Being long gold has been a hugely successful trade since the onset of the credit crisis, generating an annualized real return of 17.2% over the past five years. With the introduction of gold-backed exchange-traded funds such as the popular SPDR Gold Shares
The problem? It's impossible to establish an intrinsic value for an asset that produces no cash. In this article, the second part of my interview with the head of global macro hedge fund firm Armored Wolf, John Brynjolfsson (@Brynjo on Twitter), I wanted to understand how a professional investor who is focused on real returns thinks about the yellow metal (you can find part one of the interview here.)
I began the discussion with a provocative question: Is it possible to invest in gold rationally? Brynjolfsson was game and he gave me a carefully articulated answer:
Absolutely. Here's a framework that I look at for gold. I see gold as an insurance policy in certain ways. And if it were an insurance policy, you should ask two questions. What are we insuring against? In other words, under what circumstances does an insurance policy pay off? And two, how much do we have to pay for that insurance policy, in terms of premium.
He spoke about gold insuring three risks:
Gold insures against three things: One, it insures against inflation, or in other words, debasement of your traditional money. If there's inflation in dollar terms, or inflation in Euro terms, then gold does well as an inflation hedge.
Two, it's a crisis hedge: If you have a situation where the payment system is seizing up, where people are not 100 percent sure that they'll be able to get their money out of their banks, or they're not 100 percent sure that their bank will survive, or if they're less comfortable with the political situation.
Thirdly, gold is traditionally a hedge against higher taxation: Gold is going to go up in price when people feel that they need to store their wealth in gold, as a way to shield it from taxation, either legal taxation, or, possibly, confiscatory taxation.
And about pricing gold as insurance:
If those three perils are what gold insures against, we have to ask, how much are you paying for that insurance. The answer, I believe, is, if you think of the current gold prices as a starting point, and that if any of these policies pay off, the price goes up, the price you're paying for that insurance is how much interest you're giving up by owning gold, rather than, say, owning a bond.
The yield on 10‑year bonds is down to one‑and‑a‑half percent, in the U.S. On five‑year treasuries, it's 60 basis points. For an ounce of gold, it's 1,600 dollars. If you're only paying, let's say, one percent per year, for the next five years, as an opportunity cost to hold that, you're giving it a relatively small amount, you're giving about 16 dollars a year in opportunity cost, or 80 dollars over a period of five years in opportunity cost.
This is a very important insight that is well-supported by the data. Indeed, if we look at the evolution of the price of gold over the past five years, it is closely related to that of real interest rates. The following graph shows the price of gold (in blue, measured along the left-hand axis) and the real yield on 5-year Treasury Inflation-Protected Securities (in red, measured along the right-hand axis -- I've multiplied the yields by minus one to better depict the correlation):
The price of gold has risen as the opportunity cost of holding gold (i.e. real interest rates) has fallen. I asked Brynjolfsson how have the risks against which gold insures changed?
Six years ago, interest costs were at least five times higher. The insurance you were getting was not worth as much, either, because, prior to the financial crisis, no-one really would have thought that raising taxes in a significant way was a possibility or likelihood. Now, everybody agrees that, over the next five years, taxes are going to go up -- maybe as quickly as next December 31.
Is the risk of global crisis, either banking crisis or a military crisis going up? Absolutely. It's double or triple or quadruple what it was six years ago. Six years ago, no one really thought of that as a risk at all.
Six years ago, inflation wasn't really a threat. Now, with global money‑printing coming out of the Bank of Japan, the Bank of England, the Swiss National Bank, the ECB, the Fed -- that has to increase the inflation risk. The things that gold insures against, those perils seem to be much greater risks now than they ever were. And the cost seems to be much lower. That means that gold is easy to justify as an investment.
Nevertheless, Brynjolfsson actually prefers another precious metal to gold:
Right now, we're not long gold, we're long platinum, and we have a slight short on [in gold], as a hedge, to essentially magnify our platinum bet. As I said earlier, if you can hedge out the general trends, and focus in on the nuanced opportunities, that's a good thing to do.
How is this a "nuanced opportunity"? Statistically speaking, platinum is cheap relative to gold right now. Normally, the dollar price of an ounce of platinum is greater than that of an ounce of gold. Based on Tuesday's prices, the platinum/gold ratio is below parity, at 0.88. Prior to 2011, you have to go back to 1985 to find the ratio at that level:
Source: Author's calculations, Kitco, World Gold Council.
I ran a test on the ratio series spanning January 1970 through June 2012 and found that it is stationary with statistical significance at the 1% level. Translation: We can be pretty confident the ratio will move back toward its average over time and, as such, investors who like gold right now should like platinum even more (for reference, the geometric mean of the ratio over the period beginning in 1970 through July 31 is 1.36). Individual investors can own platinum through the physically backed ETFS Physical Platinum Shares
The right model
I think John Brynjolfsson's conceptual framework of gold as insurance is exactly the right one and any investor interested in gold would do well to give his analysis careful consideration. I don't agree, however, that forgone interest is the only cost of that insurance or that the price of gold will necessarily increase if any of the risks he describes comes to pass. When you buy an ounce of gold, the amount of wealth you're actually insuring is equal to the fair price of that ounce of gold -- any premium you pay above that amount adds to the cost of insurance. With the dollar price of gold nearly three standard deviations above its inflation-adjusted average in the post-Bretton Woods era, that insurance looks enormously expensive. Platinum certainly looks like the superior choice.
There are macro trends and there are macro catalysts: These Stocks Could Skyrocket After the 2012 Presidential Election.