Since the US Securities and Exchange Commission authorized electronic exchanges in 1999, high-frequency trading (HFT) has grown to account for over 70 percent of equity trades taking place in the US, 40 percent in Canada, 35 percent in London and rapidly growing percentages of trades in other European and Asian markets. When an activity that is secondary to the intended function of a marketplace reaches the point of eclipsing the intended activity, a serious examination of the effects of such activities is called for. This need is heightened by growing concerns that HFT may be contributing to market volatility and may, in some of its manifestations, be placing the general investing public at an unfair disadvantage.

High-frequency trading is computerized trading using programs which analyze large amounts of market data with sophisticated algorithms aimed at exploiting trading opportunities that may open up briefly, perhaps only for seconds or milliseconds. High-frequency traders (HFTs) use real-time data analysis to produce automatic trading decisions and have very short time-frames for investment transactions, resulting in the submission of thousands of orders many or most of which are rapidly cancelled. Thousands or tens of thousands of trades may be entered each day and the net positions at the end of each day are zero.

HFT is conducted mainly by proprietary institutional trading desks and is extremely sensitive to the processing speeds of markets and of access to those markets. HFTs try to take advantage of large institutional block orders by using sophisticated computer programs to look for signs that a block is in the market place and then attempting to get in front of the trade which is likely to have an effect on the market price. As a result, so-called “dark pools” have been formed away from the organized exchanges where blocks can be traded without information leaking.

Among high-frequency trading strategies are:

Market-making: high-frequency traders try to compete with traditional specialist firms and capture the spread between bid and ask, but, unlike traditional market makers, they do not have any obligation to provide liquidity under all market conditions

Tape-reading: HFTs monitor market data such as quotes and volumes looking for significant or unusual changes that they hope to take advantage of

Statistical arbitrage: HFTs exploit temporary deviations from normally stable statistical relationships between securities’ prices

Liquidity detection: HFTs determine whether or not there are large orders among traders by sending small orders to look for them (so-called “pinging”)

Low-latency strategies: HFTs try to minimize the speed of market data delivery from trading centre servers to the servers at the HFT, processing in the HFT’s computer, access to the trading centre’s servers and order execution at the trading centre. This provides minuscule, but potentially very lucrative, advantages in arbitraging price discrepancies between securities trading simultaneously on various markets. Latency is the amount of time it takes a message to traverse a system and is central in the issues surrounding HFT Those marketplaces that service a limited clientele – as opposed to the larger broader markets – can achieve faster speeds- lower latency- and can use that advanced speed to identify larger orders coming from more traditional sources and trade in front of them. It is hard to identify any generally beneficial market effects of this activity).

…HFT helps to align prices across markets, improve liquidity and lower spreads – but the latter effects are often small and transitory..

Proponents of HFT have argued that HFTs’ arbitrage strategies and market-making have helped align prices across markets, improved liquidity and lowered spreads. However the latter effects are often small and transitory as HFTs are under no obligation to always make a market or provide liquidity.

HFTs do not engage in fundamental analysis of securities’ investment merits but monitor, examine and try to act on market prices quicker than their competitors. HFTs pay for superior speed in market data delivery from trading centre servers to the servers at the HFT, in access to the trading centre’s servers and in order execution at the trading centre and this gives them a clear advantage over other investors. This raises the question of whether regulators should also allow preferential access to non-public inside information being released by public companies for payment or other expenditure and, if not, why one and not the other?

The capital markets are vital to our economic system. They are the conduits through which people’s savings are transferred to the most productive and worthy businesses and industries. These enterprises in turn generate optimum economic activity and levels of employment, contributing to social well-being. In order to function well the capital markets must be seen to be fair and free of distortions or corruption by powerful interests. The traditional goal of securities regulation has been to ensure access to investment information and market trading on as fair and equal a basis as possible for all investors. When HFT is based on quicker access to market information and trading facilities than is available to the general public, it goes against the spirit of the goals of securities regulation. Moreover in some of its manifestations HFT has been seen to actually contravene securities laws prohibiting front-running If the general public comes to believe that some investors are benefitting at their expense it will tend to discourage vitally important, fundamentals-based investing activity.

…there is a growing belief that HFT is increasing market volatility…

The sheer scale of high frequency trading raises questions about its contribution to the significant increase in volatility of market prices which has occurred since HFT came into existence and about the propriety of a kind of day-trading that can gives completely false impression of the true liquidity of securities and the numbers of investors who are interested in owning the securities of the enterprise in question While some have argued that although allowing different speeds and efficiency of execution enables some to benefit at the expense of others, we should not want to subject trading speeds to the lowest common denominator). Others have countered that if slower, more uniform trading speeds are consistent with greater fairness in the capital markets then that is a preferable circumstance, When a trader places buy and sell orders at prices not far apart within short periods of time and the increase in activity is intended to attract additional investors and affect the price, such activity has historically tended to be viewed as churning and prohibited. Today it comes close to describing what at least some HFTs are trying to do in their strategies. According to a study by Paul Woolley of the London School of Economics and Bruno Biais of the Toulouse School of Economics HFTs often place “spoof” orders designed to influence other traders to enter orders at which point they withdraw their spoof orders and take the other side of the real trades that they have influenced. In their view that constitutes a form of market manipulation. In the past traders frequently posted “spoof” orders to identify trading flows but the physical capability and the extent were limited. The question now is whether the capability of doing so and the scale of such activity have grown to an extent never anticipated and whether this is now having undesirable effects on transaction execution?

Perhaps most worrying has been a growing belief among investors – reinforced by recent academic studies – that HFT is increasing both market volumes and market volatility. A recent study by Ilia D. Dichev and Dexin Zhou of Emory University, and Kelly Huang of the University of Alabama) investigates the effect of high trading volume on observed stock volatility. Their main finding is that, controlling for other factors, there is a reliable and economically substantial positive relation between volume of trading and stock volatility. Their conclusion is that stock trading produces its own volatility above and beyond that based on fundamentals. Their paper concludes that while low to medium levels of trading can have market-wide benefits, there comes a point at which the huge amount of volume that high-frequency trading adds makes the markets so volatile that “only a small circle of traders” truly benefit.

A report released by The Bank for International Settlements in September 2011 also addressed some aspects of the issue and stated that high-frequency trading can in some circumstances accentuate market shocks.

According to a joint report of the SEC and CFTC, the flash crash of May 2010 when the Dow Jones Industrial Average dropped by 9% or approximately 1000 points, recovering those losses within minutes was due to a large mutual fund firm selling an unusually large number of E-Mini S&P 500 contracts and then high-frequency traders starting to aggressively sell, accelerating the effect of the mutual fund’s selling and contributing to the sharp price declines. The New York Times reported that as computerized high frequency traders who had hitherto been dominating the marketplace exited the stock market, the sudden withdrawal of liquidity “…caused shares of some prominent companies including Procter & Gamble and Accenture to trade down as low as a penny or as high as $100,000.”

Not surprisingly, a Toronto CFA Society website poll in October 2010 showed 72% of respondents agreeing that high-frequency trading and other forms of automated trading were leading to undesirably high levels of share price volatility.

To the extent that fundamental investors are discouraged from investing by heightened volatility or demand a higher return to compensate, real businesses and our real economy will pay a price imposed on them by HFTs.

Understandably there has been some grassroots reaction to what has been taking place in the stock markets. The following was a recent comment on an article entitled “The Markets are Mad” on the www.freakonomics.com website that triggered a large positive response:

…To the extent that fundamental investors are discouraged from investing by heightened volatility… our economy will pay a price…

“To me, the concept of a stock market is to actually invest in a company, allow the company to use your investment to produce a product and make money. The investor is rewarded with ongoing income through dividends, and over time, increased equity in the company through an increase in the value of the stock.”

“High frequency trading ignores all of the above. At the end of the day, what has been accomplished? The high frequency trader… starts the day with zero invested and ends the day with zero invested. All that has been done is manipulation of the market for personal gain. This is a worthless activity for the economy and should be stopped.”

Although it is not realistic or desirable to impose Draconian restrictions on the frequency of trading, it is hard to argue with the logic behind the sentiment expressed above And at a time when the spotlight is still on the financial industry for its role in bringing about the financial crisis, the effects of which continue to reverberate in the markets, streets and ranks of the unemployed globally, it would appear at the very least to be in the long-term self-interest of the financial industry (without getting into considerations of moral or legal responsibility) to restrain such disproportionate proliferation of high-frequency trading before governments act broadly to do so through regulation or taxation.

In Europe a government crack-down has already begun. Germany’s finance minister Wolfgang Schauble said on December 27, 2011 that he intends to press for the introduction of a financial transaction tax in the European Union next year. Mr. Schäuble said that if the tax cannot be introduced in the EU as a whole then it should commence at least in the Eurozone. The governments of France and Italy have agreed. Britain, a member of the EU but not of the Eurozone, has objected to the tax on the basis that it could harm London as a global financial centre but it has said it would support a global levy.

Mr. Schauble said that the German government wants the tax to slow down the pace of financial transactions and make some speculative business unprofitable. “The markets are.. too preoccupied with themselves these days rather than supporting the real economy,” he said. “We need to decelerate the pace of transactions.”

Under the plan, stock and bond trades would be taxed at the rate of 0.1 percent, with derivatives taxed at 0.01 percent. The tax would be imposed on all transactions in financial instruments between financial firms when at least one party to the trade is based in the EU.

…the technology of trading in the capital markets has overtaken regulators ability to monitor and regulate it…

Global securities regulators have published recommendations to help local regulators deal with the challenges posed by technological developments in the trading business, including algorithmic trading and high frequency trading.

The International Organization of Securities Commissions has published a report on regulatory issues raised by technological innovation. The report examines the most significant technological developments in recent years and their impact on market integrity and efficiency and sets out recommendations to help regulators in mitigating these effects and the risks posed to the financial system.

The recommendations set out high level guidance to address issues in two specific areas: trading venue operators and trading participants. They include recommendations that: regulators should require that trading venues provide fair, transparent and non-discriminatory access to their markets and to associated products and services; regulators should seek to ensure that trading venues have suitable controls (such as trading halts) to deal with volatile market conditions; and that the order flow of trading participants must be subject to appropriate controls, including automated pre-trade controls.

It also says that regulators should continue to assess the impact of technological advances on market integrity and efficiency, and that they should ensure that suitable measures are taken to mitigate negative outcomes that emerge, including risks to price formation or to the resiliency and stability of markets. Finally, it says that authorities should also monitor for novel forms of market abuse that may arise, and they should review their arrangements (including cross-border information sharing arrangements) and capabilities for continuously monitoring trading.

…to fail to properly regulate HFT is like allowing racing cars… on our highways without regard for the effects on essential daily traffic…

The recommendations were developed in response to a request from the G20 which has asked for further work by IOSCO by mid-2012 on the subject including development of recommendations for market surveillance.

In Canada financial institutions are among the most worried in the world about high-frequency trading, according to Seth Merrin, the founder and CEO of Liquidnet which operates a dark pool with 26 member institutions. He ascribes their high level of concern to the fact that Canadian markets are dominated by large institutions trading blocks of stock – the investors who are particularly vulnerable to the actions of HFTs.

In April 2011 The Canadian Securities Administrators (CSA) published for comment proposed National Instrument 23-103 Electronic Trading and Direct Electronic Access to Marketplaces. The proposed rule is designed to establish a regulatory framework for electronic trading in Canada. Under the proposal, marketplace participants will be required to establish, maintain and ensure compliance with appropriate controls, policies and procedures, in order to manage the risks associated with various methods of electronic trading, including direct electronic access, the use of algorithms and high frequency trading. Additionally, the proposed rule enhances the current requirements imposed on marketplaces to ensure they take an active role in managing the risks to fair and orderly trading posed by electronic trading. The proposed framework is an attempt to better manage the risks associated with electronic trading and to maintaining investor confidence in Canadian markets. Investors and market participants were encouraged to submit comments on the proposed rule by July 8, 2011.

Securities regulators throughout the world are currently ill equipped to monitor trading in today’s high-speed electronic world. They do not know until they inquire (and then it may take weeks to find out) the identity of customers behind trades and imprecise time records prevent them from accurately sequencing trades, so market manipulation often goes undetected.

In the US the SEC is proposing that a new computer system be created to monitor high-frequency trading. The new consolidated audit trail (CAT) would examine the transfer of US stocks in real time and identify the customers behind each trade as well as the time of each trade down to the millisecond. Only then could the SEC have complete, timely information to identify the causes of sudden swings in the stock market and to take action on regulatory violations.

The cost of the CAT is estimated at US$4 billion and since the US Congress has restricted funding for the SEC the issue of who will pay for it arises. Not surprisingly high-frequency traders and financial institutions generally are not keen to do so. However, failing to take action to protect the flow of vital capital in our economies from the side-effects of HFT could be likened to allowing racing cars to compete freely with each other on our highways without regard for the effects of their activity on essential daily traffic. The costs of regulatory failure now could prove vastly greater in the long run, a lesson that the history of uncontrolled financial innovation in the past decade has demonstrated all too painfully.