Here is the fifth and final installment of our special report comparing two very popular fertilizer yield investments: Terra Nitrogen (TNH), formerly a subsidiary of Terra Industries but now controlled by CF Industries (CF), and the recent upstart, CVR Partners (UAN), which is controlled by CVR Energy (CVI).
We have described the history of both Terra Nitrogen and CVR Partners. We compared the anhydrous ammonia production track record and urea ammonium nitrate (UAN) solution upgrading capability of both of these nitrogen product manufacturers.
Now we wish to decide which is the better investment for yield-oriented investors interested in U.S. domestic fertilizer-leveraged capital appreciation. In addition, we want to know if we should be in this sector at all. Let's look at product revenues, cost structures and basic profitability. Then we'll get to our price outlook for these MLP units.
Nitrogen Product Distribution and Pricing
On the revenue side, you might expect that CVR and TNH receive approximately the same prices for their nitrogen products given they both sell ammonia and urea ammonium nitrate solution to mainly agricultural markets, and both plants are situated within 60 miles of each other in the Southern Plains area, CVR Coffeyville in SE Kansas, and TNH Verdegris across the border in the Tulsa area of NE Oklahoma, respectively.
Both companies sell locally via truck to the crop growing areas of Kansas and Oklahoma, but mainly via rail or pipeline to destinations farther afield. Rail freight costs to get the product to destination should be similar, as both have access to Class I rail systems.
TNH distributes primarily through the Burlington Northern Santa Fe system but also via barge at the Port of Catoosa where it leases a shipping dock and thus has access to the Mississippi River. TNH is also on the 1,100 mile Magellan ammonia pipeline which feeds from Agrium's Borger,Texas, TNH's Verdegris, OK, and Koch Industries' Enid, OK, ammonia plants, and ends at Mankato, Nebraska, serving 13 terminals along the way.
TNH benefits from CF Industries distributing its products, with its 40 ammonia terminal network and greater logistical strength. TNH owns two ammonia terminals itself at Blair, Nebraska, and Pekin, Illinois, and leases about 3,000 rail cars but has subleased these assets to CF Industries for fixed payments.
CVR does not have ammonia pipeline access but is on the Union Pacific main rail line connecting with the Kansas City Southern and other rail systems. CVR also sells directly to customers via truck. CVR serves customers in states from as far east as Ohio across to Colorado and as far north as Minnesota down to Texas.
Both TNH and CVR should benefit from higher prices than realizable at the U.S. Gulf Coast, because importers would have to pay greater shipping charges to get product to the interior. Freight is a significant expense and both companies pass through freight charges, TNH to CF Industries and CVR to customers.
Both TNH and CVR show "plant-gate" prices realized in their financial disclosures so the prices are comparable. On the other hand, CF Industries, the 75% owner and general partner of TNH, shows prices including freight, so they are higher.
|$ Short Ton||FY 2005||FY 2006||FY 2007||FY 2008||FY 2009||Q1-10||Q2-10||Q3-10||Q4-10||FY 2010||Q1-11||Q2-11||Q3-11|
|TNH Customer Advances ($million)||$31.7||$35.3||$154.6||$45.1||$16.4||$48.6||$0.1||$33.6||$61.2||$61.2||$33.1||$3.6||$0.1|
|CF Ind. Ammonia||$316||$361||$388||$560||$514||$321||$380||$394||$452||$402||$494||$596||$552|
|CF Ind. UAN||$162||$158||$215||$321||$232||$205||$220||$188||$206||$205||$277||$323||$319|
TNH ammonia price lower than CVR's
In Q3 2011, TNH received an average $455 per short ton (st) for its ammonia versus $568/st for CVR.
Although merchant ammonia (that which is not upgraded to UAN but sold on the market) is less important to TNH than UAN sales, that ammonia price was 19.9% less than CVR's, so it behooves us to analyze the nature of the disparity.
In 2010, CVR's Q3 ammonia price was only $317/st but it jumped to $491/st in Q4 2010 and then again to $564/st in Q1 2011.
TNH's Q3 2010 ammonia price was $366/st, $49 more than CVR, but by Q1 2011 was $150 less and has remained below CVR Q2 ($75) and Q3 ($113).
|$ Short Ton||FY 2005||FY 2006||FY 2007||FY 2008||FY 2009||Q1-10||Q2-10||Q3-10||Q4-10||FY 2010||Q1-11||Q2-11||Q3-11|
|TNH Customer Advances ($mill)||$31.7||$35.3||$154.6||$45.1||$16.4||$48.6||$0.1||$33.6||$61.2||$61.2||$33.1||$3.6||$0.1|
There are at least three possible reasons for this trend. One would be differences in customer base. Another could be TNH sells more lower margin industrial ammonia. Or third, TNH went into the second half of 2010 and 2011 with a larger forward book of ammonia. Most likely, it's a combination of all three factors.
TNH Verdegris sells to national farm retailers such as Agrium's Crop Production Services which was their biggest nitrogen customer in 2009 (9% of sales) and 2010 (12% of sales). CPS would no doubt get a discount price due to volume rebates, as would other big dealers.
But Terra Industries, TNH's former owner, was known for selling to many smaller customers as well, whereas CF, the new owner, would sell to mainly its big coop prior owners like CHS and Growmark. TNH's top five nitrogen customers made up only 26% of sales in 2010 while in 2008 no customer accounted for more than 7% of sales. TNH doesn't break out ammonia versus UAN customer sales.
CVR has a more concentrated customer base, with the top five ammonia customers making up about 44% of sales, as do the top five UAN customers.
Is it possible the co-ops aren't as aggressive as Agrium in negotiating discounts from a nitrogen producer? Hard to believe but possible.
What about industrial ammonia? We know that industrial ammonia contracts often tied the price to natural gas cost, making the margins lower than agricultural ammonia during periods of low natural gas prices.
TNH sells ammonia to industry, for use, according to the 10K, in "resins, metal alloying, pulp and paper, refrigeration, detergents and pharmaceuticals," but TNH does not disclose the proportion of industrial compared to agricultural ammonia.
CVR disclosed in its April 2011 IPO prospectus that on a three year average basis, it sold approximately 83% of its ammonia to agricultural customers, and 13% to industrial customers.
If you made the case that TNH sells say half of its ammonia to industry rather than ag, you would see them realizing lower prices on half their sales over the past couple of years as agricultural ammonia pricing diverged from natural gas pricing.
The last possibility is that perhaps TNH favored forward selling of ammonia more than CVR since CF Industries took over Terra in April 2010.
CF has always favored selling nitrogen forward through their FPP (Forward Pricing Program), and considers it a strategic advantage to be able to let customers lock in nitrogen prices far in advance. Ammonia has been tight and the application window is narrow, so customers probably want to line up supply in advance.
To track forward sales (and hence forward pricing) at TNH, we tracked customer advances (deposits) on the balance sheet. which are lodged when forward sales are made. The company would try to limit forward sales in a bullish environment and add to them if it thought prices were going down. Advances are also seasonal, picking up in the fall for spring, or in the early spring for fall.
You can see from the table that TNH had fairly large customer deposits of $48.6 million at the end of Q1 2010 which would have been under Terra's ownership. Tampa quoted ammonia prices began 2010 at about $400/st but steadily declined in price and bottomed in the summer at around $325/st before starting a more or less constant rise in the second half of 2010 and through 2011 to $705 this month.
TNH worked the customer advances off by Q2 (by shipping product and realizing revenues) but built the advances up even higher under CF ownership, to $61.2 million by the end of 2010.
CVR had a deferred revenue balance, representing customer advances, of only $3.0 million at the end of Q1 2011, the first quarter we have disclosure on since their IPO. Therefore, CVR was open to spot pricing during the period when nitrogen fertilizer prices were rising dramatically.
In Q2 CVR deferred revenues rose to $26.7 million and then declined to $20.6 million in Q3. This indicates CVR ramped up their foward selling of both ammonia and UAN during the early summer, and could partly explain why Q3 average prices realized were sequentially down $6 in Q3 2011 over Q2 2011.
So forward sales of ammonia in 2010 were probably the main reason why TNH's pricing in the first half of 2011 has been far below CVR's, as sales prices booked in 2010 have been lagging the unexpectedly tight 2011 domestic fertilizer market.
The good news is TNH had worked the customer advances down to $0.1 million by the end of Q3 2011 and hopefully is forward selling ammonia at higher prices now.
Where does this leave us? If industrial ammonia is the reason for TNH's ammonia price deficit, then it could mean even lower prices on these sales going forward as natural gas prices continue their decline. But if forward selling was the main reason, we would expect TNH pricing in Q4 to be better than YTD and hopefully closer to CVR's going forward. That would be a boost for TNH's results in Q4.
UAN Prices Comparable
In Q3, TNH received $295/st for UAN, versus $294/st for CVR, almost the same, as one would expect, because UAN solution is primarily an agricultural chemical.
Volumes for the two MLPs are different as previously discussed. CVR produces about 36% of the ammonia TNH does and upgrades about 4% less of it into UAN (67% versus 71%).
Another way to look at it is TNH produces about 2.7 times more ammonia than CVR, 3 times more UAN but has only 2.45 times the merchant ammonia left to sell afterwards.
However, CVR plans to expand its UAN upgrading capacity to more than 100% of its historical ammonia production by first quarter 2013, aiming to buy the additional hydrogen required to produce it from its parent CVR Energy.
TNH is penalized by its industrial ammonia contracts during a buoyant agricultural fertilizer market as we have recently experienced, but the industrial revenues tend to be more stable than the seasonal agricultural markets.
TNH's new owner (CF Industries) is trying to reduce the cost plus industrial ammonia contracts in favor of the more lucrative agricultural applications and pricing.
It is possible CF will announce additional ammonia upgrading capability to UAN at Verdegris, thereby nullifying the current ammonia price advantage CVR curently has, and the added UAN margin it plans to obtain.
In fact, CF Industries inherited an expansion of UAN production at the sister Oklahoma plant at Woodward when they bought Terra Industries, and has some recent experience in this area. The 525,000 ton expansion was budgeted for $180 million and was completed last year.
At this point, you may be wondering whether CF has a conflict of interest in selling TNH's products, given they own other nitrogen plants. All the plants have been selling all they can make, but when times turn bad, how would CF assign sales and prices to TNH? What stops CF from selling preferentially from their other nitrogen plants? How do they allocate natural gas hedges between their seven domestic plant locations?
On September 28, 2010 CF and TNH's independent audit committee signed an Amendment to the General and Administrative Agreement Services and Product Offtake Agreement which governed their relationship and was last amended between Terra and TNH on October 23, 2007.
In this agreement, CF agreed to take 100% of TNH's production (albeit while still controlling TNH Verdegris' production rate) and pay TNH market prices for its products. The market price is basically a monthly average of whatever CF sells the Verdegris product at.
The purchase price for each type of product (the "purchase price") shall be determined on a monthly basis and shall be equal to the volume weighted average net invoice price (less any shipping, railcar, barge or other transportation costs, and after any discounts, rebates and other incentives) for Third Party Sales of such type of Product during such month. A "Third Party Sale" shall be any sale of product by CF Industries to a customer (other than CF Industries or an affiliate of CF Industries) directly from Verdigris and not from another warehouse, storage or terminal facility.
TNH Cost Structure
Let's drill down on TNH's cost of product sold and particularly the cost of natural gas in making ammonia, and since each ton of UAN requires 0.41 tons of ammonia, that product as well.
TNH Verdegris consumed approximately 40 trillion BTU of natural gas in 2010.
That probably combines both process gas (for ammonia) and non-process gas (steam production, boilers, heating etc) uses.
A large, efficient ammonia plant utilizing steam methane reforming as does TNH Verdegris uses about 33 million BTU (approximately 33,000 mcf) of natural gas per short ton of ammonia produced. A widely used rule of thumb is another $35/st for other fixed and variable costs.
At TNH's approximate $4.50 Q3 average natural gas cost, ammonia production would cost $148.50 plus $35 = $183.50/st.
The UAN process uses about 0.40 tons of ammonia, or $73.40 plus a $25 rule of thumb charge for other fixed and variable costs, for a total of $98.40/st.
At a 29% ammonia, 71% UAN split, the average cost of product for both ammonia and UAN would be $123.08/st.
The actual blended cost of product sold quoted by TNH for the quarter was $113/st.
This may be because either Verdegris is more efficient than average, or more likely, there were few or no turnaround expenses capitalized during the quarter.
Note the agricultural nitrogen prooduct sold in Q3 was based on prices booked several months before and therefore lagged spot prices due to this year's constantly rising prices for nitrogen fertilizers.
At $455 ammonia and based on our cost, ammonia gross margin was $271.50 on 97,000 tons and generated $26.3 million in Q3.
At wholesale agricultural ammonia spot prices of $550-600 for spring delivery, the margin TNH would be capturing at the Q3 gas price would be near $400/st.
NYMEX December natural gas futures dropping dramatically in November, to $3.403 per mmBTU today. Given Verdegris enjoys a discount to NYMEX pricing, at these prices TNH would make at least another $1.10/mmBTU or $36.30 per ton in gross margin.
At $295/st UAN and based on our cost, UAN gross margin was $196.60/st on 525,000 st or $103.2 million in Q3.
Our combined estimated TNH Q3 gross margin of $129.5 million compares with TNH's Q3 reported gross margin of $132.9 million.
Since you can get almost 2.5 tons of UAN from one ton of ammonia, you can see that UAN upgrading is attractive, because a ton of ammonia converted to UAN would have yielded TNH $491.50 rather than $271.50 (if priced at the average Q3 which includes industrial) or even $400 if priced solely agricultural.
Either way, ammonia upgrading to UAN would have provided incremental margin, assuming a persistent and growing market for the latter.
The key takeaways from this analysis are that TNH's cost of product sold has a high component of variable cost and that variable cost is mainly natural gas. We estimate $180.1 million or 65.2% of 2011 cost of product sold of $274.1 million was natural gas.
The remaining $94.7 million of cost of product sold would include both other variable production costs such as chemicals, but mainly fixed plant costs including labor, utilities, depreciation and amortization, property taxes, insurance and environmental compliance.
In addition, TNH agreed with CF to pay the latter its out of pocket expenses to run the MLP, as well as a $3.5 million quarterly SG&A fee indexed to inflation for all the activities required to administer the MLP (TNH has no employees). These costs and fees totaled $14.5 million and $10.7 million for the first nine months of 2011.
The fact that TNH depends on natural gas prices as a large component of its variable cost for nitrogen fertilizer production has been a blessing over the past few years. The question is, where are natural gas prices going forward?
U.S. Natural Gas Price Outlook
The U.S. produced 20.58 trillion cubic feet of dry natural gas in 2009 (EIA Natural Gas Annual, 2009). Electric power was the number one use (6.9 tcf), followed by industrial uses (6.2 tcf), followed by residential (4.8 tcf) and commercial (3.1 tcf).
A key factor going forward for natural gas prices will be the interplay between increased production of shale natural gas and a recovering U.S. industrial and electric power sector consuming it. Then there is the medium-term possibility of even compressing it and exporting it to off-shore markets as Liquified Natural Gas (LNG).
Here is a chart showing some of the prices for natural gas over the past decade. Current December futures prices are the lowest in 10 years.
It's no secret the discovery and application of hydraulic fracturing and horizontal drilling techniques to constrained shale gas deposits in the U.S. Southeast and Northeast has dramatically increased the supply of natural gas over the past few years, driven by shale gas production in plays such as the Barnett Shale in Texas.
The EIA's Annual Energy Outlook released in April 2011 calls for only slightly rising natural gas prices over the next five years or more. Consistent with this forecast, NYMEX natural gas prices are slightly rising on the forward curve, with December 2011 at $3.40, December 12 at $4.19 and December 2013 at $4.68.
It should be noted that these are average price forecasts. Cold winters and hurricanes can cause short-term natural price spikes, but they are usually short-lived.
It's hard to make a case that natural gas prices can be sustained much lower from the $3.40 price reached on December futures yesterday. Domestic natural gas storage capacity continues to be increased, allowing some gas to be set aside rather than dumped on the market, which is often the reason for really depressed pricing. Here is a chart of U.S. natural gas storage capacity growth.
The rush to invest in U.S. shale gas has been formidable, with Exxon Mobil's (XOM) acquisition of XTO Energy last year and BHP Billiton's (BHP) purchase of Petrohawk this year. These purchases must be predicated on rising natural gas demand and prices.
It is my guess shale gas and Gulf of Mexico operators will shut in gas wells (effectively adding to storage in ground) and stop drilling for more natural gas, if natural gas prices fall from current 10-year low prices, as netbacks turn negative.
Another formidable factor limiting further decline in natural gas prices is the "oil to natural gas conversion" savings for those who can switch from liquid to gasoline.
A barrel of oil has about 6 million BTU of heat value. The equivalent heat value of natural gas is currently only $3.40*6=$20.40, compared to a current $101.93!
Assuming a positive crack spread for diesel and gasoline, the savings achieved by switching to natural gas are even more attractive.
Canada is the leading exporter of natural gas to the U.S. at 2.6 tcf in 2009, but this is in decline. I would expect Canadian U.S. natural gas exports to continue to decline if NG prices dropped further, as a strong Canadian dollar and rising pipeline tolls make exports uneconomic.
Finally, in the medium-term, it is possible the U.S. will turn LNG import terminals into exporters of compressed natural gas, which could give a boost to demand.
What this means for TNH is that although the cost of its natural gas will be even lower than Q3 over the next couple of quarters, in the medium term, natural gas prices will begin rising again, in spite of additional shale gas supply.
TNH's gross margins should be in the 60%-70% range in the current (Q4) and the first quarter of next year as ammonia margins will depend on the industrial component but agricultural ammonia and UAN realizations should be excellent.
CVR Partners Cost Structure
The CVR Partners cost structure is quite different from that of Terra Nitrogen.
Whereas variable costs are a large part of TNH's operation, particularly natural gas, CVR has an expensive gasification unit to operate, requiring high fixed costs such as electricity, plant labor and outside contractors and materials to ensure high on-stream operating rates.
If the gasification unit is not working (one is operating at all times, and another is held in reserve) then CVR cannot make ammonia or UAN due to the lack of enough hydrogen available from the CVR Energy refinery to compensate.
In addition, CVR needs pure oxygen to partially oxidize the slurry of pet coke, water and additives from an on-site Linde air separation plant, under a multi-year contract. If the air separation plant is down, as happened last year, the gasifier cannot operate.
CVR estimates it currently costs about $105-110 million per year to operate their nitrogen fertilizer plant with about $85 million (or 77-81%) coming from mainly fixed Direct Operating Expenses and the rest from variable petroleum coke and hydrogen costs.
If the plant is operating with a high on-stream factor in all areas (air separation, gasification, ammonia and UAN, the plant has the ability to be highly profitable at sufficient end product prices. But if any part of the plant breaks down, the high fixed costs mean losses, as evidenced by a $9.6 million operating loss in Q4 2010. In comparison, if the TNH plant had to be partially or fully curtailed, the natural gas costs would cease, so at least about 65% of COGS would be saved.
For its variable cost of product, CVR utilizes petroleum coke, a form of carbon left over during the oil refining process, and a historically lower cost feedstock than natural gas to produce hydrogen. In addition, CVR obtains about 70% of its coke from its parent oil refinery at a preferred price (limit of $40/st).
According to CVR, a ton of ammonia requires 1.18 tons of petroleum coke in feedstock. Given a recent market price of $52.50/st for USGC 4-5% sulphur pet coke, that equates to only $62 feedstock cost per ton of ammonia.
Using the formula for ammonia from natural gas, at $3.40 NYMEX, the natural gas cost of an efficient steam methane reforming plant would be about $112.20.
In its April 2010 IPO prospectus, CVR states that at $4.00/mmBTU natural gas, its pet coke gasification process produces ammonia at a comparable cost ($194 versus $193/ton) to steam methane reformering plants such as TNH Verdegris.
But at higher natural gas prices, CVR states it has an advantage which rises to $115/st for ammonia and at $7.50 gas.
CVR also states that its UAN production cost would be lower than for a natural gas ammonia plant, with an $11 advantage ($87 versus $98) at $4.00 gas and due to their lower conversion cost of $12/ton during 2010. CVR says that $10.82 per ton of ammonia costs would not be transferable to UAN costs, another advantage.
We have to wonder how that works and also how CVR calculated the cost of UAN conversion given the explosion of the UAN reactor vessel on September 30 caused the on-stream factor for the UAN plant to average only 80.8% in 2010.
CVR also intends to buy hydrogen from the refinery to ramp up UAN production once the expansion is completed, estimated for the first quarter of 2013. The cost of hydrogen will be another variable cost.
Let's look at the record for CVR's COPS (Cost of Product Sold) and DOE (Direct Operating Expenses) as well as petroleum coke usage, prices and cost.
|CVR Partners||FY 2005||FY 2006||FY 2007||FY 2008||FY 2009||FY 2010||Q1-11||Q2-11||Q3-11||Q4-11e||FY 2011e||Forecast*|
|Coke (000 tons)||Not Avail.||439.0||449.8||451.9||483.5||436.3||124.1||135.8||131.2||130.0||521.1|
|Coke Cost per Ton||Not Avail.||$19.0||$30||$31||$27||$17||$15||$30||$43||$45||$33|
|Total Coke Cost||Not Avail.||$8.3||$13.5||$14.1||$13.1||$7.4||$1.9||$4.1||$5.6||$5.9||$17.5|
* CVR Partners April 2011 IPO Forecast for twelve months ending March 31,2012
The record seems to indicate that generally CVR costs are running in line with a forecast included on page 59 of the April 2011 IPO prospectus. This is the budget underlying the MLPs promise to pay $1.92 in distributions for that period.
However, you can see that cost of product sold, which is a combination of petroleum coke costs and (mainly railway) freight, is going to escalate to $48.3 million for the TTM Q1 2012. CVR obtains about 70% of its pet coke from its parent's refinery with a price tied to UAN, but is capped at a $40/ton cost.
However, the 30% third party coke which CVR buys on the open market has been trading at higher prices. Given the Wood River Illinois, refinery owned by ConocoPhilips/Cenovus is going to start producing 1.5-2.0 million more tons, we expect mid-continent pet coke prices to be favorable for CVR going forward.
Railway freight costs continue to be buoyant, and in particular, transportation of hazardous goods such as ammonia is becoming increasingly expensive. Unlike TNH, CVR does not have barge or pipeline alternatives to rail freight. Although freight is passed along to customers, it impacts the general value proposition.
CVR's direct operating expenses have been relatively stable, but contain some items which could be problems going forward. Electricity is the biggest expense item at $25 million. Property taxes have been running over $10 million per year but are being contested in court and there is a potential windfall of back taxes in the amount of more than $30 million if CVR prevails. Another big item in DOE (albeit non-cash) is share-based compensation, which is paid to CVR Energy executives that double as CVR Partners' senior management.
TNH Vs. CVR Margin Comparison
In a recent article we compared CVR and TNH Gross Margins (See article on Rentech Nitrogen Fertilizers). Let's look at that comparison again.
|Margin Comparison||FY 2008||FY 2009||FY 2010||9Mths-11|
For the purpose of gross margin, I have added both CVR's direct operating expense and cost of goods sold together, because the former is so significant to their cost structure compared to the latter.
It seems apparent that for 2010 and 2011, TNH has been outperforming CVR on a margin basis, mainly due to the decline in natural gas prices, and in spite of TNH realizing a lesser price on ammonia sales as discussed above.
There are a few differences in company reporting on D&A and plant turnaround expense. TNH includes depreciation and amortization in their cost of product sold, but CVR excludes this. However, CVR expenses plant turnaround expense while TNH capitalizes it and amortizes it over five years.
The TNH Verdegris plant has been depreciated over the years to the point where the billion plus dollar plant complex has an accounting cost base of $79.4 million and D&A is running at only $20 million per year, mainly due to the amortized plant expense.
Another key point is apparent: TNH is expected to have other than natural gas costs of about $94.6 million to run their plant this year albeit with no turnaround scheduled, whereas CVR will requires about $86.4 million in similar expense.
Although CVR pays a little over half the cost for feedstock to make ammonia that TNH does, the cost of running the CVR plant is almost the same as that of TNH, which has produced an average of 2.7 times more ammonia and 3.0 times more UAN over the past six years.
Another interesting point: CVR's DOE for the 9 months ending September 30 was $65.4 million up $4.7 million from $60.7 million in the prior period. It would have been $7.2 million higher without a $2.5 million insurance recovery. So CVR's direct operating expenses were up 11.9% year over year and may be inflating.
TNH To Spend More On 2012 Maintenance CAPEX Than CVR
Let's look briefly at expected maintenance and turnaround Capital Expenditures for the two plants in 2012, because both have to deduct this from available cash before distributing it.
In its Q3 report TNH reported they would spend $50 to $80 million in CAPEX during 2012. This would include $35-45 million for the biennial plant turnaround, but also environmental expense related to compliance with a recent consent decree with the EPA to reduce NOx emissions.
Since this CAPEX would come out of available cash and be equally shared with the general partner, the hit to TNH distributions will be $25-40 million or $1.35-$2.16 in less distributions for 2012.
In the case of CVR Partners, the CVR Energy 2010 10K estimated nitrogen fertilizer operations CAPEX is as follows: Environmental and Safety ($3.8 million), Sustaining/Maintenance ($4.3 million) and Turnaround ($2.8 million). Subsequently, forecasted Turnaround has been increased to $3-5 million. Therefore total CVR 2012 CAPEX is estimated at $11.1-$13.1 million.
As previously indicated, CVR already includes CAPEX in its DOE forecast. Therefore, we do not have to discount available cash flow for this expense.
However, we know that CVR has outstanding potential liabilities related to the 2007 oil spill which the EPA was pursuing in court. Therefore, we cannot exclude the possibility that at some point during 2012, there may be an extra expense for E&S that has not been forecast.
CVR expects to spend $68 million in expansion CAPEX on the UAN upgrading project during 2012, with implementation during the late 2012 plant turnaround.
Integrating and commissioning a new manufacturing line is always fraught with hiccups and potential delays, so investors should be especially cautious at that time because CVR may experience more than expected plant downtime. As we indicated above, any downtime on the plant results in operating losses due to high fixed costs.
Imported Ammonia From Trinidad, UAN From CIS, Key To Nitrogen Supply
Let's briefly outline where we stand on ammonia production, consumption and imports in the U.S. In my view, what is going on with problems in Trinidad ammonia production is key to where U.S. nitrogen prices are going to go in 2012.
The U.S. Geological Survey lists 2010 U.S. production of ammonia at 10.08 million metric tons, with apparent consumption of 16.78 million metric tons and imports of 6.73 million tons. As you can guess, much of the ammonia produced or imported is upgraded into fertilizer products such as urea and UAN solutions, or industrial products such as ammonium nitrate for explosives and chemicals such as resins and plastics.
The biggest importer to the U.S. for at least the past 10 years and blessed with extensive natural gas reserves has been the Republic of Trinidad and Tobago, located just off the coast of Venezuela and across the Gulf of Mexico from prime ammonia utilizing industry. Here is a chart showing U.S. imports of nitrogen for 2009, courtesy of The Fertilizer Institute.
Trinidadian ammonia imports to the U.S. were 4.45 million tons (2010), 3.72 (2009), 3.81 (2008) and 4.36 (2007). Canada is the next largest importer with 1.01 million metric tons in 2010, but most of this ammonia is for Agrium's phosphate operation in Idaho and therefore doesn't hit the market.
However, it is estimated that about 1 million tons of ammonia was not produced this year in Trinidad due to natural gas curtailments from the National Gas Company. This is a big reason why the U.S. market for ammonia has been so tight this year, and the ammonia shortfall has fed into price hikes for not only ammonia, but also urea and UAN solutions.
Ammonia arrivals to the U.S. during September were 730,689 st, up sharply from 579,371 st during September 2010. Most of this was from Trinidad, so it is possible the situation is improving, although the government has apparently indicated more shortfalls could occur in 2012.
It should be noted that although the U.S. does not export a material amount of ammonia, it does export significant quantities of ammonium phosphate fertilizer, led by PhosChem, which represents the Mosaic Company (MOS), PCS Phosphate (POT) and CF Industries through their KEYTRADE international distributor. Therefore, DAP and MAP prices indirectly impact U.S. ammonia demand, and India is the largest importer of U.S. phosphate.
If you are starting to realize that fertilizer prices are derived from a complex set of international variables, you are right.
Here is a good report which nutshells where we stood as of November 18 on global nitrogen fertilizers prices and netbacks, courtesy of The Profercy Nitrogen Report. (click here)
A key takeaway from this report should be the right hand column which shows how much U.S. nitrogen prices were over landed import prices at the USGC.
Urea prices softened in the U.S. and global urea demand has remained strong, so with a negative $16.70/st netback importers have little incentive to ship to the U.S., especially those in Egypt and the Middle East, although we know Agrium recently lost production at their MOPCO plant there. If the much larger Abu Qir (click here) plant complex near Alexandria ever were closed, then you would a bigger spike in global nitrogen prices.
Only Trinidad and Venezuela are competitive with the USGC right now, and Venezuela has threatened to nationalize the Koch Industries plant in the latter country (Fertinitro) so its ownership is also threatened, although it continues to ship product.
According to the above report, UAN importers could land product at $17.90 under U.S. prices, and in fact, U.S. imports of UAN, a major product of our two MLPs, have been strong.
Note that the second line of defense for our two MLPs is their freight advantage over U.S. Gulf Coast product, which adds another $20 or so to their realizations, or $345/st versus the estimated $325/st landed price from active UAN importers such as Romania, Russia, Poland and Georgia.
On the ammonia import pricing, the $50 favorable import differential disappeared as Tampa prices dropped to $655/mt last week due to Trinidadian imports supplied by Yara International and others.
A certain, not probable increase in U.S. domestic supply of ammonia in 2012 will be the restart of the PCS Nitrogen Geismar, Louisiana, ammonia plant, which has a capacity of 1,500 st/day, or about 525,000 tons including turnaround downtime.
Although plant restarts are often delayed, the original PCS expected start date of Geismar was August 2012. Geismar used to import Trinidad ammonia to produce UAN but now that ammonia will not be required and will no doubt hit the market.
Another less certain but more imminent increase in U.S. Gulf Coast ammonia production is the restart of the old Beaumont, Texas, ammonia and methanol plant, which is supposed to start producing ammonia next week, with nameplate annual capacity of 250,000 tons.
Ironically, this plant was owned by Terra Industries until it was sold to Eastman Chemical as part of a plan to convert it to pet coke gasification, the same process CVR uses now. Eastman found the cost and environmental issues too prohibitive, and unloaded the plant last May to a group called Pandora Methanol and backed by Orascom, the giant Egyptian construction company (here). Orascom just bought out the minority partners, indicating the plant start-up might be in doubt (here).
Assuming this plant eventually gets going, we could see another 200-250,000 tons of ammonia production in 2012, and with Geismar starting up in Q3 2012, would increase USGC ammonia production in 2012 by 375,000 tons and 775,000 tons in 2013. The Geismar production would be upgraded to UAN but it would also displace Trinidadian production, which if it resumed, would be additive.
Without more exact information at this time, we are conservatively forecasting a half million more tons of merchant ammonia from Trinidad to the U.S. in 2012 over 2011, which would be only 3% of entire U.S. consumption but significantly more relative to merchant ammonia because most ammonia is used internally.
All in all, we could see over 1 million tons of U.S. or close to the U.S. ammonia production increase by 2013, dampening some of the fervor over nitrogen stocks.
The risks to the TNH and CVR MLP's are obvious, as both depend on buoyant ammonia prices for about 30% to 33% of their revenues respectively.
Overall, the U.S. nitrogen fertilizer market lack of aggressive import competition in 2011 has been largely due to the indirect effects from the relative absence of the biggest urea producer and exporter in the world: China. Much of the 55 million estimated tons of urea production in China is produced from low cost coal.
Chinese demand for agricultural products caused the central government to further control and tax most nitrogen and phospate fertilizer exports a few years ago.
The tax on Chinese urea exports depends on the month, with high 110% taxes from January to June and November-December, and lower taxes during July to October. Since the low tax season just ended, the opportunity for Chinese exports to the U.S. has abated for the next seven months. Note however, the Chinese have tried to get around the rules by exporting product to free trade zones during the summer, thereby staging them for reexport later in the year.
If the Chinese ever dropped their export taxes, we could see a resurgence of price competitive urea to the U.S., which goes into UAN and competes with it.
For the foreseeable future, China is out of the picture, allowing high prices for global nitrogen to persist, and exacerbated by the problems in Trinidad, Egypt and other countries where natural gas and nitrogen production is competitive.
Fertilizer Demand By U.S. Farmers And Dealers Softening
Clearly the forward price of nitrogen fertilizer intensive crops such as corn and wheat will have an impact on farmer appetite to lock in current historically high prices for ammonia and UAN.
I remember Agrium CEO Mike Wilson saying during the summer (when corn was at $7.50/bushel) that farmers would still want to buy nitrogen even if the crop price dropped to $5.00, because the farm budget would still make it profitable to apply it (I assume he meant at then current pricing). However, nitrogen fertilizer prices are higher than they were last summer, and new crop December 12 corn futures have declined to $5.41 as of the pre-open this morning.
Wheat prices are also important to nitrogen prices, and to our two MLPs in particular, becuase they also sell into the wheat growing areas of Kansas and Oklahoma. Winter wheat planting season just finished, so perhaps its not such a crucial issue for their forward selling.
But if competitively priced substandard wheat can be a substitute for corn and DDGS in the feed market, and supplant corn use, sapping prices for the latter. We had a terrible drought in Texas and Oklahoma and as far north as Kansas this year, so the spring wheat crop might be damaged and competing with corn.
The new crop December 2012 #2 soft red winter wheat price (below) traded in Chicago on the CME is indicated top open at $6.64/bushel, only a dollar or so above the corn price, but the lower grades would discounted be closer to corn.
It is my view the grain complex has been selling off across the board due to the possibility the euro will be dispensed with, in which case Europe, and particularly France, would be at an advantage vis-a-vis the U.S. in terms of wheat exports, hence the decline in USD denominated grain prices.
Valuation And Conclusion: CVR Currently Better Value Than TNH
Both TNH and CVR Partners units have dropped in price somewhat since the beginning of November. Weakness would be expected after distributions are paid, given the next payment won't be declared until February at the earliest.
CVR is currently trading at $22.20 versus the $26.05 area before it went "ex" its Q3 57.2 cent distribution on November 3, a decline of 12.8%.
TNH is currently trading at $154.74, versus the $180 area before it went "ex" its Q3 $3.96 distribution on November 9, a decline of 11.8%.
We expect CVR to generate in available cash for distribution of $2.20 per unit over the next 12 months (Q4 2011-Q3 2012), at which point the company will implement its large and transformational UAN expansion project.
The UAN expansion scheduled in Q4 should theoretically have no impact on distributions given it is fully funded. CVR expects to generate incremental EBITDA of $24 million from the UAN at full capacity, but this would require the partnership buying some extra ammonia from the oil refinery to make up the shortfall, as 160,000 tons would of excess ammonia would be required and CVR has averaged 129,200 tons of merchant ammonia.
Since debt is not free, the expansion would require CVR to service the revolving loan of $125 million they currently have already financed.
We expect CVR available cash flow per unit to peak in Q2 2012 at a $2.57 annualized rate based on their own sensitivity analysis based on prices they are booking for Q1 2012, but then drop off based on our view nitrogen prices will weaken in the summer of 2012.
The Price/ACPU multiple on CVR we currently estimate at $22.20/$2.20 or 10.1 times.
TNH is somewhat of a "black box" compared to CVR, because the MLP does not hold investor and analyst quarterly earnings conference calls and does not provide guidance on the forward pricing program book, nor on plant turnaround timing.
CVR, being newer, has been eager to give investors guidance to support the new MLP issue. In addition, CVR is a more liquid unit with a tighter spread, while I expect there are many long-term holders of TNH, which reduces trading volumes.
We estimate TNH will generate $755.3 million in revenue in 2012 or 6.2% less than our $805.3 million estimate for 2011 due to ammonia and UAN production being lower from the plant turnaround that will occur at Verdegris sometime in 2012.
We expect TNH to generate less revenue from industrial ammonia contracts with dropping natural gas prices, but offset by higher agricultural ammonia and UASN pricing as the forward book has been depleted as of Q3.
As the natural gas price "strip" is declining for 2012 and below $4, we expect TNH gross margin to be maintained at $500 million, or down only 4.8%, with $155 million for natural gas and $100 million for other product cost expenses.
We expect $20 million in 2012 costs to run the TNHCLP partnership, as well as $15 million for fees to general partner CF Industries and other G&A expense.
We therefore expect 2012 operating earnings and net earnings for TNH of $465 million (TNH pays no income taxes and has no interest expense).
TNH expects to spend $50-80 million on CAPEX and turnaround costs in 2012.
This would be a significant increase from $26.5 million in 2010, 30.4 million in 2009 but only $8.2 milion in 2008. The last major turnaround occurred in 2009 and Verdegris' operating rate droped to 93% that year, although this could have been partly due to poor market conditions for fertilizer sales.
According to the TNCLP Limited Partnership agreement, the General Partner deducts CAPEX spending from available cash before it is split with the Limited Partners, and "any reserves establishd in such quarter in such amounts as the General Partner deems to be necessary ... including reserves for future capital expenditures."
After allowing for the CAPEX expenditures disclosed by TNH in their recent Q3 10Q, we expect TNH Available Cash to come in at $385-415 million in 2012.
Our estimate for the Common Unit share of Available Cash would then be $230-235 million ($12.43-12.70) based on an estimate given the Minimum Quarterly Distribution schedule, with a mid-point of $232.8 million or $12.58/unit, down 18.1% from our 2011 estimated of $15.36.
Assuming TNH paid out all Available Cash in 2012, our estimated distribution per Common Unit would be $12.58 down from $13.91 estimate for 2011 or (9.6%).
We expect the TNH cash flow and distribution to be greater in the first half of 2012 with less in the back half of 2012, depending on how TNH provisions for the added turnaround and other capital expenditure.
The Q4 distribution for 2011 we had previously estimated at $3.75 but based on the roll off of the forward book and a drop in natural gas prices in Q4 to date, we would expect that to be a minimum and could easily be as much as $4.00. The Q4 distribution would be declared in mid to late February and paid shortly after.
The Price/ACPU multiple on TNH we currently estimate at $154/$12.58 or 12.2 times.
Therefore, CVR is about 17% cheaper than TNH based on our 2012 projected distribution rate.
Valuation includes not only a multiple, but a risk assessment.
We have described in detail the differences in TNH and CVR cost structure.
Clearly, TNH has a higher variable cost component, and could handle a shut-down or curtailment of operations more easily, as it would stop natural gas consumption. CVR Coffeyville enjoys a cheaper feedstock but basically has a higher fixed cost to run the plant, almost as much as TNH's but 36-40% the size.
When operating full-out CVR is just as profitable as TNH, and could be more profitable in a high natural gas environment. However, we have declining natural gas prices, nullifying that advantgage. In addition, TNH Verdegris has twined plants, i.e. two of each component, whereas CVR has only one air separation plant, and only one UAN reactor, although they plan to rectify that situation.
Finally, in terms of operator, we have to give a thumbs up to TNH, which has CF Industries, the largest and most powerful nitrogen producer in the U.S., as their general partner. TNH benefits from CF's distribution network, marketing prowess, and recent deal to buy all its product at the plant gate, removing accounts receivable risk.
CVR is owned by an experienced management team held over from the Farmland days, but for how long? Those managers made signficant amounts of money when the assets were sold twice over the last decade. They share duties with CVR Energy, which has oil refining as its main line of business.
Recently, CVR Energy has turned its attention to expansion, with the acquisition of the Wynwood refinery in Oklahoma for $525 million and working capital, which will require their assumption of more debt. Hence, it is not clear CVR Partners has the stable, experienced, dominant fertilizer general partner that TNH enjoys.
CVR presents better value than TNH, has greater liquidity, and provides more guidance to investors, but we believe is a riskier bet in the fertilizer yield space.
We are not providing target prices for CVR or TNH due to the uncertain events that can and will impact their results in the future, and we are not responsible for any losses that investors may experience if they purchase these units. Although we have made a best effort to check for accuracy in the above report, we take no responsibility for any errors or omissions. Please consult a financial advisor.
Please note that non-U.S. investors of either MLP may be charged U.S. withholding taxes on distributions, which may only be refundable based on filing tax returns with the IRS. Please consult a tax consultant experienced in these matters.