Michael Lewis' Flash Boys

[Praetorian]

I am currently reading Michael Lewis' new book and find the idea behind RBC's Thor quite interesting. Nevertheless, wouldn't it be much easier to simply split a block trade into a large number of orders and place them randomised by time directly to an exchange? The HFT front running described by him seems only to work if you place a large order to a "spray" router that has different latencies to each of the connected exchanges.

[athletico]

Fun little debate on CNBC today between the IEX CEO, Michael Lewis and the BATS CEO:



http://www.cnbc.com/id/101544772

[jslade]

I wonder at Lewis as journalist. I have always enjoyed his prose and skill as a storyteller, and he is actually a friend of my extended family, but a lot of the crap he's written has just plain been factually wrong.

[Nonius]

He's a dickhead.

[starchild]

I've just finished the book. My conclusion? Michael Lewis is a tabloid-style journalist. His description of HFT is very shallow and biased. He didn't even bother to talk to actual HFTs and exchange executives.



Rigged markets? Extraordinary claims require extraordinary evidence. The book offers none.



Eliminating middle man/intermediaries? Michael Lewis needs to understand that, absent intermediaries, [intended] buy volume is never exactly equal to the [intended] sell volume in any time frame in any market. If you want to trade now (or within a specified time window), you do need intermediaries to absorb the excess liquidity. That's what HFT is for, that's what stat arb is for. That's even what Warren Buffett does (e.g., in the middle of financial crisis). If you eliminate that function, markets will disappear.



His work is a disgrace to some of the smartest people I know who make markets in this extremely competitive marketplace with an as tight spread as possible.



Sadly this is the 4th Lewis book I've read, and it will probably be the last one.

[starchild]

http://online.wsj.com/news/articles/SB10001424052702303978304579475102237652362



High-Frequency Hyperbole

Beware of critics who are 'talking their book' about trading that lowers costs.



By CLIFFORD S. ASNESS And MICHAEL MENDELSON

April 1, 2014 7:23 p.m. ET



A few nights ago, CBS's "60 Minutes" provided a forum for author Michael Lewis to announce that Wall Street is "rigged" and for the sponsors of a new trading venue called IEX to promise to unrig it. The focus of the TV segment was high-frequency trading, or HFT, an innovation now over 20 years old.



The stock market isn't rigged and IEX hasn't yet generated a lot of interest. In our profession, what we saw on "60 Minutes" is called "talking your book"—in Mr. Lewis's case, literally.



The onslaught against high-frequency trading seems to have started about five years ago when a blogger made a wildly exaggerated claim about one firm's HFT profits. Nowadays after any notable market event, and again last Sunday for no reason other than a book launch, the world gets bombarded with arcane details and hyperbolic assertions about HFT strategies. If you find the discussion overwhelming, we have some good news: The debate can be understood without knowing how equity orders are routed, matched or canceled.



Few professionals completely understand the details of market microstructure. Rather, when someone has a strong opinion about the subject, it's likely to be what they want you to believe, not what they know.





Getty Images

Our firm, AQR Capital Management, is an institutional investor, primarily managing long-term investment strategies. We do not engage in high-frequency trading strategies. Here is where our interest lies: What is good for us is lower trading costs because it translates into better investment performance and happier clients, which makes our business slightly more valuable.



How do we feel about high-frequency trading? We think it helps us. It seems to have reduced our costs and may enable us to manage more investment dollars. We can't be 100% sure. Maybe something other than HFT is responsible for the reduction in costs we've seen since HFT has risen to prominence, like maybe even our own efforts to improve. But we devote a lot of effort to understanding our trading costs, and our opinion, derived through quantitative and qualitative analysis, is that on the whole high-frequency traders have lowered costs.



Much of what HFTs do is "make markets"—that is, be willing to buy or sell stock anytime for the cost of a fraction of the bid-offer spread. They make money selling at the offer and buying at the bid more often than they have to do it the other way around. That is, they do it the same way that market makers have done it since they were making markets in Pompeii before Mount Vesuvius halted trading one day. High-frequency traders tend to do it best because their computers are much cheaper than expensive Wall Street traders, and competition forces them to pass most of the savings on to us investors. That also explains why many old-school Wall Street traders hate them.



One of the biggest headline-grabbing worries about HFTs is how fast the trades are conducted. The speed sounds unnecessary, dangerous and possibly nefarious—"These guys care about the speed of light!" For the most part, though, HFTs don't need that super speed to get ahead of the little guy or even institutional traders, but to get ahead of other HFTs. Some of the loudest complaints about high-frequency trading come from the slower traders who used to win the races.



While we like HFTs on balance for reducing our clients' trading costs, some may push the envelope at times. Some of them may negotiate advantages that might be bad for markets. Worse, these arrangements tend to be little understood by the broader range of market participants. A little more transparency would be good here, and the market venues that have been offering these deals have been moving in that direction. They should move faster.



But these concerns are occupying too much attention. The biggest concern we have with modern markets is their complexity and the associated operational risks. The market structure that enables the HFTs and provides us with their benefits may also be one that risks technological calamity.



The good news has been that regulators began to focus on this potential problem last year. Unfortunately, the recent fusillade of hyperbole about HFT practices threatens to derail this effort and refocus attention where the problem isn't. Real work is necessary to improve and safeguard a complex and still reasonably new system. We shouldn't get ourselves dragged into a hyped-up war over a matter that doesn't affect investors very much—and where, to the degree that it does, we'd argue that the effect is easily a net positive.



So why are so many people so loudly certain about the problems of high-frequency trading? Again, look to interests. Making mountains out of molehills sells more books than a study of molehills. But some traditional asset managers are also HFT critics. These managers are institutional investors like us but with different investment strategies and trading methods.



Rather than embracing electronic markets, these managers have stuck with their old methods. They think HFT costs them money. Often when they try to trade large orders quickly, they find the trades more difficult to execute in a market that has gravitated toward more frequent trades in smaller sizes, and that the price moves away from them faster now.



We doubt that these old-school managers were truly better off in the pre-HFT world, but it's hard to prove either way. And if they're right, it may be only because HFTs have made the markets more efficient, eliminating some of the managers' edge.



Well, sorry, but prices responding quickly—and traders not being able to buy or sell a ton without the market moving—is what is supposed to happen in a well-functioning market. It happens to us too. It may be that in the old days these managers were able to take advantage of whomever was on the other side of their trade, and that nowadays they find it far more difficult to gain that advantage. A more efficient market shouldn't be mistaken for an unfair one.



These big, traditional investment managers represent a business opportunity to anyone who can offer them new market venues, like IEX, that might conceivably avoid the perceived ill effects of high-frequency trading. We wish them well in that effort, and if they succeed these new exchanges and their clients will benefit. But let's allow the issue to be decided by open competition, not by politics, demagoguery and rules born of crony capitalism.



Our bet is that high-frequency trading comes out on top as it offers more investors better execution. But we have zero problem being proven wrong by the marketplace.



How HFT has changed the allocation of the pie between various market professionals is hard to say. But there has been one unambiguous winner, the retail investors who trade for themselves. Their small orders are a perfect match for today's narrow bid-offer spread, small average-trade-size market. For the first time in history, Main Street might have it rigged against Wall Street.



Mr. Asness is managing and founding principal of AQR Capital Management, where Mr. Mendelson is a principal and portfolio manager. Aaron Brown, chief risk officer at the firm, also contributed to this op-ed.

[jslade]

IEX guy took $10,000 a day slippage on cement-head trades, using his wacky band of crazy misfits to figure out that people can ... cancel their limit orders. In 2009. I recall reading exchange documents in 2009 using the miraculous powers of the internets. Not that you needed to; knowing what a limit order was would have sufficed. Either Lewis is misunderstanding things, or Katsuyama is/was dumb as a stump.

[radikal]

Pretty solid response.



On the floor, if paper comes in and asks locals:



'what's there in 1750 dec put at 20?'



A bunch of locals might show him 100 at 20 but as paper lifts locals, the others will cancel. If he tries to "sweep the entire pit", locals will explain to paper that he should **** himself. Market makers provide liquidity that's always conditional on order flows, it's weird to me that this is considered "wrong". I obviously am less psyched to sell you 20s if I think you're trying to buy 1000s of them; that's how all markets work regardless of speed.



The buy-side is supposed to pay for liquidity. The fact that 1000s of shares of MSFT are offered on various ECNs does not reflect that MMs want to provide that total amount of liquidity at that price; they're just competing to get the trade and as a price trades, they will update their price accordingly.



And no, providing liquidity does not mean "accepting the risk"; all MMs risk shift. I sell you the 1750 puts only while I think I have somewhere to go with the risk. You aren't paying me to accept general downside risk, you're paying me to accept that SPECIFIC risk regardless of how I spread it off.



There's a lot of shenanigans in the world, but I worry about the constant media onslaught against MM by people who have basically no understanding of market fundamentals. This is really not me talking my own book as I'm part of the HFT crowd that doesn't provide a ton of instantaneous liquidity. (But I do greatly appreciate those who do -- I'm largely of the opinion that the mass MM side of HFT is increasingly expensive and diminishingly profitable)

[AndyM]

Problem is: setting up HFT vs guys with thick necks presents a false dichotomy. 'You don't like HFT; do you know what it was like back in the day?!' It's a handy rhetorical device, but doesn't advance the discussion.

[filthy]

the actual presence of HFT doesn't bother me. It IS better than the days of the specialist and hearing them complain is akin to the mafia whining about police brutality.



and as someone who was a market maker i can't claim to be totally innocent in the intermediation world. But there were a few differences. First we had obligations to make markets. Second we were expected to trade on our markets. Third we couldn't play the order type arbitrage games. we had the same orders as everyone else.



but the thing that really annoys me is payment for order flow. your broker should be working for you with no conflicts of interest. also, dark pools are by definition really just a place to give bad fills to customers.