If There Are Riots In New York, You Can Blame Ethanol And Index Funds


Mayor Bloomberg is on record claiming that riots are caused by unemployment among college graduates. But I suspect that Occupy Wall Street is overweight degrees and underweight 9-to-5 employment, and it's not taking down the regime so much as it is taking down a lot of pizza. So one might question Bloomie's conclusions and seek someone with more rigorous statistical training to explain Where Do Riots Come From.

Fortunately we have that in Yaneer Bar-Yam, a physicist, complex systems theorist and general man-about-town with a pleasing CV that includes writing a book called "Making Things Work" and telling Slate that "the shortcomings of the U.N. humanitarian-response system in Haiti have a lot to do with a 50-year-old mathematical theorem known as Ashby's Law of Requisite Variety."

He and some friends wrote a paper, released yesterday, arguing that increases in food prices over the last few years can be explained almost entirely by two factors: financial market speculation and the growing use of corn for ethanol production rather than food.

Now these arguments have been madebefore, and also disputed or minimized. The new paper rejects a lot of those criticisms; more interestingly, it constructs a fairly simple four-parameter model that can pretty closely match the actual trends in food prices over the last few years:

So, neat. Also disturbing. Also also disturbing is this:

That's from their companion, non-mathy paper that proposes "that protests may reflect not only long-standing political failings of governments, but also the sudden desperate straits of vulnerable populations. If food prices remain high, there is likely to be persistent and increasing global social disruption." The main graph is food prices, the red dotted lines are incidents of rioting. Charmingly the numbers in parentheses are death tolls. They predict global doom in 2012-2013.

So that's not great. But what does it mean for your trading book? Well, besides global doom, there's this. They write:

The question of whether speculators stabilize or destabilize prices has been the subject of a large body of literature, going back to Milton Friedman, who said "People who argue that speculation is generally destabilizing seldom realize that this is largely equivalent to saying that speculators lose money, since speculation can be destabilizing in general only if speculators on average sell when the [commodity] is low in price and buy when it is high."

But their model, which they at least think closely tracks real prices, is purely momentum driven: "If the price change of the commodity is positive in the previous time step, speculators are willing to buy a quantity mu[P(t) - P(t - 1)] of commodity, otherwise they sell mu[P(t - 1) - P(t)]." Similarly they assume that investors shift from commodities to other investments when those investments' price goes up, and vice versa. Which suggests that commodity speculation is driven mainly by herd-following momentum chasing - and that when the revolution comes speculators will get burned down along with everyone else.

The Food Crises: A Quantitative Model of Food Prices Including Speculators and Ethanol Conversion [NECSI, pdf, via Scientific American]

The Food Crises and Political Instability in North Africa and the Middle East [NECSI, pdf]


The Smart Indexes Are Even Worse Than The Dumb Ones*

You may have heard that the Dow hit 13,000 today before subsiding to a shameful 12,965.69. You may not have heard this, or cared, because the Dow is for morons, being a price-weighted index of thirty semi-random companies that, gah, aren't even "industrial" any more.** There are alternative theories but those theories are wrong: Joe Weisenthal in defense of the Dow has been noting its very high correlation with other, broader, more sensible indexes. I see this as further undermining the Dow's legitimacy. If it's very different methodology were leading to some kind of meaningfully different result, then we could perhaps argue that it's adding value in some kind of way. But instead what's going on is that the Dow's creators are hand-picking which stocks to include in the index specifically with an eye toward constructing an index that mirrors the other, better indexes out there. Apple and Google, for example, aren't in the Dow and aren't doing to get in any time soon because their very high share prices would skew the index in weird ways. This just goes to show that the Dow's creators already "know" the right answer (from looking at the S&P 500 and the Wilshire 5000) and then are trying to assemble an index to create the predetermined result. Maybe! An alternative theory is maybe suggested by [Occam's razor and] this piece from the Journal this weekend about index funds that I just loved and so am now going to inflict on you at unnecessary length: