Many well-respected investors and finance writers—from Warren Buffett to Burton Malkiel (author of Random Walk Down Wall Street)—do not believe that investors should own gold as part of their investment portfolios.
However, many other investors and commentators argue that gold does belong in most portfolios.
Who’s right?
In this post, we’ll examine both arguments.
Case Against Gold
The main argument against owning gold is that gold is not an “investment.” Buffett, for example, takes this position. He argues that gold produces “nothing” because it produces no cash flows in the way that, for example, a bond produces cash flow.
Malkiel, in the 11th edition of Random Walk, put it this way:
“Gold, unlike stocks and bonds, is a sterile asset. It creates no products, earns no interest, and generates no earnings. Its value depends solely on what someone else is willing to pay for it. It is a classic example of the ‘greater fool’ theory at work.”
It is correct that gold doesn’t produce cash flow. It is also correct that cash—unless lent out—also does not produce any cash flow. However, most people do not believe that simply because cash does not produce cash flow, cash has no place in an investment portfolio.
For example, Malkiel, again in the 11th edition says, “Every investor should have a cash reserve.”
But few people, presumably including Malkiel, advise holding actual cash, as in, say, demand deposits that earn no interest. Instead, they cheat (intellectually) by advising investments in money market instruments. Money market instruments are not cash, but in normal times they function like cash.
So, people “cheat” by taking a non-productive asset – cash – and converting into a productive one by lending it out, at interest. In theory, you can do the same with gold.
Lending and Borrowing
There is a market for lending out gold, at interest. However, that market for lending is available mainly to institutions, such as central banks.
The world’s central banks are in aggregate, the largest class of owners of gold (unless you count the jewelry owned by billions of individuals as a single class of owners). And these central banks are the main lenders of gold.
You might wonder who would want to borrow gold. So did I. While there might be some small demand to borrow gold by speculators who want to sell it short, I think that is very limited. Anyone wanting to sell gold short can do so with much less hassle (and cost) by shorting gold futures.
So, who does that leave as borrowers? Gold mining companies. The main argument against borrowing gold, and borrowing dollars instead, is that the price of gold fluctuates. If you borrow gold when it is $1500/ounce, and sell it, you’ll get $1500 per ounce. You then use that to do whatever investment you were planning, and hopefully you earn enough from your investment to repay the loan, in gold.
Repayment in gold means you have to buy the gold. If your investment produces dollars, you’d have to take the dollars, buy the gold, and repay that gold. For the great majority of potential borrowers, the main consequence would be to create a new risk – exposure to the gold price – without any corresponding benefit.
And so, almost the only investment borrowers of gold are gold producers. For gold producers, one of the biggest risks in the business is the future price of gold. But if producers borrow in gold, and repay in gold (which they produce), they actually reduce the risk of changes in the gold price, because if they borrow (say) 1000 ounces, and repay 1000 ounces that they produce, the change in the gold price makes little difference.
Chances are, with respect to gold, you will have an easy time following the advice offered by Shakespeare’s Polonius: “Neither a borrower nor a lender be.”
Argument For Gold
The specious[1] argument for including gold in an investment portfolio is that “gold has had a great return” over some arbitrary period, and therefore, presumably, will continue to have a great return.
In fact, with gold over $3000 per ounce, almost no matter when you bought gold in the last 50 years, you would have earned a decent return. If you bought around 2000 near the lows (when the criminally negligent Chancellor of the British Exchequer, Gordon Brown, sold half of Britain’s gold so ineptly that a kindergartener would have handled it better), you would have earned compound returns to now better than the stock market.
Nevertheless, because it is true that gold produces no cash flow, the only investment return from gold, if any, must come from price change. And there is no long-run reason to believe that gold will outperform inflation.
Gold and Inflation
Is gold an inflation hedge? Probably the correct answer to this is: over the long run, yes. Over the short run no. Over the medium run? Sometimes yes, sometimes no.
Long Run
Over the very long run, measured in hundreds of years, gold has held its value extremely well. During the many centuries when much of the world was on a gold standard, the purchasing power of an ounce of gold fluctuated around a fairly stable level. Over long periods, the purchasing power tended to revert to the mean. If gold got too expensive (i.e. the price level fell – so-called “deflation”), the operation of the gold standard tended to reverse the downward movement of prices (i.e. the increase in the purchasing power of gold).
Similarly, if prices rose too much (i.e. “inflation” or a decrease in the purchasing power of gold), the gold standard tended to bring prices back down.
These long run empirical regularities are supported by hundreds of years of history, in the case of England going back to the 13th century.
Post-Gold, Fiat Money World
Since the world abandoned the gold standard in the 1930s, every currency on the planet has lost purchasing power over time. Even the best performing currency on earth, the Swiss Franc, has lost most of its purchasing power since the end of the gold standard.
For the period from the mid-1930s until 1971, the world, via the US and the post-WWII Bretton Woods agreement, was on a quasi-gold standard in which the official price of gold in dollars was fixed, but Americans were legally prohibited from owning gold. The US treasury “fixed” the dollar price of gold by selling off the US’s gold supply, until the drain became too big and Richard Nixon, instead of controlling the growth of the US money supply (which was driving the other nations to demand gold for dollars) caused the US to default on the gold promise.
When measured in terms of fiat currency prices, the value of an ounce of gold is quite volatile. In the US, the price of gold was allowed to float freely in 1971. Since then, the dollar price of an ounce of gold has increased by a factor of nearly 100 times.
Put differently, the purchasing power of a dollar, in terms of gold, has declined by almost 99%.
Medium Term
There are now 54 years of data since the time that Nixon cut the last official link between the dollar and gold. That period gives us some limited insight into how good, or poor, an inflation hedge gold has been over terms measured in say 5 to 10 years.
And that track record is mixed. During the 1970s, gold tended to be an excellent hedge against inflation.
Then, in the 1980s and 1990s, gold was a terrible hedge against inflation.
In first decade of the 2000s, gold was a good hedge.
In the second decade of the 2000s it was sometimes a good hedge and sometimes a bad hedge.
Short Run
Over the short run, measured in months, gold is not an inflation hedge. To be a hedge, the dollar price of gold would have to change in a way correlated with inflation. When we look at monthly CPI data, and the gold price, we find that there is little if any correlation. Thus, over the short run, gold is not an inflation hedge.
Volatility
Measured in dollars, the appreciation of gold’s price in dollars has been anything but smooth. Over the entire period since 1971, the price has risen at an average annual compound rate of about 8.7%. The annualized standard deviation has been about 17.5%.
That volatility is similar to the average long run volatility of the S&P 500. And the appreciation since the dollar was severed from gold has also been similar to, though lower than, the return to the S&P 500.
Low Correlation with Stocks
Since 1971, the average correlation of daily returns between gold and the S&P 500 has been negative. Thus, as both asset classes had strong positive returns, looking backward through a mean/variance optimization lens, gold was an excellent addition to most portfolios.
Conclusion
Is 50 years a long enough period on from which to draw conclusions? Statistically, the answer to this question is pretty clearly no for most of the inferences we’d really like to draw from past data. However, for most markets, it’s what we have.
Unless you think there’s good reason to believe that the last 50 years of gold price performance is exceptional and unlikely to be repeated, on a data-based approach, it seems like the argument for including gold in a portfolio is strong.
Contact Us
To speak with us further, reply back to this email or call us at 703-437-9720.
[1] Please pardon the pun.
Leave a Reply