William Baker On Investing in Small and Mid-Cap Growth Stocks: ATMI and Cree

Includes: ATMI, CREE
by: The Wall Street Transcript

On December 18, The Wall Street Transcript interviewed GARP Research and Securities Company Founder and President William Baker on the topic on investing in small and midcap growth companies. Baker named a number of stocks his firm recommends -- excerpts from two of them, ATMI and Cree:

TWST: Let's start with ATMI. What is it that you like there?

Mr. Baker: In the interest of full disclosure, I must also tell you we have an affiliated money manager that owns each of these stocks, and I own them all as well... ATMI supplies materials and packaging products to manufacturers of semiconductors and flat panel displays. The technology markets have not been exciting for a number of years, but ATMI has had a number of efforts internally that are producing organic growth, which in the latest quarter was 18% top line.

The real driver of this is from consumables used in manufacturing semiconductors that have copper wiring, which is about 40% of revenues; in the last quarter, that was growing about 50% year over year. About two-thirds of copper comes from cleaning, with the rest in plating. Copper is only about 10% of industry wafer starts, so as it becomes predominant over aluminum, we're going to see continued growth on the order of 25% for a number of years to come. On top of that, ATMI has been developing a number of other products that will be introduced, and it feels each of them could be a major product. There are four or five of those, so that is very interesting. This is a high margin company.

Another part of the story is that margins are expanding, so operating margins for the whole year are 15% and we believe are headed up to 24% by 2009. Besides having all this top-line growth, you've got a ton of bottom-line growth behind it.

Lastly, we look out three years in terms of our valuation. It's trading at 13 times our 2009 estimate, which is reasonable for us. I would also want readers to be aware that the company has $6 a share in cash without debt attached, so if you knock that out, the valuation is even more reasonable. We do all of our estimates here at GARP after subtracting the cost of FAS 123(NYSE:R) - a lot of technology analysts don't do that, but we do. So if you are in the camp that thinks that options don't count, the stock's even cheaper. We think it's pretty silly that the majority of Street analysts like to give out free passes, but I've been in this business for over 20 years and some things never change.

TWST: What was the second name on the list?

Mr. Baker:
The second name on the list is Cree.

Is that another technology company?

Mr. Baker: Yes. We're very optimistic about technology and certain parts of the medical space. Going forward, we think there has been a five- or six-year compression in valuation and there has been enormous investment this decade set to yield significant earnings growth over the next four or five years for both of those sectors. So GARP has populated its recommended list with a lot of stocks in those two sectors, and Cree is among the most interesting.

One aspect of our philosophy that I didn't highlight at the beginning was that we're contrarian growth oriented. In other words, we will look at companies that the Street doesn't like that we still think will be acknowledged as great high quality growth companies in a couple of years, and Cree fits that mold. It is a maker of LEDs: little electronic lights made out of silicon carbide, a diamond- hard material that is a distant cousin to the silicon used in conventional semiconductors. 40% of Cree's sales are to backlight keypads in cell phones, and that market has gotten very competitive, which has sunk the stock near its five- year plus low. Nonetheless, Cree has a decent operating margin in it despite intense competition from Taiwan.

I think that the business is really changing, and going forward, the market is seeking higher value white light applications. These will honor a strong intellectual patent portfolio, and just as this is happening, Cree is emerging with better and better technology competitively. So we think that you will probably see LEDs in laptops very shortly, and then you'll see some trickling into the higher priced television sets over the next couple of years, but we're not basing the recommendation on that. Eventually, you'll see penetration of general illumination, but that could be four or five years down the road and will start out again as a specialty item.

TWST: What's going to wake investors up to Cree? As you say, it has been out a favor for a while.

Mr. Baker: The thing that will awaken them will be the laptop market. In October, Cree signed an agreement to license a company in Taiwan called Lite-On Technology. It's one of the world's largest producers of backlighting for LCD panels. Elsewhere, one of the Sony VAIO models has been using LEDs from a Japanese competitor, Nichia.

This year, I think that LEDs were getting competitive and intriguing to laptop manufacturers, but the fluorescent light makers saw it coming and cut prices by about 30% to basically block them out of the market for a little while longer. I think you can bar the doors temporarily, but eventually LEDs are like a battering ram that is going to knock the door off its hinges - it's just a matter of when, not if. And cost isn't the only issue; LEDs consume 30% less power, a gap that will only widen over time as efficiency improves, and this saves battery life - a goal that has been elusive and very much demanded by consumers. Think about how many LEDs might be inside a laptop panel that you look at compared to the little panel in a cell phone. It is geometrically much greater. So that's a really transformational event.

But even without that, we saw a real breakthrough, which was announced about two months ago, where Cree demonstrated in a lab setting a white LED with the efficiency of 131 lumens per watt. This compares to 80 lumens per watt for fluorescent or 15 for incandescent bulbs.

They've also come out with a product this year called the EZ Bright 1000, which is in a special category known as Power LED chips. One of the big competitors of Cree is a company called Lumileds, which basically began as a joint venture between Agilent (NYSE:A) and Philips (NYSE:PHG), but in 2005, Philips bought Agilent's stake, which valued the enterprise at roughly $2 billion. Bear in mind that the market cap at Cree is only about $1.6 billion.

Lumileds has had great success, and it has been isolated from the competition in the keypad space because it makes these very large LEDs, which are known as Power LEDs because they presently burn about one watt of power and are 1,000 microns in size, making these devices 3 times the size of cell phone LEDs. So Lumileds has great operating margins and Cree has demonstrated that it has a really terrific product in that space as well. So these would be used in really high brightness lighting applications.

TWST: So it has a couple of different opportunities?

Mr. Baker: Yes, it does. There is more than one opportunity, and oddly enough, I mentioned that we really don't hold out any hope for LCD TVs any time soon. But we've heard from multiple sources that a commercial design win might be announced as early as March 2007. So the product that it has developed for that is called Colorwave, and that's very interesting.

There is another thing that I would point out, which is that you should not think of this as the conventional semiconductor, because it isn't. It's made out of silicon carbide, and as we progress, instead of getting smaller, LED chips get larger. The wafer is getting larger in LEDs and in conventional semiconductors, but with LEDs, the chips themselves are getting larger here, particularly when you get general illumination or other end uses where you want a big, bright device.

Probably 10 to 15 years ago, Cree produced its chips on 2-inch wafers. It shifted up to 3-inch wafers in Q3 of 2004, and today, about 85% of its output is now on that. In about March 2007, it will convert to 4-inch wafers. The interesting thing that has dogged the company's return on capital - even though margins are pretty good, this is a capital intensive process - is the yields. They're not anywhere near as high as what you see in traditional semiconductors, and this applies to the whole LED industry as well as Cree. Bear in mind that Cree's gross margins are still around 50% in keypad LEDs due to its technological advantage, even though this sector is hotly competitive. Back in July 2006, Cree spent $46 million to buy a company called INTRINSIC Semiconductor, which has very important technology that really improves silicon carbide yields. So we could see a dramatic competitive advantage develop, sort of like what Nichia used to enjoy before last year when Cree improved its phosphor technology and caught up in brightness. Cree could really step up and totally differentiate itself if it can implement this technology. So if it can distance itself and get to 4 inches, Cree would have a different cost structure from the industry.

There are a lot of people who have shorted this stock. The hedge fund community loves to short it because these hedge fund guys pay to fly analysts over to Taiwan where they get firsthand reports that the Taiwanese will dominate the industry. I don't know if you've ever talked to a Taiwanese businessman, but they're optimistic people and they have an axe to grind. They won't hesitate to say, "We will have $100 million of sales by 2007." Then 2007 rolls around, and they say, "Sorry, but in 2008, we will have $100 million of sales." I am not denying there is cutthroat Taiwanese competition, but it may be running its course. Taiwan's cameo role may have been mistaken for the symptoms of simply being in the last year of the trend to put LEDs into phones, sort of like what happened to red LEDs a decade or more ago.

If you're in the sixth year of competing in keypads for blue LEDs, Taiwanese competition is an important thing to think about. But if you're going into white light for laptops, where someone has to build laptops and market them into the US market, the US makers of laptops - Dell, Hewlett-Packard, and so forth - are not going to use patent-infringing LEDs. Last year, cross royalty agreements among the five leaders of this industry, reinforced by infringement lawsuits against everyone else, knocked out any sort of stealing of IP when you have big brand names marketing into the US. Some guy in the middle of China can use a laptop that infringes, but so what? If you want a Dell, a Compaq, or a Hewlett- Packard, you're just not going to have an infringing LED, and that's the market that's going to count.

So we could see all of the people short on the stock because they're worried about Taiwan looking in their laptops and saying, "Where is the infringing chip? Oops, there aren't any. Oh gee, I thought there were." Then you see Cree licensing one of the largest producers of LCD panels in Taiwan. They're going to be paying Cree a royalty. Oops, that wasn't supposed to be under control, either.