People from lower social classes were less likely to negotiate workplace rewards in research involving 11,344 participants. In a separate hiring experiment within the project, a salary request also brought a larger penalty in perceived cooperativeness when the applicant was presented as coming from a less privileged background.
The findings suggest that advice to ask for more pay addresses only part of the problem. Employees may differ both in their willingness to negotiate and in how others respond when they do.
The peer-reviewed paper, published online on July 24 in the Proceedings of the National Academy of Sciences (PNAS), combines five studies in the United States and Australia. Its authors are Ying Lin of the Hong Kong University of Science and Technology, Jackson G. Lu of MIT Sloan and Michele J. Gelfand of Stanford Graduate School of Business.
Similar qualifications, different willingness to ask
One study examined 634 MBA students from the 2017–2020 graduating classes of a leading US business school who had accepted full-time jobs in the country. About 58% reported negotiating their starting salaries.
The researchers used parents’ educational attainment as an indicator of social-class origin. Graduates from less advantaged backgrounds were less likely to negotiate, even after accounting for academic performance, admissions-test scores, industry and several demographic characteristics.
Sharing an MBA education made the participants more comparable, but did not make their experience or job opportunities identical. The sample also excluded consulting jobs, where compensation was typically standardized and nonnegotiable.
The largest study examined 7,883 employees at 1,383 Australian companies. More highly educated workers were more likely to report negotiating their initial pay. The association remained positive within the ten most common occupations, suggesting that differences between types of jobs did not fully explain it.
Education is an incomplete measure of social class. The researchers used additional measures in other studies, including income and people’s assessments of their own socioeconomic position.
The salary difference was observed, the lifetime gain was not
In the MBA sample, the adjusted analysis linked negotiation to approximately $6,450 more in starting annual salary, measured in 2022 US dollars.
The researchers illustrated how that difference could accumulate: if it increased by 5% each year, the additional earnings would total about $779,000 over 40 years.
They did not follow graduates through four decades of employment. The calculation assumes a continuing salary gap and regular percentage increases. Actual earnings would depend on job changes, career breaks, promotions and other events.
Nor does the salary comparison establish that negotiating caused the entire initial difference. People who negotiate may have stronger outside offers or other advantages. The study found no statistically significant direct association between parental education and starting salary in this MBA sample.
Fear of damaging a relationship
Among MBA graduates who did not negotiate, those from less advantaged backgrounds were more likely to cite concern about damaging their relationship with the employer.
A further survey of 995 US full-time workers linked lower social class to a weaker sense of personal power and greater concern about social backlash, meaning an unfavorable reaction to speaking up. Those factors helped statistically explain differences in negotiation, but were not experimentally established as causes.
The pattern extended beyond starting pay to benefits, promotions, raises and job responsibilities.
It also appeared when the opportunity was explicit. In an online task involving 699 US adults, participants were offered a 50-cent bonus and told they could negotiate. Only 39 asked for more. Higher social class was associated with a greater likelihood of asking, even after accounting for actual and perceived task performance.
Employers reacted differently to the same pay request
The final experiment recruited 1,133 US professionals working in recruitment, hiring or human resources. They evaluated a fictitious male applicant for a marketing associate role.
Participants were randomly assigned versions that signaled higher or lower social-class origins through background information, hobbies and other cues. They also saw the applicant either negotiate or make no compensation request.
The compensation request was the same: increase an $80,000 salary to $100,000 and add eligibility for a year-end performance bonus. The scripts were not word-for-word identical, because the lower-class version also identified the applicant as the first in his family to attend college.
Negotiation reduced perceived cooperativeness in both class conditions, with a larger reduction for the lower-class applicant. A statistical analysis connected that difference to less favorable hiring judgments through cooperativeness.
The experiment measured evaluations, not actual hiring. Among participants who saw a negotiating applicant, the researchers found no significant class difference in salary counteroffers or willingness to concede to the request.
A modest association, not a rule about individuals
Combining the first four studies produced a correlation of 0.12 between social class and negotiation propensity. That is a modest relationship, not evidence that class determines whether someone asks for more.
Most of the research took place in the US. The employer experiment involved one male candidate and one occupation, limiting what it can establish about other applicants or workplaces.
The authors propose that lower-class applicants may face stronger expectations of cooperation, making negotiation more likely to violate those expectations. The experiment did not establish that explanation conclusively.
Clearer pay rules could reduce reliance on bargaining
For employers, the findings raise questions about compensation systems that leave employees to discover what is negotiable and ask individually for better terms.
“Encouraging people to speak up isn’t the whole picture,” Gelfand said in Stanford’s account of the research, published August 27.
The researchers propose clearer salary ranges, explicit explanations of negotiation opportunities and more transparent advancement processes. They also suggest checking whether evaluations and rewards differ by social-class background. These are proposed responses, not policies tested by the five studies.
A separate 2026 OECD review of pay transparency explains how salary information can help applicants assess their options and negotiate. It also cautions that evidence on whether salary disclosure requirements narrow gender pay gaps remains limited. Establishing which policies reduce class-related barriers will require direct evaluation in workplaces.
