High-frequency trading as a portion of the U.S. equity markets has exploded over the past decade and is now estimated to account for roughly 75 percent of market volume. As the use of algorithmic trading has increased, the difficulty in actively trading a large book has dramatically risen. While proponents argue that high-frequency trading (HFT) creates liquidity, it is unwise to depend on individual firms that have no duty to be liquidity providers. Much of today's HFT represents opportunistic liquidity provision, not the liquidity provision of old, making the execution of large orders challenging.
The role of the NYSE specialist as a market maker has largely been eliminated as electronic designated market makers (DMMs) and special liquidity providers (SLPs) are now employed to provide liquidity in equities. The growth in HFT is argued to have filled any liquidity void, yet this type of provision is far different than the specialist model.
Throughout the history of the equity markets, liquidity has been understood as the ability to enter and exit positions with a limited impact on price. A crucial duty of a liquidity provider was the requirement to be on the bid or offer in highly calamitous times. Worries of specialists going bankrupt after the 1987 crash and after the collapse of Long Term Capital Management (LTCM) in 1998 were common because the NYSE requirement to step in on the bid forced specialists to take huge risks.
HFT, seen today as the main source of liquidity provision in markets, does not offer this same protection to investors. High-frequency traders - outside of DMMs and SLPs - simply owe no fiduciary duty to anyone to create markets in stocks and thus should not be depended on for liquidity.
As such, many algorithms are programmed for what is dubbed "opportunistic liquidity provision." These particular programs attempt to detect the footprints of larger players in the market and trade off of the order flow for miniscule gains back and forth. While these high-frequency traders may be on the bid or the offer, their intention is to profit by trading alongside larger players rather than by taking the opposite side of the trade.
With their opportunistic nature, high-frequency traders also seek to avoid any large losses that could result from wild gyrations in the stock market, as any player without a direct obligation would do. For example, during the Flash Crash it was reported that several major HFT firms, namely Tradebot Systems and Tradeworx, shut down their systems to protect themselves. This worked to remove critical sources of liquidity that the market has come to depend on when it was needed most.
These two major differences in today's liquidity - its opportunistic nature and its ability to shut down at any time - make the entering and exiting of large positions very difficult. Many large buy-side firms have joined the race by hiring computer scientists to write algorithms to execute their orders in hopes of hiding their order flow and causing the smallest possible impact on price. Another tool to combat the high sensitivity of price to large orders has been the growth in dark pools, which help facilitate the movement of large blocks of stock without displaying the size to the open market.
A recent study commissioned by Knight Capital Group, "Equity Trading in the 21st Century," showed that in the past five years average trade sizes on the NYSE have fallen from more than 700 shares each to just more than 300 shares by the end of 2009 (see chart, at top), while the cancellation-to-execution ratio has doubled from 15-to-1 to 30-to-1 (see chart, at right). Recent proprietary estimates by T3Live.com show this cancellation ratio to be as much as three times higher. This high ratio results from splitting up orders to hide intentions and frequently cancelling orders to maneuver around. T3Live proposed instituting a miniscule order cancellation tax in order to curb much of the cancellation trickery played by high-frequecy traders.
Revise Reg NMS?
As a result of these trends in liquidity provision, the active equity trader has been increasingly challenged. Not only has it become far tougher to read order flow, finding fills for orders has been hampered as HFT runs circles around stagnant bids and offers placed on the order book. While many recent studies have demonstrated the benefits of HFT by showing the rising liquidity in high-volume, large-cap stocks, manual traders across the spectrum have recognized how arduous the process of receiving fills on bids and offers has become in many stocks outside the most liquid universe.
Many of the current problems could be corrected by adjusting Regulation NMS, which prohibits trading through the inside bid or offer. The inability to do this slows orders and makes the footprint of large players much more observable. In general, structural changes in the market have forced many traders to reduce their position sizes and adopt larger timeframes for their trades, giving in to the execution difficulties.
While no market participant should be unfairly singled out or punished for its market activities as long as its actions are legal, designated market makers have always been necessary for the efficient functioning of securities markets. The problem in recent years arises from regulators depending on HFT for liquidity provision even though many HFT firms do not intend to be such in the traditional sense of the term - unless they are a DMM or SLP.
HFT is a fellow player in the market seeking to make money like any other player and oftentimes increasing the efficiency of the price discovery process. A rethinking of Reg NMS could solve many of these issues.