New book from Michael Lewis on High Frequency Trading

None of it is illegal. What Lewis was talking about is legal frontrunning.

A few things stuck out. This has been all over CNBC this morning. It's funny, because it is nothing new. Everyone has known about the little deal between the exchanges and anyone who wants to pay extra for a few milliseconds. The exchanges make a ton of money from it. Getting them not to take that revenue is going to be tough. It's why what those guys at IEX is doing is so cool. It's a really good sale when you sell an exchange that guarantees a level playing field -- no pay to play to get a jump on everyone else. It's why IEX has gotten people like David Einhorn behind them.

As for little investors getting screwed, indirectly, I suppose they do. The penny an execution might -- and it is a might -- cost you on one of your handful of trades a year is not a big deal. But if you own mutual funds, for example, for your retirement, it is indirectly costing you more than that, because this drives up the cost of execution. Whenever someone is making billions of dollars, you have a zero sum game, in which it is costing someone billions.

HFTs are interesting for a lot of other reasons besides the frontrunning via technology / fiber optics that Lewis was talking about. I wonder if his book covers how big a systematic risk they have the potential to be? They occastionaly move markets. Whenever you have a flash crash, or news that moves markets, they have created added volatility (to scary levels) because their computers are driving the markets. When they all start selling at once, they can create havoc, even before the circuit breakers kick in. They have the potential to bring down markets. The HFT firms are hiring guys with PhDs in mathematics and they are applying what they do to algorithms that can game markets in ways beyond the frontrunning. I have seen it in the unusual activity you see in some of the futures and derivatives markets in the middle of the night. You'll get these quick flash orders for huge sums that come and go quickly trying to move the bid and ask price -- in a sleepy market. I have no idea what is behind it usually, but I know it is the product of sophisticated computer programs that are up to all kinds of things that I can't comprehend to move prices for someone's advantage. And I know there are a handful of firms out there making boatloads of nearly no-risk money from it and what they are doing is probably a systematic risk to the financial markets.

Zero Hedge (a contrarian website) has been all over HFT firms for years. They are laughing today about how everyone woke up last night to what they have been writing about for years. If you want to know about HFT firms, read just this:

http://www.zerohedge.com/news/2014-03-10/holy-grail-trading-has-been-found-hft-firm-reveals-1-losing-trading-day-1238-days-tr

You don't have to understand what they are doing. But this HFT firm went public, and in its offering, it said that as a trading firm it had lost money in just one day out of four years (1238 days!). Reason should tell anyone that that is an impossible feat unless you have an advantage. The question is what is their advantage, and how can others neutralize it? I think exchanges that don't allow pay to play are an easy answer. You won't be able to get rid of technological advantages, but at the least you shouldn't be SELLING the ability to have those advantages. It would be great if the NYSE stepped up the plate on this and showed some balls, but they make so much money from selling access that they probably won't step out in front of this, even though it would probably benefit them in the long run.

 

One other thing. ... The notion that the stock market is "rigged" is the wrong way to think about this, in my opinion. What Lewis was talking about was about the cost of executions -- what the front running is effecting. And in that regard, the markets are less "rigged" than they used to be. People can trade stocks today for $5 a side, for example. Relatively recently, you had to call your broker, didn't get almost instantaneous executions (often costing you on price), and it cost way more than $5 or $10 a trade. So in the aggregate, people are better off today because of technology -- it is saving them money, even if technology is giving someone a else big advantage it is making money off of.

Look at it this way. If you want to trade in those markets, you are in effect stepping into a major league stadium. And if you are going to try to play your team against the New York Yankees, you know that the Yankees spend more and have better players -- better resources to win with.

 

And in the process dry up liquidity in markets in ways you aren't anticipating, creating all kinds of dangerous volatility that can wipe out trading desks in seconds, and creates price inefficiencies (because of the illiquidity your tax has created) that screws people a gazillion times worse than you think they are getting screwed already.

The natural response from people is always, "regulate it." You can not regulate a market to make everyone the same -- or else you don't have an actual market. Sometimes people have better info than the corresponding buyer or seller. Sometimes someone has an advantage because they have paid for better technology -- as is the case of HFTs.

Let the market sort itself out -- for example, in that 60 Minutes piece, it focused on Brad Katsuyama and what he is doing with IEX. He is getting a lot of support from some big institutional investors whose interests it is in not to be front run.

By all accounts this is already working itself out on its own. In that link I posted about Virtu going public, their offering showed that they made less and less money in their most successful scalping "bucket" each year. And it was making more money in its higher net income buckets with each year. The HFT model, though, is to make money on the low end scalped trades, which shows that the strategy is doing worse each year. Virtu has been making more money by taking market share from less profitably HFT firms, not by scalping pennies on more and more trades. That is a business that is slowly leaking air and it relies on volumes, which are there in an overheated stock market like the one we have had the last two years, but will not be a great business when things simmer down. Zero Hedge pointed out that the gigaHFT firm had cashed out because they realized they are being countered successfully -- it has been getting harder to make money scalping.

The explanation is that it wasn't only Brad Katsuyama who knew the game, but others have figured out how to counter what they do and they are taking away their advantage -- without hamhanded regulations like your tax, which will create more harm than good.

There have always been scalpers, discounts to certain buyers on exchanges that others don't get, and front running. Even before technology. People are "screwed" much less than they used to be. Partially because the markets have worked it out themselves -- people want better executions when technology makes it possible and that has created competition and transparency. It is why today, anyone with a small brokerage account you can see Level II quotes, multiple exchanges, get split second executions and pay $5 a trade.

Transparency is good, too, which is why that 60 Minutes piece was nice. But the large institutional traders that are the actual victims of HFT scalping already knew about this, of course, and already have been countering it on their own -- without asking regulators to try to somehow tax away the HFT advantage or regulate it. They know that that is a formula for disaster.

Only the technology is new, really, not the story. You can't regulate away advantages that people garner when they are bigger than their competition. Before fiber optics and electronic exchanges, firms got discounts for volume, for example. If you promised to route your order flow to an exchange and you were big, you could use your volume to negotiate huge discounts. But that discount one guy got was passed along to everyone else in higher prices. Were people being screwed or were the markets rigged because of that? It's not different. Only the technology is.