Risk Markets And Politics

Sunday, February 01, 2009

The gold trade isn't too crowded yet

The long gold trade isn't yet crowded because, although the pundits on CNBC have been capitulating over the past couple of months, it still seems like every other commentator has at least one of the following ideas, answered here.

1) Gold is not a true commodity. It's largely useless.

This is true, but of course the question is rather what is the use of paper money, increasing in supply, and paying no interest? Large changes in gold prices have little to do with what value it has as a commodity. It's more useful to think of gold as a (negative carry) currency. But you can't buy milk with it at the corner store? You can't do that with yen outside of Japan either, and privately issued gold-backed notes are feasible. It is easier to turn a store of value into a medium of exchange than the reverse.

2) Gold is just a safe-haven that you only buy if you think the world is going to end. With all that happened in 2008, if gold didn't go up then, when is it going to go up?

Gold had a positive return in 2008, and the drawdown of 30% was pretty muted for a "commodity." Given all of the de-leveraging, you could instead interpret this as impressive relative strength.

More importantly though, the generic "safe-haven" argument is suspect. Gold is not particularly correlated with measures of market risk like credit spreads and implied volatilities. If gold were a generic safe-haven, it wouldn't have a near-zero correlation with the VIX for example. Gold is a monetary safe-haven, and the generic safe-haven concept is largely a straw-man meant to conjure-up crazy gold bugs living in cabins stocked with guns. There are disasters that might be sufficient to take the generic safe-haven idea more seriously, but they aren't necessary for gold to go up.

3) We are experiencing deflation. Gold only goes up during inflation.

Well, first, what sort of inflation do we expect some months out? Second, gold has historically gone up during periods of deflation, and notably in those periods corresponding with multi-year boom/bust credit cycles (where the busts tend to be labeled "depressions"). The fact that gold declined during one period of disinflation of the 1980s colors many trader's view on this. When the predominant concern is inflation, gold will tend to fall with inflation expectations, but in deflation money appreciates relative to stuff, and gold, again, is closer to money than stuff. I'm not going to push this specific point too hard though because I don't think we have enough useful data on it. Most of the data we have is from the 1800s, where of course we were on the gold standard, so the comparison isn't quite fair for that and other reasons. Here is an interesting paper on the historical behavior of gold under deflation. Judge for yourself. Note though that the recent low inflation number in Europe allows the ECB to cut rates in order to fulfill their single mandate of promoting price stability. Two weeks after the HICP was released, gold broke out against the euro. Five days later it broke out against the dollar. Only the yen was left standing against gold and now in the past couple of days it's also succumbing in relative terms. What would happen if the JCB intervenes to weaken their currency?

The options on the December 2009 gold futures roughly give a 20% chance of those futures expiring above $1500/oz, and a 10% chance of them finishing above $2000/oz. If anything, the options might be a crowded trade, and an outright long is more appealing to me here. My main question is what price level will provoke some undermining public policy response, or credible threat thereof?

Anecdotally, Treasury is offering gold at over $1200/oz.