University academics are also working on new models. For example, John Finnerty, a professor of finance at Fordham University in New York, has sent a paper to the International Accounting Standards Board (IASB) – a UK-based accounting standards setter – which attempts to model the impact of early exercise and forfeiture before and after vesting. He claims his model produces values that are half of those produced by Black-Scholes. Others are looking at models that try to predict when executives with inside information will exercise their options.
Market practitioners have identified several problems with Black-Scholes, the most widely used pricing model for assigning a “fair value” to employee stock options. The model was created to value freely tradable, European-style options. But there is no market for employee stock options, and working out if and when the option will be exercised is problematic. Also, employee stock options usually come with restrictions on when they can be exercised – for example, directors may not able to exercise directly before the company announces its results. For this and other reasons, experts say Black-Scholes consistently overestimates the true value of employee stock options.
FASB first proposed that companies be required to recognise stock-based compensation in the income statement using a fair-value based method in the mid 1990s. However, in the face of strong opposition it subsequently backed down and allowed the continued use of the “intrinsic” approach to valuation, which is regarded as something of a whitewash because it generally gives a compensation cost of zero. Last year, shareholder rage over corporate accounting scandals spurred many US companies, including Coca Cola and Ford Motor Company, to voluntarily account for their employee stock options as an expense, and FASB is now looking again at the possibility of making this a standard accounting practice.
To date, however, no-one has come up with a model that can exactly price stock-based compensation. Nera’s model only seeks to give a more accurate value to the time-to-expiry variable. It fails to address another highly contentious variable, the volatility estimate. Experts told RiskNews this is almost impossible to predict with any accuracy. Moreover, there is no standard method in place for working out its value, so one company’s estimate can be vastly different from its peers.
Nera concedes that the volatility estimate is important, but insists its model provides a good first approximation. It plans to give its model away for free and make money from providing further consulting services. It also aims to create a new market standard. “We’re hoping to get the endorsement of the accounting community in the form of the big four and the IASB [International Accounting Standards Board] and the FASB," said Nera's Abrams.
The week on Risk.net, January 6–12Receive this by email