Market structure and inefficiency in the foreign exchange market

Abstract

Black, two anonymous referees, and participants at an NBER conference on exchange rate regimes for many helpful comments. All remaining errors pertain to the author. The views expressed here are those of the author and do not necessarily represent the view of the Federal Reserve System or the Federal Reserve Bank of St. Louis.

Segmentation and Time-of-Day Patterns in Foreign Exchange Markets

“…X t−1 is a 6x6 matrix containing the previous six 4-hour consecutive returns, as follows: 13 Likewise, Breedon and Vitale (2005) find that the relationship between exchange rates and forex order flow is mostly due to liquidity effects rather than any information contained in order flow. Further empirical evidence on the relevance of inventory effects in currency markets may be found in Bessembinder (1994), Cao, Evans and Lyons (2006), Flood (1994), Lyons (1995Lyons ( , 1998). …”

“…We have tested other definitions of European and UK holidays and the main results remain essentially unchanged. 20 The lowest number of non-overlapping days is 36 (euro-dollar) and the highest is 100 (yen-dollar).…”

“…Therefore we decided to consider only national holidays. 20 We analysed several combinations by including or excluding from German holidays official bank holidays in Belgium, France, Holland, Italy and Spain. For UK, we considered official holidays in England and Wales.…”

Segmentation and time-of-day patterns in foreign exchange markets

“…Also for the yen/dollar currency pair, price changes during US working hours (12:00-20:00 GMT) tend 15 In his experimental design, Flood (1994) highlights the fact that, through the "hot-potato" mechanism, inventory imbalances can give rise to foreign market inefficiency. BIS (2005) shows that the "hot-patato" trade accounts for more than 50% of daily volume in 2004.…”

How Does Risk Flow in the Credit Default Swap Market?

“…Second, an important result of our network analysis is that the trades between the strongly connected dealers represent the majority (about 60% − 90%) of traded CDS notional. Although this stylised fact was first documented by Flood (1994) and Lyons (1995Lyons ( , 1997 in the FX derivative market and identified as a result of so-called "hot potato" (Burnham, 1991) trades between dealers, it was not analysed in a network setting. A third general empirical finding is that the underlying credit risk flows from a high number of URSs to a small number of URBs, where ultimate risk concentrates in a few leading counterparties.…”