Currency composition of international portfolios - did the entrance of the Euro make a difference?

1 Introduction The introduction of a unied currency for European countries occurred at a time of ever increasing nancial globalization and raising cross-border holdings in foreign currencies (Papaioannou and Portes 2008). As forecasted by numerous antecedent studies (e.g. Wyplosz 1997, Portes and Rey 1998), the Euro rapidly reached the status of international currency: it did not overcome the US dollar as leading world currency (Goldberg 2010), but quite clearly become the second most used one (ECB 2010). Hence for those countries that do not (often because they cannot) make use of their domestic currency for international transactions, the choice of foreign currencies was altered by the creation of a liquid market for a single European currency. In particular, the choice among foreign currencies was most aected by the Euro in countries with strong linkages with Europe - trade, cultural and political ties due to the marked regional character of this currency (ECB 2010, pp.7). For some of these countries, there could even be the expectation to join the EU and its currency, a further incentive to adopt the Euro for most international transactions. However, despite the great initial interest in the process of internationalization of the Euro, not many studies exist on the eect that its introduction has had on currency composition of countries’ portfolios. The importance of currency exposure is underlined by the literature on valuation eects, which demonstrates that the impact of exchange rate movements on the value of nancial assets and liabilities can be quite sizeable (see Tille 2003 for the USA). Furthermore, theoretical models of optimal portfolio choice analyse in detail the determinants of such exposure and foreign currency usage, but empirical research on the other hand has lagged behind signicantly, mostly due to lack of disaggregated data on the currency breakdown of the various assets classes. Hence the goal of this thesis is to contribute to the empirical analysis of international portfolios, incorporating the key event of the introduction of a new international currency, the Euro, and expanding the seminal work of Lane and Shambaugh (2009, 2010) on the determi- nants of foreign currency choice. My approach is twofold. First, I examine the determinants of currency exposure and usage considering 5 major currencies (US Dollar, Euro, Yen, Pound Sterling and Swiss Franc), focusing particularly on the transition period from pre-Euro to post-Euro era (1994-2004). In doing this, I follow the work of Lane and Shambaugh (2010), who study the determinants of currency exposure, but I depart from their analysis by con- sidering bilateral relations and individual currency breakdown - for instance to see if bilateral trade ows with the European Union are related specically to exposure or usage of Euros. I consider both net positions (assets minus liabilities) and gross positions (assets plus liabilities) to capture directional exposure on the one hand and total usage of a currency on the other. 3