Overall review of your option plans and your grant strategy

In light of FAS 123(R), more companies are considering restricted stock, options with performance-based vesting, indexed options, options with profit caps, and SARs among other alternatives to the basic fixed stock option. In July 2003, Microsoft announced that the company would stop granting options and increase grants of restricted stock. Under the revised plan, the top few hundred executives receive restricted stock with performance vesting; the broader grants have service vesting. In February 2004, IBM announced that it would begin granting premium options struck at 110% of the grant date stock price to its top 300 executives. IBM's stated intention is to extend the plan to most of its top 5000 executives. The strategy of IBM appears at odds with that of Microsoft when one considers the fact that restricted stock is like an employee stock option with a zero exercise price (except for the treatment of dividends and taxes). In part, these divergent strategies may be a reaction to each company's recent history. However, the strategies are not necessarily inconsistent. Corporate decision-making relating to equity-based incentive plans is complex and in to a certain extent company-specific. It involves:

1. Determining company objectives:

   • rewarding performance

   • retaining employees

   • minimizing accounting costs

   • minimizing dilution

   • simplifying plan administration

2. Analyzing information specific to the company:

   • company specific risk (volatility)

   • industry risk

   • correlation with the market (company beta)

   • correlation with peer group companies

3. Analyzing information specific to the employee:

   • risk aversion (desire for diversification)

   • wealth concentration (need for liquidity)

   • expected term of employment

   • level in the company

   • ability to affect company performance

Based on existing research, as well as empirical evidence from survey data, some guidelines can be set forth with the caveat that there is no substitute for learning as much as possible about company specific employee preferences and performing data analysis.

Stock is easier for employees to understand and convert into the certain cash payment equivalent. Stock grants also contribute to employee retention for vesting periods usually up to five years. This is true for a range of scenarios of stock price performance. The cost and dilution resulting from grants of restricted stock is relatively certain. The incentives that stock grants create are debatable. Options are potentially more rewarding of performance, particularly when there is a clear link between employees' performance and stock price performance. When performance vesting is contracted into stock or option grants it introduces another approach to rewards that can be tailored to measure impact on earnings and accounting metrics as well as market metrics. Additionally, options are more efficient for employees that can tolerate risk and have wealth diversification. At one end there is restricted stock, roughly an option with a zero exercise price, which imposes the least amount of risk on the employee. As the exercise price increases, from discount options to standard options to premium options, the implied leverage increases imposing more risk on the employee. For a given employee with a fixed level of risk tolerance and wealth concentration, and a fixed level of influence on company performance, the preference is typically for more options at a less volatile company, and less options (more stock or cash) at a more volatile company. Poor historical stock price performance makes options less attractive to employees. Companies provide a buffer for the risk of options with salary paid in cash. In many cases, optimal equity-based packages call for a mix of stock and options with a cash buffer, with the right mix depending on the company's assessment of cost, the employee's assessment of value and the incentives that the equity creates.

Examples of Performance Metrics

Performance-based vesting does not enter into the option valuation, but it affects accounting cost. If used properly it can contribute to the alignment of employee incentives with the objectives of shareholders. The most common metrics are financial metrics related to profits. These include:

• Net income ("NI"), earnings before interest and taxes ("EBIT")

• Per share metrics, the most common being earnings per share ("EPS")

• Margin metrics, such as NI / Sales

• Return metrics, such as return on invested capital ("ROIC"), return on equity ("ROE"), and return on assets ("ROA")

• Growth metrics, such as EPS growth

Some companies use revenue or revenue growth as a performance metric rather than a profit metric. Particularly for a company or business unit in the early stages of its life cycle, profit metrics can be unrealistic. Early stage growth companies might have a negative bottom line as a result of considerable investment to develop products, services, infrastructure, distribution networks and customer relationships. Investments might consume more cash than the business can generate in the early stages.

However, most companies prefer profit metrics, which take expenses into account. The advantage of a return metric such as ROIC over NI, EBIT or EPS is that it captures the fact that the creation of value depends not simply on earnings, but on earnings per unit of invested capital. ROIC also has potential disadvantages as a performance metric, again depending on the life cycle stage of a company or business unit. By depreciating capital and harvesting investments, it is possible to increase ROIC but at the cost of growth. Alternatively, economic profit or value-based metrics, such as "economic value added" ("EVA"), measure the return on capital in excess of the weighted average cost of capital ("WACC") multiplied by the amount of capital invested:

Economic Profit = (ROIC - WACC) x Capital Invested

Economic profit captures the fact that depreciation increases ROIC at the cost of declining invested capital. The idea is that a company, or business unit creates value by earning money per unit of invested capital in excess of the cost of the unit of invested capital, so long as it does not deplete the capital. Companies that can generate and/or sustain growth through investment are increasingly drawn to metrics related to economic profits. The challenge is communicating the more complex metrics to employees.

Additionally, companies use operational and production metrics related to volume and efficiency, and customer satisfaction metrics related to customer retention, repeat business, etc.

Choosing a metric is the first step. Benchmarking the metric is the second. The levels that are associated with performance targets can be based on internally observable standards such as budgets and forecasts, year-to-year improvements, external standards such as relative performance within peer companies, or simply the discretion of the board under advisement of the compensation committee.

Options with Performance Conditions that are linked to the Stock Price

When the performance metric is directly linked to the stock price, the accounting under FAS 123(R) is different than the accounting for a fixed grant, or a grant where the performance condition is not based on the underlying stock price. Specifically, there is no discount corresponding to a grant-date estimate of the effect of the performance condition on q. There is only a true up to reflect forfeiture related to the service condition - that is, departure from the company. The main point is that the performance condition is actually modeled as part of the valuation in contrast to a modified grant date method. Although options with performance conditions related to the stock price are likely to remain less popular in the United States than elsewhere, interest is increasing. This type of instrument continues to be very common in the United Kingdom and Australia and merits a brief discussion of the typical structure.

The most common approach is to use relative total shareholder return ("TSR"). TSR refers to the appreciation on company stock assuming all dividends are reinvested. Relative TSR refers to performance relative to an index either of the broad market or a set of peer companies. There are many ways to create performance conditions based on relative TSR. The following three basic examples consider vesting that depends on the company's percentile TSR rank among a set of peer companies, where TSR is measured from the grant date:

1. The number of options that vest range from 0% to 100% based on the company TSR relative to TSR for the set of peer companies as follows. If the company TSR ranks in the lower 50th percentile (at least half of the peers have higher TSR), then no options vest. Between the 50th and 75th percentile, 50% to 100% of the options vest, increasing linearly by 2% for each additional percentile between 50th and 75th percentile. For example, if the company TSR ranks in the 60th percentile, then 70% of the options vest. Full vesting occurs if the TSR ranks above the 75th percentile.

2. If the TSR ranks in the lower 25th percentile (at least 75% of the peers have higher TSR), then no options vest. Ranking between 25th and 50th percentile, 25% to 50% of the options vest, increasing linearly by 1% for each additional percentile. Ranking between 50th and 75th percentile, 50% to 100% of the options vest, increasing linearly by 2% for each additional percentile increment. If the company TSR is above the 75th percentile, then all options vest.

3. During a period of three years from the grant date, if the company TSR ranks in the upper 50th percentile for any 90 day period, 25% of the options vest; for any 180 day period, 50% of the options vest; for any 270 day period, 75% of the options vest; and for any full year, all options vest.

Because prediction and true up as it relates to performance-based vesting linked to the stock price is not compliant with the FAS 123(R) modified grant date method, this type of performance condition should be implicit in the grant date valuation of each option. The best solution to the valuation problem is to simulate the total shareholder return of the company along with the n-1 peer group companies. The standard method uses the assumption that each stock return (with dividends re-invested) is lognormal with a covariance matrix that best describes the correlations of each return with the other n-1 returns. The covariance matrix is estimated from historical data. In the case of n peer companies, each iteration of the simulation produces n stock price paths. (The next section describes simulation of two correlated variables.) Therefore, each iteration represents a simulated state of the world in which the company TSR can be compared to the other n-1 TSRs on any relevant date or relevant period.

Just as there are many ways to create performance conditions, there are many ways to choose an index or a set of peer companies. In addition to a broad market index, e.g., the S&P 500 or Russell 2000, or a standard off-the-shelf industry index such as the S&P midcap 400 healthcare index, some companies opt for a custom index of peer companies. Peer companies can be chosen depending upon a number of criteria, including:

• SIC, GICS or NAICS classification

• Market Capitalization

• Industry Risk

• Earnings

• Cost of Capital

• Stock Volatility

• Product or Services Comparability

• Labor market Comparability

This list of criteria is not exhaustive. Furthermore, each company should decide on the best approach for choosing a set of peer companies or a market index based on historical correlations of returns and variance of returns relative to the market or a tailored set of peer companies.

Options with an Indexed Exercise Price

Indexed employee stock options are call options on company stock that have an indexed exercise price rather than a fixed exercise price. The point of an option with an indexed exercise price is to reward superlative performance in a bear market or weak sector, while avoiding rewards for mediocre performance in a bull market or a strong sector. If S₀ and I₀ are the grant date stock price and index levels, respectively, then the intrinsic value of the indexed option on any given date t is . The idea is that an option with an exercise price, which changes in lockstep with an index, rewards employees when the stock price outperforms the index. As is the case for performance conditions, discussed above, the index can be the broad market, a standard off-the-shelf industry index, or a custom index of peer companies. The standard valuation approach, as illustrated in paragraphs 311-316 in FAS 123, is to treat the index as the numeraire in place of cash. That is, the "currency" of the strike price is effectively the index. The index dividend yield is the rate of return of the numeraire in place of the risk-free rate. The volatility of the relative return, of the stock to the index replaces the stock return volatility as an input, and the exercise price is the value of an equivalent share of the index. The setup is equivalent to an option with a strike price of $1 on I₀ shares of an asset with a price at time t of $ . Note that if S and I are both lognormal, then so is :

If the expected correlation, ρ, between the stock and the index can be estimated with no more difficulty than the expected volatilities, standard valuation methods including lattices and Monte Carlo simulation can be used. Simulation is particularly well suited to valuing indexed options. The standard approach is to generate two random processes, rather than one, using a pair of independent standard normal random variables, ε₁ and ε₂. Then the second random variable is transformed using to account for the correlation
between the stock and the index. An iteration of the simulation models is now a pair of paths. The option intrinsic value at any given time is modeled as the spread between the paths, which is the difference between the stock price level, S_t and the level of an equivalent share of the index, S₀ I_t / I₀.

An option with an indexed exercise price should not be confused with an option with a performance-based vesting condition based on stock performance relative to an index. In the case of the latter, the index affects the size of the award that will vest. The effect works its way into the valuation because FAS 123(R) does not allow a modification to grant date accounting to reflect a true up of q. Disallowing this modification, which is retained from FAS 123, also makes economic sense because vesting and valuation are directly intertwined. The scenarios that are positive (negative) for vesting are highly correlated with higher (lower) option cash flows to the employee.

In the case of an option with an indexed exercise price, the index affects the option intrinsic value past the vesting date and throughout the option term. So long as an indexed option does not also have a performance-based vesting condition related to market price, p and q are measured independently. The grant date accounting is then modified to reflect a true up of q on the date of vesting.

It is interesting to note that an employee with sufficient liquid assets could independently purchase the index and hedge the indexing effect. This suggests that, if the company granted index options to an employee rather than fixed stock options, then the exchange rate should be the cost to the employee of creating the index hedge, adjusted for the liquidity constraint that the hedge imposes on the employee.

Premium Options

Options that are granted with the exercise price exceeding the grant date stock price are premium options. Premium options have a lower accounting cost per share than standard options. They also have lower value per share to employees. The rationale underlying premium options is that they require a positive return to shareholders above some basic threshold before providing any intrinsic value to the employee. IBM announced in February 2004 a plan to grant premium options to executives rather than at-the-money options.

Options with a Profit Cap

Companies can grant standard stock options with a cap on profits based on a multiple of the exercise price. For example, an option with a strike price of $50 and a 100% profit cap entitles the option holder to the minimum of intrinsic value and $100 upon exercise. This reduces the per share accounting cost. Options with a profit cap are straightforward to value as barrier options in a trinomial tree.

Stock Appreciation Rights ("SARs")

SARs give the employee the right to a cash payment, stock, or a combination of the two based on the market appreciation of the stock in excess of a stated price for a stated number of shares. The stated value, in effect a strike price, is usually the fair market value on the grant date; hence it's similar to the strike price of an at-the-money option. However, SARs are different from stock options in that the employee does not make a capital investment in the company at the time of his or her choosing subsequent to vesting. Rather, SARs provide a method of effectively paying employees a deferred bonus that is linked to stock appreciation.

From a company's perspective, SARs are a way to share the value of equity with employees without actually sharing equity in the case of cash settlement. Furthermore, the regulatory requirements and implementation costs of SARs are lower than those of options. In the case where a company is a division or subsidiary of a larger entity, SARs are a way for employees to share in equity value without actual stock.

Research into the Question of Stock vs. Options

Substantial academic literature exists on the subject of equity-based incentive instruments and their relative cost and efficiency. The components of much of the research include analysis of:

1. Value to the employee of an equity-based incentive instrument, whether it be stock or options; this is a subjective valuation and depends on the employee

2. Cost to the Company of the instrument; this is an objective market valuation

3. The incentive that the instrument creates as measured by the sensitivity of the value to the employee to changes in equity value.

The efficiency of an equity-based incentive instrument refers to the strength of incentive (iii), relative to the cost of the instrument, (ii).

For stock, objective valuation is the current fair market value of the stock without marketability discounts related to restrictions. For options, the traditional measure of objective value since the publication of FAS 123 in 1995 is Black-Scholes with expected term. The subjective value for stock is usually measured with a market model or other empirical research to arrive at discounts for non-marketability resulting from restrictions on selling. The marketability discounts that apply to restricted stock depend on the characteristics of the employee, such as level in the company, income, and wealth concentration. For restricted stock, Rule 144 of the U.S. Securities and Exchange Commission (SEC) affects the ability of most corporate insiders to sell shares after vesting leading to illiquid positions for several years. Holding period is an important factor in assessing restricted stock illiquidity discounts. For options, the subjective value is usually measured with a market model or a preference-based model.

Studies on illiquidity discounts have used a diversity of assumptions and methods to find average restricted stock discounts in the 25% to 35% range. Examples of early studies include Moroney (1973) and Trout (1977). Later studies include Silber (1992) and the Mercer (1997) "Quantitative Marketability Discount Model." The former is an empirical regression analysis of blocks of restricted stock, while the latter is theoretical and requires knowing certain inputs that are as difficult to estimate as the discount itself. Recently, Meulbroek (2001), Tabak (2002) and Kahl, Liu and Longstaff (2002) present market model approaches. Meulbroek (2001) analyzes the illiquidity discount by setting the sharp ratio of the market equal to that of the illiquid stock. This provides a lower bound on the discount for illiquidity. Tabak (2002) provides a good review of the literature and presents a market-based model of illiquidity discounts based on the equity risk premium in the market at any given time. Valuation for an employee with no outside wealth and stock with a two-year restriction would require a discount roughly in a range between 15% and 33% for stock with lower than average volatility, and between 33% and 77% for stock with higher than average volatility. As with options, the impact of stock illiquidity, and the corresponding valuation discount, depends on the level of diversification of the stockholder vs. the desired level of diversification. Clearly, the ability of the employee to diversify is a function of wealth that is already tied up in company stock, as well as the components of risk in company stock, some of which is likely difficult to diversify away.

Academic papers that criticize option contracts as being inefficient include Meulbroek (2001). She argues that, relative to the cost of options to the company, the value to the typical employee is too low to make them an efficient tool for providing equity-based incentives. Meulbroek claims that the typical employee at a NYCE company values options at 70% of what the options cost the company.

In contrast to Meulbroek's paper, Core and Guay (2003) claim that the typical employee's wealth would be exposed to market risk even if they were not employed at the company. Principal agent theory suggests that the contracts including equity-based incentives do not expose employees to more market risk than they are willing to bear. As a result, the wedge between employee valuation and company cost is generally overstated, and that conclusions cannot be drawn about optimal equity-based incentives without knowing the wealth and wealth concentration of employees.

Hall and Murphy (2002) and Jenter (2000) also claim that restricted stock dominates options with positive exercise prices, including standard options granted at-the-money and premium options granted out-of-the-money. But Lambert and Larker (2004) argue that the analyses of Meulbroek (2001), Hall and Murphy (2002) and Jenter (2000) are not complete concerning the option cost to the company, value to a variety of employee types, and incentives that result from hypothetical equity-based contracting. Lambert and Larker find that options with positive exercise prices, including both at-the-money options and premium options often dominate restricted stock rather than the other way around. The optimal mix, based on the theory of optimal contracts, as pioneered by Holmstrom (1982), depends on the specific risk aversion of the employee and the ability of the employee to influence company performance.

Valuation and Strategy

Copyright © 2008 FAS123 Solutions, LLC

Home   |    Contact Us