Want more expert insights on the topic of pay equity? Access our entire Pay Equity Deep Dive Series .
The primary objective of a proactive pay equity analysis is to identify and remediate pay disparities and prevent future ones.
In contrast, the primary objective of a reactive pay equity analysis is to address immediate legal risk.
You might assume that the process for conducting a proactive pay equity analysis is pretty much the same as the process for conducting a reactive pay equity analysis. While there are some overlaps, there are some notable differences. Let’s start with the big picture and drill down.
Why Are You Conducting the Analysis?
First, let’s discuss why you are conducting the analysis.
Are you conducting the analysis for one or more of the following reasons?
- “It’s the right thing to do”
- You want to attract and retain top talent
- You want to build trust with employees
- You want to enhance employee engagement
- You’d like to position your organization for greater pay transparency
- You’re under pressure from activist investors to share information about pay equity
- You’d like to mitigate future legal risk
If so, then a proactive pay equity analysis is likely your best choice.
Are you conducting the analysis in response to a perceived threat of legal action, a specific legal issue or area of concern, or actual legal action?
If so, then a reactive pay equity analysis is likely your best choice.
What if both apply to your organization? Then we recommend conducting separate proactive and reactive analyses.
Scope of the Analysis
One key area in which proactive and reactive pay equity analyses differ is the scope of the analysis.
The scope of a proactive pay equity analysis is typically an organization’s entire workforce. Global or complex organizations may choose to start with specific geographies and/or lines of business and expand the analysis over a period of years to encompass their entire workforce.
For a reactive analysis, it’s preferred to limit the scope to the group or groups at issue. For example, if an organization has specific concerns about potential pay disparities in one of its job functions, the pay equity analysis likely will be limited to the employees in this one function.
Legal Involvement
A proactive pay equity analysis often involves legal counsel. An organization may choose to involve internal legal counsel only or may choose to bring in external counsel as well. While it’s rare for an organization to conduct a proactive pay equity analysis without any involvement from its legal counsel, I’ve seen it on a few occasions. While a protected, proactive pay equity analysis is typically managed by counsel, much of the day-to-day work itself is delegated by counsel to the compensation function.
A reactive pay equity analysis, on the other hand, always involves legal counsel. Moreover, legal counsel is likely to be more involved in the direct management of the work as compared to their day-to-day involvement in a proactive pay equity analysis.
Methodology
The primary methodology underlying both proactive and reactive pay equity analyses is multiple regression. Recall that multiple regression estimates the relationship between an outcome of interest, such as compensation, and multiple factors that are related to that outcome.
Achieve Authentic Pay Equity With Software
Starting with compensation as our outcome, we add relevant Wage Influencing Factors (WIFs) to a regression model. Once these WIFs are accounted for, we determine if gender, race/ethnicity, and other demographic characteristics are statistically related to compensation.
Despite both types of analyses using multiple regression, the analyses themselves may look quite different.
Pay Analysis Group (PAG) Differences
As we shared in a previous blog, an important first step in conducting a pay equity analysis is segmenting your workforce into Pay Analysis Groups (PAGs). The purpose of the PAG formation process is to group employees into meaningful pools for comparison purposes. The PAGs you develop for a proactive pay equity analysis will likely be different than the ones you develop for a reactive analysis.
For example, since a primary objective of a proactive pay equity analysis is to identify and remediate pay disparities, you’ll want to ensure that your PAGs have sufficient statistical power. Without sufficient statistical power, your analysis may not be able to detect pay disparities, which may mask issues. Our statistical power rule of thumb at Trusaic is to ensure each PAG has at least an 80% chance of identifying a 5% pay disparity.
Meeting this rule might require pooling employees such that you have fewer pooled PAGs, but each one has more employees. In an earlier blog, we discussed the trade-offs between creating many differentiated PAGs and creating fewer pooled PAGs.
In a reactive analysis, you’ll likely want to ensure that your segmentation groups together employees that are highly similarly situated. For this reason, you’ll likely focus on creating many differentiated PAGs. For example, if you have concerns about a specific job, you may decide to create a single PAG that includes only employees in that job. If people in these jobs are spread across different geographic locations, you may decide to further segment by location.
These smaller PAGs may not be sufficient for conducting a regression analysis. In these cases, you may decide to use non-regression methods (e.g., Fisher’s Exact test or Mann–Whitney/Wilcoxon rank sum test) to support the position that pay differences are justified.
We do not recommend using non-regression methods for a proactive pay equity analysis. Such methods do not allow you to account for relevant WIFs, do not provide actionable results, and run the risk of “false negatives” because they have less statistical power than regression.
Wage Influencing Factor (WIF) Differences
We would also expect WIFs to differ between a proactive and reactive pay equity analysis. Recall that a WIF is a factor reflecting skill, effort, responsibility, working conditions, or location applied fairly and consistently in determining employees’ compensation. WIFs enable us to answer the question: Once legitimate, compensable factors (i.e., WIFs) are accounted for, do we find that gender, race/ethnicity, or other demographic characteristics are statistically related to compensation?
For a proactive pay equity analysis, the selection of WIFs should reflect an organization’s pay philosophy. To help identify the appropriate WIFs for each of your PAGs, we recommend creating a WIF matrix to catalog which potential WIFs are relevant for which parts of your organization. Since you may be making remedial pay adjustments based on the results of a proactive pay equity analysis, you’ll want to: (1) appropriately consolidate and refine your WIFs and (2) ensure that the models you’ve created are reliable and robust. Our earlier blog on the topic of Wage Influencing Factors and Reliability Testing covers these topics in detail.
Optimize Pay Disparity Remediation Efforts With Software
This model refinement process typically leads to more “parsimonious” models. In statistics, a parsimonious model is one that uses relatively few factors to explain an outcome, such as compensation. According to Statistics By Jim, “A parsimonious model in statistics is one that uses relatively few independent variables to obtain a good fit to the data…Analysts often think that intricate problems require complex regression models. However, studies reveal that simpler models tend to be more precise. When evaluating several models with similar explanatory power, choose the simpler one.” This precision is vital when you are using the results of your analysis to change the pay of individuals. Moreover, government regulators, such as the OFCCP, advise using more parsimonious models.
Since a reactive pay equity analysis is primarily about addressing legal risk, it’s less likely that you’ll be making remedial pay adjustments based on the findings. With that in mind, your pay models will likely include as many business-related or job-related factors that you would expect to influence employee pay for the group or groups under consideration, with less focus on parsimony. Additionally, you may choose to collect qualitative information you are not currently tracking if it helps to explain pay differences.
Frequency
We recommend conducting a proactive pay equity analysis on a regular basis. As discussed in an earlier blog on preventing pay disparities, for most organizations, it’s appropriate to conduct a pay equity analysis and remediate disparities annually (in conjunction with your merit process). In addition, we recommend monitoring pay equity throughout the year, either on a quarterly or semi-annual basis, to track progress.
A reactive pay equity analysis is conducted only as needed to address immediate legal risk.
If an organization conducts a proactive pay equity analysis on a regular basis, addresses disparities identified, and puts in place systems and processes to prevent pay disparities from recurring, then you’ll likely find that you need to conduct fewer reactive pay equity analyses.
Related Posts
