>In reference to your asset allocations analysis, how would stock
>options that are promised but not vested yet fit into this?
Great question. Let me give you a top-of-the-head response, without doing all the math. Assume for the purposes of this discussion that the employee is not considering exercising options to enhance lifestyle, but is solely concerned about building up assets. (Also, I'm not responding precisely to your "promised but not vested" condition.)
The value today of an (vested or not) employee stock option consists of the difference between the current value of the stock and what the employee has to pay for it when exercised, plus a premium that covers the likelihood of the stock increasing over the remaining time of the option, and is never negative. Whenever the option is exercised early, the employee sacrifices the so-called "time-value" premium. Such a sacrifice is warranted if (1) the employee feels strongly the stock price is about to trade significantly lower, so the "in the money" component needs protection in spite of tax consequences, or (2) if dictated by optimal asset allocation. It should be noted that the "time value" premium of a multi-year employee stock option for a high-growth, volatile company like WIND is huge, and should almost never be given away by exercising the option early. The classic Black-Scholes algorithm for pricing options (i.e. establishing the time value component) depends on the variance of the underlying stock and the amount of time remaining until expiration. Thus, it appears that the algorithm applies only to European-style options, in which the option cannot be exercised prior to date of expiration. In fact, the algorithm applies equally well to American-style options (sell any time) simply because it can be shown (under a host of reasonable conditions) to be always sub-optimal to sell early. However, to the extent that an employee has keen understanding of the company and the industry, it is possible that this understanding should credibly override the random-walk assumptions in the Black-Scholes algorithm, justifying premature exercise of the option.
Traditional asset allocation worries a lot about underlying statistical correlation between different classes of assets, or individual stocks in a portfolio. Nothing pleases a financial analyst more than discovering that two assets are negatively correlated, thereby seemingly working together as a team to reduce volatility ("risk"). Nothing is more scary to an analyst than observing two assets in a portfolio are highly positively correlated, as they give a false sense of diversification. They are fond of saying, for example, that you can increase expected return while decreasing risk by investing a portion of your portfolio in foreign markets - in companies about which you probably know absolutely nothing.
Consequently, the traditional financial analyst would warn employees against over-weighting their portfolios with their own company stock (including stock controlled through options). The reason is that if the company encounters hard times, two things can go wrong, and they are highly correlated. First, the stock will deflate in value. Second, the employee will enjoy less job security. The possibility of being out of a job with worthless stock options violates traditional views of prudent asset allocation focused on diversification.
On the other hand, just as investors sense that Warren Buffett probably really does know about investing, and may be right when he scoffs at diversification, employees must notice that lots of their predecessors in other companies have made tons of money over the years by continuing to buy company stock and by not taking quick profits on earned stock options. Microsoft has made many a millionaire of employees who just hung on to stock options and/or who purchased company stock, perhaps at a discount.
Accordingly, many WIND employees essentially face the same kind of investment question I tackled in my Asset Allocation posts. When is their portfolio of WIND stock options over-weighted, requiring them to exercise options, let's say with the intent of re-investing the after tax proceeds in other companies, bonds or other assets. How much WIND is too much?
The correct answer would approximate what I calculated for outside investors, but there are differences. First, outside investors do not consider job security implications in their analysis, although many WIND employees may be equally comfortable also ignoring job security in their considerations. Second, there are huge deferred tax benefits and time-value premiums associated with employee stock options, which would add weight to the "don't exercise strategy". (Similarly, if I had started up the asset allocation model I presented with the investor already heavily invested in WIND, sitting on sizable unrealized capital gains, the optimal allocation would have increased substantially beyond the large values calculated.)
It would be an intriguing exercise to extend the allocation model to handle a portfolio over-weighted in employee stock options, to see by how much concentration is dictated by deferred taxes and the need to preserve the time-value premium. But even without the calculations, we can safely surmise the general direction of the outcome.
Allen
PS - John Huber or anybody. Please correct me if I misstated anything about employee stock option technicalities. |
