BUSINESS INVESTMENT, PRODUCTIVITY AND BREXIT

Abstract

The United Kingdom has experienced significantly lower growth rates of business investment and labour productivity following its decision to leave the European Union, although this lacklustre performance was affected by the economic shocks caused by the COVID-19 pandemic and the Russian invasion of Ukraine in addition to Brexit. This article aims to quantify the impact of Brexit on business investment and labour productivity in the United Kingdom using the National Institute of Economic and Social Research’s Global Macroeconometric Model. We model Brexit as a decline in trade with the European Union and associated reduction in terms of trade, a decrease in productivity and a permanent increase in uncertainty. Our estimates suggest that these shocks have led to an approximately 12–13% decline in UK business investment in 2023, which gradually declines to 7–8% by 2035 as businesses adjust to the terms of trade and productivity shocks. This corresponds to a real gross domestic product (GDP) loss of 2–3% (about £850 per capita) in 2023 and 5–6% (about £2,300 per capita) by 2035. Additionally, we find that Brexit has reduced labour productivity by around 2–2.5% as of 2023, with a projected reduction of 5–6% by 2035.

1. Introduction

The decision to leave the European Union (EU) has significantly shaped the UK economic landscape since the 2016 referendum. Over the ensuing years, the United Kingdom has experienced an economic environment characterised by low growth and, for the post-pandemic era, high inflation. UK GDP projections before and after key events presented in Figure 1 show that UK GDP could have been substantially greater, had it not suffered from the major economic shocks of Brexit, COVID and the Russian invasion of Ukraine. To put it differently, the average UK resident could have earned 8–9% more (around £1,700 annually) and consumed 11–12% more (around £2,300 annually) by 2023 if the 2010–2019 trends in real income and private consumption had been maintained (Figure 2).

Figure 1. GDP projections.

Source: NiGEM database and NIESR forecasts.

Figure 2. Income and consumption (per capita).

Source: NiGEM database and NIESR calculations.

One important aspect of the lacklustre economic performance is associated with firms’ reluctance to invest given the uncertain environment following the referendum and reduced profitability of trade due to rising trade frictions after the formal separation. Lower business investment and the reduction in trade due to Brexit have further contributed to the decline in productivity growth that has been observed following the global financial crisis in 2008. In fact, with an average productivity growth of 0.3% (measured as output per hour worked), the United Kingdom ranks among the lowest performing countries within the G7 since 2016. However, the substantial decline in productivity and investment performance cannot solely be attributed to Brexit, given the notable economic damage caused by the pandemic (Mortimer-Lee and Pabst, 2022) and Russia’s war in Ukraine (Liadze et al., 2022) in the post-Brexit period. In this article, we aim to examine quantitatively how much of the falls in investment and productivity growth can be attributed to Brexit.

We build on the recent work by Hantzsche and Young (2019) and consider the final withdrawal agreement and the EU-UK Trade and Cooperation Agreement (TCA), which entered into force at the beginning of 2021. We consider three key channels through which Brexit could have affected the UK business investment and productivity. First, we argue that the reduction in trade in goods and services between the United Kingdom and the EU remains the most important channel through which the impact of Brexit will be visible over the long term. The TCA provides for the continuation of tariff- and quota-free trade in all goods, some provisions in services and cooperation between the United Kingdom and the EU in a wide range of areas, such as investment, competition, state aid, transportation, energy, data protection and social security. However, businesses—both exporters and importers—face higher trade costs, due to more stringent customs checks and forms to fill in, and higher nontariff barriers, for example, rules of origin requirements and regulatory barriers such as the loss of passporting in financial services because the United Kingdom is no longer part of the EU single market or customs union. Indeed, Clarke et al. (2023) highlighted that the primary challenges encountered by businesses in the context of the TCA revolve around issues of bureaucratic hurdles, shipping time delays and intensified controls by customs and border authorities. These costs act to reduce the UK terms of trade, making UK households either continue to purchase imported goods at higher relative prices or switch to more expensive domestically produced goods, both of which make them poorer. Lower household real income, in turn, implies lower consumption and lower business investment.

Second, we think Brexit led to a greater reluctance of businesses to invest in the United Kingdom due to the change in the business environment as a result of a fundamental rise in uncertainty. Investors and multinational companies find the United Kingdom potentially less attractive for investment as a non-EU country and this would reduce foreign capital flows in the form of portfolio and direct investment relative to the counterfactual. This is another channel that we consider that may have led to the fall in overall business investment and productivity growth in the United Kingdom.

Third, besides the indirect decline in labour productivity coming through the terms of trade and investment channels we mentioned above, we argue that Brexit led to a significant negative shock to total factor productivity. This is based on the premise that the reduction in effective competition resulting from an increase in barriers to trade, coupled with the departure of highly productive multinational companies, could result in an overall decrease in productivity in the United Kingdom over time (Mayer et al., 2021).

Another channel that could affect business investment and productivity is the potential loss of human capital. However, we do not consider this as a separate channel as international migration data show that the reduction in net migration from the EU due to the loss of free movement of people has so far been offset by the increase in net migration from non-EU countries since Brexit. Similarly, we assume no change in fiscal balances, as the impact of the reduction in the UK’s net contribution to the EU is negligible.

The literature has extensively explored the effect of Brexit on trade, productivity and investment; however, there has been less emphasis on linking these together now Brexit has taken place. Therefore, we think this work will contribute to the Brexit literature by bringing together the effects on trade, business investment and productivity. Our analysis using National Institute of Economic and Social Research’s Global Macroeconometric Model (NiGEM) suggests that the long-term impact of Brexit on UK business investment will be around 7–8%. Furthermore, we find that Brexit will reduce UK real GDP by about 5–6% of GDP or about £2,300 per capita by 2035. Labour productivity, as measured by output per hour worked, will also be affected by about 5–6% as of 2035. An extension of this study will be developing a three-country DSGE model of international trade based on Jaef and Lopez (2014), where the three countries correspond to the United Kingdom, the EU and the Rest of the World, within which we can more carefully examine the mechanisms at work and obtain further quantitative estimates of each of the effects.

The rest of this article is organised as follows. Section 2 summarises the literature on the impact of Brexit on the UK economy. Section 3 briefly describes the NiGEM model and how it captures the channels discussed above. Section 4 presents and discusses the results, and Section 5 concludes this article.

2. Literature review

In this section, we provide an overview of research on the possible consequences of Brexit for the short- and long-term prospects of the UK economy. The literature has extensively explored the effects of Brexit on the UK economy. Apart from studies that explored the direct link between Brexit and business investment or productivity, a large body of literature focuses on its impacts on trade and foreign direct investment (FDI) inflows, through which business investment and productivity can be affected. The presence of higher trade opportunities when the United Kingdom was part of the EU single market benefited both sides through lower prices and more diverse accessibility to goods and services (Dhingra et al., 2018). In addition to trade, being a part of the EU single market allowed foreign investors to leverage the United Kingdom as an export stage for accessing the EU markets. However, increasing tariff frictions, rules of origin requirements and regulatory barriers, such as the loss of passporting in financial services increase the costs of trading for businesses (Gretton and Vines, 2018). Higher barriers to trade and FDI have stifled competition and reduced opportunities for firms to exploit economies of scale, leading to lower efficiency, limited access to foreign technology and reduced opportunities for innovation (Ahn et al., 2019). As a result, the United Kingdom experienced a significant shock to trade, investment, productivity, and thus, welfare loss because of leaving the EU’s single market (Dhingra et al., 2018; Latorre et al., 2020). In this section, we provide an overview of the empirical research on the short- and long-term consequences of Brexit on the UK’s business investment and productivity.

In the run-up to the referendum, Baker et al. (2016) analysed the short-term effects of the decision to leave the EU by using the NiGEM. In order to capture the effect of rising uncertainty and deterioration of expectations in financial markets, they introduced various shocks to the exchange rate, government yields, corporate and household lending spreads and the equity risk premium for the period of uncertainty following the referendum. They found that overall impact of widening risk premium through escalated uncertainty following the referendum was a substantial initial decline in business investment of about 15% relative to the counterfactual, followed by a gradual return to the baseline. HM Treasury (2016a) estimated that business investments would be the primary source of decline in UK output growth in the short term due to increased uncertainty and financial market volatility.

Other studies using NiGEM have focused on the long-term effects of Brexit. Among them, Ebell and Warren (2016) argued that Brexit would affect the EU economy through trade, FDIs and fiscal channels. In this article, they assumed a significant trade decline with the EU, a drop in FDI inflows, as well as a reduction of EU fiscal contributions under different about how the Brexit negotiations would unfold. Their results show that Brexit would lead to a sharp decline in business investment of up to 15% following the referendum in 2016. This initial sharp decline was projected to gradually stabilise by 2030, leading business investment to be 2.5% lower than the counterfactual scenario that the United Kingdom remained in the EU. OECD (2016) found that the immediate effects of Brexit on business investment exceed 10% due to the disruption of business expectations and the precipitous depreciation of the pound. Over the long term, they estimated that the shocks to total factor productivity and migration diminish potential output and reduce the capital stock of the business sector by approximately 9%.

Hantzsche et al. (2018a) published a more comprehensive analysis following the first proposed Brexit deal, by also considering the loss of migration and decline in labour efficiency assumptions in addition to the trade, FDI and fiscal channels in Ebell and Warren (2016). Their estimates show that total investment—by companies, homeowners and the government—would be 4 to 5% lower by 2030. They also argued that the prolonged period of underinvestment leads capital stock to be 3 to 5% lower in the long run, reducing the potential output by 3 to 4%. Hantzsche and Young (2019) reached a similar decline when modelling the final proposed Brexit deal, primarily due to increased trade and migration barriers and decreased productivity growth. Further research showed that although the short-term adverse effects of Brexit can be mitigated with accommodative monetary and expansionary fiscal policies, such measures would not address the fundamental structural issues stemming from Brexit, especially changes in trade and investment relations with the EU (Hantzsche, 2019). In a recent paper, Kaya et al. (2023) revisited the assumptions made by Hantzsche et al. (2018a) and found that Brexit has resulted in UK business investment being approximately 12% lower than the counterfactual in 2023. They estimated the impact of Brexit on UK investment to be 7–8% by 2035.

Several studies focused on Brexit’s impact on firm-level investment decisions. Among them, Simionescu (2017) estimated that Brexit could cost from 65% to 90% of FDI projects in the United Kingdom based on gravity model approach based on mixed-effects Poisson model and differences-to-differences estimators. Using a similar method and firm-level data, Górnicka (2018) analysed the impact of post-Brexit potential trade costs. The results show a significant and negative influence of potential trade costs on firm investment after the Brexit referendum, although the depreciation in sterling may have contributed to the increased investment expenditure of foreign-oriented firms in the United Kingdom. Using macroeconomic data for OECD countries, Welfens and Baier (2018) found that leaving the EU single market significantly reduces the FDI inflows. Moreover, the synthetic control method by Breinlich et al. (2020) showed that UK investment in the EU increased by 17% within 1 year following the Brexit referendum because of UK firms’ aim to retain access to the EU market after formal separation. They also found that new investments from the EU to the United Kingdom fell by about 9% following the referendum.

Bloom et al. (2019) exploited the Decision Maker Panel, a survey that collects data on several thousand firms on a monthly basis, to assess the investment and productivity implications of the substantial surge in uncertainty caused by the Brexit referendum. Their findings suggest that the anticipation of Brexit was associated with a gradual decline in investment of 11% in the 3 years following the referendum. This trend unfolded more slowly than initially predicted, suggesting that the prolonged and significant uncertainty may have delayed firms’ reactions to the Brexit vote. In addition, they found that the Brexit process has reduced the productivity of UK firms by between 2% and 5% over the same period. This reduction is mainly due to within-firm effects, as firms devote significant top management time to Brexit planning, while there is evidence of a reduction in between-firm productivity effects, reflecting that more productive, internationally exposed firms were more adversely affected than less productive domestic firms. An updated version of this study revealed that the impact of Brexit on firm-level investment was even more pronounced, at around 23% by 2021, largely due to high uncertainty, which then fell significantly with the TCA, helping investment to recover in subsequent years (Anayi et al., 2021).

Springford (2022) used a synthetic control method and constructed a doppelganger United Kingdom that did not exit the EU to analyse the impact of Brexit on UK investment, GDP and trade. Despite the analysis being rooted in data from 22 other advanced economies, the doppelganger United Kingdom is primarily composed of the United States, Germany, New Zealand, Norway and Australia, countries that experienced a similar economic performance to that in the United Kingdom before the referendum. The results indicate that UK investment (gross fixed capital formation) is 13.7% lower than the doppelganger United Kingdom in the last quarter of 2021, the latest observation in the sample. The results also suggest that the impact of the COVID-19 pandemic on the UK economy would have been smaller had the United Kingdom remained in the EU. Carella et al. (2023) investigated the factors influencing business investment in the United Kingdom at both a macroeconomic and firm level. For the macroeconomic analysis, they used annual panel data covering the G7 countries over the period 1980–2022. Additionally, their microeconomic analysis was based on annual data for 5,000 UK-listed firms between 1984 and 2022. Their results show that uncertainty linked to Brexit stands out as one of the primary catalysts behind the downturn in business investment in the United Kingdom post-referendum.

In addition to studies analysing the impact on investment, a great number of studies focused on the productivity implications of Brexit. Using NiGEM, Ebell and Warren (2016) estimated that labour productivity, defined as output per hour worked, would be around 3% lower in the long run. HM Treasury (2016b) also estimated the long-term productivity impact of Brexit between 3% and 8% based on different alternatives. Latorre et al. (2020) estimated that Brexit lowers the average productivity of UK firms by 2.3% in most manufacturing sectors as less productive firms enter the market due to reduced competition and increased protectionism. Although the exact estimates vary depending on estimation techniques and assumptions, almost all papers showed that UK productivity would be lower due to Brexit. A survey of literature by OBR (2020) showed that the average of the central estimates of the potential productivity impact of Brexit is around 4%. Finally, a recent study by Fingleton et al. (2023) found that Brexit would reduce productivity by 0.6% in the short term and 0.9% in the long term across all UK regions.

Overall, the literature portrays a broad consensus that Brexit has exacerbated the UK’s long-standing problems of underinvestment and low productivity as a result of additional barriers to trade and FDI inflows. However, some studies like Booth et al. (2015) and Minford (2019) argued that United Kingdom may even gain from the Brexit because of a more ambitious deregulation, avoidance of EU budget contributions and the control of unskilled immigration.

3. Model and transmission channels

3.1. NiGEM

We use NIESR’s global macroeconometric model, NiGEM, to examine the quantitative impacts of Brexit on productivity and business investment. NiGEM is the leading global macroeconometric model widely used by central banks, policymaking institutions and international organisations for macroeconomic forecasting, scenario analysis and stress testing. NiGEM is a complete global model in which most developed and emerging market economies are modelled individually, while the rest of the world is modelled through regional blocks. This ensures overall global consistency of trade volumes by stipulating that the growth of import volumes equals the growth of export volumes on a global scale. Because all countries are integrated through trade, competitiveness and financial markets, it is well-suited for modelling the impact of Brexit on the UK economy.

Individual country models incorporate long-run relationships grounded in the New-Keynesian framework, assuming forward-looking agents while allowing nominal rigidities to impede the adjustment process to external events. Although not all countries have full specifications, all country models include variables such as domestic demand, export and import volumes, prices, current accounts and net assets. Figure 3 illustrates the interlinkages between different macroeconomic sectors and variables for full country models, including the United Kingdom. As shown in the figure, output in the short run is determined by the demand side of the economy, while in the long run, it is influenced by supply-side factors, such as capital stock, total hours worked, labour-augmenting technical progress and oil input. Hantzsche et al. (2018b) provide detailed explanations about the equations underpinning NiGEM.

Figure 3. NiGEM full country specification.

Source: Hantzsche et al. (2018b).

As discussed in the previous work, the impact of Brexit depends on factors such as the importance of the United Kingdom’s trade relations with other economies, especially those within Europe, changes in competitiveness, labour market dynamics and fiscal and monetary policy responses. Following the previous work summarised above, we have focused on three main channels through which Brexit could have a longer-term impact on the UK economy: reduced trade with EU countries leading to a fall in the terms of trade and a reduction in real incomes, a reduction in the willingness to invest in the United Kingdom and a relative reduction in productivity, that is, a decline in labour-augmenting technical progress.

3.2. Terms of trade channel

We assume that Brexit will have a negative impact on trade between the United Kingdom and the EU and lead to a fall in the UK terms of trade as exporters and importers face higher costs of trading. The UK-EU Comprehensive TCA provides for the continuation of tariff and quota-free trade in all goods, as well as provisions on trade in services. It also provides for cooperation between the United Kingdom and the EU in a wide range of areas including investment, competition, state aid, transport, energy, data protection and social security. Despite this, there has still been a shock to UK trade coming from tariff frictions, rules of origin requirements and regulatory barriers such as the loss of passporting in financial services, given that the United Kingdom is no longer part of the EU single market or customs union, all of which increase the costs of trading for both importers and exporters (Gretton and Vines, 2018). It is also reasonable to assume that the introduction of stricter border controls and the imposition of full customs requirements for exports and imports will incentivise EU firms to seek more favourable trading partners within the Union.

Within NiGEM, there are two main transmission channels through which the change in trade relationship can affect the UK economy. First, its direct impact on output through the demand side (Equation (1)), with the overall impact depending on how it affects exports and imports:

(1) $$ Y=C+G+I+\left(X-M\right) $$

where Y denotes GDP, C denotes total consumption of households, I denotes business and public investment, G denotes government consumption spending and X and M denote exports and imports, respectively. Second, reduced trade with EU countries can lead to a deterioration in the terms of trade through export and import prices. Additional tariff and nontariff costs increase the prices of imported goods while the price received domestically for exports falls. As a result, the consumption basket becomes more expensive, reducing real incomes and thus, total consumption and investment.

(2) $$ \ln C=\alpha +\beta \hskip0.4em \ln \hskip0.4em RPDI+\left(1-\beta \right)\hskip0.2em \ln \hskip0.2em \left( RNFW+ RTW\right) $$

where RPDI represents the real personal disposable income, while RNFW and RTW denote real net financial wealth and real tangible wealth, respectively. Lower consumption leads to lower aggregate demand, which in turn reduces investment and the capital stock, resulting in a long-run decline of output:

(3) $$ K=\left(1-\delta \right){K}_{-1}+I $$

(4) $$ Q=\gamma \left\{{\left[s\;{K}^{-\rho }+\left(1-s\right)\;{\left(L{e}^{\lambda}\right)}^{-\rho}\right]}^{-1/\rho}\right.\;\Big\}{\;}^{\alpha }{M}^{1-\alpha } $$

where K indicates the capital stock, L is total hours worked, $ \lambda $ is the rate of labour-augmenting technical progress and M is oil input.

Previous work on the impact of Brexit on the UK economy assumed a sharp decline in trade with the EU, based on gravity model estimates. However, bilateral trade data show that the initial sharp decline in trade with the United Kingdom has been largely reversed in the post-Brexit period. In fact, as shown in Figure 4, UK imports from the EU have been above their short-term trend since 2016, while exports to the EU are close to trend. From a longer-term perspective, the EU share of UK exports and imports has remained relatively stable, despite an initial sharp decline. In the first quarter of this year, the EU shares of UK exports and imports were 41.4% and 50.9%, respectively (Figure 5).

Figure 4. Trade in goods and services with the European Union.

Source: ONS.

Figure 5. Share of European Union in UK trade in goods and services.

Source: ONS.

Although it is still too early to fully assess the long-term impact of Brexit on trade, preliminary data suggest that the impact will be materially less than suggested by previous gravity estimates. A recent survey conducted by Dhingra and Sampson (2022) indicates that the expected long-term effects of Brexit have not yet materialised, and there has been minimal trade diversion away from the EU thus far. Portes (2022) also suggests that although Brexit has had a significant impact on EU trade, the magnitude of the trade decline remains uncertain. In addition, OBR (2023) estimates that the volume of UK trade will be 15% lower in the long term than in the alternative scenario of the United Kingdom remaining in the EU. Given the share of the EU in UK total trade, we project a 25% reduction in UK-EU trade over 15 years. Within NiGEM, this implies a reduction in the terms of trade of around 1%, which translates into a fall in real personal disposable income of around 2.5%.

3.3. Productivity channel

As outlined in OBR (2023), Brexit is likely to have a significant long-run impact on productivity in the United Kingdom through several transmission mechanisms. First, rising costs of trading with the EU may reduce the incentives for high-productivity multinationals to invest in the United Kingdom, leading to a decline in the UK capital stock. There is already evidence that more than 440 firms in the banking and finance sector left the United Kingdom just 1 year after it left the EU (Hamre and Wright, 2021). Second, leaving the EU and losing new access to the EU pool of skilled workers may reduce overall productivity in the United Kingdom. Finally, higher barriers to trade and FDI can stifle competition and reduce opportunities for firms to exploit economies of scale, leading to lower efficiency, limited access to foreign technology and reduced opportunities for innovation (Ahn et al., 2019). Therefore, we consider a direct negative shock to productivity ( $ \lambda $ ) in addition to the terms of trade shock. Transmission of this shock to the UK economy is already apparent from Equation (4). A lower $ \lambda $ means less efficient labour input, and thus lower potential output in the long run. It should be noted that, there may be some productivity gains due to the potential benefits of deregulation in certain sectors and selective migration policies that favour high-skilled over low-skilled workers (Hantzsche et al., 2018a). However, these potential gains are likely to be too limited to offset the productivity losses due to the other impacts.

Previous NIESR work by Hantzsche et al. (2018a) and Hantzsche and Young (2019) assumed a relatively smaller fall in total factor productivity of 1–1.6% in the long run. Latorre et al. (2020) estimated that Brexit lowers the average productivity of UK firms by 2.3% in most manufacturing sectors as less productive firms enter the market due to reduced competition and increased protectionism. A recent study by Fingleton et al. (2023) found that Brexit would reduce productivity by 0.6% in the short term and 0.9% in the long term across all UK regions. However, we have based our assumptions on the extensive review of the literature contained in OBR (2020), which suggests that a 4% decline in productivity in the long run is plausible. In its analysis, the OBR took the average of estimates of the impact of Brexit on long-term productivity under a typical free trade agreement. They found that a third of this long-term effect has already occurred; so, we considered an immediate decline of 1.4% in labour-enhancing technical progress in the first 2 years relative to the baseline. The decline in productivity in our simulation relative to the baseline gradually reaches 4% over 15 years.

3.4. Uncertainty channel

We also consider a permanent reduction in the willingness to invest among UK firms resulting from a fundamental rise in uncertainty. As we have discussed, the short-term impact of Brexit on investment results from increasing uncertainty following the referendum. These effects are regarded as having already materialised. However, as depicted by Figure 6, despite the waning of most Brexit uncertainty over time, some Brexit uncertainty remains even 3 years after the introduction of the TCA. This indicates a fundamental shift in the business environment which has reduced the UK’s attractiveness as an investment destination, lowered inward investment and changed business expectations, affecting investment decisions. The combination of these factors has led to an increase in the cost of capital and reduced investment demand, resulting in a significant investment gap in the UK economy. In order to model this additional uncertainty shock, we introduce a shock that increases the investment premium permanently by 110 basis points. The rise in investment premium increases the user cost of capital as in Equation (5), which corresponds to a decrease in the equilibrium capital stock in Equation (6) and so, a decrease in business investment in Equation (7):

(5) $$ \ln \left(\frac{c}{p}\right)=\alpha +\beta LRR+\sigma IPREM $$

(6) $$ \ln \left(\frac{\mathrm{K}}{Y}\right)=v+\theta \ln \left(\frac{c}{p}\right) $$

(7) $$ IB=K-\left(1-\delta \right){K}_{-1}- IG $$

where ( $ \frac{c}{p} $ ) is user cost of capital, LRR is real long-term interest rate, IPREM is the investment risk premium, IB is business investment and IG is public sector investment.

Figure 6. Brexit uncertainty index (BUI).

Source: Chung et al. (2023).

3.5. Other channels

Another channel that has been considered in the previous work is migration. Hantzsche et al. (2018a) assumed that net migration in the United Kingdom would fall by up to 100,000 as a result of leaving the EU. However, recent international migration data show that net migration has increased since the United Kingdom left the EU due to the government’s post-Brexit migration regime. Figure 7 shows that the composition of migration into the United Kingdom has changed, as lower EU migration has been offset by the increase in net migration from non-EU countries since Brexit. It appears that the post-Brexit migration system has effectively maintained stable net migration flows, particularly thanks to the substantial rise in the number of skilled worker visas granted to non-EU nationals (Portes, 2022). Therefore, we do not account for any possible fall in migration in our modelling. Similarly, we have omitted the reduction in the UK net fiscal contribution to the EU. This decision is based on Hantzsche (2019), which suggests that this shock does not have a substantial impact on GDP. Table 1 summarises our assumptions.

Figure 7. Stock of foreign nationals in the United Kingdom.

Source: ONS.

Table 1. Brexit macroeconomics channels

4. Scenario results

Figure 8 summarises the impact of the combination of our shocks on business investment under different assumptions about household and business expectations. As can be seen, the impact of these shocks due to Brexit on business investment varies between 9% and 15% by 2023 based on different assumptions. These estimates are in line with studies by OECD (2016), Bloom et al. (2019) and Springford (2022), which find a decline in business investment of more than 10% using different methodologies. With the assumption of forward-looking expectations, businesses immediately began cutting investment when the referendum results were known in 2016 by around 3%, relative to the alternative scenario in which the United Kingdom remained in the EU. After Brexit starts to have a direct effect, the response of business investment reaches around 12.4% in 2023 and then gradually declines to 7.5% by 2035. As shown in Figure 9, the long-term decline in business investment is mainly driven by the drop in productivity, introduced as a reduction in labour-augmenting technical change in NiGEM, leading to approximately a 5% decline. The loss in investment attractiveness due to elevated uncertainty reduces long-term business investment by 1.7%, while the initial strong negative impact of the terms of trade shock diminishes over time, directly reducing business investment in 2035 by 1%.

Figure 8. Response of business investment.

Source: NIESR forecasts and calculations.

Figure 9. Impact of macroeconomic channels on business investment in 2035.

Source: NIESR forecasts and calculations.

Figure 10 illustrates the trajectory of real GDP published in Bejanaro Carbo et al. (2023) against a counterfactual scenario where the United Kingdom retained its EU membership. Our analysis suggests that 3 years after the end of the transition period UK real GDP is now 2–3% lower than it would have been had the United Kingdom retained its EU membership. This corresponds to a reduction of around £850 per capita as of 2023. The negative effect of Brexit on output shows a gradual escalation, reaching 5–6% or around £2,300 per head by 2035. This protracted decline is mainly driven by the fall in real income resulting from both the reduction in the terms of trade associated with the declining trade with the EU and the fall in productivity, both of which manifest themselves over a longer period. In fact, the fall in the terms of trade associated with the reduction in trade with the EU and productivity both contribute more than 2.5 percentage points to the estimated reduction in real GDP. The initial strong negative impact of lower investment as people are less willing to invest in the United Kingdom is reduced over time, and this shock only directly reduces GDP in 2035 by 0.2%, although investment does respond endogenously to the trade and productivity shocks.

Figure 10. Real GDP impact of Brexit.

Source: NIESR forecasts and calculations.

Figure 11 shows the evolution of labour productivity under current conditions relative to the counterfactual without Brexit shock. ¹ According to our estimates, Brexit has reduced labour productivity by 2.3% as of 2023. This negative impact gradually increases to 5.5% by 2035. The decline in labour-augmenting technical change contributes most to the overall drop in labour productivity (2.8 percentage points), followed by the trade channel (2.5 percentage points). The impact of Brexit uncertainty on labour productivity is rather limited at only 0.2 percentage points.

Figure 11. Labour productivity impact of Brexit.

Source: NIESR forecasts and calculations.

Table 2 summarises the impact of Brexit on key macroeconomic aggregates. As discussed above, the impact is most pronounced on business investment, where a significant decline of 12.4% from the baseline is observed by 2023. Labour productivity, measured as output per hour worked, experiences a significant decline of 2.4% by 2023. Given the reduction in the terms of trade associated with the increase in the costs of trading for both importers and exporters, real income and consumption levels are also affected, with declines of 3.5% and 5.8%, respectively, in 2023 compared to the baseline forecast. Looking ahead to 2035, business investment continues to be hit hard, with a 7.6% decline, while labour productivity remains under pressure, 5.5% lower than in the baseline scenario. Our estimates show that, given the permanent effect on the UK terms of trade, income and consumption levels in 2035 will be 5.2% and 8.2% lower, respectively, than they would have been under the baseline scenario.

Table 2. Brexit impact on macroeconomic variables (per cent difference from base)

Overall, our estimates suggest that the negative impact of Brexit on GDP is higher by 2030, at 4.4%, than the pessimistic scenario of 3.2% in Ebell and Warren (2016). Moreover, this negative impact continues to increase to 5.7% by 2035 in our estimates. This is mainly due to their neglect of the productivity channel. In fact, when they include a negative productivity shock in their robustness analysis, they obtain a GDP decline of 7.8%. This is in line with other research, such as OECD (2016) and HM Treasury (2016b), both of which use NiGEM in their analyses. However, all these studies assumed that Brexit would lead to a significant decline in bilateral trade with the EU, partly because these studies did not consider that the bilateral trade agreement would partially offset the trade decline due to loss of the single market.

Our estimates of the GDP impact are also greater than those presented in the free trade agreement scenario by Hantzsche et al. (2018a), which found that Brexit lowered GDP by 3.9% in 2030, a figure more closely aligned with our own estimates. Although they consider a free trade deal, they still assume a significant decline in trade (around 46% in 15 years), for which we do not find evidence in recent UK-EU bilateral trade statistics. However, their assumption of a 1.3% decline in productivity is smaller than our assumption of 4% based on OBR estimates. We believe that the OBR (2020) finding that 1.4% of the productivity decline has already been observed justifies our higher productivity decline assumption. Finally, they assume a significant fall in net migration inflows, which we have not observed in international migrant flows data since the TCA came into force.

5. Conclusion

In conclusion, this study aimed to quantify the impact of Brexit on business investment and labour productivity in the United Kingdom, using NiGEM to model the impact through various channels based on the literature and our own observations. Our analysis assumes that Brexit affects the UK economy through three main channels: a trade decline with the EU leading to a reduction in the UK terms of trade, a decrease in productivity (i.e., labour-augmenting technical progress) and a permanent reduction in the willingness to invest due to a fundamental rise in uncertainty in the United Kingdom.

The results indicate that forward-looking businesses immediately reduced investment by around 3% after the 2016 referendum. As the direct effects of Brexit began to take effect, the reduction in business investment reached approximately 12.4% in 2023. According to our estimates, this corresponds to a 2.5% reduction in UK GDP (or around £850 per capita) relative to the alternative scenario where the United Kingdom remains part of the EU, despite no significant trade diversion and no notable decline in net migration flows, both of which are integral aspects of the single market. The long-term impact on business investment is 7.5%, reflecting businesses’ adjustments to the decline in terms of trade and productivity shocks, as well as higher uncertainty. However, the negative effect of Brexit on output shows a gradual escalation, reaching 5.7% or around £2,300 per head by 2035. Finally, our findings suggest that Brexit will reduce labour productivity by around 5.5% by 2035.

It is important to note that the exercise presented in this article serves as an illustration of how NiGEM can be used to explore Brexit under specific assumptions about the magnitudes of its effects on exogenous variables within NiGEM and the channels through which these effects would play out. In a future version of this article, we aim to develop a three-country DSGE model of international trade based on Jaef and Lopez (2014), where the three countries correspond to the United Kingdom, the EU and the Rest of the World. Our aim is to calibrate the fixed and variable costs of exporting for each of the three bilateral trading relationships to reflect better the direct effects of Brexit on the costs of trade. We can then analyse the impact of Brexit on business investment and productivity without having to rely on exogenous assumptions about the reductions in trade and labour-augmenting technology resulting from Brexit.