Where should gold trade? Until virtual currencies came along, this was the single most contentious question in investing. Even now, the loudest voices are at the extremes, either saying gold is the only “real money” or calling it an antiquated and societally unproductive store of value.
Our own view is that gold is a useful asset when constructing diversified portfolios. Its price is uncorrelated to stocks and bonds. Over time, gold tends to increase in value (10-year CAGR of 2.1 pct, 20-year CAGR of 10.1 pct). We don’t really care if it’s “money” or just part of the commodity asset class.
In terms of assessing gold’s price, our approach is to look at a ratio of the S&P 500 divided by the cost of one ounce of the yellow metal. Both have nice long time series to analyze, are priced in dollars, and coexist in whatever economic/monetary policy regime happens to exist at any point in time.
Ultimately, we see this as a comparison between investor confidence in 1) human ingenuity (stocks reflect corporate earnings) and 2) central bankers’ competence (gold supply relatively fixed, monetary policy variable). Two very different factors, true, but both highly relevant just now.
This chart shows the gold/S&P ratio back to 1970. The only footnote necessary is that the US came off the last vestiges of the gold standard in 1971, which left it free to trade in line with market forces. That’s the drop you see from the S&P trading for +2x gold in the very early 1970s to below that level from 1972 onwards.
While there is literally a book’s worth of observations in this chart, we’ll summarize the 3 most important ones:
#1: The average S&P/gold price ratio from 1972 – present is 1.6x. If gold trades for $1,000/oz, for example, the S&P would trade at 1,600 if it ran in line with historical averages.
#2: There are three distinct periods represented in the chart:
1973 – 1992 (the long U-shape on the left side of the graph), when an ounce of gold cost more than 1 unit of the S&P 500. The bottom for the ratio was in 1980 (at 0.2x), and that was the start of the decade’s bull market for stocks. But the hangover from gold’s spike to $800/ounce in January 1980 took a long time to unwind completely.
1993 – 2007 (the mountain in the middle), when the S&P/gold ratio soared to 5x at the peak of the late 1990s dot com bubble and remained above 2x through 2007. This, broadly speaking, was a period when investors overvalued human ingenuity. The dot com bubble merged into the housing financialization bubble leaving stocks overvalued versus gold.
2008 – present (the right third of the chart, another U-shape), when investors first lost faith in central bankers during/after the Financial Crisis (lifting gold prices to a 1x ratio to stocks from 2009 – 2013) but then developed a renewed confidence in human ingenuity (2x stock/gold multiple from 2017 – onwards). Much of that latter move is, of course, related to Big Tech’s increasing economic and US equity market dominance.
#3: At a 2.3x S&P/gold ratio today, we are right at the high end of a band that’s been in place since 2006; that makes gold cheap versus stocks. Consider:
- The 2018 peak of a 2.4x ratio (i.e., not far from today’s levels) coincided with 10-year yields at 3.0 pct and Fed Funds of 2 percent. No surprise, but gold did not do well in that environment of tighter money, falling to $1,200/oz from $1,350 earlier in the year.
- Today’s 2.3x ratio is actually slightly higher than the 2006 – 2007 period’s 2.2x. Back then, we had 5 percent Fed Funds rates, 4-5 pct 10-year Treasury yields and no bond buying. Now we have much lower rates and lots of QE. The two environments could not be more different, yet the stock/gold ratio is the same.
Worth noting: you can’t use this ratio to just buy gold or short the S&P; this is a relative value discussion, which means:
- Let’s say the “right” S&P/gold ratio is 2x – the lower end of the last 4 years.
- That means gold should trade for $2,125/oz today (S&P 4,250/2), or 13 pct above current levels.
- But we could also get to 2x by having the S&P 500 drop to 3,716 (2 x gold’s $1,858/oz), a 13 pct decline.
Takeaway (1): we like gold here in the context of a diversified portfolio. On a relative basis, it is cheap on near-term history (2006 – present) and the long-run (1970 – present) shows it is a reasonable hedge against investor unease regarding monetary policy. There’s certainly enough of that at the moment. Of course, if investor confidence in central bankers really starts to waver, then gold could do much better than the above 2x ratio scenario. History is clear on that point.
Takeaway (2): gold vs. stocks is a tug of war between fear of monetary policy failures and confidence in human ingenuity, so do virtual currencies offer the best of both worlds? Perhaps, but as we’ve noted in recent reports they remain just as volatile now as years ago despite their growth. We like them for a small, speculative position (up to 1 pct) but not as a chunky core holding. We also believe that their ongoing volatility limits their substitutability for gold.