Litigation — delay taxonomy, claim causes and forensic analysis: 16-Civ-B8 Management of Construction, December 2019, Q2

Question 2: Litigation — delay taxonomy, claim causes and forensic analysis (20 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

Delay is the single largest source of construction disputes in Canada, and almost every argument about it is really an argument about two questions in sequence: who bears the risk of the event, and what did the event actually do to the completion date. The classification below answers the first; the forensic analysis at the end answers the second. The two must be kept apart, because a delay can be perfectly excusable and still worth nothing if it never touched the critical path.

Excusable and non-excusable delays

A delay is excusable when its cause lies outside the contractor's control and outside its contractual risk allocation. Owner-directed changes, late or defective owner-supplied information, differing site conditions, unusually severe weather beyond the contract's stated threshold, strikes, and acts of a public authority are the standard examples; CCDC 2 clause 6.5 (Delays) lists substantially this set. The consequence of an excusable delay is an extension of the contract time, so the contractor is relieved of liquidated damages for the extended period. A delay is non-excusable when it flows from something the contractor took responsibility for — under-resourcing, poor sequencing, defective work requiring rework, late procurement, subcontractor default. The contractor gets no time and remains liable for liquidated damages, and it must generally accelerate at its own cost if it wishes to recover the date.

The practical test is not fault in the moral sense but risk allocation in the contract. Rain in Vancouver in November is foreseeable and priced; rain of a severity outside the historical record for that month is not. The contract's weather clause, not the contractor's disappointment, decides which one it was.

Compensable and non-compensable delays

Excusability governs time; compensability governs money, and the two are not the same question. A compensable delay is an excusable delay caused by the owner or by someone for whom the owner is responsible: a design change, a suspension order, late site access, delayed approvals, interference by the owner's other contractors. The contractor is entitled both to an extension of time and to its resulting costs — extended site overhead, idle plant, escalation, and in the right circumstances a share of head-office overhead computed by a recognised formula (Hudson, Emden or Eichleay). A non-compensable delay is excusable but not the owner's doing: severe weather, a general strike, an epidemic, a change in law. The risk is shared rather than transferred — the contractor gets time so that it does not pay liquidated damages, the owner gets a later building, and each absorbs its own costs.

Hence the familiar three-way outcome for any single delay event: time and money (owner-caused), time only (neutral event), or neither (contractor-caused).

Concurrent and non-concurrent delays

A non-concurrent delay is a single delaying event acting alone on the critical path over a period, so that cause and effect can be traced without competition. Concurrent delay arises when two or more independent delays, for which different parties are responsible, affect the completion date over the same or overlapping periods — classically an owner-caused delay running at the same time as a contractor-caused one. True concurrency, on the stricter definition adopted by the Society of Construction Law Protocol, requires that both delays be effective on the critical path at the same time; the looser “same period” test catches delays that were merely simultaneous, one of which had float and therefore did no harm.

Concurrency matters because it decides money. The generally applied outcome is that the contractor recovers an extension of time but not delay costs for the concurrent period: it should not be shielded from its own default, and it should not pay liquidated damages for a period the owner delayed as well. Where the two delays can genuinely be separated in time or in effect, an apportionment is made. Canadian courts have shown themselves willing to apportion on the evidence rather than apply a mechanical rule, which raises the value of contemporaneous records enormously.

Main reasons for delay-related claims

The recurring causes cluster into a few families. Design and information: incomplete or uncoordinated drawings at tender, a stream of changes after award, slow responses to requests for information and shop-drawing review. Site and access: differing subsurface conditions, contaminated soil, late expropriation or permitting, utilities not relocated, restricted access on urban or live-traffic sites. Owner administration: late decisions, late supply of owner-furnished equipment, slow payment forcing the contractor to slow production, interference by separate contractors. Contractor performance: unrealistic tender durations bought to win the job, understaffing, poor subcontractor management, rework. External: weather, labour disputes, supply-chain disruption, regulatory change. Underlying most of them is a common accelerant — a schedule that was never a real plan, so that when the events arrive there is no baseline against which to measure them.

Contractual modifications that reduce claims

Claims are cheapest to fight before they exist, and most of the useful measures are contractual rather than legal. Require a baseline CPM schedule submitted and accepted shortly after award, with monthly updates in native electronic format; make float an explicit project resource owned by the project rather than by whoever consumes it first, and state that fact. Impose a notice regime with a short, workable period and a defined content, so that events are recorded contemporaneously rather than reconstructed two years later. Provide a clear changes and valuation mechanism with pre-agreed rates for extra work, plus a defined process for constructive changes. Allocate risk deliberately: a differing site conditions clause with a proper geotechnical baseline report is far cheaper than the contingency contractors add when the risk is dumped on them, and a weather clause with a numerical threshold from Environment and Climate Change Canada data removes an entire class of argument. Add partnering and escalating dispute resolution — project-level negotiation, then a dispute review board or a project neutral, then adjudication or arbitration — and note that statutory prompt payment and adjudication regimes now in force in Ontario, Alberta, Saskatchewan and federally under the Federal Prompt Payment for Construction Work Act impose interim adjudication whether the parties like it or not. Finally, tender realistically: an owner who buys an impossible duration has purchased a claim.

Types of analysis needed to validate a claim

Entitlement is established by classification; quantum is established by schedule analysis. The recognised methods, in rising order of rigour and cost, are:

As-planned versus as-built. The baseline schedule is compared with what actually happened, activity by activity, and the differences are narrated. Simple, cheap and intuitive, but it demonstrates correlation rather than causation and is easily attacked where several events overlap.
Impacted as-planned. Delay events are inserted into the baseline schedule as new activities and the network is re-run. Prospective and simple, but it ignores the as-built reality and tends to overstate the contractor's case; it is generally the weakest method in a contested claim.
Collapsed as-built (“but-for”). The as-built schedule is built first, then the delay events are removed to see what the project would have achieved without them. Retrospective and evidence-based, but it depends on an as-built network whose logic is inferred after the fact.
Time impact analysis. Each delay event is inserted into the schedule update current at the moment it occurred, and the network is re-run to measure the incremental effect. This is the method the Society of Construction Law Protocol prefers for prospective assessment and the one arbitrators most often accept, because it respects the sequence in which events actually arrived, but it needs disciplined, regular updates to exist at all.
Windows or contemporaneous period analysis. The project is divided into periods, usually monthly, and the critical path and its movement are examined within each window. It is the most rigorous approach available for concurrent delay, because it shows which delay was actually driving the completion date in each window, and it is correspondingly the most expensive.

Whichever method is chosen, three supporting analyses are usually needed alongside it: a critical path and float analysis demonstrating that the delay actually affected the completion date rather than consuming float; a causation and concurrency analysis assigning responsibility for each period; and a quantum analysis converting delay days into money through measured extended overhead, disruption and loss-of-productivity claims, ideally by the measured-mile method rather than by a total-cost claim, which tribunals treat with justified suspicion. AACE International Recommended Practice 29R-03 catalogues these methods and the conditions under which each is defensible.

Topic. Construction delay: classification by risk allocation, the causes of delay claims, contractual measures that suppress them, and the forensic schedule methods used to prove or defeat them.

Key relations. The two-question framework — (1) is the event excusable, and if so is it compensable; (2) did it in fact extend the completion date. Extension of time follows excusability; delay costs follow compensability; both are worthless without critical-path causation, since $\Delta T=\max(0,\ \text{delay}-TF)$ applies to a delay event exactly as it applies to a planned activity.

Why this works. A construction contract is above all an allocation of risk, so the classification is not moral but contractual: it asks which party agreed to bear the consequence of this class of event. Separating time from money resolves what otherwise looks like an inconsistency, namely that a contractor can be excused from liquidated damages and still recover nothing, which is exactly the right outcome for a neutral event such as a strike. Requiring critical-path causation prevents the largest category of overstated claim, because a delay that consumes float costs the project nothing even though it is real, disruptive and expensive to the contractor's own operations. And insisting on contemporaneous records exists because memory and reconstruction both drift towards the interests of the party doing the remembering; a schedule update signed in month seven is evidence, whereas an as-built network drawn in year three is argument.

Common pitfall. Treating “excusable” as synonymous with “compensable”, which converts every weather day into a cost claim. Claiming delay for an event that consumed float rather than the critical path. And presenting a total-cost claim — final cost minus tender price — as though it proved damages, when it silently charges the owner for the contractor's own tender errors and inefficiency; the measured-mile comparison of an unimpacted period against an impacted one is the defensible alternative.

Source. Society of Construction Law, Delay and Disruption Protocol, 2nd ed. (2017), Core Principles 1–14; AACE International RP 29R-03, Forensic Schedule Analysis; CCDC 2 (2020), Part 6 (Changes in the Work) and cl. 6.5 (Delays); Halpin & Senior, Construction Management, 4th ed., Ch. 3 and 20.

Question 2 of 6: Litigation — delay taxonomy, claim causes and forensic analysis