Why Companies’ Attempts to Close the Gender Pay Gap Often Fail

Gender pay equity has become a big point of contention at many companies. But the most common approaches for identifying a pay gap and resolving it are full of pitfalls for the unwary. That’s because it’s a tall order: you have to calculate the gap the right way and figure out how to fix it without ballooning your wage bill, all while truly helping underpaid women, maintaining your incentive structure, and avoiding the creation of new legal liabilities. The authors have extensively researched the most common ways companies try to fix a pay gap – and how these fail or cause other problems. They’ve found that closing a gender gap without regard to cost effectiveness can be prohibitively expensive; however, only focusing on cost (as many managers do) creates more problems than it solves. They suggest first identifying which employees are contributing the most to the gender pay gap at your firm, and then allocating raises as efficiently as possible to close the gap — while working within the framework of your HR strategy and norms of fairness.

Gender pay equity has become a big point of contention at many companies. Not only have politicians and other public figures spoken out against the gender pay gap, but there has also been a rising tide of high profile lawsuits targeting major employers, most notably in the U.S., with all the bad publicity and financial liability they entail.

In response, many firms have hired external pay consultants and law firms to identify whether they may have a problem with the pay gap from either an HR or legal perspective and to offer possible remedies. But in our view, the most common approaches for identifying a pay gap and resolving it are full of pitfalls for the unwary. That’s because it’s a tall order: you have to calculate the gap the right way and figure out how to fix it without ballooning your wage bill, all while truly helping underpaid women, maintaining your incentive structure, and avoiding the creation of new legal liabilities.

We have extensively researched the most common ways companies try to fix a pay gap – and how these fail or cause other problems – and we’ve worked with several companies in different countries to solve their pay equity issues. We’ve found that closing a gender gap without regard to cost effectiveness can be prohibitively expensive; however, only focusing on cost (as many managers do) creates more problems than it solves.

Our approach focuses on first, identifying which employees are contributing the most to the gender pay gap at your firm, and second, allocating raises as efficiently as possible to close the gap — while working within the framework of your HR strategy and norms of fairness.

The gender pay gap is usually described in the popular press as a simple comparison of the average pay of women and men. The problem with this calculation is that it does not account for the many job- and person-related factors that could legitimately affect someone’s pay, such as functional role, experience, or performance.

That’s why pay analysts usually try to “control” for these other factors when analyzing pay. Although there is more than one way to do this, the most common method is linear regression. In the courts, academic research, and policy, however, there is an additional twist: log-wages are used instead of dollar wages in the regression.

You may remember the logarithm function from high school math. “Taking logs” has the effect of making things go up more slowly, as in a graph using the log of stock prices. Specifically, with wages, taking logs is like counting digits. For example, if you have employees who earn $10,000, $100,000, and $1,000,000, then after taking (base 10) logs, their wages would become 4, 5, and 6. Nearly a half-century of social science research shows that taking logs of your employees’ wages to calculate the pay gap results in a more accurate model than using their dollar wages, as it lets you measure effects that increase salary by a given percentage, rather than by a given dollar amount. This turns out to have important implications, as we explain below.

Let’s say your firm has a 10% pay gap – meaning that men make roughly 10% more than women despite equivalent job and personal characteristics. This is a pretty realistic figure. The companies we’ve worked with typically have a pay gap in the mid to high single digits, as do many countries. What should those companies do first?

One approach we’ve seen leaders take is to give every woman in the firm a 10% raise. But this can be a major cost increase, enough to put companies in the red. So a common next step is to have HR evaluate every employee’s pay and give raises to every woman (or person) who appears underpaid based on their job and qualifications. But this process lacks a clear objective way of determining who is underpaid and can be equally expensive.

Our research and experience with practicing managers have identified another way. Managers need to establish a list of defined priorities around closing the gender pay gap. Based on what we’ve heard from companies, these priorities may be things like minimizing the overall increase in the wage bill, capping raises to individual employees in percentage terms, maintaining pay differences across job categories to reflect different job responsibilities and to incentivize good performance, avoiding large discrepancies with the external job market, and paying women fairly in the context of your firm.

These priorities should then be converted into quantitative goals in a raise allocation process. The result is called a constrained optimization problem — which can be solved mathematically. We find that by targeting raises to women whose pay is driving the gap, and taking managerial objectives like fairness and equality into account, those raises can close the gap more cost effectively than simply giving across-the-board, equal raises. This also makes the pay structure more transparent and more equitable.

Even with a strong strategy in place, there are pitfalls companies should be prepared for when doing the important work of closing a gender pay gap. Here are a few to consider:

You may not help women much. If you focus on closing your gender pay gap at minimum cost, women may not benefit nearly as much as one might expect for two reasons. The first reason is that a cost-efficient allocation of raises will usually cost much less than if you had given your female employees an across-the-board 10% raise. That has obvious appeal, but communicating this strategy requires care and equal pay advocates may think you’ve pulled a fast one. The second reason is that our research also shows it is frequently efficient to give certain men raises to reduce the gap, something that surprises even trained statisticians. In other words, some men will get paid more so that women appear to be paid more equitably.

The mathematical reasons for this are complex, but the intuition is straightforward. In nearly every workforce, there will be gender patterns in terms of what jobs and qualifications women and men tend to have. For example, women at your firm may be overrepresented in certain departments or tend to have more or less education than men. If you give a highly educated employee a raise, you not only raise the average pay of the employee’s gender but also raise the importance of education as a predictor of pay. And as education (or anything else) becomes a better predictor of pay, the importance of gender as a predictor goes down.

The implication is that if, say, men tend to be more educated than women at your firm and you give a raise to a man with a very high level of education (e.g., a PhD), you may actually shrink your gender pay gap – simply by making gender less important as a factor in determining pay. Perversely, you are paying women more “equitably” by giving raises to certain men. But this doesn’t necessarily mean pay is equitable. It just means you’ve shifted the indicators which drive how pay is determined.

You may corrupt incentives. Giving a $1 raise to a low-wage employee will have a bigger percentage impact on the gender pay gap than the same raise to a high-wage employee. (This is because using the logarithm of pay in the gender pay gap calculation tends to down weight high wages; see the sidebar for more on this.) So allocating raises efficiently can result in raises disproportionately going to low-wage employees, compressing a firm’s wage structure. We have seen cases where blindly allocating raises as cost-efficiently as possible would have inverted a firm’s wage structure. This means that you could pay less qualified women (in terms of education, experience, and job responsibilities) more than more qualified women – probably the opposite of what your HR strategy calls for. A major potential problem from this is that getting promoted no longer comes with a large enough increase in pay to motivate employees. Similarly, women who perform worse may end up with higher salaries than women who perform better.

You may create new legal liabilities. We aren’t lawyers, but we have seen enough to know that liability in labor law can have a whack-a-mole quality. You eliminate your pay gap in part to avoid getting sued for gender discrimination. But if you leave a paper trail saying you cleverly reduced that gap in the most cost-efficient way possible, it could be used against you. A plaintiff could argue it was evidence you had gamed the pay gap measure and that pay equity at your firm should be assessed some other way.

Allocating raises efficiently can also result in some employees getting large raises for purely statistical reasons, which could be regarded as unfair by a court or your employees. Further, raising salaries will be received as a signal of strong performance, as opposed to salary correction. Therefore, an employee may point to the raise as evidence that she is a strong performer, even if her work isn’t meeting expectations. This creates complicated situations for managers to handle.

Thinking through the issues outlined above can help your company make progress on its gender pay gap. It can keep you from getting sued, losing your best female employees, and being blindsided with a sudden PR crisis. It can be implemented in a matter of months, even weeks, and communicated to employees as a proactive strategy to make your company a better employer.

Ultimately, however, you need to think about the HR processes that led to the gender pay gap arising in the first place. Are women disadvantaged on intake? In ongoing raises? Are there large gender disparities in representation in different parts of your firm? Are you suffering a high rate of attrition in your female employees?

Your firm may not be able to solve all these problems immediately, but if you can make even modest headway identifying and rectifying the frictions that women may face in managing their careers, you can gain a step up on the competition. That can help prevent another gender pay gap from arising down the road, boost the company’s reputation, help you recruit the best female talent, and position all your employees – male and female – to maximize their contributions. This process is ongoing and without end. Staying ahead of the competition always is.

David Anderson is an assistant professor of analytics at the Villanova School of Business. He is a co-founder of PayAnalytics.

Margrét V. Bjarnadóttir is an assistant professor of management science and statistics at Robert H. Smith School of Business at University of Maryland and a co-founder of PayAnalytics.

Cristian Dezso is an associate professor of Business and public policy at Robert H. Smith School of Business at University of Maryland.

David Gaddis Ross is the R. Perry Frankland Professor in the Department of Management at the University of Florida’s Warrington College of Business Administration.

Why Companies’ Attempts to Close the Gender Pay Gap Often Fail

Research & References of Why Companies’ Attempts to Close the Gender Pay Gap Often Fail|A&C Accounting And Tax Services
Source

Share on Facebook

Tweet

Follow us