News & Insights · 27 July 2026 · 5 min read

CMA Assets v John Holland: The Harsh Time-Bar Decision Commercial Teams Should Re-Read Annually

The court accepted the head contractor caused the delay — and still enforced the time bar. Worse: the clause excluding the prevention principle was held effective.

Malachy MullinClaims Management · Contract Administration · Infrastructure
Abstract graphic of a gate closing on a delay claim despite the delay being real

If you want one Australian case to make a project team take notices seriously, use this one. In CMA Assets Pty Ltd v John Holland Pty Ltd [No 6] [2015] WASC 217, the Supreme Court of Western Australia accepted that the subcontractor really was delayed — and that the head contractor caused much of it. The delay claims still failed, comprehensively, because the notices did not comply with the subcontract. And the decision goes a step further than most time-bar authorities: it held that a clause expressly excluding the prevention principle is effective, so the subcontractor was liable for liquidated damages even for delay the head contractor caused.

The facts

John Holland held the contract to upgrade and extend a wharf at Finucane Island, Port Hedland, for BHP Billiton Iron Ore interests — reconstructing a berth for larger vessels. In 2006 it subcontracted CMA Assets to demolish the existing berth structures, including berthing and mooring dolphins, for $10.4 million, with completion due in March 2007. The work finished around a year late.

CMA's difficulties were real: the reinforcement in several dolphins differed materially from the drawings, defeating its blasting methodology (and leaving large debris blocks on the seabed, in breach of the subcontract); John Holland failed to relocate a shiploader, blocking a phase of the works for some three months; and access to other work fronts was denied or deferred for further extended periods. CMA claimed roughly $8.1 million in variations and delay costs; its formal EOT claim sought 101 days. John Holland counterclaimed, including the cost of removing the seabed debris and liquidated damages.

The subcontract's notice regime was layered and specific. Clause 10.12 required notice as soon as CMA became aware of the likelihood of delay; notice of intention to claim an EOT within 7 days of the occurrence of the cause; and a detailed written claim within 14 days of commencement of the delay, with prescribed content — estimated delay, mitigation steps, the facts, the days claimed and the programme effect. Clause 10.13 said it plainly: a subcontractor that failed to comply with clause 10.12 "shall have no entitlement to an extension of time".

The issues

Was strict compliance with the notice regime a condition precedent to EOT and delay-cost entitlement? Did it matter that John Holland itself caused key delays — could the prevention principle defeat the time bar and the liquidated damages? And could waiver, estoppel or an implied duty to exercise the unilateral extension power in good faith rescue the claims?

What the court held

Allanson J enforced the regime as written. Compliance with clause 10.12 — as to both timing and content — was a condition precedent. CMA's first shiploader notice came well outside the prescribed time; other claims failed on content. The judge accepted the strict operation of the clauses was harsh, but not so purposeless or absurd as to compel a different construction. The claims failed even though the court found, on the evidence, that John Holland's conduct had in fact delayed CMA.

The prevention principle argument fared worse. Following the orthodox line — a contractor with an available EOT mechanism it failed to use cannot call the completion obligation into question — the court went on to hold (at [865]) that clause 10.13 expressly excluded the prevention principle and was effective in doing so. CMA was precluded from extensions and liable for liquidated damages "even where the relevant delay has been caused by John Holland". The estoppel case failed on the facts; the unilateral reserve power to extend (clause 10.14) was held unfettered, with no implied duty to exercise it fairly.

The arithmetic of the outcome deserves its own sentence: CMA recovered an agreed variation, while John Holland recovered over $5.3 million in debris-removal costs and $1.18 million in liquidated damages — the capped maximum — from the party the court accepted it had delayed.

What this means in practice

  1. "They know we're delayed" is not a notice. The principal's actual knowledge of the delay does nothing. Entitlement lives in the notice document — served in time, to the right person, with every prescribed content element. Rimfire Energy v BSF Co (No 2) [2025] FCA 384 has since made the same point at Federal Court level: content gaps are as fatal as late service.
  2. Multi-stage notice regimes need a process, not a memory. Clause 10.12 had three clocks running off different triggers — awareness, occurrence, commencement — plus content requirements. That is unmanageable by recollection. It is trivially manageable with a notice matrix, standing templates and a register reviewed weekly.
  3. Assume the prevention principle will not save you. CMA Assets confirms parties can contract out of it with clear words, and modern amended contracts routinely do. The only reliable response to principal-caused delay is the same as for any other delay: notice, on time, in full.
  4. Time bars cut twice. CMA didn't just lose its claims — it paid liquidated damages for delay it didn't cause, plus the counterclaim. When pricing a subcontract with an aggressive notice regime, price the administration to comply with it; the alternative is pricing the double jeopardy.
  5. Respondents and head contractors: the regime is your audit checklist. Every incoming claim should be tested against each notice element before any merits engagement. This decision is the template for that exercise.

Key takeaways

  • Strict compliance with notice provisions — timing and content — was a condition precedent; claims failed despite the court accepting the head contractor caused delay (CMA Assets [2015] WASC 217).
  • A clause expressly excluding the prevention principle is effective: liquidated damages ran even for delay caused by the other party (at [865]).
  • Unilateral "reserve power" extension clauses may be construed as unfettered — do not rely on them.
  • The defence is administrative: notice matrix, templates, multi-clock diary triggers and a live register.

This article is general information only and is not legal advice. For advice on a specific contract or dispute, seek legal counsel or contact Sumit Consulting for commercial and claims advisory support.

The record, delivered

The analysis, before you need it.

Commentary on Security of Payment, delay, variations and quantum — direct to your inbox as it's published. Unsubscribe any time.

Facing this issue on a live project?

The analysis above is general commentary, not advice. For your specific contract and records, talk to us directly.