Flash Boys — Michael Lewis

Amazon link

“Speed” is the buzzword in M. Lewis’ Flash Boys, which comes just four months after The Big Short. In it, he offers very much the same message: Wall St is making money from its incomprehensibility and your average Joe is being ripped off at the expense of the privileged few. For cynics who believe this to be the natural state of things, Flash Boys offers nothing new; however, for cynics interested in Wall St’s newest scheme, and everyone else, Lewis’ novel offers an accessible but roundabout* introduction to the world of high-frequency trading (HFT), dark pools, algos, and obscure trade types.

Anyone following financial news over the past decade would have already come across the fiercely divided HFT debate. On one side HFT-proponents argue that HF traders increase efficiency in the market by increasing liquidity and ‘making markets’, thus bringing stock prices closer to their real levels, on the other, opponents contend that HFT firms’ insistence on ‘co-location’ (meaning building servers physically close to the stock exchange (SE) – sometimes paying hundreds of thousands for a few meters of relative proximity) to increase their trade speed can only be explained by nefarious self-serving reasons

Lewis explains why the latter is more likely to be true. He outlines three main tactics that HF traders, armed with privileged market information that they pay SEs for (NASDAQ in 2011 earned 2/3 of its revenue from selling trade information to HFT firms), use: (1) electronic front-running: seeing an investor trying to buy something, beating him to it, then sell it to him at a higher price; (2) rebate arbitrage: using the complexity of the SE’s kickback system** to their advantage; and (3) slow market arbitrage: HF traders see demand for stock X on exchange A and uses their speed advantage to buy X on exchanges B, C, and D when somebody sells at a lower price, and sell them to the buyer at a now higher price before the market officially changes (and reflects the new price).

In light of this HFT-proponents insist that ordinary investors lose pennies at most while HFT firms fight multi-million dollar battles for information and speed ($250m was spent 2010-2012 building a microwave network that sped up information conveyance between Chicago and NYC by 0.002s from 0.0075s)… against each other. And that’s true. On the other hand, HFT firms are breaking bank: $1.2b in 2012. However, this is down from $5b in 2010. So my thoughts are kind of like: they’re taking pennies from millions of people, but legally (SEC—step up your game please), and it’s winding down… so no big deal? Well also bear in mind that HFT firms inject tons of volatility into the market and are single-handedly responsible for many flash crashes other than the more widely-known one in 2010, where 6% of stock market value was wiped out in microseconds.

My $0.02 is that HFT firms don’t add much to the market, and that’s why they don’t earn much. The whole business is contingent on information asymmetry, which is antithetical to the tenets of market efficiency. If HFT really contributed to the efficiency of the market, HFT firms would be rewarded for it accordingly. Instead, they add volatility, which they argue is liquidity… except that very same liquidity dries up when it’s needed most: in price spikes and collapses (in crashes, 85% of inventory is held by HFT firms). 

And don’t get me started with ‘market fragmentation leads to good competition’ arguments that HFT-apologists throw around. Economics teaches us that markets fail and externalities exist; so we need public goods and utilities to be provided for, not privatized and divided as has been the case with SEs since the SEC intervened in mid-2000. Subjecting a SE to for-profit considerations led to the downwards spiral of them selling trade information to HFT firms, which I would argue is where this whole debacle originates from.



*Lewis employs his usual story-telling narrative that follows ex-Royal Bank of Canada banker Brad Katsuyama as he discovers that he can’t make trades because HFT firms are always beating him to the punch (electronic front-running). Brad then goes on a crusade to find out what those firms are really doing, and as a result decides to create his own rival stock exchange, one that protects regular investors. This narrative, however, simply didn’t suit the technical explanation that HFT requires.

 **The system was originally designed to create liquidity in the market by rewarding ‘market-makers’, i.e. people who buy at bid price, and charging people who buy at offer price. (If stock X’s price is 100.5-101, the bid is 100.5, and the offer price is 101). Somebody who ‘crosses the spread’, or takes the offer price, is a ‘market-taker’. This ‘maker-taker’ system was employed by some SEs and turned on its head at others (BATS), to little obvious sense: why reward takers? BATS did this, supposedly, to attract trade volume, which increased the value of its trade information… which it sold to HFT firms.