Liquidity Supply and the Intensity of Limit Order Cancellations
Posted: 17 Jul 2013 Last revised: 19 Jul 2013
Date Written: February 22, 2013
Abstract
We document that around 95% of the limit orders submitted to the London Stock Exchange (LSE) are cancelled in a short period of time. Based on that, we look at the intensity of limit order cancellations and the effects on market liquidity.
We find that the rate of limit order cancellations increases: the greater the number of cancellations observed and the higher the market volatility is. For the case of short-lived orders, we find evidence supporting that the intensity of cancellations is higher: when the liquidity is low at the time of the order submission and when the market is becoming more liquid after the time of the order submission; the more aggressive a limit order is; as limit orders move away from the current market prices; as fewer limit orders are executed and as the depth of the opposite side of the order book is smaller. For the case of long-lived orders, we find that the intensity of cancellations is lower: when the liquidity is low at the time of the order submission and when the market is becoming less liquid after the time of the order submission; the less aggressive a limit order is; as limit orders move away from the current market prices and as fewer limit orders are executed.
We find that the higher the duration between two cancellations, the lower the impact on the market spreads. On average, long-lived limit orders improve market spreads and short-lived limit orders worsen spreads. However, when long-lived limit orders are aggressively cancelled as they are approaching the top of the order book, they actually degrade liquidity conditions, thus behaving like short-lived orders.
These results have important implications for regulators. First, our findings may help to understand the use of limit orders and under what circumstances they are cancelled. Second, our results can be used for the development of market surveillance systems raising flags to detect the presence of market abuse conducted, for example, by recurrently cancelling limit orders.
Keywords: Limit orders, cancellations, liquidity
JEL Classification: D40, G14
Suggested Citation: Suggested Citation