Description

I believe there is an interesting and timely opportunity to generate a significant, market-neutral return by going short Terra Nitrogen (TNH) and long its parent, Terra Industries (TRA). 

Quick Background / Overview

As many of you may know, TRA is a U.S.-based nitrogen fertilizer producer.  Ten years ago, it placed its largest plant (located in Verdigris, Oklahoma) into a publicly-traded limited partnership.  TRA owns 75% of the LP interests in TNH (the other 25% are owned by the public) and 100% of the GP interests.  In addition to its stake in TNH, TRA has 100% ownership of five other smaller North American plants, joint ventures in the U.K. and Trinidad and a large net cash balance.   The relative value between TRA and TNH has completely decoupled in recent months.  In fact, valuing TRA's LP stake in TNH at market value, the enterprise value of the remaining businesses is only $333 million despite generating $594 million of EBITDA of the past year!  

For many years, TNH traded between $5 and $20 per share, until the agriculture boom that started in late 2006.  TNH, with its sizeable dividend yield, was bid up by retail investors to around $150 per share in Spring 2008 as fertilizer prices soared.  Over that same period, TRA rose from $8 to around $50 per share.  Since then, TRA has fallen with the rest of the ag universe back to around $29 per share, while TNH continues to trade at around $145 per share (despite paying out all of its earnings of around $14 per share in dividends over the past 12 months), making it the only commodity stock in the world that I am aware of that is effectively trading at its all-time highs. 

As a result, TNH trades at 12x book value vs. 2.7x for TRA, 5.9x trailing EBITDA vs. 2.6x for TRA and 10x trailing EPS vs. 5x for TRA.  So basically, TNH now trades at over a 100% premium to TRA.  As shown in the chart below, it is clear that TNH is the outlier when compared against TRA and its closest peer CF Industries (CF).  Since the beginning of 2007, TNH's stock (including dividends) is up 4.4x vs. 1.5x for TRA.

While you might be able to argue that a higher EBITDA multiple is justified for TNH because it does not pay corporate income tax, much of this benefit is diminished by the fact that LP unit holders pay taxes on the dividends at their ordinary income tax rates (in fact, TNH should probably trade at a discount to TRA on EV/EBITDA since TNH LP holders actually do not even receive all of its EBITDA due to TRA owning all of the GP interest which is deducted below EBITDA, as discussed later).  

Also, TNH should trade at a lower P/E than TRA since TRA could pay out all of its earning at 15% tax rates to its holders, while TNH's earnings will be taxed at around 35% for its holders (not to mention that TRA should trade at a higher P/E given its large net cash position). 

How did this happen?

I believe this anomaly has largely occurred for the following reasons:

(1) a high trailing (and growing) dividend yield at TNH that has attracted dividend and retail investors

(2) a low float at TNH due to TRA's 75% ownership that can facilitate temporary value distortions

(3) short covering at TNH - short interest has fallen by 2/3rds in the past six months

(4) tight borrow at TNH - this has now changed somewhat as the short interest has declined substantially in the past few months (borrow costs can still run in the high teens, but as I describe below, I believe this trade should work within six months, making the borrow cost small vs. the return potential)

(5) a huge retail and day-trader following in TNH (this is one of Cramer's perennial favorites)

(6) a fundamental lack of understanding of the GP/LP economics by TNH shareholders

Why this should work soon

I think this trade should generate significant returns in the near-term.  Specifically, I think that TNH's forward dividend will be dramatically lower than the trailing dividend that is propping up its stock price.  Given declining fertilizer prices, I estimate TRA's EPS will fall to around $4 this year.  By year end, TRA will have net cash of $10 per share, so the P/E, net of cash, is around 5x EPS.  Hence, the decline in earnings this year at TRA is more than anticipated by the market in my opinion.  

However, the shareholders of TNH are asleep at the switch.  On the same earnings case as used for TRA, I expect TNH's earnings to fall to around $9 per share.  I think this sharp dividend cut will result in a re-rating of TNH, perhaps driving it to a double-digit yield when shareholders realize the lack of safety (and growth) in this dividend.  At a 12.5% dividend yield, this would put TNH at $72 per share.  At an 8x P/E for TRA + $10 of cash, TRA would trade at $42 per share.  So, I suspect there is around 50% downside on the short and around 50% upside on the long.  Of course, fertilizer prices are impossible to predict, but since the earnings of these two companies are inextricably linked, the directional outcome works the same no matter what the commodity price assumptions.

TRA Valuation

While I suspect this trade will work by TNH's share price ultimately collapsing, if it does not, it will highlight how undervalued TRA's stock is in this context.  As shown below, TRA owns a stake in TNH worth $2 billion.  When added to the net cash balance of $525 million at TRA, this amounts to almost the entire market capitalization of TRA despite most of TRA's cash flows coming from outside of the TNH LP interest.  The EV of the non-TNH assets at TRA is $333 million.  Those assets generated EBITDA last year of around $600 million.

GP Interest

In addition to TRA's 75% ownership of the LP interests in TNH, TRA owns 100% of the GP interest.  Prior to the recent agriculture boom this GP interest was effectively worthless because TNH's distributions were below the level at which the GP carry kicked-in.  In fact, years of low distributions at TNH had created a large cumulative deficit that needed to be recouped before any value could accrue to the GP.  The dramatic increase in earnings at TNH that began in 2007 allowed the cumulative deficit to be cleared by the second quarter of 2008 and TRA began receiving GP distributions.  Given the current level of TNH's earnings this GP interest has considerable value to TRA above the value of its LP holdings listed above.  Additionally, this same GP value should be a reduction to what TNH's publicly-traded LP units are worth since they do not lay claim to all of TNH's cash flows. 

TRA's GP carry is calculated off of TNH's quarterly LP distributions based on the following sliding scale:

•  0% up to $0.715 per share (with a minimum distribution of $0.605 for purposes of the cumulative deficit calculation)
•  14% between $0.715 and $0.825
•  24% between $0.825 and $1.045
•  49% above $1.045

M&A Interest in TRA

In January, CF Industries made a hostile stock-for-stock offer for TRA that was rejected by the TRA Board.  CF upped its bid and it was again rejected by the Board.  I think this dynamic is only good for the trade since I think there are three outcomes here, all of which are positive or neutral:  (1) CF raises its offer and TRA accepts, giving TRA shareholders a premium while allowing them to participate in the synergies of the combined entity (that entity would be trading at barely over 2x trailing EBITDA including synergies and an even bigger discount to TNH), (2) CF drops their interest in TRA, which leaves the trade unaffected (I don't believe there is currently any acquisition premium in TRA's stock given it hasn't outperformed its fertilizer peers lately) or (3) a third party such as Yara tries to top CF's bid for TRA (which would of course be good for the TRA long position).

Catalyst

(1) TNH dividend cut

(2) Better understanding of GP economics

(3) Some modicum of rationality in the market