The Intraday Pattern of Information Asymmetry, Spread, and Depth: Evidence from the NYSE

Authors

  • George Tannous,

    Corresponding author
    • Department of Finance and Management Science, Edwards School of Business, University of Saskatchewan, Saskatoon, Saskatchewan, Canada
    Search for more papers by this author
  • Juan Wang,

    1. Department of Finance and Management Science, Edwards School of Business, University of Saskatchewan, Saskatoon, Saskatchewan, Canada
    Search for more papers by this author
  • Craig Wilson

    1. Department of Finance and Management Science, Edwards School of Business, University of Saskatchewan, Saskatoon, Saskatchewan, Canada
    Search for more papers by this author

  • The authors wish to thank an anonymous referee for insightful comments that improved the paper significantly. Wilson acknowledges financial support from the University of Saskatchewan President's Social Sciences and Humanities Research Council. Wilson and Tannous appreciate receiving travel grants in support of this study from the research fund at the Edwards School of Business, University of Saskatchewan. The authors are indebted to Gordon Sick, Marie Racine, Hung-Ling Chen, and Hilal Yilmaz for their valuable suggestions. We thank participants at the 2010 Eastern Finance Association (EFA) Conference, the 2010 Western Economic Association International Annual Meeting, and the Edwards School of Business Finance Seminar for their helpful comments. We appreciate the staff of the Technology Support Center at the Edwards School of Business for their help with the data and analysis. Any remaining errors are our own responsibility.

George Tannous

Department of Finance and Management Science, Edwards School of Business

University of Saskatchewan

25 Campus Drive

Saskatoon, Saskatchewan

Canada S7N 5A7

tannous@edwards.usask.ca

Abstract

Studies suggest that investment flows, liquidity imbalances, and institutional trading may create intraday trading patterns and opportunities for investors to time their trades to reduce transaction costs. Motivated by these studies, we divide each trading day into 13 half-hour trading intervals and measure information asymmetry from price changes, trade sizes, and trade directions. We find that information asymmetry starts high in the morning, drops continuously until it reaches a midday low during Interval 7, rises to a midday high during Interval 10, and drops continuously after. In contrast, neither the spread nor the depth exhibit similar midday extreme values. Essentially, we identify a 90-min window in the afternoon when net valuable information arrives to the market in high frequency while liquidity is stable, and that may be an opportunity for some investors to time their trades. In addition, we show that market makers employ dynamic strategies that change the spread, the depth, or both to manage information asymmetry. This is particularly evident during the last three trading intervals, where the significant drop in information asymmetry is countered primarily by a significant increase in the depth while the spread is almost constant.

Ancillary