A 25% EPS Compounder Trading at 14× Earnings?
The German Automotive Chipmaker Quietly Beating the Auto Cycle
Key Metrics
P/E is around 14, historically above 20.
ROIC is 21.6%.
EPS CAGR over last 10 years is 24.9%.
Expected industry growth 8% CAGR
All currencies in the article are in Euro’s.
The Business
Today’s business is Elmos, active in the automotive semiconductor industry. It is a fabless semiconductor designer founded in 1984 in Dortmund, Germany.
The company partnered initially with the local university and produced the first chip for a household appliance, the iron.
With a sales strategy and the involvement of BMW, the company made the decision to design and produce exclusively for the automotive industry.
Since BMW, Elmos added many car companies to its customer list. They developed an expertise in analog mixed-signal sensors. This allowed them to become major players in small automotive niches.
Elmos has several product families. These include ultrasonic parking sensors, thermal measuring integrated circuits (IC) and LED drivers for rear lights and grill lights.
The company’s workflow is as follows. The customer has a design request. Together with the engineers at Elmos, they create a chip design. Thereafter, a foundry fabricates wafers (layers of silicon, simplified). Elmos receives the wafers, assembles and tests them. Finally, the chips are shipped to the customer.
Fabless Transition
The company’s workflow used to be slightly different, but had a large financial footprint. Instead of finding a production plant, they used to do all the production in-house.
Owning a production plant or known as wafer fab within the industry, is capital intensive and a huge strain on the cash flows. The maintenance and keeping it clean is the burden.
Owning a wafer fab for a small product offering, increases the price of produced chips. It happened that the equipment was used for less popular chips. Because of lock-in contracts, it wasn’t easy to switch to more profitable chips.
As a relatively small player in the automotive semiconductor industry, it was harder to have competitive pricing.
Situation like this made Elmos decide to outsource the production of chips.
10 Years ago, Elmos decided to go fabless which means not owning the fabrication plant. Examples of such companies are Apple, Nvidia and Qualcomm.
Fabless companies have contracts with third-party fabrication plants, what are called foundries. A well-known foundry is TSMC, the Taiwanese Semiconductor Manufacturing Company.
The former CEO started the transition to what is called fab-lite. Elmos had multiple fabrication plants and sold them off in steps. This situation created a hybrid form where one part of the production is outsourced and to other part is produced in-house, hence fab-lite.
The current CEO continued with the plans and sold off the last fabrication plant successfully in 2023. Hereby, the transition had completed.
Why would Elmos go fabless?
As explained earlier, it is very capital intensive. Owning fabrication plants are a distraction for smaller designers and sellers.
Elmos is now able to focus on R&D, customer service and marketing.
Another great benefit is for Elmos to become more competitive as it reduces the cost per chip. As foundries spread the amortized cost over multiple clients.
It prevents using expensive machinery to produce competitive chips, the lock-in. It also gives the company more flexibility to scale up or down the production.
Customers
Elmos sells its chips to tier-1 car companies and original equipment manufacturers (OEM). It started with BMW and over the years Elmos gained trust and reliability within the industry.
Building a name as a supplier for car companies is a slow and painstaking process. Their approach to design on request helps significantly.
It is unclear who Elmos supplies. The car companies nor Elmos is transparent about the supply chains.
A little digging shed some light on the question. A recent press release in November stated that Elmos shipped the 2 Billionth ultrasonic IC. And “…Elmos ICs can be found in almost all common ultrasonic applications in vehicles…” Thus, we can conclude that Elmos sells to almost all major car brands.
I also found BMW owning a minority share in Elmos. It isn’t conclusive. However would they keep the shares when they aren’t satisfied with chips of Elmos? I don’t think so.
Margins Development
The earlier described transition from wafer fab to fabless changes cash flow within the company.
Starting with the income statement, the gross profit and operating income margins improved. The whole automotive semiconductor industry experienced a boost from supply chain shortages. Revenue and margins grew for many within the industry.
The boost ended in the latter half of 2023. Car manufacturer inventories were sky-high and needed reduction. 2024 and early 2025 saw less orders within the industry.
During this period, Elmos surprisingly managed to increase revenue marginally, while competitors experienced a decrease in sales.
At the moment, inventories are low and some car manufacturers might run inventories too low to cushion a supply shock. A real treat without an international pandemic is always around the corner.
Nexperia, located in the Netherlands, is an semiconductor company providing chips to Volkswagen and others. The company was taken over by Chinese in 2018. And since a few months, the dutch government seized control and has taken management control.
The Chinese owner shut down the production process place in china as retaliation. It was a precarious situation for Volkswagen as it was very close to closing a production line.
Situations like this raised awareness for manufacturers to increase inventory again and is expected to boost 2026 sales.
Operating margins
Operating margins improved because of enormous sales growth. Also, the sale of wafer fabs resulted in fewer employees. This in turn requires less managers, reducing much of the salary.
Depreciation of the wafer fabs is also excluded from the books. The income statement will show better margins because of it.
The flip side is added outsourcing production costs covering a part of the savings.
The operating margin increased in the last decade significantly. The transition effects haven’t materialized fully yet. Thus, it is hard to estimate what normal operating margins for Elmos will be.
If we compare the operating margins to other similar fabless semiconductor companies, we see that these operating margins are sustainable.
Gold price
The gold price has a stark influence on the companies increased raw material costs. Hence the lower gross profit margins.
Elmos changes legacy designs with gold traces into copper. It is a relatively slow process as engineers have to test the new designs, rewrite specifications for chips in manuals etc.
The soaring gold price increased Elmos’ costs by more than €10 million. They expect to replace all designs by copper in 2026. Saving them more than €8 million.
Free cash flow
The sale of the last wafer fab has an impact on capital expenditures. The high maintenance capital expenditures aren’t necessary anymore. It results in a higher free cash flow available for future growth investments or return to the shareholders (yay).
In the previous 2 years, Elmos made payments regarding the unwinding of the last fabrication plant. The upcoming 2 years are definitely lower CapEx years. The CEO made this clear in a recent earnings call.
He explained that CapEx in this business is cyclical. And the increased spending happens in 2 years.
Competitor Melexis provides us with an indication as to what FCF margins we can expect form Elmos. 90% of Melexis’ revenue comes from automotive semiconductors. They design less on request and have a different product mix with high profit margins.
It results in a staggering FCF of around 17% a year. (But unfortunately don’t tend to unvest it very wisely right now)
The different products and design process lower Elmos’ profitability. But higher margins are certainly possible. FCF margins around 10 to 12% are more likely.
However, analyzing the FCF of Elmos isn’t that easy. The CapEx is cyclical. Meaning that there are years with less investments and years of high expenditures.
For example, Elmos invested heavily in testing facilities in China the last 2 years. This indicates that they expect to scale up production or new chip designs.
As investors, we like these type of investments. They fall under growth CapEx. The FCF margin doesn’t go up.
However, the growth CapEx is a part of the FCF. It brings us future returns when company’s value increases. See it as a sort of hidden FCF. When you know the story, the numbers come to life and get a different meaning.
CapEx is the same of maintenance and growth capital expenditures. Maintenance costs keeps the company in the current state. Growth expenditures obviously grows the company’s capacity to make more sales and profit from it.
Earnings per Share
The further one goes down the income statement, the less real the numbers may be. It is tricky to make investment decisions solely based on the income statement. But, I would like to look at the exceptional earnings growth over the last 10 years.
The company delivered 25% EPS CAGR (compounded annual growth rate) over 10 years. Compared to others in the industry it is unusually high. As explained earlier, the growth has to do with the supply shock and the cost reduction of the company.
If Peter Lynch would look at the stock, he would come to the following conclusion. With a current P/E of 14 and an EPS growth of 25%, the PEG would be a 0.56.
This is far below the 1, were the price per share meets the past growth. It is slightly flawed as past growth isn’t a guarantee for future growth.
As Warren Buffett would say:
In the business world, the rearview mirror is always clearer than the windshield.
Management
The current CEO, mr. Schneider, is on the board since 2014. He started as the chief finance officer. And in 2021, he took over the role as CEO and his term ends in 2030.
The previous CEO, mr. Mindl, was originally an physicist and served as CEO for over a decade. It indicates that the company looks for serious hires looking with long term tenures.
Mr. Mindl started to move the company towards a fabless future. He couldn’t finish the plan within his term. The new CEO continued execution.
It demonstrates that the company as a whole decided on a direction and not some pride project of a sitting CEO. Definitely a plus for investors!
The founders of the company are still on the board as chairman and vice chairman. They are in their 70’s and usually I am not a fan of older people in such positions. Some tend to be to conservative and hold off on progression.
However, since they are the founders they want to build a long sustaining company. And it shows with moving away from wafer fabs.
Another positive is how the CEO, executives and the board handled the pandemic. The company was having a bad year filled with lockdowns. They decided to voluntarily to reduce their salary for the year. A great sign of leadership and alignment with the company.
While most employees worked on reduced hours with full pay, they had the brightness to let R&D work full hours. Because the future of the company is dependent on R&D.
Ownership
Approximately 59% of that outstanding shares are hold by insiders. This is unusually high for a public company.
The 2 founders on the board own together 37% of the shares. Another individual coming from an entrepreneurial family holds 17%. The family is affiliated with the founding of the company too.
Then there are various executives, the current and former CEO who own shares.
This distribution leaves just over 40% for outside investors. They have no grip on the direction the company is heading to.
Normally this can cause some friction. Yet, I think that it favors investors in this situation. Management is very well aligned with the investors. They have skin in the game too.
The founders built the company from 1984 and have faced many difficulties along the way. They seem to know what they are doing.
Growth
Industry reports from Mordor Intelligence and Global Market Insights report that the industry till 2034 is likely to grow around 8% CAGR.
They recognize Asia as the fastest growing market. China is expected to grow slightly above 10% and India even 11%.
These numbers outpace the typical economy growth of 6% annually.
Elmos has an presence in Asia since the early 2000s. In the last few years, they invested heavily in test facilities in China. Chinese companies love to buy chips from local vendors.
Knowing this, Elmos set up a subsidiary in 2024 and gained many sales fairly quickly. It goes at “China Speed” as the CEO likes to say. Germans are known for slow office processes with a lot of bureaucracy.
… we operate at China Speed.
Thus seeing the CEO and management adapting quickly to a local market like that is a great for its investors.
Last year, Elmos also dipped its toe into the Indian market. Unfortunately, they have not yet elaborated on this.
I hear you thinking…
Why would the industry grow at such an incredible rate?
The automotive industry goes more and more electric. The current trends in the industry show the electrification and efficiency of passenger vehicles.
Drivers prefer more assistance systems, sensors and autonomous driving. They also want comfort premium features. These features are usually electric. Think about how you now adjust your seat with buttons instead of manually.
These trends require new and more chips. Something the semiconductor industry is happy to help with :)
Then there is the trend to go from the combustion engine to electric. Europe wants to prohibit the sale of new combustion engines. However this plan has changed a few times. Dates have changed and for who the rules apply exactly.
Car designers and manufactures want to move small local electric control units (ECU) to a central computer. Which is called software-defined vehicles. Future vehicle updates happen over the air then.
All these trends give support to a fast growing industry. Of course, markets like China and India, see an influx of drivers on the road. Prosperity allows for a larger part of the population to buy a. Car. The more cars sold, the more chips are required.
How can Elmos grow revenue?
The increasing product mix and new inventions, like the eFuse, grow sales over time. It attracts customers with new requests too.
They also move into new niches to create a larger product offering. In 2020, Elmos acquired a software company to offer a complete product of software-defined chips. Since car manufacturers move towards software-defined car designs, moving with them in the same direction creates growth.
The wafer fabs limited production and the number of offerings. The company is now free to increase both.
Competitors
Elmos has direct and indirect competitors. The industry is divided by company size. Industry giants like NXP, Infineon and Texas Instruments deliver to automotive and other industries.
They go after semiconductor productions that require larger investments and production quantities. Microcontrollers, battery management, motor control and telemetry are areas of interest to them.
The return on investment to break into small niches is too low. Such small niches are ideal for companies like Elmos, Allegro and Melexis.
Elmos has expertise in mixed-signal chip designs. They have over 40% market share in ultrasonic parking sensor ICs.
Other niches they have a meaningful market share in is LED drivers for rear and grill lights. Maybe you have seen the grill & logo lights on new models from Volkswagen. They also produce thermal sensors for EV cars. These tend to get hot and need proper sensors to control overheating.
The earlier mentioned competitors like Allegro and Melexis have other areas they excel in. But these companies also have revenue coming in from selling to other industries. Whereas Elmos is 100% focused on the automotive market.
Moat
Most semiconductor companies selling to tier-1 car manufacturers experience the same moat. The nature of the development process guaranties steady cash flow for many years to come. And once in production, no competitor can step in that easily.
The process starts with the request by the customer. The design team works alongside the customer to create a desired design. This typically takes 1 to 2 years.
When the car goes into production, the lifespan is typically 5 to 8 years before making alterations. Although it depends on the model. Afterwards, spare parts should be available for some years after the last production car rolls out of the factory.
During the model’s lifespan it is unlikely for another competitor to swoop in and become the new supplier. This protects many years of cash flow which every supplier experiences with their own products.
Elmos has another edge, which is operating in small niches and being the best. They excel in the mixed-signal expertise as explained earlier.
They developed the first ultrasonic parking sensor in the 1990s. Their 40% market share is protected by the small size and their expertise. The small size leaves less room for many competitors. Especially large companies find the return on investment low when they have to start from zero in a specific small niche.
Becoming fabless reduces the costs of their offerings. This allows them also to be more price competitive if needed.
In conclusion, I think Elmo has a narrow moat they can solidify by becoming more price competitive. Also keeping their R&D spending higher to maintain the expertise in mixed signal designs.
The Risk
Just like every company there are some standard risks related to cyber security, supply chain strains, tariffs wars and so on.
I would like to highlight the risks of going fabless as it introduces new risks for the company.
The reliance on suppliers becomes higher. During times of supply shocks, it might be harder to find available foundries to produce the chips. When producing small chip quantities, the prices are higher, than for larger quantities.
The production process in not entirely in-house anymore. The design schematics are shared with foundries. Which means an increased risk of counterfeits.
And we all know how China is a country of producing cheap copies.
It is unlikely for tier-1 manufacturers to purchase cheap copies. But another company can gets its hands on the design and start producing a competing product. This obviously reduces Elmos’ revenue.
The competing company Melexis is since the early 2000s fabless and has always managed to contain those risks. Thus it is certainly possible for Elmos manage to do the same.
Cyclicality
Everyone knows the cyclical nature of the car industry. People naturally change cars after many years of service. I recently learned, there are stark differences per European country in how long people tend to drive newly purchased cars.
Car manufacturers sell on average more cars every year. But there is a cycle where some years many cars get bought. Then there are slower years as well.
Logically, you would assume the automotive semiconductor industry experiences the same cyclicality. They are indeed not immune to it.
Yet, the industry as a whole is very resilient due to the growing number of chips required in cars, which offsets the cyclicality.
Elmos gets its revenue 100% from the automotive industry and seems to manage to grow almost every year. They do certainly better than competitors delivering to several other industries.
Conclusion
The deep dive was mainly building a case for Elmos move towards a fabless company. There is much potential for investors to profit from increased free cash flow in the coming years. It definitely allows the company to grow faster. Or return more money to the shareholders through dividends and share buybacks.
Apart from the special situation Elmos is in, the industry’s growth prospects look also promising.
Thus, I think that the coming years look bright. And think the company is an undervalued company at current prices when you take the growth potential into account.
Disclaimer
The information provided in this newsletter is for informational purposes only and does not constitute financial, investment, or professional advice. All opinions expressed herein are solely those of the author and are based on publicly available information as of the date of publication. Investments in stocks and other securities involve significant risk, including the potential loss of principal. Past performance is not indicative of future results. Readers are encouraged to do their own research and consult with a qualified financial advisor before making any investment decisions. The author and publisher of this newsletter assume no responsibility for any errors, omissions, or outcomes related to the use of this information.
By reading this newsletter, you acknowledge that the author is not liable for any decisions made based on the content provided.