KPMG report sheds light on manual robodebt refund process

By Shannon Jenkins

Wednesday August 4, 2021

A heavily redacted report prepared by KPMG has revealed more details about the processes Services Australia used to refund payments.
A heavily redacted report prepared by KPMG has revealed more details about the processes Services Australia used to refund payments. (Studio_Loona/Adobe)

A heavily redacted report prepared by KPMG has revealed more details about the processes Services Australia used to refund payments that were wrongly recovered under the welfare compliance scheme known as robodebt.

A key aspect of the robodebt scheme was income averaging — a process based on the assumption that a person would earn income at an even rate over the course of a year.

The government announced that Services Australia would no longer raise debts solely based on averaged income information in November 2019, and in 2020 said it would refund repayments made on debts based wholly or partially on income averaging.

As part of this process, Services Australia began assessing all income compliance debts to determine which debts had been raised wholly or partially using income averaging.

According to a report released by commonwealth ombud Michael Manthorpe earlier this year, Services Australia engaged KPMG to provide external assurance on the debt identification process.


Read more: Robodebt: Services Australia continued to recover debts despite risk that they were ‘legally insufficient’


The KPMG report — which has been seen by The Sydney Morning Herald and The Age — detailed how Services Australia staff manually investigated whether debts should be refunded using spreadsheets rather than computer systems.

Those staff, described in the KPMG report as ‘learners’, needed to undertake the process manually due to poor record-keeping practices. The debts were then reviewed by more senior staff, the report found.

Manthorpe’s report noted that staff guidance provided by Services Australia had explained what information its staff should consider when determining whether a debt was averaged, and whether the individual ‘agreed’ to the use of averaged data.

“Based on the guidance, it appears this process was highly manual and non-prescriptive in nature requiring presumed level of knowledge about typical income patterns, thorough investigative skills and subjective analysis to make consistent and correct decisions,” Manthorpe wrote.

“Services Australia advised that as those staff undertaking the identification process had previously undertaken income compliance reviews, they were familiar with how averaged income would be represented on the customer records.”

The process has reportedly raised concerns from Labor and the Greens that some people who are eligible for a refund may have not received their payment.

But a Services Australia spokesperson has told The Herald and The Age that 421,000 victims of the robodebt scheme have either been paid or have had their outstanding debts cancelled. They said 97% of eligible debts have been paid back.

Earlier this year a federal court judge found the commonwealth ‘unlawfully asserted’ at least $1.763 billion in debts against roughly 433,000 Australians between July 2015 and November 2019. The court approved a settlement worth $1.8 billion.


Read more: Robodebt a symptom of overreliance on temporary workforces, senate committee hears


 

About the author
0 Comments
Inline Feedbacks
View all comments
The Mandarin Premium

Insights & analysis that matter to you

Subscribe for only $5 a week

Get Premium Today