Do Women Avoid Salary Negotiations?

The subtitle of the paper is Evidence from a Large Scale Natural Field Experiment and the authors are Andreas Leibbrandt and John List.  Here is the abstract:

One explanation advanced for the persistent gender pay differences in labor markets is that women avoid salary negotiations. By using a natural field experiment that randomizes nearly 2,500 job-seekers into jobs that vary important details of the labor contract, we are able to observe both the nature of sorting and the extent of salary negotiations. We observe interesting data patterns. For example, we find that when there is no explicit statement that wages are negotiable, men are more likely to negotiate than women. However, when we explicitly mention the possibility that wages are negotiable, this difference disappears, and even tends to reverse. In terms of sorting, we find that men in contrast to women prefer job environments where the ‘rules of wage determination’ are ambiguous. This leads to the gender gap being much more pronounced in jobs that leave negotiation of wage ambiguous.

An ungated copy I do not see, does anyone?


John List does awesome work -- his dictator game experiment in which he includes negative values is pure genius -- in one simple experiment, he refuted tons of "behavior economics" nonsense

"Yet, a result that holds regardless of action set composition is that agents do not ubiquitously choose the most selfish outcome"
- from the abstract, List, 2007, "On the interpretation of Giving in Dictator Games", Journal of Political Economy, Volume 115 No 3

The crux is, it was not his idea....

But the same could be said of Darwin ... i.e., "it was not his idea" (it was Wallace's)

A variable I suspect is significant in the gender pay gap, but I haven't seen any data on, is the frequency of job hopping of men vs. women. The best way to give yourself a raise is to go to a competitor when you already have a good job. This maximizes your leverage in the negotiation. If men job hop more frequently than women then they may end up negotiating higher salaries -- not merely because they might be more aggressive negotiators -- but because they have more leverage with which to negotiate.

When I'm up for a contract with a variable advertised pay-out, I press interviewer for details as/try to nudge the price upwards. I assumed everyone did this.

Is it gated for everyone else? It's not for me on a campus network. I can throw it up on scribd or something if folks need.

Women prefer to tell their husbands to get into salary negotiations. My wife spends almost as much time nagging me about my income than on increasing hers.

Maybe she saw your OKCupid profile? Man up, dude.

A sense of humour.

Pay up, Silas.

I thought it was settled that there is no gender pay gap when considering equal type of work and benefits.

It's my understanding that unfortunately even when you control for equal type of work and benefits there is still a substantiation gender gap.

Suppose there is a gender gap of 15% of salary. I should be able to set up a service company - for example a law firm - employing only women, and I'll be able to undercut all the competition by 15%. The fact that nobody has done so would imply that there is no gender gap.

The supply of firm entrepreneurs is not infinite, and principal-agent problems permit agents to indulge in their prejudices.

Still it seems that private equity companies would arbitrage this opportunity by making a priority of replacing the men with women to as great an extent as possible. Private equity would have the freedom to search for the ideal firm in which such an arbitrage was possible, so it stands to reason that they would be able to find one.

But thinking this way gives me another idea: I once worked for a Fortune 500 company that made a priority of promoting women, because they were under-represented in management. (There was the usual petty bitterness by some of the men complaining about how the women had it easier.) But making a priority of PROMOTING women resulted in this: women were more likely to be in over their heads at the position they were promoted into. (At least accept that as a hypothesis). So the women working the same positions as men often, from a merit point of view, did not deserve as much money as the men at the same position, because they had been pushed up the ladder much quicker.

Private equity has its own special set of divergent interests, and buying a firm to dismiss half its workforce would lose much of its organizational capital.

Rather than employing women because they have a tendency to work for less than men, firms find it easier to attack the problem of how to pay people as little as possible in a more head on fashion and simply hire people who work cheap.

So... they hire only women?

"they" do, such businesses are called kindergardens.

@andrew m: the education system is running with "cheap" labor force =) 18.3% of elementary & middle school teachers are men

Cliff, if a law firm decided to only hire people who where five foot four or less, they would not end up with an all female staff even though women are shorter than men on average.

Blame the women

For their

Gender gap

In wages.


What a jackass. I'll make this as simple as possible. Women don't get paid as much (to the extent that that is even true anymore) because they are worse at everything, except one or two things that are generally not done for pay. Yeah, it's true. Get over it.

lol this was awesome

awesome in its wrongness. (I was holding out hope it was satire.)

In a whole range of occupations men and women can be equally productive. Now they may have different comparative skill advantages but gender is rarely a real impediment to getting the job done. One nifty difference I have noticed is that men tend to think more highly of their skills than women. I have heard more men than women talking frightening beyond their skill level (and they still sound authoritative) and I personally have felt ill over hard earned bonuses I received for my work. Clearly men and women are different. I don't think they should have equal pay and positions as a rule, but I do think that women's skill set should be recognized as valuable to the firm and women should be part of the decision making process. Again it's not about equity it's about productivity.

It was ignorant hyperbole and bravado.

Your point seem reasonable, but the dialogue was unnecessary.

Yesterday served up more examples of smart, successful men whose careers were stunted in part by their inability to talk reasonably about "women's issues" (which actually affect men too). What you believe in your heart if hearts is most important, but if you cannot put it in sensible words and actions then it devalues quickly. I will be thrilled when such dialogues are unnecessary.


I'm not gonna defend the blockheads in Missouri or Indiana, and Republicans clearly need to do some retooling, but I'm wondering what "women's issues" you seriously think were at stake in this election. What policy changes do you advocate that are more likely with Obama in office? How decisive were these issues to your vote?

The popular vote was close to a draw, and the People's House is firmly Republican. Yes, there are trends, but there are also ebbs and flows sometimes mistaken for trends.

44% of women somehow overcame "women's issues" to vote for Romney. Since 59% of whites voted for Romney, prolly more than half of white women did. Racism? (It's just as silly to posit a "men's issues" problem for Democrats IMO.)

I will also be thrilled when such dialogues are unnecessary.

"One nifty difference I have noticed is that men tend to think more highly of their skills than women. "

Undeniable, and probably one major reason why men are more willing to negotiate. Women are just happy to get an offer, whereas men think "I could do better."

Wasn't there a post last week saying that men tend to overestimate how much their women friends are attracted to them, and vice versa for women? Same concept.

L, Pretty scientific response. What were your controls? May you have a female supervisor with whom you can share your views.

There is a gender gap in wages, but women earn more. Stop having emotional responses to these issues and look at the data.,8599,2015274,00.html

You are really going to cite a study from 2010?

I forgot that all papers from 2010 were invalidated. Or that any paper from before 2010 is also invalid because, I mean come on, are you really going to cite a study from before 2010?

Is this like, "Blame the Redskins for the Redskins/Falcons gap in wins. Priceless"?

Are you suggesting there is no reason for investigating this possibility? Would you rather remain ignorant?

Not to mention the lack of any actual wage gap.

Their findings sound plausible to me but point to so many other questions. Why don't women negotiate? Are ambiguous wage structures just quiet forms of wage discrimination against women? (Economists are almost always the last to learn about regular patterns in behavior, so maybe male bosses have used this for a while to promote favored groups.) And could this also explain gender differences in careers...women minimizing negotiation and conflict? And what does it mean? Are women ill suited for competitive work environments or do they just need more clarity? And finally how do they get these studies through human subjects review?...I doubt the job applicants would be pleased to learn the postings were almost all fakes. Interesting paper.

I think, according to the premise of the article, they need to hire a male negotiating agent.

You could make a fortune just by taking half of the wage difference, say, by charging 10% as a negotiating. This will also reduce male unemployment: which, in itself, may tell you something: perhaps because males are more willing to negotiate, they have higher unemployment. I can see where this is going though: soon Tyler will recommend that to introduce wage flexibility, men should hire female negotiating agents.

You heard it here first.

I think we're a ways away from such proposals...and yet, making the wage negotiation explicit seems like the most natural first step.

The NBER copyright allows 2 paragraphs of quoted materials without author permission, so these are my selections from the introduction. (Not cool to only have a gated copy up of an econ working paper. Plus the abstract is not very detailed either.)

page 2-3

"In total, our job advertisements were posted in nine major US cities. Nearly
2,500 job-seekers responded to our initial job postings. Overall, we find data patterns
that share some similarities to the literature. First, we find that when there is no
explicit statement that wages are negotiable, men are more likely to negotiate than
women. However, when we explicitly mention the possibility that wages are
negotiable, women are more likely to negotiate. The first result is consonant with the
literature, but we find smaller effects than what is typically reported. Interpreted in
light of the meta-analysis of Stuhlmacher and Walters (1999), our tentative
conclusion is that in environments in which negotiations are impersonal, gender
differences in negotiation play a lesser role than when negotiations are face to face. A
tentative implication is that as modern economies evolve, and a greater number of
transactions are completed impersonally; gender differences in bargaining might play
a lesser role in wage setting.

A second result is that the gender gap in applications is much more pronounced
for jobs that leave negotiation of wage ambiguous. Interestingly, this result is driven
by men preferring jobs where negotiation of initial wages is ambiguous rather than
when negotiations are expected. This result squares well with the literature that shows that men perform much better than women in environments where negotiations
are ambiguous (see Bowles et al., 2005). It is also in accord with psychological
theory (e.g. Mischel, 1977), which argues that situational moderators such as
ambiguity systematically influence gender differently."

I think what it is is something different: peer influence and common knowledge. Because there are few women higher up on the ladder, there is less history of women earning more. If women knew, just as some men know, what their respective cohorts make, they would negotiate aggressively as well. Men may talk, or disclose, comp with each other (bragging?) while women don't. Moreover, I wonder whether women talk to each other about their salaries, while maybe men do a bit more.

I think when women know what comp levels are, they are no different in their negotiation. In a lawfirm, as a partner, I found out what other partners and associates make. When both males and females find out what their counterparts make, I have never seen a lack of women negotiating. In fact, because people know I use regressions, I'm often asked to model what their points or earnings should be relative to their peers. What may temper their negotiation is that some have had more flexible time in earlier parts of their career, and may be in effect adjusting for this.

I think there are things missing in the study. I would be more interested to know if the study: 1) identified the level of knowledge males and females had of each other's comp (I bet males know more than females) ; 2) if there was a wage differential if there was a female supervisor who supervised, for example, males and females below her, or whether there was less of a differential than if there was a male supervisor.

So in other words, this is good, valuable research that should be continued with more resources?

'The NBER copyright allows 2 paragraphs of quoted materials without author permission'
So what? They can claim whatever they like, but if, for example, I am criticizing the content, and using the content to illustrate my points, their claim is meaningless. As their lawyers undoubtedly well know - which never stops lawyers from making baseless claims which are then treated as somehow having legal force.

Fair use is a still living part of American copyright, even if a supposedly 'originalist' Suppreme Court has gutted the Constitution - 'To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries.' Article I, Section 8, Clause 8

It seems obvious that encouraging negotiation by making it explicitly an option is in itself a (very) bad way to negotiate. Can you imagine the Craigslist posting "Price negotiable"- what kind of terrible offers would you get?

My takeaway is that employers should actively and explicitly disclose that wages are negotiable to ensure a level playing field and negate this glass ceiling.

This does suggest a reasonable policy to help combat the wage disparity. But would the employer be harming their own interests by doing so? Would they just end up paying more to employees whom they could have underpaid?

First of all, this is a wage gap, not a glass ceiling. Second, it doesn't really exist. And third, what you propose would not only REVERSE the problem (men then do worse than women, supposedly), but also be directly against the firm's interest. Would you suggest a similar policy in all contract negotiations?

I think the reason for the pay gap is pretty obvious: testosterone.

Men are more likely to negotiate, more likely to engage in conflict to gain an advantage, more likely to jump ships or to move to get higher pay, more likely to sacrifice their personal lives for a career goal. I don't have data for any of this, but it seems pretty obvious if you take off your politically correct lenses.

Testosterone means we can't really legislate away the pay gap, though we might be able to minimize it.

Note for those offended: I think women are better at most modern jobs and I don't think unequal pay is fair. I just think it's reality and I think any opinion on how the world should work ought to be matched by a realization of how it does work.

Injectable/ingestible testosterone is offered discretely to corporate men and women through select clinics.

Would that help?

Do economists avoid election-related posts?

Just doing a bit of testing - it may be that the filters have been tweaked, almost as if someone is not proud of a job they hold.

And here is the term that was filtered repeatedly, both upper and lower case - General Director of the Mercatus Center

Good enough - might be other problems, because really, covering up that relation is another level entirely, compared to never actually acknowledging it.

Other terms used in the repeatedly swallowed post included whosis, .edu and .org

What exactly are you talking about?

Is it something more than academics solicit grants?


There is a flaw in your proposal to make wage negotiations more explicit as a means of leveling the playing field.

In any social situation, there are both implicit and explicit ways of negotiating with others. In fact, many jobs already have explicit, or semi-explicit, means of salary negotiation. This "explicit" form of negotiation takes of the form of the, often annual, performance review in which employees are given an opportunity to discuss their value with the firm. But performance reviews still leave space for more ambiguous, implicit forms of negotiation. Someone may decide, on their own, to renegotiate their salary outside of a normal performance review.

So, no matter how much of a negotiating process you manage to expose explicitly, there will always be implicit and ambiguous means of negotiation. Thus, there will always be an advantage for more aggressive negotiators over less aggressive negotiators that follow the explicit rules.

Posted about this in more detail here but, given that the US Supreme Court noted that Lily Ledbetter was initially paid wages similar to her male counterparts, would it make sense to apply this both to initial salary negotiations and later renegotiations (i.e. raises). Not sure, outside of a union job in which performance and wage increases may be disconnected, how you could resolve the seemingly inevitable ambiguity over salary increases in a gender-neutral manner.

Comments for this post are closed