The Going-Concern Market Anomaly

Authors

  • ASAD KAUSAR,

    1. Manchester Business School, University of Manchester
    Search for more papers by this author
  • RICHARD J. TAFFLER,

    1. Martin Currie Professor of Finance and Investment, University of Edinburgh
    Search for more papers by this author
  • CHRISTINE TAN

    1. Fordham University. This paper has benefited from the helpful comments made by Vineet Agarwal, Mike Bowe, Elisabeth Dedman, Ilia Dichev, Stuart Hyde, April Klein, Edward Lee, Clive Lennox, Lakshmanan Shivakumar, Dushyantkumar Vyas, and Martin Walker, as well as participants and discussants at the AAA 2006 annual meeting, FMA European meeting 2007, and EAA 2008 annual meeting, and seminars at New York University, London Business School, Arizona State University, University of Lancaster, University of Edinburgh, Cranfield School of Management, London School of Economics, U.K. INQUIRE, and Manchester Business School. We especially thank an anonymous referee and Douglas Skinner (the editor) for their valuable comments and suggestions.
    Search for more papers by this author

ABSTRACT

We explore the market response to announcements of first-time going-concern (GC) audit opinions and, for a subset of these cases, their subsequent withdrawal, from 1993 to 2005. We find that the market fully responds to GC withdrawal announcements but underreacts to the GC announcements themselves, resulting in a downward drift of −14% over the one-year period subsequent to the GC opinion. This result is robust to alternative explanations documented in prior literature. However, after adjusting for transactions costs, the opportunity to earn profits by trading on this market anomaly is limited. We demonstrate that despite such clear adverse signals about the firm's continuing financial viability, this information is not being fully impounded by the market on a timely basis. Our findings differ from those of others who suggest that there is no pricing anomaly associated with GC opinions in the United States. We show that this is likely due to important issues with their research methods.

Ancillary