After a long debate and a delay or two, most companies will finally start to recognize stock option costs in their earnings report. Many got ahead of the game with voluntary recognition, but most of them occupy industries that tend to grant fewer options. For example Exxon Mobil
), General Motors
) and Coca-Cola
) all voluntarily expensed options. These companies are less vulnerable to an adverse market reaction. Technology companies, on the other hand, have been resisting the rule on all fronts, including a vigorous lobby effort.
That first thing to understand about option expensing is that it matters a little or a lot depending on the business model. To illustrate this, we compared option "run-rates" in transportation to those in the computer hardware sector. The run-rate is the number of options granted in a year as a percentage of the total shares outstanding. You can see the big difference below. The top quartile in transportation grants fewer options than the bottom quartile computer hardware. The median computer hardware company grants three times as many options as its transportation counterpart (i.e., about 3% per year compared to about 1% per year.)
For the majority of these transportation companies, option expensing is a non-event. Among such companies, you would do better to analyze depreciation policies rather than option costs. As a general rule, companies that prioritize fixed asset investments are less likely to make large option grants. Companies that prioritize talent and human capital are more likely to make big option grants. Further, as long as talent wants a "piece of the action" and companies increasingly depend on intellectual assets, options and stock are going to remain popular.
The second thing you want to understand about the option expense is that it is just an estimate about the future. Under the new rules, this estimate is deducted from profits to get a reduced reported earnings number (and EPS). As an expense, many people have compared it to depreciation because it is a non-cash charge. That means the company did not actually use cash in the period. Here is a really simplified income statement:
Because option expense is a non-cash charge, a very popular reporting tactic will be to report pro forma earnings and exclude the option expense. But notice an important difference between depreciation and option expenses. Depreciation is a certain number based on cash invested in a prior period; e.g.., if a company invested $1 million last year, perhaps it depreciates, or allocates, one-tenth of that amount over each of the subsequent ten years. But an option expense is an estimate of a future cash outlay, or more precisely, a future cash transfer from shareholder to employees. This is an important difference: most non-cash charges are known allocations based on prior investments, but option expenses are estimates about the future.
As estimates, they have been remarkably bad. I write this despite great admiration for option pricing models like the Black-Scholes. Three years ago I co-conducted one of the largest empirical tests of these option pricing models.
We retroactively asked a simple question, "If we had used these option pricing models to make option expense estimates over the last thirty years, how accurate would they have turned out, given the actual stock price behaviors?" Using a large sample of over 1,500 companies, we compared the predicted option value to the actual value.
The outcome was the opposite of what I had hoped – my plan was actually to uncover the most accurate flavor of the various option pricing models. We found both models (the Black-Scholes and several variants of the binomial) tended to either grossly under-price or grossly over-price the actual, subsequent cost of the option. In other words, if you had used an option cost estimate, you could be pretty sure it was wrong, but you had no idea in which direction! Saving you the details, the best explanation seemed to be that secular market trends simply overwhelmed the accuracy of the model. For example, under a secular bear market, many option grants are simply worthless-in which case option expense estimates are way off!
Several of the large investment banks are now fine-tuning their option expense models in anticipation of the new rules – they are trying to estimate the impact of option expensing on the price-to-earning (P/E) multiple. Given the inherent inaccuracy of the models, I think the time is far better spent looking at the actual features of the new plans. For example, in reaction to the rules, many companies are installing performance-based restricted stock. What performance metrics are they using and what hurdles are they setting for the executive team? In looking at plan features, we have seen everything from ridiculous giveaways to sublime alignment with shareholders.
So, what do we do at Advisor? We look at equity overhang, also called potential dilution. To understand overhang, consider the life cycle of a stock option. First, the company receives authorization from shareholders-think of this as a pool of options that are "available" for future grant. Second, the company grants options from this (authorized) pool and into the hands of employees.
At this point, the options are outstanding but they have not yet cost anybody anything. Third, the employee exercises the option for a gain. In doing so, existing shareholders are diluted (as reflected in either an increased common share base or corporate cash spent to repurchase shares). Equity overhang is a measure taken at a particular point in time, typically the end of the fiscal year. It is the total of: options "available for grant" (i.e., "in the pool" and ready to be granted) plus the options outstanding (i.e., already held by employees). Because it is only a matter of time before this collective group of options (available + outstanding) converts into future dilution, we consider this a respectable gauge of potential dilution.
Overhang is a test in our red flag checklist. If we see a high overhang, it is a red flag and is considered in totality with the other factors and other red flags. It's hard to imagine an investment idea turning on this one flag. But we are not surprised to see that a high overhang is often accompanied by other bad governance practices.
The best thing about the new rule is that it gives us another peek into the quality of corporate governance. Boards and corporate officers tend to reveal their true colors in reacting to an accounting rule change. An unfortunately long list of companies are reacting by accelerating the vesting of outstanding options because the transition rules won't count them if they are accelerated. This is the equivalent of tucking the expense under the rug and hoping shareholders won't notice. It's even worse because the acceleration "de-incentivizes" the incentive ("you were supposed to wait four years for your option, but go ahead and have it today"). The decision to go along with this maneuver signifies a weak board or, at best, a board that suffers from accounting myopia. As another example, some companies are going to dismantle their employee stock purchase plans (ESPP) because, while such plans remain sensible and tax-effective, they will get recognized. In the vast majority of cases, this is a bad decision. Consider instead a wise decision by First Data Corp
) to retain their ESPP:
"the Company recognizes that the new accounting rules issued by the Financial Accounting Standards Board will create an expense for shares issued under the ESPP beginning in 2005...However, the Company believes that the ESPP is a valuable employee benefit that assists the Company in its efforts to attract, retain and motivate valuable employees...and these benefits are well worth the additional expense recognized for accounting purposes" (FDC Proxy, 3/28/2005)
First, the option expensing issue ranges from non-eventful in many sectors to critical in, say, software. Software companies grant the most options, by a wide margin, owing to people-intensive business models. Second, option expensing will create more metrics to consider along with GAAP reported earnings. It is fine to consider pro forma earnings, but be careful about cash flow measures. Popular metrics like earnings before interest, taxes depreciation and amortization (EBIDTA) and cash flow from operations (CFO) are incomplete and often artificially high because they exclude capital spending (investments). This is similarly true of pro forma earnings that excludes option expense; by excluding the non-cash charge, it ignores dilution altogether. If you are going to use it, you want to combine it with a look at dilution (e.g., equity overhang, cash used to repurchase shares). Third, use the rule to gain in peak into the quality of governance. Strong boards already see the rule for its opportunity: by putting all equity incentives on a the same level playing field, it grants companies the freedom to design rewards the encourage value creation. A weak board will consider options an expense, to be minimized or hidden if possible.