Escalation Clauses That Survive Contact With a Claim

For most of the two decades before 2021, material price escalation was a risk the market was content to leave with the contractor. Movements were small enough and predictable enough that the premium for carrying them was modest, and a contractor who asked for an escalation clause was signalling weakness.

That assumption has not survived. The Canadian Construction Association's guidance on risk management in construction contracts now recommends including a price escalation clause for materials subject to inflation risk, and advises that contractors avoid assuming risk for changes in law where those changes could materially affect cost or schedule. Industry associations have publicly urged public buyers to include price adjustment provisions in tenders.

The principle the CCA applies is the orthodox one: allocate the risk to the party best able to manage it. Neither party manages a sovereign tariff decision. What that argues for is not a transfer of the risk in either direction but a mechanism for sharing it on defined terms — which is what a well-drafted escalation clause is.

Escalation provisions in Canadian construction contracts tend to be built from a small number of components, combined to taste. Understanding them as components rather than as whole clauses makes it much easier to see what a particular draft actually does.

A threshold or deadband sets a movement below which nothing happens — commonly expressed as a percentage change in a named price from a stated base date. Its purpose is to keep the mechanism from operating on noise. Its effect, which is less often noticed, is that the contractor carries every movement up to the threshold.

An index link ties the adjustment to a published series, so that neither party's own records determine the outcome. A named-commodity basket does the same thing for specific materials — steel, aluminum, copper, lumber — rather than for construction costs generally. A share mechanism splits movement above the threshold between the parties in stated proportions rather than passing it wholly to one. Caps and collars bound the total adjustment in either direction. And quotation validity provisions handle the narrower problem of a supplier price expiring between tender and award.

The component worth the most attention is the trigger, because it is the one most often drafted loosely. A clause that operates on "increases in the Contractor's material costs" requires the contractor to open its books and the owner to audit them, under time pressure, in a dispute. A clause that operates on a published index or a government-published tariff rate can be operated by both parties from public information on the day.

Index selection is treated as a technicality at drafting and becomes the whole argument later. Two points are worth settling before signature.

The first is scope. Statistics Canada's building construction price indexes are published separately for residential and non-residential work, and for individual census metropolitan areas. Those series have recently diverged sharply: in the second quarter of 2026, non-residential rose 1.4 per cent in the quarter while residential rose 0.5 per cent, and Toronto residential fell 0.8 per cent. A clause on an institutional project indexed to a residential or all-construction series will systematically under-compensate. This is not a small effect, and it is entirely avoidable by naming the right series.

The second is lag. Construction price indexes are published quarterly and some weeks after the period they describe. A clause that adjusts on the index alone therefore compensates for movement that has already been paid for, and it will not reflect a sudden regulatory change — a new tariff schedule, say — until the quarter after it bites. Where the risk being managed is a discrete policy event rather than gradual inflation, a trigger tied directly to the published rate in the relevant order is faster and more precise than any index.

In practice the better clauses use both: an index for general movement and a specific trigger for defined events.

This section is advisory rather than a statement of law. I am not aware of any Canadian authority addressing the point, and it should be treated as a drafting risk to be managed rather than a settled position.

The standard forms already contain a price adjustment mechanism for changes in taxes and duties — GC 10.1.2 of CCDC 2 and its equivalents — which operates automatically from bid closing. A bespoke escalation clause introduced by supplementary conditions operates on different terms: usually conditional, usually thresholded, often capped.

Two problems follow. Where the supplementary conditions contain a precedence clause, as most do, a court or adjudicator could read the bespoke escalation provision as the exhaustive code for price movement, displacing or capping the broader statutory-duties entitlement the standard form would otherwise have given. A contractor could end up worse off for having negotiated an escalation clause.

Conversely, where both provisions are engaged by the same cost increase, an owner has an obvious double-recovery argument. A tariff-driven rise in the price of steel may be caught by GC 10.1.2 as a change in duties and by the escalation clause as a movement in a commodity index, and nothing in either provision obviously resolves the overlap.

The remedy is a sentence, not a redesign. State expressly how the two interact: whether the escalation clause is in addition to or in substitution for the taxes-and-duties provision, and how a cost increase caught by both is to be treated. Parties who address it at drafting spend a sentence. Parties who do not may spend an adjudication.

An escalation clause in a head contract that is not mirrored in the subcontracts leaves the general contractor holding the difference on every subcontracted package — which, on most projects, is most of the work.

The mirroring has to be real rather than nominal. A head contract indexed to a non-residential series and subcontracts indexed to an all-construction series will drift apart. A head contract with a two per cent threshold and subcontracts with a five per cent threshold leaves a band of movement the general contractor absorbs on every package. Base dates that differ between the head contract and the subcontracts create the same effect, less visibly.

The subcontract forms in the CCDC family already carry a taxes-and-duties provision mirroring the head contract, which is one reason to prefer them to bespoke documents when the market is volatile. Where bespoke subcontracts are used, the escalation and duties provisions should be checked against the head contract line by line at award.

A well-drafted escalation clause still fails if the project cannot produce what it needs on the day.

What it needs is generally modest: the base date and base prices fixed and agreed in writing at award rather than reconstructed later; purchase orders and supplier quotations retained with their dates; country of origin and import documentation where a duties trigger is in play; and the index values recorded at each adjustment point rather than looked up retrospectively from a series that may since have been revised.

That last detail catches people out. Statistics Canada revises published series. A calculation performed in 2029 against today's figures may not reproduce a calculation performed today, and an owner in dispute will use whichever version favours it. Recording the values used at the time, in the contemporaneous record, removes the argument entirely.

The theme is the one that runs through every claim: the clause defines the entitlement, and the record determines what is actually recovered. A strong clause with a weak record is a negotiating position. A clear clause with a complete record is an arithmetic exercise, and arithmetic settles.