Nat-Cat Claims Aggregation and Clash: How Multiple Perils in One Event Get Apportioned Under Indian Policies

An Indian monsoon cyclone is rarely one peril. When Cyclone Amphan hit West Bengal in May 2020, a single weather system delivered sustained wind above 155 kmph, storm surge along coastal districts, torrential rainfall, and subsequent river flooding. Policies covered wind, flood, and storm surge, but the treaties that protected those policies had per-event limits, hours clauses, and aggregation wordings that determined whether Amphan was one loss event or several for reinsurance purposes. The answer moved hundreds of crores of reinsurance recovery either way.

For direct policyholders, the mirror question is equally consequential. Does your fire policy's STFI (storm, tempest, flood, inundation) add-on cover one occurrence or multiple? Does the sum insured reinstate between sub-events? Does a separate deductible apply to each peril or to the aggregate? Indian policy wordings are less standardised than London market forms, and the answer depends on specific clause language that few policyholders read until after a loss.

The stakes are magnified by sub-limits. A Chennai textile unit with a 100 crore property sum insured may have an STFI sub-limit of 40 crore per occurrence. If the 2015 Chennai floods are treated as a single occurrence, the unit recovers up to 40 crore. If the flooding across November and December 2015 is treated as two occurrences (the mid-November event and the catastrophic early-December event), the unit potentially recovers up to 80 crore. The wording of the hours clause and the occurrence definition decides which outcome applies.

This post covers five layers of the aggregation problem: hours clauses and their Indian applications, single occurrence definitions under direct policies, clash coverage when multiple policies respond, reinsurance treaty aggregation mechanics, and policyholder strategies for structuring limits to maximise recovery under catastrophe scenarios.

The hours clause is the most common aggregation device in Indian reinsurance treaties and increasingly appears in high-value direct policies with sub-limits. It defines the temporal window during which losses from a named catastrophe peril are treated as one event for limit and deductible purposes.

Standard hours clause windows used in the Indian market follow London reinsurance practice: 72 hours for storm, tempest, flood, cyclone, and hurricane; 168 hours for earthquake and related volcanic activity; 72 hours for riot, strike, and malicious damage; 168 hours for terrorism in some treaties. The clock starts at the policyholder's election from the first loss, but once elected cannot be changed. Losses outside the window are treated as a separate event with a fresh limit and deductible.

Chennai 2015 illustrates the practical complexity. The November flooding began around 8 November and caused substantial damage over several days. A second, more severe round of flooding struck on 1 to 3 December. Policyholders with 72-hour clauses could elect to start the clock at either phase. Those who started at the November onset exhausted the window before the December event and treated it as a second occurrence with a fresh limit. Those who started the clock on 1 December caught only the second event under that election and had to treat November as a separate prior occurrence, potentially with sub-limits already partially eroded.

For earthquakes, the 168-hour window covers the main shock and aftershock sequence but creates ambiguity for long aftershock sequences like the 2001 Bhuj earthquake, where significant aftershocks continued for weeks. The practical rule: aftershocks within the 168-hour window are one event with the main shock, those outside are separate events.

Election timing matters strategically. Under Indian broker practice, the policyholder typically waits until the full extent of the event is known before electing the clock start, allowing optimisation of recovery across limits and deductibles. Losing this election right through premature notification or unclear clause wording can cost crores in recovery.

Direct policies in the Indian market use varying definitions of single occurrence, and the language determines whether multiple damages from a catastrophe fall under one limit or several. Three drafting approaches dominate.

The unity-of-cause approach treats all loss arising from a single originating cause as one occurrence. Under this wording, a cyclone and its consequential flooding and wind damage are one occurrence even if they unfold over several days. The approach favours insurers by consolidating losses under a single limit but benefits policyholders by ensuring a single deductible. Standard fire policy wording in India often uses this language, particularly for older policies issued under tariff regime carryover.

The time-bounded approach applies the hours clause from reinsurance to direct placement, defining a single occurrence as losses within a specified time window from the first loss. This is increasingly common in high-value property programmes, large corporate placements, and industrial risk packages. It gives clearer boundaries but requires careful election management.

The specific-peril approach treats each named peril as a separate occurrence even if they arise from the same weather system. Under this wording, Amphan's wind damage and flood damage are two occurrences, each with its own limit and deductible. Policyholders generally prefer this for catastrophe exposure because it multiplies available limits, but it also multiplies deductibles, so the economics depend on deductible size and loss distribution.

A real example: a Mumbai pharmaceutical manufacturer's policy defined occurrence as any single cause or series of causes arising from one event within 72 hours. During the 2017 Mumbai floods, the insurer initially treated the entire week's damage as one occurrence, applying a single sub-limit. The policyholder successfully argued that the flooding was a series of distinct high-tide events triggered by separate rainfall episodes, each outside the 72-hour window of the others, supporting treatment as multiple occurrences. The resulting recovery was approximately 2.5 times the single-occurrence calculation.

Read your occurrence definition before the renewal conversation, not after the loss.

Catastrophe events often trigger multiple policies simultaneously. A warehouse fire following a seismic event may trigger the fire policy, the earthquake extension, the business interruption policy, and potentially a contingent business interruption cover under a separate insurer. A factory roof collapse during a cyclone engages property, machinery breakdown, and stock policies that may be with different insurers.

Contribution principles under Indian law, rooted in Section 80 of the Indian Contract Act 1872 and equitable contribution doctrine, govern how multiple insurers share a loss. Where two or more policies cover the same interest, same peril, and same subject matter, each insurer pays rateably in proportion to its exposure. The principle prevents policyholder over-recovery while ensuring each insurer bears a fair share.

Rateable distribution mechanics vary by formula. The maximum loss formula pays each insurer's share based on their limit relative to total limits. The independent liability formula pays each insurer's share based on what it would pay as sole insurer, then adjusts for excess. Most Indian policies use the maximum loss formula with contribution language in the standard conditions section.

Clash covers, sometimes called clash excess or clash aggregate treaties, operate at the reinsurance layer and protect insurers against accumulations when the same event triggers multiple policies across their portfolio. For a corporate policyholder with multiple placements, understanding the insurer's clash exposure is relevant because it affects claim settlement speed and willingness to pay at the upper layers.

The policyholder's tactical question: if multiple policies respond, which one should claim first? Under Indian law, the policyholder can claim against any responding insurer, and that insurer then seeks contribution from the others. In practice, claim against the policy with the most favourable wording, the most cooperative claims handler, and the strongest limit headroom. Do not accept insurer attempts to delay settlement pending contribution negotiations between insurers. Indian courts have held that contribution disputes are internal to insurers and do not justify delaying the policyholder's indemnity.

Reinsurance treaty wordings typically do not bind the policyholder directly, but they shape insurer behaviour during catastrophe claims. Understanding them allows policyholders to anticipate where friction will arise.

Per-risk treaties cover the insurer's net retention on each risk up to a specified limit. They are indifferent to catastrophe aggregation because each risk is individually assessed. Per-occurrence or per-event treaties, in contrast, cap the insurer's recovery from a single catastrophe event, creating strong insurer incentives to characterise losses as either one event (if their retention is low and recovery is high) or multiple events (if their retention multiplies favourably under a working excess treaty).

Catastrophe excess of loss treaties use event definitions that mirror the hours clause wording. If the treaty defines an event as losses within 72 hours from a single natural cause, the cedent insurer will take the same position when settling direct claims. Policyholders often find their insurer's coverage position on single-versus-multiple-event aligns suspiciously with whatever maximises reinsurance recovery. Good brokers identify this pattern early and challenge it where the direct policy wording supports a different outcome.

Single event versus multiple event aggregation in the reinsurance programme creates ripple effects for the policyholder. If reinsurance exhausts on a single-event basis, the insurer has no treaty capacity to cover further losses from what it now characterises as the same event, even if the direct policy arguably permits treating subsequent losses as separate. Some insurers resist honouring otherwise valid claims in this scenario, requiring policyholder escalation through Ombudsman or consumer forum routes.

The Kerala 2018 floods illustrate the interplay. The flooding was triggered by heavy monsoon rainfall, compounded by dam releases, and continued intermittently from 8 August through 26 August. Insurer reinsurance treaties with 72-hour clauses allowed aggregation of losses in discrete 72-hour windows. Some insurers recovered reinsurance across three distinct event periods, others across one. Direct policyholder settlements varied accordingly. Policyholders with sharp brokers recovered under multi-occurrence readings of their own policy wordings where the language permitted.

Three recent Indian catastrophes illustrate how aggregation mechanics play out in practice.

Kerala floods August 2018: insured losses estimated between 4,000 and 8,000 crore rupees depending on source. The event had multiple components, heavy rainfall (peaking in the third week of August), dam release cascades from Idukki and other reservoirs, landslides in hilly districts, and glacial-lake-outburst-like surge events debated in some catchments. Direct policies with unity-of-cause occurrence definitions generally applied one limit per site. Policies with specific-peril language allowed separate limits for flood, landslide, and business interruption. Reinsurance treaties largely treated the event as one under 72-hour clauses with election mechanics, though some sub-events fell outside the window and were separated.

Chennai floods November to December 2015: insured losses estimated at 3,000 to 5,000 crore rupees. The mid-November flooding and the 1 to 3 December flooding were separated by roughly two weeks, allowing treatment as two distinct occurrences under any reasonable clause wording. Policyholders who claimed under both phases recovered two deductibles but two limits, generally a favourable outcome for high-severity exposures. Business interruption claims were particularly large, with Chennai's auto and IT corridors facing multi-week disruption. BI aggregation raised its own hours clause questions, which most policies resolved in favour of treating BI loss as arising from the underlying property damage occurrence.

Cyclone Amphan May 2020: insured losses in the 3,000 to 4,500 crore range for West Bengal and Odisha combined. A single landfall with compound perils: wind above 155 kmph, storm surge of 3 to 5 metres in coastal Sundarbans, extreme rainfall, and subsequent river flooding. Unity-of-cause policies treated the whole event as one occurrence. Some large corporate placements had specific-peril language, allowing separate limits for wind and flood damage. The 72-hour clause captured the full intensity of the event in most cases. Policyholders with adequate sub-limits recovered substantially, those with single aggregate sub-limits across multiple perils were capped at the sub-limit despite much higher actual losses.

The recurring lesson: catastrophe exposure requires policy structuring, not just catastrophe add-on purchase. Sub-limit architecture, occurrence definitions, and hours clause alignment determine recovery more than headline sum insured.

Building catastrophe resilience into your insurance programme requires deliberate choices at three stages: policy structuring at placement, documentation during the policy period, and claim strategy after a loss.

At placement, begin with a catastrophe exposure modelling. For coastal industrial sites, model cyclone and storm surge scenarios. For riverine or urban sites, model flood scenarios based on return periods. For sites in seismic zones IV and V, model earthquake scenarios. Translate model outputs into required sub-limits and structure coverage accordingly. Generic 20 percent or 40 percent of sum insured sub-limits are usually inadequate for sites with genuine catastrophe exposure.

Negotiate occurrence definitions. Push for specific-peril language where it benefits you (multiple perils from one event can access separate limits). Push for unity-of-cause language where you prefer a single deductible. Align hours clause windows across direct and reinsurance layers to avoid gaps. Ensure the reinstatement of sum insured after a partial loss works automatically rather than requiring fresh underwriting approval.

During the policy period, maintain catastrophe preparedness documentation. Update asset registers with photographs and current values. Maintain stock records accurate to the current week. Keep business continuity plans current. Document flood protection, seismic retrofitting, and other loss mitigation investments because these both reduce claim disputes and strengthen renewal terms.

After a catastrophe, the election timing for hours clauses must be strategic. Do not notify a formal clock start until the full extent of the event is understood, which may be 7 to 14 days post-event. Work with your broker to model recovery under single-event and multi-event assumptions. Where multiple policies respond, coordinate claims to maximise total recovery while respecting contribution principles. Engage independent loss consultants for complex BI and contingent BI claims where quantum disputes with insurers are probable.

AI-driven platforms like Sarvada's help by modelling catastrophe scenarios against policy wordings at placement, flagging sub-limit inadequacy and occurrence-definition ambiguity, and tracking the real-time recovery position during active catastrophe claims. The structural improvements that AI enables, particularly around sub-limit sizing and occurrence definition clarity, compound over multi-year programmes to deliver materially better catastrophe outcomes.