Page 183 - 34-2

NTU Management Review Vol. 34 No. 2 Aug. 2024

reform dummies and report them in the context, with y being the order choices, the order
t
execution quality, and the individual stock turnover (see Tables 7, 8, 9, respectively), for
discussions on the “true” impact of Reform I and II on various aspects. In the end, two
types of robustness checks, including reducing the sample size and extending the sample

period, are also considered in the current study.

3. Findings

Our findings are fruitful. First, Reform I and II improve the Limit-order Book (LOB)
transparency as well, encouraging the informed and uninformed traders to make somewhat
different reactions. For instance, increasing auction frequencies may influence the different
facets of order aggressiveness for the two types of traders. Considering the high demand

for immediacy and severe competition among informed traders, institutional investors
significantly increase their order aggressiveness in prices after Reform I and II. To avoid
excessive price impact or revealing complete trading intentions, institutional investors
engage more in order splitting for large cap stocks when auction interval was further

reduced to 5 seconds. Considering that more transparent LOB information may lower
the risk of winner’s curse (Tseng and Chen, 2015), individual investors tend to increase
their order aggressiveness in prices after Reform I. However, when the auction interval is
further reduced to 5 seconds, the ‘wait-and-see’ effect in turn dominates the ‘encouraging’

effect, causing individual investors to significantly decrease their order aggressiveness in
prices for large cap stocks. As for the small cap stocks, where the financial reports and
corporate information are less transparent, Reform I and II seem to slightly improve the
transparency of their trading environments, encouraging individual investors to place more

new order submissions. Except for the above scenario, the “time contraction” effect is
dominant and causes the significant decrease in the number of new order submissions and
cancellations.
Second, it is notable that more frequent auctions are detrimental to order execution

quality for uninformed traders. Though the probability of limit-order execution within
the present day seems unchanged, the Reform I and II decrease the possibility that orders
are quickly executed in the present auction. This finding is bad news for individual day
traders, who need immediacy and are eager to trade quickly (Chiu et al., 2017). I also find

175