December 16, 2011 - Lam Research (LRCX) and Novellus (NVLS) are pulling off the industry's second major M&A of 2011: a $3.3B blockbuster for which the industry apparently has been rooting for years. Given the semi equip sector's, ah, checkered past in M&A activity, why all the celebration this time? "You have 2 blue chip companies that are wall street sweethearts. Not quite the union of Annette and Frankie, but close," quipped Gartner's Dean Freeman. Echoed VLSI Research's Risto Puhakka: "Both Lam and Novellus are extremely well run companies. Both of them are extremely profitable in upturns and maintains profitability through downturns."
Scanning industry analyst reactions, here's a list of some consensus questions and answers:
Why does this deal make sense?
By customers, LRCX is heavy in memory and foundry, NVLS is strong at logic-- and with logic taking up more of the industry's projected capex in 2012 (>50% for Samsung-Intel-TSMC, 30% for Samsung alone) it behooves LRCX to get into those accounts, notes Citi's Tim Arcuri. NVLS can help LRCX finally establish inroads to Intel for etch and clean business, possibly displacing Hitachi High-Tech and TEL -- and now's a good time because Intel is making new equipment decisions at the 14nm node. From a customer perspective, this deal should mean better customer service and better upfront development, notes Freeman.
By technology, LRCX brings etch and clean to the table; NVLS brings thin-film deposition and wafer surface prep. Freeman also notes that Lam gets better access to etch/dep processes for client-creating unit processes (especially for multipatterning litho), and that combined the two also have a play in through-silicon vias (TSV) as an alternative to AMAT's TSV process.
Deutsche Bank's Vishal Shah points out the combination is better positioned for increased R&D spending for 450mm (Arcuri notes LRCX's run-rate for 450mm is already $100M by YE2012), and customers' growing desire for more integrated process technologies. Uncertainty about 2012 capex spending, recovering semicap stocks, and increased 450mm R&D activity from competitors (e.g. AMAT) also spurred action, he says.
Barclays' CJ Muse pulls together the likely combined marketshares:
Marketshare by segment, in US$M 2010 revenues. (Source: Dataquest, Barclays)
Several analysts point out that there's still a competitive wall around the litho processes and suppliers, who represent the lion's share of capex spending and increasing capital intensity -- that means ASML and Nikon, plus associated steps & suppliers TEL and KLAC.
What about investors?
Shah thinks LRCX "got a good deal" especially given what seems to be a low 28% premium -- half of the 55% VSEA garnered 55% from AMAT. Arcuri echoed that, saying he "always thought" NVLS would ask for ~$60/share."
Citing investor feedback, Credit Suisse's Satya Kumar suggests some see the deal as "significantly dilutive" for LRCX; that LRCX should have simply bought back stock; and that NVLS investors (and the ambulance-chasing lawyers as usual) might be happier with a higher premium, to the point that they wonder if playing-hard-to-get wasn't a better option.
To reiterate how good a deal this is for LRCX, several analysts compared the LRCX-NVLS value to other recent semicap industry M&A. Here's Kumar's comparison (all others were cash deals):
(Source: Company data, Credit Suisse estimates)
Integration concerns, combined metrics
Almost four years to the day from Lam's purchase of SEZ Group (unfortunately timed at the start of the downturn), this deal has more similarities going for it, Freeman says: the two companies very familiar with each other (partners since 1991). Still, one of the real challenges here, as in most M&A, is in the details of merging the corporate cultures ("Lam has a culture of operational excellence. Novellus will need to get use to this") and regional sales efforts ("keep the needed client relationships to be able to gain share in the areas you don't already have penetration").
This would also appear to be the swan songs (legacies?) for both companies' high-profile top execs, Rick Hill and Steve Newberry. Will their exits be a help or a hindrance to integration the two companies, both organizationally and culturally?
Several analysts pointed out that NVLS has historically underinvested in R&D, so LRCX might have to ratchet up its own opex to offset -- but "on the other hand, LRCX "could certainly use a dose of opex discipline," Kumar writes. He also foresees LRCX using NVLS' manufacturing capacity in Oregon, among other synergies. Kumar also addresses questions whether promised synergies might underestimate integration/opex risks, pointing to an upcoming strong order cycle (1H12) and a projected LRCX-NVLS value-accretion of $400M, equal to the current cost synergies.
Citi's Arcuri breaks down the combined company metrics:
EPS scenario for combined LRCX-NVLX, assuming $28B in wafer-fab equipment spending.
(Source: Company Report, Citi Investment Research and Analysis)
Responses from rivals?
As post-M&A share prices fluctuate (LRCX down, NVLS up) and the valuation shifts a bit, it's possible that TEL could sweep in with a counteroffer (possibly an all-cash deal; Arcuri notes TEL's $3.6B in cash and a favorable JPY/USD rate) but analysts seem confident that there won't be any third-party spoilers or sweetheart-deal-revisions. Kumar suggests LRCX-NVLS is a better fit than TEL if only for geographical concerns -- and if TEL was interested it would have already done something at $44/share, and now would have to dig even deeper.
Should we expect any additional M&A aftershocks? LRCX-NVLS creates a bigger, broader competitor, particularly in deposition; "AMAT will need to step up their game to make sure they don't lose share," Freeman says. TEL, he added, "loses out on having a partner for the interconnect part of the process, so they are still basically a front-end company." ASMI also seems to be on the outs and might need a partner... maybe those two phones will start ringing?