When balancing the trade off between finding liquidity and avoiding volatility around Asian open auctions, applying careful analysis to the specifics of the order generates more benefits than applying a standard approach or relying on ‘gut feel’.
As the investment process across Asia matures, trading desks at institutional investment houses across the region are looking for ways to add value to the investment process. There is therefore, a growing focus on reducing impact cost, lowering opportunity cost and intelligently sourcing liquidity. The process boils down to two fundamental questions a trader on a buy side desk needs to answer:
- Should I engage with the market at this point in time? And;
- To what degree should I engage?
There are multiple ways to address these questions, ranging from gut feel or traders’ market experience to more quantitative methods based on models that optimize the trading schedule to reduce impact cost and ‘timing’ or opportunity cost (e.g. ITG’s ACE). In this short paper we will focus on the question as it pertains to the Open auctions in Asian markets, offering a methodology that helps traders address the question of ‘engagement’ around the first few minutes of trading. The goal is to suggest an approach grounded in analysis that helps traders form a view on how to trade during and around the Open.
WHY THE OPEN? THE BALANCE BETWEEN LIQUIDITY AND VOLATILITY
The Open tends to be a ‘liquidity event’ in most markets. Accrued overnight information is funneled to price formation during the opening auction and the first few minutes of trading. As a result two things happen: the Open and the trading period right after it tend to represent a significant portion of the day’s volume, and given the price formation that takes place, markets tend to display higher volatility. The charts below illustrate these points for Hong Kong and Japan.
The traded volume around the Open in Hong Kong is around 1-3% of the daily volume while in Japan, the traded volume around the Open is around 4.5% of the daily volume (Figure 1).
The high volatility is also clearly visible (Figure 2). This has become particularly topical with the increased volatility in emerging markets, in general, and some Asian markets in particular, during Q3 2015. Note the distinct increase in Q3 volatility in Hong Kong given its China exposure, versus Japan. Not trading at or shortly after the Open not only excludes the investment manager’s input to initial price formation, but also denies access to the liquidity that can be significant at that time. However, participating around this time exposes the investment manager to price volatility and potentially disadvantageous fills if significant reversion follows.
TO TRADE OR NOT TO TRADE… HOW TO ANSWER THAT QUESTION?
Absent any stock specific news, deciding on participation depends on the direction of the trade (buying, selling or shorting) relative to the market trend: momentum, flat or reversion. Using past experience as well as tools that help predict the move on the Open can help the investment manager make a more informed decision at the start of the day.
Our analysis used a sample of about 40 investment management firms, each of which may represent multiple strategies and portfolio managers. We took the order characteristics (stock, side, size and day of trade) and modelled the prevailing execution price for those orders given a set passive participation (10%*) based on different start times: the Open, Open + 1, +5, +10, +15, +20, +25 and +30 minutes. We then compared the weighted average price the order incurred to the benchmark of the Open price to see if one timing strategy consistently delivered better results. In addition, we analyzed what percent of the trades may not complete given the later start time.
ONE SIZE DOES NOT FIT ALL
Our review of the results over our sample set suggests that for both Hong Kong and Japan starting an order at the Open would yield the better overall trading result during Q1 and Q3 of 2015, however during Q2, starting 15 minutes after the Open yields the best result in HK while a similar result is achieved if starting 30 minutes after the Open for Japan (Figure 3A and 3B). The later start causes a 2% drop in HK and 6% drop in Japan in order completion rate in our sample.
When examining individual orders in more detail over the three quarters of 2015, across the sample, some reversion after the Open is evident in both the Hong Kong and Japan markets thus it is worth looking at the distribution of best starting time by order (and not only the weighted average). The distribution is similar in the sense that starting at the Open or 30 minutes after the Open provides the best result for 62% (by value) of orders in HK and 70% in Japan (Figure 4).
It is clear then that while there is an overall pattern that can be gleaned – during less volatile conditions, taking advantage of liquidity and price formation in the Open generally improves the trading outcome, while during volatile conditions, starting trading later may help –the optimal strategy changes from one investment manager to another and sometimes from one quarter to the next. This suggests that there is merit in looking at the specifics of strategy, the manager’s trading goals and the trades themselves before coming to a conclusion of the optimal start time. Trading desks should run analyses for their specific profile, potentially by fund type or portfolio manager before converging on a specific strategy.
In this short article we demonstrated a methodology that can help traders at investment managers apply a data driven approach to help design a trading strategy for the Open. It is clear that the results are not uniform across the data set and differ from one investment manager to the next, as well as from one quarter to the next; it is very plausible results may differ by fund strategy and even portfolio manager. Therefore it is best to run a specific analysis before choosing a strategy for trading at the Open, and to continuously fine tune it as market conditions, such as volatility, change.
ITG’s Algorithm Execution Consulting team offers customized analysis to strategize the Open in Asia. The strategy can be further refined to apply various different attributes and help provide guidance to traders.
*We use a 10% participatory strategy as it is relatively passive thus suggests low impact. It is possible to repeat this exercise with higher participation rates, however one would need to account for the increased impact of such strategy.
Ofir GefenContributors Managing Director, Head of Electronic Brokerage, Asia Pacific
Mr. Gefen is a Managing Director and Head of Electronic Brokerage in Asia Pacific.
Mr. Gefen oversees ITG’s electronic and block trading, as well as algorithmic and dark pool product development in Asia Pacific. Mr. Gefen has been with ITG for over 15 years and in this time has developed experience across all regions through roles in product development and execution services.