Alex Steer

576 words | ~3 min

Why should schools have high fences, security cameras and biometric pass cards? Why should employment contracts be conditional on the supply of fingerprints and retina scans checked against a central database? Why should DNA samples of every citizen be held by the state? Why should sharp-nibbed pens be banned in hand luggage on planes?

Given any of these questions, it's easy to think of plausible answers. To prevent identity theft or hijackings or deter illegal immigration; to stop sex offenders or mad gunmen or kidnappers from getting into schools. We can come up with these 'security narratives' regardless of how high the level of risk of the things we wish to prevent, because it's so easy for us to measure the benefits of the interventions we are considering. If we know that mad gunmen typically get into schools because of lax physical security, it's obvious that tightening up physical security will directly reduce the number of man gunmen getting into schools. This is, our thought processes tell us, a Good Thing. Up go the fences, on goes the CCTV. It doesn't matter how many or few mad gunmen get into schools, because fewer is always better, so anything that will reduce the risk automatically gets the nod.

But what if the question is the opposite: why should we not fence round our schools, store everyone's biometrics, or file off the corners of everything that goes onto a plane? Why might that be a bad idea? When faced with questions like this, it's a lot harder for us to come up with simple narratives of risk and reward. If there is the possibility of mad gunmen, doing nothing just seems negligent.

Choosing between the options is not a fair process, though, because we have so few good ways of measuring the benefits of not acting. This is because there are not good enough measures of the impacts on individual wellbeing or social capital of intrusive or controlling measures. These impacts may be likened to the externalities that economists talk about when assessing the full cost of a transaction. Monetary externalities - such as the true cost of producing a t-shirt that you buy for £3 - are easier to account for than non-financial ones - such as the social cost of the damage to the wellbeing of the child that makes that t-shirt instead of going to school. There is a growing recognition among economists that better accounting of non-financial externalities is needed to assess the desirability of different kinds of transaction: some are even discussing the end of GDP as a useful metric.

While these debates are already being had about trade and globalisation, they need to be had around risk assessment as well. We are now more able than ever to control risks in our lives and in society, but we are still bad at measuring the fallout from the controlling measures we take. Until we have ways of accounting for the effects (positive and negative) on social capital and individual happiness of every CCTV camera that goes up in a town centre, or indeed every moral panic that hits the headlines, we will be unable to make good decisions on whether or not to invest in intrusive behaviours. Until then, like bad economists, we will fail to count the full cost of our social transactions and will keep taking the safe option, which may be easier to account for but not necessarily for the best.