Also, long and short positions aren't symmetric, because there is no non-margin way to enter a short position. This makes the ability of the market to stay irrational more likely to go bite you in a short position than a long one, ceteris paribus.

Isn't that a market failure then? If the founder and CEO of the company says the company is over-valued, there should be someone in a position to capitalize on that information.

Sure, but then so is pretty much every aspect of the stock market, IMHO. It's a tool with all sorts of terrible flaws when it comes to capital allocation. A single company being overvalued without a safe way to bet against mass delusion of the public isn't that large of a flaw compared to the systematic over valuations and under valuations, IMHO.

This isn't to say that it's not without its uses, but we should be aware that it has huge problems even if we don't know of a way to fix them yet.

Not really a failure. It's important to realize that people are not necessarily always paying for the 'actual value of a company' (subjective itself) but for what other agents will be willing to pay for the stock. If that value is higher than the current price, they'll buy.

> Isn't that a market failure then?

By definition, every non-ideal behavior of real markets is a “market failure”.

Real markets, as opposed to Econ 101 idealized ones, are full of market failures, both due to structures that Econ 101-level analysis ignores about basic market structures and human irrationality.