Hacker News Reader

Comment by noelsusman | Hacker News Reader

First, when you control for everything there is still a gap, it's just smaller. 94 cents on the dollar is nothing to sneeze at.

Second, when you do that you're sort of controlling the problem away. Women don't seek more work/life balance because they're not as crazy as men. They do it because they're expected to carry more of the burden of raising a child. The literature is clear that the wage gap is relatively small until children enter the picture. After that the gap skyrockets.

This is why the main practical goal of every wage gap activist is equalizing the burden of having a child. It's why the number one policy issue is universal mandatory parental leave for both genders.

View on HN

Well there are other factors you need to control for too, men are more likely to try and choose higher paying jobs because social status and attractiveness to potential mates is impacted by a man's wealth. This is not the case for women, studies show that men are indifferent to a women's economic status when it comes to mate choice and focus more on physical appearance.

There certainly is sexism in the workplace, although the court ruled different I think Ellen Pao was a victim of this. That said, I don't know if it as prevalent and widespread so as to lead to meaningful and statistically significant wage gaps when all the factors are considered. Certainly the evidence for this sort of widespread institutionalized sexism is lacking in my mind.

Still I think the conversation here is more that the wage gap numbers are not as significant as they are often presented in discourse about gender equality

I'm just going to point out that status-chasing is controlled for in what he described. The .95 cents on the dollar figure appears when you compare populations of men and women in similar roles with similar performance. You're confusing wage gap 1 (pay for performance) with gap 2 (seniority/glass ceiling).

Saying you need to control for things that have already been controlled for doesn't leave a particularly good impression.

I assume you're refering to the CONSAD reserach study which found a gap of between 95 to 93 cents on the dollar. That study did not in fact control for status chasing - nor did it compare men and women in similar roles with similar performance, they also found that the wage gap was not statistically significant enough to warrant corrective action

Yes it did. It controlled for hours worked and industry. That is, it controlled for men working harder and choosing higher paying jobs.

What makes “that the wage gap numbers are not as significant as they are often presented” an important conversation for you?

To me it seems like an uninteresting tautology. Like these others:

“False statements about immigration numbers are often presented in discourse about naturalization policies”

“Misleading financial figures are often presented in discourse about what stocks to buy.”

It’s like... yah. So?

All discourse is full of garbage. If you are interested in the truth, you weed through that and evaluate the numbers and, according to best practices, find a good estimate.

When you do that, you find there is a substantial gender-attributable pay gap in many countries in many industries.

Why that is and how we change it seems like a worthy discussion.

I am curious to why you think the existence of people with a poor understanding of the facts is a worthy topic of conversation.

There will always be people in “the discourse” saying wrong things. If that’s all it takes to halt the important conversation about how women get pay equity, then you have yourself a surefire tool to ensure that conversation never happens. Or at least that you never have to participate.

Well in this case the people who are wrong include most of academia, most of the popular media, and the former president of the United States. They were all touting the 77 cents figure I think that's a pretty big deal.

You should read some actual research if you truly think the broad academic consensus is that women make 77 cents on the dollar and that's the end of the story.

I see much more media disputing the 77 percent figure than presuming it. Don’t you?

(And as a side note, the existence of a raw 77% gap is factually true. Media would have to argue that it’s entirely due to sexism for them to step into territory you object to)

https://web.stanford.edu/~diamondr/UberPayGap.pdf

"Female Uber drivers make 7% less per hour than their male counterparts—even though the algorithms that determine pay for the ride-hailing service are gender blind, according to a multi-year study."

If you read the study, they find the gap is fully explained by two factors: 1) men drive faster 2) men work more hours, and therefore learn the platform faster.

If their conclusion is accurate, I am puzzled as to what solution can be proposed. Should we lower speed limits or cut mens’ hours?

Wouldn’t driving fast earn you less? Cab drivers for years have been accused of taking intentionally longer routes to inflate their pay. Isn’t taking a long route effectively the same as driving slow?

Apologies if I don’t understand the uber cost structure. Is there not a time component in the service charge similar to cabs?

The biggest component of the price is distance traveled so driving faster means you make the almost fare for a trip and finish the trip in less time so that you can complete more trips in the same amount of time. There is also a time component, but it is pretty small, so it wouldn't be worth the driver's time to take more time without also covering more distance.

Taxis often take this even farther and make the time component only include time where the taxi is going less than fixed, fairly low speed (in NYC it is 12 mph), so slowing down from 25 mph to 20 mph wouldn't gain them any extra money on the fare but would cost them extra time.

I remember someone argued that the algorithm must be sexist because it rewards aggressive driving behavior (although in reality the difference in speed was 1-2 mph IIRC) and it should thus pay women more to account for their different driving style.

>control for everything

The studies do not "control for everything". For example I have never seen

1. Control for the much higher levels of sick days taken by women.

2. Control for the product of hours worked and years worked. Women work fewer hours and spend more years out of the work-force. This has a multiplicative effect on "hours of work experience"

Additionally the power of "controlling for X" is vastly overstated. In reality it is only reliable when the different factors are close to orthogonal, have linear relationships and are not causally related. Typically none of the above are true.

Reply on news.ycombinator.com