Feature |

Deep Dive

Looking Beneath the Surface of the Gender Pay Gap

Over and over we hear that women in the United States who work full-time are paid only 80 cents for every dollar paid to men. This statistic is repeated by many organizations and in many publications. (See “The 20% Gender Pay Gap Is Misleading Without Context.”) So, what are all of these examples missing? While some sources talk about occupations, there is no discussion of the actual work being done, no mention of the job.



Because gender pay inequity addresses the pay of all employees, it is an obvious major issue that demands the attention of total rewards professionals. Only by exploring the issue in depth can we properly respond to inquiries from not only employees and executive management but also the inquiring press and politicians.

The articles referenced in the first sidebar all use data from the exact same source, namely the Census Bureau report, “Income and Poverty in the United States: 2016.” (See Tables 1 and 2.) What do these numbers represent? It is the ratio of the median income of more than 47,000 women to the median income of more than 63,000 men who were respondents to a Census Bureau “Current Population Survey” (CPS), no matter the industry, occupation, level of work and location.

The CPS, which originated in the late 1930s, is the longest-running survey conducted by the Census Bureau. It is a household survey primarily used to collect employment data. Data are collected throughout the year, every year. (See “Census Bureau’s CPS Methodology.”)

After the data are collected for each gender, everything is thrown into one big pot for the calculation of the median income: secretaries, bricklayers, senior IT professionals, dishwashers, company presidents, chemical engineers and so on. Most importantly, there are no questions in the survey questionnaire about the specific work people were doing. Hence, those compensation factors that may account for legitimate pay differences were neither gathered nor considered. (See “Survey Shortcomings.”)

In general, what factors do determine a person’s pay? Figure 1 identifies factors that help determine a person’s pay, including five system factors and three individual factors. The five system factors typically are the foundation for the pay range of a job. Two factors of the job that focus on the work — market pay and internal value — are combined to form a job’s pay range (usually determined by an assigned grade). Note the emphasis on the job rather than the occupation or the title: This is what is missing from the discussion of, and efforts to achieve, pay parity.

We all are familiar with the concept and laws regarding equal pay. The Equal Pay Act was enacted in 1963, and it mandates equal pay for “equal work on the jobs the performance of which requires equal skill, effort, and responsibility, and which are performed under similar working conditions, except where such payment is made pursuant to (i) a seniority system; (ii) a merit system; (iii) a system which measures earnings by quantity or quality of production; or (iv) a differential based on any other factor other than sex.” Notice the act’s focus on the work. Again, there was no reference to the work or jobs in any of the articles that would form a basis for 80%.



For all practical purposes, the 80% number is conceptually meaningless with regard to pay parity because it only indicates that many female-dominated jobs are paid less than many male-dominated jobs. From this data, we do not know how the two genders are paid within a job. As such, nothing actionable can be derived from this figure regarding parity for the work done.

This is a prime example of a statistical phenomenon known as Simpson’s Paradox. This paradox can get you into a heap of trouble if you inappropriately combine different groups into one, especially if the results are a basis for a policy decision. (See “Simpson’s Paradox.”)

After examining the example, the answer to the question, “Should I combine the data for different jobs into one figure?” is “Do so at your own peril.”



Why do the articles and pronouncements ignore the missing link of work? We can only speculate. The information in the CPS is there for all to see. One wonders, then, why those articles’ authors did not dig into the data to understand what the numbers really represent. Assuming the authors are familiar with the Equal Pay Act, why did they ignore the context of equal work in their statements? Are they so caught up in a narrative that women are undervalued that it clouds their thinking process? Or are they using these figures to advance a political agenda? These are not idle questions, given the false statements in a presidential political campaign, and given the almost universal ignorance of the missing work context.

In 2012, the ratio was 0.77 from that year’s Census Bureau survey, which was the source of a 2012 campaign TV ad stating, “President Obama knows that women being paid 77 cents on the dollar for doing the same work as men isn’t just unfair, it hurts families.” The addition of “same work” was misleading and this same misleading notion was repeated in his State of the Union address.

They’re not “lying with statistics,” but are being dishonest by not telling the whole story with the context of the figures (lying by omission).

For whatever reason, authors of the articles apparently either did not dig into the data to discover the context of what was reported, or simply chose to ignore it. But that is our task as total rewards professionals.



The 80% figure raises volatile false expectations because it creates a climate of envy, oppression and divisiveness. Improperly discussed, it can be very contentious. What might women think? “Either my company or the government is going to bring me to parity. Someone has to do something.” But that’s not going to happen, because it can’t happen.

The implication is that a woman doesn’t have to change jobs or careers. So, envy, dissatisfaction and frustration occur because of the incompleteness and subsequent misleading nature of the 80% pay statements and articles.

The “something” that must be done is to be honest with the numbers, and that includes their context. To repeat, it is incumbent on total rewards professionals to exercise their critical thinking, especially when the message to female employees is so misleading.

The actions to take are three-fold:

  • Ensure your own house is in order vis-à-vis equal pay. In my consulting experience, most companies want to pay their employees fairly for the work that they do. Employees who perceive they are paid fairly are engaged employees and that is clearly in a company’s self-interest. Most total rewards professionals have or can get the statistical tools needed to conduct internal audits and identify and correct equal pay issues.
  • Educate yourself on the source and context of the 80% figure. Focus on gender pay differences within jobs. Be prepared to tell the whole story. Truth is what we are seeking –– the truth, the whole truth and nothing but the truth.
  • Speak out about it with complete facts and don’t let the misinformation spread by the default of your silence. Silence is assent with the misleading statements.

John Davis.jpg John H. Davis, Ph.D., CCP,  is the retired president of Davis Consulting.

Have Questions?


+1 877 951 9191

USA and Canada

+1 480 951 9191

Other Countries


Email Us