In our experience, aggregation clauses are among the most heavily contested provisions in cargo and logistics insurance. They govern whether multiple losses are treated as a single claim for the purposes of applying limits, deductibles, and sub-limits, often making the difference between meaningful recovery and severe underinsurance.<\/p>\n
While insurers frequently assert aggregation where there is a run of thefts, shortages or logistics failures, English law does not permit aggregation by default. The outcome turns on\u00a0policy wording,\u00a0causation, and\u00a0the factual coherence of the losses said to aggregate.<\/p>\n
The Central Importance of Wording<\/u><\/strong><\/p>\n Aggregation is not governed by a single legal test, but depends on the policy wording selected. English law distinguishes between:<\/p>\n Courts have repeatedly emphasised that an\u00a0\u201cevent\u201d<\/em>\u00a0is something that\u00a0\u201chappens at a particular time, in a particular place, and in a particular way\u201d<\/em>\u00a0(see <\/em>AXA Reinsurance (UK) plc v Field [1996] 1 WLR 1026 (HL)<\/em>). By contrast, where a clause aggregates by reference to an \u201coriginating cause<\/em>\u201d, aggregation may extend to a continuing state of affairs or a common operational failure, as considered by Spire Healthcare Ltd v RSA [2022] EWCA Civ 17<\/em>.<\/p>\n Cargo and logistics policies frequently sit closer to the former category, even when they use hybrid language that permits aggregating a\u00a0series<\/em>\u00a0of accidents or occurrences.<\/p>\n Accident, Occurrence And Event: Distinct Concepts<\/u><\/strong><\/p>\n In logistics policies, theft or loss of cargo is typically treated as an\u00a0\u201caccident\u201d or \u201coccurrence\u201d<\/strong>, that is, a discrete fortuitous incident, often involving deliberate human conduct.<\/p>\n Where a policy allows aggregation of:<\/p>\n \u201ca series of accidents or occurrences arising out of one event in any one location\u201d<\/em><\/p>\n the structure of the clause matters.<\/p>\n In our view, the accident or occurrence remains the\u00a0primary operative unit. The insurer must first establish that multiple losses can properly be described as arising out of\u00a0one event (\u201devent\u201d being used here in the same\u00a0functional sense\u00a0as an \u2018occurrence\u2019, namely a specific happening in time and place) before aggregation can follow.<\/p>\n This hierarchy is critical, and it prevents insurers from eliding multiple independent thefts or interceptions into a single claim by appealing to a high\u2011level narrative (e.g. \u201corganised crime\u201d, \u201csystemic theft\u201d, \u201ccriminal gangs\u201d etc.) divorced from the actual incidents.<\/p>\n Series of Losses<\/u><\/strong><\/p>\n Secondly, on the above wording, insurers must prove that the multiple losses are a \u201cseries\u201d<\/em>; in other words that there is some connection between the thefts and they are not simply a number of unconnected happenings.<\/p>\n What Insurers Must Prove: The Legal Burden<\/u><\/strong><\/p>\n Where aggregation is relied upon in relation to the number of limits applicable,\u00a0the burden lies squarely on insurers to prove that the losses aggregate. To aggregate multiple cargo losses by reference to accident, occurrence or event wording, insurers must demonstrate the following:<\/p>\n Insurers must identify\u00a0one event<\/strong>\u00a0capable of unifying the losses. This cannot be merely a shared background risk or an industry\u2011wide problem. The courts have consistently rejected attempts to aggregate by reference to general conditions, vulnerabilities or ongoing criminality.<\/p>\n As the Supreme Court held in AIG Europe Ltd v Woodman [2017] UKSC 18<\/em>, losses must be connected by more than similarity or coincidence; there must be a\u00a0real inter\u2011connection\u00a0such that they \u201cfit together\u201d, rather than being linked only through external background factors.<\/p>\n An assertion that multiple thefts were committed by an organised crime group is not sufficient unless the insurer can point to a\u00a0specific operative event, for example, a single coordinated operation, decision or execution, rather than a series of similar opportunistic crimes.<\/p>\n Even if an event is identified, insurers must show that each loss\u00a0arose out of\u00a0it. This incorporates orthodox principles of legal causation.<\/p>\n Courts have repeatedly warned against treating the causal language in aggregation clauses as infinitely elastic.<\/p>\n In logistics contexts, where thefts occur:<\/p>\n it becomes increasingly difficult for insurers to demonstrate that each loss arose out of the same event rather than out of separate, self\u2011contained criminal acts.<\/p>\n Although not a rigid checklist, the \u201cunities\u201d analysis remains a powerful analytical tool where aggregation turns on event or occurrence wording.<\/p>\n Following Kuwait Airways Corp v Kuwait Insurance Co SAK [1996] 1 Lloyd\u2019s Rep 664<\/em>, courts typically examine:<\/p>\n Where cargo losses are dispersed across time and geography, and where no single coordinated operation can be demonstrated, these unities are unlikely to be satisfied. Similarity of method, or repetition of thefts along a supply chain, is not enough.<\/p>\n Many logistics policies restrict aggregation to losses arising from\u00a0\u201cone event in one location\u201d<\/em>. This is a significant limiting factor.<\/p>\n Insurers must demonstrate not only a single event, but a\u00a0single location\u00a0in a meaningful sense. Attempts to characterise an entire logistics network, route corridor, or regional supply chain as \u201cone location\u201d have little support in English law. Policyholders are entitled to insist on a\u00a0physical and geographical reality, assessed from the standpoint of a reasonable observer, not by reference to the insurer\u2019s preferred level of abstraction.<\/p>\n Common Insurer Arguments a<\/u><\/strong>nd Their Weaknesses<\/u><\/strong><\/p>\n Insurers frequently argue that multiple cargo thefts form part of:<\/p>\n\n
\n
\n
\n
\n
\n
\n