Skip to main content
Message Font: Serif | Sans-Serif
No. of Recommendations: 50
Long awaited CREE report: (Note: I know too much about this company - please use given links to do more DD and we can discuss merits more specifically to make more sense of this long post)

There are two distinct elements I look for in a tech stock and my recent reading of Geoffrey Moore's Gorilla Game is the best terminology I have found to explain this. The terms are (1) continuous innovation to the customer and (2) discontinuous innovation in the technology.

The first means that the new technology is not so complicated that customers won't adopt it. For example, one of the many reasons I am so high on Gemstar is that it is a complete and utter continuous innovation to the customer. People have watched TV excessively for generations. People have flipped through TV Guides for generations. What Gemstar does is improve the process in an effortless manner in which the customer says “cool” Not kidding host of CNBC last week when Gemstar came up said is that the TV Listing deally mcbob....”cool.” {to paraphrase}. The technology will be quickly adopted.

At the same time GMST is a discontinuous innovation technology wise as it moves the printed TV Guide – with enhanced functionality - at less cost and more profit opportunity - to the television set totally superseding the prior technology.

The invention of the cellular phone is another recent example. We use phones, not a great leap to use cellular phones. Continuous innovation to the customer, discontinuous technology to the fixed phone. Allah, another killer technology.

For a technology to Tornado in the market place these two components must exist: continuous innovation to the customer, discontinuous innovation to other technologies.

But a continuous innovation to the customer and a discontinuous innovation to the technology does not always translate into profit for the first moving company. The best example I can think of is GUI, the Macintosh like graphical operating system invented by Xerox and then first implemented by Apple. MSFT was really the third mover to GUI when it came out with Windows. And the first version of Windows was rather weak in comparison to what Macintosh produced. So why did MSFT take home most of the rewards? Value Chain.

The volume of people intricately dependent upon the inarguably inferior Windows was much larger than the chain of people dependent upon Apple's OS. From Intel to software developers to users, the switching cost from Windows to the better Macintosh was very, very, very, high. Therefore the value chain supported the later arriving inferior product. In fact the value chain is even larger than that because behind it is the value chain of dependent that IBM built with its corporate clientele. Conclusion Microsoft wins, Apple loses, Xerox gets pie in the face.

Now why start here on my CREE report? Because I am going to present CREE using the background of terms explained above. In short CREE's product is a discontinuous innovation (ie, its value proposition is to great not to transplant the former technology) which functions to the customer as a continuous innovation (its use is so continuous with current practice that its adoption by customers will not be difficult).

For those who need very detailed, scientific breakdown of CREE go to this site. I am going to gloss over things. To do some more nitty gritty DD use this site and I have about 10 more you can go to from there if you want. This site will explain in far more detail than I want to explain about what CREE does.


CREE currently has 2 products in mass production:

CREES largest product at this time is LED lighting. The lighting on the large screens in
Times Square are CREE SiCs, for Nokia owners the lighting in the new 7000 some series Nokia phone is a CREE SiC, and I have heard rumors that more Nokia phones will begin using SiC lighting as well. The lighting in the Volkswagen dashboard and in a new model of Audi is SiC (have also heard that SiC is now starting to be used in headlights as well). SiC is also being used in stop lights (such as a recent deal with China) in street lights and in billboard lighting. Some think that eventually SiC will replace the incandescent light bulb. It might, but it is sometime away.

CREE has only been producing SiC LEDS for about 18 months now. During that time it has become the dominant supplier in the world in blue and green LEDS.

Why? Because the competition in this area is sapphire. The economics of sapphire make it an inferior product both from the production stand point and the application stand point as sapphire requires more energy and more complicated connections. A larger sapphire is also needed to do the same job that a SiC LED can do. Conclusion: SiC is a discontinuous innovation in lighting. A SiC LED lasts much longer than current lighting technology, uses less energy, is smaller, and more cost effective than current lighting alternatives – in fact SiC not only performs better but it is also the lowest cost LEDs shipping in the world today. Just the $$$ saved from having to change the lighting in stop and street lights less often justify its purchase. SiC will become the predominate form of lighting in these areas. From street lights to stop lights to cell phones and in-between. Whether or not SiC actually makes it into office buildings and then houses time will tell (but I don't care cell phone lighting alone is large enough for me).

CREE's SiC are so good that Microvision (a company trying to develop a computer monitor that shines a laser straight into your eye to remove need for a physical monitor) has paid CREE to develop SiC for the purpose.

CREEs LED business alone is sufficient to justify the purchase of CREE. CREES LED business is either entering or just about to enter the Tornado as demand is booked 12 months in advance and CREE is having to triple or more its production capacity to book it.

CREE'S second product is Moissanite. I've seen it, it is gorgeous. Indistinquishable from diamonds without special electronic equipment. 1/10 the cost of diamonds, yet the jeweler gets a higher margin on the product. Mark it at ½ the price of an equivalent diamond product and buy it at wholesale for 1/10 to ¼ the cost of diamond. Big mark-up. Nice business, but really a sideline.


As expressed in a scholarly journal (of which a link is provided in the above link) “A change in technology from silicon to silicon carbide will revolutionize power electronics.”

All the remaining products involve CREE's superior ability to withstand heat and conduct electricity. I would have to write a book to go into detail but here are some examples, replacing current Silicon chips in heated environments like in cell phone base stations, automobile engines, power switching devices, satellites, aircraft, HDTV transmitters, and the list goes on and on will save literally billions of billions of dollars.

For example 30%-40% of the cost of cell phone base stations are in transistors. Replacing Si and GaAs chips reduces the number of transistors needed, increases the power per transistor and reduces the need for wiring and cooling systems. Same result in Microwave RF devices in which it has been found that SiC can tolerate 5x the power of existing transistors and operate at much higher temperatures.

Another example, using SiC on aircraft removes 90% of wires and cooling systems needed to support heat averse silicon sensors currently in existence. The weight savings and increased reliability (not to mention increased performance as SiC semiconductors can be optimally placed in heated environments) should save millions of dollars per commercial aircraft. SiC reduces the weight of an F-16 by several hundred pounds vs. having to rely on Silicon sensors. The weight of locomotives can be significantly reduced, engine efficiency increased through better SiC sensors on the engine, etc. The effect is even more dramatic on satellites as radiation shielding can also be lessened with use of SiC. This is just a sampling.

The largest pay off for SiC, if the above was insufficient, will be in power switches. I may do a write up separate on this – but power switches may be as big, if not a bigger market during the next 10 to 20 years as computer chips are today. These power chips will be made out of either SiC or out of Aluminum Nitride (which except for diamonds is the only material that outperforms SiC). Not coincidentally CREE purchased the patent on producing Aluminum Nitride.

Power switches will be the holy grail for CREE (if the aforementioned wasn't sufficient in and of itself). Currently Siemens and ABB use CREE products and CREE has been retained by the fourth largest power company in the world {hate it when I forget their name} but a Japanese firm to produce power switches on a demonstration basis. {NOTE: Everyone pays CREE to conduct research, R&D for CREE is a profit center in and of itself}.

Thus SiC is an enabling technology that is a discontinuous innovation and should replace Si and GaAs in nearly all high temperature, high power functions someday. That is beginning to happen, but won't happen in a large scale for a few years. Fortunately CREEs LED business is plenty large to propel growth today.

SiC is also a continuous technology for consumers. The use of SiC makes ubiquitous products better and cheaper without adding any burden or expense to the consumer.

VALUE CHAIN: CREEs proprietary technology is the production of SiC. Something no one has been able to do on a mass scale other than CREE. CREE's production techniques are covered by over 70 patents and according to CREE management is at least 6 years ahead of their closest competitors and growing. SiC production is therefore the proprietary and open architecture that CREE provides. Entire industries will be revolutionized by use of SiC and come to depend on SiC. CREE may also develop patents around products like power switches, blue laser, etc.

Second, CREE is the only producer in the world which can commercially mass product SiC to commercial quality. All future competitors in the field will either (1) have to license CREE production techniques or (2) produce at a much higher cost given CREEs tremendous head start.

In fact competitors will actually aid CREE because more SiC production is needed to meet demand and to increase demand going forward. I don't know if CREE will ever be able to catch up to the demand on its own.

ECONOMICS: CREES margins have been increasing strongly quarter after quarter. CREE's margins exceed such stalwarts as Cisco and NTAP.

The real kicker here is the work of geometric growth. Demand is growing at about 90% or more per year (and will probably accelerate – just for LEDs – not counting future products). Add to this that CREE has just expanded its production facilities which has 3 x the production equipment as the current production site. In addition CREE continues to increase the size of its wafers. Starting at 2” wafers, CREE is now beginning 3” wafers and plans for 4” wafers are 12 – 18 months away. Moving from 2” to 3” wafers doubles the amount of usable wafer due the surface area formula which involves r squared. Each double in size increased yield by 400%. Add to this that CREE's yields are in the mid 50% range – and CREE management has indicated that they expect to increase the yields by 300% and you have GEOMETRIC PRODUCTIVITY GROWTH x a tripling in production capacity producing for demand that CREE cannot hope to fully meet. Not a bad formula for explosive earnings growth!

Given time constraints I can only introduce you to the company in this post, but the above-provided link is an excellent source to get more details. The company is in the early stage of the first of what should be many tornadoes. The company has large barriers to entry and management that has reached every goal on time which it has set out.

In my opinion CREE is a company that I believe can be held for the next 10-20 years and just looked at during quarterly conference calls.

Not every technology spoken about above has been perfected commercially yet. Many of the projects are on-going R&D. The greatest risk is CREE's ability to follow through in its R&D goals. These risks are minimized, however, due to the current LED business which should grow enormously over the next few years to decade.

Their is a lot in CREE to be excited about. I am most excited about the Power Chip possibilities. But investing in CREE just for Power Chips is premature as they are still in the R&D stage – although I have heard that a commercially viable product is being introduced (or recently has been introduced). Haven't heard anything about CREE's ability to mass produce said product.

I believe CREE will be a mix of Intel and DELL in that it will be a Gorilla with some aspects of a King, but will have an efficiency advantage in the King aspects superior to its competitors than even DELL enjoys due to proprietary methodology, laser like focus on SiC, and many years head start on the competition.

Oh yes, the balance sheet ain't bad either. CREE is getting close to being cash flow positive. Once the new production facilities are up (which CREE is internally financing) I believe CREE, like NTAP, currently EMC, and DELL, will be a cash flow positive manufacturing enterprise.

But, since this forum is free – you can do the editing in the above (cost of free exchange of information) and maybe I'll clean up bits or answer specific questions later. Just wanted to get the bulk of my excitement about CREE out.
Print the post  


What was Your Dumbest Investment?
Share it with us -- and learn from others' stories of flubs.
When Life Gives You Lemons
We all have had hardships and made poor decisions. The important thing is how we respond and grow. Read the story of a Fool who started from nothing, and looks to gain everything.
Contact Us
Contact Customer Service and other Fool departments here.
Work for Fools?
Winner of the Washingtonian great places to work, and Glassdoor #1 Company to Work For 2015! Have access to all of TMF's online and email products for FREE, and be paid for your contributions to TMF! Click the link and start your Fool career.