Executive Compensation Goal Setting: Putting Together Your Best Toolkit
Setting rigorous executive compensation goals is critical for aligning pay with performance. Discover best practices, from metric selection to statistical modeling, that ensure fairness and shareholder confidence.
One of the most challenging tasks for many companies is setting goals for their compensation plans that effectively align executive incentives with shareholder interests. A well-designed program needs to recruit and retain top talent, incentivize executives to meet or exceed performance targets and assure investors that they are getting value for money.
Not all companies demonstrate a commitment, nor have the required tools and data, to set incentive metric goals at sufficiently rigorous levels or develop a balanced payout formula. These situations may contribute to misaligned incentives or a disconnect between pay and performance.
ISS-Corporate examined some of the most rigorous corporate executive incentive program designs across industries and identified best practices across key areas: metric selection, using statistical analysis for goal setting, understanding investor expectations and providing robust disclosure. Read our full white paper on goal setting for compensation, or read on for key takeaways from the paper on how to set effective executive compensation goals.
Choosing the Right Metrics
Properly balancing the need to incentivize executives and reward shareholders begins with metric selection. A review of peers can provide a springboard, but further discussion is required to determine how specific metrics align with company strategy and direction.
Most well-designed programs follow a formula that relies predominantly on pre-set objective financial metrics, a structure preferred by shareholders. Investors often frown on programs that rely heavily on committee discretion, especially when it exceeds about 20% of the weight for a company’s annual incentive program. Meanwhile, administrators frequently aim to use metrics in their long-term incentive plan that are distinct from the annual program and target long-term performance objectives with at least a three-year period as best practice.
Connecting Incentives and Performance
Once a board has identified appropriate metrics for its incentive programs, the next task is to set performance levels. Administrators can do this by incorporating statistical analysis into their goal-setting process with the aim of designing a program that strongly correlates incentive outcomes with expectations for performance objectives.
Using Statistical Analysis for Goal Setting
A goal-setting approach that incorporates statistical analysis can help ensure that goals are challenging but achievable and present payout opportunities that are fair for both executives and shareholders. By modeling incentive programs at the outset with statistical assumptions such as metric growth rate, volatility and correlations, companies can better align potential performance and payout levels according to their probabilities of achievement.
Monte Carlo simulation is a common statistical framework that can be used for such purposes. In short, this method aims to form a distribution of probable outcomes through a random sampling of scenarios (typically creating up to 10,000 data points), based on a set of assumptions. Using Monte Carlo analysis to simulate an incentive award program can provide valuable insights into whether targets are sufficiently rigorous and whether threshold and maximum payout levels are commensurate with their performance expectations.
Reliable Assumptions for Accurate Modeling
Modeling for an award program will only produce useful results if based on reliable assumptions. A simulation might include a growth rate assumption to integrate a metric’s rate of change. Volatility assumptions may be used to predict the variation of metric performance outcomes. Correlation assumptions may be used to account for how relationships between metrics and relative performance measures interact and affect scenario results.
Combined, these assumptions, when recreated and repeated over a multitude of scenarios, will form a distribution of probable outcomes, which can be used to determine a payout probability curve for analysis. After establishing a baseline, however, modeling assumptions should be stress-tested according to company-specific situations or economic uncertainties.
Common Pitfalls in Program Design
-
- Performance levels corresponding to higher payout probabilities may suggest “softball” goals and indicate a need for renewed attention to goal rigor.
- If payout probabilities for threshold and maximum are set in a tight band around target relative to the variance, metrics will lack adequate leverage and result in a “feast or famine” scenario for executives. A higher probability of maximum payout corresponds to a higher probability of no payout at all.
- While threshold and target performance levels may appear reasonable, a quick escalation in payout probability to maximum may signal goals are not appropriately balanced.
Meeting Investor Expectations Through Disclosure
Some companies, through insufficient disclosures, fail to demonstrate to investors that their programs are sufficiently rigorous. If forward-looking goals are disclosed, shareholders and proxy advisors may seek to understand the level of goal-setting rigor by comparing them with prior year achievement levels or targets as a baseline. However, many companies withhold forward-looking goals to protect this information from competitors, which can hinder close analysis.
There are resources available to compensation committees that can help them create a strong data-driven framework, such as benchmarking and statistical modeling analysis tools, including Monte Carlo simulations. Developing and working within such a framework will greatly ensure goal-setting outcomes that are both achievable and rigorous, and providing clear and forthright disclosure may aid in justifying metric and goal-setting decisions. ISS-Corporate’s Incentive Benchmarking and Award Simulator tools were developed to address some of these challenges.