6.2 Concurrent, Consecutive Causes and Burden of Proof

Key Takeaways

  • In consecutive causation, an unbroken chain initiated by an insured peril renders all resulting loss covered, whereas an unbroken chain initiated by an excluded peril excludes all consequential damage.

  • A novus actus interveniens (new, independent intervening force) severs the causal chain, transforming the intervening force into the new proximate cause of subsequent loss.

  • When independent concurrent perils operate simultaneously to cause distinct damage, losses must be segregated so the insurer indemnifies solely the damage attributable to the insured peril.

  • For interdependent concurrent causes combining inextricably, an insured peril combined with an uninsured peril is fully covered, but an insured peril combined with an excluded peril results in complete exclusion under the Wayne Tank doctrine.

  • The burden of proof initially requires the insured to prove a valid policy and loss caused by an insured peril on a balance of probabilities; once proven, the burden shifts to the insurer to substantiate any exclusion or warranty defense.

Last updated: October 2026

6.2 Concurrent, Consecutive Causes and Burden of Proof

In commercial insurance practice, losses rarely stem from a single, isolated event. Property damage often results from complex interactions of consecutive occurrences, simultaneous hazards, and human conduct. When multiple causes interact, claims adjusters and courts apply proximate cause principles to determine liability. Furthermore, when the factual cause of a loss is contested, the legal burden of proof determines which party must establish its claims.


Consecutive Causes and Chains of Causation

Consecutive causation occurs when events happen in chronological sequence, one following another over time.

1. Unbroken Chain of Events

When an initial peril triggers a series of natural, foreseeable consequences leading to damage, the chain is unbroken:

  • Initiated by an Insured Peril: The insurer is liable for all resulting damage, even if the final instrument of loss is an uninsured peril.
    • Example: Under an explosion policy, a boiler explosion ruptures water pipes, flooding machinery. The flood damage is fully covered because it flows directly from the insured explosion.
  • Initiated by an Excepted (Excluded) Peril: All subsequent damage is excluded, even if the final destructive force is an otherwise covered peril.
    • Example: Under a fire policy excluding earthquake, an earthquake ruptures gas lines that ignite and destroy a building. Because the excluded earthquake initiated the unbroken chain, the fire claim is excluded.

2. Broken Chain: Novus Actus Interveniens

A broken chain occurs when an independent force intervenes between the initial peril and the final loss, known as a novus actus interveniens:

  • The intervening event must not be a natural or foreseeable consequence of the first peril; it operates actively from an autonomous source.
  • A novus actus interveniens severs causation, making the new force the proximate cause of subsequent damage.
  • Example: A storm dislodges roof tiles on a warehouse. Days later, an intruder enters and commits arson. The fire loss is proximately caused by arson (novus actus), not the storm. If the policy covers storm but excludes arson, the fire claim is excluded.

Concurrent Causes: Simultaneous Perils

Concurrent causation arises when two or more distinct perils operate simultaneously to cause damage. Common law distinguishes between independent and interdependent concurrent causes.

Independent Concurrent Causes

Independent causes operate simultaneously without influencing each other, causing distinct heads of damage:

  • Principle of Segregation: Losses must be segregated. The insurer pays only for damage caused by the insured peril, while the insured absorbs losses from uninsured or excluded perils.
  • Example: Lightning (insured peril) destroys a chimney causing S$15,000 damage, while flash floods (excluded peril) simultaneously ruin S$30,000 of basement stock. The insurer pays S$15,000 and excludes the S$30,000 flood loss.

Interdependent Concurrent Causes

Interdependent causes combine inextricably to produce a single, indivisible loss. Neither peril alone would have caused the loss, or the damage cannot be separated into distinct parts. Two fundamental rules govern:

Rule 1: Insured Peril + Uninsured Peril ➔ Entire Loss Covered

When an insured peril and an uninsured peril combine inextricably to cause an indivisible loss, the insured peril prevails, and the full loss is covered. In the absence of an express exclusion, the insurer cannot escape liability.

  • Example: Under an all-risks policy with no design defect exclusion, poor design combined with wind causes a roof collapse. The entire loss is covered.

Rule 2: Insured Peril + Excluded Peril ➔ Entire Loss Excluded (Wayne Tank Principle)

When an insured peril operates concurrently and inextricably with an excepted (excluded) peril, the exclusion takes precedence, and the insurer incurs zero liability.

  • Landmark Precedent: Wayne Tank & Pump Co Ltd v Employers Liability Assurance Corp Ltd [1974].
    • Facts: A factory fire was caused by two interdependent factors: dangerously unsuitable plastic pipes (excluded as liability from goods supplied) and an employee leaving heating tapes switched on unattended (insured negligence).
    • Ruling: The English Court of Appeal held that where an indivisible loss results from two interdependent proximate causes, one covered and one excluded, the exclusion prevails. The claim was entirely repudiated.

Causation Decision Matrix

Causation ConfigurationOperational MechanismGoverning RuleClaims Determination
Consecutive: Insured InitiatesUnbroken chain flowing from insured peril to final lossInitial peril dominates sequenceFully Covered: Insurer pays for all resultant damage.
Consecutive: Excluded InitiatesUnbroken chain flowing from excluded peril to final lossInitial exclusion taints downstream damageEntirely Excluded: Insurer denies the entire claim.
Consecutive: Broken ChainNew intervening force (novus actus) breaks causal linkNew force becomes fresh proximate causeDetermined by New Force: Covered only if new force is an insured peril.
Concurrent: Independent PerilsSeparate perils act simultaneously causing distinct damageSegregation of losses into separable headsApportioned: Insurer pays only for damage wrought by insured peril.
Concurrent: Insured + UninsuredInterdependent perils combine to cause single indivisible lossInsured peril dominates in absence of exclusionFully Covered: Insurer pays total indivisible loss.
Concurrent: Insured + ExcludedInterdependent perils combine inextricably (Wayne Tank)Express exclusion overrides coverage clauseEntirely Excluded: Insurer pays nothing.

Burden of Proof in Insurance Claims

In contentious claims, proximate cause depends on the burden of proof (onus probandi). Under Singapore civil law, the standard is the balance of probabilities (>50%).

1. Initial Burden on the Insured

The initial burden rests on the insured to establish a prima facie case:

  1. A valid policy was in force;
  2. Physical loss or damage occurred; and
  3. The loss was proximately caused by an insured peril.

2. Shifting Burden to the Insurer

Once the insured establishes a prima facie claim, the burden shifts to the insurer to prove an affirmative defense:

  1. The loss was proximately caused by an excepted (excluded) peril;
  2. The insured breached a policy warranty or condition precedent; or
  3. The claim is fraudulent.

3. Unexplained Losses

If the cause of loss cannot be established on a balance of probabilities, remaining an inexplicable mystery, the claim fails. The insured carries the threshold burden of proving causation by an insured peril; if causation is unproven, the burden never shifts to the insurer.

4. Named Perils vs. All-Risks Policies

  • Named Perils Policies: The insured must prove that the specific named peril caused the loss.
  • All-Risks Policies: The insured proves only accidental physical loss during the policy term. The burden then shifts to the insurer to establish an exclusion.
Loading diagram...
Burden of Proof Allocation in Insurance Claims
Test Your Knowledge

In Wayne Tank & Pump Co v Employers Liability Assurance Corp (1974), a fire had two interdependent causes: an excluded cause (unsuitable plastic piping supplied) and a covered cause (an employee leaving heating equipment on). What did the court decide?

A

The insurer was liable in full, because employee negligence was an insured cause of the fire

B

Not liable: where an excluded cause operates with a covered one, the exclusion prevails

C

The loss was split 50/50 between insurer and insured

D

The insurer paid only for the heating equipment

Test Your Knowledge

After forensic investigation, the cause of a warehouse fire cannot be established on the balance of probabilities. The insured holds a named-perils fire policy. What is the outcome?

A

The claim fails: the insured has not proved an insured peril caused it

B

The insurer must pay, because doubts favour the insured

C

The claim goes to FIDReC for a 50% split

D

The insurer must pay unless it can positively prove that an exclusion applies

Test Your Knowledge

A storm damages a warehouse wall. Three days later, an unrelated trespasser deliberately sets fire to the building through the gap. The policy covers storm but excludes malicious damage and arson. How is the fire loss treated?

A

Covered, because the storm created the opening that made the arson possible

B

Paid at 50% because of concurrent causes

C

Paid through subrogation against the trespasser

D

Excluded, because the arson was a new, independent cause

Test Your Knowledge

Once the insured has shown a loss caused by an insured peril, what must the insurer prove to escape liability?

A

That no exclusion applies at all

B

That the insured was careless in some respect, however minor the carelessness

C

That an exclusion applies, or a warranty or condition was breached

D

Nothing; the court decides without evidence

Sections you finish are checked off in the contents.