Recent commentary has suggested the use of various risk-adjusted measures in assessing executive performance. It has been argued such measures provide better incentive alignment than the widely used Relative Total Shareholder Return (TSR) performance measure. In particular, adjusting relative TSR for risk will provide fairer rewards aligned with shareholder interests.
Most of the suggested risk-adjusted performance measures incorporate the company’s Beta (the sensitivity of the company’s share price movements to the market as a whole). Suggestions have included subtracting the cost of equity capital (on a retrospective basis) from the company’s TSR and adjusting the TSR for a measure of the company’s systemic risk. The former approach implicitly extracts the beta of the company (generally used to calculate cost of equity) from the performance analysis whilst the latter typically explicitly adjusts for the company’s Beta.
The above risk-adjusted performance measures are subjected to the significant limitations arising from the use of the company-specific Beta in the analysis. These limitations have been widely considered in academic literature and centre upon the appropriateness of Beta as a proxy for risk (as opposed to price volatility).
Specifically, beta is measured on the basis of historical returns and assuming the pattern of historical returns will repeat in the future. Further, it is assumed the pattern of historical returns will repeat in the same proportion and in the same order. More generally, Beta is calculated assuming strong form market efficiency and observations from an equilibrium market. New information is not reliably incorporated into the measurement of a company’s Beta.
We have previously incorporated risk-adjusted performance measures into the construction of Long-Term Incentive (LTI) plans. In our experience, adjusting for the risk associated with the performance measures (typically TSR) does not necessarily contribute to better aligning the relevant LTI plan.
In our opinion, the efficacy of a LTI plan continues to be heavily dependent on the selection of the most appropriate comparator group as part of a Relative TSR performance measurement. This traditional approach provides a degree of objectivity, is easily understood and not over layered with complexity or the manipulation of input variables to determine a historic Beta nor seen to be judging a company’s relative performance, and, accordingly, effective in driving desired executive behaviour.