Promotion Equity: Linking Pay Equity to Diversity and Inclusion

After completing pay equity studies — and taking steps to remedy any shortcomings — many companies are left wondering why some employees remain underrepresented among management. In truth, the link between pay equity and diversity and inclusion (D&I) mandates can be limited. Although the methodologies for studying them are similar, a pay equity study focuses on like roles within the organization as opposed to the organizational structure itself.

How can the data from a pay equity study be repurposed to advance D&I initiatives? The answer is with a promotion equity study. It’s similar in methodology to a pay equity study, but it captures differences in advancement within a firm and seeks to identify root causes for diversity issues at the management level. In this blog post, we’ll discuss what a promotion equity study seeks to identify, how it’s conducted, and what it can reveal about an organization.

Understanding Promotion Equity

Suppose that at Company A, three-quarters of non-management employees are women. The company wants to make sure they promote women from this level to the lowest management level at a rate equal to men, e.g. 50% of promotions go to men and 50% go to women. Promising as this sounds, some quick math shows that if one of 10 men gets a promotion, only one of 30 women gets one since women outnumber men at the pre-promotion level by three to one.

Promotion equity offers an opportunity to break down the promotion differentials. In Company A’s case, an examination of the data reveals two primary patterns:

• Promotion bias. Qualified women fail to be promoted at the same rate as their peers even after controlling for experience and performance.

• Promotability bias. Women are less likely to get promoted because they lack the relevant experience, performance, or other attributes that lead to promotion.

These patterns illustrate how a company should think about D&I. They also highlight the differences among roles in ways that pay equity simply can’t.

If Company A passes over women strictly because of promotion bias, the remedies may be obvious. However, promotability bias is more interesting. If women typically have less experience and tenure, is there a reason for their higher turnover? Are performance ratings biased? Are high-profile projects — the kind that can lead to promotion — going to one demographic more often than others?

By studying the relevant factors for predicting promotions, Company A can set up targeted cohort training and development opportunities that level the playing field for employees.

What a Promotion Equity Study Looks Like

Like a pay equity study, a promotion equity is a statistical model aimed at identifying differences between different groups of employees within an organization. The difference is that promotion equity identifies patterns of advancement within the firm in order to support D&I initiatives.

Mathematically, pay equity and promotional equity studies are fairly similar. Both use regression analyses that control for factors that may influence something like pay or the likelihood to be promoted. These include experience, education, performance, etc.

The key difference between the two is in the target or dependent variable. In the case of pay equity, everyone has pay that can be tested. That’s the continuous dependent variable we use to predict what everyone’s pay should be. In a promotional equity study, there are only two outcomes we look for: promotion or no promotion. Instead of picking the exact individuals who will be promoted, however, we predict a probability for each individual reflecting how likely they are to be promoted.

To do this, we use a logit or probit regression. For each individual, the explanatory variables (experience, education, etc.) implies a probability between 0 (no promotion) and 1 (absolutely certainty of a promotion).

From this analysis, we look for two outcomes. The first is promotion bias, indicated by the fact that being part of an overrepresented employee segment results in a higher expected probability of promotion. We can also look at various cohorts. For instance, if a male cohort of 100 people has a promotion probability of 30%, but 45 end up with promotions, there could be some sort of potential bias in the process.

Promotability bias stems from the difference in the average overrepresented or underrepresented segment. For instance, if the average male promotability score at Company A is 15% while the average female score is 5%, we know there are underlying factors other than gender driving these differences. More importantly, we can see exactly what these factors are because we can measure the differences due to tenure, performance, overall experience, and any other explanatory factors.

Note there may be different impacts of the control independent variables of pay equity versus promotion equity. For example, in pay equity we often find that experience in a particular role isn’t meaningful while broad experience at the firm (or elsewhere) can matter a great deal. On the other hand, for promotion equity performance (both short and long term) and tenure in a role tend to matter more. Promotion equity can also require larger data sets to see where issues are. All that being said, there’s one thing that pay equity and promotion equity always have in common: A careful and well thought analysis is critical.

Final Thoughts

In the war for talent, career opportunities are crucial. So put your pay equity data to work. Equal pay is but one stop along the environmental, social, and governance (ESG) journey, and a quantitative analysis for promotion equity can give a shot of energy to the prospects of meeting your D&I goals over time.