The 'golden constant'

Chas Craig, BridgeTower Media Newswires

“The only possible exceptions to this cash flow test are precious metals, such as gold, which is a widely recognized store of value; throughout history, for instance, the value of an ounce of gold has been roughly equivalent to the cost of a fine men’s suit.”

– Margin of Safety by Seth Klarman (1991)

To those that purchase gold on post-apocalyptic fears, you’d probably be better off loading up on canned food, bottled water, munitions and whiskey. I strongly believe those four items would prove more valuable.

However, to the point of the opening quote, gold probably has historically provided a highly variable and imperfect hedge against inflation over the long-term. This is despite gold possessing only limited practical use. Indeed, as Warren Buffett put it in his 2011 letter to Berkshire Hathaway shareholders, “what motivates gold purchasers is their belief that the ranks of the fearful will grow.”

A research article entitled “Gold, the Golden Constant, and Déjà Vu” by Claude Erb, Campbell R. Harvey and Tadas Viskanta in the 4Q 2020 edition of the Financial Analyst Journal cited earlier work by Jastram (1978) that suggests that in the long-term, the purchasing power of gold is constant, meaning the expected real (inflation adjusted) rate of return for gold is zero. Therefore, the nominal return (the real return plus inflation) that is most often quoted to investors is expected to be derived exclusively from inflation. To this point, since gold futures started trading in 1975, the real price of gold has been mean reverting in a highly volatile fashion, with no discernable upward or downward trend.

Erb and Harvey (2013) explored Jastram’s “golden constant” framework and concluded that the starting real price of gold has historically been a more important determinant of subsequent real and nominal rates of return for gold than the realized rate of inflation.

Erb and Harvey’s conclusion is analogous to the finding of Asness (2012) that concluded that the starting Shiller P/E valuation, not earnings growth, was the best explanatory variable for forward 10-year real stock market returns, on average. In this gold and stocks analogy, inflation is to earnings as the real price of gold is the Shiller P/E ratio.

The real price of gold is now nearly on par with the post-1975 highs seen in early-1980 and mid-2011. Here is an excerpt from the previously referenced Financial Analyst Journal research report: “Since 1975, periods of high real gold prices have occurred during periods of elevated concern about high future price inflation. Five years after the real price peaks in January 1980 and August 2011, the nominal (real) prices of gold fell 55% (67%) and 28% (33%), respectively. Today’s high real price of gold suggests that gold is an expensive inflation hedge with a low prospective real return.”

Sticking with our earlier assumption that the real price of gold is likely to revert to its long-run mean over time, if the real price of gold declines to its average since 1975 in 10 years, inflation will need to average about 7% per year in order to break even on a nominal basis. The 7% inflation figure compares to the little less than 2% implied by the yield spread between 10-year nominal and inflation protected Treasury bonds.

Bottom line, although J.P. explained our generally skeptical view toward gold in a July column, readers may have incorrectly inferred we favor gold when I highlighted inflation as a medium to long-term risk to financial markets in a June column. For what it’s worth, we’re not buying gold.

—————

Chas Craig is president of Meliora Capital in Tulsa (www.melcapital.com).