Accounting's Three Degrees of Separation

"What do I care about the law. Ain't I got the power?" steel magnate Frederick Vanderbilt once said.

As Americans watch an array of equally arrogant, next-generation robber barons being led away in handcuffs, scorned investors are speculating on who will be arrested next. Meanwhile, it's D-Day for big firms to recertify their number-crunching, and new math problems are surfacing. But the road from aggressive accounting to criminal fraud can be a tough one to navigate.

If the Securities and Exchange Commission requires a team of investigators and months to determine the severity of a violation, how can individual investors be expected to determine the difference between, say, the bad-but-not-criminal issues unearthed so far at AOL Time Warner and the almost-laughable criminal shenanigans of Adelphia executives? Good question.

Those distinctions are tough for retail investors to suss out before the whip comes down. What will help stem unnecessary panic-selling of many stocks, however, is a better understanding of how regulators and the Justice Department tackle these issues.

Determining the Degrees

Noncompliance with generally accepted accounting principles or even violations of SEC mandates are not necessarily criminal acts. But if these breaches are egregious enough, federal or state authorities will investigate and sometimes charge individuals with specific crimes.

As with any other criminal behavior, eventually it's the courts that make that call in determining what types of accounting shenanigans are actually criminal behavior, and which were just bad practices.

The SEC employs a three-tiered hierarchy when assessing violations. The first and most basic is simply a breach of securities law. The next category is for violations that were actually fraud; in other words, that someone deliberately attempted to break the law. The third and most serious category of violation are for those in which there was a deliberate intent to break the law, and the public either lost or were put at risk of losing a substantial amount of money.

The now-infamous string of companies currently under investigation would all almost certainly fall into the third camp. After all, any intentionally squirrelly accounting would put the public at risk of substantial losses.

Violations that fall into the second or third tier are considered to be potentially criminal behavior, although the SEC itself does not have any authority in criminal matters. Instead, it refers cases to the Justice Department or a state attorney general's office. The SEC then pursues civil means of punishment.

The SEC's hierarchy of violations includes a healthy amount of discretion, and that latitude often surfaces in criminal investigations as well. For instance, if an employee -- for fear of losing his job if he did not comply -- fudged some figures to make a quarterly report look better, he has clearly reached the third tier of violations. But if that same employee had done so for personal profit, that intent could warrant a greater civil punishment as well as less sympathy in the criminal system.

"There's no checklist as to when certain conditions are met," says Brian Sierra, a spokesman for the Justice Department. "There are criminal laws on the books regarding fraud, but nothing that says definitively when fraud has been committed as opposed to bad accounting."

That Fine, Hazy Line

Accountants are fond of saying that the methods employed by their profession are art, not science. Snicker if you will, but the assessment is accurate.

"Companies can choose among several acceptable accounting treatments," says Lawrence Revsine, an accounting professor at Northwestern University's Kellogg School of Management. "There's judgment in financial reporting that's just inescapable."

Indeed, the financial reports announce just that. "A typical footnote found in every financial report says something to the effect of 'The preparation of financial statements in accordance with generally accepted accounting principles requires management to make some assumptions,'" says Kevin Pianko, an audit partner at New York's Eisner LLP.

Essentially, the line between aggressive accounting and accounting fraud is crossed when that judgment, those assumptions, are indefensible -- completely contrary to established accounting theory.

Take, for instance, the fairly straightforward accounting issue of depreciation. (A brief Accounting 101: Depreciation is the allocation of an asset's cost over the estimated useful life of the asset. Each year that the asset is in use, a certain portion of its worth is treated as an expense.)

There are two basic methods for allocating the costs of an asset: Straight-line depreciation and accelerated depreciation. Straight-line depreciation simply allocates an equal amount of an asset's cost every year, making the depreciation amount the same every year of the asset's life. Accelerated depreciation -- as the term implies -- uses shorter life estimates for the asset, and takes a greater amount of depreciation in early years and lower amounts in later years.

A company purchasing a lot of technology should opt for the accelerated depreciation method, since technology generally loses value more rapidly in its early years. However, that company could opt instead for the straight-line method, which would mean a lower expense and lead to a pumped-up bottom line. Aggressive? Yes. Illegal? Probably not.

"That area is so gray that most of the time companies are not thought of as crossing any line," Revsine says. "They're choosing between equally permissive vehicles. One version is certainly more appropriate, but that's a hard argument to make in a court of law."

And while it seems like a rash of modern-day robber barons are flagrantly fleecing investors, true accounting fraud is historically rare, says Harvard Business School accounting professor David Hawkins. "When it has happened, it's led to significant changes in the accounting profession," he says. "But we really don't see this very often."

It's anybody's guess how much more of it we'll see.