Wednesday, 2 August 2017

All models are wrong - and some aren't even useful

On Monday, the Tory group leader in the Assembly demanded that the First Minister dissociate himself from Corbyn’s policies, claiming that they would result in around £4,000 of extra debt for each person in Wales, and that the UK would end up paying around £5.8 billion a year in additional interest payments if Labour’s plans were implemented.  It’s the stuff of good political knock-about, but without a lot more information on how they’ve done their sums (and the Tories don’t exactly have a brilliant record when it comes to financial arithmetic), it’s difficult to know what, if any, relationship exists between his figures and ‘truth’, in the mathematical sense of the word.
But, for the sake of argument, let’s suppose his figures are accurate ones.  Is it really the economic disaster as which he paints it?  Of course, £5.8 billion sounds like a very large sum of money to be paying in extra interest every year, but that’s in absolute terms.  And it makes a number of unstated but implicit assumptions.
The first comparison that has to be made is not, as the Tories effectively claimed, with the status quo, but with what the outcome would be over the same period with a Tory Government.  The implicit assumption in what Davies said is that Tory spending plans would not lead to a similar outcome, but given the way in which out-turn has varied from predictions over the last few years, and the way in which much of the (uncosted) Tory manifesto has been ditched, that looks to me like an invalid assumption.  If there is a gap between the likely outcome under a Corbyn government and the likely outcome under a Tory government (and even that is a significant ‘if’) then it is probable that the gap would be much smaller than Davies is suggesting.  All the signs are that the Tories will also increase borrowing to pay for their programme; the honest question is not how much Labour would need to borrow, but what is the difference in borrowing levels between the two.
The second question is about what proportion of GDP the debt would represent, and what proportion of expenditure any extra interest payments would represent.  Both of those are dependent on a range of assumptions and guesses about the likely level of inflation, economic growth, and interest rates.  Given the propensity of all involved to get such estimates wrong, it would be a very brave person who would claim to know the correct value of any of those variables over a five-year parliament.  But in principle, simple mathematics shows that a debt which increases in absolute value by a smaller percentage than the rate of economic growth will end up reducing the ratio of debt to GDP, which is why the absolute value being used by Davies is irrelevant.  The same mathematics also demonstrates that when interest rates are lower than the rate of inflation, paying more interest in absolute terms can still result in a reduction in the percentage of government income committed to paying interest.
What we do know is that, as things stand today (and I accept that’s a very important caveat), the UK Government is effectively borrowing money interest-free.  It’s costing us, in real terms, absolutely nothing, and given the demand from people who want to lend money to the government, there is no immediate problem in borrowing more.  Indeed, some would even argue that increasing government spending actually generates more tax income than the amount spent: the calculation all depends on the value assigned to the infamous ‘multiplier’.
Now of course it is true that different economists will give different answers to questions such as these, but that merely serves to underline that economists base their predictions on models rather than on absolute truths, and there are a number of different models available.  As the famous statistician, George Box, said, “All models are wrong, but some are useful”.  It’s a point worth bearing in mind that when politicians state categorically what the outcome of a particular policy will be for the economy they are depending on a model of some sort, whether they admit it - or even realise it - or not.
As I said at the beginning, this sort of guff from Davies is all good knock-about politics, but it’s really froth; he has no more clue than do I about the accuracy of what he says.  The real question is why one particular model – the idea that the government is like a giant household, which is used by the Tories when they come out with this stuff – is taken as gospel truth by a media which regularly demands that politicians from other parties explain themselves in the terms mandated by that model.  It would be more useful to political debate – let alone to economic policy – if the idea which underlies much of what they say was challenged more forensically rather than being simply accepted.  And it’s a shame that more opposition politicians don’t appear to have the understanding or the confidence to do that.

No comments: