Lessons on expert evidence in defending construction delay claims from Walsh Construction v. TTC & Ors, 2024 ONSC 2782

Key Takeaways

• Delay claim defendants should obtain their own expert schedule analysis evidence. Mere critique of the plaintiff’s expert’s findings runs the risk of an all-or-nothing result.
• An expert’s involvement in the contractor’s prior claims on the same project is not incompatible with that expert’s independence before the court. It does not, on its own, give rise to bias.
• Experts that display dogmatism or stubbornness risk damaging the court’s perception of their credibility and independence.

The case

TTC hired Walsh to build the Steeles West Subway Station in September 2011. The contract price was $166 million with a scheduled substantial performance date of November 5, 2014, that is, 1,154 days from award of the contract.

To say the project was delayed is an understatement. Substantial performance was achieved 953 days late, on June 15, 2018. The revenue service date was also 1,047 days late, and contract completion was 1,372 days late.

TTC accepted that it caused 411 days of delay and certified an increase of $57 million to Walsh’s contract price, however, it refused to accept responsibility for the balance of 636 days of delay.

As is typical, each party blamed the other for the balance of the delay. Walsh claimed the delay was all caused by TTC. They claimed $193 million under 23 heads of damages, $19 million of that consisting of direct delay damages. TTC counterclaimed for $22 million in liquidated damages against Walsh for missing certain contractual milestones. TTC’s argument amounted to saying that delays are to be expected in a project of this size and complexity, and responsibility should be shared between it and Walsh.

The parties’ stances on how the judge should address the gulf between them was also at odds. Walsh took the position that the court could find the number of days of TTC-caused delay was between the 411 days which TTC admitted and the 1,047 days Walsh claimed. TTC went for all or nothing. They argued that the court could not find another number of days, it was 1,047 or 411, nothing in between.

With both sides pointing fingers, the question of who was truly at fault for the delays came down to each party’s expert testimonies.

Walsh’s evidence

Walsh relied primarily on the evidence of their delay analyst, Richard Ott. Mr. Ott had a history of being retained by Walsh, including during the project, when he prepared a time impact analysis used by Walsh in its claims for delay. Notably, there was no question on whether Mr. Ott’s fee was contingent upon success. Mr. Ott testified that he investigated over 780 TTC-related change conditions to see how each affected Walsh’s work and the critical path using the software, Primavera. He assumed Walsh was responsible for all delay unless the analysis showed that TTC was responsible. He reviewed data from the project documentation including RFIs, RFQs, CDs, NOICs, Walsh’s schedules, TTC’s schedules, correspondence, daily reports, and meeting minutes. He also testified that he looked for concurrencies. The result of his analysis was that Walsh was entitled to a total of 1,047 days of compensable time extension.

Walsh also retained a second expert who rebutted TTC’s criticism of Mr. Ott’s methodology.

TTC’s evidence

After initially seeking an order for the court to refuse Walsh’s expert’s qualifications and then backing down from that approach, TTC settled on challenging Mr. Ott’s independence. They argued that he was part of Walsh’s team since he had a history of being retained by Walsh.

Importantly, TTC did not retain an expert to perform its own delay analysis. Instead, TTC’s expert limited his opinion to critiquing Mr. Ott’s methodology, alleging that Mr. Ott’s delay analysis was faulty because Walsh’s project schedules were erroneous, Mr. Ott departed from the recommended practice, and there were logic errors in his analysis.

Justice Hood’s decision on delay

Justice Hood accepted Mr. Ott’s opinion that Walsh was entitled to 1,047 compensable days of delay; all the delay was TTC’s fault and all of it was compensable. The court awarded Walsh $58 million in damages (the shortfall being largely the result of the court’s rejection of flow throughs from subcontractors) and dismissed TTC’s counterclaim.

First – TTC’s allegations of bias. Justice Hood accepted that an expert’s bias could go to the weight of their testimony. But he did not consider the fact that Mr. Ott had been previously retained by Walsh to be determinative of bias. It was not unusual for an expert to analyze work during construction and continue with the analysis if litigation ensued.

Second – the substantive issues of liability for the delay and the quantum of the delay.

Justice Hood summarized the relevant construction law principles, as follows:

1. Delay is categorized as (1) excusable or non-excusable, and (2) compensable or non-compensable. Non-excusable delay is delay for which the Contractor is not entitled to any time extension or compensation because it is a delay within its control. Excusable delay is generally viewed as delay that is beyond the Contractor’s control and for which it may be entitled to compensation.
2. Concurrent delay for which the Contractor is responsible would make excusable delay non-compensable.
3. It’s important to consider whether the delay is on the project’s “critical path”: the series of connected tasks that define the minimum overall duration for completion of a project, also known as the “longest path”.
4. For the Contractor to be compensated for delay, it must prove on a balance of probabilities that the delay was the sole responsibility of the Owner, was on the critical path, and without concurrent delay. (This was TTC’s argument, apparently undisputed by Walsh or the court.)

The court held that the delays were TTC’s responsibility because they arose out of design issues, and design was TTC’s responsibility. Contrary to what was said by TTC at the time of awarding the contract, the design was not complete. This resulted in a very large number of Requests for Information (“RFIs”) and Change Directives as well as excessive delays by TTC in responding to the RFIs. Justice Hood was of the view that the large number of RFIs were a telltale sign that the delays originated from TTC, not Walsh. Justice Hood also decided that TTC’s position, that it was only responsible for 411 days of delay, was based on some faulty analysis, as had been acknowledged in cross-examination.

Having accepted that the delay was the TTC’s fault, Justice Hood turned next to consider the number of compensable days of delay. According to Justice Hood, the Court was in “no position to determine the amount of compensable delay on its own; it does not have the expertise to do so.” [86] In a massive project like this there were a multitude of moving parts. Delay events might be closely intertwined, overlapping, concurrent or not concurrent, excusable or non-excusable, and on or off the critical path. The court needed the assistance of delay experts to quantify the number of days of delay for which a party might be entitled to compensation.

The decision hinged on whether the court should accept Mr. Ott’s evidence or not. Justice Hood stated plainly that, “I am left with only one expert opinion as to the amount of compensable delay, with two conflicting expert opinions as to whether the methodology used by the first expert was appropriate.” [123] Justice Hood lamented that the differing experts from both sides were eminently qualified and delay analysis methodology is “completely foreign” to him, yet he was tasked with “having to conclude whose opinion to accept on matters that these three experts have spent their life studying.” [130]

In deciding which expert’s evidence to prefer, he compared each expert’s attitude and approach to answering during cross-examination. He contrasted the patience and straightforwardness of Walsh’s expert, Mr. Ott, with TTC’s expert’s dogmatism and resistance to being challenged. He concluded that TTC’s expert was more prone to advocacy, and so he preferred Mr. Ott.

Lastly, Justice Hood reiterated that Mr. Ott was the only expert who actually performed a delay impact analysis and provided an actual opinion on the compensable time extension. “If TTC had presented its own delay impact analysis, then I would have been placed in a position of having to choose between the two or arriving at a different number altogether,” he noted, “[but] I was given a binary choice between 1,047 days and 411 days.” [132] Faced with that binary choice, and having accepted Mr. Ott’s evidence despite TTC’s expert’s criticisms, he found that the total number of days of compensable delay was 1,047 days. In his view, he had no basis to find another number of days on the evidence.

Commentary

First, when it comes to assessing bias, courts are cognizant that the pool of experts is often small. If the fact that a litigation expert was retained to consult during the project were to lead to automatic disqualification of that expert, litigants would encounter difficulties in finding independent and separate experts for claims and for litigation. Indeed, there may be practical advantages to having an expert who provided analysis during the currency of the project, also provide evidence before the court. As long as the expert can maintain their independence, their familiarity with the project can be an asset to the court and may be more economical to the parties than retaining a forensic expert who must learn the project and familiarize themself with documents and data from scratch. As Justice Hood opined, the nature of the retainer, in particular if compensation was tied to the success of the claim, is a stronger indicator than the mere fact of a previous retainer.

Second, experts must be independent witnesses, not advocates for their clients’ case.  Particularly for delay claims, where the requisite technical expertise is so far outside the expertise of the Court, an expert’s credibility becomes paramount. That credibility is not just a function of the expert’s written report, its readability, and coherence, but also the expert’s attitude and demeanor during cross-examination, just like any other witness. In other words, an expert’s credibility is not premised exclusively on their background, experience, and academic and professional credentials, but how they present themselves as honest individuals. Thus, an expert’s willingness to objectively consider other views and to adjust their own views in the face of contrary evidence is a signal to the court of honesty and integrity that the court will weigh heavily.

Third, Justice Hood’s decision is a cautionary tale to delay claim respondents everywhere. In the face of complex disputes and even more complicated delay analyses, a judge’s fact-finding role may be reduced to choosing between the opinions of competing experts. Thus, a defence strategy that seeks to play “spoiler” only carries significant risk. Yet, critical path delay analysis for complex projects is typically a very expensive undertaking. Is forcing a respondent to undertake a full-blown delay analysis fair? In a system that places the burden of proof squarely on the claimant, is it fair to require the respondent to prove an alternative quantification? Arguably, that is akin to requiring the respondent to prove the claimant’s case, or at the very least, fix the respondent’s mistakes.

While it has always been the respondent’s prerogative to simply cast doubt in its own defence, such an approach may be too simple for complex delay claims. Rarely is delay to large projects black or white. But once the Court is satisfied there is some compensable delay, in other words, that liability has been established, it has an obligation to “do its best” to quantify the damages, i.e. the number of days of delay.

In Perini Pacific Ltd. v. Greater Vancouver Sewerage and Drainage District, [1967] SCR 189, the Supreme Court established, long ago, that a judge must make a reasonable effort to quantify damages, even when a precise calculation is hard, holding that damages should not be denied simply because they are difficult to measure. Instead, the judge should use the best available evidence to make an informed estimate. In Justice Hood’s defence, Mr. Ott’s calculation was not just the best available evidence, it was the only available evidence.

Given the dearth in Canadian jurisprudence on construction delay claims, Walsh v. TTC is a welcome contribution. Moving forward, Justice Hood’s decision provides important guidance for plaintiffs and defendants both.