| The first essay documents strong comovement in currency spreads at both the intraday and the daily frequency. Controlling for inventory effects, this comovement remains strong. We also show that currency spreads commove with aggregate equity market spreads. Our results are consistent with the existence of market-level private information within the foreign exchange market as well as common to the foreign exchange and equity markets. They also cast doubt on the effectiveness of international diversification strategies as a means of reducing liquidity risk.The third essay documents that the incremental power of the U.S. market movements in explaining domestic returns (or the U.S. information factor) increases after a firm cross-lists. The change in domestic ownership and most specifically, the increase in the mass of informed traders, is the most significant determinant for the change in the value relevance of the U.S. market. Controlling for firm characteristics (such as size, book to market), relative liquidity between the U.S. and Canada, and issue specifics (such as NYSE/AMEX listings vs. NASDAQ listings) does not change the result. Baruch, Karolyi, Lemmon (2007) suggest that a higher fraction of trading volume arises in the foreign market if comparatively more information is generated there. We document the dynamics of the U.S. information factor in domestic prices around cross-listings and relate the change in this measure to the change in the composition of market participants in the cross-listed stock.The second essay investigates whether stock exchanges induce herding by examining a sample of firms that switch from NASDAQ to the NYSE. We find that trades for the switching firms co-move more strongly with NYSE trades and less strongly with NASDAQ trades following the switch, indicating that investors on the two major U.S. stock exchanges display herding behavior. The results are not driven by changes in the comovement of cash flows or by firm characteristics. A similar pattern is found for stock returns. While we are not able to rule out rational origins of herding, our broader results appear to be most consistent with a behavioral view of comovement proposed by Barberis, Shleifer and Wurgler (2005). |
