Commodity industry investing
May 12th, 2011As usual, this doesn’t mean shiny metals, which I have never owned, but rather microeconomic commodities — goods and services that compete only on the basis of price.
It’s a supposed truism that in such industries, perfect competition leads to forever low profits, and thus every company in such industries makes for a bad investment. This is generally right, but as investors, we make money not where things are generally right, but instead where they are occasionally wrong. When faced with a generally accepted rule, the curious and enterprising investor naturally asks, “when is this not true?” In investing, the money is often hidden in the exceptions to received wisdom.
In commodity industries, one big exception to the no-profits rule is that, in the real world, a few companies do gain sustainable cost advantages; in these rare situations, the same features that make the industry profitless for others can actually benefit the low-cost producer.
The situation is easiest to see in pictures. Imagine an industry with 3 competitors. In the graph, each colored block represents a company. The height of each block is that company’s average marginal cost per unit, while the width is the company’s maximum unit capacity. In the real world, even in a commodity industry, each will have slightly different marginal costs. We will exaggerate those differences here to make a point. Superimposed over them is the hoary supply/demand curve from Econ 101.
In this imaginary example, we start with equilibrium (what could be more imaginary than real-world economic equilibrium?). The market is exactly cleared by the combined capacity of the three companies (the width of the three boxes is their aggregate capacity, which adds up to the market-clearing number of units). 
Red has gross profit of zero — its marginal cost is exactly the same as the market-clearing price (dotted line). Yellow makes some profit, and green makes a lot.
Now, imagine that a recession shifts the demand curve down as shown in the second graph (right). Everything else is the same. Green continues to make money. Yellow scrapes by with gross profit near zero. Red is in a world of hurt, running below capacity and losing money, on average, with every unit.
Red may survive for a while by closing its highest-cost plants, which simultaneously reduces its capacity and average cost. If successful, the new equilibrium might look something like the third graph (right).
What if Red figures out how to copy Yellow, cuts costs further, and matches Yellow’s cost structure? Graph 4 (right) shows the result: supply curve is shifted down, price falls to the new, lower marginal cost, someone (we’ll say Red) steps in to satisfy increased demand at that price — but Red still makes no money, and now Yellow makes no money either.
Through all this, Green makes money, because everyone’s selling price is the same (commodity industry), and that price cannot fall below Red’s costs. So Green always makes money. Theoretically. In practice, this is rare, because it’s hard to have a cost advantage that other cannot copy. But this is the source of advantage for most of the companies that defy the odds in commodity industries — Southwest Airlines, POSCO, and so on.
As an investor, one might look at commodity industries, estimate cost per unit for each supplier, and buy a consistently lower-cost producer (only when cheap, of course).
There is a second, less powerful, but also less obvious and less rare type of competitive advantage for commodity producers: balance sheet strength, which confers advantage mainly during shocks, and thus is not given the price premium it deserves. More on that in a future post.
ADDENDUM 5/23/11: An astute reader and fellow Caltech alum points out that in practice, sustainable advantages in commodity industries are often non-scalable. For example, since bricks are heavy and low-priced, shipping is a big percentage of unit cost, so plants have a sustainable cost advantage in selling to nearby customers, but cannot extend this advantage to other markets. As a result, there is no opportunity for internal compounding, i.e. the company cannot productively reinvest in itself. This is absolutely true — the cheap, high-quality business with long growth runout is more attractive. By contrast, low-cost producers in commodity industries, as above, are appealing either as medium term investments (if mispriced) or as reliable yield engines (as long as the company is either paying dividends or buying back shares at rational prices).
it’s also useful to know what exactly enables certain companies to be lowest-cost producers. the best business models, the efficiency kings, allow for international expansion/reinvestment. what stops these companies is not shipping costs but the fact that the world is finite.