Deal, or no deal? The £2,000 question


Recent leaks of a Brexit impact study produced for the Department for Exiting the EU have reignited the debate about the costs of leaving the EU without a comprehensive trade agreement. The reported magnitude of estimated aggregate effects for such a ‘no-deal’ scenario is very similar to estimates published independently from each other prior to the Brexit referendum and also in line with updated work presented in our latest National Institute Economic Review: a loss in annual GDP relative to what it would otherwise have been of 7 to 8 per cent within the next 10 years. Put differently, annual income per head would be up to £2,000 less, compared to a scenario in which the UK remains in the EU’s single market.

What remains less clear in the debate is how these numbers came about, leaving room for political attack. This blog explains the assumptions behind our analysis.

We started off considering different scenarios. Under a ‘soft’ Brexit, Britain would remain part of a single market with the EU and trade in goods and services would be frictionless. In return, the four freedoms, including free movement of labour, would be accepted and the UK would continue to make financial contributions to the EU budget. But, alternatively, talks may fail and the UK might end up trading with the EU on WTO rules under a ‘no-deal’ scenario. In addition, several intermediate scenarios could be conceived, with a varying intensity of trade barriers and different levels of labour movement and financial contributions (‘The Great British Trade-Off’). There are other possibilities that fall outside such a spectrum of scenarios. A termination of the Brexit process, for instance through a second referendum, or a disorderly Brexit with substantial uncertainty could also be considered.

In our latest forecast, we adopt the ‘soft’ Brexit scenario as our central case. This is because the official position of the government remains to achieve the “freest and most frictionless” trade possible in an agreement with the EU. In addition, the Phase 1 negotiations resulted in the commitment not to erect physical borders between Northern Ireland and the Republic of Ireland, and between Northern Ireland and the rest of the UK. By logical extension, this implies borderless trade between the UK and the EU.

To calculate the cost of an alternative ‘no-deal’ scenario, we build on our earlier study by Ebell et al. (2016) and think of Brexit as a set of different shocks to the economy that compound each other. The response to these shocks is simulated in our estimated structural model of the world economy, NiGEM, relying on standard assumptions about the economic structure.

Which shocks did we consider?

1. Trade

Under our ‘no-deal’ scenario, exports to the EU are reduced immediately and by a substantial amount. Some services will no longer be exported to the rest of the EU, e.g. financial services that require passporting rights. We assume that the size of UK exports in EU imports drops by around 35 per cent on the day of Brexit. Over the course of the 10 years, total trade with the EU reduces by 60 per cent relative to our baseline ‘soft’ Brexit scenario. These numbers are inherently uncertain for the simple reason that we do not have relevant examples of countries leaving the EU, or for that matter any free trade arrangement, but the empirical literature agrees that free trade agreements lead to a substantial increase in trade flows.   

Some have argued that trade deals with other countries will offset the loss in EU market access. EU members enjoy comprehensive access to each other’s market in both goods and services. By contrast, there are no current trade agreements that are as comprehensive in services outside the EU. While this does not imply that new deals are impossible, the time it will take to negotiate them will be considerable. Our working assumption is that trade with the rest of the world will not increase substantially after Brexit but existing agreements between the UK and non-EU countries will remain in place.

2. Foreign direct investment

Compared to our central case, in which uncertainty somewhat dampens investment, our ‘no-deal’ scenario assumes that foreign direct investment falls by a lot more - around 20 per cent compared with our base case scenario. Recent empirical work suggests that the reduction in trade will likely make the UK a less attractive FDI destination.

3. Repatriation of EU budgetary contributions

Net contributions of the UK into the EU budget in the recent past have been around £10 billion per annum, or 1 per cent of total government expenditure. In our ‘no-deal’ scenario, we assume that the UK honours its commitments to the current EU budget until 2020. Afterwards, net contributions halve but do not fall to zero as outstanding liabilities will need to be paid and the UK may decide to continue being part of initiatives such as Horizon 2020, for which contributions will be due.

4. Net migration

Our central forecast is based on principal population projections provided by the Office for National Statistics, which already factor in a slowdown in net migration. We employ their ‘low migration’ variant in our ‘no-deal’ Brexit case whereby annual net migration is being reduced by around 100,000. We assume that the lack of a trade agreement could enable the government to restrict migration from the EU more than under a ‘soft’ Brexit. In addition, the reduction in trade and investment may make the UK a less attractive destination for workers from the rest of the world.

5. Productivity

The lacklustre evolution of UK productivity since the financial crisis has often been described as a ‘puzzle’. Similarly uncertain is the overall effect of Brexit, which likely compounds some existing factors dampening productivity growth. In the short term, a reduction in trade could lead to a rebalancing of the economy to sectors that are relatively less productive, in order to meet domestic demand. Over time, the smaller exposure to competitive international markets, the reduction in foreign direct investment and less skilled migration are likely to reduce firms’ propensity to innovate. On the other hand, a reduction in unskilled migration may provide an incentive to encourage automation and thereby increase productivity growth. Overall, we therefore consider a gradual reduction in labour productivity, which accumulates to a loss of 3 per cent over the course of a decade and, in our view, can be considered a conservative estimate.



The effect of a ‘no-deal’ Brexit relative to a ‘soft’ Brexit baseline can be analysed along several dimensions: the growth rate of GDP, the level of GDP, components of GDP and economic welfare measured by GDP per capita.

Figure 1 depicts the growth rate of GDP under both scenarios. Our results imply that the UK economy would enter a mild recession within a year of exiting the EU without a trade deal, compared to a growth rate of just below 2 per cent in the ‘soft’ Brexit case. This is because the erection of trade barriers leads to an immediate reduction in exports. Over time, the growth rate of GDP would settle at a somewhat lower rate of 1 per cent as net migration falls and the productive capacity grows at a slower pace.



The impact on the level of GDP is shown in Figure 2. The difference in annual output between both scenarios gradually increases over time and reaches between 7 and 8 per cent by 2029. This is shown by the black line. The coloured bars illustrate the effect of each of the Brexit-related shocks. Of these, the decline in trade, which can be thought of as a large demand shock, has the biggest effect initially. This is because the reduction in exports takes away a share of GDP by assumption. In addition, it leads to a reduction in export prices causing the exchange rate to depreciate. This has a small offsetting effect on the volume of exports but leads to an increase in import prices. Higher import prices raise consumer price inflation and consumption falls. Over time, effects from productivity as a supply side shock start to dominate. Similarly, the reduction in net migration contributes to a permanent fall in GDP growth. The direct effect of the reduction in FDI on economic output is small but negative. Repatriating budgetary contributions to the EU can be thought of as a fiscal expansion, but effects are comparatively small and short-lasting.



Figure 3 disentangles the combined effect of Brexit-related shocks on the components of GDP, relative to the ‘soft’ Brexit baseline. This illustrates how agents in the economy are affected: consumers, firms, or the government sector. Initially, business investment suffers most due to the reduction in foreign direct investment. Economy-wide spillovers from the reduction in trade further dampen domestic investment. Over time, consumption, which is the largest expenditure component in overall income, is hit hardest. Government expenditure also suffers from a reduction in the productive capacity of the economy. The impact on trade, i.e. the difference between exports and imports, follows an interesting pattern. As trade barriers are imposed, exports decline without delay. Over time, however, the impact of the ‘no deal’ Brexit scenario on net trade, and therefore GDP, is positive. Given that we do not assume a substantial broadening of non-EU export markets post Brexit, this effect is driven by a decline in imports. Exchange rate-induced rises in import prices deter domestic consumers from buying foreign goods in our analysis.

What does all that mean for the welfare of UK residents? A measure of the welfare impact on the average UK resident is provided by differences in GDP per capita across scenarios. The red line in Figure 2 plots the welfare cost of a ‘no-deal’ Brexit over time. Initially, some £500 are deducted from average annual income. This increases to around £2,000 over the course of a decade, or 6 per cent of GDP per capita under a ‘soft’ Brexit. Given that less migration means overall GDP is shared with fewer people, the percentage reduction in GDP per head is smaller than for total GDP, while all other channels make similar contributions. A caveat is that aggregate effects largely hide what the economic cost of Brexit will be to individuals. The National Institute recently published more detailed estimates by region, sector and income group which which confirm that, while economic effects are negative for nearly everyone, some people are likely to lose out more than others.​


Do you have any questions on this blog? Send them to @NIESRorg on Twitter with the hashtag #AskNIESR. The authors will address your queries in a Live Q&A on Friday 23rd February at 3pm.





Total views: 5,210