mishedlo -
There is no way to say in general whether option grant compensation helps or hurts shareholders, as the terms could be extreme in either direction. Plus there is no way to predict the price of the stock and how valuable the retained employees actually are. However, expensing options is always a form of double counting the negative impact on shareholders of dilution.
The following is a comprehensive discussion of the issue from a supporter of expensing options, followed by an email That I sent to him this morning.
online.wsj.com
"More Accounting Answers
For Confounded Investors
[Editor's note: Understanding accounting has become an important issue for investors. In an effort give investors a better handle on the complex financial statements, we're asking experts to explain how companies can use gimmickry to improve their reported financial results.
This is the second of two-part question-and-answer session with Charles Mulford, Professor of Accounting at the Georgia Institute of Technology in Atlanta. Last week's installment is here.
In this week's installment, Mr. Mulford discusses questions about stock-option accounting, earnings restatement and gimmickry to watch for in the future.]
WSJ.com: There has been a great deal of chatter about stock options and accounting. Do you think that compensation expense should be recorded for stock options or do you think that footnote disclosure is sufficient?
Mr. Mulford: Companies presently have a choice in how they account for compensation expense associated with stock-option grants -- an intrinsic-value approach or a fair-value approach. Under the intrinsic-value approach compensation expense is recorded for the option's intrinsic value. This is the amount by which the exercise price of an option granted is exceeded by the market price of the underlying stock on the measurement date, typically the date of grant. With the fair-value approach compensation expense is recorded for the estimated fair value of the options granted using an established valuation model such as the Black-Scholes option-pricing model. With both approaches any compensation expense recorded is spread over the estimated service period of the recipient, usually the vesting period.
Most option grants entail qualified options that afford certain tax benefits to the recipient, mainly that no income tax is due on shares obtained through stock-option exercise until the acquired shares are sold. At that point capital-gains treatment is applied. However, qualified plans require that an option's exercise price cannot be less than the market price of the underlying stock on the date of grant. Thus, such options have no intrinsic value. Accordingly, under the intrinsic-value approach there is no compensation expense to be recorded.
The vast majority of all companies use the intrinsic-value approach. While companies choosing this method will not, in most cases, record compensation expense, they still must disclose in a footnote to their financial statements an estimate of the cost of the options granted using the fair-value approach.
Thus, as long as options are granted with no intrinsic value on the date of grant, generally no compensation expense is recorded. This is not to say that the options do not have value. Recipients certainly do not consider options to worthless or they would not be inclined to accept what in many instances is a below-market cash-based rate of pay.
In the early 1990s the Financial Accounting Standards Board proposed requiring companies to use a fair-value approach to recording compensation expense for stock options. Companies, especially technology firms that had been generous in their stock-option grants, argued forcefully against such a measure. Their argument held that requiring expense recognition for stock options would affect the viability of many technology firms by significantly reducing their ability to use options as compensation and in the process would threaten our nation's technology edge. In Congress they found a sympathetic ear and pressure mounted on the FASB to back down. The compromise solution reached is our present practice of permitting a choice in the method used to account for stock options but requiring footnote disclosure of the effects on earnings of the fair-value approach if it were not used in recording compensation expense.
In the past year, as serious questions about accounting generally in the U.S. have been raised, the option-accounting issue resurfaced. The use of options as a form of compensation soared in the late 1990s and into the new decade, all without the recording of compensation expense. Were earnings being properly measured? Also putting pressure on our regulators to reconsider the issue is talk that the International Accounting Standards Board, a board that is gaining stature in its pursuit of a body of common accounting standards internationally, is considering adopting a fair-value approach. In addition, very recently Standard & Poor's announced that it would incorporate estimated compensation expense for stock options in its calculations of operating earnings.
I personally side with those who think that a fair-value approach should be used in recording compensation expense for stock-option grants. The options have value, they are being used as a form of currency to pay for services and accordingly, there is an accompanying expense to be recorded. While measuring the value of options is based on an estimate, we use estimates for many items reported in financial statements.
There are other valid arguments against the fair-value approach. In estimating the value of stock options, the Black-Scholes model incorporates assumptions for such factors as the exercise price of the option, the market price of the underlying share and the volatility of that price, the level of interest rates, the underlying share's dividend yield, and the option's term to expiration. It was developed more for publicly traded options with relatively short expiration periods. Stock options granted to employees and others are not publicly traded and carry expiration dates that often extend out as far as ten years. Whether the assumptions employed by the Black-Scholes model are appropriate for such options is open to debate. Though it would seem that even if the model were not perfect for estimating the values of stock options it is better than assigning no value to those options.
Others note that the granting of options involves no cash payment and indeed results in cash inflows for the amount of any exercise price received plus, in some instances, a tax benefit. How can there be expense with no ultimate cash payment? Many companies that issue options will use cash to repurchase their own stock in order to avoid the dilutive effects of stock-option exercises. Others may use treasury shares, purchased earlier for cash, to satisfy exercises of stock options. However, even without using cash to repurchase shares there is a foregone cash inflow for the excess of the market price of the stock over the exercise price on the date of exercise. In fact, it is roughly the present value of this excess that a valuation model such as the Black-Scholes model attempts to estimate when assigning a value to stock options on the date of grant.
Another argument against the fair-value approach is that the effects of options are seen in the dilutive effects caused by an increase in the number of shares outstanding used in the calculation of earnings per share. That is, the inclusion of the effect of options on earnings and on shares outstanding would result in a double counting of their impact. It should be noted that dilution reflects a division of equity ownership and of earnings and does not affect the absolute amount of that equity or those earnings. Stated differently, the dilutive effect of options does not appear in reported net income or shareholders' equity. Also, there is no debate about the need to record expense when shares are granted outright for services. Yet even here there is an expense effect and a dilution effect.
These are all valid arguments, but I think that they can be addressed in a satisfactory manner. Unfortunately, the issue of whether or not to record compensation expense remains a political one subject to pressure from lobbyists and political rhetoric. In such a politically charged environment it is difficult to get objective discourse. If we can move the debate away from politics, something that is important if there is to be more trust in our accounting standards, I think that there would be an excellent opportunity to make meaningful change to our accounting practices for options. Unfortunately, this may be a tall order to fill. ..."
Dear Mr. Mulford,
I found that your WSJ Online article of 6/13/02 covering stock-option
accounting was by far the most comprehensive and objective that I have seen,
but nevertheless, still contains what I would categorize as misconceptions
based on widespread economic fallacies. In addition, although it may be far
too much to ask for accounting to be entirely logically self-consistent,
methodologies that give grossly conflicting results for similar
circumstances are troubling. Finally, the question arises as to what exactly
the purpose of accounting is. My expectation would be that the accounting
function should be to comprehensively assure existing shareholders that
their interests are being protected and advanced by management. Since it is
the shareholders that are paying for the accounting service, they should be
the customer. It appears to me that many of the accounting controversies
that currently exist seem to assume that accounting exists merely to reduce
all companies to a handful of numbers that potential investors can compare
on a computer screen without having to do any actual work in understanding
what actually makes each unique company tick. If investors want this result,
they should be the ones paying for it.
From the article :
"...Another argument against the fair-value approach is that the effects of
options are seen in the dilutive effects caused by an increase in the number
of shares outstanding used in the calculation of earnings per share. That
is, the inclusion of the effect of options on earnings and on shares
outstanding would result in a double counting of their impact. It should be
noted that dilution reflects a division of equity ownership and of earnings
and does not affect the absolute amount of that equity or those earnings.
Stated differently, the dilutive effect of options does not appear in
reported net income or shareholders' equity. Also, there is no debate about
the need to record expense when shares are granted outright for services.
Yet even here there is an expense effect and a dilution effect. ..."
This paragraph contains the key, "... Also, there is no debate about the
need to record expense when shares are granted outright for services...." To
the contrary, this is exactly the pivot point upon which the question of
expensing option grants must fall. If I were to believe that stock grants
should be expensed, there is no logic in the world that could support
options not being expensed as well. Unless otherwise specified, all of my
arguments below deal with stock grants, and apply to option grants as a
logical extension.
While Warren Buffet apparently can't see it, both stock and option grants
represent offers of contingent partnership, not expenses. The entire cost is
borne by existing shareholders in dilution, as you note, just the same as
any expense is ultimately borne by existing shareholders. If you create a
fictitious public company whose assets consist entirely of $100K in cash,
staffed by a single employee, whose entire value is liquidated over ten
years by the payout of cash dividends, the net results for both the employee
and every shareholder are exactly the same to the penny, no matter whether
the salary of the employee is paid in cash or in stock or a mix of both.
Assuming, for simplicity, that the discount rate is zero, the fictitious
company's value is precisely known every year as it declines to zero. Any
thought of creating a phantom salary expense line for stock grants is out of
the question as a clear distortion and error.
If stock grants are to be expensed, should the inventor/founder sole owner
of a company who gives away a 50% ownership to a manager/partner see an
expense of 50% of the market value of the company in addition to being
diluted by 50%?
If you blindly sample a company at one year intervals, and you find that the
employees have become part owners during the previous year, in every case in
which the employees have received stock grants in lieu of salary
compensation, the exact same result for everyone can be accomplished by a
combination of a fractional stock split and having the shareholders donate
their extra shares received to the employees in the same quantities as with
the direct stock grants. Why is no one calling for expensing stock splits?
When a company pays cash salaries, the compensation is real dollars that
leave both the company and the existing shareholders and reduce the value of
the company. If Warren Buffet bids $1B for an entire company and then finds
that $100M in deferred salaries was overlooked, he would presumably lower
his bid. If instead,
he finds that the CFO had spent the previous Saturday afternoon handing out
Treasury shares for 10% of the company to every other passerby in Times
Square, he might well look for a new CFO, but his $1B bid for the entire
company should still stand, as the value of the company has not effectively
changed.
Economic Fallacy #1 -
From the article -
"...Thus, as long as options are granted with no intrinsic value on the date
of grant, generally no compensation expense is recorded. This is not to say
that the options do not have value. Recipients certainly do not consider
options to worthless or they would not be inclined to accept what in many
instances is a below-market cash-based rate of pay...."
Whatever the recipients consider the options to be worth has absolutely
nothing to do with what they cost the company. This is probably the Number
One Fallacy of popular economics. When an economic exchange is made, people
tend to believe that the parties involved have agreed that the items
exchanged are of equal value. In fact, nothing could be further from the
truth. Every time I spend a dollar, I believe that whatever I purchase is
not only worth more than a dollar to me, but that is worth more than any
other thing I could use that dollar for, including saving it. Even more, the
person who sold the item to me believes exactly the reverse, that the dollar
is worth more than the item sold and that it is the most he could have
received for it.
There is no such thing as an intrinsic economic value. All economic values
are subjective, determined on the margin, and subject to the law of
diminishing marginal utility. (see Austrian Economics, for example
mises.org )
Economic Fallacy #2 -
From the article -
"... Others note that the granting of options involves no cash payment and
indeed results in cash inflows for the amount of any exercise price received
plus, in some instances, a tax benefit. How can there be expense with no
ultimate cash payment? Many companies that issue options will use cash to
repurchase their own stock in order to avoid the dilutive effects of
stock-option exercises. Others may use treasury shares, purchased earlier
for cash, to satisfy exercises of stock options. However, even without using
cash to repurchase shares there is a foregone cash inflow for the excess of
the market price of the stock over the exercise price on the date of
exercise...."
There is no significant foregone cash inflow. Unless stock certificate #
314159 is valued as a collectible, after you grant it to an employee,
virtually the same market price can be received for # 314160 as soon as it
cools coming off the copy machine. One share is as good as another.
All economic costs are opportunity costs. The cost of preferred action A is
the value of the best alternative action B which was precluded by the
undertaking of action A. The opportunity cost of a Paris vacation was
the preclusion of a Las Vegas vacation, ranking second. Accounting costs may
bear little resemblance to economic costs, but there is likely to be some
connection in the end. Granting one or more new shares in no way limits
subsequent new share sales or grants.
Assume that famous author Tom Clancy's autograph has a market value of $500.
Does his accountant enter an expense line of $500 every time he signs his
name? Even if he does, does he submit it to the IRS? Market value and costs
have no necessary relationship, especially where there is no real scarcity
on the production side.
None of the above should be misconstrued to say that option grants are not
greatly abused. They are, but the problem is not with the option grants per
se, but rather with the use of stock buybacks at any price to disguise the
share dilution. This is where accounting should be looking for remediation,
along with making sure that option grants are fully reflected in the
reported share count dilution, possibly even 100% at time of grant until
void. There is clearly a major inherent conflict of interest between
management and shareholders when it comes to compensation of all types.
Thank you for your attention. In spite of all of the above, I found your
article both interesting and educational as it clearly exposed many facets
of the problem seldom exposed.
Regards,
Don Lloyd
Peabody, MA |
