The second commodity — mobility precipitates on the energy cell's rails

Published live (2026-07-02). The second definite treatment, and the diligence upgrade the mobility work was owed: the original four-pass search run was computed entirely without ground truth, so this run put its conclusions through the same machine as the ratchet cell — evidence fleet (one researcher of four was lost mid-run to a tooling failure; the liability leg is thinner for it, flagged below), four expansion lenses, walls classified, and three independent adversaries. Verdict: passes-with-conditions, three from three — but this time the gate discriminated inside the design: it passed the ladder and failed the headline. Both results are printed here.

What the evidence did to the prior run

The earlier mobility work concluded that the end-state is a member-owned mutual pool, and that the residual wall — the enclosure race — had a single lever: permit-mandated interoperability. Tested against the world:

• Contradicted: the "only move" claim. No jurisdiction anywhere has adopted an open-dispatch permit condition — it exists only as a 2026 law-review proposal. Meanwhile the race itself is real and reshaped: the enclosure play is not a single AV maker but Uber-as-cross-operator-aggregator (~$10B committed across ~30 autonomous partners), assembling the supply side. Which points at the move the prior run missed: the demand side is still unassembled, and aggregating it needs no politician.
• Still unverified: the liability-forces-a-capitalised-entity claim (the lost researcher's leg), the bond-at-point-of-use mechanism, and the mutual-with-demurrage end-state — each carried forward as open, not silently assumed.
• Confirmed and sharpened: the fleets are strikingly thin (Waymo ~2,500 vehicles across ten metros; Tesla's Austin service ~20 cars for a whole metro area), UK Automated Passenger Service pilots became legal this spring, and cost-parity forecasts cluster around 2030 — so the window is roughly three to four years, not a horizon.

One negative finding worth its own line: the researchers found no waitlist signature anywhere in shared mobility — UK car clubs (~339k members) show no movement toward demand-side mutualisation. Where energy had a queue, mobility has silence. The find-the-waitlist lens returned empty-handed, which is the lens working, not failing: it says this domain's bridge, if it exists, has not yet been half-built by anyone.

The precipitation finding, stated plainly

The design brief asked whether the mobility mutual could precipitate — arrive late, as paperwork catching up with trades — rather than be founded. The answer the evidence forced: pure precipitation is blocked in standalone mobility. Two structural facts, high confidence: the Section 19/22 community-transport fare layer legally cannot hold a reserve (cost-recovery cap, no distributable surplus), and the P&I-club template is in reality a fully regulated mutual insurer, not a permit-exempt body. A standalone mobility mutual cannot crystallise out of fares; somewhere it must quasi-found.

But precipitation works on a substrate: where an energy cell already exists — membership roll, half-hourly billing rail, paid originator, a Community Benefit Society already registered — mobility enters as a second metered commodity on the same rails, and the institution's arrival really is a constitutional amendment forced by the accountant, not a founding act. This refines institutions-precipitate rather than confirming it: precipitation requires rails. The first commodity pays the founding cost once; every later commodity inherits it. The cells compound — which quietly upgrades the ratchet cell from an instance to the substrate.

The ladder — the second commodity

Eight rungs, each individually rational on day-one terms; the two-books move (passengers at cost-recovery, reserve fed by non-fare streams) dissolves the legal wall the prior run never saw:

1. Cheapest miles — an existing club member's EV charges in the hydro match windows; fuel spend collapses; mobility becomes a line on the existing bill, and charging telemetry makes true per-mile cost legible for free.
2. Seats on the bill — repeat pairings only, on the valley's one dominant corridor, settled as bill credits. The only UK lift-sharing with adoption evidence is repeat-pairing and institutionally anchored (>90% of Liftshare's activity); the open-marketplace variant is a documented failure. The demand book begins as a by-product.
3. Goods before people — the shop's and pharmacy's runs ride trips already being driven. Goods carriage engages neither the stranger-in-my-car stigma nor the fare cap — so this is the reserve's legal home: cost-recovery-plus margin, lawfully accumulated.
4. The disposal intercept — households already shedding a car (a decision ~36% of UK motorists made inside two years, single-survey, flagged) are offered a strictly-dominant exit: block-rate credit with the existing licensed taxi firm, an e-bike loan (the best-evidenced low rung in the pack: 51% car-mileage reduction), optional named-neighbour hire on rented peer-to-peer rails. The buying club begins; the club never operates a vehicle and never accepts a booking.
5. The standby retainer — tail certainty as a purchasable product: an annual fee, priced far below the shed car's cost, buying response-time SLAs from the licensed operator; refunds if the guarantee can't be kept. This rung pays the one physical floor openly — idle slack against the un-forecastable peak, bought once per catchment instead of duplicated per driveway. It is also, unanimously, the weakest load-bearing rung — see the gate.
6. Anti-phase vehicle-hours — the school minibus and weekend-idle vans, pooled under one keyholder-operator with paid MiDAS drivers; the volunteer-collapse answered with wages. Scoped bridge-only, for the hard trips the cheap rungs honestly don't serve.
7. The wrapper precipitates — the accumulated reserve, escrow and contracts cross what an unincorporated arrangement can hold, and incorporation (or amendment of the energy cell's existing constitution) becomes the cheapest compliance move. Nobody is asked to want a mutual; the paperwork catches up with the trades.
8. AV arrival: re-tender the book — the cell meets autonomous supply as an incumbent aggregated buyer holding a contracted demand book, years of telemetry, the charging depot on cheap hydro, and the availability-retainer contract form — which is precisely the guaranteed-utilisation instrument a thin rural fleet wants.

The gate — passed the ladder, failed the headline

Three adversaries, three passes-with-conditions — with the same three findings surfacing independently:

1. Rung 8 does not pass on its own. A single valley's demand book is trivial bargaining weight against a ~$10B supply-side aggregator with direct consumer reach; real leverage would need a cross-valley federation, which is an organizing act the bridge test forbids — and of which there is zero early trend. The template survives only because it pre-demotes the AV endgame to a free option: every rung from 1 to 6 must pay standalone, and does. The honest sale is "individually-rational rural mobility riding energy's rails," not "the valley out-negotiates Waymo."
2. The verdict is narrowed to the substrate. On an existing energy cell (there are ~100 Energy Local-style clubs), no founding act is needed and the route-around holds. Off-substrate, rung 7 becomes a marketed share offer — a quasi-founding — and the template honestly degrades. The catchment, per the bridge-catchment discipline: this passes as a cost-saving buying club for the median member, not as median rural car-replacement; the car-shedding rungs are the least-evidenced and reliability-limited.
3. Rung 5 is the crux, three ways at once. It carries the true reason second cars survive (2am tail certainty); it is a precedent-free proposal gated on one decisive legal unknown (whether a discretionary standby pool sits below the FCA perimeter); and where rural licensed capacity is genuinely thin — the honest rural case — it may be physically unfillable at any price, in which case it refunds out gracefully and car-shedding stalls at the households who don't fear the emergency.

Conditions before any rung ships as "ready": verify the billing rail can carry a non-energy line item; confirm the lift-share insurance/tax position against primary sources; de-risk the corridor geometry with a local survey before rung 2; resolve the FCA question on rung 5; and prove the reserve's revenue streams actually earn — the legal home is solved, the income is not.

The walls ledger

What this establishes

The gate now has three runs and has behaved differently each time — narrowed a claim (ratchet cell), and here split a design down the middle, passing the ladder while failing its most quotable thesis. A gate that can partially fail a design it likes is doing real discrimination. The candidates moved too: institutions-precipitate gains its condition (precipitation requires rails — first commodities pay the founding cost, later ones inherit); the-closing-window gains a measured width (~3–4 years) and a new strategy class (concede the race, build below it); find-the-waitlist returned its first honest empty result. And one structural insight now spans both treatments: the cell is the unit that compounds, not the commodity. Energy was not the first treatment; it was the first rail.

Not yet promoted into propositions.ts. Forming register. The liability leg of the evidence fleet was lost to a tooling failure and its claims stay open; the AV-arrival thesis is demoted to a free option by the gate's own reasoning; and the Ogwen instance inherits the same first-week homework as the energy treatment, plus one more: a corridor survey of the A5 funnel.