Gas and electricity price indices play a central role in the price discovery
process in the energy market. Moreover, these indices are used to determine
prices on billions of dollars of physical and financial energy transactions.
Real money and real decisions ride on these widely employed price benchmarks.

Unfortunately, the system for collecting and disseminating price information
used to calculate indices that the industry has employed
for the past decade is broken. The past two years have seen numerous revelations
regarding the false reporting. Prior to publication, major energy companies
have already agreed to pay sums totaling $130 million to the government to
settle claims of misreporting. The Commodity Futures Trading Commission is
pursuing other cases in which penalties could exceed $100 million; in one case,
the CFTC alleges that three-quarters of the prices or volumes reported by one
company (representing thousands of deals) were knowingly incorrect. Two individual
energy traders face felony charges for reporting misleading price information
to index publishers. Numerous companies have decided to abstain from reporting
price data.

These problems have spurred congressional action. Section 333 of the Oil and
Gas Title of the draft Energy Bill recently before Congress inserted a new
Section 24 to the Natural Gas Act (15 USC 717 et seq.) titled “Natural
Gas Market Transparency.” This section would authorize the Federal Energy
Regulatory Commission (FERC) to “issue rules directing all entities subject
to the Commission’s jurisdiction … to timely report information
about the availability and prices of natural gas sold at wholesale in interstate
commerce.”

Some envision that FERC will operate its own price-reporting mechanism, but
FERC Chairman Pat Wood has expressed reservations about a direct FERC role
in the process. There is a way, however, to meet Congress’s objective
of improving energy market transparency without direct FERC involvement in
the collection and dissemination of price information. This is to create a
truly independent third party that utilizes well-established methodologies
and technologies to collect price reports from market participants, validate
the accuracy of these reports, and distribute the validated price and volume
data (stripped of any commercially sensitive information) on an open access
basis to index publishers, data vendors, and other users of energy price information.

Although it would be a new development for the energy industry, such a “
data hub” would follow well-established precedents in the financial markets.
In response to similar concerns about inadequate transparency, participants
in the government, corporate, and municipal bond markets have created single
data hubs to collect and disseminate validated price data on a timely basis.

Relying on a patched up version of the old model of price reporting to fix
the well-documented problems with energy price indices would represent, as
Samuel Johnson said of second marriages, a “triumph of hope over experience.” The
data hub can rectify the recognized problems of the legacy system efficiently
and fairly.

The Virtues of a Data Hub

A single data hub offers many advantages over two alternative proposals to
improve price transparency: fixing the existing system and permitting the creation
of numerous competing data collection and dissemination entities.

Reporting data to a single hub is more economical for market participants.
Moreover, a truly independent third-party data collector with no competing
commercial interests offers a far more effective
way to restore confidence in the reliability of price data. Furthermore, a
data hub can readily implement best-practice methods for verifying the accuracy
and completeness of price reports.

In particular, a single data hub can implement most efficiently the best way
to assure the accuracy and validity of reported prices – the matching
of buys and sells. In a matching system, the buyer and the seller independently
submit a report containing the relevant terms of a transaction to a data collection
entity. The collector verifies that the reported terms match and reports prices
and volumes only for matched transactions.

A matching system prevents any individual from influencing price data by submitting
a false transaction report, or a report on a transaction that did not occur.
A false report will not – and cannot – be matched, and hence cannot
corrupt a price index. Nor can a report on a fictitious trade. Therefore, if
only matched trades are disseminated, an individual trader or firm cannot distort
price reports and price indices by submitting false trading data, either purposefully
or accidentally.

Matching also prevents another problem that has bedeviled price reporting,
and continues to do so – double counting. If trades are not matched,
and both the buyer and the seller report a transaction, this transaction will
be double counted in calculating an index.

In brief, matching is by far the single most important measure
that can be implemented to ensure that data disseminated to the marketplace
is accurate, reflects only bona fide transactions, and does not double count
trades.

Matching is nothing new. On organized futures markets, a trade
isn’t a trade until it’s matched. Moreover, the matching process
is already computerized. There are myriad systems in existence capable of matching
computerized buy and sell reports submitted by buyers and sellers; the data
hubs for government, corporate, and muni bonds use matching to validate data.
Due to the existence of this off-the-shelf technology, matching can be implemented
in a reasonable time frame at reasonable cost.

Fortunately, there is growing recognition among market participants that matching
of buys and sells is the best way to
ensure the accuracy of reported prices. The Global Energy Management Institute
of the University of Houston has taken a leadership role in educating market
participants and regulators on
the importance of matching. Recently, FERC specifically mentioned matching
as a means of validation in a staff report on price reporting issues. Similarly,
a document setting out areas of agreement and disagreement on price reporting
issues produced by a coalition
of market participants – including major trade associations, the Committee
of Chief Risk Officers (CCRO), some financial players, and the Coalition for
Energy Market Integrity and Transparency (EMIT) – specifically mentions
matching as a means of validating the accuracy
of price data; indeed, matching is the only validation method set
out in the coalition document.

Thus, it is widely acknowledged that matching can improve substantially the
accuracy of reported prices and the indices based thereon. Unfortunately, not
all engaged in the debate have grasped fully the implications of a matching
system. When these implications are understood, the case for an independent
data hub becomes strong indeed.

There are several important logical consequences that follow the acceptance
of matching as an essential part of ensuring the accuracy
of energy price data.

First, matching is most effective if all participants that submit data submit
it to a single data hub. Submission to multiple entities – whether they
be competing data hubs or publishers, or a combination of both – is duplicative
and excessively costly. Submitting trade information to multiple entities increases
the likelihood of a release of sensitive data. Most important, if some parties
submit data to one entity and some of their counterparties submit data to another,
it will be impossible to match some trades. This means less data will be available
to incorporate into indices and, as a result, the indices will reflect market
conditions less accurately. Indeed, missing even a few trades may make some
indices unreliable altogether. A single data hub maximizes the potential for
matches and therefore is capable of producing the best data and the best indices.

The thought of a monopoly makes some uncomfortable, but it must be recognized
that the fundamental economics of the data collection and dissemination process
impel consolidation and the emergence of
a very dominant – and likely monopoly – data hub. It is better
to recognize this at the outset and create a data hub with the appropriate
legal, regulatory, and organizational safeguards to ensure that it does not
exercise market power rather than to attempt to deal with an unconstrained
data monopolist at some date in the near future; Congress and the Securities
and Exchange Commission have recognized the importance of this issue for almost
30 years in securities markets and have developed means of coping with it that
can serve as models for the energy markets.

The second logical consequence of matching relates to the reporting of counterparty
data. Some, but by no means all, market participants have expressed serious
reservations about reporting counterparty data.

Counterparty data greatly facilitates the matching process. Indeed,
it is possible that effective matching is impractical without this data. Moreover,
counterparty information greatly facilitates the ability to detect other means
of manipulating price indices, such as wash trading.

It should also be noted that if counterparty data is not reported to a data
hub, two things can happen. First, the data hub may be able to match the buyer
and seller properly despite the fact that neither party submits counterparty
information – in which case, the data hub will know the counterparties
anyway. Second, the hub may incorrectly match a buyer’s submission and
a seller’s submission because they agree on price, quantity, and location
despite the fact that this buyer and seller did not in fact trade with one
another – in which case, the data hub has created a false transaction
record. This outcome is especially problematic if, as will almost certainly
be the case, government regulators have access to hub data for investigation
and enforcement purposes.

The Energy Bill (as recently proposed) addresses another crucial issue inherent
in a matching system; how can a data hub that matches trades to ensure data
quality obtain the critical mass of participation required to produce reliable
price data that accurately reflect the bulk of market activity? For matching
to work effectively,
a large fraction of transactions must be reported to the data hub.
A mandate that market participants report price data ensures that prices reported
to and disseminated by the data hub will reflect the full breadth of market
activity.

In this context, it should be noted that transparency initiatives in the
municipal bond and corporate bond markets are not voluntary in nature. The
Municipal
Securities Rulemaking Board requires its members to report their muni bond
transaction data, and the National Association of Securities Dealers requires
its members to report their corporate bond trades. These mandatory (but self-regulatory)
systems were implemented because purely voluntary efforts to improve price
reporting in these markets did not result in high reporting rates, nor did
they produce reliable price information that truly reflected the breadth of
trading activity.

Governance of the Hub

Although implementing a data hub that uses matching system to validate price
data is essential to restoring confidence in the price reporting process, it
is important to get the organization and governance of the data hub right to
ensure that this system works for the long haul.

Information is power, and price information is particularly potent. It is imperative
that market participants have the utmost confidence that the energy data hub
is operated in the interests of the industry as a whole and is not unduly influenced
by any particular party or segment. That is, the hub should be truly independent,
with an independent
board of directors responsible for all crucial decisions and operational oversight;
the recent travails of the New York Stock Exchange demonstrate the debilitating
effects of real and perceived conflicts of interest. At the same time, it should
be recognized that industry
input is essential to help the hub operate effectively and respond in
a constructive way to technological and structural changes in the marketplace.
It would therefore be desirable to have industry advisory boards to provide
information and feedback to the independent board (which should retain all
decision-making authority). Such a governance structure is similar to that
of the PJM ISO. Similarly, securities industry data hubs also tap on industry
expertise through advisory boards.

Relatedly, the data hub should be focused on one task – the collection
and dissemination of price information – and should not provide any other
services such as publishing, brokerage, execution, or clearing. It is imperative
that the data hub should not be able to exploit control of price information
to obtain a competitive advantage over rivals in these other activities. A
narrow focus on data collection and distribution precludes this possibility.

Data accuracy is the overriding goal of the data hub, and consequently its
organizational form should be chosen to enhance its incentive to produce high
quality, accurate price reports. The economics literature suggests that the
nonprofit form of organization can create better incentives to ensure quality
when the characteristics of a firm’s product are hard to monitor directly.
Inasmuch as this condition plausibly describes the process of data collection
and dissemination, the nonprofit form is arguably the best way to organize
the hub.

Quality – and confidence in quality – can also be enhanced through
independent audit of the hub’s processes and products.

Summary and Conclusions

An independent, not-for-profit entity that collects transaction reports electronically,
matches buyers’ and sellers’ reports to validate accuracy, and
disseminates price and volume information on an open access basis would improve
transparency and help restore confidence to the energy marketplace. Therefore,
the time is now ripe to implement an energy data hub as a crucial part of the
process of restoring confidence to the energy marketplace.