Briefing

Furlough as a saving: Supreme Court offers finality on COVID-19 BI loss calculations

Summary

While the Supreme Court addressed a significant number of important issues for COVID‑19 business interruption (BI) insurance claims in the FCA Test Case in early 2021, a key outstanding question has been how furlough payments should be treated when assessing an insured’s losses.

The Supreme Court has now resolved that issue in Bath Racecourse v Liberty Mutual Insurance¹: furlough payments should be treated as savings and, accordingly, deducted when calculating recoverable BI losses.

Background

“Furlough” became one of the key terms associated with the COVID‑19 pandemic in the UK, and is shorthand for payments made by the UK Government to businesses under the Coronavirus Job Retention Scheme (CJRS). The policy rationale behind CJRS was simple: the Government would step in to help pay wages to encourage employers to retain staff during the COVID‑19 restrictions.

BI insurance indemnifies insureds where certain insured events cause interruption of, or interference with, the insured business at its insured premises. In the cases before the Court, the insuring clauses concerned interruption caused by a prevention of access resulting from government action. Earlier court decisions had established that such clauses provided cover for lost revenue resulting from premises closures due to the Government’s COVID‑19 restrictions.

A BI policy will generally set out the mechanics for calculating an insured’s loss, including how any savings to the business resulting from the insured peril should be treated. Here, the Court was concerned with two savings clauses that were variants of the standard wording published by the Association of British Insurers (ABI). The parties agreed that nothing turned on the differences between the clauses, so the Court focused on the following wording:

“If any of the charges or expenses of The Business payable cease or reduce in consequence of the Damage such savings during the Indemnity Period shall be deducted from the amount payable.”

It was common ground that “Damage” should be read in this context to refer to the insured peril (i.e., interruption caused by the prevention of access). The issue was whether the insureds’ wage costs had “ceased or reduced” in consequence of the insured peril by reason of the furlough payments. The Commercial Court and Court of Appeal found that they had.

The insureds appealed on two grounds:

On proper construction of the savings clause, the wages did not “cease or reduce” as the companies still had to pay the wages in full, then separately claim reimbursement from the Government under the CJRS scheme.
The furlough payments were not legally caused by the insured peril because (a) proof of prevention of access was irrelevant to furlough entitlement and/or (b) the payments were gratuitous, benevolent or voluntary in nature and should be treated as “collateral” benefits.

The insureds also argued that the lower courts had placed too much emphasis on the “indemnity principle” at the expense of the strict language of the contract.

The Supreme Court declined both grounds of appeal unanimously.

Construction

The Court confirmed that the language of the savings clause is concerned with the economic reality of whether a charge or expense has been reduced; it does not require a legal analysis of how that reduction occurs. A reasonable person, viewing the policy as a whole, would not draw a distinction between a situation where an employer’s wage bill is waived and one where the employer pays wages but is then reimbursed by a third party.

Much of the COVID-19 litigation following the FCA Test Case has applied the “ordinary policyholder” test set out in that case, which took the “reasonable person” to be, in the context of an SME, an “ordinary policyholder who, on entering into the contract, is taken to have read through the policy conscientiously in order to understand what cover they were getting” and not “a pedantic lawyer”.

The Court has now clarified that the objective approach should not focus on one party alone: the reasonable person is in the position of both parties. The Court nevertheless reiterated that “pedantic lawyers” remain unwelcome in a policy construction exercise.

The Court also emphasised that it is permissible to consider the overarching purpose of indemnification when construing a clause that is capable of more than one meaning, and to adopt the construction more consistent with that purpose. Savings clauses are designed to prevent over‑indemnification, and the Court found that this purpose would be undermined if only reductions in legal liability (and not factual economic savings) were considered.

The insureds’ construction of the savings clause was also rejected because it would lead to arbitrary and uncommercial results. Many insureds would have made employees redundant without furlough, reducing their wage costs and correspondingly reducing their BI claims. On the insureds’ case, however, they would be permitted to retain the benefit of CJRS support while also obtaining an insurance recovery for the same amount. The Court considered that outcome to be inconsistent with the fundamentals of indemnity insurance. The Court did not accept the lower courts had placed too much emphasis on the indemnity principle; the commercial purpose of an indemnity insurance contract was a valid consideration when construing its words.

Causation

The Court found a direct and sufficient causal connection between the insured peril and the savings achieved through furlough payments. Favouring a simpler analysis over the courts below, the Supreme Court held that interruption caused by government action predictably led insureds to place employees on furlough, reducing their net wage costs. That was sufficient. The Court rejected the insureds’ argument that a strict “but for” test should be applied, as that was inconsistent with the proximate causation analysis in the FCA Test Case, which should apply equally to both the insuring and savings clauses.

The Court also dismissed the argument that furlough payments were collateral benefits that should be left out of the adjustment. It held that furlough payments were fundamentally different from gratuitous payments from third parties because they were made pursuant to a Government scheme administered by HMRC, which conferred a right to claim if eligibility conditions were met and imposed an obligation to pay.

Further, the Court held that there was nothing to support a view that CJRS was intended to benefit insureds to the exclusion of their insurers. The scheme had broader economic and public health objectives; it was not a gift. Therefore, furlough payments were more analogous to state benefits payable as of right, which are commonly deductible when assessing an insured’s loss.

Conclusion

The judgment brings long‑awaited clarity to one of the final unresolved issues arising from COVID‑19 BI insurance litigation. It confirms that, where policies contain standard ABI‑style savings clauses, furlough payments will ordinarily fall to be deducted from recoverable BI losses.

More broadly, the judgment reinforces the Supreme Court’s commitment to a principled, commercial approach to policy interpretation, grounded in economic substance and the core indemnity nature of insurance contracts.

HFW were instructed by Aviva Insurance Limited on this claim.

Footnotes