Trading on experience

Grace Lin, managing director, Citi’s head of Electronic Execution Asia Pacific, and Morgan Dunbar, director and co-head of Electronic Execution, Nikko Citigroup Limited, examine the impact of market volatility on customer use of algorithmic tools and the information they should be seeking from their brokers to trade effectively.

There is one broad trend in the relationship between sell-side firms and their institutional clients that all global brokerage firms would recognise: the desire among investors to take increasing charge of their own order flow. This is evidenced in the growth of automated trading tools such as DMA and algorithmic trading.

Until the middle of last year, however, the trend coincided with a long-running bull market. How has the volatility associated with the subprime crisis affected client attitudes to the use of these tools? In times of trading volatility, how receptive are buy-side customers to innovative ideas or do they tend to retreat into an earlier comfort zone, where sales traders are handed the more difficult decisions?

Even in the present climate, which on some days borders on febrile, most large institutional clients would still prefer to make the trading decisions themselves. They are, however, not necessarily comfortable with the way the tools available to them function in a volatile environment.

The comfort of history
A majority of institutional brokers have long had algorithms in their armoury with time-based or schedule-based benchmarks. Schedule-based algorithms draw on an historical approximation of volume, combined with real-time market data and market activity. In conditions of higher volatility and a consequent broadening in the range of prices at which a stock trades throughout the day, a ‘hit and miss’ element may attach to their performance.

In such conditions, we would therefore expect some migration to more dynamic algorithms. This may be as simple as making use of synthetic order types that exist in other markets, but which are not available on Asian exchanges. In the US market, for example, we might be able to take advantage of wide spreads or long imbalances in the queue through hidden order types that Asian exchanges simply don’t support. We are, however, able to offer Asian customers ways to take advantage of such opportunities through our algorithms, which effectively mimic hidden order types.

The availability of these tools does not necessarily mean, however, that clients are willing to take control of their deployment from the outset. Schedule-based algorithms have, after all, been in widespread use for longer. They have the added advantage that the clients themselves may be able to recognise the historical patterns underpinning them and may have a higher comfort level as a result.

Although Citi and its many of its peers have a tested range of dynamic liquidity- responsive algorithms, that they would have no hes- itation in deploying, clients may be less confident simply because they are less familiar with these particular tools.

Client choice
In such circumstances, there are effectively three options. First, customers can stick with what they know. This may be a credible approach, depending on the objectives attached to the specific order concerned. Secondly, they may hand the order off to our sales traders; or they may look to us to explain the nuances of the tools with which they are unfamiliar and to retain control of the execution process.

The fact that a market is exhibiting volatility does not mean that customers are any less inclined to capture market opportunities. ‘Sitting out’ the storm is not a viable option for most institutional managers. Brokers therefore need to be able to anticipate and respond to the full gamut of client behaviour.

Despite the diversity among Asian equity markets, the advice we would give is more likely to be name-specific than marketspecific. For example, certain Korean names lend themselves to all day VWAP, even in more volatile markets. While market structure issues present their own challenges, we should not lose sight of the trade-bytrade situation, regardless of different market structures and queuing mechanisms. At Citi, we have noticed an increased use of both our sales trading and execution consulting services across the region, again varying by client rather than market. Traditional long-only funds may prefer the comfort provided by the sales trading desk. Many institutional clients in Asian markets, whether long-only or hedge funds, are however, more exercised by issues of anonymity (well aired in the last issue of The TRADE Asia) and would therefore be more inclined to call on our execution consulting expertise.

In so doing, they will be seeking more than a simple support monitoring service. Citi has a global team that tracks liquidity and volatility on an ongoing basis, since these inform our own strategies and feed many of the models that we provide to customers. Where volatility is increasing across the globe, customers, as part of the execution consulting service, should want to understand what is in those models, how they factor in intraday volatility as opposed to whole day volatility, what weightings are assigned accordingly, and how we ourselves are adapting and reshaping these strategies, before deciding how to use them.

There is, of course, no right or wrong answer in any trading scenario – it depends on client objectives – but if they are to retain control of their execution strategies, clients need to make sure that they know how the vehicle they choose to drive should be handled in changing weather conditions and over different terrains.