Money Matters

Brexit: Politics, not economics, has a forecasting problem

February 5, 2018
By Chris Giles.


A year ago, Andy Haldane caused a stir when he said that economists’ failure to foresee the 2008-09 crash had been the profession’s “Michael Fish moment”. Recalling the BBC weather forecaster’s inaccurate on-air dismissal of hurricane risks in 1987, the Bank of England’s chief economist noted how weather forecasters had upped their game with the application of better data and computing power. Economics, he suggested, should follow suit to resolve its own crisis.

With cross-departmental ministerial support, the government economic service duly took Mr Haldane’s advice and has built a computable general equilibrium model of the sort that all countries use in trade negotiations. It has more computational heft and uses more data than the Treasury brought to bear in its pre-referendum assessment of the long-term effects of Brexit.

How do we know this? Because Philip Hammond lauded the government’s “wholly new” tool to MPs. He boasted the model could look at Brexit questions on a country-by-country basis, examine every sector of the economy and look at the effects of bilateral trade pairings.

So long as the chancellor was not misleading parliament, we also know the whole of government, right up to the prime minister, finds the new model useful. Mr Hammond told MPs that “output from that model will have informed some of the thinking that has gone into preparing the 14 [Brexit] papers that have been published”.

The more grit you throw into a smoothly running system, the worse the outcome

It must have come as a bit of a shock to government economists that the moment some results of this new model were leaked this week, ministers rushed to deny the usefulness of the tools they commissioned. Such models are “always wrong”, declared Steve Baker, a junior Brexit minister, on Tuesday.

The model’s output showed a hard Brexit would impart an 8 per cent hit to the size of the economy after 15 years compared with the alternative of EU membership. A soft Brexit, in the single market, would still lead to a 2 per cent loss, and a Canada-style free-trade agreement with the EU27 would knock 5 per cent off national income. It also found that signing free-trade agreements are worth little. A deal with the US, for example, would provide an uplift of only 0.2 per cent after 15 years.

On one level, Mr Baker is correct. The model cannot predict the future to the accuracy of tenths of a percentage point. That is not its purpose, as the chancellor rightly stated. It is there to provide a guide to thinking and policymaking.

The ordering of the Brexit results accords with easily understood economic principles. The more grit you throw into a smoothly running system, the worse the outcome and if you ditch profitable longstanding relationships in the vague search of something new, pain is likely to exceed gain.

Nor are the findings crystal-ball gazing of the sort described by Mr Baker. They are simple conditional statements along the lines of “the thicker the clouds, the more likely it is to rain”.

Adding to their credibility, the results also tally with almost all other studies of the same question, which have been built using many different methods. Sense About Science, the successful charity that challenges misrepresentation of scientific evidence in public life, considers findings to be much more reliable when they are widely replicated by many different organisations.

Ministers now have a choice. They can opt for an honest Brexit in which they argue in public that people should pay an economic price for their policies. Or they can opt for a dishonest Brexit, pretending they have a secret plan for economic nirvana and trashing their own internal economic evidence. Ministers’ initial reaction in disowning the analysis suggests deception is the government’s central Brexit strategy.

People talk about a crisis in economics. After this episode, it is the crisis in politics that should really concern us.