Nested Interest

“Gold gets dug up out of the ground someplace. Then we melt it down, dig another hole, bury it again and pay people to stand around guarding it. It has no utility. Anyone watching from Mars would be scratching their head.”  Warren Buffett

I have never been a big fan of gold as an investment, and here’s why: it does not generate cash, it has no balance sheet or income statement, and, as Mr. Buffett points out, it does not create any kind of tangible value.  I prefer giving $X to someone who can invest it in a business that will create additional cash flow with value of $X + $Y (this is what stocks and bonds allow investors to do). As best as I can tell, gold is purely a speculative bet on a future change in price of a commodity.

Notwithstanding my views, gold has experienced quite a bull run over the past few years which makes it hard to ignore.  In the past three years, the price of gold has more than doubled.  Two reasons seem to be most frequently cited for this run up in value:

Reason #1: Pending Hyper-Inflation

Some believe that the actions of the Fed over the past few years will ultimately lead to a period of severe hyper-inflation due to an excessive money supply.  The thinking goes “we can always print more money, but there is only so much gold, so if we print too much the value of gold will increase relative to the dollar” resulting in a natural hedge to inflation.  So, in anticipation of this hyper-inflation investors have been hording stockpiles of gold.

Reason #2: Armageddon

Gold is believed by many to be a universally recognized measure of value (remember, we once used it to peg the value of our currency). Should Armageddon settle upon us, these investors believe that gold will be the one currency that will still have trading value.  So, this theory suggests that in times of financial and social distress, gold will serve as an economic hedge.

What’s The Deal?

Is there any truth to these views of gold, or is it just folklore?  Well, let’s take a look at history (beyond just the past few years, that is) and see what this information might tell us about the relationship of gold to inflation and economic distress.

Myth #1:  The Hyper-Inflation Hedge

To get a sense of how well gold performs as a hedge against inflation risk, I decided to compare the Consumer Price Index (“CPI”) to the price of gold since 1971 – take a look at the following chart:

Now, if gold was a very good hedge against inflation it would closely follow the CPI; however, this chart clearly tells us that there are many other factors affecting the price of gold than simply inflation.

Correlation

To more precisely quantify the strength of this relationship, we will apply a statistical measure called correlation.  Correlation is a statistical measure where a value of 1.00 indicates that the two variables are perfectly, positively related (so the prices of each move exactly by the same amount each unit of time), a value of zero indicates no correlation at all (i.e., the two prices move completely independent of one another with no commonality), and a value of -1.00 indicates perfectly negative correlation (so that an increase in one price always corresponds to an equal and offsetting decrease in the price of the other).

Sorry about my actuarial musings, but I think you’ll find this useful.

Using this statistical measure, the historical correlation of the price of gold and the CPI is only 15%, reflecting a very low level of correlation.

I presume at this point that the critics are thinking “not so fast – we are only concerned about hyper-inflation, when the CPI experiences large and abrupt changes.  This analysis does not really focus on these situations.”

Fair enough.

To explore this, let’s focus on only those points in history when the monthly CPI changed by more than 3 standard deviations of its historical mean (meaning it was a big and highly unusual shift).  If gold was a really good hedge against this kind of unusual inflation, then we should see a big change in the price of gold moving in lock step.

Using this approach, I found five monthly periods with inflation more than 3 standard deviations from their historical averages.  Ranked in order of the percentage change in CPI (from left to right), here are the changes in CPI and gold prices during these months.

The first three months appearing on this chart correspond to the 2007-2008 time period, the first being October 2008 followed by November 2007 and September 2008, all of which had a large decline in the CPI. The two months with large increases in the CPI are July 1973 and December 2007.

Based on this information, how well would you sleep knowing that you have secured gold to protect you from hyper-inflation?  Probably not that well, I am guessing – in two out of five of these observations, the price of gold went in the complete opposite direction than we would need it to effectively serve as a hedge.  The statistical correlation between changes in the CPI and gold prices using these data points is only 4.35%, which is even lower than the relation we saw across the entire historical period.

I think it is safe to say that history give us little reason to believe that gold provides dependable protection against inflation.

Myth #2: A Hedge to Financial Distress

Now, let’s explore the second relationship – whether gold can serve as a good hedge in times of economic distress.  To do this, I will use historical returns of the S&P 500 as an indicator of economic well-being, and compare the historical returns of the S&P 500 to the price of gold.  Take a look at the following chart:

Once again, we do not see a close relationship; the statistical correlation between changes in the price of gold and returns on the S&P 500 is a meager -0.30%, meaning there is essentially no relationship between the two.

What about periods where there was severe economic distress?  Just as we did for the comparison against the CPI, let’s isolate those months where the returns of the S&P 500 were more than 3 standard deviations from the mean and see how closely the price of gold moved with the market.  Here is a summary of these observed monthly periods:

Moving from left to right, we first have a 26% market drop that corresponds to the market crash in October 1987, followed by financial crisis in September 2008 (loss of 23%), the terrorist attacks in September 2001 (loss of 12%), and President Nixon’s resignation in August/September 1974 (loss of 11%).  In each of these situations, gold has demonstrated very little resiliency in offsetting these market losses.  The statistical correlation between the price of gold and the S&P 500 for these observations is 38%, which is a non-trivial correlation but in the complete opposite direction of what is needed from a hedge – gold would need to increase in value when the S&P decreases, and in the same amount, to effectively serve this role.

Once again, history does not demonstrate gold’s ability to consistently deliver returns in a manner required to hedge this risk.

Conclusions

Well, I ended up coming to the same conclusion as Mr. Buffett – Martians must be scratching their heads wondering what we are doing. There does not appear to be strong evidence to support the notion that gold can serve as a dependable protector of wealth in situations similar to those we explored in history.  Of course, the future will be different than the past, but history is all we have to work with, so it is difficult to buy in to the espoused virtues of gold.

So, what about the recent increase in gold prices?  I can’t support this with hard evidence, but I suspect that it is driven by speculation.  Perhaps investors, subscribing to the beliefs noted above, expect gold to continue to increase in value (or, the value of the dollar to continue to decline).  One observation: the price of gold has gone up at a dramatic rate without a commensurate change in inflation or investor wealth.  So, perhaps gold prices are more affected by the perception of future hyper-inflation or severe economic distress rather than their actual occurrence?  It is not clear to me how you would test this hypothesis in a conclusive way.

Alternatives

What alternatives should investors consider?  I suggest exploring a laddered TIP strategy or an inflation adjusted immediate annuity to protect against hyper-inflation, and using put options or collars to protect against economic distress.  These vehicles have a much more direct and dependable relationship to these risks, and are therefore much more likely to deliver the returns that you will need, when you need them most.