Peter Jones: Yes/No experiment that might work

STATISTICS, facts, and economic theories are being relentlessly poured out by the Yes and No campaigns to support their opposing arguments. How is the bemused voter, seeking unbiased information to help understand whether to vote for independence or the Union, to judge which is right and which is wrong? Perhaps Professor Brian Cox, particle physicist and television evangelist for the wonders of scientific understanding, has an answer.

Examples of highly problematic information from both sides are not hard to find. From the pro-Union camp, there was the claim in a paper published by the Treasury that Scots, after independence, would have to pay international roaming charges on any mobile phone call they made while in England, Wales, or Northern Ireland. It didn’t take a genius to spot that this is almost certainly untrue as the EU had already decided to get these charges phased out by 2015.

This looked like a straightforward mistake, though you can take it – as many nationalists of the more fervent variety do – as evidence that the UK government is engaged on a programme of outright baseless fear-mongering. This, I think, is hard to sustain as the Treasury has gone to great lengths to back up its arguments with evidence, mainly drawn from non-aligned academic sources.

Nevertheless, the point remains that when doubt is sown because of one error, it spreads over all the material published by the Treasury and other parts of the UK government.

On the Yes side, last week saw a rather worse example which I think can be fairly branded as an attempt at deception. A paper published by the Scottish government argued that pensions in an independent Scotland could both be marginally more generous and paid a bit earlier than they are likely to be inside the UK.

None of the statistics produced in support of this thesis were in themselves wrong. But each presented only a part of what is a complex story and when added together, they offered a highly misleading picture.

Central to this pensions question is the matter of whether Scotland’s population is ageing more quickly, more slowly, or at the same rate as the population in the rest of the UK. The paper said: “Between 2010 and 2035, the number of people of state pension age in the UK is projected to increase by 28 per cent. The increase is slightly lower in Scotland … projected to grow by 26 per cent.”

So it looks like a reasonable conclusion is that Scotland will have fewer pensioners and so can afford to pay them a bit more. No. The complexity here is that if you are asking about the affordability of pensions, you need to know how many taxpayers there will be producing the revenues out of which pensions are paid.

When you do these sums, you discover that both Scotland and the UK had, in 2010, the same proportion of pensioners, expressed as a percentage of the working age population – 32 per cent. But by 2035, the demographic projections are that the percentage will rise to 38 per cent in Scotland and 35 per cent in the UK. The increase is 18 per cent in Scotland and 10.5 per cent in the UK.

Demographers call these percentages the dependency ratio. The Scottish government paper did discuss dependency ratios, but not in the terms described above. It used a total dependency ratio, which lumps pensioners and children under the age of 16 together.

This is more favourable to the Scottish government’s argument. Scotland presently has fewer dependents per 1,000 people of working age than the UK does, a gap which will steadily close until by 2035, Scotland will have about half a per cent more dependents than the UK does.

But this is a nonsense. Why include the under-16s in a calculation aimed at demonstrating the affordability of pensions? The answer appears to be simply that it suits the Scottish government’s political agenda, because when academics do the demographic sums, they come to the opposite conclusion – that Scotland’s population is ageing more rapidly than the UK’s population and thus that an independent Scotland will face a harder problem than the UK does in paying its future pensioners.

There are ways of tackling it, as the pension paper points out. You can increase immigration to increase the size of the workforce, aim to raise the productivity of those workers, or increase the proportion of 18 to 64-year-olds in employment, all of which would increase tax revenues and make it easier to pay pensions.

But the basic demographic problem means that some, or all, of this has to happen just to be on a par with the UK, and then it has to happen some more for the prospect of more generous and earlier pensions to be realised. This, however, is not mentioned.

What can be done to stop politicians, on either side of the debate, from producing this kind of misleading material? I was worrying about this when I happened to watch one of Prof Cox’s latest TV programmes, on the brilliance of British science.

This pointed out that the British publishers of scientific papers had devised a system, now standard world-wide, of ensuring that the material they published was not wrong. It is called peer review. When a publisher gets a paper telling of some new discovery, they send it out to experts of equivalent standing in the relevant field and ask them to pick holes in it. If nothing serious emerges from this scrutiny, the paper is published.

We can’t expect politicians to submit their stuff to experts before they publish it, but suitable ad hoc panels of statisticians, demographers, economists and other social scientists could be assembled to review them once they are out there.

Perhaps organisations such as the Royal Society of Edinburgh, or the Royal Statistical Society, or other august bodies above the grimy political fray could do the job. If they were to do it, it should stop an awful lot of rubbish from polluting the political debate.

This is, after all, the most important political decision to be taken by Scottish voters for 300 years. If we are going to make an informed decision, we need information which is accurate, unbiased, and not deceptive.