Thoughts on Metrics and Incentives
Thoughts on Metrics and Incentives first appeared at The Activist Investor.
A brief meditation on motivating, measuring, and rewarding executive performance.
Metrics have been in the news lately:
❖Sensational accounts of how share repurchases boost EPS to benefit CEOs
❖Bennett Stewart promoting his Corporate Performance Index (CPI)
❖Corporations futzing with GAAP accounting, specifically EBITDA, to present great results.
Let’s consider the metric alphabet soup, then.
❖EPS: Earnings per Share, duh. Accounting profit divided by number of outstanding shares.
❖EBBS: Earnings Before Bad Stuff. EPS without expenses that management doesn’t like, the zenith of futzing.
❖EBITDA: Earnings Before Interest, Tax, Depreciation, and Amortization. A customary measure of operating cash flow, but based on accounting profit.
❖Adjusted EBITDA: see EBBS, call it AEBITDA
❖ROI: Return on Investment, with whatever measure of return and investment the company chooses. Highly futz-able.
❖TSR: Total Shareholder Return. Change in share price, plus any cash to shareholders as dividends. Can’t really futz with it.
❖CPI: Corporate Performance Index. The new metric, based on EVA (Economic Value Added).
How to make sense of all this in the context of recent news accounts?
For as long as investors have monitored EPS andEBITDA, companies have tried to massage it into EBBSor AEBITDA. GAAP accounting is rife with judgement, so management will seek to influence (futz with) EPS and EBITDA in subtle ways, or just dispense with it and use EBBS and AEBITDA.
Investors also know that EPS measures mostly the returns part of ROI. We also want to know the investment part.
Bennett Stewart years ago gave voice to these two concerns with EVA. It deals with the two problems ofEPS, EBBS, EBITDA, and AEBITDA: management can futz with accounting results, and thinks capital investment comes free of charge. He spent decades trying to persuade companies and investors that EVAimproves on these other metrics.
We don’t know why Stewart created CPI, which starts with EVA. It seems like he wanted something similar to but better than TSR in exec comp packages.
Many exec comp packages reward EPS or change inEPS. Lately they also reward TSR. Neither idea makes any sense.
Basic economics, and indeed cognitive and behavioral science, finds that one designs incentives to elicit the behavior one desires, or to discourage behavior one doesn’t. In this instance, exec comp incentives should pertain directly to decisions and other actions that executives can influence and control.
Executives don’t influence and control share price. TSR measures mostly share price.
On the other hand, executives control the metrics EPS,EBBS, EBITDA, and AEBITDA, in addition to controlling the decisions and other actions whose outcomes these metrics measure. That won’t work.
More generally, exec comp programs should use metrics that measure company performance, not investment performance. TSR makes sense for a PM, but not for a CEO. EVA or maybe CPI makes sense for a CEO. EPSmakes no sense for anyone.
Critics can object to share repurchases that boost exec comp. Let’s improve exec comp and the underlying metrics - reward and punish CEO decisions and other actions, and make it hard to futz with the metrics. Leave share repurchases alone.