Lies, damn lies and statistics in the referendum campaign

The referendum campaign has so far reinforced the lies, damn lies and statistics view of life.

We've learned that we can't trust Boris Johnson with numbers, as the Leave campaign got pulled up for claiming that it costs us £350M a week to be members of the EU. This figure excludes our rebate and other monies returned directly to our public sector, so is wrong. So the Leave campaign switched to the sort of accurate but deliberately misleading statement that we "send £50M a day" to Brussels, ignoring what we get back. I think we "send" the amount net of the rebate so it's still wrong as well as misleading. Odd Boris should deliberately marginalise Maggie's famous rebate, but tosh either way. So what is the number? My economics guru, David Smith, the Sunday Times economics editor,  allowing for both the rebate and other flows from the EU to our public sector, says the actual net contribution averaged £135M a week over the last 5 years. In 2014 it averaged £110M a week, less than a third of the figure  claimed by the Leave campaign. No wonder Smith said, in his column on 1 May, we should ignore any politician who uses the £350M figure and expunge it from our minds. Smith's numbers come from Sir Andrew Dilnott, head of the Statistics Commission, who said the £350M figure is "potentially misleading ". Clearly, the Leave campaign are not to be trusted on numbers.

You might say that £100M a week is still a big number, but it's less than 1% of government spending. And, in a Brexit scenario, we would still have to pay in to enjoy the benefits of the single market. Even if we didn't, the OECD notes that the transfers to the EU are small and any saving would be offset by the (even I accept) more likely than not result that our growth would be slower, at least for a number of years. Smith notes that Oxford Economics, with no axe to grind in the referendum debate, say that even the most benign Brexit outcome would produce a negligible "dividend" for our public finances. Smith concludes that Brexiteers should stop pretending that leaving the EU would free up money for other spending (Johnson has suggested the £350M a week would be spent on the NHS). He says there is "no money tree... it is an insult to voters intelligence to suggest there is".

Smith said all this a month ago but the £350M a week/£50M a day figure is still being used.

Perhaps more disturbingly, we can't trust chancellor George Osborne to use numbers properly, either. He was one of many claiming the EU benefits us to the tune of £3,000 per family per year, a figure that Full Fact says dates from a 2011 Dept for Business, Innovation and Skills report which concluded that the EU "may be responsible for income gains between 2% and 6%, that is between £1100 and £3500 per British household".  Naturally the Remain camp seized on the higher figure and it was quoted extensively by ex M&S man Sir Stuart Rose before he seemingly got decommissioned as the politicians  got involved. But the figure was widely condemned as misleading. Then Osborne got the Treasury bean counters to sharpen their pencils and he bigged up their conclusion that the economy might be 6% smaller by 2030 if we leave the EU, which represents £4,300 per household. Many sources have branded this statement as misleading, as it invites people to confuse the GDP per household statistic with household income, a completely different parameter. Even the Guardian, in its EU referendum realuty check series, said "Would each household lose £4,300? No."

The common thread here is that the numbers are right but the statements are, well, bollocks, basically.

It seems to me that the politicians are playing much faster and looser with numbers in the referendum campaign than in a general election, maybe because they feel there is less personal accountability.

It's made me conclude that I can't trust anything either campaign says about numbers or statistics. They quote numbers from reputable sources but wilfully use them in the wrong context, or invite us to draw an inappropriate conclusion from them.

Numbers are important to the debate and matter to me and my views on these matters. I often feel that an argument that isn't quantified, isn't capable of satisfactory resolution. I've always been wary of people who reach decisions because they "feel it in their waters", or some similar guff.

But I'm coming to the conclusion that the in or out decision comes down to whether one feels we should or should not be part of the EU on an emotional level.

That said, my next post will be on one aspect of the debate where unexpected clarity has appeared on the numbers: whether we could cut red tape for business if we quit the EU and left its dead hand behind us.