My blog name is "The Commodity Strategist." I chose that because I specialize in taking very duration positions in commodity related assets. In particular I am willing to buy cheap and sell dear even with no foreseeable catalyst. This usually requires patience, but I have a lot of that.
The reason I mention my style is that I've had a few very lucky SA hits recently (nickel, coal, South32), and I've accumulated a bunch of followers. This large-cap and well covered recommendation is not likely to run as hard as fast, so keep that in mind.
About three years ago I wrote a bearish piece on John Deere (NYSE:DE). I was roundly denounced by the Deere fanboys, but things worked out well for us shorts. In the last three years, DE has underperformed the S&P by about 50%! A lot of this was due to initial overvaluation, but some is because of hard times in the US farm sector. For example, corn prices on the CBOT have fallen from about $4.50 to about $3.40. Soybean prices have fallen from $13 to $9.50. This has been bad for both farmers and for their suppliers.
In my 35 or so years investing in commodities, I have experienced firsthand how mean reverting they are. In fact, I'm hard pressed to find another asset class as mean reverting. The economists have a phrase for this, the "cobweb pattern," describing how the graph of volume vs. price looks. My thesis is that we are in the undervalued part of the agriculture cycle.
Here's some graphs that I use to measure a commodity's valuation. For those who haven't seen these before, the x-axis is the price in 2016 dollars, adjusted for both US inflation and the change in value of the $US. The y-axis is the forward 10-year price change (again adjusted for inflation and the $US). The graphs are monthly, so the last point is in Oct 2006. The red line is where we are now.
I'm always surprised at how good the fits on these graphs are. We are talking R-squareds in the .70s or so. Given where prices are now, the next ten years should see about a 50% increase in before-inflation corn prices.
And here's soybeans. It's not as bullish, probably because of the competition from weak-currency countries, especially Brazil:
I've saved the best for last, wheat:
Wheat is literally off the charts.
Here's another way of showing that prices are too cheap. This table, based on USDA data, calculates how much the average farm makes or loses on their operations. I have calculated it both with and without the cost of the land. Land, of course is a derived asset based on the profit of the farm operations it supports. You can see that at current land prices, all major crop operations are unprofitable. Even with land at zero value, only soybeans are really profitable. This is not a situation that can continue indefinitely.
|Trend Yield, bu/Acre||166||45||48|
|Current Farm Price||3.18||9.28||3.77|
|Gain/Loss / bu.|
Ok, I hope I've convinced you that the major US ag products are cheap. How to play this?
I almost always advise people against either actively trading or passively investing in commodity futures or commodity ETFs. First, the forward contracts have substantial contango built into them. This makes sense; the commodity has to be stored and storage and insurance don't come for free. So that limits your gains straight off. Second, trading costs are higher than large cap equities. You not only have to buy your initial stake, but you have to pay to keep rolling it forward as well. BTW, the commodity ETFs have the same issue. You may not pay the trading costs directly, but the ETF pays them, and it subtracts from their NAVs.
Another avenue I don't recommend is investing in the farmland REITS that have started in the last few years. I believe that at current crop prices, farmland is still overvalued. The REITS have been buying their acres in the last few years at high prices. It may take quite awhile until they get back to their cost.
A better way to pick a bottom in the ags is to buy one of the suppliers. I choose Deere . DE has a tremendous franchise in its space. Many will agree that it is the class of the industry. The problem is that the industry is in trouble. When farmers' income goes down, they stop purchasing new equipment. Also, the boom times of a few years ago led to way above trend purchases of new equipment. Since much equipment is traded in, this led to a big reduction in the price of used. With all the competition from used, Deere's sales plummeted. Here are some tables from Greg Paterson, aka "Machinery Pete." First, auction prices of S680 combines.
For you city slickers, this is what an S680 looks like:
Next, here's the auction prices of a smaller machine, the 568 round baler.
Those are some pretty hefty price declines. Now there may be some non-homogeneity in the table's machinery (i.e. average age and less functionality), but it still was very bad for Deere.
The market for farm machinery is something like the truck market. There is a normal wearing-out and obsolescence of old machinery. Sometimes the production of new machines exceeds that, like it did in 2012. Sometimes it is lower, like I expect it is now. This feeds into inventories of farm equipment. Here's a graph of Deere's sales vs. manufactures' US inventories of farm equipment from the census bureau:
You can see the close correlation between inventories and DE's sales. The question is whether inventories have been drawn down enough to turn around sales. I think they have, since inventories are back to levels at the end of the great recession in 2010.
Deere gets about a third of its sales from ex-US, with Brazil being particularly important. The farming economic situation in Brazil is similar to the US, with the added kicker of an even weirder national political problem (weirder than the US, barely). I have no idea how the politics will sort themselves out, but the Brazilians have been thru worse, and I'm sure they will survive.
If I'm right, what kind of numbers can we expect from DE? If past cycles are any indication, I believe DE's revenues could rise by 25% in the next three years. Given the cost reductions put through, this could lead to yearly EPS of about 7.50 per share. Using a P/E of 14.25 (the long-term average), this would equate to a price of 107. Including dividends, that's about 11% per annum.
That alone is not too bad, but I believe there is a fair probability of a much bigger upcycle. This would required a jump in crop values - not as farfetched as it sounds. World crop yields have been well above trend for four years now. One of these seasons, we are going to get only trend yields, or maybe worse. That is the lottery ticket that would pay much higher returns.
When buying the downcycle in commodities, I always buy the strong names, the ones that have the staying power to still be there when things get better. I believe Deere is in that club.
Disclosure: I am/we are long DE.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.