Picture a small scaffolding or excavation subcontractor working a Big Build site in Melbourne's inner north. For months the jobs have come steadily. Then, with no written explanation, the head contractor stops passing work their way — and it goes instead to a firm connected to a union delegate on site, handed over with no tender, no invoice trail and no documented business case. This is not one leaked anecdote; it is the pattern described in clinical, repeated detail by the administrator a Federal Court appointed to run the construction division of Victoria's most powerful building union: contractors facing informal pressure to hand work to subcontractors with organised-crime links, with the union's authority on site standing in for a written contract. Victoria's Big Build — a construction pipeline worth more than $100 billion (approx.) across road, rail and level-crossing projects — runs on exactly this kind of subcontracting chain, tier below tier, on hundreds of sites at once. For a decade the state government that owns the pipeline treated the union enforcing it as a construction partner, not as a risk it needed to actively manage.
What the government did
The Construction, Forestry and Maritime Employees Union's Victorian branch has been the industrial power broker on the state's building sites since long before the Big Build existed, and it has been a major financial backer of the party that runs the state. The Victorian branch was affiliated to the state Labor Party and, on the Australian Electoral Commission's own disclosure returns, was reported as one of Labor's largest single donors in Victoria, contributing more than $3 million (approx.) around the 2022 federal election cycle (Australian Electoral Commission, Transparency Register). That relationship sat alongside the union's day-to-day industrial leverage on Big Build sites, where its ability to slow or stop work gave it practical control over which subcontractors got a run, largely outside any process the government itself designed or audited.
In mid-2024, a months-long investigation by Nine's newspapers and television arm — reported publicly from early July 2024 — alleged that men with bikie and organised-crime links were operating as CFMEU delegates on major building sites, including Big Build projects, and were able to steer work to favoured subcontractors. The allegations included claims of kickback arrangements and correspondence between then Victorian branch secretary John Setka and a known underworld figure. Mr Setka, who had led the Victorian branch since 2012, resigned in mid-July 2024. The federal Minister for Employment and Workplace Relations called the reports "completely unacceptable" and flagged possible deregistration of the union's construction division; the allegations were referred to Victoria Police and to Victoria's anti-corruption watchdog, IBAC, for assessment. Victorian Labor moved within days to suspend the CFMEU construction division's party affiliation and its capacity to donate.
The decisive intervention, however, came from Canberra rather than Spring Street. Parliament passed the Fair Work (Registered Organisations) Amendment (Administration) Act 2024, which received Royal Assent on 22 August 2024 and created a statutory scheme for placing a registered union's division into administration where the minister is satisfied it is in the public interest. On 23 August 2024, acting on an application from the Fair Work Commission's General Manager, the Federal Court placed the CFMEU's Construction and General Division — and its divisional branches, including Victoria — into administration for up to five years, and Mark Irving KC was appointed as independent administrator with power to run the division's affairs, remove officials and report to the minister every six months (Fair Work Commission, CFMEU Construction and General Division Administration). Ousted officials challenged the scheme's constitutional validity; on 18 June 2025, in Ravbar v Commonwealth of Australia, a full bench of the High Court unanimously rejected every ground of challenge and upheld the administration.
The strongest case
A union built around real, sometimes fatal, stakes
Put at its best, construction unionism exists because building sites remain among the most dangerous workplaces in Australia. National workplace-fatality data compiled by Safe Work Australia has consistently shown construction accounting for one of the highest shares of workplace deaths of any industry each year (approx.), and the CFMEU's institutional lineage — through predecessor unions that fought for asbestos bans, fall-arrest requirements and enforceable scaffolding standards — reflects decades of campaigning that measurably changed how Australian building sites operate. The union's right of entry to inspect safety conditions, and its role at the bargaining table in lifting wages and superannuation coverage for a workforce that is disproportionately casualised and subcontracted, are not incidental benefits; they are the core justification for construction unionism existing at all.
None of that case evaporates because a Federal Court found governance failures in one division and one branch. The administrator's own reporting has been careful to distinguish the conduct under investigation from the union's ordinary safety and bargaining functions, and the vast majority of the CFMEU's Victorian members were never personally implicated in the criminal allegations that triggered the administration. A construction sector delivering the Big Build at the pace and scale Victoria has attempted needs organised labour with a genuine safety culture on site; the case against what went wrong in the Victorian branch's governance is not a case against the existence of construction unions.
The evidence: what the administration has found
The administration's own reporting is the clearest primary record of what was happening inside the union, and it has not been kind to the pattern of governance the Victorian branch had operated under for years. The Administrator's First Bi-Annual Report, dated 24 February 2025, stated that Mr Irving had by that point uncovered information supporting "the accuracy of the allegations of criminal and corrupt conduct" first raised in 2024 (Administrator of the CFMEU, First Bi-Annual Report, 24 February 2025). The Second Bi-Annual Report, tabled on 25 August 2025, went further: it documented a pattern of intimidation directed at building-industry employers, subcontractors and rival union officials, contracting arrangements in which subcontractors with organised-crime links obtained work on major projects through informal pressure on head contractors, and financial irregularities inside branch-level accounts, including expense claims that did not appear to relate to legitimate union activity. The Administrator's office also established an information-sharing arrangement with the National Anti-Corruption Commission, supporting referrals in both directions (Administrator of the CFMEU, Second Bi-Annual Report, 25 August 2025).
That governance failure did not sit apart from the state's construction economics — it sat inside the same pipeline the Victorian Auditor-General has separately found is running well over its original budgets. VAGO's Major Projects Performance Reporting 2024, tabled 19 February 2025, assessed 113 major Victorian projects worth a combined $145.5 billion and found their total estimated investment had risen by $11.66 billion, or 8.7 per cent, in a single year, with the report's central conclusion being that "the information public entities provide Parliament and the community is not meaningful" about why costs move (Victorian Auditor-General's Office, Major Projects Performance Reporting 2024, 19 February 2025). None of VAGO's public reporting attributes a specific dollar figure of that growth to union conduct — cost inflation, scope changes and construction-market pricing are the causes VAGO's own auditors most often cite — but a subcontracting market in which access to work could be steered informally rather than competitively is not a market operating on the price discipline the government's own business cases assumed.
The critique: an institutional failure, not a rogue-actor problem
The administration's findings matter precisely because they describe a system, not a handful of bad individuals. Intimidation of employers and rival officials, informal steering of subcontracts to crime-linked firms, and financial irregularities inside branch accounts are not the profile of a few officials exceeding their authority — they are the profile of an institution whose industrial leverage over a captive market of head contractors was never subject to the competitive or probity checks that would apply to any other party controlling access to $100 billion (approx.) of public construction work. A union that can make a head contractor's project run late, over budget or unsafely picketed has genuine coercive power on a building site; the evidence now on the public record is that this power was used, in a documented pattern, to direct work outside any process the state government designed, funded or audited. The reform now under way — new officials, a court-ordered financial reporting regime, an information-sharing arrangement with the National Anti-Corruption Commission — is being imposed from outside the union rather than having come from within it, which is itself a comment on how entrenched the old governance had become.
A union used its industrial leverage over subcontractors as though it were a licensing regime — without ever being one.
The critique: a government too close to audit itself
That failure belongs to the state as much as to the union. Victoria's government is the client on nearly every Big Build project, the party that sets probity and procurement conditions, and the party that chooses which head contractors and delivery authorities are accountable for what happens on its sites. It took a media investigation, not a state audit or a state-initiated integrity review, to surface conduct the administrator later confirmed; Victoria's own response was to suspend a financial and organisational relationship only after that exposure, and to refer the matter to police and IBAC for assessment, rather than to commission its own inquiry into how a subcontracting market this compromised had operated, seemingly unaddressed, across a program the government itself had spent a decade expanding.
The historic financial relationship between the Victorian Labor Party and its largest union donor is not evidence that any individual minister acted improperly — no public finding says that — but it is a structural conflict of interest that made the government a less independent monitor of conduct on its own construction sites than an arm's-length client would have been, and that conflict predates the 2024 scandal by years. A government that depends on a counterparty for millions of dollars in political donations, and that needs that same counterparty's cooperation to keep its flagship infrastructure program on schedule, has a weaker institutional incentive to go looking for the kind of conduct the administrator eventually found than a government with no financial relationship to the union at all. Victoria has not yet said what, if anything, it will do differently as a client the next time a union — or any other party with informal control over site access — is positioned to direct public money outside a competitive process.
The risk to the future
A temporary fix on a permanent pipeline
The federal administration is, by its own statute, temporary. It can run for up to five years from August 2024, meaning it is legally capable of lapsing by 2029 — well before Victoria's construction pipeline is finished. The Suburban Rail Loop and the next generation of Big Build projects will still be under construction long after any administrator has handed the union's Construction and General Division back to elected officials. Nothing in the record to date shows Victoria has built a permanent, state-level structure — independent contractor vetting, transparent subcontractor-selection rules, or a standing probity function inside its delivery agencies — that would outlast the federal scheme and stop the same informal leverage over subcontracting reasserting itself once Canberra's oversight lifts.
That matters because the underlying economics have not changed. Victoria's major projects were already running $11.66 billion over budget in the single year VAGO most recently measured, before accounting for whatever premium a compromised, less-than-competitive subcontracting market added on top. A state government that relies on the same union structures to keep its largest construction program moving, and that has never independently audited how work was actually allocated on its own sites, is not well placed to guarantee the conduct the administrator has now documented will not simply wait out a five-year clock.