Finding an edge in commodities: a tale of two futures curves

“Show me the incentive and I will show you the outcome.” – Charlie Munger

Warren Buffett’s long-term investment partner was well known for his ‘zingers’, which neatly encapsulated his learnings from a long and successful investment management career. For commodity investors, the quote above is worth keeping in mind.

Having explained backwardation in the first instalment of this blog, let’s now look at two specific commodities, and the structural reasons why they typically display very different futures curves.

Gold: the cost of carry

When we think about factors influencing the price of gold, there are several obvious contenders: market sentiment, inflation expectations or the relative strength of the US dollar.

But when considering the typical futures curve for gold, it’s valuable to take a step back from these near-term variables that influence the spot price and to focus instead on the fundamental structural factors – the ‘incentives’, to use Munger’s phrase.

It’s estimated that around 200,000 tonnes of gold has been mined to date, while estimated reserves yet to be mined amount to around 50,000 tonnes.¹ Gold is virtually indestructible, meaning almost all of the metal that’s been mined is still around in one form or another.

For any quiz fans among you, if all the gold mined to date was melted down and turned into a cube, it’d be 22 metres along each edge.²

This is important for the future curve because it means a seller can be expected to incur costs for storage and security of these pre-existing stocks of gold, which dwarf the additional reserves that are mined each year. All else equal, the price of gold in distant-delivery months should be higher than the price for near-delivery, resulting in a drag on performance for investors using futures to access the market.

As shown above, gold typically displays this upward-sloping future curve. This doesn’t mean gold is never in backwardation, but the structural drivers explain why contango should be the base-case assumption.

Oil: why not leave it in the ground?

The fundamental structural factors influencing oil’s futures curve are very different. Unlike gold, oil is destroyed when it’s used, meaning the oil already extracted is almost entirely absent from the market.

This means oil miners must be incentivised to incur the expense of extraction and forgo the opportunity of potential rises in the spot price. Oil has been safely stored underground in the form of unexploited reserves for hundreds of millions of years, so the owner might reasonably leave it there, awaiting higher prices in the future.

As a result, buyers pay a premium for near delivery to compensate sellers.

Again, this doesn’t mean oil is always in backwardation – many other factors influence the future curve – but structural factors mean distant-delivery months often show lower prices.

The portfolio perspective

In the context of broad commodity portfolios, gold and oil provide straightforward examples of how structural factors can influence expectations of positive or negative carry.

By examining the wealth of data available on the long-term price history of various commodities in many different market environments, strategies can be constructed that attempt to optimise exposure by dynamically selecting the expiry date of futures contracts.

To end with another of Munger’s zingers, “Opportunity comes to the prepared mind.”

1. Source: https://www.mining.com/web/chart-how-much-gold-is-in-the-world/

2. Ibid.