The advice is always the same sentence. Always negotiate. It is not wrong so much as incomplete, because it skips the question that decides how much the asking is worth: whether this employer negotiates at all.
That is not a question about your nerve. It is a question about how the organisation sets pay, it was answered before your application existed, and in most cases you can find it out.
Two different processes go by the same word
Robert Hall and Alan Krueger put the question to a representative sample of American workers and published the results in the American Economic Journal: Macroeconomics in 2012. The survey, part of the Princeton Data Improvement Initiative, was carried out by Westat between 5 June and 20 July 2008. Of 2,513 people interviewed, 1,435 had been hired within the previous ten years, and those are the ones asked how the pay was arrived at.1
"A third of the respondents reported bargaining over pay before accepting their current jobs," the paper reports.1 The rest had faced what the authors call a take-it-or-leave-it offer. Almost a third already knew exactly what the job paid at the time they were first interviewed, which is the signature of a posted wage rather than a negotiated one.1
Two things about that number before it gets used for anything. It is United States data from the summer of 2008, so it is eighteen years old, and the survey's response rate was 17.9 per cent.1 And the authors head off the obvious misreading themselves. Because they sampled workers rather than hiring events, "our finding that a third of the respondents engaged in bargaining does not imply that a third of all hires involve bargaining". Their own estimate is that the true share is "well below a third", since "bargaining is less common in the high-turnover jobs held by younger and less-educated workers".1
So the take-it-or-leave-it offer is not the unusual case. It is the ordinary one.
The split is not random, and it is mostly not about you
The same survey shows where the two processes live. The education gradient for bargaining is, in the authors' word, remarkably steep: 29 per cent among people who did not graduate from high school, 57 per cent among those with professional degrees. Individual bargaining is rare in union and government jobs, which is what you would expect where a published scale exists. Bargaining was more common among minority workers and less common among women.1
Read that as a description of jobs rather than of people. A hospital pay band, a civil service grade and a graduate intake are all built so that the number is not the outcome of a conversation. When a friend tells you they got 15 per cent added to their offer, the useful thing to learn is not how they phrased the email but what kind of employer they were phrasing it to.
One more figure from that survey is worth carrying around. Forty-seven per cent of workers said their employer had learned what they were earning in their previous job before making the offer.1 Whatever else this is, it is not a process in which the two sides know the same things.
One sentence made two and a half times as many people ask
The clearest evidence that the situation matters more than the candidate comes from a field experiment by Andreas Leibbrandt and John List, published in Management Science in 2015.
They placed real job advertisements for administrative assistant roles on city job boards in nine US metropolitan areas between November 2011 and February 2012, and 2,422 people got in touch. Once someone had expressed interest, they were randomly assigned to one of two versions of the same job. Both were told the wage, usually $17.60 an hour. The only difference was whether the employer said out loud that the wage was negotiable.2
Where nothing was said, 9.4 per cent of applicants tried to negotiate. Where the sentence appeared, 23.1 per cent did. Negotiations happened, in the authors' phrasing, "approximately 2.5 times more often".2 Same job, same money, same employer.
The gender result is the part usually quoted, and it is worth quoting accurately. Where negotiability went unmentioned, women negotiated less often than men, 8.2 per cent against 10.6 per cent. Where it was said explicitly, the ordering flipped: 23.9 per cent of women against 22 per cent of men. The authors are careful about how much that reversal can carry. "We cannot say with confidence that women are more likely than men to initiate wage negotiations," they write, "but we can clearly reject the opposite."2
The limits are real. These were hourly administrative roles, the fieldwork ran in 2011 and 2012 in nine American cities, and the asking happened in writing rather than face to face.2 But the design is clean, and one thing survives it. In the condition where the employer said nothing, more than nine in ten applicants never asked. The employer had not told them the wage was fixed. It had told them nothing.
Find out which process you are in, before you need to know
The thing you want to establish is whether pay for this role comes off a scale or out of a conversation. It is a neutral question, it commits you to nothing, and most employers answer it directly.
Ask about the structure rather than about your offer. Is there a pay band or grade for this role, and where in it would this sit? An organisation that works from a scale tends to answer almost by reflex, because the information is already written down somewhere.
Be careful with the answer that sounds like the opposite. "It depends on experience" fits an employer that genuinely bargains, and it fits an employer with a fixed band who lets a manager choose which step of it you start on. Those are different situations with different amounts of room in them, and the phrase does not tell them apart. One follow-up does. Is that a step inside a published band, or is the base itself open?
The advert usually leaks it too. A single figure, or a band with a grade number attached, points at a scale. "Competitive" and "negotiable" point at a conversation. "Depending on experience" carries exactly the ambiguity above, so read it as a reason to ask the follow-up rather than as an answer. Public-sector and unionised employers, agency contracts and graduate intakes almost always have a published structure behind them. If the pay was never disclosed at all, that is a different problem with its own answer, and we have written about when an employer has to tell you the pay and when you have to ask.
Ask early, before an offer is on the table. Afterwards the same question sounds like an opening move, because it is one.
A published range is not the same as room to move
It is tempting to read the spread of published pay ranges as a straightforward gain for candidates. The evidence on what employers do once pay becomes comparable is less comfortable than that.
Zoë Cullen and Bobak Pakzad-Hurson studied US state laws protecting private-sector workers' right to discuss their pay with colleagues, and published the analysis in Econometrica in May 2023. Their model predicts that transparency "reduces the individual bargaining power of workers, leading to lower average wages", because "employers credibly refuse to pay high wages to any one worker to avoid costly renegotiations with others". Their event study finds that such laws "lead wages to decline by approximately 2%", and that the declines are smallest where workers had little individual bargaining power to start with.3
Note exactly what was studied: laws about employees talking to each other, in the United States, not rules requiring a range in the advert. This is not evidence that published ranges leave you worse off. What it does show is that once pay becomes comparable, an employer acquires a reason to hold the line with any one person, and that reason has nothing to do with you.
The same logic runs through collective agreements, which is how a great deal of European pay is set. The EU's minimum wage directive of 19 October 2022 requires any member state whose collective bargaining coverage falls below 80 per cent to provide "a framework of enabling conditions for collective bargaining" and an action plan to promote it, which tells you how central the mechanism is taken to be.5 Where an agreement covers your role, part of the number is settled before you arrive. How large a part varies: some agreements prescribe the rate for a grade, others set a floor and leave anything above it to the employer, and the directive's definition establishes only that an agreement applies to a worker, not how much room it leaves. Which makes the questions worth asking concrete ones. Which agreement covers this job, which grade am I being placed in, and does that agreement allow paying above its rate?
What moves when the wage will not
A fixed wage is not a fixed offer. Nobody has ranked the rest in order of difficulty, so treat the order below as our guess rather than a finding: the start date, the date of your first review and what it will be measured against, holiday above the statutory floor, notice period, remote days written into the contract rather than granted informally, a training or equipment budget, the job title, and a one-off signing or relocation payment that does not touch the salary line. Where a grade structure exists, placement within it is usually decided by a person rather than by the structure, and a person can justify a higher step if you hand them the evidence to do it with.
If the wage genuinely is negotiable, ask once, in writing, with a number and the reason attached to it.
The question about the band is one you will most likely have to ask in an interview, and it comes out better for having been said out loud once already. JobCraftly will run a practice round against the actual job description and give you feedback on your answers rather than general advice. What it cannot tell you is whether this particular employer's pay comes off a scale. Only they can.
The part the advice usually leaves out
"Always negotiate" is offered as though asking were free. It is not, and the cost is not evenly shared.
Hannah Riley Bowles, Linda Babcock and Lei Lai ran four experiments, published in Organizational Behavior and Human Decision Processes in 2007. In the first three, people evaluated candidates who did or did not try to negotiate their compensation. In the first two, working from written accounts, "evaluators penalized female candidates more than male candidates for initiating negotiations". In the third, using video, "male evaluators penalized female candidates more than male candidates for initiating negotiations; female evaluators penalized all candidates for initiating negotiations".4
The fourth turned the scenario around and put participants in the candidate's place, asking whether they would initiate. "With male evaluators, women were less inclined than men to negotiate, and nervousness explained this effect. There was no gender difference when evaluator was female."4 Which is the loop closing: the cost is anticipated, and the anticipation is part of who ends up asking.
All four were scenario exercises rather than real hiring decisions, with 119 students in the first and 299 college-educated adults in the second, and the work is close to twenty years old.4 Take it as evidence that the social cost of asking is real and unevenly distributed, not as a prediction about the employer in front of you. And note what it does not cover. The experiments varied whether a candidate negotiated and who was doing the judging. They did not vary when the question was asked or how it was worded, so they cannot tell you that asking early is cheaper than asking late. The suggestion earlier in this article to ask about the pay band before an offer exists is our reasoning rather than a finding, and it rests on something simpler: a question about how the employer sets pay is not a counter-offer, and it stops being possible to ask it that way once there is a number to push back on.
What none of this settles
None of it tells you what you are worth. The research says who bargains, what changes when the possibility is named out loud, and what it can cost to be the one who names it. Your number still has to come from the market you are actually in.
It does retire one habit of mind, though. If the employer said nothing about negotiability, you have learned nothing about negotiability. Worth finding out on purpose, in one sentence, while there is still nothing on the table to defend.
