printer friendly version

NGI's Statement on Natural Gas Price Surveys

The reporting of wholesale spot natural gas prices, which Intelligence Press Inc. -- more commonly known as Natural Gas Intelligence or NGI -- has been doing for nearly 20 years, has come under fire and extensive discussion in the wake of the flame-out of the merchant energy sector.

NGI has participated in some of these discussions and is aware of the suggestions which have been made to improve price reporting. In recent months, addressing the reduction in trading and price report submissions, NGI has informed market participants of changes in its data collection in NGI's Daily Gas Price Index and NGI's Bidweek Report. In this statement today we are offering our readers, those who contribute to and use our pricing information, a comprehensive explanation of our position in the controversy that has developed over the function of price surveys.

At NGI, we have always been aware of the heavy responsibility we carry as aggregators and purveyors of natural gas price information and have worked hard at delivering a price report that is as complete and accurate as possible. It is not an easy job, but we have grown up with the gas market, learning its quirks and nuances as we went along.

Going forward we are very determined to protect the trust and responsibility the price reporting function requires. While the fact that false data has been submitted to energy price publications appears to have come as a surprise to some, indexing publications have long been aware of the threat to their indexes and have developed methodologies to guard against manipulation.

In the wake of the reports of false submissions and government investigations into the nature of price discovery that publications perform, there have been suggestions for changes, for government oversight, and even for the federal government to take over price collection and reporting.

Backlash Threatens Energy Infrastructure

In this period after the collapse of the speculative stock market bubble, we realize the investigation of our function is just one part of the backlash that is affecting many companies, our government and financial institutions. We are concerned that, particularly for major companies in the energy industry, this backlash - the desire to place blame and extract retribution - may have gone too far. It is seriously damaging the operation of the fundamental energy market and the ability of the industry to reliably fuel the nation.

NGI's part in this is small, but the price discovery function is critical to the market, and we have taken the opportunity to examine our practices to look for ways to improve.

We are aware that Platts, our competitor in reporting on the natural gas industry, on Feb. 12, 2003 published its manifesto on price reporting for natural gas and electricity, which is available on its website at platts.com. We endorse much of the ideals and sentiment in Platt's proposals, but have some reservations.

Since NGI reports only natural gas market prices, while the Platts organization also reports electricity prices and the prices of crude oil and products and a number of other commodities, we have a slightly different view of what is necessary for the gas market going forward.

Essentially, Platts proposal is as follows:

• Price submissions to publications should come from a company's mid- or back office -- not from individual traders -- and should be verified as accurate and complete and signed by a senior level official such as a chief risk officer;
• Market participants should submit detailed transactional level data -- not aggregated data; and
• Market participants should provide counterparty information for each transaction.

NGI agrees that some of these are ideals which should be pursued. We always have urged our price sources to submit transactional data. Since the identified problems involving false submissions appear to have come from individual traders, the idea of a back office source and an official signature is helpful. A number of companies already are making price submissions in this format, and it is extremely helpful in compiling our price surveys.

However, NGI is very strongly of the opinion that multiple submissions and price quotes from the largest, most varied number of market participants is the primary factor in achieving accurate prices. A robust sample is the strongest defense against attempts to manipulate the market. There are other controls used in sampling the market, but this one is the most important.

NGI therefore would not agree with any attempt to install any of these standards as mandatory or to exclude submissions from survey participants which do not conform.

For instance, we understand a substantial segment of the industry is not willing to provide counterparty information. NGI appreciates receiving counterparty information from those who want to contribute it and whose contracts with others in the market do not preclude revealing that information.

But we would not recommend any practice that may serve to diminish the quantity or timeliness of price reporting.

If providing counterparty information or following any of the above suggestions is considered too onerous and discourages even one potential information contributor, that is too high a price to pay for that information, particularly in the current, very thin market.

Price Submission Volume and Variety Paramount

We note that market contracts typically contain confidentiality clauses. In order to provide counterparty information many companies would have to rewrite all their contracts, something they are not likely to do.

We can deal with potential manipulators, as we have in the past, through the use of multiple and varied sources and our close surveillance of what is going on in the market; we cannot deal with an absence of fixed price or physical basis quotes. NGI's price staff has worked hard in recent months to seek out new price sources and convince those who have dropped out of our survey to return. We are hoping we have come through the worst of the market and reporting contraction. Price quotes for the February Bidweek were much improved over the prior three or four months and we hope this is the beginning of an upward trend.

We continue to encourage market participants to do as many fixed price trades as possible and report them in NGI's price surveys. If you do not currently participate and would like to, call Mark Curran at 503-235-1174, Roger Tanner at 713-974-6345, Dexter Steis at 703-318-8848 or Phil Sano at 503-235-1175. Information about our veteran staff is available at http://intelligencepress.com/whoweare.html.

We also encourage companies not to put restrictions on their traders talking to our price reporters about market conditions and prices they have seen. It is important for us to keep in close contact with the market, and access to traders is crucial.

NGI's price survey methodology for the delivered to pipeline spot prices published by NGI's Daily Gas Price Index and NGI's Weekly Gas Price Index is available to anyone on our website at http://intelligencepress.com/methodology.html. It includes a description of our large and varied source base, our pledge of confidentiality, specifics as to type and timing of price information we use, our method of volume weighting price submissions to create an average and a description of outliers that may be excluded.

We also apply our thorough knowledge of the market to identifying and examining dubious submissions. Those devoted to strict statistical formulations have derided the use of what they call "subjective judgment" by publishing companies. We would offer that one description for this "subjective judgment is "expertise." We note that Platts threw out and did not use the 48 false trades at Sumas allegedly submitted to it by an El Paso trader, so those false trades did not affect their index.

Our price editor also has thrown out suspicious, unconfirmed and questionable data submissions in the past and will continue to do so. If these were elaborate surveys with unlimited time and total access to all information, you certainly would eliminate so-called "subjective" judgments. As it is, we at NGI will continue to put our expertise and thorough knowledge of the market to good use.

Although published prices have been beaten up in the last few months by admissions from traders that they submitted false data, we believe our methodology is able to protect the indices from any large deviations. This ability is, however, diminished as our sample size decreases. The best protection from potential index manipulators is increased participation in our surveys by the industry.

The prospect of audits being mandated on either the companies reporting transactions or the publishers, as has been suggested by some, would only serve to reduce the willingness of companies to report transactions to publications and increase the susceptibility of indexes to manipulation.

The final judges of our product are those who have continued to use it over the last 20 years. Market participants are aware of what is going on in the market and have in the past ceased using prices from publications which they believed did not consistently reflect that market. Unlike the power market, the natural gas market has developed over long years into an efficient and competitive market. Likewise the price reporting on natural gas by NGI and Platts have come to be accepted as a reasonably accurate reflection of that market.

Additional evidence of that can be seen by examining trades on the new Intercontinental Exchange. A substantial number of the trades are indexed to Platts, Natural Gas Intelligence or Natural Gas Week. Also, the New York Mercantile Exchange uses Platts' and NGI's published indices at different locations to settle some over the counter trades.

Here, it is important to note just what price reporting by publications is and what it is not. It is not a mandatory survey that collects data on every transaction and has an extended time period for reviewing all the data points.

It is what the market requires, however: a snapshot that collects as much information as possible on where the market is at the time and publishes it quickly so that market participants can react. The market requires this fast feedback in order to work efficiently.

Journalistic organizations typically have been the collectors and publishers of market price information for several reasons: (1) they are unbiased and hold no interests in the commodities traded; (2) they work smart and fast and have ironclad deadlines; (3) they can promise confidentiality of competitive information under most circumstances because of press protections written into the First Amendment to the Constitution. Prices of many commodities are tracked by journalistic organizations. As an example, American Metal Markets has been publishing prices for steel and other metals since the 1880s.

Because the science of price reporting is inexact - as most sciences are - it functions best if there are competing sources of market information - from other publications, from Nymex and other exchanges. This keeps all of our feet to the fire to put our best efforts into the enterprise. Noted petroleum industry analyst Phillip Verleger has endorsed published price indices, saying "You need at least two, maybe three or maybe four aggressively competitive press organizations" tracking the market." You could include the expanding trading exchanges as providing some of that competition.

Competition, Not Controls

Thus competition must take the place of rigid controls. It is not perfect, but it is the basis of capitalism, which has evolved as the best system mankind has so far found to direct a large society. Some who advocate other regimes have taken this opportunity, when some of the competitive system's flaws have been exposed, to move toward reinstating controls that already have proven to be a total failure.

To set up a mandatory government-run price reporting system, first would take an act of Congress, and second hundreds of millions of dollars to create a system that would work fast enough to meet the market requirements for information. While the government is good at spending large amounts of money, it works better on a longer time frame. The Energy Information Administration revises its data months, if not years, after the fact. FERC's new FERRIS electronic bulletin board has been something less than troublefree. The market is not likely to survive an extended systems shakeout period.

We would also note that the strict government oversight and reporting requirements for pricing publications, which some are advocating in the absence of a government-run survey, would slow down the process and compromise the competitive confidentiality that is a key ingredient in the surveys. NGI also questions the motivations of those who are proposing that solution.

We would not support auditing either of the reporting company's submissions or the publications practices, since they could compromise confidentiality. Also, it can be argued that audits are a needless and expensive waste of time, producing only dubious results -- as Enron, Arthur Andersen and numerous other companies have recently shown.

Regarding those calls for detailed reporting and government oversight, we are reminded of the best method for boiling a frog. If the frog is dumped into hot water, it will react and immediately jump out. If it is put in a pot of cold water, which is gradually warmed up to boiling, it won't notice the change until it's too late.

Likewise, if a government agency has put all the systems in place for collecting all the information on who is buying and who is selling, and how much and at what price, how big a step is it to start controlling those prices and that market? It's an interesting strategy - unless you're the frog.

Ellen Beswick                                 Mark Curran

Editor/Publisher                              Price Editor