The Gateway to Algorithmic and Automated Trading

Hedge Funds and Multi Asset Algorithmic Trading - Perfect Partners?

Published in Automated Trader Magazine Issue 02 July 2006

The exponential growth of the hedge fund sector has been a major feature of capital markets over the past few years. Another more recent hot topic has been algorithmic trading across multiple asset classes. Matt Simon, Analyst at the TABB Group considers whether a union between the two could be a marriage made in heaven.

Matt Simon

A new wave of low- and no-touch trading is gaining acceptance, and this time around it is not all about flashy equity algorithms focused on niche strategies. Market participants are fanning out across foreign exchange, options, futures and even fixed income markets. Leading brokers are developing high-tech offerings that are potentially appealing to buyside participants for their cost efficiency, control features and sophisticated overall trading capabilities.

There is an important terminological distinction to make before embarking upon a discussion of multi asset class algorithms. In equities, algorithmic trading is generally understood as using tools/rules that attempt to optimize the execution process - in effect synthesising the activity of a skilled and tireless human trader. In other asset classes, algorithmic trading is synonymous with black-box trading. Black box trading is the automation of the investment decision process primarily using fundamental data. For the purposes of this article, we are mostly referring to execution only algorithms and refer to black box trading as needed.

Who wins?

When looking globally, a logical question about the adoption and use of multi asset class algorithms arises: Who has the most to gain? The answer could very well lie with the hedge fund industry. In the recent TABB Group report "Hedge Funds 2006: The Quest for Alpha in a Competitive World", it became evident that hedge funds are going to be a major driver of future capital inflows for at least a few years more. Due to the larger percentage of total capital that hedge funds are now capturing and their consequent increase in daily transaction volume, hedge funds seem to be a perfect fit for advanced execution technologies.

Figure 1

Much like traditional asset managers, alternative investors are constantly looking to reduce trading costs, improve execution and find the fastest way to deploy large amounts of capital. As a result, a growing number of institutional investors, seeking greater yields, have allocated a larger portion of their assets to hedge funds. This small portion of the immense capital controlled by institutional clients is having a huge impact on the growth rate of the hedge fund industry, which exceeds 50% per year since 2000. It is apparent that hedge funds have drawn in new capital because of their ability to achieve superior benchmark returns and their willingness to incorporate advanced electronic trading into their everyday processes.

Market influence

Larger hedge funds (defined by TABB Group as those with over USD1bn in assets under management) have played a major role in the fast growth of electronic trading. In the last few years, a few large US hedge funds even developed black box trading strategies that have allowed them to trade large quantities of stock with the intention of making just a small profit on each share. Hedge funds with these black box capabilities are then routing their orders to electronic execution venues, such as ECNs and crossing networks. This phenomenon has given a relatively small number of funds a huge influence on the trading habits of the asset management business. The huge volumes of order flow these hedge funds trade will continue to have an indelible, though perhaps immeasurable impact, on market structure, including squeezing spreads, lowering commissions, and increasing quote traffic.

Given that hedge fund managers have squeezed out the inefficiencies in the micro-structure of the US equity markets, they are adopting their strategies accordingly. The black box trading community is beginning to trade across different geographies and in different asset classes.

Traditional asset managers will pick up on this trend going forward, and greater use of no- and low-touch tools should develop for trading European and Asian stocks, just as it has in the U.S. markets. The use of DMA and algorithmic trading can then be viewed as solutions to hedge funds' increased demands.


Another focus area for hedge funds will be on risk management tactics, where trading in multiple asset classes, such as equity derivatives and options, will be necessary for hedging against international equity movements. As the global markets develop, multi asset class algorithms will be brought in for the purpose of executing more international equity and exchange rate trades.

Hedge fund investors are not paying a two-tiered fee schedule for mediocre returns or a typical risk profile. The intense competition for returns, in conjunction with a flexible investment mandate, leads most hedge funds to refine their strategies regularly and expand the scope of their search for good trading ideas. Everyone wants to know where the next best idea will be captured and how to execute at the highest rate of speed, but achieving that goal is another story.

Speed and precision

Ultimately, a primary objective of any hedge fund trader is to carry out the successful trade effectively in as little time and with as much precision as possible. As hedge funds stride forward in areas of complexity, their respective market equals will continually try to mimic and one-up each other. So, using multi asset class algorithms can be viewed as a weapon for attaining the much-desired competitive edge.

The trend can already be observed today; over the last few years US hedge funds (accounting for about one out of every two funds globally) have been rapidly expanding the instruments they trade in because of the diminishing opportunities in US equity markets. Language barriers and intricate trading instruments are creating inefficiencies in the global markets, and where this inefficiency becomes apparent, therein lies the potential alpha.

No-touch, low touch

If the current use of algorithms is to find alpha, then any indication of future adoption of multi asset class algorithms can only be seen as positive for the hedge fund industry. When weighted by assets under management, more than 50% of hedge fund order flow is currently routed to noand low-touch channels, most significantly proprietary trading algorithms (see Figure 2). Traditional managers are embracing electronic trading, but instead of building systems outside their core competency, they are relying on brokerprovided or other third-party solutions. Although traditional asset managers communicate most of their order flow electronically, much is still sent to the trading desk, almost twice as much as is sent by hedge funds.

Figure 2

Hedge funds also have a great opportunity to leverage their expertise to pick-off trades that institutions are improperly valuing, and/or to create additional liquidity where it may have not previously existed. Cross-asset class trading and the availability of high-speed electronic trading are both growing at an unusually high rate. There are some regional variations in the proportion of high/low-touch preferences. The largest discrepancy between high- and low- touch is found in Europe, where traders suggested that they would prefer low-touch channels 66% of the time, whereas today they currently allocate only 6% of their flow this way (see Figure 3 ).

Figure 3

Filling the gap

So why is there this gap? In some regions, markets or institutions are not structured in a way that enables them to achieve optimal benefit from electronic platforms. In other markets, where the providers (brokers and vendors) have not made strong enough inroads into these markets to drive penetration, there are significant opportunities for advanced execution providers.

As more hedge funds employ specific multi asset class strategies, prime brokers are catering to their new favourite customer, and are developing innovative algorithmic tools for different hedge fund trading needs. For example, certain prime brokers have already been effective in providing their clients with direct access to more geographic locations or different exchanges.

Additionally, broker dealers that are providing the electronic tools for other asset classes have developed the infrastructure needed to process and distribute a variety of asset class trading tools through a single platform for their proprietary trading desks. Multi asset class strategies must not only handle concepts such as market impact, risk aversion and volatility, but also the trade management and risk implications of being unable to execute one or more of a strategy's legs.

Prerequisites and obstacles

But in the end, the ability of hedge funds to integrate multi asset class algorithms into their everyday processes may depend on vendors finding a way of quickly incorporating them into their trading applications. This is by no means a given - the ability of the technology vendors that service hedge funds to understand, adopt and incorporate new technologies quickly and affordably varies considerably.

Although existing advanced trade execution techniques are already complex, multi asset class algorithms take the level of required understanding to an even higher level. Therefore, the ability to take advantage of such algorithms will be heavily dependent on the available level of intellectual capital. Fortunately for the hedge fund industry this is exactly what it has in abundance, as well as a proven track record for leveraging that intellectual capital to produce proprietary algorithms. (As witnessed by its high percentage of proprietary algorithm order flow noted earlier).

In the short term, any move to multi asset algorithmic trading will have to contend with obstacles such as liquidity discovery and market data capacity issues, but these should recede in time as brokers and exchanges further develop their technology capabilities. The only other significant impediments to hedge funds' adoption of multi asset class algorithmic trading are their existing investment strategies, which may be too tightly defined in their prospectuses to permit this expansion. However, this might in fact spur some managers to seek investor sanction for the relaxation or redefinition of those strategies, or even to create new subfunds that are dedicated to multi asset algorithmic trading.


Even after making allowances for the hyperbole that inevitably surrounds any new concept in technology/ finance, the union between hedge funds and multi asset algorithms does seem a natural. The agility of hedge funds in responding to new technological and investment opportunities is well-proven, and their increasing inflows (particularly from institutional investors) will certainly give them an incentive to expand into algorithmic trading across multiple asset classes.

The sellside has every reason to assist hedge funds in this expansion with both market access and algorithms. The numbers speak for themselves. A hedge fund with USD8bn under management typically generates the same sellside commission as a conventional asset manager eight times its size. Furthermore, hedge fund spending on prime brokers will probably hit USD10bn in 2006. That should prove sufficient incentive…