I have spent time recently following a thread that began as an economic question and kept arriving, unexpectedly, at a much older one. Why does money create such difference among the same species? Not why are some people richer than others — that is a policy question with several plausible answers. The deeper question: why does the existence of a number attached to your name alter how long you live, how your brain processes stress, whether your government responds to your preferences, and how you feel standing in a room with people who have more or less of it?
The question matters because its answer determines what kind of problem money-driven inequality actually is. If it is primarily a distributional problem — a matter of where resources end up — then the solutions are redistributive: tax policy, transfers, public goods. But if it is something more structural, rooted in mechanisms that predate money itself and that money merely activates and amplifies, then redistribution addresses the symptom while leaving the architecture intact.
What follows is what the research found. It is not comfortable. It suggests the question has been answered across several disciplines in ways that haven't been fully absorbed — in part because the answer implicates not just particular policies or political systems but the deep structure of how our species processes social life.
Exploration · Part OneBegin with the biology, because it is the foundation everything else rests on. Robert Sapolsky spent thirty years studying wild baboons in East Africa, and the finding that anchors his life's work is this: social rank predicts physiological outcomes more powerfully than absolute resource access. A mid-ranking baboon with adequate nutrition suffers measurably worse cardiovascular health, higher chronic cortisol, weaker immune function, and shorter life expectancy than a dominant baboon in the same troop with the same diet.
The mechanism is not deprivation. It is relative position. The subordinate animal's body is running a continuous low-level threat response — not to predators or starvation, but to its place in the social order. Every interaction with a higher-ranking animal reactivates the stress pathway. The hierarchy is, in biological terms, a source of chronic illness that has nothing to do with whether there is enough food.
Frans de Waal spent decades studying chimpanzees at Arnhem Zoo and in the wild, and his research showed that our closest primate relatives engage in extraordinarily sophisticated status cognition: coalition-building, reputation management, reciprocal altruism calibrated to rank, deliberate deception, and post-conflict reconciliation designed to preserve valuable social relationships. The social intelligence of primates is not general intelligence applied incidentally to social problems. It is intelligence that evolved specifically to navigate hierarchy.
This matters enormously for understanding money, because it means money did not create the problem it is usually blamed for. It inherited one. The cognitive and affective architecture that humans later recruited to process wealth — reward anticipation when status rises, rank anxiety when it falls, the compulsive monitoring of where one stands relative to others — was already running millions of years before the first clay tablet recorded a grain debt in ancient Sumer.
Georg Simmel understood this in 1900, when he wrote that money is not merely a medium of exchange but a social relation — a crystallised claim on other people's time, attention, and deference. What money accomplished was the conversion of vague, context-specific prestige — the village elder's authority, the warrior's renown, the landowner's local dominance — into something universally legible and indefinitely scalable. A number on a screen represents status that can be transported across continents, accumulated without physical limit, and deployed instantaneously in any social context.
Traditional status markers were bounded. Physical strength degraded. Lineage became irrelevant in new communities. Religious authority didn't travel across cultural borders. Money has none of these constraints. It is, in Simmel's phrase, the universal solvent of social hierarchy: capable of dissolving old distinctions and encoding new ones with a speed no previous status technology could match.
Exploration · Part TwoIn 2008, a team led by Caroline Zink at the National Institutes of Health published a study in Neuron that directly imaged, for the first time, what happens inside the human brain when people process their relative social position. Participants engaged in a simulated social hierarchy task while lying in an fMRI scanner, and the results were precise enough to resolve a question that had previously been a matter of inference.
The brain does not process money or status as an abstract quantity. It processes them as relational signals that modulate the core motivational architecture of the organism. Higher perceived rank produced differential activation in the striatum — the dopaminergic reward circuit at the centre of approach motivation, anticipation, and the experience of pleasure. Lower perceived rank sustained activation in cortisol-related stress pathways.
Zink et al. — Neuron, 2008 · NIH National Institute of Mental Health
The implication is precise and uncomfortable: you are not making a cool assessment when you register someone else's wealth or your own. You are running a threat-detection and reward-evaluation system that evolved for a social species navigating a status hierarchy. Money registers neurologically not as "I have sufficient resources" but as "I am here in the social order relative to you."
This is why the experience of inequality is visceral rather than intellectual. It is why people who, by any objective measure, have more than they need still experience acute distress when they encounter someone with more. The brain's assessment of economic position is fundamentally comparative, not absolute — and no amount of rational reflection fully overrides it, because the circuitry doing the comparison predates the capacity for rational reflection by millions of years.
Leon Festinger's social comparison theory (1954) gave a name to the mechanism: humans evaluate their abilities, opinions, and worth by comparing themselves to relevant others. The insight that transformed this from an academic observation into a clinical finding was his further claim that the reference group is not fixed — it shifts upward. Once the immediate comparison group is exceeded, the next group becomes salient. There is no stable destination. The brain is not running a satisficing algorithm — it is running a continuous, open-ended ranking process that no level of wealth permanently resolves.
Brickman and Campbell named the outcome hedonic adaptation in 1971. Brickman's subsequent 1978 study — which tracked lottery winners and found they had returned to their pre-win baseline happiness within a year — remains one of the most replicated findings in welfare economics. The remarkable thing about this finding is not that money doesn't buy happiness, which is too simple. It is that the destination keeps moving. The brain updates its reference class, the comparison restarts, and the process that felt like progress reveals itself to have been motion on a treadmill.
If the evolutionary and neurological evidence suggests the mechanism of money-driven division is deep, the historical evidence suggests it is also extraordinarily stable. The recent archaeological and economic history literature has begun producing numbers that are difficult to absorb without a long pause.
Alfani, Bolla, and Scheidel published a study in Nature Communications in April 2025 that calculated the Gini coefficient — the standard measure of income inequality, where 0 represents perfect equality and 1 represents one person holding everything — for the Roman Empire around 165 CE and the Han Chinese Empire around 2 CE. The Roman figure: 0.46. The Han figure: 0.48.
The United States in 2023: 0.485.
Pause on that. A global superpower with an advanced financial system, universal suffrage, functioning courts, and several decades of inequality research producing policy recommendations — and the distribution of income it has achieved is, on the best available measure, nearly identical to the Roman Empire under Marcus Aurelius, when 80% of the population lived at subsistence and the Senate class held incomes approximately sixteen times the average.
The historical timeline of monetary innovation is, on close inspection, also a timeline of inequality innovation. Each new monetary technology did not merely facilitate exchange more efficiently — it created new forms of wealth that only the already-wealthy could access, compounding concentration further.
Walter Scheidel's The Great Leveler (Princeton University Press, 2017) delivers the most unwelcome finding in this entire literature: across five thousand years of economic history, the forces that have reliably reduced monetary inequality are not reform, growth, or institutional improvement. They are catastrophe: mass mobilisation warfare, transformative revolution, state collapse, and pandemic.
The most structurally important thread in this research is the political science one, because it closes the loop. If money converts reliably into political power, and political power is used to design the rules that govern money's future accumulation and protection, then the system becomes self-sealing. Reform requires using the very institutions that have been shaped by the process you are trying to reform.
Martin Gilens (Princeton) and Benjamin Page (Northwestern) published their landmark study in Perspectives on Politics in 2014. Using a dataset of 1,779 U.S. policy cases between 1981 and 2002, they tested four competing theories of how American democracy works. The statistical result was unambiguous enough that it generated immediate and ongoing controversy.
The average citizen's standardised influence coefficient — 0.01 — is statistically indistinguishable from zero. Even when 80% of the public supported a policy change, it was adopted only 43% of the time in the absence of elite support. The mechanism is not corruption in the colloquial sense. It operates through structurally legitimate channels: campaign finance, lobbying access, regulatory participation, litigation, and the sheer sophistication asymmetry between a corporation with thirty specialist lawyers reviewing a proposed rule and a consumer advocacy group with a single staff member.
Harvard Kennedy School research published in 2022 extended this to the administrative layer where most binding law is actually made. Analysis of over 260,000 comments on Dodd-Frank financial reform rules found that wealthier organisations were more likely to participate in rulemaking, advance more sophisticated comments, and enjoy more success in shifting the content of agency rules. The return on lobbying investment can be extraordinary: one frequently cited analysis found Boeing received approximately $7,250 in regulatory and tax advantage for every $1 spent on lobbying.
The University of Chicago's 2025 PNAS study added the cross-national dimension: economic inequality is one of the strongest predictors of democratic erosion — not only in developing states but in wealthy, longstanding democracies. The Milken Institute's research found that it is specifically politically connected billionaire wealth — not inherited wealth per se — that most reliably corresponds to lower democracy scores globally.
Brought together, the research across these disciplines does not describe three separate phenomena — an evolutionary curiosity, a historical pattern, and a political finding. It describes one coherent system with three analytical layers, each of which reinforces the others.
The evolutionary layer established the wiring: primates are exquisitely sensitive to relative position, and this sensitivity is not a flaw to be corrected but a feature built for a social species that needed to navigate hierarchy to survive. The historical layer shows how money was never a neutral tool overlaid on this wiring — from the first Sumerian clay tablet recording a grain debt, money was a technology for formalising and scaling existing hierarchy, with debt as its primary instrument. David Graeber's anthropological research demonstrated that the economic textbook narrative — barter first, then money to solve its inefficiencies — has no archaeological or anthropological support whatsoever. Debt came first, and with it came debtors and creditors, and with them came the first monetised class divide.
The political science layer closes the circuit. Money converts reliably into political influence through structurally legitimate channels. That influence is used to design the rules governing taxation, inheritance, regulatory access, and campaign finance — the very rules that govern money's future conversion into political influence. The system is self-sealing. And the historical record shows it has been running, with this basic architecture intact, for approximately five thousand years.
What this convergence means practically is that the standard prescriptions for inequality — transparency, campaign finance reform, voter participation drives, progressive taxation — may be necessary but are operating at the wrong level of the problem. They address the outputs of a system whose inputs are neurological, evolutionary, and structurally institutional. They are fighting with policy tools against mechanisms that predate policy itself.
This is not an argument for fatalism. Several findings in this literature complicate a purely deterministic reading. Wilkinson and Pickett's Spirit Level data shows that societies approximating genuinely equal basic stakes produce better outcomes on nearly every social metric. The neuroscience literature identifies specific levers: the volatility and unpredictability of status threat matters as much as its average level. Safety net design — reducing the cost and terror of falling — may be a more effective welfare instrument than redistribution alone, because it addresses the threat-response mechanism directly rather than trying to raise everyone's absolute position on a treadmill whose destination keeps receding.
And Scheidel's catastrophe finding carries an implicit policy question that no institution has satisfactorily answered: if voluntary reform has never durably reversed monetary inequality in five thousand years, and if the political process available for reform has been shaped by the inequality it would need to reverse, what does a society genuinely committed to both justice and stability actually do? That question does not have a tidy answer. Which is, I think, why it keeps returning.
I don't have a resolution to offer. The research I've described is real, the convergence across disciplines is genuine, and the implications are large enough that I want to be careful about reaching for conclusions before they're earned. What I'm more confident about is this: if you understand money as a social technology that inherited an ancient primate status system, scaled it to civilisational proportions, and then used the power it generated to protect itself from correction — you are looking at the right mechanism. Everything else is downstream of that.