Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Reelya]

Well, we can go off the actuall Bureau of Labor statistics and reports like the ones I linked. I made specific assertions, which you slammed as bullshit. I then backed that up with reputable sources, and now you've gone full troll rather than acknowledge the data I provided.

My original and only assertion: The Paycheck Fairness Act will have no effect, because it's targeting a portion of the raw pay gap that is miniscule (the unexplained let's call it 6%), and only the portion of that 6% that could be explained by deliberate employer sex discrimination. You'd be lucky to get a fraction of a percent improvement from that law.

And then, from your evidence merely telling women "why don't you wait 1 extra year before having your first child?" adds 9% to their raw wage. Which is perhaps 20 times the size of the effect you'd expect from the Paycheck Fairness Act.

[mainiac]

Look I tried to explain in the "math" post.  Sorry I dont know what else to do.

[Reelya]

What is your specific defence of the Paycheck Fairness Act then? Nothing in that post had any connection to it.

"Sound like a nice thing" is not a defence. "There's a raw wage gap" is not a defence either because myriad factors can be shown to be orders of magnitude greater than direct employer discrimination.

Look, the "1 year difference" first baby thing accounts for almost half the raw pay gap between genders. And I doubt any employer actually gives two shits whether any specific woman had a baby this year or next.

[mainiac]

What is your specific defence of the Paycheck Fairness Act then? Nothing in that post had any connection to it.

If: Bj < aj + uj
Then: Bj = aj + uj - cj where cj>0

So we reduce cj and we get a more efficient market, i.e. one with less discrimination.

[Strife26]

Look I tried to explain in the "math" post.  Sorry I dont know what else to do.

Respond to the point?

[mainiac]

Respond to the point?

You mean by like exactly defining my arguments and assumptions?

[Reelya]

What is your specific defence of the Paycheck Fairness Act then? Nothing in that post had any connection to it.

If: Bj < aj + uj
Then: Bj = aj + uj - cj where cj>0

So we reduce cj and we get a more efficient market, i.e. one with less discrimination.

]
Entirely circular logic, and the massive improvements from tangetial things kind of prove that direct discrimination is not an area you're going to get a large amounts of market efficiency improvements.

[mainiac]

Entirely circular logic,

No, just a basic understanding of the terminology of the paper you provided.

edit well terminology is the wrong word.  They said "unexplained" when they meant "residual"  so it's more "implied terminology".

[Reelya]

www.nber.org/papers/w11240.pdf?new_window=1
http://www.forbes.com/sites/timworstall/2015/10/01/the-gender-pay-gap-is-the-result-of-being-a-parent-not-discrimination/

Table 9 (last column) further highlights the relative importance of family responsibilities versus labor market discrimination by examining the gender gap among men and women in apparently similar lifetime family situations—namely men and women who were never married and never had a child. In this case, the unadjusted gender gap is actually positive—women earn about 8% more than their male counterparts.

This sort of underlines some of the assumptions we make when talking about this, and that those assumptions can sometimes be wrong. Never-married / no-kids women outearn never-married / no-kids men. But never-married women tend to have slightly higher education than never-married men. When that is adjusted for, the gender gap becomes so small as to be statistically insignificant. So when we compare "single, no kids" people you don't even get the 5% residual. Since the "unexplained" amount just went away, I'd say it is actually pretty well explained as collateral damage from the family situation rather than employers being dicks (or, they are only dicks to married women for some unknown reason).

[mainiac]

Generally one responds to a working paper with a meta analysis, not the other way around. 

Yeah, I admit I dont even give a shit at this point.

[Shadowlord]

Spoiler: long maniac math quote (click to show/hide)

Look, I dont know how to say this in english so I'm just going to say it in math.  Suppose you have a model of wages:

W = B1*experience + B2*overtime_willingness + B3*career_choice + a0 + u0

What you are doing is saying "discrimination is u0".  Because a0+u0="5%-7%" you conclude that discrimination is small.  But discrimination is not u0.  Let's make a new variable "D" for discrimination and then propose a model of B1:

B1 = D1*sexism_field + u1 + a1

and for B2:

B2 = D2*elasticity_of_labor_with_respect_to_earnings_gap + u2 + a2

and for B3

B3 = D3*elasticity_of_labor_with_respect_to_adjustment_of_lifecycle_discount_adjustment + u3 + a3

So you've said that "discrimination is u0" but in fact that was wrong.  The impact of discrimination is:

D = W - ( a0 + a1 + u1 + a2 + u2 + a3+ u3).

And I'm trying to point out to you that:

B1>a1+u1
B2>a2+u2
B3>a3+u3

Arriving at a precise estimate of all of these terms is like impossibly hard. For one thing a1,a2,a3 are all going to overlap each other.  And we'd need a lot more models because what I've tossed up here are just the first three variables I thought of but there are going to be dozens of important ones.  Each of them will have the fixed effects residual overlap the others.  It's theoretically possible but I wouldn't want to try.

But what I can tell you is that I'm pretty effing sure that:

B1>a1+u1
B2>a2+u2
B3>a3+u3

How did you come up with all that? What's it based on? What are D1, sexism_field, a1, u1, etc? Why isn't "sexism_field" discrimination but "a1" and "u1" are? etc.

In other news, John Kasich has released an ad which paraphrases the "First they came for the..." quote by Martin Niemöller:
http://thehill.com/blogs/ballot-box/campaign-ads/john-kasich-donald-trump-nazi-germany
Video: https://www.youtube.com/watch?v=qCQhBYEMRQI

[mainiac]

How did you come up with all that? What's it based on? What are D1, sexism_field, a1, u1, etc? Why isn't "sexism_field" discrimination but "a1" and "u1" are? etc.

a1 and u1 aren't discrimination.  sexism_field is.  a1 and u1 are the components of how experience matters after you account for how sexism affects your ability to raise your wage through experience.

Eh, I dont really know how to explain this but the best I can say is that the a terms are personal effects (besides the variables accounted for), B terms are the effects of the variables examined and the u terms are the "unexplained" things about the person.

Spoiler (click to show/hide)

So you have a worker and you have some information about them like how long they've been working, they're degree, etc.  You plug that in and make their "expected" earnings.  You see that women underperform their expected earnings.  Well some of that is due to the things the women do (work less because of childcare responsibilities) and that's the a term.  What's left after that is the effects of sexism.

In the next equation it's the effects of experience.  Experience helps some people more then others.  Some of that is inherent to the people, some are lazy, some are hard working, some have jobs with advancement, some dont.  Some is a product of the sexism of the field.  There's also just the inherent effect of more time equals more experience.  So when we have the component "how much is experience" helping the person, we have all these sub factors, among them sexism.

The other equations are similar, here's the effect of one variable, here's the other differences, here's the general case.

[MaximumZero]

You should probably use something you can thoroughly explain. If you can't explain it, you probably don't understand it, and that makes any conclusion you draw from it invalid. What you're doing is fitting the data into your answers via obfuscation and the logical fallacy of appealing to emotion. which is highly intellectually dishonest, and not the way data works. We aren't dumb people here, and it takes a very convincing, well thought out argument to win if you're going to debate on B12.

[TheBiggerFish]

It makes some sense, but we need names to terms.

[Bauglir]

We aren't dumb people here

speak for yourself, bub