Thursday, April 3, 2014

High Frequency Trading, the Grossman-Stiglitz Paradox and a Cheery Doomsday Analysis

Last year I forgot to post a link from Oxford's Austin Gerig. It has lain dormant in the link-vault until Bloomberg's Matt Levine provided the memory-jog. More after the jump.
From Bloomberg View, Apr. 2, 2014:

High-Frequency Trading May Be Too Efficient
If by some ghastly mistake I were put in charge of regulating the world's financial markets, my first order of business would be to find an organizing principle or slogan or whatever to direct my regulating. "Fairness" and "transparency" and "looking out for the little guy" seem to be popular slogans for the world's actually existing market regulators, but I think my slogan might be "the Grossman-Stiglitz paradox." This is the idea that if markets are efficient -- if market prices accurately reflect all the information in the world -- then there's no incentive for anyone to invest any time or money or effort into finding more information. And if no one goes looking for information, then there's no way for market prices to be accurate.

This is not, like, a huge thing to worry about in your everyday life, though it raises some potentially interesting questions about your index funds.1 But as a regulator I would seek ways for markets to be optimally inefficient -- that is, just inefficient enough to promote the information-gathering work that makes them efficient. And this might help organize my thoughts about insider trading, for instance, or about when activist investors should have to disclose their stakes in their target companies.2

Also it might influence how I thought about high-frequency trading. Cliff Asness, who knows a lot about market efficiency, ended his Wall Street Journal op-ed this morning like this:
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.
This is obviously right, and it is my least favorite thing about high-frequency trading. Main Street are the bad guys! "Wall Street" manages your pension funds and retirement funds and mutual funds and exchange-traded funds and index funds and really all of your good funds, and may or may not be losing out from high frequency trading. (Asness thinks not, but "can't be 100% sure.") "Main Street" is "retail investors who trade for themselves," that is, wealthy hobbyists. The way the world is supposed to work is that dumb wealthy hobbyists are supposed to lose money to subsidize the normal people who invest sensibly in mutual funds.3 It's not supposed to work in reverse, where my retirement fund subsidizes your stupid hobby.

I say this not only out of self-interest,4 but also because of my organizing principle. Retail investors are dumb. I mean, fine, whatever, you're not, you're a special snowflake who consistently beats the market, but as a category, retail investors are dumb. That's why wholesalers compete to trade with retail orders: Those orders are small and uninformed and unlikely to beat the market, so trading with them is a good way to avoid being picked off. You don't want to sell stock to David Einhorn, because if he's buying then the stock is going higher and you're on the wrong side of the trade. But you should sell all the stock you can to David Einhorn's dentist.

Big, smart institutional investors, meanwhile, get a mixed bag from high-frequency trading. On the one hand, they can generally trade lots of shares quickly, and quoted spreads (that is, explicit trading costs) are very tight. On the other hand, their trades do seem to move markets more than they used to, because high-speed traders are good at reacting quickly to big orders.5 You can characterize this as front-running or phantom liquidity, or you can characterize it as HFT firms reacting quickly to avoid being picked off by informed investors at the wrong price, but the result is the same: Prices very quickly reflect information, specifically the information that there are big informed buyers in the market.

That's good! That's good. It's good for markets to be efficient. It's good for prices to reflect information. We want that. Oxford finance-physicist Austin Gerig has written about the efficiency benefits of high frequency trading, which quickly synchronizes prices to reflect new market information. My Bloomberg View colleague Mark Buchanan summarizes Gerig's research:
Ok, so HFT helps synchronization. So what? Using a standard model of financial markets, Gerig goes on to show that price synchronization is broadly good for the market, as it makes prices more accurate and thereby reduces transaction costs. Specifically, improved price accuracy leads to cost reduction because liquidity providers – market makers who stand ready to buy or sell at fixed prices at any moment – have more confidence that they won't be “picked off” or taken advantage of by someone out there who has better information. An important implication of this is that HFT is good for the average investor, as it makes real prices more apparent, and reduces the advantages of other investors with great information gathering and analyzing resources.
On the other hand, those other investors with great information lose out. Gerig....MORE
It was at that point I asked "What else do we have from Gerig" and was confirmed in my dawning realization that I are a moron.
This was supposed to go on the blog a year ago.
From Geek:


New doomsday analysis says humans are doing better than expected
Nuclear-Bomb-Mushroom-Cloud
Predicting the end of the world is a gruesome and somewhat inexact process, but statisticians have been working on this question for years. The estimates until now have been less than encouraging, but a new analysis from physicist Austin Gerig and his colleagues at the University of Oxford gives us reason to hope.
The Doomsday Argument is an idea that holds you can predict the number of humans that will ever exist by the number that have existed thus far. This traditional analysis of humanity’s future works with the assertion that about 70 billion humans have lived on Earth. Probability suggest there is a 95% chance we are among the last 95% of humans that will ever exist. Doing the math, that means 1.4 trillion humans are likely to be born before we go extinct.

So where would that 1.4 trillion figure leave civilization? If we assume the world population tops out at 10 billion, and we have 80-year life spans, Homo sapiens has 10,000 more years. That might sound like a lot, but it’s an evolutionary blink of the eye. The Gerig analysis attempts to estimate the real danger of various existential threats like nuclear war, disease, and cosmic disasters.

This new paper is based on the position that long-lived civilizations are rare in the universe. If they weren’t, we would probably be living in one. Since we don’t see very old alien civilizations everywhere, the likelihood of survival must be low. Gerig’s approach is mostly dense statistical analysis, but the bottom line is that humanity has a better chance of survival than the Doomsday Argument would have predicted. If we as a species can be proactive, our chances for long term survival are as high as “a few percent.” Yay?...MORE
Better late than never.
Yay?