In the 1750s, the British government wanted to conduct a national census, and a bill was introduced into the House of Commons to establish one. But opponents of the government sought to block it. It wasn’t just partisanship: the opposition was self-interested — it came from regional, conservative elites, and conservative landed aristocrats — and the rural squirearchy was controlled the local instruments of state power that would be used in a census-gathering. But it was also ideological: opponents saw in the census the centralising tendencies of a government they already believed was too powerful. A census would further strengthen the power of the Crown.
The bill was defeated and it was another 50 years before the first census was conducted in Britain.
It was a different story in the United States. In the land of the free and the home of the brave, which had but recently thrown off the yoke of the British monarchy, a requirement for a regular census was written into the constitution. But the Americans had different concerns: in a rapidly expanding young nation where seats in Congress were allocated on the basis of population, it was in every state’s interest to ensure an accurate headcount of “free persons” and slaves — who counted as three-fifths of a free person for the purposes of allocating congressional seats, even though they never voted. Southern politicians were especially keen to make sure censuses counted their slaves.
Both responses reflect something that we tend to forget: the banal process of collecting statistics is innately political, and has been from the moment printing began enabling early modern governments to record and store large amounts of information, further strengthening their capacity to impose order internally and wage war externally.
If the Abbott government goes ahead and dumps the national census, as Crikey tipped yesterday, it won’t be about ideology or politics, it will be a cost-cutting measure. A compulsory national census, which is a population-scale form of coercive personal-data gathering, is enormously expensive — based on the cost of the 2011 census, it would cost over half a billion dollars to force every citizen in the country to divulge extensive personal information next year.
The government’s critics are up in arms. And, true, the Abbott government already has a rotten reputation when it comes to evidence and data: it is run by climate denialists and wind turbine syndrome spruikers and, by and large, prefers to pretend science doesn’t exist unless it suits its agenda. But would swapping a census for a detailed household survey really be that disastrous for evidence-based policy? Germany hasn’t had a national census in decades; the EU-wide census it participated in in 2011 was the first since the 1980s, and that hasn’t stopped Germany from becoming one of Europe’s richest and healthiest societies. Belgium doesn’t have a census, but manages to have one of Europe’s best-performing education systems. Canada gave up censuses in 2010 in favour of voluntary National Household Surveys. A census is unnecessary for evidence-based policymaking, if this or future governments decide they’re actually interested in evidence-based policymaking.
As the long history of attitudes to censuses shows, they are not ideology-free. And they are now more ideological than ever. We live in an era of increasing information asymmetry between citizens and governments. Governments are less and less transparent in the way they develop and implement policy. But the internet gives them more tools to monitor and accumulate information on their citizens. Police forces insist they need data retention because phone companies no longer keep phone records. But phone records now yield vastly greater information about an individual than they did in the analog era. Metadata provides far more accurate information about an individual than obtaining the content of their phone calls or emails. Even relatively small amounts of online data can be de-anonymised both by government agencies and corporations. The information balance between governments (and corporations) and citizens has swung dramatically in favour of the former.
Ah, but census advocates say, censuses allow governments to do good things for us, to provide better infrastructure and services. Surveillance advocates always tell us it’s for our own good — data retention, after all, will apparently protect us from terrorists and paedophiles. But, even if you accept the argument, those same good things can be provided without recourse to a coercive national personal data vacuuming operation — and more cheaply, as well.
The stout rural reactionaries of 18th-century Britain were instinctively wary of a central government accumulating more information. In an era when our own government is pushing hard to force us all to pay for the costs of being placed under mass surveillance, we have far more reason to be wary of any form of compulsory information acquisition, whether you think it’s a good thing or not.
The only pity is that blowing half a billion dollars on information-gathering is apparently reason enough to dump a census — but not reason enough to dump a data retention policy that will have none of the benefits a census would deliver.