• In the second part of my coal posts, I analyze two coal miners: Alpha Natural Resources (ANR) and Foundation Coal (FCL). I discuss my price, tonnage and cost per ton forecasts for the two companies, ultimately deriving projected EBITDA over the next several years.
• So are coal companies cheap as a group? My answer is no, or at least ANR was not. They’re certainly not expensive, but also not irrationally cheap. But my results were interesting. My models forecast a nasty next several years for ANR but decent ones for FCL, mainly because of my bearishness for met coal relative to steam coal; FCL’s higher proportion of contracted steam coal commitments; ANR’s currently inflated margins per ton; and ANR’s somewhat higher cost structure relative to FCL’s. Current market multiples are valuing FCL at a bit of a premium to ANR, but not by much.
• So I end up being bullish on a pair trade involving long FCL and short ANR on a 1-to-1 ratio.
• The relevant PDF materials mentioned in this post can be downloaded here: Coal-Part-2-Materials

In our last post, I wrote about the coal industry, investigating supply / demand drivers and ultimately forecasting spot coal prices for the next three years. In this email, I will use those price forecasts to analyze two coal companies: Alpha Natural Resources (ANR) and Foundation Coal (FCL). Materials are attached on each company. I’ve chosen ANR and FCL because, amongst US coal producers, they’re quite different, and give a decent flavor of the entire coal market. The eight largest public US coal producers are listed on the page titled “Comparable Coal Companies”. ANR is a large met coal producer located in Central Appalachia (CAPP) and Northern Appalachia (NAPP), while Foundation is a predominantly steam coal producer in CAPP, NAPP and the Powder River Basin (PRB).

A brief description of each:

Alpha Natural Resources, Inc. owns or leases total proven and probable coal reserves of approximately 600 million tons. It has 32 active underground mines, comprised of 26 active surface mines, and 11 preparation plants located in Virginia, West Virginia, Kentucky, and Pennsylvania, as well as a road construction business in West Virginia that recovers coal. In 2008, 42% of its production was met coal, by far the highest percentage of any public US coal company. It’s a relatively higher-cost coal producer, given its high proportion of met coal assets.

Foundation Coal Holdings, Inc has 1.6bn tons of proven and probable coal reserves spread across Northern Appalachia consisting of two underground mines in southwestern Pennsylvania; Central Appalachia comprising seven underground mines and two surface mines in southern West Virginia; and the Powder River Basin consisting of two surface mines in Wyoming, as well as an underground mine in Illinois. It has 8 mining complexes that consist of 13 individual coal mines. Steam coal was 97% of production in 2008. PRB, NAPP and CAPP are each approximately a third of sales.

The most important pages in the attached materials are my projection models for ANR and FCL, which are on pages 3-4 of the ANR materials and FCL materials. In those models, I’ve used our forecasted prices from the last post to determine what EBITDA and cash flow will be for ANR and FCL over the next several years. The three key drivers of the model are (i) assumed pricing for the coal; (ii) how much tonnage is produced; and (iii) each company’s cost per ton.

Pricing

A key feature of the coal industry is that coal is negotiated via long-term contracts. While many other commodities are sold on the spot market, the vast majority of coal is sold through 1-3 year contracts. Steam coal contracts can be longer than 1 year, and are negotiated throughout the year, though more in the second half in anticipation of the upcoming fiscal year beginning in January. Domestic met coal contracts are typically 1-year in length, and are negotiated in the November to February time frame, which again corresponds to the January to December calendar year. International met coal contracts are also 1-year in length, but are negotiated more in the February to May time frame, because the contracts typically last from April 1 to March 31st of the following year. This corresponds to the Japanese fiscal year, and since Japan is the largest buyer of seaborne met coal, the Japanese fiscal year is the standard contract term for international contracts. In met coal, BHP and an affiliate are the largest producers, with ~30% of worldwide production. As a result, producers and consumers typically monitor the BHP negotiations, and use the BHP contracts as the market benchmark. It’s possible for certain producers and consumers to deviate from the BHP contracts, especially in a year like 2009 when there is so much upheaval in the market.

Contracts are take-or-pay, which means that customers are legally obligated to purchase the coal at the specified contract price, unless they choose to merely defer the shipments. As described in the last post, coal prices spiked in the first half of 2008, and many of the 2009 contracts that were negotiated were struck at record high steam and met coal prices. Coming into 2009, producers insisted that customers would be legally required to honor the high prices even though spot prices have plummeted. What happened? Steam coal has been somewhat protected, since domestic steam coal customers are mainly utilities and are not as impacted by the global downturn as some other sectors. But met coal customers have insisted that 2009 prices be renegotiated and have dared producers to sue them for breaching contract terms. Based on recent management comments, I think it’s best to assume that met coal producers will back down and recut the prices.

So let’s talk about 2009 prices for ANR and FCL.

First, ANR. As of January 16, 2009, ANR had contracted 89% of 2009 production and 38% of 2010 production. They disclose that their contracted met coal is at $108/ton and contracted steam coal is at $70/ton. By “ton”, I’m referring to short tons, and all prices are FOB mine. I’m assuming that the $108/ton is not recut lower, though there is risk to the downside. For the uncontracted 2009 met coal, I would have assumed $75/ton per my last email, but BHP came out with pricing of $115-$125 per tonne (FOB port in Australia) last week in a deal with Nippon Steel, which could serve as a benchmark price that the rest of the industry may follow. Using the midpoint of that range of $120, we would get about $80 to $85/ton FOB mine for US producers, so our estimates were a bit low. We’ll use $85/ton for 2009 uncontracted prices in our ANR model. For 2010, I assume $80/ton for the small contracted piece, and $75/ton for the tons that have not yet been contracted, up $5 from last email’s estimates. For 2011, I use $85/ton for all of the met coal. For steam coal, I assume $70/ton for 2009 contracted steam coal and $65/ton for 2010 contracted steam coal. Per the last email, 2009 uncontracted production is assumed to be $50/ton and 2010 is at $52/ton. Steam coal for 2011 is at $54/ton.

For FCL, the Company has disclosed contracted pricing of $10.43/ton for 2009 and $11.16/ton for 2010 for the PRB and $65.58/ton for 2009 and $68.06/ton for 2010 for its Appalachian coal. It does not split pricing for NAPP steam vs CAPP steam vs CAPP met coal, so I’ve made various assumptions on FCL’s 2009 and 2010 contracted prices for each of those segments. But the overall weighted average pricing for NAPP and CAPP matches the $65.58/ton and $68.06/ton numbers that the Company has disclosed. Naturally, there is downside risk to those prices in the event that some of those contracts are repriced. FCL has so far taken the route of suing steelmakers that have tried to reneg on contracted prices, unlike ANR, which has been willing to re-negotiate. For uncontracted coal, I’ve assumed the pricing we derived in our last email.

Two key differences between FCL and ANR are that FCL has a materially higher amount of coal that is (a) contracted for 2009 and 2010 and (b) has very little met coal exposure, whose customers are more likely to try to aggressively renegotiate contracts. I haven’t come across news that utility customers have been renegotiating steam coal contracts, but coal inventories are piling up and some customers are looking to defer shipments until 2010. Met coal customers, on the other hand, have been aggressive about renegotiating 2009 contracts, in addition to deferring shipments.

Tonnage

We should next discuss projected tonnage for each company.

Based on discussions with management, ANR will reduce steam coal production by 20%-25% and met coal production by 25%-30%. Part of those reductions are due to the permanent closure of its Kingwood mine in December 2008, which will reduce total company tonnage by 1mm. But a substantial part of the tonnage decline is due to weak market conditions, as customers reduce orders and lower prices make some of ANR’s higher cost mines unprofitable. I assume that 2010 tonnage stays flat from 2009.

FCL tonnage should be flat or may actually go up slightly in 2009. It added a new longwall operation to its NAPP mines in 2008, which increased production in NAPP. I assume 2009 PRB tonnage will be 52mm tons; NAPP 13mm tons and CAPP 5.6mm tons. These are loosely based on management guidance, and there’s downside risk in the event that the Company reduces tonnage in response to market conditions. I assume a slight decline from management guidance, but not much. I’m assuming that tonnage won’t drop materially because FCL has contracted virtually all of its 2009 tonnage, and much of its 2010 tonnage at high prices. We’ll see if those prices and shipments stick. I guess for now I’m accepting their optimism that steam coal customers won’t aggressively try to renegotiate contracts. Also, FCL has guided to a weak first quarter in NAPP, due to maintenance work on its two NAPP mines. Tonnage will decline and cost per ton will rise in Northern Appalachia in Q1.

Cost per Ton

Next let’s discuss production costs for each producer. The cost of mining coal is dropping, mainly due to falling diesel and steel costs, as well as lower royalty costs and more manageable labor cost appreciation.

For ANR, I estimate that diesel comprises 13% of cash costs and steel comprises 13% of cash costs. Unfortunately, for 2009, ANR has hedged 50% of its diesel at high 2008 prices. Page 4 of the ANR materials shows my cost per ton estimates. I have cost per ton dropping 9% in 2009 and then increasing 2% in 2010. One thing I haven’t really factored into my analysis is the risk that the lower tonnage could result in fixed costs being spread across a smaller revenue base, resulting in higher costs per ton. In the fourth quarter, for example, we saw costs per ton jump to $67.95 from $64.27 in the third quarter, partly because of the reduced tonnage.

Page 4 of the FCL materials shows my cost per ton estimates for FCL. FCL actually hedged the vast majority of its diesel costs for 2009, so its diesel costs are going to come out higher this year than for 2008! Generally, FCL has guided to flat to higher costs in 2009. I have cost per ton rising 0.5% in 2009 and then dropping 8% in 2010. Much of that drop is due to lower royalty costs and diesel / steel costs, offset by moderate labor and other cost increases, and it’s also due to PRB coal increasing its share of the company’s production at the expense of Appalachian coal.

EBITDA Multiples

Having analyzed pricing, tonnage and cost trends, we can derive projected EBITDA for the next several years for our two companies. For ANR, we get EBITDA of $380mm for 2009 and $100mm for 2010. I project met coal prices to start bouncing back in 2011 and 2012, so EBITDA jumps to $123mm in 2011 and $230mm in 2012. Our estimates are materially below analyst estimates, which currently stand at $514mm for 2009 and $415mm for 2010. That’s a huge difference. Either I’m very wrong or the analysts are.

Given ANR’s low leverage, the company is projected to generate substantial free cash flow in 2009, at $160mm in 2009 and ($60mm) in 2010. In terms of EBITDA multiples, the EBITDA multiple hovers around 3x in 2009, which isn’t too bad. But given the depressed 2010 EBITDA that I’m projecting, multiples jump dramatically in 2010, before dropping back to 10x in 2011 and 5x in 2012.

So is ANR equity a buy? Two facts point to ‘no’. First, my projections just show too low of an EBITDA estimate in 2010 to merit an investment in ANR today. As well, 10x 2011 EBITDA and 5x 2012 EBITDA is not attractive. Second, my projections are much lower than analyst estimates. That’s a rough sanity check that also points to avoiding ANR equity.

Is ANR equity a short? I don’t think so. Despite my analysis, coal prices are tough to project. I could be wrong. Given the equity’s very low current valuation on LTM and 2009 EBITDA, I’m not comfortable taking the risk that I’m wrong. If I am, I could see a situation where ANR doubles and I lose a lot of money on my short. Naturally, I don’t think I’m wrong and I don’t think that ANR stock will double anytime soon. But it’s not worth the risk.

At what price would I be interested? At $8.50, more than 50% below today’s prices. At $8.50, ANR would be priced under 5x 2011 EBITDA. That’s an attractive valuation. Who knows, maybe the stock will one day get to my target price. Crazier things have happened.

With FCL, the story is a bit different. I get to $390mm EBITDA in 2009 and $420 in 2010. That’s in line with analyst estimates of $416mm in 2009 and $450mm in 2010. I project a falloff in 2011 EBITDA, seeing it drop to $230mm in 2012. The rationale behind my 2009 and 2010 numbers is that FCL has contracted a high proportion of its 2009 and 2010 production, and those commitments are with utilities, not steelmakers. Assuming utilities don’t materially renegotiate those contracts, 2009 and 2010 should be decent years for Foundation. Then in 2011, EBITDA should plummet as the previously struck contracts roll off.

In terms of multiples, we get 3x TEV/2009 EBITDA; 2.5x TEV/2010 EBITDA and 4.6x TEV/2011 EBITDA. These are pretty low multiples.

There is certainly the risk of a material drop-off in 2010 domestic steam coal prices, rendering the forecasts in my last email as too lofty. Appalachian steam coal could drop into the $40s or $30s, and PRB coal could drop below $10 (actually, it already has). I’m particularly concerned about NAPP coal, since NAPP has more industrial exposure and the premium for NAPP relative to CAPP coal could soon disappear. Maybe it gets to $50/ton, which implies $350mm 2010 EBITDA. For now, I’m a bit comforted by the fact that my price forecasts are still materially below analysts’ and futures contracts, and by the fact that recessions have historically not impacted electricity demand too much. Weather trends seem equally as important as economic cycles.

There is also the risk of poor mining execution on the part of FCL; 3Q08 and 1Q09 are examples of quarters where FCL tonnage dropped and costs increased as a result of temporary mining issues.

So is FCL a long? Maybe. But I prefer a pair trade of going long FCL and short ANR. I don’t have a tremendous amount of conviction behind the trade, because I’ve just increasingly realized the difficulty of trying to predict the macroeconomic drivers that influence the price of coal. If Chinese steel demand bounces back in 2010, met coal prices could rebound and this pair trade could go awry. Or maybe met coal really is as scarce as some people claim it is, and prices will continue to hover above $90/ton. But at this point, my EBITDA estimates for FCL and ANR, relative to analysts’, are dramatically different enough that a pair trade involving the two could be interesting. On a multiple basis, I have ANR trading at 10.1x 2011 EBITDA and FCL trading at 4.6x 2011 EBITDA. Sure, I may be off on my EBITDA projections. But am I that far off to have one company at 10x 2011 EBITDA and the other at under 5x? It’s possible, but I don’t think so.

LEGAL:

THIS COMMUNICATION IS FOR INFORMATIONAL AND EDUCATIONAL PURPOSES ONLY AND SHALL NOT BE CONSTRUED TO CONSTITUTE INVESTMENT ADVICE. NOTHING CONTAINED HEREIN SHALL CONSTITUTE A SOLICITATION, RECOMMENDATION OR ENDORSEMENT TO BUY OR SELL ANY SECURITY OR OTHER FINANCIAL INSTRUMENT OR TO BUY ANY INTERESTS IN ANY INVESTMENT FUNDS OR OTHER ACCOUNTS. THE AUTHOR HAS NO OBLIGATION TO UPDATE THE INFORMATION CONTAINED HEREIN AND MAY MAKE INVESTMENT DECISIONS THAT ARE INCONSISTENT WITH THE VIEWS EXPRESSED IN THIS COMMUNICATION. THE AUTHOR MAKES NO REPRESENTATIONS OR WARRANTIES AS TO THE ACCURACY, COMPLETENESS OR TIMELINESS OF THE INFORMATION, TEXT, GRAPHICS OR OTHER ITEMS CONTAINED IN THIS COMMUNICATION. KERRISDALE CAPITAL MANAGEMENT, LLC OR AFFILIATED ENTITIES MAY OWN SECURITIES OF OR OTHERWISE HAVE AN INVESTMENT RELATED TO ANY COMPANIES MENTIONED IN THIS COMMUNICATION. THE SENDER EXPRESSLY DISCLAIMS ALL LIABILITY FOR ERRORS OR OMISSIONS IN, OR THE MISUSE OR MISINTERPRETATION OF, ANY INFORMATION CONTAINED IN THIS COMMUNICATION.