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.

4. The Public Counter and Its Effect on Elections¶

Chapter: 01 File version: v2 (universalised) Date: 2026-06-11 · universalised 2026-09-15 Source: v6.53 §8, §9

An early-draft layer (note of 01.10.2026 after audit 040m). In §4.7 the "strict confidentiality" of the choice and the reference to "Article XIV" are the earlier wording: Article 9 of the charter is in force. In §4.8 the efficiency bonus and payment "one day before the election" have been removed: the payment arrives at once after the choice. The "rating" scale in §4.3 is to be read by change, not by level: the level depends on the approved percentage (§4.3b). In §4.9 the covering of the early premium by the late discount has not been calculated and is left to financiers. The closing of the window is a parameter, five days by default. The exact-answers sheet 1d and the charter 048m are in force.

Note of 02.10.2026. "Bonus" and "efficiency bonus" in the examples of §4.4–4.5 are also the earlier wording: the protocol has one payment. Comparing payments between territories and across years remains: the payment equals a percentage of the median and so grows with the median. "About 600 units" is an example.

On the numbers in this section. All amounts are in generic units for an economy with a median wage of roughly 40,000 units/year: a dividend of about 600 units per cycle. Substitute your own economy's median — the proportions hold.

4.1. The publication mechanism¶

During the declaration window (30–5 days before the election) the jurisdiction publishes DAILY, on its official site and on the AB-EXIT portal:

the number of citizens who chose the Dividend (Option B)
the number of citizens who chose the Vote (Option A)
the number who have not yet declared
the percentage in each category

The data is refreshed every 24 hours. Individual choices (who exactly chose A or B) remain CONFIDENTIAL. ONLY the aggregate is published.

On the length of the window. In the first version the window closed 15 days before the election; on 20 September 2026 this was changed to 5 days (the reasoning is in 4.3b: those who think need time to watch the campaign through, while candidates get enough from the first wave of the counter). The architect set a limit here: "political fine-tuning this delicate is already a matter for political scientists, not architects." The protocol sets the principle — the window opens early and closes as close to the election as the payment machinery allows; the exact number of days is chosen by each country and city, and 5 days is a default, not a norm.

4.2. Why the counter must be public¶

Without a public number AB-EXIT runs at 50%. With one, it runs at 100%. The reason is the barrier to entry for new candidates.

Without a public counter:

A potential candidate does not know how many people will vote. Maybe 200,000 (as before). Maybe 50,000 (with AB-EXIT). At 200,000 you need 100,000 votes — unrealistic. At 50,000 you need 25,000 — realistic. But the candidate DOES NOT KNOW. Uncertainty breeds fear. Fear keeps them out. The machine survives.

With a public counter:

Day 10 of the declaration window: "Dividend taken: 170,000 (72.1%). Will vote: 65,745." The potential candidate: "I need 33,000. The incumbent has about 6,000 machine votes. This is doable." The candidate runs. Four more candidates run. The machine is dead. AB-EXIT runs at 100%.

The public counter \= a calibrator of courage. Each day the number grows → the barrier falls → new candidates dare.

4.3. The dividend percentage as a trust rating for the incumbent¶

The share of citizens who chose the dividend becomes an objective rating of trust in the sitting government. This is not an opinion poll (subjective, small sample) — it is a decision with real money at stake.

50% took the dividend \= "Half consider their vote worth more than the dividend. The incumbent is DOING THE JOB."
70% took the dividend \= "70% consider the dividend worth more than their vote. The incumbent FAILED TO CONVINCE."
90% took the dividend \= "90% do not want to vote even for free. The incumbent HAS FAILED."

No existing mechanism provides a trust rating this precise, this unmanipulable, and this financially backed.

4.3b. Two camps: why the percentage alone is not a rating¶

Added on 20 September 2026. The architect corrected §4.3: "it is all more complicated here. The percentage gives no understanding: a person is fully satisfied with the authorities and takes the money — or he does not care and takes the money. These are entirely different camps." The correction is right, and §4.3 in its earlier form contradicts the rest of the repository: by the logic of the thermostat (015b.6b), exit grows under good governance — a satisfied person reckons they will manage without him. So "90 % took the sum" may mean the authorities' failure or their success, and the level cannot be read as a grade.

	Camp 1 — the satisfied	Camp 2 — the indifferent
Why they take the sum	"things are going fine, they will manage without me"	"I do not care who is there"
Do they follow the campaign	yes: they watch debates and read	no
Can they be brought back to the polls	yes — if candidates persuade them there is something to lose or to choose	no, whatever the campaign
When they declare	late: they watch the campaign first and think	at once: free money is needed now, and there is nothing to think about

Two consequences follow. First: the rating is not the level but the curve. Camp 2 does not react to campaign events at all; camp 1 does — after strong debates or a scandal the flow of declarations for the sum slows and the flow of declarations for the vote rises. So what speaks of the authorities is not how many people exited but how the counter responds to the campaign and how it has changed from the previous cycle: steady high exit with a calm counter is satisfaction or indifference; a counter that turns after debates is camp 1 changing its mind, and that is a signal no poll provides.

The second consequence concerns time. The assistant first read the architect's words the other way round and wrote that the indifferent drag on to the last day. The architect clarified: "camp 1 may think it over; camp 2 takes the money at once — they need free money now and do not want to think; and here one can count the dissatisfied by the old scheme." This changes the picture, and for the better. The counter moves in two waves. The first, in the very first days of the window, is camp 2: it is large, fast and does not respond to the campaign. The old reading of §4.3 applies to it: these are people for whom the vote is worth nothing, and the size of the first wave is a measurement of alienation, comparable from cycle to cycle. The second wave, towards the end of the window, is camp 1: it is small, slow and depends wholly on what people saw in the campaign.

Hence "time must be left", and it is a remark on the design itself. In the first version of §4.1 the window closed 15 days before the election, while the main debates usually take place precisely in the last two weeks: a satisfied, thinking person has to decide before he hears the candidates. There is no need to close the window that early. Candidates get their calibration (4.2) from the first wave: the bulk of those taking the sum is known in the first days, and the second wave changes the number of voters by per cent, not by multiples. So the window can stay open almost up to the election, taking nothing from candidates and giving camp 1 back what it is waiting for; the term in §4.1 has been changed to 5 days. With this arrangement the premium for early declaration (4.9) is needed for the purity of the first figure. The assistant had written that it is hardly needed to pull the indifferent forward; the architect kept it: "for the purity of the figure for the alienated and apathetic, so that they do not drag on to the last day; but that is already a figure to be discussed for each place." Right: most of the indifferent will come at once, but some will put it off out of mere forgetfulness and blend into the second wave, spoiling both measurements. The premium gathers them into the first days and fixes the boundary: whoever took the premium was not waiting for the campaign, and the first wave is separated from the second not by guesswork but by date. The size of the premium and the length of the "early days" are parameters each country or city chooses.

Who needs the second wave most. The architect added: "and for the underdog it is important to see the figures — to step up the campaign or not to spend money because there is no point." This is a third reason not to close the window early, and the most practical one. Today a campaign flies blind: a poll costs money an underdog does not have, arrives late and measures words rather than deeds. The counter gives him, free and daily, what large campaign staffs pay for: after a debate or a successful video the flow of declarations for the vote rose — the campaign works, and it is worth putting in one's last; the counter did not stir — the satisfied stayed satisfied, and it is better to save money and strength for the next cycle. The first wave tells him whether to run at all (4.2); the second — whether to carry on. The assistant first wrote that an incumbent has such data today as well — from closed polls and administrative channels. The architect corrected this: "polls are simply not comparable with money in accuracy." Right: a poll measures the words of a small sample, the counter measures the deeds of everyone, and paid-for deeds at that (4.3); nobody has such data today, the authorities included. So the field is levelled more than was said: the incumbent had costly and imprecise knowledge, the underdog none, and now both have the same thing, precise and free — and the authorities' informational advantage disappears entirely.

Weak point. A camp cannot be identified from the counter alone: it is an inference from the shape of the curve, not an observation. It can be tested only in a pilot — by asking those who took the sum for their reason and comparing early and late declarers. 🟡

4.3c. Two objections from the "limited good"¶

Both follow from the concept examined in 056d.5b: in archaic consciousness good is not created but redistributed. And neither is speculation: in current quantitative work on zero-sum thinking these are two measured correlates of one disposition — the stronger it is, the greater both the support for redistribution and the harshness towards outsiders (same place). For the rent part of the economy that belief is correct (same place), but it acts on the mechanism of §4.3 in two ways.

First: the rating function works more weakly than intended. §4.3 assumes the dividend percentage measures the quality of management: manage better and the share grows. But a man for whom the pie is fixed expects from a manager not growth but less theft. The mechanism is not abolished but degraded: choosing the most capable turns into choosing the least thieving.

Three consequences deserve naming. The dividend begins to reward honesty rather than competence, and these are different qualities. A manager who genuinely enlarged the base gets no credit for it: the growth is written off to prices or luck. And his own incentive shifts from "develop" to "don't take" — better than today, but less than intended.

What survives. Even "steal less" is a working criterion expressed in money, and today there is none at all. And the belief corrects itself here by the only means available: if the share genuinely grows under a capable manager, the person sees his own money, and that is the one kind of evidence that overrides a settled opinion. The objection therefore concerns the first cycle rather than the design. 🟡

Second, and sharper: the universality of the share is politically fragile. Gallyamov shows how the energy of envy converts into a demand: after the war, on his forecast, a mass demand will arise to recover compensation for Ukraine from the earnings of the war's participants — "why should we pay for their doings." The same energy applied to the dividend yields a different demand: "and those people should get no share" — emigrants, officials, the non-working, the war's participants themselves. The first politician to say it will collect votes.

For the protocol this is more dangerous than any objection about inflation. Universality is not humanism but a load-bearing part: a share made conditional turns into a benefit that someone awards, which is precisely what the protocol replaces. And the argument against exclusions must be mechanical rather than moral: any exclusion creates an office that decides whom to exclude — and that office is the one the protocol abolishes.

Hence a requirement for the referendum text (§4.7b): the unconditional nature of the share must stand in it explicitly and be harder to amend than the rest. And hence a lesson from Gallyamov's own caution: he declines to push the recovery idea now because it would create a group ready to defend the regime to the death. The same rule holds for the protocol: no group must lose more by it than it gains, or it manufactures defenders of the existing order for itself. 🟡

4.4. The effect on debates¶

The public counter creates a new kind of debate question, IMPOSSIBLE without AB-EXIT:

"80% took the dividend. Why did your citizens prefer the money to your representation?"
"The dividend in your territory is 600. In the neighbouring one, 1,100. What have you been doing for twenty years?"
"Efficiency bonus: 12. Next door: 95. Explain."
"There are 45 citizen proposals on the portal. You implemented 3. Why?"

Every question \= a specific number that cannot be answered in generalities. The incumbent is forced to account SPECIFICALLY — to an electorate in which every single person PAID for the right to ask, by giving up the dividend.

4.5. The two-phase campaign¶

AB-EXIT splits the campaign into two fundamentally different phases:

Phase 1 — "before the counter" (prior to the declaration window): campaigning for PARTICIPATION as such, not for a particular candidate. "Don't take the dividend — your vote is worth more!" This is a new kind of campaigning that does not exist in the current system.

Phase 2 — "after the first wave of the counter" (the last weeks before the election): the number of voters is KNOWN to within a few per cent (4.3b). The campaign becomes targeted: convince a specific number of motivated voters. Money loses its decisive weight — motivated voters ignore advertising and listen to arguments.

4.6. The effect on good vs bad incumbents¶

AB-EXIT does not kill incumbents. AB-EXIT kills BAD incumbents and STRENGTHENS good ones:

A good incumbent: "The dividend rose from 570 to 700. Bonus 150. I implemented 47 citizen proposals. Next door: 580. Us: 700. Only 50% took the dividend — because people WANT to vote for continuity." → RE-ELECTED with a strengthened mandate.

A bad incumbent: "The dividend fell from 700 to 630. Bonus 12. Three proposals implemented. 85% took the dividend. Five challengers have declared, each with a specific plan." → WILL LOSE. Not to scandal — to ARITHMETIC. (The chapter's thesis: the outcome is not guaranteed, and there has been no pilot.)

4.7. For the statute¶

ARTICLE XIV: TRANSPARENCY OF PARTICIPATION

Section 1: REAL-TIME PUBLICATION

The aggregate number of citizens who have

declared for Dividend (Option B) and Vote

(Option A) shall be published daily on the

jurisdiction's official AB-EXIT portal during

the declaration window.

Section 2: FORMAT

Published data shall include:

(a) Total declarations for Dividend

(b) Total declarations for Vote

(c) Total undeclared

(d) Percentage of each category

Updated every 24 hours.

Section 3: INDIVIDUAL CONFIDENTIALITY

Individual choices (A or B) shall remain

strictly confidential. Only aggregate numbers

are published. No citizen's personal choice

shall be disclosed to any party, candidate,

employer, or government official.

Section 4: RATIONALE

Public knowledge of aggregate participation

levels enables informed candidacy decisions,

reduces barriers to entry for challengers,

and provides an objective, financially-backed

measure of public trust in incumbent leadership.

4.7b. The counter belongs in the text of the referendum¶

Added on 20 September 2026. The occasion was a fork: a public counter takes away the authorities' informational advantage (4.3b), and they are left with two moves — keep it public and give the underdog the same precise data they have themselves, or classify it and thereby admit the figure is bad. The architect drew a conclusion for the design: "then the counter must be put into the referendum."

The conclusion is right, and behind it stands the Georgian lesson (048g.3b): whatever is not written into the text adopted by the people remains a slit through which a reform is dismantled by ordinary law — there it was the clause "except excise". The counter is exactly such a candidate for quiet spoiling. The payment cannot be repealed: everyone who received it holds on to it (048g.3c). But "temporarily suspending daily publication for technical reasons", publishing once a week, publishing with a delay, publishing without a breakdown by polling station — all that is possible if Article XIV exists only as a by-law; and each of these trifles hands back to the authorities exactly the advantage the counter took away.

So the text put to the referendum includes not only the sum and the right to choose but four properties of the counter: the aggregate is published for every polling station; daily throughout the declaration window; in open machine-readable form; with no right of suspension. The counter then falls under the same asymmetric lock as the payment (048g.6b): it can be changed only by a new referendum, and holding a referendum on hiding a figure from the people is a move that defeats itself.

4.8. The two payment phases¶

Amendment of 30.09.2026 (048k §3). The second phase — the efficiency bonus — is removed: the protocol has one payment. The first is made precise: whoever chose the money receives it at once and enters the counter at once; "a day before the election" is no longer the deadline. The text below is kept as the earlier wording.

Phase 1: base dividend (D_base) \= M × 1.5 × K. Paid one day before the election from a fund formed in advance.

Phase 2: efficiency bonus (B) \= max(0, Budget_plan − Budget_actual) × 30% / N, where N is the number of dividend recipients. Paid six months after the election, following an independent audit.

Total dividend per cycle: D_total \= D_base + B.

The bonus is a share of what was saved, returned to those who chose to exit. It ties a citizen's personal income to the budgetary discipline of the government rather than to its promises.

4.9. A premium for early declaration¶

Added on 20 September 2026. The architect's proposal: "if the public counter matters so much and we impose a penalty on late payments (048i.5b), then perhaps make a premium for early ones." The assistant considers the idea right and sees in it the completion of one design: the sum stops being a point and becomes a scale over time.

When the person chooses the sum	What he receives	Why the protocol wants it
In the first days of the declaration window	the sum with a small premium, paid at once	the counter fills early, when it is needed most
On the remaining days of the window	the ordinary sum	—
After the election, if he did not vote	the sum with a 20–30 % discount	a witness for every "empty" record (048i.5b)

Why early figures are worth more than late ones. The counter is a "calibrator of courage" for new candidates (4.2), but seeing the figure is not enough for a candidate: he must have time to register, gather a team and run a campaign. A figure that appears three days before the election brings nobody into the race. Early declarations are also needed by the second gauge — the number of refusals before polling day, by which vote buying is found (019d.4b).

What exactly the premium buys. A person who puts off declaring holds something of value — the right to watch the campaign through and change his mind. The premium is the price at which he sells that right. Someone who decided long ago and is merely procrastinating has nothing to sell and gets the premium for free — such people are the majority, and it is precisely their early figures the counter needs. Someone to whom the campaign really matters keeps the right and loses nothing against the ordinary sum. So the premium does not penalise the thoughtful; it pays for certainty.

Two conditions. First, the premium must be small, of the order of 5–10 %: its task is to beat procrastination, not to outbid a decision; a large premium would pull out of the election those whom the campaign might have persuaded to stay, and would break the first phase of the campaign (4.5), which is precisely where people are urged to take part. Second, the budget identity (048f.3) must balance: the natural way to fund the premium is from the late discount, and then the scale as a whole costs the budget nothing. The cheapest part of the premium is free altogether: paying the early declarer at once rather than the day before the election (4.8); for someone living from pay cheque to pay cheque, money today is worth noticeably more than the same money a month later.

Weak point. An early declaration with immediate payment is irreversible: a scandal in mid-campaign will not bring such a person back to the polls. It is a deliberate exchange, but in the first cycle it may prove a bad one for those who underestimated how the campaign would change; the size of the premium and the share of those who regret it are parameters for the pilot. 🟡

Calculations for specific jurisdictions are in Part IV. American municipal examples (including the efficiency bonus calculation) are in chapter 10.