Lauren J. Harris - Bloomberg Project

Growing up, one of the first "facts" I learned about the economy was that women CEOs were outnumbered by CEOs named John. For an in-class project on CEO pay using Compustat Execucomp data, I decided to revisit this pattern, expecting to find that little had changed. Instead, I found that female CEOs now outnumber Johns (and John alternatives), are paid more than Johns and John alternatives, and that the distribution of female CEO pay increasingly looks like the male distribution.

Compustat Execucomp data post-1994 includes the full S&P 1500 as well as companies that used to be part of the S&P 1500 and other large companies. I exclude firms with multiple CEOs listed in a given fiscal year (86 out of 52,898 observations). This drops eight women from the sample, both in the last two years, and one John (and no alternatives) in 2015. This allows me to analyze the percentage of firms in a given fiscal year with a CEO that’s either a woman or a John/Jon/Jonathan.

There are three main takeaways from these plots: 1) the share of women who are CEOs in the S&P 1500 is extremely low, 2) women began outnumbering Johns in 2016 (and Jons/Jonathans in 2018), and 3) while the share of women CEOs has been almost monotonically increasing since 1994, the share of Johns is decreasing (although less rapidly than the share of women has increased). This is due in part to the fact that male CEO names have become more diverse with time.

Of course, this is something of an absurd exercise. The fact that female CEOs now outnumber one small subset of male CEOS does not imply that women CEOs are any better off. To understand this issue a bit more, I compare how female CEOs are paid relative to the Johns/Jons/Jonathans. I follow the Compustat Execucomp definition of CEO pay: before 2006, pay is defined as the real value of a CEO's salary, bonus, restricted stock grants, option exercise value, and long term incentive plan in thousands of dollars. After 2006, restricted stock grants are replaced by the fair value of stock awards. Note that none of the following pay analyses include controls, and all pay is reported in 2012 dollars.

Comparing average female CEO pay with average John/Jon/Jonathan CEO pay reveals that on average, female CEOs are paid more than John/Jon/Jonathan CEOs. This gap has also closed at the 75th and 90th percentiles. The spike in average female CEO pay in 1998 is due almost entirely to Linda Wachner, whose pay in that year was the subject of much discussion. Winsorizing the top and bottom 10% of observations (so Linda's pay would be recoded as the 90th percentile of the pay distribution among CEOs in all years, for example) yields the following plot:

I also compare female CEO pay to male CEO pay in general over the last eleven years. The 25th and 50th percentiles of the male and female CEO pay distributions roughly track eachother over time. The 75th percentile of the female pay distribution has overtaken the 75th percentile of the male pay distribution, and the 90th percentile of the male pay distribution exceeds the 90th percentile of the female pay distribution.