高通 (QCOM) 2004 Q4 法說會逐字稿

Ladies and gentlemen, thank you for standing by and welcome to the QUALCOMM fourth quarter conference call.

At this time all participants are in a listen-only mode.

Later we will conduct a question-and-answer session.

If you have a question, you will need to press the one, followed by the four on your push-button phone.

As a reminder, this conference is being recorded November 3, 2004.

The playback number for today's call is 1-800-633-8284.

International callers, please dial 402-977-9140.

The playback reservation number is 21209991.

I would now like to turn the call over to Bill Davidson, Vice President of Investor Relations.

Bill, please go ahead.

Thank you and good afternoon.

I'm joined by Dr. Irwin Jacobs, Tony Thornley, Bill Keitel, Dr. Paul Jacobs, and Dr. Sanjay Jha.

An internet presentation and audio broadcast accompanies this call and you can access it by visiting www.qualcomm.com.

During this conference call if we use any non-GAAP financial measures as defined by the SEC in Regulation G, you will find the required reconciliations to GAAP on our Web site.

I'd also direct you to our 10-K and our earnings release, which were filed and furnished respectively with the SEC today and are available on our Web site.

We may make forward-looking statements that may differ materially from QUALCOMM's actual results.

Please review our SEC filings for a detailed presentation of each of our businesses and associated risks.

Results presented in accordance with GAAP reflect the Company's decision in the fourth fiscal quarter to cease estimating royalties before they are reported by our licensees as announced on October 28, 2004.

Pro forma results that assume the continuation of estimating royalties, before they are reported by our licensees, are presented to assist investors with evaluating financial performance during this transition period.

Please visit our Web site as we've provided extensive information to assist investors with this change in our method of accounting for royalties.

Following opening remarks from Dr. Irwin Jacobs, Dr. Paul Jacobs will discuss our plans for MediaFLO USA, and Dr. Sanjay Jha will provide an update of our QCT business.

Bill Keitel will conclude with a summary of financial highlights for the fourth quarter and fiscal year 2004 as well as provide guidance metrics.

We will then open the discussion to your questions.

Before I turn the call over to Dr. Jacobs, I'd like to mention that we are hosting an analyst meeting on November 17th in London with detailed presentations from our executive team.

The meeting will be simulcast on our Web site with audio and slide presentations.

And now it's my pleasure to introduce Dr. Irwin Jacobs, Chairman and CEO.

Thank you, Bill and good afternoon, everyone.

I'm pleased to report that we posted strong results for the fourth quarter and for the full fiscal year with solid growth in our Chip and Technology Licensing businesses.

This has obviously been an outstanding year for QUALCOMM with global demand for CDMA products accelerating very strongly.

I'd like to thank QUALCOMM employees around the world for making this possible with their continued dedication and commitment to our customers and focused execution on our business goals.

According to In-Stat, the CDMA handset market has now grown to approximately 24% of the global wireless market, up from approximately 20% in the June quarter last year.

We expect to complete the calendar year with CDMA handset sales of 168 to 172 million units, substantially higher than the 131 to 136 million units we expected when giving our guidance last November for calendar year 2004.

This exciting growth benefits everyone in the wireless value chain from equipment manufacturers, network operators, and application developers, to the consumers and business users who rely on today's increasingly sophisticated CDMA wireless devices.

The CDMA development group reported that CDMA2000 IX-EV-DO is leading in the deployment of 3-G broadband networks worldwide with an established subscriber base exceeding 10 million users globally.

To date, 13 operators have commercially launched 1xEV-DO in nine countries, including recent deployments, to name just a few, by Vivo in Brazil, Alaska Communications Systems in the U.S., BellSouth Guatemala, Eurotel Praha in the Czech Republic, and Smartcom in Chile as well as Pelephone in Israel, and soon Telecsa in Ecuador.

Our experience in the development of DO positions us well to take advantage of the opportunities for High Speed Downlink Packet Access, HSDPA, the broadband wireless technology for the WCDMA market.

We'll sample our first HSDPA chip, the MSM6275 in the fourth quarter of this calendar year.

To support the growing worldwide demand for multimedia services, we announced two evolutions of 1xEV-DO technology, called Gold Multicast and Platinum Multicast.

We also announced a new air interface innovation called FLO, Forward Link Only, for use in conjunction with CDMA2000 and WCDMA.

These innovations are designed to increase network capacity and reduce the costs associated with delivering high-quality video and audio to mobile handsets.

Additional efficiencies are provided through the MediaFLO media distribution system, which provides intelligent scheduling of content delivery during off-peak hours.

Together, these innovations form an end-to-end system that enable the continuing growth of wireless multimedia services worldwide in a user-friendly, cost-efficient fashion.

QUALCOMM will support Platinum Multicast with a CSM6800 Southside Modem and a MSM6800 chipset and system software solutions.

These chips also support the many improvements included in 1xEV-DO Rev A, Revision A, while supporting full backwards compatibility to all earlier CDMA-based devices.

QUALCOMM continues to make strategic investments in companies and technologies that advance the evolution and adoption of 3G wireless data services around the world.

By evaluating companies to determine if an investment is appropriate, QUALCOMM gains great visibility into emergent technologies.

A notable example was the investment we made in Iridigm Display Corporation nearly two years ago.

We believed Iridigm was a company that had technology that would deliver significant improvements in the ability to view displays in direct sunlight, enable lower power consumption, and increase performance relative to other display technologies.

These features fit well with our overall strategy of rapidly increasing the capabilities of wireless devices while driving down cost, size, and power consumption, particularly for video applications.

As such, we made the decision to acquire Iridigm and operate it as a wholly-owned subsidiary of QUALCOMM.

On October 19th, 2004, we announced that Siemens will build a range of wireless multimedia handsets based on chipsets from QUALCOMM's WCDMA or UMPS portfolio.

Siemens is the first major European device manufacturer to become a QUALCOMM customer for WCDMA chipsets.

Initial handsets will use the MSM6250 solution with future products to be based on QUALCOMM's expansive line of WCDMA chipsets.

This announcement underscores the increased momentum in Europe and worldwide transition to 3G mobile services.

WCDMA growth was particularly strong in Japan and Europe, and we are seeing significant progress in Australia, Hong Kong, and the United States.

We're now working with 25 manufacturers worldwide to enable faster time to market with advance and competitively-priced WCDMA equipment.

Turning to our BREW business, the entire wireless value chain continues to benefit from the ecosystem that BREW has helped generate.

Third-party BREW publishers and developers to date have earned more than $200 million from the sales of BREW-based applications.

Jamdat and Faith West have both publicly stated revenue in excess of $1 million a month from their BREW-based applications.

This achievement marks the on-going momentum for the BREW system within the rapidly expanding global wireless industry.

To date, there are 37 commercial operators in 24 countries offering wireless services based on the BREW solution, and 27 manufacturers have offered more than 150 BREW-enabled device models to consumers.

We're focused on future opportunities in the WCDMA market.

To facilitate the demand for differentiation of services and faster time to market for hand set manufacturers in the global market, QUALCOMM acquired Trigenics, a mobile user interface company, based in the U.K.

This acquisition marks a significant next step in the Company's efforts to bring enhanced 3G wireless data services, and significant additional support to operators, device manufacturers, and developers in Europe and around the world, and illustrates our resolve to support wireless operators.

This acquisition greatly improves the ability of our customers already offering BREW-based services as well as other operators and handset manufacturers to reduce time to market, and to simplify and standardize the user interface on wireless devices.

In addition to targeting the consumer segment, we're working with operators to drive data adoption for enterprises.

Our own QWBS division is using BREW for their transportation business, but we also see an opportunity for wireless enterprise applications in other vertical markets including healthcare, financial field service, and the government.

A developer community is creating a variety of useful applications for these markets, and we're working with operators to evolve wireless data offerings for the business user.

Due to the continued confidence in our business, the QUALCOMM Board of Directors voted to increase our dividend twice in fiscal 2004, bringing our annual dividend to 28 cents per share of common stock, post our recent 2-for-1 stock split.

We remain well-positioned for substantial increases in our dividend program and look forward to another strong year for CDMA in fiscal 2005.

I am confident that the employees of QUALCOMM understand and share in the focus of delivering on the goals we have set for our business.

Our challenges for 2005 are many, but they are not unfamiliar to us.

We remain focused on growing the entire CDMA market, delivering highly integrated chipset solutions that enable our customers to have the ability to offer the most advanced feature sets in their devices, lower cost, and faster time to market.

I'd now like to turn the call over to Dr. Paul Jacobs to discuss our plans for MediaFLO USA.

Thank you, Irwin.

This week we announced that we have formed a network operator subsidiary, MediaFlo USA Inc.

It will deploy and operate our wireless multimedia multicasting or mediacasting network in the United States.

MediaFLO USA's network will use QUALCOMM's FLO or Forward Link Only technology, announced in October, and the MediaFLO media distribution system announced in March.

This nationwide network will deliver many channels of high-quality video and audio programming from leading content providers to third-generation wireless devices.

This means both CDMA2000 and WCDMA operators in the U.S. will be able to offer very compelling wireless multimedia services to their subscribers without bearing the cost of deploying and operating a media casting network.

This structure will enable multimedia delivery at the lowest possible cost per bid and translate to market pricing that would enable mass adoption.

The business model of MediaFLO USA is to act as a wholesaler to the wireless operators.

The wireless operators will do the retail marketing to their customers of service packages they create using content delivered over the FLO network and content they procure directly.

MediaFLO USA will deploy its network in 700 megahertz spectrum for which QUALCOMM holds licenses, which is UHF channel 55.

We currently own sufficient spectrum to provide a seamless, nationwide footprint for the network.

Some of the Spectrum is currently encumbered by TV stations.

We are working in those markets to clear the stations, either by having them turn off analog transmissions and transition to fully digital operation, or possibly by moving to a different channel.

Let me give you a little bit more detail on the capabilities of the FLO technology deployed in our 700 megahertz spectrum.

The network will support 50 to 100 national and local content channels, including up to 15 live streaming channels and numerous clipcast and audio channels.

By clipcasting, we mean sending numerous short format clips across the network that will be stored on subscribers' phones for later viewing, according to a programming schedule.

Users access channels of content, both streaming and clipcast through the MediaFLO program guide, which looks similar to an electronic program guide on TV.

This content will be delivered in an easy-to-use and familiar format at quality levels that dramatically surpass current mobile multimedia offerings through the use of QVGA video at up to 30 frames per second and high-quality stereo/audio.

Very importantly, we have optimized the FLO technology for mobile devices so that battery life is preserved, and to ensure that channel switching times are in the range of one second so that viewers can easily channel surf on their handsets.

This performance is significantly better than competing technologies, such as DVBH.

In addition, the MediaFLO system supports alerts, so subscribers know when to watch short format content that is interesting to them, whether it's a report on a stock in their portfolio, a score by their favorite sports team, or a traffic jam on the way home.

The additional information that describes the multimedia content can be customized by each operator for differentiation and can include interactive links to BREW applications or Web pages for one-click access by the viewer, thus driving additional revenue opportunities.

The planned mediacasting network and 700 megahertz spectrum is in no way a substitute for cellular networks.

Actually, the two compliment each other, enabling U.S. wireless operators to offer interactive multimedia services to their subscribers with a broad range of high-quality content.

Our operating this purpose-built mediacasting network in cooperation with the cellular operators networks point to what we believe will be increasingly common place, heterogenous networks operating in conjunction with each other, supporting a wide variety of services, all accessed by subscribers on their wireless devices.

In this case, where the same multimedia content is being sent to many subscribers, there's no additional network cost for additional subscribers, it makes sense to share the network between multiple wireless operators.

However, the MediaFLO system works in conjunction with multimedia transmitted over an operator's 3G network.

It allows them to differentiate their multimedia services by seamlessly mixing content delivered through the FLO network with unique content they have obtained and transmitted to their subscribers over their 3G network.

We have been in discussion with content providers and broadcasters about this major new distribution channel that compliments their current offerings, enabling them to reach their audiences when they're away from the home and on the go.

Local broadcasters are attracted by the idea of sending clips featuring their newscasters via the FLO network, thus creating pull for the viewer to watch their full news show when they go home.

And while we do not have specific announcements to make today, our discussions with content companies are going so well that we have had inquiries that some wish to invest directly in this business.

We anticipate that the investment in MediaFLO USA Inc. over the next four to five years will be approximately $800 million, some of which may be funded by third parties.

We plan to trial the new network next year and begin commercial operations in 2006.

This initiative extends our core strength.

We strongly believe that the broad delivery of wireless multimedia is the next logical step in the evolution of the wireless industry.

We made the decision to obtain spectrum and form operator subsidiary to bring the FLO technology to market as rapidly as possible.

In addition, this allows us to ultimately spin the subsidiary off, creating incremental shareholder value, and enabling QUALCOMM to continue to concentrate on our central mission of innovating new technology and services for our operator partners and integrating it into our chipsets and software for device manufacturers.

I'd now like to turn the call over to Dr. Sanjay Jha for updates to our QCT business.

Thank you, Paul and good afternoon, everyone.

QCT achieved a record 845 million in revenue for the September quarter, and approximately 3.1 billion for the fiscal year 2004.

This represents an increase of approximately 29% compared to the total QCT revenue for fiscal 2003, and a sequential increase of 7% over the June quarter.

Earnings before taxes for the September quarter were 271 million, up 7% sequentially, and over 118% versus the same quarter last year.

Earnings for fiscal 2004 were $1.43 billion, up 31% from fiscal 2003.

In 2004 we saw a significant increase in demand for our chips, up 38% to approximately 137 million chips, versus approximately 99 million chips in fiscal 2003.

During the September quarter, QCT shipped approximately 39 million MSMs, a new record for us, compared to approximately 35 million in the June quarter, and approximately 20 million in the year-ago quarter.

Year-over-year, consistent with our prior guidance, we have only seen a moderate decrease in our chipset ASPs, which is partially offset by favorable changes in our product mix.

Going forward year-over-year, we expect ASPs to either remain stable or increase marginally.

We expect the effects on ASPs of continued expansion in India, Latin America, Southeast Asia, and China will be offset by higher ASP products supporting EV-DO wideband CDMA and integrated multimedia in products such as 6100, 6500, and 6250.

2004 has been an investment year.

We have invested in delivering EV-DO, DO Revision A, wideband CDMA, HSDPA, enhanced uplink extension of HSDPA, RF CMOS, integrated multimedia capability and par efficient microprocess accords in our products.

We have invested in helping our wideband CDMA OEM partners as well as our wideband CDMA carrier partners to launch services around the world based on our chipset technology.

Despite these investments, our operating margin for September quarter was approximately 32% of revenue, and our fiscal 2004 operating margin of 34% was higher than fiscal 2003 by 1%.

We continue to see strong demand for our solutions going forward.

For the current December quarter we expect to ship between 38 and 39 million MSM chips.

Demand for our infrastructure product has also continued to be strong.

During the September quarter, we shipped CSM infrastructure chips supporting approximately 5.4 million equivalent voice channels, our second highest quarter ever.

This is due primarily to transition from 2G to 3G in North America, China, India, and also because of continued demand for EV-DO in North America.

We're very bullish on EV-DO as we look to future rollouts worldwide, including Sprint and the further extension of Verizon in the U.S., plus various carrier launches in India, Brazil and China.

During the September quarter, we shipped approximately 4 million DO chipsets, another record for the division, and we believe a positive indicator for the future.

We expect to double our DO shipment next year from approximately 12 million units this year.

Last quarter I indicated that we were working hard to add capacity at existing suppliers.

QCT has made significant progress in working with its partners to insure future product supply.

These improvements include duel sourcing strategies and capacity reservation agreements with key strategic supply partners.

As part of this initiative, QCT has signed a comprehensive long-term capacity reservation agreement with IBM for guaranteed wafer starts to begin in the December quarter.

Our acquisition of Spike Technology has also provided with us engineering resources which we are now utilizing to support our duel sourcing strategy as well as to provide design sales and technical support to our customers around the world.

Finally, let me emphasize that this past year has seen QCT invest strongly in its future in order to extend its leadership position.

In addition to our core software and hardware engineering base in San Diego, we also now have engineering teams in North Carolina, Texas, Germany, U.K., and India.

These new groups will help us to continue to provide complete system solutions to an expanding customer base and worldwide marketplace.

And now I turn the call over to Bill Keitel for a financial overview.

Thank you, Sanjay and good afternoon, everyone.

As Bill Davidson mentioned, we filed our Form 10-K report today, continuing our practice of the last couple years of filing our SEC report on the same day we issue our earnings release.

As well, our Form 10-K report filed today includes a certification from our outside auditors in accordance with Sarbanes-Oxley Act Section 404.

Although we were not required to comply until one year from now, I'm very pleased that QUALCOMM's financial reporting controls have already met or exceeded the stringent requirements of Sarbanes-Oxley 404.

I'll begin with a brief recount of the accounting decision we announced on October 28th.

Since 1998, royalties from licensees, for which estimates could be reasonably made, have been accrued in the quarter when earned and then adjusted in the subsequent quarter to record the actual royalties reported.

Effective with the quarter ended September, 2004, we will report all royalties upon receipt of reports from our licensees.

In other words, we will now record all royalties one quarter after the royalties are earned.

To the extent we've seen quarterly seasonality in our royalty revenues, which has been the case from time to time, our new method of recognizing royalty revenue will reflect any such seasonality with a one quarter lag.

As a result of this change and consistent with Generally Accepted Accounting Principles, the initial three quarters of fiscal 2004 include the formal accrual estimate of accrued royalties, but the fourth quarter does not.

Therefore, our GAAP results, specifically the QTL segment, do not reflect a full year of the economic performance of the Company's licensing business.

To assist investors, we provided pro forma results that assume we continued our previous method of estimating royalties in the fourth fiscal quarter.

We also showed our new method of recognizing royalty revenues as though it had been in effect for fiscal years 2001 through 2004.

For fiscal 2004 we achieved record revenues, record net income, earnings per share, and operating cash flow.

Based on the new method of recognizing royalty revenues, fiscal 2004 revenues excluding QSI, were $5 billion, up 31% on a comparable basis to fiscal 2003.

Earnings after-tax were $1.8 billion, and diluted earnings per share were $1.06, up 53% and 49% respectively on a comparable basis year-over-year.

By comparison, the pro forma results, using our previous method of estimating royalties, showed fiscal 2000 revenues were 5.1 billion and diluted earnings per share were $1.09.

The differences are slight and due primarily to the new method of reporting royalties with a one quarter lag.

In the September quarter, revenues using the new method excluding QSI, were 1.4 billion, up 57% compared to the prior year, and earnings per share were 29 cents, up 107% year-over-year.

Using the pro forma, or previous method of estimating royalties, September revenues and earnings per share were the same amount as they were for using the new method.

Turning to our segment results, it is important to know that the change in estimating royalties in the September quarter only impacted QTL, our licensing business.

QCT, QWI, and QSI 2004 results did not change based on the new methodology.

The effect of the change on QTL was that $251 million of royalties that would have been estimated and accrued in the September quarter were not included in QTL revenues.

QTL's September pro forma revenues, which include the 251 million estimate were $402 million.

For the fiscal year, QTL pro forma revenues were 1.6 billion, including the 251 million estimate, up 58% year-over-year, and earnings before tax were 1.4 billion, up 65% year-over-year.

In our chip business for the fiscal year 2004, QCT revenues were 3.1 billion, up 29% year-over-year with an operating margin of 34%.

During the September quarter QCT shipped a record 39 million MSM phone chips, up 19 million year-over-year, and up 4 million sequentially.

During fiscal 2004, QUALCOMM Wireless & Internet Group grew revenues by 17% year-over-year to 596 million, earnings before tax grew 11%, and the operating margin was 5% of revenues.

The QUALCOMM Strategic Initiatives, or QSI segment, generated net cash of $169 million in fiscal 2004 as we continued to decrease our rate of new investment and as we monetize a portion of prior investments.

QSI contributed 2 cents per share of earnings for fiscal 2004.

Based on royalty reports from our licensees, worldwide CDMA handsets shipped in the June quarter were approximately 41 million units, up from 38 million units shipped in the March quarter, for a total of 79 million handsets shipped during the first half of the calendar year 2004.

Our most recent data suggests that approximately 41 million handsets have shipped in the September quarter, lower than our prior estimate of 46 million phones, due in part to the recent slowdown in South Korea, and due in part to CDMA handset manufacturers as a whole, holding slightly more inventory.

As well, we now expect slightly greater CDMA phone shipments in the December quarter.

Our estimate for CDMA handset shipments for the December quarter is 48 to 52 million units, up from our prior estimates for the December quarter of 41 million and 48 million units made in July and September respectively.

The average selling price of CDMA handsets was approximately $211 for the June quarter.

For handsets shipped in the 12 months ending June, 2004, which correlates to revenue recognition for our fiscal year 2004 based on the new royalty method, the average selling price was approximately $205.

The difference between this $205 estimate and the prior $209 estimate is the result of the one quarter lag in reporting royalty revenues.

It is not a result of a change in estimated average selling prices.

Turning to our forward guidance, I'll compare fiscal 2005 revenue and earnings per share to fiscal 2004 with royalties reported on the same basis as 2005.

This adjusted, historical data is available on our Web site.

For the calendar 2004 CDMA market, we now expect approximately 168 to 172 million CDMA handsets to be shipped, including approximately 23 million WCDMA handsets.

Based on the 170 million mid-point of our estimate, calendar 2004 CDMA worldwide handset shipments are anticipated to grow approximately 45% year-over-year.

Based on the current business outlook, we estimate the CDMA phone market will increase to approximately 218 to 228 million unit shipments in calendar 2005, an increase of 29 to 35% over our mid-point estimate for calendar 2004.

Based on the 223 million mid-point of our 2005 estimate, we anticipate shipments of approximately 168 million CDMA2000 units and approximately 55 million WCDMA units.

Based on the current business outlook and using the new method of recognizing royalty revenues, we anticipate fiscal 2005 revenues, excluding QSI, to be in the range of 5.8 to 6.3 billion, an increase of 16 to 26% to the comparable basis for fiscal 2004.

We anticipate a diluted earnings per share, excluding QSI, to be in the range of $1.15 to $1.19, an increase of 8 to 12% to a comparable basis year-over-year.

We estimate average selling prices for CDMA2000 and WCDMA phones combined to increase 5% in fiscal 2005 to approximately $215.

This compares to our estimate of $205 in fiscal 2004.

We anticipate the combination of R&D and SG&A expenses to increase approximately 28 to 30% year-over-year, with the majority of the growth occurring in R&D.

We anticipate an increase in our tax rate in 2005 to approximately 30%, compared with approximately 29% on the pro forma basis in fiscal 2004.

Beyond 2005, we see the opportunity for our effective tax rate to decrease below 30%.

Included in our 2005 guidance is a 2 cents per share decrease in earnings due to our acquisitions of Iridigm, Trigenics, and Spike Technologies.

Turning to the first quarter of fiscal 2005, we estimate revenues excluding QSI to be in the range of approximately 1.3 to $1.4 billion, an increase of 12 to 23% on the comparable basis year-over-year, and relatively flat on a comparable basis sequentially.

We estimate diluted earnings per share excluding QSI, to be in the range of 24 to 26 cents, a decrease of 10 to 17% to the comparable basis sequentially, and an increase 4 to 13% to the comparable basis year-over-year.

This estimate assumes shipments of approximately 38 to 39 million MSM phone chips during the quarter.

We are supply constrained on December quarter shipments of 6000 Series MSM phone chips.

As Sanjay mentioned, we expect to resolve the supply constraints in the first quarter of calendar 2005.

Given the new method of recognizing royalties, royalty revenues in the first quarter of fiscal 2005 will reflect CDMA phone shipments that occurred in the quarter ended September 26, 2004.

We are looking forward to the end of contractual arrangements in which we share royalties with two entities.

Because of the impact in fiscal 2006 will be material, and because it is a contractual arrangement, we are sharing this information with you today.

Under these agreements, the two entities are entitled to share in a percentage of the royalty revenues for certain CDMA products received by QUALCOMM from third parties.

One of these sharing arrangements expires in latter fiscal 2005, the other expires in fiscal 2006.

In fiscal 2006, we expect the termination of royalty sharing arrangements to contribute approximately 330 to $360 million to both our fiscal 2006 revenue and our fiscal 2006 pretax earnings.

We have included a modest improvement to fiscal 2005 revenue and earnings related to these, ending of these arrangements.

QUALCOMM Investor Relations Web site includes an extensive slide presentation on the many data points included in this conference call.

I look forward to sharing with you additional data points regarding our fiscal 2005 guidance, including regional handset shipment estimates at our London analyst meeting in two weeks.

The analyst meeting will be Web cast for those of you not able to attend.

That concludes our remarks.

Operator, we are ready for questions.

Thank you.

Ladies and gentlemen, we will now begin the question-and-answer session.

To queue a question, press one four.

To retract a question, press one three.

If you're using a speaker phone, please pick up your handset before pressing the numbers.

One moment for the first question.

Our first question comes from the line of John Bucher at Harris, Nesbitt & Gerard.

Please proceed with your question.

Thank you.

Question on MediaFlo USA subsidiary.

Is there any concern about attracting carriers with what appears to be proprietary forward-link?

In other words have the carriers asked for this to be compatible with DVBH or DMB or any of the other digital standards?

And second, it appears that MediaFlo USA will actually do the content aggregation.

Have any of the carriers expressed a concern that they would like to be able to do that content aggregation and rather use FLO as a pipe that they would fill rather than your parsing and aggregating the content?

Thank you.

Okay, so in terms of compatibility, you know, if there are other broadcast technologies as we do with many other technologies, we can integrate those into the chipsets.

In terms of being a proprietary technology, we're in the process of setting up a consortium to work on the standardization of the technologies.

So that shouldn't be a concern going forward.

In terms of content aggregation, the way that we look at the FLO technology, that the FLO will provide common content that could be shared by multiple operators, but they also have access to the DO Gold and Platinum Multicast as well as the standard DO for delivering unique content that they might get access to.

For example, some of the operators have relationships with some of the content providers because they also do advertising on their networks, and therefore, they might be able to get differentiated content.

The MediaFlo system allows seamless intermixing between the content that comes over the common FLO network and content that's delivered over the differentiated and unique carriers network.

And so in that way, they are able to do their own deals on content as well as to use the FLO network

In terms of them being able to use it for their own content, we have considered providing some amount of capacity over the network to operators for their own use, but in general, the bandwidth isn't really sufficient to allow multiple content of the same types, and therefore, we thought that it was more efficient to provide the common content over the FLO network and allow them to differentiate primarily on their DO networks or WCDMA networks.

Paul, the $800 million estimate for network build-up, does that also include all the backend server and production and digital handling that would be required for content?

Exactly.

So it's all the operating losses, includes capital needs, working capital and all those expenses up to EBITDA positive.

Thank you.

Our next question comes from the line of James Faucette at Pacific Crest.

Please proceed with your question.

Thank you.

Bill, I'm wondering if you can explain or just give us a little bit more insight into the ending of these licensing agreements, and how we should think about how the amounts can vary over the next couple of years just where there may be some sensitivities there?

Sure, James.

I can't give a lot of specifics because both the agreements have confidentiality clauses, but in effect, what we've been doing is sharing a substantial amount of royalties here for many years, and those arrangements, the first one begins to end at the end the latter part of fiscal '05, the second one ends in fiscal '06, so there's going to be a one-time step-up in fiscal 2006 to the tune of about 330 to 360 million.

You know, from there, our royalties in total should continue to correlate closely with the CDMA market, as they have been, you know, given the unit volumes and the ASPs.

Now, I guess just for a little more clarity, should we think about this then as an effective, I guess, for modeling purposes, an effective increase in kind of your percentage royalties that you're generating on handsets and the like?

Or is this more like just a dollar amount that's stepping up and then you kind of grow from there?

It would be in effect, it would be an increase in the net royalty rate that QUALCOMM keeps for itself.

Okay, great.

And then just housekeeping question.

With the new acquisitions being incorporated, can you give us an idea on what the, on total employee base is like, and how that breaks down in R&D versus SG&A, and how we should expect those operating expenses to grow at least over the next fiscal year?

Sure.

In terms of the three acquisitions, the acquisition with the substantial number of employees is Spike Technologies, and that's very heavily R&D oriented.

The acquisition that's driving the majority of the 2 cent decrease in earnings year-over-year is the Iridigm acquisition, due in part to the cash, the greater price for that acquisition, but then also that will be growing the resources steadily through the year.

So I think in total, they would tend to be heavily, the three acquisitions would tend to be heavily weighted towards R&D.

Okay.

And can you give us an employee count of people involved in research and development versus the SG&A line?

For total QUALCOMM?

Yeah, right.

Yeah, James, I haven't updated that number for about a year, so why don't I take that away and consider that for London so that I'd give you something I'm confident of?

Okay, great.

Thank you very much.

Our next question comes from the line of Brian Modoff at Deutsche Bank.

Please proceed with your question.

Hi, guys.

A couple of questions for you.

First, Sanjay, what's the number of WCDMA ASIC handset customers you currently have?

I believe the number is now up to 26.

26?

And how many of the top 5?

Certainly I think we've announced now Samsung, LG, and Siemens.

Any others?

None that we've announced.

Okay.

Looking at your unit numbers for next year, the 168 million in CDMA, where's that increase coming from?

And then on top of that, looking at the number you're guiding to for Q4, the 48 to 52 million, given kind of the weak numbers we're seeing out of China and Korea, what comfort level do you have with that forecast?

Yeah, Brian, let me speak to that.

First on the MSM shipment number, I think our visibility's extremely good.

I think we'd probably be guiding a bit higher if we had a little bit more supply.

On the phone numbers, the 41 million estimate, the great majority of our licensees share with us now a forecast, so that supplements our own processes.

And the 48 to 52 million estimate we've seen inventory at the full manufacturers bump up a bit, which we think could decrease into the December quarter.

And as well, we have a pretty good idea what that demand is for MSM so it's an extension out from those two.

I would add that you know, some of our competitors in the chip business tend to see you know, greater number of phones shipped in the December quarter in preparation for the Christmas season.

Brian, you were asking also about the year for the phones.

We're seeing, we expect to see growth in most of the markets around the world for CDMA2000, particularly in China, India, North America, you know, but it's pretty much every market we expect to see reasonable growth.

Could you give us an idea of the, you know, the numbers you're looking at increase-wise for China and India?

And then lastly, Sanjay, I think you qualified Atmel as the second source for your transceivers on 6000.

Is that, are they going to be up and running in Q1 to alleviate that shortage issue on the 6000?

Let me take the second one first, Brian.

Yes, that's indeed right in fact, they're up and running already.

The question is how quickly can we ramp volume?

And I think we're already ramping volume with them, but because of the lead time, most of the impact of that volume ramp will come in first quarter and that will be a significant factor.

Although I should also say that our first source there is Free Scale and they have also stepped up and supported us more heavily.

Brian, just, we are not at this point splitting out the '05 market forecast.

We'll be giving some more updates of that again in London.

I should also mention Southeast Asia as a growth market, we're certainly seeing growth there in a lot of countries, so I should add that to that list.

Okay, thank you guys.

Our next question comes from the line of Louis Gerhardy at Morgan Stanley.

Please proceed with your question.

Good afternoon.

Just wanted to ask about the 48 to 52 again.

Just to be clear on what you're saying, are you suggesting then you expect to snap back from an inventory burn and also more favorable seasonality to help you to go from 41 to 48 to 52?

And then also in that number, can you give us a sense of what you think the mix would be WCDMA versus CDMA for the December quarter?

Louis, we do, we are looking forward, we do expect an increase here for the holiday season.

We are anticipating a slight decrease in the handset manufacturers inventory, and I would add that we're seeing some recent strength here in China and India that in recent past hasn't been, we haven't seen that good strong up growth, so I think all the markets, all our regions around the world are on a good path here going into the December quarter.

And how about the mix then, WCDMA versus CDMA?

Well for two quarters now we've had the WCDMA royalties as a percent of our total royalties here in the 25 to 26% range, so I think we'll continue to see good progress there and as we noted here, we do expect the September quarter ASPs to be approximately 213, so that's a good indication of what we're expecting for WCDMA.

Okay.

And then your chipset guidance for December around 39 versus the 38 to 39 in September, that number will stay flattish despite a big pick-up in handset sales just because your capacity constraint?

Is that right?

That's right.

Our demand is certainly greater than that and we have the RF chips at capacity, we could certainly supply significantly more than that to our customer base.

I just wanted to understand the capacity, because in the semi industry we've had for six months plenty of capacity.

Is it a just a matter of timing of bringing up new facilities or is there some sort of yield issue?

No, it's not a yield issue.

Well, it's a yield issue in the ramp-up of the Atmel fab, but it's a silicon germanium fab that we're talking about and the capacity in silicon germanium fab, particularly the two fabs that we're talking about, where most of our products are, is still constrained.

Free Scale has significant demand from two of its main customers, ourselves and their other main customer, and as a result of that increased demand that fab continues to be constrained, but as I say, they are now delivering more to us as a result of, I presume some easing in their requirement from the other customer.

And as I say, we are bringing up Atmel, and Atmel fab is coming up and it's beginning to yield very well and we're ramping volume there also.

Thank you.

Our next question comes from the line of Tim Long at Banc of America.

Please proceed with your question.

Thank you.

Two questions, if I could.

First on the WCDMA numbers for next year, a little bit higher.

I know you don't want to talk too much about where you're putting those extra phones, but can you just give us a sense on how you think linearity will work throughout the year?

Do you think we will, you know, start off with a typical down march and then ramp through, or are you looking at this as if it could be just continued strength in growth all along?

Then secondly, on the expectations for September, where we went from I guess the original expectation was 46 million phones to 41, you mentioned Korea and some inventory build.

I'm just a little confused with the numbers here, Bill.

I'm assuming that would mean that the true-up would have been negative in the September quarter or that would have impacted the September quarter true-up?

Just curious as to how you would still hit that 29 cent number under the old method?

If you could clarify that, that would be great.

I'll take the question on WCDMA.

This is Sanjay Jha.

We see continued growth in wideband CDMA, though as more and more networks get launched, and as the networks begin to stabilize and as handsets based on our MSM6250, very power efficient and other handsets in similar quality of chips come to the marketplace, and the price of handsets begin to come down, we see starting from, starting next year volume increase, and clearly December next year and Christmas next year is going to be important for UMPS volume.

Tim, on your other question, to your point on the true-up, the last quarter for which we were accruing, the actual reports came in and the true-up was at $2 million to what we had accrued.

So that's for the prior quarter.

In the case of the September quarter, if you recall, several weeks ago I estimated 46 million handsets would ship and I'm reporting here our current estimate is 41.

So, had we accrued this quarter, I would have accrued substantially less than what I had expected I would have accrued as of several weeks ago.

So, we're still within the earnings range.

There's been a lot of other pluses and minuses within the business, and I think you would note there are, our tax rate for the September quarter improved a nod as well.

Okay, so that would assume that that would have been incorporated in that 29 cents?

It would have been five million phones less?

That is correct.

Assumption.

Okay, great.

Thank you.

Our next question comes from the line of Tim Luke at Lehman Brothers.

Please proceed with your question.

Thanks a lot.

I was wondering, did Bill give the number of WCDMA signs that you had thought for the September and then the December quarter out of the 41 and the 48 to 52?

Tim, we did not break that down.

What we did say was we think it was in the range of about 23 million WCDMA phones that would ship in calendar 2004.

And for the coming quarter, how should we think about your operating expenses?

It looks, judging by your guidance, that there is, you basically got a flat revenue number with an EPS number that is sequentially lower, right?

Could you just talk us through what's going on there?

Sure, sure.

Say you got a couple of things happening there, Tim.

First on the operating expenses for the December quarter, I expect operating expenses to be up in the 10 to 12% kind of range sequentially, number one.

Number two, you have the effect, the beginning the effect of the acquisitions we're making, and including some write-off of in-process R&D, and number three, you have going from a, if you look at our pro forma, a 29% tax rate to a 30% tax rate.

And could you give us a sense of what's the revenue differential in your guide, in guiding for the December quarter between the change of, versus your estimates of recognizing the royalties from September and December, rather than the December royalties in December?

Yeah, that's a good point.

So, now our December quarter under our new method of accounting for royalties will be based on what we estimated as another quarter of 41 million handsets shipped, similar to the prior quarter.

Had we stayed on the accrual method, as you're noting, Tim, that estimate of 48 to 52 million handsets would have been the basis for the December financials.

So the Christmas effect so to speak, is going to, that we normally would have seen in our December quarter is now going to be reported in our March quarter.

So it's like 205 times 7?

So you, yeah.

Yeah.

You've got that effect of 41 million handsets versus 48 to 52 that will affect both royalties and chipsets to a degree, but we'll see that now in the March quarter.

One last question, if I may.

The shape of the year next year, Bill.

If you're guiding let's say the mid-point of your guidance approximately a buck 20 or something or a buck 18 or whatever next year.

If you start at 25 cents, it's a sort of steady ramp up or your March is basically flat with December?

How does it look?

The shape of the year?

At this point I'm looking forward to kind of a steady improvement through the year.

Thank you guys.

Our next question comes from the line of Tal Liani at Merrill Lynch.

Please proceed with your question.

Hi.

I'm wondering if you can quantify the charge for in-process R&D for next quarter?

And then what's going to be the charge for the following three quarters of the year or maybe for the year itself?

And then do you anticipate further increase in operating expenses in the sequential basis throughout the year in the other quarters of the year?

So, okay, Tal, first on the total operating expenses, I expect operating expenses to grow for the total year in the range of 28 to 30%.

For the first quarter a sequential increase of about 10 to 12%, so we should be looking for a fairly steady increase in our operating expenses as we proceed through the year.

That would be number one.

On the in-process R&D for the quarter, Tal, it's, you know, if I answer it exactly, I'll have to get to a very detailed reconciliation given Reg G requirements.

I would like to just say at this point I'm looking at a 2 cent impact for the full year, and that's a little bit weighted towards the first quarter.

But that's from the acquisitions as a whole, not just the in-process R&D.

And that's inclusive of the in-process R&D.

So when I look--

Let me say this, the in-process R&D as a percent in the context of that 2 cents is not, it's not the majority of it.

Mm hmm.

It's all going to hit in the first quarter, the in-process.

For in-process, yeah.

So when I try to break down the 4 cents, if I take your mid-point of EPS guidance for next quarter and I take the 29 cents to 25 cents decline sequentially, could I assume that maybe 1 cent related to acquisitions and 3 cents will be related to Op Ex and taxes?

I'm trying to--

I would rank them in this perspective, Tal: I would rank it first as the Christmas season effect is moving into March, that would be number one, number two would be the increase in our tax rate, number three would be the operating expenses, and number four would be the effect of the acquisitions.

Okay.

One more thing.

You mentioned on the call that this quarter as well as last quarter you had royalty from WCDMA of 25% of total royalties, and the question is, are you referring to the new method or the old method?

Are you taking into account royalties for what it would have been in the old method or the new method?

Under the new method now, so--

So that means 25% of total royalties is the same, when you say two quarters in a row that means that's the same 25% you reported last quarter?

Actually, 25% two quarters ago and 26% in the recent quarter.

In the June quarter.

Yeah.

So June and September.

You're referring to---

No, March and June.

Okay.

March and June.

Got it.

Thanks.

Okay.

That's great, thank you.

Ladies and gentlemen, that was our final question.

Dr. Jacobs, I will turn the conference back over to you for closing remarks.

Thank you all very much for attending.

Sorry for the complication of this change in how we handle our royalties.

It will be this quarter, hopefully we won't get that complication now going forward.

This next year continues to look very exciting.

We're certainly doing a lot of R&D in-house now, preparing for the many developments with DO, with WCDMA, HSDPA, a lot of the new applications.

We're also looking forward to many changes occurring in several countries.

We expect to see some announcements out of China, hopefully before the end of this coming year.

Thank you all very much for attending.

Ladies and gentlemen, that does conclude the conference call for today.

We thank you very much for your participation and ask that you please disconnect your lines.

Thank you.