Freehold covenants

Freehold covenants are promises concerning freehold land. They are usually made on a sale of part: A sells Plot 1 to B and B covenants not to build above two storeys, not to use the land for trade, or to contribute to the maintenance of a private road. The immediate parties are bound by contract. The difficult question is whether later owners of Plot 1 and the retained land are bound or entitled to enforce. That question sits at the junction of contract, property, equity, and registration.

The organising distinction is between the benefit and the burden. The benefit is the right to sue on the covenant. The burden is the obligation to perform or refrain. English law is generous as to the running of the benefit, but conservative as to the running of the burden. At law, the burden of a freehold covenant does not run with freehold land. That remains the rule after Austerberry and Rhone. Equity, however, will allow the burden of a restrictive covenant to bind successors under the doctrine in Tulk v Moxhay, provided the covenant is genuinely negative, benefits identifiable land, and is properly protected under the relevant registration system.

This asymmetry is the central intellectual point. Leasehold covenants, studied last week, are largely governed by statutory regimes permitting transmission of both benefit and burden. Freehold covenants are not. The freeholder is still treated, in principle, as an owner not to be made personally liable for another person’s contractual undertaking merely because land has been conveyed to him. Equity mitigates that rule for negative restraints, because an injunction preventing prohibited use can be conceptualised as enforcing a proprietary equity against the land rather than imposing an affirmative personal obligation.

The practical importance is considerable. Modern estates depend on private regulation: building lines, user restrictions, parking controls, maintenance of roads and shared open spaces, and architectural uniformity. The law supplies strong tools for negative covenants but unsatisfactory tools for positive covenants. Conveyancers respond by using estate rentcharges, long leases, management companies, chains of indemnity covenants, or commonhold. Each is a workaround for the refusal to let positive freehold burdens run automatically.

For Durham purposes, this topic consolidates the first ten weeks of the Year 2 Land Law module. You must use the taxonomy of estates and interests from Week 1, the registered and unregistered priority rules from Weeks 2 to 4, and the distinction between proprietary and merely personal rights sharpened by licences and proprietary estoppel in Week 10. The best answers do not recite Tulk v Moxhay mechanically. They ask four questions in order: what is the covenant; who seeks to enforce; against whom; and under which system of title and priority? In problem questions, the answer usually turns on one missing link: no benefited land, no annexation or assignment of the benefit, a positive covenant, lack of registration, or an over-ambitious attempt to use the benefit and burden principle.

The law of freehold covenants developed from two opposing instincts. The first is contractual: a covenant is a promise under seal, enforceable between promisor and promisee. The second is proprietary: land use promises may be intended to regulate land permanently and to benefit neighbouring land rather than the original covenantee personally. The history of the subject is the effort to reconcile those instincts without allowing parties to create unlimited new forms of property obligation.

At common law the approach was restrictive. A covenant could bind the original covenantor personally, and the benefit of some covenants could pass to successors where the covenant touched and concerned the land. But the burden of a freehold covenant did not run with the land. The common law was suspicious of imposing personal liability on a successor who had not contracted. Austerberry v Oldham Corporation fixed the modern formulation: a freehold successor is not liable at law merely because he has acquired the land. That rule remains central, despite repeated criticism.

Equity supplied the major exception. Tulk v Moxhay arose from Leicester Square. The purchaser had covenanted to keep the square as an ornamental garden. A later purchaser, with notice, sought to build. Lord Cottenham LC restrained him. The decision is often treated as a landmark in the development of equitable property rights: equity would not permit a purchaser to take land at a reduced price because of a restriction and then disregard the restriction. The historical language was conscience and notice; the modern explanation is that a restrictive covenant may create an equitable proprietary interest in the burdened land.

Nineteenth- and early twentieth-century cases then worked out the limits. Equity would not enforce positive obligations against successors, because compelling expenditure or active performance looked too much like imposing personal contractual liability. The covenant had to benefit land retained by the covenantee; it could not be a mere personal or commercial right in gross. London County Council v Allen illustrates that point. Nor was it sufficient that the covenant was aesthetically or commercially useful. It had to be tied to an identifiable dominant tenement.

The Law of Property Act 1925 then altered the position on the benefit side. Section 78 deems covenants relating to land of the covenantee to be made with successors in title. The Court of Appeal in Federated Homes interpreted this provision strongly: it can effect statutory annexation of the benefit to the benefited land, reducing the need for express words of annexation in many modern conveyances. Section 79 deems covenants relating to land of the covenantor to be made by the covenantor on behalf of successors, unless contrary intention appears. But, crucially, section 79 does not abolish the rule that the burden of positive freehold covenants does not run. Rhone v Stephens confirmed that the 1925 legislation did not reverse Austerberry.

Registration changed the practical operation of notice. In unregistered land, post-1925 restrictive covenants are registrable as Class D(ii) land charges; failure to register may make them void against a purchaser of a legal estate for money or money’s worth. In registered land, they are protected by entry of a notice on the charges register of the burdened title. Registration does not make an invalid covenant valid. It protects priority if the covenant is valid on ordinary principles. This is a recurring examination point. Students frequently treat registration as constitutive. It is not.

The modern law is therefore layered. It contains common law privity, equitable intervention, statutory annexation, land charges, and registered-title priority. Its structure is historically explicable but conceptually untidy. That untidiness is why reform has been repeatedly proposed, most notably by the Law Commission. Yet the old division remains: negative covenants may run in equity; positive freehold burdens generally do not.

Begin with the classification of the covenant. A covenant is positive if it requires expenditure, action, repair, maintenance, or contribution. A covenant is restrictive if compliance is possible by doing nothing. Substance prevails over drafting. A covenant not to allow a wall to fall into disrepair is positive in substance, because it requires maintenance. A covenant not to build more than one dwelling is restrictive. A covenant to pay towards road maintenance is positive. This first classification often decides the problem.

The original parties are bound by contract. If A covenants with B, A remains personally liable even after selling the land, unless released by agreement or by the terms of the covenant. Likewise B may enforce as promisee. The problem of freehold covenants concerns successors: C, who buys the burdened land, and D, who buys the benefited land. Do D and C stand in the shoes of B and A?

At law, the benefit of a covenant may run if the covenant touches and concerns the land, the covenantee had a legal estate, the successor takes that estate, and the parties intended the benefit to run. Section 78 of the Law of Property Act 1925 is now central. In Federated Homes, the Court of Appeal held that section 78 can annex the benefit of a covenant to the land of the covenantee where the covenant relates to that land. Annexation means that the benefit passes automatically with the land, without needing express assignment each time the land is sold. But one must still identify the land intended to be benefited. A covenant cannot be annexed to air. London County Council v Allen remains important: the covenantee must retain land capable of benefiting from the covenant.

At law, the burden of a freehold covenant does not run. A successor to the covenantor’s freehold estate is not liable at common law merely because he owns the land. Austerberry states the rule; Rhone confirms it. Section 79 does not change this. It supplies deemed wording of intention but not a mechanism for transmission of burden. Therefore, a positive covenant to repair a roof, maintain a road, or contribute to a service charge will not bind a later freehold owner at law.

The cases form a coherent line if arranged by function. Tulk v Moxhay supplies the equitable foundation. It shows that a purchaser of burdened land may be restrained from acting inconsistently with a restrictive covenant where it would be inequitable to disregard the restriction. Modern law no longer depends simply on actual notice; registration now performs much of that priority function. But Tulk remains the starting point for the proposition that the burden of a negative freehold covenant may run in equity.

Austerberry v Oldham Corporation supplies the contrary common law baseline. The Court of Appeal refused to allow the burden of a positive obligation concerning a road to bind successors. It is the case to cite for the proposition that freehold burdens do not run at law. Haywood v Brunswick Permanent Benefit Building Society is an associated authority on positive obligations: equity will not enforce against successors a covenant requiring expenditure on repair. These cases explain the enduring distinction between restraining use and compelling action.

London County Council v Allen prevents the doctrine from becoming a right in gross. A covenant must benefit land, not merely serve a personal, commercial, or public object of the covenantee. The covenantee must own or retain land capable of benefiting from the restriction. In practice this requirement screens out attempts by a seller with no retained land, or a public body acting only in a regulatory capacity, to enforce as though it held a proprietary right.

Federated Homes is the leading modern case on the benefit. The Court of Appeal gave section 78 real force by treating it as capable of annexing the benefit of a covenant to the covenantee’s land. This is why post-1925 conveyances often require less elaborate express annexation than older cases suggested. Yet the case should not be overstated. It does not remove the need to identify the land benefited, nor does it allow covenants to float unattached over an estate.

Rhone v Stephens is the controlling modern authority on positive covenants. The House of Lords refused to enforce against a successor a covenant to maintain a roof. Lord Templeman reaffirmed the rule that equity will not compel a successor to perform a positive obligation. He also confined the benefit and burden principle. It is not a device for making all positive covenants run. It operates, if at all, where the successor chooses to take a particular benefit and the burden is genuinely conditional on that benefit.

Crest Nicholson Residential (South) Ltd v McAllister is important for conveyancing realism. The Court of Appeal refused to find enforceability where the benefited land could not properly be identified and where the covenants were not annexed in the way asserted. The case is a useful corrective to simplistic reliance on section 78. The question is not whether the word successors appears, but whether the instrument, land retained, and estate context show a benefit attached to identifiable land.

Finally, Elliston v Reacher and later building scheme cases show how mutual enforceability may arise within a defined development. The doctrine is demanding because it creates a network of proprietary control among purchasers. The court looks for a common vendor, a defined estate, substantially common restrictions, and an intention that each purchaser should take both the benefit and burden of the scheme. In exam analysis, a building scheme is a possible alternative route, not a substitute for proof.

See practice questions section below.

See further reading section below.