Is there an International Trade-Venue Clientele Effect for Canadian Shares Cross-Listed in the United States?

Abstract:     
This paper extends the work by Amihud and Mendelson (1986) and Atkins and Dyl (1997a) that in equilibrium assets with higher spreads are held in portfolios with the same or expected longer holding periods. The paper examines whether or not investors have the same holding periods for the Canadian cross-listed share trades executed in the U.S. trade venues relative to the same-firm share trades executed in the TSE, and whether or not any differences are associated with the size of their relative effective half-spreads. The results support the trade-venue clientele hypothesis for the Canadian cross-listed shares whose trades are executed in the TSE and the AMEX, the TSE and NASDAQ, and the TSE and NYSE, primarily for the period after the TSE decimalization on April 15, 1996. Holding periods (effective half-spreads) for share trades executed in the U.S. trade venues significantly decreased over time relative to investor holding periods (effective spreads) for the same-firm share trades executed in the TSE. This suggests that the TSE has consistently lost its share of executed order flow (share turnover) relative to the U.S. primary trade venues, and that this loss is associated with increased relative trade costs in the TSE.

Working Paper Series