Financial Shenanigans: How to Detect Accounting Gimmicks & Fraud in Financial Reports, 3rd Edition (13 page)

Enter AIG and its “perfect world” products. AIG created a special $15 million “retroactive” insurance policy that would “cover” Brightpoint’s unreported losses. Here’s how the policy worked: Brightpoint agreed to pay “insurance premiums” to AIG over the next three years, and AIG agreed to pay out an “insurance recovery” of $15 million to cover any losses under the policy. This sounds like your normal insurance policy, except for one big problem: there was no transfer of risk, since the policy covered losses that had already happened. You can’t insure your house after it burns down!

Brightpoint proceeded to record the $15 million “insurance recovery” as income in the December quarter (which blotted out its unreported losses). AIG recorded what amounted to bogus revenue on the insurance premiums over the next three years. Economic sense dictates that this transaction was not an insurance contract because no real risk had been transferred. Indeed, the transaction was nothing more than a financing arrangement: Brightpoint deposited cash at AIG, which AIG eventually refunded as purported “insurance claim payments.”

Accounting Capsule: Legitimate Insurance Contracts Require a Transfer of Risk

Just because two parties call an agreement an insurance contract does not mean that they can book it as such in their financial statements. In order to be considered an insurance policy for accounting purposes, an arrangement must involve a transfer of risk from the insured to the insurer. Without this transfer of risk, GAAP treats the arrangement as a financing transaction, with premium payments being treated like bank deposits and recoveries being treated like the return of principal.

The Regulators Considered This Scheme to Be a Scam
. Brightpoint got into trouble with the Securities and Exchange Commission (SEC) for inappropriately masking its problems. AIG found itself in the SEC’s crosshairs as well for knowingly structuring the insurance policy in such a way that it allowed Brightpoint to misrepresent its actual losses as “insured losses.” In November 2004, AIG agreed to pay $126 million to settle litigation with the Department of Justice and the SEC on charges that it had sold products that helped companies inflate earnings via the use of finite insurance.

Charity Begins at Home—After Helping Its Clients, AIG Finds Time to Help Itself

After several years of helping its clients smooth out rough patches in their earnings, AIG decided to use its expertise to help rid itself of nasty questions raised by its own Wall Street analysts. AIG’s September 2000 earnings release did not go well. Many analysts were unsettled by an unexpected decline in AIG’s insurance liability reserve. They were asking questions about this decline, and some worried that AIG had released reserves to make its numbers for the quarter.

To help solve its “Wall Street problem” and boost the troubling anemic reserve balance, AIG sought the assistance of Gen Re to structure a sham reinsurance arrangement. Since AIG was one of Gen Re’s largest clients, Gen Re acquiesced and actively participated in the fraud, for which it would later pay dearly. Here’s how it went down. AIG received $500 million in purported insurance premiums from Gen Re, which it used to beef up its loss reserves. At the same time, AIG paid $500 million to Gen Re to reinsure a risk. As in the Brightpoint transaction, there was no real economic substance or transfer of risk behind this exchange—it was really just a round trip of cash. However, AIG did not account for it that way. It used the transaction to inappropriately prop up its loss reserves.

Beware: Bogus Reserves Will Often Lead to Bogus Revenue or Income.
In its sham arrangement with Gen Re, AIG could just as easily have decided to record bogus revenue instead of propping up its reserves. However, AIG’s goal was to increase its reserve account, not to inflate revenue—at least not yet. If a reserve is bogus and no future payment exists, a company can easily make a bookkeeping entry that releases the reserve to report phony income.

So AIG actually improperly benefited twice from this shenanigan. First, when it created a bogus liability reserve, Wall Street analysts were mollified by the high reserve balance. Then, with the “reloaded” reserve, AIG gave itself an opportunity to “flip the switch” and release these reserves into income. The plan worked great for a while—or until the regulators started sniffing around. AIG Pays Big Time for Its Shenanigans in Its Settlement with the SEC. In May 2005, AIG announced a staggering $2.7 billion restatement, correcting misstatements for the previous five years. The following February, AIG paid another $1.6 billion to settle litigation as part of a global resolution of federal and state actions.

Peregrine Dupes Investors with Software Sales That Lack Economic Substance

Not surprisingly, creating bogus revenue from transactions that lack any economic substance extends far beyond insurance companies. Plenty of technology companies seemed to get the memo on how easy it is to employ this shenanigan. Take, for instance, San Diego–based Peregrine Systems, which got busted for a massive fraud scheme that involved recognition of bogus revenue.

The SEC charged that Peregrine improperly recorded millions of dollars of revenue from nonbinding sales of software licenses to resellers. The company apparently entered into secret side agreements that waived the resellers’ obligation to pay Peregrine, which means that revenue should not have been recorded. Employees at Peregrine had a great name for the scheme: “parking” the transaction. Sales that were near the finish line were often parked in order to help Peregrine achieve its revenue forecasts. Peregrine engaged in other deceptive practices as well to create bogus revenue, including entering into reciprocal transactions in which the company essentially
paid for its customers’ purchases
of its software. In 2003, Peregrine restated its financial results for those quarters, reducing previously reported revenue of $1.34 billion by $509 million, of which at least $259 million was reversed because the underlying transactions lacked substance.

And with Bogus Revenue Come Those Troublesome Receivables
. Peregrine obviously did not receive cash from customers on these nonbinding bogus revenue contracts, resulting in bogus receivables festering on the Balance Sheet. As we have learned, a rapid increase in accounts receivable is often an indication of deteriorating financial health. Peregrine knew that analysts would naturally begin questioning the “quality of earnings” if the bulging receivables balance remained stubbornly high. To prevent these questions, Peregrine played several tricks that made it seem that the receivables had actually been collected. These shenanigans inappropriately lowered the receivables balances, and in doing so, improperly inflated cash flow from operations (CFFO). We will break down the mechanics of this chicanery and discuss Peregrine’s Cash Flow Shenanigans further in Chapter 10.

Symbol, Of Course, Wants In on the Action

Symbol Technologies found a creative way to recognize revenue that lacked economic substance. From late 1999 through early 2001, Symbol conspired with a South American distributor to fake more than $16 million in revenue. It instructed the distributor to submit purchase orders for random products at the end of each quarter, even though the distributor had absolutely no use for those products. Symbol never shipped the products to either the distributor or any of its customers. Instead, to fool the auditors into believing that a sale had actually occurred, Symbol sent the products to its own warehouse in New York; however, it still retained all “risks of loss and benefits of ownership.” The distributor, naturally, did not have to pay for the warehoused product and could “return” or “exchange” the goods at no cost when it placed legitimate new orders for any product that it actually needed. Without a doubt, the only purpose of this charade was to give the appearance of a legitimate sale so that Symbol could record revenue.

Symbol’s Three-Ring Circus.
Symbol would do virtually anything, including losing money, just to recognize a sale. In a bizarre story, Symbol concocted a three-party circular scam to create fake revenue out of thin air. Typically, Symbol sold its product to a middleman (a distributor), who then sold the product to the actual customer (the reseller). Symbol found a way to use this structure to create bogus revenue: it improperly enticed (really, bribed) the reseller to purchase more of Symbol’s product from the distributor. The distributor would, in turn, purchase more products from Symbol. What a way to create artificial demand!

How did Symbol entice its resellers to go along with this scheme? Well, Symbol allegedly gave the resellers the cash they needed to make these purchases. Symbol also paid a substantial markup so that the resellers could turn a profit on these transactions. On top of that, Symbol paid resellers a bonus equal to 1 percent of the purchase price (referred to by the Symbol schemers as “candy”). The net result of this transaction was that Symbol purchased its own product back at a much higher price than that at which it had initially sold it. Symbol did not mind the loss, however, as the scheme was all in the name of revenue growth.

During 2000, for example, Symbol improperly recognized approximately $10 million of revenue on transactions for which it had first paid out $15 million to resellers. In this bizarre scheme, Symbol actually lost millions of dollars on this circular sham transaction. The scheme produced the desired revenue, but Symbol suffered losses by effectively buying back its own products at a higher price and paying what amounted to a bribe. This is clearly one of the more ridiculous financial shenanigans you will ever see!

2. Recording Revenue from Transactions That Lack a Reasonable Arm’s-Length Process

While recognizing revenue on transactions that lack economic substance can never be considered legitimate, transactions that lack a reasonable arm’s-length process are sometimes appropriate. But prudent investors should bet against it. That is, most related-party transactions that lack an arm’s-length exchange produce inflated, and often phony, revenue.

Transactions Involving Sales to an Affiliated Party

If a seller and a customer are also affiliated in some other way, the seller’s quality of revenue on sales may be suspect. For example, a sale to a vendor, a relative, a corporate director, a majority owner, or a business partner raises doubt as to whether the terms of the transaction were negotiated at arm’s length. Was a discount given to the relative? Was the seller expected to make future purchases from the vendor at a discount? Were there any side agreements requiring the seller to provide a quid pro quo? A sale to an affiliated party or a strategic partner may be an entirely appropriate transaction. However, investors should always spend time scrutinizing these arrangements, as it is important to understand whether the revenue recognized is truly in line with the economic reality of the transaction.

Be Wary of Related-Party Customers and Joint Venture Partners.
A representative case in point is the alleged fraud at SyntaxBrillian Corp., the Arizona-based maker of high-definition televisions. In 2007, Syntax-Brillian was flying high. Extraordinary demand in China sent sales of TVs soaring, and a new marketing relationship with ESPN and ABC Sports generated a buzz about its Olevia HDTVs. The company more than tripled its revenue in fiscal 2007, with sales approaching $700 million, up from less than $200 million the prior year. Yet one year later, Syntax-Brillian was bankrupt and under investigation for fraud.

Syntax-Brillian’s demise was not a surprise to investors who understood the extent to which the company’s reported results benefited from transactions with related parties. For example, the company’s staggering revenue growth came from a tenfold increase in sales to a related party that accounted for nearly half of its total revenue—an Asian distributor named South China House of Technology (SCHOT). Syntax-Brillian’s relationship with SCHOT was much more incestuous than a typical customer-supplier arrangement, according to a study by RiskMetrics Group. The two companies seemed to be involved in a tangled web of joint ventures (which also, oddly, included Syntax-Brillian’s primary supplier). Syntax-Brillian was close enough with SCHOT that it granted it 120-day payment terms and routinely extended those terms even further.

Syntax-Brillian described SCHOT as a distributor that would purchase its TVs and then resell them to retail outlets and end users in China. Many investors failed to question the company’s significant uptick in sales to SCHOT, as they believed that demand in China was high, with people upgrading their TV sets heading into the 2008 Summer Olympics in Beijing. Investors were also cheered by reports that the Beijing Olympic Village itself was planning to fit its facilities with Olevia TVs.

Then all of a sudden, in February 2008, Syntax-Brillian cryptically announced that the Olympic facilities would no longer be installing the TVs that the company had “sold” to South China House of Technology. Even though Syntax-Brillian had already recorded revenue from the sale of these TVs, it agreed to “repurchase” more than 25,000 TVs for nearly $100 million. The company did not need to come up with the cash because the receivable from SCHOT was, of course, still outstanding. With this significant right of return and no receipt of cash, Syntax-Brillian should never have recognized this revenue in the first place!