Abstract
This thesis consists of five separate papers, broadly within the field of International Finance. The first paper, An Empirical Analysis of the Currency Denomination in International Trade, investigates the choice of currency in international trade transactions by Swedish exporting firms. It uses an extensive dataset on payment transactions between foreign importers and Swedish exporting firms. It is the first paper to examine currency invoicing at such a disaggregated level. The main findings are that high exchange rate volatility reduces the likelihood of using the importers currency while high GDP and GDP per capita in the importing country increases the likelihood. A large market share of a third country increases the likelihood of using the third country's currency. A further finding is a decreased use of Swedish krona and a rise in the use of the euro as a vehicle currency. State Dependent Pricing, Invoicing Currency and Exchange Rate Pass-Through, written jointly with Martin Flodén, analyzes exchange rate pass-through in a dynamic model with menu costs. In the paper, we provide a link between the fixed and flexible price analyses by specifying a dynamic framework with exogenous choice of exporting currency, but with endogenous pricing decisions. We consider the pricing strategies of firms that produce in a home country, sell on a foreign market, and can change the price in response to exchange rate fluctuations, while being subject to menu costs. Our main finding is that when the exporter prefers to set price in the importer’s currency, the exporter also changes prices less frequently than if price was set in the exporter’s home currency. The intuition is that in this setting, the optimal currency choice is the one that on average minimizes the difference between fixed and flexible price profits, and thereby the frequency of price updates. When the importer’s currency is preferred it leads to limited pass-through and a low correlation between exchange rate movements and import prices. The third paper, Demand and Distance: Evidence of Cross Border Shopping , written jointly with Marcus Asplund and Richard Friberg, uses data from 287 Swedish municipalities to estimate how responsive alcohol sales are to foreign prices, and relate the sensitivity to the location's distance to the border. Typical results suggest that the elasticity with respect to the foreign price is around 0.4 in the border region; moving 200 (400) kilometers inland reduces it to 0.2 (0.1). For example, a 10 percent reduction in the Danish price of spirits causes a fall in per capita sales of roughly 4 percent at the border (Malmö). This large cross price elasticity is almost half the own price elasticity. The effect diminishes gradually as one moves further from the border, but fall in sales is estimated to drop below 1 percent only at 460 kilometer from the border. Not until we reach 1000 kilometers can we reject that the effect is zero. Common Currencies and Equity Prices: Evidence from a Political Event, uses a political event, the Swedish referendum on whether or not to join the European Monetary Union (EMU), as a natural experiment to examine the relationship between common currencies and the market value of exporting firms. If Sweden would have voted to join the EMU, exchange rate uncertainty as well as transaction costs would have been greatly reduced for many exporting companies. Prior to the referendum, these potential gains (adjusted for the probability of joining) should have been included in equity prices. The day after the referendum that probability of was zero and one would expect a decline in equity prices of exporting firms. We find evidence of statistically significant negative abnormal returns on the trading day after the election for only two out of fifteen examined industry indices. The small effects found in this study are in line with earlier research that finds a weak relationship between exchange rates and equity prices. The fifth paper, When is a Lower Exchange Rate Pass-Through Associated with Greater Exchange Rate Exposure?, written jointly with Martin Flodén and Witness Simbanegavi, we study the relationship between exchange rate pass-through and exchange rate exposure (the relation between profits and exchange rates) under flexible prices. We introduce a convex cost function and study the effects of changing the elasticity of costs with respect to output. We do this both in a model of monopolistic competition as well as in the oligopoly models used by Bodnar et al (2002). We find that increasing the convexity of costs reduces both exchange rate pass-through and exposure, both under monopolistic competition and in duopoly settings. The conclusion is thus that if industries differ mainly on the supply side, this would imply a positive correlation between pass-through and exposure. However, our extension does not affect the result in Bodnar et al. that exchange rate pass-through and exposure should be negatively correlated across industries if industries differ mainly on the demand side, more specifically in the substitutability between domestically produced and imported goods.