Jean-François Gajewski

Centralised order books versus hybrid order books: A paired comparison of trading costs on NSC (Euronext Paris) and SETS (London Stock Exchange)

This article compares the cost of trading large capitalisation equities on the hybrid order-driven segment of the London Stock Exchange and the centralised electronic order book of Euronext. Using samples of stocks matched according to economic sector, free float capitalisation, and trading volume, our study shows that transaction costs are lower on the centralised order book than on the hybrid order book. The presence of dealers outside the electronic order book favours the frequency of large trades, but is associated with higher execution costs for all other trades and higher adverse selection and inventory costs inside the order book.

Earnings Announcements, Asymmetric Information, Trades and Quotes

This paper empirically analyses trades and quotes around the times of 37 earnings announcements in the Paris Bourse. We find that trading volume is larger on announcement days, spreads are wider after announcements, and the permanent positive (resp. negative) price impact of purchases (sales) is greater around announcements. While the findings pertaining to the spread and the permanent impact of trades are consistent with the view that earnings announcements correspond to an increase in information asymmetries, the result that trading volume is larger suggests that other effects are at work.

Trading Volume, Heterogeneous Expectations and Earnings Announcements

Our objective was to conduct an experimental study of investors’ reactions in terms of trading volume to the announcement of annual earnings. Shareholders consider annual net income as the most important figure in the announced results since it is the basis for determining profit for individual accounts at the shareholders’ general meeting. In our experiment, annual earnings are announced at the end of eight rounds of exchanges. Prior to the year end announcement, a fraction of the annual income is revealed to all of the participants at the end of every two periods. Participants thus periodically revise their expectations regarding annual results. The experiment shows that the heterogeneity of expectations does not decrease when investors have more information about the final results. This heterogeneity also is the main factor behind transactions in our experimental asset markets. However, too large a dispersion in expectations prevents investors from trading. Although price changes in absolute value influence trading volume as expected, the heterogeneity of expectations has a greater impact.