Closing in on better performance
Ben Valentine, director, head of electronic execution sales, Asia-Pacific, and Ian Smith, managing director, head of electronic execution, Asia-Pacific, at Citi, tackle two thorns in the side of buy-side clients: the Hong Kong close and the lack of transaction cost analysis (TCA) standardisation.
As most of you know, the Hong Kong Stock Exchange (HKSE) close price is determined by the median of five nominal prices in the last minute of trading taken at 15-second intervals. To minimise slippage, a lot of the market participants using the Close benchmark try to time their trades at 15 seconds interval causing volume spikes around the snapshots times. The spikes in volume can cause large price movement and deviation from the final close price. Using data from the last four months for the Hang Seng Index (HSI), the last-minute VWAP deviates by about two-thirds of a spread from the close price. For certain names, the deviation is much larger and can be as large as 3.5 times the stock's average spread. The HKEx also imposes a throttle restriction which limits the number of orders it can process every second. Trading the close effectively can thus be difficult due to the combination of volume spikes, price volatility and throttle capacity.
Citi estimates that a naïve strategy trading only at 15-second intervals would underperform the close by half the spread on average, or about 10 bps for HSI names. To achieve good performance, a close strategy must trade an optimal portion of its order quantity at the snapshot times while passive quantity is pre-queued for a length of time that balances the trade-off between getting filled passively and the price moving away. Scope for change? We think most buyside and market participants would welcome a closing auction mechanism similar to others that exist, however for this to be successful there must be a reasonable level of consultation with both buyand sell-side. This is of course only one of several possible changes to market structure being discussed in Hong Kong with many hoping that progress might be made next year toward anonymity of broker IDs on the HKSE.
The other area we wanted to touch on is post-trade analysis. In an increasingly complex world where we are bombarded by vast amounts of conflicting data, having a clear, timely and unbiased source of information to benchmark and measure of an execution is vital. The old adage of 'if you can't measure it, you can't manage it' is increasingly being applied to dealing desks in an attempt to reduce costs, increase margins and help differentiate what appears at first glance to be an ever more commoditised execution arena. While most market practitioners will readily agree on the need for effective execution analysis, the market has yet to come up with a truly effective solution.
In order to find a solution that caters for the institutional client who trades multiple baskets across multiple brokers a number of problems need to be solved. First and foremost, the industry must to agree on how to measure and calculate each benchmark. This sounds easy until you consider the vast number of benchmarks in use and the seemingly minor differences between the ones that on the surface appear the same. Should you include on and off exchange prints in total volume? Which venues should be included? Should you measure against all day VWAP or interval VWAP? The list goes on.
The second major problem is one of perception and bias in who can and who should run the analysis. Sellside firms are seen (unfairly!) as defining benchmarks and calculations that make their executions look better versus their competitors. While this would be addressed by a common definition, more importantly the sell-side firms do not normally see 100% of the flow of any given basket, and find it difficult to provide a comprehensive analysis. In contrast, independent posttrade firms see all the clients flow but are often not transparent about their metrics or how they rank executions, which can inhibit the buy-side dealing desk in trying to maximise the performance from their sellside providers. Further, without independent TCA providers consulting with the sell-side on how best to optimise a client's trading, it is difficult to see how the analysis can be truly effective. Perhaps in response to these sub-optimal solutions just mentioned, we are also seeing ever more of our large institutional clients running their own posttrade analysis. While it is new territory for many with considerable pitfalls, it does make a lot of sense to invest in fully independent and highly-tailored, information- rich reports that focus on exactly what is being looked for. However, even with large institutions creating their own post-trade analysis, end-investors will still require a consistent view on performance completing the circle and leading us back to the first point.
Encouragingly, the first steps towards effective and consistent post trade analysis may have already been taken with the OpenTCA project being driven from Europe by a number of the large sellside firms including Citi that aims to get the industry starting to agree on what a good trade actually means.