Photronics: The Undervalued Photomask Giant That’s Poised for Growth
The following is a condensed report by the Long/Short Equities Team at Alumnus Capital
Thesis:
Photronics is severely undervalued due to low analyst coverage, lack of knowledge/coverage of the photomask industry, and cyclicality of the semiconductor industry. This has created a big disconnect between PLAB’s share price and company value with an opportunity to buy an excellent company for cheap. All of their valuation metrics are low historically compared to competitors. This has happened while their Revenues will be YoY up, while most if not all of the semi-industry will be down 10–20%+. This is due to the photomask industry operating differently compared to the rest of the semi-equipment market.
About Photronics
Photronics (PLAB), engages in the manufacturing and sale of photomasks and services in the United States, Taiwan, China, Korea, Europe, and internationally. The company offers photomasks that are used in the manufacture of integrated circuits and flat panel displays (FPDs); and to transfer circuit patterns onto semiconductor wafers, FDP substrates, and various types of electrical and optical components. It sells its products to semiconductor and FPD manufacturers, designers, and foundries, as well as to other high-performance electronics manufacturers. The company was formerly known as Photronic Labs, Inc. and changed its name to Photronics, Inc. in 1990. Photronics, Inc. was founded in 1969 and is based in Brookfield, Connecticut.
What are Photomasks:
Photomasks are used to transfer circuit patterns onto semiconductor wafers and FPD substrates. Photomasks are sold in sets comprised of layers, with each layer having a distinct pattern that is etched onto a different photomask. The resulting series of photomasks is then used to image the circuit patterns onto each successive layer of a semiconductor wafer or FPD substrate. The increasing complexity of chip designs means that more layers will have to be used, increasing the end cost (in 2006 photomasks could vary in price from 250–100k, but now the most advanced EUV photomasks can cost over 300k). This cost is due to the complexity in the designing and manufacturing of photomasks, but has resulted in increasing margins for companies like PLAB (this will be talked about later).
How are Photomasks manufactured:
The typical manufacturing process for a photomask involves the receipt and conversion of circuit design data to manufacturing pattern data. A lithography system then exposes the circuit pattern onto a photomask blank. The exposed areas are developed and etched to imprint the pattern on the photomask. The photomask is then inspected for defects and conformity to the customer’s design data. After the repair of any defects, the photomask is cleaned, any required pellicles (protective translucent cellulose membranes) are applied and, after final inspection, the photomask is shipped to the customer.
The key ingredient for photomasks are the high precision quartz plates (including large area plates), which are used as photomask blanks and “are primarily obtained from Japanese and Korean suppliers; pellicles and electronic grade chemicals, which are used in the manufacturing process; and compacts, which are durable plastic containers in which photomasks are shipped. These materials are sourced from several suppliers. We believe that our utilization of a select group of strategic suppliers enables us to access the most technologically advanced materials available. On an ongoing basis, we continue to consider additional supply sources. We typically enter into annual pricing agreements with our suppliers, some of which include volume-based incentives that have resulted in substantial cost savings; these agreements do not require us to purchase minimum dollar amounts or quantities of their subject materials.” 10-k
Photomask Industry:
“We believe that the demand for photomasks depends primarily on design activity rather than sales volume from products using photomask technologies. Consequently, an increase in semiconductor or FPD sales does not necessarily result in a corresponding increase in photomask sales (but it can in certain cases). In addition, the reduced use of application-specific ICs, reductions in design complexities, other changes in the technology or methods of manufacturing or designing semiconductors or FPDs, or a slowdown in the introduction of new semiconductor or FPD designs could reduce demand for photomasks ‒ even if the demand for semiconductors and FPDs increases.” 10-k
But on the contrary, if design activity continues to become more advanced and competitive, the demand for photomasks could stay flat or grow even if the end market demand for the products slows or drops.
The TAM for the total Photomask industry is currently ~7.5B which is expected to CAGR at ~4% to 10–12B in 2030. With EUV expected to grow even faster (10%+), PLAB currently controls roughly 11% of the TAM and I expect that they can continue to take market share due to their favorable position and strategic partnerships (DNP joint venture in China) and connections (UMC, Samsung, TSMC). Only 4 photomask companies can make EUV masks which are PLAB, DNP, TOPPAN, and HOYA. These masks are a lot more high-tech and therefore are a lot less commoditized; as the foundry rebounds these will also provide growth as they will become even more crucial and expensive as we go sub 3nm.
CAPEX outlook for 23 and beyond also looks to be pretty solid, with UMC forecasting Y/Y 11% CAPEX growth and TSMC forecasting around 32B. These figures will most likely continue to grow in coming years with TSMC confirming this in their recent earnings call. They also stated this might grow at a slower pace but I think this will not materially affect PLAB because it was mainly due to capacity which photomasks are not involved in. This will most likely affect ASML, KLA, and LRCX (this is another reason I like PLAB).
IC Photomask (74% of total revenue, 641m TTM) :
The photomasks for ICs are classified under two metrics, high-end (28nm and below) and mainstream 28nm (and above). The high-end photomasks consist of 28nm, and 14nm, which can be used to manufacture 28 and 14 nm chips. Another segment of high end is EUV photomasks which are 7nm and 5nm photomasks that are used to make the most advanced chips such as TSMC’s 3nm (5nm photomasks are used in 3nm chipmaking by scaling the mask 2.7x to etch the features). Mainstream is classified as anything above 28nm. Mainstream currently makes up around 70% of revenues which tend to be less cyclical and more solid growth while high-end are more cyclical and fast growing. The high-end will likely grow a lot more in future years as the ramp-up of EUV photomasks and when new EUV lithography technology is released in FY25.
FPD Photomask (26% of total revenue, 230m TTM) :
The photomasks for FPD are also classified as High-end which means Generation 10.5+, AMOLED, and LTPS display-based process technologies for FPDs. As well as mainstream for Generation 8 and below (excluding AMOLED and LTPS) display process technologies. High end makes up around 84% of total FPD revenue with mainstream at 16%, high end has continued to take share from mainstream which was at 24% of total FPD revenue in FY21.
Q2 23 & Outlook:
For Q2 23 total IC photomask revenue was up 7% Q/Q and 15% Y/Y, with high end down (9% Q/Q, 14% Y/Y) and mainstream up (13% Q/Q, 30% Y/Y), high end down due to foundry/memory in Asia. While mainstream demand resumed at a high level during the quarter.
Outlook was that “new designs by customers to gain market share and support technology roadmap contributed to improved high-end orders toward end of the quarter” This could continue as inventory levels normalize. Also “Regionalization driving investments for global chip capacity” and “ASPs continue to expand, reflecting the demand/supply balance”. The photomask supply shortage and increasing complexity are leading to high price points as stated before.
For Q2 23 total FPSD photomask revenue was up 14% Q/Q and 6% Y/Y, with high end up (14% Q/Q, 11% Y/Y) and mainstream up (18% Q/Q) and down (14% Y/Y), High-end grew as mobile display sustained high AMOLED demand and Mainstream expanded with additional write capacity. Outlook was that “technology development drives demand for higher-value masks” and “Mobile demand is expected to continue strong with the rollout of new designs”
Fort Knox Balance sheet:
PLAB has an extremely strong Balance sheet with $367m in cash and $45m in short-term investments (treasuries) and $28m total debt with ~$8m short-term and ~$21m long-term. They have net cash of $384m with a 1.5B market cap which means they could theoretically repurchase ~25% of outstanding stock. They list this Fort Knox balance sheet as “resilient to economic contraction and able to fund investments, share repurchases, and strategic opportunities”. They currently do not have a dividend and have decided to return to shareholders via repurchases instead, which may increase rapidly soon (read below).
Financial covenants expiring (The lower coverage of PLAB also means that even fewer people know about this more hidden catalyst):
“Financial covenants related to our credit facility, which expires in September 2023, include a total leverage ratio, a minimum interest coverage ratio, and minimum unrestricted cash balances. We are also subject to covenants that limit our financing flexibility, such as a limit on the amount we can spend to repurchase shares of our common stock. Existing covenant restrictions limit the amounts of dividends, distributions, and redemptions we can pay on our common stock to an annual amount of $50 million.” This combined with their large cash position could allow them to repurchase 200m worth of shares which would be 13% accretive to EPS and would leave them with 200m+ in cash to have on hand (expires September earnings in late August, last quarter they generated 70m straight into the balance sheet if continues they would have 450m in cash).
Margin appreciation & growth:
PLAB has seen major growth post-2021 but even better margin appreciation, gross margins have improved 1200 basis points (FY21-FY23) to 38% and EBIT 1400 basis points (FY21-FY23) to 28% with more appreciation in FY24 (gross possibly 400+ basis points on the higher end of guidance). This is the main result of increasing photomask prices, demand, and increasing CAPEX spending.
EPS has more than doubled from $0.89 in 2021 to $1.99E in 2023 (~12x P/E) and the high end of guidance for 24 is $2.6, which would imply a P/E of around ~9x. This is incredibly cheap (this will be covered later) considering that the EPS growth for FY 24 will be between 18%-33% YOY. Continued EPS growth from top-line growth and margin expansion will likely continue for years to come.
Competition:
Geopolitical risks (South Korea, Taiwan)
Direct write lithography (is not commercially viable so not a real threat at the moment) or other technology that can render photomasks obsolete
Photomask challenges due to 2nm and smaller could affect the growth of the industry
The Catalyst:
The company uses large amounts of excess cash post-covenant expiry to repurchase large amounts of stock
Increased analyst coverage of the whole industry and PLAB (will happen as the TAM gets bigger and PLAB’s market cap increases over time)
Earnings Beat (guidance is definitely on the lower side which could trigger a combination of beats)
