The Single Apparatus

Human Services Procurement and the Privatisation of Judgement, 2026–2037

Surreal glass corridor with glowing lines and layered document forms — A surreal glass corridor carries glowing pathways through layers of documents and geometric architecture.

Working Paper– a work in progress

Authors: Andrew Klein & Sera Elizabeth Klein
Date: October 2026

Abstract

This working paper examines a proposition: that, by 2037, Australia could have a human-services system in which government remains the principal funder, but the practical authority to determine what people need, what support they receive, and how their needs are recognised is increasingly distributed among private providers, contractual frameworks, funding rules, assessment instruments and automated administrative processes.

The paper considers seven portfolios: disability services, aged care, veterans’ services, mental health, employment assistance, childcare and social housing.

These portfolios are not identical. They have different legislative histories, funding arrangements, professional practices and populations. Nor is there evidence presented here that a single organisation or group has coordinated their development towards one predetermined outcome. The proposition under examination does not require such coordination.

Instead, it asks whether similar institutional incentives and administrative approaches can produce convergent outcomes across otherwise separate systems.

The paper uses the term “privatisation of judgement” to describe a possible transfer of practical decision-making authority away from direct professional engagement with a person and towards funding classifications, eligibility rules, contractual performance requirements, assessment tools and other administrative mechanisms.

The central question is not whether governments should ever use private providers, standardised assessments or digital systems. Each can have legitimate purposes. The question is whether these arrangements preserve meaningful human judgement, transparency, accountability and the ability to respond to individual circumstances.

The paper distinguishes documented findings from the authors’ interpretations and from questions that remain unresolved. It identifies evidence that would strengthen the proposition and evidence that would weaken or disprove it.

The preliminary conclusion is that the proposition warrants systematic investigation, but it cannot responsibly be treated as proven across all seven portfolios.

1. Introduction: The Question Behind the Systems

The starting point for this paper is a concern raised in public discussion by the account known as @criprights: that, by 2037, Australia may have a largely government-funded human-services system in which practical control has moved increasingly into privately delivered services and administrative mechanisms.

The proposition deserves examination because it draws attention to a distinction that is often obscured in policy debate.

A government may continue to finance a service without retaining direct control over every decision that determines how the service operates in practice.

Public money may fund a program, while private organisations deliver the service. Contracts may define what those organisations are expected to achieve. Performance measures may influence which activities are rewarded. Assessment instruments may determine how need is classified. Eligibility rules may limit available support. Digital systems may structure which information is considered and how decisions are reached.

The result can be a situation in which government retains formal responsibility, but the practical experience of the person seeking assistance is determined by a chain of decisions made elsewhere.

This paper asks whether that pattern is emerging across Australia’s human-services portfolios.

The question is particularly important because these services deal with people whose circumstances may be complex, changing or difficult to represent through standard categories. A person seeking disability support, an older person requiring assistance at home, a veteran experiencing psychological injury, a parent seeking childcare or a family without secure housing does not necessarily fit neatly into an administrative model.

Standardisation can improve consistency. Contracting can bring additional capacity. Digital systems can reduce delays and make decisions more transparent.

But each can also create risks if the administrative process becomes more important than the human need it is intended to address.

The concern is not that every standardised process is harmful. It is that a system can become increasingly efficient at processing people while becoming less capable of understanding them.

1.1 The central hypothesis

The hypothesis is that seven human-services portfolios may be developing different versions of a common institutional pattern:

Government retains a substantial funding and policy role.
Service delivery is increasingly mediated through providers, contracts, eligibility rules and administrative systems.
The assessment and classification of need becomes increasingly important to the allocation of resources.
Providers and agencies operate within incentives that may not always align with the individual interests of service users.
Responsibility becomes distributed across institutions, making it more difficult to identify who is accountable when a person experiences harm.
People seeking support may find it increasingly difficult to challenge decisions, explain unusual circumstances or obtain an effective remedy.

These are propositions to investigate, not findings that can be assumed in advance.

1.2 What the paper does not claim

The paper does not claim that every private provider behaves improperly.

It does not claim that every government reform is designed to reduce entitlements.

It does not claim that the use of algorithms or digital assessment tools is inherently harmful.

It does not establish that the seven portfolios are controlled by one apparatus in the literal sense of a coordinated organisation.

Nor does it establish that harmful outcomes necessarily result from deliberate malicious intent.

The proposition is more limited, and potentially more consequential: different institutions may produce similar outcomes because they operate under comparable incentives, constraints and administrative assumptions.

If that proposition is wrong, the evidence should show it to be wrong.

If it is partly correct, the evidence should identify where, how and to what extent.

If it is correct, the implications for public accountability and the delivery of essential services are substantial.

2. Method: Separating Evidence from Interpretation

A working paper examining public institutions must distinguish what is documented from what is inferred.

Without that distinction, criticism can become indistinguishable from speculation. Equally, refusing to interpret documented findings would prevent meaningful examination of the institutional arrangements that may produce them.

The paper therefore uses three categories.

2.1 Documented findings

These are statements supported by primary sources or identifiable independent research.

Examples include:

findings and recommendations of Royal Commissions;
official government descriptions of eligibility and funding arrangements;
published assessment frameworks;
Australian Institute of Health and Welfare statistics;
Australian Bureau of Statistics estimates;
documented procurement and contracting arrangements;
published provider-performance measures;
independent investigations of administrative or assessment systems.

A Royal Commission’s finding should be attributed to that Commission. An official department’s description of a policy should be attributed to the department. A journalist’s account of an expert’s analysis should not be represented as an official government finding.

2.2 Interpretation

Interpretation begins when the authors ask what documented facts might mean.

For example, evidence that an assessment tool classifies people into funding categories may establish how a particular process operates. It does not, by itself, establish that the process is designed to deny support.

The authors may nevertheless ask whether the classification process adequately represents individual need, whether its incentives are appropriate, and whether affected people can challenge its decisions.

Those are legitimate questions, but they must be presented as analysis rather than as established fact.

2.3 Open questions

Some important matters remain uncertain and require further evidence.

These include:

the extent to which automated or rules-based decisions influence final outcomes;
whether private-provider concentration is increasing across particular portfolios;
how contract incentives affect service quality;
whether people with comparable needs receive comparable support;
how frequently people successfully challenge assessments;
whether administrative efficiencies translate into better outcomes for service users;
whether the seven portfolios are genuinely converging or merely share superficial similarities.

Where evidence is unavailable, the paper should say so.

2.4 The evidentiary principle

The paper adopts two principles.

First: evidence of a harmful outcome does not automatically establish deliberate intent to cause that outcome.

Second: the absence of proven malicious intent does not excuse a system that repeatedly fails the people it is intended to serve.

These principles are complementary. They allow institutional failure to be examined without inventing motives, while ensuring that the absence of evidence of conspiracy is not used to shield policy arrangements from scrutiny.

2.5 Falsifiability

The hypothesis must remain open to correction.

It would be weakened if evidence showed that professional judgement remained meaningful, that people could effectively challenge decisions, that provider concentration was not increasing, that digital tools improved outcomes without reducing accountability, or that the seven portfolios were moving in substantially different directions.

A thesis that cannot be challenged by evidence is not a useful research proposition.

The purpose of this paper is to establish a framework for investigation, not to declare the investigation finished.

3. The Purchaser State: Funding Without Direct Control

One of the central questions in contemporary public administration is the relationship between the funding of a service and the control of the decisions that determine how that service is delivered.

In a traditional public-service model, government may fund a service, employ the people delivering it, establish professional standards and retain direct managerial responsibility.

In a purchaser-provider model, these functions can be separated.

Government may determine the policy objectives, establish eligibility criteria, allocate funding and contract with providers. A private or non-government organisation may deliver the service. An assessment organisation may determine a person’s needs. A digital system may classify information. A regulator may oversee compliance. A review body may be responsible for considering complaints.

Each participant may fulfil its formal role while the person seeking assistance experiences the system as a single, often confusing institution.

This creates an accountability problem.

When the service fails, responsibility may be distributed among the government department that designed the program, the agency that administers it, the organisation that assessed the person, the provider that delivered the service and the rules governing the available funding.

The person experiencing the failure may have little practical ability to determine which institution is responsible for correcting it.

3.1 The difference between funding and judgement

Funding is not the same as judgement.

A government can fund a service while the practical determination of need is made through a series of contractual and administrative decisions.

Consider a hypothetical person seeking support.

The person explains their circumstances to an assessor. The assessor records information according to a prescribed framework. The information is translated into categories. Those categories are connected to a funding level. The funding level determines which services are available. A provider then delivers services within the resulting budget.

Every stage may be formally compliant.

Yet the final support package may fail to address the person’s actual circumstances.

The question is not simply whether the funding was allocated correctly under the rules. It is whether the rules and the process adequately captured the need they were supposed to measure.

This distinction is central to the idea of the privatisation of judgement.

3.2 What is being privatised?

The term privatisation is often used to describe the transfer of publicly owned assets or services into private ownership.

This paper uses the term in a more specific analytical sense.

The privatisation of judgement occurs when the practical authority to determine what a person needs and receives moves away from direct, accountable human assessment and towards contractual, financial, classificatory or automated mechanisms.

That transfer may occur even when the government continues to pay for the service.

It may also occur within a public agency. The relevant distinction is not always public versus private ownership. It is the relationship between administrative authority, professional judgement and the person affected by a decision.

A public servant following a rigid classification system may have little discretion. A private professional may exercise considerable independent judgement. The institutional arrangements must therefore be examined rather than assumed.

3.3 The role of metrics

Performance measures are not inherently inappropriate. Public programs require ways to assess expenditure, quality, timeliness and outcomes.

The problem arises when a measure becomes a substitute for the purpose it was intended to represent.

A service provider may be rewarded for meeting a target that does not adequately measure whether a person has received useful assistance. An agency may reduce processing times without improving the quality of decisions. An assessment process may produce consistent classifications while failing to recognise unusual or changing needs.

In such circumstances, an organisation may become more successful according to its performance measures while the people it serves experience little improvement.

This is a familiar risk in any institution governed by narrow indicators. It must, however, be demonstrated in each portfolio rather than assumed.

3.4 The principal-agent problem

A related analytical concept is the principal-agent problem.

Government, acting as purchaser, seeks to achieve public objectives through an organisation that delivers the service. The provider may have its own financial, operational and institutional interests.

The interests of government, the provider and the service user may overlap, but they are not necessarily identical.

Contracts and oversight mechanisms attempt to align those interests. Whether they succeed depends on their design, the information available to government, the enforceability of standards and the quality of independent monitoring.

This does not mean that private provision inevitably fails. It means that the institutional arrangement creates questions that cannot be answered by examining expenditure or formal compliance alone.

A serious assessment must ask:

What does the contract reward?
What does it penalise?
What outcomes are measured?
What information is unavailable to the purchaser?
What happens when the provider’s commercial interests conflict with the service user’s needs?
Who can intervene when the contract is being followed but the intended outcome is not being achieved?

These questions apply to public and private institutions alike.

4. Seven Portfolios: Seven Different Mechanisms

The seven portfolios considered here are not interchangeable examples of one uniform process.

The working proposition becomes stronger, not weaker, if the analysis identifies the specific mechanism operating in each sector.

The following sections distinguish the available evidence from the interpretation that requires further testing.

4.1 Disability services: Displacement of individual need

The National Disability Insurance Scheme (NDIS) was established to provide support to eligible Australians with disability. Its scale, complexity and reliance on a network of participants and providers make it an important case for examining the relationship between individual needs, administrative processes and funding decisions.

Two major inquiries provide essential background:

the Independent Review of the National Disability Insurance Scheme, which published Working together to deliver the NDIS;
the Royal Commission into Violence, Abuse, Neglect and Exploitation of People with Disability, which published its final report in September 2023.

These inquiries provide substantial primary material for examining the scheme’s design, implementation, safeguards and experiences of people with disability.

The question for this paper is whether administrative and funding arrangements can displace individual need as the central organising principle of support.

The hypothesis would be supported if evidence showed, for example, that people with comparable needs experienced materially different outcomes because of administrative classifications, that decisions were difficult to challenge, or that provider and funding arrangements systematically undermined the scheme’s stated objectives.

It would be weakened by evidence showing that decisions reliably reflected individual circumstances, that participants could obtain effective reviews, and that administrative changes improved both consistency and outcomes.

The existence of problems in the NDIS would not, by itself, establish that the scheme was designed to fail. Nor would evidence of expenditure growth or provider profit, without more, establish that those profits caused particular harms.

Those claims require evidence about institutional design, financial incentives, decisions and outcomes.

Evidence base: the Independent NDIS Review; the Disability Royal Commission; legislation and administrative guidance; published evaluation and review data; procurement and provider information; and the experiences of participants.

The authors’ previously published critical commentary may help identify questions for investigation, but it should be treated as the authors’ interpretation, not as independent confirmation of the hypothesis.

4.2 Aged care: Classification and the allocation of support

Aged care provides a particularly important setting for examining the relationship between assessment, classification and funding.

The Royal Commission into Aged Care Quality and Safety delivered its final report on 1 March 2021. Its findings and recommendations provide a primary source for understanding longstanding concerns about Australia’s aged-care arrangements.

The transition to the Support at Home program creates a further question: how does the assessment of an older person’s needs translate into a practical level of funded support?

The Australian Government publishes information about the Support at Home classification system and the assignment of ongoing home-support classification levels.

These official materials should be examined alongside independent reporting and expert analysis of the assessment process.

An ABC investigation published in August 2026, titled Inside the aged care algorithm deciding support for older Australians, reported concerns raised by Professor Kathy Eagar about how information gathered through the Integrated Assessment Tool was used in determining support classifications.

The reporting raised questions about which assessment responses influenced the algorithm, how the decision process operated and whether small differences at particular decision points could produce substantial differences in funding.

Those are important issues, but attribution matters. An expert’s analysis reported by a news organisation is not automatically an official finding, and the details of a particular assessment process must be checked against the relevant technical documentation.

The authors therefore identify three questions:

Representation: Does the assessment adequately capture the older person’s circumstances, including needs that are difficult to express through standardised questions?
Translation: How is the information collected during assessment converted into a classification and funding level?
Review: Can an older person or their representative understand and challenge a decision that appears inconsistent with the person’s actual needs?

The concern is not that every algorithm is unsuitable for public administration. Properly designed systems can improve consistency and reduce arbitrary variation.

The concern is that a classification mechanism may become decisive without being sufficiently transparent, clinically appropriate or open to effective review.

A system may process information consistently while still producing an inadequate outcome if the information used, the decision rules or the available categories do not properly reflect the person being assessed.

Preliminary interpretation: aged care may provide a particularly useful case study of the relationship between standardised assessment and the allocation of public resources.

What remains to be established: the precise operation of the relevant decision rules, the extent of their influence on final funding decisions, their validation, their treatment of complex circumstances and the effectiveness of review arrangements.

4.3 Veterans’ services: Institutional layering

The Royal Commission into Defence and Veteran Suicide delivered its final report on 9 September 2024, containing 122 recommendations.

Its findings and recommendations are central to any serious examination of how Australia’s systems respond to current and former Defence personnel experiencing distress, injury or other difficulties.

Veterans’ services can involve multiple institutions and processes: Defence, the Department of Veterans’ Affairs, healthcare professionals, rehabilitation providers, compensation arrangements and other support organisations.

The existence of multiple institutions is not itself evidence of failure. Complex needs may legitimately require specialist organisations and different forms of assistance.

The question is whether the interaction between those institutions creates barriers that no single institution is accountable for resolving.

A person may have a recognised need but still face difficulty navigating eligibility requirements, medical assessments, compensation procedures and separate service arrangements.

If responsibility is distributed across institutions, each may complete its assigned task while the person remains without adequate assistance.

This paper describes the potential mechanism as layering.

Layering occurs when successive administrative and institutional requirements accumulate around a person without a corresponding improvement in coordination or outcomes.

To test this proposition, the research must examine:

how responsibilities are divided;
how information moves between institutions;
whether a person must repeatedly provide the same information;
whether decisions are coordinated;
whether there is a clearly identifiable authority responsible for resolving gaps;
whether the recommendations of the Royal Commission have been implemented and whether implementation has improved outcomes.

The Royal Commission’s report establishes an essential evidence base for the inquiry. It does not, by itself, prove every proposed connection between institutional layering and individual outcomes.

Preliminary interpretation: veterans’ services may demonstrate how accountability can become fragmented when multiple systems respond to different parts of a person’s circumstances.

What remains to be established: which administrative arrangements produce measurable barriers, how frequently those barriers occur, and whether procurement or provider arrangements materially contribute to them.

4.4 Mental health: The risk of confusing funded access with clinical need

Australia’s Better Access initiative provides Medicare benefits for eligible mental-health services, subject to the program’s rules. The published arrangements include limits of up to 10 individual and 10 group psychological therapy sessions per calendar year for eligible patients, subject to the applicable requirements.

This is a documented feature of the program.

The interpretive question is whether the structure of the funding arrangement adequately accommodates the different needs of people seeking treatment.

People experiencing psychological distress do not all require the same number of appointments. Some may benefit from a relatively short intervention. Others may require longer treatment, different forms of support or coordination across several services.

A funding limit is not automatically evidence of poor policy. Governments must make decisions about the allocation of finite resources, and treatment duration should not be assumed to be unlimited or identical for every person.

However, a standardised funding arrangement may create tension when the administrative limit does not correspond to the clinically appropriate course of treatment.

The research should examine whether the available evidence shows that funding arrangements affect:

continuity of care;
access to treatment;
waiting times;
treatment completion;
out-of-pocket costs;
access to alternative services when funded sessions are exhausted;
outcomes for people with more complex needs.

The paper must not assume that the session limit causes poor outcomes merely because the limit exists. That causal claim would require appropriate evidence.

It must also avoid assuming that all unmet mental-health needs arise from funding limits. Workforce shortages, geographic disparities, affordability, service availability and other factors may contribute.

The proposed mechanism is capping: a standardised funding limit may constrain the duration or form of assistance available to an individual.

What remains to be established: the actual effect of funding limits on clinical outcomes, how often alternative support is available, and whether the effects differ across populations and regions.

4.5 Employment assistance: Substitution of measurable activity for meaningful outcomes

Employment-assistance programs aim to help people obtain work and improve their participation in the labour market.

Australia’s Workforce Australia arrangements provide a case study in the use of providers, performance ratings and administrative measures to assess service delivery.

The Department of Employment and Workplace Relations publishes Workforce Australia Services Provider Performance Ratings. The June 2026 material provides a documented example of the government’s use of performance measures to assess providers.

The existence of ratings does not establish that they are inappropriate. Public purchasers need information about provider performance, and transparent measures may help identify poor service.

The central question is what the measures actually reward and how accurately they represent the outcomes that matter to participants.

A provider’s performance may be assessed through measures that are administratively convenient to collect. Yet meaningful employment outcomes can depend on factors that are difficult to capture through simple indicators, including a person’s health, skills, caring responsibilities, local labour-market conditions and the availability of suitable work.

If a performance system rewards measurable activity without adequately accounting for the quality or durability of employment outcomes, providers may face incentives that do not fully align with participants’ needs.

That possibility must be investigated rather than presumed.

The research should establish:

the actual performance measures used;
the weight given to each measure;
the contractual consequences of different ratings;
the extent to which ratings influence provider funding or continued participation;
how outcomes are adjusted for participant circumstances;
whether participants report meaningful improvements;
whether independent evaluations support the claimed effectiveness of the program.

The relevant question is not whether performance measurement should exist. It is whether the chosen measures are valid indicators of the program’s public purpose.

The proposed mechanism is substitution: a measurable administrative indicator may substitute for a more complex outcome that is difficult to measure.

What remains to be established: whether this occurs in practice, how it affects provider behaviour, and whether alternative performance frameworks produce better results.

4.6 Childcare: Market dependence and geographic inequality

Childcare differs from disability support, aged care and veterans’ services because access depends not only on eligibility and funding but also on the availability and location of services.

The Australian Competition and Consumer Commission’s Childcare Inquiry examined prices, costs, profits, competition and the operation of the childcare market.

Its findings and recommendations provide an important starting point for examining whether market arrangements deliver the outcomes expected by families and governments.

The inquiry identified deficiencies in how the market served some households and communities, including issues relating to underserved markets and the unequal effects of the system on lower-income families.

The Australian Government’s Three Day Guarantee, which commenced on 5 January 2026, is another relevant policy development. Its operation should be considered alongside the supply of childcare places, affordability and the geographical distribution of services.

The central issue is that a subsidy cannot, by itself, create a childcare place where no viable service exists.

A family may be eligible for financial assistance but still be unable to obtain a suitable place because of local shortages, operating hours, transport, workforce availability or the cost of providing services in a particular area.

The market may respond to demand where demand is financially viable while providing insufficient services in locations where the costs are higher or the customer base is less profitable.

That possibility is not unique to childcare. It is a general risk in services where public policy relies on private or market-based supply to achieve universal or geographically broad access.

However, the degree to which this occurs must be established through regional and provider-level evidence.

The research should compare:

the distribution of childcare places;
local prices and household costs;
provider entry and exit;
workforce availability;
waiting lists and unmet demand;
the effects of subsidy rules;
the effectiveness of supply-side measures in underserved areas.

The proposed mechanism is market-location dependence: access depends partly on whether a service can be supplied on commercially viable terms in the location where it is needed.

What remains to be established: how large the geographic and socioeconomic gaps are, how they change over time, and which policy arrangements most effectively reduce them.

4.7 Social housing: Fragmentation between need, eligibility and supply

Social housing is different from the other portfolios because the availability of a service depends heavily on the existence of a physical resource: an appropriate dwelling in an appropriate location.

The Australian Institute of Health and Welfare’s Housing Assistance in Australia 2026 reports that approximately 206,000 households were on the consolidated social-housing waiting list in June 2025.

The AIHW also reports on households living in the principal social-housing programs.

Separately, the Australian Bureau of Statistics estimated that 122,494 people were experiencing homelessness on Census night in 2021.

These figures describe different things.

The waiting-list figure refers to households recorded on a consolidated social-housing waiting list at a particular date. The homelessness estimate refers to people classified as experiencing homelessness on Census night in 2021. The figures must not be added together or treated as interchangeable measures of the same population.

Nor should the 2021 homelessness estimate be represented as a current count.

The AIHW cautions that waiting-list numbers are affected by eligibility rules and allocation policies. A waiting list is therefore not a simple, complete measure of underlying housing need.

These qualifications are important because the interpretation must not exceed the evidence.

The broader question is how responsibility for housing assistance is distributed across governments, community-housing organisations, other providers, eligibility systems and funding arrangements.

Where a household meets the relevant criteria but no appropriate dwelling is available, the problem cannot be solved by administrative processing alone.

A household may be assessed, categorised and placed on a list without receiving a home. The administrative process may be functioning according to its rules while the underlying need remains unmet.

This paper describes the potential mechanism as splitting: the assessment and administration of need may be separated from the availability of the material resource required to meet that need.

This is not necessarily a consequence of privatisation. Housing shortages, land costs, planning decisions, construction capacity, public investment and allocation rules may all be relevant. The analysis must examine those factors rather than attribute every shortfall to one institutional cause.

The research should investigate:

the supply of social housing relative to assessed need;
waiting times and allocation rules;
differences between states and territories;
the role of public and community-housing providers;
the distribution of funding and ownership;
the availability of appropriate housing for people with disability, older people and families;
the relationship between eligibility decisions and actual housing outcomes.

The proposed mechanism is splitting: the system may be able to identify and classify a person’s need without possessing the resources or authority necessary to meet it.

What remains to be established: the contribution of procurement and provider arrangements to unmet housing need, compared with shortages in supply and other structural factors.

5. Comparing the Seven Portfolios

The comparison becomes more useful when it identifies the mechanism under investigation in each portfolio rather than treating every problem as evidence of the same cause.

Portfolio	Proposed mechanism	Central question	Evidence required
Disability services	Displacement	Does administration or funding classification displace individual need?	Participant outcomes, decision rules, review data, provider arrangements
Aged care	Classification	Does the assessment-to-funding process accurately represent individual need?	Assessment rules, algorithm documentation, validation, review outcomes
Veterans’ services	Layering	Do multiple institutions create gaps in coordination and accountability?	Royal Commission follow-up, service pathways, implementation and outcome data
Mental health	Capping	Do standardised funding limits constrain appropriate treatment?	Clinical outcomes, access data, waiting times, evaluation of funding rules
Employment assistance	Substitution	Do performance measures represent meaningful employment outcomes?	Contracts, rating methodology, participant outcomes, independent evaluations
Childcare	Market-location dependence	Does reliance on market supply leave some communities without adequate access?	Regional supply, prices, provider concentration, unmet demand
Social housing	Splitting	Is assessed need disconnected from the availability of appropriate housing?	Waiting times, supply, allocation rules, funding and provider data

These mechanisms are related, but they are not identical.

Aged care raises questions about how assessments are translated into funding. Childcare raises questions about the geographical distribution of supply. Social housing raises questions about the availability of a scarce physical resource. Employment assistance raises questions about whether performance measures reflect the public purpose of a program.

The fact that all seven portfolios can be examined through the relationship between administrative arrangements and individual outcomes does not prove that they share one underlying cause.

The proposed commonality is narrower: in each portfolio, we should investigate whether the administrative and institutional arrangements preserve or weaken the connection between a person’s actual circumstances and the assistance ultimately provided.

That is the point at which comparison becomes analytically useful.

5.1 A common structure without a common conspiracy

A system can exhibit common characteristics without being controlled by a single coordinating authority.

Different departments may independently adopt contracting models, standardised assessments, performance targets and digital systems because these approaches are promoted as ways to improve efficiency, consistency, accountability or cost control.

Providers may respond to the financial and operational incentives created by those arrangements. Departments may respond to budgetary pressures and demands for measurable performance. Governments may favour approaches that are easier to explain, monitor and scale.

These developments can reinforce one another even where no central plan exists.

This is a hypothesis about institutional convergence, not evidence of secret coordination.

To establish whether convergence is occurring, the research must compare the actual arrangements across portfolios. Similar language in policy documents is insufficient. The comparison needs to examine funding, contracts, decision-making authority, data systems, performance incentives, review rights and outcomes.

5.2 The danger of forcing the evidence

A common analytical framework should help identify differences as well as similarities.

If the evidence shows that one portfolio is improving professional discretion while another is reducing it, that difference must be retained.

If private provision improves outcomes in one setting but weakens accountability in another, both findings must be reported.

If a digital assessment tool demonstrably improves the accuracy and consistency of decisions without reducing meaningful review rights, the paper should acknowledge that result.

The purpose of the framework is not to make seven portfolios fit a predetermined theory. It is to test whether a common pattern exists and identify where the proposed pattern fails.

6. Financial Incentives, Procurement and Accountability

The next stage of the investigation concerns the relationship between public funding, provider incentives and the allocation of responsibility.

It is not sufficient to show that private providers operate within a publicly funded system. The analysis must establish what those providers are contracted to do, how they are paid, how performance is assessed and what happens when the intended public outcomes are not achieved.

6.1 Follow the money, but do not stop there

Financial analysis is necessary, but expenditure alone cannot explain institutional behaviour.

The research should identify:

the total public expenditure associated with each portfolio;
the proportion spent on direct service delivery;
the proportion spent on administration, assessment, coordination and intermediary services;
the value and duration of major contracts;
the distribution of contracts between providers;
the ownership structures of significant providers;
the contractual relationship between payments and performance;
the costs incurred by people using the service;
the mechanisms through which poor performance can be challenged or remedied.

The objective is to establish where public money flows and what public value is delivered in return.

A private provider receiving public funding is not, by itself, evidence of extraction. Nor does a provider making a profit establish that the service is ineffective.

The relevant questions concern the relationship between payment, risk, performance, service quality and outcomes.

A provider may deliver substantial public value while earning a reasonable return. Conversely, a provider may comply with its contract while the contract itself fails to protect service users adequately.

The latter possibility is especially important. Contractual compliance and successful public policy are not necessarily the same thing.

6.2 The architecture of accountability

The paper proposes examining accountability through five questions.

First: Who defines need?

Is need determined by a professional who can consider individual circumstances, by a standardised assessment, by an administrative eligibility rule, or by some combination of these?

Second: Who determines the available resources?

Does the person receive support according to assessed need, a funding classification, a contractual entitlement or the availability of a service?

Third: Who delivers the service?

Is delivery undertaken by a public agency, a private company, a not-for-profit organisation or a network of providers?

Fourth: Who measures success?

Are outcomes assessed through the experiences of service users, independent evaluations, administrative indicators, financial measures or provider-reported results?

Fifth: Who is responsible when the outcome is inadequate?

Can the person identify an institution with both the authority and the obligation to resolve the problem?

These questions help distinguish formal responsibility from practical accountability.

A system may have numerous regulators, review bodies and complaint channels while still leaving a person unable to obtain a timely and effective remedy.

That possibility must be investigated through the actual operation of the system, not inferred merely from its complexity.

6.3 Data and the allocation of authority

Digital administration raises a further issue: who controls the information that determines a person’s access to support?

A person’s circumstances may be recorded by one institution, assessed by another, processed through a government system and used by a provider to determine the services that can be delivered.

The person may not know which information was decisive or how to correct a mistake.

This does not mean digital systems should be rejected. Well-designed systems can make decisions more consistent, provide useful audit trails and improve the accessibility of information.

The question is whether the system makes decisions more transparent and accountable or merely makes the administrative process more difficult for an individual to challenge.

The investigation should therefore examine:

what information is collected;
who can access it;
how information is translated into decisions;
whether automated or rules-based processes are used;
whether the decision logic is documented;
whether errors can be corrected;
whether affected people can obtain an understandable explanation;
whether there is an effective route to human review;
whether independent oversight examines the system’s performance.

The paper treats these as questions requiring evidence, not as established defects across all seven portfolios.

6.4 The Robodebt lesson

The Royal Commission into the Robodebt Scheme provides an important Australian case study of administrative decision-making, legal responsibility, the use of data and the consequences of institutional failure.

It is relevant because it demonstrates why administrative systems must be examined in terms of legality, evidence, accountability and the treatment of affected people.

However, Robodebt should not be used as proof that every contemporary digital assessment or automated administrative process is equivalent to that scheme.

The appropriate lesson is methodological: when a system uses rules, data and administrative processes to make decisions affecting people’s lives, its legal basis, evidentiary foundations, safeguards and review mechanisms require careful scrutiny.

The comparison should be based on demonstrated similarities and differences, not rhetorical association.

7. Testing the 2037 Hypothesis

The proposition that Australia could develop a largely government-funded but increasingly privately mediated human-services system by 2037 is a prediction about institutional direction.

To assess it, the research needs a baseline against which change can be measured.

Without a baseline, statements about increasing privatisation, concentration or automation may remain impressions rather than findings.

The following six indicators provide a framework for testing the hypothesis.

7.1 Indicator One: Funding distribution

The research should establish how public expenditure is distributed between government-delivered services, private providers, not-for-profit organisations, assessment bodies, administrative contractors and other intermediaries.

It should then examine changes over time.

Relevant evidence includes government budget papers, annual reports, procurement records, contract disclosures and available financial statements.

The analysis must distinguish expenditure on service delivery from expenditure on administration and assessment. It must also account for differences in how each portfolio records and reports costs.

Evidence supporting the hypothesis: a documented increase in the proportion of expenditure channelled through external providers or intermediary arrangements, particularly where this coincides with reduced direct public capacity or diminished public control.

Evidence weakening the hypothesis: stable or declining externalisation, increased direct public capacity, or evidence that external provision improves outcomes without weakening accountability.

An increase in private expenditure alone would not establish a harmful transfer of judgement.

7.2 Indicator Two: Provider concentration and ownership

The research should examine whether contracts and service delivery are becoming concentrated among a smaller number of organisations.

It should identify ownership structures, related entities, market entry and exit, contract renewals and the distribution of public funding.

The purpose is not to assume that large providers are necessarily harmful. Larger organisations may have economies of scale, specialist expertise and the capacity to deliver services across wide areas.

The question is whether concentration reduces meaningful choice, increases dependence on particular providers, weakens competitive pressure or makes the system more difficult for government to supervise.

Evidence supporting the hypothesis: increasing concentration combined with weaker choice, limited transparency or reduced public capacity to intervene.

Evidence weakening the hypothesis: effective competition, meaningful alternatives, transparent ownership and contracting, and demonstrated improvements in outcomes.

7.3 Indicator Three: Automated and rules-based decisions

The research should establish where assessments, classifications, eligibility decisions and funding allocations are influenced by automated systems or rigid rules.

It should distinguish between systems that merely store information, systems that assist professional judgement and systems that determine or substantially constrain the available outcome.

These are different functions and should not be conflated.

Relevant evidence includes technical documentation, procurement specifications, assessment guidance, validation studies, audit reports and information about human-review arrangements.

Evidence supporting the hypothesis: growing reliance on automated or rules-based decisions that materially constrain individual judgement, particularly where the logic is opaque or review rights are weak.

Evidence weakening the hypothesis: systems that improve consistency and accuracy while preserving meaningful professional discretion, transparent explanations and effective review.

7.4 Indicator Four: Professional discretion and review rights

The research should examine whether the people making decisions have the authority and capacity to respond to circumstances that fall outside standard categories.

It should also examine whether affected people can challenge decisions effectively.

The relevant questions include:

Can assessors consider information that does not fit the standard framework?
Can decision-makers depart from a default outcome when justified?
Must reasons be given?
Can a person obtain an independent review?
Are review processes timely and accessible?
Can a successful review produce a practical change in the assistance provided?

Evidence supporting the hypothesis: reduced discretion, weak review mechanisms, or evidence that formal review rights rarely produce meaningful remedies.

Evidence weakening the hypothesis: stronger professional discretion, accessible review processes and demonstrated correction of inappropriate decisions.

7.5 Indicator Five: Data governance and transparency

The research should establish who controls the data used to assess people, how that data is processed and whether the decision-making process can be examined independently.

The existence of a digital system is not itself evidence of reduced accountability.

The relevant issue is whether digital administration makes the decision more understandable, accurate and contestable, or whether it creates a process whose operation cannot be meaningfully explained to the person affected.

Evidence supporting the hypothesis: increasing dependence on systems that cannot be adequately explained or audited, especially where the resulting decisions materially affect access to essential services.

Evidence weakening the hypothesis: transparent rules, effective auditing, clear explanations, accessible correction mechanisms and evidence of improved decision quality.

7.6 Indicator Six: Outcomes for people

This is the most important indicator.

The paper should not evaluate the architecture of a human-services system solely by examining its contracts, funding arrangements or administrative efficiency.

It must ask whether people receive the support they need.

Relevant outcomes vary by portfolio.

For disability services, the inquiry might examine access to appropriate support and participants’ experiences of decision-making.

For aged care, it might examine whether assessed needs translate into suitable care and whether older people can challenge inappropriate classifications.

For veterans’ services, it might examine access, continuity, coordination and the effectiveness of assistance.

For mental health, it might examine access to appropriate treatment, continuity of care and outcomes.

For employment assistance, it might examine the quality and durability of employment rather than administrative activity alone.

For childcare, it might examine affordability, availability and geographical access.

For social housing, it might examine access to appropriate accommodation, waiting times and housing outcomes.

Evidence supporting the hypothesis: deterioration or persistent inadequacy in relevant outcomes associated with identifiable institutional arrangements, despite administrative or financial claims of success.

Evidence weakening the hypothesis: sustained improvements in outcomes, effective accountability and evidence that the arrangements under examination contribute to those improvements.

No single indicator can establish the entire thesis. The argument would need to be supported by converging evidence across the relevant dimensions.

8. What Would Disprove or Weaken the Thesis?

A credible working paper must specify what evidence would cause its authors to revise their conclusions.

The term single apparatus is a hypothesis about a possible common structure across multiple portfolios. It must not become a label applied indiscriminately to every administrative failure.

Several findings would substantially weaken the proposition.

8.1 The portfolios are not converging

If the evidence shows that the seven portfolios are developing in materially different directions, with no consistent movement towards common arrangements for funding, assessment, contracting or accountability, the claim of convergence would need to be narrowed or abandoned.

Similar problems can arise independently. Their existence does not necessarily establish a shared institutional architecture.

8.2 Private provision improves outcomes without weakening accountability

If independent evaluations demonstrate that external providers consistently improve service quality, accessibility and outcomes while preserving effective public oversight, that evidence must be acknowledged.

The purpose of this paper is not to establish that private provision is inherently harmful. It is to examine the conditions under which institutional arrangements may separate public responsibility from practical control.

8.3 Standardisation improves decisions

If assessment systems are shown to improve accuracy, consistency and fairness, while accommodating complex circumstances and preserving effective human review, their use would weaken the claim that classification necessarily displaces individual need.

The distinction between a tool that assists judgement and a system that substitutes for judgement is essential.

8.4 Accountability becomes stronger

If affected people have meaningful opportunities to challenge decisions, receive understandable explanations and obtain timely remedies, the claim that administrative systems are progressively removing practical accountability would need to be revised.

Formal review rights alone are insufficient to establish that accountability is effective. But evidence that review works in practice would be important.

8.5 Outcomes improve

If the relevant services produce sustained, independently documented improvements in outcomes, the authors must examine whether those improvements contradict the proposed relationship between institutional design and harm.

Improved outcomes would not automatically resolve every question about accountability or fairness. They would, however, count against any argument that assumes externalisation or administrative standardisation necessarily produces worse results.

8.6 The proposed mechanisms cannot be demonstrated

The paper identifies displacement, classification, layering, capping, substitution, market-location dependence and splitting as distinct mechanisms.

If research cannot demonstrate these mechanisms in the portfolios where they are proposed, the relevant claims should be withdrawn or revised.

A common analytical vocabulary is useful only if it clarifies the evidence. It becomes counterproductive if it merely gives different problems the appearance of unity.

8.7 The 2037 prediction does not materialise

The forecast should be assessed against a defined baseline and observable indicators.

If public institutions retain or strengthen their capacity to make decisions, provider concentration does not increase, professional discretion remains meaningful, accountability improves and outcomes strengthen, the predicted trajectory would be substantially weakened.

The paper must be capable of reaching a conclusion that the original hypothesis was mistaken.

That is not a concession. It is a requirement of serious investigation.

9. Research Priorities: What Must Be Established Next?

The existing public record provides a starting point. It does not yet supply every piece of evidence required to test the complete hypothesis.

The next stage should concentrate on primary documentation, measurable changes over time and the experiences of people using the services.

9.1 Establish a procurement baseline

For each portfolio, compile a record of major contracts, funding arrangements and external service-delivery relationships.

Relevant sources include:

Commonwealth budget papers and portfolio budget statements;
departmental annual reports;
the AusTender procurement database;
GrantConnect;
publicly available contract notices and contract documents;
relevant state and territory procurement records;
financial statements and ownership disclosures for major providers.

The baseline should identify contract value, duration, provider, service type, renewal arrangements, performance requirements and available information about outcomes.

Where contract details are unavailable, the paper should record that limitation rather than infer the missing information.

9.2 Establish provider concentration

For each portfolio, identify the distribution of contracts and public expenditure across providers.

Where the data permit, calculate concentration measures and examine changes over time.

The research should distinguish between:

the number of providers;
the number of genuinely available providers in a region;
the share of expenditure held by the largest providers;
ownership links between apparently separate entities;
barriers to entry;
the ability of government to replace an underperforming provider.

A large number of registered providers does not necessarily mean that a person has meaningful choice in their locality.

Equally, a small number of providers does not automatically establish that a market is failing. The available alternatives, quality of service and consequences for users must be examined.

9.3 Obtain primary evidence on automated decisions

For relevant assessment and classification systems, seek:

system and assessment specifications;
decision rules and classification criteria;
procurement documents;
validation studies;
independent audits;
documented error rates;
information about human intervention;
review and appeal statistics;
evidence of how complex or exceptional cases are handled.

The research must distinguish between a system that supports an assessor and one that effectively determines the available outcome.

Where a department or provider cannot supply the necessary information, that absence may be relevant to transparency, but it does not automatically prove that the system is defective.

9.4 Examine contracts and performance measures

For employment assistance, disability services, aged care and other externally delivered services, obtain the relevant contractual frameworks wherever possible.

Determine what providers are paid to do, how performance is measured and what consequences follow from poor results.

The key question is whether the measures align with the public purpose of the program.

If a program is intended to improve people’s lives, the research should seek evidence of those improvements rather than relying exclusively on the administrative indicators used to manage the program.

9.5 Test the relationship between funding limits and outcomes

The mental-health and aged-care cases provide opportunities to examine how formal limits, classifications and funding decisions relate to actual needs.

The research should seek evidence that allows comparison between people with similar circumstances but different decisions or outcomes.

Any causal claim must account for alternative explanations. For example, people with greater needs may receive more services and still experience worse outcomes. That would not establish that additional services caused the poorer outcome.

Likewise, a person receiving fewer funded services may have better outcomes for reasons unrelated to the funding arrangement.

The analysis must be designed to distinguish these possibilities.

9.6 Incorporate lived experience

Administrative data can reveal patterns, but they cannot always explain what those patterns mean to the people affected.

The research should invite contributions from:

people with disability and their families;
older people and their carers;
veterans and their families;
people who have sought mental-health support;
jobseekers;
parents and childcare workers;
people seeking social housing;
professionals and frontline workers;
former public servants and people with direct knowledge of procurement or assessment systems.

Contributions should be collected with appropriate regard for privacy, consent and the risks of disclosing personal information.

Lived experience should neither be dismissed as anecdotal nor treated as sufficient proof of a system-wide conclusion.

Individual accounts can identify mechanisms that require investigation, reveal the practical consequences of policy and show where administrative data may be incomplete. Broader claims should then be tested against additional evidence.

9.7 Examine legislative and review arrangements

The research should examine the legislation, regulations and administrative rules governing each portfolio.

It should also consider explanatory memoranda, parliamentary debates, court and tribunal decisions, review reports and relevant oversight findings.

This material can help establish what governments intended a system to do, what powers were legally available and what safeguards were contemplated.

It is important to distinguish stated policy objectives from actual institutional effects.

A policy document may describe an intended benefit without proving that the benefit occurred. Equally, a harmful outcome does not, by itself, establish that policymakers intended it.

10. Required Documents and Source Register

The following register identifies core documents for the working paper. These are starting points for the investigation, not a claim that every relevant source has already been collected or fully analysed.

10.1 Disability services

Independent NDIS Review

Working together to deliver the NDIS.

The review provides a major source for examining the design and operation of the NDIS and the reforms proposed to address identified issues.

Source: Independent NDIS Review — Working together to deliver the NDIS

Royal Commission into Violence, Abuse, Neglect and Exploitation of People with Disability

Final Report, published September 2023.

The report provides primary material on the experiences of people with disability, institutional arrangements, safeguards and recommendations for reform.

Source: Disability Royal Commission — Final Report

Further documents required

NDIS legislation and relevant amendments;
current administrative and operational guidance;
published review and appeal data;
provider and procurement information;
evaluations of participant outcomes;
relevant government implementation reports.

10.2 Aged care

Royal Commission into Aged Care Quality and Safety

Final Report, published 1 March 2021.

Source: Aged Care Royal Commission — Final Report

Department of Health, Disability and Ageing

Support at Home classification system.

Source: Support at Home classification system

Department of Health, Disability and Ageing

Ongoing home support classification levels assignment.

Source: Ongoing home support classification levels assignment

ABC News

Inside the aged care algorithm deciding support for older Australians, 17 August 2026.

This investigation reports expert analysis and concerns about the Integrated Assessment Tool and the translation of assessment responses into support classifications. Claims attributed to the investigation should be checked against technical documentation and any subsequent official review.

Source: ABC News — Inside the aged care algorithm

Further documents required

technical documentation for the assessment and classification process;
validation and evaluation studies;
information about the decision rules;
review and appeal statistics;
relevant procurement documents;
official responses to reported concerns;
evidence of actual funding and care outcomes.

10.3 Veterans’ services

Royal Commission into Defence and Veteran Suicide

Final Report, published 9 September 2024.

The report includes 122 recommendations and provides the central public record for examining the systems affecting Defence personnel, veterans and their families.

Source: Royal Commission into Defence and Veteran Suicide — Final Report

Further documents required

Department of Veterans’ Affairs annual reports;
implementation reports responding to Royal Commission recommendations;
rehabilitation and service-provider contract information;
relevant provider performance data;
evidence about the coordination of services;
independent evaluations of veteran-support outcomes.

10.4 Mental health

Australian Government Department of Health

Better Access initiative.

The program’s published information describes eligibility and Medicare arrangements, including limits on subsidised sessions.

Source: Better Access initiative

Further documents required

current Medicare Benefits Schedule requirements;
mental-health workforce and geographic-access data;
waiting-time information;
independent evaluations of Better Access;
research on continuity of care and treatment outcomes;
evidence about out-of-pocket costs and access to alternative services.

10.5 Employment assistance

Department of Employment and Workplace Relations

Workforce Australia Services Provider Performance Ratings — June 2026.

The ratings provide a documented example of the government’s provider-performance framework.

Source: Workforce Australia — Provider Performance Ratings

Further documents required

current Workforce Australia contract documents;
payment schedules and performance requirements;
the methodology used to calculate provider ratings;
independent program evaluations;
participant outcome data;
evidence about the relationship between ratings, payments and contract renewal.

10.6 Childcare

Australian Competition and Consumer Commission

Childcare Inquiry Final Report.

The report examines market structure, prices, costs, profits, competition and policy settings.

Source: ACCC — Childcare Inquiry Final Report

Australian Government Department of Education

Three Day Guarantee.

The government’s published information describes the guarantee that commenced on 5 January 2026.

Source: Department of Education — Three Day Guarantee

Further documents required

childcare supply and demand data by region;
provider entry, exit and ownership information;
local price and affordability data;
workforce availability;
waiting-list information where available;
evaluations of subsidy and supply-side measures;
evidence about the effects of the Three Day Guarantee.

10.7 Social housing

Australian Institute of Health and Welfare

Housing Assistance in Australia 2026.

The report provides data on social housing, households and waiting lists.

Source: AIHW — Housing Assistance in Australia 2026: Households and waitlists

Australian Bureau of Statistics

Estimating Homelessness: Census, 2021.

The report provides the ABS estimate of homelessness on Census night in 2021. It should be cited as a 2021 estimate, not a current count.

Source: ABS — Estimating Homelessness: Census, 2021

Further documents required

state and territory social-housing reports;
housing-provider registers;
public and community-housing funding agreements;
social-housing supply and allocation data;
waiting-time data;
information about ownership and management arrangements;
evidence about the availability of appropriate housing for different populations.

10.8 Cross-portfolio administrative accountability

Royal Commission into the Robodebt Scheme

Report, published in 2023.

The report provides a primary source for examining administrative decision-making, data use, legality and institutional accountability in the Australian public sector.

Source: Royal Commission into the Robodebt Scheme — Report

Further documents required

relevant privacy impact assessments;
automated decision-making policies;
procurement and technical specifications;
audit and oversight reports;
information-sharing arrangements;
relevant legislation and explanatory memoranda;
court and tribunal decisions;
independent assessments of administrative decision systems.

11. A Note on the Authors’ Earlier Work

Andrew Klein has published critical commentary on the NDIS and the relationship between public funding, commercial incentives and disability support.

The following articles may help explain the development of the present hypothesis:

NDIS — When Care Becomes a Commodity and Lives Become Profit, 6 July 2026.
The NDIS Extraction — How Australia’s Insurance Scheme Was Designed to Fail the Disabled and Profit the Few, 26 July 2026.
The Great NDIS Betrayal, 3 August 2026.
The Algorithmic Lens — How the Australian State Replaced Human Judgment with Algorithmic Governance in Human Services, 29 September 2026.

These articles should be identified as the author’s own commentary.

They are relevant to the development of the research questions and the interpretation proposed in this paper. They are not independent evidence that the hypothesis is correct.

The distinction matters because the paper seeks to establish a credible research framework that can be evaluated by people who do not share the authors’ initial conclusions.

The public record must remain the basis on which the proposition is tested.

12. Conclusion: Who Controls the Decision That Shapes a Life?

The question at the centre of this paper is straightforward.

When an Australian seeks assistance from a human-services system, who ultimately determines what that person receives?

The answer may involve a government department, an assessor, a professional, a private provider, a contract, a funding formula, an eligibility rule or an automated administrative system.

Often, it may involve several of these at once.

The fact that responsibility is distributed does not necessarily mean that the system is defective. Complex services require different forms of expertise, administration and oversight.

But distribution creates a problem when no participant has sufficient authority, information or obligation to ensure that the final outcome meets the person’s actual needs.

A system may satisfy its contractual requirements while failing its public purpose. An assessment may be internally consistent while overlooking important circumstances. A provider may deliver the service specified in a contract while the contract fails to secure the outcome the person needs.

These possibilities are not proof of a single apparatus. They are reasons to investigate how public funding, private provision, administrative systems and human judgement interact.

The seven portfolios considered here present different questions.

In disability services, the issue is whether individual need is displaced by administrative and funding arrangements.

In aged care, it is whether classification systems translate assessment into appropriate support.

In veterans’ services, it is whether institutional layering creates gaps in coordination and accountability.

In mental health, it is whether funding limits adequately accommodate different treatment needs.

In employment assistance, it is whether performance measures represent meaningful outcomes.

In childcare, it is whether market arrangements provide adequate access across different communities.

In social housing, it is whether identifying and classifying need is disconnected from the availability of a home.

These are distinct problems. Their similarities must be demonstrated, not assumed.

The proposition advanced by this paper is that they may reveal a broader institutional question: can a government continue to fund a service while progressively losing practical control over the decisions that determine whether that service fulfils its public purpose?

That question is important enough to warrant investigation.

It is also important enough to demand more than assertion.

The next stage must establish the relevant contractual arrangements, financial flows, decision-making rules, accountability mechanisms and outcomes. It must compare developments over time. It must seek evidence from people who use the services and from those who work within them. It must distinguish institutional failure from deliberate intent and avoid treating every adverse outcome as proof of a common cause.

Most importantly, it must remain open to findings that contradict the original proposition.

The aim is not to demonstrate that every public-private arrangement is harmful. Nor is it to defend existing arrangements simply because they are established, legally authorised or administratively convenient.

The aim is to determine whether Australia’s human-services systems are structured to meet human need—and, where they are not, to identify the reasons and the available remedies.

A public system should ultimately be judged by more than the amount of money it spends, the number of assessments it completes or the number of contracts it administers.

It should be judged by whether people receive appropriate assistance, whether decisions can be understood and challenged, whether institutions accept responsibility for failure, and whether the public can establish that the system is achieving its stated purpose.

If the evidence supports the hypothesis of a converging institutional architecture, the public should be able to see the evidence and debate its implications.

If the evidence does not support it, the hypothesis should be revised.

Either outcome would advance understanding.

The purpose of this paper is not to close the argument. It is to make the argument testable.

13. Invitation to Readers: Help Test the Hypothesis

We invite readers, researchers, service users, frontline workers, professionals, public servants and people with relevant experience to examine this working paper critically.

We are particularly interested in evidence that helps establish what is happening in practice, rather than evidence that merely confirms a preferred interpretation.

We welcome:

primary documents that clarify how decisions are made;
contracts and procurement records that illuminate institutional incentives;
independent evaluations of service quality and outcomes;
evidence about the operation of assessment or classification systems;
examples of effective human review and accountability;
documented experiences that reveal gaps between formal policy and practical outcomes;
corrections to factual claims or interpretations;
evidence showing that the proposed mechanisms do not operate as suggested.

We also welcome disagreement.

A reader who believes that a particular portfolio has been characterised incorrectly should identify the relevant claim and provide evidence that enables it to be tested. A reader who can demonstrate that a reform has improved outcomes should make that evidence available. A person with direct experience of a system may identify questions that formal reports have overlooked.

We ask readers to distinguish between documented facts, personal experience and interpretation, just as we have attempted to do here.

No individual account should be dismissed merely because it is personal. Nor should a single account, however compelling, automatically be treated as proof of a system-wide conclusion.

The task is to assemble a clearer and more complete picture.

We recognise that this paper contains propositions that require further evidence. The source register identifies documents already relevant to the inquiry and areas where additional primary material is needed. As further evidence becomes available, the paper should be revised accordingly.

We make no claim that the investigation is complete.

We claim only that the questions are important, that the evidence warrants careful examination, and that public institutions should remain open to scrutiny when the people they are intended to serve report that the system is failing them.

Andrew Klein & Sera Elizabeth Klein
October 2026

Final research note

This is a working paper, not yet a completed empirical study.

The paper is presented as a working hypothesis, not a predetermined verdict. Its purpose is to examine whether different parts of Australia’s human-services system are moving towards a common model in which governments continue to fund services while private providers, contractual requirements, performance measures and administrative systems increasingly shape what people receive.