Hong Kong Regulation
With Emma Quinn, Head of Asia Pacific Trading, AllianceBernstein
How do the SFC Algo regs look six months on?
From a buy-side perspective it meant firms were minimising their administrative burden by reducing the number of brokers that they use. My personal view is that people will now start adding more brokers, if they haven’t already, now that they’re over the hump of the administration.
I’m not one of those people but I do think others will take that initiative.
Will people start innovating again?
No: I think people are going to pause and realise that they are quite happy with the more vanilla product that they’ve got. Responsible officers will want to make sure they understand every aspect of the algorithm that they use and are unlikely to want to do anything with more exotic algos and experimental technology.
SFC first or innovation already paused?
When people feel personally responsible for a spike in the market, for example, from a small change to an algo, it can scare you.
People should be responsible and know what they’re using; there’s no doubt about that. If you have 10 providers that each provide 10 algorithms, the chances of you understanding a hundred algorithms in enough depth is just not possible. No one can know it in that much detail. In my experience you find that traders tend to use a maximum of 10 anyway: from the general algo list you might use two IS ones that you like, and two sniper ones (and maybe an over the day algo as well), but at the end of the day you don’t need too many varieties.
Unintended consequences that the SFC may not have thought about?
I haven’t seen any unintended consequences that the SFC may not have thought about: I think the SFC intended people to be more responsible for their algorithms. The result has been that people taking responsibility for their own actions within the market. One aspect that may not have been considered fully is the implications for those trading out of Hong Kong into other jurisdictions that are captured by these rules.
And how does that feed into the current conversations around circuit breakers and closing auctions?
It’s interesting that there was only one party that was left out of the SFC algo regulations, and I would like to think that the Hong Kong stock exchange would do what most other developed markets in the world have done and introduce circuit breakers. The risk is that the buy-side do all the right things, the broker does all the right things and something happens between when the broker sends their message from their system to the exchange. The exchange is your last line of defence. There are four participants in the chain, including vendors, and one of those, the exchange, has been left out of this completely.
How does that feed into the dark pool regulations?
This has obviously got a lot more publicity after the “Flash Boys” book and the way exchanges would like volume to be 100% back on their markets, but it’s not an issue in Hong Kong. You only have to look at the numbers; not a huge percentage of the market is done in the dark.
How about retail flow going into the dark?
I don’t think that they will allow retail in there. If I was a retail person, I would of course like to improve my pricing with the use of dark pools. If trading at mid you’d always do better than the offer or better than the bid, depending on which side you’re on. With that being said, dark pools are more about block trading; they are not about the small orders. It is more for doing the big blocks in an anonymous form.