Tag Archives: statistics

Government by numbers – statistics and lies

The headlines yesterday included the news that the UK unemployment rate has fallen to 7.1%, apparently the best since early 2009. But the comments and discussions that followed are a reminder of the dangers of using statistics as a shorthand for government achievements.

For a start the 7.1% is a national figure and disguises significant variations. For those under 24 not in education, 18.1% are out of work and more than a quarter of these have been looking for a job for over a year.

Nor does the headline figure mean that all problems have been solved for those people who have found work. The newsworthy figures don’t show how many more jobs are only part-time, temporary or on zero-hour contracts. If there’s no guarantees of actually work in a role that can be whipped away at a moment’s notice, it hardly counts as a success. The resulting low wages also mean that for the first time since records began the majority of people in poverty are actually in working families, according to a report from the Joseph Rowntree Foundation published last December. And Ed Miliband claims that average wages have gone down £1,600 a year since the 2010 general election, making it tougher for everybody. Employment figures only tell part of the story of how well we’re doing.

Government figures always need to be taken with a pinch of salt. They can be selective and misleading after ministers apply their own spin. But they can also be just plain inaccurate or self-fulfilling.

Last week the UK Statistics Authority, the independent body who’s aim is to safeguard the production of official statistics, downgraded the crime figures for England and Wales because of a belief that the police figures may not be reliable. In its Statistics on Crime in England and Wales report the organisation refers to comments from the Office of National Statistics (ONS) that the problem in crime figures could be due to the fact that ‘a growing number of crimes reported to the police are not being captured in crime recording systems’.

Just today, the National Audit Office released a report into NHS waiting times for elective care in England which found significant errors and inconsistencies in how NHS trusts record waiting time, masking ‘a good deal’ of variation between trusts in actual waiting times.

Even if reported government figures are correct, they may not be all that they seem. My own favourite example also relates to the NHS and the government’s aim of ensuring that patients can see their GP within 48 hours. In the case of my local surgery, the initial response was to ensure they fulfilled the target by not allowing appointments to be booked more than two days in advance. If there were no slots available, tough. It resulted in a Glastonbury-like phone scramble every morning as people speed-dialled to try and book an appointment – heaven help those too ill to participate. But it helped the government to claim that the 48-hour rule was being complied with.

Another example of figures not quite being what they seem is the recent panic over immigration, now that Bulgaria and Romania are in the EU and their citizens entitled to work in Britain. Leaving aside the way the government has spread panic about an invasion of workshy foreigners, it turns out that the goal for reducing immigration is actually based on net migration, i.e. those arriving minus those leaving the country. So the target can equally be met by getting more people to leave the UK, which means there are alternative solutions. One springs to mind. Bankers are continually threatening to leave the country if their bonuses are cut or taxes increased. So with one piece of legislation we could simultaneously prevent the increasing wealth inequality in the UK and at the same time reduce immigration. A win-win.

When you hear the claims of government success based on a set of figures, it pays to look more closely. Check, if you can, to see if the numbers are accurate, what they actually mean and if they tell the whole story. Remember, the police can’t even get the crime figures right.