Arbitrage in International Markets

(ProQuest: ... denotes formulae omitted.)IntroductionConsider a model with two equity markets, home and foreign, each arbitrage free and endowed with an investor. There are no barriers to financial trade between markets, no taxes or transaction costs. In such setting possibility of arbitrage between those markets, from perspective of home investor, exists. The empirical results suggest that a simple measure, difference in Sharpe ratios between home and foreign markets as perceived by a home based investor, taken to be a U.S. investor, indicates existence of arbitrage opportunities and seems to influence portfolio investment.Arbitrage is defined as the simultaneous purchase and sale of same, or essentially similar, security in two different markets for advantageously different (Sharpe and Alexander, 1990). Arbitrageurs, by trading to profit from price differentials between similar assets in different markets, eliminate such mispricing and bring prices in line with fundamental values (Shleifer and Vishni, 1997).The idea of arbitrageurs profiting from every opportunity and eliminating all price differences is not realistic, partially due to risky nature of arbitrage operations (Shleifer and Vishni, 1997), to point that sometimes arbitrageurs leave opportunities untouched (Kondor, 2009). Along these lines, Pontiff (2006) recognizes possibility of mispricing even in presence of arbitrageurs, mostly due to costs arising from idiosyncratic risks. Gagnon and Karolyi (2010), in a study encompassing 506 cross listed stocks, recognize economically small but very volatile deviations from price parity. In agreement with Pontiff (2006), they find highest cost for arbitrageurs to be related to idiosyncratic risk. When arbitrage opportunities are left untouched, equalization of prices for similar assets across markets will likely not occur, implying that prices of these assets do not necessarily reflect fundamental values.Several of studies on arbitrage in international markets focus on differences in price, when expressed in a common currency, of cross listed stocks. Examples of such studies include: Gagnon and Karolyi (2010); Chan, Hong, and Subrahmaniam (2008); Bluin, Hail, and Yetman (2009); and De Jong, Rosenthal, and Van Dijk (2009). These studies agree on existence of arbitrage across markets for stocks in their samples, giving as reasons for persistence of small mispricings existence of costs ranging from tax differentials (Bluin et al. 2009) to idiosyncratic risk (De Jong, Rosenthal, and Van Dijk 2009). Licht (2001) discuss effects on arbitrage of laws in different countries; he posits equalization of prices provided by arbitrageurs to reflect, among other factors, differential effects of laws in countries where stock is listed.Following a different approach, Delbaen and Shirakawa (1996) use an equivalent probability measure to generate a theoretical no arbitrage condition in currency markets. It is worth noting that their result is not dependent on a subset of currencies, rather it is a general no arbitrage condition that applies to all currencies. Following their path this study focuses on markets, proxied by their indexes, and presents a general, though simple, measure of arbitrage independent of any particular stock.Since approximately 1990s, several financial markets in developing countries opened to trade with rest of world (Bekaert and Campbell 2000, Licht 2001). Possibly related with opening of these new markets, global portfolio investment increased. Data from Coordinated Portfolio Investment Survey (CPIS), conducted by International Monetary Fund (IMF), shows value of total world portfolio investment in equities to have increased from 1.6 trillion U.S. dollars in year 2001 to 5.3 trillion U.S. dollars in year 2012, a compounded growth of around 8. …