Opportunities in commodities come in cycles. For a few years there will be a lot of attractive investment opportunities and then there might not be any for 5 years. But taking advantage of these opportunities can add significantly to your bottom line.
I will split this post into 3 parts. The first part will look at valuing commodities, the second part will look at some historical examples and part three will review the current commodities market.
Reversion to Mean
To keep the model simple we will assume that prices eventually revert back to their mean. We need to use a long term mean with this assumption that will take out or at least reduce the cyclical nature of commodities. For this we can use a long moving average. I like to use 2000 days (8 years of trading days). The chart below shows a 2000 day moving average on Crude Oil:
The chart shows almost 20 years of prices. I could have chosen any commodity for this and ou would have seen a similar pattern. Prices moving over and below the long term average but always crossing back over. It's useful to use this to see where we are in relation to the historical norm.
So if we assume that prices revert back to the norm all we need to do is buy when they are below and sell when they are above? Pretty much, but we have a few more things to add to the model.
As commodities tend to trend for extended periods of time it's important that we wait for a signal to enter. The chart below shows Sugar:
As you can see the price moves below the 2000 day average and continually breaks it's 100 day low support level. The downtrend is clear and we are better waiting for a break out to a 100 day high which happens at the red arrow. After the upside break out the price does fall back again but then increases to return to its 2000 day average within about 12 months from the break out. Entering too early, even when the prices look cheap, can be dangerous as the downtrend may continue for some time and the overall return on the investment can be much lower. It's better to be safe and wait for some kind of signal that the down trend is changing first.
When entering a commodities trade it's important to have a price target as prices are often volatile and can move very fast.
For a price target I will use the 2000 day moving average plus an upper Bollinger Band of 1.282 standard deviations based on the 2000 day moving average. 1.282 standard deviations capture 80% of the data in a normal distribution with 10% of the data above this and 10% below. As you may know commodity and stock prices don't strictly follow a normal distribution but for our model it will suffice.
The chart for Crude Oil below includes the purple price target line of the 2000 moving average plus 1.282 standard deviations:
This purple line will act as our price target.
As a conservative estimate of the annual returns on a commodity investment we will set a target of achieving the price target within 10 years (in reality the average time is about 1 year).
So the estimated annual returns will be:
((Target Price / Entry Price)^ (1/10))-1
So let's say the current price is 60 and your target is 100 your estimated return will be:
= 5.24% a year