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 Market Price of a Vote Has Already Been Measured: Moldova 2024 and Other Countries¶
Chapter: 04 — Electoral Dynamics (paired with §19b) File: 04_019d · v1 · 19 September 2026 Source: the architect's and the assistant's analysis of the question "will the elderly take money instead of a vote, and are polls needed for that" (048g.3c); Moldovan police data on the vote-buying scheme of autumn 2024; data on the price of a vote in Bulgaria and the Philippines.
How to read this chapter (note of 02.10.2026). The text contains wording that is easy to misread: sums and formulas with coefficients are worked examples: the size of the payment is set as a percentage of the median income approved by referendum; a "reversible" choice means "at the next election": within one election exit is final. The exact-answers sheet 1d and the charter 048m are in force.
1. A question usually answered by a poll¶
The protocol's main empirical assumption is that a sizeable share of voters will prefer the sum D to the vote. The standard objection: "people will not sell the right to vote, it is a matter of dignity; ask them — they will refuse." A poll is a poor instrument here: to the question "would you sell your vote?" people answer the way one is supposed to answer. In Bulgaria in 2024, in a pre-election survey, only 10 % admitted they might vote for pay or under pressure, and another 12 % were unsure.
But there is a measurement more reliable than a poll — the market. Votes are bought at real elections for real money, and the price at which the deal is struck is known from criminal cases and journalistic investigations. That is revealed preference, not stated preference.
2. Moldova, autumn 2024¶
According to the Moldovan police, ahead of the referendum on European integration and the presidential election, 39 million dollars were transferred to the accounts of Moldovan citizens through a sanctioned Russian bank: 15 million in September and 24 million in October. Investigators established the identities of about 138 thousand recipients. The average sum was about 280 dollars per person. The scheme was coordinated through a network of messenger bots: a person was told how to open an account and which tasks to carry out. The announced rates: 2,000 lei for minimal tasks and 5,000 lei if the polling station voted the required way.
A comparison with the protocol:
| The 2024 scheme | The protocol | |
|---|---|---|
| The sum | 2,000 lei base, up to 5,000 | D by the formula at K = 1 % — 2–2.5 thousand lei (033c.9c) |
| Legality | A criminal offence for both sides | A lawful payment |
| Who pays | A foreign bank under sanctions | One's own state |
| What is demanded in return | Vote as instructed, carry out tasks, report back | Nothing: only not voting in this cycle |
| How it is arranged | A bot, an account in a foreign bank, the risk of exposure | A button in the same channel as voting |
| How many agreed | About 138 thousand — a number comparable to a tenth of all who came to the polls | — |
Two thousand lei is exactly D. For that sum more than a hundred thousand people accepted criminal risk, someone else's errand and shame. The protocol offers the same sum without all three costs. The estimate "55–65 % will take the sum" (015b.6) remains a forecast, but its lower bound is no longer speculative: a tenth of those voting took this money under the worst conditions imaginable.
3. Who took it and why: the pension¶
Moldova has 672 thousand pensioners — about 28 % of the population. The average old-age pension is about 4,400 lei, the minimum 3,055–3,525 lei. The sum D is from a half to three quarters of the minimum pension; it reaches a whole minimum pension at K of about 1.5 %. But measuring against the whole pension is the wrong yardstick: almost all of a minimum pension goes on food and utility bills, and a few hundred lei a month remain as free money. D is a pensioner's free money for roughly half a year, received at once. The architect: "a whole pension for medicines and presents for the grandchildren".
On the other pan of the scales lies not "civic duty" but its actual content. The disciplined elderly voter often votes ritually, for a candidate who has not won in decades: "or they will go and vote for Zyuganov for the tenth time, who never wins." A person knows the expected result of such a vote from experience. His choice is not between duty and money but between a ritual without consequences and half a year of free money.
And the architect's observation on where this sensitivity comes from: a low pension is not an accident but the condition under which a pre-election top-up works. A system that keeps a pensioner on three thousand lei has itself created the voter for whom two thousand decides everything — and has itself left him open to anyone who pays, a foreign buyer included. The protocol turns this against it: the same sensitivity to a sum is captured not by the one handing out the favour, nor by the vote-buyer, but by the person himself.
4. Other countries: D is above the market price¶
The architect supposed that in other countries the ratio would be comparable. A check on two countries with well-documented vote buying shows that Moldova 2024 is the upper bound: the buyer was geopolitical and paid generously. Usually the market pays less, and D turns out noticeably higher than the market price.
| Country | The market price of a vote | D by the formula at K = 1 % (the assistant's estimate from the average wage) | The ratio |
|---|---|---|---|
| Moldova, 2024 | 2,000–5,000 lei | 2–2.5 thousand lei | D ≈ the base price |
| Bulgaria, 2024 | 100–300 leva (50–150 euros) | on the order of 300 leva | D at the upper end of the price |
| The Philippines, 2025 | 250–500 pesos in mayoral races; in the hottest contests sums of up to 15,000 pesos were reported | on the order of 3,000 pesos | D six to twelve times the usual price |
The consequences. First: vote buying becomes dearer after the protocol. To buy a vote that a person can lawfully exchange for D, the buyer must pay more than D — otherwise the seller takes the sum from the state and risks nothing; this is the floor price of a vote (048h.5). Where the market paid a third or a tenth of D, it simply closes. Second: a foreign buyer, as in Moldova, now has to outbid not zero but D; but this does not stop a rich buyer and has a reverse side — see §4b, where the first wording of this point is corrected. Third: the protocol does not create a market in votes — it exists, is documented and serves the worst buyers; the protocol makes the state the buyer of last resort and the seller the one who gains.
4b. Does the protocol solve "the Shor problem": an honest count¶
The architect's conclusion: "so AB-EXIT solves, simply by protocol, the real Shor problem that the security service of an entire country" could not prevent. This is half true, and the other half must be written down, because it is exactly what an opponent will find.
An error in the first version of §4. It says the buyer will have to outbid D "among the motivated who stay". That is wrong: the buyer recruits not the motivated but the same poor people as before — only now from among those who were going to take the sum: "do not take D, take more from me and vote as required." The choice is reversible, and the protocol cannot prevent this.
The arithmetic for a rich buyer. Before the protocol 39 million dollars yielded about 138 thousand recipients among 1.5 million voting — on the order of nine per cent. After the protocol the price of one vote rises: to the former 2,000 lei one must add the D the person forgoes, that is, about 4–5 thousand lei. But those voting become two and a half times fewer, and each bought vote weighs correspondingly more. The same 39 million at 5,000 lei is on the order of 130 thousand votes, now out of 600 thousand, that is, about a fifth. The floor price of a vote closes the cheap market — the local buyer with buckwheat and two hundred lei — but it does not hinder a geopolitical buyer with tens of millions, and the shrinking of the pool helps him. This is the same narrow-pool vulnerability as in 039b.3.4 and 015c.7b, and it must not be hidden.
What this arithmetic left out. The architect objected: "the sum paid per purchase is twice as large — so it is twice as visible and conspicuous, both in cash and in transfers. How does that help the buyers?" The objection is fair in two respects, and the second matters more than the first.
The first is conspicuousness. The total volume of money at the same budget is the same, but each payment is twice as large: five thousand lei to a pensioner on three and a half is a transfer that stands out from his usual receipts more than two thousand does, and in cash it is twice the banknotes, couriers and meetings. To the trace in the money is added a trace in the choice itself (the table below). Buying becomes not cheaper and quieter but dearer and louder.
The second is what the calculation above held constant and what will not stay constant: the buyer's free base. In 2024 about 740 thousand people voted against European integration (from the assistant's memory: the referendum result was 50.4 % to 49.6 %), while the recipients of money numbered 138 thousand: four fifths of that result cost the buyer nothing — they came from television, habit, handouts and sincere conviction. The bought votes were a top-up to a free base. After the protocol it is exactly this base — the poor, the elderly, the rural, the viewers of broadcast television (048g.3c) — that mostly takes the sum and leaves the polling stations. The buyer has to pay for what used to come free. To get half of the 600 thousand who stay, he needs about 300 thousand votes; if a hundred to a hundred and fifty thousand of the free base remain, he must buy another hundred and fifty to two hundred thousand at five thousand lei — that is 45–60 million dollars instead of 39, and the bought will make up not a tenth but a quarter or a third of all those voting. Such a share cannot be hidden either in transfers or in the statistics of refusals by polling station.
A correction to the conclusion. The phrase "the arithmetic favours the buyer" is true only for one additional bought vote: its weight really does grow. For the campaign as a whole the arithmetic is the reverse: the protocol takes the free majority away from the buyer and leaves him only a paid one, dear and conspicuous. The estimates in this paragraph are an order of magnitude built on the assumption that it is above all the poor and the elderly who leave; the symmetrical check — how many free votes the opposite side loses — only a pilot will give. The narrow-pool vulnerability remains for small constituencies, where what has to be bought is not two hundred thousand people but two hundred (039b.3.4).
What the protocol really changes — and what the security service does not have today.
| Today | After the protocol | |
|---|---|---|
| When the buying becomes visible | After the election, from bank transfers; in 2024 identities were established when the money had already been handed out | Before the election: a recruit is a person who by every indication should have taken the sum and did not; a surge of such refusals at the polling stations of one district is visible on the day the choice closes, before the vote |
| Whom to check | All recipients of transfers from abroad | A narrow group: those who refused the sum where refusals are anomalously many |
| What holds the influence network together | Handouts: top-ups for pensioners, "social shops" — gratitude is expected at the polling station | The same sum comes from the state unconditionally; the handout ceases to be the only source of free money |
| What the seller loses if exposed | Nothing but the risk of a fine | The lawful sum he could have received without risk |
A threefold narrowing of the search. The architect: "all the same, the sample in which one has to look will be several times smaller and simpler." This is true and does not contradict the line drawn below: the line concerns whom to treat as a suspect, not where to look. Today the security service has no map: buying may be going on anywhere among a million and a half voters, it is visible only in transfers and only after the fact — in 2024 identities were established when the money had been handed out and the votes cast. After the protocol the search narrows three times over.
| Narrowing | Today | After the protocol |
|---|---|---|
| By people | Any of the 1.5 million voting | A recruit can only be among those who refused the sum — about 600 thousand, most of them ordinary motivated voters who create no anomaly |
| By place | The whole country | Polling stations where refusals are noticeably more numerous than the composition of the population implies: in a village where twenty people should refuse and a hundred and twenty did, one has to look not at thousands of stations but at dozens |
| By time | After the election, sometimes months later | On the day the A/B choice closes — before the vote, when the result can still be protected |
The security service gets what it did not have: an answer to "where and when to look" without knowing a single surname. At an anomalous station one looks for the organiser — a courier, a bot, the accounts from which payments flow; in the 2024 scheme such networks left more than enough traces once it was known where to dig. A person enters the case only later, by court order and on evidence from the money trail, not by the fact of his choice. So a "burnt" poor man who returned to vote does not come under scrutiny: if many returned at his station, the check finds no transfers and is closed.
The buyer's dilemma. Knowing that a surge of refusals at a station gives him away, the buyer cannot recruit densely: cheap recruitment through one village elder for the whole village becomes the most conspicuous kind. What remains is to spread recruitment thinly across the country — and that means more intermediaries per bought vote, dearer and less reliable: every extra intermediary is an extra person who may talk. Dense is visible, thin is expensive; the statistics of refusals leave him no third option.
A line that must not be crossed. The architect saw a powerful tool here: "mathematics comes into force, and working out who was bought is very simple; the security service can then check his accounts." The first half is right, the second is dangerous, and the assistant's objection is worth recording in full.
The mathematics reveals an anomaly at a polling station, not a guilty person. A poor person who refused the sum and went to vote is not only a recruit. He is also the "burnt" voter — the one the government drove to it and who came back with his vote; by the protocol's logic his return is the real sensor (040b.2b). If refusing the sum on a low income makes a person a suspect, the one placed under scrutiny is exactly the one the protocol was built for, and button A ceases to be free: for the poor it begins to cost not only D but a visit from an investigator. That breaks the protocol's first rule — the choice is voluntary and without consequences — and creates the very selectivity the statute forbids.
The second objection is the mirror one. A register of "who refused the money and votes", in the hands of an unscrupulous government, is a list of its opponents, compiled by the state itself and already sorted by vulnerability: the poor on it are the most defenceless. A tool that today catches those bought by Shor catches dissenters tomorrow. The protocol must not hand it to any government.
So the line is this. Only aggregates are published and analysed: the share of refusals at a polling station and its deviation from the forecast. An anomaly triggers a check not of voters but of the buyer: money flows, recruitment networks, bots, organisers — what in 2024, too, was provable without questioning a single pensioner. A particular person's account is checked only by court order and only on evidence unconnected with the very fact of his choice. The choice of A or B is itself never grounds for suspicion. That way the tool keeps its force — a network with thousands of transfers is visible in aggregates too — and does not turn into a search of the poor at the door of the polling station.
The upshot. The protocol does not replace a security service and does not make vote buying impossible; against a buyer with a state's budget, price is no defence. It does something else: it turns hidden buying into conspicuous buying (recruitment leaves a trace in the A/B choice itself) and deprives handout networks of their monopoly on the poor voter's free money. For the first to work, the statute needs two norms: publication of the number of refusals of the sum by polling station before polling day, and a threshold at which an anomaly triggers a check. Without them conspicuousness remains an opportunity nobody is obliged to use. An indicator for a pilot: the share of refusals of the sum by polling station compared with the forecast from the composition of the population. 🟡
5. Weak point of the section¶
The vote-buying market measures the price of a vote cast as instructed — the person does not merely abstain but votes as told; that is a different service, and readiness for it may be lower (shame, risk) or higher (someone will take money for a vote he would have given that candidate anyway). The average sum in the Moldovan scheme is above the base rate because some recipients were organisers, not voters; the figure of 138 thousand is recipients of transfers, not proven sold votes. The estimates of D for Bulgaria and the Philippines were made by the assistant from the average wage, not the median, and serve only as an order of magnitude. The pensioner's subsistence minimum in §3 is given from memory, unchecked. The share of each group that will take the sum will still be shown only by a survey with real stakes or a pilot. 🟡
Related: 048g.3c (two watchmen, the pension) · 019b (the poor as the most expensive voter) · 048h.5 (the floor price of a vote) · 033c.9c (the order of the sums) · 015b.6 (the forecast of the exit share) · 053 (Moldova) · 019c (reform by subtraction)