The concession agreement is the master contract that governs a healthcare public-private partnership (P3) for its entire term — typically 25 to 35 years. It binds a public authority and a privately financed special-purpose entity, transfers defined lifecycle and performance risks to the private side, and ties the public side's payments to the facility being available and performing to standard rather than to construction milestones. This Article explains the commercial spine of the model: what the concession agreement covers, which risks move and which stay, and how the availability-payment mechanism converts performance into cash flow.
In a healthcare P3 (usually a Design-Build-Finance-Operate-Maintain, or DBFOM, structure), the public authority — a health system, hospital district, NHS-style trust, provincial health authority, or state agency — does not buy a building and a separate operations contract. It grants a single long-term concession to a private consortium to design, build, finance, and then operate and maintain (the hard side, sometimes soft side too) the facility for a fixed term, after which the asset reverts to the public owner in a contractually defined condition.
The concession agreement (also called the project agreement or partnership agreement) is the governing instrument. It is a single, integrated document — often hundreds of pages with dozens of schedules — that sets out:
Because the agreement bundles design, construction, finance, and decades of operations into one contract, it forces a whole-of-life perspective: the party that designs and builds is the same party that must keep the facility available and pay to maintain it for 30 years, which aligns capital and lifecycle decisions that traditional Design-Bid-Build (DBB) separates.
A defining feature of the concession agreement is that the public authority specifies outputs and performance, not detailed design. Instead of issuing fully designed construction documents, the authority writes a set of output specifications describing required functional and environmental outcomes — for example, the number and type of clinical spaces, required adjacencies, air-change rates and pressure relationships consistent with ASHRAE 170, redundancy of essential electrical systems under NFPA 110, temperature and humidity bands, and minimum availability of each functional area.
This matters for risk transfer: by specifying outcomes rather than means, the authority hands the private partner responsibility for achieving the result. If the design solution the consortium chooses fails to deliver the specified ventilation, redundancy, or availability, that is the consortium's risk to remedy at its cost — not a change order the public side must fund. The trade-off is that the output specifications must be written with great care, because anything not specified is, in effect, not the private partner's obligation. Healthcare output specs are particularly demanding because clinical, infection-control, and life-safety requirements (FGI, ASHRAE 170, USP 797/800 for compounding pharmacies, EMTALA-driven access, ADA/ABA accessibility) are detailed and non-negotiable, and must be embedded so that compliance is the consortium's contractual duty for the full term.
The justification for a P3 is that it transfers risk to the party best able to manage it, and that the value created by better risk management outweighs the higher cost of private finance. The concession agreement is where that allocation is made explicit, risk category by risk category. The principle is allocate each risk to the party best positioned to control or absorb it — not simply to move as much as possible to the private side, which only raises the price without improving outcomes.
A typical healthcare P3 allocation looks like this:
| Risk category | Usually transferred to private partner | Usually retained / shared by public authority |
|---|---|---|
| Design adequacy | Yes — consortium owns the design solution against output specs | Public owns clinical brief / output specs themselves |
| Construction cost & schedule | Yes — fixed-price, date-certain; cost and delay overruns are the consortium's | — |
| Latent defects & quality | Yes — defects affecting availability reduce payment | — |
| Lifecycle / major maintenance & replacement | Yes — consortium funds and schedules whole-of-life renewal | — |
| Hard FM availability (building systems uptime) | Yes — unavailability triggers deductions | — |
| Financing / interest rate (post-close) | Yes — consortium's lenders bear it | — |
| Inflation on FM payments | Shared — typically indexed | Authority bears the indexed escalation |
| Clinical operations & medical staffing | No | Retained by the health authority (clinical care stays public) |
| Demand / volume (patient throughput) | No (in availability-based deals) | Retained — payment is for availability, not utilization |
| Change in law (general) | Often shared | Authority frequently bears discriminatory or healthcare-specific changes |
| Site conditions / contamination (pre-known) | Often retained or shared | Authority may carve out unknown ground conditions |
| Force majeure | Shared per defined relief regime | Shared |
| Political / authority default | No | Retained by public side |
Two allocations are especially important in healthcare and distinguish the model from infrastructure P3s in other sectors: