It seems that not a day goes by without another Brexit economic forecast – whether it is one from the Treasury, the OECD or Economists for Brexit. Some say it will cost Britain to leave; others say it will be beneficial to the UK economy.
If a report favours remain, the leave side is quick to criticise it as fear mongering and politically motivated. If it favours Brexit, the remain side is fast to tarnish it as unscientific and politically motivated. So who should we trust? Can we trust any of them at all?
Most of the forecasts estimate what UK income levels would be in 2020 and in 2030. For the sake of comparison, there are three main types of forecasts and we can refer to them by their average headline effects: plus 4%, zero effect and minus 7%.
At one extreme, Economists for Brexit predict that the main economic consequence of Brexit is that UK incomes in 2030 will be about 4% higher.
In the middle, there are various studies that suggest that UK incomes by 2030 will be will be unaffected. In this light, Brexit and/or the UK membership in the EU is pretty much immaterial.
At the other end, various studies (including the Treasury, the LSE, the OECD, and the National Institute for Economic and Social Research reports) indicate substantial losses to the UK economy, of about 7% by 2030.
To be more precise, the Treasury, LSE, OECD and National Institute predict short-term income losses of about 3.6%, 2.6%, 3.3% and 2.3%, respectively, and long-term losses of about 6.2%, 7.5%, 5.1% and 7.8% respectively. The Bank of England and the IMF have spoken about the potential costs of Brexit but have not presented forecasts.
Mind that such apparently small figures can be misleading: as Nobel-prize winning economist Paul Krugman notes in this context “2% is a lot”. Plus, these latter group of estimates are also in line with the net benefits the UK historically enjoyed from its membership in the EU which are estimated to be around 8.6% in its first ten years.
Transparency and grounding
So what is the essential difference between all these forecasts? Clearly, there is only one that predicts a positive effect. In the middle ground there are a number of slightly older studies mostly authored by think tanks. Supporting the view that Brexit would entail substantial economic losses, there are quite a few studies.
These forecasts differ in two fundamental ways. They differ in the transparency of their method and the grounding of their key assumptions.
To trust a forecast, it is necessary to know how it is made. If estimates cannot be replicated and if you do not know how to arrive at a certain figure, you have far less reason to trust it. It can be an imaginary or subjective number that an “expert” or a politician likes to think is a good approximation to the future.
It is abundantly clear that studies in the “minus 7%” group are superior to the others in this regard. They provide extensive details of how their figures are arrived at so that, everything else being the same, one can trust them more.
The second factor that helps make a forecast more trustworthy is the quality of the assumptions it uses. Are these realistic? Are the numbers being used as inputs into the modelling confirmed by previous research? Do they fall within what most people believe is a reasonable range of values?
The three groups of forecasts vary significantly in this regard and a most illustrative example is how the costs of regulation and EU membership are treated. The “minus 7%” group often assumes these to be very small. The “zero effect” studies tend to set the costs of regulation at about the same size of the benefits from EU integration, such that they cancel each other out. This yields a small range of values and may look balanced and serious, but when you consider that these figures do not often come with methodological details, you better be suspicious.
The “plus 4%” study uses costs of regulation that are absurdly large, of the order of 6% of the UK’s GDP. The problem is, in the real world, these figures are much smaller – less than a sixth of this is a self-professed “conservative” estimate of Economists for Brexit. Clearly, the larger the costs assigned to EU regulation, the better the Brexit option looks. But this is not grounded in the bulk of research and how they arrive at these large costs is unclear. So this lack of transparency impedes proper judgement of the quality of this assumption.
Predicting the future of human actions (and interactions and expectations) is not easy. The leave campaign likes to single out the forecasts made at the time the UK was considering joining the euro. What is seldom mentioned is that nobody at the time of the eurozone’s formation expected those within the monetary union to allow large chunks of their currency trade to take place outside of the eurozone. Thanks to a ruling by the European Court of Justice, the UK’s membership of the EU is a key reason why euro clearing houses remain in the UK.
The verdict from forecasts that use reasonable assumptions and are transparent about their methodologies is clear: Brexit will make the UK permanently poorer. They draw on lessons from history – including the benefits brought in 1973 when the UK joined the EU – and use the available information to conclude that the expected economic losses from Brexit will be indeed substantial.
But, though the economic argument has been won, this does not mean it will be heeded. The debate is clearly moving on to issues of sovereignty and migration, and, if the past is any guide, is likely to become rather more unpleasant.