Welfare recipients won't find out about robodebt refund until new year

Paul Karp
Photograph: Mick Tsikas/AAP

The 734,000 welfare recipients who have received robodebt notices will have to wait until the new year to discover if they are entitled to refunds because debts were unlawfully calculated using income-averaging.

At a Senate committee hearing on Monday, Department of Human Services officials refused to provide any estimate or running total of the unlawful debts – despite government services minister Stuart Robert’s assurances that only a “small cohort” are affected.

Ros Baxter, the deputy secretary of integrity and information, told the Senate community affairs inquiry that the department expects to have identified all affected welfare recipients “early in the new year” and is preparing to write to those it has already found to explain why it has frozen debt recovery against them.

Related: The robodebt horror was all about boosting the budget. That's the brutal truth | Katharine Murphy

Officials also revealed that – although the department has paused the use of income-averaging for raising new debts – existing purported debts could still be enforced through debt collectors and garnisheeing of tax returns until recipients are proven to be victims of unlawful averaging.

In November the Morrison government abandoned sole reliance on income-averaging ahead of a landmark federal court case in which it conceded averaging was not a proper basis to claim the plaintiff, Deanna Amato, owed the commonwealth a debt.

The changes triggered a massive review of purported debts that have already been issued – with the promise of refunds for all those calculated in whole or part using income average alone.

On Monday, officials revealed that at least half of the department’s 1,500 compliance staff have been reallocated to find recipients of dodgy debts and recalculate them.

Baxter said it was “tricky to know” how much the clean-up will cost until the full cohort had been identified. In the midyear economic and fiscal update, the government noted the “refinement” to the program but said the effect on the value of recovery is not yet clear.

Baxter repeatedly deflected questions about what proportion of debts are affected, saying only that at 31 August there were 734,000 online income compliance reviews which resulted in a debt, not all of which used income-averaging, and not all of which would have been calculated differently if it hadn’t been. Staff have been told more than 220,000 debts may be affected.

Asked how many debts using income-averaging have been identified so far, the general manager of information and integrity, Jason McNamara, replied “we haven’t been counting [that] dataset”.

Baxter said she would “hate to give an inaccurate perception” about how many debts had been identified for recalculation when the department is only part-way through the review process.

Sign up to receive the top stories from Guardian Australia every morning

Anthony Seebach, a general manager at Services Australia, said that its approach “is to continue debt recovery” until debts are identified in the affected cohort – meaning the Australian Taxation Office could garnishee tax refunds this year of people who will later be entitled to a refund.

Nor have debts referred to “external collection agents” been recalled – unless and until they are identified as affected – he said.

Baxter offered to “take on board the suggestion” that the department should change its process so that when the ATO asks for permission to garnishee tax refunds, a separate check be undertaken so income-averaged debts are frozen at that point.

Earlier, the ATO second commissioner, Jeremy Hirschhorn, revealed that from 2016-17 it had garnisheed 225,000 Australians’ tax refunds to recover a total of $205m for unpaid Centrelink debts.

Garnisheeing of tax refunds rose from 40,000 cases worth $32m in 2017-18, to 74,000 recoveries worth $63m in 2018-19 and 66,000 in the year-to-date for 2019-20, worth a total of $72m.

Hirschhorn said the ATO cannot independently check whether income-averaging was used on a debt and the validity of a debt “is a matter for DHS”.

“Whether it is lawful is relevant – but we can only assume that a government agency issues lawful debts.”

Baxter noted the legality of robodebts is still being challenged through a separate federal court case and a class action. She took on notice questions about when the department received legal advice, and suggested the minister may make a public interest immunity claim to block disclosure.

Baxter conceded the department had taken recent litigation into account when it changed the robodebt rules, but painted the move as part of an “ongoing” and “iterative” process of improvement that predated the cases.