Beyond Backdating: Fair-Value Fraud Risk

Companies that try to minimize past experience in gauging the value of their option awards are at risk for fraud, the nation's auditing regulator advises.

Hard as it is to believe, backdating may not be the only risky area in stock-option accounting. Independent auditors should also be able to sniff out fraud when their corporate clients start monkeying with the metrics they use to gauge the fair value of option grants, the Public Company Accounting Oversight Board thinks.

Since there aren’t any markets employers can use to measure what the options they issue are worth, companies use pricing models to measure the options’ fair value. Such models, including Black-Scholes and the lattice method, are packed with assumptions and estimates that are ripe for improper manipulation and intentionally misstated accounting, the PCAOB suggests.

But taking a stab at the market value of an option is no longer optional. In 2004, the Financial Accounting Standards Board issued a revised standard, FAS 123R, that required companies to use the grant-date fair value of an options award in recognizing the grants as a compensation cost in their financials. And FAS 123R began to apply to financial statements of companies with fiscal years ending on or after June 15, 2006.

To help auditors find their way through the minefields of their clients’ fair-value options estimates, the PCAOB issued a question-and-answer guidance on October 17. Following hard on the heels of a July staff alert telling auditors which backdating problems they should be focusing on, the board’s fair-value guidance highlights possible client booby traps in compensation estimates.

Auditors should raise their eyebrows when they see evidence that their clients seem to be softening the impact of historical data on their models, according to PCAOB. There is, for instance, a higher risk of fraud when a company’s assumptions cut the fair value of a grant “below what it would have been had the company based the assumption on unadjusted historical information,” according to the guidance.

Similarly, there’s reason for an auditor to be suspicious if a client company keeps data from a certain historical period out of the calculations of the options-pricing estimate, “especially when the effect of that exclusion is to lower expected term or expected volatility,” the board warns. Since an option on a share with higher volatility is generally worth more than one with lower volatility, a company could presumably slash its recognized options expense by keeping volatile historical data out of its pricing calculations.

Unusual options-exercise estimates should also get an auditor’s attention, the audit board says. An auditor should do more probing, for instance, if a client that has averaged a term of seven years for its option grants offers up an estimate of five years for a new grant.

Employers that pick the lowest expected term or expected volatility estimate as the most likely outcome in a range of possible estimates might also be trying to finagle lower reported compensation costs, PCAOB suggests. Further, those that select expected term and expected volatility estimates lower than the historical averages should get further scrutiny.

Other clients worth probing further are those that haven’t consistently adjusted data on option exercises or share-price volatility when their adjustments should have been consistent, according to the board.

“I appreciate that many [audit] firms currently are auditing the fair value of share options and have developed their auditing approaches using the general principles in the existing auditing literature,” Tom Ray, the PCAOB’s chief auditor and director of professional standards, said in a release accompanying the guidance. “The staff’s questions and answers were developed to help auditors apply the existing auditing standards to this area appropriately and consistently, and I encourage auditors to review this guidance as they plan for calendar year-end audits.”

Discuss

Your email address will not be published. Required fields are marked *