Abstract

We provide micro-econometric evidence that, following the large and persistent sterling depreciation after the Brexit referendum, on impact, exchange rate pass-through (ERPT) was complete for transactions invoiced in producer currency and low for sales invoiced either in a vehicle or in the destination market currency. Yet these differences strikingly narrowed within six quarters. A weaker currency did not translate into a persistent gain in price competitiveness for UK exports. At a granular level we find that UK exporters invoice in multiple currencies—even when shipping a product to the same destination—and switch currencies over time. Remarkably, we fail to detect significant changes in the relative shares of invoicing currencies in response to the Brexit shock. Last but not least, we find that UK firms price-to-market, i.e., adjust markups to bilateral exchange rate and CPI movements, only when they invoice sales in the destination-market currency.

Highlights

  • International economists have long noted that aggregate exchange rate pass through (ERPT) is significantly correlated with the currency in which most international trade transactions are invoiced.1 In recent years, the availability of large datasets has boosted this line of research; leading papers have identified empirical regularities that have propelled the frontier of open macro and trade theory forward (see, e.g., Gopinath, Itskhoki and Rigobon (2010); Goldberg and Tille (2016); Amiti, Itskhoki and Konings (2018); Gopinath et al (2020); Auer, Burstein and Lein (2021))

  • We provide micro-econometric evidence that, following the large and persistent sterling depreciation after the Brexit referendum, on impact, exchange rate pass-through (ERPT) was complete for transactions invoiced in producer currency and low for sales invoiced either in a vehicle or in the destination market currency

  • We use transactions data to carry out an investigation into the invoicing currency patterns of UK exporters, and a comparative analysis of the dynamics of ERPT and pricing-to-market when sales are invoiced in different currencies

Read more

Summary

Introduction

International economists have long noted that aggregate exchange rate pass through (ERPT) is significantly correlated with the currency in which most international trade transactions are invoiced. In recent years, the availability of large datasets has boosted this line of research; leading papers have identified empirical regularities that have propelled the frontier of open macro and trade theory forward (see, e.g., Gopinath, Itskhoki and Rigobon (2010); Goldberg and Tille (2016); Amiti, Itskhoki and Konings (2018); Gopinath et al (2020); Auer, Burstein and Lein (2021)). Our second fact is that around 15% (50%) of extra-EU export transactions (value) originate from firms that use more than one currency to invoice sales of the same product in the same destination in a given year. Our econometric analysis of ERPT in the wake of the Brexit depreciation documents significant differences in the dynamic responses of British export prices according to the currency in which UK firms invoiced their cross-border transactions.. Expanding the dataset to include UK exports with the US invoiced in US dollars, we find that markup adjustment becomes more pronounced, accounting for up to 70% of incomplete exchange rate pass through; this is consistent with evidence on the dollar price stability of US imports (Gopinath, Itskhoki and Rigobon (2010)).

Facts about invoicing currencies and their trade shares
The UK’s trade is dominated by firms invoicing in more than one currency
Firms switch the currency of invoicing over time
Slow-moving trends in aggregate shares of invoicing currencies
The empirical model for the event study
PCI 2010-2015 PCI 2016
Price responses to the Brexit depreciation
Invoicing and markup adjustment
Econometric models
Results excluding UK trade with the US
Results extending the analysis to UK trade with the US and the EU
Trade with the US
Trade with the EU
Conclusion
16.6 GBP USD
Convert concordance tables
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call