Terra Industries (stub) TRA/TNH S
June 19, 2007 - 10:26am EST by
2007 2008
Price: 6.40 EPS
Shares Out. (in M): 0 P/E
Market Cap (in $M): 2,465 P/FCF
Net Debt (in $M): 0 EBIT 0 0
Borrow Cost: NA

Sign up for free guest access to view investment idea with a 45 days delay.


This write-up is brief and, hopefully, reflects the fleeting nature of the market inefficiency.

Son of Palm?


TRA owns 75% of TNH and consolidates its (TNH’s) results.  TNH is set up as a partnership and consists essentially of one asset, a fertilizer plant in Verdigris, Oklahoma, that boasts a great geographic and cost position.  TNH is unmistakably the crown jewel in TRA’s overall fleet of plants.  Today, there is a significant valuation disparity between TNH and the balance of TRA’s holdings, non-TNH holdings that I’ll refer to as “Core”. 


Holy Fertilizer!


Consider the following:


  • Core generated $55MM and TNH generated $35MM in EBITDA in the most recent quarter.
  • Core has generated more EBITDA than TNH in 32 of the last 35 quarters and has generated almost 3x as much cumulative EBITDA as TNH over the last eight years.
  • Core has 2.5x the capacity and 3x the revenues of TNH.




  • Core (the part with more capacity, greater revenues, more historical cash flow, and more recent cash flow) has an enterprise value of less than half that of TNH.


Note that for the purpose of this analysis I am ignoring 1) the pension obligation of roughly $150MM at the TRA corporate level (even as TNH shoulders some pro rata portion of it, say $25MM worth and hence a net $100MM hit to Core) 2) all contribution from TRA’s Trinidad JV which generated $17MM in net income for TRA in 2006 and reasonably should be valued in excess of $200MM 3) TRA’s interest in the GP of TNH, which in public market valuations is frequently worth in excess of the LP interests (I do not pretend to endorse these valuations, but see e.g. ETE/ETP, XTXI/XTEX, and HPGP/HLND).  Hence, I believe these numbers are conservative and understate the valuation discrepancy.


Historically, TRA and TNH have traded roughly in-line with one another from a business valuation perspective, and the securities exhibited a correlation of over 50% and had nearly identical performance from 1/1/05 through 3/1/07.  For all of 2005 and 2006, Core had an enterprise value of 2x that of TNH.  Since March 1 of this year, TNH is up 160% and Core is down 45%, so Core now trades at 44% of TNH’s enterprise value.  Part of this is due to slightly better EBITDA growth from TNH and the company’s higher exposure to agricultural end markets (markets which have experienced some higher demand for fertilizer through increased corn plantings relative to TRA’s industrial markets), but part is surely due to TNH’s inclusion in momentum screens. 


Historically, EBITDA margins have been similar for TNH and TRA, within 100 bps over the last decade.  In 1Q07, the spread in TNH’s favor spiked to 1200 bps due to short-term agricultural demand and natural gas pricing on the supply side.  One should expect mean-reversion over the intermediate term to a more normalized 500 bps premium (the premium of the last two years) as supply and demand move into balance.


TRA has the right to buy in the 25% of TNH that it does not own based on the average of the past 20 days of trading prices.  This seems unlikely to occur in the short-term, since, in a conversation with the company, I was led to believe that management appreciates the current misvaluation:  “TNH is kind of crazy.”


TRA has $120MM of net debt and a market cap of $2.3 billion.  TRA just refinanced its debt and saved about 500 basis points.  TNH has net cash of $100MM and a market cap of $2.2 billion.  Core is simply TRA less 75% of TNH, so Core has $200MM of net debt and a market cap of $700MM.  To put the Core stub trade on and pick up Core for a little over 4x EBITDA, one would go short 0.13 shares of TNH for each share of TRA long.  As with other stub trades, this setup is pretty compelling.  However, since Core and TNH are highly comparable, I’ve chosen effectively to go long Core and short TNH by buying TRA and shorting TNH in equal dollar amounts.  In so doing, I am long Core at 4x LQA EBITDA and short TNH at 15x LQA EBITDA.


    show   sort by    
      Back to top