The controversy embroiling Centrelink, the Australian government agency that provides welfare payments, shows no sign of abating as a public and political backlash continues over its apparent mishandling of a debt recovery programme.
The trouble for Centrelink – and thousands of its clients - started when it began using a computer program to go back up to six years to find discrepancies to issue more notices for overpaid money to be repaid. What the programme did was identify debts by matching the benefit payments made to clients against their Australian Tax Office (ATO) records.
Since July 2016, Centrelink increased the number of debt notices it issued from around 20,000 per year to 169,000 between July and December alone. This might be all good and well, except that it has been shown that up to 20 percent of those notices were based on false assumptions.
These mistakes have caused consternation and outrage. One Centrelink client told an Australian news organisation that he nearly had a heart attack when his account incorrectly showed him owing $9,000. Another explained in this article that she had been unfairly billed for her ex-partner’s debt. One woman said Centrelink wiped her $7,800 debt after she indicated she would take her case to the Ombudsman's office.
The details of the fiasco are not yet entirely clear, but they are expected to become so with an enquiry by the Australia’s Commonwealth Ombudsman. What is apparent is the problems that have arisen were not caused by the matching of information between the ATO and Centrelink. They are instead down to the processes Centrelink has chosen to use in assessing the personal information and in deciding how to act on the results.
The process of sending out assessments of amounts owed by individual clients had also been changed to take out the manual checking done by staff. An internal process has been credited with correcting 37.5 percent of the notices generated in the previous system.
Claims have also been made in various news media suggesting Centrelink has not been using all of the information available to it, and it seems a major part of the inaccuracies in the system is due to the income earned by an individual in part of a year being averaged out over the entire year.
Instead of taking reasonable steps to make sure the information was correct before issuing debt notices, Centrelink appears to have shifted the onus of correction onto its clients. While this might have up-front savings in terms of staff processing time, the change in approach will certainly have added other costs to the agency.
Centrelink had planned that any problems with the notices would be handled through its website. But the website has proven unreliable and often crashed under the increased workload, increasing the stress on clients trying to dispute their debt notices.
Counting the cost
This has put added pressure on Centrelink staff who will now have to engage directly with clients to fix the errors. There’s a human resources cost to this and Centrelink will also have to deal with a rising volume of complaints against it.
There is a substantial reputational cost. The latest loss of public trust and confidence is another blow to the agency’s reputation after this earlier Ombudsman’s investigation in 2011.
But the most significant damage is the avoidable waste of expense, time, and stress imposed on clients and staff because of the new error-filled process.
In New Zealand
By comparison, New Zealanders fare better. Various protections for individuals were designed into the information matching provisions of the New Zealand Privacy Act in 1993.
Information matching programmes by government agencies are approved by statute and the Privacy Commissioner has a regulatory role to monitor the use of this data matching. There are several steps to get right when matching information held by one government department with information from another.
These protections continue to matter. As the Government invests more in automating its processes and in making greater use of data sets, our Office has a role to play in ensuring that the risks and benefits are assessed up front. This is done by using privacy impact assessment and privacy by design principles to prevent New Zealand agencies from stumbling into the same sort of problems experienced at Centrelink.
Image credit: Centrelink via Wikipedia