Are you a fan of Michael Lewis? Wondering about Flash Boys? The book itself would do just fine, but perhaps you are like “What’s this High Frequency thing?” To assist I’ve created a review of two books: Flash Boys by Michael Lewis and All About High Frequency Trading by Michael Durbin. Flash Boys is a terrific narrative of a fundamental change that has happened in the financial markets. And change is happening elsewhere. The ideas underpinning high frequency trading are now found in other industries. So I’d like anyone who’s interested in Flash Boys to consider All About High Frequency Trading as a primer. An equally easy to read book but more of an educational work.
Summary: Flash Boys is an eye opener. The type of book you can read in a week and realize when you’re done the world is different now. The very first chapter sets the tone as it describes the investment to connect NYC and Chicago, places connected for some time. The entrepreneurs spent millions of dollars to create the shortest path possible between NYC and Chicago. A straight line. They’re doing it so they can sell time, specifically milliseconds. And they have many buyers.
There’s some that say the financial market is gambling. Others who’d say if gambling is exciting, you’re doing it wrong. High Frequency traders believe the latter. All About High Frequency Trading is an excellent introduction of what’s involved with High Frequency Trading (HFT). The author, Michael Durbin, balances the line of over simplification and detail. His goal isn’t to be exhaustive. If he was a piano teacher this would be his Chopsticks. Someone else can worry about The Flight of Bumble Bee.
Why do I care? I’m in the financial services industry. I have a business, ops, and technology background, I’m not a trader. The work described in these books was a learning opportunity. And despite the increased attention High Frequency Trading is still for an exclusive few. And that is why this is so interesting, the whole model for market trading has changed in the course of ten years. And we can probably fit all the experts on a single jumbo jet.
Level Set: We typically think of financial markets as a place where you go to buy stocks and bonds and hold onto them for some period. Sometime ago, maybe the ’80s, we started to hear about people called day traders. The name itself conjures a sort of comparison to stunt men and crocodile keepers. It just seems risky. And in the 30 years since the day trader evolved. Their strategies did too. To help explain Durbin creates four archetypes. Together they make the market more efficient and competitive. They are:
- The investor – This is the group most of us belong and consider the core of the financial market. Investors typically buy and sell based on their portfolio and longer timescales. Often times they want dividend payments and other forms of compensation for owning the security.
- The Market-Maker – This group is present in the market to buy and sell. It’s said they are in the moving business, not storage. They work on a shorter timescale than the investor and they are compensated by making the spread i.e. buying low and selling high. The key to both of these books is the number of Market-Makers. As they compete they look for advantages, whether its speed, information, or payment models.
- The Arbitrageur – This group works to keep pricing in equilibrium. They are compensated by buying at one exchange and selling at another. They do this by identifying a point in time when there is a price discrepancy and racing to capitalize. By doing this they create supply and demand across exchanges to move the prices back to equilibrium.
- The Predictor – This group uses statistics and history to predict where prices are headed. They’ll analyze a security and find when it’s outside an expected price range. Statistical arbitrage. They are compensated by finding these differences and make money when the security moves back into alignment with the Predictor’s expectations for price.
Why High Frequency Trading? To be profitable in this space you have to be good at creating a feedback loop where you are constantly sampling markets and movements looking for patterns and trends. Rapidly learning. The pace of learning is astonishing. Smart firms identify a pattern or trend and create a strategy with it, test it, and see if it’s a candidate for use. If yes, they put it in with the rest of their strategies. Now they constantly assess. Markets change and usefulness does too. New patterns emerge and old ones die out. Each has it’s own survival of the fittest lifespan, a lifespan which could be a matter of seconds.
Durbin does a good job of describing a general architecture for a HFT system. It is:
- Thinkers – Take direction from humans and convert it into instructions for other components.
- Listeners – They sit as close as possible to the market data feed and take in market data to share with the other components. They are fast.
- Pricer – Constantly, in real time, calculating the price of all the securities. They consider both alpha (the risk adjusted return beyond general market performance) and security inventory (what they own).
- Traders – Connected straight to the exchanges to submit orders and quotes.
- Managers – Control the work of the other components. They monitor inventories and can act when high risk events occur.
This is their structure:
So how did we get here? Everyone has the image of guys on the floor of the NYSE yelling out numbers like “$64 bid for 500″ or “500 offered at $10.50.” It seems like pure chaos and to the outsider it is. However, these brokers do this day in and day out so they know who buys, who sells, and how to settle up.
But over the course of the early 2000s a couple of things changed. The first was subtle and it was the cost of computing got very cheap. The internet boom in the 90s went bust and all kinds of capacity was now looking for a use. Hardware, software, and networking was advancing at the speed of light and costs were so low.
The second change was the Securities and Exchange Commission (SEC) enacting a well intentioned regulation call National Market System aka Reg NMS. Reg NMS was put into law to ensure prices were always transparent. Price transparency helps markets to work more efficiently. To make this happen Reg NMS created the NBBO, the National Best Bid and Offer. This is what is published and brokers must guarantee this price, typically the midpoint between the bid and offer to their customers. What’s good about this idea is the average investor can establish an expectation about what they’re getting. It opens up trading to a larger pool of potential investors. The downside is speed. The NBBO is updated throughout the day from different exchanges and data feeds. Speed is where High Frequency Traders make their hey.
Based on these two changes the Market-Maker, Arbitrageur, and Predictor found opportunities. They use computing speed and fine tuned strategies to take advantage of information gaps and the rules for trading to make their profits. Durbin does an excellent job laying out the strategies. I particularly like pairs trading and tow the iceberg, but recommend everyone turn to page 49 and learn about the different strategies employed by the Investor, Market-Maker, Arbitrageur, and the Predictor. Below is a starter list described in the book:
Changing gears. Lets focus more on the narrative telling Michael Lewis employs with Flash Boys.
Flash Boys is Lewis’s version of Stephen King’s horror novel It. Both deal with memories of how things were, trauma, and what’s lurking behind the facade. Also like It, there is a shape shifter. It isn’t a clown, it’s the algorithm.
Flash Boys can be read in a matter of days. That is the joy of Michael Lewis, real clean and crisp writing. It flows from word to word and page to page (and the font is big, who doesn’t love big font?).
The book rises and falls on two men, one is featured more than the other, but both show the value of what’s at stake with a change from trading on the floor to trading in the boxes. The two men are Sergey Aleynikov and Brad Katsuyama.
Aleynikov, a russian immigrant, has since 2009 been the subject of arrests, trials, and legal proceedings which anyone from the general public would after hearing the reason would say “huh?”
Katsuyama, a Canadian citizen, on the other hand has acted as a detective of sorts for RBC and discovered an opaque world of trading. He reacted to what he found by offering an alternative trading approach and technology. What should have been a slam dunk turned out to be a luke warm reception. He realized the interests and motivations of the different players weren’t always aligned. As discussed above, the Market-Maker is in the market to make money from buying low and selling high. These players bring liquidity to the market (making spreads closer) but to do so they benefit by using speed to profit from incongruities in information. Also discussed above.
The irony Lewis highlights, the reason Aleynikov was jailed, is because he stood accused of taking computer code, code that Goldman Sach’s claims in the wrong hands could be used to “manipulate markets in unfair ways.” Raising the question, if the code, the algorithm, is that powerful, why is it any better in the hands of Goldman Sachs?
Back to speed. Lewis early on in Flash Boys writes about Spread Networks. This is a company which spent millions of dollars creating the shortest pass possible between Chicago and NY. They did this to cut the time it takes for data to travel between these cities by 13 milliseconds. To keep supply low, those that want speed are willing to pay millions of dollars for it. These fractions of a second are very valuable.
Now when you combine speed with algorithms you can exploit the market. On page 171 of the hard cover of Flash Boys Lewis tells the story of the Puzzle Masters. This is a team Katsuyama built to figure out the algorithms the High Frequency Traders were using. The goal of the group is reverse engineer the logic so they could foil it. What they found was exactly the strategies Durbin explains in greater detail in All About High Frequency Trading. The three Lewis writes about are:
- electronic front running – learn who is buying and selling, race to acquire the position, and act as a middle man making a little profit as the go between.
- rebate arbitrage – because of the fragmentation of exchanges there are rebates and kickbacks incenting traders to bring their business to one place or another. The High Frequency Trader may trade even all day, but make their money on rebates. This is the way an exchange may pay for liquidity.
- slow market arbitrage – with the advent of so many exchanges the prices they list do not stay synchronized. Price discovery is something High Frequency Traders can do very efficiently, so they act as the means to make sure exchanges stay in price equilibrium, making a profit on these small variances along the way.
To adjust to these strategies Lewis describes the rise of dark pools. Dark pools are set up typically by brokers or banks to allow their clients to trade without transparency. Because there are different players in the markets some have big blocks of stock. When they want to trade, move in and out of a position, due to the size of the move the price can be negatively influenced. Lots of supply and no new demand. Suppose I know a big pension management company wants to buy a million shares of Coca-Cola. If I’m fast I can go out and buy the million shares first and then sell them to the institution for more than I paid, making a profit on the knowledge. So dark pools allow the institutions to avoid this scenario and not tip their hand (artificially raising the price).
Katsuyama takes all his learnings and creates IEX – a dark pool built to negate particularly types of exploits. Market-Makers can provide liquidity on this exchange but speed will only help so much. And to cater to their customer IEX only allows certain order types. They are market, limit, midpoint peg and IEX Check (a type of fill or kill order).
The irony in the book is their success was largely due to Goldman Sachs the same firm prosecuting for stolen High Frequency Trading code.
I’ll end this review with an excerpt from Lewis. It typifies his opinion of a broken financial market. I don’t agree. The market does have vultures, scavengers, and foragers. And there may be more of them than you’d expect. But it’s important to remember why you’re there and who you are, an investor.
The arguments against the high-frequency traders hadn’t spread nearly so quickly – at any rate, Brad didn’t hear them from the SEC. A distinction cried out to be made, between “trading activity” and “liquidity.” A new trader could leap into a market and trade frantically inside it without adding anything of value to it. Imagine, for instance, that someone pass a rule, in the U.S. stock market as it is currently configured, that required every stock market trade to be front-run by a firm called Scalpers Inc. Under this rule, each time you went to buy 1,000 shares of Microsoft, Scalpers Inc. would be informed, whereupon it would set off to buy 1,000 shares of Microsoft offered in the market and, without taking any risk of owning the stock for even an instant, sell it to you at a higher price. Scalpers Inc. is prohibited from taking the slightest market risk; when it buys, it has the seller firmly in hand; when it sells, it has a buyer in hand; and at the end of every trading day, it will have no position at all in the stock market. Scalpers Inc. trades for the sole purpose of interfering with trading that would have happened without it. In buying from every seller and selling to every buyer, it winds up: a) doubling the trades in the marketplace and b) being exactly 50 percent of that booming volume. It adds nothing to the market but at the same time might be mistaken for the central player in that market.
This state of affairs, as it happens, resembles the United States stock market after the passage of Reg NMS. From 2006 to 2008, high-frequency traders share of total U.S. stock market trading doubled, from 26 percent to 52 percent – and it had never fallen below 50 percent since then. The total number of trades made in the stock market also spiked dramatically, from roughly 10 million per day in 2006 to just over 20 million per day in 2009.