Worstall @ the Weekend We had a distressing little incident here recently among the commentards over the meaning of the efficient markets hypothesis. So I'll explain what it is, using a current little story to prove that one version of it must be wrong as well.
That little efficient markets hypothesis (EMH) incident was an insider trading case concerning shares in Chipotle, a less-bad-than-usual purveyor of fast food to the colonials.
The misunderstanding was that one thought that this EMH referred to all markets: that it claimed that markets were, by definition, efficient in whatever it is that markets do. This isn't quite so: even a resolutely pro-free (or free-ish if you prefer) market person like myself does not make this claim.
Vast areas of even impeccably free market economics is dedicated to proving where and when simple unadorned markets will not get us to where we want to be: they're not efficient in that sense. Examples abound: uncosted externalities mean we might want a carbon tax or cap and trade. The public goods nature of invention means we'll need something like copyrights and patents even if not exactly the system we do have. That market labour incomes fall “too low” for us to be comfortable with for at least some people means there's always going to be a welfare state of some kind or another.
It's which sort and how generous it might be that is at issue.
No, what our EMH refers to is the idea that markets are efficient at calculating, from the information available, what prices in those markets should be. This is a much narrower claim, obviously. It also has some interesting corollaries: the one that brings the most opprobrium is that you shouldn't try to set prices in a market.
This is obviously annoying to those like Nicolas Maduro, president of Venezuela, who really does know what a bog roll should cost and is most upset to find out that none exist at that price he has insisted upon*. It also seriously pisses off those economic planners who absolutely know how your pension should be invested, for the signals given by markets rather often conflict with their desires as to how your money should be invested.
The one that's much more important to us mere mortals is that it tells us that we're just not going to beat the markets with our own investments. Low cost no-load index funds for us it is, then.
But, given that the EMH is this specific about information processing efficiency, the most obvious concerns relate to the financial markets. A stock or a bond is only worth what the market says it is worth because that market is efficient at processing the information about what that price should be.
This theory comes in three flavours: weak, semi-strong and strong. The weak one is obviously correct: no one still thinks that the bureaucrats at GOSPLAN have a better idea of what a company might be than the actual market. At least, we all hope to buggery that no one outside Pyongyang does. The semi-strong flavour tells us that all publicly available information is processed and encapsulated in those prices, while the strong flavour says that all information is.
You'll find just about every economist agreeing with the weak, most stop with the s-s and it's really only true market fundamentalist die-hards who insist upon the strong flavour. And, as far as we can tell, they're wrong.
Which brings us to this fascinating little story of trading in options upon the shares of that burrito joint, Chipotle.
This is a quite delightful little piece of work, in fact. A couple of people working inside a credit card processor started to check how much consumers were spending at various retail outlets. After all, they've got the stream of information (hmm, perhaps data is better, information being what was pulled from the data?) humming through the computers they're working on. So, see whose sales are going up and whose are going down. Buy options on the various companies (calls if rising sales, puts if falling) before the results are announced and just like the Underpants Gnomes, Profit!
In one sense there's not actually anything wrong with this. Most of that wholesale financial industry is really about trying to get those scraps of information that can be gleaned about what's going to happen. And it's actually one of the proofs that some version of the EMH must be true. We don't see people consistently beating the market even though there's hundreds of thousands of people trying to gain such information edges. So, if these information edges don't lead you to consistent outperformance then most of that information must already be in prices. QED.
There's certainly some pretty odd examples of such information gleaning: a favourite is the people who buy satellite photos of store parking lots to see who is busy and who isn't.
In another sense, of course, there is something wrong as our information processors had broken their contracts by trading on the information they were processing. And thus they've been nicked for insider trading.
There's just the one last piece of information we need to be able to prove that that strong version of the EMH isn't true:
That's amazing! These two, like, customer-support guys at Capital One were seemingly running an incredibly successful fundamental research-driven long/short equity hedge fund. A small fund, but still. The average equity hedge fund returned 25 percent - total, not annual - during that period. You sometimes see insider-trading cases where someone makes like a thousand-percent return in a few days by buying call options just before a merger. Every so often a network of tippers will yield multiple big scores like that. But to do hundreds of searches and trade multiple stocks over three years based entirely on raw consumer spending signals, and to make 1,819 per cent doing it, is just phenomenal. Even if the consumer spending signals were, you know, stolen.
That's also known as making a super profit. And the strong EMH says that you can't do that by collecting information because it's already in the stock prices anyway. But they did: therefore this information was not in stock prices and thus the strong EMH is wrong.
We're back down to the semi-strong version, then. Some set of information is not already included into what prices are in those financial markets. In this case the data is there, in the payment processors' computers, but it has not been extracted as information and also is unknown to the markets and thus doesn't influence prices. Note that we're still not saying that GOSPLAN would do a better job. Only that the extreme version of this idea seems not to be true.
It is a nice little proof of the contention as well. I know everyone insists that economics is not a science because it doesn't use the scientific method. Yet we are using it here. A logical and necessary prediction from the strong version of the EMH is that you can't make super profits by collecting information that isn't already included in prices, because all information is already included in prices. These guys did; therefore that strong version is wrong and we're back down to the semi-strong version, which insists that only all publicly known information is efficiently processed.
At which point, of course, we're all thinking of what we're going to do about this. El Reg is a decent enough place to put out a call for people who work as data analysts at credit and debit card companies. I guess? The Editor is preparing a corner of the lair underneath the volcano out at sea as a dealing room and I'm hunting around for jurisdictions with slightly more relaxed rules about insider trading. Knowing what people are spending their money on a little before everyone else does really is quite valuable you know.
Oh, and as to why GOSPLAN and the like works so badly? They're using data on what people were buying five years ago to plan what they will be permitted to purchase in 5 years time. That's several layers of more stupid than believing that markets might be efficient is. ®
* A small diversion here. There's absolutely nothing at all wrong with the aim of making that benighted country a little more equal. But the way they've tried to do it is simply pants. Don't fuck around with the markets and prices, instead if you want the poor to have more money to buy stuff then, well, give them more money to buy stuff with. The solution is welfare, not a planned economy.
from ffffff http://go.theregister.com/feed/www.theregister.co.uk/2015/02/01/insider_trading_on_chipotle_and_proving_the_efficient_markets_hypothesis_wrong/
via IFTTT
Entradas
0 comentarios:
Publicar un comentario