Research Report | Dec 2019
Overlap of Executive Incentive Plan Performance Measures: Is the Concern Warranted?
Our statistical analyses of the data uncovers whether or not the increase in proxy advisor scrutiny on “redundant” incentive plan measures is justified.
The selection of goals and metrics related to incentive plans is a core element of effective executive compensation plan design. Pearl Meyer advises clients to identify their unique combination of goals and metrics that will drive value creation for the organization, both in the short term and in the long term. In some cases, that might necessitate the use of one particular measure across multiple time horizons.
However, over the past six to seven years, there has been an increase in scrutiny from both major proxy advisors, Glass Lewis and ISS, when the same performance measures are used in both short- and long-term incentive plans. There is some concern that this practice may allow for a high level of pay-out (or lack thereof) for performance against similar metrics or may overly focus executives on a single dimension of performance.
Given the criticality of performance measure selection in compensation plan design and the absence of data on the issue from the proxy advisors, we felt it was important to understand if this idea of “redundant metrics” or “performance measure overlap” is in fact a problem or a fallacy.
Relying on data from Main Data Group and working with the Institute for Compensation Studies (ICS) at Cornell University’s Industrial & Labor Relations School, we set out to answer whether or not investors should be concerned with what we refer to as performance measurement overlap. At a high level, this research study focuses on four specific questions:
- How has performance measurement overlap changed over time?
- Is there any difference in pay based on the use of overlapping performance measures?
- How much pay is actually tied to overlapping performance measures?
- Are there any differences in firm performance based on the use of overlapping performance measures?
Our hypothesis was two-fold: first, that statistical modeling would not support the proxy advisors’ position that this practice is potentially problematic and second, that we would find no statistically significant difference in firm performance for companies who used overlapping performance measures.