The rule, exactly. Before an election each citizen chooses: to vote — or to take a payment and not vote in that election. The payment is a percentage of the median income, set by referendum; there is no fixed sum. One ballot is one vote, with no multiplier in the count; the vote is amplified only in that each ballot's share grows as others step out. The budget pays under law; a candidate never pays. Only a referendum of all citizens — simple majority, no quorum — introduces, changes or repeals the rule.
The protocol has been introduced nowhere and no pilot has been run: shares, turnout and outcome figures in the chapters are estimates, and the protocol promises nobody an election victory. If a chapter says otherwise, Exact Answers and the Charter are correct. For a candidate: ten questions and ten steps. For a citizen, a mayor, a finance officer, a donor, a journalist, a scholar, a lawyer: answers by role. Everything in force in one file: llms-full.txt.
The Laws of Economics in Politics: Why the System Does Not Need Clever Participants¶
Chapter: 03 — Theoretical Foundations (paired with §13) File: 03_013e · v1 · 19 September 2026 Source: the architect's reply to the objection in 013d.5 ("informed does not mean choosing the best manager"): "AB-EXIT is in essence a repetition of the business environment and the economy, and there the stupid and the greedy created everything in this world; well, the clever helped them too. AB-EXIT is the laws of economics in politics, and that is all." A check against the literature on markets with non-rational participants.
1. The argument¶
The objection in 013d ran: refusing the sum selects the motivated and informed, but not necessarily the wise; a voter who stays may be self-interested or blinded. The architect's answer turns the question over: the economy never required wise participants. Bread, houses, aircraft and telephones were created not by an assembly of sages but by a multitude of people each of whom wanted to earn, and many of whom were mistaken. The result comes not from the quality of the participants but from the construction of the environment: a price, a personal stake, competition, the possibility of leaving, and the ruin of the loser. If the construction is carried over into politics, the wisdom of participants ceases to be a condition.
2. What is known about this¶
This is not a metaphor but a tested result, and in the harshest possible setting.
- Gode and Sunder (1993). In a double-auction experiment human traders were replaced with "zero-intelligence" programs: they quoted prices at random. The only restriction was a budget — no selling below cost and no buying above value. Allocative efficiency turned out close to one hundred per cent. The authors' conclusion: a market's efficiency derives largely from its structure, not from the motivation, intelligence or learning of participants; the market is "a partial substitute for individual rationality".
- Alchian (1950) showed the same for firms (from the assistant's memory): even if decisions are taken at random, the environment keeps those who happened to be right, and from a distance it looks like intelligent behaviour. Selection does the work that is attributed to intellect.
- Becker (1962) (from memory): the demand curve slopes downward even with irrational buyers — because the budget is limited.
- Smith and Hayek. We get our dinner not from the butcher's benevolence; and a price transmits knowledge that no single participant possesses.
What all four share: the desired property of the system arises from constraint and selection, not from the quality of people. The key words in Gode and Sunder's experiment are budget constraint: a zero-intelligence trader cannot trade at a loss to himself. Remove that constraint and the efficiency disappears.
3. The transfer: what politics lacked and what the protocol adds¶
| An element of the economy | In politics today | After the protocol |
|---|---|---|
| A price | The vote is free and therefore says nothing about how much its owner needs it | The vote has a price — the sum that must be forgone |
| A personal stake | None: voting costs nothing | Everyone who votes has paid by refusing D — that is the budget constraint of Gode and Sunder's experiment |
| Exit | The dissatisfied have only voice; one can leave only by not turning out, which nobody sees or counts | Exit is lawful, paid and counted: the exit share is a public number |
| A profit-and-loss signal | No common number showing whether governance was good | The sum tied to the median, and the return of people with their votes (040b.2b) |
| The ruin of the loser | A bad government does not go bankrupt: its income is compulsory (018.41) | The thermostat: a bad government loses voters and weight, a good one gets them back (015b.6b) |
| Competition | Exists, but for the sympathy of the indifferent | For the vote of someone who counts money (019c.3) |
The protocol's name is literal here. In Hirschman ("Exit, Voice, and Loyalty", 1970) a dissatisfied customer has two means — to leave for a competitor or to speak up — and organisations deprived of the first channel deteriorate. Politics was a domain with voice only. The protocol adds exit — not from the country but from voting — and makes it measurable (013).
4. What this changes in the estimates of 013d¶
The objections of 013d.5 — "the informed may be self-interested", "the motivated may be blinded" — remain true as facts but lose weight as objections: a market too consists of the self-interested and the blinded. The question moves from people to construction: does the voter have a personal stake, and does the system have selection by result. The first the protocol gives for certain; the second — through the thermostat, with a lag of one cycle. The assistant keeps the estimate "those who stay choose more accurately than those who leave" at 80–85 %, but changes its meaning: even if that assumption proves weaker, the system may work through its construction, like a market with "stupid" traders. The probability that the protocol improves the selection of government by at least one of the two routes the assistant puts higher — at about 90 %.
5. Where the analogy ends¶
"And that is all" is stronger than can be defended, and three differences are worth naming ourselves.
First. In a market a person bears the consequences of his own choice: bought badly — lost his own. In an election the winner governs everyone, those who left included. Refusing D is a stake at the door, not a loss for a bad choice. The first version of this point concluded that the constraint "screens out those who do not care rather than punishing those who were wrong". The architect objected: "your candidate will be booed and never elected again, you are humiliated; that is no punishment for the apathetic, but for strategists it is a disgrace." The objection is accepted: the one who stays faces three punishments for a bad choice, and today's voter faces none of them.
| Punishment | Why it falls precisely on the one who stays |
|---|---|
| The forgone sum | He paid for the vote by refusing D and got a bad government for it; his neighbour took the money and lost nothing |
| His own damage | Those who stay are more often the ones who pay levies, run a business, raise children (019c.3): bad governance hits them first and hardest |
| Disgrace | Those who stay are a small and attentive circle where results are watched and it is remembered who backed whom; with an open verifiable vote (048i) the choice is public altogether. An apathetic voter suffers nothing when his candidate fails — he does not even remember whom he voted for; for a strategist it costs reputation |
So the constraint is softer than the market's — no money is debited from an account for a mistake — but it is not zero, and it is selective: it presses on the one who takes the decision, not on the one who declined to decide. To the third row the architect added: "the strategist will punish himself, understanding how he erred: he thought he was clever, and here is a failure. He does care — he put money on the candidate." This is a separate, fourth punishment — an inner one, and the most reliable, because it needs no witnesses: a person who considers himself calculating and has paid for his calculation experiences failure as a blow to his self-esteem, not as someone else's misfortune.
The first version of this point set a caveat here: one who has paid for a choice sometimes does not admit the mistake but doubles down. The caveat stands, but measurements say that money at stake works rather the other way. In the experiments of Bullock, Gerber, Hill and Huber (2015) supporters of two parties were asked factual questions about politics; ordinarily the answers diverge along party lines — people "cheer" for their own side even on facts. When a small sum began to be paid for a correct answer, the gap between the parties' supporters shrank sharply. The authors' conclusion: a considerable part of partisan blindness is not sincere error but a costless expression of loyalty; when a mistake costs something, a person starts answering what he actually thinks. This is the same mechanism as the protocol's: a free vote can be cast "for our side" out of spite and principle; a vote that has been paid for is cast with calculation. Doubling down happens, but it has a limit of a cycle or two, because the bill for bad government comes to the same person.
Second. The first version of this point said: a market works badly with public goods and costs imposed on third parties, and politics deals precisely with those. The architect: "again a comparison with an ideal; public administration today is at schoolchild level and utterly inefficient." The objection is accepted, and the error was double. First, the comparison must be not with an ideal provider of public goods but with the one in force: roads, schools and hospitals are today allocated by a government chosen by an indifferent majority and pressed by the recipients of levies (019c.2); the bar to clear lies low (015b.6b). Second, the protocol does not hand public goods to the market at all. Decisions about them remain political; what changes is only to whom the decision-maker answers. What is carried over is not market provision but market discipline over those who provide.
What remains of the objection. Those who stay also decide on the goods used mainly by those who left: the village school, the district hospital, the bus. The temptation to economise on what is someone else's exists. But those who left have three levers that today's non-voter lacks: everyone votes in referendums on the rules (048g.6b); one can return with a vote in any cycle, and a mass return of the "burnt" is the loudest signal a government can receive (040b.2b); and the sum is tied to the median, so the impoverishment of those who left hits a number everyone sees. This is not a guarantee but a testable claim: an indicator for a pilot is spending on services in districts with a high exit share, before and after.
A comparison with today. The assistant presented this remainder as a risk of the protocol: after it the government may "write off those who do not vote". The architect: "again a comparison with an ideal; and today everything is fine for these people, is it?" The reproach is fair: the risk was described as if it did not exist today. Today an emptying village is already being written off — slowly, without money and without choice: the school is closed not because someone did the sums but because there is nobody to teach; the road goes unrepaired for years; the young have left; between elections the village is forgotten, and before elections a bus is delivered. That is precisely "taken away with nothing given in return", stretched over twenty years.
| Today | After the protocol | |
|---|---|---|
| What the village gets from the government | A handout before elections and oblivion between them | A sum to every adult in every cycle — for certain and unconditionally |
| Who counts what the village costs | Nobody | Whoever answers to a voter who counts money |
| What is offered when a service closes | Nothing | A replacement or money (by the logic of this section) |
| How the village can respond | In no way: its vote has already been bought | By returning with its votes; by voting in referendums on the rules |
| The probability of the village being written off | It is already happening | By the assistant's estimate 10–15 % — and with three levers against it |
So this is not a weak point of the protocol but a place where it improves things less than it could, should those who stay behave badly. The indicator for a pilot remains — spending on and access to services in districts with a high exit share — but it must be read against the same district before the protocol, not against a district that does not exist.
Hard decisions. The architect went further: "populism saves not those who should be saved but those whom the apathetic will enjoy seeing saved. Perhaps these people should not be saved at all but resettled for the same budget money; but without AB-EXIT nobody will go against populism for the sake of budget efficiency." There are two thoughts here, and they deserve to be accepted differently.
The first is a diagnosis, and it is exact. Today spending goes not where need is greater or the return higher, but where an indifferent voter will notice it: a top-up before an election, a school of twelve pupils as a symbol that "the village was not abandoned". Proposing otherwise — merging schools and bussing children, giving a family money to move closer to a hospital and work instead of maintaining emptying infrastructure — is suicide for a politician, even when it is better for the people themselves: the child gets a proper school, the old man a doctor half an hour away rather than three hours. Such decisions today cannot even be uttered, let alone taken. This is exactly what the protocol changes: for a voter who counts money the question "what does it cost and what do we get" ceases to be forbidden.
The second — "perhaps they should not be saved at all" — needs a boundary, and it is worth writing down before an opponent writes it. Those deciding about someone else's school will mostly be people who do not live in that village, while the villagers have mostly taken the sum. This is the very case the previous paragraph warned about, and efficiency is no excuse here. The boundary is this: moving is an offer, not an order; closing a service in order to force people to leave means coercing those who have no vote in this cycle. Money for the move, a guaranteed school place and housing — yes; a cancelled bus as an argument — no. And the villagers keep an answer they do not have today: to come back with their votes. If those who stay cut what is not theirs, and at the next cycle the whole village refuses the sum and turns out to vote — that is the loudest signal a government can receive (040b.2b), and at the same time an honest test of whether the decision was an offer or coercion.
The architect's objection to this boundary: "so the state and everyone else will pay for what these people cannot provide for themselves, and they take the money as well? A poor grandmother went off into the forest by herself and shouts from there: help, give me a house, a bus and a hospital?" The objection has a true core and a false picture, and the assistant does not lift the boundary.
The true core: nobody is owed an urban set of services in a village of thirty people at everyone else's expense. The boundary forbids forcing people to move by switching things off; it does not oblige anyone to maintain everything that was once built. The honest order is this: the cost of a service per resident is published; a minimum is guaranteed — a passable road, an ambulance, a visiting doctor, transport for children; everything above the minimum is decided by those who pay; and those for whom the minimum is not enough are offered a move with money and housing. That is precisely the decision that cannot be uttered today.
The false picture: the grandmother did not go anywhere. She was born in that village, worked and paid for forty years, and the village emptied around her because the young left; her pension of three thousand lei was set by the same state. She is not asking for something new — she is losing what there was.
And the main thing — why the boundary must not be lifted even for the sake of efficiency. It is not a matter of pity but of the protocol's integrity and the arithmetic of its adoption. The protocol promises that whoever takes the sum loses no right except the vote in this cycle (033c.1). If in practice it turns out that "you took the money — your bus is the first to be cut", the opponent's strongest attack comes true: "take it and you lose the right to everything" (040c.3). And it is precisely those who will take the sum who must adopt the protocol at a referendum — they are the majority. A protocol whose author tells them "perhaps you need not be saved" they will not adopt, and they will be right. The protocol delivers budget efficiency through the repeal of levies and competition for the money-counting voter; it has no need to extract it at the expense of those who trusted it.
A clarification after the next objection. The architect: "you are thinking like a populist again. One can close schools and remove buses that carry one person." One can, and the text above does not forbid it; but since it read that way, the boundary was named imprecisely. It runs not between "close" and "keep" but between replacement and abandonment without replacement. A school for one pupil is the worst thing one can give that pupil: closing it and taking the child to a proper school is cheaper for the budget and better for the child. A bus for one passenger is replaced by an on-demand car at a tenth of the cost. This is neither cruelty nor coercion but the ordinary work of a manager, which nobody today dares to do because a closed school is a ready-made television story. Coercion begins where a service is removed with nothing given in return, in the expectation that the person will leave by himself. The criterion is simple and verifiable: after the decision the child has a school, the old man has access to a doctor, and both have a way to get there; if so, it is efficiency; if not, it is squeezing out. The architect added: "and after AB-EXIT this will have to be done, and it saves the budget." Yes: this is not a side risk of the protocol but one of the things it is needed for — the same class of decisions as the repeal of levies (019c), only on the spending side: there one stops taking what is superfluous, here one stops spending on what serves nobody. And the money released is part of what the sum and the return of taxes are paid from (019c.6f).
How to replace: not with a programme. The architect objected to this too: "you do not even see that you write like a populist government that needs only a budget flow to pinch from; where it goes does not matter to them." The reproach lands on a specific spot. "A guaranteed minimum", "a visiting doctor", "an on-demand car", "housing on relocation" — all of these are worded as new state services, that is, new procurements, contractors and posts. A school is closed — a transport programme is opened; the flow is preserved, the channel changed, and one can go on pinching from it. That is how a government department reasons, and the assistant reasoned the same way.
The protocol already has the consistent answer, and it is the same as in the sum D itself: money goes to the person, bypassing the intermediary (056e.2). A school for one pupil is closed, and the per-pupil funding is given to the family: to drive the child, to rent near the school, to move — the family decides, and will decide more cheaply and better than any education office. A bus for one passenger is cancelled, and part of the saving goes to the passenger himself. The criterion of result remains — the child has a school, the old man has a doctor — but it is met not by a new line of spending but by a line disappearing while the person receives money. The flow through officials shrinks twice over: by the saving itself and by the part that used to settle along the way.
A calculation: a bus or a house. The architect: "a bus for three children over 50 km costs per year as much as building these people a house next to the school — work it out; populists do not even count this." The assistant's rough estimate for Moldova, a dedicated route, 175 school days; all figures are an order of magnitude, not a budget.
| Item | Per year |
|---|---|
| Mileage | 100–200 km a day (the bus waits at the school or returns empty) — 17.5–35 thousand km |
| Fuel (12 l per 100 km, 22 lei a litre) | 46–92 thousand lei |
| Servicing and tyres (1.5 lei per km) | 26–53 thousand lei |
| The driver, with taxes | about 170 thousand lei |
| Depreciation of the bus (60 thousand euros over ten years) | about 120 thousand lei |
| Insurance, parking, other | about 20 thousand lei |
| Total | 380–450 thousand lei — 19–23 thousand euros a year |
A modest 70 m² house at the construction cost from 019c.6c (425–715 euros per m²) plus an inexpensive plot in the village with the school comes to 37–57 thousand euros. So the architect was off by no more than a factor of two: the house pays for itself not in one year but in two to three years of the route. And the route lives as long as the children are at school — nine to twelve years: that is 170–270 thousand euros, four to six houses for three families, and after the children leave school a house remains, whereas what remains of the bus is a written-off bus. Per child the bus costs 6–8 thousand euros a year — by the assistant's estimate several times the entire per-pupil funding (the funding norm itself was not checked here).
Caveats. A dedicated bus for three children is an extreme case; usually a route collects several villages, and the extra cost of an added 50 km is smaller. The second caveat of the first version ran: "not every family will want a house by the school: it has a farm, land, elderly parents". The architect: "the profit from the farm goes to them, while the bus is paid for by the state." True, and this is not a caveat but one more argument: the income from living remotely is private, while the costs of remoteness are shifted onto everyone. As long as the bus is free, the family does not know whether its farm is really profitable: it may bring in less than the route costs and survive only because others pay for the route. The fair principle is simple: the state owes the child an education and pays the same for every child wherever he lives; the premium for remoteness follows from the family's choice, and is paid by whoever earns from that choice. A farm that can bear its own road to school will remain; one that cannot was loss-making before as well, only the loss was not visible.
So the conclusion is not "build houses instead of buses" but the same as above: the sum a service costs must be known and named to the family — "your route costs twenty thousand euros a year; here is that money, decide for yourselves." Some will move, some will buy a car, some will arrange things with a neighbour. The first version said that today nobody works out this sum; a check showed this to be wrong: experts and ministries have been working it out for more than ten years (053b). It is not worked out by whoever decides, because there is nobody to demand it of him: those who decide elections do not care, and those who run the route benefit from its existing.
Three claims by the architect and what is accepted of them. "Today the votes of these remote people are bought with handouts of rusty buses from the authorities. And the fact that they take the money proves that their farm makes no profit and that there is no economic sense in their living there. And who was born and lived where is no argument at all; the availability of housing in the new place — that is an argument."
The first is accepted in full and explains why nobody works out the sum. A loss-making route is not a planning error but payment for votes: the authorities buy a village with a bus just as they buy a pensioner with a top-up before an election (019d.3). As long as the vote is free and goes to whoever brought the bus, there is nobody to count its cost and no reason to. After the protocol the village mostly takes the sum, the bus ceases to be an electoral instrument and can for the first time be assessed as a bus.
The second is not accepted: it is a logical error, and it is worth naming before an opponent names it. Taking the sum says how much a person values his vote, not whether his farm is profitable. The sum will be taken by a successful farmer indifferent to politics and by a city programmer alike. The connection is only indirect — two thousand lei mean more to a poor person — but "took the money, therefore lives irrationally" does not follow, and the protocol must not draw such a conclusion: choice B carries no consequences, labels included (019d.4b). The architect insisted: "if I have a good farm, then either it is large, or a non-working wife drives the children, or I pay for the bus myself without the authorities; and if I took the money, then everything is the other way round." The assistant did not agree, and the disagreement is recorded as it stands. By Bayes (013d), taking the sum is evidence, not proof, and weak evidence at that: the sum is taken by 55–65 % of all voters, including a great many who are not poor and simply indifferent to politics, so it raises the probability of "poor" only slightly; it says far more about attitude to politics. A well-off farmer who does not care who the mayor is will take two thousand lei as readily as his poor neighbour. And the main point — the argument is not needed: that a route is loss-making is proven by direct count, what it costs and how many children it carries, with no guesses about who chose what. To build a conclusion about a person on his choice of A or B is to pay with the protocol's integrity for something already visible in the budget sheet. The disagreement was then removed by the architect himself: "it does not matter whether he took the money or not; if he is a good farmer and lives VERY far away for that reason, then he has worked out from his profit how he will get the children to school." With this the assistant agrees completely, and it is the same principle as above: whoever chose remoteness for the sake of income includes its price in his own calculation; the sum D has nothing to do with it either way.
The third is accepted, and it matters more than it looks. Where a person was born is indeed no ground for demanding other people's money. But "the availability of housing in the new place" is precisely the real reason people stay in emptying villages. A house in such a village is worth a few thousand euros, a flat in a town ten to twenty times as much; having sold everything, one still cannot move. And urban housing costs two to four times its construction not because of concrete but because of the status of land, permits and waiting (019c.6c). The result is a closed circle, both links of which were made by the state: artificially dear housing locks people in where there are no jobs or services, and then the budget pays to run a bus out to them. Subtraction in housing — the possibility of building a house cheaply and lawfully near a district centre — does more for resettlement than any resettlement programme: people do not need to be transported; they need somewhere to go at a price they can pay.
A calculation: the whole village or flats. The architect: "and if one adds the bus, and gas, and electricity, and road repairs, and the school, and the hospital — perhaps it is cheaper to buy them flats?" The assistant's rough estimate for a notional village: 20 households, about 40 residents, three schoolchildren, 5 km of access road. All figures are an order of magnitude.
| Item | Per year |
|---|---|
| The access road: upkeep and a share of future repairs, 5 km | 25–75 thousand euros |
| The school bus (see above) | 19–23 thousand euros |
| A medical post or a visiting paramedic | 8–12 thousand euros |
| Networks — electricity, gas, water: the share of upkeep that other consumers pay today through the tariff | 3–6 thousand euros |
| A share of the commune's costs: the mayor's office, lighting, refuse | 3–5 thousand euros |
| Total | 58–121 thousand euros a year — 3–6 thousand euros per household |
A 45 m² flat in a district centre (not the capital) is on the order of 27–40 thousand euros; for twenty households — 540–810 thousand euros. Payback: from four to five years in the case dear to the budget to fourteen in the cheap one. So yes, over a horizon of ten to fifteen years flats are cheaper than upkeep, and the gap grows: each year there are fewer residents in such a village, while the road and the networks cost the same.
What must be kept in mind in this arithmetic. Part of the costs will not vanish when people leave: the road also leads to fields, the line also feeds a farm. Not everyone will want to go, and nobody may be forced (the boundary above). Mass purchase of flats by the state will raise prices in district centres if one cannot build freely there — it runs into housing again (019c.6c). And the main point, by the architect's own argument: "buy them flats" is a programme, that is, a procurement, a contractor and a flow to pinch from. The consistent solution is the same as with the bus: work out what the village costs and offer each household its ten-year share as money for housing — on the order of 30–60 thousand euros, which is just the price of a flat. Whoever agrees leaves with the money and chooses where to go; whoever stays, stays on the minimum. When a household leaves, its share of costs disappears from the budget for good.
The architect's refinement of the calculation: "and if the village has a farm, and there is work, and people — the grandmother need not be resettled. Nor need she be bought a flat in the centre of Chișinău: just in a bigger village." Both refinements make the argument more exact and stronger. The first sets the criterion: the question arises not for any village but only for one where the cost per household is several times that of its neighbour; a village with a farm, work and people costs little per head, and nobody touches it. The second changes the arithmetic for the better: a house in a bigger village ten to fifteen kilometres away, which already has a school, a doctor and a bus, costs, by the assistant's estimate, 8–15 thousand euros rather than 27–40. For twenty households that is 160–300 thousand euros against 58–121 thousand a year in upkeep — a payback of one and a half to five years, not up to fourteen. And such a move is more humane: the same neighbours, the same land within a bicycle ride, the same way of life. It is not resettlement to a city but what villages did by themselves for centuries — drawing together to where life is.
And one more consequence, named by the architect: "flats can be built — and construction in the country lifted." Money that today goes into fuel for an empty bus and into repairing a road to twenty households turns into orders for houses: work for bricklayers and carpenters in those same districts, demand for local brick and timber, and instead of a cost that burns up within a year there remains housing that stands for decades. In a country where about two thousand permits for residential buildings a year are issued for 2.4 million inhabitants (019c.6c), even a few thousand such orders are a noticeable addition. The condition is the same as everywhere in this section: the building should be done by private builders on the order of families who received the money, not by a state programme with a contractor; and building must be possible — without land conversion and years of permits, otherwise money for housing will lift not construction but prices. Subtraction in housing and money following the person work only together.
Why this is not done today has been said above: a loss-making village is bought votes, and the offer "take forty thousand and move" is a ready-made headline "the authorities are evicting the elderly". A calculation that on paper benefits all three sides — the household, the budget and the other taxpayers — is politically impossible as long as elections are decided by whoever watches that headline on television.
A caveat for a campaign. The phrase "AB-EXIT wants to resettle villages" is a ready-made headline for an opponent. In public language this is a subject not for the protocol but for the programmes of candidates, who will for the first time be able to raise it; the protocol says nothing about it and should say nothing.
Third. In a market whoever has more money has more votes. The protocol does not carry this over: those who stay have one vote each, and the sum is the same for all. This is not a flaw in the analogy but a deliberate boundary — yet it means that "the laws of economics" are not taken whole.
Rights and support are two different things. Comparing the repository with classical designs for a state, the assistant called it a gap that "what protects a minority from those who stayed to vote with a tenfold weight is not analysed." The architect replied sharply: "you are thinking like a populist again. They must not be wronged — but there is no need to make everyone happy. If a minority persuades the pragmatists that it needs support, it will be given support — but not at the others' expense merely because they dyed their hair pink and walk on stilts. It is the populist who needs the pretty picture of himself personally rescuing a panda from a helicopter for a million from the budget; that is not a panda, it is propaganda at the people's expense."
The analysis shows that two things were stuck together in the assistant's objection, and the architect is right on both. The first is rights: not taking away property, liberty, the vote. Today these are protected not by voters but by the constitution and the courts, and the protocol does not touch that; the yardstick is the usual one (compare with the existing system) — the protection stays what it was. Moreover, those who stay to vote are the better informed, and in survey data since Stouffer (1955) it is education and awareness that are most strongly tied to support for the civil liberties of unpopular groups; so the gap, if there is one, shrinks rather than grows. The second thing is support: money, programmes, privileges. Here the architect states a rule that matches the rest of this section: support goes to whoever persuaded those who pay, not to whoever looks best on camera. Today's system works the other way round: help goes where it yields a picture and votes, which is why a panda gets rescued from a helicopter while the water main goes unrepaired. "Do not wrong them" is a duty without conditions; "support them" is a decision that has to be earned by argument. The protocol does not weaken the first and makes the second honest. 🟡
So the exact formula: the protocol carries into politics not the market but three of its load-bearing elements — a price, a personal stake and measurable exit. That may suffice — the experiment with "stupid" traders says that little is enough — but only the first working cycle will show it. 🟡
Related: 013d (Bayes: one quantity and percentages) · 013 (game theory, Hirschman) · 015b.7 (greed, stupidity, conscience) · 018.40–41 (the pizzeria test) · 019c (reform by subtraction) · 040b.2b (two sensors) · 015c (a change of sign)