). Those two figures should cancel each other. While there always is a margin of error, that margin has widened sharply in recent months.

In December, the agency reported total new gas supply at 87.8 billion cubic feet a day and total demand of 80 billion, leaving 7.8 billion cubic feet unaccounted for—a margin of error of 10%.

"It's getting ridiculously large," said Ben Dell, an analyst with Sanford C. Bernstein. "When you have a 10% gap, that's somewhat making a mockery of the data."

Mr. Dell in January wrote a report raising questions about the mismatch. In that report, he focused on October numbers that showed a 12% margin of error.

"We think that most would agree that a 12% margin of error makes a data set tough to rely on, to say the least," Mr. Dell wrote in that report. Rather than gas supply being flat or slightly down as the data suggests, Mr. Dell wrote, he believes production is actually falling.

When that gradually becomes apparent, gas prices will be pushed "much higher," he says.

When told of the expected changes, Mr. Dell said: "It's good that they are actually paying some attention."

Mark Papa, chief executive of EOG Resources, a Texas-based gas producer, has long criticized the data, and sought a meeting with EIA officials because it is "a serious enough consistent data error we need to bring to their attention."

The "erroneously high" numbers have depressed prices, Mr. Papa said.

On April 30, the EIA is scheduled to release its natural-gas monthly report for February. In the report, the agency will use the new methods to estimate gas supply and revise its January numbers. The numbers for 2009 won't be updated until late fall.

In the upcoming report, the agency also will use more recent data—six to 18 months old—to estimate production by companies that aren't included in the survey. The current model uses data that are two to seven years old, the EIA says.
—Brian Baskin
contributed to this article.

No comments: