Michael Lewis hits back | Hacker News Reader

Michael Lewis hits back

salon.com

69 pointsrberger75 comments

>> “What is true,” says Lewis, “is that the sums of money involved — if you spread it across the market — are trivial, it’s a penny a trade. But it’s offensive as hell that rich people are stealing from middle-class people — even if it’s just a penny. There’s also the issue of what HFT does to the stability of the marketplace as a whole. And then there’s the question of the whole screwed-up model it creates for success in life, when the guys who get rich do so by skimming on the market, so all of a sudden the young people at the best schools want to go skimming in the market — like that’s a noble career path.”

>> “I don’t understand that argument. The little guy is the big guy. The little guy is not the day trader on eTrade; it’s all the money packed up in pension funds and college endowment funds and mutual funds. The savings of the country is in big institutions.”

Michael Lewis seems like a solid guy and I love how there's endless attempts to portray him as a sinister, self-interested actor.

Lewis may be a solid guy -- I don't know him, so I can't say.

I can say with a great deal of certainty that he hasn't read, or doesn't understand, any of the contemporary literature on market quality. If he did, he'd know that the efficient realization of market impact (you buy and the price goes up) is a property of effective markets. In the absence of actors to ensure this process of price discovery, all traders will pay unfair prices as small changes in supply and demand remain undigested.

I would also suggest that "young people at the best schools want to go skimming in the market" is reductio ad absurdum at best and ad hominem at worst, which sours my opinion of his ability to make a cogent argument without resorting to the Wall-Street-bogeyman trope. Then again, I suppose that's his job as a writer -- a story about the meek Canadian David battling the establishment equities market Goliath makes for a better read than a nuanced look at the role of various players in a complex ecosystem.

(Throwaway because my views are not necessarily those of my employer.)

If an exchange was front-running customer orders, that would be unethical and illegal. So why is it suddenly OK for exchanges to enable front-running by third parties and profit on their actions by charging insane amounts for colocation and data services?

And - please do answer this - if what we care about is market efficiency and liquidity and there are no problems with front-running in principle, what is gained through the current division of labour except legal obfuscation? Shouldn't it be more efficient to let exchanges screw over their customers directly?

I'm not sure I understand your argument. Arbitrageurs help with price-discovery and are useful agents, though much maligned. HFT, as Lewis describes it, is not arbitrage, it's front-running. This doesn't add to market quality, it's just information theft.

HFTs are making price impact happen faster than big players can stop it. That's improving efficiency. That's exactly what Lewis is complaining about.

It's not front running since the HFT is not your agent.

http://www.chrisstucchio.com/blog/2014/fervent_defense_of_fr...

I'm not sure i follow the argument that introduction of information by a middleman is adding to market efficiency. Your note expressly suggests liquidity be be witheld...when trading with people who have information.

As a result, market makers try to focus on providing liquidity to people with consumption preference while avoiding (or charging higher prics to) the informed.

Only to usurp the information to benefit the market maker (or other intermediary party...fronting or not) at the expense of the person introducing information. Or, are you trying to say that person is actually 'witholding' information and by 'appropriating' it, you are then responsible for 'introducing' it?

Big players are hiding info about their trades and changes to the demand curve, at least until after their trade finishes. Predatory traders are revealing this info.

>I love how there's endless attempts to portray him as a sinister, self-interested actor.

I've found most people I talk to do not think he's sinister at all. I (and may others) think he's one of the best non-fiction writers of our time. But for this particular book, he seemed in over his head, which made him susceptible to believing misinformation or at least misinterpreting the significance of some information he came across. I still think he's great, and I'm sure I will devour his future books with as pleasure as I did with Liar's Poker and Moneyball.

Maybe he's not a sinister person, but he tells narratives in which his side is unambiguously good. What happened in this book was lazy research and poor journalism. There have been plenty of factual critiques, which should set off flags around a book that purports to be an expose. But he spends more of the book moralizing and appealing to pathos than actually explaining how things are. As an immigrant, I thought it was cheap how often race was brought up in the book (examples: at some point one of the protagonists had a "this is how we do it in America" moment, whereas several HFT programmers were noted to be Russian, and as I recall Lewis saying, Russia helped them because in addition to forcing them to write clever, efficient code, they had learned to take advantage of the corrupt system. I mean, come on. The one guy from the SEC who spoke up? An indian quant. I don't know how Michael Lewis surmised that he was a quant, rather than just someone else working for the SEC. But there you have it).

Used to not bother much about HFT's but after someone referred to it as "legalised scalping", there is no doubt on its effect.

So if the one particular strategy Lewis describes in his book -- inter-exchange arbitrage that inadvertendly beats someone's split up order -- was stopped/prevented, is that the end of it? There are many, many strategies across a million different products that HFTs run that have nothing to do with trying to "front-run" someone's order, many of them market making a product that only trade on one exchange (so no inter-exchange arbitrage to front run anyone) on products that trade an order of magnitude more dollar volume than US equities, e.g. E-mini S&P futures, US or german bond futures complex, etc. Are all HFT strategies bad, or does Michael Lewis have a hard on for this single, particular strategy? My guess is that the overwhelming majority of profits of any HFT firm do not have anything to do with the particular strategy described in his book.

Well,

One could argue that giving someone latency for money is an inherent disservice to every other investor - whatever strategy you are following, you're getting opportunities before everyone else. One approach would be to have exactly one kind of connection to an exchange with a fixed delay level that everyone has to live with. Another approach is to have a small on every kind of transaction (Edit: other claim HFT pay a small fee and I am no expert but it's clear one could decrease HFT by raising transaction fees).

IEX does neither. They introduce artificial ~350 microsecond delays on their orders, among other things.

The book alleges that the HFT firms actually lobbied foreign countries to open new exchanges some distance apart from their existing ones, precisely to allow for inter-exchange arbitrage.

What do you mean inadvertently? No way is that by accident.

When one exchange's offer or bid is taken out (fully traded), there is no guarantee that the same offer or bid will continue to trade on all remaining exchanges. If you took out the remaining shares on all remaining exchanges every time any one exchange's offer or bid traded out, you will lose money -- I've tested this exact strategy before.

Has anyone read the criticism of the book in question? The Amazon review linked to in the article is very informative. I just finished the book and it read like an advertisement for IEX.

http://www.amazon.com/review/R3PJO6KJGRMWUE/ref%3Dcm_cr_pr_v...

The review is fantastic. From a quick skim it is spot on in every point - I can detect far fewer factual errors in the review than in the hype about Lewis's book.

Out of curiosity, since you're a strong defender of HFT here, have you read Lewis' book yet? You mention the hype, but I've found the hype and the substance of the book to be quite different. A lot of the hype on both sides seems to start and end around the first few chapters (one of which was excerpted publicly) and the "60 Minutes" piece.

When I finally finished the book, I didn't get the sense that his target was HFT alone. He heaps plenty of criticism on the bankers, the dark pools, the brokers selling order flow of their customers, the SEC's revolving door and apathy, and (implicitly, with exercise left to the reader) the conflicts of interest of the people selling the SEC the systems (MIDAS) that purport to keep the market fair.

It's the defensiveness of the HFTs and exchanges (like Direct Edge's O'Brien) that has me a little bewildered at why they feel persecuted in particular, when a better strategy might be acknowledging that there are some systemic flaws and look to fix them before another ill-thought out regulatory approach is put into place.

I don't see IEX as a pure white knight, but I do give them props for getting into the arena with a solution for one part of the problem instead of complaining bitterly, spreading FUD and launching ad hominem attacks on authors who are shining light on a whole slew of questionable practices as a lot of the people in the hype machine have done on the financial news channels since the book has been out.

I haven't read the book. That's why I'm not attacking the book, but only specific claims which people who read the book repeat. Most notably, confusing people that the "little guy" is David Einhorn, confusing everyone (including Mark Cuban) that HFT's have some ability to "jump in front" of other orders (they don't), and obfuscating the fact that HFT's pulling their orders is just a way of price discriminating against big guys.

spreading FUD and launching ad hominem attacks on authors

The only person spreading FUD is Lewis. I'm unaware of any ad-hominem attacks on him - all I see is "Lewis is wrong in X,Y,Z, did that idiot even talk to an HFT?" (Not an ad hominem Lewis being an idiot is orthogonal to the real argument, which is X,Y,Z.)

I agree, it appears to be well thought out. I have it bookmarked, and will return to it once I finish the book.

Folks interested in HFT might like this - HFT firm Virtu has a nearly flawless daily trading record over the past 3.4 years:

http://www.businessinsider.com/virtu-hft-only-one-losing-day...

Prospectus:

https://www.sec.gov/Archives/edgar/data/1592386/000104746914...

Looking at the S&P 500, it was positive for 55.8% of the trading days between 2010 and 2013:

http://www.crestmontresearch.com/docs/Stock-Yo-Yo.pdf

BofA had 10 trading day losses in 2013:

http://blogs.marketwatch.com/thetell/2014/02/28/goldmans-tra...

US equity HFT revenues are around $1.3 billion, making it a pretty small niche compared to wider market activity: "TABB Group estimates that US equity HFT revenues have declined from approximately $7.2 billion in 2009 to about $1.3 billion in 2014."

Law of large numbers. This is not surprising.

EDIT: Law of large numbers. Not central limit theorem. I blame alcohol.

Um, what? If anything, applying the theorem in this case would say that profits should be normally distributed, whereas it's fairly common for market makers to have perfect/near-perfect trading records.

Sorry, that was vague. My point was that market participants who make many trades will rapidly converge on their mean return, with little fluctuation. If you sample a normal distribution 10M times with a mean of 0.01%, you will very rarely see a negative sum of the samples.

The theorem says nothing of the sort. By your argument, if I flip a coin 10M times, where heads is 1 and tails is -1, and get a mean of 0, then I will very rarely see tails. You're confusing mean and variance.

Sorry. I said some pretty silly things.

See the new version of the posts. Hopefully it's a bit more coherent now.

Ok, I think your main idea was that since they've achieved consistently positive returns, then they're likely to keep it up. Even if that were true, the 800lb gorilla question is how they've managed to achieve such consistency in the first place.

They have slightly positive return per trade, and they make millions of trades per day. Their PnL is the sum of their trades -- the expected values grows like O(n), while the volatility grows like O(sqrt(n)), so for n = several million, the probability their PnL will be negative is extraordinarily low.

I will admit that Lewis has rustled my jimmies. But if he's interpreting this "punching in the balls" as a good guy punching the bad guys, he's mixed up. Electronic trading is the present, and it should be the future. In fact, every market that still exists as a network of people calling one-another or standing in a pit would be made many factors better by opening them up to the public and allowing trades to take place on electronic exchanges.

I get scared of what Lewis is saying in the same way it scares me when I hear of school districts banning the teaching of evolution. I'm not scared that Lewis is right. I'm scared that his misinformed screed will actually gain traction.

Electronic trading is the present, and it should be the future.

You know HFT isn't identical to any electronic exchange, right. I haven't evaluated in great depth whether Lewis is correct or not. But I'm pretty sure you can't defend HFT solely on the basis of "this is inevitable" (after all exchanges makes elaborate and special allowances for HFT, it's not a strategy you can sit at home and do). Any market is based on rules. If, for example, every transaction, no matter how quick or small, involved a small fee, I don't think HFT could exist as it does now.

Again, the point isn't that I've prove HFT should be banned but rather that like any other construct, it needs to justify it's existence rather than talk about inevitability - sort of like you defend patents if you want but you can't defend based on property rights.

Tangentially, the particular construct you describe ("transaction tax") has proved it is not worth it. Canada tried it out.

http://qed.econ.queensu.ca/pub/faculty/milne/322/IIROC_FeeCh...

It hurt the little guy, raised costs, and allowed the big guy to more easily hide their orders.

Every transaction does have a small fee. In US equities, there is an SEC fee of 22.1 dollars per million dollars sold, and for most (all?) futures exchanges, where HFT is just as prevalent, there is a fee both for both sides of a trade (unlike the maker/taker model in US equities).

I have not seen any reasonable way to decouple having a speed advantage from electronic trading. IEX is a large step backwards towards closed networks of non-public markets.

That was a rather transparent straw man argument you came up with there. Lewis is not against electronic trading. And he is certainly not trying to bring back the pits. He is against high frequency trading, or the ability of some insiders to obtain an advantage over the ordinary investor.

...obtain an advantage over the ordinary investor.

Big investor. If you are a retail investor, the HFTs don't know about you until after your 2 lot trade is finished.

> Big investor. If you are a retail investor, the HFTs don't know about you until after your 2 lot trade is finished.

Doesn't that obfuscate the fact that most retail investors interact with the market via various funds and intermediaries - which in turn collectively makes them a "big" investor?

Big investors = agents of little guys + rich people.

Little investors = all little guys.

Lewis is shilling for a group which is disproportionately not the little guy.

Further, skewing the market in favor of the big guy is just a way to ensure that the big guys can rip off the little guy. Once the little guy is forced to subsidize liquidity for the big guys (as Lewis wants), he might as well just pay a mutual fund the 50bps management fee rather than managing his own 401k.

Many big investors, e.g. mutual funds, trade on behalf of retail investors, so everyone is screwed. That was one of Lewis's core arguments in the book.

> the ability of some insiders to obtain an advantage over the ordinary investor.

What are you talking about? Who are these 'insiders'? Anyone is free to build a trading system that uses speed as an asset if they have the will and ability to. Speed of information used to be considered a noble thing for traders to invest in.

I don't think anyone has an issue with electronic trading. It's high frequency trading that has a lot of people (including myself) very worried.

What is your definition of HFT? Why exactly does it worry you?

You're the one misinformed here. He never said he's against electronic trading. In fact, IEX, the very firm he champions, is an electronic trading firm. He's against he practice of exchanges like BATS getting kickbacks from HFT firms for sending orders to them at inferior prices, screwing over institutional investors in the process.

The quotes from HFTs are NOT inferior to existing quotes. At worse, they are identical to the best quotes already available on that exchange.

Did you even read the book? The entire point is that some HFTs are given priority fills on orders that no one else has access to, even when the HFT quotes are worse, in return for kickbacks to the exchanges.

Everyone on that exchange has the same access to the same types of orders. HFTs only get priority if someone gives them priority. If these order types are so powerful, why aren't everyone using them?

Also, I will reiterate: HFTs NEVER get fills on quotes that are worse than any other participant on the same exchange. That violates RegNMS, and is straight up illegal.

> If these order types are so powerful, why aren't everyone using them?

An allegation made in Dark Pools was that the existence of the order types was initially deliberately concealed from most actors in the market. The claim was well substantiated, from what I could tell.

False - HFT firms have access to special order types. Example is Hide Not Slide, which is mentioned in the book and also here [0].

[0] http://online.wsj.com/news/articles/SB1000087239639044398920...

Why do you believe only HFTs have access to hide not slide? That order type is plainly listed on DirectEdge's website, e.g. https://www.directedge.com/Portals/0/01Trading/Trading%20Pro...

As doktrin said, they were originally concealed.

When the orders were added, there was a modification to the FIX specification for the exchange. For example: http://www.directedge.com/Portals/0/docs/Direct%20Edge%20Nex.... It is everyone who trades on that exchange's responsibility to stay up to date with what's available, and how it interacts with their trading strategy.

"concealed", in this context, just means "too lazy to carefully read the documentation".

[citation needed]

any exchange member may submit a hide-not-slide order. there is nothing secret about them.

They were originally concealed.

Tangential question :

What are the barriers to entry to HFT?

Or, put differently - how well funded would a venture have to be to realistically compete in this space?

Assuming talent (e.g. market knowledge, quantitative skills, ML competence) was not an issue - how realistic would it be to get access to the fiber & real estate necessary to compete with the very few market leaders?

Fees million at most. Real estate + connections are commoditized - tens of thousands a month for starting out.

> tens of thousands a month for starting out.

I couldn't find any reports of prices that low. According to one article, at least, Spread [1] charges ~$300k a month.[2]

[1] http://spreadnetworks.com/network-map/

[2] http://www.motherjones.com/politics/2013/02/high-frequency-t...

My NDA prevents me from discussing pricing. If you look up my resume and do some sleuthing, you can probably piece together the details of where to go for it.

spread has an interconnect between Chicago and New York but that is not generally useful for HFT stock trading since nearly every execution venue is in New Jersey.

The phrase "hits back" implies he somehow addressed the arguments against him, as opposed to merely saying "yay publicity".

Umm, all week he and Katsuyama have been in the media addressing the criticism, e.g. the CNBC debate [0].

[0] http://www.cnbc.com/id/101544772

Billy Beane still hasn't been to a World Series let alone won one. Michael Oher was illegally recruited by an Ole Miss booster who skirted NCAA rules by adopting him.

Michael Lewis is Malcolm Gladwell with normal hair. He does seem like a solid guy who really knows what he's talking about. That's how he sells books.

That's strange logic. Whatever happened or not with Billy Beane, he's absolutely right about the HFT-bank-broker strategies that he singles out in his book. Most of the critics of the book seem to not have read it, sadly..

I suppose my comment was silly and snarky. Clearly Michael Lewis is a gifted writer. But he's not a journalist, and he takes sides. He uses his enormous talent to demonize some and lionize others, when often the facts of the case could easily be viewed from another perspective. The problem isn't simply that Lewis's critics don't have the facts on their side; it's that they lack his eloquence and his podium.

To that end, referring to The Blind Side, the fact that Lewis was a schoolmate of Tuohy at an elite prep school in New Orleans, and the NCAA did investigate the Tuohy-Oher relationship, its not beyond the realm of possibility that Tuohy got away with something in adopting Oher.

As to Billy Beane, thanks to Michael Lewis he's touted as a visionary who literally changed the game of baseball, except the teams he put together have never been to the World Series, and over his GM career, the A's are slightly over .500. It's possible that he's not so revolutionary.

This is not to say the Lewis gets his facts wrong. He doesn't by and large. But non-fiction is like photography in that it is not a clinical representation, and the same event from different perspectives can be understood very differently.

I applaud Lewis' effort and maybe it will sink in this time, but this is not really news for a lot of people.

The story is that Wall Street big powerful trading firms have an inside edge against retail (middle class) customers.

It's not exactly new. This has been happening for decades.

They did it before computers with information, now they are doing it with fiber cable.

> this is not really news for a lot of people.

As a counterpoint, the fact that the NSA was monitoring electronic communication wasn't "news" either.

Shifts in opinion and popular consciousness is often times a matter of momentum & critical mass. This book will probably do more to stimulate popular interest in this issue than just about anything that's come before it.

As a side note, I'm currently reading Flash Boys and have previously read Dark Pools. So far, I would definitely consider the Lewis book to be the better of the two.

If they're using their edge to reduce the bid/ask spread of the retail customers to a penny instead of the 25 cents or more that it used to be, how is that hurting retail customers?

If that was all they were doing, that'd be fine. But read the book.

Some exchanges have 150 types of orders that are mostly undocumented and unknown to most regular players, created specifically so that HFT can not do what they publicly appear to be doing, or take incentives without providing liquidity, etc. The public price that everybody seems is outdated compared to the private prices that HFTers see, so they can risklessly front-run people because they already know if a price has dropped or rise, etc. All that stuff isn't just fast market-making.

Undocumented? http://cdn.batstrading.com/resources/membership/BATS_FIX_Spe...

The "public price that everyone else sees" is the same price as the HFTs see. These are available on the direct feeds from the various exchanges, and there's no discrimination against non-HFTs. Anyone who pays for it can get it. It's an equivalent advantage to having a Bloomberg - more data, faster.

It isn't. The SIP is much slower than what the HFTers can put together. Read the book, it's in there.

No one is arguing that all HFT firms are bad. But the ones who give kickbacks to exchanges to fulfill orders at worse prices aren't doing anybody a service.

Are you referring to decimalization? If so, that doesn't really have any direct relation with HFT. The NYSE famously used to list prices in fractions of 1/8th, and converted to the decimal system around 2001.

edit : coolio, I misunderstood the parent's comment.

No, he's referring to the lower bid/ask spread that HFTs cause.

http://qed.econ.queensu.ca/pub/faculty/milne/322/IIROC_FeeCh...

Related, how do you feel about sub-penny slicing as mentioned here: http://www.nanex.net/aqck2/3519.html

I wish there were more debates between HFT/exchange folks with Hunsader out there because I think it would be illuminating. Both sides seem to stand on their soapboxes and shout, but never truly debate the details.

That's a weird article. Nanex of all people is complaining that people are jumping ahead of HFT with negligible price improvements, making the hfts investment in low latency useless.

I mean they pretend its about retail, but the people who lose from subpenny trades are hfts. That's who investor b is, most likely - the fastest HFT.

Personally I believe we should re-decimalize - allow quotes at 0.01 cents. This will induce HFT to focus less on latency and more on price improvement.

Reply on news.ycombinator.com