Australia’s Robodebt scheme: A tragic case of public policy failure

As Australia's Royal Commission into the Robodebt Scheme publishes its damning report, MPP student Chiraag Shah examines how a political culture of scapegoating welfare recipients led to one of Australia’s most egregious and tragic public governance failures. 

Estimated reading time: 3 Minutes
Laptop surrounded by credit cards
Photo by Dylan Gillis on Unsplash

Earlier this month the final report of the Royal Commission into the Australian Government’s Robodebt Scheme was published. Commissioned by the newly elected Labor Government in 2022, the report is a scathing and unflinching exposé into how a prevailing political culture that scapegoated and negatively portrayed welfare recipients led to one of Australia’s most egregious and tragic public governance failures. 

Introduced in 2015 as part of the then Liberal-National Coalition Government’s strategy to reduce the fiscal deficit, Robodebt was a compliance and debt recovery programme to claw back supposed overpayments to welfare recipients. The policy was framed as a ‘welfare integrity’ measure, with the implication that recipients were somehow ‘cheating’ the system. Robodebt relied on automated data-matching between income tax and social welfare data and a process known as ‘income averaging’, where employer-reported income is divided up evenly and allocated on a fortnightly basis over a financial year to assess income and entitlement to benefits. When the system identified a discrepancy between the average income and the income that people actually reported while they were receiving payments, a debt notice was automatically issued to welfare recipients.

What went wrong?

First, the income averaging method assumed a person had completely stable earnings over the period when Robodebt was in force, and ignored the realities of insecure or casualised employment with more variable income. Income averaging was also not consistent with the social security legislative framework which required entitlements to be calculated based on actual fortnightly income. 

Second, the automation of the process from data-matching through to online self-service and repayment, which was sold as a cost-saving and efficiency measure, had the effect of reversing the onus so welfare recipients had to disprove overpayment. It also reduced their ability to have recourse to case officers if they wanted to dispute the debt, and failed to account for changing personal circumstances or differing levels of digital literacy.

Less than a year after the full Robodebt scheme launched it became apparent that the system was a failure, with the distressing impact on welfare recipients becoming apparent. In response to the mounting criticism, personal information of Robodebt victims was released to journalists in a campaign to deter victims from speaking out.

A devastating outcome

Most distressingly, the Royal Commission heard from the mothers of young men who died by suicide after having debts raised against them and struggling to clear their names; the Commission was confident these were not the only tragedies of this kind linked to Robodebt.  According to the Royal Australian and New Zealand College of Psychiatrists (RANZCP), the stigmatisation of people who received debt recovery notices as ‘welfare cheaters’ could have led to other forms of trauma, anxiety, distress and mental ill-health

A total of AUS $746 million was wrongfully recovered from 381,000 individuals and was later refunded. And as a result of a class action claim from welfare recipients, the government wrote off debts totaling AUS $1.75 billion in May 2020. However, the financial hardship and distress caused, and the effects of people’s ability to access credit, are very likely much more far-reaching. 

Robodebt shows how public officials lost sight of their obligation to serve the public interest.

The Commission concluded that ministers did not fulfil their obligation to ensure the program they administered was lawful. Senior public servants were found to have misled Cabinet by not disclosing the reliance on income averaging, nor including the legal advice about the unlawfulness of income averaging under the current social security legislation in the original policy proposal. The final report includes a sealed section recommending the referral of named individuals for civil and criminal prosecution; some of these officials who misled Cabinet may face criminal charges and the former department head has been forced to resign.

What lessons can policymakers learn from this failure?

  1. Policies should be designed with those who may be affected and their experiences in mind. The tragedy of Robodebt shows what can happen when this foundational principle of public policy is not followed.
  2. The importance of robust public governance processes: effective ‘gatekeepers’ in central government departments, as well as properly resourced and independent ‘referees’, such as ombudsman, privacy commissioners and auditors, play an important part in stopping poorly thought-through or unlawful programmes from proceeding.
  3. The rush to introduce AI and automated decision-making systems into public service delivery needs to be tempered. The efficiency benefits may be substantial, but legal and ethical issues must be addressed first. Those subject to automated decision-making systems should be entitled to know or understand the reasons behind decisions; the underlying algorithms should be made available for public scrutiny.

The Robodebt scheme is a shameful episode in Australian public administration and a demonstration of the potentially tragic consequences when ethical and robust public governance with citizens’ interests at its core is discarded for expediency and populist scapegoating. The Royal Commission’s report is a harrowing read and a sobering lesson for all who work in public policy.