Gold has been doing well since the start of the year, but observers remain divided as to what exactly it is that is moving gold.
Let's review the traditional arguments by gold bulls (from Goldcore):
- Diversification; as gold has a negative or weak correlation to other financial assets.
- Liquidity; it's a liquid market
- No liabilities; gold is an asset that doesn't represent someone else's liability and therefore its value isn't determined by an ability to repay
- Safe storage is available for those that do not want to store at home
- Protect against uncertainty; as in the stagflationary 1970s and the increasing instability in the 2000s
- Better than paper money; in an environment of loose monetary policy and competitive devaluations, gold can offer protection.
More specifically, it is believed that gold is a good hedge against:
- Systemic risk
However, there is pretty strong evidence against all of these.
Gold as an inflation hedge
This might have worked in the great 1970s stagflation. There is also some reason to believe that the post financial crisis surge was in anticipation of an accelerating inflation, the supposed product of loose monetary policies.
However, that inflation acceleration never came, which is why gold topped at over $1900 per ounce and has since moved back to a low of $1050 in December last year.
However, these periods seem to be the exception, rather than the rule (Reuters):
In England, over the past four centuries, gold lost its purchasing power in every period of inflation; by anything from 21 percent (in both 1675-1695 and 1752-1776) to 67 percent (1897-1920). The story is similar for four periods out of five in the U.S., from 6 percent (1861-1864) to 70 percent (1897-1920). The outlier here is the period of 1951-1976 when gold's purchasing power did increase in the U.S., by 80 percent as the price corrected after its sustained period pegged to $35/ounce.
Systemic analysis doesn't show it conclusively (Forbes):
Authors Claude Erb and Campbell Harvey, a professor at Duke University's Fuqua School of Business, conclude that the assumptions of most investors - that gold rises during times of inflation, or serves as a hedge against a collapsing dollar - don't measure up.
Inflation hedge? Nope. A chart of gold prices versus year-to-year changes in inflation - one measure of unexpected inflation - shows a random cloud. Another chart showing the rolling 10-year return on gold versus Consumer Price Index also shows a random pattern
That same article also produces some reason for concern (emphasis ours):
One way to evaluate the price of gold is to express it as a ratio of the nominal price to the level of the CPI. On that scale, gold traded at 3.36 times the CPI after futures trading was legalized again in 1975 in the U.S. It soared to nearly 9x the CPI in 1980, then plunged, hitting a low of 1.46 times in 2001. Since then it has skyrocketed again, to a current 7 times the CPI. The average over that period has been 3.2, meaning unless inflation heats up (and the caveats above prove false) gold's price likely has a lot more reverting to do before it hits the mean. Over the very long term, the authors note, gold does hold its value. The problem is, none of us will live that long.
So either inflation takes off, or gold is likely to revert to the mean in its relation with the CPI. Actually we think this crucially depends on what kind of time scale one is considering, there doesn't seem to be any immediate pressure for either to happen anytime soon.
Gold as a deflationary hedge
This is more difficult to test as we haven't really had much deflation, especially since the dollar came off gold in 1971.
Goldbugs argue that gold was a good hedge during the Great Depression (from Goldcore):
President Roosevelt revalued gold by some 70% and devalued the dollar by some 70%, from $20.67/oz to $35/oz. It is worth remembering that the Dow Jones fell by 90% from peak to trough and property prices fell by more than 50% in the early years of the Great Depression.
We aren't convinced by this. For starters, it seems odd to present gold as both a hedge against inflation and deflation.
Second, while there was indeed rampant deflation during the Great Depression, the rise in the gold price was the result of an administrative decision to escape the constraints of the gold standard. Still, according to Reuters:
In each of the four deflationary periods since the 17th century in England, gold has increased its purchasing power, by between 42 percent (1658-1669) and 251 percent (1920-1933).
In the U.S. there have been three recorded deflationary periods - and gold increased its purchasing power in each of them - by between 44 percent (1929-1933) and 100 percent (1814-1830).
It could understandably be argued that these results stem from periods when the gold price was fixed, but Leyland notes that while a fixed price could help in the short term, there is a mass of economic experience showing that a price cannot remain fixed in the face of overwhelming fundamental forces.
Gold as hedge against systemic risk
Systemic risk is the possibility of large crashes of financial markets, like the ones we experienced in 2008. Does gold offer protection? Not so (Daniels trading):
When a market crashes everything gets sold except for "flight to safety" investments, and sometimes those are not even safe. When the markets were falling in late 2008, the only assets that appreciated were the USD and Treasuries. Gold, viewed by some as the ultimate "safe" investment, traded to annual lows into the $600s during the last stock market crash.
We can't really infer much in the way of conclusions from the Great Depression, as much of the world was on the gold standard, and this functioned as a considerable constraint as we argued above.
So there isn't really much in the way of evidence. However, recessions can be seen as mild systemic risks, and here there isn't much of a correlation with gold prices either (WSJ):
But Mssrs. Barro and Misra say they've been crunching the numbers and found gold moves independently from whatever the economy is doing. Were gold to be a good hedge against recessions, its price would shoot up as economic growth and consumption plummet. Yet, between 1836 and 2011, they showed no correlation.
Our preliminary conclusion is that if correlations exist, they operate at a way too large time scale to be of much use for investors.
Simple and temporary heuristics
Something else caught our attention in these studies, and it's this (Forbes):
Authors Claude Erb and Campbell Harvey, a professor at Duke University's Fuqua School of Business, conclude that the assumptions of most investors - that gold rises during times of inflation, or serves as a hedge against a collapsing dollar - don't measure up. The most likely explanation for why gold prices go up is because gold prices are going up... Gold, like homes during the housing bubble, displays what economists call "positive price elasticity." When the price is rising, investors are attracted to gold and buy more.
That is, the gold price seems to be subject to some self-reinforcing dynamic, at least for a while. This suggests the working of some heuristics, a rationalization that serves to induce people to buy gold.
As many people are probably not familiar with these studies suggesting no correlation with inflation, deflation, or systemic risk, many can be induced to buy when some of these conditions seem to be nearing.
The market for gold can be moved by a relatively small group with homogeneous expectations and strongly held beliefs.
For instance, after the financial crisis central banks embarked on unprecedented monetary policies. There is a group of investors, people who are drawn to Austrian economics, who are deeply suspicious of active business cycle intervention during good times, let alone the unprecedented monetary expansion post financial crisis.
Many people in this group almost tend to equate 'money printing' with inflation, and strongly belief that currency wars and resulting 'currency debasement' are upon us.
This would seem to explain the great surge above $1900 per ounce in the gold price post-financial crisis, and the collapse to $1050 last December, when these expectations failed to materialize.
So the inflation hedge heuristics was active in that period, until evidence became overwhelming that it wasn't working as it was supposed to, the expected accelerating inflation (let alone hyperinflation) didn't materialize.
If long-term correlations aren't all that useful and in the short-term, gold prices are more the result of simple heuristics that tend to self-reinforce, can we tease from recent gold price reactions what particular heuristic is governing gold?
For instance, one could expect the systemic risk heuristics to have been operative during two recent periods of market turmoil:
Two such periods:
- The sharp fall in the S&P500 in the second half of August 2015 (the Chinese devaluation scare).
- The market turmoil from the end of December 2015 to mid February 2016.
How did gold perform during these periods?
As one can see in the chart above, both periods displayed a rise in the price of gold. Especially the second period there was a sustained rise from roughly $1075 to $1250, a rise over 15%.
The fall in May seems to be related to the revived expectations of a Fed hike, and this suggest yet another short-term heuristic operative, gold as a hedge against deflation.
The latter is corroborated when the June jobs figures where surprisingly strong and, gold dropped almost 2% on the day these figures were made public on July 8.
While systematic studies have not confirmed the role of gold as a hedge against inflation, deflation or systemic risks, the relatively small size of the gold market and the presence of investors with strong views can nevertheless create fairly strong temporary heuristics that tend to feed on themselves.
The self-feeding nature isn't actually surprising, as gold lacks a clear valuation metric (like yield or price-earnings ratios) of other asset classes, so it tends to become self-referential with the heuristics providing rationalization.
While these heuristics can work on a temporary basis, they are not without risk. Once again, here is the Forbes article citing systemic research by Barro and Misra:
In fact, Mssrs. Barro and Misra's figures show that gold prices swing around almost as much as stock prices-meaning it isn't the stable investment many think it to be-and yet their return is much closer to what ultrasafe U.S. Treasurys offer. From 1836 to 2011, the standard deviation of annual gold returns, which gauge how much prices swing compared to their average, was 13.7%, just 3% short of the temperamental U.S. stock market.
While having no yield, gold is almost as volatile as stocks. This isn't really a hedge we would be comfortable with longer-term.
For now, gold as a deflation hedge might work, but even that is under threat from a tightening US labor market and inflation creeping up.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.