Very efficient markets

As I was browsing through my bookmarks, I found this very old article from The Economist: Dismal science, dismal sentence. This, surreal at first glance, court case still comes back to me when anyone mentions the Efficient Markets Hypothesis.

In recent days two blogs that I like to read posted about the subject (The Money Illusion and Noahpinion). Both posts are well worth a read as they provide a few nice examples, which at face value are actually quite compelling.

But the problem with this theory is not so much its assumptions or implications but the way it has been tested. The most common tests include plotting performance of a large sample of portfolio managers against the performance of the stock index and concluding that – after accounting for management fees – the community does not beat the index. This obviously is very convenient for the guys at Vanguard etc but at the same time it is very illogical.

Let’s take an example of a market where we have only ten investors. All of them will be trading between themselves and if you think of it – there will be equal amount of longs and shorts in the market. Therefore, it is not difficult to conclude that these investors will on average have the same return as the market itself. When I think of it like that, it reminds me (in some aspects) of the Heisenberg’s Uncertainty principle. In physics, by measuring properties you change them. In finance, the measurement and the property in question are exactly the same thing. To cut the long story short, people trying to prove the Efficient Markets Hypothesis invariably arrive at the conclusion that the market cannot beat itself. Big deal!

This is not to say that the market is inefficient. I find it to be efficient most of the time (even if it stops me out). That said there are a lot of people who really take the EMH to the extreme, just like the judge in the article from the Economist. They conclude that there is no point giving anyone money to manage as the cost will definitely make it suboptimal to buying an index.

I don’t directly manage money but I know a lot of people who do. And the most impressive individuals, who happen to have pretty high Sharpe ratios and limited drawdowns ¬†all have similar characteristics. Firstly, they don’t start the day with “it’s going to be a risk-on day, let’s buy 200m AUD”. This Coffey-esque approach to investing is both ridiculous and pathetic. Instead, they usually have a whole bunch of uncorrelated trades and quite rigorous risk management. Directional, Hail-Mary trades are very seldom and fully controlled. These guys tend to beat the market.

Actually, scratch that last sentence. They don’t beat the market. They generate returns which are not depending on the market direction. Therefore, comparing them to returns achieved by the broad market is pointless.

So while the EMH is quite compelling and serves as a good baseline for analysis, testing it and drawing conclusions about performance of the aggregate industry is very much self-contradicting.