Limit Order Submission Risks, Order Choice, and Tick Size (Summary) (2024)

The author uses the Tokyo Stock Exchange (TSE) order book and transaction dataset to obtain data for 426 stocks. These data include the recent high-frequency trading environment following the 2010 introduction of the TSE’s “arrowhead” trading system. The dataset allows the author to pair stocks that were subject to the minimum tick size change with those stocks that were not. Specifically, he matches each stock with a control stock in the same industry, in the same range of trading frequency, and with a close parallel in relative tick size. This process results in 30 stock pairs for the first tick size reduction period (2 December 2013 to 21 February 2014) and 52 pairs for the second period (9 June 2014 to 29 August 2014).

In testing whether a reduction in minimum tick size decreases non-execution risk but increases picking-off risk, the author compares the non-execution rates at the best price and the rates of being picked off at the best price before and after two dates: 14 January 2014 and 22 July 2014. These dates fall in the middle of each tick size reduction period examined. The study then applies a Wilcoxon signed-rank test to discern any difference in the target or control stock rates before and after the tick size modification.

The author also conducts four tests on the minimum tick size reduction’s impact on order aggressiveness. From most aggressive to least aggressive, order aggressiveness is ranked as follows: market order, limit order inside the quotes, limit order at the quotes, limit order outside the quotes, and cancellation. For these tests, the author uses a similar test to compare the before-and-after results and also runs ordered probit regressions.

The author’s primary finding is that when the minimum tick size is reduced, non-execution risk decreases but picking-off risk increases for limit orders. This study is unique because the author is able to compare stocks that were and were not subject to tick size changes as well as control for other characteristics. This result has implications for evaluating minimum tick size reduction policies.

The author also analyzes how the two limit order submission risks affect order aggressiveness. Here he finds a less conclusive result: The impact is positive for picking-off risk but negative for non-execution risk. Consequently, reducing the minimum tick size increases traders’ tendency to place aggressive orders for some stocks but elicits the opposite reaction for other stocks.