Disclosed: The Pay Gap Between CEOs and Employees

Nearly three years after Congress ordered public companies to reveal their chief executive officer-to-worker pay ratios under the Dodd-Frank law, the numbers still aren�ft public. The provision was included to deter excessive compensation schemes that, in the words of U.S. Senator Robert Menendez (D-N.J.), �gwere part of the fuel that led to the financial collapse.�h Since then, the requirement has been parked at the Securities and Exchange Commission, which must develop a rule on how to calculate and report the ratio. Questions remain: Do companies have to determine their median employee compensation by an actual count or would statistical sampling suffice? How should global companies reconcile differences in wages and benefits from one country to the next? For that matter, how should investors interpret differences in compensation across industries?

Those who oppose publishing this ratio have seized on some of these questions to argue that the requirement be dropped. �gWe don�ft believe the information would be material to investors in making investment decisions,�h says Tim Bartl, president of the Center on Executive Compensation, the advocacy arm of a Washington nonprofit called the HR Policy Association.

To get a sense of what such ratios could reveal, we conducted an experiment. It compared the disclosed CEO compensation mandated by the SEC—including salary, bonus, perks, changes in pension accruals, and the value of stock-based awards—with U.S. government data on average worker pay and benefits by industry. (Most companies don�ft disclose actual payroll information for employees.)

In addition to using the industry-specific averages for workers�f compensation, this ratio differs from what Dodd-Frank requires in at least one other respect: It compares CEO pay with the average for all rank-and-file employees in the U.S., while the law calls for using the median of all employees worldwide, including executives other than the CEO.

Others who�fve calculated pay ratios, such as the AFL-CIO, didn�ft differentiate worker pay by industry or include employee benefits in their math. Bloomberg News did, which tended to make the ratios smaller. (The AFL-CIO�fs average CEO-to-worker multiple at big U.S. companies is 357. Bloomberg�fs average ratio for Standard & Poor�fs 500 companies is 204; the average of the top 100 companies on our table is 495. That is, CEOs of the companies on that table averaged 495 times the income of nonsupervisory workers in their industries.) There�fs no question that using industrywide averages as the denominators is not a perfect substitute for the real pay ratios Dodd-Frank calls for. If you�fre a fast-food chain CEO who pays line workers well above minimum wage plus full health benefits, your ratio would still have the same low denominator as the skinflint chain that pays only the minimum.

Every company on the list was asked to comment on the ratio—and to provide their own. Only one in the top 100 came up with a number: Wynn Resorts (WYNN), which says its ratio is 251. �gThe outdated and incorrect figures being used, together with a flawed methodology, results in a distortion that is insulting to our employees,�h Hugh Burns, a spokesman for Simon Property Group (SPG), said in an e-mail. (Simon Property is No. 3 on the Bloomberg list of the largest ratios, and CEO David Simon�fs $137.2 million in compensation for 2011 was 1,594 times what the average �gfunds, trust, and other financial vehicles�h worker is paid.) Noting that the pay reported last year is contingent on years of future performance, Burns said, �gThe survey creates a completely misleading result that grossly overstates and inaccurately portrays David Simon�fs compensation and makes any comparison meaningless.�h

The SEC has yet to set a deadline for the rule that would make pay-ratio disclosure mandatory. Commissioner Luis Aguilar, a Democrat, suggested publicly in February that companies should voluntarily disclose their ratios until the agency acts. The other four commissioners, including Chairman Mary Jo White, who took office in April, declined to comment. Representative Bill Huizenga, a Michigan Republican, has introduced language to repeal the disclosure requirement. The ratio �gdoesn�ft do anything other than play politics,�h he said.

Peter Drucker, the celebrated management theorist, certainly thought the CEO-to-rank-and-file multiple mattered. Starting with a 1977 article and until his death in 2005, Drucker considered 25-to-1 or even 20-to-1 the appropriate limit. Beyond that, he indicated, it�fs bad for business. In his view, excessively high multiples undermine teamwork and promote a winner-takes-all, �gdid-it-because-I-could�h culture that�fs poison to a company�fs long-term health. �gI�fm not talking about the bitter feelings of the people on the plant floor,�h Drucker told a reporter in 2004. �gThey�fre convinced that their bosses are crooks anyway.�h He meant the people in middle management who become �gincredibly disillusioned�h by runaway CEO compensation. On big executive payouts that coincide with layoffs, Drucker was unequivocal. That, he said, was �gmorally unforgivable.�h

To view the Top CEO Pay Ratio chart, visit http://go.bloomberg.com/multimedia/ceo-pay-ratio/.