Taking stock of market efficiency

One of the most powerful arguments for markets is that they are the most efficient way to price assets and allocate resources. In macroeconomics, the efficient markets hypothesis of financial markets suggests that open markets strongly mitigate or even eliminate opportunities for arbitrage because market-determined prices already account for all known information about an asset.

With that in mind, an interesting paper released today by the NY Fed (HT: Mark Thoma) examines what type of news actually moves stock, bond, and FX markets. The authors find that only a few types of information (nonfarm payroll numbers, the GDP advance release, and a private sector manufacturing report) have a persistent and significant effect on market pricing. As Free Exchange wonders out loud, does that really mean that everything other piece of news is irrelevant or redundant?

In another sort of market, the contracts for Obama and McCain election victories are now roughly even. Interestingly, the contracts closely reflect the national polls. The problem is, the market ought to know that the US presidential election is anything but a national race. If you look at the state-by-state polling, many analysts currently have Mr. Obama with a much more considerable lead in the Electoral College; he’s even winning by a bit on Karl Rove’s latest map. It’s hard to surmise that the prediction market has adequately taken this fact into account, especially given the timing of the price rise on Mr. McCain’s contract (nearly mirroring his rise in the national polls.)

Furthermore, as a commodities watcher, I’ve seen many situations where the price does not reflect any logical interpretation of the supply/demand fundamentals. Go try and make sense of the international sugar market over the last two years if you don’t believe me.

What gives?