Tariffs, domestic import substitution and trade diversion in input-output production networks: an exercise on Brexit

ABSTRACT

In this paper we employ the World Input–Output Database to develop a multi-sector inter-country model that allows us to identify the channels through which the trade effects of Brexit would propagate. The inclusion of global value chains and indirect Brexit effects in the model leads to estimates that diverge with the results of the main literature. Indeed, we found that Brexit could be risky and costly not only for the UK but also for many EU countries. Furthermore, we develop a second model and present the first empirical analysis on the consequences of domestic import substitution and trade diversion policies in Input–Output schemes. We found that allowing sectors and countries to partly substitute foreign products, leads to significantly lower losses for both macro-regions: the UK and EU27 would lose, at worst, the 0.28 and 0.5 percent of value-added, respectively.

KEYWORDS:

• Brexit
• trade barriers
• tariffs
• input–output analysis
• value chains
• import substitution
• production networks

JEL CLASSIFICATIONS:

• C67
• R15
• F13
• F14
• O21

Acknowledgements

Part of this research has been performed while I was visiting the Department of Economics at the University of Essex, therefore, I thank Sheri Marina Markose who provided insight and expertise that greatly assisted the research. I also thank Katie Chapman and Inacio Manjama for assistance and comments that improved the manuscript. I also to show my gratitude to the organisers and participants of the University of Trento’s XIX Trento Summer School in Adaptive Economic Dynamics, for useful comments and suggestions, as well as participants at the 2018 ESCoS conference in Naples, and the 2018 DySES conference in Paris. Finally, I am immensely grateful to Erik Dietzenbacher, Nadia Garbellini, Bart Los, Neville R. Norman, Anthony Philip Thirlwall, and Ariel Wirkierman for their insightful comments and suggestions at the very early stage of this research, although any errors are my own and should not tarnish the reputations of these esteemed persons.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Matrices are indicated by bold capitals, vectors by bold lowercases and scalars by italic lowercases. Diagonal matrices are indicated by a hat over the vector containing the elements on the main diagonal. Primes indicate transposition.

2 In their model 3, Alatriste-Contreras and Fagiolo (2014) present a similar approach to explain the propagation of economic shocks in IO networks.

3 In section 4.2 we compare our approach to that by W. Chen et al. (2018). Differently from our exercise, W. Chen et al. (2018) do not try to predict the impact of Brexit, rather they merely measure the UK value added content of trade with EU countries. In this vein, it should be stressed that the criticism made by Dietzenbacher et al. (2019) does not apply to W. Chen et al. (2018) and that comparisons between numbers in W. Chen et al. (2018) and those herein are inappropriate.

4 For more insights and comparisons between IO, CGE and econometric models, see Rose (1995, 2004) and West (1995) among others.