As a deep value investor, there's nothing that makes me giddier than seeing the stock of a great company fall nearly 25% in a matter of one week. That's precisely what has happened in the potash sector (SOIL), with long-time stalwarts such as Potash of Saskatchewan (POT), Mosaic (MOS), and Agrium (AGU), as well as American up-start Intrepid Potash (IPI), all getting hammered last week. Any time there's a plunge on that level, it's worthwhile to investigate. Unfortunately, the evidence suggests that the entire potash sector is still overpriced, even in spite of the recent drop.
The freefall in the potash sector was initiated by a surprise announcement from Russian producer Uralkali (OTC:URALL), which announced that it would end export sales through the Belarusian Potash Company; a partnership with Belarusian producer, Belaruskali. The announcement signaled the end of one of the world's two dominant potash cartels, as Uralkali decided to embark on a strategy of selling a higher volume at a lower price to boost profitability.
There are many reasons why this sudden plunge in potash stocks, on the face of it, could conceivably be great news for value investors. The P/E multiples are attractive for almost all of the potash stocks now. Likewise, the cartel's disbanding ultimately has less of an impact on pricing than commonly perceived.
Yet, while the market may be 'over-reacting' to this particular bit of news, it may be under-reacting to the broader trend of weakening demand in the potash market. Potash miners have been huge beneficiaries of China's massive debt-fueled boom. As China's problems became greater, it's likely that demand for potash will weaken significantly. Indeed, Uralkali's decision to end the cartel may be a sign that this is happening already.
The potash miners are still intriguing because regardless of what happens in China, there will be growing demand for these agricultural minerals in other parts of the developing world. Yet, there is a huge potential for margin destruction in the next few years. Margins are near historic highs throughout the entire sector, making almost all of these companies extraordinarily vulnerable to falling prices. That's why even after the 25%+ plunge, potash stocks still look expensive.
A History of Margins
To understand why potash miners may be vulnerable, let's take a look at historic gross margins in the sector. We'll start out with Potash of Saskatchewan, which has the longest history as a publicly-traded company. Below, you can see Potash's gross margins dating back from 1995.
The blue line shows the average gross margin for this entire time frame. As you can see, Potash's 2012 margin of 43% is well above the historical average of 29.5%. That suggests there's significant room for margin contraction.
Mosaic, another large potash producer, has a data series that spans back to 2003 so we can compare POT vs. MOS from '03 onwards. You can see the gross margins of both below, as well as the average gross margin from 2003 - 2012.
(click to enlarge)POT's margins have almost always been higher than MOS's, with one brief exception in 2009. It's clear that even using the average margin from 2003 onwards (which is mostly dominated by a boom market), both POT and MOS are well above historical margin levels. POT's average margin in that time frame was 33.9% (vs. 2012 margin of 43%), while MOS's average margin was 20.0% (vs. 2012 margin of 27.8%).
Finally, we can compare POT, MOS, and Intrepid Potash from 2006 onwards. You can see those figures below.
I removed the averages from this chart to make it a bit more readable, but even using this abbreviated time frame, gross margins still look inflated. The average margins from 2006 - 2012 for POT, MOS, and IPI were 39.1%, 24.5%, and 29.7% respectively. That's compared to the 2012 margins of 43.0%, 27.8%, and 37.7%. As you can see, whether we use the long-term averages (dating back to 1995), or take a more short-term approach (2006 onwards), gross margins appear inflated by historic norms. Hence, there is a very real danger of mean reversion here that could hit the potash producers harder than the market is expecting.
Even using the time frames above, we may still be overstating long-term margins. The Potash of Saskatchewan data series begins in 1995, which appears to have been a bit of a mini-boom year for commodities. That data ignores the weak commodities environment that existed from 1990 to 1994.
I could not find freely available data on the web showing potash prices prior to 1995, but I have explored this issue with copper before. In my 2011 article, "Copper Producers Could Still Have a Long Way to Fall," I showed how copper prices began to fall rapidly after the end of Japan's asset bubble in 1989, before a brief rebound around 1994 and 1995.
The chart below examines CPI-adjusted copper prices from 1981 to 2011. If copper struggled from 1990 to 1994, I'm willing to wager that potash did to some extent, as well.
If a similar trend happened with potash, it may suggest that long-run margins are even lower than the 17-yr average for POT. That means that even our rather pessimistic long-run average for POT may be too optimistic.
Spot Prices vs. Margins
We can also take a look at KCI spot prices for potash. The chart below shows spot prices since 2002.
As you can see, prices began to skyrocket from about 2008 onwards. The only issue I have with this chart is that the price movements become so extreme after 2008, that's it's difficult to compare with margins, without averaging the info out. Using data from infomine.com, I attempted to piece together average prices since 2002. The results are below.
Note that I have two estimates in the chart above. One is the highly scientific "eyeball test." The other is a mid-point average between the low and the high. The mid-point method is obviously more objective, but I used the "eyeball test" as a way to check the results. The two estimates are very close with 2009 being the only exception.
Let's use the mid-point average for KCI spot prices and compare with margins at POT and MOS. Below you can see POT's gross margins vs. KCI spot price from 2001 onwards.
While the connection isn't perfect, you can certainly see that margins generally have expanded on a similar scale to the rise in potash prices.
One other issue to explore is the high amount of CapEx at many of these companies. CapEx tends to peak in the mining sector when prices are very high. Oftentimes, miners are rushing to develop new sources of supply, just as demands starts to normalize, creating even more downward pricing pressure. The chart below demonstrates that almost all the major producers have had significant capital expenditures in the past three years.
Even the more conservative POT is spending about 60% - 70% of operating cash flows ["CFOs"] on capex. IPI is spending from 70% - 130% of CFOs over the past three years, and a whopping 282% of income in the last fiscal year! This doesn't automatically suggest that prices will fall, but certainly, if they do fall, it could make things even uglier for certain potash producers.
Falling Prices and Earnings Sensitivity
All this suggests to me that significant margin contraction is very likely in the potash sector. As I showed with copper miners in my 2011 article, a slight bit of margin contraction can result in a huge downward pull on earnings. For that reason, we should try to ascertain how declines in potash prices could impact earnings for Potash of Saskatchewan, Mosaic, and Intrepid Potash.
Obtaining absolute accuracy in this exercise is impossible. In the real world, there are a myriad of complex factors affecting profitability for each individual producer. Companies might suspend operations at higher-cost facilities or reduce production, which can make predicting earnings difficult.
For this exercise, we'll simplify things and assume that all other factors remain constant. Supply stays the same and demand stays the same, but prices fall. While this isn't 100% realistic, it does give us a sense of how vulnerable income and cash flows will be.
First off, let's establish our baseline figures. In the chart below you can see FY 2012 results for POT and IPI and FY 2013 (the most recent past fiscal year) for MOS.
Based on these figures, we'll make a few changes to see how it appears to impact profitability. In the first scenario, we'll assume that prices fall 10%.
You can see that estimated earnings for POT fall from $2.76 per share to $2.10, a 24% decline. MOS suffers an even more dramatical plunge from $4.39 EPS to $2.14 (over 51%!). IPI falls from $1.16 to $0.92, a more modest 21% drop. While MOS is affected more than POT and IPI, you can see that all three companies take a significantly larger hit to earnings than to sales.
For the next scenario, let's take a look at 20% and 25% price declines.
You can quickly see how these price drops obliterate earnings. A 20% decline in sales price results in a 49% drop in earnings for POT, a gargantuan 91% plunge for MOS, and a 59% fall for IPI.
Once again, these are simplified estimates that miss a lot of factors. As you can see that MOS's earnings turn negative in the 25% price decline scenario. At some point, some of these companies would begin to close mines or slow down production, meaning that MOS's estimates are likely overstating the impact. Nevertheless, it's easy to see from this exercise how small price declines can result in much larger drops in earnings for the potash miners.
We'll take a look at two final scenarios. Instead of starting out with revenue price declines, I backed in using gross margins. These two scenarios look at margins that are closer in line with historical ones and assume that the producers adapt a little bit, rather than simply mindlessly selling at lower prices.
You can see that even in these scenarios, there are significant declines in earnings for all three producers, with EPS falling 48% for POT, 67% for MOS, and 83% for IPI in the more conservative scenario. While there are flaws in all of these models, they do consistently show that small declines in prices and margins will likely wreak havoc on earnings.
The 10% - 20% sales price decline scenarios seem fairly realistic given how elevated potash prices are in relation to historical norms. While it's impossible to fully account for all the changes that could occur, we can get a sense of how earnings will be impacted, and take a look at P/E ratios based on this.
There are three charts below. The first shows the current prices, EPS figures, and trailing P/E multiples for POT, MOS, and IPI. The second series look at the revenue decline scenarios and how P/E multiples would shift based on those EPS estimates. The third series look at the custom scenarios and the P/E multiples based on that.
All three look relatively cheap if you simply look at trailing P/E with POT selling at 10.5x, IPI at 10.4x, and MOS at 9.3x. However, this quickly shifts with price declines. A 10% price decline could result in MOS's forward P/E being closer to 19x. A 20% price decline makes all the potash miners look expensive, with POT the cheapest at around 20x and MOS falling near break-even (P/E multiple of 105).
The custom scenario is a little less harsh on MOS, but tougher on IPI. In the Custom #2 model, POT's forward P/E is 20x, MOS is at 28x, and IPI is at 60x.
It's difficult to say what the most realistic scenarios are for each miner without a lot more research, but based on what I've examined, I think an EPS decline to $1.00 - $1.25 is somewhat likely for POT and MOS, while a decline to $0.40 - $0.70 is possible for IPI. That would make POT and MOS potentially begin to look attractive under $20, while IPI might need to fall into the $5 - $8 range.
This analysis seems to suggest that while all three of these companies are highly vulnerable to deteriorating margins in the sector, POT is in significantly better shape than MOS and IPI. That could make POT attractive in a pair trade on the long side, with either MOS or IPI on the short side, if one believes that potash prices will decline in the next few years.
Remember, this is a simplified analysis to show how margin contraction can destroy earnings for these producers. However, all three of the producers highlighted here, as well as the others in the market, have different product mixes that would ultimately affect results in varying ways. For instance, Intrepid Potash is more of a pure-play on potash, while Potash of Saskatchewan has a significant presence in nitrogen and phosphate markets, as well.
One reason I chose to focus on POT, MOS, and IPI was because all three tend to generate a significant amount of sales from potash. I excluded Compass Minerals (CMP), another potash producer, because its financial results are more tied to salt, which tends to have much more stable pricing and returns, but it still generates about 25% of its revenues from the fertilizer segment.
In spite of the massive plunge in the stock prices of potash miners, such as POT, MOS, and IPI, I would keep away from the sector. Earnings are extremely vulnerable to margin contraction and the currently low P/E ratios may be very deceiving if we see potash price return to historic norms.
Out of the bunch, POT seems to be the most insulated from a drop in prices, and hence the least vulnerable. For this reason, a long / short pair trade could make sense here if one expects prices to fall further. In that scenario, going long on POT and short on either MOS or IPI could make a winning combination.
Otherwise, I'd have to see the prices of all these producers fall quite a bit more before I would view them as tempting value investments. I would wait till POT and MOS drop below $20 before I'd consider going long. With IPI, I'd be even more conservative and stay away unless it dropped below $7.