Purpose
– The purpose of this paper is to empirically investigate whether call auctions which batch orders for simultaneous execution, may restrain stock market volatility.
Design/methodology/approach
– The authors use high-frequency data to investigate volatility changes following the suspension of opening and closing call auctions on the National Stock Exchange (NSE) of India in 1999. The authors evaluate this issue by considering both modelled and realised volatility. Using a GARCH approach the authors model intra-day volatility for the trading days preceding and succeeding the auction suspension. The authors also scrutinise return distributions to look for volatility changes during different parts of the day.
Findings
– When interpreted collectively, the empirical results suggest that the auction suspension was followed by reduced volatility particularly in the middle of the trading day and at the closing.
Practical implications
– Given that auctions are often incorporated in trading systems with the aim of curtailing volatility, the main conclusion, that the auction suspension was followed by lower volatility, has important practical inferences. Auctions cannot be automatically relied on to reduce volatility. The intricacies of the auction protocol and their interaction with ancillary market microstructure features may impact on auction efficacy.
Originality/value
– The paper adopts a novel approach towards assessing the effectiveness of auctions by considering an unusual occurrence of an auction suspension. The empirical setting enables a clear comparison of the respective regimes since these periods do not materially differ in other subsidiary aspects. This is a noteworthy factor, since the empirical contexts considered in prior studies, often feature several simultaneous changes.