Advanced Energy Powered Down For Now, But Still Undervalued On An Eventual Semiconductor Recovery

About: Advanced Energy Industries, Inc. (AEIS), Includes: CHLDF, MKSI, XPPLF
by: Stephen Simpson, CFA

A weak quarter and lower guidance is pretty much par for the course with semiconductor-exposed names now, but AEIS's weak revenue in industrial applications was disappointing.

AEIS has been through cycles before and the long-term outlook for semi-equipment is fine, but I believe AEIS should start considering more aggressive capital deployment (M&A) options.

AEIS shares look undervalued on that eventual recovery, but there could be another downswing from here as the cyclical looks for its bottom.

Down a little bit from my last write-up (during which time the stock swooned another 25% only to regain most of that), Advanced Energy Industries (AEIS) is a tempting target in a sector undergoing serious pressure and downward revisions as semiconductor companies, particularly in the memory space, slash their capex budgets and semiconductor equipment companies cut their orders for critical components like power supplies, matchboxes, and plasma generators.

This year (2019) is likely to be ugly, and I’m not ruling out the possibility of another round (or two) of guidance revision, but at some point the sector will bottom out, and I think AEIS’s leadership in power subsystems is worth more than the share price currently reflects and I think management has the capability to do something a little more dramatic on the capital allocation side.

A Weak Quarter And Another Cut To Guidance

AEIS’s fourth quarter wasn’t so surprising relative to heightened awareness of just how weak the semiconductor equipment sector has become, and the company’s 14% yoy (and 11% qoq) decline in revenue was only about 1% worse than the Street had expected. I was, however, disappointed to see both the semiconductor and industrial businesses down by similar amounts (13% qoq and 14% qoq, respectively), and it underlines the struggles the company has had in truly diversifying the business and applying its technologies and capabilities to industries outside of its traditional semiconductor and panel core.

I would argue that gross margin has held up decently relative to the revenue pressure, with non-GAAP margin down 560bp year over year and 60bp qoq, but the damage to operating earnings was more severe given the company’s going SG&A and R&D spending. Operating income (again, non-GAAP) fell by more than half from the year-ago level and by more than a third from the third quarter, with almost 16 points of year-over-year margin contraction and seven points of quarter-over-quarter contraction.

AEIS’s guidance for the first quarter was broadly similar to what most chip companies guided, which for revenue to decline around 7% qoq at the midpoint. That was considerably worse than expectations, though, and the 11% revision relative to the prior average estimate was certainly disappointing even if not entirely surprising in light of MKS Instruments’ (MKSI) guidance.

Memory Not So Persistent…

The biggest issue for AEIS, and semi equipment makers is general, is the sharp decline in orders from memory chip manufacturers, and NAND manufacturers in particular. Memory is about two-thirds of AEIS’s semiconductor business, split roughly in half between NAND and DRAM, and weaker sales of end-use products like smartphones, coupled with rapidly-improving yields as chipmakers travel up the learning curve with advanced architectures, have hammered demand for new chipmaking equipment.

What makes that situation even more challenging is that large semi-equipment OEMs don’t like to (and to have to) carry significant levels of component inventory on hand. That tends to lead to even more whipsawing for component providers like AEIS and MKS, and it also partly explains why the company’s business with Applied Materials (AMAT) and Lam Research (LCRX) has shrunk in recent quarters (though I think there could also be some share loss/pressure at Lam from Comet (OTC:CHLDF) ).

Longer term, all of the arguments that used to apply to AEIS in the good times still do apply. The transition from planar NAND to 3D NAND increased the company’s content by 2.5x due to higher power needs as well as demand for ancillary products like sensors, matchboxes, electrostatic chuck (“e-chuck”) and so on. With about one-third of the NAND market (in bit volume terms) still using planar architecture and opportunities to gain content as chipmakers transition to higher numbers of layers, there’s still incremental content growth opportunity for AEIS. This also applies, to some extent, to opportunities in logic, with more sophisticated architectures demanding more processing/manufacturing steps and more components.

Moreover, AEIS remains well-positioned in the market. AEIS appears to have about twice the share of MKS in its core semi-equipment power market, and efforts to broaden its base seem to be working, with the company significantly improving its market share with a large Korean customer (probably Hynix or Samsung) and winning ion implant business away from a competitor.

Of course there are still threats in the market. MKS, XP Power, Daihen Kyosan (Tokyo Electron’s primary supplier), and Huttinger aren’t going away, and Comet is looking to stretch beyond its market-leading positions in matchboxes and vacuum capacitors into plasma generation.

Time To Do Something Bigger?

AEIS management has talked for some time of its desire to expand beyond semiconductors (and panels) and repurpose its core technology for other markets, while also acquiring other useful and complementary technologies. There are applications for AEIS’s technology (including its relatively recent Lumasense acquisition) in areas like medical and analytical equipment, both of which are healthy markets, as well as robotics, but the reality is that the progress here has been slow.

The company has a decent amount of cash and I expect it to remain firmly cash flow positive during this downturn, so it’s time to ask if that cash could be put to better use. Both Comet and XP Power (OTCPK:XPPLF) could be “gettable”, though both would require debt and/or equity to do the deal. Comet would add leadership positions in complementary adjacent markets (like matchboxes) within semiconductors, as well as capabilities outside of semiconductors in X-ray and CT scanning and ebeam surface treatments. XP Power would add significant diversification in power supplies outside of semiconductors.

I don’t follow either company closely, and neither may be inclined to sell, but I do believe it underlines the general notion that there are some assets out there that AEIS could look to acquire. At the same time, I can understand why AEIS management may be reticent – the company’s foray into solar inverters was a failure and it’s easy for an analyst to say “you should buy something”; if it doesn’t work, the analyst moves on to another idea and the company’s management has to deal with the fallout.

The Outlook

Although I think a deal for Comet or XP could be value-additive under the right circumstances, I make no assumption of M&A in my model. Speaking of my model, AEIS ultimately came in within 2% of my midyear estimate for 2018 revenue, but missed on FCF by about 25%. Between that experience, what I see going on in the semi-equipment market, and management’s own guidance, I’m reducing my near-term revenue and margin assumptions.

I expect a double-digit decline in revenue next year, but I think FCF will hold up relatively well. Longer term, I expect mid-single-digit revenue growth and high single-digit FCF growth.

For the valuation, I still believe long-term cash flow can support a fair value around $70, but the near-term outlook is tempered by the weaker revenue and margin outlook for 2019. Stocks in the semi-equipment space tend to trade closely on current/near-term margins and AEIS isn’t all that cheap relative to the margin trough I see in 2019. On the other hand, using an average of the last three years and the next two years (my estimates), I believe a fair value in the mid-$50’s is fair. Admittedly that’s a non-standard approach, but I view it as a reasonable compromise between using a trough or peak number.

The Bottom Line

That big dip in the fourth quarter of 2018 might have been panic selling as the last holdouts finally accepted the reality that semi-equipment demand really was falling, but I’m surprised at the speed of the rebound from those December lows. I can’t rule out another valley, though, as some analysts and investors do seem too cavalier about taking a second half recovery for granted. With that, there’s a risk of another cut to numbers here (and maybe more than one) that takes the stock down again. Still, I do believe the market will recover at some point, and the long-term potential is enough that I’m increasingly willing to take the risk of being too early to buy in near a bottom.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.