Our forecasts must be prepared on the basis of current government policy. Before the EU referendum, that policy was to remain in the EU, so that was the basis for our March 2016 forecast. While we made no assessment at that stage as to what the economic and fiscal impacts of Brexit might be, in every forecast we highlight particular risks and uncertainties around our central projections. This box discussed Brexit as a particular source of uncertainty, by highlighting some pieces of external analysis which showed a wide range of views as to the possible impacts on trade, productivity and GDP growth.

The Government has announced that a referendum will be held on 23 June to determine whether the UK should remain a member of the European Union (EU) – and the Government is arguing that it should. Parliament has told us to prepare our forecasts on the basis of the current policy of the current Government and not to consider alternatives. So it is not for us to judge at this stage what the impact of ‘Brexit’ might be on the economy and the public finances.

Outside analysts have of course addressed this question. For example, a study published by the Centre for Economic Performance estimates that leaving the EU would result in lower trade and therefore lower GDP. It presents a ‘pessimistic’ scenario where incomes could fall by close to 10 per cent.a Conversely, a study published by the Institute of Economic Affairs argues that leaving the EU could increase UK GDP by 13 per cent.b The range of estimates in part reflects sensitivity to assumptions about what exactly would replace the current rules that are attached to EU membership. That was also apparent in the views presented at the National Institute of Economic and Social Research conference on the ‘Economics of the UK’s EU Membership’ last month.c

These estimates are as large as they are in part because they incorporate ‘dynamic’ effects, reflecting for example long-term changes in UK productivity. As well as being highly uncertain, these take many years to materialise, with IMF research suggesting that it takes around 10 years for half the effect of changes in the trade share of GDP to be seen in income levels.d So even if we were to base our central forecast on an assumption of ‘Brexit’, the full impact would not show up within our five-year forecast horizon. A study by Open Europe modelled a scenario in which the UK leaves the EU in 2018 and found that GDP could be 2.2 per cent lower or 1.6 per cent higher by 2030, depending on the arrangements for trade and regulation that follow ‘Brexit’.e It argued that much of the transition to either of these levels would take place beyond 2020.

Leaving aside the debate over the long-term impact of ‘Brexit’, there appears to be a greater consensus that a vote to leave would result in a period of potentially disruptive uncertainty while the precise details of the UK’s new relationship with the EU were negotiated. For example:

  • Goldman Sachs expects that delayed business investment spending would have a “significantly negative” impact on UK growth;f
  • a JPMorgan study uses a VAR model to estimate that the uncertainty following a ‘leave’ vote could cause a 1 percentage point reduction in GDP growth in 2016.g Deutsche Bank predict a similar effect on GDP growth in the two-to-three years after a vote to leave;h
  • Scotiabank predicts that GDP growth could slow by 2 to 5 per cent over a one-to-two-year horizon, due to a “sharp drop” in consumer confidence and lower consumption;i
  • Bloomberg Intelligence modelled a fall in demand of 1.5 per cent of GDP, accompanied by an increase in credit spreads and a sterling depreciation. It argued that Bank Rate would be lower over our forecast period, with inflation higher initially but lower by the end of our forecast due to a persistent negative output gap;j and
  • a number of forecasters suggest that uncertainty could lead to a significant sterling depreciation (especially given the UK’s large current account deficit). Nomura estimate that sterling could depreciate by between 10 and 15 per cent following a vote to leave.k

There were only tentative signs that uncertainty regarding the referendum result was affecting business and consumer confidence and spending intentions by the time we closed this forecast.l But it may have contributed to recent financial market movements (and thus to some of the conditioning assumptions that underpin it). For example, sterling fell to a 7-year low against the dollar shortly after the date of the referendum was announced. That period fell within the 10-day window over which we have averaged market assumptions for this forecast.

This box was originally published in Economic and fiscal outlook – March 2016

a Centre for Economic Performance, The costs and benefits of leaving the EU, May 2014.
b Institute of Economic Affairs, Should Britain leave the EU? An economic analysis of a troubled relationship, February 2016.
c NIESR conference summary: Economics of the UK’s EU Membership, held in February 2016.
d IMF working paper, The long-run effects of trade on income and income growth, February 2003.
e Open Europe, What if…? The consequences, challenges and opportunities facing Britain outside the EU, March 2015.
f Goldman Sachs Economics Research, Brexit: The uncertainty shock of leaving the EU, March 2016.
g JPMorgan Economic Research, Brexit: What impact might uncertainty have on UK GDP?, February 2016.
h Deutsche Bank Research, The UK & EU: Exit emergency, February 2016.
i Scotiabank, Brexit – market and economic impact, February 2016.
j Bloomberg Intelligence, Brexit special: Modeling a surprise exit, February 2016.
k Nomura Economic Insights, Brexit carries a recessionary risk, February 2016.
l Investment intentions in the latest EEF Manufacturing outlook were at a six-year low, with the EU referendum cited as a possible cause.

Publications