26 November 2018

The problem with the latest Brexit growth forecasts


The latest Brexit analysis from the National Institute of Economic and Social Research (NIESR) is the usual worthy read, but says nothing new. It doesn’t provide any compelling evidence for or against staying in the EU, or accepting No.10’s proposed deal instead of “no deal”, despite what the usual suspects are claiming.

Let’s first deal quickly with the funding issue. The accompanying press release acknowledges that the NIESR report “has been prepared for the People’s Vote campaign”, which has prompted some to question its independence. I think that is unfair. The results are not much different from what NIESR has said before and the analysis is clearly explained.

It is also worth noting that NIESR’s forecasts of the immediate economic impact of the 2016 vote were at least more realistic than the Treasury’s pre-referendum analysis. NIESR predicted that UK GDP would be 2.3 per cent lower than otherwise in 2018, compared to a decline of 3.6 per cent in the Treasury’s “base case” and 6.0 per cent in the “downside scenario”.

But that’s enough professional courtesy from one economist to others. In my view, the NIESR analysis is still too pessimistic this time as well.

The report assumes that No.10’s Brexit deal eventually results in a relatively comprehensive Free Trade Agreement (FTA). In this scenario, the level of GDP in 2030 is projected to be 3.9 per cent lower (or approximately £100 billion), relative to a baseline where the UK remains in the EU.

NIESR has also modelled two alternative scenarios; one with the Irish backstop remaining in place, where staying in the customs union is assumed to reduce the hit to 2.8 per cent, and an “orderly no deal” scenario, where GDP is 5.5 per cent lower.

All these figures are in the same ballpark as Whitehall’s own preliminary estimates for these scenarios, and other assessments from the likes of the OECD, IMF and LSE. But that’s no surprise, since these studies all make similar assumptions and judgements about what Brexit will mean in practice, and apply these inputs using similar methodologies to similar models. If these inputs and tools are wrong, or simply incomplete, the conclusions are likely to be wrong too.

In this case, NIESR’s headline 3.9 per cent (£100bn) fall in GDP for the proposed Brexit deal comes mainly from three sources (with each being more negative in the “no deal” scenario).

The least contentious is a potential hit to trade, though this still assumes both a large and damaging increase in barriers to trade with the EU, and only a small reduction in barriers to trade with the rest of the world. These assumptions are debatable.

More controversially, NIESR’s results depend on a big and harmful reduction in net migration from the EU (not offset by increased migration from the rest of the world). That’s a possible result of Brexit, of course, but it isn’t inevitable. The outcome would depend on policy choices still to be made by this and future governments. It certainly seems odd for opposition politicians to cite this as an argument against Brexit, unless they never expect to be in power again.

Underlining the importance of this assumption, NIESR also provides estimates of the hit to per capita GDP, thus taking away any reduction in the size of the economy due simply to there being fewer people. Interestingly, the estimated hit under the “orderly no deal” scenario is then only a little larger (3.7 per cent) than with the “deal + FTA” (3 per cent).

In addition, NIESR assumes a big hit to productivity, especially in services. Part of this seems to follow from the assumptions about trade and migration (more open economies tend to be more productive). But again, these are only assumptions.

It seems just as likely that Brexit, if done well, would boost productivity. New trade deals with the rest of the world could open up more of the economy to competition, while new approaches to regulation could make the economy more dynamic.

Finally, it can’t be stressed enough that the 3.9 per cent is an estimate of how much lower the level of GDP might be in 2030, relative to a rosy view of what could happen if we remain in the EU. Even if correct, the UK economy would still be a lot larger (and incomes higher) than it is now. If we make a conservative estimate that baseline annual growth would average 1.5 per cent between now and 2030, GDP would rise by around 20 per cent. This increase would therefore still be around 16 per cent, even if Brexit lowers growth by a cumulative 4 per cent.

That’s a crucial point, because many Remain supporters are presenting NIESR’s results as evidence that Brexit would cause another recession, with resulting job losses, budget cuts, and so on.

David Lammy, for example, has gone straight for “Brexit will make GDP shrink by 3.9 per cent annually” and warned of “further chronic underfunding of hospitals, schools and other public services, countless businesses closures, and a broken social contract for the young”. I know he’s no economist, but that’s pure Project Fear.

Julian Jessop is Chief Economist and Head of the Brexit Unit at the Institute of Economic Affairs.