Description

Abitibi is one of several Canadian lumber/newsprint companies that have suffered through a miserable five years of increasing energy costs, a significantly appreciated Canadian dollar, and unilaterally imposed tariffs.  As a result, the valuation of ABY has trended ever lower, despite Abitibi’s efforts to pay down debt and otherwise restructure.  Abitibi bears the distinction of being the worst performing stock in the TSX composite index over the last two decades.

Despite, or perhaps because of this, several noted value investors have waded into this distressed industry, establishing major positions in several stocks:

• Third Avenue Management owns 6% of Abitibi, 38% of Catalyst Paper Corp (CTL CN), and 15% of Canfor Corp (CFP CN).
• Jimmy Pattison, one of Canada’s wealthiest billionaires, increased his stake in Canfor to 25%, which is approximately ten percent of his estimated net worth, according to The Globe and Mail.
• SFK Pulp Fund (S FK-U CN) is 21% owned by funds affiliated with noted Canadian value investor Peter Cundill.

Given the pedigree of investors, it’s no surprise to see an odd mix of assets accompanied by a general tone from the analyst community that ranges between appropriate caution amid the newsprint industry’s perils and a hopefully generous tone of wanting to like the stock, but once burned, twice learned.  It is no surprise to see such symptoms of investor fatigue in a long struggling stock.

Abitibi is best thought of in terms of two assets: the newsprint/paper/lumber manufacturing and their hydroelectric power assets.  My primary interest in this company was to see if their hydroelectric assets were being masked by the distress in their core operating division, so let’s begin there, since Abitibi is in the midst of a partial spin-out of their hydro assets.

First, a word about valuation:

I admit to having a tough time with writeups that postpone the valuation metrics to the end—rightly or wrongly, I usually prejudge it as a sign that there is no compelling metric on which to hang one’s hat.  So I’ll start with an overview, but caution that it is tricky to do a valuation on such an out-of-favor business in a battered industry.  There are very few positive metrics to lean on in order to generate meaningful multiples.  Currently, ABY has a market cap of C$1.21 billion, with an EV of C$5.09 billion, making it highly leveraged at the current valuation.  A small shift in EV provides a lot of bang for the market cap buck—in either direction.

ABY’s enterprise value is a fraction of the company’s current assets at cost of C$8.3 billion.  The company is clearly trading below the historical replacement cost of its assets, and perhaps deservedly so.  However, as the industry removes capacity, the survivors might well be able to generate decent returns on these assets.

Meanwhile, the balance sheet situation is not as intimidating as might first appear, given the EV/market cap ratio, and the stock seems to be discounted for investor stress and fatigue.  I admit that having a stock trade so close, on a percentage basis of EV, to having its market cap wiped out makes me question the downside margin-of-safety in this stock and wonder if the market is trying to instruct me. 

And perhaps it is.  While there is no doubt that management’s efforts at restructuring and their monetizing assets for debt reduction has bought the company time, it will take a reduction in several headwinds for the company to return to significant profitability.  Chief among these is the damaging effects of the strong Canadian dollar on operations.

However, I have been looking for a way to play the unwinding of some of the excesses of the “commodity indexation” era.  Given the enormous flow of funds into commodity indices and other real assets, many of which have significantly negative carry and roll characteristics, it makes sense that a stock which benefits from the unwind of some of the indexation era’s secondary effects—like the strong Canadian dollar, would be something worth further investigation.

I’ll walk through the facets of several valuation models, incorporating various aspects of replacement cost, comparable asset sales, normalized EBITDA and distressed EBITDA, and try to mangle them together in a sum-of-the-parts way.  Unfortunately, the attributes of the different divisions and their interdependencies across operations does not readily lend itself to an easy summary.

Since 2001, the company has experienced a negative impact to EBITDA from lumber duties (since rescinded), fiber costs, energy costs, and foreign exchange that totaled C$1.8 billion for the last four full fiscal years.  Foreign exchange, namely the strength of the Canadian dollar versus the US dollar, is by far the most significant of these components.  With the froth in commodities and energy prices somewhat diminished, there is a decent possibilities that the company’s costs will rapidly decline as the Canadian dollar weakens.

Hydroelectric assets:

The annual report does a good job of summarizing this.  At-a-glance hydroelectric power accounted for five million megawatts, of which their ownership share was 3.8 million MWh.  Here is the complete breakdown:

While it is clear that the 2005 impact on EBITDA was a positive $108 million, what is the appropriate value for these assets?  What is the current market value for 3.8 million MWh of Canadian electricity? 

If we attach an 8% yield to the hydroelectric assets, these would be worth approximately C$1.35 billion.  Attaching a 7% yield would C$1.54 billion.  Granted, the monetization of these assets through an IPO will incur underwriting fees and some concession, but only about one-eighth of these assets are actually being sold in the proposed offering.

Due to regulatory delays, the announced IPO of an income fund to hold a 47.5% minority interest in its Ontario hydroelectric assets has been delayed until the first quarter of 2007.  Abitibi intends this fund to be their growth vehicle in energy generation.  While I expect that the recent tax ruling will decrease the proceeds of the offering, it isn’t a dramatic adjustment, given both the EV of Abitibi and the attractiveness of the assets being offered.

This IPO will have a collective capacity of 137MW, equivalent to 25% of the aforementioned 544MW share of capacity.  Working proportionately from the figures in the table above, which is conservative, given that Ontario power rates are the highest in Canada, the 25% of C$108M in EBITDA, using an 8% earnings yield, puts a conservative estimate for gross proceeds near C$160m. 

The proceeds will be used to exercise its option to acquire the remaining interest in its Augusta, GA newsprint mill at around 5x EBITDA.  From a cash flow perspective, the company would generate approximately C$40m in savings as it would no longer pay a quarterly dividend to its non-controlling shareholder.

Newsprint and paper manufacturing:

Abitibi is the largest manufacturer and marketer of newsprint in North America.  Newsprint, if you recall, is used to make these things called newspapers, which evidently used to be popular before Diet Coke met Mentos on YouTube. 

Abitibi is the number one forest products company in Canada in terms of total certified woodlands, with 16.8 million hectares of certified woodlands. 

They have low cost mills, energy sufficiency, and good fiber supply. 

Market position:

• Largest producer of newsprint in North America (3.8 million tons of capacity).  This generates 53% of Abitibi’s revenues.
• Largest producer of commercial printing papers (2 million tons of capacity).  This generates 33% of revenues.  It’s 33% market share is more than twice the capacity of runner-up Catalyst Paper Corp (CTL CN).
• One of the largest recyclers of newspapers and magazines.
• Largest lumber producer in eastern Canada, fifth largest in North America (2.2 billion board feet of capacity).

Over the last three years the company has closed one million tons of capacity. 

Abitibi is self sufficient in fiber: the company’s extensive cutting rights and long-term supply agreements provide about 91% of the company’s virgin fiber requirements.   In addition, it provides for approximately 40% of its own recycled fiber requirements.

Energy self sufficiency: the company owns about 544 MW of low-cost hydroelectric assets, which generate about 25% of Abitibi’s total energy requirements.  Mentioning this, of course, provides some overlap (and context) to the hydroelectric assets mentioned above.

In their 2005 shareholders letter, Abitibi reviewed three targeted areas for improvement:

• Put newsprint mills in the best half of the cost curve.
• Achieve $175m in improved EBITDA through cost and productivity measures.
• Target an additional $75m in improved EBITDA through higher margin products and the potential re-launch of a Texas mill.

Of course, the shift to higher margin products through conversions and re-launches may improve EBITDA, but it would have also required Capex, so those were ditched.  Instead, they shut down three mills representing 434,000 tons of annual newsprint production, and gave notice at two other mills for cutbacks that removed an additional 120,000 tons of newsprint annually (by shutting down one machine at each mill). 

On top of that, they sold their 50% share in PanAsia Paper, enabling the reduction of $1 billion in debt.  In addition, they sold timberlands near Thunder Bay, Ontario for $55 million (485,000 acres or 196,000 hectares).  At year-end 2005, long-term debt had declined to $3.8 billion from $6.1 billion in 2001.

How to value any of this?

The timberlands are difficult to value because there simply aren’t many comparable sales.  Most timberland is owned by the Canadian government.  Abitibi received US$96/acre when it sold its Thunder Bay, Ontario timberland last year.  It was simultaneously closing a sawmill and taking an impairment charge nearly equal to the sale proceeds, so $96/acre, while modest, might also be seen as the floor valuation for timberland that can’t be presently monetized in a profitable way. 

In New England, land values at the time typically ranged from US$200 to $350 per acre.  A comparable Canadian transaction for timberland associated with an operating pulp mill was the May sale by Neenah Paper of 500,000 acres for $140m, or US$280/acre.  Lending further credence to the comparability is that both the Neenah transaction and the Abitibi transaction involved the same purchaser (Wagner Forest Mgmt).

On the other hand, as part of the Augusta, GA transaction, Abitibi will be selling 55,000 acres of Georgia land and is expected to receive approximately $1,250/acre. 

How can one value the actual mills themselves, given how much mill capacity has been shuttered?  One thing to keep in mind is that the mills that have been shut were the least economically viable.  As with the timberlands that are located amidst unprofitable mill locations, mills that have profitable characteristics and lower costs still retain positive value.

Since I am not overly familiar with the newsprint manufacturing industry, but noticed no discernable howls of protest from the analyst community, I’ll take as a baseline the company’s announced transaction to buy the minority interest in their Augusta, GA mill.  Abitibi is paying $190m to buy 47.5% of 421,000 tpy capacity (equivalent to $400m for the whole mill).  In the first half of 2006, the mill had a six month EBITDA of $39m, making the purchase price 5.1x EBITDA.

Given the dismal company-wide EBITDA and lack of individual breakdown on a mill-by-mill basis, I’m left to guess at how to value the company’s 5.8 million tons of capacity.  The Augusta transaction implied a valuation of $950 per capacity ton.  It seems quite a stretch to apply this value across the company’s entire capacity, which would suggest a valuation of US$5.5m.  On the other hand, 5.1x EBITDA is not an outrageous valuation, so that tempers the assessment.

Net book value was C$4.26 billion at year end 2005.  This represents assets at cost of C$8.3 billion and accumulated depreciation of C$4.0 billion.  The company is clearly trading below the historical replacement cost of its assets, and it is interesting to see the historical cost and current book value of these assets.

The book value of the hydroelectric plants is C$411 million, 77% of historical cost.  The value of the sawmills, buildings, pulp and paper mill production equipment is C$3.66 billion, 49% of cost, despite the write downs for shuttered facilities.  Furthermore, the net book value of idled and permanently closed facilities is just $317m, or less than 10% of the aforementioned equipment.

If we took current book value on the equipment and added to it the market value of the hydroelectric assets, we get to the current EV.  That leaves out the value of the timberlands not to mention the rights managements for 16.5 million hectares.

I see the valuation story as being of two parts.  In the first part, the hydroelectric assets cover the intermediate term balance sheet risks (there is no significant risk for 2007), as does the lumber duty rebate.  In the second part, the rejuvenation of the newsprint and paper manufacturing business occurs amidst the reduction of industry capacity and through the reduction of the industry’s headwinds.

Headwinds:

What has turned Abitibi’s earnings into such a miserable story has been the continually strengthening Canadian dollar, combined with the unparalleled increase in energy costs.  In 2005, the Canadian dollar appreciation over foreign currencies was 7.4%, and resulted in an unfavorable impact of $252 million for the year.  Natural gas prices increased by 30% in 2005, although this was somewhat mitigated by the fact that Abitibi generates 34% of its own power needs.  However, the energy price increase still resulted in a substantial increase in manufacturing costs.

One of the contrarian features to this stock is that it is a major producer of newsprint.  However, while the travails of the newspaper industry have had an impact on newsprint, actual newsprint usage is fine and newsprint prices have had a significant increase, albeit much of that might well be attributed to energy costs.

However, the newspaper debate and its attendant affect on valuations of the companies involved—both the newspapers and online advertising platforms, is only a part of the story involved in appraising the newsprint industry.

Managements focus has been on being a low-cost producer, and their size gave them a broad base of assets to assess, and they ranked their properties on the ability of the mill to become low-cost with minimal infusion of capital while still generating net cash.  Older machinery, access to inexpensive energy and fibre, and proximity to market were the ultimate determinants in the selection of continuing operations.

Abitibi now has the lowest SG&A, as a percentage of sales, in the industry.  In addition, they have made a dramatic shift from truck to rail.

Something I find somewhat encouraging is that management has no reticence in addressing the issue of shareholder disappointment.  Undoubtedly, they’ve had much practice in this area.

As management puts the question:

“If your five-year plan was well conceived and well executed, and you have achieved much of what you set out to do, why hasn’t this been reflected in an increase in shareholder value?”

“[Since 2001…]  It had all the makings of ‘The Perfect Storm’ – a combination of elements coalescing without warning. The four factors included: our largest trading partner unilaterally imposing an import duty on lumber; the spike in the price of fibre; the rapid increase in energy costs; and the unprecedented strengthening of the Canadian dollar.”

To that end, here is how they quantify the aforementioned “perfect storm”:

Softwood Lumber Duty: with all manufacturing facilities in Canada and almost 2/3rds of sales to the U.S., the 31% import duty on lumber, levied in 2002, resulted in a negative C$270m since then.

Fibre: increased stumpage and harvesting costs have driven up prices, so that today, wood chips produced in Quebec are the most expensive in the world. (Cumulative impact since 2001 of C$219m.)

Energy costs: energy represents approximately 25-30% of the total manufacturing cost.  Abitibi generates 34% of their own power, which is a higher degree of self-sufficiency than many of its competitors.  Another 35% is purchased from power companies in regulated jurisdictions, while the remaining 31% has free market exposure.  (Post-2001 negative impact of C$243m.)

Foreign exchange: the continual strengthening of the Canadian dollar from 63 cents in Jan02 to 86 cents in Jan06 (for those of us used to looking at it the other way, the fall of the US$ from $1.59 to $1.16), the cumulative negative impact has been over C$1 billion.  In 2005, the negative EBITDA impact was an extraordinary C$598m.

Total impact upon cash generation since 2001 has been C$1.8 billion.

What might make one cautiously optimistic about the future?

• The softwood lumber dispute has been settled, and Canadian exporters will be refunded 80% of the duties collected (plus interest).  So they received $235m in Q4 (just last week).
• Bond debts maturing in 2006 and 2007 are just $76 million.
• Monetization of the hydroelectric assets, combined with the cash flow savings of the Augusta, GA purchase will net out nicely for further debt reduction.

Taking a stab at normalizing EBITDA:

It’s a bit of a leap to call something “normalizing” when it assumes a significant decline in a foreign currency, that’s easy enough to admit.  Still, it’s worth reviewing what happens if the Canadian dollar were to fall. 

What would happen if the Canadian dollar declined from $0.89 to $0.75 (or for those of us who think in terms of U.S. dollars, the $US rally from $1.12 to $1.33)?

For the sake of attribution and to use a general case, Prudential Equity does a good job of hypothesizing just such a scenario.  Their model produces a US$7.00 price target at 6x EV/EBITDA.

Of course, the scenarios in which the Canadian dollar were to decline might also suggest other tweaks to the model, such as the cost of energy inputs, etc.  My suggestion was only to provide a glimpse of a “reversion to the mean” so-to-speak, in the Canadian dollar might impact the stock.

Abitibi provides a sensitivity analysis for Foreign Exchange, among many other sensitivity analyses.  On net earnings per share, the cash and non-cash impact respectively from a US$0.01 change to the C$ would produce $0.044 and $0.082 per share.  What we have here is an asset play with a very interesting and potentially explosive put on the Canadian dollar!

Disclaimer:

This is not meant to be a buy or sell recommendation, and my firm frequently has both long and short positions in many of the securities mentioned on a regular basis.   

Catalyst