Case Study 39.2: The Meritocracy Research — What Social Science Actually Finds
Background
Meritocracy is one of the foundational beliefs of contemporary Western societies, particularly in the United States. Most people, when asked, say they believe that hard work and talent should determine outcomes — and that, at least roughly, they do. This belief functions not just as a description of how the world works but as a moral claim about how it should work: outcomes that reflect merit are legitimate; outcomes that don't are not.
This case study examines what social science actually finds about the relationship between merit and outcomes — and, perhaps more importantly, about the psychological and social effects of believing strongly in meritocracy regardless of whether it's true.
The findings are frequently counterintuitive, and some of them are directly relevant to the ethics of luck argument in Chapter 39.
Part 1: Does Merit Actually Predict Outcomes?
The Returns to Education
The most common meritocratic metric in contemporary society is educational attainment. We use education as a proxy for demonstrated effort and ability, and we use educational credentials as a primary sorting mechanism for economic opportunity.
The research on the return to education shows that education does matter — substantially. College graduates earn significantly more than high school graduates over a lifetime, and the wage premium has grown over the past forty years. So far, meritocracy would seem vindicated.
But examine the details and the picture complicates. Research by Raj Chetty and colleagues at Opportunity Insights found that:
- At elite universities (the "Ivy Plus" group), students from the top 1% of the income distribution are more than twice as likely to attend as students from the bottom 60%, even after controlling for test scores.
- Legacy preferences at elite universities meaningfully increase admission rates for children of alumni, independent of merit measures.
- Attendance at an elite university substantially increases earnings in domains like finance, consulting, and law — largely through network access, not through additional skill development at those schools.
In other words: the credential that supposedly signals merit is obtained through a process that is substantially influenced by structural luck (family wealth, legacy status, access to test prep, quality of secondary education), and the credential then produces returns largely through social network dynamics rather than skill. The meritocratic sorting machine is, empirically, substantially sorting by class background as well as by ability.
The Research on "Who Gets the Job"
Beyond education, research on hiring decisions reveals consistent patterns that are difficult to square with a pure meritocracy story.
Bertrand and Mullainathan's 2004 study "Are Emily and Greg More Employable than Lakisha and Jamal?" sent identical resumes with randomly assigned white-sounding or Black-sounding names to employers. White-sounding names received 50% more callbacks for interviews. The qualifications were identical. The difference was a name — a form of constitutive luck entirely outside the applicant's control.
This has been replicated across racial and ethnic groups, across countries, and extended to gender, age, and other socially salient characteristics. A meta-analysis by Lincoln Quillian and colleagues (2017) examining audit studies across 24 separate studies found no significant decline in racial discrimination in hiring from 1989 to 2015, despite substantial legal and cultural changes in the intervening decades.
The conclusion: the baseline assumption that equally qualified candidates are evaluated on their qualifications is empirically false in systematic ways that correspond to characteristics outside applicants' control. Meritocracy, as practiced, incorporates substantial structural luck preferences.
The Luck of the Timing
Research by Oreopoulos, von Wachter, and Heisz (2012) examined the long-term earnings trajectories of college graduates who entered the labor market during recessions versus graduates who entered during expansions. The finding: graduates who entered the market during a recession earned significantly less than equally qualified graduates who entered during an expansion — and the gap persisted for up to ten years.
The magnitude is substantial: a 1.5 percentage point increase in the unemployment rate at graduation was associated with a 9% reduction in earnings that persisted at roughly 7.5% after five years and 4% after ten years. This is purely timing luck — the economic conditions at the moment of graduation, which graduates did nothing to cause and could not control.
This finding — sometimes called the "graduation timing lottery" — shows that structural macroeconomic luck has measurable, durable effects on individual outcomes independent of individual merit.
The Network Premium
Research on social capital and economic outcomes consistently finds that network access — who you know, who knows you, how well connected your network is — substantially predicts career outcomes independent of formal qualifications.
Studies of job search processes find that 25-50% of jobs are filled through referrals that never appear in public job listings. LinkedIn's own data suggest that about 85% of jobs are filled through networking. The people with access to those informal channels are, systematically, the people who are already relatively advantaged — they know people, or know people who know people.
This is what Granovetter identified as the "strength of weak ties" — and it creates a feedback loop in which structural luck in network access produces more access, which compounds over careers. Early network luck is amplified over time.
Part 2: The Psychological Effects of Meritocracy Belief
Perhaps the most important — and most counterintuitive — body of research for our purposes concerns not whether meritocracy is real but what believing in it strongly does to people's behavior.
The Jost et al. Findings
John Jost, Samuel Blount, Jeffrey Pfeffer, and Gyorgy Hunyady published "Fair Market Ideology: Its Cognitive-Motivational Underpinnings" in 2003, examining the relationship between meritocracy and just-world belief. Their research found:
- People who more strongly endorsed "fair market ideology" (the belief that markets distribute rewards fairly based on effort and ability) showed less support for redistributive social policies.
- The mechanism was activation of the "just world hypothesis": believing the market is fair leads to believing outcomes are deserved, which reduces the moral urgency of addressing disadvantage.
- This effect persisted even when controlling for political affiliation, suggesting it is a psychological phenomenon rather than purely a partisan one.
The implication: the ideology of meritocracy produces less charitable social behavior — not more — regardless of individual intent. The mechanism is cognitive: the belief that the system is fair eliminates the perceived need for additional help.
The Castilla and Benard Paradox
Emilio Castilla and Stephen Benard's 2010 study "The Paradox of Meritocracy in Organizations" produced one of the most striking findings in this area. They conducted experiments with participants playing the role of managers who had to allocate bonuses among employees.
When managers were told they worked in an organization that explicitly valued meritocracy, they showed more favoritism to male employees over equally performing female employees — not less.
The finding: the meritocracy label gave managers license to feel the system had already handled fairness, reducing vigilance against their own biases. They trusted the system to be fair, which made them less careful about being fair themselves.
This has practical implications beyond gender. Any organization that declares itself a meritocracy and assumes this declaration is sufficient may, perversely, produce more biased outcomes than organizations that acknowledge ongoing bias and actively work to mitigate it.
The Immovability Research
Research by Michael Kraus and colleagues at Yale has documented that people who grow up in higher social-class backgrounds perceive themselves as more in control of their lives and more responsible for their own outcomes than people from lower social-class backgrounds. This "sense of personal control" has psychological benefits — it is associated with better mental and physical health.
But it also correlates with reduced empathy for those who are struggling. People with high-class backgrounds who have strong senses of personal control tend to underestimate the degree to which circumstances beyond individuals' control affect outcomes — because their own experience was one in which circumstances were relatively favorable and their choices correspondingly meaningful.
The result: the very people most advantaged by structural luck — who have the most resources to understand, acknowledge, and address it — are systematically less likely to perceive it, because their experience has not equipped them to see it.
Part 3: How Belief in Meritocracy Varies Across Societies
Meritocracy belief is not uniform across societies. Countries differ substantially in the degree to which their citizens believe that effort determines outcomes, and these differences correlate with social policy choices in predictable ways.
The United States is notably high on meritocracy belief relative to peer nations. The Pew Research Center and the World Values Survey both document that Americans are more likely to attribute success to hard work and less likely to attribute it to luck or structural factors than citizens of most other wealthy democracies.
This is not uniformly matched by the empirical reality. Research by Miles Corak (2013) and others using "intergenerational income elasticity" — essentially, how much a father's income predicts a son's income — finds that the United States has lower actual social mobility than most of the countries where people report believing strongly in social mobility. The Nordic countries, Canada, and Germany have substantially higher rates of actual intergenerational mobility than the United States — and their citizens express somewhat lower belief in meritocracy.
The "Great Gatsby Curve," identified by Miles Corak, shows that higher inequality is correlated with lower intergenerational mobility across countries. The United States has both high inequality and high meritocracy belief — a combination the research suggests is partially self-sustaining: high meritocracy belief reduces support for redistributive policies, which preserves inequality, which makes social mobility harder.
Part 4: The Honest Reckoning
What does this body of research actually establish?
First: Meritocracy, as practiced, is not pure. It mixes genuine merit-relevant factors with systematic preferences for luck-determined characteristics (race, family network, timing, geography) in ways that substantially shape outcomes independent of underlying ability.
Second: Believing strongly in meritocracy has measurable negative effects on prosocial behavior, bias awareness, and support for structural correction — the very things that would make meritocracy work better in practice.
Third: Societies with stronger institutional supports for equal opportunity — more redistribution, more equal early childhood investment, stronger social insurance — tend to have higher actual meritocratic outcomes (more mobility) than societies with weaker supports and stronger meritocracy belief.
Fourth: Individual effort is not irrelevant. Studies of immigrant communities, of individuals who rise substantially from disadvantaged starting points, and of natural experiments in access (what happens when disadvantaged students are admitted to selective schools) all show that individual behavior within structural constraints matters. The conclusion is not that effort is meaningless — it is that effort operates within a field shaped by luck, and the field matters too.
The most accurate summary of the social science is: effort and skill are real and consequential; structural luck is real and consequential; the two are entangled in ways that make simple attribution of outcomes to either one alone empirically unjustified.
This is exactly what the ethics of luck chapter argues. Social science is not, here, a rubber stamp on ideology. It is a check on the tempting simplifications — in both directions — that we reach for when the complicated truth is uncomfortable.
Discussion Questions
1. The graduation timing research shows that entering the labor market during a recession has persistent effects on earnings for up to a decade. From a luck egalitarianism perspective, what, if anything, should be done about this? What practical interventions would reduce this luck-based inequality?
2. The Castilla-Benard paradox suggests that explicitly declaring meritocracy may backfire, producing more bias rather than less. What does this imply for how organizations should talk about fairness and bias? What is a better approach?
3. The finding that meritocracy belief is higher in the United States than in more mobile societies raises a provocative question: is high meritocracy belief partly caused by lower mobility, as a kind of ideological compensation? Or is it an independent cause of lower mobility? What evidence would distinguish these possibilities?
4. Research on racial hiring bias finds no significant decline from 1989 to 2015, despite decades of legal and cultural effort. How should this finding affect our expectations about voluntary efforts to reduce bias (like Scott's giving or corporate DEI commitments) vs. structural solutions (like anti-discrimination law enforcement, blind application processes, or quota requirements)?
5. The research described shows that the people most advantaged by structural luck (high social-class backgrounds) are systematically less likely to perceive it. If this perceptual gap is systematic, what does it imply for how we design conversations about privilege and luck? What conditions or experiences tend to break through it?
6. The Great Gatsby Curve suggests that high inequality and high meritocracy belief tend to co-occur and may reinforce each other. Is there a version of meritocracy belief — a different framing of "effort and talent should matter" — that is compatible with strong support for redistribution and structural correction? Or is meritocracy belief inherently corrosive to redistributive politics?
Extension Activity
Find the most recent data you can (within the last two years) on intergenerational income mobility in the United States, compared to two other countries. Use the data to evaluate the following claim: "In America, if you work hard enough, you can achieve anything."
In 300-400 words, describe what the data show, assess the claim for accuracy, and explain what a more empirically accurate version of the claim would sound like — one that preserves the genuine truth in the meritocracy ideal while acknowledging the structural luck the original claim ignores.