Misconceptions about algorithmic trading

28 March 2014


Today marks the end of my first month as Secretary General for FIA EPTA. In only a month, my eyes have opened to the many misconceptions I see reflected in the media about what our member firms actually do. Although I came from a principal trading company myself, I was frankly not very aware of the public dialogue around trading firms and market structure. Principal (or proprietary) trading firms have a reputation for being secretive. In my experience, it’s no one’s plan to be secretive; rather, these firms are filled with busy entrepreneurs who aren’t pushed by customers (because they don’t have any) or public shareholders (because in most cases they are privately held companies) to publicize and explain everything they do. So they just get on with it – because solving an intellectual software development issue, in practice, is more urgent than talking to the press. But it’s now clear how many people outside this world are curious about what we do, and in absence of us talking about ourselves, a lot of incorrect information goes around online, so it’s time to change.

I’d like to start by trying to clarify a subject rife with misunderstanding: high frequency trading (“HFT”). Not many people understand what HFT is, but have a vague, uneasy sense that it represents a  certain type of market participants, who are at best opaque and at worst up to no good.

If I can impart one message, it’s this: high frequency trading is nothing more than a technique for sending orders, like e-mail is for sending messages. It’s a tool used by banks, brokers, asset managers, or firms like ours, but … not all principal traders are high frequency traders! Likewise, not all high frequency traders are principal trading firms!

EPTA stands for European Principal Traders Association. The main characteristic uniting our members is that they trade exchange-traded financial markets as principals, not as agents for clients. That means they only trade using their own capital, for (and at) their own risk. This is not “Wall Street.” These are small groups of entrepreneurs spread out across the world: in the US – Austin, Chicago, Houston, Kansas City, L.A., San Francisco and more; in Europe – Amsterdam, Hamburg, London, Prague. Our firms are active in a variety of asset classes such as futures, equities, foreign exchange, and fixed income, and on a variety of exchanges, globally. This is key: they engage in automated, manual and hybrid methods of trade generation and execution encompassing a wide variety of strategies. 
Generally our member firms do all use algorithms, like so many modern businesses today. Amazon does, to generate recommendations of what you should buy; Facebook does, to rank your friends’ status updates; Google does, to prioritize the relevance of pages to your search topic; and UPS does, to help deliver packages in the most efficient way possible. An algorithm is nothing more than a step-by-step method for doing a job. 

Many different companies active in the financial market use algorithms to enter trading orders on electronic platforms. The algorithm executes pre-programmed trading instructions, making decisions based on variables including the timing, price, or quantity of the order. Algorithmic trading is widely used by investment banks, pension funds, mutual funds, and other buy-side (investor-driven) institutional traders, as well as sell-side traders, such as principal trading firms, market makers, and hedge funds.

HFT is a sub-set of algorithmic trading. It is a technique that ensures that information, orders and trades are processed quickly and efficiently. It’s characterized by low latency infrastructure and high messaging frequency. Any company that operates computerized algorithms, whether on the buy- or sell-side, might also elect to use HFT techniques for some or all of their trading.

HFT is a strategy-neutral technique. E-mail is a much faster medium than a hand-written letter, but it doesn’t imply anything about the content. Likewise, HFT can be used to facilitate many different strategies by different types of market participants. Most of the strategies conducive to the HFT medium are in fact quite simple arbitrages which could previously have been performed at lower frequency, but which now, due to evolving technology, have to be executed faster in order to be timely. For example, a basic arbitrage strategy evens out price discrepancies in a particular security trading simultaneously on different markets. This makes sure no investor pays more for a share than the markets think it’s worth.

Likewise, market makers use HFT to provide liquidity to the exchange. Market makers are in effect providing a public service to the exchange world. They provide liquidity, reduce transaction costs, and facilitate trade by standing ready to buy from anyone wanting to sell and sell to anyone wanting to buy. They don’t have an opinion on whether prices will go up or down. They simply quote both a buy and a sell price on a financial product and hope to make a profit on the difference between them (the ‘spread’).

Being a market maker is a highly complex profession. Market makers have to understand not only the products for which they make a market, but also their relationship to other financial products across many other markets. Competition and technology have changed the job of being a market maker. HFT techniques enable market makers to manage their risk by processing all publicly available information as quickly as possible and updating quotes regularly in real-time to avoid that their prices become stale. Market makers have to stick to risk controls laid down by the European Securities and Markets Authority (ESMA) and their own risk management departments. That might mean that from time to time, they have to exit the market, for example when they've hit limits that prevent them from re-entering bids, or because they have to build down positions that have created too much exposure or demanded too high a margin call for the company. But as their bread and butter is providing liquidity, they post quotes again as soon as it’s safe (for both the markets and the firms) to do so.

In general, HFT has been proven by numerous academic studies to be a benefit to investors, for example through the tightening of the bid/ask spread. Since the advent of HFT, price discovery has improved, and liquidity has increased. Historically, the founders of these firms are the same “locals” who fought against the monopolistic pricing on exchanges back in the day when you had to buy a seat to be able to participate. Nowadays, they’ve formed their own firms and celebrate the more democratic market made possible by technological innovation. Their role in the markets, today, is to even out inconsistencies in the prices of a product, no matter where it’s trading. That means investors can now more or less always be assured that the prices being quoted encompass all the latest information available. 

Of course, any tool can be misused by the willing. There have been instances of different types of financial companies abusing HFT techniques to facilitate manipulative strategies. Market abuse can take many forms; it can also be perpetrated by manual or low latency trading. Regardless of how market abuse is carried out, it should be punished, and EPTA firms stand behind regulatory and judicial authorities that pursue and enforce against those who damage the integrity of public markets through manipulative and disruptive trading practices. 

In future blogs, we’ll explore other aspects of algorithmic trading, with the aim of shedding light on the fair & competitive investment decisions trading firms make.

The views expressed in this blog post are the personal opinions of the author and do not necessarily reflect the official policies or positions of the FIA European Principal Traders Association or the Futures Industry Association.

  • Blog