Automating something that couldn’t be automated is a promise that hurt. Shutterstock/thegreenhead.com
Campaign promises can seem like so much noise and distraction; this morning’s announcement replaced by another by the afternoon and forgotten before bedtime. But as the “robo-debt” saga showed, such announcements can have devastating consequences if not properly scrutinised, once the election rush has passed.
In December 2015 in the lead up to the mid-2016 election the Coalition announced a “crack-down” on welfare overpayments, to be brought about by matching taxation data with Centrelink records.
In the leadup to the election Scott Morrision, then social security minister, spoke of it as a “more bespoke way of dealing with people’s arrangements”. It would “cut red tape, and ensure that mistakes are minimised”.
It began as a promise
Data matching had long been used, but only to identify possible overpayments. Centrelink staff then verified and properly calculated any such debts, using its powers to compel banks or employers to provide precise fortnightly earnings records if the person hadn’t kept them.
The new scheme “automated” that key stage.
Instead of finding out what a person actually earned each fortnight so their rate of social security could accurately be worked out, it robotically apportioned to each fortnight the annual employment income reported to the tax office. Each fortnight the person receiving benefits was said to have earned form employment one 26th of what they had earned over the year, making it look as if they had been working all year even if they had not.
Debts were asserted and put in the hands of debt collectors unless the person could produce pay-slips or other fortnightly earnings records, often from as far back as 2010. The Centrelink website had only advised people to keep records dating back six months.
Government oversight was weak
The scheme was to begin on July 1, 2016, which as it happens was the day before the election. The election victory was narrow, and for a while uncertain, meaning the bureaucracy was left with more responsibility than usual for making sure it was delivered as promised.
And as events transpired, it proved to be unable to design a scheme that completely met the tests of being lawful (it reversed the onus of proof on the esistance of debts), accurate (the bulk of its assessments were either totally false or grossly inflated ), meeting the integrity standards expected of government (hiding from public scrutiny details about appeals that overturned its decisions), or meeting the pub test of morality.
It frankly beggars belief that standards of government implementation could fall so far and so low.
And Centrelink staff must have known it
It cannot possibly have been unknown within government that it was legally impossible to reverse the onus of proof of establishing a debt. The legislative provision was crystal clear that debts can be raised “if and only if” the law creates it as a debt.
Even the lowest-level employee within Centrelink knew that most people on benefits had several different jobs, of varying durations and hours with erratic and fluctuating earnings and that the law required the person’s rate to be determined every fortnight, not on the basis of some extrapolated “average” over a year.
Hardly anyone could not know that basic maths tells us that an “average” never speaks to its constituent parts unless a person has a single job at an unchanged pay rate.
And every senior bureaucrat ought to have understood their obligation to behave as a “model litigant”, which includes not continuing to raise debts that regularly overturned as illegal on appeal, and not to “hide” those decisions by never once challenging them by appealing to the next level where decisions would become public, or reportedly seek to “settle” a Federal Court challenge to the process rather than have it come to a public hearing.
All of these things were part and parcel of this botched implementation.
Bull-headedly, government continued with it 18 months after it became clear several things were wrong.
Promises have consequences
The economic cost to citizens of a program which effectively uses the might of the state to frighten them into paying up what the Ombudsman’s report found to mainly be non-existent or highly inflated “debts”, is now estimated as A$3.7 billion over the budget estimates.
The cost to the government’s reputation for integrity is incalculable, because the irony is that automation is the way of the future, and competently designed and implemented it can benefit social security clients and the public.
Elections are a time when not only the promises but also the professionalism and integrity of their implementation are on public display. If you don’t take note, you might end up getting something that looks appealing (such as a “crackdown” on welfare fraud) but ends up targeting you in a way that is illegal or immoral. You might even have voted for it.
This article was written by:
This article is a syndicated news item via