Energy Executive Compensation: The Growth of TSR Plans
"By assessing performance relative to peers, a Total Shareholder Return plan normalizes the measurement to remove the effects of conditions affecting the entire industry. It places less focus on short-term stock price swings and requires sustained performance over time.” – said Opportune’s Shane Randolph and Petar Tomov in Oil & Gas Investor...see below to read the full article.
During the most recent energy downturn, many companies are considering how to best compensate executives for their performance. Compensation generally includes a base level of compensation combined with some form of incentive compensation that is tied to performance. It can be very challenging to develop a compensation program that links entity value growth with management’s performance. This is compounded when attempting to develop multiyear performance goals for management in a volatile business environment.Companies have many alternatives for incentive compensation programs. Stock options and restricted stock have their unique challenges and there is not necessarily a high correlation between executive performance and the value the executives receive as compensation. Executives may receive value as long as the company remains viable and the recipient does not quit or is not terminated. These forms of incentive compensation are also frequently tied to a predetermined plan or budget.
Associating compensation to someone’s forecast of future results, rather than delivered results, has its own set of limitations. For these reasons, many energy companies in the current environment have focused incentive compensation plans around relative total shareholder return (TSR) plans. The following is a discussion of TSR plans, the associated accounting implications, and the valuation challenges that they pose.
A TSR plan measures the company’s return against the performance of a peer group of competitors over a given period of time. The results of the company against the peer group are then used as the metric to determine whether vesting of the award is achieved and at what payout level. The intent of the plan is to compensate executives for their performance relative to peers, where executives are encouraged to match or beat the performance of competitors.
By assessing performance relative to peers, a TSR plan normalizes the measurement to remove the effects of conditions affecting the entire industry. It places less focus on short-term stock price swings and requires sustained performance over time.
The recipients can realize value from the award as long as the company outperforms the peer group.
There are various components to a TSR plan design such as the peer group, the performance period considered, stock price averaging formulas, payout structures and free cash flow requirements. The peer group is one of the most important elements. The peer group companies are generally public companies that allow for the use of publicly available stock price closes and other financial information. The group provides a readily available metric for companies to access, and it frequently reduces subjectivity and the measurement burden.
While a company can determine which companies are included in the peer group, the establishment of the peer companies can be challenging. The peer companies do not have to be identical (i.e. same basin, natural gas/crude asset base, midstream assets composition, etc.), but there is generally a high level of commonality among the companies. In addition, companies in the peer group may not exist by the end of the performance period due to merger, acquisition, or bankruptcy.
The US GAAP accounting treatment for TSR awards falls under the guidance of ASC 718, “Compensation – Stock Compensation.” Forms of compensation where the amount of compensation received is tied to the market price of company stock generally fall under this accounting guidance. The primary forms of share-based compensation considered under the guidance are stock options, stock appreciation rights, restricted stock and restricted stock units.
TSR awards are performance-based restricted stock units, otherwise known as performance shares, where the TSR serves as the performance metric. The accounting treatment for performance shares depends on whether the metrics are based on a performance condition (e.g. EPS, net income, etc.) or a market condition (e.g. a company’s stock performance relative to a peer group or an index). The general accounting objective is to record the fair value of compensation expense over the periods in which services are performed. This requires determining the fair value of the compensation and the period of time in which participants perform the services.
The accounting treatment of TSR awards will depend on whether the awards are recorded as equity or as a liability. If they are recorded as equity, the amount of compensation is estimated at the grant date at fair value. This amount is expensed over the service period. If they are recorded as a liability, the fair value must be re-measured every reporting period until settlement. Fluctuations in the fair value of the liability award are recorded as increases or decreases in compensation cost, either immediately or over the remaining service period, depending on the vested status of the award.
The determination of whether the TSR awards are recorded as equity or a liability can be a complicated analysis. TSR plans are generally structured in a manner that allows equity classification. However, the following is a list of characteristics that would likely trigger liability accounting.
- Awards are required to be cash-settled or require cash settlement on the occurrence of a contingent event that is considered probable;
- Awards can be settled in cash or stock at the option of the employee at any time or on the occurrence of a contingent event that is considered probable;
- Conditions other than service, performance, or market conditions, that affect their fair value, exercisability, or vesting;
- Instruments that are equity in form, but the employer has a practice of cash-settling the instruments; and
- Awards where the employer can choose cash or share settlement, but cannot control delivery of shares.
- A service condition is a condition that requires the individual to remain employed by the company for a stated period of time in order to earn the right to the related equity instrument (i.e., to vest);
- A performance condition is a condition that is based on the operations or activities of the employer. The condition may relate to the performance of the entire company, a division, or an individual employee; and
- A market condition references the exercisability or other terms of payment that may be dependent on achieving a specified stock price or a specified return on the stock price (e.g., price appreciation plus dividends).
Compensation cost is only recognized for awards earned or expected to be earned. Compensation cost is not recognized for awards forfeited or expected to be forfeited because a service or performance condition is not achieved. Unlike a service or performance condition, a market condition is not a vesting condition. Rather, a market condition is factored into the grant date- fair value-based measure of an award. Accordingly, regardless of whether the market condition is satisfied, an entity may still be required to recognize compensation cost for the award.
There are two accounting classifications of performance metrics. The first is performance conditions, such as EPS, revenue target or market share. The second is market conditions, such as relative TSR, stock price, absolute TSR or internal rate of return. Therefore, for accounting purposes, relative TSR awards are restricted stock units with a market condition where the vesting is contingent on the share price performance over a given period.
The following is an example of performance shares, with a performance condition:
- The number of performance shares earned is based on EPS target at the end of three years;
- Grant date stock price is $12;
- 1,000 target performance shares are granted;
- At the grant date, the company expects 100% of the performance shares to be earned. The total expense is $12,000. Therefore $4,000 in expense is amortized each year over three years; and
- After two years, the company expects 150% of the performance shares to vest. Therefore, a true up amount of $4,000 is recognized immediately, and $6,000 is amortized over the last year.
- The number of performance shares earned is based on relative TSR over a three-year performance period;
- Grant date stock price is $10 and estimated grant date fair value is $12;
- 1,000 target performance shares are granted;
- The total expense is $12,000. Therefore $4,000 in expense is amortized each year over three years;
- At the end of the performance period, the actual performance earns 150% of the performance shares (or $15 per share); and
- No change to expense would be made because the grant date fair value accounted for the probability of earning between 0% and 200% of target.
In valuing a plain vanilla option, a Monte Carlo model and a Black-Scholes model would yield the same value conclusion if identical assumptions are used. However, the Monte Carlo model provides more flexibility, as it allows the user to customize the model in ways not possible with a Black-Scholes model. This customization characteristic is also what makes the Monte Carlo model more difficult to implement, as each model is prepared on an individual basis in accordance with the terms of the awards valued.
A Monte Carlo simulation is a generally accepted procedure for sampling random movements of future share prices, which is often implemented using a computer program to generate a large number of possible results. The simulation model forecasts future movement of share prices using several key inputs combined with empirical knowledge of the stochastic (random) process, which governs stock price behavior. The most widely used stochastic process for modeling stock price behavior is Geometric Brownian Motion, whose analytic solution is shown below.
Here’s how each input impacts the valuation:
In the above equation, St represents the forecasted stock price at a future point in time (t). S0 is the starting stock price input used for the simulation analysis, and it is based on the stock price at the valuation date, or t=0. The valuation date, or grant date of the TSR award, may not always coincide with the beginning of the performance period. In fact, most of the time, the performance periods as defined in TSR plans start a few months prior to the grant date of the awards. In such cases, the fair value of the TSR award, as estimated using the Monte Carlo simulation model, will be impacted by how the company stock price has performed relative to the peer group between the beginning of the performance period and the grant date. For example, if the company’s stock price has underperformed relative to the peers in that time period, there will be a greater probability of a lower payout, which will result in a lower fair value conclusion.
The expected average annual rate of return of the stock price appreciation is represented by Mu (µ). For US companies, the annual rate of return, or risk-free interest rate, is determined by observing the U.S. Treasury zero-coupon yields over a period commensurate with the time remaining in the forecast period or the performance period of the relative TSR award. For foreign companies, the risk-free interest rate is based on the implied yield available on zero-coupon government issues denominated in the currency of the market in which the shares primarily trade.
Although an increase in the risk-free interest rate will result in a greater probability of a higher last day stock price, it will also result in a higher discount rate used to present value the estimated future value of the award. Furthermore, because the risk-free interest rate assumption is the same for forecasted future stock prices of both the company and all of its peer companies, the final payout in a TSR award will not be affected by movements in the risk-free interest rate. Therefore, the net effect of an increase in the risk-free interest rate on the fair value of a TSR award is zero.
The volatility of the stock price, or Sigma (σ), represents the expected degree of variation in stock price over the time remaining in the performance period of the TSR award. In accordance with ASC 718 guidance, expected volatility should be measured using a combination of historical volatility (based on standard deviation of actual returns) and implied volatility (based on volatility of traded options). However, when estimating the volatility assumption for relative TSR awards, practitioners tend to rely strictly on historical volatility. This is done to ensure that the volatility assumption is on the same basis as the correlation coefficients assumptions, which can only be estimated using historical data.
An alternative approach might be to prepare two price paths in the simulation model, where the first path uses a volatility assumption based on historical volatility alone, and the second path uses a volatility assumption based on a combination of historical and implied volatility (per ASC 718). The first path would be used to calculate the company’s relative TSR rank compared to the peer companies, while the second path would be used for purposes of projecting the company’s final stock price at the end of the performance period.
An increase in the volatility of the subject company, when the volatilities of all peer companies remain the same, will result in a decrease in the fair value conclusion due to a lower probability of stock price appreciation. Conversely, an increase in the volatility of the peer group companies, when the volatility of the subject company remains the same, will result in an increase in the fair value conclusion due to a greater probability of higher payouts.
Random Numbers and Correlations
In the above equation, Epsilon (ε) is a random number that is normally distributed with a mean equal to zero and standard deviation equal to one. The random numbers in the Monte Carlo simulation are adjusted using Cholesky Decomposition to arrive at correlated random numbers. The Cholesky Decomposition is a matrix of pair-wise correlation coefficients, which reflect the historical relationship between the stock returns of the company and the peer companies over a period commensurate with the performance period.
As such, the correlations are incorporated into the random numbers to govern the behavior of future stock prices for the company and the peer group, in accordance with historical observations. The greater the number of peer companies, the bigger the correlation coefficient matrix, and the longer it will take the Monte Carlo simulation model to run. The direction and magnitude of the future stock price movement at each time step is influenced by the correlated random number.
When forecasting the future stock prices for relative TSR awards, the treatment of dividends should be carefully assessed. In most cases, the calculation of the payout percentage, as defined in TSR plans, assumes dividends are reinvested into the company and each peer company as they are paid. This can be incorporated into the Monte Carlo simulation by assuming a zero percent dividend yield, which is mathematically equivalent to the assumption that dividends are reinvested as they are paid out.
Additionally, practitioners must evaluate whether the holders of the TSR awards are eligible to receive payments of dividends over the performance period and evaluate the form of those dividend payments (cash or equity).
If the treatment of dividends is different from the payout percentage, then a separate price path should be generated for the subject company’s final stock price, incorporating the treatment of dividend equivalents during the performance period. If dividends are to be paid in cash, they are considered a liability and should be tracked separately and likely marked to market quarterly. As companies develop incentive compensation awards during volatile market conditions, the use of relative TSR plans will likely continue to increase.
Companies should carefully consider the accounting and valuation implications of TSR plan terms. The developed plan terms will greatly impact the ongoing accounting requirements as an equity or liability classification. In addition, the valuation process for complex terms may require valuation expertise not housed within the company.
Shane Randolph is a managing director at Opportune LLP. He has deep risk management, derivative valuation and hedge accounting experience for natural gas, crude oil, NGL, electricity, agricultural products, metals, foreign currencies, and interest rates.
Petar Tomov is a manager at Opportune. He focuses on the valuation of stock based compensation, complex financial instruments, debt and equity, and contingent claims for privately held and publicly-traded companies.
Both Shane and Petar can be reached via opportune.com.