‘Flash Boys: A Wall Street Revolt’ by Michael Lewis (W. W. Norton & Company; March 31, 2014)
Table of Contents:
PART I: THE MYSTERY: ELECTRONIC FRONT-RUNNING
1. Brad Katsuyama
2. Brad Katsuyama’s Problem (and the Stock Market’s Problem)
3. Going Beneath the Surface: Katsuyama’s Detective Work Begins
- 3a. Background: How the U.S. Stock Market has Changed Over the Past Decade
- 3b. The Detective Work: Katsuyama’s Experiments
4. Thor, the Hammer!
PART II: SOLVING HIGH-FREQUENCY TRADING
5. The Detective Work Continues: Investigating HFT
6. HFT Unraveled
7. Market Turbulence, and the Flash Crash
8. The Hammer Gets Fixed
9. Thor’s Sales
PART III: COMBATING HFT (AND WALL STREET)
10. The Evolution of the Corruption on Wall Street: 2005-Present
- 10a. Corruption in the Public Exchanges
- 10b. Corruption in the Big Banks and Their Dark Pools
11. Katsuyama’s Trip to the SEC
12. A New and Improved Stock Exchange
13. An Ungameable Stock Exchange
PART IV: THE MOTHER OF ALL STOCK EXCHANGES: THE IEX
14. The IEX Takes Flight
15. IEX’s Data
16. Goldman Sachs Backs IEX
Over the past 20 years, and particularly in the past decade, the stock market has undergone some significant changes. The most visible change is that much of the action has now become computerized. For example, whereas stock markets used to consist of trading floors, where floor traders swapped stocks back and forth, we now have computer servers where sellers and buyers are connected automatically. Now, on the one hand, this automation has led to some substantial efficiencies, as once necessary financial intermediaries have now largely become obsolete (this has led to savings not only because the old intermediaries earned an honest commission for their dealings, but because their privileged position sometimes led to corruption).
It is not that the new stock market has done away with intermediaries entirely. Take brokers, for example. Brokers are still used by large investors to help them move large chunks of stock where the market may not be able to fill the order immediately. The brokers take some risk in this action, and provide liquidity in doing so—since they help move capital to its most useful location—and thus brokers still provide a very useful service.
While brokers have always existed, the new stock market has also added a new breed of intermediary. This new breed of intermediary is known as the high-frequency trader (HFT). The high-frequency trader operates on speed, relying on location and advanced communications technology to learn about the movement of the market before others, and uses this knowledge to make winning trades.
To give you an indication of how important high-frequency trading has become, consider that at least half of the trades now being made in the United States are coming from high frequency traders.
Those who defend high-frequency trading argue that these quick trades actually help move money through the stock market, and thus add liquidity to the system (the way brokers do); and that, therefore, high-frequency traders provide a valuable service.
However, just how high-frequency trading works has largely remained a mystery to anyone outside of the industry itself; and many have become concerned that at least some forms of high-frequency trading are not so much liquidity-contributors as a way of scalping money off of trades that would have happened anyway.
In Flash Boys: A Wall Street Revolt, Michael Lewis follows one man who made it his mission to find out what was going on at the heart of HFT. That man is one Brad Katsuyama, a broker from the sleepy Canadian bank RBC.
Katsuyama’s interest in the mystery began back in 2007, when he found that the trades he was trying to make from his desk at RBC were not being executed in the way they once had. In short, Katsuyama was being ripped off. And that’s not all. Katsuyama soon found that other brokers were also being ripped off—which meant that the big investment firms who were entrusting their money with the brokers were being ripped off too. And since the investment firms manage your money and mine, we were being ripped off as well!! This was big.
Katsuyama’s dogged persistence eventually led him (and a growing band of fellow mystery-solvers) to find that it was indeed the high-frequency traders who were ripping him (and everyone else) off (though the HFTs were not the only culprits involved). What’s more, Katsuyama’s team also discovered just how the HFTs were doing it. The long and the short of it is that the HFTs are just gaming the technology. And in a way that is not only ripping others off, but making the system more volatile, and prone to errors and disasters as well (witness the flash crash of May 6, 2010).
Rather than deciding to join the HFTs at the trough (which would have been easy enough to do), Katsuyama and his team decided to fix things. Specifically, the team decided to start their own stock exchange: a stock exchange (called the IEX) that was designed to be immune to advantages in technology, and hence fundamentally fair to all (it was either that or wait around for the SEC to do something—which may take forever).
Now, you would think that a stock exchange that is fundamentally fair to all would be a big hit. But then again, a whole heck of a lot of people have no interest in making things fair to all. Which side will win? The fate of the IEX (which opened in October of 2013) has yet to be determined…
Here is Michael Lewis on Charlie Rose explaining how he got the idea for his new book:
What follows is a full executive summary of Michael Lewis’s Flash Boys: A Wall Street Revolt.
In 2007, Brad Katsuyama was working as a stock broker for RBC, in New York (actually, Katsuyama was the head of RBCs brokerage team, which consisted of some 20 traders [loc. 356]).
Katsuyama’s main role was to mediate between large investment firms and the public exchanges (such as the NYSE and Nasdaq). Specifically, Katsuyama bought large chunks of stock (tens of thousands, hundreds of thousands, and even millions of shares at a time) from the large investment firms, and then turned around and sold them on the public exchanges (loc. 414-18).
The reason investment firms do not sell these large chunks of stock directly to the market themselves is because often the exchanges only carry demand for a fraction of the shares the investment firms are looking to sell. So, for example, an investment firm (let’s call them INV) might be looking to sell 3 million shares of Intel, while the exchanges show demand for only 1 million shares (at a best bid price of $3.72 per share). Offloading the full 3 million shares directly on the market would take INV time and effort, so they get people like Katsuyama to do it for them (for a price, of course).
It would be unclear, of course, just what price Katsuyama would be able to get for the 2 million shares he would have to sell over and above the 1 million the market had a demand for immediately; therefore, Katsuyama would be taking a risk by buying the lot en masse. As such, the price per share that Katsuyama would be willing to pay INV would have to be lower than the current market price of the share. So let’s say the price per share that INV and Katsuyama agreed on was $3.66 per share. Katsuyama would sell the first 1 million shares to the market for $3.72, and then “work artfully over the next few hours to unload the other 2 million shares” (loc. 418). So long as Katsuyama could sell the remaining 2 million shares for an average price of $3.63 per share, he would at least break even. (Obviously, any price higher than that would earn him a tidy profit, and any price lower than that would earn him a hurtful loss [hence the risk]).
By buying and selling shares that might otherwise have sat still, Katsuyama was helping move money through the stock market from where it was less valuable to where it was more so, and hence providing liquidity—which is, of course, valuable in any capitalist system (loc. 418).
At the moment, though, Katsuyama was having a problem. Specifically, when he went to sell the first million shares of Intel for $3.72, he found that he was not able to make the full sale. Rather, he might sell a few hundred thousand shares for $3.72, and then the price would collapse (loc. 429). This was not supposed to happen. Sure, the price of a share might legitimately move lower after a large sale of 1 million shares had occurred, but that was not what was happening here. Instead, the price of the share was collapsing before the full trade of 1 million shares had even gone through. The market was reacting to Katsuyama not after he had made his move, but in the middle of doing it (loc. 429-58).
This happened time and again, and both the size of the original order that went through, and the price collapse were largely unpredictable; thus, in effect, Katsuyama no longer had an accurate indication of what the market actually was (loc. 422). This was a major problem. As Lewis explains, Katsuyama “had been supplying liquidity to the market; now, whatever was happening on his screens was reducing his willingness to do it. Unable to judge market risks, he was less happy to take them” (loc. 421; see also loc. 440).
Katsuyama wasn’t the only broker at RBC running into this problem; indeed, all of them were (loc. 467). And this was costing RBC big money: “at the end of 2007 Brad [Katsuyama] conducted a study to compare what had happened on his trading books to what should have happened, or what used to happen, when the stock market as stated on his trading screens was the market he experienced. ‘The difference to us was tens of millions of dollars’ in losses plus fees, he said. ‘We were hemorrhaging money.’” (loc. 470). (Lewis refers to his characters by their first names throughout the book. I prefer to stick to the more formal approach, and will use last names only).
At first, Katsuyama thought that the problem lay in RBCs equipment (loc. 436, 463). However, it quickly became clear that this was not the case; for the more Katsuyama poked around, the more he found that many others (far removed from RBCs equipment) were experiencing the same problem. For example, Katsuyama went to a friend of his working for SAC Capital who was using technology provided by Goldman Sachs and Morgan Stanley, and found that his friend was running into exactly the same problem he was. As Lewis explains, “right away [Katsuyama] saw that, even though his friend was using technology given to him by Goldman Sachs and Morgan Stanley and the other big firms, he was experiencing exactly the same problem as RBC: The market on his screen was no longer the market. His friend would hit a button to buy or sell a stock and the market would move away from him. ‘When I see this guy trading and he was getting screwed—I now see that it isn’t just me. My frustration is the market’s frustration. And I was like, Whoa, this is serious.’ Brad’s problem wasn’t just Brad’s problem. What people saw when they looked at the U.S. stock market—the numbers on the screens of the professional traders, the ticker tape running across the bottom of the CNBC screen—was an illusion” (loc. 555).
Katsuyama recognized now that the market had to be set-up, and that the set-up was occurring through the technology, he just didn’t know how: “‘That’s when I realized the markets are rigged. And I knew it had to do with the technology. That the answer lay beneath the surface of the technology. I had absolutely no idea where. But that’s when the lightbulb went off that the only way I’m going to find out what’s going on is if I go beneath the surface’” (loc. 555).
Katsuyama was now determined to get to the bottom of the mystery, and since RBC was losing so much money, they were willing to back him in the effort. As Lewis explains, “Brad persuaded his superiors at the Royal Bank of Canada to conduct what amounted to a series of science experiments in the U.S. stock markets. For the next several months he and his team would trade stocks not to make money but to test theories—to try to answer his original question: Why was there a difference between the stock market displayed on his trading screens and the actual market? Why, when he went to buy 20,000 shares of Intel offered on his trading screens, did the market only sell him 2,000? To search for an answer, RBC agreed to let his team lose up to $10,000 a day” (loc. 636).
3a. Background: How the U.S. Stock Market has Changed Over the Past Decade
Before we launch into Katsuyama’s experiments, it is important to review just how the American stock market works nowadays, for it has in fact undergone some enormous changes in just the past number of years.
*For prospective buyers: To get a good indication of how this (and other) articles look before purchasing, I’ve made several of my past articles available for free. Each of my articles follows the same form and is similar in length (15-20 pages). The free articles are available here: Free Articles