Tuesday, November 3, 2009
Gold is in the news today, setting a new all-time high of $1085/oz. as of this writing. The first chart here shows nominal prices, while the second shows gold in constant dollar (real) prices. Either way you look at it, gold has enjoyed a pretty spectacular run since early 2001. At the risk of slighting the obvious geopolitical risks that motivate gold buyers these days, gold has for the most part benefited from years of accommodative monetary policy from almost all of the world's central banks. Easy money helped inflate the housing bubble several years ago by keeping rates artificially low. Easy money works by effectively lowering the hurdle rate for purchasing hard assets. People must always ask themselves this key question before buying gold, commodities, or real estate: "Will gold (or oil, or copper, etc.) prices in the future rise by more than the interest rate on safe assets?" The lower the interest rate, the easier it is to answer in the affirmative.
Easy money thus erodes the demand for money and boosts the demand for hard assets, resulting in rising tangible asset prices. Money loses its value relative to things, and that's what inflation is all about.
Rising gold prices are thus a signal that interest rates are too low and monetary policy too easy. There is more money in the system than the system wants, and that is the fundamental monetarist equation for inflation. That we haven't seen inflation show up in the CPI (well, at least not very much so far) is simply a reflection of the long and variable lags between monetary policy and the real world.
I've always thought that the primary objective of a central bank that chose to adopt interest rate targeting as its method for implementing monetary policy (as all major central banks have done) should be to pick the interest rate that leaves the market indifferent between buying government bonds and tangible assets such as gold and real estate. In practice, this could be described as a type of gold standard: the central bank should simply raise or lower interest rates (by selling or buying government bonds) in order to keep the price of gold within some specified range. Shrinking or expanding the money supply in this fashion would automatically keep the market indifferent between buying financial assets and tangible assets, because it would avoid monetary excesses or deficiencies, and thus deliver an essentially zero rate of inflation. Such a policy would inevitably lead to a very low and stable interest rate environment. And that, according to supply-side tenets, would be the best way for monetary policy to stimulate the economy.
If the Fed were to do this today, what should its target price for gold be? That is a question that has no definitive answer, but I'm going to guess that it should be somewhere in the $400-500/oz. range. As it happens, the real price of gold over the past 100 years has averaged about $450. If $450 is the price of gold that corresponds to "neutral" or zero-inflation monetary policy, then gold today is trading for a premium of over 100%; buying protection against inflation in the gold market is very expensive. Put another way, the Fed is going to have to continue to stand pat, and/or inflation is really going to have to accelerate just to keep gold from falling. If the Fed were to tighten policy sooner than expected, and with vigor, gold prices could tumble dramatically.
I'm not arguing against an investment in gold today, since rising gold prices seem to be the path of least resistance for now, considering that with one single exception (the Australian central bank), the world's major central banks have given every indication that a tightening of monetary policy is unlikely in the near future. My point is that buying gold is a very risky proposition now that it is trading at these lofty levels. In a best-case scenario for gold, we might see it revisiting its 1980 high in today's dollars (about $1800), but in a worst-case scenario it might fall back to $400. That's a very lopsided risk/reward proposition (aka an extremely speculative investment).
What could drive the worst-case scenario? Gold prices are extremely vulnerable to the mere suggestion that the Fed might begin reversing its liquidity injections. Gold prices are also very vulnerable to signs of stronger-than-expected growth, since the market can easily put two and two together and realize that a stronger economy means an earlier and more aggressive monetary tightening.
And while on the subject of vulnerable gold prices, these same arguments hold for T-bond prices. Yields on Treasuries are very low, mainly because the Fed is expected to be very easy for a very long time. Even the slightest change in those expectations could result in a sharp rise in Treasury yields, and a significant decline in T-bond prices.
Posted by Scott Grannis at 2:07 PM