And they’re nearly always disappointed when that historical pattern fails to materialize in the future.

Data: It is very hard to beat the markets.

Because it is very hard to beat the markets, it is “almost true” that it is impossible to beat the markets, as proponents of the EMH might assert. But the fact that this implication of the model is “almost” (i.e., not) true suggests that the model fails to capture some vital element of reality.

I often see the assertion that markets are impossible to beat being conflated with the EMH, but they are not the same thing. EMH has always struck me as being redundant: social phenomena in general are so hard to predict that it is not surprising that markets are hard to predict. A market-specific model is not required to explain this observation.

You miss the point. True, there is no specific asset pricing model tied to the EMH. But *some* model is necessary. Without an asset pricing model the EMH is meaningless. The EMH is an explanation for the RWH. And the explanation is that news is random and all relevant information is “fully reflected” in the prices. It is meaningless to talk of information being “fully reflected” in the prices in the absence of a specific model.

Not at all, that has always been the EMH, there is no specific asset pricing model tied to it. You may be thinking of CAPM, or ABT, but they are, not the same thing as the EMH, they are separate models.

LoL. Isn’t that the reason it’s called the market–because it is defined by the majority of transactions? Isn’t that the “magic” of capitalism–that everyone gets to pretend he or she is special while actually never doing or thinking anything special at all?

So, are you saying that the asset pricing model bit is unnecessary? The EMH is just a Random Walk Hypothesis? Surely this is a major re-statement of the EMH. In this re-statement
of yours. where does the “efficient” in the “Efficient Market Hypothesis” come from?

TH: “If it was obvious that the stock market would rise tomorrow, investors would buy shares immediately and the stock market would rise today instead.”

Can you explain how this happens in the presence of model heterogeneity, the difficulty of always being able to distinguish information from noise, investors trying to exploit others whom they perceive to be noise-traders, herding, investor short-termism, limited capital, the failure of the law of iterated expectations to hold (eg Keynesian beauty contests)?

You could say that your claim only holds in the absence of these issues. True enough. But then what have your claims got to do with the world we inhabit?

Do you believe it is ethically all right for economists, in their role as public intellectuals, to make claims that are demonstrably at variance with the existing state of knowledge in their discipline?

You could defend yourself by saying that you are neither an academic economist nor an intellectual, merely a journalist. Still, I’ve always thought the FT held itself to higher ethical and intellectual standards than, say, the economics department of the University of Chicago.