1

Suppose I alone know that the price of a commodity will double next month, and will then stay at that level for years. There are two simple ways I could exploit this knowledge:

  • Buy the physical commodity (e.g. ingots, bushels, physical commodity ETFs, etc.)
  • Buy shares of companies that produce the commodity.

When considering common and easily-stored commodities such as gold, silver and palladium, both investment options are available (owning the physical metal and buying shares of mining companies). However, there are lots of other commodities where owning the physical commodity is highly impractical, which means that one has to buy shares of production companies to gain exposure to the commodity. For example, petroleum products, uranium, molybdenum, zircon, cattle, etc. Is there usually a difference in returns between owning a basket of the commodity production companies versus owning the physical commodity?

My intuition says that buying the physical commodity will yield superior returns because the exposure to commodity prices is unhedged. In contrast, the commodity production companies may have internally hedged their operations against falling commodity prices (e.g. by shorting commodity futures), which will blunt their returns when the commodity price rises. Is this the case?

1

OP, your two items are not really correct...

There are three things you can do

  1. buy the item physically

  2. buy and trade commodities futures

  3. as you mention, buy stocks relevant to the commodity (miners, agri-business, pipeline building companies, etc)

Far and away the most common is #2.

If someone says "I trade gold for a living" what they mean is they trade the commodities markets, i.e. the enormous commodities futures markets.

A tiny number of people actually do #1.

Regarding #3 it's a general sort of approach with no firm connection to the commodity price.

Company prices can go up and down for a zillion reasons.

(Consider this ... traders will say things like "Since political party A will probably win the election, sector S will probably go up", or, "Due to a cold winter, it's likely sector Z will go down". It's just a "trading thought". There is no direct or firm connection to the commodity price. And don't forget too there's the issue that markets already price in stuff that will happen tomorrow. If you have the idea "the S&P will decline due to covid", it is too late to trade, it has already declined due to covid. Note too that with say miners, it all depends on the mine operations; you trade mining stocks based on "I think hole in the ground X will be special" or "I have quietly heard they are about to buy area A" etc. Finally note that, indeed, some mining companies are structured so that the company itself is not sensitive to the price of the commodity; they hedge that away in both directions or (for example) pre-sell all the output: the company is only about the operations and not price fluctuations.)

So, if you literally as in your example knew the price of gold was going to go up: Without a doubt, you'd simply trade that (i.e. on the futures markets). There would be no point whatsoever in buying physical metal.

You ask about leverage, in the three approaches

  1. There is no leverage, it's the pure price

  2. You can simply choose mathematically how much or little leverage you wish by using different instruments (different targets, series, options, etc)

  3. There's no direct connection to the price, it's more of a "sector bet"; anything can happen.

Hence to literally answer your question

Suppose I alone know that the price of a commodity will double next month... There are two simple ways I could exploit this knowledge. Buy the physical commodity. Buy shares of companies that produce the commodity

That's incorrect, there are two ways: (A) buy the metal (B) simply trade it (i.e. on the futures markets).

Without a doubt, you'd just trade it on the futures markets, because you could have leverage (indeed, choose the leverage you wish).

Regarding buying shares of companies in the sector, it's neither here nor there, anything can and does happen with stock prices.

And yes, the final sentence of your last paragraph is 100% correct, as I mention above. Some (not all) commodities-related companies explicitly hedge-away the price aspect in various ways, by company structuring, explicit hedging or the like. But do note that even setting aside that issue, you can't bet on commodities prices by betting on production-related companies; it's just a soft connection, "a trading idea" like any other trading idea; stocks can and do move for a zillion reasons. (As you can probably tell, actual traders - i.e., commodity traders :) - think stocks are a flakey medieval pile of mysticism with no rational behavior!)

3
  • 1
    Great answer - the only thing I suggest to add, is that you imply, but don't outright state, that some companies explicitly do not hedge their operations. This decision to hedge or not the commodity price impact of operations should be a publicly known statement, probably in every annual and quarterly financial statement discussion, so you shouldn't need to assume whether the company hedges its commodity risk or not. – Grade 'Eh' Bacon Aug 21 '20 at 12:56
  • 1
    G.B. - very true. You know, I think the thrust of my answer is this ... OP thought that there were two ways one might invest in "a gold price move" - metal or stocks. In fact, both are uncommon. In 99.999% of cases to invest in a gold price move, you just trade the commodities markets. So, that is what I wanted to emphasize to our fine OP. Fortunately other, possibly more sober, contributors such as yourself have added fine detail in the comments! :) – Fattie Aug 21 '20 at 13:04
  • 1
    I'd also add options on commodities futures... – D Stanley Aug 21 '20 at 13:13
1

The highest return you can get would be by using options. Say we are talking gold. GLD trading at about $185, and you think it will double? The $280 option is priced at 5 cents.

At $370, each $5 contract will be worth $9000.

You betting $5000? Congrats on the $9000000. Now pay Uncle Sam his cut.

4
  • A wonderful example! (Wonderful to think about! :) ) For the edification of our OP, it's a great "paper trading" example. You know, I think it would be likely impossible to buy 1000 of those today; a handful would fill and then the price would skyrocket. No market maker would put out such a line, and if some civilians had a few offers (trying to collect the odd buck) they'd never have enough margin to do more than a handful at their broker! But yes, people can and do win 100x or even 1000x bets on options (the bastards). – Fattie Aug 21 '20 at 19:51
  • (I googled "famous stories of huge options trading wins" and found things like cnbc.com/2015/06/26/… and bloomberg.com/news/articles/2019-10-17/a-guy-on-reddit-turns-766-into-107-758-on-two-options-trades ) – Fattie Aug 21 '20 at 19:54
  • I honestly don’t know how many contracts it would take to really impact the price. Years ago, 1999, I had 1000 contracts for an option, $15000 total cost. The stock flew, and it was worth $750K at the top. By the time I sold, I got $50K. This was my best and worst option trade ever. – JTP - Apologise to Monica Aug 21 '20 at 20:01
  • A H H fantastic story ! – Fattie Aug 21 '20 at 20:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.