Dominion Let Off Hook in FERC Probe of Storage Error
|Thursday, February 10, 2005
Officials at FERC's Office of Market Oversight and Investigation (OMOI) said Thursday that they have found no evidence that Dominion employees had prior knowledge of the erroneous storage data that a Dominion "clerk" mistakenly submitted to the Energy Information Administration (EIA) last November.
The mistake, which was not caught in time by EIA and ended up increasing the weekly storage withdrawal reported by the agency, led to a $200 million to $1 billion increase in U.S. gas costs.
OMOI staff said that it conducted interviews with nine Dominion employees, examined written testimony from 22 others and made a careful analysis of "thousands" of Dominion emails. It also examined Dominion trading from mid November 2004 to after the Dec. 2 EIA storage revision.
"We are closing the investigation, and we have found no evidence that any Dominion employee, including traders working for Dominion Energy Clearinghouse, had any improper or advance knowledge of the mistake that had been made by Dominion Transmission in sending the wrong data to EIA," said Ted Gerarden, managing counsel in FERC's Division of Enforcement.
"We found no indication that any trading strategy was premised on knowing there had been a mistake in the Nov. 24, 2004 EIA weekly storage report," he said. Gerarden also said a review of futures and cash trading by others "did not indicate any pattern of trading activity indicative of someone knowing there had been a mistake and planning their trading strategy accordingly."
Prior knowledge of the mistake might have provided an opportunity for tremendous profits. The incorrect email attachment sent by a Dominion clerk led to an erroneous storage report on Nov. 24. EIA staffers missed the error and reported a much larger than expected gas storage withdrawal, which sent gas futures prices sharply higher in the moments following the report's release. Prices rose 60 cents/MMBtu in the first half hour of trading following the storage report and then ended the day up $1.183 from the day prior.
Because of a scheduling change to account for the Thanksgiving Day holiday on Thursday of that week (the day on which EIA normally releases its storage report), EIA moved up the storage report to 12 p.m. on Wednesday Nov. 24. The scheduling change put the report's release on the last trading day of the December futures contract.
As a result, the sharp price increase affected the final December futures price, all of the contracts that were tied to it and monthly cash price indexes. Baseload gas contracts for the entire month of December were affected, with the impact falling more heavily in the Texas-Louisiana production area and in the East, where more deals are tied directly to the futures price.
"The $200 million to $1 billion range of cost increases was not to [end-use] customers," said Stephen Harvey, OMOI's deputy director. "It was an effect to basically dollars moving throughout the system. That's not to say there weren't effects on customers. It was a movement in value across the whole system.
"A piece of that clearly is just direct movements on the Nymex. Another piece of that would be sort of other related financial instruments, swaps, over-the-counter kinds of deals. Another piece of that would be related to the important relationship between the Nymex and monthly index prices. The degree to which Nymex activity that day affected the levels of monthly index prices is a pretty big effect on many kinds of buyers, but particularly on distribution company buyers," said Harvey. "That's where a lot of the customer effect is from."
As an example of that, PECO Energy informed NGI last December that it would pass through to its customers about $8 million in costs that were directly related to the EIA/Dominion Transmission storage error. KeySpan, one of the largest gas distribution companies in the Northeast, estimated that the error would cost its customers $7 million. ConEdison, another major northeastern LDC, said its gas purchases were not significantly exposed to the storage error.
"A good portion of our gas is tied to that index," said PECO's Eric Helt, manager of gas acquisition and planning, in an interview with NGI at the time. "That will raise our gas costs somewhat, and it will have to be explained [to regulators] in our [purchased gas cost adjustment] filing in March..."
OMOI's Harvey said FERC has not taken a closer look at the total cost impact on LDCs and residential, commercial and industrial customers. The Commission apparently is ready to just put this event behind it and move on. Evidently it's not clear what more could be done.
FERC Chairman Pat Wood has said that more information needs to be collected and made available on supply and demand. Additional information could help prevent problems with the storage data from having such a large impact. FERC is considering requiring storage operators to post daily storage levels on their websites. The problem with that is that a lot of storage operators are non-jurisdictional, so the posting would provide an incomplete picture of storage levels, which could be misleading.
The EIA in early January posted a notice asking for comments on its storage revision procedure. As it turned out, EIA could have informed the market about the error on the very same day that it was released, but because of its revision policy, it had to wait until its next regularly scheduled storage report a week later on Dec. 2. After the market was told that the prior week's reported withdrawal was 32 Bcf too high, gas prices subsequently tumbled back down to where they had been prior to the error.
FERC staff filed comments on EIA's revision policy on Feb. 7. "We encouraged EIA to adopt a procedure for interim revisions -- interim being between the regularly scheduled weekly posting -- to provide more accurate information to markets when there is an episode such as this one where an error resulted in a correction by EIA," said Gerarden. However, it's not clear that an earlier EIA revision, even if released later on the same day as the error, would have prevented the huge price run-up. Once the erroneous storage data was out and the futures market closed, the damage was done.
The American Gas Association (AGA) and Natural Gas Supply Association told EIA in their recent comments that the most important thing is to make sure that everything that could be done to eliminate errors is done, so there would be no revisions (see Daily GPI, Feb. 8). Dominion, in a press release following FERC's announcement, said the incorrect file was clearly dated with the wrong date for submissions being accepted that week. The EIA had two days to catch the error before the storage report was released, but somehow still missed it.
AGA suggested that EIA document its verification process to give industry more confidence that the reports are accurate. NGSA went further, offering pre-announcement procedures for checking accuracy, including:
The agency also could have avoided such as significant impact on gas costs by simply making sure that no storage report is ever released on a day when a near-month futures contract is expiring. The agency still has not learned that lesson. It released a storage report on Jan. 28, 2005, the same day that the February 2005 gas futures contract went off the board. Luckily for gas consumers, there was no error in its report that time around.
(Copyright 2004 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.