Skip to main content

Advertisement

SpringerLink
Log in

Quo Vadis, Britain? – Implications of the Brexit process on the UK’s real economy

  • Original Paper
  • Published:
International Economics and Economic Policy Aims and scope Submit manuscript

Summary

Using the Panel Data Approach (PDA) of Hsiao et al. Journal of Applied Econometrics, 7(5), 705-740 (2012) in combination with the LASSO method, this article aims to measure the effect of the Brexit process on the United Kingdom’s real economy up to 2019Q2. The results are twofold: Firstly, compared to the existing literature, the PDA improves the measurement of the impact of Brexit on the real economy regarding computation intensity, the feasibility of statistical inference and a wider application area. Secondly, the estimated counterfactuals for the UK show that the Brexit process has played a crucial role in the UK’s economy, leading to lower GDP (growth rates), lower private consumption, lower gross fixed capital formation (GFCF) and higher exports. On average, GDP growth has declined between 1.3 and 1.4 percentage points, whereby the cumulative loss ranges between 48 and 54 billion British pounds. Moreover, private consumption in the UK has declined 4.7 billion British pounds quarterly on average. The predicted counterfactuals show that the impact of the Brexit process on GFCF has begun in 2018Q1, whereby the average treatment effect amounts to −2.9 billion British pounds. The UK’s exports increased since the referendum, most likely due to the depreciation of the British pound post-Brexit. The average quarterly effect of the Brexit process on exports is estimated here at 4.8 billion British pounds.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12

Similar content being viewed by others

Data availability

No

Notes

  1. In case of the SCM, the probability distribution of the predicted pre-treatment outcome is not easily derivable, so that statistical tests cannot be performed (Hsiao et al. 2012, p. 711).

  2. This behaviour of the model selection methods explains the smaller predictive mean squared errors of the LASSO method, since a large number of regressors increases the variance of the estimation leading to poorer predictive accuracy (Li and Bell 2017, p. 69).

  3. The export ranking of the UK is calculated using the World Integrated Trade Solution (WITS) database of the Worldbank. The exclusion of the member countries of the European Single Market and the consideration of the countries with high foreign trade activity with the British economy serves to prevent endogeneity problems and to estimate the post-treatment period using countries with strong affiliation to the UK economy in order to better reflect relevant shocks stemming from these countries. Nevertheless, as an anonymous referee has pointed out, it is important to be aware that an affiliated country could theoretically be affected from policy changes emanating from the UK. However, the exclusion of any possible endogeneity problems is practically impossible and lies in the “nature” of impact evaluation methods, which is also well-known in the literature (Wan et al. 2018, p.123): “For PDA and SCM to yield reasonable estimates of counterfactuals, the control units must not be affected by the intervention. It could be hard to find a control group that is invariant to such disruptions. For instance, it is not that easy to find control groups to measure the impact of the Iranian revolution on the Iranian economy”.

  4. For the empirical study, the „lasso“function of MATLAB R2013b is used. The calibration set ΛL = {λ1, …, λL} comprises a geometric sequence with 100 λ-variations. The largest number λL is set to result the first non-null model, where all coefficients are shrunk to zero.

  5. Desciptive statistics and unit root test result can be found in the Appendix in Tables 13 and 14.

  6. In this article, all figures given in British pound sterling are in terms of 2016.

  7. Cointegration relationships have apparently not been considered in previous empirical applications of the PDA in use of non-stationary variables (see e.g. Ke et al. 2017). In the opinion of the author, this could lead to spurious post-treatment projections. The present modeling strategy tries to avoid such problems and is therefore a more conservative and precautious implementation of the PDA. As a result, the LASSO is used to enhance the prediction accuracy, whereas the cointegration is checked for non-stationary variables to provide predictors with common stochastic trend.

  8. For all HAC estimations the Bartlett kernel density and the lag selection parameter of Andrews and Monohan (1992) are used.

  9. Regarding the post-Brexit referendum period, the standard deviation of the GDP growth of Turkey and Brazil are 4.6 and 1.5 times higher, respectively, than the mean of the standard deviation of the growth rates of the donor pool. Turkey’s economy suffered from US sanctions and tariffs in 2018 and also from the offensive into north-eastern Syria in 2019. The recent country-specific political, legal and economic turmoil in Turkey are discussed in Grübler (2017, pp. 11–12). Between 2014 and 2017, Brazil’s economy slumped into a recession due to a political crisis, high fiscal deficits and a collapse in commodity and oil prices.

    Since remarkable country-specific developments should be excluded as far as possible from the counterfactual prediction, results after these exclusions have been considered more reliable in the subsequent section. It would be not quite surprising, that these exclusions from the donor pool lead to differing results, in particular when crucial country-specific economic developments have taken place in the post-Brexit period. Also for this reason numerical results should be viewed with a certain degree of caution.

  10. As the BIC is known to be very strict in selecting the number of variables to be included, it might also be possible to overcome the problem of the autocorrelation of residuals by including further time lags instead of correcting the variance-covariance matrix. However, due to the closeness of the Brexit vote, the limited number of observations for the post-treatment period may make it difficult to estimate the autoregressive model.

  11. The following Tables of results (Tables 2-12) can be found in the Appendix.

  12. The delayed impact of the Brexit process on investment could be explained by the changed expectations for soft and hard Brexit after the EU-UK negotiations started in the second half of 2017. Another reason could be the reorganisation of production, particularly in supply chains, where planning and implementation may show some delay.

  13. Since in these settings only six observations for the treatment effects are present, the use of an AR-model is limited. Hence, only the ATE using Newey-West standard errors is calculated in order to deal with the serial correlation.

  14. Monthly exchange rate data are extracted from the Monthly Monetary and Financial Statistics of the OECD database and recalculated to compile quarterly data by taking the mean of three months.

  15. The impact of Brexit on British pound exchange rates has been investigated by Korus and Celebi (2019). They find that particularly the Brexit vote and “bad”/“hard” Brexit news have led to a depreciation of the British pound exchange rate against both the US dollar and the euro.

  16. See, for example, the Office for National Statistics (2019), section 7 (“Revision to GDP”).

  17. Note that Springford (2019) also uses European Single Market member countries in the donor pool, which could bias the estimated figures as in Born et al. (2019).

References

  • Abadie A, Gardeazabal J (2003) The economic costs of conflict: a case study of the Basque Country. Am Econ Rev 93(1):113–132

    Article  Google Scholar 

  • Andrews DWK, Monohan JC (1992) An improved Heteroskedasticity and autocorrelation consistent covariance matrix estimator. Econometrica 60:953–966

    Article  Google Scholar 

  • Bai J, Ng S (2002) Determining the number of factors in approximate factor models. Econometrica 70(1):191–221

    Article  Google Scholar 

  • Baier FJ, Welfens PJJ (2018) BREXIT and Foreign Direct Investment: Key Issues and New Empirical Findings. Int J Financ Stud 6(46)

  • Barro R (2001) Human capital and growth. Am Econ Rev 91(2):12–17

    Article  Google Scholar 

  • Born B, Müller GJ, Schularick M, Sedláček P (2019) The costs of economic nationalism: evidence from the Brexit experiment. Econ J 129(10):2722–2744

    Article  Google Scholar 

  • Brakman S, Garretsen H, Kohl T (2017) Consequences of Brexit and options for a "global Britain". CESifo Working Paper Series 6448, CESifo

  • Demmou L, Stefanescu I, Arquié A (2019) Productivity growth and finance: the role of intangible assets - a sector level analysis. OECD Economics Department Working Papers, No. 1547, Paris

    Google Scholar 

  • Devereux M, Love D (1994) The effects of factor taxation in a two-sector model of endogenous growth. Can J Econ 27(3):509–536

    Article  Google Scholar 

  • Dhingra S, Huang H, Ottaviano G, Pessoa JP, Sampson T, Van Reenen J (2017) The costs and benefits of leaving the EU: trade effects. CEP Discussion Paper 1478, Centre for Economic Performance, LSE

  • Douch M, Edwards H, Soegaard C (2018) Estimating the trade effects of the Brexit announcement shock. The UK in a Changing Europe Available at: http://ukandeu.ac.uk/estimating-the-trade-effects-of-the-brexit-announcement-shock/. Accessed 05 Januar 2020

  • Engle RF, Granger CWJ (1987) Co-integration and error correction: representation, estimation, and testing. Econometrica 55(2):251–276

    Article  Google Scholar 

  • Froot KA, Stein JC (1991) Exchange rates and foreign direct investment: an imperfect capital markets approach. Q J Econ 106(4):1191–1217

    Article  Google Scholar 

  • Gardeazabal J, Vega-Bayo A (2017) An Empirical Comparison Between the Synthetic Control Method and HSIAO et al.'s Panel Data Approach to Program Evaluation. J Appl Econ 32(5):983–1002

    Article  Google Scholar 

  • Graziano A, Handley K, Limão N (2018) Brexit uncertainty and trade disintegration. NBER working paper no. 25334, National Bureau of Economic Research

  • Grübler J (2017) Wirtschaftliche Konvergenz trotz politischer Unsicherheit. The Vienna Institute for International Economic Studies, Forschungsbericht 7

  • Henkel M, Seidel T (2019) A spatial perspective on EUROPEAN integration: heterogeneous welfare and migration effects from the single market and the BREXIT. Econ Inq 57(1):333–352

    Article  Google Scholar 

  • Hsiao C, Ching HS, Wan SK (2012) A panel data approach for program evaluation: MEASURING the benefits of political and economic integration of Hong KONG with mainland China. J Appl Econ 27(5):705–740

    Article  Google Scholar 

  • IMF (2015) Fiscal policy and long-term growth. IMF Policy Paper, USA: International Monetary Fund

  • Jungmittag A, Welfens PJJ (2016) Beyond EU-US trade dynamics: TTIP effects related to foreign direct investment and innovation. EIIW Discussion Paper No. 212, European Institute for International Economic Relations, University of Wuppertal

  • Ke X, Chen H, Hong Y, Hsiao C (2017) Do China's high-speed-rail projects promote local economy?—new evidence from a panel data approach. China Econ Rev 44:203–226

    Article  Google Scholar 

  • Korus A, Celebi K (2019) The impact of Brexit news on British pound exchange rates. Int Econ Econ Policy 16:161–192

    Article  Google Scholar 

  • Li KT, Bell DR (2017) Estimation of average treatment effects with panel data: asymptotic theory and implementation. J Econ 197(1):65–75

    Article  Google Scholar 

  • Lucas R (1988) On the mechanics of economic development. J Monet Econ 22:3–42

    Article  Google Scholar 

  • Meinshausen N, Yu B (2009) Lasso-type recovery of sparse representations for high-dimensional data. Ann Stat 37(1):246–270

    Article  Google Scholar 

  • Newey WK, West KD (1987) A simple, positive semi-definite, Heteroskedasticity and autocorrelation consistent covariance matrix. Econometrica 55(3):703–708

    Article  Google Scholar 

  • Oberhofer H, Pfaffermayr M (2018) Estimating the trade and welfare effects of Brexit: a panel data structural gravity model. Department of Economics Working Paper Series, 259. WU Vienna University of Economics and Business

  • Office for Natinal Statistics (2019) GDP quarterly national accounts, UK: April to June 2019. [Online] Available at: https://www.ons.gov.uk/economy/grossdomesticproductgdp/bulletins/quarterlynationalaccounts/apriltojune2019. Accessed 23 December 2019

  • Rebelo S (1991) Long-run policy analysis and long-run growth. J Polit Econ 99(3):500–521

    Article  Google Scholar 

  • Serwicka I, Tamberi N (2018) Not backing Britain: FDI inflows since the Brexit referendum. Briefing Paper 23, UK Trade Policy Observatory, University of Sussex

  • Springford J (2019) CER INSIGHT: The cost of Brexit to June 2019. [online] Available at: https://www.cer.eu/sites/default/files/insight_JS_16.10.19.pdf. Accessed 23 December 2019

  • Tibshirani R (2011) Regression shrinkage and selection via the lasso: a retrospective. J R Stat Soc Ser B Methodol 73(3):273–282

    Article  Google Scholar 

  • Wan S, Xie Y, Hsiao C (2018) Panel data approach vs synthetic control method. Econ Lett 164:121–123

    Article  Google Scholar 

Download references

Acknowledgements

I am grateful for the helpful comments, suggestions and criticism of Andre Jungmittag (Frankfurt University of Applied Sciences and DG Joint Research Centre of the European Commission), Paul J.J. Welfens (EIIW/University of Wuppertal), Ronald Schettkat (University of Wuppertal), Tian Xiong (EIIW/University of Wuppertal), participants of the 22nd Annual INFER Conference, participants of the 13th FIW Research Conference in International Economics and seminar participants at the Chair for Macroeconomic Theory and Policy and at the Chair for Economic Policy, University of Wuppertal. I would like to express my sincere thanks to David Hanrahan (EIIW) and Kennet Stave (EIIW) for editorial assistance and also gratefully acknowledge comments and suggestions by two anonymous referees that have significantly helped to improve this paper. Remaining errors are my own.

Author information

Authors and Affiliations

  1. European Institute for International Economic Relations at the University of Wuppertal, Rainer-Gruenter-Str. 21, D-42119, Wuppertal, Germany

    Kaan Celebi

Authors
  1. Kaan Celebi
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Kaan Celebi.

Ethics declarations

Conflict of interest

Not applicable.

Code availability

No

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix

Appendix

Table 2 Counterfactual prediction of the UK’s real GDP growth rates in percent using all donors (first estimation)
Full size table
Table 3 Counterfactual prediction of the UK’s real GDP growth rates in percent after removing Turkey from the donor pool (second estimation)
Full size table
Table 4 Counterfactual prediction of the UK’s real GDP growth rates in percent after removing Turkey, Brazil, India and Mexico from the donor pool (third estimation)
Full size table
Table 5 Counterfactual prediction of the UK’s GDP (in billion British pounds, CVM) (first estimate)
Full size table
Table 6 Counterfactual prediction of the UK’s GDP (in billion British pounds, CVM) after removing Japan from the donor pool (second estimate)
Full size table
Table 7 Counterfactual prediction of the UK’s private consumption (in billion British pounds, CVM)
Full size table
Table 8 Counterfactual prediction of the UK’s GFCF (in billion British pounds, CVM) (first estimate)
Full size table
Table 9 Counterfactual prediction of the UK’s GFCF (in billion British pounds, CVM) after removing Brazil from the donor pool (second estimate)
Full size table
Table 10 Counterfactual prediction of the UK’s GFCF (in billion British pounds, CVM) after removing Brazil from the donor pool and with cut-off point 2018Q1 (third estimate)
Full size table

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Celebi, K. Quo Vadis, Britain? – Implications of the Brexit process on the UK’s real economy. Int Econ Econ Policy 18, 267–307 (2021). https://doi.org/10.1007/s10368-021-00493-7

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10368-021-00493-7

Keywords

JEL classification

Search

Navigation