My post last week on why the "gender pay gap" is mostly bogus generated a great discussion in the comment box. In the hopes of keeping it going, this week I'd like to talk about some of the discrimination or quasi-discrimination issues we do occasionally find.

That nasty remaining five percent or so* that can't be explained by personal choice.

*Completely unscientific percentage.

If your pay audit uncovers one of these issues, you ought to be thinking about fixing it:

The Lilly Ledbetter Effect. Lilly Ledbetter says she received relatively poor performance reviews in the 1980's because she was a woman. Whether she is correct or not, I am sure that her employer, Goodyear, would have had a tough time disproving this* because she didn't sue until after she had retired, and any records were probably long gone by that time.**

*Yes, I know the burden of proof in a discrimination case in court is on the employee, but as a practical matter it's on you, the employer. And if you're a federal contractor, you know that the Office of Federal Contract Compliance Programs will not give you the benefit of the doubt. If they find a disparity, you will have to prove to their satisfaction that it's not a result of discrimination.

**This is a reason why employers should not destroy records, but that's a topic for another blog post.

But, anyway, let's take Ms. Ledbetter's word for it. So, she got lousy reviews only because she was a woman in a man's world, and pay increases were based on performance ratings. Just for the sake of argument, let's say Goodyear gave a 5 percent increase to everybody who got "Exceeds Expectations," and 3 percent to everybody who got "Meets Expectations." And just for the sake of argument, let's say Ms. Ledbetter deserved an "Exceeds" rating in 1982 but got only a "Meets" because her male chauvinist pig of a boss thought women were good for only one thing. And I don't mean making tires. Meanwhile, Ms. Ledbetter's male counterpart, Joe, got "Exceeds" in 1982 even though he had exactly the same performance as Lilly, just because he was a man.

"Women can't do no work. Dey ain't as smart as us guys is!"

Let's say the male chauvinist pig boss retires in 1983, and from that point on, Lilly begins reporting to Alan Alda, who gives her the performance ratings she really deserves every year after that. Which means that she and Joe BOTH get "Exceeds" in every year until Lilly retires, many years later.

"I love all women. And I don't mean that in an inappropriate way."

Not only will Lilly never catch up with Joe, but also the gap in their pay will widen over time, even though they received exactly the same percentage pay increase every year from 1983 forward.

If you review your compensation and find an unexplained pay disparity involving long-term employees, you should investigate whether it's a result of the "Lilly Ledbetter Effect" -- the lingering effects of a long-past discriminatory decision. If so, then go ahead and make an adjustment to catch her up. Don't wait for her to sue you, or (more likely) for an audit by the OFCCP.

Be sure to visit Stephanie Thomas's The Proactive Employer Blog for this month's Employment Law Blog Carnival, Spring Is In Bloom Edition! Stephanie, thank you for hosting!

The "Retro Career Choice" Effect. This is similar to the Lilly Ledbetter Effect, except that it's technically not a "discrimination" issue. However, it is a fairness issue that creates the appearance of discrimination.

In this scenario, you have a long-term female employee (let's call her Flo) in a supervisory or management position who is paid less than her male counterparts with similar time in the position. Flo has been a great employee, and you don't have any male chauvinist pigs working for you, so you know that she's been rated fairly her entire career.

Flo is on the right.

When you look back at the beginning of her employment, you see that Flo started out as a machine operator and worked her way up to supervisor 20 years ago. Her male counterparts started out as skilled mechanics before working their way up to supervisor 20 years ago. Flo and the boys all got a 10 percent pay increase when they were promoted to supervisor 20 years ago.

OK, Robin - we see where you're going with this. So why not give everybody a "flat" pay increase when they get promoted?

Glad you asked! The reason you use a percentage rather than a "flat" increase is that you want to make sure employees don't lose money when they're promoted from an hourly position with overtime into a management position without overtime. The only way to guarantee this (apart from giving everybody, like, a $100,000 pay increase when they're promoted to supervisor) is to base the employee's supervisor pay on his or her prior hourly pay.

Oh, OK. Thanks. Now, back to our story. Flo freely chose the machine operator job (contrast with "The Channel Effect" below), but the machine operator job didn't pay as much as the mechanic job, so when Flo was promoted and got her 10 percent increase, she got less real money than the guys did. And her future raises would have been a percentage of each year's salary. Which means that, even if Flo and the guys got exactly the same percentage increases every year after their promotions, as with Lilly Ledbetter, the pay gap would continue to widen, and after 20 years Flo would be earning significantly less than her male counterparts. Even though she is every bit as good a supervisor as they are, and even though your company has treated her in a completely non-discriminatory manner.

Doesn't seem right, does it? So, even though this is technically not discrimination, we do normally suggest that employers give their "Flo's" an adjustment to bring them into line with their counterparts.

The Channel Effect. Thanks to commenter Randy Martinez for bringing this up last week in connection with sales positions. Anyone with eyes to see and ears to hear knows that women (statistically speaking, of course) tend to choose certain jobs and that men (statistically speaking, of course) tend to choose other jobs. If these choices are truly voluntary, then there is nothing wrong with it -- whatever the government or the National Organization for Women may tell you.

"Oh, yeah?"

But occasionally the employer will take it upon itself to "guide" women into certain jobs and "guide" men into different jobs that pay more money.

In recent history, supermarkets have been accused of channeling women into cashier positions and men into meat department positions, which pay a lot more because they involve things like cleaning fish guts and cutting cows up into steaks. (I'm not saying supermarkets actually channel -- only that they've been accused of it.)

The OFCCP is onto channeling, and has announced that it will be on the lookout for it during compensation audits.

If you're "channeling" based on assumptions, stereotypes, or even past experience, about what men and women prefer to do or are better at, then stop it! Right now! This is sex discrimination, and it is illegal.

Then, once you've stopped, figure out a way to offer the "male" jobs to your female employees. Many will prefer to stay where they are, but those who prefer the "male" jobs ought to have the opportunity. Once you find out who those women are, consider whether you also need to make some pay adjustments.

The "I Have No Flipping Idea" Effect. Finally, you might come across this one. You or your lawyer finds that a female is paid less than similarly situated males. You try to figure out why. You investigate all of the non-discriminatory reasons we discussed in last week's post, and none of those fit. You investigate all of the discriminatory or quasi-discriminatory reasons discussed above, and none of those fit, either. You talk to everybody who has been involved in hiring these people, as well as those involved in giving raises. Nobody has a clue why this woman is making less money.

"Hmm. I have no flippin' idea why Evangeline is making less money."

If you have a pay disparity that you cannot explain, then go ahead and make an adjustment. Your utter lack of an explanation will be held against you.

Next week, I'll conclude this series with a post on preventive steps that employers can take to avoid creating pay disparities in the first place.