It almost seems like “conventional” wisdom; you should have a portion (typically 5%) of your portfolio in gold. The reasons for owning gold vary, but they tend to be highly emotional as opposed to rational. A recent article in the trade publication, Investment News, questions the recommendation by Jerry Wagner, the manager of the Gold Bullion Strategy Fund that investors should have 20% of their portfolio in gold (the author fails to bring up the possibility of selfish motives).
I question the allocation of any assets into gold (owning a well-diversified portfolio means that you will own gold shares anyway). It has rarely generated much in the way in real returns (after inflation) over long periods.
So, imagine my surprise when I the author quotes Mr. Wagner’s claim that gold has posted “a 40-year compound annual growth rate of more than 7.5%… second only to equities, at 8.2%.” I smelled a whiff of data mining and said so in the article’s comment section:
“An annual return of 7.5% for gold? Wow! Talk about data mining. Gold is not an asset from which an investor should ever assume a return greater than the increase in other commodities' values. Pick a time period in the last 2000 years and you'll find that an ounce of gold would have purchased a nicely tailored man's suit (or toga). If you're really concerned about the total collapse of the global economy, would you be better off with some hunks of shiny rock or a cache of really big guns and several years of Costco freeze-dried food?”
I probably would have stopped there had Mr. Wagner not chosen to respond directly to my comments:
“Yes, although it was actually a CAGR of 7.6%, second only to Equities 8.2% in the same time frame. Hard to call it data mining though – our research covers the results for all forty years since the price of gold was made free floating after Nixon’s devaluation of the dollar in the seventies. I know it’s not back to roman times, as you would like, but neither are the statistics that we all use most of the time for equities. And speaking of equities, what’s the value of ancient Rome’s coliseum gladiator (or toga) concession today?
“I agree that your suggestions are much better that (sic) gold if what you are worried about is a “total collapse of the global economy," especially if you have the gold in an ETF, mutual fund or distant bank box. But our paper was addressing the role of gold in a modern portfolio. Last I checked most portfolio managers worry about making money for their client’s in the market environments we have been exposed to over the last, say, forty years, and we assume the present financial system continues.”
Not one to back down from a spirited argument, I replied in some detail:
“‘From 1836 to 2011, the average real rate of price change for gold in the United States is 1.1% per year and the standard deviation is 13.1%, implying a one-standard-deviation confidence band for the mean of (0.1%, 2.1%). The covariances of gold’s real rate of price change with consumption and GDP growth rates are small and statistically insignificantly different from zero.’
“The same conclusion was reached in this 2011 paper from the National Center for Policy Analysis:
“‘• Gold purchased in 1975 at $129 an ounce would have sold at the end of 2010 for $1,420.25 for an annual real rate of return of 1.56 percent.
• By contrast, the real average annual rate of return for Standard & Poor’s 500 Composite Stock Index from 1975 to 2010 was nearly 8.4 percent.’
“With gold currently trading at about $1284, that real return number will be even less today. In fact, it appears that the historic REAL rate of return for gold will likely vanish after investors pay the 1.46% annual fee for the Gold Bullion Strategy Investor fund.
“The biggest argument of “gold bugs” is gold’s value as a means of storing wealth. I have no problems with that argument. However, to imply that a lump of shiny metal has the means to actually generate real returns is absurd. Gold cannot, in and of itself, become more valuable. It doesn’t grow. It doesn’t reproduce. It doesn’t create. It’s a rock to which we have ascribed totally arbitrary value.
“Businesses, in general, can grow more valuable. The global economy, which those aggregated businesses represent, can (and historically has) grown more valuable. While economic growth does occasionally pause, if it ever ceases permanently, I don’t think gold will bail us out.
“Many investors grew excited about the prospects for gold back in the early 1980s. Had someone purchased an ounce of gold at $600 (then watched it explode up to $850) and held it until today, they would have LOST money on a real return basis after 34 years (pre-inflation average annual return: 2.25%). And, yes, that, too is data mining.
“To my point about data mining, I totally forgot to add:
“You state ‘our research covers the results for ALL forty years since the price of gold was made free floating...(emphasis added)’ There have only been THREE 40-year periods over which to measure gold prices since Nixon allowed gold prices to float in August 1971. Just THREE! Implying that you ran the numbers for multiple 40-year periods is disingenuous. Try running your numbers over 20 or 30-year rolling periods and the results will look VERY different.”
The only people who actual generate occasional real returns from gold are those who mine it or sell it. Those who guess well and buy it at the right time, and then manage to get lucky enough to sell after is rises might turn a profit. They, however, are no different than the person who picks the right horse at the track, lucky.
Gold looks nice on someone’s finger or neck, but I have never seen a ring or necklace increase in size or weight.