Thursday, August 16, 2012

What Goes Up, Must Go Up Forever? Commodity Prices

There appears to be a prevailing market belief that commodity prices can only rise in the long-term, despite short-term fluctuations. Inelastic demand from developed countries and unremitting demand growth from emerging markets are the most commonly cited reasons for this. As a result, investors are putting companies in this space on a pedestal, taking recent earnings growth of such firms for granted. Value investors must be careful not to get caught up in this game of rising commodity prices leading to rising earnings expectations; it can result in portfolio disaster.

One commodity in particular that well-represents the general consensus of that of commodity bulls is the oil market. Demand for oil is indeed inelastic in the short-term; one cannot design a more fuel-efficient car overnight, and nor does one upgrade his car overnight in response to a change in the price of oil. But over the long-term, the market for oil is like any other market in that it responds to price signals.

To see this in action, consider US oil imports over the last several years (click to enlarge):


In the chart above, it is clear that oil consumers react to high oil prices, it just takes them time. In both the early 80's, and in the last five years or so, high oil prices resulted in down-trending consumption in the ensuing years. Though it may be argued that part of the reason for the fall in consumption is thanks to the recession, it is worth noting that US GDP today is around 50% higher than it was in the mid to late 1990's, which was the last time the US was importing so little oil according to the chart.

Of course, making up for this demand are emerging markets like China and India, where economic growth is strong. Despite this, however, global "proved" oil reserves are actually increasing despite these draw-downs.

But what happens if these emerging markets have not conquered the business cycle? If growth slows or a recession is experienced in these economic behemoths, expect there to be a lot of oil for sale without a lot of takers.

Furthermore, as oil prices currently remain high, a number of new technologies are about to make it possible to consume less oil without lowering our cushy living standards. For example, hybrids and electric vehicles continue to improve in fuel efficiency and price. Since transportation fuels make up more than 70% of oil consumption, expect technology changes in this space to meaningfully reduce oil consumption in the coming years.

Though I am clearly a long-term bear on oil, the end result of all this may very well be that oil (and other commodity) prices will continue to rise as they have over the last decade. But hopefully it is clear that whether this will happen or not is not clear at all. The future is uncertain, and considering the cyclical nature of this industry, it would be very dangerous to extrapolate the last ten years into the future.

For example, can you really predict what an oil services firm will earn five or ten years from now with any reasonable standard of deviation? Probably not. But if you pay 10 times earnings for such a company, you are implicitly saying that you can.

Commodity prices are volatile and difficult to predict over the next week, let alone over the next few years. As such, it is best for investors to stay away. Why place a bet when the odds are uncertain? Place your bets when the odds are in your favour.

6 comments:

Anonymous said...

The real price of oil has been declining for the past 100 years.

Anonymous said...

Saj,

Wouldn't it be reasonable to assume that commodity prices will rise at roughly the rate of overall inflation over the long term? In the modern world "stable prices" is taken to mean "stable inflation", let's say at 2% per year. So over the long term, it seems undeniable that commodity prices, just like all other prices, can be expected to rise over the long term.

- aagold

Saj Karsan said...

Hi aa,

I don't really agree with that. While the money supply will likely be grown such that the price of a basket of consumer goods will rise at 2%/year, there's no reason why every commodity should come in at that level. Due to unpredictable technological changes on both the demand and supply side, I'd expect various commodities to come in wildly above or below that rate of growth.

Anonymous said...

Saj,

Wouldn't your logic apply to any product or service produced by any company? For example, let's say a company is in the business of making office furniture. I'd say it would be reasonable to assume that the price of office furniture will tend to rise with the overall rate of CPI inflation over the long term. Does that mean I know the precise rate of office furniture inflation? No. Is it possible that due to "unpredictable technological changes on both the demand and supply side" that the price trajectory of office furniture may be significantly different than CPI inflation? Yes. But if we have no other information to work with that biases our expectations of the future, then I think it's reasonable to use the overall level of price inflation as a going-forward estimate.

Now I do acknowledge that commodity prices are more *volatile* than manufactured goods. So over the short term I agree that commodity prices are unpredictable. But over the long term, commodity prices (particularly a diversified basket of commodities) should rise with the money supply and overall CPI inflation.

- aagold

Potato said...

Your oil imports graph confounds changes in demand and production. High prices curtail demand, but also lead to additional domestic supply coming on-stream. So it looks like a ~25% decrease in imports from 2006-2010 there, but the decrease in demand was only about 10%.

Saj Karsan said...

Hi aa,

I agree with you to a point. Over the long-term, even goods with stable prices are unlikely to match the CPI. But for commodity products that occurs even in the short-term thanks to the relatively extreme volatility you refer to.