|Shares Out. (in M):||51||P/E||0.0x||0.0x|
|Market Cap (in $M):||890||P/FCF||0.0x||0.0x|
|Net Debt (in $M):||290||EBIT||0||0|
At first glance, Horsehead looks like a below-average zinc producing company with little to no upside. By just relying on the numbers without doing further research, the company even seems to be a compelling short. However, I disagree strongly. I think the company has more like an 60-80% upside potential over the short term (12-18 months) and a lot going for it in the longer term.
With an improving US economy, the implications of the shale gas boom for US steel industry underway, signs of an reindustrialisation in the US due to cheap energy, China’s still modest zinc consumption per ton of steel compared to developed countries, and up to 12% of the total zinc mines coming offline globally until 2016, it is hard to believe in lower zinc prices without a global recession crushing the demand. Not needed for the long thesis to play out well, one could even argue for higher than current zinc prices after looking at the supply and demand situation of the commodity. However, more on that vague front only at the end, if the thesis would just rest on a rise in zinc prices, I would not bother you with this long recommendation.
Description of the business
Horsehead’s core business is the production of zinc metal and zinc oxide from recycled zinc feedstock. Horsehead is quite unique - it is the only zinc producing company that uses close to 100% recycled feedstock in the US. The feedstock for the conversion into zinc metal comes primarily from Horsehead’s own electric arc furnace (EAF) dust recycling facilities that provide environmental services to the steel industry by collecting and recycling EAF dust for them. The dust is a zinc containing by-product of the EAF steel production and considered hazardous waste by the US Environmental Protection Agency (EPA) since 1988. From that point onwards, the EAF steel facilities had to take special care of the zinc containing dust. Horsehead’s predecessor company created the market for EAF dust recycling in the US in the early 1980s and provided the steel mini mill producers with an attractive alternative to chemically treating the dust and eventually dumping it into a landfill. This recycling technology has since been designated by the EPA as “best demonstrated available technology” for processing EAF dust and became the industry standard with little dust left ending up in landfills. To date Horsehead remains in an unmatched dominant position in EAF dust recycling in the US, which acts as the backbone of its significant cost advantages to produce zinc metal when combined with its new zinc metal production facility.
Since the recycling of EAF dust is not only the cheaper option for the steel companies, but also eliminates any environmental liability, less and less EAF dust ends up in landfills. Nowadays only around 10% of EAF dust ends up being shipped to landfills, compared to around 1/3 ten years ago. The end product of an EAF dust recycling facility is crude zinc oxide (CZO) also called waelz zinc oxide (WOX) which is an intermediate product with a 55-65% zinc content, that is ultimately further processed into zinc metal (98-99% purity) or other zinc end products. Before I get into the implications of the transformation and finally put everything into context with numbers, a little bit more on the EAF dust supply side and an explanation of the competitive landscape in the recycling business is relevant to fully understand and evaluate Horsehead’s unique position as a global low cost zinc producer.
To date the electric arc furnace (EAF) is the only widely used furnace besides a blast oxygen furnace (BOF) to produce steel. From the two, BOF steel production is the more traditional method, especially to produce steel from iron ore. The BOF method is also referred to as integrated steel mill production, because it is generally a huge facility where iron ore, coke and lime stone come together with heated oxygen in one very large blast furnace to produce steel. In comparison to an integrated plant, an electric arc furnace plant is very small and therefore called mini mill. The mini mills are primarily fed with scrap metal instead of ore, which is a distinctive difference. Broadly speaking, the process is more competitive (cheaper, more flexible and more environmentally friendly) than steel production in an integrated steel mill. The EAF steel production in the US was pioneered by Nucor (Horsehead’s largest supplier of EAF dust - 30-40%) in the 1960s. The company first entered the low quality rebar market and then managed to move up the quality ladder of steel products over the years. Ever since Nucor entered the steel industry with the first mini mill, the company took market share from the traditional steel mills and literally killed them little by little in every quality segment they entered (the company provides an interesting case, but it is obviously too long for this write up). Today the market share of mini mills that provide Horsehead and other recycling facilities with EAF dust is around 60% of the total steel production in the US. The market share of the mini mills has steadily increased every decade by roughly 10% since the 1960s. With the cost advantages of mini mills still in place today, their market share is estimated to grow to 70% over the next 5-10 years. There are currently quite a few mini mill projects announced that support this view.
Today the barriers to entry are higher than in the past. The market is more saturated, the EAF dust supply is mostly secured by long term contracts and less business can be taken away from landfills to justify a recycling facility for a new entrant. So this side looks rather protected, but what about the suppliers themselves? They could have some bargaining power over Horsehead. However, considering incentives the risk seems also low. A ton of steel currently sells somewhere at $700, 2% dust is created per ton and recyclers are paid $70 for the collection – so this area represents roughly 0.2% of steel production costs. Before the long term contracts where set, Nucor briefly wanted to build something up in a joint venture, but skipped plans shortly afterwards and decided to take the long term contract route.
Horsehead successfully increased and secured its EAF dust supply and capacity in the past, and I think it is sensible to consider it a base case that Horsehead will gain market share and increase its utilisation in its existing facilities (which would lower feedstock costs further) from this point onwards.
Besides the core business - which I consider as collecting EAF dust, recycling, converting it to crude zinc oxide and finally producing zinc metal - Horsehead has two additional businesses worth mentioning, Zochem (normalised EBITDA $15m) and INMTECO (normalised EBITDA $20m), both very good recent acquisitions. To put it into context directly, the maintenance capex for the whole company is around $10-12m.
INMTECO is a special nickel-bearing waste recycling facility, that provides environmental services to the stainless steel industry and it is also a leading recycler of batteries in the US. INMTECO is the only facility of this kind in North America. Apart from the technical details and differences in input and output products, the business model is relatively similar to EAF dust recycling. INMTECO serves all major stainless steel companies whose only alternative are landfills. To date 60% of the business is based on long term tolling arrangements (no commodity price exposure) where the recycled input from the stainless steel companies is directly sold back to them. The other 40% revenue comes from the recycling of batteries and other nickel-bearing waste materials. This part is exposed to the nickel price, as nickel is the main end product. The company has been operating basically at full capacity since its acquisition in December 2009 and generated on average $15m EBT per year since then. Management is currently working on expanding the capacity by 20% for $10-12m due to anticipated increased stainless steel production (a new company entered the market). Besides increased capacity, the investment should also increase margins slightly since one and the same facility will have a greater throughput due to higher power levels (I have no exact number though).
INMTECO was acquired from Vale Inco, a subsidiary of Vale, for $38m valued at 10 times the current EBITDA and around 2 times forward EBITDA in December 2009. Horsehead got its money back in 2 years. I addressed this with Horsehead and Vale and according to them Vale Inco sold because they wanted to focus on mining and smelting only and apparently they wanted that quickly. Strange things happen - maybe there is more to it, but I couldn’t figure it out. What I can say with reasonable assurance is that based on the nature of the business a negative EBITDA contribution is unlikely, 2008/09 was positive. Going forward INMTECO should contribute a normalised EBITDA of $20m, based on a $17,000 nickel price.
Zochem produces zinc oxide (72 ktons capacity) from zinc metal, a product that is used in tyres, chemical, ceramics and plastics industry. The roughly 200 customers are different from zinc metal costumers, but the demand is still cyclical, primarily driven by tire manufactures. The business buys zinc metal (SHG) at LME zinc price plus premium and sells zinc oxide on a zinc contained basis for a LME zinc price plus premium. Again, it’s unlikely that EBITDA gets negative. Zinc oxide supply would quickly dry up until the market balances and premiums improve again.
The acquisition of Zochem was very beneficial for Horsehead in terms of pay back and strategy. Horsehead shut down its 80-year old smelter (more on that later) and the old facility also produced zinc oxide (90 ktons capacity) besides zinc metal, which the new facility won’t. At the time of the acquisition in 2011, the zinc oxide business was oversupplied and the market was controlled by Horsehead. Zochem had little earnings and could be bought for $15m. Shortly after that, Horsehead started to balance the market and to increase premiums for zinc oxide again. Zochem’s capacity was recently increased from 50 to 72 ktons for $15m. With the shutdown of the old facility (Monaca), and one new entrant (the first one in over 20 years) with 40 ktons capacity this still represents an overall capacity take out of 25-28 ktons. In 2014 Zochem will contribute around $15m EBITDA according to my estimates – not bad for the $30m spent. Going forward, including the expansion, a normalised EBITDA of $15m seems balanced for this business.
The game changing transformation
Horsehead’s 80-year old smelter is replaced by a new facility that uses a different process to produce zinc metal - one that is very favourable when fed with crude zinc oxide from EAF dust. There are currently only 2 other zinc producing facilities of this kind in the world as production with traditional zinc containing ore from mines makes little sense. One facility is in Japan and the other one is in Namibia, converting mined zinc in an oxide form (a very rarely found substance to be commercially mined) into zinc metal. This is one of the cheapest ways to produce zinc because it eliminates a whole process step. However, the open pit mine will be shut down in 3 years and with it probably the facility. Horsehead’s management team visited that facility twice before they decided to build one in the US. Fortunately, Horsehead’s low cost “EAF dust mine” is unlikely to deplete any time soon. The total cost came in at $525m and it took 3 years to complete it. There were unanticipated cost over runs, but at the end the majority of the money was spent wisely.
Horsehead’s old smelter is already history. It is written down to 0 and Shell is paying all demolition costs of the facility and has an option to buy the property at the end (potential value $10m).
So what are the major changes with the new facility?
The new EBITDA break-even of the core business (recycling and producing metal) will be in the $0.40s per lb for zinc. That represents an EBITDA margin of 55-60% at current zinc prices. The list of cost savings and benefits that support the new low break-even point is long.
Conversion costs decrease due to lower labour intensity - 250 people at the new facility compared to 500 workers previously: savings of around $20-25m ($0.07-0.08/lb), lower energy costs: savings of roughly $25m (0.08$/lb). The new facility produces zinc with higher purity (SHG instead of PW) and there is a difference in premium: higher revenues ($0.03-0.04/lb). Previously unrecoverable by-products (silver and lead) will now be able to be recovered: cost savings of roughly $10-30m ($0.03-0.9/lb) depending on lead and silver prices.
Feedstock cost will come down as well. The recovery rate of zinc increases from 92% to 97-98% with the new facility, thus less CZO will be needed for the same amount of end product. Additionally, a whole conversion step (creating zinc calcine from CZO) is not required anymore. This lowers the cost by roughly $18m ($0.06/lb). It all adds up and brings Horseheads EBITDA break-even close to $0.40/lb.
In addition to the improved profitability, some more benefits arise. The maintenance capex of $7-8m for the old plant will come down to roughly $3m at the new facility. New maintenance capex for the whole company will be $10-12m. The hedging costs averaged $9m per year over the last 5 years. Once the new facility is up running at full capacity (base case in 5-6 months), management will only protect the business from disasters with far out of the money put options to remain cash flow positive. Being a global low cost producer is a solid protection to begin with. The new product (SHG zinc) will have a much larger (10 times) market than the previously produced PW Zinc. 75% of zinc used in the US is imported. Horsehead will supply around 15%. Currently there is no concern that the company might not be able to operate at full capacity. According to IR, some customers are already reserving purchases for the second half of the year to be sourced from the new facility. The location of the business is also favourable in terms of transportation costs. Most of the imported zinc comes from Canada and Horsehead can provide the same product at cheaper costs to customers.
Would somebody like to copy Horsehead’s core business? Probably yes. To my estimates the capital employed needed (if the core were built from scratch) is around $900m-1b and normalised EBIT based on current forward zinc prices (at last 5 and 10-year avg. prices) is around $180-200m. However, the chances for this to happen are slim - Horsehead’s facility consumes 75-85% of crude zinc oxide available and without it, there are no such returns.
Is the liquidity adquate to get through the transition phase? What if the new facility ran into troubles during the ramp up? I think Horsehead is save on this side. At the end of Q1, cash on hand was $65.5m and liquidity including revolvers was $80m. Zochem und INMETCO would be unaffected and yield roughly around $8m cash flow per quarter. With the new facility struggling, Horsehead could sell its crude zinc oxide and zinc calcine (it already sold a little in Q1). Feedstock costs for the new facility are calculated at $0.35/lb, of which 87% is supplied internally (140,000 tons zinc contained basis) and 13% comes from third parties for a cost of $0.50-57/lb (at 55-60% LME zinc price). Horsehead’s internal costs are somewhere close to $0.30/lb. Assuming Horsehead can also sell at the low end of its purchase prices, the company could probably generate $14m per quarter - hence in total around $24m. On the outflows, additionally to $5.5m in SG&A plus $2.5-3m maintenance capex plus $6m interest expense per quarter, the fixed costs of the new facility have to be covered. The new facility has 65% variable costs (mostly energy, reagents and supplies) and 35% fixed (labour, maintenance, overhead). Total conversion cost of the new facility including lead and silver recovery is $0.265/lb of which $0.1/lb is fixed. In total this is $30m a year ($192/ton * 156,000 tons) or $7.5m per quarter in fixed costs. So leaving aside changes in working capital, this represents a slightly positive inflow of $2 per quarter. Total capex for 2014 is budgeted at $85m. Besides maintenance capex, $55m is for the new facility, $12m for INMETCO expansion and $6m for Zochem expansion. $48m was spent in Q1. Zochem is finished, so there are roughly $15m of the $55m for the facility (lead and silver recovery unit) and $12m for INMETCO left to be spent, though INMETCO could be delayed. Overall the liquidity position could support some problems at the new facility but not a lot could go wrong without concerns coming up. Right now, the company is ramping up the new facility smoothly and by doing so, the business becomes a lot more cash flow positive. With additional working capital requirements and the full capex program, management expects to stay at this liquidity level until the end of Q4. From that point onwards, the new facility should be up running at full initial capacity.
Talking about numbers
Horsehead, like any other commodity business, is difficult to value accurately so I won’t try too hard.
But here are some of my numbers and assumptions to end up at a normalised FCF yield of 18%:
Nickel price: 17,000/ton
Zinc price: $0.95/lb (5-year avg. = $0.9/lb & 10-year avg. = $0.95/lb) below forward prices and industry consensus.
EBITDA break-even constant at $0.42/lb
SG&A: $23-24m, interest expense: $26m, maintenance capex: $12m, tax: 30%
TEV = $1.18b ¦ Net debt = $290m ¦ Market value: $890m
Normalised FCF = $160
Year 2014 2015 2016 2017
Gross profit: 120 230 260 260
SG&A: 22 23 23 24
EBITDA: 98 207 237 236
D&A: 40 40 40 40
Net Income: 30 100 128 128
FCF: -13 155 170 170
TEV/EBIT (normalised) 14 6 5.3 5.3
2015 will be the first year when strong cash flow kicks in. Sometime in 2015 I expect Horsehead to refinance and pay down debt. I also expect an increase in capacity to 175,000 tons (positive for conversion costs). This can be done without meaningful investment (I think less than $5m, the facility is already designed for it). The 2016 figures reflect this increase.
I’ll provide only a small representation of facts/figures. Industry analyst consensus for zinc aims at higher prices ($1.1-1.2/lb going forward) than I assumed to end up at a normalised FCF yield of 18%.
Total zinc metal production: 13.5m tons (2013)
Around 2m tons of supply is announced to go offline or already coming offline (including some of the biggest open pit mines with low costs) until 2016.
Average zinc head grades declined from 7% in 2000 to 5.5% in 2013.
Broadly speaking new zinc mine supply, replacing old bigger open pit projects, probably shift the cost curve upward. Average ore grades are lower, projects smaller in scale.
Zinc is more a by-product of Au, Ag, Cu mining than a primary mining ore, falling other metal prices are a drag on zinc supply/costs.
Most exploration over the last years was focused on precious metals, lack of new zinc exploration due to poor economics.
Roughly 50% of supply comes from major producers, 50% from small mines in China and Latin America. If China tackles its pollution problem the number of these mines might decline/costs increase (5m tons 40% of total supply comes from china).
Canadian mine closures, where most of the zinc ore supply for the US import comes from, will come in at 0.3m tons (total supply 0.65m tons in 2013). New projects seem to be smaller in scale and have higher costs.
Roughly 50% of zinc goes into galvanising steel. Chinas zinc galvanising consumption per ton of steel is still below US and Europe (It stood at 1/3 in 2011).
Roughly 25% zinc demand comes from transportation; Chinas level of car ownership is like the US in 1920-30s, not to mention India and other developing countries. Huge transportation budget in place in China. Living and transportation are zinc intensive, spending in China and in other developing countries should go up in this areas. Housing is not as zinc intensiv (more ungalvanised steel is used like reabrs etc.)
|Entry||06/26/2014 10:47 AM|
A couple questions on INMETCO... this always struck me as a hidden gem within Horsehead. There was talk a couple years ago about creating a JV in Japan to utilize the INMETCO process. The plan seemed to get put on the back-burner while the new zinc plant was built. Any talk to reviving this type of expansion?
Are capacity expansions at the current facility, or building other US-based facilities, strictly limited by the size of the domestic market for nickel-bearing waste streams? Any idea what % of the market they currently process?
Thanks, great write-up.
|Subject||Confidence in new plant breakeven|
|Entry||06/27/2014 09:14 AM|
Thanks for the interesting write-up.
What gives you confidence in the ~$0.40-0.45/lb breakeven for the new plant? Can these projections be verified with the results from the two other facilities of this kind or are these merely "projected" costs which could have substantial variability once the plant is up and running?
Also, any idea why 15% of the float is short (and increasing)? Is there a credible/widely-distributed short thesis on the name?
|Entry||06/27/2014 01:42 PM|
I just realised that I posted the numbers with $0.98/lb instead of $0.95/lb. So it is of by $10m compared to $0.95/lb
With $0.95/lb the core contributes around $182m (155.000 tons * $1180 per tons). + $15m from Zochem + $20m from INMETCO - 24 SG&A, EBITDA is getting closer to $195m instead of $207m.
However, the current forward prices for 2015 are already above $0.98/lb so it is again closer or above $210m. We will only find out in 2016 at what price 2015 averaged, so trying to be too precise is not worth the effort I think.
The rising zinc prices are one reason why I think my 2015 estimate might be higher than previously given ones by others.
A clarification on the EBITDA difference between 2013 and 2015: EBITDA was close to $30m in 2013 ($27.4m restructuring charges), zinc prices that year averaged $0.87/lb the difference compared to my assumption is around $35m, so in total $65m. + $100m contribution from the new factory (midpoint of the guidance) is $165m. Zochem & INMETCO had below avg. years in 2013 and less capacity (30% and 20%) compared to 2015. Add around $15m for both and you are at $180m EBITDA. Take the $110m of the guidance and the difference becomes around $5-10m or below 5%.
I calculated it a bit differently, more on a normalised basis, with the break even point of the new facility at $0.42/lb and I added Zochem and INMETCO. This way it is easier for me to evaluate the business. Maybe I am slightly off but I think it is accurate enough, especially with the high impact from changes in zinc prices.
|Subject||RE: Confidence in new plant breakeven|
|Entry||06/27/2014 02:39 PM|
2 good questions.
First on the short thesis, I am looking for a thesis for quite some time now but I just can't find one... If others are aware of a short thesis, I would be happy to get the information. Maybe I miss something material.
Until I know of a short thesis I have a different theory for the high short interest. However, I am not experienced in this area so I could be easily wrong. Based on historic data, Horsehead is richly valued. The numbers look terrible, the company is loss making, debt and capex are increasing but it just doesn't help...
So my guess is that Horsehead is picked up by quant based shorts and the high short interest is a result of many small quant short positions in the relatively small marketcap. In total not a lot of money is short, $80m that grew to $120m (short interest relatively since the beginning of the year). Thus, I don't rule it out until somebody experienced in this area tells me that this very unlikely.
My other guess is that some of the longs might have bought put options to create a more appealing risk/reward situation for them, but again this is a bold guess. The biggest shareholder (12.5% owner) has around 30% of his portfolio in this stock so maybe he owns quite some puts...
I think it is sensible to trust the guidance (base case).
Some parts like lower labour and energy costs or the elimination of a process step are pretty safe, others are more at risk but management spent a lot of time and money to make sure the guidance is reliable. One example is the silver and lead recovery unit which I think is the least reliable variable of the cost reductions. The company run a set of tests and initial tests results were good but in some circumstances, the tests results showed that the anticipated recovery rates were lower than expected. If I remember it correctly Management then spent additionally $30m just to make sure that they will achieve the lead and silver recovery rates with a higher certainty.
Hope that helps.
|Subject||Is there a potential pair you would recommend?|
|Entry||06/27/2014 02:45 PM|
Is there any name in the space that you would recommend shorting against this to take out some of the general market/metal risk/sector sentiment risk? Thanks.
|Subject||RE: Is there a potential pair you would recommend?|
|Entry||06/29/2014 06:09 AM|
Sorry, I don't have a short recommendation. I generally don't really follow mining companies. I'd guess a good way to hedge/take out risk is to buy zinc puts directly. I wouldn't short zinc or even poor zinc mining companies. I think it would be too painful if the zinc bull thesis plays out. Prices skyrocket in 06/07 and I'd say there is a small chance for this to happen again.
|Entry||06/29/2014 08:18 AM|
Sorry for the late reply.
There are currently no new talks about creating a JV in Japan, but I wouldn't rule out an international JV completely - might be something for 15/16. I think management has (good) plans/opportunities for INMETCO, but so far IR was very cloudy in this area. I still lack some information on the current expansion and market situation in the US, but since INMETCO is not central to the thesis and the downside seems protected, I could live with it.
However, the company still needs to get back to me on some of my INMETCO questions, so hopefully I have something more concrete for you next week, apart from what you can find in the 10k's and so on.
My current guess is, is the market size limits the possibility of another facility of INMETCO's size in the US, at least from a tolling business perspective. I have no market share numbers though.
|Entry||10/02/2014 11:20 AM|
just quick thoughts here on the 12% sell off on news that the new factory is running into some start up problems.
I remember an interview with Michael Price not long ago where he said one of his favorite investment set ups was when companies had temporary production problems due to factory shut downs etc. I believe he named some pharmaceutical co as a current example (at the time) and pretty much guaranteed that the stock would be 50% higher in 2-3 quarters when production was back on line because temporary shut downs have nothing to do with normalized earnings power.
This is a little bit different, but as i see it the same general premise applies.
the stock has traded off on other news in recent weeks (china possibly slowing, general small cap weakness etc) but i don't think the supply side of the zinc equation has changed at all, and looking out 2-3 years the demand side is likely reasonably intact barring major economic slow down etc.
|Entry||09/22/2015 10:47 AM|
any updates here? recent selling can only be 1) absolute panic or 2) Pabrai trying to liquidate his entire position as fast as possible to squeeze it in before having to file on it.
based on his public statements regarding how he invests #2 seems unlikely.
As for #1, reason for panic seems to be liquidity concerns, which may be warranted. however, it is not hard to argue that in a "normal" zinc pricing environment the stock should trade in the mid $20s assuming they can get the new plant online at some point, so even if you factor in serious dilution, the stock looks cheap if you can handle the near term volatility...
|Subject||Re: Re: Re: liquidity|
|Entry||09/23/2015 09:12 AM|
agree - seems extremely unlikely. i think it is fair to say that his estimate of intrinsic value has likely been reduced by his estimate of likely dilution, but based on "normal" earnings power of the completed facility and "normal" zinc prices (obviously both of these are difficult to get a handle on) the stock still appears cheap even if there is in fact a dilutive event... but obviously it has been and will continue to be a very messy ride, which most investors are not set up for / capable of handling.
|Subject||Re: Re: Re: Re: liquidity|
|Entry||02/14/2016 06:04 PM|
|Entry||08/20/2016 02:58 PM|
Bids should be in on Aug 26th per the process letter filing (Docket 1455)
|Subject||Re: Re: BK|
|Entry||08/31/2016 01:58 PM|
What is the hurdle for equity holders to be in the money?
|Subject||Re: Re: Re: BK|
|Entry||09/01/2016 02:34 PM|
would this be the math?
thus for equity to have positive value the enterprise must be worth $401m
I have heard $80-$90m in EBITDA projections at current zinc price
at midpoint this means 4.7x multiple on $85m
should $85m be worth 6x, there would be $110m in eq value or $1.82 per share
these are all best guesses - this is a zero if the creditors get their way