Tick sizes and stock market volatility: A regression discontinuity approach

Sammanfattning: Increasing the tick size has been suggested as a countermeasure to high volatility on stock markets caused by growing high-frequency trading. To identify the potential effects of such an increase, we isolate the effect of the tick size change on volatility through a regression discontinuity design (RDD), utilizing the sharp increase in the tick size at 100 SEK where the tick size changes from 0.05 SEK to 0.10 SEK. The study is restricted to the Stockholm Stock Exchange, the 30 largest stocks on the market, OMXS30, and a range of share prices between 90 and 110 SEK. Volatility is measured in two ways: the level relative high-low range and the logarithm of the range, both as daily measures. While no direct causality between the tick size change and the level daily range could be isolated, the results for the log range indicate an average increase of about 10% following an increased tick size. This effect is corroborated by earlier panel studies isolating this effect. With the exception of types of volatility not controlled for in the daily range measure, such as clustered volatility, the results indicate that stock markets would not benefit, as hypothesized, from a marketwide increase in tick sizes.