The Fall of Enron: What Actually Went Wrong, and Why It Mattered

Enron filed for bankruptcy in December 2001, fifteen months after its share price peak and weeks after restating five years of accounts. At the time it was the largest corporate bankruptcy in United States history and it took Arthur Andersen, one of the five global audit firms, with it.

The case is taught constantly and understood unevenly. The usual summary — that Enron hid debt and lied about profits — is true and explains almost nothing, because it does not say how a company could do that for years while audited, rated and analysed by people whose job was to notice. The mechanisms are the lesson.

Three mechanisms, not one fraud

Mark-to-market accounting on long-term contracts

Enron obtained permission in 1992 to use mark-to-market accounting for its energy trading business. Under it, the present value of expected future cash flows from a contract is recognised when the contract is signed, not as the cash arrives.

For a liquid traded instrument with an observable price, this is unremarkable. Enron applied it to twenty-year energy supply contracts with no observable market, which meant the reported profit was the output of the company's own model of what energy prices would do for two decades. The estimate was the earnings.

Two consequences followed, both structural. Because profit was recognised at signature, growth required ever larger contracts each year simply to stand still. And because the estimates were never trued up against outcomes in any visible way, a contract that went wrong could be quietly offset by a new one booked more aggressively.

Special purpose entities and the three per cent rule

The second mechanism moved debt and loss-making assets off the balance sheet. Under the accounting rules then in force, a special purpose entity did not have to be consolidated if an independent third party held at least three per cent of its equity and that party bore the risks of ownership.

Enron created hundreds of such entities. The difficulty was that the outside equity was frequently not independent — it was funded, guaranteed or indemnified by Enron itself, often in Enron shares. The entity therefore carried no genuine external risk, and the condition for non-consolidation was satisfied in form and not in substance.

The structures became self-referential. Several entities were capitalised with Enron stock and used to hedge Enron's own investments, so the hedge and the thing hedged fell together. When the share price declined, the entities required more shares to stay solvent, which diluted the stock further.

The auditor relationship

Arthur Andersen earned roughly $25 million in audit fees from Enron in 2000 and approximately $27 million in consulting fees. It also staffed part of Enron's internal audit function, and a steady flow of Andersen staff moved into Enron finance roles.

No single one of those is improper on its own. Together they produced a situation in which the firm was auditing structures its own people had helped design, for a client whose consulting spend exceeded its audit fee, with the reviewing partners' careers tied to retaining the relationship. Andersen was convicted in 2002 of obstruction of justice for destroying documents; the conviction was overturned by the Supreme Court in 2005 on the grounds of a faulty jury instruction, by which time the firm no longer existed.

Why it was not caught earlier

Four defences failed at once, and the pattern is more instructive than any of them separately.

The disclosures were technically present. Related-party transactions were described in the notes, in language that was accurate and close to unreadable. Disclosure is not communication, and a note that is precise and incomprehensible discharges a legal obligation while concealing the same facts it discloses.

The board approved the structures. The audit committee waived the company's own code of conduct to allow the chief financial officer to run the entities he was transacting with. A governance structure that grants waivers to its own rules for the people the rules exist to constrain has stopped functioning as a control.

Analysts reported what the company said. Coverage was overwhelmingly positive until weeks before the collapse, partly because the business was genuinely hard to model and partly because firms with investment banking relationships had little incentive to look harder.

And internal warning was raised and absorbed. Sherron Watkins wrote to the chairman in August 2001 setting out the accounting concerns directly. The letter produced an internal review by the company's own external law firm, conducted with a narrow scope, which found no serious problem.

The business underneath the accounting

It is worth separating two claims that are usually run together: that the accounting was misleading, and that there was no real business. The first is established. The second is not, and getting it wrong weakens an answer.

Enron began as a gas pipeline company and built a genuinely innovative trading operation. Creating a market in gas contracts, and later in electricity, bandwidth and weather derivatives, was a real commercial idea, and parts of that business were sold as going concerns after the bankruptcy. The trading floor was not a fiction.

What the accounting concealed was that this business consumed enormous capital and produced thin, volatile cash returns, while the reported figures showed steady growth in earnings. The gap between reported profit and cash from operations is visible in the published accounts for several years before the collapse, and it is the number a sceptical analyst would have started from. Profit is an opinion; cash is closer to a fact, and a company whose profit rises while operating cash flow does not is asking a question that deserves an answer.

That is the transferable lesson, and it is more useful than the specifics of any structure. The structures were exotic and are now prohibited. The signal — accruals growing faster than cash — is ordinary, available in every set of accounts, and still works.

What changed afterwards

The Sarbanes-Oxley Act 2002 followed within a year, and its main provisions map closely onto the specific failures.

Section 302 requires the chief executive and chief financial officer to certify the accounts personally, which converts an accounting failure into a personal legal exposure. Section 404 requires management to assess and the auditor to attest to internal control over financial reporting, which is the provision that generated most of the compliance cost and most of the complaints about it. Section 201 prohibits auditors from providing most consulting services to audit clients. Section 203 requires rotation of the lead audit partner. The Public Company Accounting Oversight Board was created to inspect audit firms, ending self-regulation of the profession in the United States.

The accounting standards changed too. FIN 46 replaced the three per cent bright line with a control-based consolidation test, asking who bears the majority of the risks and rewards and not who holds a nominal slice of equity. The principle — substance over form — was not new; what was new was writing it into a rule that a structuring team could not satisfy artificially.

The employees, and the part that is easy to skip

Around 20,000 people lost their jobs, and a large number lost their retirement savings as well, because the company's pension arrangements were heavily weighted towards Enron stock. A lock-down period during the administrative change of plan provider prevented employees from selling while the price collapsed, at a time when senior executives were not similarly constrained.

This matters to the analysis and not only to the moral of the story. Concentrating employee retirement savings in the employer's own shares means a single failure destroys income and savings together, which is precisely the diversification failure the savings were meant to avoid. The Pension Protection Act 2006 later restricted the practice in the United States, and the episode is the standard illustration in any discussion of why occupational schemes limit holdings of sponsor stock.

What the case is actually used to teach

Three uses, and they call for different emphasis.

In audit, the case is about independence in appearance as well as in fact. Andersen's judgements may have been defensible individually; the position it occupied made those judgements impossible for an outsider to trust, and the profession's response was to regulate the position and not the judgements.

In financial reporting, it is about substance over form, and about the fragility of any rule stated as a bright line. A three per cent threshold tells a structuring team exactly what to build. This is the strongest available argument for principles-based standards, and the honest counter-argument is that principles are harder to enforce and easier to litigate.

In governance and ethics, it is about the difference between a control and a control that operates. Enron had an audit committee, a code of conduct, an internal audit function and a whistleblowing route. Every one of them existed. The code was waived, the committee was briefed by the people it was meant to supervise, internal audit was partly outsourced to the external auditor, and the whistleblower's letter was investigated by a firm with a relationship to protect.

That last point is the one worth carrying out of the case. The fall of Enron was not a failure to have controls. It was a failure of every control to be independent of the thing it was controlling, which is a design question and not a compliance question, and one that any organisation can answer badly while its compliance register shows green.