Records and expert evidence in Australia

That combination. A homegrown procedural innovation with no reported footprint in the field it should most obviously help, and a records failure that decided the leading case instead. Is the real shape of Australian delay evidence. Start with where concurrent evidence came from, because the corrected history matters more than the folklore.

Concurrent evidence is an Australian invention, and the folklore is wrong

Concurrent evidence did not begin in the "Australian Competition Tribunal". It began in the Trade Practices Tribunal under Lockhart J: that body was renamed the Australian Competition Tribunal only in 1995, years after the technique was already in use. The first courtroom use is recorded as Rogers J, in 1985. And Sir Laurence Street was already running concurrent evidence in arbitrations by 1992, which matters more than a historical footnote, because it means the technique entered Australian arbitration practice as early, and as natively, as it entered court practice. That is the forum where most Australian construction delay disputes are actually resolved.

How hot-tubbing changes a delay case, in principle

Concurrent evidence removes counsel's fragmentation advantage. In sequential cross-examination a delay expert is taken through their assumptions one at a time, often days apart from the opposing expert, with no one in the room to call out an inconsistency in real time. In a hot tub the two experts sit together, and peer accountability instantly exposes the three moves that do most of the work in a contested programming analysis: re-logic, calendar swaps, and window-boundary choices made to produce a preferred number.

But the dominance risk is worse for delay evidence than for most other expert fields, for a structural reason: the tribunal cannot verify an as-built critical path from the bench the way it might sanity-check a valuation or a medical opinion. A confident expert with a wrong baseline is harder to catch in real time than a confident expert with a wrong number, because the tribunal usually lacks the independent means to know the baseline is wrong at all.

The practitioner evidence on how the technique is actually run is not reassuring. A survey of experts found that 85% consider concurrent evidence useful in principle, but over half rarely or never received an issues list before the session, and over half never questioned the other expert during it. The technique is being used well below its design intent: it works when both experts arrive briefed on the same issues and are prepared to test each other, and in a large share of Australian practice neither condition is met.

No Australian construction or delay judgment using or commenting on concurrent evidence was located for this corpus. The most plausible explanation is structural, not accidental: Australian delay disputes are predominantly arbitrated under forms that mandate arbitration, so the reported court record simply does not see them. And in the leading reported delay-evidence case that is available. White Constructions. The court did not run a hot tub at all. It rejected both parties' experts and appointed its own under UCPR r 31.54, bypassing concurrent evidence entirely at the moment it might have mattered most.

The governing rules, jurisdiction by jurisdiction

Concurrent evidence and expert conduct are governed differently across Australian courts, and the differences are wide enough to change how a report should be prepared before it is known which forum will hear it.

The Queensland entry deserves particular care. Standard reference works still describe Queensland as having no expert code of conduct. That description is obsolete. A delay expert instructed on a Queensland matter under the old assumption risks a communication approach the current rules do not permit.

Admissibility: the Makita and Dasreef line

The foundational admissibility test for expert opinion evidence in Australia is Makita (Australia) Pty Ltd v Sprowles, at [85] of Heydon JA's judgment. It sets four requirements: the expert's specialised knowledge. Based on training, study or experience. Must be identified. The opinion must be wholly or substantially based on that specialised knowledge. The facts and assumptions on which the opinion proceeds must be stated, and the reasoning must be exposed, showing how the specialised knowledge applies to the facts to produce the opinion.

Two errors circulate about this case and are worth correcting explicitly. It is sometimes headed as a High Court of Australia decision. It is a NSW Court of Appeal decision. And at least one published source gives the citation as "[2001] 52 NSWR 708". The correct report is (2001) 52 NSWLR 705.

The reasoning must be exposed. How the specialised knowledge applies to the facts to produce the opinion.

That fourth limb is the one that decides most contested delay-expert admissibility arguments, because it targets exactly the failure mode a weak delay report produces: a stated conclusion about the critical path without a traceable chain from the contemporaneous record to that conclusion.

Dasreef Pty Ltd v Hawchar supplies the related "basis rule" question. Whether the Evidence Act requires the factual basis of an opinion to be proved, not merely stated. Sources on this point conflict: the NSW Civil Trials Bench Book reads Dasreef at [41] as meaning proof is not required, while at least one barristers' paper takes the opposite view. This corpus resolves that conflict in favour of the Bench Book's reading, supported by later authority at [24]: the basis rule survives as a requirement of exposition, not of proof. An expert does not have to prove every assumption is correct. The assumptions have to be stated clearly enough for the tribunal and the other side to test them.

No Australian decision has been located applying Makita or Dasreef specifically to a delay expert. Limb 2 of Makita. Whether the opinion is wholly or substantially based on specialised knowledge. Is accordingly an unrun argument, and a strong one. Much of a typical delay report is narrative chronology and scheduling-software output rather than expert opinion in the Makita sense, and a party facing an unfavourable delay report should expect that argument to be run against it before it is run against them.

How exclusion actually happens

Australian courts do exclude expert evidence, but almost never on the ground of delay methodology itself. Exclusion is concentrated on qualification and on reasoning.

Around that pattern sits the code acknowledgement. UCPR r 31.23(3) requires an expert to acknowledge the Harmonised Expert Witness Code of Conduct, and a discretionary line of authority. Refusing admission in some cases, enforcing the requirement strictly in others. Shows courts prepared to act on non-compliance. But Chen v R softened that position: non-compliance now bites through the Evidence Act's discretionary exclusion provisions, sections 135 and 137, rather than through automatic inadmissibility.

The honest summary is the same one this corpus reaches in the United States: Australia receives and disregards. The code acknowledgement is the one genuinely distinctive Australian gate, and even its force is diminishing as courts prefer to discount weight over excluding evidence outright.

Records: the lesson of White Constructions

White Constructions is the reported case that should anchor how any Australian delay claim is prepared, and its lesson is about records, not methodology. Both parties' delay experts were rejected, and rather than choose between them the court bypassed the process entirely and appointed its own expert under UCPR r 31.54. Behind that rejection sat a records failure: the site diary recorded what work was done, not what work was prevented, and why. That is the single most transferable practice point in Australian delay evidence. A diary that logs progress but not obstruction cannot support a causation case, however well the expert who later relies on it writes the report.

There is no Australian adverse-inference doctrine for a party's failure to keep records. The consequence of a records gap in Australia is not a presumption against the party that failed to keep them. It is simply the ordinary onus of proof, left where it always was, on the party bringing the claim.

Cessnock and the under-pleaded shift in onus

That ordinary-onus position has one significant, apparently unused, counterweight. Cessnock City Council v 123 259 932 Pty Ltd is not a delay decision on its facts, but it is the closest Australian analogue to a facilitation-of-proof doctrine, and it is powerful. Where the defendant's own breach caused or increased the evidential uncertainty a claimant faces, a rebuttable presumption of recoupment arises and the onus shifts to the defendant to disprove the loss. Described in the judgment as "a fair wind, not a free ride".

The fact pattern where this bites hardest in a delay dispute is one where the party controlling the record-generating machinery is also the party whose conduct is under challenge. For instance, where a Superintendent is found to have acted in concert with the Principal rather than independently. On facts of that shape, Cessnock gives the claimant a real answer to "your records are inadequate": the inadequacy was caused by the defendant's own conduct, and the presumption should run the other way. This appears never to have been run on Australian delay facts, which makes it one of the more significant unused tools available to a contractor whose record base was compromised by conduct it did not control.

Whether the programme is a contract document: deliberately left open

One recurring assertion in Australian delay practice is that the programme is deemed a contract document under a standard AS 4000-1997 clause. That proposition is doubted here rather than adopted: treating the programme as a contract document would make redundant the clause's own wording requiring the contractor not to depart from it "without reasonable cause". Wording that would do no work if the programme already had contractual force in its own right. The primary clause text could not be independently verified for this corpus, and the point is deliberately left unresolved rather than stated either way. Do not plead either position as settled.

Why the doctrine is so hard to see: arbitration

Australian construction delay disputes are predominantly arbitrated. The AS standard forms mandate it, and that fact explains much of what is missing from the reported case law on both concurrent evidence and expert admissibility. No statistics on the scale of that arbitral caseload were located. What is visible is the structural evidence of the docket: of roughly twenty-five reported construction decisions across 2024–26, about a third concerned security of payment, and exactly one concerned delay analysis.

One plausible mechanism ties that back to concurrent evidence specifically. ACICA Rules 2021, Art 39.3, imposes a nine-month award deadline on institutional arbitrations, and concurrent evidence is plausibly the fastest way to resolve competing expert cases within a fixed timetable. If that is right, hot-tubbing is not absent from Australian delay disputes at all. It is simply invisible, because it is happening inside arbitrations whose awards are not published. The same "arbitration capture" pattern this corpus has identified in delay disputes elsewhere in the common-law world applies here too.

None of this changes what actually wins an Australian delay case. It is still the record that connects a cause to a date, and an expert whose reasoning survives being read line by line. Concurrent evidence, wherever it is actually used, does not change what the record needs to say. It only changes how quickly a weak record gets found out.

About this material Library content is general information about construction claim practice, not legal advice. Entitlement, deadlines and procedure are governed by your own contract and by the law of the place the work is performed. Reviewed September 1, 2026 · Report a correction