To Truly Address Gender Pay Gaps, Companies Must Dig Below the Surface

In most cases, you can’t begin to fix a problem until you understand why it exists in the first place. This is certainly true for the complex challenge of addressing gender pay equity.



In media and government reports, it's common to hear that women make 20% to 30% less than men, and while statistics of this nature are true in a broad sense when comparing all full-time working women to all full-time working men across the entire US workforce, they can obscure the true scale and specific drivers of gender pay gaps within individual organizations.

We don't say this to minimize the issue. Pay inequities between women and men are a real and persistent issue and we are actively working with many of our clients to address this challenge head-on. However, for business leaders and HR professionals, the path to truly eliminating pay inequities often starts with understanding the limitations of overly-broad and simplistic comparisons of pay. In other words, approaching your first gender pay equity analysis like just any other compensation planning project isn't a recipe for long-term success.

Ultimately, to change policy and behavior, far deeper analysis is needed with statistical controls in place to account for the impact of geographies, job types, job levels, individual performance, education, work experience, inherited new-hire pay levels and a wide range of corporate policies, among other variables. Only after examining all of these inputs through the lens of a multiple regression analysis can companies begin to quantify real gender pay gaps (i.e., situations where the only variable left that accounts for differences in pay between people in similar roles at similar levels is gender itself) and then fully assess the relative influence of compensation, talent management and HR policies on producing such outcomes.

We like to think of this exercise as peeling back the layers of an onion to expose the specific, and often numerous, issues that contribute to aggregate differences in pay. Some of these layers will point to justifiable or intended differences in pay outcomes for all employees, not just women, and some of these layers will expose problematic practices where fixes in policy, behavior or both are needed.

Once this complex mix of factors is taken into account, the data often reveals gender pay gaps of less than 5% inside individual companies, with some variation across industry lines and by company size. However, while gender pay gaps of only a few percentage points may sound innocuous compared to figures cited in the press, they can still have severely detrimental and lasting impacts on wage progression and lifetime earnings. So again, we don't say any of this to minimize the issue; rather, our aim is to advise companies to look at all of their data much more carefully and to work outside the confines of basic pay comparisons to find and fix specific root causes. Once companies actually start to dig deeper, we tend to find the following issues are the most significant drivers of both real and perceived gender pay gaps:

Workforce Composition

In 2017, Radford began to collect data on gender as part of our standard survey submission process. The collection of this data remains a work in progress and is optional for all companies, meaning our data is still preliminary in nature and not necessarily a complete picture of market trends. However, early patterns in our technology sector data highlight that workforce composition is a key factor to consider.

HR jobs, for example, are paid relatively less than core technology jobs and are heavily skewed toward women; 73% of HR incumbents reported are women. Other functions are more balanced, like finance and marketing, and have nearly equal distributions of women and men. Finally, some functions, like product development, are paid very well and are heavily skewed toward men (i.e. women only represent 19% of incumbents). Naturally, if functions with a larger male population pay more than other functions, this can have a dramatic effect on overall appearance of gender pay equity across an organization.

Importantly, this does not mean internal systems for setting pay are necessarily unfair. It only highlights the need to examine data across multiple fronts. When factors like geography, job types and job levels are introduced into the workforce composition mix, you can quickly see how it's possible to get a hundred different answers for why women and men might be paid differently. As such, looking at workforce composition is usually the best place to start a gender pay equity analysis, but it should only be the beginning.

Ultimately, if a comprehensive analysis of your compensation and talent management programs suggests your approach to setting pay is fair, yet perceived pay gaps at your company are large and driven heavily by workforce composition, a different set of questions must be raised. Are there specific recruiting practices or behaviors that lead certain functions and teams to be male dominated? Are there certain functions where women and men are hired at fairly equal rates, but turnover among women is much higher? And what can you do to increase female employment in underrepresented functions and roles over time?

Employee Leveling and Salary Structures

Like workforce composition, seeing where employees sit within your employee leveling and salary structures can go a long way toward explaining perceived pay gaps. Often, aggregate gender pay gaps of 20% or more across an entire organization drop well below 5% once you begin to compare women and men in the same job families and at the same pay grades. Still, as we noted before, even slight variations in pay can be very troubling, and poorly defined policies around how employees are slotted into job families and pay grades in the first place can create significant legal issues. When it comes to both assessing and addressing gender pay equity issues, building effective and consistent employee leveling and salary structures is critically important.

There is a strong legal basis for thinking in this manner. Most new state laws related to pay equity essentially require companies to compare employees across jobs that are “similar” in terms of skills requirements, responsibility, working conditions, and other factors. “Similar” is a vague term of course, which means it is currently up to employers to interpret these laws in their own way— hopefully with a mindset of doing the right thing. However, given this ambiguity, companies can go wrong in a number of ways. If grades and job families are defined too broadly, employers put themselves at risk because pay equity analyses based on large groupings may point to problems that don't actually exist. Conversely, if grades and jobs are too granular, employers are at risk because pay equity analyses may not be able to identify real pay gaps when they do exist.

As such, it's important to find the right balance between specificity and flexibility. A good place to start is leveraging established job classification systems that are based on employee categories, levels, functions and job families. At Radford, we begin every pay equity project by reviewing how employees at the client company are matched to our globally consistent job leveling and job family system. Getting this step right ensures your analysis has the best chance of identifying real issues accurately and sets the table for long-term consistency in applying go-forward pay decisions.   

Starting Salaries

Exemplified by laws in California and Massachusetts, the notion that the process of setting starting pay can drive gender pay gaps is gaining real momentum. After all, we know what it feels like when a recruiter asks us what we currently make. Some may be tempted to add a little something on top, but fundamentally it's an uneasy exercise. On the one hand, we fear we might be asking for too much, and on the other hand, we risk pegging ourselves to a salary rate below what the company might otherwise offer. It is the latter case that some states, by virtue of making it more difficult for companies to ask for salary history, are attempting to correct. In theory, these laws force companies to determine the fair market value of jobs with less input from candidates, which in turn should drive companies to make more consistent offers to all candidates.

In our work with clients, we find the thinking behind these laws is generally sound. Starting salaries, for both women and men, go a long way toward explaining long-term pay outcomes within organizations, and in some extreme cases, explain as much as half of the difference in pay between women and men. In our view, companies who embrace the opportunity to create stronger mechanisms for determining fair starting salaries for all employees, with less input from job candidates, will enjoy a competitive and legal edge by better aligning pay offers to both internal and external labor markets.

Connecting Performance and Pay

In our experience, it's rare to find cases where poorly implemented or executed performance management systems explain most of, or the entire pay gap between genders or ethnic groups. This is the good news. However, when we do find situations like this, companies have a real problem on their hands. How do you rationally explain why one demographic group might systematically underperform another? As such, a consistent performance rating gap that drives pay gaps is a real red flag.

Although performance management should be more straight-forward for teams with objective performance measures, such as sales quotas or customer satisfaction scores, issues can still occur. In theory, things like commissions payments shouldn't be an issue if there is a strong formulaic link between sales wins and pay outcomes. Yet, companies increasingly face questions from attorneys and government agencies when women and minorities systematically generate lower win rates. The key question then becomes, are these groups put in an unfair position by virtue of being assigned less-favorable or smaller sales territories, which in turn creates less-favorable pay outcomes even if people perform well?

Leave of Absence Policies

In some cases, we find that leave of absence policies have an outsized impact on gender pay gaps. While employees on leave are legally protected from discrimination, this does not mean compensation, performance and promotion decisions made during or after a leave of absence are always fair. The simple fact is, a lot of managers struggle with this issue, often due to a lack of coherent and consistent policies on how to fairly assess the impact a leave of absence should or should not have on pay decisions. Companies need to ask themselves, when should bonuses be prorated or paid in full, when should promotions be on or off the table after a leave, how should a leave impact annual merit increases, if at all, and when should exceptions be made?

Given the potential for leave of absence policies to drive gender pay gaps, we recommend clients pay close attention to this issue when analyzing gender pay equity. If employees, both women and men, with extended leaves of absences in their work history have noticeably slower pay progression, this could be a sign that their leave has become a drag on pay that may be hard to justify to employees and regulators.

Manager Bias

Manager bias is always hard to quantify directly, but in some cases, it's just obvious. When above-average gender pay gaps within an organization are isolated to specific functions, teams, or even a handful of managers (e.g., male managers with female supervisees), it's reasonable to assume that manager bias or a lack of training may play a role. Truly assessing the degree to which bias, conscious or unconscious, has played a role in determining pay outcomes for women vs. men usually requires a qualitative touch that data alone cannot address. And most often, the outcomes of such studies are recommendations for greater investments in diversity and inclusion training and better leadership coaching on career development, active listening and pay planning, as opposed to disciplinary action. Still, in every pay equity analysis, this is an angle that must be considered and screened for.

What's Next?

Even though gender pay gaps inside individual companies are often less than 5% once other variables are properly accounted for, these relatively small differences in pay are, first and foremost, unfair, and can add up to significant long-term gaps in wage growth, lifetime pay and retirement savings for women. What's more, even small differences in pay can be signs of systemic cultural issues within an organization— the types of issues that can create significant legal risk and tarnish your reputation as a destination of choice for diverse talent. This is why we tell clients that getting pay equity right requires much more than just crunching the compensation numbers. It requires a holistic view into all of your people data to examine the complex mix of issues than can create and sustain gender pay gaps. And often, the questions you ask yourself after looking at the data are what matters most.

To learn more about our point of view on gender pay equity and our comprehensive approach for delivering lasting change for your people and organization, visit our Pay Equity Matters site.