Investment Thesis

Basic economic theory contends it is impossible to sustain excess profits in a perfectly competitive industry, as market entry will eventually lead to increased supply and falling returns on capital.  To an outside observer, it might appear as if this theory does not apply to Terra Nitrogen (TNH) given the company’s current profitability.  In reality however, Terra’s current earnings are much greater than justified in any sustainable market equilibrium and thus are only temporary. Supply in Terra’s market will increase substantially in the next several years and thus erode the company’s profits.  Adding only currently funded and permitted projects to the industry supply curve suggests normalized earnings are 60% below their 2012 level. Based on justified/historical multiples, this implies TNH shares are worth $75-$110 versus $204 today. Shares are available to borrow in size at almost no cost.

TNH trades at about 11.5x trailing earnings, so if you’re going to buy this you need to believe that current record earnings are (at least mostly) sustainable. But they’re not, and it not even close. To get a sense of just how distorted this market is, in 2012 US natural gas prices were approximately $3.30/mmBTU implying ammonia production costs of about $150/ton. However, the US imported ammonia from suppliers in Europe who purchased oil-index gas for about $14/mmBTU and had total costs (including shipping) of nearly $600/ton. The only reason this extremely inefficient situation exists is because it takes a few years for new capacity in the US and other low cost countries to come online.  This capacity is in the process of being built, and once it is operational both nitrogen prices and TNH’s earnings will fall substantially. 

Business Overview

Terra Nitrogen is a master limited partnership that owns a single nitrogen fertilizer-processing complex in Verdigris, OK. The Verdigris plant has annual production capacity of 1.1MM tons of ammonia and 2.0MM tons of UAN. CF Industries (NYSE: CF) is the general partner and owns 75% of the common units. TNH distributes substantially all cash flow as quarterly distributions. The common unit share of distributions decreases as the level of distributions increases. In each of the last two years, common units received 56% of total distributions.

Besides being expensive and a taking a few years to build, there is nothing really special about a nitrogen fertilizer plant. Anyone with a couple billion dollars, a supply of natural gas and a few hundred acres can build one. Some plants are slightly more efficient (use less gas), in better locations (closer to farms), or produce different types of fertilizer, but at least for domestic plants these differences have a rather minor impact on overall profitability. 

3 Things we Know for Sure

What we know #1: Prices are Determined by the High Cost Producer:

Natural gas is the primary input in nitrogen production, accounting for over 70% of TNH’s cash production cost. Not surprisingly then, variations in nitrogen supply costs are almost entirely determined by differences in natural gas costs.  As such, according to a recent Potash Corp presentation “The floor for the nitrogen market is established by high-cost producers that are primarily located in Ukraine and Western Europe”  because in these countries “Gas prices on a spot basis and those linked to oil prices are well above most other nitrogen producing regions of the world” (emphasis added). This dynamic has historically lead to a strikingly strong relationship between the cost of producers using oil-linked gas and world nitrogen prices. As such, its not surprising that a simple liner regression of weekly oil prices to gulf ammonia prices from 2006 to today shows a positive relationship with a 99% confidence interval and an r-squared of .55. You can also look at POT's annually published estimates of Ukrainian midwest delivered ammonia cost and it will always be nearly identical to TNH's realized ammonia price.

What we know #2: Economics of New Supply Are Compelling:

Terra operates in a classic “perfectly competitive” industry with a commodity product and no barriers to entry. Even so, there has not been a new nitrogen facility built in the United States in over ten years, as previously high natural gas prices and lower fertilizer prices made such projects uneconomical. This is no longer the case however; now, the economics of new nitrogen capacity are extremely compelling.

To illustrate just how compelling these economics are, let’s look at Terra. To replicate Terra’s assets would cost about $2.2 billion and take three years. Last year, Terra had $561MM in net income. Even if you assume $4 gas (versus 2012 realized cost of $3.31), earnings would be about $530MM, implying an unleveraged pretax IRR over 20% for a new facility. This is a compelling return, even before considering the advantageous funding options available to these projects, such as the tax-free bonds being issued by OCI to fund a new fertilizer plant in Iowa. If we assume a 5% interest rate similar to what OCI is paying on its Iowa bonds, 70% debt financing and a 35% tax rate, the after-tax return on equity grows to over 30%.

What we know #3: Significant Amounts of New Supply are Anticipated:

Given the attractive economics, it should not be surprising that, in the next three to four years both governmental organizations such as the UN FAO and independent consultants expect that about 5–6MM metric tons of new low-cost capacity (on an elemental N equivalent basis) will come online in the United States alone, with an additional 5–6MM globally (excluding China). Much of this will come from expansions or restarts of current facilities. This estimate includes only projects that are currently funded and permitted. The lengthy planning and construction process for these assets means that supply increases can be estimated with reasonable conviction During this same period, demand should grow by about 5–6MM tons.

3 Things we Don’t Know (but don’t really matter)

What we don’t know #1: Natural Gas Prices:

A $1/mmBTU change in natural gas prices is equal to about a $1.15 change in EPS TNH. While this is not immaterial, it is small relative to the overall variability of TNH’s profitability. So even if gas prices were to fall well below $3, the thesis here does not change.

What we don’t know #2: Demand:

In order to be effective, nitrogen must be applied every season. Because of this, nitrogen demand is largely a function of agricultural acreage and is thus fairly stable, generally increasing 1%–2% per year. Given the recent record acreage levels, it seems reasonable to believe demand will not exceed this growth rate over the next few years.

What we don’t know #3: If Projects are Actually Completed:

Of course, there can be no guarantee that the planned supply increases will ever be completed, as even in-construction developments can always be stopped. Indeed as CF emphasized in their recent analyst day, actual capacity increases are often significantly less than “announced” projects would indicate. However, never before have nitrogen companies sustained the record earns we’ve seen in the past couple years. As such, the only plausible reason for this expanded capacity not to materialize would be a change in the underlying economics due to either decreased nitrogen prices or higher natural gas prices. In the event that nitrogen or natural gas prices change enough to kill planned expansion projects, TNH’s earnings will be similarly impacted, and the short position will profit even faster.

What New Supply Means For Prices

Given the forthcoming supply increases, it’s reasonable to believe that prices of nitrogen fertilizer will be much lower in a couple years than the levels realized today. As noted earlier, the marginal producers in this market are European firms who use oil-indexed gas as their feedstock.  However, oil-indexed gas accounts for only approximately 3.5 million tons annually or 3% to total supply. By any reasonable estimates, new low-cost supply growth will outpace demand growth by much more than this amount of oil-indexed production, thus rendering these factories obsolete.  Once oil-indexed producers are eliminated, the marginal producer becomes those who buy gas on the European spot-market. In 2012 these producers had total costs of about $350/ton.

Last year, net US nitrogen imports ex Canada and Trinidad were approximately 2.4MM metric tons. Near-term domestic supply increases will make these imports unnecessary and actually place the North America + Trinidad into an export situation. As such, the domestic price will equal the export (global) price less applicable transportation. Assuming $35/ ton shipping costs to South America, domestic ammonia prices would be about $320/ton on current European gas costs. Even if European gas prices go to $12 (an annual level never seen before), domestic ammonia prices would be $400.

The idea that nitrogen producers are currently over-earning is not exactly controversial. Not only would you be hard-pressed to find an informed party that disagrees with this thesis, but nitrogen producing c-corps (non-MLPs) such as CF and Yara trade at about 7.5x earnings implying that the market thinks future earnings will be lower than trailing numbers. 

So What’s it Worth?

Using these normalized price levels of $320 - $400/ton and a US natural gas price of $4.00 - $4.50, TNH would earn approximately $5 - $9 per common unit. As an MLP, TNH pays out nearly all income as distributions. As such, the stock’s dividend yield is substantially the same as the earnings yield. Over the last 7 years TNH averages about an 8.5-9.5% yield which seems justified by a high-single digit cost of capital and minimal long-term growth expectations.  This equates to a normalize earnings power value of about $56 - $105 per share. Adding the value of excess earnings over the next three years, yields an intrinsic value of $75 - $110 per share.  

Replacement Cost Provides “Sanity Check”

Assuming that common units are entitled to 60% of total economics (versus 56% in the last two years), then the market currently values Terra at 3x replacement cost (market value of common units of $3.8B divided by 60% less net cash / $2.1B replacement cost). As discussed earlier, if 2012 did indeed represent “normalized earnings,” then the value of TNH’s earnings would be much greater than the replacement cost.  Since we know that in the absence of barriers to entry the value of a business’s normalized earnings should equal its replacement cost, this provides further evidence that Terra is currently over-earning.

Other Issues

Those who know this market (and are still reading) will probably realize that I’ve ignored two issues that some believe are important; China and the different forms of nitrogen fertilizer. So far, I’ve discussed ‘nitrogen’ as a homogenous product, but in reality prices of the primary types of nitrogen fertilizer (UAN, urea and ammonia) can diverge. However, the ability of both suppliers and suppliers to change products makes a significant long-term price divergence unlikely. Furthermore, if anything I view TNH’s primary product (UAN) and even more vulnerable than other nitrogen products given the expected increases in UAN supply and the greater expense of shipping UAN internationally.

As for China, China is the world’s largest producer of nitrogen as well as a significant exporter. Currently, Chinese nitrogen exports are subject to fairly hefty export taxes during certain times of the year. This policy is intended to lower fertilizer costs for domestic Chinese farmers.  As such, it seems plausible that further export restrictions could decrease supply in this market.

While the actions of the Chinese government are unknowable, it seems unlikely that the supply of Chinese exports will decrease materially. For starters, during the high-tariff season (November – July) taxes are already rather high at over 75%. Furthermore, China has significant excess nitrogen capacity with most facilities having a fairly high production cost. As such, higher taxes would not be beneficial to the Chinese economy as they can never lower prices beyond this relatively high cost of production of Chinese firms, but such taxes would still hurt the fertilizer industry by not allowing exports of the excess capacity.  Thus, I don’t see Chinese taxes as material risk to this short.

What’s the risk?

The main risk is that feedstock costs for future marginal suppliers (European spot gas and Chinese coal) rise enough to keep global prices elevated. This risk is acceptable, as discussed earlier, even if costs rise above all-time highs, TNH’s intrinsic value is still less than half the current price.