Mercer operates three pulp mills:
Canfor Pulp Products owns 49.8% of the operating subsidiary Canfor Pulp Limited Partnership (CPLP), with the other 50.2% being owned by Canfor Corporation (CFP). Canfor Pulp was spun out of CFP in 2006. CPLP owns three NBSK mills in British Columbia, Canada with a combined capacity of 1.1 million mt, making them the third largest producer of NBSK market pulp. One of the mills also processes 140,000 mt of kraft paper (15% of revenues).
(Note all currency figures in reporting units — MERC in Euros and CFX in Canadian Dollars — unless otherwise noted.)
Both Mercer and Canfor have high capacity utilization regardless of market pricing levels:
They are both on the lower end of the production cost curve so it makes sense for them to keep running even as marginal producers take downtime during periods of lower NBSK pricing. (Canfor did take some downtime for capital projects with depressed market prices at the end of 2008.) NBSK market pricing is the primary determinant of revenues, and realized pricing is typically about 85% of list price.
Aside from market pricing for the commodity there are two other issues to consider — exchange rates and energy revenues.
Exchange rates can have a large impact on MERC and CFX revenues. NBSK prices are quoted in U.S. dollars, while MERC operates primarily in euros and CFX in Canadian dollars. So any change in exchange rates flows straight to the bottom line, with both companies benefiting from a stronger US dollar. Canfor estimates that a 1 cent change in the exchange rate of U.S. to Canadian dollars has the same impact ($6 million in EBITDA) as a $10 (U.S.) change in the price per ton of NBSK. If the recent strength of the USD against the EUR and CAD holds, it will offset about half of the recent price declines in NBSK.
The pulp production process generates large amounts of biofuels as byproducts. These fuels can be used as internal power for the mills as well as sold externally. In the last several years MERC has invested in expanding its external energy sales. They are aiming to generate 700,000 megawatt hours in external sales and are close to that run rate in the most recent quarter. That should generate about €55-60 million in revenue, which has little incremental operating expense. The energy revenues are key to offsetting some of the volatility in NBSK pricing and effectively lowering the cash operating cost of the mills.
Canfor is a bit farther behind in terms of energy sales, having only $5.4 million in energy revenues in 2010. Canfor is increasing their biofuel generation as part of the “Green Transformation Project” largely funded by the Canadian government. Energy sales will be several million dollars higher in 2011. Management is aiming to grow energy revenues to $50 million, but that is at least several years off.
The major operating expense for pulp makers is wood fiber, which is the raw material used to make pulp. Fiber makes up 40-50% of variable costs. British Columbia has an abundant supply of wood fiber, and has seen an uptick in harvesting levels in the last few years due to the pine beetle infestation. However, balancing the abundant supply is the reduced sawmill activity with the downturn in the construction markets. Pulp producers have not been able to source their fiber needs from residual sawmill wood chips alone and have had to use more expensive whole logs. In Germany, Mercer has also faced competition for wood chips from producers of various biofuels who are heavily subsidized by European governments. Mercer’s fiber costs per ton of pulp have reason 33% since 2006.
Mercer possesses the most modern asset base in the industry as well as some of the largest scale mills in Stendal and Celgar (chart from September 2011 Mercer presentation):
The young mill age helps to lower maintenance capex and higher up-time while the large mill size gives MERC economies of scale. From a mill operations standpoint, MERC should be on the lower end of the industry cost curve. Canfor also has some of the larger and most cost competitive mills in the industry albeit older ones.
This chart compares MERC EBITDA margins with Canfor EBITDA margins in their pulp segment (with corporate level costs allocated by percentage of total revenue):
We see that Canfor has typically had higher margins, mainly due to the larger headwinds Mercer has faced with wood fibers.
Capital Structure and Allocation
Mercer has 733 million euros of debt and 287 million of equity (including 132 million in cash). However, there are several points that mitigate the leverage risk:
Mercer has used most of its free cash flow for deleveraging and discretionary capital projects. Capex has averaged in the €30 million range the last several years, although most of that is growth capex as the mills are all relatively young with low maintenance requirements. They have also instituted a USD25 million stock buyback plan and have bought back $10.6 million so far this year.
The debt against the Stendal mill has a reserve fund that requires a year of principal and interest payments to be put in before cash generated by the mill can be distributed to Mercer. On top of this, in 2009 Mercer fell behind on principal payments on the Stendal debt and had to restructure the debt. MERC reached an agreement with the Stendal lenders to defer €164 million in principal payments provided that no cash from the mill operations could be distributed to MERC until the €164 million was put into the reserve fund. What this all means is that MERC equity will not see any cash from Stendhal, which is 43% of their production capacity, for quite a few years. MERC needs to put about €220 million euros into the reserve fund and the fund balance is currently €31.7 million. Stendal generated €36.7 million in cash from operations in 2010, and €47.6 million through nine months of 2011.
The Canfor Partnership has $114.3 million in debt and $466.4 million in equity ($25 million in excess cash), which appears to be a sustainable amount of leverage. CFX had been set up as an income trust (something akin to an MLP in the U.S.) prior to this year when Canada eliminated the tax advantages associated with the structure. CFX converted to a regular corporate structure, but still plans on distributing the after tax free cash flow from the Partnership to shareholders. Canfor management has guided to maintenance capex of $10-$15 million and normalized capex of $15-$20 million. Capex has been higher this year (an estimated $67 million for the year net of government contributions) due to GTP environmental projects that are being primarily funded by the Canadian government.
Earnings Power and Valuation
The key question in terms of earnings power is of course NBSK pricing. We can look at what would happen in a range of cases. Current NBSK pricing is $846/mt (USD) in Europe and $890/mt in North America. NBSK peaked over $1,000 and fell close to $600 in the recent trough in 09. Given that operating costs have risen since 09 I think a reasonable range to look at is $700-$1,000. Both MERC and CFX look like they would be at breakeven on a cash basis on pulp sales in the $625-650 range, so more marginal producers would halt production at much higher levels.
Please see my model here for all the details but the major assumptions are:
In terms of valuation multiples, commodity paper stocks generally trade in the 4-6X EV/EBITDA range. For Mercer that is probably the more relevant metric than P/E or FCF yield given their leverage and the fact that most of the free cash for the foreseeable future will go to the debt holders. For CFX, FCF yield is very relevant as they will be paying almost all of it out to shareholders. The cash yield on CFX has bounced around as NBSK pricing has moved more quickly than the quarterly dividends. It is currently 16.3%. At the peak of NBSK in the summer the yield was about 8.5%.
At current NBSK prices I project $201 million in annual EBITDA for CFX and €191 million in EBITDA for MERC. At a 5X EBITDA multiple that would come out to a $12.85 share price for CFX and a $8.80 share price for MERC. On that basis both stocks are undervalued (CFX by 25% and MERC by 33%), so the market might be pricing in further NBSK pricing deterioration. It is worth noting that these scenarios assume the large recent gains of the USD hold.
Here are the scenarios:
Clearly both MERC and CFX are extremely leveraged to NBSK pricing. The big question is whether the NBSK run up to $1,000 was an aberration (caused by the Chile earthquake, dissolving pulp diverting capacity, and a slew of China paper starts) or a “new normal” given the permanently shuttered NBSK capacity and growing Chinese demand. At some point I think you have a margin of safety given that demand for NBSK is not going to evaporate and MERC and CFX are both relatively low cost producers. But given the sharp downward trend in pricing in the past few months and elevated producer inventories it seems like there is still some time to wait for a bottom. (It will also be interesting to see whether CFX can maintain their .40 cent quarterly dividend. At current pricing it still looks safe, but probably not if NA prices fall to $850.)
MERC or CFX?
Given the current valuation gap between them is not large, I like CFX better given the higher pulp margins, lower leverage, and large cash yield. The main advantage of MERC is the large energy revenues, which will help to smooth out earnings. MERC also has newer assets, and I like that they are buying back stock. But MERC still needs to delever (and all of the Stendal cash flow will be going to that end) so for the time being it looks like equity holders will be better rewarded with CFX.
Disclosure: No position in stocks mentioned.
Elie Rosenberg runs a value investing research website at valueslant.com. Sign up here to get his free value investing ideas and analysis by email and get his free ebook, "16 Ways to Find Undervalued Stocks."