Cognitive Dissonance Man and the Compensation Data Collection Tool

I'm sure you've heard by now that OFCCP published the Advanced Notice of Proposed Rulemaking on the compensation data collection tool (If you haven’t seen the ANPRM, you can read it here, or head over to Carla Irwin’s blog, AA Insights - she’s got a great summary of the questions on which the OFCCP is seeking comment).

According to the ANPRM, the purpose of the new tool is "to provide insight into potential problems of pay discrimination by contractors that warrant further review or evaluation by OFCCP or contractor self-audit."

OFCCP should be commended for its efforts at creating equal opportunity workplaces. Undoubtedly, there are real cases of compensation discrimination and OFCCP can - and should - play a role in ending discrimination. On its face, the compensation data collection tool appears to be one way to do this. But a closer reading of the Advanced Notice leads me to believe OFCCP has been visited by Cognitive Dissonance Man - they're trying to do the right thing, but for the wrong reasons:

Women still earn only 77 cents for each dollar earned by a man... Potentially non-discriminatory factors can explain some of these differences. Even so, after controlling for differences in skills and job characteristics, women still earn less than men. Some scholars find that these differences can be explained, to some extent, by differences in education and prior labor market experience. Others identify job segregation as an important cause of the pay gap. Ultimately, the research literature still finds that an unexplained gap exists even after accounting for potential explanations. Moreover, research literature finds that the narrowing of the pay gap has slowed since the 1980s. To the extent that these factors, such as type of job or amount of continuous labor market experience, are also influenced by discrimination, the "unexplained" difference may understate the true effect of discrimination.

Even though OFCCP acknowledges the vast body of research that indicates occupation, industry, labor market experience, and other factors that explain a large portion of the gender pay gap, the Advanced Notice seems to dismiss the research. The Advanced Notice also suggests that some of these explanatory factors themselves, such as continuous labor market experience and occupational choice, are influenced by discrimination. 

Let's think about continuous labor market experience. Research shows that compared to men, women tend to experience greater periods of absence from the labor market. Part of this is due to physiology. Right now, women are the only ones who can become pregnant and give birth (advances in medical technology may change that in the future, but let's concentrate on what we know to be true today). Most women prefer not to give birth underneath their desks at the office, so they take time off from work to have their babies in more appropriate surroundings.

Physiology - not discrimination - explains why women have greater career interruptions for childbirth, barring discriminatory intent on the part of God, evolution, the aliens who engineered us, or however else you think we ended up here in our current Human form. More data, increased regulation and stiffer penalties won't change this, nor will it change the fact that compensation and labor market experience are positively correlated. 

Women also have greater periods of absence from the labor market because of care-giving responsibilities. Whether right or wrong, women in American society spend more time caring for children, ill or disabled family members, elderly parents, etc. This might be because of discrimination at the societal level and our cultural norms about the roles women should play. Should employers be held liable for discrimination in cultural norms?

A similar argument can be made for differences in occupational or industry choices. While it's true that "back in the day" women's choices were limited - women could be nurses but not doctors, teachers but not administrators, and secretaries not CEOs. Occupational segregation does still exist, but not to the extent that we once saw. Women have come a long way, baby.

Arguing that women's occupational choices are pre-determined because of discrimination fails to recognize the achievements of hundreds of women who are not only working but succeeding in historically male-dominated occupations. It also undermines the validity of the choices of those women who want to work in historically "female" occupations.

Compensation discrimination is a real problem that needs a real solution.  The compensation data collection tool would provide OFCCP with an enormous volume of data which could be used to detect patterns of discrimination. But I think we're looking in the wrong places. We've been studying compensation data for years now, and still have not been able to successfully predict or prevent pay discrimination.

We need to take a new approach and start looking at the "non-metrics" side of things- things like compensation expectations, willingness to engage in compensation negotiations, the cash/benefits trade-off, and the reasons why people make the career choices they do. This avenue is a lot more difficult to study, but I think it's an avenue that will pay off in terms of really understanding differences in compensation by gender and race. Doing the right thing for the wrong reasons won't give us new insights into an old problem.

Image courtesy of Forest For the Trees Blog

Stephanie R. Thomas is an economic and statistical consultant specializing in EEO issues and employment litigation risk management. Since 1999, she's been working with businesses and government agencies like the DOJ and FBI, providing expert analysis. Stephanie's articles on examining compensation systems for internal equity have appeared in professional journals, and she has appeared on NPR to discuss the gender wage gap. Stephanie is the founder of Thomas Econometrics and is the host of The Proactive Employer Podcast. Follow her on Twitter at ProactiveStats.

Comments

[Shannon]

Just as not every woman has children and childcare/household responsibilities pulling them away from work more often than men, the following comment does not apply to men across the board either. But a clear trend I have seen consistently in businesses is that the men are pulled away from focusing on work much more often than women in order to play golf, watch several-hours-long sports events, have extended lunches, and attend happy hours. While of course this is at times related to building business relationships, in some ways that makes it even more frustrating to see--i.e., not only are women putting their heads down at work to get through those responsibilities efficiently so they can get out of the office at a decent time to get home for evening family responsibilities, but they're also often missing out on developing those internal and external relationships that are critical to career success and progression.

[Stephanie R. Thomas]

Thanks for the comment Shannon. You are absolutely right that different women are affected differently by family and care-giving responsibilities, as are men. I also agree with your point about differences by gender in building internal and external relationships and career success. There is no doubt that some inequities still exist in our workplaces. I think the answer lies in asking the difficult questions relating to the very kinds of issues you raise, not looking in the same old place over and over again, expecting to find new answers. We need to change our focus on these issues and develop solutions that empower all employees to be successful - regardless of gender.

[E. James (Jim) Brennan]

The American Compensation Association conducted a survey of member pay using a regressed questionnaire job evaluation approach. Analyst Howard Risher, PhD, reported that when the variables of age, experience, education, number of jobs, position level, organization type and company size were accounted for, women earned much less than their male peers. When I first read that published member survey study in 1981, seeing how pay increased for factor after factor, until the very end, I was stunned when I read the final instruction to subtract 14.3% from the predicted total compensation figure if you were female. That survey was never repeated.

Even though actual experience time was equalized, the fact is that "mommy-breaks" may have occurred to disrupt the continuity of those absolute experience years. For example, while a man may have 5 years of work experience from 2007 to 2011, the female peer may have spent a year outside the trade raising a family and hence her 5 years job experience spanned a longer period. Regardless, after all the legitimate reasons have been accounted for, that nagging 14.3% residual differential remains as a serious issue.

[Stephanie R. Thomas]

Jim,
Your comment reinforces my point. If I'm understanding correctly, the 14.3% differential you reference remains after controlling for age, experience, education, number of jobs, position level, organization type and company size. Collecting more information on earnings and age, experience, education, position level, etc., is - in my opinion - not likely to shed any light on the residual differential. We can't keep running the same model over and over again expecting to get a different answer. I agree that residual differentials in pay are a serious issue. We need to look outside of the conventional models - and the same old legitimate factors we've been looking at - if we hope to make any headway in understanding why residual differentials exist.

Thanks for weighing in on this issue and sharing your expertise!