Large construction disputes are regularly resolved privately through contractually-mandated or subsequently-agreed upon processes, including arbitration, dispute boards, mediation, and even structured negotiations. There are a number of reasons for this shift towards private dispute resolution processes, including cost effectiveness, court backlogs, construction expertise, and privacy (which reasons are not the subject of this article). However, due to the increased use of such processes, it has become less common for courts to be presented with complex delay and disruption claims, meaning that there are limited opportunities for them to offer new guidance on how such claims are viewed from the bench.

Fortunately, the Superior Court was recently presented with such an opportunity in Walsh Construction v. Toronto Transit Commission et al., 2024 ONSC 2782 (“Walsh”), and in response, has a offered a number of valuable takeaways for the construction industry and construction law practitioners, including counsel and arbitrators, regarding what can be expected in the resolution of a complex construction claim.

Notably, Walsh addresses a number of issues regularly found in complex construction disputes, including design issues, allegations of excessive scope changes, expert schedule and quantum analysis, flow-through of subcontractor claims, liquidated damages, and allegations of contract mis-administration, among many other issues.

Given the length of the Court’s analysis – totaling nearly 850 paragraphs – this case comment only covers a handful of the case’s key issues, rather than offering an exhaustive analysis.

Below, we provide an overview of the case, as well as outline the case’s key takeaways and key questions raised.

Factual Background

In 2011, the Toronto Transit Commission (the “TTC”) and Walsh Construction Company Canada (“Walsh”) entered into a design-bid-build contract for the construction of the Steeles West Subway Station, with a contract value of $165,925,000 inclusive of taxes. However, the contract value increased over time to a certified value of $223,267,711 (an increase of more than $57,300,000), of which amount Walsh had been paid $213,695,610 by the time of trial.

Substantial performance was originally scheduled to be achieved by November 5, 2014, while final completion was be achieved by February 4, 2015. In fact, substantial performance was not achieved until June 15, 2017 (953 days later than planned), while final completion was achieved on November 7, 2018.

TTC and Walsh disagreed as to the cause(s) of the delay and the apportionment of responsibility, with each side unsurprisingly attributing liability to the other. Walsh’s position was not an uncommon one – broadly speaking, it claimed that the TTC-provided design was incomplete when the contract was signed (notably, the TTC also third-party claimed against its designers), and that this design issue was then exacerbated by a lack of full site access and the TTC’s improper administration of the contract, which then created delay and consequently necessitated acceleration efforts. This in turn led to a multitude of change-related claims by Walsh. The TTC, on the other hand, was of the view that Walsh was fully responsible for the delay and all of the associated costs.

The situation ultimately culminated in litigation, in which context Walsh claimed for approximately $193,000,000. The amount of Walsh’s claim was comprised of a number of components commonly seen in large construction claims, including delay and disruption damages, acceleration costs, overhead, currency fluctuation, unpaid or underfunded change orders and change directives, bonding costs, subcontractor claims, and an unpaid Contract amount of $9,500,000.

The TTC counterclaimed for approximately $22,000,000 in liquidated damages associated with missed contractual milestones.

The Superior Court’s Decision

In introducing its reasons, the Court began by reflecting upon the significant amount of time and resources needed to administer the trial, including the following:

  • 161 trial days, consisting of a 1-day site view, 4 days of opening submissions, 151 days of evidence (specifically, cross-examinations) and 5 days of closing submissions;
  • Over 50 witnesses;
  • Affidavits – used to deliver examinations-in-chief in place of viva voce evidence, as is commonly the case in large construction disputes – which were hundreds of pages long, with thousands of pages of exhibits;
  • Over 1,400 additional exhibits marked during the course of the trial; and
  • Written closing submissions that were thousands of pages long, with several thousand attachments appended thereto.

The trial of the matter was therefore, to say the least, a Herculean effort by the Court and counsel. In that regard, the Court’s judgment canvassed a host of issues; for the purpose of this case comment, we focus on two of them – the delay analysis (including the expert evidence provided in respect of delay), and the claims of Walsh’s subcontractors. We consider each issue under separate headings below.

Delay Analysis

On the issue of delay, each party attributed all of the delay to the other party. Somewhat more unusual, though, was the binary manner in which the issue was framed: on the one hand, Walsh took the position that the Court could find any number of days of TTC-caused delay other than the amount Walsh claimed; however, the Court observed that there was no basis to do so on the evidence before it. On the other hand, the TTC argued that the Court could only choose between the amount of days claimed by Walsh (being 1,047 days), or the amount of days acknowledged by the TTC as its responsibility (being 411 days). As a result, the Court determined that it could only choose between these two options as presented by the TTC.

The Court observed that the design was not completed by the time Walsh’s trades were required to start construction, and that the TTC’s witness acknowledged in cross-examination that the initial design was not constructible. According to Walsh’s design expert witness, the volume of changes (e.g. through change orders, change directives, RFIs, etc.) remained excessively high over the life of the project, and according to a third-party consultant report commissioned by the TTC during the project, the average time for responding to RFIs was also excessively high, with many changes attributable to the incomplete design. On its face, this does not appear surprising to these authors, as early design issues (whether a design is incomplete, inadequate or otherwise problematic) will typically require extensive changes (and such changes will give rise to lengthy RFI exchanges).

For its part, however, the TTC argued that it was “not surprising that on a project of this size and complexity, there would be issues leading to cost and time overruns”, and that as a result, both it and Walsh should bear responsibility. However, the Court did not accept this argument, given that the TTC bore the sole responsibility for issues arising from the design.

Among many other findings as to the causes of delay, the Court also observed that Walsh was entitled to (but deprived of) possession of the entire construction site, that such deprivation in turn deprived Walsh of sufficient lay-down areas and made available areas more crowded, and that increased crowding of such areas can (and did) lead to inefficiencies and resultant delays. As a result, the Court determined that several significant causes of delay were attributable to the TTC.

With respect to determining the amount of compensable delay, the Court was required to consider the evidence of Walsh’s delay expert, as atypically, the TTC did not put forward a delay analysis of its own (but rather provided expert evidence critiquing Walsh’s schedule analysis).

In that regard, the TTC abandoned an attempt to disqualify Walsh’s expert, instead arguing that Walsh’s expert’s evidence could be admitted but should be given no weight. On this point, the TTC argued that Walsh’s expert lacked independence and was biased, primarily on the basis that he had a history of being retained by Walsh (the expert had been involved in Walsh’s prior claims on this same project, although it was unclear if he had also been retained by Walsh on other projects).

The Court rejected this argument, confirming that it is not unusual for someone like Walsh’s expert to analyze work during construction, and to continue with this analysis if litigation ensued. In other words, the Court appears to have accepted that a claims consultant hired for a construction project to help with claims analysis during the project can potentially stay on as a delay expert through trial or other potential hearing (although the Court did not detail the precise pre-litigation scope of retainer for Walsh’s expert).

In any event, the TTC did not offer an expert delay analysis of its own, but rather offered expert evidence criticizing the methodology of Walsh’s expert – specifically, that Walsh’s analysis (1) did not insert fragnets for Walsh or subcontractor delay, (2) relied upon Walsh’s erroneous project schedules, (3) used “reverse logic”, (4) contained an excessive number of changes, (5) failed to account for concurrent delays, and as a result, ultimately failed to conduct a proper analysis. In response to this criticism, Walsh delivered further expert evidence affirming the propriety of the methodology. Given the lack of a competing delay analysis from the TTC, the Court’s analysis was therefore restricted to assessing the methodology of Walsh’s expert.

Ultimately, the Court rejected the TTC’s contention that because Walsh’s methodology departed from the AACE’s Recommended Practice 29R-03, it was therefore erroneous; to the contrary, the Recommended Practice allows for various adjustments and options, relies upon professional judgment and expert opinion, and usually requires subjective decisions by the expert. On the totality of the evidence, the analysis performed by Walsh’s expert was sufficiently credible that it was to be preferred over the TTC’s position.

Accordingly, the Court found the TTC responsible for the entirety of the 1,047 days of delay (bearing in mind that the Court was only presented with two options, as noted above). In this regard, the Court’s finding of full responsibility for delay may be facially  problematic insofar as it represents a departure from the usual practice of apportioning delay, but it was unavoidable in this case given the binary nature of the TTC’s approach. Presumably, Walsh’s proposed approach of apportioning delay could have yielded some relief to TTC.

Subcontractor Claims

Consistent with many large delay claims, Walsh’s claim was partially comprised of a flow-through of claims made by its subcontractors on the project; in this case, Walsh’s claim included its subcontractors’ delay costs and acceleration costs. As is also not uncommon, Walsh entered into agreements with its subcontractors which released Walsh from liability. These agreements took two forms:

  • assignment liquidating agreements, in respect of which Walsh paid an amount to the subcontractor that represented all of the contract payments due and payable to the subcontractor. In exchange for payment, the subcontractor assigned its claim to Walsh and released Walsh from liability; and
  • non-assignment liquidating agreements, in respect of which Walsh paid to the subcontractor an amount representing the payments due and payable to the subcontractor in exchange for a release (but the subcontractor did not assign its claim). If Walsh were able to recover anything from the TTC with respect to the subcontractor claims, Walsh would pay it to the subcontractor minus a percentage for overhead, markup and costs incurred. If there were a global recovery, the subcontractors would receive a pro-rated recovery as solely determined by Walsh.

The most salient aspect of these arrangements was, as noted above, the fact that Walsh was released from liability.

As a preliminary point, the TTC sought to have this aspect of the action stayed on the basis that Walsh’s agreements with the subcontractors transformed their relationships from adversarial ones into cooperative ones, thus fundamentally altering the litigation landscape and thereby requiring disclosure (pursuant to Handley Estate v DTE Industries Limited, 2018 ONCA 324 and its subsequent case law).

The Court rejected this argument, finding (among other things) that none of the subcontractors were parties to this action, and a stay can only apply to an agreement made between parties to the same proceeding. Notably, this appears to have been the first time this line of reasoning has been applied in the construction context, so it remains to be seen whether this point will be expanded upon in subsequent case law; at first blush, it would seem reasonable to conclude that a contractor settling with several of its subcontractors would change the dynamic of a proceeding (certainly from the contractor’s perspective), particularly given the complex web of claims that can arise on large construction projects.

On the substantive issue of whether Walsh could flow through these subcontractor claims, the Court determined that Walsh could not, and these claims were therefore dismissed. In reviewing the paucity of Canadian case law on the topic of flow-through claims, the Court observed that in order to sustain a flow-through claim, the contractor must have actual (or a least potential) liability to its subcontractors, which of course does not exist in circumstances where the contractor has obtained a release from its subtrades.

This was equally true in the US case of Severin v. United States, 99 Ct. Cl. 435 (1943), where the court held that a general contractor cannot flow through a subcontractor’s claim against an owner if the general contractor has no liability to the subcontractor for the damages at issue; this case led to the use of liquidating agreements, which is what Walsh had attempted to rely upon in this case, and which are considered under American law to be a valid means of flowing through subcontractor claims. In American practice, such agreements typically acknowledge that the contractor remains liable to the subcontractor for the subcontractor’s damages to the extent that the contractor prevails against the owner, thus creating a nexus of liability. It is not clear in Walsh how the liquidating agreements in question differed (if at all) from this practice, although the existence of the release appears to have been the most significant factor in this case.

In any event, the Court in Walsh rejected the validity of such claims, and determined that the liquidating agreements in this case could not overcome the fact that (1) Walsh had no liability to the subcontractors in question, and (2) the owner had no privity of contract with the subcontractors. In other words, Walsh took assignments of subcontractor claims that had no legal existence (in relation to the assignment liquidating agreements), while the other subcontractors (the parties to the non-assignment liquidating agreements) had no claims in contract against TTC.

Here, the Court concluded that Walsh could not flow through its claims as a result of the releases, which extinguished the claims of the subcontractors against Walsh – the entity with whom they were contracted – particularly in circumstances where Walsh “failed to scrutinize any of the subcontractor delay claims”; for those subcontractors with whom Walsh did not settle, the Court concluded that these were better dealt with as part of the subcontractor’s actions against Walsh, in respect of which Walsh had added the TTC via third-party claim.

Ultimately, in the result, the Court awarded judgment to Walsh in the aggregate amount of $52,160,563 (plus interest, bonding costs, and HST on certain items), while dismissing the TTC’s counterclaim in its entirety.


As noted above, Walsh offers a rare instance of a complex delay claim being litigated all the way through to judgment; as such, the Court’s judgment offers valuable guidance for parties advancing such claims in the construction context. That being said, the case also raises a number of questions in respect of the two issues considered above – some of which may need to await appeal for clarification.

With respect to the issue of expert witnesses and the manner in which their evidence is presented, the Court’s discussion of the schedule experts is instructive. First, and as noted above, the fact that only Walsh delivered an expert delay analysis (noting again that TTC only provided a critique) allowed Walsh to effectively frame the delay issue, which fact was emphasized by the further retainer by each party of additional experts to critique the methodology of Walsh’s schedule expert (or, in the case of Walsh’s secondary expert, to affirm the methodology of Walsh’s schedule expert).

In that regard, electing not to proffer a schedule analysis of one’s own may be a “high risk, high reward” endeavour since, as was the case here, such an approach might force a court or arbitrator into the binary choice of apportioning delay on an all-or-nothing basis. In our view, best practice is to take the default approach of proffering one’s own schedule analysis, subject to the possibility of exceptional circumstances dictating otherwise.

Here, it is worth considering that delay can be (and most often is) awarded on a spectrum. In other words, there could have been a middle ground finding of delay attributable to TTC – somewhere between 411 days and 1,047 days of delay that was not available to the TTC in the circumstances. While it is not clear from the decision as to why this approach were taken, the authors assume there was good cause in the circumstances.

In addition, the Court offered useful commentary on the issue of repeated retainers of the same expert by the same party. As readers will appreciate, it is not unusual in the construction industry for parties to appoint the same experts more than once over a period of several years. This is partly due to the fact that full-time expert witnesses (i.e. professionals who provide reports and/or testimony for a living) are common in the construction industry – particularly given the growth of large professional services claims firms – and partly due to the fact that some issues are so specialized that there are only a handful of experts in North America (or the world). Accordingly, it is not necessarily the case that repeated retainers of the same expert, in and of itself, should automatically disqualify an expert.

Furthermore, the fact that Walsh’s expert was not disqualified (and did not have his evidence discounted in weight) based on his involvement with Walsh’s prior claims submitted to the TTC on the same project is also consistent with industry practice on large construction projects. So long as the expert in question retains their independence, impartiality, and lack of bias over the course of their involvement (i.e. both before and during dispute resolution), then an extended familiarity with the project may be beneficial, insofar as it increases the expert’s knowledge of the project. On this note, it is important for counsel, experts and clients to remember the importance of maintaining the independence of experts and taking all steps necessary to do so – if the desired outcome is to rely on that expert for an independent analysis.

With respect to the issue of flowing through subcontractor claims, the result is more challenging insofar as it arguably creates a tension between the existing case law and the practicalities of managing large claims on construction projects.

In particular, the practicalities of large delay claims frequently dictate a different approach – particularly where the contractor is faced with dozens of subcontractor claims and the prospect of managing a tsunami of subcontractor litigation, all while attempting to motivate the subcontractors to finish building the project. In such circumstances, it is not uncommon for the contractor to resolve the subcontractor claims and then take an assignment of those claims, in order to streamline the overall litigation and put some funds into the pockets of the subcontractors notwithstanding the fact that payment still remains disputed as between owner and subcontractor. Such steps taken by contractors can be highly beneficial to both the contractor and the owner, given that they can potentially avoid large lien actions or other messy multi-party disputes.

Indeed, this latter point may be particularly important in the context of very large claims which take years to resolve. If contractors are disincentivized from settling with subcontractors for fear of being unable to pass those costs on to the owner (who the contractor believes is liable), then it might be the case that subcontractors go several years without receiving any payment, thus increasing the risk of their insolvency  and potentially having a negative effect on the construction of the project itself. In Walsh, for example, the action was commenced in 2017, seven years before judgment was rendered. It is entirely plausible, particularly in the case of smaller subcontractors, that seven years of non-payment could seriously impact their ability to carry on business.

In that regard, it therefore bears noting that one of the key purposes of the Construction Act is to ensure the flow of funds down the construction pyramid to subtrades, suppliers, and so forth. Accordingly, it would seem that Walsh exposes a more fundamental tension between the first principles of contractual liability in relation to flowing through subcontractor claims, and the first principles of the Construction Act. Ultimately, the issue of flowing through subcontractor claims is a challenging one with no clear answer at present.

Given that the parties experienced mixed success in Walsh, we await with interest to see if one or both parties appeal, and if so, the outcome of such appeal(s).

Learn More About Construction and Infrastructure Law

We are a preeminent Canadian construction and infrastructure law firm. Our peers and clients recognize our lawyers as the best in the construction industry.