Robodebt is dead and buried, or is it? Is the heartless automation at the core of Robodebt not really dead?A subscriber to this newsletter sent us a note about our post on Centrelink telephone wait times:
Twice we have had our couple’s part pension cancelled and just a note sent saying that it was done. I suffered a cut off of my telephone call [to Centrelink] after an hour with no warning or explanation. I rang again and, after a 30-minute wait, I was attended to. (..) The explanation of the automatic cut-off of the pensions was: the automatic machine did it during a routine scan. This means that Robodebt is alive and well under another name and applied to pensioners.
Is this true? Are pensioners subjected by Centrelink to Robodebt-type practices?
The short answer is no.
The long answer involves a look at the corporate culture found in the government agencies administering Australia’s social security system.
How did Robodebt work?
But first a look at Robodebt, which was proposed in 2015 to recover supposed overpayment of unemployment benefits going back to 2010.
Robodebt checked with the Tax Office how much employers had paid to social security recipients and then compared it with what had been declared to Centrelink by the recipient. This is known as ‘data matching’.
If the Tax Office’s data didn’t match Centrelink’s information, recipients were asked to explain. If the recipient did not respond or could not explain the difference, Robodebt would swing into action and raise a debt.
However, Robodebt went even further. It averaged out income declared to Centrelink over the period reported to the Tax Office by the employer. It did so regardless of whether the social security recipients had worked during the whole period. This was known as ‘income averaging’.
For example, if a recipient had worked full time for four weeks in a financial year, their pay would be, say, $25 an hour or $4,000 for four weeks. The fortnightly unemployment benefit is currently $750, or $1,500 over two fortnights. Our recipient loses 50 cents of every dollar they earn through work, so should have missed out on their $750 unemployment benefit twice.
Our recipient would therefore lose $1,500 in unemployment benefits over the year.
However, average out $4,000 over the 26 fortnights in a year, our recipient’s unemployment benefits would be cut by $77 each fortnight.
Our recipient would therefore lose $2,000 in unemployment benefits over the year.
That’s $500 more, and Robodebt would raise that $500 as a debt, even though our recipient did not owe this $500. If our recipient could not explain what happened in the year in question, the debt would stand. If not paid on time, collection would be farmed out to companies specialising in that field of endeavour.
Robodebt and pensioners
It’s easy to see that Robodebt had generally no effect on pensioners as few pensioners have employment income. Robodebt was the subject of a Royal Commission and has been found to be illegal. It has been abolished and in excess of $1 billion has been paid in compensation by the federal Government. Then Prime Minister Scott Morrison, also the architect of Robodebt, apologised to the victims.
However, the aspect of Robodebt that causes the reader who sent us a note (and others) to feel ill-treated by Centrelink is usually: automation. It is frightening to know that an automatic machine (..) during a routine scan can stop your pension, without that action being checked by a human and without giving advance warning.
Add to that, the perverse amounts of time callers to Centrelink spend waiting, and a picture emerges of a social security system where there is no respect for the people who depend on that system. Heartless automation is but a manifestation of that lack of respect.
The federal Government has accepted the Royal Commissioner’s recommendations which relate to automation at Centrelink. Specifically, it has agreed to put in place clear review procedures for automated decisions and to the establishment of a body to monitor and audit automated decision-making.
We add to this that the detection of social security fraud is unlikely to ever turn up anything of note if it concentrates on data matching of official records. Social security fraud typically occurs off the books. For that reason alone, Robodebt was doomed to failure.
However, social security fraud in Australia is probably bigger than assumed on the basis of data matching, which cannot cover off-the-books activity. It has to be.
The JobSeeker Payment currently stands at $375 a week. As advocacy organisations keep saying: you can’t live on that. But are – say – cash-in-hand jobs really social security fraud, or are they a de facto part of the social security system?
Also read:
Mystery: cashless transactions and demand for bank notes hit record highs
Centrelink call wait times worse than ever
A subscriber to this newsletter sent us a note about our post on Centrelink telephone wait times:
Twice we have had our couple’s part pension cancelled and just a note sent saying that it was done. I suffered a cut off of my telephone call [to Centrelink] after an hour with no warning or explanation. I rang again and, after a 30-minute wait, I was attended to. (..) The explanation of the automatic cut-off of the pensions was: the automatic machine did it during a routine scan. This means that Robodebt is alive and well under another name and applied to pensioners.
Is this true? Are pensioners subjected by Centrelink to Robodebt-type practices?
The short answer is no.
The long answer involves a look at the corporate culture found in the government agencies administering Australia’s social security system.
How did Robodebt work?
But first a look at Robodebt, which was proposed in 2015 to recover supposed overpayment of unemployment benefits going back to 2010.
Robodebt checked with the Tax Office how much employers had paid to social security recipients and then compared it with what had been declared to Centrelink by the recipient. This is known as ‘data matching’.
If the Tax Office’s data didn’t match Centrelink’s information, recipients were asked to explain. If the recipient did not respond or could not explain the difference, Robodebt would swing into action and raise a debt.
However, Robodebt went even further. It averaged out income declared to Centrelink over the period reported to the Tax Office by the employer. It did so regardless of whether the social security recipients had worked during the whole period. This was known as ‘income averaging’.
For example, if a recipient had worked full time for four weeks in a financial year, their pay would be, say, $25 an hour or $4,000 for four weeks. The fortnightly unemployment benefit is currently $750, or $1,500 over two fortnights. Our recipient loses 50 cents of every dollar they earn through work, so should have missed out on their $750 unemployment benefit twice.
Our recipient would therefore lose $1,500 in unemployment benefits over the year.
However, average out $4,000 over the 26 fortnights in a year, our recipient’s unemployment benefits would be cut by $77 each fortnight.
Our recipient would therefore lose $2,000 in unemployment benefits over the year.
That’s $500 more, and Robodebt would raise that $500 as a debt, even though our recipient did not owe this $500. If our recipient could not explain what happened in the year in question, the debt would stand. If not paid on time, collection would be farmed out to companies specialising in that field of endeavour.
Robodebt and pensioners
It’s easy to see that Robodebt had generally no effect on pensioners as few pensioners have employment income. Robodebt was the subject of a Royal Commission and has been found to be illegal. It has been abolished and in excess of $1 billion has been paid in compensation by the federal Government. Then Prime Minister Scott Morrison, also the architect of Robodebt, apologised to the victims.
However, the aspect of Robodebt that causes the reader who sent us a note (and others) to feel ill-treated by Centrelink is usually: automation. It is frightening to know that an automatic machine (..) during a routine scan can stop your pension, without that action being checked by a human and without giving advance warning.
Add to that, the perverse amounts of time callers to Centrelink spend waiting, and a picture emerges of a social security system where there is no respect for the people who depend on that system. Heartless automation is but a manifestation of that lack of respect.
The federal Government has accepted the Royal Commissioner’s recommendations which relate to automation at Centrelink. Specifically, it has agreed to put in place clear review procedures for automated decisions and to the establishment of a body to monitor and audit automated decision-making.
We add to this that the detection of social security fraud is unlikely to ever turn up anything of note if it concentrates on data matching of official records. Social security fraud typically occurs off the books. For that reason alone, Robodebt was doomed to failure.
However, social security fraud in Australia is probably bigger than assumed on the basis of data matching, which cannot cover off-the-books activity. It has to be.
The JobSeeker Payment currently stands at $375 a week. As advocacy organisations keep saying: you can’t live on that. But are – say – cash-in-hand jobs really social security fraud, or are they a de facto part of the social security system?
Also read:
Mystery: cashless transactions and demand for bank notes hit record highs
Centrelink call wait times worse than ever
