The EU Debate and Averages

Don’t you hate it when you hear that the EU costs/benefits each UK citizen £x per annum? They make it sound like everyone pays/gets exactly the same every year, like either a personal membership fee (anti EU) or a John Lewis discount voucher (pro EU). Of course any cost or benefit is going to be distributed unevenly across people in the UK. So the spin is all about how you measure averages, and how averages can be abused.

The latest number that’s being quoted is £450 benefit per person per annum, according to Lord Rose. That’s at the smaller end of the spectrum of quoted numbers. It’s also positive, when in reality the EU could well be a cost. But anyway let’s work with this number.

That £450 is going to be a lot to some people. And pretty insignificant to others. The claimed total benefit is about £29billion, just multiplying £450 by the number of people in the UK. But what if that benefit is divided differently across people? Let’s say, for example, that all employees benefit by the same percentage? Back-of-an-envelope calculations then give a benefit of 3.7% to your salary. Pretty good, an extra £5,000 p.a. if you’re on £150k. Obviously quite a bit less than the £450 if you are down at the lower end of the pay scale though. That’s just a first trivial example of a different average. And at least those numbers are also positive. It’s very easy to play around with averages to see how benefits for the worse off can be non existent, or even negative, i.e. a cost.

You can’t just divvy up £29billion according to whether the UK is in or out of Europe. The calculations are made more complicated by the two million EU citizens working in the UK. And the surprisingly similar number of 1.8 million British unemployed. Adding or subtracting EU citizens from the calculations will change those averages.

We often hear how fantastic EU and Rest-of-the-World employees are compared to the layabout Brits. “Every non-Brit adds £y to the economy while Brits cost £z in unemployment benefits,” is the classic line. How does that make the average British person look? But can you see how that is an example of spin, with an indirect use of averages?

As a thought experiment start with zero EU citizens working in the UK and full employment. Now bring in a couple of million EU workers. They undercut the original workforce who end up on the dole. And what’s the end result? Exactly what we see. EU citizen in employment, and out-of-work-and-benefit-taking British. And how does that get spun? As hero EU workers and lazy Brits. This is great if you are an employer like er, Lord Rose, since you’re employment bill has just gone down. But not so great for the newly unemployed and now insulted Brit. Not to mention the impact of a larger population on social services, housing, healthcare, etc. And the reliance on imported skills. For the worse off there’s no £450 benefit p.a., there’s a real cost: No job.

How does all that figure in the “average”? Do the following calculation: Decrease the pay for a couple of million people down to benefit levels, take that total amount and add it to the £29billion you first thought of, and distribute the result across all employed people. Do this correctly and the picture changes completely.

Assuming that all of those EU citizens in the UK are at the lower end of the pay scale, that’s about £16billion they are earning. In reality it will be much higher. There are also welfare benefits, again in the billions. All of those numbers are of the same magnitude. All of them can be moved around and re-interpreted in any number of ways. Each one will give a different “average.” All will be affected whether the UK is in or out of the EU in some highly non-linear and time-dependent fashion. The simplest way to abuse them though is to quote an average and pitch it as a fixed amount per person.

Reminds me rather of everyone’s least favourite risk measure, Value at Risk. A single number meant to capture the downside in a portfolio in a way that dumb management could understand. How’s that working out, guys?

Be the first to comment

Leave a Reply