“The law of unintended consequences is among the most potent laws in existence. Governments, for instance, often enact legislation meant to protect their most vulnerable charges but that instead end up hurting them. Consider the Americans with Disabilities Act (ADA), which was intended to safeguard disabled workers from discrimination. A noble intention, yes? Absolutely – but the data showed that the net result was fewer jobs for Americans with disabilities. Why? After the ADA became law, employers were so worried they wouldn’t be able to discipline or fire bad workers who had a disability that they avoided hiring such workers in the first place.”
Levitt & Dubner, Superfreakonomics.
They don’t cite their proof for this, which is a shame because I’d like to know the extent to which there is any. Maybe the ADA made more applicants disclose disabilities in the hopes of getting a fair shot. Maybe employers are more concerned with the (often perceived) risk of sickness absences, maybe they didn’t like the thought of paying for necessary adjustments. Maybe they just assume disability means the candidate won’t be as good at his job. UK research has demonstrated that all of these factors are relevant to disability discrimination here. It just seems to me that the idea that a significant enough number of employers cried the discrimination version of “‘elf and safety” and got the wrong end of the stick after coping for years with laws protecting against discrimination on grounds like gender and race seems more than a bit odd. It tells me less about disability discrimination than it does about putting your faith in material you read. When I find myself questioning a claim on a subject I know a bit about (and it’s fair to say I only know a bit about disability discrimination in the American context), it also makes me more skeptical of claims relating to issues I haven’t previously considered myself.
That, in turn leads me to think about lies, damned lies and statistics. The current British government and media are absolute masters in the art of twisting information and data to suit their own ends.
They use data over shorter or longer periods of time to suit their purposes (for example looking back at benefits increases over a period of 30 years because it conveniently shows that, at times, they have increased above the inflation rate).
They draw comparisons when it suits them and ignore relevant information which fails to help their case (for example saying the UK Legal Aid budget is higher than other countries while ignoring things like an increased number of criminal offences on the statute books and more complex trials for offences like terrorism in the UK or ignoring migration of British workers to other EU countries when talking about immigration or claiming work capability assessments are reducing benefits fraud because some people stop claiming ESA (which is actually used by many people who will get better and my experience was I’d have been off Incapacity Benefit faster if I hadn’t had to wait so long for medical treatment at every single step)).
They refer to increases in numbers in percentage terms over whichever period they care to pick without mentioning the underlying total number.
Equally, when it suits them, they do the opposite and talk about absolute numbers instead of percentages (for example by saying x number of people are in work and that is more than were in work at a point in time when there were less working age people in the UK).
They also avoid giving figures at all when that’s the most convenient option (for example, when talking about large families on benefits or multi-generational unemployment, it would be impossible to generate the kind of outrage against all benefits recipients and unemployed people if they told people exactly how many people fall into these two categories nationwide).
They only give part of the picture (for example, returning to employment, by not saying how many jobs are actually full time permanent jobs or by trying to avoid saying how many people became unemployed over a more relevant period of time).
They flip flop their arguments (but not necessarily their policies) when their own data turns against them like Frankenstein’s apocryphal monster (fortunately for George Osborne, he’s already got the mad scientist who’s been locked in the basement of his castle with his very own Scottish Igor, tinkering away with forces beyond his comprehension bit down pat).
They spin the ever loving crap out of the data they can’t avoid (hurrah: we’ve avoided the first ever British triple dip. Let’s conveniently ignore we’ve failed to meet our targets and that the technical economists’ definition of a recession has little relevance to people who are suffering as a result of our failures anyway).
They talk about “real terms cuts” when it suits them (for example in party political broadcasts in the run up to local elections) and absolute increases when the fact that there is actually a real terms cut would be unpalatable to the public (for example increases in the minimum wage, benefits and national average salaries).
They’re willing to discuss anecdotal evidence when it suits them (for example, to support the argument that migrant EU workers are “stealing” British resources) but refuse to take them seriously when they don’t (for example when confronted with real stories about real people harmed by social security reforms).
In short: their data stands up to so little scrutiny that it’s easier to take every new statistic as “shit we made up because you’re gullible enough to lap it up”. By using numbers to make their case, they take many people in. I don’t know if the problem has gotten worse under the current government or whether we just have more ways of catching them out. Thank God for the Freedom of Information Act and the rise in fact check blogs. Believe it or not, I wrote everything up to this point last weekend. It was with a sense of deep satisfaction that I saw that a complaint against Iain Duncan Smith was upheld by the UK Statistics Authority. Every time they spin statistics, politicians are doing it for one reason: to make it easier to do what they want to do. Those of us who oppose those plans would say that they’re deliberately trying to hide the truth in order to ensure that the silent majority stay silent. “Go about your business [if it hasn’t been closed down as a consequence of our awful economic policies]. Nothing to see here. This doesn’t concern you. All will be well for you.” The problem is that complaints like mine often sound like the paranoid ranty left, which is why it’s enormously important that when it’s possible to catch them lying, going beyond their usual obfuscation, that’s done and widely reported.
Update 29 May 2013
A cross party committee has urged Government to present statistics in a more honest way.
Update 2 June 2013
This article from the Guardian addresses the bare faced lies and proposes it’s time to get tougher on politicians who lie about statistics.