There was someone on HN a while ago who hired several engineers, and said that his company pays women less than men because he makes the same starting offer to several people, and almost all the men asked for more money, yet none of the women did. So the offers the company made were "fair", but the end results were very skewed wages.
There are entire books about this subject, such as "Women Don't Ask: The High Cost of Avoiding Negotiation"
Companies could fix this overnight by disallowing negotiation (Didn't Fog Creek do this?), or making all employees salaries transparent (countries like Norway do this, and we do this for C-levels often in the US, just not plain old employees).
I suspect nothing institutional will change though because it would ultimately cost companies money. They benefit too much from workers who negotiate poorly.
If this was a conversation about whether a new drug worked, the conversation would be over.
In fact given that the article we're allegedly discussing is about how there is no pay gap in tech, there's reason to think that this sort of outcome may not apply across the board.
or maybe this?
1 - more roleplay
2 - perceptions of women in management, says nothing about earnings, concludes people may actually prefer female managers.
3 - Says women who alter their behaviour in certain ways actually do better than men.
I hope by mentioning it that other people might do similarly.
Employees, take it to heart.
Employers, you have been warned.
A great rule of thumb is if people aren't saying no you didn't ask for enough.
The first time I negotiated I shot way over, apparently even the lawyer didn't make that much, my reply was that the lawyers work usually wasn't scalable across multiple clients and that any code I made was a capital good that could be resold again and again generating value in the company for years, then I pulled out a sales spreadsheet showing how the code I wrote for various product lines brought additional revenue to the company and then multiplied by the average length on the contract. (Make a really big number and ask for a small part of it).
Needless to say I was so successful in my negotiations that I was no longer allowed to have access to sales data.
The best part is once you negotiate a good salary you can use it in future negotiations as an indication of skill to get even better salaries. (eg. Does the competitor want to steal the $50K per year coder, or the $250K per year coder?)
I think your best course of action is to gather up salary data based on the job responsibilities of the job you're applying for. There are lots of online resources for researching salaries and you'll be able to grab data points like median, and quartiles. You can then massage that number up or down based on extra criteria like your unique skills, additional education, etc.
In any negotiation, finding objective criteria that you can show will always put you in a stronger position than creating semi-arbitrary bargaining positions.
The other best advice I could offer to would-be negotiators is to avoid setting any initial price until it's no longer reasonable to avoid it. If you have to name a price, name a price you're going to be very happy with.
Objective criteria a horrible way to negotiate, look up the accounting term 'good will', you create 'good will' by not using objective criteria.
Name a price they will say no to, not one you're happy with, if they counter below your BATNA, tell them the mininum you'll accept is something above your BATNA, give them a timeframe to come correct, get up from the table and leave.
I could careless how badly the average, or 5th percentile employee is paid. If they want to retain me they'll a price better than my BATNA.
BATNA isn't a bargaining position, it's your fallback option in case bargaining fails. I think you're talking more about your "bottom line", which is the lowest number you're willing to accept. A good negotiator goes in knowing his/her BATNA and bottom-line, but they're separate concepts.
I also completely disagree with you disregarding objective criteria. I used my recent purchase of a 2-year old vehicle as a way to gain some extra experience in negotiating and work on different approaches with two separate dealers. My initial approach with the dealer I ultimately bought from, was to derive my bottom-line number from objective criteria, but to show no strong evidence to how I got the numbers I was offering. I got nowhere with that approach. The sales manager didn't even come out to stop me from leaving when negotiations with the salesman got nowhere. We were only $1,000 from my bottom-line (which he didn't know) and about $1,800 away from our stated positions.
I came back a few days later with reasonable evidence based on market data of the particular model I was buying, as well as reasons why I'd go spend more money on another manufacturers' similar, but nicer model in the same class. They ended up coming down the full $1,800 after I was about to walk away again.
This works time and time again. Unless you're in a position where you completely don't care if you complete a purchase/sale/agreement, then holding onto unrealistic positions will never get you very far. My father-in-law likes to buy and sell stuff, but he pretty much only deals in unrealistic offers, and he rarely ends up buying or selling anything, even though he spends a decent amount of time looking.
Do not tell them what your previous salary was.
Whatever their number is, ask for $15k more
If they are a strong no, try and trade off that $15k for increased benefits in some way
If they are a weak no, lower your ask.
They will almost certainly give you $5k above what they quote, maybe $10k
And keep in mind: if you're a typical software engineer going through a typical full-day on-site software engineering interview, they've dumped the better part of $1000 worth of dev-hours just to interview you (double that if they flew you out!) so seriously, asking for an extra $5k is no skin off their back. Just do it.
-- signed, underpaid female. :-)
A small example of this is when the US froze wages during World War 2, thus employers started offering perks like company-paid health insurance. That has since evolved into the mess we have now - that sure worked out great for everyone, didn't it?
Gets trickier once if you're immense and have very specific skillset requirements.
Fix what? What are they "fixing"? They seem to be fixing the problem of people being able to do better than default. Exiting times we live in...
I think that instead of disallowing negotiation at all, we could look to university job courses (the ones that currently teach resume and interviewing skills) to teach negotiation skills and techniques. It would be something that any student could find valuable. Not only would it make young people entering the job market more comfortable with the idea of negotiation, but it would make people on the company side of things more familiar and comfortable with the notion of women negotiating (correcting that prejudice, instead of accepting/ignoring it as the Bergeron-solution does).
If that proves unsuccessful, then less "personal" negotiation processes could be explored before the idea of negotiation was written off entirely. Perhaps sending a form letter with every offer letter saying something along the lines of "I'm sorry but I have to decline [position] because I've received a better offer of [$] from [company]" could work. Some sort of more forumalic approach to negotiation that reduces the human element (and therefore hopefully, the opportunity for human prejudices to manifest. I think that it is best to combat the prejudices however, not work around them).
Much of at least the interview portions of those courses involved splitting off into groups and grilling each other for several hours a quarter. The idea was to help people get over the jitters and feel comfortable bragging about themselves to others. I think that at least in that capacity, the course was effective.
Some people did not take those courses seriously, but that was likely because my university was heavily a co-op/internship school, so everybody taking those courses junior/senior year had already had dozens of interviews (1-15 each round, with up to 3 rounds for each co-op cycle, and either 2 or 3 co-op cycles. Granted interviews for extended internships are not quite the same as regular interviews, lacking most negotiation particularly).
I'm not sure it will work, but it seems to be the wage-side proposal of what e.g. Saturn is trying to do with cars: just say, look, this is our price, take it or leave it. Rather than the traditional car (and salary) pricing model where the real price you're willing to agree to is hidden, and arrived at via a negotiation process. That often results in whoever's better at negotiating getting the better deal. Now if it's a negotiating job that might even be a good thing, because the better negotiators are more valuable to you. Or if it's a job where the skill you're looking for correlates highly with being good at negotiating. But if that's not the case, just making the final offer of what you're willing to pay up front, rather than starting lower and negotiating upwards, could be better idea. Then rather than negotiating per-candidate, you just adjust your offer policy until it's high enough to get the kind of candidates you want.
I'd also suspect younger people negotiate less and people who are newer in the cycle of generational wealth (don't have someone helping them understand the process). Maybe since there are fewer women in engineering and friendships tend to generally be segregated by sex they don't have people to talk to about this kind of negotiation which further compounds the issue.
And don't underestimate the effect of networking among talented professional women either. Another friend of mine casually mentioned she was uncertain about a new boss during a ski trip and she immediately had solid options in better environments, one of which she took.
Other industries may lag, but it's fairly well established here that employers in technical and engineering fields are stuck with an increasingly difficult challenge of attracting and retaining talent. If you're not doing this proactively across the board already, then I'd love to know why.
When asked by friends in my industry what I make, I openly tell them if I am close to them, but I guard that information much more closely with anybody that I know to be making much less than me in other industries.
Wages are public in Sweden (e.g. http://ratsit.se) and that doesn't make wages in Sweden high or raise faster. Salaries are, on average, much higher in neighbour Norway than in Sweden.
So there are other - more important - factors at play than knowing the competitions salaries.
Not really a world I want to live in.
Taking one decision in isolation, it might be better to pay someone $10k more to keep them from leaving. But it might be better structurally to come up with a different policy for allocating raises than "allocate them to the people who negotiate most hardball". Unless, maybe, it's a sales job, where being good at negotiating is precisely the skill you want to reward.
Then your initial offer wasn't made in good faith. No-negotiation forces the person making the offer to actually consider what that hire would be worth, and offer accordingly.
As an employer, how do I calculate this ahead of time? It's next to impossible. This is not a hard science. We may find that gender biases exist here as well, as worth is very subjective.
If you mean "expected revenue potential for the company" then they pretty much have to try to lowball to some extent, because an employee that makes exactly what they bring in is not a net asset.
--edit-- I probably sound a bit pompous. Maybe there is a way to calculate this stuff ahead of time, I genuinely have no idea. I know in sectors unlike ours, where there is a shortage of positions rather than a shortage of people, that non-negotiable salaries are indeed used. In that situation you can easily say "We need a person to do X, this pays Y" and you'll get lots of worthwhile applicants.
> "As an employer, how do I calculate this ahead of time? It's next to impossible. This is not a hard science."
Perhaps if the prospective employer guesses the prospective employees worth incorrectly, the prospective employee could politely correct them. ...wait, that's negotiation re-invented. ;)
Then how do you decide how high to go in negotiations? Its exactly the same thing, except you start there.
You discuss it with the prospective employee. You make a guess at what they are worth to your organization. If they disagree, they explain why they think that they are worth more. Explanations may include new information about what other companies in the market believe that person to be worth.
Without negotiation, you make a guess, the prospective employee tells you that they disagree.. and you're done. You've wasted that entire recruiting and interviewing cycle.^ It doesn't matter if both parties want to work together; without negotiation if the hiring party makes an incorrect guess then they do not get to work together. Any correction or attempt to re-guess falls under the umbrella of negotiation. Perhaps they could reapply, go through the interview process again, and have you attempt to guess a good salary again, but that is nothing but inefficient negotiation.
You could publish the offered salary before the beginning of the interview process, making that information known during recruiting, but then any attempt to communicate with failed recruiting attempts would be again another form of negotiation. Recruiter contacts a developer, says "I've got an position for you to look at, it offers $120k/yr" and the developer responds "I am worth at least $140k/yr"? If the recruiter relays that information to the hiring party, the hiring party would not be able to act on that information and adjust the offer because that is just pre-interview negotiation.
"Make the initial guess the maximum you would be able to pay somebody in the most extraordinary of situations" is very obviously not a viable strategy. I hope I don't have to explain why.
^ Note that while this is bad for the prospective employer, it will often be even worse for the prospective employee, who in many cases will be looking to become hired as soon as possible as they are either in an unsatisfying job (in which they are forbidden from asking for a raise...) or worse, unemployed. Without any negotiation, an unemployed person would be placed in the wildly unfair position of being forced to immediately settle for an unsatisfactory salary, or spend more time without income looking for and applying to another job.. hoping that that one doesn't accidentally lowball the salary. Most unemployed prospective employees would tend to settle immediately since some income is generally better than no income. In this way, bans on negotiation screw over individual employees and place unemployed people in an even tighter bind than they are in currently.
On a more serious note, the way we choose a salary is to look at various similar companies in the area and get an idea of what a competitive salary for the position/experience level is. If people are happy with this (they almost always have been, at least since I started) they take the offer; if they aren't, they don't. Its pretty simple, and avoids the issue of paying people for being good at negotiation, which as far as I can tell has no correlation with being a good developer.
Our companies can safely refuse to negotiate because 1) they can let other companies in the area negotiate to set prices for them (without negotiation, determining the value of an employee becomes massively inefficient, mostly at the expense of the unemployed, who will tend to accept positions lower than they could negotiate for, lowering the overall regional salary), and 2) because generally for standard dev positions, they are aren't looking to hire a specific person. Any person that fits the roll will do, and there are plenty.
You want a college grad to start training on your software stack? Well sure then, figure out the regional salary for recent college grads, offer lots of college grads that salary, and take those who accept. So long as you have enough grads to choose from and eventually the positions get filled, who cares right?
You want recent Standford grad Joe Smoe who recently made waves online demoing that really cool shit that accidentally compliments [Top Secret Project]? Without negotiation, fat chance. There are no regional stats for Joe Smoe's expected starting salary. You either negotiate or your gamble.
If you are finding an employee to fill a position, you can get away with no negotiation if others in the industry still negotiate. If you are creating a position for a person, you have to negotiate.
If every company were forced to acted as our companies do, and for every position that they were filling, it would be catastrophic.
No hiring process is perfect I guess.
Don't know if I agree with this. Maybe I can hire a somewhat experienced female engineer for $110k, but I know that sooner or later a recruiter or friend or GlassDoor query will tip her off that her peers are making 20% more. When that happens her morale will suffer, or she'll quit, or both.
When I've hired candidates who truly did not know their own worth, I made them offers above their asking to be inline with market rates.
The thing we need to do is inform folks that screwing women (or anyone) over comes back to bite you. It isn't advantage to "take advantage" of anyone. The real costs are non-obvious and may be somewhere down the line but they do exist.
Have an(other) upvote. :-)
If an explicit salary negotiation is banned, people will find other things to negotiate instead.
You will likely lose potential hires because you refuse to budge on salary. It is an inadvertent [negative] signal if, for example, the company is not even willing to add $5 to a salary. That would definitely be a deterrent to me.
Finally, this potentially penalizes all of those who do negotiate salary to attempt to appease those who won't. It sends a message that (in this context) women are weaker and cannot possibly learn to negotiate. A far better solution is to change the culture to help all, men and women both, feel comfortable and informed when negotiating terms of a new position.
Also worth stating: negotiation is a skill that helps throughout life, and rather then attempt to bury it in a specific context, it is far better to raise everyone up so that they can reap its rewards elsewhere.
Never really understood this "statistic" still dont. I do wonder if things have changed? I have a couple of girl friends in tech and have to do a fair bit of ego and confidence building before they work up enough balls to tell their bosses they are bored and need new responsibility and a raise to boot. Most guys I know just ask for more money in passing til it happens or they leave.
Disclaimer, I dont know shit about the inner workings of the female mind all I have is anecdote and the up most respect for girls/women in tech or otherwise.
There IS a gender gap, but with it a matching skills/abilities gap.
I have experienced this first hand with dozens of women. A rare couple of them were top notch developers. They were ambitions and had a real passion for development and technology. So did about 20-25% of the men I worked with (vs less than 10% of the women). The other woman and men had base skills, but rarely showed ambition or passion to learn new thing and improve their skills, habits, tools, ETC. They were kind of just grinders. They never excelled or impressed. They weren't the kind who would tinker on the weekends with OSS or read development or tech blogs or books when they had free time.
The top notch women were promoted and made similar salaries as the top notch men. Overall though, the women lagged behind the men percentage wise because of the reasons mentioned.
I would be curious to know if others have seen this pattern. This is based on my 15 years in the software industry.
I agree. The fact I have a penis attached to me means that I don't get to participate equally with all conversation partners. So while I'm all for equality, I keep quiet about things I both disagree and agree with, because my opinion is branded patriarchal bullshit a priori.
Such people aren't interested in having a constructive discussion in the first place, make that perfectly clear in the same sentence as in which they express their opinion, and then proceed to act indignant if the other side responds in kind...
When Adria Richards tweeted that guy's picture in DongleGate, he got fired, and she got called a hero and fashioned herself Joan of Arc (no really). When Feminist Software Foundation appeared on Github, Ashe Dryden tweeted the list of people who starred the repo and suggested their employers should be told. Shanley Kane's medium.com posts are full of blaming "white men", "privilege" and "patriarchy" for this or that, with usually nothing to back it but anger and indignation.
To me, the phrases "twitter witch hunts" and "being branded patriarchal bullshit" seem entirely accurate.
Or would you call that being interested in a "constructive discussion"?
That is just incredibly presumptive.
A stereotype has no role to shift the status quo because the token exists to demonstrate that which is to be corrected.
If you're a good token, you're just being used to buy another round.
So studying more and harder and partying less is considered a privilege?
Don't women get better grades than men on average?
Something during that birth-18 age range also seems at least partially responsible for gender disparities in major enrollment. Male:Female ratios for particular classes in CS departments seem to get worse as each year (a class that starts off 1:4 might drop down to 1:20 by senior year), but the fact that they start off bad suggests that something is going wrong earlier in the process, likely during highschool.
Identifying exactly what is going wrong in primary/secondary schools is paramount, but I see rather little to address it. There are many extracurricular programs designed to correct gender imbalances that appear during primary/secondary schooling, but those programs are addressing symptoms, not attempting to identify the cause. What exactly is it about highschools that make boys struggle with literacy more than girls? What exactly is it about highschools that set the stage for disparate university and program enrollment numbers?
If I had to venture a guess, without any sort of evidence, I would say that zero-tolerance policies probably affect teenage boys more than teenage girls, and therefore probably play a roll in teenage boys receiving a poorer quality education (particularly since punishments typically involve being removed from lectures... rather than receiving supplementary instruction.) Areas where self-education is viable and common, such as CS, may not be hit by this as hard.
Among social justice academics, Asian success in majority white countries is largely ignored. It upsets the Marxist narrative of oppressor/oppressed, where the oppressors are white and the oppressed are everybody else. In academic circles, "Asian privilege" is considered a myth and an attempt to derail a conversation.
> Social justice is largely a white male self-loathing thing
I guess my point is that white people are bona fide official second class citizens, as classified by taxpayer funded research. Asians aren't given special privileges like other minorities, but neither is their success considered evidence of oppression.
Basically, in a study of federal workers, a 7% unexplained gap exists between male and female worker's wages when controlling for occupational, experiential and educational differences, among other things.
Especially relevant data is on pages 84-86.
(edited for clarity)
Discrimination is an important discussion in general, but most of the time it's extremely hard to prove that someone didn't get a job because the hiring officer was prejudiced. This wage gap stuff is easier to quantify and is a visible problem to solve.
I said there's a 7% unexplained gap in salary amongst government workers, implying that even controlling for most factors salary discrimination still exists.
But why would we ever expect them to be 0% different? Recall that natural unemployment is ~3%. The market and real world are not perfectly efficient and making them perfectly efficient would make them less efficient in the process.
I wouldn't expect uncontrolled wages to be entirely equal. I would expect that eventually controlling for employment history, experience, education etc. would make the difference approach zero, though. I think that'd be good.
I don't think it's hard for people to imagine that there is less variance in starting salaries. Reading into this data and thinking it applies to anything beyond that is a guess and should be treated as such with that guess being tested in future studies.
Even if you demonstrate there is no wage discrimination over the course of people's careers, it doesn't account for the huge gender gap in the industry.
"7% gender gap in starting tech salaries" would have been a fairer headline.
I wonder how how it would look after that age range.
I am guessing from my anecdotal experience that most professional women usually wait until around that sort of age to have children, and how much having children is a factor in gaining higher salaries.
In addition to that, I am guessing that people early in their careers are less likely to negotiate salary. I certainly was. Now I have a fair bit of experience, I can see that I am more knowledgeable and valuable than some of my colleagues, and so I am in a position to negotiate, with good reasons to back it up. Graduating in the dot-com bust, I was happy just to get a job after nearly a year and a half of searching. I didn't feel I was in any position to negotiate then.
The problem with controlling for all those things is that you leave out other factors; for example, if men were preferentially hired over women, that wouldn't show up in this data.
Such a case would definitely be discrimination, it just wouldn't be germane.
My wife was out on unpaid leave for nearly two years when we had our first child. That meant that she missed two salary scale steps, which are earned based on satisfactory performance over a period of at least 1,500 paid hours/year. In her case, that means that a male who started the same day as her makes approximately $2,500 more than her, assuming the same career path. (She started at level-x and is at level-x+y.)
For a nurse, or techie, salary is driven by the market in most cases.
That's how it works, if your family is more valuable to you than your career, then you'd be happy with getting more family at the tradeoff of less career - and that explains why the positions at the very top get a disproportionally large number of divorces/etc - if someone is consistently neglecting family because of their career, then they do get to the top easier than someone who'd sometimes actually prefer going home to their kids.
It is interesting to discuss whether there is more pay equality in STEM fields, specifically engineering, because females in these fields are also more comfortable with (and better at) negotiating higher salaries for themselves. At the moment there isn't research to suggest this but it could be a contributing factor.
"When controlled for all factors other than gender, the earnings difference between men and women is about 6.6%, something most people don’t know."
Strip the gender politics out of it and the answer is very obviously no.
Consider two women, A doesn't want children and B does. How do you think A will feel having worked hard for a year while B was on maternity leave, only to find that B's seniority etc has kept up with zero effort or contribution to the company on her part?
Regression analysis was used to estimate wage differences, after controlling for the following choices and characteristics: graduates’ occupation, economic sector, hours worked, employment status (having multiple jobs as opposed to one full-time job), months unemployed since graduation, grade point average, undergraduate major, kind of institution attended, age, geographical region, and marital status.
That's really misleading. Some of those variables at least might have gender baked into them. For one example, let's say that women are more likely to have longer unemployment spells after college. (I don't know if this is true.) Then there could be zero gender correlation in the model not because there is none, but because that input factors it out.
With 11 variables, some of those being categories, you're going to almost certainly have "small data" issues on real world datasets, where the inputs (empirical "design matrix") are not even remotely orthogonal. So pardon me if I don't buy it.
The issue isn't that companies explicitly choose to pay women poorly. It's that many tech companies end up with shitty cultures that exclude women, which leaves them out of the loop and causes them to end up on shitty projects. It doesn't happen by intention. The leadership of most tech companies is criminally negligent on a lot of cultural issues, but not explicitly sexist.
ETA: it's also worth pointing out the "women don't negotiate" argument. I think this is because they get worse results. My wife is a better negotiator than I am, in terms of skill level, but whenever we have to haggle on the price of something, I go out... and I often get more, not because I am better, but because a 6' male is more intimidating.
Wake me when there's justice.