Once I was thinking about a scene.
A man stands in his private museum, surrounded by paintings he has bought — each priceless, together worth perhaps the GDP of a small country. His time for looking at them is finite; at most he takes in a few a day, and in a year he cannot get through half a hall. Then the museum closes, and he returns to another of his houses — or another boat, another plane, another place he rarely visits but must possess. These assets do not enter his life. They do not generate his experience. They are simply... there. Growing. Representing the inflation of some number, a number that no longer bears any physical relation to his stomach, his sleep, his twenty-four hours.
And then, on another street in the same country, someone reduces their insulin dose because they cannot afford it, and one night does not wake up.
These two scenes hold true, simultaneously, in the same country in the same year. And I am caught on a question: why do we suppose that what the former holds and what the latter needs enjoy the same "inviolable" moral standing?
I am not against wealth. I am against a strange phenomenon: a person who is no longer capable of converting wealth into his own life nonetheless retains the legitimacy to accumulate it without limit.
A person's stomach is only so big. A person's sleep is only so long. A person's day is only twenty-four hours. Wealth can grow without limit, while a human life remains at the same scale. Perhaps what is truly strange is not poverty, but why we should suppose that a finite life can rightfully possess infinite wealth.
This is not a socialist proposition. It is closer to a classical philosophical question: when wealth has already exceeded the range that a person can directly experience, directly use, directly "live within" — when it is no longer a resource for living and has become something else, institutional power, ranking chips, the capacity to constrain the behavior of others — can the relation between it and its holder still be called "property"?
I concede that wealth incentivizes innovation. I concede that entrepreneurs deserve ample reward. I even concede that inequality in itself is not the problem. The real problem is that we have conflated two things: "rewarding the creation of value" and "permitting unlimited possession." The former is legitimate; the latter demands questioning.
Wealth Undergoes a Phase Transition
My core thesis is not a number but a qualitative judgment: the function of wealth does not vary continuously — at some point it undergoes a phase transition, from a resource directly lived in by a person to a control variable constraining the feasible set of others.
Begin with "direct use."
A private jet, a private island, a private museum, even a rocket — these are direct uses. You can sit in the jet and fly, stand on the island, walk into the museum and look at the paintings, watch the rocket rise. What they generate is direct personal experience, part of your life.
But buying twenty listed companies is not direct use. You cannot "experience" the cash flows of twenty listed companies. You cannot "walk into" a portfolio of ten thousand patents. You cannot "live inside" a hedge fund's short position. These things do not enter your life or generate your experience — they are tools, instruments, the media through which you exert influence upon others.
So the real question is not "whether consumption has an upper bound" but whether the capacity for direct use has an upper bound. The answer is: it does. And this bound is determined by the physical attributes of the creature that is the human being, not by preference or taste.
Once direct use is saturated, additional wealth can no longer be converted into personal experience. It can only flow elsewhere — toward altering the choice space of others: influencing tax rates, shaping regulation, defining standards, dominating talent, restricting competition. Beyond this transition point, wealth changes from "something you live within" into "something that shapes how others may live."
Wealth does not automatically become power. But once it can no longer improve the life of its holder, its most valuable use gradually becomes influencing what others are able to do.
The existence of this transition point does not require a precise value. You need not know whether it lies at one billion or five billion — you need only know that it exists. Just as you need not know whether the freezing point of water is 0.0001 or −0.0002 degrees; you need only know that water freezes.
The "Band" of Legitimacy
With the transition point in hand, a structure of legitimacy can be drawn.
The lower bound is the line of subsistence and opportunity security: the level of wealth sufficient for a person not to be disadvantaged under fair equality of opportunity, and to cover the foreseeable risks of subsistence, medical care, and education across a lifetime. Those below this line are the proper objects of social-justice concern.
The upper bound is the direct-use saturation point: the threshold at which wealth begins to slide from a resource for living into a control variable.
The band squeezed between them is the legitimate region. Inequality falling within this band — some having more, some less — is morally acceptable provided it simultaneously satisfies Rawls's difference principle (that is, that it benefits the least advantaged). It reflects differences in ability and the need for incentives, and does not infringe upon the basic rights and opportunities of others.
And beyond the upper bound? That is the core of what this essay argues: the legitimacy of wealth held beyond the bound is no longer self-evident.
"Usable Influence": Harm Without the Blow
You may say: the wealthy have influence, but as long as they do not actually intervene in politics, the political liberty of ordinary citizens is not infringed. Possessing a capacity is not the same as exercising it.
This objection does not hold in game-theoretic terms.
The political game is a strategic interaction. Legislators, regulators, and ordinary citizens, in formulating policy, anticipate the possible actions of the various parties and adjust their strategies accordingly. A person holding excess wealth can at any time convert it into political action — funding an opponent, launching a campaign of public opinion, backing an alternative candidate. This "constantly available" capacity constitutes a credible threat: even if he does not act at present, the other players, anticipating that he may, will adjust their strategies on their own.
A legislator who knows that a certain policy may provoke tens of millions of dollars in counter-lobbying will self-censor at the proposal stage. He need not actually be lobbied — he need only know that the possibility of lobbying exists, and his behavior has already changed.
This is a fundamental theorem of game theory: in equilibrium, a credible threat changes the outcome even when the threat is not carried out. A player with "punishment capacity" need not actually punish. The political deterrent force of wealth is the same.
So the "usable influence" of excess wealth — even if it is never exercised — has already altered the equilibrium of the political game, substantially weakening the effective political influence of ordinary citizens. The infringement does not occur at the moment of exercise; it occurs at the moment the capacity exists.
Someone will object: an educated person has greater influence than an uneducated one, a charismatic leader is more persuasive than an ordinary person — these differences in influence also have "usability"; why do they not count as unjust?
The difference lies in three structural features of the influence of wealth, none of which belong to education, charisma, or organizational capacity:
- Transferability: wealth can be transferred to agents — lobbying firms, super PACs, media organizations — so that influence is exercised without the holder's own participation. An educated person cannot transfer his persuasiveness to another. A charismatic leader cannot outsource his charisma.
- Accumulability and inheritability: wealth can be accumulated without limit and transferred across generations without decay. A family's political influence can keep growing over generations, forming structural barriers of power. Education and charisma dissipate with the individual — they cannot be inherited.
- Quantitative scaling of magnitude: the influence of wealth grows linearly or even super-linearly with the amount — ten times the lobbying budget means ten times the lobbying capacity. The "influence" of education and charisma is bounded by the individual's time and energy and cannot be expanded proportionally by "doubling the input."
These three features make the influence of wealth structurally distinct from the influence of personal ability. The "usability" of personal ability is bounded by the individual's presence, time, and lifespan; the "usability" of wealth is not bound by the individual — it can be transferred, accumulated, inherited, and expanded.
Rent-Seeking: When the Marginal Return of Innovation Loses to Power
A second independent line of argument is anchored in Rawls's difference principle.
The standard defense of the right is that permitting inequality can incentivize innovation and investment, that economic growth benefits everyone, including the least advantaged. Within the legitimate band this mechanism does indeed work — moderate inequality does incentivize productive innovation. But once wealth crosses the transition point, the mechanism reverses.
The chain of reversal runs as follows.
Link one: beyond the transition point, the rational strategy of the holder shifts from "productive innovation" to "constraining the feasible set of others." This is the central insight of public choice theory. The return on productive innovation is uncertain, long-term, and constrained by market competition. The return on changing the rules — influencing tax rates, erecting regulatory barriers, restricting market entry, or simply eliminating competitors through mergers — is relatively certain, short-term, and directly increases the preservation of the holder's wealth. Once direct use is saturated, the holder has no need to obtain further experiential resources through innovation; his sole aim is to preserve and expand his existing wealth. Under this aim, the marginal return of constraining others is systematically higher than the marginal return of innovation.
Link two: rent-seeking does not create new value; it redistributes existing value (zero-sum) and, in the process, consumes resources — legal fees, lobbying expenditures, bureaucratic approval costs — that reduce total social value (negative-sum).
Link three: as a larger share of social resources is channeled into rent-seeking rather than innovation, the growth rate of total factor productivity falls.
Link four: slowing productivity growth is transmitted to the least advantaged through three channels — slower job creation and lower wage growth; a smaller tax base sustaining a thinner social safety net; less public investment in education, health care, and infrastructure.
So: excess holding, through rent-seeking → resource misallocation → productivity decline → harm to the least advantaged. This arrangement of inequality not only fails to benefit the least advantaged, it damages their long-term prospects. The difference principle is violated.
For every link in this causal chain, you can find corroboration in the empirical record of the United States over the past forty years. Saez and Zucman have documented the sustained decline in effective tax rates at the top and the sustained rise in wealth concentration. Over the same period, the rate of new firm creation has fallen, market concentration has risen, and the pace of leader turnover across industries has slowed — the attenuation of business dynamism is highly synchronized with the climb in wealth concentration. This is no coincidence. If the incentive myth held — if excess wealth ignites innovation — you should not be seeing the pattern of "the more concentrated the money, the more enervated the innovation." The empirical facts run exactly opposite to the myth's prediction.
The Collapse of the Incentive Myth
At this point the ultimate line of defense on the right still stands: "Even granting everything you have said, if you tax high wealth the innovators lose their motivation. Growth slows, and in the end everyone suffers — including the least advantaged."
This is the core of the incentive argument. Let me dismantle it.
But first, a story. A billionaire who inherited a great deal of real-estate wealth ran for president on a promise to "drain the swamp." Once elected, he packed his cabinet with billionaires and Wall Street executives and passed one of the largest tax cuts for the wealthy in American history — the corporate tax rate was slashed from 35 percent to 21 percent. The promise of the cuts was that the released profits would flow back into investment and employment. What actually happened? Most of it went into stock buybacks, pushing up share prices and making those who held large amounts of stock — that is, himself and his class — richer. The real wages of workers barely moved. The swamp was not drained; it was filled with gold.
This is not an isolated case; it is a systematic regularity of the excess zone. The incentive myth tells you that money left in the hands of the rich becomes innovation. The fact is that money left in the excess zone becomes rent-seeking.
The incentive argument requires two premises: first, that inequality is a necessary spur to productive innovation; second, that any reduction of the excess portion necessarily destroys that spur.
Consider premise one. Once wealth crosses the transition point, the character of "incentive" undergoes a qualitative change. Within the legitimate band, competition for rank is conducted chiefly through innovation — innovation is the most effective means of raising rank within the legitimate band. But in the excess zone, the most effective means of raising relative rank shifts from innovation to rent-seeking. For the return on innovation is uncertain — after heavy investment in R&D one may still be overtaken by a rival, and rank may even fall. The return on rent-seeking is relatively certain — lobbying for tax breaks, mergers to eliminate competitors, political influence to set up entry barriers, all raise one's rank directly and reliably.
When the aim is "to have more money than others" rather than "to create this much value," a rational agent will choose the more reliable means of raising rank. It is not that the rich have grown worse; it is that the optimal strategy in the excess zone has systematically swerved.
More seriously, excess wealth not only weakens the innovating drive of the holders themselves but systematically suppresses innovation from below. This is a two-way suppression — the top end loses its drive, the bottom end loses its room. The mechanisms include: killer acquisitions (buying up potential competitors and then shelving their technology); "kill zones" (when innovation appears in some area, the giants use their capital advantage to copy at scale or to dump with subsidies, so that the innovator cannot survive); patent thickets (using vast patent portfolios to block new entrants).
The empirical facts run exactly opposite to the prediction of the incentive myth: over the past forty years, the share of top wealth in the United States has kept rising, yet the rate of new firm creation has fallen, market concentration has risen, and the pace of leader turnover has slowed. If "excess wealth ignites innovation" held, you should not be seeing this pattern.
Now consider premise two. The incentive argument requires that "any reduction necessarily destroys the incentive." But this "necessity" can be refuted. Consider a simple proof: if a uniform proportional reduction is imposed on the excess portion, no one's rank changes — because multiplying everyone by the same positive constant does not alter the ordering. The utility of a rank-player depends on relative rank rather than absolute amount, so rank-preservation means that the incentive structure is preserved intact.
The rank incentive is not destroyed. The "necessity" premise of the incentive argument collapses.
Excess wealth is not the engine of innovation but its double fetter: it strips the holder of the drive to innovate while strangling the innovating room of the challenger.
When Excess Holding Meets Deprivation of Subsistence
The preceding four sections have argued for the transition point and for a pro tanto injustice. But there is a still sharper question. It is not a theoretical question; it is a fact visible the moment you open your eyes.
When some people hold wealth far beyond the transition point — wealth that can no longer enter their lives and can only be used to alter the choice space of others — while at the same time other people die foreseeable deaths for lack of the basic means of subsistence, what does this coexistence signify?
This is not a hypothesis. It is the present reality of the United States.
Case and Deaton call them "deaths of despair" — suicide, alcoholic liver disease, drug overdose. The mortality rates for these three causes began to climb in the late 1990s, accelerated in the 2010s, and by the 2020s had wiped out half a century of life-expectancy gains among the white American working class. In 2023, drug-overdose deaths in the United States exceeded 100,000 — a figure greater than the combined American military deaths in the Vietnam, Iraq, and Afghanistan wars. Fentanyl accounts for most of them. A synthetic opioid, fifty times cheaper than heroin and fifty times more lethal, is systematically killing those cast off by economic transformation.
Meanwhile, on the streets of America's wealthiest cities — San Francisco, Los Angeles, Seattle — there are tent encampments. The homeless are not people who are "too lazy to work"; a large proportion of them hold full-time jobs — they simply cannot afford the rent. A software engineer earning $150,000 a year may still be unable to afford a one-bedroom apartment in San Francisco, while a few miles away someone sleeps on cardboard.
Life expectancy — the most composite indicator of a society's health — declined in the United States for three consecutive years. There is no precedent for this in a developed industrial country at peace. And it declined while the wealth of the top 0.1 percent was rising, while the S&P 500 was rising, while corporate profits were rising.
These two things happen in the same country, in the same year: someone dies because they cannot afford insulin; someone spends four hundred million dollars on a yacht.
Henry Shue argues that the right to subsistence is a "basic right" — it is the precondition for the enjoyment of all other rights. A person dying of a preventable disease cannot exercise freedom of speech, the right to vote, or property rights. Thomas Pogge goes further: global poverty is not a natural disaster but "the foreseeable consequence of an institutional order we impose." When an institutional order systematically permits excess accumulation while systematically leaving those at the bottom to lose the means of subsistence, the beneficiaries of that order are not innocent bystanders — they bear the negative duty "not to maintain an institutional order that foreseeably kills people."
Someone will say: financial wealth (stocks, bonds) and real resources (food, medicine) are not the same thing, and a billion dollars of stock cannot be directly turned into bread. But financial wealth is a claim on real resources. A billion dollars of stock represents an indirect claim on the labor, land, and capital controlled by real enterprises. Concentrated financial wealth, through capital-market price signals, corporate-governance voting rights, and the direction of investment, indirectly but substantially influences the direction of resource allocation. The real resources shaped by the excess holder serve his already saturated needs; the same resources could have been directed toward the production of the means of subsistence — medicine, housing, insulin. The ownership of financial claims is not a law of nature; it is the product of institutional arrangement.
So when excess holding coexists with deprivation of subsistence, the pro tanto injustice escalates into a structural infringement of the basic right to subsistence. At this point, the property-rights claim of the excess holder loses its moral protection — it no longer enjoys an "inviolable" standing.
This is not a utilitarian calculation — not "because the marginal utility of wealth to the rich tends to zero and the marginal utility to the poor tends to infinity, therefore transferring wealth maximizes total utility." Precisely such interpersonal comparisons of utility are what Rawls objected to. The criterion is not "the increase or decrease of total social utility" but "whether basic rights are infringed." The rank-utility of the rich — the satisfaction of relative standing that excess wealth brings — does not enter the reckoning at all. While some are still striving for rank in wealth, others are dying of a preventable drug overdose. What the former pursue cannot, in moral weight, be spoken of in the same breath as the right to subsistence of the latter.
Of course, the opponents will not be silent.
Nozick's theory of justice in holdings will say: provided the process of acquisition and the process of transfer were just, the resulting holding is just. But the process by which the excess holder acquires wealth depends deeply on the institutional channels of big government — patent monopolies, licensing, tax breaks, subsidies, government procurement contracts, regulatory barriers. You cannot say on the one hand "the subsidies of big government are a legitimate avenue of acquisition" and on the other "the redistribution of big government is a violation of self-ownership" — that is self-contradictory.
Some say that the holding of excess wealth is to guard against doomsday risk. Then truly leave it idle — do not plunge it into the financial markets in pursuit of increase, do not use it for political influence. And in reality almost no excess holder does any such thing.
Still others appeal to the "trickle-down effect": the investments of the excess holders will eventually benefit those at the bottom. But over the past forty years, the share of top wealth has kept rising while the real income growth of the bottom has stagnated. If trickle-down worked, this should not be happening. Trickle-down is not a source of legitimacy for excess holding; it is the empirical evidence of its failure.
Diagnosing the Structure, Not the Number
I have not given a precise value for the transition point. Nor do I intend to.
Precise quantification depends on empirical data and situational parameters, while the normative argument requires only a weak proposition to hold — "there exists such a threshold." You need not know whether it lies at one billion or five billion; you need only know that wealth undergoes a phase transition.
But if an existence argument offers no empirical guidance whatsoever, its practical relevance approaches zero. So I offer a directional judgment: the median wealth of the top 0.1 percent in the United States is on the order of hundreds of millions of dollars, and their annual consumption expenditure falls well below the point of consumption saturation. At the same time, total annual spending in the American lobbying market is about $3.5 billion, and a single super PAC may spend tens of millions to hundreds of millions — which means that individuals holding wealth on the order of hundreds of millions already possess the capacity to cross the threshold of political influence. The intersection of these two data points indicates that the wealth level of the American top has reasonably fallen within the transition band: the consumption function is saturated, the power function activated.
[Editor's Note] This essay and The Tyranny of the Future share a deep structure: both ask "by what right." By what right does the future rule the present? By what right does infinite wealth enjoy moral protection? The answers, too, are similar: not because some authority has been bestowed, but because a structure of argument has been running covertly, hiding the burden of proof. You say "for the future," and you are absolved from argument; you hold a billion in wealth, and others must instead prove that you "should not" hold it. The two essays do the same thing: they dig out the hidden burden of proof and put it back where it belongs. Once it is put back, you find that the bill was always meant to be paid by the party claiming the power.