The recent impact of high-frequency trading on European and US equity markets has sparked debate among investment institutions about whether the current market structure now favours technologically superior market participants.
Towards the end of 2008, wild volatility in US and European equity markets caused traditional market participants to retrench their activity, while high-frequency strategies thrived on the market uncertainty.
Most high-frequency trading in the US and Europe is thought to be derived from electronic market making, i.e. providing two-sided markets lasting for a matter of milliseconds in the hope of profiting from the spread.
The main worry is that high-frequency traders can hit bids and offers ahead of other market participants with slower execution capabilities and therefore force laggards to aggressively cross the spread more frequently to get executed and avoid having their intentions exploited by electronic market makers.
Upon first viewing it seems as though this kind of practice can only be to the detriment of non high-frequency asset managers, hedge funds and mutual fund investors, but the overall impact of electronic market making tells a different story.
“While the traditional market participant has to cross the spread more often, it is not necessarily a bad thing. Spreads are narrower compared to ten years ago, in large part because high-frequency traders have made the market more efficient,” comments Justin Schack, vice president, market structure analysis at boutique US brokerage Rosenblatt Securities. “It does affect brokers however, as they have a harder time getting their limit orders filled and have to pay to take liquidity more frequently.”
By having to use aggressive orders more often, brokers will have to pay higher trading venue fees for removing liquidity, instead of receiving rebates for posting passive orders; costs that are rarely reflected in commissions paid by the buy-side.
“The effect on their execution-fee bills is probably a big reasons why some larger brokers have expressed concerns over high-frequency trading, and could explain some of the thinking behind the London Stock Exchange’s decision to move from maker-taker to volume discount pricing earlier this year,” added Schack.
Moreover, buy-side traders consider that the advent of high-frequency trading can help to stabilise the market in periods where there is reduced activity.
“High-frequency trading is somewhat of an enigma that some off the buy-side haven’t really looked at in depth,” comments Steve Wood, global head of trading, Schroders. “During the financial crisis last year, we found that high-frequency traders actually added liquidity to the market, which is in marked contrast to 1973 when liquidity completely dried up.”
Nevertheless, some buy-side traders are worried that by subscribing to the high-speed data feeds of trading venues, such as Nasdaq OMX’s ITCH feed or BATS Exchange’s FASTPITCH, high-frequency traders are able to act upon trading opportunities before those getting the regular consolidated feed.
For example, a high-frequency trader that uses a venue’s ultra-low latency data feed may spot a long-only institution placing a bid on that venue at a better price than is currently available, microseconds before it is viewed on the consolidated tape.
Using an algorithm, a high-frequency trader could then post the same bid on one or more different trading venues and receive an execution ahead of the long-only firm, as price-time execution priority does not apply across orders posted on multiple trading venues. The long-only investor potentially faces the risk of missing his an execution opportunity at the expense of a high-frequency trader.
The ability for a select number of firms to view market data before anyone else bears similarities to the flash order debate that is still ongoing in the US. Flash orders allow markets to delay routing orders to other venues that show a better price – as required under RegNMS – to give selected customers the option to execute against it first. This could leave buy-side traders disadvantaged as the market may have moved unfavourably during the time an order is being ‘flashed’.
However, advocates of high-speed data feeds would point out that they are available to anyone, do not send out specific orders and therefore would not disadvantage end customers.
For traditional buy-side institutions that have sat and watched while the technological arms race unfolds in front of them, transparency is the key. Many feel that if they are given a choice when it comes to execution against high-frequency flow, and have the tools to measure its impact, any opportunity to improve execution should be taken.
“As long as we on the buy-side are aware of what we are transacting against and the cost of trading with these types of market participants then we have no problem executing against high-frequency flow,” says Wood. “Dynamic transaction cost analysis and measuring the visible cost of trading will be crucial tools in measuring the influence of high-frequency trading on execution performance.”
To vote in the new poll on high-frequency trading, please click here.