Advertiser Disclosure

What is High Frequency Trading and Who’s To Gain From HFT?

October 11, 2012
Investing
At NerdWallet, we adhere to strict standards of editorial integrity to help you make decisions with confidence. Some of the products we feature are from our partners. Here’s how we make money.
We adhere to strict standards of editorial integrity. Some of the products we feature are from our partners. Here’s how we make money.

Usually the traditional long-term investors wait until stock market crashes to come after the high-frequency traders with torches and pitchforks.  So it’s unusual to see stories attacking high-frequency trading (HFT) just as equity markets are trading near multi-year record highs.  Barry Randall’s High Frequency Trading Must Die story does just this.

The media is increasingly taking on high frequency traders: very short-term investors who are generally trying to exploit very small moves in a stock’s price by working very quickly. Their trades last for a few minutes, a few seconds and sometimes even less.  Does this work, and is this safe?  NerdWallet breaks down the basics of this approach to trading.

Who – or What  Are High-Frequency Traders?

Some high-frequency traders operate their programs manually – making a conscious decision to engage in every trade.

Other traders program what we call “black boxes.” These are just computer bots that enter buy and sell orders based on a specific trading strategy. There are investors and even institutions that use these high-frequency trading platforms almost exclusively. They program in a trading strategy, buy and sell criteria, a couple of limit orders – and then they just let their trading strategy run on autopilot.

Does HFT work?

It certainly works for the broker and dealers that serve the high-frequency traders. They earn a commission for every trade. For many brokerages, high-frequency trading represents the majority of their business. Some broker/dealers set up their businesses primarily to attract high-frequency and program traders.

There is something to be said for the ‘black box,’ in theory. According to the most recent DALBAR survey, investors only realize a fraction of the return each year that a passive investor in the S&P 500 index would make. And this is true survey after survey, year in and year out. This is because when traders get emotional – and most of them do – they tend to make bad decisions.

By entrusting the whole trading program to an emotionless computer program, the black box trader hopes to eliminate emotion as a factor in investing. He relies on the computer to add alpha – that is, return over and above that explainable by market moves as a whole.

Traders who don’t use a black box approach, on the other hand, are relying on their own ability to see trends and find and exploit anomalies in the market themselves, while resisting the temptation to give in to emotions.

Is HFT profitable in practice?

There are entire firms that do quite well for themselves with algorithmic trading – that is, trading by computer program alone. Some institutions also use high-frequency trading as a kind of radar or sonar to navigate the market with. By placing a constant stream of small trades, an institution can provide itself and the markets with a current price on any traded security. They can also sense if there is a large order out there. If a trade usually takes 20 minutes to execute, and then their “sonar” senses that all their offers are being snapped up in a heartbeat, that tells them there is a large order in the market somewhere. Someone is bullish on the security.

Then, by releasing a stream of market orders at increasing prices, they can potentially discern the other party’s maximum offering price. They can then release a larger trade with this information. So in this case, the high-frequency trading is not the main event, but serves to make the larger trades more profitable.

On the other hand, many smaller investors, with less programming skill or market insights, have been wiped out.  Meanwhile, high-frequency traders rack up transaction costs and taxes like crazy. In most cases, each trade costs a few dollars with a discount broker.

Other Expert Opinions: Is HFT Safe?

We turned to the experts to ask whether HFT needs more safeguards and regulation.

  • Jeffrey Chang, Principal Consultant at EXIQ, explains how HFT has improved efficiency and driven down pricing, but warns that algorithmic trading can pose risks:

“High-frequency trading needs to be differentiated from algorithmic trading a bit. HFT is market-making and providing liquidity — if you want to sell your stocks now instead of waiting for a buyer, a market maker saves you time and makes some money in the process by being the buyer of last resort.  High frequency trading has driven down costs. It’s the main reason most people will only pay a 1c bid/ask spread for a bunch of stock. Before computerized trading, they would’ve paid over 12x that. Additionally, as they push the systems to scale and force competition between markets, the per order and per share fees are driven down.  They’ve also made the market more efficient.

Algorithmic trading, on the other hand, is susceptible to a lot more issues – but when things fail, it’s the people who wrote and used the code that suffers. If there are regulations involved, they should be about making sure who suffers the liability of irresponsible trading in general, computer-assisted or not, and making that explicit to the everyday people and pensioners who don’t want their money invested in weird financial instruments.”

  • Joe Fox, Founder and CEO of the online broker Ditto Trade, explains why HFT has created an unfair playing field and how things can be leveled out:

“High frequency trading definitely needs to be reigned in, because it puts the individual investor on an unfair playing field. When the individual investor feels that the stock market is unfair, the stock market is broken.  However, it will be difficult to find an approach that works. It looks increasingly likely that Europe will institute a transaction tax that will limit the profit potential of such activity, but the result will likely be that those firms double-down on their trading on US markets, where I don’t believe there will ever be support for such a tax, further increasing the risk of more flash crashes. Other ideas involve giving regulators 100% transparency into trading activity. That’s fine for figuring out what’s happened in the past, but it’s hard to see that as preventing problems in the future.

Previously, exchanges’ number one priority was the investor. Now there number one priority is profits, even if in generating those profits it hurts investors. It makes no sense. It’s caused exchanges to lose site of what is important. I believe the best way to fix the problem is to give companies the ability to dictate where their shares are traded, preventing them from happening in “dark pools” or some of the other alternative exchanges where high frequency traders congregate. I think that kind of control would go a long way to reducing high frequency trading’s impact.  All of the above issues point to a new conundrum: investors need to be in the market, but if they can’t pay attention to what’s going on, then they shouldn’t be in the market. We started Ditto Trade to as a solution to this problem.”

  • Jay Hinton, Global Product Manager with the trading platform Mantara, builds the case for regulation of HFT as well as other types of trading:

“HFT, like all other trading, requires regulation if we are to maintain stability and fairness in the capital markets. The shifting of exchanges into for-profit companies has made this trickier, as it pits exchanges against the regulators.  As we see with other high tech industries, the regulators struggle to maintain ahead of the curve, and often are left trying to make stopgap measures. This issue gets murkier when you consider that there are really several kinds of HFT strategies, some of which address legitimate market inadequacies, and others that seek to game the system, and are essentially manipulating the market.

The answer here is twofold: first, the exchanges need to adjust the incentives so that HFT market makers are rewarded for contributing liquidity when volatility increases. Second, the regulators need to focus on issue such as tick size reform, and increasing surveillance for quote stuffing, and other strategies that are essentially market manipulation.”

  • Peter Pham, author of The Big Trade, explains why regulations alone won’t solve the problems posed by HFT:

“Regulation will not solve the issues surrounding HFT.  The funds should be free to deploy their capital at their discretion.  The solution should come from the exchanges themselves, creating breaks on the rate of order flow.  Smoothing out the rate of change of the bids and asks coming in with prevent a number of the abusive flash crashes and stop-loss raids taking place now.”

  • Merlin Rothfeld, instructor at the Online Trading Academy, explains how the quote stuffing tactics used by HTF firms can greatly disadvantage the average investor:

“High frequency is simply an exploitation of technology. It does offer firms a significant speed advantage by collocating their execution servers closer to the exchanges servers. This puts the average investor at a significant speed disadvantage. In itself, the above situation is bad, but that’s not the worst part. The next layer is quote stuffing by HFT firms. They intentionally send in massive numbers of orders attempting to jam up the data feeds, thus giving the computers which are collocated closer to the exchanges servers huge advantages while the average trader is stuck with data lags. The pinnacle of the problem is not HFT, but the exchanges that give firms who send in large blocks of shares preferential treatment with regards to order routing. It’s all algorithm based, and the larger your order the higher in the queue you go. So the average trader does not stand a chance. Clearly not a fair and level playing field, but what do you expect the exchanges (publically traded companies looking to make big profits for shareholders) to do?

Should it be regulated? Absolutely. You should have to pay for orders you send to the market.  If a HFT firm wants to send 100 thousand orders for a total of a million shares which they have no intention of actually executing, they would still have to pay 10k just for the effort. Regulators need to make the playing field fair.”