September 24, 2013

Dodd-Frank Strikes Again

New regulations on CEO compensation disclosure are both costly and inane.

This past week, the Securities and Exchange Commission announced that it is preparing a full new set of regulations on CEO compensation that are in keeping with the Dodd-Frank law. The SEC’s report on the matter puts three issues on the table. First, the firm must disclose “the median of the annual total compensation of all employees of an issuer.” Second, it must disclose the annual “total compensation” of the CEO. Finally, it must state the ratio of the compensation of the median employee to that of the CEO.

epstein
Illustration by Barbara Kelley

These regulations are the brainchild of Senator Robert Menendez, a New Jersey Democrat, who believes that investors deserve to know “whether public companies' pay practices are fair to their average employees, especially compared to their highly compensated CEOs." The seductive search for “fairness” is, however, both mischievous in its choice of ends, and perverse in its selection of means.

The first question to raise in evaluating regulations generally is whether their benefits to firm value are greater than the costs, both public and private, of compliance. At no point did Senator Menendez or anyone else examine this question to see if the benefits of the regulation outweigh its costs. The answer is that they do not. By any rational measure, this part of Dodd-Frank should be struck down as unconstitutional.

The Dangers of Disclosure

The first point to note about any new SEC regulation is that it will be piled on top of other corporate disclosure obligations, including those that already deal with the thorny issue of executive compensation. On this score, current SEC regulations require the pay disclosure of five persons: the CEO, the CFO, and three of “certain other high-ranking executive officers” of each public company.

The gains from that provision are murky at best. If this information were valuable to investors and helped them to better value the corporation, we should expect firms to make voluntary disclosures in response to market pressures. But this 2006 rule was driven by forces inside government, and the 370-page SEC report explaining these requirements is long on detail, but utterly devoid of any justification for the rule.

The SEC’s more recent report is far worse. Consider the Wall Street Journal account that identifies the champions of the new rule—labor unions and political activists—who claim that the rule will allow investors to make “better decisions” by increasing “investors’ sensitivity” to CEO wage increases. Translation: using one big number is an easy way to inflame public opinion against the firm in a thousand different political, bargaining, and regulatory settings. Remember the AFL-CIO and other union groups don’t use this public information to drive up the market value of a company’s shares. Instead, they use it to unionize a workforce or to gain bargaining leverage against the target firm.

The ever-present misuse of sensitive information drives the creation of the modern law of trade secrets. In countless contexts, privatefirms are careful to restrict the parties to whom they give sensitive information, and to limit the ways it can be used. Nondisclosure agreements are standard features in most business settings, born of the powerful awareness that information about the internal structure and operation of any business must never fall into the hands of the firms’ competitors, suppliers, and customers. Confidentiality agreements allow the selective disclosure of firm information to advance the firm’s position, without letting outsiders use that information to the detriment of the firm.

One reason public disclosure in all investment contexts is a dangerous tool is that it undermines the salutary practice of selective disclosure for limited use. Once the information is given publicly to the investors, it is necessarily given as well to those whose interests are adverse to the firm. Any evaluation of the net benefits of the disclosed information requires a comparison of the benefit that shareholders receive against the harm that they suffer when that information is used by others.

It therefore follows that rational investors frequently would rather not receive information about their firm if it comes at the price of having it fall, as it were, into enemy hands. It is therefore perfectly rational for shareholders in a pharmaceutical company to not want to know the particular lines of company research, which could direct its competitors to the same area. Nor do shareholders want their firm to disclose the location of valuable deposits of oil or copper, if it lets competitors bid on nearby properties.

It is just this logic that persuaded the First Circuit Court of Appeals in Phillip Morris v. Reilly in 2002 to strike down, as a taking of private property (e.g. trade secrets), Massachusetts legislation that required tobacco companies to publish a list of tobacco additives in the order of their relative concentration. The partial disclosure of this trade secret is worth a lot more to competitors than to smokers, who know how cigarettes taste by smoking them.

Medians versus Means

The second reason why these disclosure ratios are useless is that they measure the wrong thing. As Gretchen Morgenson wrote recently in the New York Times, references to CEOs of peer firms often count as “a better way” to measure CEO pay. Worse still, it is unwise for an internal comparison to focus on the compensation of the median employee. The bizarre statutory pay ratio does not correlate with any known measure of corporate efficiency. It would be far better for the company to disclose the ratio of the average,or mean, employee compensation against the CEO’s. The mean can be calculated for wages and salaries by dividing the total compensation paid by the number of employees—all of which is available right now from public information.

There is no valuable information that requires the use of medians relative to means. And once we look at the general trend, there is really no pressing need for further disclosure. It is well known that the gap between CEO pay and that of the average worker has grown since 1970, but has shrunk since about 2000. That ratio was about 100 to 1 in 1990, spiked to about 550 to 1 in 2000, and now stands at about 300 to 1.

The same ratios can be compiled from public data for every firm. Using the median actually gives less information because it focuses on the compensation of one employee in the middle of the distribution, without giving any indication of any movement of total compensation either above or below that number, or why those movements matter.

Thus a nationwide company in the restaurant business that hires a large number of low-wage employees is likely to have a very low median wage. Its astute CEO should be praised if he can expand the rate of new employees who might otherwise not get jobs elsewhere. It doesn’t make sense to act as if the higher ratio of total CEO compensation to the median compensation indicates corporate misconduct.

In other situations, CEO compensation can be highly variable over years, for often a board will award additional compensation in cash, stock, and options to reflect the superior performance over past years. The SEC ratio could easily vary extensively from year to year. One sensible modification of the basic position could be to run CEO compensation over multiple years. But that alternative raises questions of its own if the company hires a new CEO or acquires or divests of some division, at either home or abroad. There is little reason to refine this measure—and every reason to scrap it. The analysts can do just fine with available data.

Compliance Costs

The final piece of the cost/benefit analysis requires a closer look at the extensive compliance costs these new regulations, most of which are not apparent to the naked eye. There are some modest difficulties in determining CEO total compensation with both stock options and fringe benefits.  But the real challenge is figuring out both median wage and median benefits for firms with tens of thousands of employees spread across the globe. The SEC allows firms some degree of flexibility in designing, for example, their sampling procedures. But that concession only makes cross-firm comparisons less useful than before. Nor, until the rubber hits the road, does anyone know just how flexible the SEC will be for major multi-national firms that use multiple currencies and that must comply with a wide range national and local labor regulations.

Yet, to determine that elusive median, the SEC contents itself with the pious observation that “an appropriate and most cost effective methodology would necessarily depend on a registrant’s particular facts and circumstances,” which ushers in a laundry list of issues including workforce composition, firm complexity, pay stratification, and the different tax, accounting and payroll schemes used world-wide under unspecified sampling protocols. While it is easy to figure out the total amounts paid, finding the median benefits package must monetize the full range of healthcare, pension, vacation, and other benefits routinely supplied.

A Constitutional Escape?

The unwise selection of the median total compensation benchmark is yet another mindless feature of Dodd-Frank bill that will be with us for the foreseeable future. Both Congress and our administrative agencies never think hard about the costs and consequences of regulations before laws are enacted. All too often, the proponent of some legislative gimmick, like Senator Menendez with pay ratios, casually observes that additional disclosure might generate some benefit, and then lets the agencies to struggle with how to implement that off-the-cuff statutory scheme.

Unfortunately, the courts eagerly defer to the will of Congress and the (supposed) expertise of the SEC. This judicial buck-passing has got to stop. The new regulations could cost billions and yet provide no useful public information, in light of the massive amounts of information already available. Surely, the median income requirement is so arbitrary and irrational that the federal courts need not show deference in the face of a regulatory train wreck.  If it passes constitutional muster, the champions of regulation will win and everyone else will lose. It’s time to bring this downward cycle to a halt.


Richard A. Epstein, Peter and Kirsten Bedford Senior Fellow at the Hoover Institution, Laurence A. Tisch Professor of Law at New York University, and senior lecturer at the University of Chicago, researches and writes on a broad range of constitutional, economic, historical, and philosophical subjects. He has taught administrative law, antitrust law, communications law, constitutional law, corporate law, criminal law, employment discrimination law, environmental law, food and drug law, health law, labor law, Roman law, real estate development and finance, and individual and corporate taxation. His publications cover an equally broad range of topics. His most recent book, published in 2013, is The Classical Liberal Constitution: The Uncertain Quest for Limited Government (2013). He is a past editor of the Journal of Legal Studies (1981–91) and the Journal of Law and Economics (1991–2001).


Letters to the editor may be sent to definingideas@stanford.edu. Editors reserve the right to reject or publish (and edit) letters.