In effect the other participants are buying ahead of the fund, benefiting from the inevitable uplift that the large order will have on the price and taking a margin for ultimately selling their short-term purchases back to the fund at a slight premium. The fund may end up achieving the large purchase that it wished to achieve, but not without moving the market away from the price at which it wanted to execute the trade.
By digression there is an alternative scenario that is worth brief discussion which also illustrates the way in which order flow can be interpreted by market participants. This is the so-called “pump and dump” strategy in which a large fund or trading desk is keen to show to the market that it has a particular interest in a large order. After revealing its intention for all to see, let us assume that it is a large buy order, the market reacts to the move by following through with positive price action thinking that the buyer must have some superior knowledge about the attractiveness of the particular security that is being purchased. In fact the buyer is hoping to sell unwanted inventory into the strengthening market. This highlights a theme that we shall return to repeatedly which is that nothing influences price development more than price development. Another saying that seems apropos is the beautifully ironic remark that Wall Street is the only place that puts its prices up when it wants to have a sale.
Returning to the concerns that large institutions have had about exposing their orders to the market, a new type of automated process has been developed to disguise the true intent of these large fund managers. The process, known as algorithmic trading, not only facilitates the more efficient execution of large orders, but can even introduce subtle false signals into the procedure which are designed to confuse the markets about the underlying transaction objectives. For example, if a fund wants to buy a large quantity of a particular stock, the order is “sliced and diced” into a series of much smaller sub-orders and then executed over a period of time where the objective is to achieve actual price executions at the optimal cost. In otherwords, the algorithms are capable of scattering the original trade objective into a fragmentary process which should no longer be transparent to other market players. As part of the procedure the algorithms can also throw off contrarian trades that will from time to time reverse the original motivation by, for example, creating a selling phase within a large buy order:
The most common type of algorithm, called Volume Weighted Average Price (VWAP), slices the parent order into a series of child orders over a certain time frame, attempting to conceal the true size of the parent order. These algorithms are dynamic and in response to current market conditions, cancel and replace live orders. Each time an order is canceled and replaced, the information becomes part of the market data environment. Therefore, the use of algorithms has not only increased the number of trades that occur, but it has increased the amount of intraday market data.
One of the consequences of this innovation is that the microstructural behavior of markets is changing. There is far less transparency at the order book level and even when a series of orders do appear on the Level 2 or DMA screens there is a real question mark as to how firm these “orders” really are. Access to the order books was originally seen as a giant step forward in increasing market transparency and leveling the playing field for smaller traders, but as with most innovations there are usually ingenious techniques designed to defeat the purpose. Traders, both large and small, welcome transparency as a great principle but in practice they would rather be able to operate anonymously and stealthily in the marketplace (other than in the “pump and dump” mode we discussed).
There has been a lot of innovation regarding the complexity of the algorithms that buyside traders are now using and the motivations have extended beyond the original desire to “hide” large trades. Another important driver of the trend is the changing landscape between the buy-side (i.e. the large pension funds, mutual funds etc.) and the sell-side (i.e. the large brokerage operations that are focused on taking a small (and smaller) margin or commission from executing the trades of the large players on the buy-side). Issues such as the competitive nature of commission arrangements, the separation of research and trading costs and activities and the confidentiality of trading motives are also pushing this agenda. According to the TABB Group in late 2005, more than 60% of buy-side managers were experimenting with algorithmic trading techniques.
We need to clarify the significance of these newtechniques and to differentiate them from the more “traditional” notions of computerized trading known as “program trading”. Algorithmic trading has very different objectives to program trading which was a technique pioneered in the 1980s designed to exploit temporary arbitrage opportunities that arose in the trading of cash instruments such as the S&P 500 cash index and its major constituent stocks, and the futures contracts that trade in parallel with the cash market. When the derivative (the futures contract) and the cash index (“the underlying”) become misaligned a risk-free arbitrage opportunity arises and program trading takes advantage of these temporary spread discrepancies:
Algorithms are a step up from the more familiar program trading, which institutions for years have used to buy or sell bundles of 15 or more stocks worth a combined $1 million. Algorithms handle trades in individual stocks, and the exchanges don’t ban their use when trading becomes extremely volatile, as they have done with program trades since the 1987 market meltdown. As the use of algorithms moves from hedge funds and Wall Street’s trading desks to mutual- and pension-fund managers, it will account for more than 40% of total U.S. equities trading on all markets by 2008, up from about 25% today, according to Boston-based researcher Aite Group.
To highlight this realignment of the workflow between the major market players, the brokerage and investment banking business, which, largely pioneered the algorithmic trading technology and uses these platforms for conducting its own proprietary trading activities, is morphing its role with respect to large buy-side players:
Many bulge-bracket firms – the major brokerage houses that underwrite and distribute securities as well as produce research – are taking on a consulting role, helping buy-side customers choose algorithms. Brokers say they’ll advise buyside firms on which electronic strategies to apply for particular trading styles and develop customized algorithms, as well as pre- and post-trade analysis tools, for clients.
In February, Goldman Sachs began providing a framework, known as the orderexecution “Cube,” to help buy-side customers classify their orders and segment their flow by methodology and venue. “The Cube maps orders into different execution strategies based on order size, liquidity, and trade urgency,” says Andrew Silverman, head of U.S. algorithmic trading at Goldman Sachs, who explained the concept in April at a trading technology conference.
Why should the individual trader be concerned about this issue? Surely it is only of relevance to the largest institutional players and has little bearing on the activities or concerns of the smaller fund manager and individual trader. But we would argue that because of these fundamental changes to the manner in which volume is recorded, and the fact that the use of algorithms has not only increased the number of trades that occur, but also the amount of intraday market data, there have been radical changes to the ground rules that are the basis for many technical indicators that are widely followed by practitioners of technical analysis. A substantial amount of the legacy indicators in technical analysis have assumptions about volume, money flow and other measures of accumulation and distribution. Can these be as valid today, given the nature of the obfuscatory intent of algorithmic trading, as they were when the traditional trading workflow paradigm was in place?
Read More : Algorithmic Trading
