Rules, Institutions, and Economic Instruments

6.1From operating picture to operating authority

The sensing layer described in Chapter 5 gives a city visibility. This chapter describes how visibility becomes authority – the ability not merely to observe what is happening but to set the conditions under which it happens, and to enforce those conditions with credibility. A city that has built a sophisticated operating picture but lacks the legal authority to act on what it sees, the institutional model to coordinate responses across agencies, or the economic instruments to create the incentives it needs, has invested in knowledge but not in governance.

Institutional reforms create the authority a city needs to govern CAMCAMConnected and Autonomous Mobility — the commercial deployment of autonomous vehicles operating connected to data and control networks. The paper uses CAM rather than "AV" to foreground the network and governance dimensions that determine how the technology lands in cities. (Ch.1 §1.1; Ch.2 §2.1.) operations effectively. Economic instruments deploy that authority through pricing signals, quantity controls, and fiscal mechanisms that shape operator and vehicle behavior in ways consistent with public goals. Neither is sufficient alone: authority without instruments is capacity that cannot be used; instruments without authority are unenforceable.

6.2Institutional reforms: the operating authority uplift

Authority clarity. The most basic institutional requirement for CAM governance is also the one most often missing: a clear and legally defensible answer to the question of who has the power to do what. In most cities, the authority landscape for CAM-relevant decisions is fragmented across multiple agencies and tiers of government in ways that make coherent governance almost impossible. Curb management is typically a highway authority function. Fleet licensing is a transport or vehicle licensing function. Land use and depot siting is a planning authority function. Grid coordination sits with energy regulators. Data rights are a combination of licensing conditions and data protection law. No single agency owns the whole picture, and the default when two agencies’ remits overlap is either conflict or paralysis. Establishing authority clarity – which means mapping the existing remit of each relevant agency, identifying the gaps and overlaps, and either filling gaps through legislative means or creating a coordinating structure with clear decision rights – is the foundational reform. CAM will not wait for a governance review to complete.

Licensing and service obligations. The license is where the Governability StackGovernability StackThe four-layer model the paper uses to organise what cities must govern: Flows (movement), Space (kerb, road, depot), Power (grid and energy), Data (sensing and authority). Each layer has its own institutional surface and timescales. (Ch.3.) becomes a contract. Transport for LondonTfL (Transport for London)London’s integrated transport authority, licensing over 100,000 private hire vehicle drivers alongside running the network. Its private hire licensing regime is the closest operational analogue the paper finds to what CAM governance will require. (Ch.2.)’s private hire vehicle licensing regime provides the closest operational precedent: operators must meet fit-and-proper tests, submit structured trip data, comply with zero-emission-capable vehicle requirements, and pass annual safety checks.^EV33EV33Licensing Regime Analogues for CAM GovernanceAviation, maritime, rail licensing frameworks cited as CAM governance analogues; tiered certification, operator liability, incident reporting as…Open in the Evidence Base → Non-compliance results in license suspension or revocation. TfL has used this framework to exercise authority over Uber – one of the world’s largest and best-capitalized technology companies – including two license non-renewals, each eventually resolved on appeal, and a renegotiated operating agreement with strengthened data-sharing and safety conditions. The framework works because the licensing conditions are specific, measurable, and backed by enforcement powers that the operator cannot simply ignore. New York’s TLCTLC (Taxi and Limousine Commission)New York City’s regulator for taxis and for-hire vehicles, licensing roughly 200,000 drivers. Cited alongside Transport for London as one of the few bodies with an operational template for conditioning a large private fleet. (Ch.4.) (Chapter 2) is the complementary model: accessibility mandates on the existing fleet, and a hard cap on its size^EV33EV33Licensing Regime Analogues for CAM GovernanceAviation, maritime, rail licensing frameworks cited as CAM governance analogues; tiered certification, operator liability, incident reporting as…Open in the Evidence Base →.

For CAM, the minimum viable licensing model should specify: fleet size limits with review triggers, geographic coverage obligations (including equity coverage in lower-income areas), mandatory data-sharing conditions with defined format, frequency, and API specification, accessibility standards, safety reporting protocols, workforce transition contributions, and fee-based cost recovery for oversight. TfL’s micromobility concession framework – which includes all these elements plus independent data verification and a three-part fee structure^EV33EV33Licensing Regime Analogues for CAM GovernanceAviation, maritime, rail licensing frameworks cited as CAM governance analogues; tiered certification, operator liability, incident reporting as…Open in the Evidence Base → – provides the closest existing analogue to what a CAM licensing framework needs to look like. The template is fully specified; it needs to be adapted rather than invented.

The licensing framework is also the instrument through which cities can invite operators to co-develop governance conditions. When Amsterdam faced the rapid growth of short-term rental platforms disrupting housing availability, the city did not simply ban the practice or wait for national legislation. It negotiated directly with Airbnb to co-design a registration and data-sharing framework – resulting in Airbnb itself collecting and remitting tourist tax, enforcing night limits through the platform, and providing city authorities with aggregate occupancy data for planning purposes. The process was not frictionless, but co-development produced a regime both sides could work with. Register-backed verification made it enforceable.

For CAM, the equivalent approach is to bring the major operators into the licensing design process early to give them the opportunity to demonstrate that they can operate well under formal governance, and to build the data-sharing approach jointly rather than adversarially.

Data rights and audit authority. The information asymmetry identified in Chapter 3 – operators knowing more about how the city works than the city does – is primarily a legal and institutional failure rather than a technical one. The technical capability to collect and transmit fleet operational data exists. The legal obligation to do so does not, in most jurisdictions. Making data sharing a mandatory condition of licensing – not an optional partnership or a discretionary concession – is the instrument. MDSMDS (Mobility Data Specification)An open standard for how mobility operators report trip and vehicle data to cities, originating with Los Angeles and now widely adopted for micromobility. The paper treats MDS 2.1 as mature enough that no city need invent its own schema. (Ch.4.) 2.1 provides the technical standard; licensing provides the legal mechanism.^EV3280EV3280MDS: Open-Source Digital Infrastructure for Right-of-Way Governance at Global ScaleMDS open standard, LADOT-originated, OMF-stewarded Nov 2019; 6 APIs, v2.0.2 June 2025; 200+ cities, 21 countries, 50+ MSPs — regulatory relationships…Open in the Evidence Base → The combination closes the information asymmetry that underpins the value at risk across all five levers.

Compliance and enforcement modernization. Static compliance instruments – annual license renewals, periodic inspections, complaint-driven enforcement – are inadequate for governing a technology whose behavior varies by hour, corridor, and day-of-week. Real-time compliance monitoring, automated anomaly detection, and graduated enforcement powers that can respond proportionately to specific failures are what the CAM context requires.^EV113EV113California AB 1777 — DMV "Notice of Autonomous Vehicle Non-Compliance" Enforcement Regime Adopted, Live 1 July 2026DMV rules (adopted 29 Apr, effective 1 Jul) let police ticket the manufacturer directly, with 72h/24h reporting duties and permit suspension as the…Open in the Evidence Base → London’s Grid Smarter Cities curbside management pilot demonstrates the direction: virtual loading bays activated by demand, with e-ink signage updated in real time and automated booking and charging replacing the physical parking meter and the periodic parking inspector.^EV32EV32Dynamic Curb Pricing and Allocation FrameworksReal-time curb pricing models balance loading, AV drop-off, micromobility; revenue recycling and equity access remain core governance leversOpen in the Evidence Base → The same structure – dynamic authorization backed by automated monitoring – is the model for CAM compliance.

Inter-agency and utility coordination. CAM governance inherently spans domains that are currently managed by different agencies with different mandates, different political accountabilities, and different data systems. The depot siting decision requires inputs from transport (routing implications), planning (land use compatibility), energy (grid hosting capacityHosting capacityHow much additional load or generation a given point on the electricity network can accept without reinforcement. Publishing it is what turns depot siting from a series of private negotiations into a planning question a city can participate in. (Ch.2.)), and public health (emissions and noise). No current governance framework coordinates all four. A cross-agency governance wrapper – with explicit decision rights, formal data-sharing protocols, and joint planning processes covering at minimum the transport, energy, and planning domains – is the institutional framework this requires. The UK’s Local Transport Plan framework provides a legal basis for multi-agency transport planning; the question is whether it can be extended to cover energy infrastructure in a way that gives it teeth.

Procurement for uncertainty. Cities that are already contracting for transport services – bus operations, dial-a-ride, curbside management, traffic signal operation – can embed CAM governance requirements into those contracts in ways that create durable obligations without requiring primary legislation. The key design principles are: shorter contract terms with performance-triggered extension options, technology-neutral specification that describes required outcomes rather than specified means, break clauses tied to operating picture metrics, and data-sharing obligations that cannot be waived without city consent. The alternative – a ten-year contract signed in 2026 without these provisions – is a decade-long commitment to a governance baseline that will be demonstrably inadequate by 2028.

6.3The economic toolbox

Institutional authority establishes what a city can do. Economic instruments specify what it is worth doing – the set of price signals, quantity controls, and fiscal mechanisms that shape the behavior of operators, vehicles, and the market in ways that serve public goals. Every instrument below has a precedent in a comparable context.

Price instruments. Dynamic curb pricing is the most immediately deployable instrument in the toolbox, with the lowest political risk and the clearest operational precedent. New York’s Smart Curbs pilot uses sensor-based occupancy data to price curb access in real time, showing that the design for deployment at scale exists. Washington DC’s curbFlow reservation system has processed over 15,000 commercial vehicle reservations, demonstrating operator willingness to pay for managed curb access when the alternative is congestion and enforcement uncertainty.^EV32EV32Dynamic Curb Pricing and Allocation FrameworksReal-time curb pricing models balance loading, AV drop-off, micromobility; revenue recycling and equity access remain core governance leversOpen in the Evidence Base → Beyond the curb, congestion pricing is proven across six programs on three continents^EV35EV35Congestion Pricing Outcomes: Evidence from Operational SchemesStockholm, London, Singapore schemes cut traffic 10-22%, raised revenue, improved air quality; pricing level and exemptions shape equity outcomesOpen in the Evidence Base →, with consistent evidence of sustained traffic reduction and public acceptance. Stockholm achieved 20 to 25 percent traffic volume reduction that has persisted for nineteen years. London’s congestion charge generates $290 million (£229.8 million) annually in revenue, about two-thirds of it net of operating costs, while maintaining central London’s economic performance.^EV35EV35Congestion Pricing Outcomes: Evidence from Operational SchemesStockholm, London, Singapore schemes cut traffic 10-22%, raised revenue, improved air quality; pricing level and exemptions shape equity outcomesOpen in the Evidence Base → New York’s central business districtCBD (Central Business District)A city’s core commercial area. In the paper it usually refers to New York’s CBD tolling programme, launched January 2025 — the first US congestion charge and the most recent large-scale evidence on pricing. (Ch.6.) (CBD) tolling program, launched in January 2025^EV35EV35Congestion Pricing Outcomes: Evidence from Operational SchemesStockholm, London, Singapore schemes cut traffic 10-22%, raised revenue, improved air quality; pricing level and exemptions shape equity outcomesOpen in the Evidence Base →, reduced daily vehicle entries by 11 percent^EV35EV35Congestion Pricing Outcomes: Evidence from Operational SchemesStockholm, London, Singapore schemes cut traffic 10-22%, raised revenue, improved air quality; pricing level and exemptions shape equity outcomesOpen in the Evidence Base → in its first year and generated $548 million in net revenue after operating costs^EV35EV35Congestion Pricing Outcomes: Evidence from Operational SchemesStockholm, London, Singapore schemes cut traffic 10-22%, raised revenue, improved air quality; pricing level and exemptions shape equity outcomesOpen in the Evidence Base →. Milan’s Area C program has achieved 30 to 38 percent traffic reduction sustained over a decade^EV35EV35Congestion Pricing Outcomes: Evidence from Operational SchemesStockholm, London, Singapore schemes cut traffic 10-22%, raised revenue, improved air quality; pricing level and exemptions shape equity outcomesOpen in the Evidence Base →. The evidence and the playbook are now strong enough that the remaining barrier to congestion pricing is political.

Quantity instruments. Fleet size caps, permit systems, and reservation systems complement price instruments by controlling the total volume of fleet activity directly, without touching price. The advantage of quantity instruments is predictability: a cap on total licensed vehicles provides a hard ceiling on the allocation and demand failure modes that price instruments address more softly. The disadvantage is rigidity: a cap set at the wrong level, or without adequate review mechanisms, becomes a constraint on beneficial deployment rather than a ceiling on harmful excess. New York’s 2018 for-hire vehicle (FHV) pause^EV33EV33Licensing Regime Analogues for CAM GovernanceAviation, maritime, rail licensing frameworks cited as CAM governance analogues; tiered certification, operator liability, incident reporting as…Open in the Evidence Base → worked as an emergency brake but proved difficult to calibrate as a long-term management tool. Fleet caps work best as backstop instruments, combined with occupancy-linked licensing conditions that create dynamic flexibility within the overall ceiling.

Market design. Where price signals and quantity controls are most effective is in structuring competition among operators for access to constrained public resources – curb slots, service territory, charging infrastructure. Competitive concession processes for geographic service areas, curb slot auctions for high-demand locations, and tradable access rights for fleet operators are market design instruments that introduce price discovery and prevent the rent-seeking that occurs when public resources are allocated through administrative discretion. The design of these mechanisms requires institutional sophistication, but the underlying economics are well understood from analogous domains – spectrum auctions, port concession agreements, and airport slot allocation.

One mechanism that translates well from analogous contexts is the competitive permit token model. In micromobility, several cities have moved beyond simple license caps to allocate operating permits through a scored competitive process – where operators bid for a fixed number of deployment tokens, with scoring weighted across price, service quality commitments, equity coverage obligations, and data-sharing standards. Portland and Louisville have used variants of this approach for shared scooter allocation since 2019. The mechanism works because it creates competition on dimensions of public value – not just who will pay the most, but who will serve the most, report the most accurately, and extend coverage furthest into under-served areas. Applied to autonomous fleet licensing, the token model could allocate service territory permits across multiple operators with differentiated commitments – creating a market structure in which public value is priced in at the point of market entry, rather than negotiated retrospectively after deployment has established path dependencies that reduce the city’s leverage. First-round allocation can only score commitments, since it precedes the operating picture. It stays disciplined through short initial terms with renewal contingent on delivered performance. The competitive process itself is the anti-capture instrument: scored allocation is harder to bend than bilateral discretion.

Fiscal stabilizers and operator-fee model. The fiscal transition challenge – replacing declining parking and enforcement revenue before the decline creates a budget crisis – requires proactive revenue design. The first instinct of most cities, drawn from a generation of transit-corridor financing precedent, is to reach for the value capture toolkit: tax increment financing (TIF), special assessment districts, business improvement district (BID) contributions, Mayoral CIL (Community Infrastructure LevyCIL (Community Infrastructure Levy)A charge on new development in England and Wales, applied by local authorities and the Mayor of London, used to fund supporting infrastructure. The UK counterpart to the US value-capture instruments in the same menu. (Ch.6.)), enterprise zone retention. These mechanisms work when the public investment creates a measurable, attributable, geographically bounded uplift in property value – corridor scenarios, station catchment areas, mixed-use district redevelopment. CAM, in its dominant deployment form, has none of these properties. A roving private autonomous fleet is not corridor-bound; its benefits are dispersed across a service area rather than clustered around stations; and the relationship between fleet activity and property value is too indirect to support the financial-instrument design that value capture requires. Cities applying corridor-style value-capture instruments to roving CAM fleets are likely to find the revenue base too diffuse, the attribution arguments too contested, and the political case too weak to defend against the operator counterargument that the city is taxing benefits the city did not produce.

The more durable structure for the dominant deployment case is transaction- and deployment-based fees. The worked example is Portland’s micromobility scooter framework, designed by the Portland Bureau of TransportationPBOT (Portland Bureau of Transportation)The City of Portland’s transport agency, cited as an early mover on curb regulation and data-sharing conditions. (Ch.6.) (PBOT) and operating since 2019. Each licensed operator pays an annual deployment fee per vehicle authorized to operate in the city, plus a per-trip fee that scales with utilization. The framework has three properties that value capture lacks for CAM. First, it tracks the activity that generates the externalities the city is governing. Second, it is operator-agnostic, corridor-agnostic, and asset-class-agnostic, so the same instrument functions identically for a roving robotaxi fleet, a depot-based logistics service, or a hybrid passenger-and-freight operation. Third, it scales linearly with the externalities: when fleet activity doubles, the fee revenue doubles, providing automatic fiscal stabilization as deployment grows. A municipal fiscal model combining a deployment fee, a per-trip fee, and (where applicable) a curb-event fee replicates for CAM the structure that has stabilized micromobility revenue in Portland, San Francisco, Washington DC, and Austin.

Value capture retains a role, but a narrower and more disciplined one. Where CAM deployment is corridor-bound, the conditions for value capture are satisfied and the instrument applies. This could happen in, for example, dedicated depot precincts, AVAV (Autonomous Vehicle)A vehicle capable of driving itself without a human operator. The paper mostly avoids "AV" in favour of CAM, because "autonomous vehicle" frames the subject as a machine when the governable object is a networked commercial service. (Ch.2.)-transit interchange hubs, district-scale conversion-ready parking redevelopment, mixed-use district reactivation projects. The recommendation for cities and their fiscal advisors is to lead with transaction and deployment-based fees as the primary fiscal model for CAM operations, and to deploy value-capture instruments in the specific spatial contexts where their underlying logic holds, never as the default. The fiscal design question is which spatial pattern of deployment the city expects to govern. The instrument follows from the pattern.

6.4Sequencing: what comes first

No city can deploy every instrument simultaneously, and no city should try. The sequencing principle that follows from Chapter 1’s path dependencyPath dependencyThe property that early decisions constrain later ones, so that what is cheap to choose now becomes expensive or impossible to change afterwards. The paper’s central mechanism: the terms of the first operating licences become the baseline for every subsequent negotiation. (Ch.1.) logic is: start with decisions that preserve the most options, build institutional capacity before locking in economic frameworks, and use the operating picture to calibrate instruments progressively as evidence of real behavior accumulates.

The minimum viable institutional toolkit – authority clarity, licensing conditions with data-sharing requirements, and a cross-agency coordination wrapper – should be in place before any fleet licensing decision is made. These are the preconditions without which everything else is either ineffective or unenforceable. Price instruments should begin with curb pricing, which carries the lowest political risk, generates immediate revenue, and builds the institutional capacity and public acceptance that makes the subsequent introduction of congestion pricing and VMT charges easier. Quantity instruments – fleet caps, occupancy requirements – should be designed as part of the initial licensing framework so that they can be activated without requiring legislative amendment when operational evidence suggests they are needed.

Table 6.1 is a vocabulary, not a sequence. The right combination and timing for a given city depend on the city’s Institutional ODDInstitutional ODDThe governance counterpart to the technical ODD: the conditions a city has to put in place — in rules, instruments, sensing, and authority — for CAM to operate inside its boundaries on terms the city can live with. Answers where the city can govern. (Ch.2, Box 2.1.) score, its revenue profile, its political environment, and the specific deployment scenarios it faces. What the menu provides is the set of tools from which that combination is drawn, each with known properties, known precedents, and known political risk profiles. Curb management, depot consent and planning conditions sit under powers most cities already hold; congestion and VMT pricing require state or national enabling; and in pre-emption states several rows of the table are unavailable at any price. Read the menu with Chapter 2’s authority map beside it – the tier that holds the power determines which instruments are yours to deploy.