The SEC has released proposed rules for disclosing pay for performance, based on the Dodd-Frank law requiring that each company provide “information that shows the relationship between executive compensation actually paid and the financial performance of the issuer, taking into account any change in the value of the shares of stock and dividends of the issuer and any distributions.”
As my regular readers know, “pay for performance” can be usefully evaluated only when we have appropriate definitions of “pay” and “performance.” Appropriate definitions depend on one’s purpose for making the evaluation. Pay-for-performance can be either an exercise in costing, i.e., seeing if shareholders are getting what they are paying for, or for determining alignment, i.e., seeing if management rewards are consistent with shareholder value creation. These can be different analyses, with the same underlying figures yielding very different conclusions.
Dodd-Frank pretty much dictates that “performance” will be measured as total shareholder returns (TSR). That may be reasonable as long as we are looking at (a) returns over a long enough period of time, or (b) our company’s returns relative to those of our sector peers. Both of these conditions are more or less contemplated by the proposed rule.
The proposed SEC definition of “pay” is closer to what one would properly associate with measuring alignment, but with some important differences.
If we wanted to determine the degree to which an executive’s financial interests are aligned with those of her shareholders, we would look at the total company-derived compensation realizable by the executive. This would equal:
- Actual cash compensation received, plus
- The change in the realizable value of vested and vesting equity, plus
- The net change in value of any defined benefit plan(s).
In contrast, this is how the SEC is proposing to define “pay”:
- Actual cash compensation received, plus
- The realizable value of vesting equity, plus
- Service cost of defined benefit plans (e.g., pensions).
The difference between these respective items #2 can be illustrated as follows: Consider a situation where an executive, during a difficult year for the stock, vested in $1 million worth of new shares, but saw the value of his existing company shares (i.e., those vested before the beginning of the year) drop by $1 million. At the end of the year, when the executive was looking at his personal net worth, he would not feel one dollar wealthier (in terms of equity) after a difficult year of work. That might reflect a reasonable alignment between the executive and his shareholders. Under the proposed SEC definition of “pay,” the executive would look $1 million richer, which may be difficult for investors, if unaware of the executive’s net worth loss, to reconcile with a drop in the stock price.
This ‘underlying portfolio’ effect becomes much less pronounced the longer the period we are reviewing. Over the full tenure of the executive, including any up front equity awards, this effect would disappear, and we could reliably track their alignment.
Relating to the items #3, the SEC proposes to include the company’s cost of defined benefits (i.e., the service cost) in a measure otherwise tracking the executive’s realizable pay. This is in odd contrast to total compensation in the Summary Compensation Table in the proxy statement whereby the SEC mandates the change in the realizable value of benefits in a measure otherwise tracking the compensation cost to the company. I guess that’s what happens when rules are made via a political tug of war between competing interests instead of via mathematically clear distinctions regarding purpose and effect. But this confusion over defined benefits is small potatoes in most years.
As we can see, the real value of the SEC proposed pay-for-performance disclosure rules will depend heavily on the period over which “actual pay” is compared to TSR performance. The proposed rule would mandate that companies ultimately track pay and performance over five years, which research suggests is the minimum period over which such a comparison is likely to indicate a valid relationship. It is, in fact, quite possible to see discontinuities between pay and performance year-by-year, and perhaps every single year, yet see pretty good alignment over a five year period.
Interestingly, in contrast to the proposed SEC performance standard of TSR, Bloomberg announced just last month that they are using a completely different measure of company performance against which to compare pay—Economic Profit (EP). Bloomberg’s rationale is that EP is a better focus for management than TSR, and therefore what companies ought to be paying for. In theory, they’re right. In practice–especially a practice with significant legal implications–TSR provides and objective, uniform comparability across companies and time, while EP must be tailored to each sector, and ought to adapt to changes in one’s business model over time.
The SEC proposal at least provides a better definition of pay, and therefore a better standard in evaluating corporate pay for performance, than the existing standards of proxy advisors like ISS. (Yes, that is a really low bar.) I will submit my comment letter to the SEC to improve its “pay” measure, but am doubtful that I will prevail upon them to incorporate a definition of realizable pay consistent with what was outlined above, given that most most issuers will consider my suggestion even more complicated than what the SEC is proposing, most compensation consultants won’t understand the distinction upon which my suggestion is made, and most executive pay critics don’t care because they remain fixated on costs rather than alignment.
At the end of the day, I don’t think the SEC’s rejection of my suggestion will matter much. All of the compensation information that sophisticated investors needs to know is already in the disclosures. This is just another exercise in handholding by the SEC aimed at less sophisticated investors looking for a magic screen, or another point of outrage. One would think that if someone is going to take a neophyte by the hand, they should at least know where they are going. Alas, that is not the world of public company governance at the federal level.