Ask HN: How to determine equal pay for genders at a company?

3 points by JBlue42 ↗ HN
I have a question sparked by the NYT Gender Equality link on the main board as well as a throw away comment by a former co-worker yesterday.

With a few women at my past work, I've had the discussion about equal pay for experience and, since the majority of payroll is opaque, the general assumption is that they get paid less, even though that workplace claimed to be "very progressive".

The general assumption for pay was that it was whatever you were able to negotiate when you came in.

Beyond that though, how would you figure out if a company actually was or was not paying people equally for equal amount of experience, regardless of gender?

I guess the easiest answer would be to have enough trust and honesty with some of your co-workers that you could freely discuss it. Since I was not in the same profession as them, I had no trouble discussing my salary but, especially in the US, it's seen as something that's not done.

I don't think this previous company were consciously nefarious, only that the operated like a company does, getting people to work for as little as they'll accept.

Curious what others think.

8 comments

[ 3.0 ms ] story [ 11.6 ms ] thread
There's no solution other than anonymity. Humans are and will forever be inherently biased. Therefore the solution is to remove information we don't want to influence us. In your case, this means gender.

Other than anonymity, which is the general solution, you could just not allow negotiation or adjust wages (which will be to you/your company's detriment).

It seems like the salary band, determined on experience, would at least give room for negotiation or adjusting wages, at least the hiring level.

In terms of a company with the wage structure it currently has, you would have to the executive level be interested in "normalizing" the wages, which would probably also cause a ruckus.

Not to mention if you have people that, due to client relationships, work on more monetarily significant projects.

Closest I've seen anyone get to success on this front is to have per-title salaries with very narrow pay bands, like "Developer 1: $80K-$90K". Not really practical in a small shop though. If there's no next rung, better people will immediately know that they've topped-out and just leave, but at least it's honest.

The problem with the secrecy surrounding this topic is that instead of hashing out the differences in compensation and provided-value up-front, the acrimony gets dragged-out over time, and under opacity people tend to assume the worst.

Yeah, it seems like it would open up such a can of worms which, in my mind, just seems to be a part of the difficulties of addressing the equal pay for equal work stuff.

Not to mention that taking a full maternity leave puts women six months behind on things which maybe they're penalized for, whether consciously or subconsciously, when men might be able to take a shorter leave (even if they did want longer).

I know this is all anecdotal, but most people are A-OK with someone making more than them if that person is also obviously providing more value.

With the ad-hoc thing we do in tech, where everyone is working in the same pen and has a similar title, but some are making more and some are making less, and the ones making more are often providing less value and only have a higher salary because they were hired during a better market, you end up with a lot energy being expended on intrigue/sabotage/oneupsmanship that could have been spent on the product.

Take sales for example. You can make a decent estimate of what another salesman on the team is making just by looking at the board and multiplying, and if it comes out higher than what you pulled, it is really hard to pin that on anyone but yourself.

> of addressing the equal pay for equal work stuff.

The "problem" in tech is you have super keen beans who do side projects and have been coding since they were 12.

You also have people who happily take on more responsibility rapidly.

Etc etc.

It makes comparison very difficult.

I read some years ago about this problem in a paper. They solution they had was this:

1. Make buckets of "pays", say 80-85k, 85-90k, 90-95k, and so on, you can decide bucket length

2. you have two people at hand. we will come to know if there is a wage difference between these two.

3. put box in a room, send one person in, with a chit, ask them to put the chit in the box with the label of their salary.

4. get the person out, send second one in, ask them to open the box of their salary range and tell if they find a chit or not.

If they find the chit, they have same salary and the information is complete. if they didn't find the chit, there is a difference.

I am sorry I couldn't find the link or search it back. Hope the grammar/english wasn't too shitty.

I'd wonder if given the prevalence of the bias - you could simply make it be a function of company size.