Mr. Chairman and members of the Committee, I want to thank you for inviting me to present my analysis of the accounting issues that led to Enron's downfall. I am honored to be given this opportunity.

I am Bala Dharan, professor of accounting at the Jesse H. Jones Graduate School of Management, Rice University, Houston. I received my PhD in accounting from Carnegie Mellon University, Pittsburgh. I have been an accounting professor at Rice University since 1982. In addition, I have taught accounting as a professor at Northwestern University's Kellogg School of Management, and as visiting professor at the Haas School of Business at University of California, Berkeley, and the Harvard Business School. I am also a Certified Public Accountant and a Registered Investment Advisor in the state of Texas. I have published several articles in research journals on the use of financial accounting disclosures by investors.

The Enron debacle will rank as one of the largest securities fraud cases in history. Evidence to date points to signs of accounting fraud involving false valuation of assets, misleading disclosures and bogus transactions to generate income. I have had several invitations to speak on Enron's accounting issues over the last few months. In my talks and lectures, I am asked two questions most frequently: One, how could this tragedy have happened while the company's management, board of directors and outside auditors were supposedly watching over for employees and investors? Two, what can we learn from this debacle so that we can avoid future Enrons? Undoubtedly the first question will be the focus of the many investigations currently under way, including your Committee's efforts. In my testimony, I will focus on what we can learn from the accounting issues related to Enron's use of mark-to-market (MTM) accounting and special purpose entities (SPEs). These two issues are very closely related, especially as they were practiced by Enron. In addition, I will address the related accounting issue of pro-forma disclosures, and also how Enron's failed business strategy contributed to the accounting errors. I hope other invited panelists addressing before this Committee will talk about the critical roles played by Enron's management, board, auditors, lawyers, consultants, financial analysts, and investment bankers in Enron's fall. I conclude with recommendations for regulatory changes and improvements in the accounting and auditing rules governing special purpose entities, mark-to-market accounting, and financial disclosures in general. 

1. Loss of Investor Trust

My analysis of the Enron debacle shows that Enron's fall was initiated by a flawed and failed corporate strategy, which led to an astounding number of bad business decisions. But unlike other normal corporate failures, Enron's fall was ultimately precipitated by the company's pervasive and sustained use of aggressive accounting tactics to generate misleading disclosures intended to hide the bad business decisions from shareholders. The failure of Enron points to an unparalleled breakdown at every level of the usual system of checks that investors, lenders and employees rely on - broken or missing belief systems and boundary systems to govern the behavior of senior management, weak corporate governance by board of directors and its audit committee, and compromised independence in the attestation of financial statements by external auditor.

Enron started its transformation from a pipeline company to a "risk intermediation" company in the 1980s. It adopted a corporate strategy of an "asset-less" company, or a "frictionless company with no assets." The company's Chief Financial Officer said in a 1999 interview to a management magazine (which awarded him "CFO Excellence Award for Capital Structure Management") that the top management transformed Enron into "one engaged in the intermediation of both commodity and capital risk positions. Essentially, we would buy and sell risk positions." What this description of the company implies is that unlike any other major company in the US, Enron's corporate strategy was virtually devoid of any boundary system that defined the perimeter of what is an acceptable and unacceptable investment idea for managers to pursue. Since any business investment basically involves some risk position, this strategy is not really a strategy at all but an invitation to do anything one pleases. Enron's top management essentially gave its managers a blank order to "just do it", to do any "deal origination" that generated a desired return. "Deals" in such unrelated areas as weather derivatives, water services, metals trading, broadband supply and power plant could all be justified and approved by managers under this concept of an asset-less risk intermediation company. The company even briefly changed its tagline in a company banner from "the world's leading energy company" (which implies some boundary system for investments) to "the world's leading company." It is no wonder that this flawed business strategy led to colossal investment mistakes in virtually every new area that the company tried to enter.

While bad business strategy and bad investment decisions can and do contribute to a company's fall, it is a company's desperate attempt to use accounting tricks to hide bad decisions that often seals its fate. My analysis of cases of major stock price declines shows that when news of an unanticipated business problem, such as a new product competition or obsolescence of technology, is released to the market, the company's stock price does take a hit, but it often recovers over time if the company takes appropriate and timely management actions. However, when a company loses the trust and confidence of the investing public because of discoveries of accounting wrongdoings, the net result on the company's stock price and competitive position is mostly devastating and long-lasting. This is because accounting reports are the principal means by which investors evaluate the company's past performance and future prospects, and a loss of trust effectively turns away investor interest in the company.

My analysis also suggests that it is not possible to recover from a loss of investor confidence by some quick management actions. Before re-admitting the company to their investment portfolios, investors would demand and seek evidence that the accounting numbers are again reliable, and this process of rebuilding of trust often takes place through several quarters of reliable financial disclosures. If the company's finances are not fundamentally sound to begin with, then it is quite likely that the company would not survive this long trust-recovery phase intact. This is exactly what happened in the case of Enron. Burdened with dozens of failing investments and assets hidden in special purpose entities whose very existence and financing often depended on high stock price of Enron's shares, the company quickly entered a death-spiral when investors questioned its accounting practices and pushed its share price down to pennies. 

2. Use of Pro-Forma Earnings

Enron's loss of investor faith started with the company's 2001 third quarter earnings release on October 16, 2001. As earnings releases go, this one must rank as one of the most misleading. The news release said in an underlined and capitalized headline, "Enron Reports Recurring Third Quarter Earnings of $0.43 per diluted shares." The headline went on to reaffirm "recurring earnings" for the following year, 2002, of $2.15 per share, a projected increase of 19% from 2001. But an investor had to dig deep into the news release to know that Enron actually lost $618 million that quarter, for a loss of ($0.84) per share. A net loss of $618 million loss was converted to a "recurring net income" of $393 million by conveniently labeling and excluding $1.01 billion of expenses and losses as "non-recurring".

The practice of labeling certain earnings items as non-recurring or "one-time" has unfortunately become widespread in the US, and has corrupted corporate disclosure environment to the detriment of investors and the public. Companies ranging from General Motors to Cisco mention some form of pro-forma earnings in their earnings disclosures. Of course, there is nothing "one-time" or "non-recurring" about the $1.01 billion of expenses and losses that Enron chose to label as such in its 2001 third quarter earnings release. In other words, neither accountants nor managers could assure that what they call non-recurring would not recur.

            My ongoing research also shows that the adoption of pro-forma earnings reporting is often a company's desperate response to hide underlying business problems from its investors. As an example, Enron did not always use pro-forma earnings in its news releases. Its earnings release as late July 24, 2000, for 2000 second quarter, did not contain any reference to recurring earnings. In its 2000 third quarter earnings release on October 17, 2000, Enron started using the recurring earnings in the body of the news release. We know from the Enron board's internal report dated February 1, 2002, that this was also the time when the senior management started worrying about the declining value of many of their merchant investments. By the following quarter, recurring earnings had been elevated by Enron to news headline.

Not all companies, of course, use pro-forma earnings or use them in blatantly misleading way. Companies like Microsoft do report their earnings without having to resort to misleading pro-forma disclosures. However, we need to ensure that misleading pro-forma disclosures are halted altogether. In a recent speech, the chairman of the Securities and Exchange Commission has warned companies that pro-forma earnings would be monitored by the SEC for misleading disclosures. However, this does not go far enough. The SEC should recognize all pro-forma disclosures for what they really are - a charade. They may differ from one another in the degree of deception, but the intent of all pro-forma earnings is the same - to direct investor attention away from net income measured using generally accepted accounting principles, i.e., GAAP earnings.

Enron's 2001 third quarter earnings press release on October 16, 2001, contained another major shortcoming - lack of information about its balance sheet and cash flows. While the company's press release provided information on net income, the company failed to provide a balance sheet. This is inexplicable - we teach in Accounting 101 that the income statement and the balance sheet are interrelated ("articulated") statements. This essentially means that we cannot really prepare one without preparing the other. Not surprisingly, almost every major company's earnings release contains the balance sheet along with its income statement. Financially responsible companies would also provide a cash flow statement. Analysts and investors puzzled with Enron's lack of balance sheet disclosure had to wait until after the markets closed on October 16, 2001, when the senior management disclosed in response to a question during the earnings conference call that it had taken a $1.2 billion charge against its shareholders' equity (a balance sheet item), including what was described as a $1 billion correction of an accounting error. The experience suggests that along with reforms on pro-forma earnings usage, we should mandate a fuller, more complete presentation of financial statements in the earnings news releases so that investors can truly be in a position to interpret the quality and usefulness of the reported earnings numbers. 

3. Special Purpose Entity Accounting

3.1. Business Purpose of SPEs

Enron's internal report released on February 1, 2002, makes clear that Enron used dozens of transactions with special purpose entities (SPEs) effectively controlled by the company to hide bad investments. These transactions were also used to report over $1 billion of false income. Many of these transactions were timed (or worse, illegally back-dated) just near end of quarters, so that the income can be booked just in time and in amounts needed, to meet investor expectations. However, SPEs were not originally created as mere tools of accounting manipulation. Surprisingly, the SPE industry did start with some good business purpose. Before discussing the accounting issues related to Special Purpose Entity (SPE) accounting, it would be useful to have a brief description of what these entities are and how they arose.

The origin of SPEs can be traced to the way large international projects were (and are) financed. Let's say a company wants to build a gas pipeline in Central Asia and needs to raise $1 billion. It may find that potential investors of the pipeline would want their risk and reward exposure limited to the pipeline, and not be subjected to the overall risks and rewards associated with the sponsoring company. In addition, the investors would want the pipeline to be a self-supported, independent entity with no fear that the sponsoring company would take it over or sell it. The investors are able to achieve these objectives by putting the pipeline into a special purpose entity that is limited by its charter to those permitted activities only. Thus a common historical use of SPE was to design it as a joint venture between a sponsoring company and a group of outside investors. The SPE would be limited by charter to certain permitted activities only - hence the name. Such an SPE is often described as brain-dead or at least on auto-pilot. Cash flows from the SPE's operations of the project are to be used to pay its investors.

In the US, the use of SPEs spread during the 1970s and 1980s to financial services industry. In the early 1980s, SPEs were used by the financial services firms to "securitize" (market as securities) assets that are otherwise generally illiquid and non-marketable, such as groups of mortgages or credit card receivables. Because they provide liquidity to certain assets and facilitate a more complete market for risk sharing, many SPEs can and do indeed serve a useful social purpose. 

3.2 Accounting Purposes of SPEs

These examples illustrate that SPEs can be motivated by a genuine business purpose, such as risk sharing among investors and isolation of project risk from company risk. But as we have seen from the Enron debacle, SPEs can also be motivated by a specific accounting goal, such as off-balance sheet financing. The desired accounting effects are made possible because of the fact that SPEs are not consolidated with the parent if they satisfy certain conditions. The accounting effects sought by the use of SPEs can be summarized into the following types:

1.   Hiding of Debt (Off-Balance Sheet Financing). The company tries to shift liabilities and associated assets to an SPE. The main purpose of forming the SPE in this case is to let the SPE borrow funds and not show the debt in the books of the sponsoring entity. The so-called "synthetic leases" are examples of this type of SPEs. In the 1980s SPEs became a popular way to execute synthetic lease transactions, in which a company desiring the use of a building or airplanes tries to structure the purchase or use in such a way that it does not result in a financial liability on the balance sheet. Though Enron's earlier use of SPEs may have been motivated by this objective, the key SPEs formed by Enron since 1997, such as Chewco, LJM1 and LJM2, were intended more for the other accounting objectives described below.

2.   Hiding of Poor-Performing Assets. This objective has a major factor in several SPE transactions of Enron. For example, Enron transferred poor-performing investments such as Rhythms NetConnections to SPEs, so that any subsequent declines in the value of these assets would not have to be recognized by Enron. In 2000 and 2001 alone, Enron was able to hide as much as $1 billion of losses from poor-performing merchant investments by these types of SPE transactions.

3. Earnings Management - Reporting Gains and Losses When Desired.  This accounting objective has also been a fundamental motivation for several of the complicated transactions arranged by Enron with SPEs with names such as Braveheart, LJM1 and Chewco. For example, Enron was able to transfer a long-term business contract - an agreement with Blockbuster Video to deliver movies on demand, to an SPE and report a "gain" of $111 million.

4. Quick execution of Related Party Transactions at desired prices. Enron's use of SPEs such as LJM1 and LJM2, controlled by its own senior managers, was specifically intended to do related party transactions quickly and when desired, at prices not negotiated at arms length but arrived at between parties who had clear conflicts of interest.  For example, the above Blockbuster deal was arranged at the very end of December 2000, just in time so that about $53 million of the "gain" could be included in the 2000 financial report. (The rest of the gain, $58 million, was reported in 2001 first quarter.) The purpose of this and several similar transactions by Enron seems to have been to use these transactions with SPEs controlled by its own senior executives to essentially create at short notice any amount of desired income, to meet investor expectations. 

There are three sets of accounting rules that permit the above financial statement effects of SPEs. One deals with balance sheet consolidation - whether or not SPEs such as synthetic leases should be consolidated or reported separately from the sponsoring entity. The second deals with sales recognition - when should the transfer of assets to an SPE be reported as a sale. The third deals with related party transactions - whether transfers of assets to related parties can be reported as revenue. Of these, the accounting problem that needs immediate fixing is the one dealing with consolidation of SPEs. This is addressed next. With respect to sales recognition rules and related party transaction rules, the problem may lie more with Enron's questionable accounting and corresponding auditor errors, rather than the rules themselves. However, Enron's revenue recognition from SPE transactions often depended on the so-called mark-to-market accounting rules which gave Enron the ability to assign arbitrary values to its energy and other business contracts. These rules do have certain problems that need fixing, and this issue is addressed in section 4. 

3.3. Consolidation of SPEs

Despite their potential for economic and business benefits, the use of SPEs has always raised the question of whether the sponsoring company has some other accounting motivations, such as hiding of debt, hiding of poor-performing assets, or earnings management. Additionally, the explosive growth in the use of SPEs led to debates among managers, auditors and accounting standards-setters as to whether and when SPEs should be consolidated. This is because the intended accounting effects of SPEs can only be achieved if the SPEs are reported as unconsolidated entities separate from the sponsoring entity. In other words, the sponsoring company needs to somehow keep its ownership in the SPE low enough so that it does not have to consolidate the SPE.

Thus consolidation rules for SPEs have been controversial and have been hotly contested between companies and accounting standards-setters from the very beginning. In the US, the involvement of the Financial Accounting Standards Board (FASB), the accounting standards-setting agency, in SPE accounting effectively started from 1977 when it issued lease capitalization rules to control the use of off-balance sheet financing with leases. Corporate management intent on skirting around the new lease capitalization rule appeared to have led to the rapid development of SPEs to do the so-called "synthetic leases". In the first of several accounting rules directed at SPEs, in 1984 the Emerging Issues Task Force (EITF) of the FASB issued EITF No. 84-15, "Grantor Trusts Consolidation." However, given the rapid growth of SPEs and their ever-widening range of applications, standards-setters were always a step or two behind and were being reactive rather than proactive in developing accounting rules to govern their proper use.

            The question of whether a sponsoring company should consolidate an SPE took a definitive turn in 1990 when the EITF, with the implicit concurrence of the SEC, issued a guidance called EITF 90-15. This guidance allowed the acceptance of the infamous "3 percent rule", i.e., an SPE need not be consolidated if at least 3 percent of its equity is owned by outside equity holders who bear ownership risk. Subsequently, the FASB formalized the above SPE accounting rule with Statement No. 125, and more recently Statement No. 140, issued in September 2000.

            An analysis of the development of the 3 percent rule suggests that the rule was an ad-hoc reaction to a specific issue faced by the FASB's Emerging Issues Task Force and was intended as a short-term band-aid, but has somehow been elevated to a permanent fix. More importantly, the rule, in many ways, was a major departure from the normal consolidation rules used for other subsidiaries and entities. In the US, we generally require full consolidation if a company owns (directly or indirectly) 50 percent or more of an entity. Thus the 3 percent rule is a major loosening of the normal consolidation rule. The motivation for this seems to have been that the SPEs were restricted in their activities by charter and thus the parent company could claim lack of control. The parent company only had to show that some other investors did indeed join the SPE venture with a significant exposure (signified by the 3 percent rule) in order to make the SPE economically real and thus take it off the books.

            Clearly the accounting for SPE consolidation needs to be fixed, starting with the abandonment of the 3 percent rule and its replacement with a more strictly defined "economic control" criterion. The need to fix consolidation rules has also been amply recognized by the FASB, which has been working for several years on a comprehensive "consolidation" project. However, the Enron debacle should give our standards-setters the needed push to rapidly complete this critical project and issue new rules for the proper consolidation of SPEs whose assets or management are effectively controlled by the sponsoring company. The rules should emphasize economic control rather than rely on some legal definition of ownership or on an arbitrary percentage ownership. Economic control should be assumed unless management can prove lack of control.

 4. Mark-to-Market Accounting and Earnings Management

In the US, financial assets, such as marketable securities, derivatives and financial contracts, are required to be reported on the balance sheet at their current market values, rather than their original acquisition cost. This is known as mark-to-market (MTM) accounting. MTM also requires changes in the market values for certain financial assets to be reported in the income statement, and in other cases in the shareholders' equity as a component of "Accumulated Other Comprehensive Income" (OCI), a new line item that was required for all public companies by FASB Statement No. 130 from 1997.

MTM was implemented in FASB Statement No. 115, issued in 1993, for financial assets that have readily determinable market values, such as stocks and traded futures and options. In 1996, FASB Statement No. 133 extended MTM to all financial derivatives, even those that do not have traded market values. For some derivatives, a company may have to use complex mathematical formulas to estimate a market value. Depending on the complexity of the financial contract, the proprietary formulas used by companies for market value estimation may depend on several dozen assumptions about interest rate, customers, costs, and prices, and require several hours of computing time. This means that it is hard, if not impossible, to verify or audit the resulting estimated market value. Of course, a consequence of this lack of verifiability is that MTM accounting can potentially provide ample opportunities for management to create and manage earnings. Thus MTM accounting represents the classic accounting struggle of weighing the trade-off between relevance and reliability - in this case the relevance of the market value data against the reliability of the data. In the end, the accounting standards-setters took the position that the increased benefit from reporting the market value information on the balance sheet justified the cost of decreased reliability of income statement and the earnings number.

            It will be useful to consider an example of how Enron recognized with MTM accounting, in order to understand how MTM can be easily manipulated by a company to manage earnings, especially with respect to financial contracts that do not have a ready market. Assume that Enron signed a contract with the city of Chicago to deliver electricity to several office buildings of the city government over the next twenty years, at fixed or pre-determined prices. The advantage to the city of Chicago from this "price risk management" activity is that it fixes its purchase price of electricity and allows the city government to budget and forecast future outlays for electricity without having to worry about price fluctuations in gas or electricity markets.

Enron sought and obtained exemptions from regulators to allow it report these types of long-term supply contracts as "merchant investments" rather than regulated contracts, and obtained permission from accounting standards-setters to value them using MTM accounting. Without MTM, Enron would be required to recognize no revenue at the time the contract is signed and report revenues and related costs only in future years for actual amounts of electricity supplied in each year. However, MTM accounting permits Enron to estimate the net present value of all future estimated revenues and costs from the contract and report this net amount as income in the year in which the supply contract is signed. The idea for such an accounting treatment seems to be based on the notion that the financial contract could have been sold to someone else immediately at the estimated market value, and hence investors would benefit from knowing this amount in the balance sheet and correspondingly in the income statement. Enron used similar MTM procedures to not only value merchant investments on its books but also to determine the selling price, and hence gain on sale, for investments it transferred to the various SPEs it controlled.

            A major problem with using MTM accounting for private contracts such as the one described above is that the valuation requires Enron to forecast or assume values for several dozen variables and for several years into the future. For example, the revenue forecasts may depend on assumptions about the exact timing of energy deregulation in various local markets, as well as 20 years of forecasts for demand for electricity, actions of other competitors, price elasticity, cost of gas, interest rates, and so on.

            While there are strong conceptual reasons to support MTM accounting, the Enron crisis points to at least some need to revisit and revise the current accounting rules for reporting transactions and assets that rely on MTM values. In particular, MTM rules should be modified to require that all gains calculated using MTM method for assets and contracts that do not have a ready market value should be reported only in "Other Comprehensive Income" in the balance sheet, rather than the income statement, until the company can meet some high "confidence level" about the realization of revenue for cash flows that are projected into future years. Normal revenue recognition rules do require that revenue should be recognized after service is performed, and moreover that revenue should be "realized or realizable", meaning that cash flow collection should be likely. In the absence of satisfying this condition, revenue rules (such as those explained in SEC Staff Accounting Bulletin 101) normally compel a company to wait until service is performed and cash collection probabilities are higher. Extending this logic to MTM accounting would protect the investing public from unverifiable and unauditable claims of gains being reported in the income statement.

 

5. Recommendations

The Enron Meltdown is a result of massive failure of corporate control and governance, and failures at several levels of outside checks and balances that investors and the public rely on, including an independent external audit. In my testimony, I have focused on the accounting issues, and in particular on the possible changes we need to make in these areas in order to prevent future Enrons. My recommendations are summarized below.

1.   The SEC, the New York Stock Exchange (NYSE) and the Nasdaq should adopt new rules severely restricting the format and use of pro-forma earnings reporting. All earnings communications by companies should emphasize earnings as computed by Generally Acceptable Accounting Standards. Any additional information provided by the company to highlight special or unusual items in the earnings number should be given in such a way that the GAAP income is still clearly the focus of the earnings disclosure.

2.   Companies should be reminded by regulators and auditors that the use of terms such as "one-time" or "non-recurring" about past events in earnings communications implies certain promises to investors about future performance, and therefore should not be used except in rare cases.

3.   Companies should present a complete set of financial statements, including a balance sheet and a cash flow statement, in all their earnings communications to the general public, in order to permit investors evaluate the quality of the reported earnings numbers.

4.   The FASB needs to accelerate its current project on consolidation accounting, and in particular, fix the consolidation rules in the accounting for Special Purpose Entities to prevent its continued abuse by corporations for earnings management. The current consolidation rules, including the "3 percent" rule for SPEs need to be abandoned and replaced with an "economic control" rule. The new rules need to emphasize economic control rather than rely on some legal definition of ownership or on an arbitrary percentage ownership. Economic control should be assumed unless management can prove lack of control. Similar rules should be extended to lease accounting.

5.   The FASB and the SEC need to consider requiring new disclosures on transactions between a company and its unconsolidated entities, including SPEs. In particular, more detailed footnote disclosures on the sale or transfer of assets to unconsolidated entities, recognition of income from such transfers, and the valuation of transferred assets should be required.

6.   The mark-to-market accounting methodology, while theoretically sound, needs to be modified in the light of what we have learned from the Enron meltdown. Traditional revenue recognition rules, such as the realization principle, should be extended to the recognition of gains and losses from MTM accounting. Forecasted cash flows beyond two or three years should be presumed to have a low level of confidence of collectibility. Gains resulting from present values of such cash flows should be recorded in the Accumulated Other Comprehensive Income in the balance sheet, rather than the income statement, until the confidence level increases to satisfy the usual realization criterion of collectibility.