By “uncertainty”, I mean what former US defense secretary Donald Rumsfeld called “unknown unknowns” – dangers which cannot be adequately quantified.
To see why this matters for gold, imagine you were very uncertain about the economic outlook. What would you do? You wouldn’t want to hold many equities for fear of recession. But you wouldn’t want many bonds either for fear of inflation. Gold, then, becomes attractive. This is especially so because in uncertain times our main motive is simply to preserve wealth, rather than increase it. And gold helps to do this; if the economy booms, gold loses less than bonds but if it slumps, gold loses less than equities. Gold is, says Dirk Baur of the University of Technology, Sydney, a minimax asset; it minimizes our maximum loss.
Yes, gold is volatile in itself, as we have been reminded recently. But this volatility is largely idiosyncratic. Gold protects us from a systematic danger – that of increasing uncertainty.
This implies that the gold price should rise and fall with uncertainty. When investors are uncertain, they’ll want gold and its price will be high. But as uncertainty falls, so will demand for gold and its price.
So goes the theory. What of the evidence? A team of researchers led by Nick Bloom at Stanford University have constructed an index of uncertainty about US economic policy. If the above is right, this index will be positively correlated with the gold price.
And it is. In monthly data since 1985 – when the Stanford index begins – the correlation between the two is a hefty 0.65. Low uncertainty in the mid-90s was accompanied by a low gold price and rising uncertainty in the 00s was accompanied by a rising gold price. In fact, this uncertainty index along with the real interest rate can explain almost three-quarters of the variation in gold since 1986*.
And here’s the thing. The uncertainty index has fallen sharply so far this year. It’s now half its post-1985 average, and close to its lowest level since 2007.
Of course, uncertainty about US economic policy is only one aspect of overall uncertainty. But it’s not obvious that other aspects of uncertainty have risen to offset this decline. If anything, the opposite. There’s probably less uncertainty about the euro area now than there was a few months ago, in the sense that there’s less speculation both about a possible immediate break-up and about a dramatic resolution of the crisis.
Gold has fallen, therefore, in part because demand for protection against uncertainty has fallen.
You might object that this is just the wisdom of hindsight. This is not a serious charge. It’s better to be wise after the event than not at all. And in this instance it’s not true. Back in September I wrote: “If uncertainty recedes…then (ceteris paribus) gold could fall as maximin demand for it declines.
Herein, however, lies a reason not to dump the metal entirely. It’s quite possible that uncertainty will re-emerge – though by definition we cannot predict how or when. If or when this happens, gold could well be in demand again. In this sense, paradoxically, the recent fall in the metal has actually reminded us of its value to investors.
* I’m measuring the real interest rate as the average yield on index-linked gilts with a maturity of five years or more, which is a rough proxy for the world interest rate. The idea here is that low interest rates mean that the opportunity cost of holding gold is low because there’s less yield on bonds to give up when you hold the metal. This means that gold prices should be high when interest rates are low.
