The other day a client asked which current accounting requirement is the worst from a U.S. GAAP standpoint. There are a few poor standards out there, but to me the answer is easy: FAS123R, now known as ASC 718, accounting for stock compensation. It’s been around eight years, and it’s not getting any better with age!

The idea of FAS123R, which replaced stock compensation rules under APB 25, is that all stock grants have a value to the employee, and that should be accounted for as compensation. Consequently, on each stock option grant, there’s a charge to expenses over the vesting period of the grant. Under APB 25, a charge arose only when the fair value of the grant was greater than the grant price, so most grants did not give rise to a charge. Under FAS123R, the expense varies depending on a number of factors, the two most important of which are the fair value of the stock at the time of grant and the volatility of the stock.

Here’s why I think the FAS123R is a bad accounting standard:

Inconsistent and arbitrary outcomes. Take two similar companies: Company A’s stock price is $10 and Company B’s is $5. Both grant an employee 1,000 stock options vesting over 4 years. All else being equal (stock price volatility, expected life of the stock, dividend yield and risk aspects), under the current methods, Company A’s amortized stock charge is double the charge for Company B. That makes no sense. Why does a higher stock price at the time of grant give rise to a bigger charge? If anything, the grant in Company B should result in a bigger gain, as any gain will be a higher percentage of its stock price than for Company A.

The bottom line: the charge is misleading and arbitrary no matter how you look at it. If the stock price rises, that is the real compensation, but the true gain is not reflected anywhere.

In the same vein, if the stock price stays flat or decreases, the employee would have no gain and would not exercise the option. In effect, the grant recipient is not receiving any compensation, so there shouldn’t be a charge to the accounts as FAS123R requires.

Sticker shock. The inclusion of the charge can make a good operating performance look average or poor, and the charge can vary a lot from period to period based on what is happening with the company’s stock price.

Doesn’t reflect reality. You have to ignore the charge to get a good view of the underlying business. Analysts back the actual and expected charges out of their models so they can look at them on a cash basis. If they don’t need to see the charges, why do we? More and more companies are presenting adjusted EBITDA in their earnings press releases. These calculations back out the FAS123R charge for exactly the same reason analysts do—it’s a meaningless charge that mathematicians like but that users of accounts don’t need.

Most private companies ignore it. Who can blame them? There is no value added in accounting for it, and all it does is cost money in systems, review of the numbers and so on. An audit adds even more expense.

It makes budgeting hard. Have you ever put together an annual plan with FAS123R charges in it and then tried to hold people accountable to their budgets? It’s not easy, and most people won’t do it.

If you do want to do it (and it makes sense to have budgets that align to your financial accounts), to estimate the charge you need a crystal ball to estimate your future stock price at the time of the future grant, which you then need to combine with your estimated stock grants and headcount changes, as well as the residual charge from previous grants that are still vesting.

As a CFO, if someone asked you what your stock price will be in 6 months’ time you’d never answer (unless you enjoy SEC investigations), so why make this prediction internally to calculate the expected charge? And it’s impossible to hold managers accountable for their actual charges against the budgets for that expense. It’s also not wise to tie compensation to managing budgets if you have FAS123R in the compensation—at the end of the year the manager will be very happy or very unhappy, depending on which way the variance goes based on events totally out of their control.

So what’s the solution?

I believe FAS123R in its current form should be scrapped, and that only real gains, at the time of exercise, should be accounted for, and only in the notes to the accounts. By removing that expense from the accounts, you can then analyze, assess and compare companies based on their true operating performance, not some arbitrary performance.

Unfortunately, I don’t see any changes taking place soon—but the fact that more and more companies produce numbers that exclude FAS123R charges says that the FASB has gone too far in the accounting requirements, and that accounts are becoming more meaningless when presented under GAAP. Getting rid of FAS123R charges from the income statement would be a good first step to more meaningful accounts.