How Wide is the Euro?

AbstractThis paper examines the impact of Euro invoicing on Italian exports to non-EU countries. In addition to examining the role of currency invoicing on the intensive and extensive margin of trade, we introduce the ‘entrenched’ margin of trade. We define the entrenched margin of trade as the number of transactions between two countries of a particular good. With highly disaggregated data, we use a two-stage methodology to predict the probability of Euro dominated Italian exports and then use that predicted probability on the intensive, extensive and entrenched margin of trade. Results show that the probability of Euro dominated trade invoicing reduces all three margins of trade. Specifically, a 10% increase in probability of Euro dominated Italian exports has roughly the same impact as additional 1532 km on the intensive margin of trade, 1096 km on the extensive margin of trade and 1314 km on the entrenched margin of trade. The negative effect of Euro invoicing is most consistent with lower-middle income trading partners and more thinly traded goods. We surmise that these results are due to varying access to financial instrumentation among Italian trade partners and a trade diversion effect of Italian exports to EU countries versus non-EU markets.KEYWORDS: Trade marginscurrency invoicingEuroexportsdistanceJEL Classifications: F10F14 Disclosure statementNo potential conflict of interest was reported by the author(s).Notes1 See Ventura and Witte (Citation2016) and Arioldi et al. (Citation2022).2 It would not be impossible for a good forensic analysis to identify major Italian firms and gather a variety of sensitive trade information with this data.3 Region dummy variables are used in place of country dummies for the same reasons that 2-digit good dummies are used instead of 4-digit dummies: overidentification and multicollinearity in predicting a variable that is often zero or one. Overidentification and multicollinearity are major problems with predicting the currency denomination of trade because, as far as the invoicing currency is concerned, there isn’t much statistical difference between many countries (e.g. the Gulf Cooperation Council countries, the Central African Franc countries) or between 4-digit good designations (e.g. 3901-3909 are likely produced by the same two companies: Vinavil and Versalis). This simplification is used in other studies of the same Italian trade data.4 Notable exceptions are the low-middle income countries with either the East African franc or West African franc. These include Senegal, Cote d’Ivoire, Benin, Cameroon and Congo which are not major importers of Italian goods.