They paid down $426M in debt and boosted gross margins to 47.6% in a flat revenue year. Imagine what happens when volume actually shows up.
Thesis: MKS is a classic 'coil spring' setup. They used the downcycle to optimize the cost structure, delivering 190bps of margin expansion on flat sales via procurement and factory efficiency. They are not a pure AI play; they are a leverage play on the memory cycle and packaging complexity. With NAND inventory normalizing and 'green shoots' appearing, any top-line recovery will drop heavily to the bottom line due to the leaner cost base. The debt paydown ($426M in 2024) removes the bear case's structural overhang.
Verdict: LONG — Conviction: MEDIUM
Catalyst: NAND inventory normalization converting to upgrade orders in 2H 2024, driving high-margin RF power sales.
Key Risk: Net leverage at 4.3x leaves them vulnerable if the memory recovery is delayed or WFE contracts further.
The Tell: When pressed on AI exposure, John Lee admitted that AI servers are only '10%, 15% of the entire IC substrate market' and that the other 85% is PCs/non-AI servers. It reveals the 'AI' growth story in packaging is largely a bet on a broader PC/Server cycle recovery, dressed up in a leather jacket.
Friction Level: MODERATE_FRICTION — Bulls see structural growth in packaging chemistry (50 layers below the chip). Bears see a legacy PCB business rebranding itself as AI to justify a multiple re-rating.
Report not found
The report data is no longer available. Please return to the archive.