Public Transport Operating Models Explained: From Direct Provision to Private Contract

Introduction

When I teach contract management for the UITP Academy, there is one slide that stops the room every time which comes from an UITP report. Five boxes, arranged left to right, from a transport authority running everything itself to a private operator answering to nobody but the market. Every participant, wherever they are from, points at a box and says: that’s us. And then, often we’re moving.

That slide is the subject of this guide. Whether you are a transport authority deciding how to structure a new procurement, or an operator trying to understand the market you are bidding into, the operating model question sits at the centre of everything. This guide explains the five principal public transport operating models, how gross cost and net cost contracts differ, and what the choice of model means in practice for authorities and operators.

The operating model decision is often treated as settled before the real implications are understood. Authorities choose a model for political or historical reasons without fully working through the risk allocation consequences. Operators bid into contracts without understanding whether the authority or the operator carries revenue risk—which directly affects how the commercial proposition should be structured. These are not abstract questions. Getting the model wrong, or misunderstanding the model you are bidding into, creates financial exposure that persists for the entire contract term.

Surbon Consulting has supported transport operators and authorities across bus, rail, tram, ferry and micromobility contracts in the UK and internationally, including advisory work on franchising structures, contract mobilisation and market engagement. I am a lead trainer on UITP’s contract management diploma programme, covering operating models, contract structures and performance regimes with transport professionals from across the world.

One thing that rarely gets discussed openly: the operating model shapes the mobilisation challenge as much as it shapes the commercial bid. A gross cost contract gives the authority control but demands more active contract management. A net cost contract transfers revenue risk to the operator but changes the incentive structure entirely and can be financially catastrophic in a global pandemic. Understanding those dynamics before the contract is signed is where the best advisory work happens.

Public transport operating models: what are the choices?

Let’s start by clearing up a word. In public transport, ‘outsourcing’ is not about handing your logistics to a large logistics company. It is the decision by a transport authority not to directly operate services itself, and instead to have them delivered by an operator under varying degrees of contractual oversight and risk allocation. This is a structural governance question, not a procurement efficiency question.

Every transport system sits somewhere on a spectrum. At one end, fully integrated public operation: the authority plans the network, sets the fares, employs the drivers and takes the fiscal consequences. At the other, fully deregulated private operation: operators run whatever the market will bear, and the authority’s role shrinks to safety and licensing. Where a system sits on that spectrum determines who plans routes, who sets fares, who takes revenue risk and who owns the assets.

The distinction between direct operation and contracted services is the fundamental one, and the dividing line within contracted services—whether the authority relies on administrative regulation or contractual regulation—is what gives us the five-model framework that UITP uses, and that I use in training. It is the most practical way I have found to explain a public transport service delivery model to a mixed room of authorities and operators, and it structures the rest of this guide.

The five public transport operating models

The UITP spectrum runs from full public control (Model 1) through to unregulated private operation (Model 5). The key dividing line is whether the authority uses administrative regulation or contractual regulation to shape service delivery. Most modern systems of interest to our clients sit in the Models 3–4 range: contracted services delivered by either a public or private operator.

Model	Operator type	Oversight type	Revenue held by	Typical examples
1: Direct Provision	Transport authority itself	None—authority is both planner and operator	Authority	Some municipal bus operations in Middle East
2: Public Sector (Administrative)	Public sector agency	Administrative oversight	Authority	Arm’s length public body, no formal contract
3: Public Sector (Contract)	Public sector agency	Contractual oversight	Authority or operator	Transport for Wales (rail)
4: Private Operator (Contract)	Private sector operator	Contractual oversight	Authority (gross cost) or operator (net cost)	London buses, Singapore MRT, Sydney Metro
5: Private Operator (No Contract)	Private sector operator	General regulation only	Operator	Flixbus, UK open access rail (Lumo, Grand Central)

Table 1: The five public transport operating models

One thing the table cannot show: nothing here is permanent. Requirements change over time, and systems move along this spectrum in response to political priorities, fiscal pressure and service quality concerns. The UK bus sector is the clearest example, oscillating between deregulation after 1985 outside London and re-regulation through the Bus Services Act 2017 and Bus Services Act 2025.  That history matters, because it tells you the operating model decision is never made once—it is revisited every time the politics or the money changes.

Model 1: Direct provision

Under direct provision, the transport authority is both the planning body and the operator. There is no separation between commissioner and provider: the same organisation designs the network, sets the fares, runs the depots and answers for the results.

The advantages are real. Full control over service design, fares and quality; no procurement overhead; and any surplus stays within the public sector. The disadvantages are equally real. There is no competitive pressure on cost or innovation, poor performance carries no contractual consequence, and fiscal risk sits entirely with the authority. When something goes wrong there is no contract to enforce—only an internal conversation.

This model is now rare in developed markets, though it survives in some municipal transport operations. In my experience it tends to persist where a system pre-dates the contracting era and nobody has had a compelling reason—or the political appetite—to change it.

Model 2: Public sector (administrative oversight)

In Model 2, services are delivered by a separate public sector agency—typically a municipal transport company or arm’s length body—but the relationship is governed by administrative rather than contractual mechanisms. The authority directs the operator through ownership and political levers, not through a formal contract.

This preserves public sector ethos and avoids procurement complexity, and for some systems that is enough. What it lacks is the accountability and performance incentive structure that a formal contract provides. If the operator underperforms, the remedy is a difficult meeting rather than a defined regime. Genuinely pure examples of this model are surprisingly hard to find—when I run training exercises on it, the room usually struggles to name more than a handful if that.

Model 3: Public sector (contractual oversight)

Model 3 introduces the discipline of a formal contract while keeping delivery in public hands. A contract governs the relationship between the authority and a public sector operator, bringing performance accountability, KPIs and financial consequences for failure.

Transport for Wales is the standout UK example, holding the Wales and Borders rail services under contractual arrangements with a publicly owned operator. Scotland’s CalMac ferry network and ScotRail follow a similar logic while Great British Railways will be running trains for DfT Operator Limited (DFTO) is the government’s rail owning group and delivery partner for the public ownership programme 

The attraction is obvious: the authority maintains public sector delivery while gaining the structure of contractual regulation—defined obligations, measurable performance and a clear basis for intervention. Some systems also use this as a transitional model, building contract management capability before opening services to competitive tender.

Model 4: Private operator (contractual oversight)

Model 4 is the most common outsourcing model in modern public transport, and the fastest growing. A private operator is selected through competitive tendering and delivers services under a formal contract—referred to as a franchise or a concession depending on the jurisdiction. The authority retains control over service design, fares and network planning; the contract specifies performance obligations, a KPI regime and a revenue structure.

The examples span the globe: London buses under TfL’s gross cost model, Singapore’s MRT, Sydney Metro Northwest, Stockholm’s bus contracts and in the UK rail franchises between 1992 – 2020.  

That revenue structure—gross cost or net cost—is the single most consequential decision within Model 4, which is why it gets its own section below. This is also the model where most of Surbon Consulting’s advisory work sits, on both sides of the table: authorities designing the contracts and operators bidding into them.

Model 5: Private operator (no contract)

At the far end of the spectrum, private operators run services on commercial terms without a formal contract with any authority. Route planning, fares and scheduling are driven by commercial viability rather than public need, and regulation is limited to safety, licensing and general consumer law.

UK examples include Flixbus, National Express in the coach market and rail open access rail operators—Lumo, Grand Central and Hull Trains—which run alongside the contracted network. The model is far more common in bus and coach than in fixed infrastructure, for the simple reason that rail assets are hard to enter and exit commercially.

Why authorities use contracts: the case for contractual regulation

Why does the sector keep converging on the contracted models—Models 3 and 4? The UITP framing is the one I keep coming back to: regulation is about combining the benefits of market competition with the coordination benefits of integrated public planning. A pure market focuses on profitable routes and fragments the network; a pure hierarchy breeds monopoly, inertia and cost. The contract is the instrument that lets an authority take the best of both and public transport is a social good and need.

At its simplest, a contract answers three questions: who does what, who is responsible for what, and who should do what. But the value runs deeper. I suggest the top 10 reasons for a contract are: 

• It navigates the law—the contract sits within, and gives effect to, the regulatory framework.
• It ensures payment—both parties know what is owed, when and for what.
• It minimises disputes—ambiguity is resolved on paper before it is tested in service.
• It protects intellectual property—data, systems and know-how have defined owners.
• It limits liability—each party knows its exposure.
• It creates certainty—for planning, investment and financing.
• It covers the full scope of the relationship—not just the timetable, but assets, people and handover.
• It improves customer service accountability—performance is measured, not assumed.
• It builds trust—a well-drafted contract is the foundation of a working partnership, not a substitute for one.
• It manages expectations—on both sides, from day one to end of term.

In the public transport context there is an eleventh function that matters more every year: the contract is the vehicle through which the authority embeds its policy priorities—decarbonisation targets, accessibility standards, social value commitments—into binding operator obligations. A policy without a contractual home is an aspiration. A policy written into a KPI regime more often than not gets delivered.

Gross cost vs net cost contracts: how revenue risk is allocated

Within Model 4 sits the decision that most shapes the commercial reality of a contract: who takes the revenue? The revenue model determines who carries the risk of passenger demand being higher or lower than forecast, and who benefits if demand grows. Under a gross cost contract, the authority collects the fares and pays the operator to run the service. Under a net cost contract, the operator keeps the fares and lives with the consequences though there can be mechanisms to add to encourage profit share but that is another article. Everything else about the two structures flows from that difference.

Gross cost: the authority carries revenue risk

Under a gross cost contract, the operator is paid all costs plus a management fee, and the authority collects farebox revenue. The operator has no direct stake in whether passenger numbers grow or fall—which is precisely why these contracts rely on incentive and penalty regimes to maintain service quality and drive performance. Many also include a patronage incentive: additional payment if growth in patronage exceeds a benchmark level, giving the operator some skin in the demand game.

The advantages: the operator focuses on production efficiency, which is what it controls; the authority controls fares and network revenue, which is what it wants; and accountability is clean. The disadvantages: the authority bears all revenue risk and must actively work to grow demand itself, and a poorly designed incentive regime can push an operator towards cost minimisation at the expense of service quality.

London buses and Sydney Metro are the classic gross cost examples—and it is no coincidence that gross cost is the structure behind the fastest-growing contracting model worldwide. Authorities under fiscal pressure, post pandemic like knowing exactly what a network will cost.

Net cost: the operator carries revenue risk

Under a net cost contract, the operator collects fare revenue and takes the risk that it will be higher or lower than forecast, typically paying profit above an agreed cap back to the authority as revenue share. The Melbourne tram franchise is a long-running example.

Pure net cost is increasingly rare, and most modern contracts soften it with risk-sharing mechanisms. A cap and collar mechanism shares the pain and the gain: if revenue falls significantly short, the authority absorbs part of the loss; if it significantly exceeds forecast, the upside is shared. A revenue reset mechanism periodically adjusts contract payments to reflect actual revenue performance against forecast, keeping the deal fair when the world changes underneath it.

The advantages: the operator has a direct commercial incentive to grow patronage and improve the service, and subsidy requirements can be lower where demand is strong. The disadvantages: operators may concentrate on revenue-generating routes at the expense of social obligation routes; revenue uncertainty makes pricing the contract genuinely difficult; and external shocks expose the operator to serious financial risk. The pandemic made that last point brutally concrete—operators holding revenue risk watched their commercial model evaporate in a fortnight, through no fault of their own.

Choosing between gross cost and net cost: key considerations

Neither model is universally superior. The right choice depends on context, and in my experience four factors do most of the work:

• Fiscal risk appetite. If subsidy certainty is the authority’s priority, net cost transfers demand risk to the operator—but at a price premium, because operators price the risk into their bids.
• Market maturity. Established markets with good patronage data make net cost pricing tractable. Greenfield or newly restructured networks are very hard to price on a net cost basis, and asking bidders to do so produces either inflated bids or reckless ones.
• The policy environment. Where fare levels are politically controlled, operators cannot manage revenue risk effectively—so asking them to hold it is asking them to price the politics.
• Market appetite. Some operators simply will not bid for net cost contracts in high-risk markets. A perfectly designed contract that attracts one bidder cannot be described as t a competition.

The discipline I recommend is: scenario-test before you commit. Model the financial outcomes under different demand scenarios for both structures. 

What constrains the operating model choice?

The choice of operating model is not made in a vacuum, and the theoretical best answer is often not available. Three constraints do most of the shaping.

The regulatory framework. In the UK, bus franchising powers require mayoral authority or Secretary of State approval and a full business case process; rail services are governed by their own franchise and concession legislation; different rules again apply to light rail, tram and metro. The first question is never ‘what model do we want?’ but ‘what model are we allowed?’

Policy preferences. Some authorities carry political or institutional commitments to public sector delivery; others are bound the opposite way. These preferences are legitimate—they reflect electoral mandates—but they narrow the option space before any analysis begins. The creation of Great British Railways reflects exactly this kind of policy shift, towards greater public coordination of the rail network started by a Conservative government but taken forward under labour.

Market engagement. Even where an authority has the legal power and the political will, the market may not respond as expected. A greenfield Model 4 procurement in an emerging market can attract limited bids if operators perceive the demand risk or political risk as too high. This is why serious pre-market engagement belongs at the start of the process, not as a courtesy at the end. I have seen well-designed procurements stumble simply because nobody asked the market whether it would be interested in bidding. 

Hybrid and dual management models in practice

Few systems sit at a pure model extreme, and some of the most interesting arrangements deliberately combine models to achieve specific objectives.

Barcelona: The Àrea Metropolitana de Barcelona (AMB) runs a dual bus management model: TMB, the public operator, manages the core urban routes, while private operators deliver suburban connections under contracts of up to ten years. The benefit of the arrangement is the benchmarking it creates—public and private delivery running side by side. The authority gets accountability through competition without full privatisation.

The UK provides its own hybrids. London’s bus network is private operators under gross cost Model 4 contracts to TfL—private delivery inside a completely publicly planned network. Greater Manchester’s bus franchising has been a progressive rollout of Model 4 since 2023 rather than a single switch-over. And open access rail runs Model 5 services alongside the contracted network.

The pattern across all of them: hybrid models let authorities keep public control of strategic assets and core routes while harnessing private sector efficiency for the rest. The question is not public versus private. It is which parts of the network need which treatment.

International case studies: what good looks like

Four contrasting systems, all Model 4, all successful, all structured differently. The comparison below draws on UITP case study data I use in contract management training.

System	Contract size	Duration	Bus ownership	Revenue risk	Key performance focus
Singapore	300–400 buses	5+2 years	Authority	With authority	Passenger satisfaction, waiting times, punctuality
Stockholm, Sweden	100–160 buses	10+2 years	Private contractor	Partly with contractor (passenger numbers)	Punctuality, customer satisfaction, reliability
Sydney, Australia	2,500 buses (metro-wide)	7 years with options	Private contractor	With authority	Punctuality, driving style
Gothenburg, Sweden	100 buses (average)	10 years	Private contractor	Partly with contractor (passenger numbers)	High passenger satisfaction; cancellation penalties

Table 2: International Model 4 contract structures—selected comparisons

Asset ownership varies: Singapore’s authority keeps the buses, the Swedish systems transfer them to the contractor. Contract duration lengthens where the operator bears capital investment risk—ten years in Gothenburg against Singapore’s five-plus-two. Revenue risk allocation reflects local market conditions and deliberate policy choice, with both Swedish systems sharing it and both Singapore and Sydney keeping it with the authority.

None of these systems is the ‘right’ answer. Each reflects a deliberate choice about how to balance risk, accountability and incentives—and each works because the choice was made consciously. The role of an adviser is to help authorities and operators understand these trade-offs and identify the model that fits their specific context, rather than importing someone else’s answer.

Frequently asked questions

What is the difference between a gross cost and net cost transport contract?

The difference is who keeps the fare revenue, and therefore who carries the risk. Under a gross cost contract, the authority collects fares and pays the operator its costs plus a management fee—revenue risk sits with the authority. Under a net cost contract, the operator collects fares and carries the risk that revenue falls short of forecast, usually with profit above a cap shared back to the authority. Gross cost buys the authority certainty and control; net cost buys the operator an incentive to grow the marke but at a price.

What operating model does London use for its bus network?

London uses Model 4 with gross cost contracts. Transport for London designs the routes, sets the fares, collects the fare revenue and owns the bus stops and supporting infrastructure. Private operators—Arriva, Go-Ahead, First, Metroline, Stagecoach and others—bid competitively for route bundles and are paid to operate them, with a quality incentive regime driving performance. TfL retains all patronage risk. This is the model Greater Manchester has adopted for its franchised bus network.

What is bus franchising and how does it differ from deregulation?

Under deregulation—the status quo outside London since 1986—private operators decide which routes to run, at what frequency and at what fares, subject to registration requirements. Under bus franchising, the authority decides the network, specifies the service and competitively tenders contracts for operators to deliver it. Franchising powers were introduced nationally through the Bus Services Act 2017 and require a business case process. Greater Manchester is the first area outside London to franchise its network, with services progressively transitioning from 2023, and a further wave of mayoral authorities is now following.

How do transport authorities decide which operating model to use?

Three primary factors shape the decision: the regulatory framework (what powers the authority actually has), policy preferences (what its political and institutional goals are) and market engagement (what operators are likely to bid, and on what terms). A structured selection process should define the authority’s objectives for the contract, assess the risk allocation implications of each model, test market appetite through pre-market engagement and scenario-test financial outcomes under different demand assumptions—before the tender documents are drafted, not after.

What is an open access operator in UK rail?

Open access operators—Lumo, Grand Central and Hull Trains—are Model 5 in the UITP typology: commercial rail services run without a contract with any public authority. They are licensed by the Office of Rail and Road and pay track access charges, but routes, fares and schedules are commercial decisions. Open access exists alongside the contracted network and is constrained by the principle that it should not cause ‘abstraction’ of revenue from publicly subsidised services.

How Surbon Consulting advises on operating model decisions

The operating model question is where Surbon Consulting’s advisory work on procurement strategy, bid support and mobilisation all meet—and we work on both sides of it.

Supporting transport authorities on model selection and procurement design

We advise transport authorities on operating model choice, procurement structure and contract design in UK and international contexts—from franchising implementation to greenfield procurements in emerging markets. The value of adviser input at this stage is helping authorities understand the downstream implications of their model choice before tender documents are issued, not after. The model determines the risk allocation, the risk allocation determines who bids and how they price, and by the time the ITT is published those decisions are locked in. Pre-market engagement, scenario testing and honest advice about what the market will bear all belong at this early stage.

Supporting operators bidding into contracted models

From the operator’s perspective, understanding the operating model is essential to pricing and bidding correctly. A bid team that does not understand whether revenue risk sits with the authority or the operator cannot produce a credible commercial proposition—and I have reviewed bids where that misunderstanding was visible on every page. We help operators understand the model they are bidding into, structure the bid around it and plan for the mobilisation that follows, because the model shapes the transition as much as it shapes the contract. Gross cost demands you demonstrate production efficiency and quality; net cost demands a credible revenue growth story. Knowing which game you are playing is the first requirement of winning it.

Navigating an operating model decision or preparing a bid for a contracted transport service? Surbon Consulting advises transport authorities and operators on model selection, procurement design and contract mobilisation. Get in touch.