Dewey & LeBoeuf: Revenue Fraud and Law Firms

Revenue frauds have long been among the most popular forms of financial reporting manipulations. The SEC, for example, has traditionally identified revenue recognition infractions as being either the top or the second most common variety of the management and accounting frauds it has had to pursue.

[i] The ACFE, a professional organization of fraud examiners, has documented similar findings in its bi-annual Report to the Nations.[ii] Although other techniques are also able to distort reported earnings, e.g., concealing routine expenses (as was done by WorldCom, for one prominent example) and investment losses (as done by Olympus) by charging them to unrelated accruals, or by pushing expenses or losses off the financial statements (e.g., through employment of “special purpose entities” [now called “variable interest entities”], as done by Enron), these other methodologies often lack the market-moving impact of distorting revenues.

The propensity to make use of fraudulent revenue schemes is explained by the fact that many stakeholders in law firms and corporate entities are particularly sensitive to the amounts and period-by-period changes in a reporting entity’s revenues. For publicly held companies, stock prices often react sharply and quickly to changes in actual or projected revenues. In the case of private enterprises, banks and other creditors may reevaluate risk based on these same factors, and in some instances – such as at Dewey & LeBoeuf, which had raised over $150 million in a private debt offering to several major insurers – loan and bond covenants may require that certain levels of revenue be maintained to avoid triggering a default or other covenant violations.

Dewey & LeBoeuf, the product of a law firm mega-merger in 2007, recently boasted about 1,000 lawyers and 26 offices around the world. It ranked #28 in 2012 on the closely followed Am Law 100 list, which is compiled based on law firm gross revenue, revenue per lawyer, profits per partner and other metrics. However, its performance, particularly its growth in revenues and profits, had plateaued by 2011, which might have been part of what motivated the alleged accounting improprieties.

The risk of revenue-related fraud has long been recognized, and is so great that the Auditing Standards Board, formerly charged with establishing guidance for auditors of both publicly held and private enterprises, in adopting a key standard, SAS No. 99,[iii] cited this as one of only two factors (management override of controls being the other) that must be presumed to represent a fraud risk in each audit engagement, regardless of how well other categories of risk have been controlled.[iv]

The actual workings of the Dewey & LeBoeuf alleged fraud have yet to be proven, or even fully explicated; the criminal trial involving firm leadership charged in this matter has just begun in NYC as of this writing. However, prosecutors have stated that the scheme formulated by the indicted executives led to falsely reducing expenses, falsely increasing revenue, and falsifying invoices, among other improper entries made in the books and records of the firm. Presumably the seven cooperating witnesses, who may include the lower-level employees who actually made the false entries, will provide mechanical details about this multi-year fraud, which left one of the nation’s most preeminent law firms bankrupt.

Given the relatively simple accounting that is required for a law firm or other professional firm, it is suspected that overstatement of revenues and thus also of receivables formed a central aspect of this alleged fraud. The possible means of accomplishing such manipulations are therefore reviewed in the following paragraphs.[v]

Revenue frauds usually involve either accelerated recognition of real, but not yet realized, revenues, or the creation and recognition of entirely fictitious revenues. In the former instance, the “borrowing forward” of future revenue inevitably leaves a shortfall in the subsequent reporting periods, and this is often dealt with by committing yet another timing fraud at the end of those periods, often supplemented by further “borrowing forward” to make those succeeding fraudulently distorted periods’ results demonstrate the desired increases from that of their antecedents. Obviously, this cannot continue indefinitely, inasmuch as the ever-growing amount of prematurely recognized revenue should become rather apparent. In particular, the aging of period-end receivables will raise concerns from the independent auditors, and may even become obvious, indirectly, to non-accountant users of the financial statements.

For commercial enterprises producing tangible goods, a commonly observed scheme is to engage in so-called “bill and hold” transactions, whereby actual customers are induced, perhaps by being offered reduced prices, to place otherwise not-yet-necessary orders, allowing the producer to bill and record sales but holding the inventory until the later date when the customer calls for it to be delivered.[vi] The SEC has pursued a number of these frauds over recent decades, including cases involving such once-household names as Nortel and Sunbeam.

Another variation occurs when revenue is recorded while significant uncertainties still remain. Under current GAAP, this requirement should preclude revenue recognition until the uncertainties have been resolved. This deception may not always be apparent, particularly to auditors who only sample from the population of the period’s revenue transactions and may not fully appreciate the conditions stipulated in each sales transaction. However, cash collections from such sales will, upon closer inspection, seemingly lag behind the normal pattern, because the customer will only pay when all agreed-to conditions have been met. Although somewhat subtle, this should be detectable during a properly planned and conducted audit, if such transactions are material in amount.

When the sales involve multi-element arrangements, which commonly occur in such businesses as software development and also in some construction contracts, revenue may not to be recognized until customer acceptance has occurred, which often requires that later elements that are needed in order to give value to the earlier-delivered elements also be delivered. In practice, detecting premature recognition may be somewhat difficult under these circumstances, unless detailed attention is directed to the individual contracts and any related correspondence. This will only become a more common issue under the recently promulgated revenue recognition standard,[vii] since that standard essentially imposes a percentage-of-completeness model on transactions that hitherto had not been subject to such accounting. One prominent case involving multi-element arrangements outside the software development industry was the late 1990s–early 2000s matter pertaining to Xerox, which misapplied GAAP to accelerate revenue recognition from equipment leases that also incorporated such elements as supplies and maintenance agreements.

According to a study using 2007-2008 SEC enforcement data, premature revenue recognition frauds accounted for about 23% of all revenue frauds, and those involving recognition before all conditions of sales had been met accounted for about another 17% of those frauds.[viii]

More specifically, premature revenue recognition frauds may involve shipping goods before the sale (not easily accomplished in a service business, however) or making a partial shipment but booking a full one; recording revenues while material uncertainties remain, including a right to return that cannot be fairly estimated until the privilege expires; offering extended return or other special terms in order to “channel stuff” (common to such frauds as Sunbeam, Donnkenny, and Bristol-Myers Squibb, but again not readily achieved for law firms and other service providers); and simply holding the books open after year-end (the well-known “December 45th” year end strategy), commonly observed in service businesses as well as for sellers of goods.

Somewhat more ambitious than mere premature revenue recognition are those that involve the recordation of fictitious revenue – that is, treating as real those revenues that will never become legitimate, and will never, therefore, be collected. Because a constantly growing balance of increasingly old receivables is likely to become painfully obvious to auditors, those perpetrating fictitious revenue fraud must find a way to create the appearance of movement – cash collections, perhaps sprinkled with a typical rate of bad debt write-offs – or must engage in more elaborate devices to transfer receivable balances to other asset accounts. According to the aforementioned study of recent SEC enforcement data, these accounted for about 23% of all revenue frauds.

In contrast to premature recognition of revenues, where, subject to normal risks of non-collectibility, the receivables will eventually be converted to cash, concealing fictitious revenue will later necessitate that fabricated receivables be somehow eliminated from the balance sheet. In some cases, these will be written off as uncollectible receivables, which of course will not truthfully convey the essence of the fraud, nor does it restate prior periods’ results of operations to remove fraudulent revenues. Stale accounts receivable will inevitably draw audit attention and demands for bad debt recognition, reducing future earnings, even if the auditors do not detect the actual fraud. Astute stakeholders and employees might also notice the rising ratio of receivables to sales, a “red flag” of financial reporting fraud, which might also lead to the scheme’s unraveling.

Achieving a successful fictitious revenue fraud will therefore necessarily involve “refreshing” the old receivables by transferring the balances to other customers, who may be equally fictitious. In this way the age of the bogus customer obligations can be maintained within a historically normal range. Transferring bad receivables balances to substitute customers will require non-cash entries to the books and records, usually being made in the so-called general journal, where they should, but sometimes do not, draw auditors’ attention. Any entries in the general journal, apart from routine period-end adjustments such as the recording of depreciation, are suspect, and indeed most financial reporting frauds leave a trail in the general journal, albeit not always perceived accurately on a timely basis.

An alternative strategy is to engage in “lapping” the receivables, or crediting collections on real receivables against the bogus ones. This leads to a never-ending pattern of applying later collections to cover prior misapplications of collections to the fraudulent receivables. Here, too, the need to continue this practice over an extended period increases the likelihood of eventual discovery, again with probable disastrous consequences.

Common variations on the fictitious revenue fraud include reporting non-revenue exchanges, such as exchanges of assets or gains on asset dispositions or debt restructurings, as being revenue from customers; massive fraud involving wholly bogus revenues; misrepresenting other credits, such as purchase discounts, as being revenues from customers; and, particular to those industries using percentage of completion revenue recognition, deliberately mis-estimating project status at period ends.

In most instances, bogus revenue and receivables schemes collapse of their own weight, but this can take years, and the affected entities by then may be mortally wounded.[ix] Again, although the specifics of the apparent Dewey & LeBoeuf revenue fraud are not yet known, that firm’s collapse after several years of engaging in the alleged fraud would be a typical dénouement for such conspiracies. In the long run, just as “no tree grows to the sky,” bogus revenue schemes cannot continue forever, not merely because the odds of discovery multiply with extended time periods, but also because the scale of the fraud must increase in order to create the illusion of persisting growth in the entity’s revenues from period to period, which is generally the primary objective of the perpetrators.

Historically, revenue frauds have mainly afflicted manufacturers and merchandisers, in part because the opportunities associated with the physical movement of goods have been available and tempting. Perhaps the Dewey & LeBoeuf case, once fully revealed in terms of mechanical details, will give auditors a renewed enthusiasm for auditing service firms’ revenues, as well.

End Notes

[i] In most years, revenue recognition frauds were the most common, typically accounting for about 40% of detected frauds, with manipulations of “reserves” (i.e. expense accruals) being second most popular and occasionally the most common device employed. See, e.g., Deloitte, Ten Things About Financial Statement Fraud – Third Edition: A Review of SEC Enforcement Releases, 2000-2008.

[ii] According to Report to the Nations on Occupational Fraud and Abuse (Association of Certified Fraud Examiners, 2014), although financial statement frauds are the least common of the range of abuses (which prominently include asset misappropriations and various forms of corruption) addressed, the median losses resulting from such frauds are by far the largest, roughly ten times that of other popular schemes.

[iii] SAS No. 99, Consideration of Fraud in a Financial Statement Audit [codified as AU §316 and then subsequently, in the clarified standards, as AU-C §240], which became effective for audits of calendar year 2003 and fiscal 2004 financial statements, was the last of a sequence of four auditing standards addressing the obligations for planning and conducting audits so as to control the risk of material, undetected financial statement fraud at a low level.

[iv] As a result of the Sarbanes-Oxley Act of 2002, a new, nominally private-sector organization, the Public Company Accounting Oversight Board (PCAOB), was created to, inter alia, develop auditing standards applicable to public company (“issuer”) audits. As an interim step, most of the then-extant universally applicable standards, including SAS No. 99, were adopted by PCAOB. As of mid-2015, this standard has not been superseded by a PCAOB-created rule.

[v] The allegations concerning Dewey & LeBoeuf go beyond revenue recognition, and appear to extend to mischaracterizing non-revenue inflows, such as partner capital contributions, as being reductions in expenses. This article does not address those matters, which have not yet been expansively described in the press or by prosecutors.

[vi] It should be noted that not all “bill and hold” transactions connote fraud. However, under accounting standards the risk of ownership must transfer to the customer, and the “hold” must be at the customer’s request, not the selling entity’s.

[vii] Financial Accounting Standards Board, Accounting Standards Update to Accounting Standards Codification Section 606, Revenue from Contracts with Customers, issued in 2014 but not effective until at least 2017.

[viii] Deloitte, op. cit.

[ix] According to a study conducted by Deloitte, bankrupt companies are three times more likely than non-bankrupt ones to have been cited by the SEC for a financial reporting fraud. [Deloitte, Ten Things About Bankruptcy and Fraud, November 2008] Thirty-five percent of the companies charged with fraud later filed for bankruptcy protection, more than twice the rate for companies not so accused. Thus, financial reporting fraud is a leading indicator of eventual corporate failure.