Enron and the Raptors

By Mark P. Holtzman, Elizabeth Venuti, and Robert Fonfeder

In Brief

SPEs that Flourish in Loopholes

Many of Enron’s troubles can be traced to a number of special purposes entities (SPE), dubbed the Raptors, that Enron established to shield itself from mark-to-market losses in its growing equity investment business. When these investments went south, Enron’s attempts to shore the Raptors up with its own stock proved to be a temporary solution at best. The presence of Enron’s CFO on the board of directors of the SPE that funded the Raptors, LJM2, probably ensured that the entire house of cards would eventually come down. The authors detail the byzantine structures of these SPEs and demonstrate how existing GAAP requirements, though somewhat ambiguous, should have led to different treatment.

On October 16, 2001, Enron’s routine announcement of $0.43 recurring third-quarter earnings per share led to a series of events that would bankrupt this global Fortune 10 corporation, inaugurate an unprecedented series of large-scale accounting scandals, and permanently change the dynamics of the auditing profession. The end of the press release included a carefully worded announcement of an after-tax nonrecurring charge of $1.01 billion, or $1.10 per share, attributable to asset impairments at a subsidiary, Azurix Corp. ($287 million); the restructuring of another division, Broadband Services ($180 million); and “$544 million related to losses associated with certain investments, principally Enron’s interest in the New Power Company, broadband and technology investments, and early termination during the third quarter of certain structured finance arrangements with a previously disclosed entity.”

A year earlier, Enron stock had been trading at $90 per share. With suspicions now raised by the company’s announcement, investors quickly drove the price down from $34 to $29. Three days later, the Wall Street Journal delivered the next barrage of bad news. Secret limited partnerships controlled by Enron’s own executive vice president and CFO, Andrew S. Fastow, allegedly realized millions of dollars in profits from doing business with Enron. One of the partnerships wrote put options on Enron’s own stock at a loss of $10.5 million to Enron. By October 24, Fastow was placed on leave of absence, and a new CFO was appointed. On November 8, Enron announced restatements of its financial statements and, a day later, a proposed merger with Dynegy. By December 2, merger talks had broken off and the company filed for bankruptcy. (A few months later, Dynegy sank into a scandal of its own.)

In the short term, investor confidence in financial reporting and market investing has been shaken. Given the heightened scrutiny, more companies, such as Tyco and WorldCom, have withered or even collapsed. And even blue chips such as IBM and General Electric have suffered major sell-offs. Furthermore, the Enron scandal has led to a series of events that in the long term will probably change the scope of financial reporting, the standards-setting process, the function of the auditor, and audit regulation.

Enron managers went to extraordinary lengths to circumvent accounting rules in order to artificially increase earnings through a series of schemes involving the “Raptors”—a group of entities designed to both buffer Enron’s earnings from mark-to-market write-downs and to pay millions of dollars to a handful of Enron executives and their friends. Congressional testimony and documents released by Enron, its auditor Arthur Andersen, and federal regulators, and the Report of the Special Investigation Committee by Enron directors William C. Powers, Jr., Herbert S. Winokur, Jr., and Raymond S. Troubh (the Powers Report) provide the foundation for understanding the Raptors.

Special Purpose Entities

A special purpose entity (SPE) is a trust, corporation, limited partnership, or other legal vehicle authorized to carry out specific activities as enumerated in its establishing legal document. The SPE provides its sponsor with financing and liquidity while offering creditors protection against the sponsor’s bankruptcy. A sponsor would typically create an SPE to carry out a specific function through some type of asset transfer. For example, an SPE might borrow cash from a third-party creditor. In exchange for that cash, the sponsor sells an asset to the SPE and then leases it back under an operating lease (sale-leaseback). Under certain circumstances, the debt used by the SPE to acquire the asset would be its own liability and would not appear on the sponsor’s balance sheet.

Companies also use SPEs to access capital markets and manage risk. For example, an SPE might issue debt or equity, using the proceeds to acquire financial instruments (such as home mortgages) from its sponsor. Such a transfer of financial assets, if it qualifies as a sale or purchase, lowers the sponsor’s cost of capital, because it isolates the assets from the risk of sponsor bankruptcy. Loan underwriters and credit-rating agencies often require business entities involved in synthetic leasing and commercial mortgage-backed securities to be SPEs. Furthermore, the transferor generally derives a tax advantage because the SPE is a pass-through entity that does not pay its own taxes. The benefits of lower financing costs, lower taxes, and off–balance sheet financing can often outweigh the cost of establishing and maintaining the SPE. In short, the SPE was designed, in part, to minimize risk, but also to bypass accounting treatments that would otherwise increase leverage and decrease earnings.

GAAP for SPEs

Three different sets of standards were in place for SPE accounting before the collapse of Enron: SFAS 125, SFAS 140, and a series of Emerging Issues Task Force (EITF) releases. SFAS 125 criteria for sale treatment do not apply when the assets held by the qualifying SPE are financial assets created either through converting nonfinancial assets, such as real estate or servicing assets, into a financial asset or by recognizing previously unrecognized financial assets (EITF 96-20). Because the Raptors’ assets fell into these categories, SFAS 125 criteria would not have applied. SFAS 140 is effective for transfers and servicing of financial assets and extinguishments of liabilities occurring after March 31, 2001, making it inapplicable to the Enron transactions. Therefore, GAAP for the Raptor transactions is based on a series of EITF releases (EITF 84-30, 90-15, 96-21, and Appendix D-14). These releases permit nonconsolidation and sales recognition by the sponsor when an independent third party meets the following conditions:

An Elaborate Magic Trick: LJM2

As Enron increased its portfolio of equity investments, company managers found that mark-to-market losses were having an increasingly adverse effect on net income. Accordingly, they began to design SPE transactions to offset these losses. For example, Enron’s investment in Rhythms NetConnections had increased from $10 million in March 1998 to $300 million in May 1999. Managers wanted to lock in the $290 million gain, even though the company was prohibited from selling its shares before December 1999. In June 1999, an Enron SPE, LJM Swap Sub LLP, controlled by Enron CFO Andrew Fastow, issued a put option on the Rhythms shares in exchange for a cache of restricted Enron stock and various notes. This put option would entitle Enron to recover any loss in the value of the Rhythms shares. The SPE would pay for such losses out of the value of the restricted Enron stock it received. In short, any mark-to-market losses would be shifted from Enron to the SPE.

Subsequently amended numerous times, the deal appears to have clearly violated GAAP SPE requirements. First, there was no substantive capital investment; the equity level dropped below 3% of assets. Second, the SPE was controlled by Enron’s CFO, who could not possibly have been an independent third party to the transactions. Third, the numerous amendments made to the agreement—before it was wound down—suggest that Enron shielded LJM Swap Sub LLP from the full losses that it had initially contracted to incur. Worst of all, the SPE could not provide an effective hedge against mark-to-market losses.

As Enron continued to experiment with SPEs, it found that the 3% third-party equity requirement prevented it from creating larger, more effective hedges. Because previous SPE equity-holders had earned exceptionally high returns, managers decided to use this strong investment record to lure outside capital for new, large-scale SPEs. Higher equity levels would help meet the 3% equity requirement and permit Enron to effectively hedge a larger number of risky investments. Enron established LJM2 Co-Investment LP (LJM2), issuing a private placement offering and raising $394 million in limited partnership equity. The general partner was an entity owned by Fastow and subordinate Enron employee Michael Kopper.

Raptor I/Talon. The Raptor deals were designed to use appreciated Enron stock to hedge against volatile assets. In April 2000, Enron and LJM2 set up an SPE named Talon LLC. Harrier, an Enron subsidiary, invested $1,000 equity in Talon for “special interest limited partnership equity,” entitling Enron/Harrier to execute equity swap transactions with Talon, and, under certain circumstances, to all of Talon’s profits exceeding $41 million. Source documents, however, indicate that Talon was never designed to earn profits: it was designed to absorb Enron’s mark-to-market losses.

LJM2 contributed $30 million of equity to Talon, with the understanding that it would receive a return of $41 million on this investment before Enron/Harrier could proceed to trade derivatives with Talon. To fund future hedge transactions, Enron/Harrier issued to Talon approximately 3.7 million restricted shares of Enron stock, and rights to an additional 3.9 million shares to be issued under certain circumstances in future years. All of these shares had a current face value of approximately $537 million. In ex-change for these shares and options, Enron received net notes from Talon worth $350 million. See Exhibit 1 for a hypothetical reconstruction of Enron’s journal entries for this arrangement.

The initial transaction was designed to provide LJM2 with its required $41 million return. Talon sold Enron a put option for $41 million, granting Enron the right to sell Talon 7.2 million Enron shares at a strike price of $57.50 per share, exercisable six months later. At the time of issuance, Enron stock traded at $68 per share. If Enron’s share price did not drop by more than $10 per share by October, the put option would expire and Talon could pocket a $41 million gain, a 273% annualized return on investment.

Two months before the put option expired, Enron agreed to terminate the put early in exchange for a $4 million payment from Talon. Enron booked the loss as a reduction in equity, based on the nature of its settlement and EITF 2000-07 requirements. Talon, recording a $37 million gain (370% annualized return on investment), delivered $41 million to LJM2, and was ready to hedge with Enron. (For a visual description of LJM2 and Talon, see Exhibit 2 and Exhibit 3.)

Mark-to-Market Losses Disappear

As an example of the many derivative hedges that followed, the Powers Report describes Enron’s August 3, 2000, hedging of its investment in publicly traded Avici Systems. At the time, Avici was trading at an all-time high, $162.50 per share. The Powers Report concluded that the contract was actually entered into in mid-September, when the stock was trading at a significantly lower price, then backdated to August 3, 2000. By the September 30 quarter-end, Avici was trading at $95.50 per share, and Enron avoided recording any mark-to-market losses.

As Enron cherry-picked its bad investments and then hedged them, and as the Enron stock supporting these money-losing hedges dropped in value, Raptor I rapidly lost the credit capacity it needed to support further hedge activity. As a crutch for the beleaguered SPE, Enron entered into a “costless collar” contract with Talon. If Enron’s stock price fell below $81, endangering the SPE’s credit capacity, Enron would pay Talon the loss, directly bailing out the SPE. If Enron’s stock exceeded $116, Talon would pay Enron the gain. If Enron’s stock price remained between $81 and $116, then the costless collar required no action.

Raptor II/Timberwolf and Raptor IV/Bobcat. Having exhausted the credit capacity of Raptor I, Enron established Raptor II, called Timberwolf LLC, on September 22, 2000, and Raptor IV, called Bobcat LLC, on December 28, 2000. Both SPEs were modeled after Raptor I, with a $30 million initial investment by LJM2, the transfer of Enron stock to the SPE, put options on Enron stock for $41 million, and costless collars. While Timberwolf (Raptor II) functioned in much the same way as Talon (Raptor I), Bobcat (Raptor IV) never traded derivatives with Enron, but, in a transaction explained below, supplied additional credit to support its fellow Raptors.

Raptor III/Porcupine. The third Raptor SPE, Porcupine LLC, is perhaps the most puzzling. This transaction was virtually identical to the other Raptors, except, instead of supporting the hedge transactions with Enron stock and a costless collar, Porcupine received stock warrants of New Power Holdings, the investment on which it was writing puts in the first place. In other words, Porcupine would fund any loss in Enron’s New Power stock warrants out of the remaining value of its own New Power stock warrants. The Powers Report explains this as the equivalent of a “doubling-down.” If New Power’s price increased, then the hedge agreement would shift the gains from Enron to Porcupine. Furthermore, Porcupine would receive gains from its own New Power holdings. If New Power’s price deteriorated, however, it would be faced with funding the stock puts out of its own deteriorating investment in New Power. Reviewing Congressional testimony and supporting materials for the Powers Report, the authors found nothing to explain this arrangement. Perhaps the managers were confident that the stock warrants would substantially appreciate, so that the income recorded from two long positions—that is, income recorded twice—could later be used to offset and fund the losses incurred by other Raptor transactions.

Here’s what happened. On September 27, 2000, Porcupine received a $30 million investment from LJM2 and purchased New Power warrants from Enron for $10.75 each. On October 5, the New Power initial public offering price was $21 per share, increasing to $27 at the close of its first trading day. Porcupine cashed in a sufficient number of warrants to return $39.5 million in cash, the orally agreed-upon amount, to LJM2. Now able to engage in hedging activity with Enron, Porcupine sold a total return swap to Enron covering 18 million shares of New Power at $21 per share. This hedging contract would force Porcupine to compensate for any fluctuations in the value of Enron’s investment in New Power. By December 31, the price of New Power dropped below $10. Porcupine had insufficient reserves to support its hedge agreement with Enron, because it could cover losses only from its own near-worthless investment in New Power. Nevertheless, Enron booked a gain of approximately $370 million.

By the end of 2000, after several months of money-losing hedge transactions buttressed only by falling Enron stock prices, Raptors I, II, and III were suffering severe credit crises; only Raptors II and IV had any excess credit capacity. If these SPEs were unable to absorb Enron’s losses, the company would be forced to book reserves against net gains in excess of $500 million. On December 22, 2000, Enron managers engineered a 45-day cross-guarantee agreement, merging the credit capacity of all four Raptors. Excess credit in Raptors II and IV could be used to support the other Raptors, and Enron would not have to book any reserves against its gains on the hedge transactions. From an accounting perspective, it is difficult to understand how a temporary cross-guarantee agreement would provide assurance that Enron would collect hedging gains underwritten by Raptors I and III. Such an agreement offered only a crutch so that Enron could utilize the remaining credit in Raptor IV to engage in more hedging transactions.

By March 2001, the Raptors’ total credit shortfall increased to $504 million. To eliminate this shortfall, the Raptors entered into a permanent cross-collateralization agreement, which would permit the Raptor SPEs to collect credit deficiencies from each other. In order to relieve credit deficiencies in Raptors II and IV (which could also now be used to relieve problems in I and III), Enron agreed to deliver 18 million shares of its own stock and warrants exercisable for an additional 12 million shares in 2005, in exchange for notes totaling $828 million. To absorb the effects of any additional fluctuations in Enron stock on the Raptors, Enron entered into additional costless collars with the Raptors. According to the Powers Report, although these actual transactions were dated April 13, 2001, they were made to be effective on March 26, 2001, a few days before quarter-end, so that Enron recorded a credit reserve loss of only $36.6 million.

By the third quarter of 2001, it became clear that the Raptors (with a combined deficiency of over $500 million) were no longer sustainable, and Enron agreed to buy these SPEs from Fastow’s LJM2 partnership for $35 million. To cover the Raptors’ deficiencies, Enron would be forced to record a $544 million loss (net of tax effects). This write-off represented the bulk of the nonrecurring charges in Enron’s October 16, 2001, earnings announcement.

To account for notes receivables from the Raptors, Enron recorded an asset. Because the notes were secured with Enron stock and stock rights, the company should have reported them as contra-equity items. By the third quarter of 2001, the accounting staff at Enron had reclassified $1.2 billion from assets to a reduction in shareholders’ equity.

Accounting for the Raptors

With regard to GAAP, the following five questions appear to be at issue:

Did an independent third party make a substantive (3%) capital investment in the SPE? It is difficult to understand how a partnership controlled by Enron’s executive vice president and CFO could be construed as an independent third party. However, at initiation, limited partners contributed $394 million to LJM2, so that one could argue that LJM2 and Talon initially were capitalized adequately at 3% of equity. The capital investment may have been substantive, but it clearly was not independent.

Did the independent third party have control over the assets transferred to the SPE? The LJM2 private placement provides evidence that LJM2 would not have full control over the transferred assets, stating that: the partnership expects to make Investments in portfolio companies over which Enron [acquires or retains] ownership or control. The partnership may not have the power, acting alone, to control a portfolio company’s board of directors, management, or operations. In addition, the partnership may not have the ability, acting alone, to cause a portfolio company to take, or refrain from taking, certain actions, or to cause a portfolio company to engage, or refrain from engaging, in material transactions which conceivably could have an adverse effect on the partnership’s investment, and the partnership may not have the ability, acting alone, to control the timing of the liquidation of its investment.

Furthermore, judging from the large number of backdated, money-losing transactions, not only did the SPE not have control over its own assets, but also it had no control over when and how it would acquire assets and assume liabilities.

Did the independent third party possess the substantial risks and rewards of ownership of the SPE assets, including any residuals? The sweetheart deals of selling put options that would probably never be in the money (for $41 million), the costless collars, the cross-collateralization agreements, and the final purchase of the then-worthless Raptors for $35 million strongly suggest that Enron shielded LJM2 equity holders from risks of loss. This shielding of LJM2’s equity holdings was ambiguously worded in the partnership’s private placement offering: [T]he partnership may be forced to rely on the fact that Enron will possess some or all of the foregoing control rights and that the interest of the partnership and Enron will be sufficiently aligned such that Enron will exercise those rights in a manner that will protect the partnership’s investment. [Nevertheless,] Enron will have no obligation to align its interests with those of the partnership.

In other words, because, for some unspecified reason, the interests of Enron and LJM2 were “aligned,” Enron was expected to act in the partners’ best interests, even though it had no obligation to do so.

Were disclosures about the nature of the related-party transactions adequate (SFAS 57)? SFAS 57 provides guidance for disclosing the nature of transactions with related parties. For purposes of SFAS 57, “related parties” of an enterprise include management (members of the board of directors, CEOs, COOs, vice-presidents, and other persons without formal titles), affiliates (any party that directly or indirectly controls, is controlled by, or is under common control with, an enterprise), and even members of managers’ immediate families. Following these guidelines, an SPE may qualify for related-party treatment even if it is not required to be consolidated into its sponsor’s financial statements.

If a related-party relationship exists, disclosure must include the following:

If the reporting enterprise and one or more other enterprises are under common ownership or management control and the existence of that control could influence operating results or the financial position of the reporting enterprise, the nature of the control relationship must be disclosed. This holds true even if there are no transactions between the parties. Furthermore, the SEC requires companies to identify the dollar amounts of related-party transactions on the balance sheet, income statement, and cash flow statement [Regulation S-X, rule 4-08(k)].

In its December 31, 2000, notes to the financial statements, Enron disclosed many details about the nature of its transactions with LJM2 (dubbed “the Related Party”) and Andrew Fastow (referred to only as “a senior officer of Enron”). Many of these details, however, contradict subsequently revealed information. For example, the notes state that “management believes that the terms of the transactions with the Related Party were reasonable compared to those which could have been negotiated with unrelated third parties.” Yet the $41 million put options on Enron stock clearly had unreasonable monetary values attached. Furthermore, the notes to the financial statements do not disclose the dollar values associated with the transactions. Enron Executive Vice President and Chief Accounting Officer Rick Causey told the commission preparing the Powers Report that he did not recall why the disclosures omitted dollar amounts; he thought that SFAS 57 did not require disclosure of the amounts of related-party notes.

Were guarantees properly disclosed and recorded (SFAS No. 105 and 125)? There were no known explicit guarantees made to Fastow, his LJM2 partners, or the individual Raptor entities. The Powers Report suggests, however, that tacit guarantees were made to Fastow and partners. Otherwise, why did Enron pay $35 million to bail out the Raptors? First, the existence of any such guarantees would require that the sponsor consolidate the SPEs in question under EITF 90-15 and D-14. Furthermore, SFAS 107 defines such guarantees as financial instruments or liabilities, requiring either disclosure of the fair value of such guarantees or, if estimation of fair value is not practicable, descriptive information pertinent to estimating the financial instrument’s value. SFAS 125 requires transfers of such financial instruments to be initially recognized at fair value.

Proper Accounting Treatment

LJM2 and the Raptors were not controlled by an independent party that possessed the substantial risks and rewards of ownership. Therefore, these entities were, in substance, part of Enron, and should have been consolidated into Enron’s own financial statements. Any gains or losses recorded by Enron, including the hedge transactions described above, should have been eliminated from Enron’s financial statements. The equity shares of Fastow and other partners should have been accounted for as minority interest holdings. Furthermore, Enron management should have provided greater detail about the related-party transactions with Fastow, including the specific amounts of transactions recorded, and the nature of any specific guarantees made to Fastow and other equity holders.

The scope of accounting principles. University of Chicago Professor Richard Leftwich has commented, “It takes the FASB two years to issue a ruling and the investment bankers two weeks to figure out a way around it.” From the Raptor transactions, and numerous others described in the Powers Report, Congressional testimony, and newspaper reports, Enron may have paid out well over $300 million—in the form of cash, investments, and Enron stock— to advisors and SPE equity holders in order to sustain its network of off–balance sheet financing entities. By comparison, the Financial Accounting Foundation spent just $22 million to generate and maintain its FASB and GASB standards-setting programs. As a result, it is not difficult to see how determined companies can run rings around GAAP, exploiting technicalities and loopholes to create financial statements that even the most sophisticated investors cannot understand.

The question has been raised as to the adequacy of the standards governing SPE accounting. The old patchwork of EITF releases and FASB standards was, at best, confusing and inconsistent. Such problems may be resolved by new changes introduced by FASB. SFAS 140, Accounting for Transfers and Servicing of Financial Assets and Extinguishments of Liabilities, offers more specific definitions of SPEs and guidance for related transactions. FASB Interpretation 46, Consolidation of Variable Interest Entities, identifies “variable interest entities” as legal structures that either do not have equity investors with voting rights or have equity investors that provide insufficient financial resources to support the entities’ activities. This definition includes most types of SPEs. Such variable interest entities must be consolidated with their “primary beneficiaries,” companies which either bear the risk of loss from the entity’s activities or are entitled to receive a majority of the entity’s residual returns.

New standards and increased regulation brought about by the Sarbanes-Oxley Act now define the post-Enron landscape. The profession remains faced with a challenging task ahead: to restore public confidence in the reliability of financial reporting.