Thursday, March 31, 2016

The Gender Pay Gap

How big is it, really?

Numbers are great, data in the form of statistics can help us make a point, prove a theory or just move a conversation along in an organized, factual way. However, those same numbers can have a dark side. They can be used to sway arguments in one direction or another, depending on your motive or position on the subject.

"Simple" statistics, like gross percentages, are particularly prone to this kind of misuse, I think. Enter the "gender pay gap" argument. The most commonly stated gap is that women earn only 78% of what men earn – a 22% difference. There are many who use this as ammunition for the passage of more and more federal, state and local laws and regulations to "address" pay discrimination. There are at least a couple of problems with that number. The figure compares all working women with all working men. It also implies a statistical correlation or relationship that is less than direct (if you’re into statistics, you might be familiar with the terms spurious correlation and spurious relationship).

Comparing all working women with all working men ignores several, very important factors. Among them are age, job title, time in workforce, industry, employer, education, experience and geography. All of these factors can and do influence rates of pay. That gross number of 78% does not control for any of these relevant factors. So, what happens when you do control for those factors? Well, a recent survey by Glassdoor tells us.

When controlling for all those factors, the gap shrinks to something like 5.4%. Glassdoor looked at data from 5 countries. As you’ll see the results from those countries are fairly consistent.

"To drill down further into what’s causing the gender pay gap, we look at the numbers in a slightly different way. We divide the overall gap into an "explained" part due to differences between workers, and an "unexplained" part due either to workplace bias—whether intentional or not—or unobserved worker characteristics. In all countries, most of the gender pay gap is explained. The "unexplained" part is 33 percent in the U.S. and is less than half in every country, suggesting overt discrimination alone does not explain most of today’s gender pay gap."

What is the main cause?

"But the data show that while overt forms of bias may be a partial cause of the gender pay gap, they are not likely the main driver. Instead, occupation and industry sorting of men and women into systematically different jobs is the main cause."

Take a look at the study. It’s very thorough and does a good job of explaining the numbers in a factual and intellectually honest way.

Yes, gender pay discrimination occurs; I’ve experienced it myself in my career. But what this study and others shows me is that more and more laws (that truly don’t offer any more protection) do nothing to address the situation. I covered this in a recent post here. Forcing business to collect and report mass amounts of gross data and not drilling down to the relevant factors will not address any real pay discrimination that occurs. There are already several federal (and state counterpart) laws that address discrimination and provide appropriate penalties for violations. More is not better. And using numbers to mislead and garner support for your cause without fully examining or explaining those numbers is dishonest.

No comments:

Post a Comment