A deliberately practical, simple, technically-ready scenario — the opposite pole from the Saudi-legal run. It is also the worked application of held-value-not-reachable: the private car is the textbook frozen asset, and this tests whether the canon survives contact with a real configuration.
The scenario
Self-driving cars operated as on-demand services (robotaxi fleets, Uber-without-the-driver), displacing private car ownership. The CE claim: a private car is used ~10 hours a week and sits idle for the other ~158 — and when idle it is not merely dormant value, it is a net hindrance: parked on the street it removes road capacity and worsens congestion for everyone. A shared autonomous fleet could deliver the same journeys from far fewer vehicles, thawing the frozen material value (metal, manufacture, embodied energy) and returning the road and kerb space to the commons.
Why it could be strongly positive-sum (the CE core)
- It thaws
held-value-not-reachabledirectly: capital frozen in idle private cars becomes reachable mobility delivered to many from few. The same journeys, a fraction of the frozen metal. - It returns a second frozen asset most people forget: kerb and road space. Parked cars are stored private value occupying shared space; freeing it widens the reachable set for everyone (cycling, walking, transit, housing, greenery).
- Access widens for the carless — disabled, elderly, young, poor — who are excluded by the ownership model.
- It is technically ready or nearly so (operating robotaxi services already exist), so this is not a far-horizon bet.
The tensions that must be tested (so this is not a cheerlead)
- Who owns the fleet is the whole game (
route-around-the-chokepoint,cleanliness-and-scale). If one platform owns it (a Waymo/Uber/Tesla monopoly), the frozen value is not thawed to the commons — it is re-enclosed as platform rent, and the kerb-space dividend accrues to a private toll-keeper. Municipal / cooperative / open-network ownership vs platform enclosure is the fork. - Induced demand and empty miles (the throughput trap). Cheaper, easier, door-to-door mobility can increase total vehicle-miles — more trips, plus empty repositioning miles between fares. Does it actually reduce congestion and throughput, or just change who is stuck and add zombie miles? Throughput is a cost, not an achievement.
- Less to none. Displaced drivers (taxi, private-hire, eventually freight — millions of livelihoods); the equity of access (served or priced out); and the risk of cannibalising public transport — buses and trains move people far more space-efficiently than even shared cars, so a fleet that strips transit ridership could be net-negative for option-space.
- The rebound on the original asset. If the fleet makes car-travel so frictionless that total car-travel rises, the kerb-space and congestion dividend can evaporate even as private ownership falls.
The sharp questions
- Does an autonomous shared fleet genuinely thaw the frozen value to greater-life-to-all, or does it re-enclose it as platform rent and induce enough new vehicle-travel to forfeit the congestion/kerb dividend?
- Who must own it for the thaw to reach the commons rather than a toll-keeper — and is that ownership configuration reachable?
- What makes it complementary to public transport rather than a parasite on it, and how is the just transition for displaced drivers handled, so it passes less-to-none?
Confidence note
Operating robotaxi services, private-car utilisation (~4–5% of the time in use), parking's share of urban land, and induced-demand evidence from ride-hailing are largely verifiable — for White to anchor. Fleet-size-replacement ratios and congestion outcomes are model-dependent and contested — flag accordingly.