Financial Summary

Recent Updates

Highlights

On February 25, 2011, Ferrotec released an upward revision to the planned full year dividend payment from 12 to 20 yen per share. The 20 yen per share includes a 30th anniversary commemorative dividend of 2 yen per share.

On February 14, 2011, the company released Q3 FY03/11 results.

Back to Top

Trends & Outlook

Quarterly Trends

FY03/11 Q3 Results

On February 14, 2010, the company announced Q3 FY03/11 results (see table above).

The company offered the following comments on Q3 performance (sales and operating profit are for cumulative Q3 FY03/11).

Equipment-Related Segment: sales were 20.2 billion yen (+104.8% increase), operating profit was 2.3 billion yen (vs. operating losses of 782 million yen in Q3 FY03/10).

• Vacuum feedthroughs: capex spending by semiconductor, FPD, and other manufacturers have increased due to demand for electronic products, particularly from companies in Taiwan and South Korea. Although a temporary adjustment took place, sales remain strong.
• As device manufacturers have kept their capacity utilization rates high, demand has been strong for consumables used in the manufacturing process, such as quartz and ceramic products.

Electronic Device Segment: sales were 5.2 billion yen (+100.7% YoY), operating profit was 852 million yen (+1252.4% YoY).

• Sales in thermoelectric modules remained solid mainly due to strong automobile sales in China, despite the end of many government support programs for the automobile industry in other countries.
• Demand was strong for thermoelectric modules and products used in private-sector products like air conditioners due to the approaching end of eco-point subsidies within Japan. Sales were also solid for semiconductor manufacturing equipment, medical testing equipment, and high performance products for optical communications and other areas.

Photovoltaic Segment: sales were 12.9 billion yen (+56.7% YoY) and operating profit was 1.0 billion yen (+21.2% YoY).

• Shipments of silicon ingot manufacturing equipment increased following a recovery in orders.
• Sales were strong in crystal silicon ingots for photovoltaic applications after the company began producing and selling photovoltaic wafers.
• Sales of quartz crucibles increased following an increase in capacity.

Topics

On February 1st, the company published a press release detailing orders received for products such as silicon ingot manufacturing equipment. According to the release, it had received 430 orders for silicon manufacturing equipment from January through June 2010. The company subsequently received a number of orders, centered on enquiries from 10 Chinese companies, including 870 new orders for silicon manufacturing equipment and 70 orders for its new product, block-cutting machines. Based on customers’ requested delivery dates , these orders will be booked as sales in FY03/12.

FY03/11 Q2 (1H) Results

On November 12th, 2010, Ferrotec announced its Q2 FY03/11 results (see table above) and announced an upward revision to its full-year forecasts. As a percentage of the company’s revised full-year estimates, the results were as follows:

• Sales: 51.1% (Full-year forecast: 46.5 billion yen)
• Operating profit: 54.7% (4.1 billion yen)
• Recurring profit: 55.7% (3.4 billion yen)
• Net profit: 58.5% (2.3 billion yen)

Sales increased 77.0% YoY during 1H of FY03/11 and operating profit reached 2.2 billion yen (vs. operating losses of 543 million yen in 1H FY03/10). Leading the growth were Equipment-Related sales, where the company realized YoY increases of 117.8% and operating profit of 1.4 billion yen (vs. operating losses of 824 million yen in 1H FY03/10). Furthermore, sales in the Electronic Device segment grew by 101.9% YoY, with operating profit of 411 million yen (vs. operating losses of 72 million yen in 1H FY03/10). Although sales in the Photovoltaic segment grew 34.7% YoY, operating profit declined 17.9%, finishing at 357 million yen.

The company offered the following comments on the 1H performance:

Equipment-Related Segment

1) Vacuum feedthroughs: Sales growth exceeded company expectations.

• Led by demand in developing nations, semiconductors remained strong in Asian and US markets. The domestic Japanese market has been recovering also.
• The European LED market remained strong.
• FPD has entered a stable phase.
• Sales for PV thin-film equipment have been increasing gradually.

2) Quartz products

• Taiwan memory market became more active leading to more capital investments.
• Orders from Asian and domestic users as well as major US OEM firms have increased rapidly.
• Production was unable to keep pace with the increase in orders for LED equipment.

3) EB-Guns/Deposition Equipment

• Vapor deposition equipment for LEDs was strong.

Electronic Device Segment

Thermoelectric module products

• Temperature-controlled automobile seats recovered following government support for the automobile industry in various countries, and new car sales in the US and Asia also contributed.
• In non-automotive applications, sales to the semiconductor industry were strong, and sales to customers in optics and biotech also grew.

Photovoltaic Segment

• Shipments of silicon ingot manufacturing equipment fell YoY but orders have rapidly recovered. Repeat and new orders have been increasing from new and existing clients.
• Payment from a major client was delayed, so the company recorded a charge during the second quarter which lowered segment earnings. Customers typically pay in three parts: 30% with the order, 60% on shipping, and 10% after inspection, but the company used different terms for the client in question.
• Sales were strong in single crystal silicon ingots for photovoltaic applications due to increased demand in the photovoltaic market. The company started production of photovoltaic wafers, which are now on sale.
• Strong orders for crucibles led the company to continue capital investment and significantly increase capacity. The reputation of the company's products has improved among Chinese clients and orders were strong.

Topics

The company issued unsecured convertible bonds to Tokyo Marine and Nichido Fire Insurance through a third-party allocation on November 5th, 2010. The issue raised approximately 2 billion yen, and had dilution potential of 7.72%.

The company said it would use the funds for the following investments:

• 650 million yen: capacity increase of wafer production equipment at Shanghai plant (planned for December 2010 to September 2011)
• 450 million yen: expansion of building that houses the Shanghai wafer production plant (planned for February 2011 to September 2011)
• 888 million yen: investment in square vessel operations for polycrystalline photovoltaic cells at Hangzhou (planned for November 2010)

FY03/11 Q1 Results

On August 13, 2010, the company announced Q1 FY03/11 results (see table above). Simultaneously, it announced upward revisions to its forecasts for the 1H of the same year. As a percentage of the revised 1H company estimates, results were as follows:

• Sales: 50.9% (vs. 1H forecast of 21 billion yen)
• Operating profit: 56.4% (vs. forecast of 1.8 billion yen)
• Recurring profit: 70.3% (vs. forecast of 1.3 billion yen)
• Net profit for quarter: 79.3% (vs. forecast of 830 million yen)

Q1 FY03/11 sales grew 64.2% YoY. Operating profit was 986 million yen (vs. operating losses of 238 million yen in Q1 FY03/10). Equipment-Related sales led company performance, rising 108.4% YoY. Additionally, operating profit was 558 million yen (vs. operating losses of 401 million yen in Q1 FY03/10). Furthermore, Electronic Device sales increased 108.3% YoY yielding operating profit of 152 million yen (vs. operating losses of 51 million yen in Q1 FY03/10). While sales in the Photovoltaic segment increased 15.0% YoY, operating profit was down 7.3% YoY to 243 million yen.

The revised 1H FY03/11 forecast was as follows:

• Sales: 21 billion yen (previous forecast: 18.8 billion yen)
• Operating profit: 1.8 billion yen (previous forecast: 1.4 billion yen)
• Recurring profit: 1.3 billion yen (previous forecast: 1.2 billion yen)
• Net profit: 830 million yen (previos forecast: 700 million yen)

The company cited “favorable conditions” when explaining the 1H upward revision. It left the full year forecast unchanged due to uncertainty in foreign exchange fluctuations and overall macroeconomic factors.

Full Year (FY03/11) Outlook

On February 8, 2011, the company revised FY03/11 estimates upward. Achieving these new forecast numbers would mean that sales, operating profit, and net income reach all-time highs.

The company gave the following reasons for the revision to the full-year forecasts.

1. In the Equipment-Related segment, although there has been a temporary adjustment period in the semiconductor and FPD industries, sales are expected to remain roughly unchanged (vs. 1H).
2. In the Photovoltaic segment, silicon ingot manufacturing equipment is expected to continue to perform well. The high level of market demand for crystal silicon ingots for photovoltaic applications exceeded expectations and photovoltaic wafers also performed strongly.
3. In the Electronic Device segment, the impact of the anticipated slowdown in thermoelectric module products was minimal and sales remained solid.

In the Equipment-Related segment, the company initially expected the semiconductor and FPD industries to slow down in 2H, but in actuality semiconductors sales were practically unchanged and the slowdown in FPD seems to have been short. Additionally, in the Electronic Device segment, sales of thermoelectric modules were expected to slow because vehicle sales (the main application for the products) had been forecast to decrease following termination of purchase incentives. But given factors such as the strong sales of vehicles in China, it now seems likely that results could exceed the forecast.

Capital expenditures, depreciation expenses

The company plans full year capex of 3.5 billion yen (vs. an initial amount of 3 billion yen) and 2.7 billion yen of depreciation expenses.

Shareholder return

The company plans to pay a full year dividend of 12 yen per share.

Longer Term Outlook

The company does not provide a medium-term business plan.

In estimating the company’s performance, key points worth considering are the semiconductor cycle and both the market dynamics and the company’s position in the LED and photovoltaic markets, which SR Inc. believes to be other important growth drivers.

FY03/12

(SR Inc. typically does not make forecasts, but offers the following discussion as reference for better understanding the company.)

While the Equipment-Related segment made up 61% of operating profit in 1H FY03/11 results, the Photovoltaic segment only contributed 16%. However, SR Inc. estimates that the Photovoltaic segment should contribute around 50% of the operating profit in FY03/12. SR Inc. notes that new capacity added in 2H FY03/11 and potential demand from China could see Photovoltaic segment sales grow to 32 billion yen. Using FY03/10 OPM as a guide and assuming increased depreciation, this could mean about 3.2 billion yen of operating profit. In addition to firm orders for single crystal furnaces during 2H FY03/11, production capacity for quartz crucibles and photovoltaic wafers (which have seen a lot of interest) should be higher in FY03/12 than the year before. Additionally, the company plans to start selling square vessels (see "Business Description") again. Depreciation expenses will probably increase, but in absence of the special provisions taken in FY03/11, there could be significant YoY growth.

In the Equipment-Related segment, trends in semiconductor-related products have been difficult to predict. However, even if semiconductor-related products enter a slowdown in 2H FY03/12, the present level would be maintained until 1H FY03/12. Also, considering that the uses of LED-related and FPD-related products are expanding, FY03/11 results should be achievable in FY03/12 (sales of 26 billion yen and operating profit of 2.8 billion yen; FY03/11 estimates also by SR Inc.)

Changes to demand following the end of automobile purchase incentives are an unknown variable for the Electronic Device segment. However, in December 2010, policymakers in the US decided to implement an income tax reduction of 110.0 billion dollars (around 0.8% of the US GDP). Many automobiles in the US have temperature-controlled seats that use Ferrotec's thermoelectric modules, so the impact of tax changes on US consumer spending may be important. Also, automobile sales in China are expected to remain strong. In other areas, it seems reasonable to conclude that sales of products eligible for eco-points will fall after the program ends (especially considering a final rush to buy just before the program stops). SR Inc. thinks that it will be possible for Electronic Device segment to reach nearly the same level in FY03/12 as they do in FY03/11 (SR Inc.’s FY03/11 estimates: 6.7 billion yen in sales, 800 million yen of operating profit).

Considering the above scenarios, SR Inc. thinks consolidated FY03/12 sales could reach 65 billion yen with 6.4 billion yen of operating profit.

Future direction of the Photovoltaic Segment

Ferrotec announced a 36 billion yen sales goal at a presentation regarding its Photovoltaic products in July 2008. The 36 billion yen goal was broken down into 20 billion yen for silicon manufacturing equipment, 5.4 billion yen for quartz crucibles, 8 billion yen for silicon ingots, and 1.5 billion yen for maintenance and other equipment. In January 2011, the company said that it was possible to achieve the 36 billion yen sales goal in the next two or three years.

Back to Top

Business

Business Description

Ferrotec was established in 1980 as a technology venture company. With two core technologies, thermoelectric modules (which were attracting attention as a new promising component), and magnetic fluids (born from US NASA projects), the company provides products and services to the electronics, semiconductor, machine tool, and consumer equipment industries. The company’s three main business segments are: Equipment-Related, Photovoltaic, and Electronic Devices.

Equipment-Related Segment (37.5% of FY03/10 sales)

Products used in the manufacture of semiconductors, FPDs (flat panel displays), LEDs (light emitting diodes), manufacturing equipment, etc. are sold in this segment. The principle products are vacuum feedthroughs, quartz products, ceramic products, EB (electron beam) guns, vapor deposition equipment, and silicon wafer processing.

• Vacuum feedthroughs

This is the company's main product. The primary use for vacuum feedthroughs is in manufacturing equipment for semiconductors, FPDs, and LEDs. According to the company, its worldwide market share was approximately 70% in FY03/10.

(Technology background: Vacuum feedthroughs and magnetic fluids)

Vacuum feedthroughs use magnetic fluid technology. Among those that have been developed for practical applications, magnetic fluids are the only liquid substances that have viscosity in response to magnetic fields. For that reason, the liquid can be kept from spilling by controlling the level of magnetism. Magnetic fluids were developed during the Apollo program, which was underway in the 60s at NASA (National Aeronautics and Space Administration) as a way of delivering fuel to the rocket engines in a weightless environment without spilling. After achieving mankind's first-ever moon landing, two of the men involved with the development founded Ferrofluidics (currently Ferrotec USA). According to the company, magnetic fluids can be produced in any standard laboratory. However, the three main components have to be blended as magnetic microgranules, surfactants, or base solutions (depending upon the environment they will be used in) and the company has that blending expertise. The company claims a worldwide share of approximately 88% (as of FY03/10). Other firms working on magnetic fluids include Eagle Industry Co., Ltd. (TSE 6486) and a firm in China.

Because ultra-precision films are produced during the semiconductor wafer and FPD manufacturing processes, those processes take place in sealed vacuums. If air, gas, vapor, microscopic particles, or other contaminants find their way in, circuit pattern quality can degrade. In addition to isolating the sealed space where the processing takes place from the outside, vacuum feedthroughs also play the role of transmitting various sorts of movement needed during the process taking place in the sealed environment. The process used in semiconductor manufacturing equipment is either batch or wafer, but the company's products are used primarily in wafer type production.

In recent years, there have been a lot of inquiries from LED and photovoltaic-related industries. Vacuum feedthroughs are used in the MOCVD (Metal Organic Chemical Vapor Deposition) devices that produce LEDs. To produce LEDs, the gas that serves as a raw material has to be introduced into the environment to produce many different layers on a substrate. Because the gas that is the raw material is a very toxic specialty high-pressure gas, it is extremely important that a sealed environment be maintained.

(Technology background: LED)

Broadly speaking, the LED manufacturing process can be divided into the chip process and the packaging process. The chip process can be further divided into the MOCVD, edging, electrode formation, polishing, and the dicing processes. The MOCVD process is the most important part of the chip process and it is the process in which the gas feed is used to deposit GaN (gallium nitride) or another substance onto a sapphire substrate (or a silicon carbide substrate). Aixtron of Germany and Veeco Instruments of the USA have the largest share of the MOCVD equipment market.

In addition to Ferrotec's products being used in photovoltaic cell-related industries in single crystal furnaces (three per device), they are also sold externally for use in photovoltaic cell manufacturing equipment.

Vacuum Feedthrough

Vacuum Feedthrough

Ferrofluid

The breakdown of the company's customers by industry in FY03/10 was: semiconductors 25%, LEDs 24%, FPDs 19%, and photovoltaic cells 16%.

Principle semiconductor clients include Applied Materials (USA), Novellus (USA) and Tokyo Electron (TSE 8035). Some LED clients include Aixtron (Germany) and Veeco Instruments (USA), while FPD clients include ULVAC (TSE 6728), Canon ANELVA (unlisted), TOKKI (JASDAQ 9813), and Doosan (South Korea); photovoltaic cell customers include ULVAC (see above).

The following three characteristics make the company’s competitive position in vacuum feedthroughs relatively unique. First, the company has a large share of the market for the magnetic fluids used in vacuum feedthroughs. Next, the product is needed when a client develops new manufacturing equipment. For that reason, the company is involved with the client from the design and development stages, which means that it has a grasp of that client's needs ahead of the competition. Lastly, the vacuum feedthrough market is not that large in size, so other firms do not feel that great an incentive to enter.

According to the company, prices and quantity tend to move in the same direction (it can raise prices when demand increases). For that reason, when looking at the product's potential profitability, one needs to look at the cycles of the client industries, which include semiconductors, LEDs, FPDs, and photovoltaic cells. Also, the stainless steel (a raw material) price trends also impact the cost to some degree.

• Quartz products

The company began selling quartz products as a way to leverage its number of blue-chip clients and its share of the vacuum feedthrough market.

Quartz glass boasts high purity (purity: 99.99%), is heat resistant, and is not easily affected by chemicals. The many processes used in the production of semiconductor wafers and LEDs take place in an environment of high temperatures and chemical changes brought on by activated gases, requiring high-purity quartz glass products. More specifically, it is used in fittings and parts like the various core tubes used for diffusion in CVD (chemical vapor deposition). The company purchases high-purity quartz glass from Momentive Performance Materials and processes and sells it to semiconductor manufacturers. Additionally, the company has a full line-up of quartz products needed in semiconductor manufacturing processes.

The sales breakdown by industry for 1H FY03/11 is OEM 52%; end users 26%; LEDs 10% and PVs (Photovoltaics) 8%. OEM involves supplies for semiconductor manufacturing equipment. End users are primarily foundries (semiconductor subcontracting firms) in Taiwan and elsewhere. Shipments to the semiconductor industry accounted for about 78% of combined sales to OEMs and end users.

Ferrotec said it sales volumes were about average compared to the rest of the industry, and estimated its share at about 10%. The company said that it works in areas that major firms, like Shinetsu Chemical (TSE 4063), do not.

• Ceramics

The company started selling ceramic products for similar reasons as its quartz business.

Ferrotec’s ceramic products can be divided into fine ceramic products and machinable ceramics. Fine ceramics are used in all of the semiconductor and liquid crystal manufacturing processes (wafer production, processing, assembly, inspection) and they require high purity, high strength, and high precision. They are also used by other industries for their heat and wear resistance. Machinable ceramics have superior electrical insulating, heat insulating, and heat resistance properties. Because it is “machinable,” it can be fabricated to precise specifications (it’s surface can be shaved, cut, etc. whereas typical ceramics are too brittle). They are used to reduce lead time when going from design to prototype in the manufacturing of semiconductors and liquid crystals.

Ferrotec acquired shares of Sumitomo Ceramics & Quartz, a subsidiary of Sumitomo Metals (TSE 5405), in September 2008 and made it into a consolidated subsidiary. In doing so, Ferrotec added machinable ceramics production to its internal manufacturing capabilities.

Principle clients include Japan Electronic Materials (TSE 6855) and Micronics Japan (JASDAQ 6871).

Ferrotec explained that its market position is similar to quartz products, relatively smaller when compared to the bigger market leaders like Kyocera (TSE 6971).

• EB guns and vacuum deposition equipment

Electron Beam (EB) vapor deposition is a method of depositing and building up metals and oxides on the surface of a substrate in a vacuum using an electron beam. There are no upper limits on heating temperatures, so advantages include the ability to vaporize any substance and a capacity for a high degree of precision control. The method employed in the semiconductor manufacturing process is like sputtering, in which a laser fires against a rigid substrate. However, in LED manufacturing, chemical semiconductor compounds are used on soft substrates, so the method used is vaporization onto the substrate. Basically, it is a vapor deposition system for chemical semiconductor compounds and is used in the process that follows MOCVD in the LED manufacturing process.

The company originally sold EB guns through a German subsidiary, but recognized that its existing supply relationships in the LED market (mostly vacuum feedthroughs) could lead to sales growth. The company was already selling vacuum feedthroughs, ceramic and quartz products to LED manufacturers, and moved to expand its product offering to include vacuum deposition equipment also needed to make LEDs. In January 2010 the firm’s US subsidiary bought the vacuum deposition-related business from Edwards Vacuum UK. It then combined its EB guns with vacuum deposition equipment and began selling packaged electron beam deposition systems.

• Wafer processing

The company processes silicon wafer substrates, which are indispensible in the manufacturing of LSIs (large-scale integrated circuits) and other devices, with its partner, Covalent Materials Co., Ltd. (unlisted). Wafer processing is performed largely in Shanghai, China. Traditionally included in CMS Operations (which existed through FY03/10), the firm started to produce wafers under its own brand. Wafer processing was reported in the Equipment-related segment starting in FY03/11. Monthly production capacity is approximately 400,000 units, of which, approximately 300,000 are sold to Covalent Materials, and most of the rest sold under the Ferrotec label.

Photovoltaic Segment (33.5% of FY03/10 sales)

This division manufactures, develops, and sells photovoltaic cell-related products. The main product is silicon (single and poly) crystal manufacturing equipment, silicon products for photovoltaic cells, and quartz crucibles. All production is done by Chinese subsidiaries.

(Technology background: Photovoltaic cells; more detail can be found in the "Market and Value Chain" section)

Photovoltaic cells can be divided into crystalline silicon types, thin-film silicon types (a thin-film silicon membrane is deposited onto a glass substrate, which uses less silicon) and chemical compound semiconductor-types (cells in which semiconductor compounds of elements such as copper, iridium, selenium and gallium are used in place of silicon).

As of 2010, approximately 85% of all photovoltaic cells were the silicon type. Crystalline silicon cells can be further divided into single crystal (monosilicon) and polycrystalline (polysilicon) types. Generally, single crystal silicon photovoltaic cells have a superior conversion efficiency (electrical energy output divided by the solar light energy) of 17 to 20%, but their manufacturing costs are high. Conversely, the polycrystalline type has a conversion efficiency of 15 to 18%, which is slightly worse than the single crystal type, but the manufacturing costs make it preferable. As of 2010, the polycrystalline type was mainstream, accounting for approximately 50% of the entire market. In contrast, single crystal silicon took up approximately 35%.

Among the thin-film and chemical semiconductor compound (non-crystalline silicon types), First Solar currently has the top share in the world. First Solar appears to focus on the chemical compound type, primarily cadmium tellurium (CdTe). Other types include pigment-sensitized and organic thin-film types, but these have not reached the point of forming significant markets.

Single crystal silicon photovoltaic cells were developed in 1954 at Bell Laboratories in the US. In single crystal silicon, atoms are aligned regularly so they are able to convert light into electricity efficiently. The high-purity silicon used in single crystal silicon photovoltaic cells is produced using the same sort of manufacturing process used to make silicon for the semiconductors used in personal computers. Silicon used in semiconductors (“semiconductor grade”) has a purity of at least “Nine 9s” (meaning 99.999999999% pure), and is very expensive. Silicon used in photovoltaic applications (commonly referred to as "SOG-Si"), has a purity of at least 99.9999% and is typically less expensive.

The manufacturing process for silicon wafers used in single crystal silicon photovoltaic cells is outlined below (underlined sections are products handled by Ferrotec).

1. Silica (raw material for silicon) is reduced to metallic silicon in an electric furnace and refined into high-purity silicon (seed crystal) through gasification and other means.
2. Silicon that has been refined to a high degree of purity is heated in a dish (some of which are quartz crucibles) inside a carbon crucible known as "hot zone" in a vacuum furnace heated to around 1,500 °C. The molten liquid is slowly turned and drawn upward, using a single crystal pulling device to produce a single crystal silicon ingot (a solid piece of silicon).
3. A block cutting machine is used to bevel the ingot and thinly slice it with a wire saw. The slices are then processed into wafers.

Conversely, polycrystalline silicon photovoltaic cells were developed to resolve problems associated with the process of producing single crystal photovoltaic cells (like the time and cost of producing single crystal silicon ingots and difficulties with mass production). The process of manufacturing the silicon wafers used in polycrystalline silicon photovoltaic cells is outlined below (underlined sections are products handled by Ferrotec).

1. Silica (raw material for silicon) is reduced to metallic silicon in an electric furnace and refined into high-purity silicon (seed crystal) through gasification and other means. (Same as single crystal silicon.)
2. Refined, high-purity silicon is melted in a melting pot (polycrystalline silicon production equipment) at about 1,000 °C and poured into a mold (square vessel). Cooling it in that state allows many crystals to gather together, producing a polycrystalline silicon ingot.
3. A wire saw is used to cut the ingot into thin slices, which are processed into wafers. (Same as single crystal silicon.)

Moreover, the amount of silicon used in thin-film silicon photovoltaic cells can be kept to 1/100 of that used in polycrystalline silicon, reducing production costs. Unlike crystalline silicon, which is made by melting at high temperatures, thin-film silicon is produced by heating to around 200 °C. The most common method involves the use of a vacuum furnace known as a plasma chemical vapor depositor (CVD) device. This is where Ferrotec's vacuum feedthroughs are used. Silane (SiH4) and other gases that serve as the raw material are broken down by electrical discharge, which causes the silicon to bind chemically to the glass or other type of substrate.

The photovoltaic cell products handled by Ferrotec are shown below broken down by material. Note also that Ferrotec does not work with products from the post-processes subsequent to the photovoltaic cell manufacturing process.

Single crystal silicon photovoltaic cells

Single crystal furnace (internal vacuum feedthroughs), quartz crucibles, carbon crucibles, single crystal silicon ingots, wafers, block cutting machines

Polycrystalline silicon photovoltaic cells

Polycrystalline furnace, square vessels, wafers, block cutting machines

Thin-film silicon photovoltaic cells

Vacuum feedthroughs

(Source:Company Data Processed by SR Inc.)

Single Crystal Furnace

Polycrystalline Furnace

Quartz Crucible

Crystal Silicon Ingot

Source: Company Data Processed By SR Inc.

• Silicon crystal manufacturing devices

The main products in the photovoltaic cell-related industries are "single crystal furnace", or "polycrystalline furnace." These devices produce single (multi) crystal pieces (ingots) of silicon used in photovoltaic panels.

The company estimates its global share of single crystal furnaces at approximately 20%. Also, it has a base established in an area of China where the majority of the world's production of silicon ingots is located. Restricted to that region, the company's share rises to approximately 30%. The company can manufacture between 60 to 70 units per month, with a turnaround from order receipt to shipment of about four months.

Single crystal furnaces and polycrystalline furnaces produce ingots in a vacuum at high temperatures. Control of the vacuum environment affects ingot yields and ultimately, the conversion ratio of the finished product. The process of making the silicon ingots, particularly with single crystal silicon, is the focal point of improving yields when drawing (producing) ingots. In addition to the use of Ferrotec's core technology "vacuum feedthrough" in the rotating part of their "single crystal furnace," there are a number of other technologies in use that were developed in the production of semiconductor manufacturing equipment. (Previously, Ferrofluidics worked on semiconductor manufacturing equipment, but demand fell off as silicon manufacturers switched to internal production, forcing them out.) Ferrotec states that its superior quality makes it stand out among local equipment manufacturers in China.

Conversely, Ferrotec admits that its share of the polycrystalline furnaces, at several percent, is low. That particular product has been produced by competitors GT Solar (USA) and ALD Vacuum Technologies (Germany) for many years who collectively boast a large share.

Ferrotec’s production capacity for these devices is five units per month, with a turnaround time of approximately five to six months from receipt of order to shipment. Due to Ferrotec’s small market share and the lack of a vacuum feedthrough in polycrystalline furnaces (see discussion on manufacturing process above), it would seem that this equipment is not as profitable as single crystal furnaces.

• Crucibles

"Quartz crucibles" are used as the container for melting the raw silicon in the vacuum chamber of the single crystal furnace. High-purity quartz is used in the production of single crystal silicon because it is heat resistant and there is no chemical change that takes place in the presence of the active gas. One quartz crucible is consumed each time the single crystal furnace operates. Ferrotec's crucibles are not always used in Ferrotec-branded silicon production equipment, but it is making efforts to sell its crucibles to sites where it has delivered manufacturing equipment. Ferrotec estimates it has a worldwide share of approximately 30%, making it the number one manufacturer of quartz crucibles for photovoltaic equipment in the world. The company estimates that it can produce about 18,000 units per month (as of January 2011).

As with its single crystal furnaces, Ferrotec claims that superior purity and other quality concerns set it apart from competition. More specifically, the development of materials is something that is very important in terms of technology. In contrast with semiconductor crucibles, where only one type of sand is used, the level of quality must be kept consistent by blending different sands together for photovoltaic cells.

In the longer-term, the company thinks that more growth will come from consumables like the quartz crucibles than manufacturing equipment. In other words, once the manufacturing devices have been set up, they will not be replaced until they get old or the technology changes. On the other hand, there is always a fixed level of demand for consumables. This reminds SR Inc of the relationship between copiers and toner to the copier manufacturers. Once a copier has been installed, it is going to be used until it stops working, but the toner cartridge has to be changed frequently. It is well known that Canon (TSE 7751) and the other copier manufacturers make their money on toner.

Additionally, Ferrotec is planning to start selling square vessels (large-size square crucibles made from ceramics). These are used in the manufacturing of polycrystalline silicon, and are produced by a joint venture formed with Covalent Material (unlisted).

• Photovoltaic Wafers

Ferrotec creates, cuts, and polishes its own ingots, processing them into wafers. The company manufactures the single crystal silicon that is used for solar cell substrates in Shanghai. The company put renewed emphasis on wafer sales in FY03/11 because it said it needs to have a deeper understanding of the complete processes in order to provide value-added solutions to its wafer-producing clients.

• Other

Ferrotec is also involved with the development, manufacture, and sale of carbon crucibles, block cutting machines, wafer inspection devices, and other peripherals products and equipment.

Electronic Device Segment (11.9% of FY03/10 sales)

Principle products are thermoelectric modules and magnetic fluids.

• Thermoelectric modules

A thermoelectric module is a semiconductor-cooling element that can instantaneously warm or cool an object by producing a temperature differential by inducing an electric current. Because they are small, light, and do not make noise or vibrate, they are used in a wide variety of fields, from growth areas, such as electronics, optical communications and biotechnology to leisure goods and home appliances to maintain constant temperatures. In recent years, one expanding area of use has been in temperature-controlled automobile seats. Ferrotec supplies its "thermoelectric modules" for temperature-controlling units in automobile seats developed by Amerigon (USA), a US automotive parts manufacturer. Ultimately, Amerigon's (USA) products are installed primarily in the seats of luxury automobiles at major manufacturers in the US, Japan, and South Korea.

(Technology background: Thermoelectric modules)

A thermoelectric module is a semiconductor cooling element that uses the "Peltier effect," in which direct current flows, causing heat to move so that one surface absorbs heat (cools) and the opposite surface radiates heat (warms). Reversing the polarity of the power supply allows the heat absorption and radiation to be switched and reversed easily. It is possible to control the temperature between 0° C and 100° C in increments of 0.1° C. The advantages of using these products are that they are not only compact, light and CFC-free, but also that have very precise temperature control. Thermoelectric modules are best suited for close proximity uses, not large spaces. The principle behind the "Peltier effect" was discovered in 1833, but arguably only came into practical use in the 20th century. Ferrotec's President Yamamura has been involved with the production method as a researcher since the 1970s.

Examples of uses for this product in home appliances include air purifiers, hair dryers, and air conditioners. For instance, in recent years, air purifiers with negative ion generators have become a popular item, and thermoelectric modules are being used to produce negative ions.

Thermoelectric modules were originally used for military applications such as air conditioning on submarines. SR Inc. understands that the manufacturing process is generally difficult to automate, so the technology is ill-suited for mass market use such as consumer electronics. Automation is challenging because semiconductor dies have to be lined up, and if this is not done correctly, the module would not perform (temperature control) fully. Ferrotec solved this problem by building an integrated production line in China. Additionally, the manufacturing process starts by producing ingots with bismuth, tellurium, antimony, or other ores and attaching electrodes before slicing and assembly. Also, the ores used as raw materials can be procured in China. Although market forces affect raw material prices, labor is a larger cost component, so marginal profit margins are high.

Ferrotec estimates its market share of thermoelectric modules used in automobiles at 100%, but it can’t provide an estimate which includes every application, because some markets (such as military) are not disclosed. Still, it seems safe to assume that Ferrotec is a market leader in thermoelectric modules. The company commented that maintaining product quality and cost controls while making a profit serve as barriers to entry for competitors.

• Magnetic fluids

Magnetic fluids are explained under "Vacuum feedthroughs". In addition to being used in the company's vacuum feedthroughs, they are also sold to manufacturers of home and automobile speakers. According to Ferrotec, there is a wide range of applications for magnetic fluids, but because the amounts used in each product are small (just a few drops), the size of the market is limited.

Other

For the most part, these are operations that produce products for other manufacturers under subcontracting agreements. The products seem to be quite varied, but Ferrotec refrains from discussing details due to restrictions stemming from confidentiality agreements with clients. Some examples that the company did provide include machine tool manufacturing and equipment parts cleaning.

Business Model, Profitability, Financial Ratios

Business Model

Ferrotec provides goods and services to various industries, including photovoltaic cells, semiconductors, electronic devices, LEDs and FPDs. The company has a large market presence in its core products, but started to expand into peripheral goods. While trying to expand sales, the company has also been taking steps to manage costs, like using global manufacturing capabilities (particularly in China).

One can get a rough idea of the company's sales by following trends in the industries mentioned above and examining Ferrotec’s market share for particular products. Watching the market cycles of each industry is an important key to understanding Ferrotec’s sales trends. Specifically, semiconductor manufacturers make up a lot of Ferrotec’s clients, and swings in capital expenditures by these companies can be significant. The effects of semiconductor manufacturers’ spending should gradually decline as the size of Ferrotec’s photovoltaic business increases. Sales to the semiconductor industry were about 36% of total sales in FY03/10, so the spending trends remain a key sales driver for Ferrotec.

SR Inc. performed a linear regression examining the relationship between the company's expenses (CoGS, and SG&A) and sales, and based on the figures since FY03/05, the firm's fixed cost ratio would be approximately 40% (see note below for detail). In other words, the ratio of fixed costs is high so that the level of capacity utilization would tend to have a significant effect on operating profit.

Note:

Dependent variables: CoGS, SG&A

Independent variable: sales

Linear regression equations:

CoGS (y) = Sales x 0.72 - 371 (R-squared = 0.99)

SG&A (y) = Sales x 0.19 + 982 (R-squared = 0.82)

Interpret the results as follows: on average, 72% of CoGS were variable (remaining 28% were fixed), and 19% of SG&A were variable (remaining 81% were fixed). Calculating the total percentage of fixed costs in both CoGS and SG&A indicates that about 40% of total costs are fixed. SR Inc. examined data beginning in FY03/05 because it is likely that the company restructured costs following several years of losses from FY03/02-FY03/04.

Profitability, Financial Indicators

The operating profit margin seems volatile, and SR Inc. thinks that high fixed costs are likely the cause.

Strengths & Weaknesses

Strengths

• Exposure to growth markets. Ferrotec supplies a number of important products (such as vacuum feedthroughs) to developing industries with promising growth prospects. Two such industries are LED manufacturing and the photovoltaic cell industry. Ferrotec has been leveraging its market position to increase involvement in these markets, and seems well-positioned for future growth. If the company is able to establish stable long term businesses in these areas, it could help offset its dependence on the semiconductor market, and open up additional growth possibilities.
• High market share in core products. Ferrotec is the market leader in vacuum feedthroughs, and uses this position to get close to clients during early stages of their product development. From that position, the company is able to proactively sense what its customers will need vs. other companies who are more reactive. For example, when the company was working on vacuum feedthroughs for LEDs, it saw that demand for vacuum deposition devices was growing, and was able to position itself to benefit. SR Inc. notes that the company’s leadership in vacuum feedthroughs could continue to be advantageous, considering the development of the photovoltaic cell industry and needs for single crystal furnaces.
• Management with financial and engineering focus. Many manufacturers in Japan seem to focus on quality as a way to achieve ultimate success – build the best mousetrap and you will become the market leader, irrespective of costs. As a result, it is not uncommon to find manufacturers that can boast impressive product quality, but rarely do the firms also have compelling profitability. Ferrotec seems to be an exception. The company’s roots are firmly in cutting-edge technology, but SR Inc. thinks that earlier challenges have instilled a since of financial discipline in management (a lesson corroborated by President Yamamura). The decision making process seems to include equal parts engineering focus as well as profitability.

Weaknesses

• Volatility in key industries. Clients in the semiconductor industry represent a large portion of sales, and the speed and size of capex ebbs and flows can be significant. These shifts in spending mean that Ferrotec is to some extent constrained to take risk in other businesses because sales to core clients can be unpredictable over the longer term. Although the company has been expanding into non-semiconductor clients and consumables businesses, it still has a ways to go before these develop into strong pillars that can offset the risk posed by the semiconductor industry.
• (By extension) small size = cannot diversify volatility risks away. One of the advantages that larger technology firms possess is the “portfolio effect” of many different businesses. Due to Ferrotec’s relatively smaller size, it is unable to assemble enough different businesses to imitate this. This is a relative weakness when considering management’s focus on financial performance, whereas competitors may not face the similar constraints.
• Expertise in polycrystalline silicon a work in progress. Ferrotec’s strength in the photovoltaics market involves single crystal silicon cells. Although single crystal silicon cells have the highest conversion efficiency (see Business Description), other technologies like thin-film or hybrid cells may prove to be winners for other reasons. Ferrotec has a relatively small share outside of single crystal silicon, which means that it would be at a relative disadvantage if this were to happen. SR Inc. notes that Ferrotec is improving its capabilities outside of single crystal silicon, but until successful, its reliance on one type of silicon could be considered a relative weakness.

Global Presence

The firm has sales offices in Japan, China, Taiwan, South Korea, Singapore, the US, and Europe (England, France, Spain, Italy, Germany, and Russia).

Ferrotec’s manufacturing capabilities are concentrated in China, but also has production facilities in Japan, China, the US, Germany, and Russia.

• Equipment-Related segment: Vacuum feedthrough sites are in Japan, Hangzhou (China), Taiwan, South Korea, and the US. Quartz products are produced in Japan and Hangzhou. Ceramic products are produced in Japan and Hangzhou.
• Photovoltaic segment: Silicon (single and poly) crystal furnaces for photovoltaic cells are produced in Shanghai and Hong Kong. Silicon products for photovoltaic cells are produced in Shanghai and crucibles and made in Hangzhou.
• Electronic Device segment: Thermoelectric modules are produced in Hangzhou.

Sales By Region

Sales in Japan were 47.9% of FY03/10 sales (overseas sales were the remaining 52.1%). SR Inc. thinks that the higher proportion of overseas sales comes from Ferrotec’s China-centric global manufacturing capabilities and the large number of overseas companies engaged in photovoltaic and semiconductor industries.

Group Companies

The corporate group consists of Ferrotec, 19 consolidated subsidiaries, and 5 equity-method affiliates. Ferrotec develops, manufactures and sells vacuum feedthroughs, develops and sells thermoelectric modules and develops, manufactures and sells magnetic fluids. The other principal group companies are listed below. (Locations and ownership percentage shown in parentheses.)

• Hangzhou Dahe Thermo-Magnetics Co., Ltd. (Hangzhou, Zhejiang Province, China, 100%): Manufacture and sale of vacuum feedthroughs; manufacture and sale of quartz products; development, manufacture and sale of ceramic products and manufacture of thermoelectric modules.
• Advanced Quartz Material (Hangzhou) Co., Ltd. (Hangzhou, Zhejiang Province, China, 100%): Development, manufacture and sale of crucibles.
• Shanghai Shenhe Thermo-Magnetics Co., Ltd. (Shanghai, China, 100%): Development, manufacture and sale of silicon products for photovoltaic cells, manufacture of thermoelectric modules, processing and sale of silicon wafers, etc.
• Shanghai Hanhong Precision Machinery Co., Ltd. (Shanghai, China, 88%): Development and manufacture of silicon (single and poly) crystal manufacturing equipment for photovoltaic cells, etc.
• Ferrotec Apollo Holding Co., Ltd. (Hong Kong, 88%): Development and manufacture of silicon (single and poly) crystal manufacturing equipment for photovoltaic cells, etc.
• Ferrotec (USA) Corporation: (New Hampshire, USA, 100%): Development, manufacture and sale of vacuum feedthroughs, sale of quartz products, sale of thermoelectric modules, etc.

Back to Top

Market & Value Chain

(Note: For the Market and Value Chain discussion, SR Inc. focused on the company’s PV segment because it seems to be the most likely growth driver for the company going forward.)

Market Overview

Due to the problem of global warming and soaring crude oil prices in recent years, sources of clean energy that can replace fossil fuels have been getting attention. Although nuclear power is frequently discussed, some concerns (such as safety) make it an inferior solution compared to clean energy. One such promising source of energy is photovoltaic power generation (PV). Solar power uses photovoltaic cells, which produce electricity by converting sunlight into electrical power. It emits no carbon dioxide.

The figure above shows the past performance and future forecasts for the photovoltaic cell market. In 2008 in Spain, upper limits on the number of solar power generation facilities allowed each year and reductions in feed-in tariffs (FITs a system whereby power companies are required to buy all power produced using sunlight at a uniform price) were announced, causing a boom in demand with a sudden surge in last minute applications. The size of the market in 2008 was 5.98GW, of which, the Spanish market was approximately 46%. However, the global economic crisis of 2008-2009 caused investment funds to dry up, and the market for photovoltaic cells stopped growing. However, moving into 2010, in Europe and other regions including Japan, the introduction of FITs and other preferential systems resumed and demand increased accordingly. The 2010 market for photovoltaic cells was 16.3GW (estimates from Solarbuzz, an industry research and consulting firm, are used here and below), of which 80% was said to be in Europe. However, interest from China and other new markets seem to be increasing, pushing up worldwide levels of demand. The amount of corresponding photovoltaic cell production in China and Taiwan was approximately 50%. About 80% of the photovoltaic cells made in those regions was bound for Europe.

One potential concern for the demand outlook in 2011 is the impact of lowering FITs in Germany. However, Germany accounted for about 27% of the global market for photovoltaic cells in 2010, which is lower than Spain's level in 2008. Another difference between the market in 2011 vs. 2008 is the expected appetite from the US and China.

In July 2009, China announced a joint "Golden Sun Model Project" shared by the Ministry of Finance, the Science and Technology Department, and the National Energy Bureau. This project is supposed to supply subsidies of 50-70% of the amount of the investment in projects related to solar power production, ranging from the production of silicon used in solar power production to systematic cooperation. In China, there has been a steady stream of construction projects involving power production using renewable energy sources resulting from this support.

Japan introduced its own version of FITs in November 2009. These “Japanese FITs” established the purchase price for electricity generated from photovoltaic sources at double the household electric power rate (48 yen/kWh vs. 24 yen/kWh). The increased rate had the effect of instantly lowering the hurdle for introducing solar power production systems in Japan.

According to Solarbuzz forecasts, the photovoltaic cell market in 2011 will reach 20.4GW. Additionally, it expects the market will reach 56.0GW by 2020 (growing about 13.1% per year). SR Inc. feels that these projections are probably reasonable when considering the declining prices of photovoltaic cells and expanding potential uses, all contributing to growth.

Regarding pricing, the cost of solar power generation will fall as installation becomes cheaper, technology improves, economies of scale become possible from mass production, and "grid parity” extends globally. (“Grid parity” refers to the situation when electric power generated using solar power costs the same or less than electric power from the regular power grid.) In addition to being affected by the amount of sunlight and other factors, the cost of solar power generation varies with differences in electric power rates, countries, and regions. Therefore, grid parity will likely be realized at different times in different regions. First Solar achieved grid parity in the US, and other companies in Europe (like Italy and parts of southern Europe) have also achieved grid parity. As for other manufacturers and other regions, it is generally believed that grid parity will be achieved by 2020 (Source: Development Bank of Japan). Theoretically, after grid parity has been achieved, it will be less expensive to build solar power generation facilities than to purchase electricity off the grid, so demand could make a quantum leap both in terms of size and speed of growth. At the same time, as grid parity becomes more realistic, FIT and other government subsidies will probably disappear.

Suppliers

• Single crystal furnaces and polycrystalline furnaces

Ferrotec uses some of its own products, such as vacuum feedthroughs, in manufacturing and assembly of the units.

• Quartz crucibles

The principal raw material is crystal and quartz found in nature. Ferrotec points out that although there are multiple places to obtain quartz, material prices and other market forces can influence the price.

Substitutes

• Comparison of clean energy types

Clean energy types that do not require fuel to generate power and can reduce greenhouse gases, include solar energy, wind power, geothermal, and wave power. However, because it is also possible that these types will offset each other, there could be trade-offs among the various types of clean energy. With geothermal and wave power, disadvantages include the fact that the structures are complex and the number of installation sites is limited. In terms of limited sites, wind power is the same way (in the US and other countries, the percentage of wind power plants is high, but this type of power is not suitable for areas where the wind does not travel in one direction). Currently, it is probably fair to assume that photovoltaic cells are most likely to see the widest practical use. The amount of energy available on Earth is estimated at 42 trillion kcal/sec for sunlight, 500 million kcal/sec for hydro power, 700 million kcal/sec for tidal power, 7.7 billion kcal/sec for geothermal, and 88.0 billion kcal for wind, wave, and ocean currents (Source: Japan Photovoltaic Energy Association)

• Comparison of photovoltaic (PV) cells

As stated in the "Business Description" there are a number of types of photovoltaic cells. SR Inc. understands that single and polycrystalline silicon have the largest impact on Ferrotec’s earnings (followed by the other types). Given this, SR Inc. thinks that the company’s earnings trends will vary with changes in demand for different types of silicon.

The following factors can be used to differentiate between types of photovoltaic cells:

1. Conversion efficiency
2. Effectiveness of source material
3. Possibility of price declines
4. Diversity of range of applications

Each type has unique advantages and disadvantages, so SR Inc. thinks that there isn’t necessarily a “one size fits all” solution for the market.

1. The conversion efficiency is an index that measures photovoltaic cell performance and in essence, the higher it is the better. The conversion efficiency of cells using single crystal silicon is the highest among all types, about 17-20%.
2. Effectiveness of source material. Sand and rocks are made of silicon, and in some sense is an unlimited resource (at least 28% of the Earth's crust). On the other hand, there is very little production of rare metals used in compound semiconductor photovoltaic cells, such as gallium and indium, and it is possible that these metals may be difficult to obtain in the future.
3. Potential price declines are indispensable to increased usage (and demand) of photovoltaic cells. However, there seems to be a number of problems preventing cost reduction for the most promising silicon photovoltaic cells. These problems include the fact that manufacturing high purity silicon is expensive, obtaining silicon can be difficult, and the overall silicon manufacturing process is complex. Although polycrystalline cells require a lot of silicon and the initial equipment investment is high, thin-film type photovoltaic cells seem to have a relatively high manufacturing efficiency and are lower cost.
4. Diversity in the range of applications indicates how many different ways the photovoltaic cells can be used when considering the materials, design, and environmental constraints. Silicon photovoltaic cells are at a relative disadvantage due to reduced conversion efficiency if the temperature exceeds about 30° C. Compound-type cells are heavy, tend to crack, and use toxic chemicals, creating a potential environmental impact upon disposal. Organic thin-film photovoltaic cells are light and pliant, opening up a wide range of potential applications, but there could be a tradeoff with stability because they use liquid.

Customers

• Single crystal furnace

Over 90% are manufacturers of wafers and modules for photovoltaic cells in China. The number one manufacturer of wafers and modules for photovoltaic cells in the world is China's COMTEC, who uses Ferrotec's products.

• Polycrystalline furnace

Manufacturers of wafers and modules for photovoltaic cells in China and Taiwan.

• Quartz crucibles

Manufacturers of wafers and modules for photovoltaic cells in Japan, South Korea, China etc.

Barriers to entry

The process of manufacturing photovoltaic cells is similar to the process for semiconductors or liquid crystals and when one considers the entry of manufacturers that are active in this area, it would seem that the barriers to entry aren’t very high. However, because photovoltaic production requires a different grade of silicon, developing the proper supply chain is actually a barrier to potential entrants. It seems fair to say that the technical challenges in the photovoltaic and semiconductor industries (bigger surface area for higher energy yields vs. boosting density to make faster processors) are different enough to conclude that although the business models appear similar, the keys to success are quite different. Furthermore, to stay in the market, new competitors would have to increase profitability after entry.

Competitors

• Photovoltaic cell-related manufacturing equipment

The main competitor is JYT (Beijing Jingyuntong, China), which has a global share of approximately 30%, making the company the global leader. Ferrotec has about a 20% global share, making it the worldwide #2, closely followed by Tianlong Photoelectric (China). There are about another ten manufacturers in China, but Ferrotec claims that these aren’t really in the same league. Ferrotec guarantees a “material use rate” of 96% (meaning that only about 4% of inputs, from raw materials through ingot production, are unavoidable waste) whereas other competitors guarantee about a 75% rating. Ferrotec claims that it maintains the highest material use rate in photovoltaic cell production equipment. Not surprisingly, there apparently is a 2x price gap between Ferrotec and other competitors. For that reason, the wafer module manufacturers try to make efficient use of Ferrotec's products and those that have a lower rate of material use.

Equipment efficiency (higher material use rates) can be an important consideration for manufacturers. For example, the minimum purity for photovoltaic cells is silicon with “Four-9s” (99.9999%), but “Six-9s” is the minimum requirement for single crystal silicon wafer manufacturing (but “Seven-9s” is ideal). When manufacturers use raw materials with low purity (Four-9s), yields are less important so typically equipment with lower material use rates are used. Similarly, when using raw material with high purity (Seven-9s) materials costs are higher so yields become more important. Ferrotec analysis shows that its products are mostly used with higher purity raw materials.

• Polycrystalline furnace

Ferrotec's share seems to be low, at just a few percent. The largest firm is GT solar (USA), with German firm ALD Vacuum Technologies in second place.

• Quartz crucibles

The firm has 30% of the worldwide market, making it the world's No. 1 manufacturer of crucibles used in photovoltaic cell manufacturing. Many of its competitors are in China. According to the company, wafer module manufacturers tend to use Ferrotec’s products when working with raw materials with high purity levels, a usage pattern similar to the company’s single crystal silicon manufacturing devices.

Strategy

Clearly summarizing the company's strategy in a few words is difficult because there have been many shifts in key industries in the past. Ferrotec’s current approach might be described as "create value by getting close to the customer and meet their evolving needs." This differs from firms that develop as narrow technology specialists in a certain area, and subsequently rise and fall with the particular technology’s lifecycle. Ferrotec’s overriding approach is to remain flexible in how it serves its customers and where it can find opportunities for growth, apparent when considering how the company has evolved throughout its history.

One clear area of importance for the company is the photovoltaic space. The company hopes to move beyond selling manufacturing equipment, incidental consumables, and other peripheral offerings like maintenance and wafer processing. The company’s wants to be involved in every stage of its customers’ businesses so it can offer optimized End-to-End solutions to its customers instead of single pieces. SR Inc. notes that Applied Materials (NYSE AMAT) successfully followed a similar strategy and became the global leader in semiconductor equipment manufacturing in the process.

In recent years, as the calls for reduced CO2 have been increasing, the demand for LEDs has increased (LEDs do not use much power) so the company is expanding its business in that area. It is also growing new businesses into the core businesses. The company’s acquisition from Edwards Vacuum, which allowed Ferrotec to sell a comprehensive solution to its customers, could be an example of this.

Historical Financial Statements

Summary

Expansion of Photovoltaic products helped offset weakness in other areas in FY03/09 and could be next area of long-term growth.

Balance sheet has had ample liquidity from FY03/06-FY03/10, with current assets nearly covering total liabilities for each year.

Strong operating cash flow supports investments in infrastructure and expansion of production capabilities.

Back to Top

Income Statement

Sales

Sales grew +35.8% in FY03/07 due to strong growth from all segments. An increase in capex by customers drove Equipment Related sales upward by 4.1 billion yen (+35.3% YoY) with equal strength in both vacuum feedthroughs and quartz products. Sales in Contract Manufacturing Service increased by 2.6 billion yen (+29.1% YoY) mostly due to silicon-related CMS and a new factory coming online. Increased demand from Japanese and foreign car manufacturers helped push sales in Electronic Devices up by 1.9 billion yen (+55.1% YoY).

In FY03/08, growth slowed in Electronic Devices and Contract Manufacturing Services, but was offset by performance in Equipment-Related sales. In Equipment-Related sales, vacuum feedthroughs were flat YoY following capex reductions by FPD and semiconductor manufacturers, but sales of quartz products increased 1.9 billion yen (+36.0% YoY), driving growth in the segment. Contract Manufacturing Services grew by approximately 388 million yen (+3.4% YoY) mostly due to success in the Photovoltaic business (wafer processing and other CMS areas were lower, but Chinese demand for Photovoltaic equipment offset the weakness). Electronic Devices grew 322 million yen (+6.1% YoY), driven mostly by demand from car manufacturers for the company’s thermoelectric modules used for car seat heating.

Sales in FY03/09 were flat YoY, but this obscures the fact that the company’s Photovoltaic products were the only area with growth (note that the company reclassified some segment sales and created the Photovoltaic segment in FY03/09). Photovoltaic sales increased by 6.4 billion yen (+137.6% YoY) due to demand from all areas. Photovoltaic production equipment sales were the main driver (specifically demand for polycrystal silicon production), but other areas helped. Equipment-Related sales declined 3.0 billion yen (-17.6% YoY); the biggest reason being falling demand for semiconductors used in cars and home electronics due to the global financial crisis. Sales for Electronic Devices declined 1.3 billion yen (-22.3% YoY) following production cuts by car manufacturers. Sales for Contract Manufacturing Service fell 2.1 billion yen (-22.6% YoY) mostly due to the global slowdown in the semiconductor industry, affecting wafer processing and equipment cleaning demand.

Sales in FY03/10 were down across all segments. Equipment-Related sales fell 2.4 billion yen (-17.1% YoY), mostly related to lower sales of vacuum feedthroughs (-1.5 billion yen; -27.3% YoY). When discussing FY03/10 results, the company commented that it thought the domestic market bottomed during the year, and that demand from both LED and FPD manufacturers was recovering. Demand from Taiwanese manufacturers for the company’s quartz products helped the segment recover in 2H to end the year down slightly (-310 million yen; -8.2%). Sales for Contract Manufacturing Service fell 1.6 billion yen (-23.2%). Demand for wafer processing was flat, but machine tool production and other CMS was sluggish for most of the year before showing signs of life near year-end. Sales of Electronic Devices fell 605 million yen (-13.9% YoY) after sales bottomed between the 2H of FY03/09 and 1H FY03/10. The main source of demand for Electronic Devices was automobile manufacturers, who were particularly hard-hit by the global economic recession in 2008-2009 (US’ firm GM declared bankruptcy and several global car manufacturers received government bailouts). Sales in the Photovoltaic segment declined 440 million yen (-4.0% YoY) following fewer orders in FY03/09 related to the global financial crisis and price competition. The company commented that it thought that sales in the Photovoltaic segment started to recover in 2H of FY03/10.

Operating Profit

The company’s operating profit peaked in FY03/08 and subsequently declined through FY03/10. The reason for the decline was the global economic crisis of 2008-2009 which negatively affected sales. Because SG&A expenses are relatively fixed, the key factor in the company’s operating profit is gross profit, which itself is a function of capacity utilization. Higher production volume means that per-unit fixed costs are lower and gross margin expands.

The decline in operating profit in FY03/09 was a result of higher SG&A expenses (the company absorbed subsidiary Ferrotec Ceramics). Although the company reduced SG&A spending in FY03/10, it was unable to scale back spending at a rate matching the fall in sales, which meant that operating profit margin was at the lowest levels since FY03/03.

Historical Performance vs. Estimates

The FY03/10 operating profit was below initial estimates due to lower gross profit and an increase in the SG&A to sales ratio. Gross profit fell 2.7 billion yen (-24.3% YoY) in FY03/10 because sales volumes, and therefore utilization, declined, affecting gross profit and margin. SG&A spending in FY03/10 was only reduced by about 655 million yen (-7.7% YoY, less than the fall in gross profit), so the SG&A to sales ratio increased and reduced operating profit.

Back to Top

Balance Sheet

Assets

The company’s balance sheet has had ample liquidity from FY03/06-FY03/10; current assets nearly matched total liabilities from FY03/06-FY03/09, and exceeded them in FY03/10. The mix between short vs. long-term assets has shifted over the past several years: initially most assets were long-term (about 60% of total assets in FY03/06), but this reversed by FY03/09. The increase in the company’s short-term assets has been a result of an increase in working capital from 5.5 billion yen in FY03/06 to nearly 11 billion yen in FY03/10.

Examining the working capital turnover ratio (sales / working capital) can show if changes in working capital have been efficiently used. From FY03/06-FY03/09 the company’s working capital turnover ratio has been relatively stable (from 4.3x - 5.0x), however declined to 2.9x in FY03/10. Accounts receivable increased significantly in FY03/10 (from 7.9 to 10.1 billion yen, +28.2% YoY), and the company commented that the increase was related to higher sales activity. Sales actually declined about 14% YoY, but looking at the 1H vs. 2H an improvement seems apparent: sales fell in the 1H by -31.3% YoY, but the reversed in 2H, growing +5.9% YoY. SR Inc. noted that the 1H FY03/11 results showed working capital turnover dropping to about 1.8x, mostly driven by increases in accounts receivable. When discussing FY03/10, the company suggested that overall sales were recovering. SR Inc. therefore thinks that it might be too soon to tell if the change in working capital is temporary due to improving sales or a new reality going forward.

Liabilities

The company’s liabilities have been mostly short-term obligations from FY03/06-FY03/10. Although the company has used long-term debt, its short-term debt has been a larger and more volatile component of total debt. Short-term debt nearly doubled in FY03/09 when the company needed to finance the purchase of a subsidiary, Ferrotec Ceramics. SR Inc. thinks that given the ample liquidity of the company’s balance sheet, the mix between short vs. long-term debt is less relevant than total debt load. Debt used to fund the acquisition of Ferrotec Ceramics pushed total debt up to 18.4 billion yen in FY03/09, but then decreased by about 3 billion yen in FY03/10, so it appears that the company is actively managing its total debt level.

Shareholders’ Equity

Changes in shareholders’ equity have generally reflected net income and dividend payments. The company raised approximately 3.1 billion yen in FY03/10 from a private placement of warrants (the number of shares increased by approximately 3 million shares).

Per Share Data

Increases in number of shares:

• FY03/08: exercise of warrants added about 1.1 million shares
• FY03/10: exercise of warrants added about 3.0 million shares

Back to Top

Cash Flow Statement

Operating Cash Flow

Depreciation and non-cash items have a significant impact on the company’s reported operating cash flow. Depreciation and amortization exceeded net income each year from FY03/06-FY03/10, so accounting earnings do not reflect cash earnings.

Working capital changes have also had an impact on stability of operating cash flows. Working capital needs have increased from 5.5 billion yen in FY03/06 to 10.9 billion yen in FY03/10; please see the Balance Sheet section for a discussion of working capital.

Investment Cash Flow

The company’s investment cash flow has reflected key characteristic of the business: the capital-intensive nature of its production process.

SR Inc. thinks that the company’s need for fixed-asset investment will likely remain a theme as more advanced products (like Photovoltaic technology) are developed.

Financing Cash Flow

The increase in financial cash flow in FY03/09 was mostly short and long-term debt used to pay for the acquisition of subsidiary Ferrotec Ceramics.

Simple Free Cash Flow

The company has been mostly consuming simple free cash flow from FY03/06-FY03/10. SR Inc. thinks that this illustrates the company’s policy of investing in future growth: despite volatility in accounting earnings, capex levels have been relatively more consistent. Although a certain level of capex is required to replenish the capital base, SR Inc. notes that if the company is able to invest in creating a dominant position in the growing Photovoltaic industry, first mover advantages could be significant.

Back to Top

Other Information

History

The company was established as a Japanese corporation of Ferrofluidics (USA) in 1980. Subsequently, the company experienced multiple crises, including conflicts with the former parent firm, Ferrofluidics, and the decline of major products. In 1987, Ferrofluidics attempted to sell off Ferrotec, but Ferrotec purchased the company in an MBO (management buyout). In 1991, the company established a corporate presence in the US. In 1992, the company attempted to revitalize its US business when the Ferrofluidics vacuum feedthrough patent expired, but Ferrofluidics brought suit and a legal battle ensued. Ferrofluidics attempted to buy out its now-larger subsidiary, but was unsuccessful. In 1999, Ferrotec purchased its previous parent, Ferrofluidics, through a TOB (takeover bid) and turned it into its subsidiary, Ferrotec (USA) Corporation. Another crisis began in 1980, when a change in hard disk technology meant that demand for a key product, computer seals (a magnetic fluid product used in hard disks), plummeted.

Looking at the company’s timeline, the company began manufacturing and selling magnetic fluids and related products (computer seals, vacuum feedthroughs), which would become the firm's core technology, in the 1980s. The 1990s saw the company escape a legal crisis while expanding overseas: entering the US in 1991, Hangzhou, China in 1992, Shanghai, China in 1995, Singapore in 1997, and Europe in 1999 (through the acquisition of Ferrofluidics). Since the year 2000, the company has continued to grow and establish new profit sources. M&A has been a key tool for profit growth, as have alliances and restructurings. The following serves as a short list:

Strategies for M&A and Alliances

Back to Top

News & Topics

Back to Top

Top Management

Akira Yamamura (Representative Director and President)

After completing a master's degree at Northeastern University in 1969, Akira Yamamura worked in R&D of magnetic fluids and thermoelectric modules at companies like Cambridge Thermionic in the US. Later, in 1979, he joined Ferrofluidics (now Ferrotec USA). He has served as president of the company since the founding year, 1980, and continues in that role today.

Back to Top

Employees

The company employs 145 staff at the parent company level (4,373 consolidated). Average age is 39.3 (parent company), average salary is 4.91 million yen (parent company) – data as of March 31, 2010.

Back to Top

Investor Relations

The company maintains an IR website, and publishes information in both English and Japanese. The company holds results presentations semi-annually.

Back to Top

By the Way

Back to Top

Latest Q&A

Back to Top