A funder advancing development finance, a purchaser buying the finished building, and a tenant signing an agreement for lease all have the same problem: none of them has a contract with the contractor, the subcontractors, or the design team who actually built the thing. They sit outside the chain of contracts entirely. If something goes wrong – a defective curtain wall, an underperforming M&E installation, a structural design error – their only route in is a collateral warranty or a third-party rights clause, and getting either one wrong, or late, is a problem that lands on the QS’s desk as much as the lawyers’.
Why this mechanism exists at all
English law’s privity rule means a contract only binds and benefits the parties who signed it. On a typical development, the employer contracts with the main contractor, who subcontracts the envelope, the services and the structure, and separately appoints an architect, a structural engineer and other consultants. A funder, a purchaser or a tenant brought in later has no direct relationship with any of them. Without one, their only fallback is a claim in negligence – which means proving a duty of care was owed to them specifically, and then running into the courts’ long-standing reluctance to let pure economic loss (the cost of fixing a defect, as opposed to physical damage it causes) be recovered that way. Two mechanisms close that gap: a collateral warranty, or a third-party rights clause relying on the Contracts (Rights of Third Parties) Act 1999.
Collateral warranties: what’s actually in one
A collateral warranty is a standalone contract, signed separately from the building contract or professional appointment it sits alongside, in which the contractor, a subcontractor or a consultant promises a named beneficiary – the funder, purchaser or tenant – that it has performed, and will go on performing, its underlying obligations with the skill and care expected of its trade. Strip away the bespoke drafting and most warranties are built from the same handful of clauses:
- The principal covenant – the warrantor confirms it has exercised, and will continue to exercise, reasonable skill and care in carrying out its obligations under the underlying contract.
- A “no greater liability” proviso – the warrantor’s exposure to the beneficiary can’t exceed what it already owes under the main contract or appointment; insurers generally insist on this before they’ll price the risk.
- Step-in rights – letting the funder or purchaser take over the employer’s position and keep the contract running, most commonly if the employer becomes insolvent, rather than losing the benefit of the contract along with the employer.
- Assignment – the right to pass the warranty’s benefit on, typically capped at one or two further assignments, which matters most to a purchaser who may sell the building on.
- An insurance undertaking – committing the warrantor to maintain professional indemnity or other cover at a specified level for the liability period.
On JCT contracts this is formalised into standing templates rather than drafted from scratch each time: JCT publishes dedicated warranty deeds for a funder (CWa/F) and for a purchaser or tenant (CWa/P&T), with parallel versions for sub-contractors, all periodically updated. Most are executed as deeds rather than simple contracts, partly because a deed needs no consideration from the beneficiary and partly because a deed carries a 12-year limitation period in England and Wales rather than the 6 years that applies to a simple contract – though that only holds if the warranty is drafted to run from the same point as the underlying contract. In one widely discussed English case, a warranty signed well after practical completion was still held to apply retrospectively to the whole contract period, because its wording and its “no greater liability” clause showed the parties intended the beneficiary to simply step into the employer’s existing position – not to acquire a separate, later-starting set of rights. It’s a useful reminder that vague drafting on timing is a real risk, not a technicality.
The statutory alternative: third-party rights
The Contracts (Rights of Third Parties) Act 1999 offers a shortcut: if a contract expressly says a named person, or a described class of person, may enforce one of its terms, that person can enforce it directly – without a separate document, a signature, or even necessarily being identified by name at the time the main contract is signed. On NEC4 contracts this is switched on through secondary option Y(UK)3, with the enforceable terms and the beneficiaries (or classes of beneficiary, such as “any purchaser or tenant of the works”) set out in Contract Data Part One; selecting Y(UK)3 and entering “none” is how a party excludes third-party rights altogether rather than leaving the option out. JCT’s main contracts offer an equivalent route through a negotiated third-party rights schedule that can stand in for a CWa/F or CWa/P&T deed, built to cover the same funder and purchaser/tenant protections, step-in rights included.
The appeal is speed and cost: one agreed schedule can cover every future beneficiary in its class, so there’s no chasing a subcontractor for a signature every time a new tenant takes a floor. The trade-off is rigidity – a schedule agreed at contract signing is hard to renegotiate later for one particular funder’s bespoke requirements – and, historically, reluctance from banks and insurers more comfortable with a document they can see has actually been signed. That resistance has thinned over time, but it’s why most projects still default to collateral warranties as the fallback even where third-party rights are available.
| Mechanism | How it’s created | Where you’ll meet it |
|---|---|---|
| Collateral warranty | Separate deed, individually negotiated and signed by the warrantor and each beneficiary | JCT CWa/F, CWa/P&T and sub-contractor equivalents; bespoke warranties on NEC and FIDIC projects |
| Third-party rights schedule | Built into the main contract at signing; no separate signature needed once a beneficiary falls within the agreed class | NEC4 secondary option Y(UK)3; a JCT third-party rights schedule used instead of CWa deeds |
Why this lands on the QS’s desk
None of this is purely a legal drafting exercise. A funder will often make drawdown conditional on warranties being in place, and a purchaser’s solicitors will routinely hold up completion until the full schedule has been signed – so an unexecuted warranty is a cash-flow and programme risk a QS should be tracking the same way as an unresolved bond or an expired insurance certificate, not leaving to sit in a legal correspondence file. A few things worth building into the administration of any project with warranties or third-party rights in play:
- Keep a live tracker of who owes a warranty to whom, by what date, and whether it’s actually been executed – not just requested.
- Check warranty wording against the main contract and any relevant subcontracts before anyone signs; a mismatch in liability caps, net contribution clauses or limitation wording between the two is where disputes actually start.
- Flag step-in rights to the programme team, not just the lawyers – a funder or purchaser stepping in after a contractor’s insolvency changes who the QS is actually answering to, and usually comes with its own notice periods.
- Remember that performance bonds and parent company guarantees protect the employer against the contractor’s default – they don’t give a funder, purchaser or tenant any rights of their own, so one is never a substitute for the other.
Where the underlying contract itself is NEC, JCT or FIDIC changes the default mechanism you’re working with as much as it changes anything else about risk allocation – worth revisiting alongside a comparison of how the three suites handle time, money and dispute risk more generally. And if a dispute over a warranty’s scope or a step-in notice does end up contested, the route is usually the same fast-track process used for any other construction dispute: adjudication, since most warranties incorporate the underlying contract’s dispute resolution provisions by reference.
The Australian picture
Australia has no broad, generally applicable equivalent to the Contracts (Rights of Third Parties) Act 1999 covering commercial construction. The clearest statutory exception sits in New South Wales: sections 18B and 18D of the Home Building Act 1989 impose non-excludable statutory warranties on residential building work, and extend the right to sue the head contractor and subcontractors – even without a direct contract – to the original homeowner and to later successors in title. Outside residential work covered by that kind of regime, Australian funders, purchasers and tenants rely on collateral deeds of warranty in essentially the same way UK parties did before the 1999 Act existed, which is one more reason a QS moving between the two markets shouldn’t assume a mechanism available on a UK project will have a ready equivalent on an Australian one.
The legal detail belongs to the lawyers drafting the clauses. The programme and cash-flow consequences of a warranty that’s late, inconsistent with the main contract, or simply forgotten belong to the QS – which is reason enough to know what’s actually in these documents before the first one lands on your desk for review. NEC’s own guidance on how X8 warranties and Y(UK)3 third-party rights work in practice is a useful primer if you want the contractual mechanics in more detail: see NEC’s note on collateral warranties and third-party rights.