Vilfredo Pareto, an Italian economist and sociologist, formulated around the turn of the twentieth century a standard of efficiency that still forms the foundation of neoclassical welfare theory today. The Pareto criterion states: one situation is preferable to another if at least one person is made better off by it — and no one worse off. A state is Pareto-optimal when no such improvement is possible any more.
In 1954 Arrow and Debreu proved with mathematical exactness that, under certain conditions, the free market leads to a Pareto-optimal equilibrium. This proof is correct. The question we are asking is a different one: what is the Pareto criterion worth as a definition of social welfare?
I. What Pareto Says — and What He Leaves Unsaid
The Pareto criterion deliberately defines efficiency without reference to justice. Pareto believed that no scientifically grounded statements could be made about questions of justice — so he left them out. That was meant in an intellectually honest way. The consequences are far-reaching.
A concrete picture: ten people. One owns 91% of all resources, nine own 1% each. This state is Pareto-optimal — because in order to make one of the nine better off, one would have to take something away from the one, which would make him worse off. The theory calls this efficient. About its justice it says — nothing. That is not a gap. It is a methodological decision.
It follows that there are infinitely many Pareto-optimal states. From complete equality to the most extreme concentration — all of them are, under the model's assumptions, equally "efficient". The criterion cannot say which of these states is desirable. That question is kept out of the model as unscientific.
The Pareto optimum is not a measure of welfare. It is a measure of the fact that no mutually agreed exchange is possible any more — regardless of how unequal the starting position from which one exchanges.
The Second Fundamental Theorem of Welfare Economics tries to repair this: if the initial endowments are redistributed through transfers, any desired Pareto-optimal equilibrium can be reached. Distribution would therefore be correctable in principle. That sounds like a solution. Until one asks: who decides on the transfers? By what criterion? With what power? The model is silent. It delegates the question of distribution to the political sphere — where, as we can observe every day, it finds no adequate answer.
II. The Ratchet — How the Mechanism Perpetuates Inequality
The deepest objection to the Pareto criterion is not its indifference to distribution in the static sense. It is its effect in the dynamic process: combined with innovation and growth, it works like a ratchet — a mechanism that turns in only one direction.
The sequence has three stages.
Stage 1 — Freezing. The existing distribution is set as the starting point. How it came about — through work, inheritance, extraction, historical asymmetries of power — is irrelevant to the model. It is the reference point from which Pareto improvements are calculated. Existing property is legitimate by definition.
Stage 2 — Extraction outside the theory. Innovation takes place. New productivity, new markets, new technologies come into being. At the moment they arise, they lie outside the existing equilibrium — they are new capital not yet assigned to any owner. Who appropriates it is not governed by Pareto theory. It happens through patents, through market power, through network effects, through political influence, through the mere fact of who owns the means of production. The theory does not observe this process. It looks away.
Stage 3 — Freezing at a new level. As soon as the appropriation is complete, the result becomes part of the new status quo. The distribution of the gains from innovation is now the starting distribution for the next round. And this new state is again Pareto-optimal — no one may be touched, not even those who have newly appropriated.
This is not a spiral. It is a ratchet. It turns in only one direction. Every round of innovation potentially widens the gap between those who are able to extract and those who are not. The theory does not depict this movement — it legitimises every end state after the round as efficient and every intervention as a violation of property rights.
The Megamachine does not need this framework explicitly. It needs only the selection conditions it creates: an economics that externalises distribution as a political problem while internalising property rights as an economic fact. This is not a conspiracy. It is the consequence of a methodological choice of direction made more than a century ago.
III. Rawls — A Different Definition of Welfare
In 1971, in "A Theory of Justice", John Rawls formulated the direct counter-design. He came not from economics but from philosophy. That is no coincidence.
Rawls's starting point: how would people design a just society if they did not know what position they would occupy in it? Behind the "veil of ignorance" — without knowing whether one will be born rich or poor, gifted or disadvantaged, into good or bad circumstances — no one would choose a system that permits extreme inequality. For one might find oneself at the bottom.
From this follows the difference principle: social and economic inequalities are justified only if they benefit the worst off. Not: if overall efficiency rises. Not: if no one is made worse off. But: if the situation of the worst off is improved.
This is the direct antidote to Pareto. Pareto asks: are there still mutually agreed improvements to be made? Rawls asks: how are the worst off faring? The first question is silent on distribution. The second makes it the yardstick.
The difference principle would reverse the logic of the ratchet. Gains from innovation would not automatically be the property of whoever appropriates them — they would be tied to the condition that they improve the situation of the worst off. That does not rule out profit. It rules out profit arising at the expense of those who already have the least.
Rawls has remained largely without consequence in economic policy. The Pareto criterion has won — not because it is more convincing, but because it serves the selection conditions of the Megamachine better: it legitimises what is, instead of asking what ought to be.
IV. What This Means for Institutional Architecture
To make increasing general welfare possible with decreasing political redistribution, we need an architecture that eliminates the ratchet of growing injustice and concentration of wealth in the dynamic sectors of the economy.
This is not a call for redistribution in the political sense — but a cybernetic observation: a system without built-in negative feedback on distribution drifts structurally towards concentration. Not because people are evil. Because the selection conditions make it so.
The ratchet needs a counterpart — a mechanism that works in the other direction. Not once, through a political decision, but structurally built in, acting automatically. What this means in concrete terms:
Progressive taxation is negative feedback on the concentration of wealth — but only if it is genuinely progressive and not hollowed out by tax-planning loopholes. The GmgV (Gesellschaft mit gebundenem Vermögen, a company with locked-in assets) is negative feedback on the extraction of capital: wealth that cannot be distributed cannot be accumulated. Genuine separation of powers is negative feedback on political power — it prevents economic concentration from being converted into political concentration, which then protects economic concentration further.
Rawls's difference principle would be the strongest institutional anchor: the question "Is the situation of the worst off improving?" as a binding criterion for economic policy decisions — not as a wish, but as a measurable target value with automatic consequences if it is missed. A reference variable in the cybernetic sense, with a defined controlled variable and a built-in controller.
This is the opposite of Pareto logic. Pareto asks: are there still mutually agreed improvements to be made? That question leaves the ratchet untouched. Rawls asks: how are the worst off faring? That question intervenes in the mechanism.
We do not need a welfare criterion that perpetuates injustice. We need one that structurally brakes concentration — not by appeal, but by design. A system without this braking distance ends like a tumour: it grows until the host dies.