The Split Incentive Solved: Who Pays, Who Saves

You own the roof. Your tenant pays the electricity bill. So when you spend £185,000 putting a 250kWp array on the building, who actually captures the saving? This is the split incentive, and it is the single question that decides whether a commercial solar project is worth doing for a landlord. Get the ownership and lease structure right and the array pays for itself and lifts the asset. Get it wrong and you have funded an improvement that benefits only the occupier, with no rent, no recovery, and an EPC gain you cannot monetise. This guide sets out the five routes a commercial property owner can take through the split incentive, what each one does to who-pays and who-benefits, and how to choose between them by asset type.

What the split incentive actually is

The split incentive is the structural mismatch between the party that pays for an energy efficiency measure and the party that benefits from it. In commercial property it follows directly from the lease.

Under a full repairing and insuring (FRI) lease — the standard institutional form for single-let and most multi-let property — the tenant takes a direct supply contract and pays the electricity bills. The landlord owns the building fabric and the roof but consumes almost no power within the demised areas. So if the landlord funds rooftop solar, the generation displaces grid electricity that the tenant was buying. The tenant pockets the saving; the landlord carries the capital cost.

That is the trap. It is why so many otherwise sound buildings have bare roofs. The economics are real — self-consumed solar displaces grid power at roughly 24–28p/kWh in 2026 against a generation cost well under that — but under an unmodified FRI lease the value lands on the wrong side of the demise.

Two things change the calculation, and both push owners to act rather than wait:

• MEES. It has been unlawful to let commercial property in England and Wales below EPC E since 1 April 2023 (the rule first applied to new lets from 1 April 2018, and now binds even sitting tenancies). Solar typically lifts a commercial EPC by one to three bands. A proposed EPC B minimum by 2031 for privately-rented non-domestic buildings over 1,000 m² — set out in the Government’s interim response of 18 June 2026, cost-effectiveness tested and still subject to secondary legislation — would sharpen this further. (Note the earlier “EPC C by 2027” target was scrapped, and “EPC B by 2030” was never law.) See our MEES and EPC guide for the full regulatory position.
• Asset value. Energy performance is now priced into rent and capital value. In prime central London offices, JLL associated a single EPC band improvement with +4.2% rent and +3.7% capital value (2017–21); read that as directional, not a guarantee, and not a read-across to secondary stock.

So the question is not whether to put solar on the building. It is which structure lets you, the owner, capture the value.

The five routes through the split incentive

There are five clean ways an owner can structure a commercial solar project. They differ on who funds the capital, who consumes the power, who benefits, and how much control you retain. The right one depends almost entirely on how the building is occupied.

Route A — Common-parts / landlord-controlled supply

The landlord installs solar to power the parts of the building it controls and it pays for: lifts, common-area and external lighting, communal HVAC and pumps, car-park lighting, EV chargers, and landlord-metered plant. Here there is no split incentive at all — the landlord both funds the array and consumes the output, displacing grid power it was already buying through the service charge.

This is the cleanest starting point for any multi-let building. The landlord supply is modest (typically 10–25% of total building load), so the array is smaller, but every kilowatt-hour generated displaces the landlord’s own cost, and the EPC and ESG benefits accrue to the owner directly. It also cuts holding costs during void periods, when common-parts energy is an unrecovered drain. Full detail in our common-parts landlord supply guide.

Route B — Landlord–tenant PPA (behind-the-meter, private-wire or sleeved)

The landlord funds and owns the array, then sells the generated electricity to the occupying tenant under a power purchase agreement (PPA) priced below the grid tariff. The tenant saves on every unit it buys from the array; the landlord earns a generation revenue that repays the capital and produces an ongoing return.

Behind-the-meter (the solar connects on the tenant’s side of the meter) and private-wire arrangements are the usual forms; a sleeved PPA routes the power through a licensed supplier. The tenant gets cheaper, greener power with no capital outlay. The landlord retains ownership of the asset and a long-dated income stream — but takes on the role of a power seller, with metering, billing and a PPA contract to administer. Our landlord–tenant PPA and private-wire guide covers the contract mechanics, pricing and metering.

Route C — Sell-the-roof / airspace lease

The owner leases the roof or airspace to a third-party developer, who funds, installs, owns and operates the array. The owner takes a rent for the roof and usually a discounted power offer to the tenant; the developer takes the generation economics over a 20–30 year term.

This is the lowest-capital, lowest-effort route — the owner puts in no money and carries no operational burden. But it is also the lowest-return, and it is a registrable interest. A roof lease creates a legal estate with SDLT and Land Registry consequences, and it needs mortgagee and insurer consent. The structural and reversion issues are real: “no sun, no power” means the developer controls the roof, and at lease end the array must be dealt with. Compare it head-to-head with the PPA and licence options in our roof lease vs PPA vs licence guide.

Route D — Green lease contribution

The landlord installs the array and the lease is varied (or a new green lease is granted) so the tenant contributes to the cost — but the contribution is capped at the tenant’s own energy savings. The tenant is no worse off; the landlord recovers part of the capital from the party that benefits from the cheaper power.

Green leases work within the framework of the BBP Green Lease Toolkit and are the natural mechanism on a single-let FRI building where the tenant is engaged on ESG. The cap on the tenant’s contribution is what makes it fair and lettable. It needs a willing, sophisticated tenant and a lease event to hang the variation on. See our green leases and solar guide.

Route E — Owner-occupier

The owner occupies the building, funds the array and consumes the power. There is no split incentive because owner and occupier are the same party. The owner captures 100% of the economics — self-consumption savings, export income, capital allowances, business-rates exemption and the EPC and ESG gains.

This is the simplest and strongest case for solar, with the fastest payback (often 3–5 years on a high daytime-load site). If you own and occupy, there is nothing to structure around. Our owner-occupied commercial property page sets out the full economics.

The five routes compared

The routes are not mutually exclusive. The most common real-world structure on a multi-let building is A plus B: the landlord powers the common parts directly (no split incentive) and sells the surplus to tenants under PPAs. We design the combination to fit the building.

The service-charge truth: capex is not recoverable

A landlord cannot recover the capital cost of new solar plant through the service charge. This is the most common and most expensive misunderstanding in the market, and it is settled by the RICS professional standard.

The RICS Service Charges in Commercial Property (2nd edition, in force 31 December 2025) draws a clear line between operating expenditure (recoverable) and the initial capital cost of new plant and equipment (not recoverable). Solar PV is new plant. Putting it in next year’s service-charge budget and apportioning it across the tenants is a breach of the standard, and tenants advised by surveyors will reject it.

So the service charge is the wrong tool. The capital has to be funded another way, and that is exactly what the five routes do:

• Route B (PPA) recovers the capital through the price the tenant pays per kilowatt-hour — a supply charge, not a service charge.
• Route C (roof lease) funds it off the owner’s balance sheet entirely; the developer pays.
• Route D (green lease) recovers a capped contribution through an agreed lease term, not the service-charge apportionment.
• Routes A and E are funded by the owner and repaid through the owner’s own consumption savings, with no recovery from tenants at all.

The running costs of solar — maintenance, monitoring, inverter servicing — can sit in the service charge once the asset is operational, because that is operating expenditure. The capital cannot. Structure for that from the outset.

Choosing the right route by asset type

The right route falls out of how the building is occupied. Work through it in this order.

1. Do you own and occupy? → Route E. No split incentive, 100% of the economics, simplest case. Stop here.
2. Is it a single-let FRI building with an engaged tenant? → Route B (PPA) if you want to own the asset and earn a generation income; Route D (green lease) if there is a lease event to hang a capped contribution on and the tenant prefers that. Either captures value for both sides.
3. Is it multi-let? → Route A first (power the common parts you control — no split incentive, immediate EPC and void-cost benefit), then layer B to sell surplus to tenants under individual PPAs. This is the workhorse structure for offices and mixed-use.
4. Is it a large industrial or logistics shed where you want zero capital and zero operational burden? → Route C (sell-the-roof lease) — accept the lower return and the registrable-interest mechanics in exchange for a hands-off rent. Just clear the mortgagee, insurer and reversion points first.
5. Is the asset vacant or about to re-let? → Route A on the common parts cuts holding costs during the void and lifts the EPC for the next letting — the array works for you before the new tenant arrives.

Whatever the answer, the owner-side due diligence is the same and it runs in parallel: structural roof-loading survey, roof lifecycle alignment (re-roof and solar in one project if the membrane has under 15 years left), lender and insurer consent, dilapidations and reversion treatment, and the capital-allowances position. Solar PV is a special-rate (integral features) asset; the 100% first-year relief comes from the Annual Investment Allowance (£1m, permanent), not from full expensing — and assets bought to lease cannot use full expensing at all, so roof-lease and PPA structures rely on AIA plus the 6% writing-down allowance. We map all of this before a single panel is specified.

That is the order of work that matters: structure the ownership and the lease first, design the system second. A technically perfect array on the wrong commercial structure is a stranded cost. Get a structured assessment of which route fits your building with a free quote.

Frequently asked questions

Why can’t I just put solar on my let building and recover the cost from the service charge?

You can’t because the RICS Service Charges in Commercial Property standard (2nd edition, in force 31 December 2025) treats the initial capital cost of new plant — which solar is — as non-recoverable through the service charge. Only the operating costs once the asset is running (maintenance, monitoring) can go in the service charge. The capital has to be funded another way: a tenant PPA where repayment comes through the per-unit power price, a roof lease where a developer funds it, a capped green-lease contribution, or the owner’s own consumption savings. Trying to push £185,000 of new plant through the service-charge budget will be rejected by any tenant taking surveyor advice.

My tenant pays the electricity bills, so don’t they get all the benefit of solar I install?

Under an unmodified FRI lease, yes — and that is precisely the split incentive the five routes are designed to solve. If you do nothing structural, landlord-funded generation displaces grid power the tenant was buying, so the tenant captures the saving while you carry the capital. The fix is to change who consumes or who pays: power the common parts you control directly (Route A, no split at all), sell the power to the tenant under a PPA at a price below grid (Route B, you earn the income), or agree a capped green-lease contribution (Route D). Each one routes the value back to the owner.

What’s the difference between a roof lease and a tenant PPA?

A roof lease (Route C) leases the roof or airspace to a third-party developer who funds, owns and operates the array; you take a rent and carry no capital, but it is a registrable legal interest with SDLT and Land Registry consequences and it needs mortgagee and insurer consent. A landlord–tenant PPA (Route B) keeps you as the owner of the array — you fund it and sell the electricity to your tenant below grid price, retaining the asset and a long-dated income. The roof lease is lower-effort and lower-return; the PPA keeps the upside and the asset on your balance sheet. Our roof lease vs PPA vs licence guide compares them in full.

Which route gives the best return for a commercial landlord?

For an owner-occupier, Route E is unbeatable — 100% of the economics and the fastest payback. For a landlord, ownership-retaining routes (A and B) give the strongest long-run return because you keep the asset and the income, where a roof lease (C) trades return for zero effort and zero capital. The genuinely best answer, though, depends on the building: a multi-let office is usually A plus B combined, a single-let shed with a passive owner often suits C, and a single-let building with an ESG-engaged tenant can work well as a green lease (D). The return is set by the structure, not the panels — which is why we design the ownership and lease first.