Description

High quality/defensive: We want to be owners of Entegris (ENTG) as it’s a high-quality, defensive, EPS compounder and will out-grow semi volume growth due to trends of increasing purity and material intensity in semis.

CMC Acquisition: We think the benefits of the CMC acquisition could surprise on the upside and is not fully appreciated by the market.

Inflecting financials: Financials are poised to improve: EPS growth will accelerate meaningfully in late 2023 driven by revenue growth, high incremental margins, cost saves, and continue into 2025 as debt is paid down.

Agentcooper2120 did a great write-up on ENTG in February 2021, much of the core points of the thesis still stand, the aim of this post is to provide an update with focus on the CMC acquisition and near-term inflection points.

Description:

• Entegris (ENTG) is a high-quality chemicals and filtration company, it just happens to sell into the semiconductor market, and is therefore covered and classified as a semi-cap equipment company

  • ENTG supplies specialty materials and advanced filters, both of which are becoming increasingly important to semiconductor advancement as Moors Law scaling has become tougher

• The business is volume-driven as consumables are 80% of revenues which are tied to wafer starts / volume growth in semiconductors

• Post CMC acquisition ENTG reports in 4 segments:

  • Specialty Chemicals and Engineered Materials (15% of profits)

    • Provides high-performance and high-purity process chemistries, gases, and materials and delivery systems to support semiconductor and other advanced manufacturing processes

  • Micro-contamination Control (38% of profits)

    • Solutions to filter and purify critical liquid chemistries and gases used in semiconductor manufacturing processes and other high-technology industries. One of only 2 players

  • Advanced Materials Handling (17% of profits)

    • Solutions to monitor, protect, transport and deliver critical liquid chemistries, wafers and other susbstrates

  • Advanced Planarization Solutions (29% of profits)

    • Sells slurries, pads, conditioners and electronic chemicals used in chemical mechanical planarization (CMP). CMP is using chemical oxidation and mechanical abrasion to remove material and achieve very high levels of planarity in the making of semiconductors

• Top 10 customers approximately 45% of sales

  • TSMC > 10% of sales, Samsung Electronics is 9-10% of sales

Reasons to own Entegris:

Business Quality:

• We consider ENTG a high-quality company

  • 20%+ operating margins with consistent attractive returns, in an industry notorious for high levels of deflation

  • Consistent FCF generation provides ability to pay down the debt. Historically FCF has been used for accretive acquisitions. Since 2014, the management team have been good allocators of capital

  • Low competitive intensity in most of the businesses with Entegris having #1 / #2 market shares

    • E.g. in the filtration side (38% of profits), Pall Corp, owned by DHR, is the only large competitor. Pricing is less of a threat

    • Most of ENTG’s competitors are verticals of large companies, that have displayed discipline and return focus over the years

M&A:

• We think benefits from the CMC acquisition could surprise on the upside and is not fully appreciated by the market

  • We see potential acceleration in market share as CMC means a step-change in scale, although this is unlikely to become evident until 2024

Growth:

• We like the long-term growth opportunity:

  • Estimate potential through the cycle 15% pa EPS growth over the next four years:

    • Underlying semiconductor industry growth likely to improve meaningfully from here. Historically semi downturns have lasted less than a year

    • Entegris will outgrow semi volume growth driven by trends of increasing purity and material intensity, coupled with node changes

Valuation:

• Derating of stock on the semiconductor cycle is another misunderstanding in our opinion as ENTG is an industrial consumables company that happens to have semi manufacturers as its customers

  • ENTG is impacted by the cycle, but it is a highly defensive company with 80% exposure to consumables

• Financials are poised to improve:

  • EPS growth will accelerate meaningfully in late 2023 driven by revenue growth, high incremental margins, cost saves, and continue into 2025 as debt is paid down

  • Capex is peaking, and will decline going forward aiding FCF

M&A:

Some background:

• Entegris was founded in 1966 as Fluoroware, merged with Mykrolis in 2005. Entegris then spent a decade being unremarkable, until they acquired ATMI in 2014

  • ATMI deal was important as it created a leader across multiple chemistries, at the time that Moore’s Law scaling was getting harder

• Subsequent investments in R&D and M&A used to add products to different steps in the semi manufacturing process

• Entegris has the most complete solution in the industry

• 2022 CMC materials acquisition only adds two product steps, but they are important ones and have relatively large addressable markets

CEO / Entegris learning curve completed:

• Current CEO, Bertrand Loy was promoted from COO in November 2012. Stock had essentially traded sideways for a decade. (Gideon Argov prior CEO for 8 years.)

  • Entegris’ history of M&A was poor. In the prior years, the company had spent $1Bn in M&A and not created any value (EPS or stock price)

    • Loy joined the company in 2005 as part of the merger with Mykrolis, where he had been the CFO

  • Under Loy, the company has focused on R&D, built an attractive niche selling materials to the semi industry

• M&A has become and increasingly important driver of growth

• CEO / Entegris has become good at M&A:

  • They buy high profitability / cash generating companies that are technology leaders serving the ENTG customer base

  • According to Loy, the Mykrolis merger suffered from inefficient integration – he was in charge of the integration efforts

    • His objective: to have 90% of the functional integration plans done by the time of closing. Within a week of closing he wanted all employees to now their job status and to provide those who would not have a long-term role with specific facts of their transition plan, severance package and benefits

• Taking the average of the deals over the M&A model is as follows:

  • Pay 3x revenues, integrate the business as fast as possible

  • Better distribution, customer relationships and broadening product assortment gives much of the upside

  • Accretion: for every $100MM in revenues, approx. 20c accretive to EPS after two years

M&A has been a compounder:

• We think M&A has become a core competency for ENTG

• Generally it takes a couple of years for acquisitions to accrete to EPS fully

  • Revenue synergies and cost synergies

  • Incremental margins: 35% incremental EBITDA margins

• M&A has been focused on tuck-ins. Over the last 5 years deals have added value on multiple levels:

  • EPS, due to benefits from scaling distribution and lower operating costs

  • Higher organic growth, guarantee of future design-in as materials evolve.

  • Improved customer relationships, strategic importance

  • Higher market share

• E.g.:

  • Sinmat: leader in the design and production of Chemical Mechanical Planarization (CMP) slurries used for polishing ultra-hard surface materials including SiC (silicon carbide) and GaN (gallium nitrate)

    • SiC and GaN are substrates utilized in power electronics and advanced communications

  • Anow: Chinese filtration company acquired in Q3 2019

    • Brings new membrane technology and liquid filtration and manufacturing capabilities in China

    • Growth linked to node transitions in advanced logic and advanced memory, and we assume better relationships with FAB buildoutds in China

    • Had been a long-time supplier of some membranes to Entegris

CMC is a little different

• The most recent acquisition of CMC was large (unusual) and materially increased ENTG indebtedness – something to be aware of/conscious of

• For the CMC deal management announced:

  • $75MM of run rate cost synergies expected to be realized shortly after the migration to a common ERP platform by early summer 2023

  • ENTG management’s target is $200MM of revenue synergies by 2028. Generally, we are highly skeptical of revenue synergies, but in this case we find it more believable than we usually would.

    • Within a 2- to 5-year timeframe, co-optimized new products  will be developed and introduced to the market

    • These revenue growth benefits are included in ENTG’s organic growth algorithm of 3 to 6 points of revenue growth above the market

• In 2019, Entegris offset cyclical declines in revenues by increased level of acquisitions, perhaps CMC is a similar bet

  • Historically ENTG has managed to do accretive deals, while seeing high incremental EBITDA margins, and an increasing company-wide RoE. In early 2019, ENTG walked away from the Versum acquisition despite a compelling strategic and financial rational due to a Merck counterbid. ENTG showed focus on returns, they could/would not pay up.

• Are CMC benefits under-appreciated? This is a key bet generating upside to EPS estimates / owning Entegris.

• We think that the CMC deal is a step change in ENTG establishing a scale advantage over competitors that will give it the opportunity to accelerate growth over the next three years

  • Similar to ENTG, CMC has a customer collaboration model to address the most critical CMP challenges from leading customers

  • Opportunities:

    • Provide better and integrated and optimized CMP solutions, which is becoming critical with increased complexity of requirements

    • Improve margins and profitability of CMP, especially the consumables of slurries and pads

    • Content per wafer is likely to accelerate further with integrated solution with CMP for node transition

    • Increase CMP market share in pads: 10% share in pads and a distant second position

    • Optimize solutions with CMC. This should increase the moat of ENTG, leveraging scale and cross sell

      • Competitors are focused on individual end markets and are often divisions within large organizations

Some CMC Materials background:

• CMC had good financials before they did a large (KMG) acquisition, CMC then struggled with execution until December 2021, when it was bid for by ENTG

• In 2021, pre-takeover announcement by ENTG, CMC had underperformed ENTG by 52% and SOXX by 40% and was trading at a 20% P/E discount as compared to a premium in the past

• ENTG paid mostly in cash for CMC, with ~$3.8Bn out of the $5.7Bn purchase price funded by debt; ENTG also assumed $700MM of CMC debt. The remainder was share funded

• CMC had been struggling due to the KMG acquisition in FY19, which was paid for by increasing CMC debt, plus the pandemic impact

  • We think the main reasons that CMC struggled were poor execution, lack of ability to increase margins, and underwhelming growth in the Electronic Materials segment which was underperforming peers especially ENTG

  • Pipeline and Industrial Materials (PIM) segment (10% of sales) was struggling mainly driven by weak Drag Reduction Agent (DRA) demand used in oil industry.

    • This division has now been sold by ENTG (Octobe 2022), to Infineon for $240MM (EBITDA ~$25MM)

• While CMC is leader in CMP slurries, it has been successful in pads, which were only 8% of revenues

  • This is where we believe an opportunity lies for ENTG to improve/integrate the product in its solution and increase market share

• CMC does bring advantages to ENTG namely end-to-end capability and scale

• ENTG will be the only company having this end-to-end capability, which should enable ENTG to shorten the time to solution

• Having shorter time to market in the semis space is critical for Entegris’s customer to gain market share

Growth:

• ENTG benefits from industry growth in volume, historical volume CAGR of 5%

• 2019 and 2023 had negative growth, but they are likely only bumps in the road, which we are attracted to

  • Historically, recession driven semi downturns have lasted 1 year

  • 2019 downturn was not very severe, 2023 may be much more severe, but difficult to call

  • Forecasts suggest a muted volume recover, given strong digitization tailwind we doubt this will be correct over the long term

• Over the past five years annual wafer starts have been ~2ppts higher than silicon million of square inches. This is the relevant number for ENTG (~7% CAGR)

The Covid Semi-Cycle:

• The cycle is now normalizing with the traditional inventory glut and reduced production occurring in 2023

  • Fabs are generally unable to sustain utilization rates of 80% for any extended period of time, Covid was the perfect storm

• Covid demand:

  • Buying surge for laptops and other electronic devices

  • Strong cycle of demand for chip upgrades driven by new technologies such as cloud services and AI services

• Covid supply:

  • Global lockdowns/labor shortages that forced chip production facilities to shut, depleting inventories

    • It has taken a long time to reach the other side – we are now getting back to 80% capacity utilization

    • Some sectors decision making had a particular influence on them. E.g. autos – who cancelled orders at the beginning of the pandemic. By the time the auto industry saw an increase in customer demand and reinstated their chip orders, the foundry space had already filled with other products

Market growth drivers:

• Cyclically there may be a pause in the growth, but long-term new end point growth means that semi growth is approximately 2x GDP

• The average life cycle of a typical semiconductor device is approximately three years, which includes introduction, design-in, production, low-volume, end End-of-life (EOL) phases

• Clearly not all semis will get replaced over 3 years, however the volume of semis needed to keep infrastructure up to date will continue to grow

  • E.g. faster growth markets such as Data Centers must grow dramatically to keep up with the volume of data being collected. 5G plus IoT means that connected chip-enabled devices are expected to grow at 20%+

• US-China tensions continue to have repercussions. The last few years have been particularly strong for announcements relating to new capacity expansions in less risky geographies:

  • Intel, Samsung and TSMC have pledged to build more than $300Bn of new capacity in the coming years, much of it in Europe and the US

• The US Chips Act:

  • Originally introduced in June 2020

  • $39Bn manufacturing incentive program to revitalize the US chipmaking ecosystem

  • 25% advanced manufacturing investment tax credit as a complement to the manufacturing incentive program. Together these incentives will lower the cost gap between investing in the US and investing abroad

• The European Union

  • In February 2022, the European Commission began formal consideration of “the EU Chips Act,” which includes up to $43Bn in targeted support for Europe’s semiconductor sector

• South Korea:

  • In May 2021, South Korea unveiled the “K-Semiconductor Belt” strategy aimed at building the world’s largest semiconductor supply chain by 2030

Smaller nodes is a secular tailwind

• Materials science is increasing in importance to semiconductor companies

• Scale: smaller dimensions (nm node reduction) require more pure error free chemicals / finer filters

  • 10^-9 types of purity needed

• Moore’s Law running out of steam:

  • Certain parts of semis now scaling via stacking/3D manufacturing, rather than just because of Moore's Law

  • Node shrinks are more expensive and performance increases are coming form other factors, rather than just Moore’s Law

    • E.g. adopting new device architectures, such as FinFET transistors and 3D-NAND to increase transistor and bit density

• As each technology node becomes more complex, to enable improvements and to maximize yields, manufacturers require the development and application of new materials.. This is a tailwind for ENTG

  • Note that R&D is 8% of ENTG revenues. This is necessary to stay relevant as semis evolve their product and increases barriers to entry as local R&D and collaboration with customers is critical

  • Newer layered architectures increase the number of process steps required to manufacture semiconductors

  • More lithography, deposition, CMP, etc. required to manufacture them = more chemicals

Higher content per chip

• Each generation of processes require higher level of purification

  • Structures are becoming more complex, more prone to contaminants

  • Extremely difficult to reach the purity levels that advanced nodes are requiring (10^-9)

  • The risk of yield loss grows with the incremental manufacturing steps needed to produce these devices (e.g. bowing in 3D NAND)

• Smaller nodes, new materials and greater materials intensity, means that more process steps are being used to make new chips

  • This means the industry is seeing proliferation in points of filtration across the upstream ecosystem and supply chain benefiting ENTG

  • The largest portion of material increase is in deposition.

• Over the next few years there should be a tailwind to ENTG as production increases at newer nodes

  • TSMC Arizona has started the construction of a second fab which is scheduled to begin production of 3nm process technology in 2026

  • Arizona: Fab 1 is on track to begin production of N4 process technology in 2024

  • The overall investment for the 2 fabs will be approximately $40Bn (2022 price)

Valuation:

Discussion on the debt profile:

• The company has gone from low debt levels to “bears quite a lot of consideration” levels of debt due to the CMC acquisition

• Our position is that the quality of the business means that it is manageable, but even ending 2025 ENTG will still be 2.2-2.5x ND / EBITDA

  • However, the maximum balance sheet risk is in 2023 if the cycle downturn is a lot worse

  • The second negative is that accretive growth via M&A is ruled out for at least the next three years

• Currently: Net debt of $5.2Bn

  • Equals 4.2x pro-forma EBITDA including the $75MM cost synergies

  • PIM sale ($240MM) is not included. Deal expected to close shortly.

• ENTG has hedged a portion of the floating rate debt beginning in January 2023.

  • The initial hedge amount is $1.95Bn and ramps down to zero over the next three years. (80% of term loan)

  • Also has a $275MM, 364-day bridge loan. ENTG expects to pay down the remainder of the bridge loan between now and year-end 2023

• Target for 2025 is to pay down $1Bn in debt from FCF. Plus additional paydown from unit divestitures, which is $240MM so far

  • We estimate $1Bn debt reduction is just possible, assuming dividend stays at $60MM

  • Because of this, interest payments fall from ~$332MM in 2023 to $264MM in 2025

  • There should also be a small benefit in blended rate as the most expensive debt is the bridge loan followed by the term loan

    • Bridge loan has a rate of little over 7%

Some additional points on the assumptions:

• Incremental margins are ~35% in our upside scenario. We used this three years ago in our assumptions and don’t think anything has changed

  • We think that scale will enable a very modest improvement to gross margins.

  • Plus, typically in semi-downturns memory fabs shift volume to the leading edge where they make better margins, and ENTG has slightly higher materials input

• Capex peaks in 2022

  • We have done our best to estimate proforma numbers, as ENTG has been in a capex cycle

    • ~$500MM capex guidance for CY23, also ~$115MM in non-cash incentives committed by state/local authorities

• There is negative sales impact related to China trade restrictions. ~$50MM.

• Segment margins are not comparable with the past:

  • Many higher margin products have been transferred out of SCM to APS and low margin products have been added from legacy CMC

• Working capital turns will be less efficient due to an increase in raw materials inventory, (sector-wide for the near future), we would expect this to normalize over the longer term.

• Management guidance:

  • Organic sales growth (2022-2025) semi unit growth: 2x GDP, pus 3-6ppts from share gains

    • Higher materials content per wafer (new nodes)

    • More wafers produced at leading edge

  • 40% incremental EBITDA margins

• M&A off until 2026-2027, because of debt levels

Valuation:

• Upside:

  • Our bottom line is that street EPS estimates are too low for the out years

  • Once the market realizes that ENTG can compound EPS at ~15% CAGR for the next three years, we think the multiple will remain at 20-22x given the defensive consumable revenue

  • Applying 20x PE to numbers that demand excellent execution and a reasonable economy

  • Upside estimate: $5.70 * 20x P/E = $114 (~40% upside)

• Downside:

  • $3.90 in “normalized” EPS, thereafter EPS growth of 15% fading towards 10%

  • Downside estimate: $3.90 * 18x P/E = $70 (~10% downside)

Catalyst

• CMC integration
• Improving financials by YE
• Growing demand for complex/tech intensive/niche consumables
• Market share gains