Who owns the float?

A contractor plans to finish two weeks early. Midway through the works, the employer issues a late instruction that absorbs ten days of that cushion — but the project is still forecast to complete within the contract period. Is the contractor entitled to an extension of time, or to any compensation, for those ten days?

The answer turns on a question that has generated argument for as long as critical-path programmes have existed: who owns the float? It is deceptively simple, and it sits at the precise point where planning meets legal entitlement. Get the answer wrong — or fail to address it in the contract — and otherwise sound delay claims can collapse.

This article sets out what float actually is, the three competing theories of ownership, what the English courts and the Society of Construction Law have said, how the major standard forms deal with it, and why, in practice, the credibility of the programme decides the argument more often than the law does.

WHAT "FLOAT" ACTUALLY MEANS

Precision matters here, because float is not a single thing. Total float is the amount of time an activity can be delayed without delaying the project completion date. Free float is the amount an activity can be delayed without delaying any succeeding activity. An activity on the critical path has, by definition, zero (or negative) total float; non-critical activities carry positive total float that measures their relative criticality.

Two further categories matter to the ownership debate. Project float — in NEC terminology, terminal float — is the gap between the date a contractor plans to finish and the contractual completion date: the contractor's own buffer at the end of the programme. Time risk allowances are durations a contractor builds into individual activities to cover risks it carries. These are conceptually distinct from the network float that emerges from the logic, and, as we will see, the standard forms treat them very differently.

This taxonomy is not academic. AACE International's recommended practices on forensic schedule analysis treat the identification and classification of float as a foundational step, precisely because the label attached to a piece of "spare time" often decides who benefits from it.

THREE THEORIES OF OWNERSHIP

When a delay consumes float that would otherwise have protected the completion date, three positions are possible.

The contractor owns the float. On this view, float is the product of the contractor's planning effort and method, built to give it flexibility and to manage its own risks. Any employer delay that erodes that float deprives the contractor of something of value and should attract an extension of time — even if the completion date is not yet threatened.

The employer owns the float. The opposite view treats float as a project resource that the employer has, in effect, paid for within the contract price and period, and which the employer may therefore consume to absorb employer-risk events without extending time.

The project owns the float. The third position is that float belongs to neither party exclusively. It is a resource of the project as a whole, available to whichever party first needs it — "first come, first served." On this approach, an employer delay that merely erodes float, without pushing completion beyond the contractual date, generally gives no entitlement to an extension of time.

Each theory can be made to sound reasonable in isolation. The question is which one the law and the contract actually adopt.

THE ENGLISH POSITION: ASCON v McALPINE

The leading English authority is Ascon Contracting Ltd v Alfred McAlpine Construction Isle of Man Ltd (1999) 66 Con LR 119. McAlpine, the main contractor on a building in Douglas, Isle of Man, had roughly five weeks of float in its programme. The works finished late, consuming all of that float and more. McAlpine sought, in effect, to appropriate the float — to use it to cancel delays for which it or its subcontractors were responsible, while still claiming the benefit of the float as against the employer.

His Honour Judge Hicks QC rejected that approach. The float could not be claimed by the main contractor as its exclusive property, to be deployed selectively against whichever subcontractor it chose. In a frequently cited passage he observed that "[s]ix sub-contractors, each responsible for a week's delay, will have caused no loss if there is a six-weeks' float. They are equally at fault, and equally share in the 'benefit'." The court held that the contractor could not, while taking the benefit of the float as against the employer, then recover from a subcontractor the hypothetical loss it would have suffered had the float not existed.

Ascon is generally read as authority for the proposition that, where the contract is silent, float belongs to the project and is consumed on a first-come, first-served basis, rather than being owned by any one party to allocate as it sees fit. More than two decades on, it remains the English starting point.

THE SCL PROTOCOL

The Society of Construction Law's Delay and Disruption Protocol (2nd edition, February 2017) adopts the same logic and expresses it as a core principle. The Protocol's position is that float "is not time for the exclusive use or benefit of either the Employer or the Contractor." It follows that an employer-risk event which merely erodes float does not, of itself, generate an extension of time.

The Protocol translates that principle into an operative rule: unless there is express provision to the contrary in the contract, where there is remaining total float in the programme at the time of an employer-risk event, an extension of time should be granted only to the extent that the employer delay is predicted to reduce the total float below zero — that is, to the extent it is predicted to delay completion beyond the contract completion date. In short: float first; an extension of time only once the float is gone and the completion date is genuinely threatened.

The Protocol is guidance, not law, and it is careful to subordinate itself to the contract. But its treatment of float is consistent with Ascon, and tribunals across common-law jurisdictions routinely treat it as a persuasive statement of good practice.

WHERE THE CONTRACT SPEAKS: NEC, AND WHERE IT DOES NOT

Both Ascon and the Protocol begin with the same caveat: this is the default, and it applies only where the contract is silent. Increasingly, contracts are not silent.

The NEC suite is the clearest example of express allocation, and it distinguishes sharply between categories of float. Time risk allowances — the contingencies a contractor builds into its activity durations for its own risks — are owned by the contractor; they are its protection and are not available to the employer. Terminal float — the gap between planned Completion and the Completion Date — is likewise reserved to the contractor. By contrast, the free float within the network is shared and consumed on a first-come, first-served basis. The practical effect is significant: under NEC a compensation event is assessed against the contractor's planned Completion, so an employer-risk event can move the planned completion date even where contractual Completion is not yet threatened — a materially more contractor-favourable position than the Ascon default.

Other widely used forms, including the FIDIC and JCT families, are generally silent on float ownership. They neither allocate float expressly nor displace the default. On those forms, the Ascon and SCL position — project-owned float, first come, first served — tends to govern, unless the parties have amended the conditions or the accepted programme to say otherwise. The lesson is straightforward but routinely overlooked: the first place to look for the answer to "who owns the float?" is the contract, not the case law.

WHY THE PROGRAMME USUALLY DECIDES THE ARGUMENT

In practice, float disputes are won and lost less on doctrine than on the integrity of the programme that generates the float in the first place. Float is an output of the network logic; if the logic is unreliable, the float figure is meaningless, and any argument built on it is fragile.

Several recurring problems undermine float in forensic analysis. Open-ended activities and missing logic links produce artificially high float that does not reflect how the works were actually sequenced. Excessive or undisclosed lags and constraints distort the critical path and can conceal float — or manufacture it. Time risk allowances buried invisibly inside durations create "hidden" float whose ownership cannot be tested. And a baseline that was never realistic, or never properly accepted, gives a tribunal little reason to rely on any float it shows.

This is why experienced practitioners insist on the discipline that the SCL Protocol, AACE International and the leading texts all advocate: a baseline built on transparent, justifiable logic and realistic durations; float that is identified, classified and — where it represents risk allowance — disclosed; and updates maintained honestly throughout the project. As Pickavance and others have long observed, the evidential weight of a float argument is only ever as good as the programme behind it.

CONCLUSION

"Who owns the float?" has a layered answer. As a matter of English law and SCL Protocol guidance, where the contract is silent the float belongs to the project and is shared first come, first served, so an employer delay generates an extension of time only once it has exhausted the float and begins to push out completion — the position established in Ascon v McAlpine and codified in the Protocol. But that default yields to the contract: forms such as NEC expressly reserve time risk allowances and terminal float to the contractor and change the outcome entirely.

For contractors and employers alike, the practical conclusions are the same. Read the contract first; it may already have decided the question. Build and maintain a programme whose logic and float can withstand scrutiny, because float arguments live or die on the credibility of the baseline. And address float ownership expressly at the outset, rather than leaving it to be fought over, with hindsight, when the completion date is already at risk.

REFERENCES

1. Society of Construction Law, Delay and Disruption Protocol, 2nd edition (February 2017) — core principle and guidance on float.
2. Ascon Contracting Ltd v Alfred McAlpine Construction Isle of Man Ltd (1999) 66 Con LR 119 (HHJ Hicks QC).
3. NEC4 Engineering and Construction Contract; CECA, NEC4 Bulletin No. 10, "Float: Types & Ownership" (2021).
4. AACE International, Recommended Practice No. 29R-03, Forensic Schedule Analysis.
5. K. Pickavance, Delay and Disruption in Construction Contracts (4th edition).
6. J. Keane & A. Caletka, Delay Analysis in Construction Contracts (2nd edition).