On Hedging Inflation, Part I: Gold

golden nickelGold is not a good inflation hedge. Yes, it has been touted as one in the past and continues to be touted as one now, but there is little evidence supporting the claim. What’s more, it is rare that any wealth manager or gold bug or supporter of such an idea will tell you why or how it is an effective hedge. It is typically presented as nothing less than a simple statement of fact, a truth everybody should already be familiar and comfortable with. If any reasoning is provided, it might sound something like this – fiat currencies are not backed by anything anymore, unlike before when they were backed by gold. Hence, as central banks the world over pump paper into the system to salvage their ravaged economies, it is inevitable that they will ultimately devaluate their currencies and trigger inflation in the process; they will overdo it, overshoot the mark, mismanage, screw up. People will “wake up” and realize that paper is just paper, that the overzealous “printing” of it has created too much of it chasing too few goods, and they will lose confidence in their central banks and in the currencies they manage. As confidence is lost, so will be value and purchasing power. Furthermore, paper is not like gold, which is a physical thing that has a limited supply (there’s only so much of it that is accessible, anyway), a thing that must be mined, processed, and stored, whereas paper is just…paper.

Under this model, gold has “real” value, while currencies really don’t. Without diving into the intricacies of the fiat monetary system, this specific idea is problematic on its own – the value of something is only “real” if people agree to that item having value in the first place. Currently, people agree that both gold and currency have “real” value, so gold’s value cannot be considered more “real” than that of fiat paper money – the proponents of gold expect this to change but so far it has not. Moreover, gold’s value is still expressed in terms of some currency, so in that case, currency will continue to have “real” value until we begin to express the worth of gold in terms of something else. Again, gold’s value is no more “real” than currency’s. What’s more, under different circumstances, values change – in a modern apocalyptic or post-apocalyptic world, a seed may have more value than some weight of gold, which in turn may become useless.

Nevertheless, suffice to say that the model and its ideas are commonplace. When people realize the worthlessness of their paper money, when they see that their purchasing power is decreasing, they will flock to the “real” store of value, gold, driving the price up. Or perhaps the proponents believe that in a period of inflation, gold, like everything else, will cost more in nominal dollars. If you are already holding gold going into a period of high inflation, then what took you less than $1,100 dollars an ounce to buy today will be worth some value much higher in dollars per ounce when inflation hits, while you’ll still only have the same number of dollars that will now buy you less if you don’t quickly invest them in something of “real” value like the yellow metal. The proponents argue that the time to buy is now, of course, before inflation hits, otherwise you’ll be caught chasing a runaway train.

The main issue with the concept of gold as a functional hedge against inflation is that there is no consistent historical data to support it. In the past, there have been as many instances of gold not hedging inflation as hedging inflation, perhaps even more cases where it has not. In other words, gold’s behavior in periods of inflation is entirely inconsistent as a hedging instrument. What good does a hedge do you if you can’t count on it to hedge risk on a more or less continuous basis, i.e. at least regularly or most of the time for some given situation? With an inconsistent hedge, you are now gambling, gambling that gold will work the way you wish it to. That’s not sound risk management.

In 2012, Claude B. Erb and Campbell R. Harvey published a study entitled, “The Golden Dilemma“. In it, they set out to understand how gold should be treated in terms of portfolio asset allocation and, among other things, seek to understand what drives gold prices by investigating common arguments for owning gold. They acknowledge that one such justification that has been presented over the years is the one that gold hedges inflation. However, in their paper, the two find little to no evidence that gold is an effective hedging instrument against inflation, i.e. that inflation drives the price of gold.

True, leading up to and then during the horrendous inflation of 1978-1981 in the United States, with the peak value in 1980, the real price of gold, as they calculate by dividing the nominal price of gold by the CPI value month-by-month, spiked, making gold investing appear to have really paid off as an inflation hedge during those years. What this means is, the real price of gold during that time period more than held its purchasing power, better than did the US dollar. That’s great, but if gold is a good short-term hedge of inflation, then the real price of gold should hold steady with little price variability, it should remain more or less constant. However, as the authors point out, if one looks at the entire period of 1975 (when futures trading in gold began) through the time of publication of the paper, 2012, it becomes immediately evident that the real price of gold is anything but constant, landing all over the place with a great amount of variability, as the chart sourced from the study below shows.


Erb and Harvey go on to compare the actual change in the inflation rate year over year with the percentage change in the gold price and once more observe essentially no correlation (only the year 1980 stands apart, once again), indicating that gold does not even hedge unexpected inflation. Below is a view of the unpredictable and irregular relationship between the two. Notice the lack of a trend of gold returns correlating with changes in the inflation rate; in many cases, negative changes in the inflation rate correspond with positive percentage changes in the nominal price of gold, and vice versa.

exhibit 3 the golden dilemma gold versus unexpected inflation

Even in historical cases of hyperinflation, such as that involving Brazil from 1980-2000, Erb and Harvey don’t uncover evidence that gold successfully hedged inflation historically, actually showing in the Brazilian case the real rate of return on gold ownership to be negative over that period. You wouldn’t have seen as much erosion of your wealth if you had been in gold instead of the Brazilian real by the end of that time period, but you wouldn’t have escaped unscathed either.

The authors further compare 10 year gold returns (nominal and real) against the 10-year CPI inflation rate, resulting in no real relationship and signifying that when 10 years is considered the long-term investment horizon (as opposed to, say, 30 years), gold does not effectively act as a long-term hedge against inflation either; they review the mean reversion of the price of gold; they compare US nominal disposable personal income against per capita nominal disposable personal income expressed in ounces of gold; they look at military pay in ounces of gold over time; and they investigate yet other things, seemingly trying to exhaust through whatever means possible any way of testing whether gold serves as a reasonable inflation hedge. More or less they come back with little to no evidence that gold is an effective hedge whatsoever against any type of inflation.

An additional and rather interesting way to look at the interaction between inflation and the price of gold is to treat inflation as just another asset and see how well-indexed percentage changes in the price of gold are to percentage changes in the inflation rate. For instance, if the inflation rate has a positive 100% change, does gold? Here, for a perfect hedge, we would expect to see a 1-for-1 change, or a ratio of 1. Well, forget about a 1-for-1 change, gold does a poor job indeed of even moving in the same general direction as inflation, as revealed by the next chart for 1976-2014. Observe how the average value (green line) falls below zero and how many data points clump around zero, implying that the yellow metal is likely to move in the opposite direction of inflation or unlikely to respond at all, respectively.

gold inflation performance index

In the end, you may choose to hold gold for any reason, perhaps because its performance often correlates poorly with the performance of other asset classes such as equities and bonds and this is appealing for those wishing to form a well-balanced portfolio. But forget about hedging inflation. There are better ways to address that risk. The argument that the yellow metal will consistently do the trick is fools’ gold.