The MSM enthusiastically distributes the annual AFL-CIO study comparing big company CEO pay with all workers — informally the “400 to 1” comparison. But the dishonesty of the labor union report is obvious, breathtaking — and completely ignored. Below is a detailed debunking of this “study.”
Perhaps the biggest “error” (it’s no error!) is the inclusion of part-time workers in the worker average pay. Another factor is that the CEO’s “pay” includes all their benefits, but the compared workers’ pay is only their paycheck — the substantial benefits are ignored. While the union is intentionally vague about what they count as worker pay, it appears that they don’t count overtime wages — even though every CEO in America works way more than 40 hours a week. The report doesn’t compare CEO pay with workers in the same company — such a comparison reduces the disparity significantly. And finally, this is a cherrypicked CEO group — the TOP CEO’s are used to represent all the CEO’s. Indeed, when you count all the CEO’s, then the average compensation ratio drops to under 4 to 1!
American Enterprise Institute
Lots of problems with AFL-CIO’s ‘CEO-to-worker pay’ analysis including small sample size and using part-time worker pay
The AFL-CIO has come out with its annual exaggerated and distorted CEO-to-worker pay comparison, which this year is a whopping 373-to-1 using some questionable statistical methods. As I reported recently on CD here, a comparison of all US “chief executives” to all US workers produces a CEO-to-worker pay gap of less than 4 (only 3.83 in 2014). Below is a review of that particular issue, along with some other statistical problems with the AFL-CIO’s comparison of CEO-to-worker pay.
1. Extremely Small Sample Size. The AFL-CIO compares a small number of the highest-paid CEOs who head some of the world’s largest multi-national corporations that are among companies in the S&P 500. This year, it’s not even clear how many CEOs are part of the AFL-CIO’s analysis, notice the ambiguity here (emphasis added):
In 2014, CEOs of the S&P 500 Index companies received, on average, $13.5 million in total compensation—an increase of 15.6% from the previous year—according to the AFL-CIO’s analysis of available data.
Last year, the AFL-CIO reported an average CEO pay (total compensation) of $11.7 million for 2013 based on “available data from 350 companies.” This year, the AFL-CIO fails to mention the exact number of companies it is using to determine that average CEO compensation in 2014 was $13.5 million.
And as I have reported before, BLS data show that there are actually 246,240 “chief executives” in the US, who earned an average annual salary of $180,700 in 2014. Using the BLS’s more narrow job classification “chief executives of companies and enterprises,” the 21,550 CEOs in that group earned $216,100 in 2014, an annual increase of only 1.3% from 2013 and far below the 15.6% increase in “CEO pay” reported by the AFL-CIO for 2014.
Using the BLS average of $216,100 for all American CEOs of “companies and enterprises” in 2014 (less by the way than the base salary of $272,00 salary for AFL-CIO chief executive Richard Trumka), and the AFL-CIO’s estimate of worker pay ($36,134 in 2014) would produce a “CEO-to-worker pay” ratio of only 6-to-1, far, far below the AFL-CIO’s 373-to-1 biased ratio using only the top 1-2% of the highest paid CEOs in the US.
2. AFL-CIO Compares Total CEO Compensation to Workers’ Cash Wages. The AFL-CIO compares average total CEOcompensation ($13.5 million in 2014) for a very small sample of the highest paid chief executives to the base cash wages of the average worker ($36,134 in 2014), not the total compensation for the average worker. That is, the AFL-CIO reports a “CEO-to-worker pay” ratio, which is actually a “CEO total compensation to worker base pay” ratio and not a “CEO cash pay to worker cash pay” or a “CEO total compensation to total worker compensation.”
The AFL-CIO’s measure of total CEO compensation includes base salary, plus any payments for bonuses and profit-sharing, the value of any deferred compensation in the form of stock or option awards, retirement or pension benefits, and other non-cash benefits like health care, personal use of company cars and airplanes, country club memberships, etc. On the other hand, the AFL-CIO only considers monetary wages for workers, and ignores the value of their fringe benefits (health care, employer contributions to 401(k) programs and other pension/retirement benefits, etc.) and the value of any profit-sharing or bonus payments. How significant are those non-wage benefits?
Consider that hourly autoworkers at Ford will receive $6,900 this year in profit-sharing payments, following even larger payments of $8,800 last year and $8,300 in 2013. GM workers could receive up to $9,000 in profit-sharing payments this year, following payments of $7,500 in 2014 and $6,800 in 2013. Autoworkers also receive generous fringe benefits that include education tuition assistance (up to $5,000 per year), book reimbursement, personal development assistance, retirement benefits, and health care benefits (including prescription drug coverage, $20 co-pay office visits, wig benefits, hearing aid coverage, dental care coverage, retiree health coverage, mental health care and substance use disorder treatment coverage, disability coverage, medical leave coverage, etc.). While UAW worker fringe benefits may be more generous than those received by the average US worker, it still illustrates the point that the AFL-CIO is making an “apples vs. oranges” comparison by considering total compensation for CEOs including every possible fringe benefit to only the cash wages of workers, while completely ignoring any worker fringe benefits.
3. AFL-CIO “Worker Pay” is for Part-Time Workers. To calculate the average annual cash-only payments of $36,134 for the “average rank-and-file worker,” the AFL-CIO use the average hourly pay of $20.60 in 2014 for Production and Nonsupervisory Employees and the average number of weekly work hours of 33.7 for those workers in 2014 ($20.60 per hour X 33.7 hours per week X 52 weeks = $36,100 per year). Note that the BLS considers weekly work hours below 35 as part-time status, and therefore the AFL-CIO is comparing the total compensation of CEOs working full-time to the cash wages of part-time workers!
4. AFL-CIO Doesn’t Compare CEO Compensation to Worker Pay at the Same Company. The AFL-CIO compares the average compensation of a few hundred CEOs to the average pay of about 100,000 “part-time” production and non-supervisory employees, while perhaps thousands of those workers don’t even work for a company headed by one of the few hundred CEOs in the AFL-CIO’s sample. Why not compare CEO compensation to the typical pay of employees who actually work at the CEO’s company? That comparison of “CEO-to-worker pay” is actually calculated by Pay Scale for the 100 largest publicly-traded US companies, as Jim Pethokoukis reported on the AEIdeas blog yesterday. The Pay Scale analysis also corrects for the flaw discussed above in Item #2 by comparing the “Total Cash Compensation” for CEOs to the “typical median worker pay” at the same corporation.
According to Pay Scale:
Data on CEO Salary was obtained from Equilar’s “100 CEO Pay Study,” an annual study of CEOs at the 100 largest U.S. publicly-traded companies. We summed Salary, Bonus, and Other compensation to get non-stock compensation. This measure is most similar to employee data from the PayScale website. We provide the overall typical median pay for workers at 100 companies, as well as the ratio of CEO-to-worker pay.
Based on the Pay Scale analysis for 2013 for the 100 largest US companies by revenue, the average CEO “total cash compensation” was about $5.5 million, the average pay for employees at those companies was $70,700, and therefore the average CEO-to-Worker Pay ratio according to Pay Scale’s analysis was only 84-to-1. The historgram above shows the distribution of CEO-to-Worker pay ratios for those 100 companies, which range from as low as 6-to-1 for Hewlett-Packard to a high of 422-to-1 for CVS. Part of that difference is explained by the fact that the typical worker at HP makes $84,500 while the average CVS worker makes only $28,700. As Jim pointed out, only five companies had ratios over 200-to-1 in 2013, and there were only two companies over 300-to-1 (Goodyear and CVS). The historgram also shows the average of 84-to-1 is influenced by a few outliers on the high end, which is corrected by calculating the median CEO-to-worker pay ratio of only 70.2-to-1 (half of the companies are above the median, and half below) for the companies in the Pay Scale database for 2013.
Bottom Line: To summarize the problems with the AFL-CIO’s “CEO-to-worker pay” ratio:
a. It is based on a small sample of companies that represent only 1-2% of all companies with CEOs.
b. It includes bonuses and all forms of non-cash compensation and fringe benefits for CEOs, while ignoring those forms of compensation for employees.
c. It compares CEOs working full-time to part-time “rank-and-file” workers (33.7 average weekly work hours).
d. It doesn’t compare CEO compensation with employee pay for workers at the same company.
Given all of these statistical flaws (the biggest I think being that the AFL-CIO’s average worker pay is actually for part-time workers), it’s somewhat troubling that the media seems to mostly blindly accept and report the AFL-CIO’s annual distorted and exaggerated “CEO-to-worker pay” ratio without ever questioning the analysis. One exception I could find is this WSJ article “Top CEOs Make 373 Times the Average U.S. Worker” that points out Item #2 above about comparing CEO total compensation to cash wages for employees. And as I have pointed out many times, a comparison of the average pay of ALL CEOs to the average pay for ALL full-time workers is less than 4-to-1, nowhere close to the nearly 400-to-1 figure used by the AFL-CIO (and the media). Call it a “statistical falsehood-to-truth ratio” of almost 100 times for the AFL-CIO’s CEO-to-worker pay ratio!