Case Commentary: MDS Inc. v. Factory Mutual Insurance Company; Future Considerations for Insurance Coverage

May 2020

Rory Barnable, B.A., LL.B., LL.M., Barnable Law P.C.1


The COVID-19 pandemic we are currently experiencing has spawned numerous discussions ranging from conspiracy theories as to virus origin, and the use of disinfectant to ward off illness, to less controversial topics such as appropriate government response, economic incentives, contractual breach and force majeure clauses. Particularly early in the pandemic much discussion involved whether there could be insurance coverage for businesses’ economic losses.

Some policyholders had pandemic coverage written into their policies, while others had policies specifically referencing viruses and shutdowns due to order of civil authority. Many articles reviewed the application of such insurance wordings to the current pandemic-induced business losses, and speculated how coverage could link to shutdowns due to property loss, communicable disease and interruption by civil authority. Often, the necessary link to insurance coverage stumbled over the absence of property damage, as the relevant policies often tie the coverage to some manner of property loss.

At the end of March, the Ontario Superior Court of Justice released MDS Inc. v. Factory Mutual Insurance Company, 2020 ONSC 1924, a trial decision which many authors then subsequently linked to the COVID-19 pandemic. The link arises due to the decision’s commentary on resulting physical damage and how that phrase may be applied in all-risks insurance policies. However, readers should approach the case and related commentary with some caution. While the case does provide commentary on resulting physical damage in the insurance context, it is not a case about the pandemic.

As well, as we expand upon below, the decision’s commentary on physical damage may not be the most impactful development from the case. Rather, for future litigating policyholders and insurers the more meaningful aspect of the decision may be its discussion concerning interest — and the Court’s recognition that common approaches to prejudgment interest (1) often fail to compensate an insured’s true costs due to an insurer’s denial of a covered claim; and (2) potentially reward an insurer who prudently invests what is ultimately an insured’s due indemnity.

The decision is lengthy and complicated, and it is under appeal. Therefore, perhaps all we may conclude now about MDS v. FM is that it is a comprehensive examination of insurance coverage in an interesting factual scenario, with at least one more chapter yet to be written.


MDS Inc. and MDS (Canada) Inc. c.o.b. MDS Nordion (collectively, “MDS”) is a health science company that processes radioactive isotopes and sells them for medical use. MDS purchased isotopes from Atomic Energy of Canada Limited’s nuclear research universal reactor (“reactor”), located in Chalk River, Ontario.

FM insured MDS through a worldwide all-risks policy, covering losses from all risks of physical loss or damage except as excluded. The policy included Contingent Time Element (“CTE”) coverage to a limit of $25,000,000. The CTE coverage is a type of business interruption insurance covering if there was an interruption in the business of a supplier affecting MDS’s profits. It covers loss of profits if a supplier is unable to furnish product.

In May, 2009 the reactor suffered an unexpected shutdown. The reactor was initially shut down for thirty-six hours, which then extended to fifteen months. The reactor was found to have a leak of heavy water within it. The leak was eventually attributed to unexpected corrosion at specific locations. The shut down limited MDS’s isotope supply, impacting MDS’s sales. Some readers may remember the shutdown was newsworthy in 2009 as the reactor supplied a high percentage of the world’s medical isotopes.

MDS sent a claim for loss of profits to FM on May 21, 2009. FM denied coverage on August 4, 2009. MDS sued FM in October 2010. Many facts were agreed prior to or at trial, including that MDS’s loss of profits for the shutdown exceeded $121 million, although its claim against FM was for the $25 million in CTE limits. The trial proceeded over twelve days in March and September, 2019 and the decision was issued on March 30, 2020.

Factually, the lengthy decision provides an interesting overview of how the Chalk River reactor worked, how the leak and shutdown occurred, and the causal assessment of the leak. As a substantive legal analysis, the decision refers to almost every interpretative principle and rule of construction one should apply when assessing insurance coverage. It addresses the purpose of all risk insurance policies, principles of fortuity, questions of policy ambiguity, considerations on exclusions, exceptions and shifting onus, and the importance of thoroughly understanding the factual context against the respective policy at issue.

Coverage Assessment

In generally considering “all-risks” insurance, the Court observed that the purpose of an allrisks insurance policy is to provide peace of mind to an insured who hopes to transfer the risks of unforeseen events to an insurer. The Court acknowledged that “all risks” policies, are not “all loss” policies, but do protect against fortuitous losses unless otherwise excluded. In this manner, one may begin to assess coverage for an incident giving rise to a loss by identifying whether the incident was expected, or non-fortuitous, as opposed to unexpected, unknown, accidental or fortuitous. The function of insurance, and thus the “all-risks” policy, is to respond only to the fortuitous loss.

As this was an “all-risks” policy the limits of the coverage are derived from policy exclusions. FM contended the corrosion, nuclear radiation and idle period exclusions all excluded coverage. Because it was relying upon exclusions, FM had the onus to prove either exclusion applied.

The Court assessed the corrosion exclusion, which was: This policy excludes … corrosion … but if physical damage not excluded by this Policy results, then only that resulting damage is insured. The Court first assessed whether the corrosion exclusion applied, and then whether the exception to the exclusion for resulting physical damage could apply. The discussion of the cause of the corrosion, whether or not it was fortuitous, and what constituted resulting physical damage, were all central to the decision.

The reactor’s internal system was subject to known, “non-fortuitous” corrosion, which was monitored over the reactor’s life. The parties agreed that generalized, known corrosion was not a covered loss, as the corrosion exclusion applied to it. However, the reactor was also found to be subject to a previously unknown corrosion that was accelerated and localized.

MDS contended this other, unanticipated corrosion, was not captured by the corrosion exclusion, and that it caused the leak leading to the shutdown, the extended delay, MDS’s inability to get isotopes, and therefore its lost profits. FM asserted it would be nonsensical to interpret the corrosion exclusion as applying only to known and anticipated corrosion, as by its very nature such corrosion would be expected, and therefore not fortuitous, and so would not fall within “all-risks” coverage to begin with. Rather, FM claimed the corrosion exclusion must have applied to all manner of corrosion, including whatever caused the reactor’s leak.

However, the Court noted FM’s approach ignored the fact that the “corrosion exclusion” also listed within it other types of generalized degradation, including “…deterioration, depletion, rust, … erosion, wear and tear…”, all of which involve expected and anticipated manner of degradation. The Court accepted MDS’s contention that corrosion that was unanticipated and unexpected was not the type of corrosion the exclusion was meant to capture.

The Court assessed the detailed expert evidence from both parties as to the cause of the leak and found as fact that it was due to fortuitous pitting corrosion. The precise manner of corrosion was by unanticipated microscopic attack, and distinct from the anticipated manner of known and generalized corrosion to which the exclusion was expected to apply.

The Court also interpreted the insurer’s evidence at discovery and trial as confirming “that the corrosion in this case is fortuitous and that not all corrosion is excluded under the Policy.” This conclusion became central to whether to apply the corrosion exclusion.

The Court found that it was the unexpected type of corrosion that caused the leak. The corrosion exclusion was applicable only to anticipated corrosion and therefore did not apply. Only expected and known corrosion would be captured by the Policy’s corrosion exclusion. However, the unknown, fortuitous corrosion that occurred here to cause the leak was outside the scope of the exclusion.

Although the Court concluded the corrosion exclusion did not apply, the Court still assessed whether the exception to the exclusion for resulting physical damage would apply. An insured holds the onus of proving an exception to an exclusion, and such exceptions are broadly interpreted.

In addressing the issue of resulting physical damage, the Court considered whether the phrase should be narrowly applied to “require actual physical damage” or “broadly to include loss of use.” As well, in this case determining the actual physical loss or damage was important because that determination also helped delineate the policy’s period of indemnity.

The phrase “resulting physical damage” was not defined in the policy, or in Canadian textbooks. The Court noted two conflicting lines of cases that considered the phrase; one suggesting a narrow approach requiring actual, tangible damage, while the other suggesting a broader approach based on a property’s impaired of use or function.

Without a “bright line” test dictating which line to apply, the Court considered the factual context against this policy and concluded the broader approach was appropriate in this instance. The leak required a shutdown of the reactor and a “convoluted” series of events involving identifying the problem and making repairs all unfolded before the reactor could be restarted. This impairment of function or loss of use was treated as resulting physical damage arising from the leak caused by the unanticipated corrosion.

The Court found in an almost summary fashion that the other exclusions, such as an idle period exclusion, and various nuclear exclusions, also did not apply.

Decision on Interest

Since this decision was released there has been much focus on the Court’s interpretation concerning the aforementioned exclusions and resulting property damage. Many authors have speculated as to how a broad interpretation of resulting physical damage encapsulating a functional impairment could apply to pandemic coverage claims.

However, often overlooked in these discussions is the decision’s approach to awarding interest. MDS not only claimed a breach of contract from FM’s denial of coverage, but it also claimed compensation from FM’s contractual breach based on MDS’s actual cost of borrowing and sought recovery on a compounded basis; not merely satisfied with set pre-judgment interest rates.

Section 130(1) of the Courts of Justice Act (“CJA”) permits courts to use discretion to impose interest rates that are higher than stated pre-judgment interest rates. Section 130(2) sets out the criteria a court should utilize when making this determination. In this case, the Court considered emerging trends suggesting a departure from the CJA rates of interest, options to award interest on a compound basis, and using interest awards as a compensatory measure, rather than a punitive one.

The Court noted “bright line” cases, where a party’s wrongful conduct warrants a departure from the CJA interest rates. But the Court also noted other cases that awarded compound interest as a compensatory measure, and referenced the example of compound interest in the statutory regime governing accident benefits in automobile accidents (known as SABS), which encourages insurers to compensate for actual losses sustained. The Court recognized that while the SABS program differs, “… both situations involve relationships of utmost good faith and situations with a vulnerable insured.” Arguably, even the recognition that a commercial insured who is denied insurance coverage can constitute a “vulnerable” party facing down an insured is a significant development

In this case, MDS made no assertion of bad faith, and certainly an insurer, like FM, is entitled to deny coverage. However, the Court observed that the “battle lines” were drawn early, with the denial in August 2010, and never wavering thereafter. The Court also observed that FM’s early commitment to re-review coverage if additional facts came to light “rang hollow”.

Ultimately, FM was not successful in maintaining its no-coverage denial after what the Court described as a “costly and protracted” ten-year legal battle. Correspondingly, the Court found it was fair and just that MDS be adequately compensated for its actual loss, based on undisputed evidence about its cost of borrowing, rather than simply an interest award based on the CJA.

The Court also accepted that FM had specific knowledge of the “unique facts and particular vulnerability” of MDS. When the shutdown occurred FM and MDS were in the midst of negotiating increasing this very CTE coverage. FM’s knowledge was based on a long term relationship with MDS, and included awareness of the importance of the isotope supply to MDS’s income, its inability to procure isotopes elsewhere and its corresponding vulnerability to supply changes, and that its actual losses were much higher than claimed under the policy. Therefore, the Court concluded that with all this information, it was clearly foreseeable to FM that MDS would need to borrow funds at market rates to cover such an extensive loss to its business. The Court went so far as to infer that because FM was the reactor’s insurer as well, it had evidence about the reactor that was not available to MDS.

The Court also rejected FM’s position that compound interest was not within the reasonable expectation of the parties. While the Policy did not stipulate an interest rate, as the controlling author of the Policy FM could hardly be expected to reference compound interest. As well, MDS’s statement of claim (from 2010) made clear it was claiming for loss of use of the insurance funds.

The Court also commented that FM set aside no reserves to cover this loss, and therefore had “free use of the funds” since the date of loss. FM’s use of the funds (that the Court ultimately found were due to MDS) over the 10-year period was calculated at $17,119,647, which was not contested. The Court compared that figure to the sum of $1,668,368, which was based on a simple interest rate calculation in accordance with the CJA.

The Court also referenced the “Pareto efficiency principle”, an economic theory whereby, when applied to contract law, a contractual breach puts neither party in a worse position. Applied here, the Court noted that compound interest rate on MDS’s rate of borrowing equates to $12,580,436 which is $4,539,211 less than FM’s calculated profit.

The Court awarded MDS prejudgment interest based on its actual cost of borrowing since the date of loss until payment of the policy limits of $25,000,000. The Court concluded this section with:

“[742] To order otherwise means that the Insurer, though losing the lawsuit on all fronts, has won. It breached the Policy with the Plaintiffs, has had the benefit of the use of the Plaintiffs’ funds totaling millions of dollars over the years. To order otherwise allows the Insurer to make a considerable profit on the amounts withheld in breach of the Policy, at their client’s expense.”

It will certainly be interesting to see if this paragraph gains traction in insurance coverage disputes.


The decision is under appeal. The notice of appeal focusses on the Court’s conclusions pertaining to the corrosion exclusion, including whether it was meant to exclude all types of corrosion, limiting the exclusion to non-fortuitous corrosion, and findings concerning FM’s witnesses conceding that not all corrosion was excluded. The appeal also contests the Court’s conclusions on the resulting physical damage exception on these facts, and its application to the loss of use of the reactor. The idle period exclusion, nuclear radiation exclusion and interest conclusions are also under appeal.


With the complex and unique fact scenario further applied to the nuances of policy interpretation, the appeal decision will certainly provide further substantive commentary on most of the above-mentioned issues. But ultimately, the decision’s commentary on interest could avoid those complexities and so may have a longer lasting, and thus greater impact, than the case’s comments on particular exclusions, or the phrase resulting physical damage.

With this decision, the Court makes clear that awarding compound interest in a commercial context need not necessarily be linked to punishing a party for wrongdoing. Rather, by awarding compound interest, and at a rate higher than the CJA, courts could better serve to compensate insureds for the true losses suffered, while contemporaneously discouraging profit through the refusal to pay covered claims. By awarding compound interest, this decision gives further warning to be cautious about coverage denials.

¹ Rory Barnable is an insurance coverage lawyer and the principal of Barnable Law P.C.