Please see disclosures on page 123 Carnegie Securities Research This report was completed and disseminated 02 May 2018: 07:27 CET c Norway Materials Initiating Coverage BUY (since 02 May 2018) 02 May 2018 Share price: NOK31.7 Target price: NOK37.0 Elkem Silicone rally Research analysts: Morten Normann +47 22 00 93 58 [email protected]We are initiating coverage of Elkem with a BUY recommendation and NOK37 target price. Our target price is based on Elkem’s closest peers, with a 15% discount. From being a purely metallurgical company, Elkem, through acquisitions of silicon assets from its parent, has become a specialised chemical metallurgical company. Elkem is today a fully integrated global player from quartz to silicones. About 50% of revenue is derived from specialised products. Elkem enjoys a number 1–3 market position in all four of its business areas. In China, the largest and also the strongest growing market in silicones, Elkem is the largest player. In its upstream divisions, it has a leading cost position, as two of its Norwegian furnaces are the second and third most cost-effective facilities in the world. We expect EBITDA to increase by 26% in 2018 on a combination of continuing growth in Silicones and cyclical higher prices in the other divisions. Silicon metal and ferrosilicon prices have increased through 2017 and 2018, and Elkem’s results will gradually reflect these higher prices. For 2019 and 2020 we expect EBITDA to grow by 6–8% per year. Elkem’s long-term growth area is Silicones, where we expect a revenue CAGR(17–20) of 8.2% with EBITDA growing by around 13% in the same period to around NOK2.16bn. The combination of asset-light downstream capex and high grading the Chinese product portfolio could increase Silicones EBITDA by more than 60% on top of our 2020 estimates. Elkem will announce Q1(18) results on 8 May. We expect Q1(18) EBITDA of NOK970m and EPS of NOK0.76. Compared to Wacker Silicones’ Q1(18) result and consensus estimates for Ferroglobe, we believe that there is upside potential to our estimates. On our estimates and target price of NOK37, the shares trade at a P/E of 11.2x for 2018e and 9.2x for 2019e. The corresponding EV/EBITDA is 6.8x and 6.0x, respectively. Changes in this report From To Chg EPS adj. 2018e 3.3 EPS adj. 2019e 4.0 EPS adj. 2020e 4.5 Upcoming events Key facts Source: Carnegie Research, FactSet Key figures (NOK) 2017 2018e 2019e 2020e Sales 16,658 23,503 24,981 26,446 EBITDA 2,098 3,979 4,295 4,577 EBIT 1,355 2,729 3,045 3,327 EPS n.a. 3.32 4.00 4.46 EPS adj. n.a. 3.32 4.00 4.46 DPS 0.00 1.33 1.60 1.78 Sales growth Y/Y 17% 41% 6% 6% EPS adj. growth Y/Y n.a. n.a. 21% 11% EBIT margin 8.1% 11.6% 12.2% 12.6% P/E adj. n.a. 9.6 7.9 7.1 EV/EBIT 18.2 8.2 7.0 6.0 EV/EBITA 18.0 8.2 7.0 6.0 EV/EBITDA 11.8 5.6 5.0 4.4 P/BV n.a. 1.9 1.7 1.4 Dividend yield 0.0% 4.2% 5.0% 5.6% FCF yield 6.0% 8.1% 11.0% 12.5% High/Low (12M) NOK31.7/27.8 Equity/Total Assets 33.6% 35.3% 39.4% 47.2% Perf. 3M 6M 12M YTD ROCE 8.5% 15.9% 17.4% 17.9% Abs. 14.0 14.0 14.0 14.0 ROE adj. 13.9% 21.3% 22.5% 21.1% Rel. 8.5 8.2 -9.7 8.1 Net IB debt/EBITDA 3.0 1.0 0.7 0.4 27.5 28.0 28.5 29.0 29.5 30.0 30.5 31.0 31.5 32.0 Mar 2018 Mar 2018 Apr 2018 Apr 2018 Apr 2018 Apr 2018 Share price -5Y Elkem OSEBX(No) (Rebased) Q1 Report 08 May 2018 Q2 Report 15 Aug 2018 Q3 Report 01 Nov 2018 No. shares (m) 581.3 Market cap. (USDm) 2,315 Market cap. (NOKm) 18,428 Net IB Debt. (NOKm) 4,098 Adjustments (NOKm) -65 EV (2018e) (NOKm) 22,460 Free float 0.0% Avg. daily vol. ('000) 671 Risk Medium Risk Fiscal year end December Share price as of (CET) 30 Apr 2018 16:26
Transcript
Please see disclosures on page 123 Carnegie Securities Research
This report was completed and disseminated 02 May 2018: 07:27 CET
c Norway
Materials
Initiating Coverage BUY (since 02 May 2018)
02 May 2018 Share price: NOK31.7 Target price: NOK37.0
We are initiating coverage of Elkem with a BUY recommendation and NOK37 target price. Our target price is based on Elkem’s closest peers, with a 15% discount.
From being a purely metallurgical company, Elkem, through acquisitions of silicon assets from its parent, has become a specialised chemical metallurgical company. Elkem is today a fully integrated global player from quartz to silicones. About 50% of revenue is derived from specialised products. Elkem enjoys a number 1–3 market position in all four of its business areas. In China, the largest and also the strongest growing market in silicones, Elkem is the largest player. In its upstream divisions, it has a leading cost position, as two of its Norwegian furnaces are the second and third most cost-effective facilities in the world.
We expect EBITDA to increase by 26% in 2018 on a combination of continuing growth in Silicones and cyclical higher prices in the other divisions. Silicon metal and ferrosilicon prices have increased through 2017 and 2018, and Elkem’s results will gradually reflect these higher prices. For 2019 and 2020 we expect EBITDA to grow by 6–8% per year.
Elkem’s long-term growth area is Silicones, where we expect a revenue CAGR(17–20) of 8.2% with EBITDA growing by around 13% in the same period to around NOK2.16bn. The combination of asset-light downstream capex and high grading the Chinese product portfolio could increase Silicones EBITDA by more than 60% on top of our 2020 estimates.
Elkem will announce Q1(18) results on 8 May. We expect Q1(18) EBITDA of NOK970m and EPS of NOK0.76. Compared to Wacker Silicones’ Q1(18) result and consensus estimates for Ferroglobe, we believe that there is upside potential to our estimates.
On our estimates and target price of NOK37, the shares trade at a P/E of 11.2x for 2018e and 9.2x for 2019e. The corresponding EV/EBITDA is 6.8x and 6.0x, respectively.
Changes in this report
From To Chg EPS adj. 2018e 3.3 EPS adj. 2019e 4.0 EPS adj. 2020e 4.5
Appendix 117 Management 117 Board of directors 118
Disclosures and disclaimers 123
c
Elkem
4 02 May 2018
Investment case Elkem is a leading manufacturer of silicones, silicon materials, foundry, and carbon products headquartered in Oslo, Norway. Over the past 114 years Elkem has been one of the cornerstones in Norway’s industrial revolution. Following the merger with Bluestar Silicones in 2015 and the acquisition of the two large Chinese assets in 2017, Elkem is today a fully integrated global player from quartz to silicones. The company has a leading cost position in many of its upstream production facilities driven by access to cheap Norwegian hydropower. With these operating at full capacity, Elkem is ideally positioned to pursue capital light expansion in the downstream silicones market, leveraging its number one position in China. We forecast 16.7% growth per year in EBITDA in 2017–19 driven by full integration of the silicones division and a continued cyclical upturn in the silicon materials division.
From being a pure metallurgical company Elkem has become a specialised chemical and metallurgical company. About 50% of revenue is derived from specialised products. We pencil in an improving EBITDA margin from 14.8% to 17.3% in 2017–20e due to increased utilisation and a higher degree of specialised volumes in the Chinese assets. Two-thirds of silicone volumes in China are commodity driven, which we forecast will decline to one-third as the share of specialised silicone volumes is increased.
We anticipate Elkem’s silicone business to grow revenue by a CAGR of 8.2% in the period 2017–20e. The market for silicones grows at double the GDP growth in western countries and has a huge potential in China. Currently China consumes only USD2.5 per capita of silicones compared to USD11.5 in the US. We expect the global silicone demand to grow 5.1% per year and Chinese silicone demand to grow 7.4%, meaning Elkem can capitalise on its strong foothold in the Chinese market to warrant growth above the market average.
We expect global demand for silicon material to increase by 5.6% per year in 2017–20e. With a relatively high operating rate among non-Chinese producers and China effectively shut out of the US and EU markets through trade regulations, we expect silicon prices to rise 6% on average in the coming three years.
Elkem has a leading cost position. Two of its Norwegian furnaces are the second and third most cost effective facilities in the world. Cost focus has been at the backbone of the organisation since 1999 driven by the Elkem Business System (EBS). This this mentality will be a defining factor in the successful integration of the new Chinese facilities.
In its smaller divisions, Foundry and Carbon, Elkem has a strong market position by focusing on niches with high willingness to pay and strong customer retention through continuous specialisation. Exposure to growing speciality steels and cast iron foundry make these divisions contribute to underlying growth for the group.
We value Elkem using comparable peer pricing. We view Wacker (silicones) and Ferroglobe (silicon-related) as Elkem’s closest peers. Taking average EV/EBITDA and P/E for those two companies into consideration, we end up at a value per share of NOK44. We apply a 15% discount to this to reflect that Elkem has no earnings history given its ongoing transformation. This gives a target price of NOK37. If we expand the peer universe and look at global chemical companies, silicon/steel-related and Nordic capital goods, the valuation upside is greater. However, we find it premature to use the overall sector valuation at the moment.
Specialisation will drive margin expansion
Double GDP growth in silicones demand
6% rise per year in silicon prices in 2017–20e
Leading cost position
Profitable niches in the smaller divisions
Target price NOK37
c
Elkem
02 May 2018 5
Executive summary In the following section we outline the key factors covered in this report. The section gives a brief overview of Elkem, market trends, estimates and valuation.
Overview of Elkem Elkem has four divisions: 1) Silicones, 2) Silicon materials, 3) Foundry Products; and 4) Carbon. Below we highlight the key products in each division.
• Elkem’s Silicones division produces silicones from its own siloxane production facilities in France and China. Silicone is the end product that is sold to customers. Silicones can be found in applications ranging from airbags and baking paper to wound dressing and batteries for electric vehicles.
• Silicon Materials produces silicon metal with varying levels of silicon content (Si). The level of silicon content determines what the silicon metal is used for. High purity silicon, known as polysilicon, is used in solar and semiconductor applications. Silicon is also used in the production of silicones by the chemical industry. Thirdly, silicon is used in aluminium production. A by-product of silicon metal production is microsilica which can be used in applications such as cementing.
• Foundry Products produces ferrosilicon, nodularisers (such as magnesium ferrosilicon) and inoculants (ferrosilicon mixed with other elements). Standard ferrosilicon is primarily used by the steel industry as a deoxidiser and speciality ferrosilicon is used in the production of for example stainless and electrical steel. Nodularisers and inoculants are typically used in foundry. Nodularisers are used in ‘nodularising’ cast iron and producing ductile iron. Inoculants are used in making grey iron and ductile iron castings. These are products that are added to the production process to give the end product a certain property. Microsilica is also a by-product of ferrosilicon production.
• The Carbon division produces electrodes to be used in furnaces when producing silicon metal and ferrosilicon, cathode blocks, ramming paste and recarburisers. Carbon electrodes are designed to transport the heat to the heart of a furnace. Applications of the carbon division’s products can be found in the chemicals industry, aluminium, steel and foundries.
Together the divisions make Elkem a global integrated player spanning the entire value chain. Its silicone division is the largest with 43% of group revenue. Elkem Silicon material makes up 27% and foundry products 18%. The smallest division is the carbon division with 7% of revenue.
In the diagram below we illustrate Elkem’s value chain. Quartz is the main ingredient in silicon production. Elkem has three quartz mines that can supply its silicon material and foundry divisions. Another key input in silicon and ferrosilicon production is carbon electrodes that transfer heat to the heart of the furnace. Elkem can source this from its own carbon division. To produce silicones Elkem produces siloxane, which is a process of mixing methyl chloride and silicon. Silicon is supplied from its silicon materials division and external suppliers. Through a
Vertically integrated production from quartz to silicones
c
Elkem
6 02 May 2018
chemical process the siloxane is used to produce silicones with different properties. At each step in this process Elkem can also produce various other products that are not consumed in its own production, such as foundry alloys, microsilica, and other carbon products.
Elkem’s value chain
Source: Carnegie Research, Elkem
No single division stands out as low margin or high margin. Rather the EBITDA margin between divisions will largely depend on the raw material pricing and where we are in the commodity cycle. Hence, Elkem has continuously worked to isolate itself from these commodity cycles through vertical integration and increased specialisation in its product mix.
The chart below showing Elkem’s EBIT margins since 1970 illustrates this. Until the mid-1990 Elkem was exposed to strong commodity cycles through its silicon, ferrosilicon and (at the time) manganese and ferrochrome segments. The cycles typically followed a 5-year pattern and left Elkem in a distressed situation at the beginning of the 1990s. After this Elkem started an impressive turnaround, where strong focus on higher-value products was a cornerstone. As a result of this strategy Elkem sold off its manganese assets to Eramet in 1992 and closed its ferrochrome production in 2002. EBIT margins from the mid-1990’s until the financial crisis in 2008 hovered around 10%. In this period Elkem also acquired Sapa (a global aluminum extruder) and traded its share in two Norwegian aluminum smelters with Alcoa’s soft alloys
15.1%
12.5%
16.6% 17.4%
14.8%
0%2%4%6%8%
10%12%14%16%18%20%
Silicones Silicon Materials Foundry Carbon Group
EBIT
DA
mar
gin
EBITDA margins across divisions, 2017
Source: Carnegie Research, Elkem
Continued specialisation opens up for margin expansion
Continuous improvement and specialisation has reduced cyclicality
considerably
c
Elkem
02 May 2018 7
(aluminum extrusion). After 2008 Elkem has sold off its power assets in Salten/Bremanger. In 2005 Elkem was acquired by Orkla, which later sold Elkem (ex-power assets and Sapa) to China National Bluestar. Since being acquired by China National Bluestar Elkem has continued on its specialisation strategy that started in the 1990’s. The merger with Bluestar Silicone in 2015 followed by the acquisition of two large Chinese assets in 2017 is a giant leap in this context.
Production takes place in 27 plants around the world (in addition to one under construction in Paraguay) with 6,100 employees in 40 countries. The group is the third largest silicones manufacturer in the world (measured by capacity) and the clear number one player in China. Elkem also holds the number three position globally in silicon material and number one in foundry products (excluding standard FeSi). The four different divisions operate in markets with fairly concentrated supply; for example, the five largest players hold 54% (foundry) of global supply and 71% (silicon materials) excluding China.
Elkem’s production facilities give it a global footprint
c
Elkem
8 02 May 2018
One of Elkem’s key strengths is the continuous cost focus in every part of the organisation. This is supported by the internally developed program called Elkem Business System (EBS), a lean manufacturing principle built on simple guidelines and rules. In short, EBS can be summarised in three core principles: 1) eliminate waste through value-added initiatives reducing cost; 2) empower people to promote decentralised decisions by those closest to the situation; 3) make to use to maximise yield. These guiding principles have been the backbone of Elkem’s operations since 1999. This, together with the benefit of being located in Norway with cheap electricity, significant economies of scale and captive raw materials sources puts the main silicon material plants very low on the cost curve relative to peers.
Elkem has seven R&D facilities with over 400 R&D workers (about 7% of total workforce). This enables it to target new applications with own developed products and collaborate with customers for tailor-made products. Elkem’s silicone division holds over 1,200 patents and each year files an additional 20 patents. Over 4% of turnover is dedicated to R&D. In 2016, 20% of silicones sales were generated by products less than five years old. As such, Elkem’s R&D efforts are important for both staying competitive and driving revenue growth. R&D initiatives lead the drive towards specialisation which supports the company’s overall strategy. Elkem strives for sustainability in every step of the process with the ultimate goal of becoming a carbon neutral metal producer, primarily by incorporating biocarbon and pyrolysis in the production process as well as continuous innovation.
Elkem serves end-markets that produce everyday products that can be seen all around us. No customer represents more than 2% of total sales. The graph below is illustrative in order to understand where the different silicon metals and alloys are used.
Source: Carnegie Research, Elkem
Overview of production facilities, across divisions
Low-cost producer through lean manufacturing, captive raw materials and
access to cheap hydro power
R&D initiatives drive growth with the aim to become a carbon neutral producer of
silicon metal and ferrosilicon
Elkem’s products can be found in applications all around the world
c
Elkem
02 May 2018 9
Elkem’s divisions supply a wide range of end-markets ranging from energy to consumer goods. Exposure to different end-user markets across its four segments is illustrated in the table below. Construction and transportation markets are the largest sources of demand. In this table it is important to understand the varying degrees of specialisation in the exposures. This is key to Elkem’s strategy as it tries to shift its exposure to more specialised products that have higher barriers to entry and fewer commodities driven price volatility. China will be a big driver in this sense. Currently China consumes only USD2.5 silicones per capita compared to USD11.5 in the US. As the country continues its economic expansion this will drive the demand for silicones and especially so for speciality silicones.
If we split the two main segments, Silicones and Silicon Material, into its end-user markets we get the following two graphs:
Elkem's segments and applications
Source: Carnegie Research, Elkem, Design: FerroglobeNote: The relative size of pie is only indicative
Sales by end-market(1)
End
us
e Construction Transportation Consumer goods Energy ElectronicsHealth / personal
careChemicals
Shar
e of
sale
s1
● Silicone
dh i
● Airbags ● Labels ● Solar panels ● LED ● Liquid Silicones ● Defoamers
Increasing the share of specialisation in end-markets
c
Elkem
10 02 May 2018
Construction is a large end-user segment where the application of silicones comes through adhesives, sealants, coatings and cables. Microsilica and related products are used to increase strength and durability in bridges, marine structures, high rise buildings, tunnels and other higher performance applications. A by-product of silicon metal production, microsilica, is often used as an additive in fibre cement.
The second largest segment is transportation where both silicone and silicon materials are used by the car manufacturers in airbags, automotive parts and cables. Typically silicon is used as an alloying agent to increase strength and reduce weight. New markets with great potential are emerging such as the electric cars that have a wide range of applications in the silicone operations. Current tests are being done on the positive impact by silicones in electrical batteries for cars. The third large end-market is consumer goods where silicone is used in labels, food grade papers, personal care products and kitchen ware. Silicon is an input in silicones production which can be found in lubricants, cosmetics and medical equipment.
Market in brief Silicones The global market for silicones is expected to grow by a CAGR of 5% towards 2021, reaching a total value of USD18.5bn. The Asia-Pacific region is the fastest growing market for silicones and we believe this warrants above-market growth for Elkem with its leading position in China and targeted downstream growth in the region.
Supply of silicones is relatively consolidated. In the 2016, there were five dominant players of silicones, which together accounted for 74% of the market. Dow Corning is the largest of the five, holding about 27%. Elkem holds 11% with its no. 1 position in China and top 3 position globally in terms of production capacity.
Energy5% Transportation
14%Chemicals
5%
Construction30%
Health & personal Care8%
Consumer Goods33%
Electronics5%
Silicones end-user markets
Source: Elkem (preliminary & unauditet incl. chinese assets), Carnegie Research
Energy17% Health & Personal
Care3%
Transportation17%
Construction32%
Consumer Goods16%
Electronics12%
Chemicals3%
Silicon Material end-user markets
Source: Elkem (prelimiary & unaudited incl. Chinese assets), Carnegie Research
14.4 15.2 15.9 16.7 17.6 18.5
-
5.0
10.0
15.0
20.0
2016 2017e 2018e 2019e 2020e 2021e
USD
bn
Global silicones market value development
Source: Freedonia Group, Carnegie Research
5.1% CAGR
80
100
120
140
160
180
200
06 11 16 21e
2006
= 1
00
Global Silicones market growth break down (indexed)
Global Volumes (tons) Global Average Price (USD/t) Total market growth (USDbn)
Source: The Freedonia Groupm, Carnegie Research
Volume CAGR(16-21e)
of 3.1%
Sales priceCAGR(16-21e)
of 1.9%
Market CAGR(16-21e)
of 5.1%
Note: 2006,11,16 are actuals, interim years are interpolated
Growth in electric cars could fuel a new market for Elkem
5% CAGR to 2021 in the global silicones maker
Consolidated supply in silicones
c
Elkem
02 May 2018 11
Silicon materials Aluminium is the largest demand driver for silicon metals, accounting 43% of the output. Silicones demand about 33% of the market, followed by solar/electronics at 24%. Historical growth has on average in the past 35–40 years been 6% annually, which we believe will hold up also towards 2021e.
Silicon metals are enjoying strong growth in China, where production has risen more than five-fold in the past 17 years (right chart below). Production in the world (ex. China) has been stable in the same period. Supply is consolidated, with the top 10 non-Chinese producers holding 96% of the capacity. In comparison, the top 10 Chinese producers have a 35% market share.
Raw material price increases and an appreciating Chinese currency relative to the USD should make it more difficult for Chinese producers to lift export of silicon metals to the global market. Additionally, trade restrictions effectively keep Chinese silicon metal out of both the US and EU, narrowing potential end-markets further. This leads us to believe that the supply and demand balance will be tighter. The two charts below outline price forecasts from CRU and Roskill, both made in October 2017. Since then, spot prices have risen substantially, probably driven by trade restrictions in the US and higher raw material (coal) prices.
Consolidated supply ex-China and very fragmented supply in China
Demand and supply balance to get tighter
c
Elkem
12 02 May 2018
Foundry products The ferrosilicon market has for decades been plagued with low industry operating rates. Figures from 2016 suggest an overall operating rate of 47% driven mainly by overcapacity from China. The industry is very fragmented with the 20 largest producers holding a global market share of just over one third. The global market for ferrosilicon amounts around 6.5-7.0 mill tons, and the growth since 2000 is caused by expansion of Chinese capacity. Demand for ferrosilicon is driven by the steel market, which consumes around 70% of global ferrosilicon production. The other two markets are iron casting (17%) and magnesium (13%).
There seems to be a consensus among market participants that the low capacity utilisation and strict regulatory clampdowns in China will lead to limited new production facilities in the region. There are two facilities in the pipeline (one belongs to Elkem). Together these will add 60ktpa new supply (i.e. only 1% of total supply).
Elkem’s production is made up of relatively little standard ferrosilicon and more speciality, nodularisers and inoculants. Since 2000 consumption of ferrosilicon has had CAGR 4%, but in recent years this has fallen due to a decline in carbon steel production. Stainless steel, however, has grown at CAGR 5.5% since 2000 with only a slowdown in 2007–09. For a producer of ferrosilicon it is more attractive to be exposed to these markets with speciality FeSi than the standard FeSi used in crude steel. Speciality steel is expected to grow by 3% per year and cast iron by 2% per year. This will be a general growth driver for Elkem’s foundry segment.
Carbon products There is no way to tell what the exact demand for carbon products is. From other producers’ statements, such as SGL Group, the message seems to have been that in the recent years cathodes, furnace linings and carbon electrode sales have done well. According to Elkem, the key customers in its Carbon division are ferroalloy, silicon and aluminium producers. As such, demand for carbon products will be driven by these industries, which we have covered above.
-
500
1,000
1,500
2,000
2,500
3,000
0
500
1,000
1,500
2,000
2,500
3,000
3,500
4,000
2008 2010 2012 2014 2016 2018e 2020e 2022e
EU
R/t
USD
/t
Silicon metal price forecasts
Europe spot USD/t - Roskill Europe spot (EUR/t) - CRU
FeSi production has been driven by Chinese expansion which has recently
fallen back
Production shifting more towards speciality products
Elkem stands for 43% of global production of carbon electrodes
c
Elkem
02 May 2018 13
Estimate summary Following Elkem’s acquisition of Xinghou (the leading silicone producer in China) we see an overall CAGR in EBITDA around 13% over the next 3 years – increasing from NOK3.15bn in 2017 to more than NOK4.5bn in 2020e. The major share of the EBITDA improvement is fuelled by continuing strong demand growth in silicones combined with a higher share of sales from speciality products in China. Elkem’s EBITDA margins in silicones are around 4%-points lower than Wacker Silicones, which we regard as the closest peer. Elkem is in our opinion five years behind Wacker with regard to specialization and as Elkem’s share of speciality products in China is only 33%, there is great upside to EBITDA in this division, which now constitute close to 50% of Elkem’s revenue and EBITDA.
205
476
129
143
43%
27%
30% 100%
0
100
200
300
400
500
Elkem Top 2-5 Other Total
Tonn
es p
er a
nnum
(100
0's)
Global electrode paste production ex. China, 2016 (1,2)
Source: Carnegie Research, Elkem1Excluding China as Chinese market is highly fragmented, includes mostly domestic focused producers
Earnings over the next years will also benefit from a cyclical upturn. Although Elkem has a substantial share of sales from specialized products in its other divisions, we see a CAGR EBITDA growth around 14%. Both benchmark market prices in Silicon Materials and Foundry Products have risen substantially through 2017 and continued this year. Elkem’s achieved prices are lagging these benchmark prices so the positive EBITDA effect will fully start to kick in in 2018.
Valuation We value Elkem using comparable peer pricing. We view Wacker (silicones) and Ferroglobe (silicon-related) as Elkem’s closest peers. Taking into consideration the average EV/EBITDA and P/E for those two companies, we end up with a value per share of NOK44. We apply a 15% discount to this to reflect that Elkem has no earnings history given its ongoing transformation. This gives a target price of NOK37. If we expand the peer universe to include global chemical companies, silicon/steel-related and Nordic capital goods, the valuation upside is greater. However, we find it premature to use the overall sector valuation at the moment.
The table below outlines the consensus estimates for Wacker and Ferroglobe.
Based on EV/EBITDA, P/E, EV/EBIT and FCF yield, the chart below gives an implied equity value for Elkem for 2018–20e. One of the challenges using multiples in valuation methods is what weight to use on each multiple and for which year(s).
Looking at EV/EBITDA, this translates into an equity value of NOK 34–36 per share. The corresponding value, based on P/E, gives NOK 54–56 per share. The average, based on 2018e earnings, is thus NOK44. We apply a 15% discount to this to reflect all uncertainties given that today’s Elkem is a new company in the stock market. This gives a target price of NOK37.
Finally, we compare Elkem’s multiples with its broader peer group. Broadly speaking, it screens favourably compared with the total peer group. There is scope for multiple expansion, but we find it premature to use sector multiples on our earnings estimates.
-
10
20
30
40
50
60
EV/EBITDA P/E EV/EBIT FCF Yield
Implied value NOK/share
2018e 2019e 2020e
Source: Carnegie Research
Peer group Chemical Silicon/steel Nordic Elkem Elkem tgt2018 comparison Global related Cap.goods Wacker Ferroglobe NOK31.6 NOK37
Macroeconomic outlook We/the consensus believe 2018 will be a year of above-normal GDP growth in most countries. The main restriction for further growth acceleration is related to utilisation rates in areas where a recovery started early. We therefore still argue that business investments will be the fastest-growing component of GDP. Over the course of the year, growth will gradually slow and inflation and interest rates will rise. The pace of rising inflation and interest rates will determine the risk of a sharper slowdown. However, our base case is that a solid performance is most likely in 2018.
Global perspectives GDP Growth at constant prices
Source: Thomson Reuters Datastream, Carnegie Research
The global macroeconomic environment has not looked as solid as it does today since the financial crisis in 2008–09 (perhaps with the exception of the start of the recovery in 2010, although that was a false positive, as the Eurozone, in particular, had not dealt with its more vulnerable members’ public finances, nor with banks’ capitalisation and non-performing loans). Today, even though not fully resolved, the situation looks much better, with lower unemployment rates, lower financial risk premiums (e.g. default spreads) and more solvent banks, etc.
c
Elkem
02 May 2018 17
There is still an ongoing increase in employment in the major economies, which continues to support private consumption. However, with unemployment at low levels in the US and some other countries that entered the recovery early, global employment growth should slow. US consumption will be receiving an extra boost by tax cuts at a time when consumption was otherwise set to slow visibly. In the Eurozone, consumers were cautious, until the last three years, and there is still pent-up demand that should sustain retail demand. However, overall, we see limited upside potential for private consumption growth, especially as interest rates will trend upwards over time.
Advanced economy employment and Major countries growth in personal consumption
Source: Thomson Reuters Datastream, Carnegie Research
If a low unemployment rate restricts employment and, therefore, consumption growth, it is having a positive effect on investments. Firstly, investments are driven by corporate profitability, which has been restored and is still growing. Secondly, bottlenecks created by strong demand lead to falling productivity, which will give rise to efficiency investments. Thirdly, as the last investment cycle ended in 2008, new equipment may be needed both for replacement and expansion. Lastly, as the ongoing process of digitalisation as a means of changing business model is reaching new levels – with Industry 4.0, Big Data, AI, etc. – most companies need to consider how and what to do. This should mean more investments.
0.0
1.0
2.0
3.0
4.0
5.0
6.0
Jan-
16
Feb-
16
Mar
-16
Apr
-16
May
-16
Jun-
16
Jul-1
6
Aug
-16
Sep-
16
Oct
-16
Nov
-16
Dec
-16
Jan-
17
Feb-
17
Mar
-17
Apr
-17
May
-17
Jun-
17
Jul-1
7
Aug
-17
Sep-
17
Oct
-17
Nov
-17
Dec
-17
Jan-
18
Feb-
18
Consensus GDP forecast development
US 2018 US 2017 Eurozone 2018
Eurozone 2017 Asia/Pacific 2018 Asia Pacific 2017
Source: Carnegie Research, Consensus Inc.
The low unemployment rate is gradually causing reduced consumption
growth
There are several reasons for business investments to grow faster than the
overall economy
c
Elkem
18 02 May 2018
US non-residential investment and profit and major countries’ growth in gross fixed capital formation
Source: Thomson Reuters Datastream, Carnegie Research
Hence, we still argue that investments will continue to outpace GDP growth for at least the next year. Another consequence of increased utilisation and difficulties in finding skilled labour is that inflationary pressure increases. Admittedly, the US economy saw even lower CPI inflation in 2017 than in 2016, but we still argue that the inflationary trend is upwards and that the consensus inflation estimates have an upwards risk bias.
Global inflationary trends
Source: Thomson Reuters Datastream, Carnegie Research
Despite being more or less invisible, we think the upside potential to
inflation is increasing
c
Elkem
02 May 2018 19
Interest rates and FX The interest rate cycle has clearly turned upwards, as the Federal Reserve is now at five rate hikes since it started in December 2015, of which three took place during 2017. In addition, of the other major central banks, the Bank of Canada, the Bank of England and the Bank of Korea have all raised their rates at least once in the past year.
Central bank policy rates and 10-year government bond yields
Source: Thomson Reuters Datastream, Carnegie Research
Since unemployment is low and material prices are already adding to factory gate prices, inflation seems to be increasing gradually. There are several reasons to expect higher US bond rates, however, of which inflation is but one. To begin with, the Fed has started a slow winding down of its balance sheet, which suggests the supply of bonds (or net borrowing interest) will gradually rise over the next year. In addition, the recent tax reform is underfinanced and, as it seems unlikely that dynamic effects will neutralise lower government income, there will be more borrowing needs from the US government (albeit with a limited or reversed impact in 2018 depending, for example, on when repatriated profit is moved into the US).
Outside the US, the ECB and the Bank of Japan are unlikely to raise rates in 2018, unless there is a significant increase in inflation, although some tightening will occur gradually as Quantitative Easing is reduced. However, considering the current strength of the global economies, it would be surprising if the language used by President Draghi and Governor Kuboda were not to signal upcoming policy changes in H1(18). Still, the main conclusion is that we expect gradually higher rates, but no dramatic moves over the coming six, probably 12, months.
Monetary policy changes are more likely to occur during 2018 for the Nordics than the Eurozone, to which Finland belongs and Denmark has pegged its currency. In other words, both Sweden and Norway are more likely to see interest rates raised; Sweden, with its negative deposit rate, should be expected to see its central bank raise rates at least once during 2018. Current consensus expectations are for one hike in Sweden during Q3(18). The view about a move by Norge’s Bank during 2018 is split – and we argue the probability of a hike is high.
FX is notoriously difficult to forecast and is not an area in which we purport to have expertise. However, we believe our/the consensus forecast for continued global growth speaks in favour of a further weakening of the USD. The USD has weakened in five of the six most recent rate hike cycles; the one when it strengthened was coupled with a large federal deficit, following tax cuts and military spending during the Reagan presidency. The recently passed tax reform will add to the government budget deficit and debt, but to a limited extent in 2018. Hence, it is difficult to draw comparisons with US bond markets in the mid-1980s, at least for 2018, so we lean more towards a weakening USD, as Europe, Asia and emerging markets suck up a larger share of global investments and liquidity over the next year.
The monetary tightening cycle has started to gain traction
Inflation is not the only reason bond rates should be rising
The ECB and BoJ are unlikely to move rates but will still change their language
during 2018
Nordic rates are caught between strong global growth, causing tight
domestic economies and households with high debt, and falling house prices
It seems reasonable to believe in a further weakening of the USD, even though our conviction rate is limited
c
Elkem
20 02 May 2018
USD during tightening cycles
Source: Thomson Reuters Datastream, Carnegie Research
Finland is a Eurozone country and Denmark has its DKK pegged to the EUR, so the loose ends in the Nordics are Sweden and Norway. Both the SEK and the NOK have been very weak over the past few years, more than should be expected, in our view, considering that both are cyclical currencies with drivers pointing to FX strength. Historically, the SEK has been correlated with large moves in the PMI, but not in the past year when the it has weakened instead. In Sweden, this weakness was initially related to the country’s very lax monetary policy; more recently, concerns about the Swedish housing sector have taken their toll. Norway clearly saw an effect from lower oil prices in its currency value until 2015, but it started to recover in 2016. In 2017, it also started to see lower house prices hurting confidence and, perhaps more importantly, inflation was lower than anticipated for most of the year. Expectations of rate increases were postponed until a few weeks ago, when the CPI figures rose above expectations.
Sweden SEK vs. PMI and Brent oil price and the NOK
Source: Thomson Reuters Datastream, Carnegie Research
c
Elkem
02 May 2018 21
Considering their starting points, we would expect the SEK and the NOK to strengthen during 2018. As long as the global economy is doing well, the Swedish housing situation should stabilise in H1(18) and ease some concerns. The NOK has already started to track the oil price again, but it is currently undervalued and, unless oil prices suddenly reverse, the NOK should strengthen further.
Main risks to our assumptions In our opinion, by far the most important risk relates to the US economy, since bull markets have historically been ended by tighter monetary policy, as GDP moves clearly above trend and causes employment to slow, inflation to rise and higher interest rates. The usual warning sign is the yield curve turning negative, but this is typically more of a preannouncement about an upcoming recession. Stocks have peaked a few times at the same time the yield curve became negative, such as in early 2000 – and the recession did not begin until 2001. For the yield curve to turn negative, inflation would have to rise further, or growth would, for some reason, have to falter and bond rates drop, which could come more abruptly and would be a bear flattening scenario. We find both these scenarios unlikely in H1(18) and would not position ourselves to allow for them yet.
The recently presented US tolls on washing machines and solar panels, stronger comments about the NAFTA treaty, and renewed concerns about intellectual property theft in comments from the Trump administration could, at some point, lead to reactions in China or elsewhere. The risk of trade war still seems unlikely, but escalating excise tolls or taxes may eventually hurt global economies.
There are definitely risks remaining in China as well, not the least as the US trade message is largely directed there. We would look more at China’s focus on ‘quality of growth’ rather than actual growth. The ‘blue skies’ policy has led to a reduction of production in heavy and air-polluting industries over the winter, and this reduction may become a negative spiral even if the winter season will soon be over. The aim to bring down speculation, including housing, may also lead to some overambitious reduction in lending that could stall construction activity, which has been a strong contributor to the global recovery since 2016. This also seems less likely, as there is still need for more homes, albeit affordable ones, and there are limited unsold inventories that we can detect.
The SEK and NOK are undervalued and are more likely to appreciate than depreciate, unless the global economy
surprises on the downside
c
Elkem
22 02 May 2018
Market outlook In this section we give a detailed account of the key market trends to which Elkem is exposed. A tighter supply and demand balance in the silicon metal markets should support prices in Elkem’s silicon materials division. Elkem’s silicones division will benefit from strong growth in silicones in China. The ferrosilicon market is plagued with low capacity utilisation and declining use of silicon in crude steel production. Elkem’s exposure to more specialised niches such as cast iron foundry provides a positive backdrop for continued growth in the foundry products division. Demand for carbon products is linked to the silicon and ferrosilicon segments but also to aluminium and other carbon-consuming foundry sectors. With the addition of the Chinese assets to Elkem’s platform, the company is in an ideal position to increase earnings on the back of strong growth from the region. Although the markets are isolated there are many common trends and feed-through mechanisms that contribute to overall growth. With the addition of the Chinese assets to Elkem’s platform the company is in an ideal position to increase earnings on the back of strong growth from the region.
Silicones The Freedonia Group, a specialised industry research group, expects the global market for silicones to grow at CAGR 5% towards 2021, reaching a total value of USD18.3bn. GDP growth and more favourable economic conditions are the general drivers, while market share capture from substitutes and increased market footprint in different industries are potential growth accelerators. Volume growth from higher consumption contributes most of the growth, 3.1%, while a normalised price growth per year adds another 1.9% to the overall growth rate. The Asia-Pacific region will be the fastest expanding market for silicones due to continued rising incomes and higher living standards, driving demand for higher quality silicone-based consumer goods over rubber-based substitutes.
The electrical and electronics, health and personal care, and automobile end-markets are expected by the Freedonia Goup to drive most of the growth. Silicone is a vital factor in the creation of fluids and gels used in a range of end-products in these target industries, such as solar panels, LEDs, elastomers in wind turbines, skin adhesives and car components, as well as in industrial lubricants and pharmaceutical products.
80
100
120
140
160
180
200
06 11 16 21e
2006
= 1
00
Global Silicones market growth break down (indexed)
Global Volumes (tons) Global Average Price (USD/t) Total market growth (USDbn)
Source: The Freedonia Groupm, Carnegie Research
Volume CAGR(16-21e)
of 3.1%
Sales priceCAGR(16-21e)
of 1.9%
Market CAGR(16-21e)
of 5.1%
Note: 2006,11,16 are actuals, interim years are interpolated
c
Elkem
02 May 2018 23
Asia-Pacific the largest end-market The Asia-Pacific region is the largest end-market for silicones demand, accounting for 47% of the global market in 2021e – up from 44% in 2016. The regional CAGR(16–21) is estimated by The Freedonia Group to amount to 6.2%. India and China are largest growth contributors, being the two largest countries and both with an increasing middle class demanding higher standards of living and higher quality products, and both with a large manufacturing industry.
The US and Western European markets have historically been the largest end-markets and the most intensive markets in terms of silicone consumption per capita. This is due to the wide adoption of silicones in multiple industries, comparing to a lower widespread adoption in e.g. the Asia-Pacific region. The US is expected to grow by a 4.4% CAGR(16–21), driven by increases in construction, electrical and electronic products, health and personal care, and transportation industries. The Western European region declined in demand between 2011 and 2016, most likely as a consequence of slower construction markets following the weaker economic situation, but is expected to revert to growth again.
14.4 15.2 15.9 16.7 17.6 18.5
-
5.0
10.0
15.0
20.0
2016 2017e 2018e 2019e 2020e 2021e
USD
bn
Global silicones market value development
Source: Freedonia Group, Carnegie Research
5.1% CAGR
9,770
12,480
14,245
18,250
-
2,000
4,000
6,000
8,000
10,000
12,000
14,000
16,000
18,000
20,000
06 07e 08e 09e 10e 11 12e 13e 14e 15e 16 17e 18e 19e 20e 21e
USD
m
Global silicones demand by region (USDm)
North America Central & South America Europe Asia/Pacific Africa Global aggregate
Source: The Freedonia group, Carnegie Research
c
Elkem
24 02 May 2018
The remaining regions will have limited impact on global demand due to their relatively low size, despite growing at a similar pace as the global average. Compared to global standards, the intensity of use is also expected to remain low.
Prices to grow at 2% Prices for silicones vary by region, but are often denominated in USD. The highest prices are found in North America, Western Europe and Japan, while developing regions generally have much lower prices. There is also a quality difference between the regions, with the higher-income countries often tending to use higher quality and more sophisticated silicone products, while industrialising countries use more basic silicone fluids and elastomers.
The transportation, electrical and electronic products end-markets have the highest average sales prices for silicones due to their need for outstanding quality and performance. At the other end of the range are the construction and chemical markets, which demand less complex silicones products – hence they also have the lowest average sales prices among the major industries.
It is fair to assume that the decline in market prices in 2011–16 is at least partly due to USD appreciation versus other currencies, depressing average prices and lowering demand in value terms. The actual demand in volumes did not decline.
Global silicones demand by region (USDm)
Region 2006 2011 2016 2021e (06-11) (11-16) (16-21e)
USA 2,750 2,757 3,534 4,385 0.1% 5.1% 4.4%
Canada & Mexico 364 426 449 545 3.2% 1.1% 4.0%
Central & South America 307 466 328 415 8.7% (6.8%) 4.8%
Western Europe 2,728 2,946 2,898 3,435 1.5% (0.3%) 3.5%
Eastern Europe 301 405 389 490 6.1% (0.8%) 4.7%
China 960 2,360 3,375 4,815 19.7% 7.4% 7.4%
Japan 1,026 1,133 1,090 1,215 2.0% (0.8%) 2.2%
Other APAC 1,090 1,644 1,840 2,500 8.6% 2.3% 6.3%
Africa/MENA 244 343 342 450 7.0% (0.1%) 5.6%
Global aggregate 9,770 12,480 14,245 18,250 5.0% 2.7% 5.1%
Source: The Freedonia Group, Carnegie Research
CAGR
North America27%
Central & South America
2%Europe
22%
Asia/Pacific47%
Africa2%
Global silicones demand by region*, 2021e
*Measured in value (USDbn). Source: The Freedonia Group, Carnegie Research
Silicone prices by region (USD/kg)
2006 2011 2016 2021e (06-11) (11-16) (16-21e)
Global average price 6.10 6.65 6.45 7.10 1.7% (0.6%) 1.9%
By region
North America 7.65 8.90 9.20 10.35 3.1% 0.7% 2.4%
Western Europe 7.00 8.25 8.30 9.35 3.3% 0.1% 2.4%
Asia/Pacific 4.90 5.45 5.20 5.75 2.2% (0.9%) 2.0%
Other regions 4.75 5.55 4.80 5.40 3.2% (2.9%) 2.4%
Health & Personal Care 5.35 6.15 6.50 7.25 2.8% 1.1% 2.2%
Chemical 5.60 6.10 5.80 6.35 1.7% (1.0%) 1.8%
Other regions 5.40 5.90 5.65 6.15 1.8% (0.9%) 1.7%
Source: The Freedonia Group, Carnegie Research
CAGR
c
Elkem
02 May 2018 25
As illustrated above and below, the price in North America can be almost double of that in the Asia-Pacific and other regions. The price variations are not as large between industries, but we note an average 50% premium between electrical end-markets and construction end-markets.
We expect prices to increase about 2% per year. We expect price increases to be biggest in the industries and regions where prices also are lowest.
Industrial breakdown The construction end-market is the largest silicon market, accounting for about 20% of annual demand and also expected by the Freedonia Group to grow the fastest. Construction, transportation and electrical markets, the top three industries, amount about 55% of the total market, amounting to USD10.3bn.
Despite the Asia-Pacific region being the largest market overall, value demand in the US, Canada and Western Europe is more or less equal due to pricing and quality differences. This is predominantly the case in the transportation equipment and personal care markets, where the difference is driven by three major factors:
• Higher personal incomes boosting demand for more advanced, more durable vehicle components and quieter interiors, stimulating silicone use over cheaper substitutes.
• The size of vehicles varying dramatically between the US, Canada and the Asia-Pacific region, with the former being much larger and demanding much more silicone per unit.
• In the personal healthcare industry, higher incomes leading to more use of silicone-based medical consumables and prosthetics, rather than cheaper plastic substitutes.
Global Average Construction Electrical & ElectronicTransportation Health & Personal Care ChemicalOther regions
Source: Freedonia Group, Carnegie Research
Construction20%
Electrical & Electronic
18%
Transportation17%
Health & Personal Care12%
Chemical9%
Other24%
Global silicones demand market distribution, 2021e
Source: The Freedonia Group, Carnegie Research
c
Elkem
26 02 May 2018
In the other industries, especially construction, chemicals, and electrical and electronic products, Asia-Pacific is dominant due to high industrial activity. So although intensity of silicone use in volume terms relative to construction spending is similar between the US and Canada and Asia-Pacific, the industry in Asia-Pacific is about 5x larger, implying significantly higher demand.
Construction markets The construction end-market is estimated by Freedonia Group to grow by a CAGR of 5.6% towards 2021, reaching a total value of USD3.7m. With an expected average sales price of USD6.25 per kg, this implies a total market of nearly 600,000 metric tonnes, up from 510,000 metric tonnes in 2016.
A rebound in activity in Europe and Central and South America should be the primary driver behind this growth, followed by continued strong increases in construction spending in North America and the Asia-Pacific region. The latter has been fastest growing region in the past decade, driven by a rising middle class.
The Asia-Pacific region makes up slightly more than half of the total market for silicones used in construction applications. This market distribution is not expected to change significantly.
The construction markets are cyclical by nature, which naturally brought a declining demand the years after the financial crisis around 2008. After normalisation, demand has in most regions been growing by around 4%, nearly double the rate of GDP.
Demand for silicones in adhesives and sealants makes up most of construction industry demand. However, high gloss and water repellent architectural coatings – the dominant applications in other segments – should grow slightly faster, fuelled by continued high construction spending in the US and a rebound of construction activities in Western Europe.
Silicone adhesives, sealants and coatings are used in a number of construction applications, with many proven advantages:
• thermal stability over a wide temperature range;
• weather resistance;
• moisture resistance in interior and exterior applications;
• bonding to wide variety of materials;
• long lifespan.
North America21%
Western Europe18%Asia/Pacific
53%
Other regions8%
Construction market silicones demand distribution, 2021e
Global construction markets silicons demand volatility
North America Western Europe APAC RoW World Average
Source: The Freedonia Group, Carnegie Research
c
Elkem
28 02 May 2018
Electrical and electronic markets Demand for silicones from electronic markets is estimated to grow at a CAGR of 4.7% towards 2021, reaching a total market value of USD3.3bn. At an average cost CAGR of 1.6% in the period, reaching USD8.4/kg, this implies a mere 400,000 metric tonne-market by the end of the period. This is up from 341,000 metric tonnes in 2016.
Being the largest manufacturer of electrical and electronic products, the Asia-Pacific region is projected by the Freedonia Group to grow the fastest (5.4% per year), maintaining its position. The slowest growing region is Western Europe, which Freedonia Group expects to grow slightly faster than GDP (3%). Europe is coming out of a stagnating period, but is expected to recover in line with economic growth.
Global electrical and electronic silicones markets demand (USDm)
Electronic products are by far the largest outlet for silicones compared to other segments in the industry. Silicones have superior abilities in this market due to clarity, heat and light resistance, and good weather resistance. LED manufacturing applications are one such area that values the use of silicones greatly, and which is experiencing strong growth.
On the other side, a move towards renewable energy sources and the increasing use of solar panels and wind turbines is another positive accelerator for silicones demand. These applications often make use of silicone adhesives, sealants and lubricants.
Transportation equipment markets Demand for silicones in the transportation equipment markets is estimated by the Freedonia Group to grow at CAGR 5.4% towards 2021, reaching a total market value of USD3.2bn. At cost CAGR of 1.6% in the period, reaching USD9.25/kg, it indicates a market size of 343,000 metric tonnes by the end of the period. This is up from 287,000 metric tonnes in 2016.
Besides North America, all regions are forecast by the Freedonia Group to accelerate demand in the coming five-year period as greater economic stability lifts vehicle purchases and production levels. Vehicle sizes and interior quality should also increase in the developing countries, further lifting demand for silicon intensity of use.
Cars and other mother vehicles account for 80% of the total silicone consumption for transportation purposes, although aerospace and marine applications are showing somewhat higher growth. Silicone has gradually replaced organic rubber due to its compression strength, durability and high temperature range.
In contrast to the other top-three end-markets, the Asia-Pacific region is not yet as dominant when it comes to demand for silicones used in vehicle manufacturing and other transportation equipment applications. This is gradually changing as vehicle production and consumption in the North American region are slowing and moving to the Asia-Pacific area.
Health and personal care markets Demand for silicone used in the health and personal care industries is projected to grow by a 5% CAGR towards 2021. This is a combination of a 2.2% expected price growth and 2.7% volume growth over the period. The market is expected to reach a total size of 294,000 metric tonnes, with a total value of USD2.1bn.
Asia-Pacific is the strongest growth region, outgrowing Western Europe in size and becoming the second largest region. North America is expected to grow at a CAGR of 4.7%, maintaining its position as the largest region for health and personal care.
North America accounts for just over 40% of the global demand for silicones for this use, while Western Europe and Asia-Pacific each account for about 25%. The market is predominantly made up of personal care products, but more and more medical products and other segments are gradually converting to silicones, lifting this segment’s percentage of the total.
The growth is driven by increased shipments of personal care products worldwide. This market has matured in many Western countries, and the industry has recently found few new opportunities for silicone-based products. Healthcare, however, is on the rise, as the average population is getting older. Developing countries are also growing steadily, and are expected to increase spending in both healthcare and personal care following personal income growth.
Healthcare innovations are pushing through with silicone medical products, among them custom prosthetics, which has been made more readily available through technologies such as 3D printing.
Chemical markets Demand for silicone in the chemical industries is projected to grow in line with the total market, hence a 5.1% CAGR towards 2021. Most of this growth is expected to come from volume demand increases, which will contribute 3.1% growth, whereas prices are estimated to increase by 1.8%. The total market will reach 250,000 metric tonnes, equivalent to about USD1.6bn.
Global health and personal care silicones markets demand (USDm)
2006 2011 2016 2021e (06-11) (11-16) (16-21e)
Health & Personal care demand 1,155 1,550 1,670 2,130 6.1% 1.5% 5.0%
By region
North America 463 577 710 895 4.5% 4.2% 4.7%
Western Europe 355 433 424 521 4.1% (0.4%) 4.2%
Asia/Pacific 223 356 405 555 9.8% 2.6% 6.5%
Other regions 114 184 131 159 10.0% (6.6%) 4.0%
By segment
Personal care products 888 1,119 1,164 1,450 4.7% 0.8% 4.5%
Medical products and other segments 267 431 506 680 10.1% 3.3% 6.1%
Health and care ASP (USD/kg) 5.35 6.15 6.50 7.25 2.8% 1.1% 2.2%
Health and care demand (000 ton) 216 253 258 294 3.2% 0.4% 2.7%
Source: The Freedonia Group, Carnegie Research
CAGR
North America42%
Western Europe24%
Asia/Pacific26%
Other regions8%
Health & personal care market silicones demand distribution, 2021e
The Asian-Pacific region will continue to outgrow other regions, stimulated by ongoing growth in the output of products such as crude petroleum, foam insulation, plastics and pharmaceuticals. Western Europe is expected to recover from the decline of recent years, and to grow by a moderate 2.8% CAGR.
Silicones in plastics manufacturing is estimated to advance at a robust growth rate of 5.6%, as plastics continue to replace materials such as glass, metals and wood. Plastics have for a long time made up 58% of the total chemicals market, but will reach 60% by 2021, according to Freedonia Group’s estimates. In the oil & gas industry, demand is forecast to grow at a continued steady-state 4%, reaching a total value of USD275m by the end of the period.
Global chemical silicones markets demand (USDm)
2006 2011 2016 2021e (06-11) (11-16) (16-21e)
Chemical silicones demand 845 1,015 1,240 1,590 3.7% 4.1% 5.1%
By region
North America 328 315 390 485 (0.8%) 4.4% 4.5%
Western Europe 212 223 220 252 1.0% (0.3%) 2.8%
Asia/Pacific 221 369 520 720 10.8% 7.1% 6.7%
Other regions 84 108 110 133 5.2% 0.4% 3.9%
By segment
Plastics 486 586 724 950 3.8% 4.3% 5.6%
Oil & Gas 162 190 227 275 3.2% 3.6% 3.9%
Other segments 197 239 289 365 3.9% 3.9% 4.8%
Chemical ASP (USD/kg) 5.60 6.10 5.80 6.35 1.7% (1.0%) 1.8%
Chemical market silicones market demand distribution, 2021e
Source: Carnegie Research
c
Elkem
02 May 2018 33
Miscellaneous markets Demand for silicone used in the chemical industries is projected to grow at a 4.8% CAGR towards 2021, reaching a total value of USD4.3bn. This segment comprises 24% of the market, and measured 600,000 metric tonnes in 2016. Volumes are forecast to grow by 3% per year towards 2021, while average sales prices are expected to grow 1.7%.
The miscellaneous market comprises a wide range of industrial and consumer uses, from machinery, textiles and paper applications to vehicle and furniture polishes, toys and houseware uses. The Asia-Pacific region is the largest market, and also the fastest growing end-market, given the rising living standards and the region’s leading position as manufacturer of machinery, textile, paper and toy products.
Automation is the largest driver of machinery developments, which again fuels consumption of silicones in components such as gaskets, seals, lubricants and tubing. Silicones perform better than natural rubber in harsh environments, where speed and durability are important performance measures.
In textile processing, silicones are used in synthetic fabric or an additive that improves natural materials (cotton or leather). Elkem’s Bluesil is such a product, using silicone as a fluid water repellent spray to protect leather and other textiles. The textile market is expected to grow by 4.1%.
Paper and pulp processing is a smaller market and expected to grow more slowly than the general market due to continued digitalisation. Growth is coming mainly from paperboard and cardboard boxes, driven by growth in e-commerce and more package shipping.
Products Silicone is generally divided into four major groups: elastomers, fluids, resins and gels, and other products. Elastomers make up more about 56% of the end-uses, followed by fluid forms accounting for 31%.
Elastomers are expected to maintain a growth rate faster than the aggregate market, as they are widely used in multiple industries and regions. Elastomers account for more than two-thirds of the silicones demand in Asia-Pacific, but slightly less than half the demand from developed regions. Fluids, on the other hand, are seldom used in construction industries, but mostly in industrial processing in personal care, chemical processing and transportation equipment lubricants, and therefore expected to grow slightly below the general market. Most of the demand for fluids comes from Asia-Pacific regions.
Global miscellaneous silicones markets, by product category
Total market 9,770 12,480 14,245 18,250 5.0% 2.7% 5.1%
Source: The Freedonia Group, Carnegie Research
CAGR
Elastomers56%
Fluids31%
Resins8%
Gels & Other5%
Global silicones demand by product category*, 2021e
*Measured by value (USDbn). Source: The Freedonia Group, Carnegie Research
c
Elkem
02 May 2018 35
Elastomers Elastomers are defined as a polymer with viscoelasticity – having both viscosity and elasticity – and are primarily used for seals, adhesives, and moulded flexible parts. Application areas cover segments as diverse as tyres, shoe soles, and damping and insulating elements. The total market for elastomers is expected by Ceranasa Research Group to reach USD56bn in 2020, while Freedonia Group estimates that silicone elastomers will make up about USD10.3bn (20%) of this. This implies a 5.3% growth CAGR, driven by 3.3% volume growth and 1.9% price growth.
Transportation and construction industries are expected to be the dominant growth drivers of silicone elastomer demand due to silicone-based components having superior water and heat resistance properties of other rubber substitutes.
In general, the better performance and environmental advantages of silicone elastomers over other plastic substitutes will force a higher demand for silicones. We understand the main industries that are transitioning to silicone-based materials include:
• Industrial components (e.g. gaskets, seals, hose, and tubing);
• Medical consumables (e.g. catheters, respiration masks, prosthetics, and valves);
• Electrical and electronic products (e.g. sensors, monitors, keypads, and mobile phones);
• Mother vehicle components (e.g. air conditioning gaskets, radiator gaskets, and spark plug boots).
Fluids Silicone fluids encompass both basic siloxane fluids and more specialised products such as functionalised fluids, fluorosilicone fluids, lubricants and emulsions. These products have high commercial interest because of their thermal stability and their lubricating abilities. Demand growth is expected at 4.7% towards 2021, driven by 1.9% price growth and 2.7% volume growth. The total market is estimated to reach USD5.6bn.
The market for fluids is expected to grow from an increase in cosmetics and toiletry shipments, and from growing electronics manufacturing activity in Asia-Pacific. In developing countries where income has been low, market penetration is expected to grow in line with rising incomes and economic activity. Chemical companies are expected to buy more silicone fluids for the production of products such as mould release agents, water repellents, and urethane foam surfactants, while transportation equipment markets, where products such as vehicle polishes and lubricants are key, are expected to grow particularly well in developing countries.
5,118
6,768
7,970
10,300
-
2,000
4,000
6,000
8,000
10,000
12,000
2006 2011 2016 2021e
USD
bn
Global silicones elastomer market by region
N. America W. Europe APAC RoW Total
Source: The Freedonia Group, Carnegie Research
4.8% CAGR (06-21e)5.2% CAGR(16-21e)
1,342 1,748 2,170 2,845
1,234 1,563 1,680 2,060
978 1,255 1,525 1,995
1,564 2,202 2,595 3,400
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2006 2011 2016 2021e
USD
bn
Global silicones elastomer market by region
Construction Electrical & Electronics Transportation Other markets
Source: The Freedonia Group, Carnegie Research
c
Elkem
36 02 May 2018
Resins In polymer chemistry resins are referred to as a solid or highly viscous substance, plant or synthetic that is typically convertible into polymers. A typical user application is as coatings, e.g. on marine equipment, to inhibit growth of algae and other damageable flora and fauna. Resins are also frequently used as coatings in motor vehicles, and as an architectural coating. Silicone resins are particularly valued for their high shine and moisture resistance.
Demand is expected to grow by 5.5%, faster than the aggregate market, reaching a total value of USD1,395m by 2021. Volume demand amounts 3.7% of this, while prices on average are expected to grow 1.8%. The fastest end-market is electrical and electronics, followed by transportation. As a consequence of this, Asia-Pacific is also expected to be the fastest growing region for demand.
Gels and other The silicone gel market, coupled with the miscellaneous smaller end-uses not defined, is expected to grow by a CAGR of 4.8% towards 2021. This segment records the highest expected price increase in the industry of 2.1%, while the lowest expected volume growth of only 2.7%. The combined market is estimated to reach USD940m by the end of the period.
Gains in gel demand are closely related to the electrical and electronics product market, which accounts more than 60% of the general market. Increased production of LEDs and solar panels is a clear growth driver, overtaking some of the lower demand for traditional electronics applications where miniaturisation trends continue to reduce utilisation of gels. However, use of protective pastes in electronic appliances is rising, which is positive for gels.
In the other two large end-user markets – transportation and healthcare – silicone foams are expected to grow well, driven by products such as noise dampening products (in motor vehicles and other open-motor appliances) and wound dressings/bandages.
3,305 4,052
4,465
5,615
-
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
2006 2011 2016 2021e
USD
bn
Global silicones fluids market by region
N. America W. Europe APAC RoW Total
Source: The Freedonia Group, Carnegie Research
3.6% CAGR (06-21e)4.7% CAGR(16-21e)
666 852 823 1,015
550 645 770 985
306 368 393 515
375 451 565 725
1,408 1,736 1,914 2,375
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2006 2011 2016 2021e
USD
bn
Global silicones fluids market by region
Health & Personal Care Chemicals Electrical & Electronics Transportation Other markets
Source: The Freedonia Group, Carnegie Research
832 962
1,065
1,395
-
200
400
600
800
1,000
1,200
1,400
1,600
1,800
2,000
2006 2011 2016 2021e
USD
bn
Global silicones resins market by region
N. America W. Europe APAC RoW Total
Source: The Freedonia Group, Carnegie Research
3.5% CAGR (06-21e)5.5% CAGR(16-21e)
278 295 308 410
155 187 235 315
92 109 120 165
307 371 402 505
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2006 2011 2016 2021e
USD
bn
Global silicones resins market by region
Construction Transportation Electrical & Electronics Other markets
Source: The Freedonia Group, Carnegie Research
c
Elkem
02 May 2018 37
Silicon materials Silicon metal and ferrosilicon have many similarities. The production process is basically the same, with differences in amounts of raw materials and the energy need. The greatest difference between the two is demand.
515
698 745
940
-
200
400
600
800
1,000
1,200
2006 2011 2016 2021e
USD
bn
Global silicones gel & others market by region
N. America W. Europe APAC RoW Total
Source: The Freedonia Group, Carnegie Research
4.1% CAGR (06-21e)4.8% CAGR(16-21e)
343 455 457 590
77 102 125 145
95 141 163 205
515 698 745 940
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2006 2011 2016 2021e
USD
bn
Global silicones resins gels & other market by region
Electrical & Electronics Transportation Other markets Total
Energy required (KWh/t) per tonne 12.000-13.000 8500-9000
Market size
Total size (mill tons) 2.7MT FeSi 6.5MT
… ex China 1.7MT 2.4MT
China export 0.9MT 0.4MT
Elkem's market position
Elkem capacity (kt) 215 383
Share revenues 2017 30% 20%
Share EBITDA 2017 25% 22%
Elkem's share (ex China) 15% 23%1
… position 3 11
Top 5 share of market 71% 54%
(1) In foundry products Source: Various industry sources, Roskill, Carnegie Research
c
Elkem
38 02 May 2018
Demand for silicon Demand for silicon comes from three main sectors: aluminium, silicones and solar/electronics. Aluminium is the largest sector, with around 43% of silicon demand, followed by silicones (around 33%) and solar/electronics (24%). As all three sectors exceed global GDP growth, the overall historical growth rate has been high. Demand for silicon materials (silicon metal with Si content >99.99%) has grown steadily over the past 35–40 years. At the beginning of 1980, demand for silicon metal hovered around 350,000t. Today, global demand has increased by 8x, implying a CAGR of close to 6%. For the first 20 years, until year 2000, growth rates averaged around 5%, while the CAGR over the past five years (2012–17e) – depending on the source – has averaged between 5.3% (Roskill) and 6.3% (CRU).
We expect continued brisk growth of 5–6% over 2016–21e. We expect silicon demand from the solar sector to still exceed 10% per year for the foreseeable future. In the silicones sector, we see demand growth around 4–5% annually, fuelled by increased Chinese consumption. The aluminium sector will grow based on underlying growth in the industry and structural growth from higher use of aluminium content in cars. In the next three sections we will briefly go through each of these three sub-sectors.
In the tables below we have outlined demand estimates from Roskill and Cru. These estimates were made in October 2017. Since then, global GDP growth has continued to surprise on the upside.
China is the largest source of demand for silicon, accounting for 37% in 2017 according to CRU. In 2021 this share is expected to grow to 42%.
Silicones (MCS)33%
Solar/electronics (TCS)24%
Aluminium and other43%
End-use demand for silicon
Source: CRU, Carnegie Rsearch
0.0
0.5
1.0
1.5
2.0
2.5
3.0
2000 2005 2010 2015
World silicon metal consumption by end -use, 2000-2016 (Mt)
Aluminium Silicones Polysilicon Other
Source: Carnegie Research, Roskill
World silicon demand by intermediate end-use (MMt)
Solar 17% 21.4% 7.5% Solar/electronics (TCS) 24% 13.8% 10.0%
Semiconductors/oth 5% 6.6% 1.6%
Total (MT 2017e) & CAGR 2.97 5.5% 4.0% Total (MT 2017e) & CAGR 2.74 6.3% 5.6%
Source: Roskill, Carnegie Research Source: CRU, Carnegie Research
c
Elkem
02 May 2018 39
Aluminium demand for silicon The aluminium sector is the largest end-user for silicon metal. Aluminium sector end-use totals around 43%, or 1.2mt, of annual demand of around 2.7mt for silicon metal (CRU figures). The chart below shows growth in the aluminium sector has historically been robust, and with China’s ‘entry’ as the growth machine around 2000, the growth rate since (CAGR 5.9%) has been higher than in 1990–2000 (CAGR 2.4%).
In aluminium, the transport sector (about 30% of total aluminium demand) is the most important driver for silicon demand, totalling around 60% of silicon metal demand.
Source: Metallgesellschafrt, CRU, Metal Bulletin, Carnegie
0
5,000
10,000
15,000
20,000
25,000
30,000
35,000
40,000
2000 2002 2004 2006 2008 2010 2012 2014 2016
Global primary aluminium production (kt)
World ex China production China
Source: CRU, Carnegie Research
c
Elkem
40 02 May 2018
In the transportation sector, motor vehicles total around 80%. If we use North America as an example, passenger cars equal around 75% of the aluminium consumption in motor vehicles, trucks/trailers around 12.5% and buses around 12.5%. Passenger cars are the most important market driver for aluminium.
As we see it, three trends in the automotive industry will affect demand for silicon metal:
• Aluminium content per car has increased and we believe it will continue to do so. To save weight, aluminium (as well as other lightweight materials) will continue to take market share from steel. Stricter emission targets will further reinforce this trend. This will benefit the overall demand growth for aluminium in the transportation sector.
• Relatively lower share of cast aluminium in total aluminium production. As a large portion of the incremental demand from higher aluminium content will be in the form of extruded and rolled aluminium products, this will reduce the market share of cast aluminium. Cast aluminium typically contains 8% silicon metal, while non-cast aluminium only around 0.5%.
• China’s growing market share in car production and relative diminishing construction demand versus total aluminium consumption. China is gradually nearing the developed world in aluminium content in cars. Lower relative demand from the construction sector is also a positive factor as construction typically constitutes around only 6% of silicon consumption in aluminium alloys.
Since the oil embargo in 1973, car producers have increased the aluminium content per car. Today, according to Ducker Worldwide (2016), the average aluminium content for a European produced vehicle varies from 62kg for the Smart Fortwo up to 610kg for the Range Rover Sport. The average car is around 150kg. For the US, the corresponding figure is around 180kg. Some 40 years ago the average US-produced car contained around 40kg of aluminium. In our opinion, aluminium content per car will increase even more. The reasons are generally more safety and entertainment equipment on board, and increased features, adding to the weight. In keeping the weight and CO2 emissions low, increased use of lighter materials such as aluminium will be part of the solution. Ducker Worldwide expects that the average US produced car will contain ~20% more aluminium in 2020 than it did in 2015 and a further 11% more in 2025. With all the extra equipment, most car models are becoming heavier. We have included the development of the iconic Fiat 500 as an example of how the auto industry has developed over time. This little car weighed 470kg 40 years ago, and now is 940kg.
Transportation60%Machinery
12%
Electrical9%
Consumer durables7%
Construction6%
Containers2%
Other4%
Silicon metal consumption in aluminium alloys
Source: AlloyConsult, Roskill,Carnegie Research
c
Elkem
02 May 2018 41
Stricter emissions standards and regulations will further fuel car producers’ appetite for aluminium. 2020 emission targets for the main car producing regions, except Japan, are more than 20% lower than emissions in 2015.
Though higher aluminium content per car is positive for aluminium demand, we expect that the effect for silicon metal consumption will be a little bit more muted. This is because we expect the extra volume of aluminium per car to come foremost from sub-sectors other than cast products. The reason why transportation holds such a large market share for silicon metal-derived consumption is because a large portion of the aluminium for the transportation sector is cast aluminium, which typically contains around 8% silicon metal, while non-cast aluminium consumes only 0.5% silicon metal.
Auto production (m) 2014 2015 2016 2017e 2018e 2019e 2020e
North America 16.0 16.4 16.7 15.9 16.3 16.3 16.3
South America 3.8 3.0 2.7 3.3 3.7 3.9 4.0
America 19.8 19.4 19.4 19.2 20.0 20.2 20.3
Western Europe 13.3 14.2 14.7 14.8 15.0 14.8 14.6
Eastern Europe 6.8 6.7 6.8 7.4 7.6 7.8 8.2
Europe 20.1 20.9 21.5 22.2 22.6 22.7 22.8
Japan 9.0 8.6 8.5 9.0 8.7 8.6 8.5
China 21.7 22.8 26.0 26.6 26.8 28.1 29.5
Rest of Asia 12.1 12.4 12.5 12.6 13.1 13.5 14.1
Other (MEA) 1.9 1.9 2.3 2.5 2.8 2.9 3.1
Global 84.6 86.0 90.2 92.1 94.0 96.0 98.3
Source: IHS Automotive, Carnegie Research
50
70
90
110
130
150
170
190
China USA Japan EU
CO2 emissions from passenger cars (g CO2 per km)
2010 2015 Target 2020
Source: The International Council on Clean Transportation,Carnegie Research
Cast66%
Extruded11%
Forged5%
Rolled18%
Aluminium content in European cars by semi product
Source: Ducker Worldwide, Carnegie Research
c
Elkem
42 02 May 2018
China’s entry into the aluminium market and its share of overall aluminium production has gone hand in hand with lower silicon metal unit consumption. In 2005, when China’s share of global primary aluminium production equalled 10%, silicon metal consumption of around 0.7mt equalled about 3% of global primary aluminium production, or about 34mt. Today, China’s share is around 57% and unit consumption has fallen to around 2.2%. As China’s market share in car production (around 34% today) is significantly lower than its overall market share, this translates into relatively limited consumption of silicon metal. Now we believe that the country’s lower dependence on the construction sector and the continuing growth in both car production and aluminium content in cars will, isolated, stall the negative trend in unit consumption.
Overall we expect growth in silicon metal consumption from the aluminium sector of 3.1–3.9% over the next years, though in a falling trend. We have based this scenario on production growth in the primary aluminium sector 4–4.5%. Our estimates are basically in line with those of CRU and Roskill (both from October 2017). We believe that there is upside to our estimates due to the continued relatively strong global growth.
Silicon metal consumption in aluminium (mill tons)
Silicon consumption Growth
Source: Roskill, estimates by Carnegie Research
0.0%
1.0%
2.0%
3.0%
4.0%
5.0%
6.0%
2017e 2018e 2019e 2020e
Silicon metal consumption growth in aluminium
CRU Roskill Carnegie
Source: CRU, Roskill,Carnegie Research
c
Elkem
02 May 2018 43
Demand for silicon material from the solar and semiconductor industry The solar industry has enjoyed tremendous growth in the past years as prices on solar equipment have dropped significantly, leading to electricity prices from solar power plants being on a par with and in many cases below traditional energy sources. A solar module is made from a solar cell which is made from a wafer, whereas the prime ingredient is polysilicon. Polysilicon is basically purified silicon material, but not as pure as silicon material used in the semiconductor industry. In technical terms the polysilicon for the solar industry is 99.999999% pure, while the semiconductor-based silicon is 99.999999999% pure. The majority of polysilicon is made through the Siemens method where silicon material is transformed into silane gas and then inserted into the Siemens reactor, where a rod builds with pure polysilicon. After a three-week period the rods are harvested and the silicon is broken into pieces that are shipped to the ingot/wafer manufacturers. A few polysilicon producers (Wacker, REC Silicon, GCL Poly) are using fluidised bed reactors where silane gas is turned into polysilicon granulars. Both methods use about 1.5 kg of silicon material to make 1kg of polysilicon.
There are two ways of illustrating the historical use and the future need of silicon material: 1) the capacity development of the polysilicon manufacturers of the world; and 2) the expected growth of solar installations worldwide. We argue that end-user demand, namely the expected growth of solar power, will give the best view of the need for silicon material, but we will also provide some graphs showing the capacity development in the polysilicon industry.
Solar power is an old technology; for example, even Thomas Edison back in the early 1900s was looking for ways to invest in the power of the sun. The problem was high costs in manufacturing leading to very expensive electricity. Around 2003 Germany introduced attractive feed-in tariffs by which developers of solar power plants would get a certain amount of money per kWh produced in order to narrow the gap between traditional electricity and solar power electricity. This was followed by several other European countries, and the cost of producing solar cells and furthermore solar electricity started coming down rapidly. In 2011 the US, led by California, begin with tax benefits for solar power, and in 2012 China entered the scene and soon became both the largest installer of solar power and the largest manufacturer of modules, cells, wafers and polysilicon.
In 2009, 21.8 GW of solar power was installed throughout the world. This equalled 43.530 GWh of electricity every year, which represented roughly 0.2% of the world’s annual electricity consumption. By the end of 2017 this number had grown to 397 GW of installed solar power – a CAGR of 145% – or roughly 3% of the annual worldwide electricity demand. This clearly points to further growth of solar installations as long as the prices remain low. At the same time, with annual installations of 100 GW of new solar power, solar electricity is close to representing 40% of the world’s annual need of new electricity; in other words, the growth in electricity consumption. To that extent it is not necessarily that we will continue to see the big jumps in annual demand Y/Y, but instead that annual installations will flatten out.
Future demand is to a large extent about China. In the current five-year programme from China the target in 2020 is 105 GW of installed solar power. It reached this target by the summer of 2017, as the cumulative figure of solar power in China stood around 130 GW by the end of 2017. The question is therefore what the Chinese government will do for future demand. If it tries to slow the installation rate, this will have a profound impact on the global demand for solar equipment since China accounted for 50% of all solar panels installed in 2017. Also considering the fact that around 80% of all solar equipment in the world is manufactured in China, it seems strange to expect a major push from the government to slow down installations. If, therefore, we keep the Chinese installations fairly stable in the coming three years, we get the following demand curve for solar power measured in GW:
c
Elkem
44 02 May 2018
This demand scenario also includes a slowdown in the US market following the import duty on Chinese modules introduced by President Trump this year (expected to increase the prices on solar modules in the US market and thereby the price on solar electricity). We assume 7GW of new solar power capacity in the US in 2018, which is down from 8 GW in 2017. India is the next big market on the horizon, with President Modi’s target of 100 GW in solar power installed by 2022e. India currently stands at 27 GW of installed power and hence will increase, and our numbers incorporate India reaching its target, implying 15 GW of new power installed in 2018e compared to 14 GW in 2017.
To calculate the end demand for silicon material there are a few other assumptions to be made. Firstly the use of polysilicon per solar watt manufactured. One of the reasons the solar installations have boomed is the extreme cost reductions in the industry. This has happened throughout the whole value chain, but thinner wafers have obviously helped in the process. Back in 2006 the assumption was around 10g of polysilicon per watt measured in the finished wafer. This dropped to 7g by 2010, 5.25g in 2015 and currently 4.9–4.7g. We assume this will drop further, towards 4.5g per watt by 2020e, which seems to be what the largest wafer manufacturers are aiming for. A new method of wiring is being adopted in 2018/2019e and is supposed to be a major contributor to reach this goal.
Secondly we have assumed 5% growth for the semiconductor industry, and thirdly we have assumed that manufacturing 1kg of polysilicon and semiconductor silicon consumes 1.5kg of silicon material.
With these assumptions we end up with the following demand scenario for silicon material from the semiconductor and solar power industries.
Total North-America Total Europe/ EU Total Asia/RoW ex. China China
Source: Carnegie Research
c
Elkem
02 May 2018 45
Another way of looking at future demand for silicon material from the solar industry is capacity increase in the polysilicon industry. If we assume full utilisation, which is what the companies aim for given the economies of scale in running these plants, the output of polysilicon is set to increase 2.7% per year in 2017–20e. The companies that are increasing production are the Yulin Joint Venture between REC Silicon and Shaanxi Non-Ferrous Tian Hong New Energy, and East and Daqo New Energy – both Chinese based companies. The graph below shows the development measured in polysilicon year-end capacity, and the increase among new companies that entered the industry in 2008 when the solar industry started booming.
The buyers of silicon material for use in the polysilicon industry are fairly fragmented, with the largest player representing 20% and the five largest accounting for 65% of the supply market. The main players are shown in the graph below.
Silicon material demand from the solar and semiconductor industry
Polysilicon Semigrade silicon
Source: Roskill, Carnegie Research
0
100,000
200,000
300,000
400,000
500,000
600,000
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018e
2019e
MT
Polysilicon supply
Established producers New initiatives
Source: Carnegie Research, company reports
c
Elkem
46 02 May 2018
Supply-side factors As in many raw materials, the supply sector is dominated by China. Silicon metal is no exception. China holds around three-quarters of the global silicon metal capacity of around 5.3mt. Its global share of production is somewhat lower at around 64%.
There is a big difference between the Chinese and non-Chinese silicon metal sector. The top 10 non-Chinese producers hold 96% of the capacity; the Chinese silicon metal industry is very fragmented. The top 10 producers have a 35% market share, while the 30th largest player has a capacity of only 10,000t. Adding up producers with capacity of 10,000t or below gives 2mt.
GCL Holdings16%
Wacker20%
OCI Company16%
Hemlock5%Daqo New Energy
4%
REC2%
Hankook Polysilicon9%
Others28%
The largest polysilicon manufacturers 2015e
Source: Carnegie Research
China75%
Europe ex Russia9%
USA4%
RoW12%
Global silicon metal capacity - total 5.3 mill tons
Source: Roskill, Carnegie Research
China64%
Europe ex Russia15%
USA6%
RoW15%
Global silicon metal production- total 2.7 mill tons
Source: Roskill, Carnegie Research
Ferroglobe37%
Elkem17%
Dow Corning12%
Rima7%
Rusal6%
Simcoa4%
Wacker4%
GS Enegy3%
Ligas de Alumino3%
Minasligas3%
Others4%
Non-Chinese silicon metal producers share of capacity
Source: Roskill, Carnegie Research
-
0.50
1.00
1.50
2.00
2.50
3.00
3.50
4.00
4.50
World ex China China
China and non-China silicon metal capacity (mill tons)
Total Top 10 pruducers Top 5 producers
Source: Roskill, Carnegie Research
c
Elkem
02 May 2018 47
While global silicon metal production ex China has been relative stable for the past 17 years (right chart below), China’s production has more than quintupled.
Thus China is the global number one. In fact its silicon metal industry has been export oriented from the start. At the beginning of the decennium, China exported 85% of its silicon metal production. With brisk demand growth, the export share is now close to 40%.
Despite the positive price trend for silicon metal, we doubt we will significant new amounts of Chinese silicon metal on the export market. The strong increase in raw material prices is one of the reasons as thermal prices have more than doubled since YE(15).
Silicon metal market (MT) Implied Implied
Region Capacity Production Op.rate Demand net export net import
Implied World ex China import of silicon metal (kt)
Source: Roskill, estimates by Carnegie Research
c
Elkem
48 02 May 2018
Chinese export price have historically bottomed out at global average cash cost. With the recent run in Chinese silicon metal prices one can argue that this is merely a reflection of higher raw material prices. Chinese thermal coal prices have risen close to 50% over the last 7–8 months, which isolated will add around USD260/t to Chinese silicon metal production costs. Adding to producers woes is the appreciating Chinese currency. Since the peak at the beginning of 2017, the currency has strengthened around 10% vs. USD, making life even tougher for an exporter.
The third factor is trade restrictions that effectively keep Chinese silicon metal out of the EU and US (see next section). The table below outlines possible new non-China capacity. The only plant that can come into action is United Silicon. We understand it is idle waiting for a buyer. Apart from that, we expect very little new capacity coming on stream the next two years.
Ferrosilicon ovens can always be converted to silicon metal. We believe that capex is USD,1000–1,500/ton. Elkem’s recent acquisition of Rana (~80,000t ferrosilicon) and placing it in the Silicon Materials division may indicate such potential conversions. With the recent rock bottom price gap between silicon metal and ferrosilicon, we doubt we will see any material conversions.
New potential silicon metal capacity Country kt On-stream Comment
PCC Iceland 35 2019-20 Greenfield
Thorsil Iceland 50 After 2020 Greenfield - on the drawing board
Silicor Iceland 20 After 2020 Greenfield - on the drawing board
United Silicon Iceland 23 Greenfield - idled
Total 128
Source: Alloyconsult, Carnegie Research
c
Elkem
02 May 2018 49
Generally the overall operating rate in the silicon metal industry is low, so scarcity has historically been a non-theme. Now we believe we will move towards a tighter supply and demand balance. Capacity in China has been flat for the past four years, operating rates among non-Chinese players are relative high (we assume around 70%), trade restrictions in the US can probably push out Brazil, and higher costs coupled with a stronger currency for Chinese producers all point to higher prices.
The left chart above outlines price forecasts from CRU and Roskill in October 2017. Since then, spot prices have risen substantially. Trade restrictions in the US are most likely the catalyst but higher raw material (coal) prices add to the momentum.
1.40
1.60
1.80
2.00
2.20
2.40
2.60
2000 2002 2004 2006 2008 2010 2012 2014 2016 Current
Silicon metal / Ferro silicon price ratio
Source: Bloomberg, Roskill,Carnegie Research
-
500
1,000
1,500
2,000
2,500
3,000
0
500
1,000
1,500
2,000
2,500
3,000
3,500
4,000
2008 2010 2012 2014 2016 2018e 2020e 2022e
EU
R/t
USD
/t
Silicon metal price forecasts
Europe spot USD/t - Roskill Europe spot (EUR/t) - CRU
Source: CRU, Roskill, Carnegie Research
1000
1500
2000
2500
3000
3500
2008 2010 2012 2014 2016 Current
Spot silicon metal prices (USD/t) - US & EU
Europe spot (USD/t) USA spot (USD/t)
Source:, Bloomberg,Carnegie Research
c
Elkem
50 02 May 2018
We think the greatest downside risk is lower Chinese coal prices once the winter heating season ends 15 March. Therefore we are a little uncertain whether current prices (USD3,040/t in the US and USD2,684/EUR) will hold once the Chinese environmental efforts ends. US and EU AD duties will also effectively prevent these two markets from Chinese silicon metal, but there may be a spill-over effect into other countries, mainly in Asia. Therefore we expect European prices around EUR2,100/t and US around USD2,800/t. As we expect a relative tight market compared to the recent past, we expect prices to increase by 1–2% in the next 2–3 years. An overview of the historical price trend is shown below.
Foundry products The ferrosilicon market has been plagued by low industry operating rate for decades. Figures from 2016 suggest an overall operating rate of 47%. In general, market growth is lower than for silicon metal and silicones, due to a mature market and lower silicon unit consumption in steel. China is by far the largest producer, totalling close to two thirds of global production and it is also the largest exporter. The industry is very fragmented with the 20 largest producers holding just over a third of the global market. We view the market as supply driven, in which changes in operating costs for the marginal player(s) dictate the price outlook.
Production of ferrosilicon Global production of ferrosilicon doubled in 1985–2016 at 2% CAGR, which is considerably below the growth rate of silicon metal. The global market for ferrosilicon is 6.5m–7.0m tons. In 1985-–2000 growth in ferrosilicon output declined slightly, but since then growth has picked up considerably with the onset of Chinese production. From 2002 to the peak in 2013 growth in annual production had a CAGR of 6%.
After 2002, ferrosilicon production growth was driven by expansion of Chinese capacity. According to Roskill, the Chinese share rose from 41% in 2002 to 74% in 2009, when it peaked. Since then it has fallen, which has contributed to a drop in overall global production. Chinese ferrosilicon production has fallen since the peak in 2011, but global production (ex. China) has been fairly stable.
Roskill estimates that in 2014 there were over 1,000 individual producers of ferrosilicon in China, which fell to 920–930 as of 2017 due to government action to curb pollution. This has contributed to a fall in overall production. Capacity utilisation has fallen from 48% in 2014 to 41% in 2016 because production of ferrosilicon has fallen by more than capacity. The low utilisation is a telling indication of Chinese overcapacity. Many small-scale producers have been shut down, but the majority is still producing less than 10,000t per year
Consensus among market participants seems to be that the low capacity utilisation and strict regulatory clampdown in China will lead to limited new production facilities there. Erdos (the world’s largest ferrosilicon producer, operating out of China) has expanded its footprint in recent years by acquiring other plants and increasing utilisation. However, this has been offset by widespread closure of other plants (about 850,000t per year net reduction). If there is going
World Ferrosilicon production 2000-2016 (Mt, basis 75 % Si content)
China Rest of world
Source: Carnegie Research, Roskill
c
Elkem
52 02 May 2018
to be additional supply in the ferrosilicon industry, it is expected that it will come from outside China. For instance, Malaysia has had a strong uptick in production since 2014, driven by two new plants coming on stream. They have not yet reached full capacity and they are expected to add some new supply. Given this and falling consumption, very few new ferrosilicon plants have been commissioned. Of the two, one is owned by Elkem in Paraguay (still under construction) and the other is an expansion project by Mechel. Together they will ad 41,000t per year, which is very limited given the total market supply of 6.4mt per year in 2016 (i.e. 1% of total supply).
Roskill estimates there was 13.55mt per year ferrosilicon capacity in the world in 2017, implying that global capacity utilisation was 47% in 2016, down from 54% in 2014. This low capacity utilisation is basically driven by China, where production has fallen faster than capacity. Excluding China, capacity utilisation was 65% in 2016, which is still down from 75% in 2014.
Consumption of ferrosilicon Compared to silicon metal, which serves three growth markets, ferrosilicon demand is driven mainly by the steel market, which consumes around 70% of global production. The other two markets are iron casting (17%) and magnesium (13%). One of the reasons that consumption of ferrosilicon has grown more slowly than silicon metal is the lower amount of silicon content in steel than in silicon metal applications.
New ferro silicon capacity (kt) Country kt On-stream Comment
Since 2000 consumption of ferrosilicon has grown at CAGR 4%, primarily driven by the early 2000s, as consumption has fallen since 2010. In 2016 around 6.5mt was consumed. The drop in overall consumption differs from silicon metal. The largest contributor to growth has been China in the early 2000s, but as its consumption has fallen in the past seven years, so has total world consumption. Since 2016 China has accounted for about 60% of world consumption.
Besides being a deoxidising agent typically used in carbon steel, silicon can also improve the electrical conductivity of steel. Hence, it is used in electrical steels where the silicon content is much higher than normal steels. Additionally, silicon can also improve strength and corrosion resistance, which means that there is generally more silicon in stainless and alloy steels. This is very important to note in terms of Elkem, as its production is made up of relatively little standard ferrosilicon and more speciality, nodularisers and inoculants. We will outline the demand drivers for different steels and show why being exposed to more speciality steel is beneficial, beyond the fact that they consume more silicon per tonne steel.
World Ferrosilicon real consumption by end use 2000-2016 (Mt)
Stainless steel Carbon & other alloy steel Iron castings Magnesium Other
Source: Carnegie Research, Roskill
0
1
2
3
4
5
6
7
8
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
Mill
ion
tonn
es
World apparent consumption of ferrosilicon by region compared with real consumption, 2000-2016 (Mt)
China Japan Other Asia Europe Americas Other
Real consumption
Source: Carnegie Research, Roskill
c
Elkem
54 02 May 2018
The declining amount of ferrosilicon used in steel applications is partially due to declining intensity of use in carbon steel in China, according to Roskill. As the left chart below shows, the production of crude steel stagnated in 2014 due to a lower share of Chinese crude steel. The biggest consumer of crude steel is the construction sector (49%).
Stainless steel, however, has grown at CAGR 5.5% since 2000 with only a slowdown in 2007–09. The primary driver has been China at a CAGR of 24%. Japan and Europe have grown very little, but kept annual production roughly unchanged. For a producer of ferrosilicon this suggests it is more attractive to be exposed to these markets with speciality FeSi than the standard FeSi used in crude steel. Unlike crude steel no major end-market is served by stainless steel. The largest is metal products and white goods (39%) followed by mechanical engineering with 25%. This picture is fairly similar with electrical steel, but growth has been somewhat lower. We have not found any split in relative demand by end-uses.
Stainless steel14%
Electrical, valve & spring steel
12%
Other alloy steel5%
Carbon steel69%
World ferrosilicon consumption, by steel group
Source: Carnegie Research, Roskill
-15
-10
-5
0
5
10
15
20
600
800
1,000
1,200
1,400
1,600
1,800
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
2011
2013
2015
year
-on-
year
cha
nge
(%)
Prod
uctio
n (M
t)
World production of crude steel, 1981-2016
year-on-year change Production
Source: Carnegie Research, Roskill
Construction49%
Machinery17%
Automotive14%
Metal hardware7%
Oil & gas5%
Others8%
World steel consumption by end-use
Source: Carnegie Research, Roskill
c
Elkem
02 May 2018 55
Going forward production of speciality steel is expected to grow at CAGR 3% in 2016–21e. This is slightly lower than the rate of around 5% in 2012–16. The main driver will continue to be China, but the rest of the world (ex. China, Europe and Americas) is also expected to be a strong contributor, growing at CAGR 4.5%.
Cast iron Cast iron accounts for 17% of global ferrosilicon consumption. However, in Elkem it probably accounts for over 50% due to its skew towards speciality end-uses. Consumption of ferrosilicon in cast iron rose around 1.5x in 2000–16, particularly rapidly in 2000–07 driven by Chinese consumption. Despite this, the share of ferrosilicon consumption by cast iron declined in the same period, reflecting very strong growth in steel consumption relative to cast iron. Since 2009 this trend has shifted and the share has increased somewhat, but still not exceeding the 20% share in the early 2000s. Growth since 2011 has been muted, reaching around 1.1mt in 2016.
0
5
10
15
20
25
30
35
40
45
50
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
Prod
uctio
n (M
t)
Production of stainless steel by region, 2000-2017
China India Japan Other Asia Europe USA Other
Source: Carnegie Research, Roskill
Metal products & white goods
39%
Mechanical engineering
25%
Transport15%
Construction13%
Electrical machinery8%
World consumption of stainless steel by application
The industry specialist CRU expects that production of cast iron will grow at about 2% in 2016–21e. This is marginally up from unchanged since 2012.
Carbon products Market data for carbon products is extremely limited partly because of: 1) production by many players, such as Elkem and Ferroglobe, is characterised by captive capacity; 2) players such as Graphite India make both carbon and graphite products, tilting towards graphite; and 3) the market is fragmented with many small and unlisted players supplying a limited number of smelters. Hence the market data that we present below is mainly based on comments from outlook statements in annual/quarterly reports and Elkem’s estimates of the market size.
The market data below is based on global production of electrode paste, ex. China. We have not found any official markets or extensive data for consumable electrodes, ramming paste, cathode blocks and recarburisers. Annual Global (ex. China) production of electrode paste was around 476,000t per year in 2016 according to Elkem. Of this, the company accounted for 43%. Note that this figure does not match Elkem’s total production in the carbon division as Chinese domestic sales are excluded and other carbon materials. Most of its production is for internal use. Most Chinese producers are privately owned and there is limited information about their production capacity.
0%
5%
10%
15%
20%
25%
0
200
400
600
800
1,000
1,200
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
kt
World consumption of ferrosilicon in cast iron, 2000-2016
FeSi consumption in cast iron (left axis) cast iron % of world FeSi consumption (right axis)
Europe North America South America China Japan Other Asia CIS South Africa
Source: CRU, 2017
0% CAGR
c
Elkem
02 May 2018 57
The largest producer, Yangguang Carbon, states on its website that it has production capacity of 1,000,000t per year, but we think most of this is taken up in the Chinese market.
There is no way to tell what the exact demand for carbon products is. However, from SGL Group’s annual reports since 2010 we can pick out some market trends from comments made. Mainly, the message seems to have been that in the past years cathodes, furnace linings and carbon electrodes have done well. Graphite electrodes have struggled due to price pressure and surplus materials, but Elkem does not produce this. Additionally, demand for cathodes went through a difficult period in 2010–12 due to peaking investments in the aluminium industry.
According to Elkem, the key customers in its Carbon division are ferroalloy, silicon and aluminium producers. Demand for carbon products is therefore driven by these industries. For an overview of demand drivers see the Silicon materials and Foundry alloys sections.
Roskill provide an indexed (2007=100) series of prices for baked and paste electrodes. Typically, ferrosilicon producers will use paste electrodes, while silicon metal producers use baked. At the start of 2017 per tonne price of baked electrodes was 3x higher than paste electrodes. Since 2013 the prices of both have fallen, and the price of baked electrodes has fallen by relatively more than paste.
205
476
129
143
43%
27%
30% 100%
0
100
200
300
400
500
Elkem Top 2-5 Other Total
Tonn
es p
er a
nnum
(100
0's)
Global electrode paste production ex. China, 2016 (1,2)
Source: Carnegie Research, Elkem1Excluding China as Chinese market is highly fragmented, includes mostly domestic focused producers
SGL Group statements from annual reports
Year Statement
2016 "Business with cathodes, furnace linings and carbon electrodes continued to be good as expected"
2015 "We expect our business in cathodes, furnace linings, and carbon electrodes to continue to perform well"
2014 "EBIT decreased more than proportionately to sales revenue … mainly attributed to price pressure in both graphite electrodes and cathodes"
2013 "Modernization of existing aluminium capacity will be associated with a shift in demand to higher quality graphitized cathodes, which is likely
to benefit our Business Unit Cathodes & Furnace Linings"
2012 "Since mid-2011 we have seen a slight recovery of cathode demand in the aluminium industry"
2011 "Still slow demand for cathodes for the aluminium industry"
2010 "Anticipated investment pause in the aluminium industry [affected] the demand for our cathodes"
Source: Carnegie Research, SGL Annual Reports
c
Elkem
58 02 May 2018
Trade restrictions Both silicon metal and ferrosilicon are subject to anti-dumping (AD) duties in the US and the EU. In the US, Chinese and Russian imports are met with various duties:
There is a general 139% AD duty on Chinese silicon metal (first imposed in 1991), which is up for review. With the Trump administration in place, we doubt that the evergreen AD duty will be changed. Two producers have particular AD duties of 14-49%.
The Russian producers Rusal (two plants) and Mechel with combined capacity of 75,000t are subject to AD duties of 62–87%. Generally, US silicon prices have consistently been higher than those in the EU (see chart below), mainly because Chinese producers are effectively being kept out of the US market.
In the EU, five Chinese producers are subject to a 16–19% AD duty. As Chinese domestic prices are lower than those in the EU, in combination with fairly low AD duties, China still has a fair share of the European market. While European demand for silicon metal totals around 750,000t, imports from China totals some 15–%. Also, it is possible to circumvent the AD duties in EU if the raw materials are re-exported as part of a finished product to a country outside EU. Dow Corning is one of the players that exploit this duty drawback.
Last year, the US began a new investigation into imports from Brazil, Norway, Australia and Kazakhstan. We do not have the numbers for Kazakhstan, but Norway exported 10,000t, Brazil 71,000t and Australia 17,000t in 2016. October last year the US Department of Commerce (DOC) announced the preliminary rates (see below).
Recent announcements from the US International Trade Commission have concluded that the import from these four countries have not harmed domestic production. This is in line with Canada’s stance, which recently dropped AD duties against Brazilian (as well as Kazakhstan, Laos, Malaysia, Norway and Thailand) based on the findings that this import was not harmful.
As for ferrosilicon, only the EU imposes AD duties. While the global no.1 ferrosilicon producer, Erdos (capacity 1.5mt), is met with a 16% AD duty, other Chinese producers face a 16% duty. Russian imports have 18–23% duty. These duties (typically lasting for five years) were imposed in 2014 and will last well into 2020. As a result, Europe imports insignificant volumes from China and Russia.
US Preliminary Silicon Dumping & Subsidy AD duty Subsidy % Total
Brazil - DC Brazil 56.8% 3.7% 60.5%
Brazil - LIASA 134.9% 52.1% 187.0%
Brazil - Rima 56.8% 3.7% 60.5%
Brazil - Others 56.8% 3.7% 60.5%
Norway - all producers 3.7% 3.7%
Australia - Simcoa 20.8% 16.2% 37.0%
Kazakstan - Tau-Ken Temir LLP /others 120.0% 120.0%
Source: Carnegie Research, US DOC
EU anti-dumping duties on ferro silicon producers
Country AD duty Comment
China 15.6%-31.2% 3 producers
Russia 17.8%-22.7% 2 producers
Source: Carnegie Research
c
Elkem
60 02 May 2018
Financial forecasts We expect revenue growth of 7.4% per year over the next three years. On the back of increased specialisation in Silicones and higher prices in Silicon Materials and Foundry Products, we expect EBITDA margins to increase by close to 2.5%-points over the coming years, reaching 17.3% in 2020e. This implies an EBITDA CAGR of 13.2%. We do not see this as a stretched target as Elkem’s non-Silicones divisions all had EBITDA margins above 19% in 2015. Following the acquisition of Xinghuo and Yongdeng at EV NOK9.2bn, NIBD/EBITDA(17) is 2.5x. Post-money, we estimate NIBD/EBITDA will come down to 2.2x. We estimate an NIBD/EBITDA around 1.4x at YE(18). We find the balance sheet healthy and, with cash generation (operating cash flow relative to EBITDA) around 75%, Elkem will have high dividend capacity (we factor in a 40% pay-out ratio). We estimate capex around NOK 1.3bn (in line with depreciations) and further reduce NIBD by around NOK 1.2bn per year.
2016 and 2017 review In conjunction with the IPO, Elkem will acquire two companies (Xinghou and Yongdeng) from its owner at an EV value of NOK 9.2bn. To understand Elkem’s result, we will go through the transition from ‘old’ Elkem to the new set-up.
The 2016 and 2017 combined figures include Xinghua and Yongdeng and form the basis of the financial reporting structure post-IPO. The combined figures are not audited.
Historical & proforma P/L Elkem actual China acquisition Elim. Elim. Combined
EBITDA Group 2,207 1,618 2,098 -81 1,060 1,536 3,154
Silicones 470 413 705 -11 810 402 1,515
Silicon materials 907 673 534 -71 270 602 804
Foundry products 717 503 707 503 707
Carbon products 275 275 274 275 274
Other/Elim -162 -246 -122 -25 -246 -146
Source: Elkem, Carnegie Research
c
Elkem
02 May 2018 61
Changed structure Through the acquisition of the Chinese companies, Elkem will be far more tilted towards the end-user market, as silicone’s share of the 2017 pro forma P&L is 45%, up from around one-third in 2015. In 2017 Silicones EBITDA (NOK705m) totalled 33% of the Elkem Group EBITDA of NOK2098m. The (combined) Silicones EBITDA of NOK1515m in 2017 will account for around 48% of Elkem Group’s EBITDA of NOK3,154m. The starting point for Elkem in 2018 is that the company will have around half of its revenue and EBITDA coming from the Silicon segment, where margins in general are less volatile than in the remaining Elkem.
Result changes 2016 – actual Elkem EBITDA(16) fell 30%, mainly due to falling prices in Silicon Materials and Foundry Products.
For Elkem Silicones, the result fell by 12% from NOK470m to NOK413m. Sales were 5% lower due to a maintenance shutdown, but a weaker NOK vs EUR(-3%) and USD (-7%) and lower raw material costs contributed positively.
EBITDA for Silicon Materials fell from NOK907m in 2015 to NOK673m in 2016 – or 26%. The main reason was a 24% plunge in silicon metal reference prices. European spot silicon metal prices (source CRU) fell from EUR2,091/t in 2015 to EUR1,675/t in 2016, below 2009 (EUR1,725/t). At the end of 2016 Elkem acquired Rana Metall, which at that time totalled around 80,000t of the division’s combined silicon metal and ferrosilicon capacity of 240,000t.
EBITDA for Foundry Products fell by around 30% from NOK717m in 2016 to NOK503m in 2016, mainly due to reference prices falling by 19%.
EBITDA in Carbon Products was unchanged at NOK275m. The division suffered pressure on sales volumes and mix effects, but sales were flat as the new carbon plant in Malaysia (capacity 28,000t) came on stream in H1(16).
Silicones33%
Silicon matr.33%
Foundry25%
Carbon9%
Operating income Elkem consolidated 2017 (ex Other/Elim)
Source: Elkem, Carnegie Research
Silicones45%
Silicon matr.29%
Foundry19%
Carbon7%
Operating income Elkem combined 2017 (ex Other/Elim)
Source: Elkem, Carnegie Research
Silicones32%
Silicon matr.24%
Foundry32%
Carbon12%
EBITDA Elkem consolidated 2017 (ex Other/elim)
Source: Elkem, Carnegie Research
Silicones46%
Silicon matr.24%
Foundry22%
Carbon8%
EBITDA Elkem combined 2017 (ex Other/elim)
Source: Elkem, Carnegie Research
c
Elkem
62 02 May 2018
Bridge old Elkem – current set-up For 2016, total EBITDA for Xinghou (Silicones) and Yongdeng (Silicon Materials) totalled NOK-81m. The two companies had combined EBITDA of NOK1,060m in 2017:
In the table below we have outlined the divisional bridge between the old Elkem and the acquisition of the two Chinese companies.
Result changes 2017 – combined company In 2017, combined EBITDA more than doubled from NOK1,536m to NOK3,154m. The greatest improvement was in Silicones, where EBITDA rose from NOK402m to NOK1,515m. The major explanation was the ramp-up at Xinghuo, where volumes increased by of 23%, in combination with streamlining, higher efficiency, and increased specialisation. Much of sales is still commodity-type products and prices rose during the year.
In Silicon Materials, EBITDA rose 34% from NOK602m to NOK804m, mainly due to higher silicon prices following the temporary downturn in 2016. According to CRU, benchmark silicon metal prices in Europe rose 19% in 2017, but were still around 14% below the 2015 average. A weaker NOK vs EUR (-3%) and USD(-7%) also contributed positively. We believe Elkem Rana achieved low margins, but we estimate a small positive effect from this acquisition.
EBITDA in Foundry Products rose from NOK503m in 2016 to NOK707m in 2017. The result was positively affected by a weaker NOK and a continued higher share of speciality products, up from 78% in 2017 to 81% in 2017, as well as 2% higher sales volumes. Standard ferrosilicon prices in Europe rose 18% (source Bloomberg). Elkem also states that the improving margins are a reflection of lower costs.
EBITDA in Carbon Products were virtually unchanged at NOK274m in 2017. Higher raw material prices were offset by 11% higher volumes. Higher EBITDA in Other/Eliminations in 2017 reflected higher revenue in Sales and Logistics and Elkem Power and no power/FX losses compared to 2016, when losses were around NOK100m.
Revenues & EBITDA split 2017 Xinghuo Yongdeng Total
In conjunction with the IPO, Elkem will acquire Xinghuo and Yongdeng for an EV of NOK9.2bn, including NIBD of NOK5.0bn. The 2016 and 2017 combined balance sheet reflects as if Elkem has owned the companies both years. As a result NIBD for the combined company is NOK7.9 at YE(17). As the expected IPO proceeds of NOK5.1bn exceed the NOK4.2bn for the shares, the opening balance sheet after the IPO gives NIBD of NOK7bn.
Divisional outlook As we do not have any quarterly results, the uncertainty increases when estimating future results. Parts of Elkem’s sales are cyclical where earnings momentum, through price changes, during the year is important for understanding the full-year earnings potential. Not having quarterly results for each division reduces visibility. Elkem also sells a large share of its production as speciality products. The margins through the cycle are clearly above more standard products, but a question is how this margin evolves in line with the standard products. The accuracy in the annual estimates is therefore lower than they would be if we had had quarterly divisional results.
Balance Sheet Elkem Elkem Elkem China China Combined Combined Opening
Figures in NOKm 2015 2016 2017 2016 2017 2016 2017 Post IPO
NET WC (from bal sheet) 3,717 3,682 3,988 -249 -17 3,433 3,967 3,967
NET WC (adjusted) 3,441 3,710 3,868 3,868
… in % of revenues 21.9% 18.1% 18.1%
NIBD/EBITDA 1.4 4.7 2.5 2.2
Source: Carnegie Research
c
Elkem
64 02 May 2018
Silicones After the acquisition of Xinghuo, Silicones is the largest division for revenue and EBITDA. The table below shows our estimates for Xinghou and ‘old Elkem Silicones’ results in 2017. There are large margin differences between them. In 2017, Xinghou had an EBITDA margin of 18%.
Compared to Elkem Silicones, Xinghuo’s EBITDA margins are about 50% higher despite a relatively low share of speciality products. The reason is high prices for commodity products in 2017, but also low raw material prices. Compared to silicon metal prices in Europe, Chinese prices were around 20% lower. For 2018 we see higher EBITDA margin for Xinghuo compared to 2017 due to higher share of speciality products – up from 33% in 2017 to 37%.
For ‘old Elkem Silicones’, we expect margins to hover around 13%, in line with the adjusted EBITDA margin for 2017 when we also take into consideration 100% capacity utilisation. We expect margins to be stable, as we see pressure from rising silicon metal prices.
The margin play is how fast Elkem can manage to high-grade its portfolio towards more specialized products. The potential is obvious highest in China and we forecast that Xinghou will have 45% of sales as speciality products in 2020, up from 33% last year. The goal is to reach the ‘old Elkem Silicones’ share of 67% within five years. Given our best guess that speciality products as a group have 15-18%-point higher margins than commodity products in China, we estimate every 5%-point increase in speciality share of sales increase EBITDA by NOK100m.
We expect Silicones sales to grow rapidly. China is the growth machine and Freedonia expects a CAGR of 7.4% for the 2016–21e. Western Europe and the US are expected to grow 3.5–4.4% over the same period. In our estimates we expect Xinghou’s volume to growth by 5% and ‘old Elkem Silicones’ by 3%. With the EBITDA margin going from 15.1% in 2017 to an estimated 17.1% in 2020, we expect EBITDA to increase by 43% in 2017–20e.
Elkem’s EBITDA margins are between Momentive and Wacker Silicones. Dow Corning, the largest silicone player, most has likely higher margins than Wacker Silicones. Silicones is reported in Dow Corning’s Performance Materials & Coating segment, which had an EBITDA margin of 24.3% in 2017. We believe that the margin for Silicones alone is higher than this.
We believe that Wacker Silicones is the best peer for Elkem Silicones. As Elkem has more or less completed ramp-up of its upstream assets, focus will be to high-grade the portfolio. Wacker Silicones is probably five years ahead of Elkem regarding specialisation. Since 2013, its EBITDA margin has increased from 13.8% to 20.3%, which is barely above Elkem’s medium-term goal. If Elkem reaches this target in 2020, there is 19% upside to our EBITDA estimate of NOK2164m
Source: Elkem (historical figures), est. by Carnegie Research
c
Elkem
66 02 May 2018
Silicon Materials For 2018 we forecast a strong increase in EBITDA, up from NOK804m in 2017 to NOK1,181m in 2018e. There are several reasons for this. Elkem sells a relative large share of its production as speciality products, but the sharp price development for both benchmark silicon metal and ferrosilicon in 2017 (see charts below) will contribute positively to the results.
Generally, silicon metal and ferrosilicon prices are lagging six and three months before they affect Elkem’s earnings. In 2007 silicon metal/ferrosilicon prices rose 14%/30% respectively (source Bloomberg). Compared to the lagged prices in 2017 (silicon metal EUR1,758/t and ferrosilicon EUR1,149/t) current spot prices in Europe are 23% and 47% higher.
The price change for 2018 alone would have increased Elkem’s results by around EUR84m if all its sales were exposed to the spot market. We have assumed that only 50% of the theoretical price change will feed into the results. Furthermore, a stronger EUR will add around NOK40m to EBITDA in 2017. We assume that higher costs will shave off around 40% of the combined price/FX effect and that Elkem will achieve around NOK100m higher contribution from Rana Metall due to the improvement in 2017. As Elkem is virtually producing flat out, we expect only 1% production growth through capacity creep.
We expect limited price increases compared to the current price level, 1–2% per year. Although the operating rate among non-Chinese players is relative tight historically and Chinese producers are effectively shut out of the US and the EU, history has shown that some extra production has always found its way to the market when prices are high enough. The recent, sharp increase in silicon metal and ferrosilicon may have tempted producers with idled capacity.
Elkem acquired Rana Metall at the end of 2016, which is a ferrosilicon producer. The reason why this plant is placed in Silicon Materials and not Foundry Products is that Elkem intends to convert it to Silicon Metal over time. We have not taken this into our estimates.
After a strong rise in EBITDA in 2018, we forecast a stable performance.
Source: Elkem (historical figures), est. by Carnegie Research
c
Elkem
68 02 May 2018
Foundry products In line with Silicon Materials we expect a strong increase in EBITDA in 2018 for the Foundry division. Though a large part of the foundry alloys are based on long-term contracts fixed for 6–12 months with no link to ferrosilicon prices, the speciality and standard ferrosilicon for the steel industry are partly linked to CRU’s ferrosilicon price with a time lag around three months. We assume around 50% of these volumes are linked.
Standard ferrosilicon prices have increased substantially since the trough at the start of 2016. In percentage terms, coal prices which typically totals 30% of production costs, have increased even more. For 2018 we expect Elkem’s average achieved prices to increase by 8%. With the corresponding rise in raw material prices, the net effect translates into an expected 20% increase in EBITDA. The result in 2018 will also be affected by production from the newly started plant in China, which came on stream in December 2017. The plant (capacity 30,000t) buys ferrosilicon and re-melts it, thus we see lower margins here than the average Elkem portfolio. Our best guess is that the new entity will add some NOK10m in EBITDA for 2018.
We expect basically the same EBITDA margins with volume growth around 2%.
Carbon Products As previously stated, the results in Carbon Products have been flat in the past three years with EBITDA at NOK275m. Prices for carbon products are generally stable. Despite 11% volume growth in 2017 due to the start-up in Malaysia and higher production in Brazil, EBITDA stayed flat, mainly due to higher raw material prices.
Using Australia Newcastle Thermal coal price as a benchmark, coal prices rose 33% in 2017. As Elkem is the global leader in electrode paste (market share ex. China 47%) it can pass higher costs onto customers, but with a time lag. Therefore we expect the EBITDA margin to improve in 2018. As we expect 2018 coal prices to increase (current prices are around 14% higher than the average for 2017), we expect an EBITDA improvement of only 1.5%-point.
10%
11%
12%
13%
14%
15%
16%
17%
18%
19%
20%
800
900
1,000
1,100
1,200
1,300
1,400
1,500
1,600
1,700
1,800
2015 2016 2017 2018e 2019e 2020e
EB
ITD
A m
argi
n
Ferr
o si
licon
-E
UR
/t
Foundry Products EBITDA margins & FeSi75 (EUR/t)
EBITDA margin FeSi75 EUR/T - lagged by 3mth
Source: Bloomberg, Carnegie Research
3,674 3,642
4,247
4,9055,204
5,520
19.5%
13.8%
16.6% 17.4% 17.4% 17.4%
0%
5%
10%
15%
20%
25%
0
1,000
2,000
3,000
4,000
5,000
6,000
2015 2016 2017 2018e 2019e 2020e
EB
ITD
A m
argi
n
Rev
enue
s (N
OK
m)
Foundry products
Revenues EBITDA margin
Source: Elkem (historical figures), est. by Carnegie Research
Australia Newcastle Port Thermal Coal 6000kcal/kg FOB Spot USD/t
Source: Bloomberg, Carnegie Research
-
20
40
60
80
100
120
2014 2015 2016 2017 YTD
Annual average coal prices (USD/t)
Source: Bloomberg, Carnegie Research
c
Elkem
70 02 May 2018
We expect EBITDA margins of around the 2018 level and with around 2% volume growth, EBITDA should increase by around 5% after 2018.
Sanity check – Silicon Materials, Foundry Products and Carbon Products We have compared our joint estimates for Silicon Materials, Foundry Products and Carbon Products with Ferroglobe’s consensus estimates. Ferroglobe is the largest non-Chinese silicon/ferro-alloy producer. In general, Elkem has a larger share in speciality products than Ferroglobe. Furthermore, about 30% of Ferroglobe’s volumes are manganese alloys, which historically are more volatile than silicon metal and ferrosilicon.
The charts below compare our EBITDA estimates for Elkem Silicon Materials, Foundry Products and Carbon Products with consensus estimates for Ferroglobe.
Source: Elkem (historical figures), est. by Carnegie Research
Silicon materials52%Foundry products
35%
Carbon products13%
Elkem :Silicon/Foundry/Carbon - operating income 2017 NOK 12.2bn
Source: Carnegie Research
Silicon metal37%
Silicon-based alloys33%
Manganese-based alloys30%
Ferroglobe - distribution of volumes
Source: Carnegie Research
c
Elkem
02 May 2018 71
Ferroglobe’s results are generally more volatile than the three combined Elkem’s divisions. When Elkem’s EBITDA margin fell from 19.3% in 2015 to 13.4% in 2016, Ferroglobe’s EBITDA margin (adjusted) went from 22.4% to 4.6%. Its actual EBITDA in 2015 and 2016 were also about USD200m and USD126m below adjusted EBITDA.
What we find interesting now is the large difference between our expected EBITDA growth for Elkem and the corresponding consensus expectations for Ferroglobe. While we expect EBITDA to increase by 32 % in Elkem, consensus is that Ferroglobe will increase EBITDA by 135%. Note also the strong increase in Ferroglobe’s EBITDA in 2017. If Elkem has anything near this quarterly performance, we believe there is big upside to our estimates as the lagging silicon metal and ferrosilicon prices will strongly boost the Q1(18) and Q2(18) results.
Source: Ferroglobe, Elkem, Bloomberg, Carnegie Research
1,200
1,300
1,400
1,500
1,600
1,700
1,800
1,900
2,000
0
20
40
60
80
100
120
140
Q1(16) Q3(16) Q1(17) Q3(17) 1Q(18e) 3Q(18e)
ASP
_ U
SD/t
Adj
uste
d E
BIT
DA
-U
SDm
Ferroglobe - quarterly adjusted EBITDA & ASP
Source: Bloomberg, Carnegie Research
c
Elkem
72 02 May 2018
Other/Eliminations Elkem forecasts normalised EBITDA of NOK-100m in Other. Eliminations totalled NOK-27m in 2017 and we have assumed NOK-30m/year for 2018–20e.
Estimates for each division are shown in the divisional sections before the P&L section. We expect EBITDA(18) of NOK4bn, up from NOK3.15bn in 2017. Our estimates imply margin of 16.9%, marginally above the 1.5–2.0%-point margin improvement Elkem expects. Our EBITDA estimates will increase gradually, mainly due to strong growth in Silicones with more sales of speciality products, but also growth at stable margins in the remaining divisions.
Total volumes (kt) Silicones 127 262 300 318 332 346
Silicon materials 133 203 278 280 282 284
Foundry products 233 255 260 286 292 298
Carbon products 255 255 284 289 295 301
2015: actual Elkem AS. 2016 actual combined. 2017 preliminary combined Source: Carnegie Research
c
Elkem
74 02 May 2018
The other elements in the P&L are:
• Depreciation. With capex around NOK1.3bn, Elkem has guided for depreciation slightly lower than this. We have used this guidance as our estimates
• Elkem will have NIBD of NOK7.9bn prior to the IPO, but after the acquisition of the Chinese companies. Elkem has said it will refinance the debt, and stated that weighted average interest cost will be 2.95%
• Compared to most industry companies, Elkem’s tax rate around 18% is low. The reason is tax-loss carried-forwards in China and France, with no tax in these countries for the foreseeable future.
Balance sheet Following the inclusion of the two Chinese assets, Elkem’s equity ratio goes from 42 % at the end of 2015 to 33% at YE(17) based on the combined balance sheet. The other two significant changes are NIBD – going from NOK2.9bn to NOK7.9bn – and net working capital (NWC) falling from around 25% in 2015 to marginally above 18% at YE(17). The latter is a result of low inventories and account receivables in the Chinese companies. We do not see any dramatic changes. We estimate NWC as a percentage of revenue will increase. As Elkem increases its sale of speciality products in Silicones , we believe it will build inventories closer to customers. As we see the strongest rise in silicon prices in 2018 compared to 2019 and 2020, we see the steepest increase in NWC in 2018. Thereafter NWC should gradually increase but at a slower pace.
We expect higher operating cash flow in 2018 due to stronger results, but we expect cash generation (operating cash flow in % of EBITDA) to decline due to the jump in NWC; we estimate NOK451m in 2017 compared to NOK167m in 2017. With a lower rise in NWC from 2019 and lower net finance costs, we expect cash generation of 76–78% of EBITDA. This gives ample financial flexibility, even after paying out an expected 40% of the net result in dividend from 2019 (based on 2018 result). NIBD was NOK7.9bn at YE(17). After the IPO of NOK5.1bn, NIBD will fall to NOK7bn, which gives NIBD/EBITDA 2.2x. At annual capex of NOK1.3bn over the next two years NIBD/EBITDA should be just above 1.0x at YE(19).
Maintanance CAPEX Growth CAPEX CAPEX in % of revenues
Source: Carnegie Research
0.0%
1.0%
2.0%
3.0%
4.0%
5.0%
6.0%
7.0%
8.0%
9.0%
0
100
200
300
400
500
600
700
800
900
1,000
2015 2016 2017 2018e 2019e 2020e
Depreciation & impairements (NOKm)
Depreciatiuon & impairements Depreciation in % of revenues
Source: Carnegie Research
c
Elkem
76 02 May 2018
Other considerations FX Elkem has operations all over the globe and hence significant FX exposure. Its main currency exposures are EUR, USD, NOK, CAD, ISK, BRL, and CNY.
60%
65%
70%
75%
80%
85%
-
500
1,000
1,500
2,000
2,500
3,000
3,500
4,000
4,500
5,000
2015 2016 2017 2018e 2019e 2020e
EBITDA, operating cash flow & cash generation
EBITDA Operating cash flow Net cash flow Cash conversion
Source: Carnegie Research, Elkem1. External revenue, group figures are on a consolidated basis
c
Elkem
02 May 2018 77
At group level in 2017 about 55% of revenue was in USD or EUR. Some of this positive effect is offset by costs in the same currency, but Elkem generally has a net long position to EUR and USD. Its largest cost position is in Norwegian kroner (NOK), which made up NOK3.8bn in 2017. It is also net short in Canadian dollars (CAD), Icelandic kronas (ISK), and Brazilian Real (BRL). The cost positions reflect where Elkem has its production facilities. In Chinese renminbi Elkem has a fairly balanced position, driven by production facilities in the country and a large end-market. A large part of this is the silicones division, but silicones also have exposure in EUR, USD and a small portion in BRL. Having a balanced currency position is an added competitive strength as highlighted in the carbon division where Elkem has a balanced currency position that leads to less FX volatility.
Based on the countries that Elkem has production facilities in, we have provided a calculated currency exposure table below. This is illustrative and not exact. For instance, Elkem splits geographical sales by continent, so we do not know how much of European sales are in EUR compared to NOK, for example. In the table below we have assumed all European revenue is in EUR. Similarly, the nominal cost of running a plant in France may be less than in Norway, and so our relative cost estimate based on production quantities should be seen only as illustrative.
Elkem’s current practice is to hedge forecast net cash flows for the next 0–3 months on a 90% basis. Thereafter, its forecast net cash flows for the next four 4 to 12 months are hedged 25% to 45%. Elkem’s Board approved a mandate in 2017 to hedge net cash flows in USD up to 75% in 2018. Below is a table showing the hedged positions for 2018. They mean that the FX exposure in 2018 is limited. There are no hedging contracts for 2019. According to Elkem and based on spot rates for YE(17) the company has unrealised losses of NOK96m in EUR and NOK2m in an unrealised gain of NOK37m in JPY.
Currency positions
Long position Short position Balanced position
Euro (EUR) Norwegian kroner (NOK) Chinese renminbi (CNY)
Note: We have excluded Paraguay from the calculation Source: Carnegie Research, Elkem
Silicones Silicon materials Foundry Carbon
FX exposures hedged in 2018
Amount Currency Avg. Rate
91m USD 8.13
281m EUR 9.56
Source: Carnegie Research, Elkem
c
Elkem
78 02 May 2018
Financial targets The table below highlights Elkem’s financial targets.
Q1(18) preview Elkem will release its Q1 report on 8 May. It has not disclosed any quarterly figures for 2017. As we expect improving results from the non-silicon areas over the coming quarters, we believe the Q1(18) will be slightly less than 25% of full-year 2018. For Q1(18) we expect EBITDA of NOK970m and an EPS of NOK0.76.
Elkem's financial targets
2017 Factors driving target 2018 target Medium term targets
● Continued specialisation
● Improved supply/demand balance
● Positive impact from price lag going into 2018 and
continued supportive development in end-market
prices
● Continued efficiency improvements
● Effect of operational improvements at Rana
● Benefits of operational leverage from growth
Reinvestment capex
NOK0.9bn● Reinvestment capex slightly higher than in 2017
Strategic capex
NOK0.4bn● Strategic capex somewhat higher than in 2017
Dep
reci
atio
n
NOK1.2bn ● In line with 2017 ● In line with 2018
Tax 18% ● 18%-19% of earnings before tax
● Slightly decreasing effective tax rate due to deferred
tax assets
Source: Carnegie Research, Elkem
● Current asset base largely sufficient to deliver
revenue growth targets● In line with 2018
● Gradually increasing through specialisation and
continued focus on efficiency improvements
● Targeting an EBITDA margin improvement in the
area of 1.5-2.0% compared to 2017
● Approximately double GDP growth● Approaching double digit revenue growth
Wacker released its Q1(18) numbers on 27 April. EBITDA for the Silicon division of EUR149m totalled 33% of full-year 2017 EBITDA and was the highest ever, both in nominal terms and the margin itself. Our estimate for Elkem Silicones (NOK435m) totals 28.9% of full-year 2017, which implies that we might be a little careful.
Doing the same comparison with Ferroglobe and Elkem ex. Silicones shows that consensus expects Ferroglobe to report EBITDA of USD86m in Q1(18) (46% compared to full-year 2017), while our Elkem ex. Silicon estimate is 32%. In general, Elkem’s results are more stable and less sensitive to changes in market prices due to its high share of speciality products, but if consensus is correct about Ferroglobe we believe that there is upside to our Elkem estimates.
Valuation We value Elkem using comparable peer pricing. We view Wacker (silicones) and Ferroglobe (silicon-related) as Elkem’s closest peers. Taking into consideration average EV/EBITDA and P/E for those two companies we end up with a value per share of NOK44. We apply a 15% discount to this to reflect that Elkem has no earnings history given its ongoing transformation. This gives a target price of NOK37. If we expand the peer universe to include global chemical companies, silicon/steel-related and Nordic capital goods, the valuation upside is greater. However, we find it premature to use the overall sector valuation at the moment.
Peer group valuation The estimates we use for our valuation are shown below. The estimates, including divisional results, are explained in the section ‘Financial forecasts’.
In general, there are three methods in a valuation: a peer group comparison, DCF and comparable company transactions. We have used peer group in our valuation, with a DCF valuation as a sanity check. To our knowledge, there have been few recent M&A transactions relevant as a benchmark. The only incident we know of is the ongoing acquisition of Xinghuo (silicones) and Yongdeng (silicon materials) at an EV/EBITDA(17) of 8.7x and a P/E(17) of 17.4x. As Elkem bought the companies from its owner, it is difficult to use these multiples in our valuation.
Therefore we use a peer group valuation. The table below outlines the broad peer group.
NIBD 7,000 5,607 4,446 3,174 CE 17,328 17,536 17,987 19,636
Source: Carnegie Research
c
Elkem
02 May 2018 81
After the ongoing acquisition, Elkem will broadly be 50/50 silicone/silicon-related. We therefore consider its closest peers to be Wacker in silicones and Ferroglobe in silicon. They (along with others) are broadly described in the section ‘Competitive positioning’. The table below outlines the consensus estimates for Wacker and Ferroglobe.
Based on EV/EBITDA, P/E, EV/EBIT and FCF yield, the chart below gives an implied equity value for Elkem for 2018–20e. One of the challenges using multiples in valuation methods is what weight to use on each multiple and for which year(s).
Looking at EV/EBITDA, this translates into an equity value of NOK 34–36 per share. The corresponding value, based on P/E, gives NOK 54–56 per share. The average, based on 2018e earnings, is thus NOK44. We apply a 15% discount to this to reflect all uncertainties given that today’s Elkem is a new company in the stock market. This gives a target price of NOK37.
Finally, we compare Elkem’s multiples with its broader peer group. Broadly speaking, it screens favourably compared with the total peer group. There is scope for multiple expansion, but we find it premature to use sector multiples on our earnings estimates.
-
10
20
30
40
50
60
EV/EBITDA P/E EV/EBIT FCF Yield
Implied value NOK/share
2018e 2019e 2020e
Source: Carnegie Research
-
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
21/7/16 21/11/16 21/3/17 21/7/17 21/11/17 21/3/18
P/E 12mth forward
WCH GSM
Source: Carnegie Research
4.0
5.0
6.0
7.0
8.0
9.0
10.0
11.0
12.0
21/7/16 21/11/16 21/3/17 21/7/17 21/11/17 21/3/18
EV/EBITDA 12mth forward
GSM WCH
Source: Carnegie Research
c
Elkem
02 May 2018 83
Peer group Chemical Silicon/steel Nordic Elkem Elkem tgt2018 comparison Global related Cap.goods Wacker Ferroglobe NOK31.6 NOK37
Valuation NOKm WACC calculation Terminal period EV 36,008 Risk-free rate 4.00% % of value 51%Net debt -5,461 Debt risk premium 1.00% FCF growth 2.0%Associates 210 Equity risk premium 4.00%Minority -102 Equity beta 1.25Equity Value 30,655 Cost of equity 9.00%Time adjustment -1,719 After-tax cost of deb 3.38%Dividend 0 Weight of equity 80%Current equity value 28,936 WACC 7.88%… per share 50
Source: Carnegie Research
c
Elkem
02 May 2018 85
Elkem’s strategy In the following section we give a brief history of Elkem from its start in 1904 through to its merger with Bluestar Silicones and the acquisition of two Chinese assets in conjuncture with the IPO. Next we outline Elkem’s strategy with focus on the historical improvements that can drive earnings in the two Chinese assets through its lean manufacturing principle that has been at the backbone of the organisation since 1999. Furthermore, Elkem has a history of increased specialisation and this will contribute to growth as the Chinese market matures. Lastly, we highlight the potential for bolt-on acquisition to increase its foothold in China, and also in areas where it does not have production facilities such as North America.
A brief history of Elkem Elkem was founded in 1904 as a result of Sam Eyde’s electrical innovations and the Wallenberg family’s money. Since then it has been one of the cornerstones in Norway’s industrial revolution built on access to cheap electricity similar to Norsk Hydro and Yara. The breakthrough came in 1916 when the company developed the Söderberg electrode, which is still an important component in smelters around the world. It first went public in the 1920-30s.
The 1940–70s were characterised by mergers and acquisition. The start-up of aluminium production in 1958 led to a partnership with Alcoa in 1963. The merger with Christiania Spigerverk in 1972 made Elkem Norway’s largest industrial company. In 1980, it acquired seven ferrosilicon plants in Norway, the US, and Canada and became the world’s largest producer of ferroalloys. In the late nineties it started buying shares in Sapa – a large Swedish downstream aluminium company. At the same time and into the early 2000s Aloca and Orkla fought over Elkem. At the time Orkla was the Norwegian industrial conglomerate. In 2005 Orkla ended up with 100% of Elkem when Alcoa decided to sell its shares. This incidence led to a mandatory bid for Sapa in Sweden, which led to Orkla becoming one of the largest manufacturers of aluminium, both up- and downstream. One of the reasons Orkla was interested in Elkem was its exposure to the solar industry through a 35% ownership in Renewable Energy Corporation (REC) and its development of metallurgical polysilicon (Elkem Solar) for the solar industry. After the Orkla acquisition Elkem was split into several segments. The Aluminium business was sold to Alcoa in 2009. Elkem’s Silicon Material, Carbon, Foundry and Elkem Solar was sold to Bluestar in 2011. Sapa was kept by Orkla until 2017 when it was sold to Norsk Hydro.
Bluestar was founded in China in 1984 by Ren Jianxin. He was a driving force behind the development of ChemChina (China National Chemical Company) as Bluestar took control over at least 100 troubled state-owned chemical factories in China while the Government retained ownership. The Xinghuo plant, which was merged with Elkem in 2017, was established in 1968 and was probably part of this integration. In 2014, the State Council of the People’s Republic of China approved a merger of companies formerly under the Ministry of Chemical Industry as the China National Chemical Corporation (ChemChina), in which Ren Jianxin became CEO before becoming Chairman in 2014. China National Bluestar, which currently own 100% of Elkem, is owned 53.7% by ChemChina, while other financial investors own the rest.
Bluestar had ownership interests in downstream silicone production through its subsidiary Bluestar Silicones. Bluestar Silicones dates from 1928 when the company Rhône-Poulenc was established in France. The Chemical division in Rhône-Poulenc was the start of Rhodia, which was spun-off in 1998 when Rhône-Poulenc merged with the German company Hoechst. Rhodia – which became a large chemical conglomerate – sold its silicone production to Bluestar in 2007. The Silicone business from Rhodia was merged into ‘old’ Elkem in 2015, while the Chinese plants in Xinghuo and Yongdeng were merged into Elkem in 2017, creating the ‘new’ Elkem of today.
Founded in 1904
Long history of M&A
Bluestar Silicones and Elkem merged in 2015
c
Elkem
86 02 May 2018
Elkem has therefore changed from being a pure metallurgical company with focus on upstream silicon material and ferro-alloys to becoming a more specialised chemical and metallurgical company where a large part of the exposure is downstream specialised products and a lesser part is pure commodities. We believe this specialisation process will continue, especially in China.
Vertical integration Elkem is a vertically integrated producer of silicones. An integrated value chain in the production of siloxane creates a natural hedge to fluctuations in the price of raw material inputs and reduces the cyclicality in the business model. At each step Elkem can also make various other products that are not consumed in its own production, such as foundry alloys, microsilica, and other carbon products.
Elkem focuses strongly on lean manufacturing across divisions. At its backbone lies the Elkem Business Systems. This has been a guiding principle since it was established in 1999. When operation efficiencies become ingrained in the organisation, we believe they can add long-term improvements, especially when integrating less efficient plants such as the Rana acquisition illustrates. Increasing operation efficiencies over time may expand margins by 1–2%. Part of Elkem’s EBS road map is increasing yield. As demonstrated in Elkem’s 2016 Sustainability Report, off-grade microsilica has historically been a large contributor to waste as they have been considered useless. Elkem started turning them into briquettes that can be used as input when charging furnaces, which reduces the waste and increases yield. This is demonstrated by Elkem’s plants (we do not know how many use this) improving their waste to landfill from 12% to 5% and increasing the quartz production yield by 5–6%. The drive for lean manufacturing improves Elkem’s environmental profile by reducing waste and increases the yield in production.
Compared to peers, Elkem has a leading cost position. Its two furnaces in Salten and Thamshavn are the second and third most cost effective facilities in the world. Their leading positions come from access to cheap hydro power and Elkem’s drive to cut costs through the EBS principle. This mentality will be a defining factor in the integration of the Chinese facilities.
One of the key factors contributing to the leading cash cost position is Elkem’s access to cheap hydro power. As the chart above right illustrates, Norway has one of the lowest per tonne electricity costs in the world, so the assets in Norway should continue to benefit from a leading cost position. Furthermore, by having in-house supply of electrodes and quartz Elkem has a natural hedge against fluctuations in these inputs.
Ope
ratin
g co
st (U
SD/t)
Cumulative production (kt)
Global major silicon smelters' cost curve 2017*
Source: Elkem, Carnegie Research*Excluding China
2000
1500
1000
500
00 100 200 300 400 500 600 700 800
c
Elkem
02 May 2018 87
Specialisation Perhaps the most important aspect of Elkem’s strategy will be its move towards more specialised products and fewer commodities. This is particularly important in the Chinese silicone operations, but it has also been a push in the other divisions. For example, in foundry products over 80% of production is geared towards speciality products. Market growth rates in these have been higher than in standard ferrosilicon, which has been plagued by falling production of crude steel. Elkem’s specialisation strategy can isolate profitable niches with more favourable demand and supply dynamics and higher entry barriers. We expect that the next step will be to convert the Rana plant from ferrosilicon to silicon metal production.
In terms of specialisation the largest potential is in the Chinese silicone operations. At present, around one-third of silicone production is specialised, while the remainder is commodities. This was the case for Elkem’s European plants, but over time it has shifted production towards more specialised products, and in Europe they represent two-thirds of production now. Therefore one of the key efforts will be to shift production towards two-thirds specialties in China. The market for silicones is growing at twice GDP growth in western countries and has huge potential in China. Currently China consumes only USD2.5 silicones per capita compared to USD11.5 in the US. As China continues the economic expansion, this will enable Elkem to capitalise on its number one position in the region.
0
500
1,000
1,500
2,000
0.0 0.5 1.0 1.5 2.0 2.5
Ex-
plan
t ca
sh c
ost
(US$
/t)
Cumulative production (Mt)
Global cash cost curve for silicon metal, 2017
South China plants (average)
Non-Chinese plants
North China plants(average)
Non-Chinese plants
Source: Carnegie Research, Roskill
0
400
800
1,200
1,600
2,000
Norway Brazil EU US &Canada
ChinaNorth
Russia Australia ChinaSouth
SouthAfrica
USD
/t
Silicon metal ex-plant cash costs by region and component, 2017
Electricity Reductants Quartz Electrodes Labour & other
Source: Carnegie Research, Roskill
0.0
2.0
4.0
6.0
8.0
10.0
12.0
0 10,000 20,000 30,000 40,000 50,000 60,000
Silic
ones
con
sum
ptio
n (U
SD/c
apita
)
GDP per Capita (USD)
Silicones consumption per capita USDm
Source: Freedonia, June 2017
China
USA
c
Elkem
88 02 May 2018
New markets and M&A Elkem recently made huge strides into the Chinese market with the acquisition of Xinghou and Yongdeng, which will enable the company to capitalise on the fastest growing silicone market in the world. It acquired Fesil Rana Metall AS in Norway at the end of 2016, which strengthened its position in speciality ferrosilicon and microsilica, and the potential of additional products. Furthermore, it acquired Minex in India to add to its foundry product presence in Asia. Through the acquisition of the iron foundry business of Minex Metallurgical Co. Ltd. Elkem is getting better access to the lucrative and fast growing Indian market with a local manufacturing footprint, thus achieving a leading positon there. India is considered one of the fastest growing foundry alloy markets.
We think Elkem’s M&A strategy is focused on downstream expansion. To capitalise on growth in silicones in China and the rest of Asia, the most likely targets will be downstream silicone plants. As the upstream part of the organisation is working at close to full utilisation, the largest value adding acquisitions are downstream with capital-light initiatives. According to Elkem the relative capital intensity for downstream and upstream is 1/8 to 1/10.
We do not think that Elkem is planning upstream acquisitions. If an attractive asset comes in play the company could be opportunistic with a bolt-on acquisition in the USD100m range, but the primary driver should be capital-light growth further downstream. Note that Elkem does not have any silicon material production assets in the Americas and this is probably where it would be looking to grow, if it takes the opportunistic route. We do not think Elkem has any plans for acquisitions in its foundry product division, but if the right asset comes up with a profile like Rana we believe it could consider it. One way to reduce the capital intensity would be to adopt the model of Shizuishan and the Minex plant. They source ferrosilicon locally, which means the capital intensity is low and it gives Elkem a stronger footing to supply the Chinese and Indian markets. However, this would only be if it could shift production towards more speciality foundry applications, as the standard ferrosilicon market is burdened with low capacity utilisation.
c
Elkem
02 May 2018 89
Business overview Silicones Elkem Silicones is the downstream division of Elkem that produces speciality products. Silicones are present in many products in everyday life from electronics to baking paper. In 2017 the Silicones division generated NOK10bn in revenue (up from NOK7bn in 2016) and had an EBITDA margin of 15%, implying the division accounts for 44% of group EBITDA and making it by far the largest contributor.
Bluestar Silicones was merged with Elkem in 2015 making it vertically integrated through the whole silicone value chain. Silicon is used as a raw material for making silicones. A very simplified process for producing silicones from silicon can be illustrated in three steps. First, silicon is converted to chlorosilanes by passing chloromethane through heated silicon and then distilling it. Generally, one tonne of silicon and four tonnes of chlorosilanes can produce two tonnes of siloxane. Second, the chlorosilanes are hydrolysed. This is where siloxane is created. Finally, chains of siloxane polymers are formed through condensation, which form different types of silicones when cross-linked. The chemist can affect the physical traits of the silicone (fluids, gels, elastomers, resins) by altering the amount of cross linking between the polymers.
The silicones value chain
Source: Carnegie Research, Elkem
In the process above, the main input is silicon, which Elkem supplies internally. This means it can produce silicon for the silicones division. To produce silicone, silicon qualities of above 99% Si are typically used (for example Elkem products Si99 and Silgrain). It is important to note that Elkem does not sell any of the siloxane to external customers. All of its production is internally consumed and only the silicones are sold. Internal sales from the silicones division are negligible.
4,985
7,619
10,025
9%
5%
15%
0%
2%
4%
6%
8%
10%
12%
14%
16%
0
2,000
4,000
6,000
8,000
10,000
12,000
2015 2016 2017
EBIT
DA
mar
gin
Reve
nue
(NO
Km
)
Silicones
Revenue (NOKm) EBITDA (%)
Source: Carnegie Research, ElkemNote: Figures are incl. Bluestar Silicones from 2016
Silicones44%
Silicon Materials24%
Foundry21%
Carbon8%
Other-3%
Silicones, EBITDA contribution (2017)
Source: Carnegie Research, Elkem
c
Elkem
90 02 May 2018
To produce siloxane, Elkem has two siloxane plants, one in France and one in China, with upstream capacity of about 300,000tpa. With the integration of the Xinhou plant, Elkem’s capacity has risen drastically. After siloxane is produced, it is carried downstream and used to produce silicones in one of seven downstream plants that make speciality and commodity silicones. The downstream plants produce different types of silicones, depending on the structure of the polymers supplied. It is a clear strategy for Elkem to grow in this part of the value chain.
In 2017 Elkem produced 300kt of silicones for sale to end-markets. This is an increase from 127kt in 2015. The reason for the doubling of production is the integration of Xinghuo into Elkem. Furthermore, production increased by 15% from 2016 to 2017 due to the Xinghuo plant ramp up. As of YE(17) Xinhou is operating at near full capacity.
46% of the silicone production in 2017 was for specialised end uses while 54% was for commodity silicones. Elkem has a clear strategy to increase the percentage of this. We do not have historical figures for the development of this figure, but management has indicated that before the integration of Xinghou, it was higher. Hence, there lies a clear potential to increase earnings on the back of increased specialisation in this plant. For comparison, in Europe the speciality mix is two-thirds versus one-third commodity. We expect that more of the volumes coming out of Xinghou will be geared towards reaching a similar ratio.
Silicones
Plant Descprition Current Capacity (ktpa)
Saint-Fons, France Silicones production 140 (intermediates), 60 (silicones)
Santa Perpetua, Spain Heat cured elastomer and silicone fluids plant 15
York, USA Fluids & elastomers 20
Joinville, Brazil Silicone fluids 12
Rousillon, France Siloxane 100
Lubeck, Germany High value elastomers and liquid silicone rubbers plant 1.5
Caronno, Italy Elastomers (RTV, HCR) 5
Shanghai, China Fluids & elastomers 15
Xinghuo, China Siloxane 220 (siloxane), 120 (downstream)
Source: Carnegie Research, Elkem
126,795
262,439
300,440
0
50,000
100,000
150,000
200,000
250,000
300,000
350,000
2015 2016 2017
Volu
me
(mt)
Silicones
Source: Carnegie Research, Elkem
Combination Elkem / Xinghou
Note: Figures are incl. Bluestar Silicones from 2016
Xinghou ramp up
c
Elkem
02 May 2018 91
Specialisation
Source: Carnegie Research, Elkem
Silicones range in price from EUR3/ton to EUR250/ton, which means that gearing production towards high value speciality silicones has great potential. The optimal place to operate for Elkem is in the volume specialities part, as demonstrated in the chart above. This maximises the quantity and price ratio to get the highest total value for each product. An example of this is airbag coatings. Although a lot of airbags are produced annually, the market has not been fully commoditised. Explanations for this could be that there is a large amount of regulation in the production of airbags and new product testing takes many years for approvals to be in place, and switching supplier can also come at great cost if the product does not function properly. For Elkem this means that it can supply a large quantity of airbag coatings at a relatively high price. Elkem expects that the greatest growth in speciality silicones will come from Asia as rapid urbanisation and increased standards of living drive the demand for higher value products. To meet this trend Elkem will clearly be on the lookout for bolt-on acquisitions to complement its current product offering as well as to expand its downstream production capacity in these areas. Elkem’s current product range serves a range of industries such as consumer goods, construction, transportation and energy. The largest end-market is consumer goods, which accounts for 33% of sales, followed by construction (30%) and transportation (14%).
Energy5% Transportation
14%
Construction30%
Consumer goods33%
Health and personal care8%
Electronics5%
Chemicals5%
Silicones sales by end-market1
Source: Carnegie Research, Elkem1External revenue
c
Elkem
92 02 May 2018
Typical products in the consumer goods industry that use silicones are labels, food grade papers (such as baking paper) and kitchen ware. Applications in the construction industry are products such as silicone adhesives, sealants and coatings. Other products are, for example, airbag coatings and lubricants. The silicones are preferred in these applications as they make the products stable, aesthetically and texturally pleasing, easier to use and longer lasting.
Silicon Materials Elkem Silicon Materials produces and sells silicon, silicon metal and microsilica. These products have varying degrees of silicon content, but all are over 95% Si. Annually the division has capacity to produce 295,000 tonnes of silicon metals and 300,000 tonnes of microsilica. In 2017 Elkem generated microsilica sales of NOK1.3bn and around NOK5bn of sales in other silicon metals. Together these accounted for 30% of group revenue in 2017 and 25% of group EBITDA. Since 2016 Elkem’s silicon materials division has grown on the back of two assets entering production (Yongdeng and Rana).
Production Silicon metal is produced through a metallurgical process in which quartz and carbon (coal, charcoal or wood) are heated in a furnace, which reduces the quartz (SiO2) to silicon (Si). Electrode paste (from e.g. the Carbon division) is added at the top of the furnace and heats up and goes into the heart of the furnace, ensuring that the quartz is heated to the right temperature. Silicon metal can be used in the production of extremely pure silicon (polysilicon). Through a leaching process from the standard silicon metal production Elkem can further refine the product to produce its own proprietary Silgrain containing over 99% Si. The silicon metal can also be used to produce silicon for chemical processes, such as in the production of silicones, or in producing aluminium. In the production process ultra-fine (~150nm) spherical particles can be collected as a by-product known as microsilica. This is a by-product that is created when about 15% of the silicon produced in a furnace exits the pipe as dust. Elkem sells this to industries such as construction where it is often used in cement to avoid fracturing. The production process is illustrated below.
Silicon Materials: Production process
Source: Carnegie Research, Elkem
There are a few key parts of this process that are important in terms of Elkem’s strategy. Firstly, Elkem controls the entire production process from quartz to silicone metal and to specialised
4,759 5,269
6,412
19%
11%13%
0%
5%
10%
15%
20%
25%
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
2015 2016 2017
EBIT
DA
mar
gin
Reve
nue
(NO
Km
)
Silicon materials
Revenue (NOKm) EBITDA (NOKm)
Source: Carnegie Research, Elkem
Silicones44%
Silicon Materials24%
Foundry21%
Carbon8%
Other-3%
Silicon Materials, EBITDA contribution (2017)
Source: Carnegie Research, Elkem
c
Elkem
02 May 2018 93
silicones downstream. Electrodes are sourced from the Carbon division and (as seen in the table below), quartz can be sourced from the company’s three mines in Norway and Spain. This captive supply means that Elkem is less prone to price fluctuations in the raw material inputs. For the actual production Elkem operates five silicon material plants with a total of 13 furnaces.
Elkem does not source energy and coal/biocarbon internally. However, it has long-term energy contracts in place, which provide higher visibility on input prices. Many of the plants are fuelled with hydro power. In total, Elkem uses 6–8TWh in its production annually, of which 3.8TWh are in its Norwegian plants. Additionally, Elkem is continually working to reduce its carbon footprint, which means developing sustainable sources of biocarbon for the raw material use. For additional information on Elkem’s carbon footprint see the CSR section. Elkem has a leading cost position in its Thamshavn and Salten production facilities.
In 2017 Elkem produced 278,000tpa silicon materials, about twice that of 2015. The increase was mainly driven by the inclusion of the Yongdeng plant from 2016 and the acquired Rana plant from December 2016. The Rana plant does not produce silicon metal but ferrosilicon and microsilica. Rana is included in the Silicon material division as Elkem is aiming to specialise the production process towards silicon materials. At full capacity the two add 135,000tpa. We do not think Elkem has any plans for more acquisitions in these divisions. If an attractive asset appears, the company could be opportunistic with a bolt on acquisition in the USD1m–100m range; however, the primary driver should be capital-light growth further downstream. We note that Elkem does not have any silicon material production assets in the Americas and this is probably where it would be looking to grow, should it choose the opportunistic route.
Its production of silicon materials is both internally consumed and externally sold. Due to its large production capacity in Europe the company has a net long production. This means that out of the 160,000tpa it produces, 50,000tpa can be consumed by its French plant. In contrast, the Chinese operations are larger downstream than upstream, which means that of the 55,000tpa
Rana, Norway Ferrosilicon and Microsilica plant 2 80
Thamshavn, Norway Silicon and Microsilica plant 2 45
Bremanger, Norway Silgrain and Microsilica plant 1 40
Erimsa, Spain Quartz n.a. 300
Tana, Norway Quartz n.a. 800
Mårnes, Norway Quartz n.a. 150
Yongdeng, China Silicon plant 5 55
Source: Carnegie Research, Elkem
133,473
203,241
278,366
0
50,000
100,000
150,000
200,000
250,000
300,000
2015 2016 2017
Volu
me
(mt)
Silicon materials
Source: Carnegie Research, Elkem
Rana (80ktpa) and Yongdeng (55ktpa)capacity added
c
Elkem
94 02 May 2018
silicon metal the Yongdeng plant can produce, all of it can be consumed by the Xinghou siloxane/silicones plant and the plant will still be short 55,000tpa, which it needs to source from the market.
Elkem aims to increase the share of its production going to specialised processes. At present, 50–60% of the silicon materials portfolio is for specialised end uses. Silgrain is a product extracted from silicon metal by dissolving the metal in liquid and then extracting the product. By doing this the company is effectively transforming a commodity into something with higher value. Silgrain is a product that has been developed and marketed for the production of polysilicon in the solar and electrical industry. In addition, by gearing its products towards more complex downstream applications Elkem can extract a higher value further downstream.
The largest end-market for the Silicon Materials division is the construction industry, which accounts for 32% of sales. A typical application in construction is microsilica, which is used to increase the strength in bridges and high rise buildings. The transportation sector, where silicon is often used in aluminium production to reduce weight and increase strength, accounts for 17% of sales. Energy accounts for the same percentage of sales as Transportation and a major end use is polysilicon in the solar industry.
Foundry Elkem Foundry produces and sells a mix of nodularisers (e.g. ferrosilicon-magnesium), inoculants (ferrosilicon mixed with some other element) and standard/speciality ferrosilicon. Ferrosilicon contains iron and has silicon content of 15–75%. Similar to silicon materials, microsilica is a by-product of the foundry production process. In 2017 Elkem sold foundry products for NOK4.2bn (up from NOK3.7bn in 2015) with an EBITDA margin of 17%. Since 2015 Elkem’s Foundry division has grown on the back of increased specialisation and the Nagpur plant in India entering production (12,000tpa). Together, Foundry accounts for 20% of group revenue and 22% of group EBITDA. The division has capacity to produce 300,000tpa tonnes of foundry alloys and ferrosilicon and serves end-markets such as automotive, engineering and pipes. About 50% of sales are linked to the production of standard and speciality ferrosilicon, with a skew towards speciality. The remainder is nodularisers and inoculants.
Making foundry products is similar to the silicon materials process. Elkem combines quartz, electrodes and coal/biocarbon in a furnace which yields ferrosilicon through an intermediary refining process. It controls the production process from quartz to ferrosilicon and specialised silicones downstream. Electrode paste is sourced from the Carbon division and quartz can be sourced from the company’s three mines in Norway and Spain. This captive supply means that Elkem is less prone to price fluctuations in the raw material inputs. Additionally, Elkem can add alloying elements in the furnace stage or in the refining process to produce foundry alloys.
Foundry products production process
Source: Carnegie Research, Elkem
3,674 3,642
4,247
20%
14%
17%
0%
5%
10%
15%
20%
25%
3,300
3,400
3,500
3,600
3,700
3,800
3,900
4,000
4,100
4,200
4,300
2015 2016 2017
EBIT
DA
mar
gin
Reve
nue
(NO
Km
)
Foundry
Revenue (NOKm) EBITDA (%)
Source: Carnegie Research, Elkem
Silicones44%
Silicon Materials24%
Foundry21%
Carbon8%
Other-3%
Foundry, EBITDA contribution (2017)
Source: Carnegie Research, Elkem
c
Elkem
96 02 May 2018
Ferrosilicon is supplied to the steel and steel related industry. It is generally used for alloying and to remove oxygen from steel (deoxidiser). It is an essential product for a steel producer to improve the steel’s quality (strength and wear resistance, elasticity, etc.). Besides standard grade ferrosilicon, Elkem can make a specialised ferrosilicon that demands higher value for speciality steels such as stainless. This product includes low aluminium, low carbon and high purity (low Titanium) FeSi. All grades are 65–75% FeSi content (silicon metal has 99% FeSi content).
Elkem Foundry Products also produces foundry alloys. These can be split into inoculants and nodularisers. Inoculants are FeSi based alloys used in the production of cast iron. Nodularisers are used to make ductile iron castings. These MgFeSi alloys create nodules that provide ductile iron flexibility. In comparison to inoculants (for instance in grey cast iron), MgFeSi alloys used in ductile cast iron are more resistant to cracking.
Inoculants, nodularisers and speciality ferrosilicon are considered the speciality products in the foundry product range. In line with the strategy, the focus is on specialisation. Unfortunately Elkem does not provide an EBITDA split between standard grade ferrosilicon, speciality ferrosilicon, inoculants and nodularisers. However, in terms of volumes, speciality volumes in foundry products have increased by a CAGR of 17% since 2013, which has increased speciality products’ share of total foundry product volumes from 64% to 81%. It is essentially the speciality FeSi that has driven this increase with a CAGR of 27%. Inoculants and nodularisers have also contributed a solid 10% annually.
Elkem produces ferrosilicon, foundry alloys or microsilica in seven plants globally with 13 furnaces. Ferrosilicon is produced in Iceland, Norway (Bjølefossen) and Canada (Chicoutimi). Inoculants are produced in the Bremanger plant and in Chicoutimi, Canada. Nodularisers are produced in Bjølvefossen, Chicoutimi and Shizuishan. In addition to these, the Rana plant also produces ferrosilicon, but as Elkem plans to upgrade the facility to silicon metal it is placed under the silicon materials division.
55 55 5787
14159 63 67
74
85
35 37 38
43
52
0
50
100
150
200
250
300
2013 2014 2015 2016 2017
mt
'000
Foundry products speciality sales volume 1
Specialty FeSi Nodularizers Inoculants
Source: Elkem1Including Rana
36% 30% 30%22% 19%
64% 70% 70%78% 81%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2013 2014 2015 2016 2017
Foundry products speciality share of total volume1
Standard Specialties
Source: Elkem 1Including Rana
Foundry
Plant Descprition Furnaces Capacity (ktpa)
Grundartangi, Iceland Ferrosilicon and microsilica plant 3 110
Chicoutimi, Canada Foundry alloys, ferrosilicon and microsilica plant 1 45
Limpio, Paraguay Ferrosilicon and foundry alloys 1 11
Bremanger, Norway Foundry alloys, ferrosilicon and microsilica plant 2 30
Bjølvefossen, Norway Foundry alloys, ferrosilicon and microsilica plant 2 65
Shizuishan, China Foundry alloys plant 2 30
Nagpur, India Foundry alloys plant 2 12
Source: Carnegie Research, Elkem
c
Elkem
02 May 2018 97
Three recent developments have boosted production. Shizuishan, China, was started in December 2017 with a 30,000tpa foundry alloys production capacity. Furthermore, the plant in India was acquired in 2016 (Minex acquisition) with a foundry alloys production capacity of 12ktpa. Both of these plants source ferrosilicon locally, which means the capital intensity is low. Furthermore, the plants give Elkem a solid footing to supply the Chinese and Indian markets. Lastly, Limpio, Paraguay is under construction. It has a capacity of 11ktpa and was built to serve the Brazilian market. To go with the specialisation strategy the company has plans to expand the foundry alloy capacity in the medium term.
An important point to make in the production of foundry products is that the cost to the customer is very small relative to all other inputs. However, the product is essential for the end product to function properly. For instance, ferrosilicon is an important addition in the steel production process as it removes oxygen from the steel. Aluminium can do the same job, but the price has consistently been above ferrosilicon and therefore the latter is preferred.
The main markets for foundry products are construction and transportation, which account for 78% of sales. Compared to the other end-markets in Elkem’s division, the exposure to transportation is much larger at 46% (Carbon is the next largest with almost half the exposure). Ferro alloys enable the production of light – yet strong – and temperature resistant castings, which makes it ideal for use in the automotive industry.
232,605 255,163 260,194
0
50,000
100,000
150,000
200,000
250,000
300,000
350,000
400,000
2015 2016 2017
Volu
me
(mt)
Foundry
Source: Carnegie Research, Elkem
Energy9%
Transportation46%
Construction32%
Consumer goods6%
Health and personal care1%
Electronics3%
Chemicals3%
Foundry products sales by end-market1
Source: Carnegie Research, Elkem1External revenue
c
Elkem
98 02 May 2018
Carbon Products Elkem Carbon produces products to be used in the production of silicon metal, ferrosilicon and foundry alloys. Together these generated revenue of NOK1.6bn (up from 1.4bn in 2015) and an EBITDA margin of 17% in 2017. Carbon makes up 8% of the total group revenue and is therefore the smallest of the divisions. However, the division is very important as it secures supply of raw material inputs in the other divisions. 70% of the production by Elkem Carbon is internally consumed.
Electrode paste is essential in the production of Elkem’s ferroalloys, such as ferrosilicon and ferromanganese. The most common type of electrode paste is the Søderberg electrode paste for submerged arc furnaces. Elkem has been producing this for over 90 years. Annually its plants can produce 260kt of Søderberg paste. This paste is used in Elkem’s furnaces to ensure that heat gets transferred to the heart of the furnace. Electrode paste is produced from calcined petroleum coke, calcined pitch coke, calcined anthracite, or coal tar pitch/other green binders. To produce silicon Elkem needs to use pre-baked electrodes due to the importance of purity.
Carbon production process
Source: Carnegie Research, Elkem
Plants producing Søderberg electrode paste are Carboindustrial, Fiskaa, Ferroveld and Shizuishan. The Caboderivados plant also produces a range of oils and pitches. Elkem has a production capacity of 260ktpa Søderberg electrode paste and 105ktpa other products. The Sarawak plant in Malaysia started production in 2016 to meet demand from metal producing industries in Sarawak. The plant contributed an annual capacity of 28kt.
1,388 1,375
1,577 20% 20%
17%
0%
5%
10%
15%
20%
25%
0
200
400
600
800
1,000
1,200
1,400
1,600
1,800
2015 2016 2017
EBIT
DA
mar
gin
Reve
nue
(NO
Km
)
Carbon
Revenue (NOKm) EBITDA (%)
Source: Carnegie Research, Elkem
Silicones44%
Silicon Materials24%
Foundry21%
Carbon8%
Other-3%
Carbon, EBITDA contribution (2017)
Source: Carnegie Research, Elkem
Carbon
Plant Descprition Capacity (ktpa)
Carboderivados, Brazil Coal tar distillation facility 60
Carboindustrial, Brazil Electrode and ramming paste, and pre-baked carbon electrodes plant 70
Fiskaa Kristiansand, Norway Electrode and ramming, paste plant, and other 110
Shizuishan, China Electrode and ramming paste plant 65
Sarawak, Malaysia Electrode paste mixing plant 28.5
Ferroveld, South Africa Electrode paste plant (50% ownership) 60
Source: Carnegie Research, Elkem
c
Elkem
02 May 2018 99
Oils and pitches are produced based on the distillation of coal tar. These include liquid/spheroidal pitches (refractories) used as binding agents for anodes, carbon pastes/electrodes used by the aluminium industry, aromatic oils as feedstock for carbon black production (car and other forms of tires), and naphthalene for the chemical industry.
Other carbon products include cathode blocks, materials for lining silicon, ferrosilicon and silicomanganese furnaces to increase efficiency and life of the furnace. Fiskaa in Norway produces taphole products for use in silicon, ferrosilicon, silicomanganese and ferromanganese furnaces. Ramming pastes are also produced in most of the plants, which are used in furnaces to ensure that the metal and bath does not penetrate the cathode in the furnace. A key difference between a high quality cathode block and e.g. electrode paste is that the cathode block has a useful life of seven years (according to SGL Group), which means it is a capital good rather than a commodity.
Pure carbon and graphite are used in the foundries/steel industries to finalise the final castings in automobile and engineering applications. Due to the close link with foundry products, this is distributed under that division.
255,141 255,485 283,650
0
50,000
100,000
150,000
200,000
250,000
300,000
350,000
400,000
2015 2016 2017
Volu
me
(mt)
Carbon
Source: Carnegie Research, Elkem
Brazil and Malaysia
Energy12%
Transportation25%
Construction36%
Consumer goods13%
Health and personal care2%
Electronics9%
Chemicals3%
Carbon sales by end-market 1
Source: Carnegie Research1External revenue
c
Elkem
100 02 May 2018
R&D Elkem operates in-house R&D centres on a global scale for its Silicones, Silicon Materials, Foundry and Carbon divisions. Over 400 R&D workers are employed (about 7% of total workforce), which enables Elkem to target both new applications with own developed products and to work together with customers for tailor-made products that enhance the customer experience.
Elkem’s silicone division holds over 1,200 patents and each year the company files an additional 20 patents with an average of 10 new products being developed annually. Group R&D totals 2.1% of revenue and in Silicones 3.2%. In 2016, 20% of sales were generated by products that were less than five years old. As such, Elkem’s R&D efforts are important for both staying competitive and driving revenue growth. The back bone of the R&D initiatives is a strong drive towards specialisation, which supports the company’s overall strategy. From here on, focusing on R&D in its Chinese operations will become increasingly important as growth in speciality silicones increases on the back of urbanisation, digitalisation, a sustainability push and other global megatrends.
In its silicon materials division, Elkem has two development centres. A key part of its R&D strategy is to increase the share of specialisation in its products. A good example of this is the Silgrain product line. This is a product that has been developed for high purity/consistency applications. The Silgrain process was developed in the late 1960s and is geared towards applications such as polysilicon production (electronics and solar), aluminium brazing (typically used in heat exchangers in the automotive industry), and speciality silicones (downstream production). This is a differentiated product that contains a higher degree of Si than for instance Si-NINE (another product range) which can be used for refractories.
Furthermore, this product is constantly being refined to meet new areas of demand. As an example, the Silgrain e-Si is a specialisation of the Silgrain product for manufacturing silicon-based anodes for rechargeable lithium-ion batteries. Silicon can hold ten times more lithium ions than the graphite based lithium-ion batteries used today, which makes it ideally suited for use in applications that require low weight batteries, such as electrical cars. This product was developed over several years together with the research centre for battery material development, AIST and launched in late 2014. A commercial application in use today is, for example, the watch maker Breitling’s use of it in its watch batteries.
R&D and other types of competence facilities
Silicones Employees Type
Beijing, China 34 Silicone R&D
Shanghai, China 891Silicones R&D
Lyon, France 135 Silicone competence centre: Technical service group, chemicals lab, process
engineering lab and labs for elastomers, fluids and coating
Silicon Materials
Trondheim, Norway 6 Silicon product development centre: Customise and develop new applications for
speciality products
Kristiansand, Norway 15 Microsilica product development centre
Foundry
Kristiansand, Norway 7 Central laboratory facility: New product testing, join R&D projects with customers
and universities
Carbon
Fiskaa, Norway 9 Main research centre with specialists in physical and chemical analysis of carbon mate
1 Both in silicones plant and R&D centre Source: Carnegie Research, Elkem
c
Elkem
02 May 2018 101
A key part of Elkem’s R&D initiatives is also dedicated to improving Elkem’s corporate social responsibility profile. For instance, Elkem Carbon aims to phase out products that contain polycyclic aromatic hydrocarbons (PAH). Elkem Carbon produces carbon electrode and lining materials. To make these products Elkem uses coal tar pitch as a binder, which contains PAHs. In this process Elkem is developing green binders for carbon products that could one day completely phase out PAH containing products.
Sales and distribution Elkem has over 40 sales offices that can market its products globally. In the upstream part of the business, being vertically integrated enables the company to have a close relationship with customers through know-how of their processes. As it is a user of many of the products it produces (such as electro paste and silicon), it can continually develop these and test them in its own R&D facilities, using its own production to ensure that the best possible product is marketed towards its customers. Globally Elkem serves thousands of customers through its four divisions. Its silicones business has the most fragmented customer base with over 2,000 customers and with the top five constituting only 6% of revenue in 2017.
Contract structures in the silicone division are typically long term, with flexible agreements for shorter durations or single purchase orders. Partnership agreements with customers are common to develop tailor-made products. We believe Elkem will put a particular focus on
1. Silicones includes only top 5 Source: Carnegie Research, Elkem2. External revenue3. 2017 figures
c
Elkem
102 02 May 2018
developing its sales and marketing organisation in China in line with its specialisation strategy. Elkem operates a global sales network in its Silicon Materials division. This ensures that it can meet customer demand and capture long-term customers as well as servicing large customers. Given that 35% of sales in the Silicon Materials division are from the top 10 customers, this is extremely important. With specialised silicon products, such as Silgrain, Elkem can raise the barriers to entry and encourage long customer relationships. Contract structures typically depend on the application. For the chemical and polysilicon segment contracts are typically large and long-term with prices linked to CRU silicon market price. The aluminium segment has more short-term contracts that are linked to the CRU market price index. Other products, such as Microsilica, are annual frame contracts or small order purchases with prices not directly linked to an index but influenced instead by the state of the construction industry.
Elkem’s smallest customer base is in the carbon division, which has 265 customers. This probably reflects that its products are also used internally in silicon and ferrosilicon production. Carbon products have short-term contracts that are renewed every three to 12 months. As Elkem operates its own smelters it is able to give customers of the Carbon division in-depth customer service for the smelting and production process and to advise them on which carbon products would best suit their individual needs. Furthermore, it allows for new inventions and specialised carbon products. Foundry and Carbon together have a strong distribution profile.
Sales in Foundry are based on a continual push towards tailor-made products for customers. The illustration below demonstrates this process. Elkem’s regional sales offices and marketing teams work together to understand customer needs, which can then be implemented in the foundry production facilities. This creates a push-and-pull mechanism between the sales and marketing teams to meet customer demand and new product innovations. Foundry’s customer base is characterised by a large number of customers. Foundry alloy contracts are typically in the 6–12 month range and fixed. Speciality and standard ferrosilicon for the steel industry follow the monthly CRU price based on long-term contracts or quarterly contracts with fixed prices. Standard ferrosilicon is directly linked to reference CRU prices; however, this makes up a small part of the portfolio. Speciality ferrosilicon prices are correlated to the CRU reference price. These two combined make up about 50% of the Foundry division’s sales.
Foundry products sales and marketing flow
Source: Elkem, Carnegie Research
Elkem has a global sales profile with the majority of its sales coming from Europe, Asia and the Americas. The geographical sales split across divisions is not uniform and largely depends on where the company has production capabilities. For the Silicones division, having a close connection to customers through local hubs is important. Elkem can serve European customers
c
Elkem
02 May 2018 103
through its French siloxane plant and customers in China through the Xinghou plant. A key goal for Elkem is to strengthen its marketing and sales network in China and this will probably fit into its downstream expansion strategy. Building up a strong marketing and distribution network in China will be important to meet an expected rise in demand for speciality silicones.
In Silicon Materials, 63% of Elkem’s revenue comes from Europe, 8% from the Americas, 28% from Asia and 1% from Africa. The revenue profile is a little different to the Foundry division as there is less exposure to the Americas because of the lack of assets in the Americas in the Silicon Materials division. We believe Elkem might expand in the Americas if an asset was in play, but that it is primarily aiming to expand downstream operations.
Foundry sales reflect the foundry production capacity. The two plants in Norway produce for the European market. Icelandic production can be placed in both Europe and the Americas. The plant in Chicoutimi produces mainly for the Americas and the new plant in Paraguay will supply the Brazilian market. Chinese and Indian plants supply the Asian markets, but we assume that some of the production from the other markets (probably Europe) also reaches Asia as the production capacity there does not match the sales exposure. The Chinese and Indian plants produce only ferro alloys with a somewhat higher and more stable price than the ferrosilicon plants, which could also explain some of the difference.
30%
0%
13%
58%
0%
63%
1%
8%
28%
0%
49%
0%
25% 25%
0%
19%
9%
43%
27%
3%
0%
10%
20%
30%
40%
50%
60%
70%
Europe Africa Americas Asia RoW
Perc
ent o
f tot
al s
ales
Sales by geography and division1
Silicones Silicon Materials Foundry Carbon
Source: Carnegie Research, Elkem
Europe Africa Americas Asia RoW
1. External revenue
c
Elkem
104 02 May 2018
Corporate social responsibility Elkem has signed the UN Global Compact and its CSR strategy follows the ten UN Global Compact principles.
Elkem’s corporate social responsibility pyramid
Source: Elkem Sustainability Report 2016, Carnegie Research
CO2 emissions and energy consumption Elkem’s CO2 emissions mainly stem from silicon and ferrosilicon plants. As of 2016, the company was working on R&D projects that could potentially make its production carbon neutral. These projects involve the use of biocarbon instead of carbon in the production process and the closure of the top of the furnace to produce CO inside the furnace (which is a product used for industrial and energy purposes) instead of the CO2 emissions that are created today. Elkem’s goal is to increase the use of biocarbon to 40% in all its Norwegian smelters by 2030 and in the longer term to achieve carbon-neutral production of silicon and ferrosilicon. In 2016 Elkem recorded 1,841,000 tonnes of CO2 emissions. Part of the increase Y/Y is due to Elkem acquiring the Rana facility in Norway. Like-for-like emissions increased by 3%.
To compare with peers, we use a ratio of CO2 emissions to sales. The peers we compare with are OSEBX listed Rec Silicon, Norsk Hydro, Yara, and Borregaard. After the inclusion of the Rana plant, Elkem’s CO2 ratio is in line with peers at about 130x. Prior to this the ratio had been steadily falling since 2012.
Health and safety Health and safety is very important for any chemical and metallurgical company wanting to produce top results. For the materials industry, sustainability related incidents typically are related to work place safety as many employees work in plants and facilities where the risk of accidents can be high. For Elkem, the number of lost time injuries per million work hours increased in 2016 to 2.4 from 1.3. Compared to our peer group average, the average is still lower, but it is natural to assume Elkem will want to reduce this. 2017 figures are not yet available. Based on figures from the early 2000s, the likelihood of injuries in Elkem has been reduced by 90%. Elkem’s sick leave is in line with its peers at 3.8% in 2016.
Employees There are over 5,000 employees in the various production and research facilities that Elkem operates. As illustrated below, the Carbon division is the smallest, followed by Foundry and Silicon Materials. Silicone is the largest division driven by about 1,650 employees in the Xinghou plant in China. This means that there is a large number of employees across a wide range of geographies. Adding employees from Xinghou, Yongdeng and Rana in 2016 mean that new challenges are created in terms of integrating the workforce and making sure that all employees follow Elkem’s code of conduct. To ensure common practice and sustainability, Elkem employs its EBS methodology.
131x
170x
128x
43x
188x
132x
0x
50x
100x
150x
200x
250x
Elkem Norsk Hydro Yara Borregaard REC Silicon Peer average
CO2 emissions / sales
2012 2013 2014 2015 2016 2016 average
Source: Carnegie Research
311x
614x
780x
360x
677x
548x
0x
100x
200x
300x
400x
500x
600x
700x
800x
900x
1000x
Elkem Norsk Hydro Yara Borregaard REC Silicon Peer average
Energy consumption (MWh) / sales
2012 2013 2014 2015 2016 2016 average
Source: Carnegie Research
3.9%3.7%
3.4%3.7% 3.8%
4.0%
3.3%
2.9%
3.6%3.9%
0.0%
0.5%
1.0%
1.5%
2.0%
2.5%
3.0%
3.5%
4.0%
4.5%
2012 2013 2014 2015 2016
Health: Sick leave (%)
Elkem Peer average
Source: Carnegie Research, Elkem, Company reports
1.8x
0.8x
2.1x
1.3x
2.4x
1.9x
2.5x
2.9x
3.5x
2.7x
0.0x
0.5x
1.0x
1.5x
2.0x
2.5x
3.0x
3.5x
4.0x
2012 2013 2014 2015 2016
Health: Accidents with sick-leave (number / million hours worked)
Elkem Peer average
Source: Carnegie Research, Elkem, Company reports
c
Elkem
106 02 May 2018
Historically, Elkem’s industry has been very male dominated due to the exposure to heavy manual labour. However, as Elkem notes in its 2016 sustainability report, plants and processes are increasingly automated, making them more attractive for female workers. In 2016, the percentage of females in Elkem’s organisation was 22%. 35% of female employees are white-collar workers while 6% are blue-collar. In Elkem’s trainee and leadership programmes the company states that the female share is around 50% or above for many years. Compared to other companies on the OSEBX, which are arranged by sector in the chart below left, Elkem ranks in the middle of the pack. We note that Elkem ranks above other materials peers. Elkem also has a higher percentage of women in the total workforce than the broader peer group, as illustrated below right.
Source: Carnegie Research, Elkem(1) Excluding employees not at a specific plant
1,649
385 300143 89 67 46 45 34 11 9
0200400600800
1,0001,2001,4001,6001,800
Xin
ghou
, Chi
na
Sain
t-Fo
ns
Lyon
, Fra
nce
Rous
illio
n
Shan
ghai
, Chi
na
York
, USA
Car
onno
, Ita
ly
Sant
a Pe
rpet
ua
Beiji
ng, C
hina
Lube
ck, G
erm
any
Join
ville
, Bra
zil
# of
em
ploy
ees
Employees in the silicones division
Source: Carnegie Research, Elkem
522
188155
103 101
15 60
100
200
300
400
500
600Yo
ngde
ng
Salte
n
Tham
shav
n
Rana
Brem
ange
r
Kri
stia
nsan
d
Tron
dhei
m
# of
em
ploy
ees
Employees in the silicon material division
Source: Carnegie Research, Elkem
160 155
94 89 8472
61
7
020406080
100120140160180
Icel
and
Bjøl
vefo
ssen
Brem
ange
r
Chi
cout
imi
Nag
pur
Shiz
uish
an
Para
guay
Kri
stia
nsan
d
# of
em
ploy
ees
Employees in the foundry division
Source: Carnegie Research, Elkem
109 104
8164
53
229
0
20
40
60
80
100
120
Shiz
uish
an
Car
boin
dust
rial
Fisk
aa
Ferr
ovel
d
Car
bode
riva
dos
Sara
wak
Fisk
aa r
esea
rch
cent
er
# of
em
ploy
ees
Employees in the carbon division
Source: Carnegie Research, Elkem
c
Elkem
02 May 2018 107
Seafood
Shipping
Tech & Telecom
Health care
Materials
Real estateFinanceConsumer goods
Oil serviceCapital goods
Oil
Utilities
Elkem
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
0% 5% 10% 15% 20% 25% 30%
Shar
e of
wom
en in
ove
rall
com
pany
Share of women in top management
Share of women in top management vs overall company, 2016
Source: Company reports, Carnegie Research
19%21% 21%
22% 22%21%
18% 18% 18% 19%
0%
5%
10%
15%
20%
25%
2012 2013 2014 2015 2016
Total company, percent women
Elkem Peer average
Source: Carnegie Research, Elkem, Company reports
c
Elkem
108 02 May 2018
Competitive position Global supply in Elkem’s core operations is fairly concentrated. Its main competitors are shown in the figure below and we will give a short description of some of these in the following section.
Silicones In 2016, there were five dominant players of silicones, which together accounted for 74% of the market. Dow Corning is the largest of the five, holding about 27%, while Elkem holds 11% and has a leading position in China and a top three position globally in terms of production capacity. In terms of silicone sales, a study by Freedonia Group indicated that Elkem held a top five position and 4% market share in 2016, but this was prior to the acquisition of the Chinese plant.
Silicones Silicon Materials Foundry
Source: Carnegie Research
Dow Corning27%
Wacker Chemie12%
Elkem11%
Momentive9%
Shin-Etsu Chemcial8%
Other33%
Global silicones production capacity (2016)
Source: IHS 2017, Carnegie Research
c
Elkem
02 May 2018 109
Dow Corning was established in 1943 with the specific aim to explore and develop the potential of silicones and is a US-based company headquartered in Michigan. It operates in more than 25 countries with major operations in Belgium, China, Japan, the UK, Brazil, Germany, Korea and the US. Historically Dow Corning was a 73-year old joint venture between Dow Chemicals and Corning, but in June 2016 Dow Chemicals became 100% owner in a transaction that seems to have been done at around an EV/EBITDA multiple of 9x before synergies.
Wacker is a German chemical company divided into four divisions: Silicones, Polysilicones, Polymers and Biosolutions. The Silicones division was founded in 1947 and represents nearly 45% of the company’s total sales with an 18.1% EBITDA margin in 2016e and a target of 20% in 2017e at high-single digit revenue growth. Both Wacker and Dow Corning are represented in China through a joint venture, which opened up in 2010 with a capacity of 210,000 metric tons.
Momentive is a US-based company with headquarter in Waterford, New York, and is, besides one of the leading silicone manufacturers, a global leader in the development and manufacture of products derived from quartz and speciality ceramics. The company was formerly known as GE Silicones and was formed in 2006. Silicones represent around 90% of sales and around 94% of EBITDA. In 2016 the company achieved an EBITDA margin of 10.8% but this improved to 12.1% in the first nine months of 2017e (last known report).
ShinEtsu is a Japanese company founded in 1926 with annual sales of YEN1,237bn, of which silicones represent 15%. The other divisions are PVC/Chlor-Alkali, which accounts for 33% of revenue, Speciality Chemical, which accounts for 9%, and Semiconductors, which accounts for 20%. ShinEtsu began basic research in silicones back in 1949 and is present in all major markets. Shinetsu’s silicones business enjoys solid EBITDA margins coming in at 24.5% in 1H(17) – an increase of 4%-points Y/Y. In the past seven years, ShinEtsu’s silicone business has booked 5% annual sales growth with an EBITDA margin on average at 22%.
Evonik Industries is the smallest of the top competitors, and is not among the top five producers. However, as it is in the top six, it is worth mentioning. Evonik’s key region is in Europe, and it is the only major company that is not backward integrated with siloxane production. It specialises in organically modified silicones, and it aims to expand production of these types in the Chinese market. In Q1(16) construction of a silicone plant in Shanghai commenced, and production began in Q3(17).
Silicon materials Supply of silicon material outside of China is even more consolidated than the silicones market with five suppliers controlling 71% of the market. Ferroglobe is the largest with 29% of global production of silicon material, followed by Dow Corning and Elkem in third place. The graph below shows the market shares when excluding the Chinese market.
Silicone markets company overview
Company Market Share (cap) Key region
Dow Corning 27% Global
Wacker Chemie 12% W. Europe
Elkem 11% W. Europe/APAC
Momentive 9% N. America
Shin-Etsu 8% APAC
Evonik 4% W. Europe
Source: Companies, Freedonia, Carnegie Research
c
Elkem
110 02 May 2018
China is the largest producing market with 1,690 metric tonnes of total capacity. The graph below shows total production of silicon metals separated into regions. Hence, when looking at the market including China, supply becomes considerably more fragmented.
Ferroglobe is the result of the merger between FerroAtlàntica and Globe Speciality Metals in 2016. These two companies were both the results of several mergers and acquisitions during the past twenty years. Ferroglobe is one of the largest producers of a wide variety of metal alloys and other metallurgical products and hence the largest silicon materials producer outside the Chinese market. Total capacity in the company is 1.464m metric tons divided between Silicon Metal (473,000 mt), Silicon based alloys (FeSi) (472,000 mt), Silicon-Manganese (215,000 mt), Ferro-Manganese (206,000 mt) and Silica Fumes (99,000 mt). The company is low on the cost curve in silicon metal, but not below Elkem.
29%
16% 15%
6% 5%
29%
0%
5%
10%
15%
20%
25%
30%
35%
Ferroglobe Dow Elkem Rusal ShinEtsu Other
Global silicon production ex. China, 2016 (1)
Source: Carnegie Research, CRU 20171Excluding China as Chinese market is highly fragmented, includes mostly domestic focused producers
0
200
400
600
800
1000
1200
1400
1600
1800
China EU North America Production Australia
'000
ton
s
Worldwide production incl. China
Series1
Source: CRU, Global Trade Atlas, Carnegie Research
c
Elkem
02 May 2018 111
Dow Corning (see above) also has 207,000 mt capacity of silicon metal and 120,000mt capacity of Silica Fumes. Dow Corning appears a more specialised silicone company where we believe excess capacity of silicon is sold in the market, but unfortunately there is limited information on Dow Corning given the previous ownership structure.
Rusal is a Russian company primarily engaged in the production of aluminium and alumina. The company was founded in 2000 and became the global industry leader in 2007 after merging with SUAL and the alumina assets of Glencore. It has two silicon factories located in Russia with about 70,000 tonnes in capacity. The primary product is metallurgical silicon, but the company is working to master the production of refined silicon.
ShinEtsu – as Dow Corning – is primarily a silicone industry company but is mentioned as one of the larger silicon material companies and has a large semiconductor silicone division. As such the company does not seem to compete head-to-head with Elkem in the silicon metal industry.
Foundry Ferrosilicon can be split into applications for the steel and the foundry industries. For the steel industry standard ferrosilicon (typically for deoxidation) or speciality ferrosilicon is used in the production of special steels such as stainless. Foundry applications typically use magnesium ferrosilicn (nodulisers) in cast and ductile iron and inoculants that can be added to cast iron to improve its properties. The barriers to substitution between the products vary depending on the product. As the EU merger procedure from 2011 noted (in conjunction with the Bluestar/Elkem transaction) participants can switch at a relatively small cost between production processes of products for these applications by adjusting what materials are put into the furnace. The report further notes that there are probably some larger switching costs between steel and foundry applications and further costs in the sub segments; for example, in the production of magnesium ferrosilicon a producer would need to invest in the magnesium process of production. Hence, it is not a case of perfect substitution and the more specialised processes create larger barriers.
The global foundry market (defined by Elkem’s own product mix) is less analysed than the silicone and silicon materials market, but based on Elkem’s knowledge of its own production and the total size of the market, the following graph should act as a fair proxy of the competitive landscape. Note that this chart does not include ferrosilicon.
Foundry alloys such as inoculants (FeSiMg) are primarily used in cast iron foundry.
Elkem23%
Ferroglobe12%
Osaka Steel8%Toyo Metallurgicals
6%Zhufeng Alloy
5%
Others46%
Foundry - global market shares
Source: Elkem, Carnegie Research
c
Elkem
112 02 May 2018
If we include ferrosilicon then the picture changes as demonstrated below. Ferrosilicon is an industry that has been plagued with over capacity coming out of China and less demand from steel over the past couple of years. Therefore, it is positive to see that Elkem has positioned itself in the more profitable market niches by focusing on specialised products.
We do not have a revenue split based on what percentage of revenue stems from standard/speciality ferrosilicon and inoculants. However, we know that only the Icelandic and Rana plants produce ferrosilicon and we would assume that production is skewed towards speciality ferrosilicon. Ferrosilicon is primarily used in the production of steel.
To make most types of steel producers need both manganese and silicon. A producer can generally choose between using only one manganese based alloy, silicomanganese (SiMn), or ferrosilicon (FeSi) mixed with ferromanganese (FeMn). The chart below left illustrates the relative use of these in steel making.
Ferroglobe produces both of these products so a key question from a competitive point of view is which product is it better to be a producer of. Firstly, as the chart above left illustrates, since 2000 the general trend has been to use less FeSi and more SiMn; however FeSi is still being used
Silicon units contained in global steel output by source, 2000-16
FeSi SiMn
Source: Carnegie Research, Roskill
0
500
1,000
1,500
2,000
2,500
3,000
3,500
Jan-
00
Jan-
01
Jan-
02
Jan-
03
Jan-
04
Jan-
05
Jan-
06
Jan-
07
Jan-
08
Jan-
09
Jan-
10
Jan-
11
Jan-
12
Jan-
13
Jan-
14
Jan-
15
Jan-
16
Jan-
17
US$
/t
Price of SiMn versus equivalent quantity of FeSi and HC FeMn
SiMn FeSi + HC FeMn
Source: Carnegie Research, Roskill
c
Elkem
02 May 2018 113
more. The question most likely comes down to price. In the chart above right, Roskill compares the relative price of SiMn with FeSi and FeMn combined. The combined quantities are slightly more expensive over time than the SiMn substitute in the European market, which would make it relatively more attractive for a steel producer. We think there is clearly a case for the SiMn production, as the chart above left demonstrates. However, the prices also largely track each other, which would make them equivalent to a producer of the silicon/manganese alloy, and FeSi still makes up the largest portion of consumption.
Ferroglobe has been described above in the silicon metals section. It has production of ferrosilicon (472,000 mt), manganese based alloys and silicon-manganese (215,000 mt), and ferro-manganese (206,000 mt).
Osaka Steel is primarily a steel company located in Japan. There is limited information on the company’s sales and profitability on foundry products, but we assume that the excess capacity is sold in the merchant market
Toyo Metallurgical Limited was established in 1963 and is a manufacturer of the whole range of metallurgical chemicals for the ferrous and non-ferrous foundry and refining industry. The manufacturing facilities in Toyo are spread in and around Mumbai.
Carbon In the production of electrode paste (ex. China), Elkem has a leading position, holding almost 43% of the market share. The industry is consolidated with the top five producers holding 70% of the market share.
China regulations are leading to an electrode shortage taking place currently, according to Ferroglobe’s investor presentation from late 2017. Many of the producers of carbon electrodes also produce graphite electrodes. These typically serve electric arc furnaces for steel making. The graphite market has been affected by an imbalance in supply and demand due to declining production of electric arc furnace steel, which has led to price pressure according to Tokai Carbon and SQL. Players such as SQL have been loss making in this area of its business due to this. The chart below illustrates the challenging environment these players have been in. In 2012–16, ca. 200ktpa (20% of production) of graphite electrode was shut down.
205
476
129
143
43%
27%
30% 100%
0
100
200
300
400
500
Elkem Top 2-5 Other Total
Tonn
es p
er a
nnum
(100
0's)
Global electrode paste production ex. China, 2016 (1,2)
Source: Carnegie Research, Elkem1Excluding China as Chinese market is highly fragmented, includes mostly domestic focused producers
c
Elkem
114 02 May 2018
SQL Group worked from 2015 until late 2017 to restructure its operations. It is no longer a direct peer to Elkem’s carbon division as it has chosen to focus on carbon fibres and graphite materials while disposing of its performance products graphite and carbon electrodes, cathodes, and furnace linings units. Triton acquired its cathodes, furnace lining, and carbon electrode interests in late 2017 for an EV of EUR250m. What multiple the transaction implies is difficult to deduct as these businesses were part of a larger division and so we are not able to isolate the results.
Yangguang Carbon is China’s largest producer of electrode paste and has supplied about 90% of ore-smelting furnaces (above 25,500kva) in China, according to the company. The company has a production capacity of about 1,000ktpa which is primarily supplied in China, but exported to Myanmar, Indonesia, Iran, Kazakhstan, South Kora, Vietnam and other countries.
262 255
105 9880
100
30 30
162 160139
9880
64
30 30
0
50
100
150
200
250
300
GrafTech SGL Carbon ShowaDenko
GraphiteIndia
HEG TokaiCarbon
NipponCarbon
SEC
1000
's to
nnes
pro
duct
ion
Shutdowns in graphite electrodes production
2012 2016
Source: Carnegie Research, Graphite India Limited
c
Elkem
02 May 2018 115
Risk factors Environmental risk Elkem’s operations may pose a risk to the environment and employees. The everyday business operations at Elkem involve heavy processing and handling of molten metals, high voltage equipment and chemical substances. The environment at Elkem could therefore potentially inflict injuries on the employees or contractors. Additionally, risk is present in relation to plant and equipment impairment. Insurance policies are in place covering the impact on the environment, property and general business. Regardless, Elkem has shown impressive improvements in health and safety measures, such as injury frequency and sick leave. Also, the company has a strong commitment to making further improvements, including targeted plans, systematic methods and an involved management.
Financial risk Financial risk most prominently relates to currency fluctuations and in Elkem’s case the EUR and USD against the NOK present the highest risk, despite 41% of sales taking place in Asia. The reason is that the majority of Asian sales are to China with a fairly balanced RMB exposure since costs are in the same currency. From a divisional perspective, 57% of Silicones’ revenue comes from Asia (China), with 30% from Europe and 13% from the Americas. Silicon Material makes 63% of its sales to Europe, 8% to the Americas and only 28% to Asia.
Elkem has a policy to hedge 90% of net cash flows occurring within three months. Net cash flow in the following 4–12 months is usually hedged on average at 45%, but this can vary between 25% and 75%.
Operational risk Elkem operates in the field of highly specialised silicon materials and advanced chemical processes for silicon production, which calls for considerable production precision and expertise to meet the end users’ expectations. Consequently, a secure, consistent and stable working environment must be ensured. Elkem Business Systems is the corporate culture foundation the group uses to achieve its goals. The capabilities in the R&D division play a substantial role in process improvement.
Europe57%
Africa1%
America19%
Asia23%
Revenue split in 2017, by geography 1
Source: Carnegie Research, Elkem(1) Group revenues on a consolidated basis
c
Elkem
116 02 May 2018
Price risk Elkem’s products, silicon, silicones, ferrosilicon and carbon related materials, are exposed to fluctuations in the supply and demand balance in the market. The prices the company achieves may fluctuate in relation to the international production capacity. The demand will follow the economic cycles as Elkem’s product offerings are in general terms used in other industries. Through significant R&D and technology capabilities, Elkem offers speciality products with less exposure to general cyclicality. Furthermore, the group’s long-term and strong customer relationships create possibilities in tailed product offerings.
Raw material risk The furnace operations at Elkem give leeway for increased raw material risk due to the high electrical energy requirement. In addition to energy, quartz, biocarbon, coal and other strategic raw materials are needed in Elkem’s operations. Quartz, however, is sourced from wholly owned mines. The base volumes as well as coal and other strategic materials are secured on long-term contracts. Flexibility in the operations is necessary to achieve the optimal outcome and as such some energy volumes are placed on short-term contracts.
Regulatory risk With operations all over the world, Elkem may be exposed to regulatory changes in the area of trade and environmental measures. The group addresses the risk by distributing its operations between emerging and developed markets as developed markets usually present a more stable business environment. Likewise, Elkem’s raw material sourcing is located in different geographical regions. The group has low capex investments in emerging markets, generally organic growth and sources raw materials form local and differing markets. Elkem is involved in associations that encourage a fair, sustainable, yet competitive industry.
Strategic risk The strategic risk is mainly related to Elkem’s inorganic growth scenarios, but also to normal business operations. Naturally, not all investments and strategic targets end up being a good fit. However, in order to retain its market share or grow into new markets strategic risk arises. Elkem has cross functional teams placed throughout all its business segments and considerable verification stages to mitigate strategic failure. Also, a project steering committee is assigned to monitor and look out for any deviations. Specific strategic risks are the downside potential of an undershooting in costs related to acquisitions, fewer synergies from acquisitions than initially expected and changed market conditions.
c
Elkem
02 May 2018 117
Appendix Management Helge Aasen, CEO Helge Aasen has been with Elkem for 25 years and has been the chief executive officer at Elkem since 2009. He has international and national experience from various fields in the industry. Before being appointed CEO of Elkem he was the CEO of Elkem Solar AS and prior to that division director in Elkem Carbon AS from 2000 to 2007. He holds a Master of Science in engineering from the Norwegian University of Science and Technology (NTNU) in Norway. In addition, Mr Aasen has participated in IMD’s executive leadership development programme in Switzerland.
Morten Viga, CFO Morten Viga came to Elkem in 2001 and was appointed the chief financial officer in 2006. Prior to this, Mr Viga was the financial director of the Silicon Materials Division and has 29 years of international experience from a range of industries, such as materials, metals & mining and oil & gas. He holds an MSc in Economics and Business Administration from the Norwegian School of Economics (NHH).
Frédéric Jacquin, Senior Vice President of Silicones Mr Jacquin joined Elkem in 2004 and has 28 years of industry experience in speciality chemicals. Frédéric Jacquin was appointed the SVP of Silicones in 2015. Mr Jacquin obtained a master of business administration from the ESSEC and a Magistere in marketing and communication from La Sorbonne French University (Celsa).
Trond Sæterstad, Senior Vice President of Silicon Materials Trond Sæterstad brings extensive experience to Elkem with over 31 years in the metals & mining industry. Mr Sæterstad first joined Elkem in 1997. Prior to being appointed the SVP of Silicones Materials, he was plant manager for both silicon and ferroalloy. Trond Sæterstad holds an MSc in chemical engineering from the Norwegian University of Science and Technology (NTNU) in Norway, as well as a business economics degree from the Norwegian Business School, BI. Furthermore, he participated in the leadership development programme in IMD, Switzerland.
Jean Villeneuve, Senior Vice President of Foundry Mr Villeneuve has been with Elkem since 1979 and was appointed senior vice president of Foundry in 2011. Prior to that, he was the plant manager for Elkem Chicoutimi. Jean Villeneuve holds a Bachelor’s degree from the University of Chicoutimi, Quebec, Canada, in electrical option engineering. In addition, he has been part of the leadership programme at the Center for Creative Leadership.
Asbjørn Søvik, Senior Vice President of Carbon Mr Søvik came to Elkem in 1995 and was appointed the SVP of Carbon in 2007. He brings 25 years’ worth of industry experience to management. Prior to the vice president position, Mr Søvik was at plants in the US, Brazil and Norway. He holds a Master of Science in engineering from the Norwegian University of Science and Technology (NTNU) in Norway and an MBA from IESE, Spain.
Inge Grubben-Strømnes, Senior Vice President of Business Development Mr Grubben-Strømnes came to Elkem in 2005 and has worked in the strategy and business development division. Before his time in Elkem, he spent four years at McKinsey & Company. He holds a Master of Science in Zoology and a Bachelor in Biology from the Norwegian University of Science and Technology (NTNU), as well as an MBA from EM-Lyon. Mr Grubben-Strømnes has leadership experience from his CEO position at Elkem Solar AS between 2012 and 2016 and was appointed the SVP of Business development in 2008.
c
Elkem
118 02 May 2018
Dr Louis Vovelle, Senior Vice President Innovation and R&D Dr Vovelle was appointed the R&D Senior Vice President at Elkem in 2015. From 2007 to 2010, he held the R&D director position at Elkem Silicones International. Later, Dr Vovelle was the vice president of Elkem Silicones International’s innovation and strategy division. Thus, Dr Vovelle has extensive international R&D and leadership experience. He has a degree in chemical engineering from the Ecole Nationale Supérieure de Chimie de Paris, as well as a PhD from Ecole Centrale de Lyon in Physics and was a participant in the general management programme at CEDEP-INSEAD in France.
Håvard Moe, Senior Vice President of Technology Håvard Moe was appointed the Senior Vice President of Technology in 2008. Mr Moe has been part of all of Elkem’s complex industrial projects since 2008, such as Elkem Solar, two furnace upgrade projects in Salten and two energy recovery projects. Prior to joining Elkem, he gained valuable management experience from ABB and Statoil. Mr Moe has an MSc in Mechanical Engineering and a PhD in Chemical Engineering from the Norwegian University of Science and Technology (NTNU). He has also participated in the transition to general management programme at INSEAD, France.
Katja Lehland, Senior Vice President of Human Resources Katja Lehland joined Elkem in 2006 after being the HR Director at Nokia, United Biscuits and Schindler. She also brings extensive board experience to the Elkem management and has completed high-level executive management development programmes. Mrs Lehland holds a Bachelor of Science in Economics and Marketing.
Board of directors Michael Koenig, Chairman Michael Koenig has been the chairman of Elkem AS since 2016 and has been with Bluestar since 2006. He is currently the CEO of China National Bluestar (Group) Co., Ltd. With 27 years in the international chemical industry as board member of Bayer, head of the Polycarbonates Business unit of Bayer MaterialScience and the CEO of Bayer Group Greater China, Mr Koenig brings extensive industry and board experience to the table.
Helge Aasen, CEO See above
Sverre T. Tysland, Director Mr Tysland works at the attorney office Steenstrup Stordrange DA.
Olivier de Clermont-Tonnerre, Director Mr Olivier de Clermont-Tonnerre is Chief Strategy and Corporate Development Officer at Bluestar representative of Bluestar at the Board.
c
Elkem
02 May 2018 119
Marianne Færøyvik, Director / Employee representative
Einar Støfringshaug, Director / Employee representative
Youngen Ge, Director Mr Youngen Ge is a board member and representative from Bluestar.
Dr Zhigang Hao, Director Dr Zhigang Hao is a board member and representative from Bluestar.
Sales 0 0 0 0 14,541 14,226 16,658 23,503 24,981 26,446COGS na na na na -6,847 -6,899 -8,126 -11,751 -12,428 -13,091Gross profit 0 0 0 0 7,694 7,327 8,532 11,751 12,553 13,355Other income & costs 0 0 0 0 -5,487 -5,709 -6,434 -7,772 -8,258 -8,778EBITDA 0 0 0 0 2,207 1,618 2,098 3,979 4,295 4,577Depreciation PPE 0 0 0 0 -674 -718 -776 -1,250 -1,250 -1,250Other amortisation 0 0 0 0 -221 52 49 0 0 0EBITA 0 0 0 0 1,312 952 1,371 2,729 3,045 3,327GW amortisation & Impairment 0 0 0 0 -2 -12 -17 0 0 0EBIT 0 0 0 0 1,310 941 1,355 2,729 3,045 3,327Net interest 0 0 0 0 33 50 -8 -407 -246 -216Other financial items 0 0 0 0 0 0 0 0 0 0Share of earnings in ass. comp. 0 0 0 0 21 22 34 43 46 49EAFI 0 0 0 0 1,364 1,013 1,381 2,365 2,845 3,160Other EO items 0 0 0 0 -7 0 0 0 0 0Pre-tax profit 0 0 0 0 1,267 947 1,281 2,365 2,845 3,160Taxes na na na na -425 -190 -269 -438 -519 -569Post-tax minorities interest 0 0 0 0 -33 -36 -39 0 0 0Net profit na na na na 809 721 973 1,928 2,326 2,592EO & Impairment adjustments na na na na 9 12 17 0 0 0Tax on EO items 0 0 0 0 0 0 0 0 0 0Adj.Net profit na na na na 818 733 990 1,928 2,326 2,592
Sales growth Y/Y na na na na +chg -2.2% 17.1% 41.1% 6.3% 5.9%EBITA growth Y/Y na na na na +chg -27.4% 44.0% 99.1% 11.6% 9.3%EBITDA margin nm nm nm nm 15.2% 11.4% 12.6% 16.9% 17.2% 17.3%EBITA margin nm nm nm nm 9.0% 6.7% 8.2% 11.6% 12.2% 12.6%Tax rate na na na na 33.5% 20.1% 21.0% 18.5% 18.3% 18.0%
Goodwill 0 0 0 0 244 343 326 326 326 326Other fixed intangible assets 0 0 0 0 644 892 911 911 911 911PPE 0 0 0 0 5,602 11,410 11,950 12,000 12,050 12,100Shares & participations na na na na 163 210 210 210 210 210Other fixed financial assets 0 0 0 0 324 67 90 90 90 90Other fixed assets 0 0 0 0 258 512 508 518 529 539Fixed assets na na na na 7,235 13,434 13,995 14,055 14,116 14,176Inventories na na na na 3,302 3,792 4,099 4,536 4,846 5,157Receivables na na na na 1,864 1,952 2,518 2,797 2,998 3,200Other current assets 0 0 0 0 770 2,594 3,144 3,207 3,271 3,336Cash & cash equivalents na na na na 1,306 1,320 1,751 3,044 3,205 3,477Current assets na na na na 7,242 9,658 11,512 13,584 14,320 15,171Total assets na na na na 14,477 23,092 25,507 27,639 28,435 29,347
Shareholders' equity na na na na 6,044 5,743 8,463 9,620 11,015 13,607Minorities 0 0 0 0 123 88 102 145 191 241Sub-ordinated loans 0 0 0 0 0 0 0 0 0 0Convertibles 0 0 0 0 0 0 0 0 0 0Deferred tax na na na na 124 114 105 105 105 105Other IB & Non IB provisions 0 0 0 0 262 506 426 435 443 452LT IB debt 0 0 0 0 3,052 5,113 4,585 3,835 3,085 2,335LT non-IB liabilities 0 0 0 0 1,319 986 824 823 874 926LT liabilities na na na na 4,757 6,719 5,940 5,197 4,508 3,818ST IB debt na na na na 328 4,204 3,647 3,397 3,147 2,897Payables na na na na 1,449 2,311 2,650 2,915 3,098 3,280Other ST non-IB liabilities 0 0 0 0 1,774 4,027 4,706 6,366 6,476 5,505Current liabilities na na na na 3,551 10,542 11,003 12,677 12,721 11,682Total liabilities na na na na 14,475 23,092 25,508 27,639 28,435 29,347
Source: Carnegie Research
c
Elkem
122 02 May 2018
Share data & key ratios
Per share data (NOK) 2011 2012 2013 2014 2015 2016 2017 2018e 2019e 2020e
Adj. no. of shares in issue YE (m) 0.00 0.00 0.00 0.00 0.00 0.00 0.00 581.3 581.3 581.3Diluted no. of Shares YE (m) 0.00 0.00 0.00 0.00 0.00 0.00 0.00 581.3 581.3 581.3EPS na na na na na na na 3.32 4.00 4.46EPS adj. na na na na na na na 3.32 4.00 4.46
CEPS na na na na na na na 5.39 6.07 6.52DPS 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.33 1.60 1.78BVPS na na na na na na na 16.5 18.9 23.4
BVPS ex. GW na na na na na na na 16.0 18.4 22.8NAVPS na na na na na na na 16.5 18.9 23.4NIBDPS na na na na na na na 7.05 5.05 2.86
P/E YE na na na na na na na 9.6 7.9 7.1P/E adj. YE na na na na na na na 9.6 7.9 7.1P/E average na na na na na na na 9.6 7.9 7.1P/E adj. average na na na na na na na 9.6 7.9 7.1P/CEPS YE na na na na na na na 5.9 5.2 4.9P/BV YE na na na na na na na 1.92 1.67 1.35P/BV ex. GW YE na na na na na na na 1.98 1.72 1.39Dividend yield YE na na na na na na na 4.2% 5.0% 5.6%Dividend Payout Ratio na na na na na na na 40.0% 40.0% 40.0%
EV/EBIT YE na na na na na na na 8.2 7.0 6.0EV/EBITA YE na na na na na na na 8.2 7.0 6.0EV/EBITA adj. YE na na na na na na na 8.2 7.0 6.0EV/Sales YE na na na na na na na 1.0 0.9 0.8EV/EBITDA YE na na na na na na na 5.6 5.0 4.4
Share price YE 31.7Share price high 31.7Share price low 27.8Share price average 30.3
ROE na na na na na 12.2% 13.7% 21.3% 22.5% 21.1%Adj. ROE na na na na na 12.4% 13.9% 21.3% 22.5% 21.1%Adj. ROCE pre-tax na na na na na 8.2% 8.6% 15.9% 17.4% 17.9%Adj. ROIC aft-tax na na na na na 5.8% 6.5% 12.8% 14.0% 15.0%Adj. ROA pre-tax na na na na na 5.9% 6.2% 10.4% 11.0% 11.7%FCF yield na na na na 4.2% 1.4% 6.0% 8.1% 11.0% 12.5%
Inventories / Sales 1.0% 1.0% 1.0% 1.0% 1.0% 1.0% 1.0% 1.0% 1.0% 1.0%Receivables / Sales 1.0% 1.0% 1.0% 1.0% 1.0% 1.0% 1.0% 1.0% 1.0% 1.0%Payables / Sales 1.0% 1.0% 1.0% 1.0% 1.0% 1.0% 1.0% 1.0% 1.0% 1.0%NWC / Sales na na na na na 25.1% 22.2% 17.8% 18.3% 18.6%Asset turnover na na na na na 0.76 0.69 0.88 0.89 0.92Sales / Capital invested na na na na na 1.08 1.00 1.35 1.40 1.45OCF / Capex na na na na 1.75 1.28 1.86 2.15 2.56 2.77Capex / Sales nm nm nm nm 7.1% 6.3% 7.7% 5.5% 5.2% 4.9%Capex / Depreciation PPE nm nm nm nm 1.53 1.25 1.65 1.04 1.04 1.04Dividend payout ratio na na na na na na na 40% 40% 40%
Equity / Total assets na na na na 43% 25% 34% 35% 39% 47%Net IB debt / Equity na na na na 28% 136% 75% 42% 26% 12%Net IB debt / EBITDA na na na na 0.8 4.9 3.0 1.0 0.7 0.4EBITDA / Net interest nm nm nm nm n.m. n.m. 262.3 9.8 17.5 21.2EBITA / Net interest nm nm nm nm n.m. n.m. 171.4 6.7 12.4 15.4
Net IB debt na na na na 1,750 7,930 6,391 4,098 2,937 1,665Net working capital (NWC) na na na na 3,717 3,433 3,967 4,418 4,746 5,077Capital employed (CE) na na na na 9,933 15,768 17,328 17,536 17,987 19,636Capital invested (CI) na na na na 10,207 16,078 17,154 17,655 18,033 18,414Enterprise value YE (EV) na na na na na na na 22,460 21,346 20,123
Source: Carnegie Research
c
Elkem
02 May 2018 123
Disclosures and disclaimers Carnegie Investment Bank AB Carnegie Investment Bank AB (publ.) is a leading investment bank with a Nordic focus. The Carnegie group of companies, together “Carnegie”, generates added value for institutions, companies and private clients in the areas of trade in securities, investment banking and private banking. Carnegie has approximately 600 employees, located in offices in seven countries.
Ratings and risk assessment structure Current rating system as of October 2011 Potential conflicts of interest
Carnegie, or its subsidiaries, may from time to time perform investment banking or other services for, or solicit investment banking or other business from, any company mentioned in this report. Any such publicly announced business activity, during the past 12 months, will be referred to in this report. A set of rules handling conflicts of interest is implemented in the Carnegie Group. Investment Banking and other business departments in Carnegie are surrounded by arrangements (Chinese walls) to restrict the flows of sensitive information. Persons outside a Chinese wall may gain access to sensitive information only after having observed applicable Chinese wall crossing procedures. The remuneration of persons involved in preparing this report is not tied to investment banking transactions performed by Carnegie or a legal person within the same group. Confidential and non-public information regarding Carnegie and its clients, business activities and other circumstances that could affect the market value of a security (“sensitive information”) is kept strictly confidential and may never be used in an undue manner. Internal guidelines are implemented in order to ensure the integrity and independence of research analysts. In accordance with the guidelines the research department is separated from the Investment Banking department and there are no reporting lines between the research department and Investment Banking. The guidelines also include rules regarding, but not limited to, the following issues; contacts with covered companies, prohibition against offering favourable recommendations, personal involvement in covered companies, participation in investment banking activities, supervision and review of research reports, analyst reporting lines and analyst remuneration.
Research Disclaimer Carnegie Investment Bank AB is responsible for the preparation of this research report in Sweden, Finland, Denmark, and the UK. Carnegie AS is responsible for the preparation of this research report in Norway. Carnegie Inc. is responsible for this research report in the US. The information in this report was obtained from various sources. While all reasonable care has been taken to ensure that the information is true and not misleading, Carnegie does not guarantee its accuracy or completeness. If the report has been disclosed to the issuer and amended following this disclosure before its dissemination, it will be referred to in the company-specific disclosures. Carnegie, its subsidiaries and any of their officers or directors may have a position, or otherwise be interested in, transactions in securities that are directly or indirectly the subject of this report. Any significant financial interests held by the analyst, Carnegie or a legal person in the same group in relation to the issuer will be referred to in the company-specific disclosures. This research report is prepared for general circulation and general information only. It does not have regard to the specific investment objectives, financial situation or particular needs of any specific person who may receive this report. Investors should seek financial advice regarding the appropriateness of investing in any securities or investment strategies discussed or recommended in this report and should understand that statements regarding future prospects may not be realized. Past performance is not necessarily a guide to future performance. Carnegie and its subsidiaries accept no liability whatsoever for any direct or consequential loss, including, without limitation, any loss of profits arising from the use of this report or its contents. This report may not be reproduced, distributed or published by any recipient for any purpose. The document may not be distributed to physical or legal entities that are citizens of or domiciled in any country in which such distribution is prohibited according to applicable laws or other regulations. Carnegie Investment Bank AB is a bank incorporated in Sweden with limited liability regulated by the Swedish Financial Supervisory Authority (Finansinspektionen). This report is distributed in Sweden by Carnegie Investment Bank AB. Carnegie UK is the UK Branch of Carnegie Investment Bank AB which is authorised by the Swedish Financial Supervisory Authority and subject to limited regulation by the UK Financial Conduct Authority. This report has been approved for the purposes of Section 21 of the Financial Services and Markets Act 2000 by Carnegie UK and issued by it in the UK. In Finland this report is issued by Carnegie Investment Bank AB, Finland Branch. The Finland branch is authoriszed by the Swedish Financial Supervisory Authority and subject to limited regulation by the Finnish Financial Supervisory Authority (Finanssivalvonta). In Norway this report is issued by Carnegie AS, a wholly -owned subsidiary of Carnegie Investment Bank AB. Carnegie AS is regulated by the Financial Supervisory Authority of Norway (Finanstilsynet). In Denmark this report is issued by Carnegie Investment Bank AB, Denmark Branch. The Denmark branch is authoriszed by the Swedish Financial Supervisory Authority and subject to limited regulation by the Danish Financial Supervisory Authority (Finanstilsynet). This report is distributed in the US by Carnegie, Inc., a US- registered broker-dealer. This report is provided for informational purposes only and under no circumstances is it to be used or considered as an offer to sell, or a solicitation of any offer to buy any securities. Any US person who wishes to effect transactions based on this report should contact Carnegie Inc. Investors in the US should be aware that investing in non-US securities entails certain risks. The securities of non-US issuers may not be registered with, nor be subject to, the current information reporting and audit standards of the US Securities and Exchange Commission.
Buy (B), upside of at least 10% to the target price and with an attractive risk/reward profile Hold (H), the stock is trading close to its target price and is fairly valued Sell (S), unattractive risk/reward ratio as the stock is trading above its target price Not rated (NR), Under review (UR), Under bid (UB). The investment rating, if any, has been suspended temporarily.
Risk assessment The risk assessment is based on the analyst’s evaluation of the company’s equity beta based on the business risk (asset beta) and financial risk (gearing). Low risk estimated equity beta <0.75 Medium risk estimated equity beta 0.75 to 1.25 High risk estimated equity beta >1.25
Valuation, methodology, and assumptions
Target price Carnegie publishes a target price for most of the stocks in our Research Universe. The target price is the analyst's assessment of expected total return over the coming six to 12 months based on various fundamental valuation methods. A commonly used method is DCF valuation, where future cash flows are discounted to today. Analysts may also use different valuation multiples, e.g. P/E ratio and EV/EBIT multiples, relative to industry peers to obtain a target price. For companies where it is appropriate, a target price can also be based on the analyst’s assessment of a fair ratio relative to the net asset value of the company. Target prices are revised when earnings and cash flow forecasts are changed. Thus, changes to estimates are a key risk to the target price. Other reasons for revising target prices include changes in the underlying value of a company’s assets and when factors affecting the required rate of return change, which can also be seen as risk factors to the target price. Information You will find detailed information about the valuation or methodology and the underlying assumptions on Carnegie Edge (www.carnegie-edge.com). The complete history of research reports and previous recommendations can also be found on Carnegie Edge.
Research coverage Carnegie’s research analysis consists of case-based analyses, which implies that the frequency of the analytical report may vary over time. Unless otherwise expressly stated in the report, the analysis is updated when considered necessary by the research department, for example in the event of significant changes in market conditions or events related to the issuer/the financial instrument.
Analyst certification The research analyst or analysts responsible for the content of this report certify that, notwithstanding the existence of any potential conflicts of interests referred to herein, the views expressed in this report accurately reflect the research analyst’s personal views about the companies and securities covered. It is further certified that the research analyst has not been, nor is or will be, receiving direct or indirect compensation related to the specific ratings or views contained in this report.
c
Elkem
124 02 May 2018
Elkem price, rating and target price history (ELK.OL)
Target Price changes
Our Target Price Horizon is 6–12 months
Source: Carnegie Research, Factset
Company specific disclosures The following disclosures relate to relationships between Carnegie Investment Bank AB (with its subsidiaries, “Carnegie”) and the subject company.
Within the past 12 months Carnegie (refer to definition in disclaimer text) has received compensation for investment banking services regarding EMGS, Ferronordic, Gaming Innovation Group, Jyske Bank, Mr Green, NEL, Next Biometrics, Nordax, Nordic Mining, Northern Drilling, Norwegian Air Shuttle ASA, Otello Corp, Ringkjøbing Landbobank, Stillfront, Telia Company, Thinfilm and Victoria Park. Within the past 12 months Carnegie (refer to definition in disclaimer text) has managed or co-managed a public offering of securities of AcadeMedia, Balco, BioArctic, Bonesupport, Boozt, Catena Media, Collector, Crayon, Elkem ASA, Handicare, Instalco, Medicover, Medivir, Munters, Orphazyme, Rovio Entertainment, Saferoad, Sdiptech, Stillfront, TCM Group, Terveystalo, THQ Nordic and Trianon. Please see Carnegie’s website for a full list of shares owned by employees of Carnegie AS in relation to potential conflicts of interest: www.carnegie.no, Verdipapirhandel, Analyse, Disclaimer (Norwegian). www.carnegie.no, Securities, Research, Disclaimer (English). Carnegie AS may also own shares in connection with trading. This disclosure is made to meet Norwegian best practice. Since September 30, 2010 Carnegie Investment Bank AB acts as a market maker in the FastPartner share. Since September 30, 2010 Carnegie Investment Bank AB acts as a market maker in the Kindred share. Since 17 November, 2016 Carnegie Investment Bank AB acts as a market maker in the TF Bank share.
Stock rating distribution in the previous 12 months
Ratings
Carnegie coverage universe % of total
Investment banking services* % of total
*Investment banking services provided by Carnegie in the previous 12 months
Elkem is one of the world’s leading manufacturers of silicones, silicon materials, foundry and carbon products. Its head office is in Oslo, Norway. Since its formation in 1905, Elkem has been one of the cornerstones of Norway’s industrial revolution. Since the merger with Bluestar Silicones in 2015 and the acquisition of the two large Chinese assets in 2017, Elkem is a fully integrated global player, from quartz to silicones. The company has a leading position in many of its upstream production facilities driven by access to cheap Norwegian hydropower.
Major shareholders
Carnegie Investment Bank AB Tel +46 8 676 88 00 Fax +46 8 676 88 95 Carnegie AS Tel +47 22 00 93 00 Fax +47 22 00 94 00 Carnegie Investment Bank, Denmark Branch Tel +45 32 88 02 00 Fax +45 32 96 10 22
Carnegie, Inc. Tel +1 212 262 5800 Fax +1 212 265 3946 Carnegie Investment Bank AB, Finland Branch Tel +358 9 618 71 230 Fax +358 9 618 71 720 Carnegie Investment Bank AB, UK Branch Tel +44 20 7216 4000 Fax +44 20 7417 9426
EPS na na 3.32 4.00 4.46EPS adj. na na 3.32 4.00 4.46EPS adj. Growth na na na 20.7% 11.4%CEPS na na 5.4 6.1 6.5DPS 0.00 0.00 1.33 1.60 1.78BVPS ex. GW na na 16.0 18.4 22.8NIBDPS na na 7.0 5.1 2.86
Ratios 2016 2017 2018e 2019e 2020e
P/E na na 9.6 7.9 7.1P/E adj. na na 9.6 7.9 7.1P/BVPS na na 1.9 1.7 1.4P/BVPS ex. GW na na 2.0 1.7 1.4P/CEPS na na 5.9 5.2 4.9EV/Sales 1.84 1.48 0.96 0.85 0.76EV/EBITDA 16.2 11.8 5.6 5.0 4.4EV/EBITA 27.6 18.0 8.2 7.0 6.0Dividend yield 0.0% 0.0% 4.2% 5.0% 5.6%FCF yield 1.4% 6.0% 8.1% 11.0% 12.5%
