This is very close to the blog post I would write if I were going to write a critic of Flash Boys. I'd add the following:
IEXT has a few main selling points. 1 transparency, 2 simplicity, 3 looking out for "investors". The problem is that they aren't any of those things and it is obvious from a simple reading of their technical documents.
First transparency. Their marketing pitch is that as opposed to other dark pools, they are transparent about what is happening on their markets. Yet the first thing they say in their technical documents is "IEX will commence operation as a fully NON- ---DISPLAYED venue." What this means is that their order books is not available for outside audit. They are meeting the lowest hurdle to regulatory compliance by providing best bid/offer and no more. If you think that is ok, let me ask you. If I told you that there was 1 buyer willing to buy at 10.00 and 1 seller willing to sell at 11.00 what would you think the "real" price for that instrument is? What if I told you that there was 1 buyer at 10, 1 seller at 11, 100 million buyers at 9.99 and zero sellers at 11.01? If you think those markets should be priced the same way, IETX is for you. If not, realize that IETX would tell you those markets are the same thing.
As far as simplicity, Flash Boys and the IETX marketing materials make a big deal about all of the order types that provide advantages to market makers, yet they remain silent on those order types that are explicitly there for large institutional investors. Lewis rages against ALO orders yet doesn't even mention iceberg orders which the sole point of is to hide the amount of shares a participant wants to sell. Then he extols the virtue of IETX for only offering 3 order types (which has since risen to 4), yet doesn't explain that in those 3 order types are many variants that only benefit large investors. IETX has literally removed all of the order types that allow market makers to do their jobs, while leaving all the order types that allow large institutions to hide their order flow. One of the most controversial order types out there is an ISO order, which is literally a legislative exemption for large block buyers that allows them to circumvent regulation about the national ticker. IETX offers those sorts of orders, but you will be happy to know they don't support "esoteric" orders.
The most egregious problem in Flash Boys is Lewis' use of the term "investor". He uses this term to make his readers think he is talking about them, after all anyone who is not an HFT, who is long holding a stock is an "investor". Yet the practices he describes as predatory, are really simple market dynamics. If an "investor" wants to buy or sell an extremely large block of shares, that should move the market price. No other marketplace would think differently. If McDonalds buys all the cows in America, beef should cost more. We all understand that. But the giant hedge funds of the world hate this. They want to use their privileged information of demand to make money, and not change the prices of the things in demand. They used to be able to accomplish this because the markets were dumb. They aren't anymore. Price discovery is an extremely efficient (and cut throat business). IETX does its gosh darn Canadian best to undercut that. Literally every rule they have is an attempt to hide demand and undermine price discovery. It's possible that bad pricing is an overall societal good. I'm not smart enough to weigh in on that. I will ask, if they are looking out for the little guy though, why do they have a special order type that insures that big investors don't trade with them (min. qty orders)? They will respond that it is protection against predatory HFT. I ask you though, if the huge hedge funds of the world don't want to trade in small volume with the rest of the informed markets, at the lowest prices conceivable, who do they want to trade with? The answer is large dumb funds, who don't care about trading at inflated costs. I.E. pension funds.
As much as there are a lot of people in the world (many of them right here on Hacker News!) who don't care about the mechanics of how liquidity is provisioned in markets and consequently have kind of backwards views of HFT most of these people aren't, you know, trading stocks.
So here's my question: since the people that are trading do tend to understand these things (if for no other reason that they'll go bust if they don't), won't liquidity providers run away from IEX as fast as they can? How do Brad Katsuyama and his partners think they'll actually scare up enough liquidity to get maintain a functioning market?
I've actually been thinking a lot about this. So, every other dark pool in existence, has started with the same premise as Brad and his merry band. That is, "man our big investors are getting killed due to good pricing in the markets". So the first thing they do is come up with a variety of schemes to obfuscate the order flow of these big investors. Invisible order books, ice berg orders, preferential routing, min. qty. orders etc.
These schemes succumb to 1 of 2 problems
1) they are trivial to game by market makers. That is, the exchanges make their money on fees, the market makers make their money on finding inefficiencies in the markets, who is better positioned to find the places that a market protection scheme falls down? Of course it is the market makers. This is why you see stories of HFT "pinging" black pools, or creating routing tables, or whatever other wild information advantage that they can come up with. At the end of the day, big brokers/ibanks still get to eat if their schemes don't work, market makers don't.
2) If an exchange actually does find a way to limit market maker profits, they simply stop trading there. The liquidity dries up and the dark pool has to figure out some way to justify it's existence. Usually by relaxing their rules and allowing the tricks described in 1) or by negotiating a preferential deal with a couple of HFT firms that have some sort of price protection involved (which may be as shady as, only screw my clients I don't care about).
Brad, Lewis, and the IETX backers have figured out another strategy. Convince enough of the "dumb" money to demand that their trades be routed through IEX that they can create a self sustaining dark pool. They actually don't care about retail investors. If a couple of thousand of them call up and demand that their broker get on IETX that's fine. They've made the barrier to entry cheap enough that it doesn't cause a lot of overhead. But what they are really looking for is the uninformed institutional investor. The East Gary Fireman's Booster Union, needs to unload 65k of GOOG that some shady Morgan broker sold them. They think of IETX first. There sits Einhorn and his fund ready to hit it for all he's worth. No intelligent liquidity provider there to take any of his execution costs.
It's sort of brilliant and shows why I'm a dumb HFT market maker technologist instead of a billionaire (or millionaire in Brad's case) Wall Street type.
You think they'll succeed with this third strategy? Are there enough gullible East Gary Fireman's Booster Unions to make it work? It still seems a little far fetched to me. How much will regulations protect Mr. East Gary from shooting himself in the foot when the IEX price moves away from everyone else?
PS: You've probably already watched it, but if you haven't you should really watch the YouTube video on the iextrading.com homepage. It's...really something.
I don't know what their success criteria is. Do I think that they will sweep the markets and become the dominate exchange? Absolutely not. Do they have a chance to drive enough volume to the dark pool to keep their backers execution costs + investment less than their execution costs on other exchanges? Maybe. If they do that though they may have another class of problem. Other investors looking to back them. If broaden that pool it gets harder and harder to keep their execution costs down.
IEXT has a few main selling points. 1 transparency, 2 simplicity, 3 looking out for "investors". The problem is that they aren't any of those things and it is obvious from a simple reading of their technical documents.
First transparency. Their marketing pitch is that as opposed to other dark pools, they are transparent about what is happening on their markets. Yet the first thing they say in their technical documents is "IEX will commence operation as a fully NON- ---DISPLAYED venue." What this means is that their order books is not available for outside audit. They are meeting the lowest hurdle to regulatory compliance by providing best bid/offer and no more. If you think that is ok, let me ask you. If I told you that there was 1 buyer willing to buy at 10.00 and 1 seller willing to sell at 11.00 what would you think the "real" price for that instrument is? What if I told you that there was 1 buyer at 10, 1 seller at 11, 100 million buyers at 9.99 and zero sellers at 11.01? If you think those markets should be priced the same way, IETX is for you. If not, realize that IETX would tell you those markets are the same thing.
As far as simplicity, Flash Boys and the IETX marketing materials make a big deal about all of the order types that provide advantages to market makers, yet they remain silent on those order types that are explicitly there for large institutional investors. Lewis rages against ALO orders yet doesn't even mention iceberg orders which the sole point of is to hide the amount of shares a participant wants to sell. Then he extols the virtue of IETX for only offering 3 order types (which has since risen to 4), yet doesn't explain that in those 3 order types are many variants that only benefit large investors. IETX has literally removed all of the order types that allow market makers to do their jobs, while leaving all the order types that allow large institutions to hide their order flow. One of the most controversial order types out there is an ISO order, which is literally a legislative exemption for large block buyers that allows them to circumvent regulation about the national ticker. IETX offers those sorts of orders, but you will be happy to know they don't support "esoteric" orders.
The most egregious problem in Flash Boys is Lewis' use of the term "investor". He uses this term to make his readers think he is talking about them, after all anyone who is not an HFT, who is long holding a stock is an "investor". Yet the practices he describes as predatory, are really simple market dynamics. If an "investor" wants to buy or sell an extremely large block of shares, that should move the market price. No other marketplace would think differently. If McDonalds buys all the cows in America, beef should cost more. We all understand that. But the giant hedge funds of the world hate this. They want to use their privileged information of demand to make money, and not change the prices of the things in demand. They used to be able to accomplish this because the markets were dumb. They aren't anymore. Price discovery is an extremely efficient (and cut throat business). IETX does its gosh darn Canadian best to undercut that. Literally every rule they have is an attempt to hide demand and undermine price discovery. It's possible that bad pricing is an overall societal good. I'm not smart enough to weigh in on that. I will ask, if they are looking out for the little guy though, why do they have a special order type that insures that big investors don't trade with them (min. qty orders)? They will respond that it is protection against predatory HFT. I ask you though, if the huge hedge funds of the world don't want to trade in small volume with the rest of the informed markets, at the lowest prices conceivable, who do they want to trade with? The answer is large dumb funds, who don't care about trading at inflated costs. I.E. pension funds.