![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
A lot of blogs are commenting on the differential in the growth of men's and women's median wage* since the 1970s. In brief: Until 1970, male median wage moved exactly in synch with GDP, but has essentially stagnated since then, even though the GDP has continued to grow at the same rate. Meanwhile, women's median wage continues to grow roughly in synch with GDP. Obviously this is a complicated issue with a number of contributing factors, but the one I want to talk about is tournament style compensation.
There's two ends to the spectrum of compensation:
There's some blur between the two: being a programmer at a large company is effort-indexed, working at a start up is tournament style. And you never know, you could be such a good nurse a wealthy patient leaves you billions. But the categories stand as useful.
The interesting thing is that there's a lot of evidence that when you move something from effort-indexed to tournament style, men compete harder and women compete less.** There's no research into the causality that I know of, but it seems likely to be related to the fact that reproductive success is effort-indexed for women but tournament-style for men.
Given that, you could add a lot of money a pool, but if it's divied up tournament style, it will make minimal difference to the median wage. Since men are more likely to be working in tournament-style systems, it would make sense that adding more money to each pool will affect women's median wage more.
*Brief primer: median means line up everyone in order of the trait of interest and select the person in the middle. This is distinct from the mean, which adds all the numbers together and divides by the total number of people. The median can be the more relevant in certain cases involving data with long tail distributions.
**See this paper for both a general overview of the research, some specifics on how to counteract the effect, and really fascinating info about how the menstrual cycle affects willingness to compete. It's worth noting that the pre-menstrual period, the yimr that makes women unfit to be president or CEO or head surgeon, is when their hormones profile most closely resembles the typical male profile. I'm just saying.
There's two ends to the spectrum of compensation:
- effort-indexed, where your rewards are roughly a function of the talent effort you put it, regardless of how much effort others put in. A good example is nursing: the best nurse makes more money than the worst nurse, but not millions more. Mathematically, you can think of the compensation as a roughly linear or maybe even logarithmic function of quality of work. It's worth noting that I made up the term effort-indexed because the official term doesn't exist and/or is hard to find.
- tournament style, where whoever does something the absolute best gets a ton of money, and everyone else gets very little or none. Think being an author, or professional athlete. Mathematically, it's an exponential curve.
There's some blur between the two: being a programmer at a large company is effort-indexed, working at a start up is tournament style. And you never know, you could be such a good nurse a wealthy patient leaves you billions. But the categories stand as useful.
The interesting thing is that there's a lot of evidence that when you move something from effort-indexed to tournament style, men compete harder and women compete less.** There's no research into the causality that I know of, but it seems likely to be related to the fact that reproductive success is effort-indexed for women but tournament-style for men.
Given that, you could add a lot of money a pool, but if it's divied up tournament style, it will make minimal difference to the median wage. Since men are more likely to be working in tournament-style systems, it would make sense that adding more money to each pool will affect women's median wage more.
*Brief primer: median means line up everyone in order of the trait of interest and select the person in the middle. This is distinct from the mean, which adds all the numbers together and divides by the total number of people. The median can be the more relevant in certain cases involving data with long tail distributions.
**See this paper for both a general overview of the research, some specifics on how to counteract the effect, and really fascinating info about how the menstrual cycle affects willingness to compete. It's worth noting that the pre-menstrual period, the yimr that makes women unfit to be president or CEO or head surgeon, is when their hormones profile most closely resembles the typical male profile. I'm just saying.
