This rather puzzles me, this solution that Dean Baker is offering to the high frequency trading problem as identified by Michael Lewis. Baker (we usually disagree it should be said, but on cordial terms) is arguing that we must do away with high frequency trading (HFT) altogether as a result of the way that some people are using it to, purportedly, front run the larger trades. This strikes me as being very much throwing out that baby with that mythical bathwater.
Yesterday I showed the figures to prove that HFT has significantly reduced trading costs for everyone in the markets:
We’ve come down nearly two orders of magnitude in that bid ask spread in the past 30 years. From around 0.20 percent to some 0.002 percent. That’s saved every investor huge amounts of money over all the trades they’ve done. It’s simply not true to say that HFT has been filling the pockets of the traders at the expense of investors.
This has reduced, to everyone who participates in the markets, the costs of participating in the markets. We usually regard this as a good thing, reducing the skim taken by the financial system through greater efficiency. I also detailed the fact that the new HFT companies are making considerably less (although perhaps more regularly) than the old market making companies used to when margins were wider. So we can see that the consumers are benefiting and also the HFT companies, both at the expense perhaps of the older market makers.
So, what does Baker suggest as a cure for whatever problems there are?
There are many complicated ways to try to address this problem, but there is one simple method that would virtually destroy the practice. A modest tax on financial transactions would make this sort of rapid trading unprofitable since it depends on extremely small margins. A bill proposed by Senator Tom Harkin and Representative Peter DeFazio would impose a 0.03 percent tax on all trades of stocks, bonds, and derivatives. This would quickly wipe out the high-frequency trading industry while having a trivial impact on normal investors.
What?
Let's accept, just for this moment of argumentation, that people really are front running stock orders. They're not doing it to the small investor because it's just not worth it. And anyway, the small investor, as Felix Salmon has noted, never goes anywhere near the main markets where the HFT is going on. Your and my orders are filled immediately at best market prices without any shenanigans at all. It's possible, and we'll accept it for right now, that front running makes some institutional orders fractionally more expensive than they would be without the front running.
But before we abolish HFT altogether by taxing it out of existence we need to work out whether that's actually a good deal on balance. And if it means that the spread balloons out from the current 0.002% back to the old, pre-HFT, 0.2% it's very difficult indeed to see that it will be. For no one at all is suggesting that the skim from front running is 0.198%, are they? Which is what it would need to be for the absence of HFT in toto to make sense.
HFT increases liquidity, increased liquidity reduces the spread and a reduced spread saves money for everyone in the market. It's possible that there is front running going on but we don't want to lose those savings that all enjoy just to get rid of the front running. So, if it is a problem that needs to be dealt with this isn't a good solution for it.
It's also possibly true (although I vehemently disagree) that there should be a financial transactions tax. But the abolition of front running isn't an argument for it either. And nor, actually, is deliberately making trading more expensive for everyone a good one either. As Baker himself will agree, he being a very good economist (even though we often disagree). For if trading spreads blow out to 0.2% again then that puts the incidence of the tax not on the financial system or markets, which is what everyone wants to tax, but upon the cost of raising capital and of saving through investment. That is, the incidence of the tax would be lower wages for the workers through lower capital investment and lower pensions for those saving in their various pension plans through higher transaction fees in the market. Which is a rather harder sell than let's tax the bankers and the stockbrokers if we're to be honest.
The FTT remains, at least to me, a solution looking for a problem. But even if it were a solution to a real world problem front running just isn't one of those that it should be used to solve.
