Investors continue to search for alpha in one of the most challenging environments in nearly a decade. Due in part to over two decades of massive debt accumulation, which ended in a housing bubble and deep recession, the Fed’s zero-interest-rate policy has prompted investors to search beyond stocks and bonds.

More than ever, investors are embracing alternative investments, including commodities. This had led to an expansion of commodity exchange-traded funds, where commodity investing is relatively easy. Easy perhaps, but to be successful, there are several key issues investors must understand.

What factors affect the price movements of commodity ETFs? Are commodity funds suitable to hold long term? Finally, how well do commodity funds track their benchmarks? We will address these and other questions as we take an in-depth look at ETFs that attempt to track the price of crude oil, natural gas, gold, silver, platinum and palladium.

The Assumptions

To answer these questions, we decided to look under the hood at these commodity ETFs. To be included in this analysis, we decided the funds needed to meet certain criteria. First, the ETF must have a minimum of $100 million in assets. Second, the fund must have at least five years of history. Third, its prospectus benchmark must be a single commodity, rather than a diversified group of commodities. Six energy-related ETFs and eight precious metals funds met the criteria.

Crude Oil and Natural Gas ETFs

Before investing in the energy sector, one should become familiar with some basic terminology.

  • Upstream – Companies engaged in extracting the commodity (i.e., oil and gas drillers)

  • Midstream – Companies that transport and store the commodity (i.e., pipelines)

  • Downstream – Companies that refine and sell the commodity (i.e., refineries and gas stations)

  • Integrated – Companies engaged in both upstream and downstream activities

  • Spot price – The current price of the commodity

  • Contango – When the future price is greater than the spot price

  • Backwardation – When the future price is less than the spot price

The great American oil boom was largely the result of technological advances such as hydraulic fracturing, or fracking, from shale rock. The process of fracking involves injecting liquid into subterranean rock at high pressure to extract oil or gas. Because fracking is more expensive than other extraction techniques, as oil prices fell, it became less profitable.

Thus, the collapse in oil prices has led to an increase in bankruptcies, unemployed workers, loan defaults and a host of additional problems, especially in areas where the boom was strongest. From the peak of oil on June 20, 2014, through April 1, 2016, the average stock market capitalization of the U.S. energy sector has fallen 60.8%.

When investing in an oil or gas ETF, contango and backwardation are two of the most important terms to understand. Contango is a performance headwind and backwardation is a tailwind. Moreover, it is possible for both to exist during a multi-month period.

To explain, let’s assume the spot price of oil in the first month (M-1) is $50, the price in month two (M-2) is $49, and the price in month three (M-3) is $51. Thus, backwardation exists between M-1 and M-2 and contango is present between M-2 and M-3.

Contango and Backwardation

To understand the effect of contango and backwardation, we must first explain how the typical oil or natural gas ETF functions. Rather than own the physical commodity — which would require transportation and storage — these funds will purchase one or more futures contracts. When you purchase a futures contract, you are agreeing to purchase the commodity at a predetermined price and take delivery when the contract expires.

Because these funds never intend to take delivery of the actual commodity, the fund manager will sell the contract prior to expiration and roll the proceeds into new futures contracts. This is the process followed by United States Oil (USO). The ETF United States 12-Month Oil (USL) attempts to reduce the risk of contango by purchasing contracts over a 12-month period. Although this approach can be beneficial, if contango rises too high, investors will still have significant risk.

Effect of Contango, Backwardation on WTI, USO

To illustrate the effect of contango and backwardation on an oil ETF, let’s look at the chart above. It contains daily data from Dec. 31, 2007, to Sept. 30, 2015. Contango and backwardation (left axis) are marked in blue and separated by a red horizontal line (contango is above the line, backwardation is below). Contango and backwardation is the difference between the price of WTI in month two (CL2) minus the current or spot price (CL1). On the right axis, the black line represents the daily price of West Texas Intermediate Crude (WTI) and the yellow line is the United States Oil fund (USO).

Notice how well USO tracked WTI in 2008. This was during a period when backwardation and contango were low. After rising in 2008, WTI ultimately fell 53.52% and USO declined 56.31%. Then, in early 2009, contango spiked dramatically. As a result, WTI and USO deviated significantly.

From Dec. 23, 2008, to June 11, 2009, the price of WTI surged 140.06%. USO only gained 32.27%. Because USL reduces the risk of contango by purchasing contracts over the ensuing 12 months, its return (36.6%) was slightly higher. Though USL was 4.33% higher than USO, it still lagged WTI by more than 103%. The divergence between oil and the oil ETFs was due to the spike in contango. As contango rises, an oil ETF’s ability to track the price of crude deteriorates.

It is also worth mentioning that during this entire period, contango was present 78.8% of the time. When is the best time to purchase this type of fund? The best conditions are when contango is low or backwardation is present and crude oil or natural gas are about to rise.

Our Oil and Natural Gas ETFs

We just discussed the mechanics of these two funds and explained some of the forces that affect their performance. Now we will examine the rest of the group to see how well they tracked their commodity.

Total Return of Oil ETFs, WTI

The chart above includes the four ETFs that attempt to track the price of WTI. United States Brent Oil (BNO) is not included since its benchmark is Brent Crude. In addition, United States Natural Gas ETF is also not on the chart. The period referenced in the chart begins Dec. 6, 2007, the inception date of USL, the newest of the WTI-tracking funds. This period was unfavorable for oil prices, as the price of WTI declined 57.9%. During the same time, USL fell 67.9%, PowerShares DB Oil (DBO) lost 76.9%, USO dropped 85.9%, and iPath S&P GSCI Crude Oil (OIL) lost a staggering 90%.

BNO, the fund that attempts to track Brent Sea Crude, seemed to track its benchmark slightly better. However, contango has been relatively tame since BNO’s inception. United States Natural Gas (UNG) was launched on April 18, 2007. From UNG’s inception through March 28, 2016, the price of natural gas fell 76.4%, while UNG lost a whopping 98.4%.

For the reasons we have discussed, a buy-and-hold strategy is not suitable for this type of fund. They are more appropriate for a short-term hold or an active trading strategy.

Precious Metals ETFs

The first gold ETF created, SPDR Gold Shares (GLD), began Nov. 18, 2004, as a joint venture between State Street Global in Boston and the World Gold Council in London. Two months later, Jan. 21, 2005, BlackRock introduced its first gold ETF, iShares Gold Trust (IAU).

On April 21, 2006, BlackRock launched an ETF to track the price of silver, iShares Silver Trust (SLV). A third ETF sponsor, PowerShares, created its gold ETF (DGL) in January 2007. Finally, ETF Securities entered the fray in 2009 with ETFS Physical Silver (SIVR) and ETFS Physical Swiss Gold (SGOL); and in 2010 with ETFS Physical Platinum (PPLT) and ETFS Physical Palladium (PALL).

Unlike the oil and natural gas ETFs discussed earlier, these funds do not purchase futures contracts. Instead, they all purchase the physical commodity, which backs 100% of their outstanding shares. (See “Shiny Storage Lockers” sidebar below for more on where and how the precious metals ETF sponsors store their loot.)

Insurance and Expenses

It is important to realize that the goal of these funds is to reflect the performance of the underlying commodity, minus any expenses associated with operating the fund. Most expenses are included in the fund’s expense ratio, such as vault storage fees, transportation costs, marketing costs, trustee fees, etc. It is also important to note that the company may or may not insure the commodity stored in its vaults. According to its prospectus, GLD does not maintain insurance on the gold backing its fund. ETF Securities also does not insure its gold. The iShares Gold Trust (IAU) does insure its gold, but may discontinue its coverage with adequate written notice (for more information, see page 27 of IAU’s prospectus). If insurance is important, you should contact the specific ETF sponsor.

Gold ETFs

Total Return of Gold ETFs

The chart above compares the performance of the four gold ETFs with physical gold, from Sept. 9, 2011 — the inception date of the newest gold ETF — to March 31, 2016. During the first few years, each fund seemed to track the price of gold very closely. However, sometime around the latter part of 2011, DGL began to lag the others.

The chart, covering 2,395 days or approximately 6.56 years, includes the four gold tracker ETFs and the spot price of gold.

During the entire period, the price of gold rose 22.16% (cumulative return), IAU surpassed gold by a single basis point, returning 22.17%, SGOL and GLD returned 21.31% and 21.18% respectively, and DGL lagged the group, rising 12.99%. Why was that the case?

In translating their cumulative return shown in the chart above to an annualized return, we find gold and IAU averaged approximately 3.08% per year, SGOL and GLD averaged 2.95% annually and DGL’s average annual return was 1.88%. Therefore, DGL lagged physical gold by about 1.20% per year (9.17% cumulative), which is 47 basis points more than can be explained by subtracting its current expense ratio from gold’s annual return.

Silver and Palladium ETFs

Although the chart is not included here, both silver funds tracked their commodity very well. In fact, their correlation with the price of silver from July 24, 2009, to April 1, 2016, was 0.998. The platinum fund from ETF Securities (PPLT) tracked its commodity extremely well. Since its inception, compared to the S&P GSCI Platinum Index (not its official target benchmark), the index lost 40.7%, while PPLT dropped 40%. ETF Securities Palladium fund (PALL) has the same inception date as PPLT. Over the same period, it gained 14.79% compared to the S&P GSCI Palladium Index (not its official benchmark), which gained 13.57%.

So Why the Disparities?

After a thorough review, the crude oil funds did a poor job of tracking oil over a long period, primarily because they purchase oil futures contracts rather than the actual commodity. The same is true for the natural gas fund we reviewed. Contango has a particularly negative effect on these funds, and the higher contango rises, the worse it gets. Therefore, the crude oil and natural gas funds are best suited for a short-term, tactical investment or a sound trading strategy.

Conversely, most of the precious metals funds we studied did very well tracking their benchmark. This is because they back the funds with the physical commodity. Therefore, investors in these funds can expect a return close to the actual commodity, minus the funds’ expenses. There was one small exception noted earlier.

Armed with the knowledge of how these funds operate we now have a better understanding of what to expect. To be successful investing in precious metals ETFs, one needs an understanding of how the commodity is used and what factors affect its supply and demand. However, that is a subject for another time.