The Problems With Gold

Includes: GLD, TLT
by: Robert Singarella Jr.


Many of the commonly believed properties of gold, such as being a hedge against inflation, are false.

Gold is at least as volatile as equities, but has generally had lower returns over long time periods.

It's not worthwhile for investors to have significant gold holdings. The risk is too high given the expected return.

Gold can still be useful (in small proportions) when trying to diversify a portfolio.

If you had invested in gold in 2004, you would have earned about 10.4%, annualized. Yet, I am writing this article to say that you should not consider gold to be a good long-term investment and you should not have a significant proportion of your portfolio in gold. In order to understand why gold is not a good investment, it's best to cover the topic from several different angles. Loosely speaking, I have organized this article around two major themes that are important to understanding gold: the narrative and the risk profile. I'll start with the narrative, then cover the risk profile after that.

When I say "the narrative," with respect to gold, I mean the many things gold is purported to do. There are some widely held beliefs about gold that are simply false; however, they persist. The first, and most common belief, is that gold is a hedge against inflation.

The charts above don't require a great deal of discussion. It's clear that gold is far, far more volatile than inflation and the correlation is weak. There are some who would challenge the CPI data itself, the theory being that the Federal Reserve fakes the numbers to make inflation seem lower. While I cannot, of course, disprove that directly, the burden of proof is on the accuser in this case. It's already been shown that some of the more prominent websites making this claim are faking their own data (see The Trouble With Shadowstats).

What about hyperinflation? It turns out that gold doesn't matter much during hyperinflation either. That's because hyperinflation is not always a monetary phenomenon. To explore this further, I used a paper by Cullen Roche: Hyperinflation - It's More Than Just a Monetary Phenomenon. By studying the historical examples of hyperinflation, he lists the primary causes as:

  1. A ceding of monetary sovereignty (usually in the form of foreign denominated debt, a currency peg, etc.)
  2. Extraordinarily unusual social circumstances (war, regime change, etc.).
  3. Very low levels of faith in government during regime change (high public mistrust).
  4. Ineffective government response or rampant corruption.
  5. Combustible political environment.
  6. A collapse in the domestic economy.
  7. A breakdown in the tax system.

Hyperinflation is not caused by money printing, rather, money printing is the response to the events leading to hyperinflation. Hyperinflation is also not the same thing as very high inflation, there are specific social/political/economic conditions required to trigger hyperinflation. Getting back to gold now. Gold is part of the financial system and therefore, its price and liquidity are fundamentally connected to the rest of the system. Hyperinflation occurs when the economic (therefore financial) system is falling apart. Unlike a currency undergoing hyperinflation, gold would retain some of its value; however, most physical assets would retain at least some value. If only because people outside the country in question still value those assets (in a stable currency). There is no reason to believe that gold would have a significant advantage in such a situation. If you have enough wealth for it to matter, you would likely be on your way out of the country anyway. Not enough people own gold for it to return to being used as a currency in such a situation. During historical periods of hyperinflation, the barter system returned to prominence, not gold. Moving on from inflation, consider another of gold's mythical properties; that it's a safe haven asset.

It's easy to find problems with gold's so-called safe haven status. The chart above shows the price of gold during the financial crisis. We can't say it did horribly (compared to stocks, for example), but it was extremely volatile during the crisis.

GLD Chart

GLD data by YCharts

TLT data by YCharts

Gold was not totally crushed during the crisis, but it was clearly not as good a safe haven as Treasuries (I'm using TLT as a proxy for Treasuries and GLD as a proxy for gold). TLT increased in value during the course of the crisis and was at no point down more than a few percent. Gold was down more than 10% several times during the crisis even if it did finish with a higher price at the end. A safe haven is going to face some risk and some potential losses, but it's problematic to consider something a safe haven if it can fall more than 10% during the time you are relying on it. In a more general sense, gold only benefits from its normal lack of correlation with other asset classes. That's not the same as being a safe haven.

There are other elements of the gold narrative that can be considered; however, having looked at the two biggest elements of the story I think now is a good time to move onto gold's risk profile. I organized the risk factors as follows:

  1. Volatility.
  2. Risk of price stagnation.
  3. Black swan.

Volatility is the easiest, so let's start there. It's not hard to show that gold is more volatile than the S&P 500 index most of the time. The 1-day log return chart shows that very clearly.

Source: Data from Federal Reserve Bank of St. Louis database.

The red line is the S&P 500 (SPX) and the blue is gold. Based on a quick look at the graph, I'd say the volatility for both is of the same order of magnitude, but gold seems to have more high volatility periods (on a side note, both time series exhibit volatility clustering, in other words, conditional heteroskedasticity). The pattern continues for the rolling 1-quarter log returns and 2-year log returns.

Source: Data from Federal Reserve Bank of St. Louis database.

Now, the 10-year log return chart tells us something different. Specifically, it shows signs of the next risk factor, price stagnation.

Source: Data from Federal Reserve Bank of St. Louis database.

Notice that the red line marks zero. In the late '80s to early '90s, gold entered a bear market until about 2004. What's interesting is that the price didn't really decline much over that period, rather, it didn't change very much. When you look at the long-term price chart, it's apparent that the period of stagnation was longer than the 10-year chart shows.

Source: data from Federal Reserve Bank of St. Louis database.

Starting in the early to mid-'80s, gold entered into a long, very slow decline. I would say stagnation is a better term because the rate of decline was so slow. There has been some speculation that the gold ETF, GLD, helped break the price out of its slump by allowing individual investors greater flexibility. In any case, this has happened before and could happen again. I want to be clear, I don't see any reason for the price to stagnate in the same manner, and I think the possibility should be considered. There is at least no obvious reason why it will not happen again.

The last risk factor is a black swan event. There isn't much to say about this because it really could be anything. I don't see any value in trying to guess what sorts of events could be a black swan for gold. The important thing is that it can happen and investors must simply accept that risk. This is no different than for stocks, so I would say the two assets face equal risk from black swans.

After all of that, what's left is the conclusion that gold and equities have a similar risk profile. Gold has a lower long-term return than equities and it provides no cash flow. There are often storage costs associated with gold unless you physically own the gold. The 10-year log return chart above clearly shows that equities provided a better return for a longer period of time than gold has. That doesn't mean gold is worthless to investors though. The concept of holding a small amount of gold in a portfolio for the sake of diversification has been widely discussed, so I will not go into detail. I will point out the tendency for gold to move somewhat inversely relative to stocks in many of the return charts above. That's an example of the non-correlated movement that investors want to take advantage of. I think I have shown that the risk profile of gold does not make it worthwhile for investors to hold a greater proportion of it in their portfolios.

Disclosure: The author has no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.