Subject: File No. 4-573
From: Stephen T Smith
Affiliation: Controller, Conectiv Energy

November 11, 2008

November 11, 2008

I serve as the controller of the generation and wholesale marketing division of Pepco Holdings, Inc., a large, regional utility holding company. I have been involved with accounting for derivatives for more than 10 years. I have served on working groups at EEI, CCRO, and PJM that have addressed various aspects of derivatives accounting and reporting. My comments reflect my observations and opinions of mark-to-market accounting, and not necessarily those of Pepco Holdings or its management.

The wholesale energy sector relies heavily on energy derivatives to hedge its exposure to commodity markets which have become increasingly volatile since the roll back of the traditional state regulatory models of the past. On January 1, 1999, EITF 98-10, Accounting for Contracts Involved in Energy Trading and Risk Management Activities, went into effect. This set of industry-specific guidelines filled a void, and ushered mark-to-market accounting into widespread use in the energy industry.

Unfortunately, early mark-to-market rules were abused by a few large industry participants, and many large, long-dated contracts were being valued under a mark-to-model concept. In a number of instances, both parties were estimating large Day 1 gains on these contracts. Modeling techniques tended to be aggressive, and favor the company performing the valuation. Accounting firms did not have the expertise, or perhaps the incentive, to refute those rosy valuations. In 2001, SFAS No. 133, the long-awaited standard for derivative accounting and hedging, arrived to save the day.

Rules-based SFAS No. 133, at 875 pages and still growing (including amendments, examples and interpretations), is the largest project ever released by FASB. In their attempt to clarify every possible derivatives accounting situation in the standard, FASB created a new discipline within the accounting profession: the derivatives accountant. The standard is so complex and confusing, that it requires its own staff, or even department, in most public companies to manage it. Public accounting firms embraced the standard because it provided black-and-white rules for everything, or so it seemed.

As time went on, those of us in the trenches realized that FAS 133 wasnt all that it was supposed to be. Hedge accounting was an administrative nightmare, and mark-to-market accounting had broken through the borders of traditional financial products and was seeping into everything from inventory management to power grid tariffs. In the meantime, many aggressive companies were still marking complex contracts to internal models and running the bountiful harvest through current income.

In an effort to contain mark-to-model accounting, FASB issued SFAS No. 157, effective January 1, 2008. This supposed step back to principles-based accounting would require companies to disclose modeled valuations, and require them to used market-based data wherever available. The disclosure requirements of FAS 157 were very similar to the SECs FRR-61 disclosure requirements that had been in place since 2002. The standards major flaw, and its departure from being principles-based, is its one-size-fits-all definition of fair value. Exit price, the only permitted method of fair valuation under FAS 157, is a perfect theoretical measurement however, it assumes that you are willing and able to exit the market immediately. Unlike previous GAAP, it ignores the intent of the parties, or the structure of the contract itself, in regard to when the exit from the market will likely take place. In an illiquid marketplace, the exit price may be artificially depressed and only represent fire sale prices. Companies with the ability to hold an instrument through times of short-term volatility can be economically harmed by the requirement to write-down assets below their actual realizable value.

Future valuation guidelines should permit companies to use valuation techniques that fit the facts and circumstances of the instruments they hold. FASB doesnt have the ability to determine what is best in all cases. Using market-based data makes good sense, when it is available and reliable. When market data isnt available or reliable, audit firms should strongly encourage the use of traditional modeling techniques over proprietary models for fair valuations. If a firm is uncomfortable with an in-house model being used by a client, they have to draw the line and require corroboration under an alternate technique.

Occasionally, industry accountants find an area of GAAP that hasnt been filled in with a rule by FASB. However, before any of us can breathe a sigh of relief and apply common sense to fill the void, we check with our audit firm to see if they agree. Imagine our surprise when, instead of opening a rational discussion of whether our assessment makes sense, we are told that the major audit firms got together and decided that that they are all going to follow one common practice, regardless of the circumstances. Over the past few years, this is an all too familiar experience. This may be a result of PCAOB scrutiny, or fear of lawsuits, but it is not good accounting.

GAAP should continue to move toward convergence with IFRS. Our economy is becoming a global economy, and there should be one set of accounting standards. These standards should address high-level concepts. They should be principles based, so they will be flexible enough to permit similar, but not necessarily identical, accounting treatment of similar items across different companies. Below the standards, industry-specific guidance should be provided in greater detail. Departure from this guidance should be permissible, but rare, with disclosure of material differences from the norm. Audit firms should assist clients with interpretation issues, but not force them into accepting a result that does not fairly reflect the companys financial position, because of some obscure rule. The goal of the financial statements is to paint the most accurate financial position of the company as possible, with the understanding that accounting involves a great deal of estimation.

In conclusion, mark-to-market accounting (aka fair value accounting) is not the problem. Modern financial analysis techniques need real, current economic data to produce accurate results. The key is not to bog the process down with inflexible, complex rules. Todays dynamic marketplaces change too quickly for hard-and-fast rules to keep up. The principles of conservatism, and substance over form should be the overarching guidance used by industry, auditors and regulators when reporting financial information.