[ NNSquad ] CLWR: Benefiting From Android Open Development

CLWR: Benefiting >From Android Open Development

Yesterday's introduction of the first Android-powered device further advances the cause of open-based innovation. We believe WiMax will be disproportional beneficiaries, given its inherent openness. The launch, on October 22, adds to list of catalysts lining up for CLWR during 2H08: WiMax market launches, 3Q results, closing the Sprint plus investment transaction, and now Android launches. Our 2008 and 2009 revenue and EPS estimates remain $215M/($4.30) and $287M/($3.75). We reiterate our Buy rating and $17 price target.

Yesterday, T-Mobile introduced the first Google phone; the G1 from HTC, with the launch is slated for October 22nd. Google has been the force behind the Open Handset Alliance, the developers of Android, the open operating system powering the G1.

We believe the G1 launch heralds in an era of more robust innovation in the mobile marketplace. We have been long persuaded that open devices and open networks attract the most innovative developers for both new network applications as well as new devices. We see the burgeoning list of members of the WiMax Forum and the Open Handset Alliance as an indicator of the accelerating pace of development, long stymied by the carrier community. (Ironically, the G1 is initially an exclusive device for T-Mobile.)

We believe Clearwire will benefit greatly from strong acceptance of Android platforms. Directly, we understand that Clearwire is working with several manufacturers and application developers on Android-based products for its network, likely for 2009 launch. We believe that there are likely several efforts under way that are beyond Clearwire's purview that will be launched on their network, given its open nature. Indirectly, we expect all open networks to benefit from the coming wave of Android-based development, noting the relative ease of moving any 3G-oriented device or application to the open WiMax environment in short order.

The biggest concern from the developer perspective is the size of the near-term WiMax network, which will be steadily addressed by the Sprint Xohm / Clearwire merger, and the rapid fire launches of Baltimore (October 8^th), Chicago, Washington, Portland, Atlanta, Grand Rapids, Las Vegas, Boston, Dallas, and Philadelphia, plus the 2009 conversion of Clearwire's current Expedience (pre-WiMax) markets covering 16.6M US PoPs.

As we look through year-end, we add October 22^nd to the list of CLWR catalysts: WiMax market launches, the posting of 3Q results, and the closing of the Sprint plus investment transaction, in our view. With each successful launch, skepticism of Clearwire and WiMax should fade.

Clearwire's improvement in quarterly results should continue, with initial market EBITDA margin up from 2Q's impressive 34%, perhaps nearing 40%. We view these initial market results as a model for the national New Clearwire network.

WiMax vs. LTE, continued

In assessing the LTE/WiMax "competition," we reinforce a point we have previously discussed. The primary reason that incumbent operators, ex-Sprint, have chosen LTE as their 4G platform is lack of near-term availability. While incumbents tend to fog up the issue as to how they chose LTE over WiMax, we believe the decision was made on a single compelling fact: LTE will be available for broad deployment, in the best case, two to three years later than WiMax. We believe that the incumbents' investors have no appetite for another major capital program just as the 3G build is wrapping up.

Further, the incumbents will likely not have sufficient spectrum to deploy a competitive data product until at least the next major spectrum auction is completed, likely several years in the future. Viewed in this context, not only is LTE's tardiness of benefit to the incumbents' business plans and investor story, but any of the almost guaranteed slips in LTE availability will likely be welcomed by the incumbents' leaders—while blaming the vendors' slips. Eventually, of course, we believe that when it is obvious that Clearwire is taking away the most desirable subscribers with a network and business model the incumbents cannot attack, we will learn about the incumbents' "Plan Bs."

Finally, LTE has a problem with TDD vs. FDD. Time Division Duplex (TDD) and Frequency Division Duplex (FDD) define how network bandwidth is managed, notably between uplink and downlink. FDD splits uplink and downlink into fixed channels, which works well with the roughly symmetric traffic flows common in voice networks. In data networking, traffic flows are vastly more imbalanced and unpredictable, with more data traveling down than up. For example, a few bytes might be sent up requesting a web page, with a multi-kilobyte or even multi-megabyte response sent down.

In TDD systems, a single channel is used for bi-directional communications. Therefore, TDD supports inherently imbalanced, and bursty, data traffic much more gracefully – dynamically in real time. This agility is missing from FDD systems. While FDD adherents may suggest setting its channel sizes unequally to accommodate the uplink/downlink traffic flows, this makes equipment design and system management considerably more complex, and still creates a static allocation unable to adapt to changes in traffic dynamics. The obvious conclusion is that TDD is far superior for data traffic or voice + data traffic, while FDD is adequate for voice traffic only.

All 4G systems are designed first as mobile data networks. WiMax is a TDD system; most LTE designs are FDD designs, except for the China Mobile (and partners) trial. Again, LTE is significantly disadvantaged relative to WiMax on spectral efficiency – and this doesn't account for the wasted spectrum consumed in the guard bands separating FDD's uplink and downlink. For these reasons, we believe LTE will ultimately settle on a TDD implementation, at the cost of still further delays in system availability and interoperability.

How to model innovation?

We note that the full impact of the WiMax ecosystem will be far more dynamic than any investor is currently modeling, in our view. The fruit of the ecosystem should be a cornucopia of products and services offered on an open basis across the Clearwire network. Indeed, we believe the company is working with several thought leaders in new service development, while many more highly innovative firms are working independently. We are especially intrigued to see what emerges 12 months after Clearwire launches service in Silicon Valley.

Our model does not yet account for the merged Clearwire/Xohm, expected in December, instead accounting only for the independent Clearwire. Further, we fail to fully account for the prospect of this rich ecosystem – notably including Android-fired innovation – creating compelling products and services to leverage this unique, open network. How should we model such unbounded innovation?

**************************************************************

If you would like to be removed from this email distribution, please contact nsinghania@thinkpanmure.com.

Please see the attached note for disclaimers.