Census bungle a learning opportunity

Illustration: Eric Lobbecke

Who is to say Bob Katter wasn’t right to claim that “the poof population of north Queensland was less than 0.001 per cent”?

Having spent more evenings than most under the privacy of the starlit heavens bonding around the campfire, Katter must surely understand the passions that fester beneath an Akubra.

Katter’s remarks date from 1993 however, more than a decade before Heath Ledger lifted the tent flaps on cowboy life in America in Brokeback Mountain and 18 years before the Australian Bureau of Statistics performed the same service on this side of the Pacific.

Thanks to the census we know that on the night of September 7, 2011, there were 91 same-sex couples living in Katter’s electorate. Those 182 individuals alone would take the gay population to 0.12 per cent, suggesting the member for Kennedy’s gaydar badly recalibrating. And that’s not counting the unattached, feckless, coy or confused who, one imagines, would push the tally higher.

In an era where facts are too frequently deemed irrelevant in public debate, the near-catastrophic failure of the 2016 census is no trivial matter. No other set of government data produces such an intimate picture of Australian life or bequeaths us such a quantity of fine-grained socioeconomic data that can be analysed down to the tiniest hamlet.

Despite the abundance of information available in the digital age, the exercise of counting each citizen once, and only once, and recording details of their everyday lives is irreplaceable. It reveals a country more diverse than may be imagined from the comfort of a bentwood chair in an artisanal bakery in Fitzroy.

What are we to make of the fact that Melbourne, the only federal seat occupied by a Greens MP, houses the largest number of psychologists (430), the second highest number of barristers (307) and the fourth highest number of journalists (667) in the country?

Should we be surprised that same-sex marriage might be a burning issue for Adam Bandt and Labor’s David Feeney? Bandt represents the highest number of cohabiting men in Victoria (779) while Feeney represents the neighbouring seat of Batman, the preferred constituency for Victorian all-female couples (415). Kevin Andrews, on the other hand, may not be so out of touch with the modern world as his critics smugly suppose. The member for Menzies’ office is less than 25km from the Sircuit Bar but the constituency is home to fewer cohabiting same-sex couples (67) than Kennedy.

The amount of administrative data collected already by government and the private sector in the course of doing business has inspired thoughts that the census might be scrapped.

Canada, Britain and New Zealand have investigated the idea, but in each case they have been persuaded against it.

There would be too many gaps and it would impossible with most data sets to drill down to the small-area details the census allows. The socioeconomic profile of postcodes, information vital to the administration of a range of government services, would be hopelessly incomplete. Making funding adjustments would be even more clumsy than the Gonski scheme, hard as that may be to imagine.

For the ABS to have put the integrity of this year’s census at risk with its ham-fisted switch to online collection is unforgivable but sadly predictable. For all their presumed superior powers of planning, bureaucrats are Luddites when it comes to innovation.

The list of expensive digital blunders by government agencies around the world continues to grow; Healthcare.gov in the US crashed in its first week, severely compromising the launch of Obamacare in 2013 and costing at least $US200 million to fix.

Last month the British government pulled the plug on Care.data, a controversial National Health Service initiative to store all ­patient data on a single database. An official report concluded that a naive disregard for privacy combined with a hopelessly incompetent technical rollout made the project unsalvageable.

Sound familiar?

The habits of public servants are inimical to innovation. Imagine how miserable vacuum cleaning would be today if James Dyson had worked in a top down, risk averse, process-driven ­bureaucracy that refused to ­acknowledge failure, let alone learn from it.

There is no room for excuses, however. If banks, airlines and retailers are able to make life easier for their customers through digital transformation and reduce costs at the same time, government agencies must learn to do the same.

Progress, however, requires a considered debate about privacy. It may be true that Google, Pay-Pal and Spotify know more about you than your mum does, but citizens are considerably less relaxed about sharing information with governments than with the private sector.

And with good reason. The returns on investing private information with a commercial company are clear and immediate: greater choice, greater range and greater convenience. What’s more, businesses, unlike bureaucrats, are accountable. A privacy breach by Facebook could kill the business; a breach by a government agency would be greeted with a shrug.

The privacy concerns sparked by the ABS’s ill-considered decision to hold names and addresses attached to census data for four years illustrates how little trust ­bureaucracies command. It is ironic, since an Australian-style stand-alone census safeguards privacy more effectively than an invisible census collated from administrative data.

Countries such as Denmark, Norway and Sweden that have successfully transitioned to an ­administrative data-based census have taken 20 to 30 years to do so, and the cost saving may be as little as 10 per cent.

The trust they demand from their citizens, however, is considerable. Synchronising data across departments and across levels of government requires every citizen to be listed on a register with a unique identifying number.

There are obvious benefits for security — no small matter in these terror-ridden times — and in the detection of fraud. But would Australians be willing to accept this tax file number on steroids for the sake of avoiding completing a census form every five years?

The Hawke government’s brave but ultimately futile attempt to introduce the Australia Card in 1985 suggests they’ll need some persuading.

Nick Cater is executive director of the Menzies Research Centre.