Magnachip Semiconductor (ticker: MX) has finally fled the forests of the “complicated” and “difficult to value” and is now frolicking in the fields of simple pure-plays. To be sure, MX has a checkered past…it was initially spun from Hynix Semiconductor in 2004, filed for bankruptcy in 2009, and capped it off with a fraudulent CFO who employed a panoply of accounting tricks to overstate revenue/earnings from 2013-2016 (MX eventually settled with the SEC).
Since then, multiple protracted attempts to explore strategic alternatives has left long-suffering shareholders bruised and battered, even as operating results weren’t bad.
Now, with the spectacularly successful sale of their Foundry businsess complete, and with all their woes finally behind them, the story at MX has gone from remarkably complex (leveraged small cap with a tainted history and lots of operating leverage tied to half the business with extremely cyclical end markets) to remarkably simple (pure-play fabless semi-conductor company operating in secularly growing high-growth end markets with no debt, no pension, and in pole position to capture ever more share via their attractive OLED technology).
MX’s equity has been written up 3 times on VIC (please see previous write-ups by aviclara81 and greenshoes93 for great background), but it has been almost 4 years since the last write-up and a lot has happened since. When the company reports Q3 results sometime in October, MX will suddenly hit screens as a fast growing secularly advantaged semi company with ~45% of its market cap in cash.
Sale of Foundry Business –
In early 2019, MX launched a review of strategic alternatives for its foundry business. A little over a year later, after the market had all but lost faith in mgmt’s ability to get a deal done, MX announced the sale of the Foundry to a private equity consortium for $350m in cash plus the assumption of substantially all of the company’s significant pension liabilities, as they were tied to employees of the Foundry being sold.
In addition to fetching a price well above even the bulls’ blue sky scenario, the sale of the Foundry accomplished the more strategic imperative of separating the highly cyclical foundry business from the secularly growing standard products business. Several quarters of stellar operating results from both businesses and a stagnant stock price made it clear that the market was simply never going to understand or properly value the mixed business.
With the sale of the Foundry having closed one week ago, MX has now emerged from its torturous, messy history and is a much cleaner, simpler investment opportunity for the market to digest.
Pro-forma numbers –
Unfortunately, the announcement and conference call that should have been a major victory lap left shareholders somewhat confused as to the earnings power of the remaining business. MX had reported segment results on a gross profit basis, but investors weren’t sure how to allocate all the opex to the two businesses, and instead of being ready to discuss this issue in detail, mgmt. informed the market they would be hosting an analyst day in November, where they would lay out longer-term growth/margin targets for the remaining business.
Thus, investors have been left to their own devices to speculate about the margin profile of the remaining display / power business.
Remain Co (Standard Products Group) –
The remaining business is divided into two business lines:
Display Solutions – provides panel display solutions to major suppliers of large and small rigid and flexible panel displays, and mobile, automotive applications and home appliances.
Power Solutions – provides discrete and integrated circuit solutions for power management in communications, consumer and industrial applications.
Embedded in the Display business is MX’s OLED business. The widespread adoption of this advanced display technology across all manner of consumer electronic devices has resulted in 40+% CAGR in this business over the past two years. The adoption by consumers worldwide of electronic devices for their daily use continues apace, as consumers increasingly access a wide variety of rich, media content. With the company’s headquarters in S. Korea, MX sits close to their largest customers and operates at the core of the global electronics devices supply chain.
Margin profile –
While MX’s CEO (YJ Kim) has given some color on margins going forward, his comments have been vague and, at times, entirely unhelpful. He has a long history of sandbagging guidance in order to “beat” estimates, but MX’s small size and minimal analyst coverage has rendered this strategy pointless. He appears to have understated yet again the margin profile of the remaining business, hence part of the opportunity.
Investors will get more color at the analyst day in November, but in the interim we can make some educated guesses. Below, we annualize depressed revs/ebitda for the Standard Products Group (depressed because of Covid in China) and apply incremental margins of 25% to 2019’s more normalized revenues to arrive at an additional $10.5m in Ebitda. In addition, Q1 2020’s Ebitda number still has some corporate overhead allocated to it from the Foundry that will go away now that the sale has closed. We estimate this at an additional $5m annually. IR has told us the above assumptions are directionally correct. Hat tip to Aviclara81 for his work here in previous comment threads.
Complicating matters further, YJ gave out an adjusted operating income margin target that is measured against total revenues, which include some inter-company revenues with no margin attached to them because they have offsetting COGS.
We expect a cleaner and simpler presentation of the margin profile of the business going forward in November, and expect significant margin expansion as their many recent design wins achieve scale, although we don't contemplate any of that in our model below.
Pro-forma capital structure –
We use the treasury stock method to reflect stock options and treat the convert on an as-if converted basis since it is in the money. We also give affect to the Foundry proceeds taxed at 15%, per YJ’s comments on the call. Finally, we ignore the pension in our net debt calculation since the remaining liability after the Foundry sale is minimal and we are not adding back pension expense to Ebitda. Cash is currently 44% of the market cap.
MX is cheap:
We think the above estimates and multiples embed conservative assumptions about margin potential and the growth runway. We take further comfort in several recent insider buys by officers that have not bought stock in almost three years.
Why does this opportunity exist?
- Small cap
- Minimal analyst coverage
- History of accounting irregularities
- Based in S. Korea (where they keep 90% of their cash)
- Poor communication from mgmt / understated margins
Bottom line, going forward ~55% of this business will be OLED, which grew at >100% in Q4 2019 as Covid began its full frontal on humanity. That such a business, with such an operationally competent mgmt. team, pristine balance sheet, and long growth runway should trade at a mid-single-digit multiple of near-term Ebitda makes no sense to us.
I do not hold a position with the issuer such as employment, directorship, or consultancy. I and/or others I advise hold a material investment in the issuer's securities.