The thesis
Australia already has the hard parts of a digital border: a universal visa requirement since 1994, no visa labels, mandatory airline pre-reporting of every passenger, biometric SmartGates at every international airport, and an unbroken arrival and departure movement record going back to 1981. What it lacks is a spine. The visa, traveller, identity, risk and compliance systems were built in different decades for different agencies and are held together by people, paper and workarounds.
The Parkinson Review (2023) called the migration system "not fit for purpose" and said its ICT "urgently needs modernisation" [S1]. The Nixon Review (2023) found the system "is being exploited when identity and criminal history is not verified" [S2]. The APSC Capability Review (2024) found that over 40 per cent of the department's 480 business systems have reached end of life, that it has no single executable plan for its ICT or data, and that it "is currently not positioned to deliver beyond maintenance" [S5]. The ANAO (2023) recorded the department's own words: core systems "over 30 years old", several programmes "still require paper-based visa lodgement" [S3].
The transformation case therefore writes itself. The harder question is how to do it without repeating Systems for People, the Global Digital Platform and the Permissions Capability, three consecutive attempts that each promised transformation and each ended with the legacy core still running [S1][S4][S24]. The answer this POV argues for: treat the border as one end-to-end process from intent to departure, lay identity and data before functionality, deliver in three tranches that each retire something real, and keep the legacy alive in parallel until the last tranche closes it.
High that the diagnosis is settled: four independent bodies (Parkinson, Nixon, ANAO, APSC) plus two parliamentary committees converge, and Home Affairs has agreed the substance on the record. Moderate on the size and timing of current funding: the 2025-26 Portfolio Budget Statements record "Nil" material changes to the visa and border-management programmes, and no line item for a core visa platform replacement was found in what I checked [S10]. Low on several 2026 claims circulating in migration-industry media about an "AI-driven" visa platform; they are single-source and sit awkwardly with the department's own CIO saying in June 2026 that Home Affairs is "at the start" of AI adoption [S14][S22].
The system at scale
Two things in these numbers matter for design. First, the border side is already largely automated at the point of clearance (three in five travellers never speak to an officer) while the visa side is not: 47 per cent of permanent applications lodged in 2024-25 were still open at year end, and the bridging population that waiting creates grew by a quarter [S8]. Second, the department cannot fully trust its own reporting: the annual report footnotes both traveller-facilitation targets, and several others, with "the department was unable to provide assurances over the flow of data from these systems into the aggregated reports that are used to calculate the performance result" [S9]. That the disclaimer is boilerplate makes the point worse, not better. A transformation that does not fix the data layer will not be able to prove it worked.
What the reviews found
Matt Hay, CIO and first assistant secretary of digital delivery services: "I think right now our biggest challenge is prioritisation. Demand, particularly for IT-enabled change, always exceeds our ability to resource that and fund it." The department is "very much, I think, right now at the start in the assisted space", has folded data and AI capabilities into the technology group, and is looking for an "incubator" model to prove use cases. Separately (July 2026) the department has begun scoping the replacement of its SAP ECC6 ERP by December 2030 [S14].
Read: the constraint the reviews described in 2023-24 is still the operating condition in mid-2026. It also directly contradicts industry-media claims that a nationwide AI decision platform went live in March 2026 (see gap report).
The Integrated Cargo System is "the sole channel" for reporting, clearing and risk-assessing all cargo. Its design "is based on a pre-digital business model" and "creates limitations and complexities on efforts to modernise and streamline trade regulations". Modernisation "is expected to take incremental delivery across 7 to 10 years" with coordinated change "across legislation, policy, business processes, operating arrangements and technology". Market sounding only; not a procurement [S12].
Read: the goods lane has the same shape of problem as the people lane, and the department is already framing it as a decade-long incremental programme, which is the right instinct and the template the people lane should share.
Processing improved in several lanes: second-stage skilled permanent down from 304 to 206 days median, Partner permanent median down from 279 to 271 days, Partner temporary down from 18 to 16 months at the 75th percentile, and visitor and Working Holiday visas under a day. The tail is where it bites: Partner permanent still took 14 months at the 75th percentile, unchanged year on year. The department reports that its risk and integrity capability has grown through "innovative use of data and technology", and that an Immigration Compliance Group was established on 27 February 2024 to manage risk "at all stages of a traveller's journey" [S8].
Read: the "journey" framing is now the department's own language. The transformation should adopt it rather than a system-by-system framing.
Recommendation 16 asks the department to build internal mechanisms to assess the effectiveness of measures introduced to "reduce the visa backlog; modernise the visa system; and embed simplification". Recommendation 17 then asks the Auditor-General to audit whether those measures worked; biometric streamlining sits at Recommendations 6 and 7, and applicant, agent and sponsor visibility of application progress at Recommendation 18. The ANAO has accepted, listing a potential 2026-27 performance audit, Implementation of improvements to visa processing, "in response to Recommendation 17" [S7].
Read: Parliament has put the modernisation question on the Auditor-General's forward programme. Whatever the department builds next will be audited against "backlog, modernise, simplify".
Note on a common error: the ANAO's own audit page, and most secondary coverage, attribute the backlog-and-modernisation substance to Recommendation 17 itself. The report text shows that substance is Recommendation 16; Recommendation 17 is the request for the audit.
- "Over 40% of the 480 distinct business systems have already reached end of life, and the department carries significant risk of failure"; the resources to sustain and replace them "are yet to be fully funded".
- "The department does not have a single, shared and executable plan for the future of its ICT or data assets." The endorsed Technology Strategy is four years old.
- Technology group supports the existing footprint plus "almost 200 active projects" with a roughly 50:50 contractor-to-APS ratio.
- "The department lacks an endorsed data and analytics strategy" despite data holdings that are "a major asset".
- "Two attempts to modernise visa platforms (Global Digital Platform and Permissions Capability) have been terminated since 2017." "The department is currently not positioned to deliver beyond 'maintenance', let alone improve its ability to radically transform its technology and use of data."
- On culture: integrity "needs to mature from being managed as a compliance process, to being a culture of integrity, relevant to all staff and through all areas of their work".
- Priority areas: fix the capital plan and budget; "establish an executable vision, plan and governance to deliver required future ICT and data assets" [S5].
Read: this is the most recent whole-of-department diagnosis and the most damning. It shifts the problem from "old visa system" to "no plan, no data strategy, no funded capital path".
Found inadequate due diligence and inter-agency information sharing in procurement. Recommendation 1: Home Affairs "should enhance its integrity risk process and culture to better inform procurement and contract decision-making for regional processing arrangements". The government agreed or agreed in principle to all 20 recommendations [S16].
Read: not a technology review, but it is why any multi-year platform programme in this department will be procured under unusually close scrutiny. Design the commercial model for that.
The department "departed in a number of respects from the approach it had set out in the RFT" and its conduct "reduced its ability to demonstrate that value for money was achieved". Five work orders with Accenture totalled $61.5m. By June 2022 "only four per cent of the base capability had been delivered"; the Digital Passenger Declaration was "an inflexible product with unresolved defects"; the reusable base capability due 30 June 2022 "was not delivered". The two shortlisted bidders from the cancelled Global Digital Platform had a competitive advantage from 24 months of "deep engagement". The department disagreed with the value-for-money finding [S4].
- Finding 5: the system "is being exploited when identity and criminal history is not verified". Biometrics were collected for only 33 subclasses and in 53 countries; the strategy "when fully implemented, will collect facial images from all travellers before they enter Australia". A 2025 FOI release confirms collection is still at 53 countries [S17].
- Finding 1: ABF had "an investigations capability of 120 across the full ABF jurisdiction" and "there is currently no compliance or investigative capability within the Department's Immigration Group". Intelligence that would disrupt exploitation was "not able to be shared under each agency's legislative frameworks".
- Finding 6: protracted processing motivates abuse. Onshore protection visa: 2 years 3 months median; AAT protection caseload up 340 per cent to 37,025; 89 per cent of onshore protection applications refused; stacking pathways "would add up to almost a decade" of lawful stay.
- Recommendation 15: a targeted data-matching exercise "to compare information holdings across Commonwealth agencies for private VET providers", then "a permanent data sharing mechanism". Recommendation 28: "strengthen identity verification requirements in key immigration systems", with two-factor authentication for end users as the action under it; public-facing systems "have not had the necessary investment or updates" to include it.
- Government agreed in full or part to 24 of 34 [S2].
Read: integrity is an identity-and-data problem before it is an enforcement problem. Nixon's remedies (biometric anchoring, data matching, shared holdings, MFA) are platform capabilities, not policy settings.
- "Australia's current core visa processing ICT systems are built on technology from the 1980s and 1990s. The main visa processing system (referred to as ICSE) is approaching 25 years of age, while other foundational systems that operate in the background are approaching 40 years of age." (Written in early 2023; on the same reckoning ICSE is now near 29.)
- The visa operating model is "supported by an ecosystem of almost 70 business systems"; reform "added layers of additional systems and layers of complexity, rather than modules".
- "There are only three major ICT reform 'release cycles' per year… Each cycle is oversubscribed." Development "is almost exclusively dedicated to system upkeep requirements and the legislative change cycle." Efficiency from minor upgrades "reached their limits"; it was staffing that cut the 2022-23 backlog.
- Identity still rests on "documents such as passport details and birth certificates provided by the applicant"; mismatches have meant decision-makers "not taking into account an applicant's full history". "This situation is entirely avoidable today."
- "'Big bang' efforts have tried to do too much too quickly"; the list is "the Global Digital Platform, Permissions Capability and the earlier Systems for People". They were funded by "efficiencies and budgetary offsets", and their failure produced "the harmful and circular takeaway message 'Government ICT is not worth investing in'."
- Lessons: "government should lead the delivery of new visa ICT systems" with industry as supplier; reform must create "a modular technology ecosystem, which enables each element of the system to be individually and progressively refreshed", sustaining existing systems in parallel, because "any ICT changes will need to be delivered in tranches over multiple years".
- Around 100 visa subclasses; subclass 408 alone has 21 streams; the Panel records departmental staff commenting on how the framework's complexity "limits the Department's capacity to automate assessments" [S1].
Read: the review already specified the delivery shape (modular, tranches, sustain-in-parallel, government-led). This POV adopts it rather than inventing a new one.
Home Affairs told the ANAO in November 2022: "IT systems remain a considerable impediment to the efficient and effective delivery of migration programs… core systems are over 30 years old and several programs still require paper-based visa lodgement", with "complexity and poor integration of its ICT systems which result in dispersed client information and inadequate support for decision-makers". Parent, Child and Other Family visas were paper-only. IRIS held 215,291 cases in October 2022, with offshore posts running local IRIS instances holding cases "dating back to 1993"; a three-year decommissioning began in 2022. The department had no consistent efficiency metrics [S3].
Unanimous first recommendation: the department "develop a long-term strategy" for its visa processing system to improve efficiency, reduce complexity, cut waiting times and give applicants transparency [S6]. Two years later the APSC found no executable ICT plan existed [S5].
The Integrated Cargo System went live on 12 October 2005 "more than two years late and more than eight times over budget" and ports faced "massive backlogs of freight" within days (ANZSOG case, 2007) [S21]. The ANAO's own audit of the project corroborates the substance: imports were scheduled for April 2003 and not implemented until October 2005, and the project "encountered considerable delays, cost overruns" and caused "substantial disruption to the movement of cargo at Australia's major ports and airports"; the "eight times" multiplier is the ANZSOG case's figure, not the ANAO's [S21]. The Seamless Traveller programme ($94m) trialled Vision-Box arrivals gates at Canberra and Perth in 2018 and paused them in 2019 as "not fit-for-purpose" [S15]. I cite these only because 2026 sources corroborate the pattern: the CRST paper calls the ICS "pre-digital", and the SmartGate fleet was only fully replaced in 2024. The findings themselves are old and should not be presented as current-state evidence.
Why the last three platforms died
The pattern across Systems for People (2006 to 2011), the Global Digital Platform (2017 to 2020) and the Permissions Capability (2020 to 2022) is consistent enough to treat as a design constraint rather than bad luck. The three are not the same kind of failure, and the differences are as instructive as the pattern.
| Programme | Ambition | What happened | Lesson the evidence supports |
|---|---|---|---|
| Systems for PeopleMay 2006 to 2010-11 · historical | Post-Palmer and Comrie: "one of the biggest business transformation programmes ever undertaken by the Australian Government", 250+ projects, role-based portals giving "a single view of a client's dealings with the department"; $495m over four years with IBM as strategic partner. Deliberately a wrap over ICSE, not a replacement [S24]. | Compliance, case management, detention, client search (113m records), biometrics and the Visa and Citizenship Wizards shipped; the visa-processing piece (Generic Visa Portal) came last, and at close in 2010-11 all subclasses were still "to be progressively rolled into the platform" over "the next two years". Budget reported at $598.1m (2007-08) and $659.1m including related projects (2008-09); a 2011 ANZSOG case calls it "more costly than initially planned" and quotes an interviewee: "One just does not talk about Systems for People" [S24]. | A wrap that leaves the core in place buys time and process discipline but not a modern platform; fifteen years on, ICSE is still the core [S1]. Fund the wrap only as a bridge with a dated retirement plan for what it wraps. |
| Global Digital Platform2017 to Mar 2020 | Outsource a ~$1bn platform to "process 90 percent of all visa applications" and let people "apply for visa at the point of booking travel" [S15]. | $92m spent on advisers and co-design before the tender was cancelled after a Senate inquiry called it effective privatisation via an "opaque" process [S15][S4 App. 3]. | Ownership of the core is a sovereignty and trust question, not just a commercial one. Parkinson: government leads, industry supplies. |
| Permissions CapabilityOct 2020 to Aug 2022 | A whole-of-government reusable "permissions" engine for visas, citizenship, customs and clearances, seeded with two use cases: a simple visa and a Digital Passenger Declaration [S4]. | Funding covered the DPD but not the simple visa. The DPD shipped in Feb 2022, rated 1.3 stars, was scrapped on 6 July 2022; 4 per cent of the base capability delivered; contract ended by a deed of agreement and release on 12 August 2022 [S4]; ship and scrap dates and the app rating are from press coverage [S15]. | A generic platform with no funded anchor use case has nothing to prove itself on. Sequence the anchor before the abstraction. |
| Seamless Traveller2015 to 2019 | "Zero-touch" arrivals using face as token, replacing the incoming passenger card [S15]. | Trial hardware not fit for purpose; reverted to incumbent supplier; Gen 3 gates only fully rolled out in 2023-24; paper card survived until the 2026 Australia Travel Declaration [S9][S13]. | Border hardware without the data and declaration layer behind it just moves the queue. |
| Funding model | Each was funded by "efficiencies and budgetary offsets" and each deprioritised the legacy estate while it ran; when they failed the department was "left not only with unmet expectations, but with a raft of other issues" [S1]. | Fund transformation and sustainment as separate lines. Never trade legacy upkeep for future savings. | |
Systems for People: what the record actually says
Parkinson groups Systems for People with the two later failures as a "big bang" effort that "failed to deliver the transformation that was promised" [S1]. The primary record supports a more precise verdict, and it matters because the programme was the last time the department attempted whole-of-enterprise change rather than a visa platform.
It was a wrap by design, not a big bang. A month before the 2006 Budget the Secretary said publicly that replacing the legacy was "too risky" and the plan was "preserving the existing systems but using modern technology to put over the top, and around it, a system that allows us to interrogate the systems as if they were one" [S24]. Judged against that intent, the wrap was delivered: portals for compliance, case management and detention (the specific Palmer and Comrie failures), a consolidated client search over 113 million records, centralised movement alerts, biometrics and identity management, and the award-winning Visa and Citizenship Wizards [S24].
Where it fell short was the visa core, which was always last in the sequence. The Generic Visa Portal arrived as the final component in 2010-11 with every subclass still to migrate; ICSE remained the onshore processing system, IRIS the offshore one, and both were still running in 2022-23 [S3][S24]. The cost climbed in the department's own annual reports from $495m to $598.1m to $659.1m (the later figures include related projects, so not strictly like for like), IBM's liability was capped at $150m, and the 2010-11 Budget added $169.6m to sustain it [S24].
The change outcome was credited; the technology outcome did not hold. The 2008 Proust evaluation gave the post-Palmer reforms "a positive score card" while warning that it was "too early to pass a verdict on Systems for People" and that management should watch whether it "remains within its funding envelope" [S24]. The department itself says the capabilities built (business-process-led development, service-oriented architecture, change management) became "core organisational capabilities" [S24]. What did not survive is the architecture: layering modern services over an unretired core is exactly what produces Parkinson's "layers of additional systems and layers of complexity" [S1].
One fairness note. Parkinson's footnote for the "failed" characterisation cites the ANAO's 2008 biometrics audit, which was published two and a half years before the programme closed, was scoped to biometrics, and does not assess Systems for People delivery. No ANAO performance audit of the programme itself was found in what was checked [S24]. The "failure" label rests on the long-run outcome (ICSE still core), not on an independent audit of the programme.
Recent corroboration: only two points have 2020+ support: that ICSE was still the main visa system in 2023 [S1], and that later reviews group the programme with the two later failures [S1]. Everything else above rests on 2006 to 2011 sources.
What is already moving (2024 to 2026)
The department is not standing still. These are the live pieces a transformation should absorb rather than restart.
- Migration Strategy, December 2023. Eight actions; Action 8 is "simplifying the migration system": "abolish unnecessary and duplicate visas", "embed simplification as a key objective of all actions", and median service standards of 21 days for the Skills in Demand visa and 7 days for its Specialist Skills Pathway [S11]. H
- Australia Travel Declaration. Piloted from October 2024 on Qantas flights into Brisbane, Sydney and Melbourne (450,000+ passengers); $56.1m over four years announced July 2026; every capital-city airport by end 2026, then all international air and sea ports over 12 to 18 months; web form first, app later, paper retained; QR pass works offline and is scanned by ABF; biosecurity and customs questions included [S13]. H
- SmartGate. Generation 3 gates at all airports in 2023-24; the eligible age recently lowered to 7+ for all travellers; eligible-traveller uptake 76.8 per cent; average inbound clearance 72.7 seconds [S9]. H
- Immigration Compliance Group established 27 February 2024 to manage risk "at all stages of a traveller's journey" [S8]; Nixon Recommendation 2 had called to "re-prioritise an immigration compliance function" [S2]. H
- Cargo Reporting System Transformation market sounding closed 3 July 2026 [S12]. H
- Budget 2026-27 (handed down 12 May 2026): $167.4m over four years "to strengthen migration integrity and border security" (secondary sources do not state the first year; on convention it runs from 2026-27) and $270m supplementary funding for ABF in 2026-27; $75.1m to modernise Trades Recognition Australia [S23]. M Neither the 2025-26 PBS nor the 2026-27 measures I found contain a core visa platform line item. H (absence claim, scoped to what was checked)
- Enterprise systems: SAP ECC6 replacement readiness study under way, target December 2030 [S14]. M
- Legal scaffolding: Digital ID Act 2024 commenced 1 December 2024 with the Australian Government Digital ID System open to Commonwealth and state entities [S20]; the Attorney-General's automated decision-making framework (Robodebt Royal Commission recommendations 17.1 and 17.2, both accepted) went to consultation November 2024 to January 2025 and has not yet been introduced as legislation in what I checked [S20]. H
Migration-industry sites report an ImmiAccount "live queue tracker" launched 18 March 2026 and an "AI-driven assessment platform" (one names a "Unified Digital Processing Environment") auto-granting low-risk cases from 25 March 2026, built "in partnership with Services Australia" [S22]. Site-restricted searches across homeaffairs.gov.au, the minister's site, the ANAO, iTnews, InnovationAus and The Mandarin return nothing for either claim, and the department's CIO in June 2026 placed it at "the start in the assisted space" [S14]. Graded L.
One thread does resolve, and it explains the rumour. An "Automation and Artificial Intelligence Strategy 2025-27" is real, but it belongs to Services Australia, not Home Affairs: the OAIC's January 2026 transparency reporting cites it as a Services Australia case study [S22]. Several industry sites appear to have attached a Services Australia strategy, and a Services Australia partnership, to Home Affairs. Treat any claim that Home Affairs has an AI visa-decision platform in production as unsupported until the department says so itself. If it turns out to be true, it belongs in Tranche 1 below; either way it shows why the department needs to own the public narrative on its own modernisation.
What other borders show
| Jurisdiction | What they did | Result | Implication for Australia |
|---|---|---|---|
| Singapore H | Passport-less clearance for residents and departing visitors across all Changi terminals from 30 September 2024, face and iris as the token [S19]. | Average clearance cut from 25 to 10 seconds; nearly 127 million passport-less clearances in 2025. | Proof that face-as-token works at scale when enrolment is done upstream and the traveller record is unified. Australia's Gen 3 kiosk-plus-gate still needs the chip. |
| European Union M | Entry/Exit System phased in from 12 October 2025 to full coverage 10 April 2026; ETIAS pre-authorisation to follow [S19]. | One-to-four-hour queues at peak, up to seven hours reported at Lisbon; outages and kiosk unavailability; 122 of 156 passengers missed a Linate to Manchester flight on 13 April 2026; nine member states wrote to the Commission in July 2026 asking to extend the partial-suspension flexibility, which runs to September 2026. | The cautionary case. First-arrival biometric enrolment at the gate, thin staffing and a hard cut-over produce exactly the queues the technology was meant to remove. Enrol before travel; keep the paper and officer fallback. |
| United Kingdom M | Digital-only eVisa status for all new visa holders from 15 July 2025; ETA for visa-exempt visitors from April 2025 [S19]. | The 3million's October 2025 report (1,877 user reports) documents wrong data on eVisas, failed document updates and locked accounts. Under the heading "Home Office system down for all ten million eVisa holders" it records sustained outages "lasting multiple days" in late August 2023, which the Home Office attributed to AWS rather than its own system. The report's estimate is that its 1,877 reports represent well under one per cent of those affected. | Digital status without a resilient offline proof and without a route to correct errors becomes a rights problem, not a service problem. Note also who carries the risk: an outage in a contracted cloud service lands on the visa holder at the boarding gate. Australia's VEVO is already digital-only; the target state needs an offline-verifiable credential and an error-correction SLA. |
| New Zealand H | New Zealand Traveller Declaration replaced the paper arrival card from August 2023 for air, October 2023 for cruise; customs, immigration and biosecurity in one declaration [S19]. | More than 2.8 million digital declarations by air travellers since launch, and over 58 per cent of New Zealand air travellers declaring digitally in the month to December 2024 (NZ Customs); Customs positions it as a biosecurity data asset and a targeting tool. | The closest comparator for the Australia Travel Declaration and the obvious partner for a trans-Tasman pre-cleared corridor in Tranche 3. |
Brisbane 2032: a deadline, not a volume shock
The Games are the fixed date this programme can be planned against, and the government has already used them as the justification for border spending. But the case for modernising by 2032 is not that the Games will overwhelm the border. On the Australian evidence, they will not. It is that they will put the border on television.
The link is official, not boosterism
The joint ministerial release funding the Australia Travel Declaration says so directly: the $56.1 million "will support border agencies in managing strong growth in traveller volumes post-Covid and ahead of the 2032 Brisbane Olympic and Paralympic Games" [S25]. Industry has made the same argument: in March 2026 the Tourism and Transport Forum, the Australian Airports Association and the Sydney, Melbourne and Brisbane airports jointly asked that Passenger Movement Charge growth revenue, forecast at more than $380 million by 2030, be held for border modernisation [S25]. The Games are therefore already functioning as the political deadline for this work, whether or not the volume case stands up.
The volume case does not stand up
Sydney 2000 is the only Australian precedent and the ABS recorded what happened. September 2000 brought 402,970 short-term visitor arrivals against 353,130 in September 1999: a 14.1 per cent rise, about 49,800 extra arrivals across the whole country, or roughly 13,400 a day. December 2000 then delivered 559,930 arrivals, 39 per cent more than the Olympic month. The Games month was not even that year's peak [S25].
Against today's baseline the increment is smaller still. Australia cleared 45.4 million air travellers in 2024-25 [S9]; a Sydney-2000-scale uplift is a rounding error against normal seasonal peaks that the border already absorbs every December. Brisbane Airport alone handled 7.08 million international passengers in 2025-26, up 7.9 per cent in one year [S25]: ordinary growth is adding more volume than the Games will.
And the tourism legacy is contested by the proponents' own evidence
The KPMG study prepared for the Queensland Government, itself a bid advocacy document, cites Giesecke and Madden (2007) finding "no support for the existence of an induced tourism effect" for Sydney 2000, and Baade and Matheson (2016) finding arrivals to the United Kingdom were six per cent lower in the London 2012 Games year, with a 30 per cent decline for Beijing [S25]. Paris 2024 fits the same pattern: 11.2 million people took part in Games activities in greater Paris but 85 per cent were French, foreign overnight visitors rose by roughly 150,000 on the prior year, and regional tourism over the period fell 4.7 per cent, with the Louvre down 22 per cent and Versailles down 25 per cent [S25]. Displacement, not surge.
Deloitte Access Economics' "$70 billion" is a net present value of a GDP opportunity over 2032 to 2052, self-published, disclosed as prepared independently of Deloitte's position as a Worldwide Olympic and Paralympic Partner, and explicitly distinguished by its authors from an impact estimate. KPMG's "$17.6 billion" is proponent-commissioned bid advocacy over 2022 to 2042. Both are claims about the Games' economy, not evidence about border volumes, and neither belongs beside the hard 2024-25 actuals above [S25].
So what does 2032 actually change
- It sets the date. Olympics 23 July to 8 August 2032, Paralympics 24 August to 5 September 2032 [S25]. Working back, the border must be demonstrably modern by early 2032, which means Tranche 2 complete and Tranche 3 under way before the Games, not because of them.
- It concentrates scrutiny. A queue at Brisbane in August 2032 is an international news story in a way that the same queue in December 2031 is not. Brisbane Airport's chief executive said in October 2025 that on current systems "we wouldn't be able to accommodate things like major events, including the Olympics, because we just don't have the capacity to process people in acceptable times" [S25]. That is a processing-capability statement, not a volume one, and it is the real risk.
- It shifts the peak into one gateway and one fortnight. The national increment is small; Brisbane's is not. Peak-hour arrival concentration at a single airport, plus accredited cohorts arriving together, is a scheduling problem the current model handles by adding officers.
- It creates a hard accreditation question nobody has answered publicly. Under the Olympic Charter the Olympic Identity and Accreditation Card operates with a passport in lieu of a visa. No published Australian arrangement for 2032 was found, nor any record of how Sydney 2000 was handled [S25]. That is a Tranche 2 design input and should be settled early, because it is exactly the kind of bespoke permission type a rules engine should absorb rather than a one-off workaround.
- Cargo needs no new mechanism. Games freight, broadcast equipment and event goods already fall under section 162A of the Customs Act, which permits temporary importation for up to 12 months duty-free against an ATA Carnet [S25]. Treat any claim of a Games cargo surge as unsourced until a logistics plan is published.
The honest framing for a business case: do not argue that the border will break under Games volumes, because the evidence says it will not and a reviewer will find that out. Argue that 2032 is an immovable date by which the work has to be finished anyway, that it removes the option of another decade of deferral, and that the cost of being visibly behind Singapore in front of a global audience is reputational rather than operational.
End-to-end process view
The border is one process with nine stages and two lanes (people and goods). Today each stage is owned by a different system, several by a different agency, and the traveller re-establishes who they are at almost every step. The target state is a single traveller record that accretes evidence from intent to departure, with every downstream stage reading from it instead of re-collecting.
| Stage | How it works today | Evidence of the gap | Target state |
|---|---|---|---|
| 1. Intent and eligibilityMigrant, visitor, employer | Around 100 subclasses; the website plus a "visa finder"; agents used for 15% of all applications and 65% of business visas. | Parkinson: complexity "necessitates the use of migration agents" and staff say it "limits the Department's capacity to automate assessments" [S1]. Migration Strategy commits to abolishing duplicate visas [S11]. | A guided eligibility service that resolves to a small set of permission types, pre-fills from a verified identity and prior history ("nudges" in Parkinson's words), and tells the applicant the expected decision date before they pay. |
| 2. Apply and decideImmiAccount, ICSE, IRIS, eMedical, biometrics | Online lodgement for most streams into ICSE; paper for some family streams; risk checks, watchlist, health and identity in separate systems (Parkinson Figure 60 maps ~70 of them); three change windows a year. | ANAO: "dispersed client information and inadequate support for decision-makers" [S3]. Nixon: identity anchored on applicant documents; biometrics in 53 countries only [S2][S17]. Bridging population 369k and rising [S8]. | A rules engine that holds visa criteria as configurable data, not code, so legislative change is a rules release not a system release; risk-tiered straight-through grant for low-risk streams with human decision on the rest; one client identity with a biometric anchor reused across every application. |
| 3. Book and check inAdvance Passenger Processing, PNR | Every international carrier must report passengers and crew through APP before boarding; PNR collected under the EU agreement and domestic law; the airline receives a board or do-not-board directive. | Works, and is the strongest asset in the chain. the OAIC assessed staff access to PNR data in 2025, with fieldwork in June [S19]. Global Digital Platform wanted "apply at the point of booking"; never built [S15]. | APP becomes the trigger that opens a "trip" on the traveller record; entitlement is checked against live status; carriers get an API, not a message queue; ETA-style permission offered in the booking flow for visa-free nationals. |
| 4. Declare before departureAustralia Travel Declaration | Paper incoming passenger card for most travellers; ATD pilot on Qantas into three cities; national rollout 2026 to 2028; up to 72 hours before departure; QR code, offline-capable, paper fallback [S13]. | 2022 DPD failed on usability and delivery [S4][S15]. Minister Burke on paper cards: data "on the third shelf in the fourth corridor of some shed" [S13]. | Declaration merges into the trip record and is pre-filled from it; biosecurity and customs questions become structured risk features; app and web share one backend; declaration state visible to the SmartGate so a clean declaration means no stop. |
| 5. In flight: pre-arrival riskABF targeting, Movement Alert List, DAFF | APP and PNR are risk-assessed while the aircraft is in the air; Movement Alert List and "model profiling" flag persons of interest; DAFF targets biosecurity separately. | Nixon: intelligence "not able to be shared under each agency's legislative frameworks" [S2]; ABF investigations capacity 120 [S2]. APSC: no data and analytics strategy [S5]. | One pre-arrival risk score per traveller fusing visa history, biometrics, APP, PNR and declaration, with DAFF as a co-owner; the score drives the gate outcome (self-clear, kiosk question, officer, secondary) before the aircraft lands. |
| 6. Arrive and clearSmartGate Gen 3, primary line, secondary | Kiosk plus gate using the ePassport chip photo; 61% of all air travellers self-clear; 72.7 seconds inbound average; ~1m maritime passengers on separate processes [S9]. | 21% of travellers ineligible for SmartGate; 23% of eligible travellers still go to an officer [S9]. Singapore clears in 10 seconds without a passport [S19]. EU shows what happens when enrolment is done at the gate [S19]. | Face as token for enrolled travellers (enrolled at visa grant, at prior arrival, or via Digital ID), passport as fallback; declaration and risk state already resolved so the gate is a confirmation not an interview; same model at cruise terminals. |
| 7. Stay: status and conditionsVEVO, PRISMS, sponsor obligations | VEVO gives employers and providers a live check; education providers report non-attendance through PRISMS into ICSE with automatic cancellation triggers; sponsors report changes; compliance is largely complaint-led. | Nixon: no compliance capability in the Immigration Group (since created) and no permanent cross-agency data-sharing mechanism on education providers [S2][S8]. UK eVisa outage shows the cost of digital-only status without resilience [S19]. | Status as a live entitlement with an offline-verifiable credential (Digital ID wallet), event-driven compliance (course withdrawal, employment cessation, tax data) that changes a risk score rather than triggering a form, and an error-correction pathway with a published SLA. |
| 8. Change: extend, switch, resolveOnshore applications, bridging, cancellation, ART | A new application from onshore creates a bridging visa; long queues in partner, parent and protection streams; merits review at the ART; ministerial intervention. | Nixon: pathway stacking "would add up to almost a decade"; 89% of onshore protection claims refused [S2]. Partner permanent 14 months at the 75th percentile [S8]. Bridging holders up 24% [S8]. | Pre-filled onshore applications from the existing record; decision-time commitments visible in the account; bridging visas issued with conditions computed from risk; review-body case data flowing back into the same record so the department and tribunal see one history. |
| 9. Depart and close the loopOutbound SmartGate, movement records since 1981 | Outbound SmartGate at 39 seconds [S9]; arrival and departure records dating from 1981 held in the Movements Reconstruction database; Passenger Movement Charge collected [S19]. | Overstay estimate (77,700) is "derived by matching visa grants, arrivals and departures across many departmental systems" and is "prone to error" because people "arrive and leave on different travel documents and under different names" [S8]. | Departure closes the trip on the same record it opened on, so the unlawful non-citizen figure is a count not a reconciliation; departure history feeds the next visa's risk score and the next arrival's gate eligibility; aggregate movement data becomes a reliable input to migration planning and to the outcome evaluation Parkinson found the system cannot currently support [S1]. |
| Goods laneICS, CRST, trade single window | 148m air and 7m sea consignments a year reported through the 2005 ICS; DAFF a co-user; 101,921 physical examinations with 21% enforcement yield [S9][S12]. | CRST: "pre-digital business model", 7 to 10 years to modernise [S12]. Simplified Trade System found 200+ regulations and 145 ICT systems across the trade environment; no single window yet [S19]. | Same pattern as people: one consignment record from pre-arrival report to release, shared risk with DAFF, single window on top. Runs as a parallel programme sharing the identity, data and integration layers built in Tranche 1. |
Design principles the evidence forces
- Government owns the core; industry supplies components. Parkinson's stated lesson from the Global Digital Platform, and the only model that survives Senate scrutiny [S1][S15].
- Identity and data before features. Nixon, Parkinson and APSC all locate the root cause in fragmented identity and absent data strategy [S1][S2][S5]. No stage in the journey improves until the traveller record exists.
- Modular, and each tranche retires something. Parkinson: "modular technology ecosystem… delivered in tranches over multiple years" [S1]. A tranche that adds a layer without switching one off is the pattern that produced 70 systems.
- Sustain the legacy in parallel, funded separately. Three failures were offset-funded and starved the estate they were replacing [S1]. Two lines: transformation and sustainment.
- Simplify the law in step with the build. Complexity "limits the Department's capacity to automate assessments" [S1]; the Migration Strategy's abolition of duplicate visas is a technical dependency, not a policy nicety [S11].
- Automate the low-risk majority, decide the rest by hand, and say so in law. Section 495A has permitted computer decisions since 2001; the post-Robodebt ADM framework is pending [S20]. Build to the framework before it lands: logged, explainable, reviewable, with a human on every refusal and cancellation.
- Enrol before travel, never at the gate. The EU's 2025-26 queues are the direct cost of gate-side enrolment [S19].
- Keep a fallback at every stage. Paper declaration, passport at the gate, officer at the line, offline credential in the wallet. The UK eVisa outage is what "digital-only" costs [S19].
- Measure the journey, not the system. Journey-level targets (time from intent to decision; share of travellers who touch no officer; overstay count accuracy) replace the per-programme measures the ANAO found inconsistent [S3] and the annual report cannot assure [S9].
Tranches 1, 2, 3
Each tranche is defined by what it lays down, what it visibly delivers to travellers and employers, and what it retires. Windows are indicative; the department's own cargo estimate of 7 to 10 years for one lane is the realistic envelope for both [S12].
Spine and visible wins
Lay down
- One client identity: master traveller record across visa, traveller, identity and compliance holdings; biometric anchor at grant and arrival; Digital ID System interoperability.
- Data platform and an endorsed data and analytics strategy (the APSC gap), with the shared risk features DAFF and ABF both read.
- API layer wrapped around ICSE (strangler pattern) so new services read and write through it while the core stays lit.
- The executable ICT and data plan, capital line and governance the APSC asked for; in-house core engineering team; contractor ratio brought down.
Deliver
- Australia Travel Declaration national rollout completed and wired to SmartGate and biosecurity targeting.
- Decision-date transparency and MFA in ImmiAccount (Nixon Rec 28).
- First permission type on the new rules engine: the already sub-one-day streams (visitor, ETA, Working Holiday), proving straight-through grant end to end.
- Nixon's cross-agency data matching on education providers made a permanent mechanism (Recommendation 15).
Retire
- IRIS and its offshore instances (finish the 2022 decommission); paper lodgement for the last family streams; the paper incoming passenger card at capital-city airports.
Permissions and pre-arrival at scale
Lay down
- Rules engine carrying the simplified visa framework (post duplicate-visa abolition); legislative change becomes a rules release.
- Case management for skilled, family and protection streams with risk-tiered allocation; sponsor, employer and education-provider portals on the same record.
- Single pre-arrival risk score fusing APP, PNR, declaration, visa history and biometrics, co-owned with DAFF.
- ADM assurance layer built to the Attorney-General's framework: decision logging, explanation, human review on every adverse outcome.
Deliver
- Straight-through grant for the low-risk majority across student and Skills in Demand; the Migration Strategy's 21-day and 7-day median service standards met by design rather than staffing surges.
- Face-as-token clearance for enrolled travellers at Gen 3 gates, passport fallback retained; SmartGate eligibility extended past the current 79 per cent.
- Cruise and maritime clearance on the same traveller record.
- Cargo: CRST discovery and pilot; consignment record and shared risk built on Tranche 1 data layer; single-window MVP for one commodity chain.
Retire
- Most of the 70-system visa ecosystem consolidated behind the API layer; separate DAFF and ABF pre-arrival targeting queues; the three-release-a-year change cycle.
Seamless and continuous
Lay down
- Status as a live entitlement: event-driven compliance, offline-verifiable credential in a Digital ID wallet, published error-correction SLA.
- ICSE decommissioned; the last ICSE-dependent streams migrated.
- Cross-agency identity reuse (Parkinson's myGovID point): biometrics and documents captured once, reused for tax, health, licensing.
- Movement and visa data feeding migration planning and the outcome-evaluation framework Parkinson found missing (Chapter 4) [S1].
Deliver
- Pre-cleared corridors with trusted partners (New Zealand first, then Singapore-style bilateral) where clearance completes before boarding and arrival is a walk-through.
- Permission offered inside the booking flow for visa-free nationals; a known-traveller lane for frequent movers.
- AI-assisted triage and evidence summarisation for officers on complex cases, inside the ADM framework, with the human decision preserved.
- Cargo Reporting System live; trade single window across agencies.
Retire
- ICSE; the paper passenger card everywhere; separate visa and traveller identities; the annual reconciliation that produces the overstay estimate.
Sequencing logic in one paragraph
On these windows, Tranche 1 runs to about 2028, Tranche 2 to about 2030 and Tranche 3 from 2030 onward, which places the Games inside Tranche 3. That is the right shape: the Games should land after straight-through processing and face-as-token clearance are already in production and boring, not during their rollout. The EU's experience of switching on new border checks close to peak travel is the case against doing it any other way [S19]. Working back from 23 July 2032, the last safe date to finish Tranche 2 is roughly mid-2031, leaving a full year of steady-state operation before the world arrives.
Tranche 1 is deliberately unglamorous because the three dead programmes all started with the glamorous part. It spends its visible-win budget on things already in flight (the declaration, ImmiAccount transparency) so the political sponsor has something to point at while the spine is built. Tranche 2 is where the money and the risk concentrate, and it is only safe because the rules engine can be proven on the sub-one-day streams in Tranche 1 and because the law has been simplified in parallel. Tranche 3 is the future-of-border story (passport-free, pre-cleared, continuous status) and is only credible if Tranche 2 has already retired the legacy core it would otherwise be layered on.
Counterarguments and risks
The strongest case against this POV
The department has buried three platforms in fifteen years, has no funded capital path, runs at 50:50 contractors, has just been told it cannot "deliver beyond maintenance", and is about to spend its scarce delivery capacity on an ERP replacement and a cargo system. A fourth large programme in that environment is more likely to fail than to succeed, and every failure raises the cost of the next attempt (Parkinson's "circular takeaway message"). The honest response is that the evidence does not support a large programme; it supports a small permanent capability that ships a retired system every year. If Tranche 1 cannot be funded as a distinct line with its own sustainment budget, the recommendation is to do only Tranche 1's spine work and the declaration, and stop.
Specific risks
- Funding reality. No core-platform appropriation found in 2025-26 or the 2026-27 measures I checked [S10][S23]. Moderate confidence there is none; it may sit in a measure I did not see.
- Facial recognition politics. The PJCIS sent the Identity-matching Services Bill back in 2019 for lacking privacy and oversight safeguards [S20]. Face-as-token at the border is narrower than that scheme, but the design must carry consent, retention limits (fingerprints are held 75 years under current authority [S17]) and independent oversight from day one.
- Automation law. The ADM framework is unlegislated. Building straight-through grant ahead of it is defensible only if the assurance layer is built as if the framework already applied.
- Queues at cut-over. The EU's experience is the base case for any gate-side change without upstream enrolment [S19].
- Digital exclusion. The UK eVisa outcomes and the 2022 DPD app both show what happens when the fallback is removed too early [S19][S15].
- Biosecurity is a second sovereign owner. DAFF co-owns the declaration, the cargo system and pre-arrival targeting; a Home Affairs-only programme will stall at the first shared interface [S12][S13].
- Procurement scrutiny. Richardson and the Permissions Capability audit mean every component buy will be examined for value for money and probity [S4][S16]; the commercial model should assume that and prefer many small competitive buys over one large deed.
Gap report
- Searched
- Parkinson, Nixon and Richardson reviews and government responses; ANAO Home Affairs portfolio audits 2022 to 2026 and the 2026-27 work programme; APSC capability review; Senate LCA (2022) and JSCM (2024) reports; Home Affairs annual reports 2023-24 and 2024-25, PBS 2025-26, the 15th edition programme report, the CRST discussion paper, the 2025 biometrics FOI release, ABF traveller-movement pages; ministerial and ACS reporting on the Australia Travel Declaration; iTnews and InnovationAus on GDP, Permissions Capability, DPD, SmartGate and current ICT; ABS arrivals and departures; Singapore ICA, EU EES coverage, UK eVisa reports, NZ Customs; Digital ID Act, Migration Act s495A, AGD ADM consultation, PJCIS 2019; Budget 2026-27 migration measures.
- Found
- A consistent four-review diagnosis (2023-24) with current-state corroboration to mid-2026; hard volumes for visas, travellers and cargo; the failure history with audited dollar figures; the live initiatives and their funding; four international comparators including two cautionary ones.
- Not found
- Any Home Affairs, ministerial, ANAO or technology-press confirmation of the March 2026 "AI-driven platform" and ImmiAccount queue tracker claims, on site-restricted searches of every outlet that would carry them. A visa-platform capital line in the 2025-26 PBS or the 2026-27 Budget measures. The first year of the $167.4m measure: every secondary source says "over four years" without naming it, and no primary Budget paper line was located. Current (2025-26) count of biometrics-covered visa subclasses (the country count of 53 is confirmed; the subclass count is not restated). On Systems for People, the budget year of the Seamless Traveller $94m, an independent restatement of the $598.1m intermediate budget figure, and free copies of ANZSOG case parts A and B, from which the "one just does not talk about Systems for People" interview quote and the 113 million client-record figure are drawn. Home Affairs' response to the APSC review beyond "an action plan has been developed".
- Contested claims
- Where a source is an advocate, the page says so: the KPMG and Deloitte Games studies are labelled as proponent-commissioned and self-published respectively, and their headline dollar figures are deliberately kept out of the statistics panel. The Games tourism-legacy claim is rebutted using the proponents' own cited literature rather than an outside critic.
- Old findings
- The Systems for People account (2006 to 2011) is built from contemporaneous primary sources (DIAC annual reports, Hansard, ANAO 2008) and is boxed as historical, with the two points that have 2020+ corroboration named explicitly. ICS 2005 launch failure and the 2015-19 Seamless Traveller trials are cited as history only, each paired with a 2024-26 source that corroborates the present-day consequence. No pre-2022 finding is relied on without such a pairing. An ANAO student-visa audit returned in search was DIAC-era and was excluded.
- Verification
- Every numbered, dated, quoted and attributed claim on this page was independently re-checked against the primary sources, and the corrections found have been applied. One claim was removed outright: an asserted UK "View and Prove" outage on 11 August 2025 could not be found in the report it was attributed to, whose actual finding is a multi-day outage in late August 2023 blamed on AWS; the page now says that instead. Three more are worth naming because they change what the page says: the APSC says "two attempts", not "two major initiatives", to modernise visa platforms; the backlog-and-modernisation substance sits at JSCM Recommendation 16, not 17, despite the ANAO's own page saying otherwise; and the "Automation and Artificial Intelligence Strategy 2025-27" that industry sites attribute to Home Affairs is in fact Services Australia's.
- Confidence
- High on diagnosis and volumes. Moderate on the funding position. Low on 2026 industry-media claims about AI processing, which this POV treats as unverified.
Sources
- S1H Parkinson, M. et al., Review of the Migration System: Final Report, March 2023, esp. Ch. 4, 18, 19. homeaffairs.gov.au
- S2H Nixon, C., Rapid Review into the Exploitation of Australia's Visa System, March 2023 (declassified); and Government Response. homeaffairs.gov.au
- S3H ANAO, Management of Migration to Australia: Family Migration Program, Auditor-General Report No. 16 2022-23. anao.gov.au
- S4H ANAO, Procurement of the Permissions Capability, June 2023, incl. Appendices 3 and 4. anao.gov.au
- S5H APSC, Department of Home Affairs Capability Review Report 2024, endorsed by the Australian Public Service Commissioner, May 2024. apsc.gov.au
- S6H Senate Legal and Constitutional Affairs References Committee, The efficacy, fairness, timeliness and costs of… family and partner reunion, March 2022; recommendation wording via HRLC (M).
- S7M Joint Standing Committee on Migration, Migration, Pathway to Nation Building, 2024 (73 recommendations); H ANAO potential audit Implementation of improvements to visa processing, 2026-27 work programme. anao.gov.au
- S8H Department of Home Affairs, The Administration of the Immigration and Citizenship Programs, 15th edition, October 2025. immi.homeaffairs.gov.au
- S9H Department of Home Affairs, Annual Report 2024-25 (Targets 63, 64, 65; cargo tables) and Annual Report 2023-24 (Target 33). homeaffairs.gov.au
- S10H Department of Home Affairs, Portfolio Budget Statements 2025-26, Programs 2.2, 3.2, 3.4. homeaffairs.gov.au
- S11H Australian Government, Migration Strategy, 11 December 2023 (overview); eight actions via KPMG flash alert (M).
- S12H ABF and DAFF, Cargo Reporting System Transformation: Discussion Paper, released 22 May 2026. homeaffairs.gov.au
- S13H Minister for Home Affairs, Digital passenger cards to streamline Australia's borders, July 2026; M ACS Information Age, Australia's digital arrival cards to roll out nationally, 13 July 2026. ia.acs.org.au
- S14M iTnews: Home Affairs' ICT modernisation needs backed by another review (27 May 2024); Home Affairs opens internal "conversation" on adopting three types of AI (9 June 2026); Home Affairs starts scoping SAP ECC6 replacement (27 July 2026).
- S15M iTnews, Home Affairs spent $92 million on failed visa platform outsourcing (May 2020); InnovationAus, Permissions denied (July 2022) and Digital passenger card shelved after scathing reviews (July 2022); Executive Traveller / InnovationAus on Seamless Traveller and the 2022 SmartGate replacement.
- S16H Richardson, D., Review of Integrity Concerns and Governance Arrangements for the Management of Regional Processing Administration, October 2023, and Government Response, February 2024.
- S17H Department of Home Affairs, Biometric Collection and Exchange, FOI release FA 25/09/01553 (2025). homeaffairs.gov.au
- S18H ABS, Overseas Arrivals and Departures, Australia, 2024-25 financial year. abs.gov.au
- S19 International: H Singapore ICA media release on passport-less clearance (2024) and 2025 volumes; M Euronews and ETIAS.com coverage of EES delays (Dec 2025 to Jun 2026); M the3million, The Digital Status Crisis, 28 October 2025 (advocacy source, 1,877 user reports); ITV News and Euronews on the Linate incident, 13 to 14 April 2026; Travel Tomorrow on the nine-state letter, July 2026; H NZ Customs, New Zealand paper arrival card goes digital (2023) and More New Zealanders embracing digital travel declarations (2025); H OAIC, Staff access and monitoring of access to passenger name records (fieldwork June 2025); H Australian Border Force, Monitoring people movement (Advance Passenger Processing, PNR, Movement Alert List, Movements Reconstruction database, Passenger Movement Charge); M Austrade STS summary of 200+ regulations and 145 ICT systems.
- S20H Migration Act 1958 s495A; Attorney-General's Department, Automated Decision-Making Reform consultation (Nov 2024 to Jan 2025); Department of Finance, Digital ID Act 2024 commencement 1 December 2024; M ABC News, Parliamentary security committee rejects identity-matching bill, 24 October 2019.
- S21historical ANZSOG case study, Cargo and culture: the Customs Integrated Cargo System (A), 2007; ANAO Audit Report No.24 2006-07, Customs' Cargo Management Re-engineering Project.
- S22L VisaHQ, VisasUpdate, TIA Lawyers and similar migration-industry sites on the 18 March 2026 ImmiAccount tracker and a 25 March 2026 "AI-driven" or "Unified Digital Processing Environment" platform. Single-source, unconfirmed by any government or technology-press outlet found.
- S23M Baker McKenzie, Budget Bites: Immigration Measures (May 2026); Refugee Council of Australia and Visa Envoy 2026-27 Budget analyses; Home Affairs PBS 2026-27 landing page.
- S25 Brisbane 2032: H ABS, Overseas Arrivals and Departures data cubes (Sydney 2000 monthly series; 2024-25 and 2025-26 actuals) and Year Book Australia 2002, "A look back at the Sydney Olympics and Paralympics"; joint media release of the Ministers for Trade and Tourism, Home Affairs, Infrastructure and Agriculture, Digital passenger cards to streamline Australia's borders, 13 July 2026; KPMG for the Queensland Department of Tourism, Innovation and Sport, Brisbane 2032: Preliminary economic, social and environmental analysis, 2021 (proponent-commissioned advocacy, cited here for the studies it reports); Deloitte Access Economics, Going for Gold, 22 July 2025 (self-published; Deloitte is a Worldwide Olympic and Paralympic Partner); Brisbane Airport Corporation newsroom, FY26 passenger results and Master Plan 2026 approval, 16 June 2026; Australian Olympic Committee and International Paralympic Committee Games dates; Australian Border Force, Temporary imports and Carnets (Customs Act 1901 s162A). M Travel Weekly, 14 October 2025 (Brisbane Airport chief executive on processing capacity); Australasian Leisure Management, 24 March 2026 (industry call on the Passenger Movement Charge) and 27 May 2026 (accommodation shortfall); The Sports Examiner, 26 August 2025, reporting the Paris je t'Aime follow-up report. Not found: any official international visitor forecast for the Games, any peak-day arrivals expectation, any published Australian arrangement for the Olympic Identity and Accreditation Card, and any Games cargo or customs plan.
- S24historical Systems for People: H DIAC Annual Reports 2006-07 (pp.13-14), 2007-08, 2008-09, 2009-10 (p.151), 2010-11; Senate Legal and Constitutional Legislation Committee Estimates Hansard, 22 May 2006; DIAC Question on Notice 81, Budget Estimates May 2009; ANAO, DIAC's Management of the Introduction of Biometric Technologies (2008); M 't Hart, P. with Tindall, K., Transforming an agency in crisis: driving change at the Immigration Department (A, B, C), ANZSOG cases 2011-120.1 to 120.3; Canberra Times, 3 April 2006 (reprinted in ANU Press, Australian Department Heads); The Australian, 7 February 2008; Delimiter, 18 April 2011; Refugee Council of Australia 2010-11 Budget summary. L Anonymous DIAC employee piece, Crikey, 15 May 2008 (paywalled, second-hand), not relied on.