The UK’s vote to leave the EU in June 2016 has sparked concerns that Brexit may be leading UK firms to redirect investment to other countries.
In particular, there is substantial anecdotal evidence that the threat of reduced access to the EU market after Brexit has pushed UK firms into setting up subsidiaries or acquiring companies in the remaining EU member states. For example, media reports have documented that both large UK companies such as Barclays, HSBC and EasyJet, and smaller companies such as Crust & Crumb, a Northern Irish pizza maker, have invested in the EU27 in response to Brexit.
In new research, we study whether the anecdotal evidence of increased EU investment by UK firms is representative of a systematic change in outward FDI. The form that Brexit will take remains uncertain, but it is likely to lead to higher barriers to trade and migration between the UK and the EU. Higher trade barriers would make it more costly for UK-based firms to export to the EU; consequently, there is an incentive for UK firms to invest in other EU countries as an insurance policy against Brexit.
We measure FDI activity through a count of announced greenfield and M&A transactions; greenfield activity refers to investments that create new establishments or production facilities from scratch, for example setting up a new factory, while M&A transactions refer to the acquisition of existing facilities. Our analysis focuses on the period from 2010 to 2018, during which we observe about 100,000 transactions in total.
A cursory look at our data does indeed suggest that UK companies have been increasingly investing in the EU27 after the 2016 referendum. For example, the evolution of FDI into the EU27 prior to June 2016 was similar for the UK and the non-EU OECD countries, with transaction counts showing an upward trend until 2016. But while FDI from the non-EU OECD countries stagnated after 2016, FDI from the UK rose sharply in the second half of 2016 and the first half of 2017 before falling in lockstep with non-EU OECD FDI afterwards.
Synthetic control method
We employ the synthetic control method to analyse the impact of the Brexit vote more formally. This involves constructing a synthetic control, or ‘doppelganger’, that can be interpreted as the expected trajectory of UK FDI if there had not been a 'leave' vote. The doppelganger is calculated as a weighted average of FDI transactions between other developed countries, with the aim of matching UK FDI flows into the rest of the EU as closely as possible before the referendum.
In practice, our doppelganger algorithm assigns the largest weights to FDI flows into the EU27 from Switzerland and the US. This makes intuitive sense as, like the UK, both Switzerland and the US are advanced economies that have a close economic relationship with the EU and are important origin countries for FDI into the EU27.
Once the doppelganger has been constructed, it can be compared to the actual evolution of UK-EU27 FDI flows. If the referendum outcome had no discernible impact on UK FDI behaviour, then the doppelganger and the actual series of FDI counts should be similar not only before, but also after the referendum.
Outward UK investment
The number of FDI transactions from the UK into the EU27 did indeed go up substantially after the second quarter of 2016 compared with the doppelganger, which remained at 2014 and 2015 levels. Overall, our results suggest that UK-EU FDI activity in the post-referendum period was about 12% higher than it would have been had the UK opted to 'remain'.
In terms of value, we estimate that these additional FDI outflows from the UK to the EU27 were worth approximately £8.3bn ($10.85bn) in total by the third quarter of 2018. Moreover, the persistence of the gap between the actual and doppelganger series of FDI count indicates that the referendum effect has not yet died away, meaning the increase in outward FDI due to Brexit is likely to grow further as more data becomes available.
Did this increase in outward FDI occur in all sectors? We split our sample between the manufacturing and services sectors and constructed a separate synthetic control for each sector.
We find a sizeable increase in outward FDI for the services sector but none for the manufacturing sector. This suggests that the aggregate effect of growing UK FDI into EU27 is entirely driven by services. This result is consistent with the view that the UK government has prioritised the interests of manufacturing over services in the Brexit negotiations by focusing on reducing customs frictions, while ruling out membership of the EU’s single market.
Alternatively, it could simply indicate that it is easier to set up new foreign affiliates in services industries rather than in manufacturing, where more expensive capital investments are needed.
Is the increase in FDI from the UK specific to the EU as a destination, or do we observe similar changes in UK investment flows to other countries? If so, our results may simply show that UK firms have become more internationally oriented and outward looking since the referendum. To evaluate this possibility, we constructed a synthetic control for UK investment into non-EU OECD countries (Australia, Canada, Chile, Israel, Iceland, Norway, Japan, Mexico, New Zealand, South Korea, Switzerland, Turkey and the US).
In contrast to the EU as a destination, we do not observe an increase in UK investment activity into non-EU OECD countries. That is, UK investment in advanced economies outside Europe has not experienced a post-referendum surge. We find no sign of a ‘Global Britain’ effect.
The opposite direction
The threat of a loss of market access after Brexit could also have led to more investment by European firms in the UK. To see whether this has happened, we constructed a synthetic control for FDI from the EU27 to the UK.
Relative to the synthetic control, investment from the EU27 to the UK went down by about 11% after the referendum, amounting to £3.5bn of lost investment. This finding is consistent with previous research by Ilona Serwicka and Nicolò Tamberi from the UK Trade Policy Observatory, who use the synthetic control method to present evidence that the referendum led to a decline in total UK FDI inflows, not only from the EU27 but also from the rest of the world.
This asymmetry highlights how the UK and the EU are differentially exposed to the effects of Brexit. Put simply, because the EU is a much bigger market than the UK, access to the EU27 is more important than access to the UK.
We show that the Brexit vote has led to a 12% increase in the number of new investments made by UK companies in EU27 countries. The increase in UK investment in the EU27 is entirely driven by the services sector. Although it is not possible to be certain about the reasons behind companies’ investment decisions, our results are consistent with the idea that UK companies are offshoring production to the EU27 because they expect Brexit to increase barriers to trade and migration, requiring them to have a local EU presence to continue to do business there.
By contrast, investment in the opposite direction from the EU27 into the UK has declined by 11%, highlighting the fact that Brexit is more of a problem for UK than EU companies.
Finally, we find no evidence of a ‘Global Britain’ effect. UK firms have not increased their investment in OECD countries outside the EU27.
Holger Breinlich is a professor of economics at the University of Surrey, a research fellow of the Centre for Economic Policy Research (CEPR) and a research associate at the Centre for Economic Performance (CEP) at the London School of Economics. Elsa Leromain is a research economist at the CEP. Dennis Novy is associate professor of Economics at the University of Warwick, a CEPR research fellow and a research associate at the CEP. Thomas Sampson is an associate professor of Economics at the London School of Economics and a research affiliate of the CEPR. This article is based on research funded by the Economic and Social Research Council (research grant ES/R001804/1) and published as part of the CEP Brexit Brief series.