Prolongation and delay costs in Canada

The Canadian delay-quantum corpus is old, provincial and largely unreviewed on appeal. Outside Quebec it is built on 1980s British Columbia and Alberta trial decisions, one 1964 Supreme Court case and one 1978 Ontario case. There are exactly two authoritative Canadian data points on head-office overhead, they are 1964 and 1978, and they use different methods. A practitioner arriving from England or the United States should expect fewer rules and more room. Canadian courts have not closed off routes that English law closed off decades ago, and have not adopted the American apparatus either.

Quebec is carved out of everything below. It is a civil-law jurisdiction with its own contract form, its own prescription regime and no doctrine of concurrent delay or time at large, do not extend any proposition on this page across that border. Start instead at the Quebec extension-of-time hub. For the mechanism stripped of jurisdiction, see the neutral prolongation and delay costs page. For entitlement, notice and concurrency upstream of quantum, see extension of time in Canada.

The heads of delay cost, and which are contractual and which are damages

Two routes, and they are not interchangeable. Under CCDC 2 – 2020 GC 6.5.1 the Contractor is reimbursed "reasonable costs incurred by the Contractor as the result of such delay" where the Owner or Consultant caused the delay contrary to the Contract Documents. GC 6.5.2 does the same for a stop-work order by a court or public authority. GC 6.5.3 gives time only and no money. Everything outside those three limbs is a common-law damages claim for breach.

Note how narrow GC 6.5.1 is. The delay must be contrary to the Contract Documents, not merely attributable to the Owner. That is narrower than JCT's "any impediment, prevention or default" and narrower than FIDIC Sub-Clause 8.5(e). A late instruction that the Owner was contractually entitled to give may extend time without opening the money limb at all. In which case the claimant is thrown back on breach, and must find the breach. GC 6.5.4 requires written notice within 10 Working Days of the commencement of the delay (not of its effect) and Canadian courts enforce notice provisions strictly, without requiring the owner to prove prejudice.

Layer: jurisdictional, and in three rows a recorded negative finding. The rows rest on secondary sources. No Canadian judgment and no consolidated statutory text was read in original form for this corpus.

One structural point that changes the answer before any of this is reached. There is no statutory prohibition on no-damage-for-delay clauses anywhere in Canada, federal, provincial or territorial. This is the sharpest single contrast with the United States, where a substantial minority of states void or limit such clauses by statute. A Canadian no-damage-for-delay clause is attacked, if at all, through the three-stage exclusion-clause framework in Tercon Contractors Ltd v British Columbia 2010 SCC 4. Read the supplementary conditions before the general conditions: supplementary conditions almost universally amend GC 6.5.

Extended site overhead, and how a Canadian court wants it proved

Extended site overhead (site supervision, site accommodation, temporary works, plant kept on site through the extended period) is recoverable in principle in Canada as a reasonable cost of the delay under CCDC 2 – 2020 GC 6.5.1, or as damages. No Canadian source gives published rates for it, none for equipment standby and none for financing cost. There is no Canadian table to point at and no judicially blessed percentage.

That absence dictates the method. A Canadian claimant proves extended site overhead from its own records: the resources that were actually on site, for the days they were actually there, at rates traceable to payroll, invoices and plant registers. The evidential standard is the one set in Schindler Elevator Corp v Walsh Construction Company of Canada 2021 ONSC 283, the postponement must be broken into its component parts so that time, responsibility and cost can be apportioned. A lump claim for "site overhead over the delay period" invites exactly the objection that decomposition was never done.

The corollary is that the period must be the critically delayed period, not the whole overrun. The Canadian authorities indicate that courts apportion where multiple parties contribute overlapping delay, rather than applying the English rule of full relief for concurrency. An inference from the decisions, not a proposition any Canadian court has stated in those terms. So a claimant who cannot separate its own delay from the owner's is likely to see the recoverable period reduced rather than the claim dismissed. Where the contract is cost-plus, note also the record-preservation burden discussed in Ace Burger v G & I Construction 2025 SKCA 82, a single-source and therefore UNVERIFIED authority, but the only recent Canadian appellate statement located on the point.

Unabsorbed head-office overhead. The two Canadian data points, corrected

Canada has exactly two authoritative decisions on unabsorbed head-office overhead, and both are commonly mis-stated. Shore & Horwitz Construction Co Ltd v Franki of Canada Ltd [1964] SCR 589 is Supreme Court of Canada authority for recovery at a rate per unit of time. But computed from actual cost, not from a formula. Ellis-Don Ltd v Parking Authority of Toronto (1978), 28 BLR 98 (Ont HCJ) applied a tender-percentage formula, and it is an overhead case, not a disruption case.

Take the corrections one at a time, because both are circulating in Canadian practice in the wrong form.

The "Shore & Horwitz formula" is not the Supreme Court's holding. The circulated formula (damages = (overhead and profit × days of delay) ÷ days of contract duration) is a practitioner gloss. What the Court actually did was award overhead on actual audited cost: monthly overhead cost multiplied by months of delay, with head-office overhead assessed at 4.99% of other, out-of-pocket costs. Two independent sources agree on this, and neither contains a contract-value-over-duration ratio.

Shore & Horwitz is Supreme Court of Canada authority for recovering unabsorbed head-office overhead at a rate per unit of time, computed from the contractor's actual overhead cost over the contract period. And not authority for a contract-value ÷ duration formula. Presenting the formula as the Court's holding is a citation risk.

Ellis-Don is not a disruption or loss-of-productivity authority. The citation is right. The characterisation attached to it in circulating commentary is wrong. It is an unabsorbed head-office overhead case. There is no measured mile in Ellis-Don, no loss-of-productivity finding, and no general statement of the law of disruption. What it contains is arithmetic.

Layer: jurisdictional. One Supreme Court of Canada decision and one Ontario first-instance decision. The 1978 figures are 1978 dollars. Both rows rest on secondary sources. No Canadian judgment was read in original form for this corpus.

There is a CONFLICT on which named formula Ellis-Don applied. One source calls it Emden. A percentage derived from historical company-wide performance, another calls it Hudson, and quotes the judge quoting Hudson's assumption of "no change in the market, so that work of at least the same general level of profitability would have been available … at the end of the contract period". Hudson is the more likely of the two, because the 3.87% came from the contractor's own bid, which is the defining Hudson input, whereas Emden derives its percentage from audited company-wide accounts. The safe way to plead it is as "the tender-percentage (Hudson-type) formula applied in Ellis-Don", which sidesteps the label. Tell a Canadian court that the leading case applied Emden and opposing counsel with the report will say Hudson.

What a claimant should actually run: lead with actual cost, because that is the Supreme Court's method, and offer a formula only as a cross-check. That ordering is not a stylistic preference. It is the difference between a method with binding domestic pedigree and a method whose Canadian pedigree is a gloss.

Formulae in Canada: Hudson, Emden, and the one with no Canadian pedigree

Hudson and Emden both have a Canadian footprint. No Canadian judgment applies the Eichleay formula. Eichleay circulates in Canadian consultancy practice and is criticised in the Canadian claims literature, but it has no Canadian judicial pedigree at all, which matters if an expert report built on it is put before a Canadian judge who has never seen it applied.

The Canadian critique of each formula is worth knowing in both directions, because it is the cross-examination.

Layer: illustrative. This table records the critique found in the Canadian practitioner literature, not judicial holdings. The Eichleay row also records a negative finding. Sources are secondary.

The practical consequence is a hierarchy. Actual audited cost first, on Shore & Horwitz. A tender-percentage cross-check second, on Ellis-Don, pleaded without committing to a formula name. Eichleay only if there is a reason to think the tribunal is US-trained, and never as the primary method in a Canadian court.

Total cost, modified total cost and the measured mile

Canada is more receptive to total cost than England is. In W A Stephenson Construction (Western) Ltd v Metro Canada Ltd (1987), 27 CLR 113, the total cost method was accepted, on a strict three-part proof. There is a CONFLICT on the level of court in that litigation. An unreported Court of Appeal decision of 20 March 1986 and the 1987 report both exist in the same style of cause, so there are at least two decisions and the 1987 report may be a subsequent instance. Establish the level of court before citing it. This page deliberately does not state one.

The three limbs the contractor must prove are:

1. its bid price was reasonable,
2. its actual cost of completing construction was reasonable, and
3. the increase in cost was due to delay for which the owner was responsible.

Two negatives govern the rest of this section and neither should be hedged. There is no Canadian modified-total-cost authority, total cost adjusted for contractor-caused and bid-error components is standard consultant practice with no Canadian judicial endorsement. And there is no Canadian measured-mile authority. A disruption claim in Canada is therefore built without the method that carries a productivity claim in the United States, which puts the weight back on daily labour, plant and materials records. See the neutral delay analysis methods page for what the measured mile requires where it is available.

A third negative closes off the composite route. No Canadian judgment uses the phrase "global claim", and there is no Canadian analogue to Walter Lilly or John Doyle v Laing. A Canadian global claim is therefore argued on English and Scottish authority by persuasion only, before a bench with no domestic framework. In practice the total-cost route accepted in W A Stephenson is the better-supported Canadian vehicle for a composite claim, precisely because it is domestic.

Acceleration and its costs

Constructive acceleration is recognised in Canadian law, but thinly, and the leading authority is Golden Hill Ventures Ltd v Kemess Mines Inc 2002 BCSC 1460, described as the primary Canadian case explicating both actual and constructive acceleration. The Canadian practitioner formulation has three elements: an excusable delay exists. The owner refuses to accept it and presses for adherence to the original completion date. And the contractor expends additional resources without an explicit acceleration order. Note what is missing against the fuller US formulation. There is no "timely EOT request" limb.

Morrison-Knudsen Co v British Columbia Hydro & Power Authority is useful for the definition (acceleration as "speeding up the work" or "increasing the rate of performance") and for nothing else. No citation for it was obtained, and none is printed here. The commonly repeated court and year could not be confirmed. Use the definition and attribute it loosely, do not put a citation or a court in a pleading until it has been checked against the reporter.

Acceleration costs were awarded in Walsh Construction Company of Canada v Toronto Transit Commission 2024 ONSC 2782, where the owner caused delay through design delays and repeated changes. One further proposition attributed to that judgment. That a valid contractor's effort to accelerate, supported by contemporaneous records, is sufficient to establish constructive acceleration even absent a direction to accelerate, is UNVERIFIED and rests on a single source. If it is right, Walsh is the most recent and most important Canadian constructive-acceleration authority. Verify it before deploying it.

Three practical points on running the claim. First, lead with mitigation: acceleration is independently recoverable as a reasonable response to owner-caused delay, which does not depend on an underdeveloped doctrine. The contractor needs the owner's breach and the reasonableness of its response, not the constructive-acceleration elements. Second, plead constructive acceleration in the alternative on Golden Hill. Third, where the owner refused or sat on a valid extension application to force acceleration, consider the good-faith limb: an owner administering an extension-of-time clause is exercising a contractual discretion, which after Wastech Services Ltd v Greater Vancouver Sewerage and Drainage District 2021 SCC 7 must be exercised reasonably and consistently with the purpose for which it was conferred. That argument has no English equivalent, and UNVERIFIED. No Canadian case deciding it on an extension-of-time refusal was located.

On the contract mechanics, CCDC 2 – 2020 contains no dedicated acceleration clause. Directed acceleration is priced as a Change in the Work under GC 6.1/6.2, or, failing agreement on price, as a Change Directive under GC 6.3 valued on the GC 6.3 rules. That routing is UNVERIFIED. It is an inference from the form's architecture rather than a sourced rule, and no bulletin stating it expressly was located. Treat it as reasoning, not as a rule. And note for Ontario specifically: there is no Ontario decision on acceleration, or on disruption and loss of productivity as a distinct head, in the period 2022–2026.

Records and expert evidence, what actually wins

Contemporaneous records decide Canadian delay-quantum cases, and the delay expert must clear an admissibility threshold before any of it is heard. White Burgess Langille Inman v Abbott and Haliburton Co Ltd 2015 SCC 23 puts independence, impartiality and absence of bias into the threshold. A question at the qualification stage, not merely a matter of weight. The expert's duty is to give "fair, objective, and non-partisan opinion evidence". The threshold is low: exclusion follows only where the expert is unable or unwilling to comply, and residual concerns go to weight.

That bites harder on delay experts than on most, because the normal Canadian pattern is a delay consultant who worked on the project during the works, which exposes the expert at the threshold rather than only on weight. In Walsh v TTC the bias challenge nevertheless failed: the judge rejected it despite the expert's prior retention by the contractor during the project, and treated the decisive factor as whether compensation was tied to the success of the claim. Prior project involvement, repeat retainers, providing analyses during construction and direct client communication did not equal bias. It is worth flagging in argument that the leading Canadian delay-expert case contains no express reference to White Burgess. A real doctrinal gap, though that absence is itself UNVERIFIED and rests on consistency across the sources reviewed rather than on a reading of the judgment.

The records that carried the quantum findings in the Canadian cases are specific, and a claim file should be built to produce them from mobilisation:

• a schedule updated at every change, not at every quarter,
• dated correspondence tying each delaying event to a named activity, which is Schindler's enumeration: work tickets, emails, incidents, shop drawings, safety shutdowns, monthly reports and schedules with clear dates and operational connections,
• an RFI and submittal log with response times, because the owner's own turnaround is provable and was decisive in Walsh v TTC, where the owner's own consultant had found the average RFI response time excessively high,
• daily labour, plant and materials records, which are the only route to a disruption claim in the absence of any Canadian measured-mile authority,
• schedule data at the level actually used. In Walsh v TTC the accepted expert divided the project into 48 windows and reviewed more than 780 owner-related change conditions in Primavera, on a CAD 165.925M contract, and the court found 1,047 compensable days against the owner's conceded 411.

One defensive point that is the defendant's trap rather than the claimant's: never run a critique-only defence. Because the owner's expert in Walsh v TTC offered only critique, the judge was left with a single opinion on compensable delay quantum and held he could not determine it himself without guesswork. The result was a binary outcome at 1,047 days rather than 411. Compare the general position on records at the neutral records and evidence page.

What the lien will and will not secure

A construction lien is not a security for delay damages. In Ontario the line is price versus damages: prolongation costs are lienable to the extent they are the price of extended supply (extended equipment, extended labour caused by owner default) and are not lienable to the extent they are damages, which is where overhead absorption, lost opportunity, lost profit on other work and tort damages fall. Equipment standby is lienable in Ontario, head-office and on-site administrative overhead generally is not. That is an Ontario line, and it should not be stated as the Canadian one.

Alberta asks a different question and has not settled the answer. The Alberta test is proximity to the liened work, not the character of the sum: in Krupp Canada Inc v JV Driver Projects Inc, 2014 ABQB 259 damages including delay damages form part of the lien only where they "relate directly to the work that is the subject of the lien". So lost productivity on the liened project may be secured while inability to work elsewhere is not. CONFLICT on what Krupp decided: one Alberta reading has it holding delay damages can be lienable, another indexes the same judgment for the opposite headline. PME Inc v Enerkem Alberta Biofuels LP, 2021 ABQB 889 is later and narrower, confining the lien to conditionally admitted holdback and invoice amounts and excluding quantum meruit. And it does not cite Krupp. There is no appellate resolution, and the Alberta line is factual rather than categorical, which makes it harder to dispose of on a lien-reduction application. Both decisions predate 29 August 2022 and are cited to the Builders' Lien Act, RSA 2000, c B-7, not to the renamed c P-26.4. A citator query filtered to the new chapter loses both.

That distinction should drive how the claim is pleaded and split. A claimant with a mixed prolongation claim gains nothing by asserting a lien over the whole of it, and may lose credibility on the part that is plainly damages. The better course is to identify the extended-supply component, secure that, and pursue the balance as a money claim in the same action. The provincial rules vary, and so does the test itself as well as the preservation and perfection clocks. The position by province, including the Ontario price-versus-damages line, the Alberta directly-related test and their authorities, is set out on the delay regimes by Canadian province page.

Two cautions. First, no post-2018 Ontario appellate decision on the lienability of delay or prolongation costs was located, and there is no Alberta appellate resolution of Krupp and PME either. The governing authority in both provinces is first-instance, which is a real doctrinal thinness for a question that decides whether a contractor holds security or an unsecured claim. Second, the lien analysis is independent of the contractual analysis: a cost recoverable under CCDC 2 – 2020 GC 6.5.1 as a "reasonable cost incurred as the result of the delay" is not thereby lienable, and a cost that is lienable is not thereby recoverable under GC 6.5.1. Run both tests separately, and preserve the lien on the timetable of the province where the improvement is, not on the timetable of the claim.

À propos de ce contenu Le contenu de la bibliothèque est une information générale sur la pratique des réclamations de construction, et non un avis juridique. Le droit à réclamation, les délais et la procédure sont régis par votre propre contrat et par le droit du lieu où les travaux sont exécutés. Révisé le 8 août 2026 · Signaler une correction