Description

Steel companies have had a great run, but it’s not over yet so don’t despair if you missed it.  As I’ll try to demonstrate below, the steel industry has gone through a fundamental transformation that bodes well for companies positioned to take advantage of growth and higher levels of profitability.

 

My Best Steel Idea: One such company is Ternium (TX).  TX is one of the best positioned and most profitable steel companies in the world, but it trades at a significant discount to its peers and its future cash flow generation potential.  TX is a steel producer with plants in Mexico, Argentina, and Venezuela and annual production capacity of 12 million tonnes.  The majority of shares are controlled by the Rocca family, which has an excellent reputation as an acquirer and operator of steel mill assets.  (The Rocca family also controls Tenaris (TS), one of the world’s most profitable producers of steel tubes used by the energy industry.)

 

TX was created in 2005 when the Rocca family consolidated their ownership of assets in Argentina and Venezuela, and acquired Hylsamex in Mexico.  TX shares (ADRs on NYSE) started trading in January 2006 in the low $20s and have traded between $20 and $30 since then.  Current price (June 2007) is $29.

 

Current Valuation: At $29 per share, TX’s current EV is $7.5 billion.  The company’s LTM EBITDA was $1.84 billion, yielding an EV/EBITDA multiple of only 4.1!  This is very low, even for a steel company.  A peer group of 45 publicly traded global steel companies has an average EV/EBITDA multiple of 7.3 at the time of this writing.  I’ll explain why I think TX is cheap later, but it is not because of profitability.  TX’s EBITDA margin (LTM) is among the highest in the world at 27%, 700 basis points higher than the average of the international peer group. 

 

TX’s excellent EBITDA performance is due to its low cost structure, which is partly a function of the fact that it is backward integrated into iron ore in Mexico and has attractive natural gas and iron ore pricing under long-term contracts in Venezuela.  The company is also well run and has very strong market positions in its main markets. 

 

TX also has a strong balance sheet.  In the last couple years, excellent cash flows have allowed the company to pay off $2 billion in debt.  Currently, TX has $985 million of cash and $960 million of debt on its balance sheet.  

 

Why so cheap? Can you say Hugo Chavez?  Venezuelan country risk has been an issue since the company was established and was brought to the fore earlier this year when Chavez has threatened to nationalize SIDOR, the Venezuelan steel mill of which TX owns 60%.  (The other 40% is split evenly between the government and the employees).  The government’s stated concern was that SIDOR is exporting too much steel instead of supporting the domestic Venezuelan market, and that the steel SIDOR does sell in Venezuela is priced too high. 

 

Paolo Rocca, chairman of TX, recently flew to Caracas and met with the government.  According to published reports, it appears that an agreement has been reached that avoids nationalization.  TX will have to sell more steel domestically in Venezuela and at lower prices.  TX is also required to make some capital investments in SIDOR to make sure steel is available to meet the needs of Venezuelan domestic markets.

 

My view is that the risk of SIDOR being nationalized has gone down significantly following the recent agreement between TX and the government, but I don’t believe this is a critical assumption.  In fact, Venezuela is irrelevant to the valuation of TX.  OK maybe not quite irrelevant, but not as important as the market seems to think.  According to management, SIDOR currently accounts for 25% of TX’s consolidated EBITDA.  Let’s assume the worse case prevails in Venezuela - the government does nationalize SIDOR and gives TX nothing in return.  TX’s EBITDA would decline by 25%, from $1.84 billion (LTM) to $1.38 billion.  At its current EV of $7.5 billion, TX would be trading at 5.4 X the lower EBITDA - with no exposure to Venezuela!  This still represents a significant discount to the average of 7.3 X for other steel companies.  More to the point, with the Venezuelan risk out of the way, a more rational valuation of TX would likely prevail, and may support a multiple of more than 5.4 X EBITDA.  Bottom line: The privatization of SIDOR would not be a disaster for TX.  In fact, it might even drive the valuation higher once investors take a more rational look at the business. 

 

In the analysis that follows, I assume that the agreement between TX and the government avoids nationalization of SIDOR.   

 

Growth Prospects: TX is poised to grow revenue and cash flow significantly over the next five years.  At the end of April 2007, TX announced it had reached an agreement to acquire another Mexican steel producer, Grupo IMSA, for $1.7 billion.  IMSA is a well respected steel company and TX’s principal competitor in the Mexican flat rolled sheet market.  The IMSA acquisition will adds over 3 million tonnes to TX’s finished steel production capacity, give TX a dominant position in the Mexican flat roll market and its first presence in the southern and western regions of the United States.

 

Aside from the IMSA acquisition, TX is in the process of a 3 year, $1.7 billion capital spending program that will increase steel making capacity by close to 2 million tonnes.  Together, the IMSA acquisition and the capital expansion plan will boost total capacity from 12 million tonnes this year to 17 million tonnes by 2010.

 

Selling 5 MT of additional steel capacity will not be as difficult as it might seem at first glance.  First, IMSA is well established player in Mexico, with significant downstream distribution and fabrication assets.  It sold over 3 MT of finished steel products last year, not only in Mexico, but also in Central America and the United States.  So 60% of the increase in sales will come via the IMSA acquisition.

 

The rest will come via expanding sales in TX’s existing markets, which are growing.  Mexico’s steel market grew by 6.5% per year for the last five years, while the South American market grew by 3% per year during the same period.  The International Iron & Steel Institute (IISI) expects steel demand in Central and South America to grow by close to 4% per year through 2015.  At 4% annual growth, steel demand in Mexico, Argentina, and Venezuela alone should grow by a combined 7 MT by 2010.  In addition, TX has established market positions in other central and South American countries where it can sell some of the additional production.

 

Future EBITDA.  Here’s a back of the envelope view of the EBITDA generation potential of TX out to 2011.  As you can see in the table below, it’s not difficult to imagine TX generating $2.0-$2.5 billion of EBITDA per year, even if EBITDA per tonne declines steadily, which I’ve assumed to be conservative.

 

	‘000 Tonnes	EBITDA	EBITDA
	Sold	$/Tonne	$ Million
TX 2005 (actual)	6,600	$ 259	$ 1,709
TX 2006 (actual)	9,035	$ 230	$ 2,074
TX 2007 Q1x 4 (annualized)	10,000	$ 210	$ 2,100
TX + IMSA 2008 (projected)	13,500	$ 180	$ 2,430
TX + IMSA 2009 (projected)	14,500	$ 165	$ 2,392
TX + IMSA 2010 (projected)	15,500	$ 150	$ 2,250
TX + IMSA 2011 (projected)	15.500	$ 150	$2,250

 

A couple of notes.  I assume the IMSA deal gets done in late 2007 and adds 3.5 million tones to TX shipments starting in 2008.  Then, in 2009 and 2010 the capital expansion projects finish and TX boosts shipments by 1 million tonnes per year in 2009 and 2010.  To be conservative, I assume a steadily declining spread in TX’s EBITDA per tonne, from $210 in Q1 2007 to $150 per tonne by 2010.  This is obviously a critical assumption.  To understand why I think my EBITDA assumption is reasonably conservative, read the next section, after which you may want to add $25 per tonne to my EBITDA assumptions and check out how much additional EBITDA TX could generate. 

 

The new old steel industry: Margins in the steel industry are a function of the “metal margin”: that is, the spread between raw material costs and finished steel selling prices.  From 1990 to 2003, the metal margin ranged between $150 and $250 per tonne.  (Note this spread is not equivalent to a steel company’s operating or EBITDA margin because on top of raw material costs, steel companies incur the cost of converting raw materials to a finished product.  These costs, such as labor, electricity, and supplies, range from $100 to $200+ per tonne depending on the mill and the technology used.)

 

The metal margin jumped up significantly in 2004 and has settled in at a higher level - between $300 and $450 per tonne.  For a number of reasons discussed below, I believe we’ll see a future metal margin of $250 to $350 per tonne.  This metal margin should allow TX to hit the EBITDA per tonne estimates I’ve projected in the table above, with a reasonable margin of safety.

 

Given that all good value investor believe in mean reversion, why do I think the metal margin will remain above the high end of the historic range and not revert to the long term mean of $200?  The answer is simple:  The steel industry that most of you know, the one that lurched from crisis to crisis in the 1980s and 1990s, was an aberration.  For most of its history, steel has been a growth business that generated good margins and significant wealth.  Think Andrew Carnegie!  I’m here to tell you the good old days are back.

 

·        After taking a pause for a couple decades, the steel industry is growing again, by 5%-6% per year.  For those of you who are students of history you’ll remember that the steel industry grew 6% per year from the end of WWII until the late 1970s and that capacity chased demand for much of that period.  So much so that President Eisenhower considered nationalizing the steel industry to ensure that sufficient capacity would be built to support domestic manufacturing.  The global industry is back on the growth track, driven by industrialization in much of the developed world.  Unless economic development stops, steel demand is poised to grow 5% per year for decades, just like it did from 1945 to 1978.

·        With the global industry growing at 5% per year, steel making capacity and raw material availability will be chasing demand and we are unlikely to experience the excess capacity problem we had in the 1980s and 1990s.  In those dark days the industry operated with 25% excess capacity (200 million tonnes of excess capacity with 800 million tonnes of demand), which was a disaster.  Now the excess capacity is less than 100 million tonnes and demand is over 1 billion tonnes, so excess capacity is closer to 10%.

·        The steel industry is consolidating rapidly and behaving in a more disciplined manner.  In many regions of the world, the top 3 steel companies have a combined market share of 70% or more for certain steel products and the consolidation process is in its early stages.  This is a far cry from the fragmented industry of the 80s and 90s. 

·        Much of the global capacity is now in the hands of private sector owners who want to make a profit, not governments who want to maintain employment. In the 1980s and into the 1990s, over 2/3 of all steel making capacity was owned by governments including much of the capacity in Western Europe and South America and all of the capacity behind what we called the iron curtain.  Today, government ownership of steel assets is insignificant outside China and a few other places.

·        There is much more multi-national ownership of assets now than in the past.  This ownership structure discourages the worst excesses of predatory pricing we experienced in the 1990s and early 2000s and should lead to more consistent profitability and higher prices and margins at the bottom of the market cycle.

 

I know, I know.  China is building huge amounts of capacity and is on the verge of becoming a major next exporter of steel products.  Well, before you accept that conventional wisdom, consider a couple facts.  One, Chinese steel mills import the majority of their iron ore requirements and pay world market prices to do so.  They also pay world market rates to transport ore from Brazil and Australia to China.  Two, Chinese steel mills pay similar rates for energy as mills in other parts of the world.  Third, while Chinese mills do have an advantage in hourly labor rates, labor costs for efficient steel companies outside China account for less than 15% of total costs, so China’s real labor advantage is minimal. 

 

Bottom line: Chinese steel mills are at best in the middle of the global cost curve and not more than $50 below the average producer in the Western hemisphere.  Given that it currently costs around $100 per ton, or 20% of the current price of a basic hot rolled coil, to ship steel from China to the Americas, how big of a long-term threat can Chinese mill be?  TX is further protected because its mills are at the low end of the global cost curve, in the bottom quintile for sure and maybe even the bottom decile.  TX’s manufacturing costs per tonne are costs are certainly lower than even the best mills in China.        

 

Bottom Line: The steel industry is poised to perform much better over the next two decades than it did over the last two.  Well-positioned companies, with low cost operations in growing steel markets and the conviction to take advantage of the growth opportunities, will do well.  TX is such a company, and it is on sale because the Venezuela risk is being exaggerated by Mr. Market.  TX could easily double from $29 per share, given its growth prospects, strong balance sheet, and potential to generate free cash flow. 

 

How much free cash flow?  Using the numbers from the table above, EBITDA over the next five years (2007 to 2011) is projected to be $11.4 billion.   Subtract $1.7 billion for the IMSA acquisition and $1.7 billion for capital expenditure, which is management’s estimate. In addition, subtract $1.4 billion for taxes over the next five years (2006 tax rate) and another $600 million for increase in working capital and other sundry items.  The net result is $6 billion of free cash flow over the next 5 years, or $1.2 billion per year.  At current EV of $7.5 billion, this suggests a multiple of 6.2 X FCF.

Catalyst

What’s the catalyst? What’s going to cause the market to recognize the value of TX? The recent announcement that TX reached an agreement with the government of Venezuela to avoid nationalization is a possible catalyst. The closing of the IMSA deal should also help investors realize that the importance of Venezuela is being diluted by TX’s growth in other countries. In addition, I would not be surprised to see an additional acquisition announced as well, maybe of assets located in the US. Such an acquisition would raise the company’s profile among US investors and further dilute the Venezuelan risk. Finally, continued generation of $2 billion of EBITDA per year won’t hurt the company’s case.