POST OF THE DAY
QUALCOMM, Inc.
Creative Slidesmanship at 3GSM WC

Format for Printing

Format for printing

Request Reprints

Reuse/Reprint

By InCards
February 18, 2005

Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light. How are these posts selected? Click here to find out and nominate a post yourself!

QUALCOMM's "3GSM CANNES" presentation slide set (31 slides) is here:

http://www.qualcomm.com/ir/PDF/3GSM0205.pdf

It is a good set of slides -- I consider it to be an excellent set of slides -- but I do find humor (for lack of a better word) in Slide 20 which indicates that 28% of handsets sold into channels last year were 'CDMA' (CDMA or WCDMA).

I'm sure that slide went over BIG with the audience that attended their briefings at the 3GSM World Congress, and that nobody noticed that along with CDMA2000 Release D 1xEV-DV disappearing with the removal of the CSM6M700 and MSM6700 from their chipset roadmap, so did the disappearance of ~91 million of the GSM (or PDC, TDMA, iDEN) handsets that were sold into channels last year.

QUALCOMM's source for that 28% technology share claim on which they receive royalties is their own 'CDMA' handset sell-in estimate and Strategy Analytics' March 2004 handset forecast of 589 million handset units.

The problem with the slide, and one that I'm sure was immediately picked up on by most of their audience of analysts, journalists, and prospects is that in calendar year 2004 Strategy Analytics upped that forecast 3 times after making their March forecast before publishing their preliminary estimate of handset sell-in on January 26, 2005, and their final estimate on February 5, 2005. I'm sure QUALCOMM's conundrum this year was in finding any reputable firm to back up their long-standing "GSM is going out of business" thesis, so they dug deep in the archives, to find something to advance the thesis, lending weight to the old adage that leopard's simply can't change their spots.

Here is QUALCOMM's considerably less than credible global view of 2004 handset sell-in extrapolated from slides 19 & 20 presented to the 3GSM World Congress which consciously relies on the obsolete 11 month old March forecast from Strategy Analytics:

QUALCOMM 2004 'CDMA' Sell-In Estimates v. Strategy Analytics (3-04)

                                                                     
CDMA        143m of 589m = 24.3% of all and 86% of 'CDMA' in 2004
3GSM WCDMA   23m of 589m =  3.9% of all and 14% of 'CDMA' in 2004
            ====
'CDMA'      166m of 589m = 28.2% of total 2004 handset sell-in


Here is Qualcomm's credible view of 2004 'CDMA' sales balanced against the up-to-date Strategy Analytics estimate of 2004 handset sell-in:

QUALCOMM 2004 'CDMA' Sell-In Estimates v. Strategy Analytics (2-05)
                                                                    
CDMA        143m of 680m = 21.0% of all and 86% of 'CDMA' in 2004
3GSM WCDMA�  23m of 680m =  3.4% of all and 14% of 'CDMA' in 2004
            ====
'CDMA'      166m of 680m = 24.4% of total 2004 handset sell-in


It should be noted that while Qualcomm estimates that 23 million WCDMA handsets were sold into channels in 2004, Strategy Analytics estimates 20 million which would make 3GSM WCDMA handsets 2.9% of 2004 handset sell-in.

Here is Qualcomm's credible view of 2004 'CDMA' sales balanced against other research agencies that have reported their estimates of 2004 handset sell-in presented in descending order:

QUALCOMM 2004 'CDMA' Sell-In Estimates v. iSuppli (2-05)
                                                                    
CDMA        143m of 703m = 20.3% of all
3GSM WCDMA   23m of 703m =  3.3% of all
            ====
'CDMA'      166m of 703m = 23.6% of total 2004 handset sell-in
                                                                   
QUALCOMM 2004 'CDMA' Sell-In Estimates v. Lehman Brothers� (1-05)
                                                                   
CDMA        143m of 683m = 20.9% of all
3GSM WCDMA   23m of 683m =  3.4% of all
            ====
'CDMA'      166m of 683m = 24.3% of total 2004 handset sell-in
                                                                   
QUALCOMM 2004 'CDMA' Sell-In Estimates v. InStat (2-05)
                                                                   
CDMA        143m of 678m = 21.1% of all
3GSM WCDMA   23m of 678m =  3.4% of all
            ====
'CDMA'      166m of 678m = 24.5% of total 2004 handset sell-in
                                                                   
QUALCOMM 2004 'CDMA' Sell-In Estimates v. IC Insights (1-05)
                                                                   
CDMA        143m of 670m = 21.3% of all
3GSM WCDMA   23m of 670m =  3.4% of all
            ====
'CDMA'      166m of 670m = 24.7% of total 2004 handset sell-in
                                                                    
QUALCOMM 2004 'CDMA' Sell-In Estimates v. IDG's IDC (1-05)
                                                                   
CDMA        143m of 664.5m = 21.5% of all
3GSM WCDMA   23m of 664.5m =  3.5% of all
            ====
'CDMA'      166m of 664.5m = 25.0% of total 2004 handset sell-in


As a somewhat humorous (for lack of a better word) aside, Tony Thornley of Qualcomm used an almost identical slide (although at that time they hadn't reduced their 2004 handset estimate so 29% rather than 28% appeared in the balloon denoting 'CDMA' market share) at the Lehman Brothers Technology Conference on December 9, two days after Tim Luke and Stuart Jeffrey who sponsored the event updated their 2004 handset forecast to the total above.

Because sell-in estimates do vary, here is Qualcomm's credible view of 2004 'CDMA' sales balanced against a composite average of the 5 research agencies listed above that have reported their estimates of 2004 handset sell-in:

QUALCOMM 2004 'CDMA' Sell-In Estimates v. Industry Consensus (2-05)
                                                                   
CDMA        143m of 679.8m = 21.0% of all
3GSM WCDMA   23m of 679.8m =  3.4% of all
            ====
'CDMA'      166m of 679.8m = 24.4% of total 2004 handset sell-in


QUALCOMM's variance to this consensus is 91 million units and attempts to inflate market share of the technologies they promote by 3.6% full percentage points (15%) above and beyond a reasonable estimate of that share.

It should also be noted that Gartner's preliminary estimate of 2004 handset sell-thru to end-users for 2004, as opposed to sell-in into channels, is 660.2 million units.

Since Nokia is the only company I am aware of that reports their view of handset sales by technology every quarter for the preceding quarter I do not have many data points for comparison but this was Nokia's view of Q4 2004 handset sales, and it should be noted that their view of 192 million units of sell-in for the quarter, while higher than Motorola's and other handset vendors, is quite consistent with what the research agencies have reported as a consensus.

Nokia: Global Device Market in 4Q04
                                                                   
TDMA      2M   (-31% QoQ)    1.0%
iDEN      4M   (+26% QoQ)    2.1%
PDC       5M   (-25% QoQ)    2.6%
WCDMA     8M   (+83% QoQ)    4.1%
CDMA     36M   (+19% QoQ)   18.6%
GSM     137M   (+23% QoQ)   70.6%
Other�    2M      N/A        1.0%
        ====               ======
        194M               100.0%
                                                                   
Total 'CDMA' [CDMA + WCDMA]= 44M � 22.7% of total unit sales
                                                                   
Rounding? Some analog? Missed CDMA or WCDMA?


Qualcomm estimated ~47.5m 'CDMA' handsets sold-in Q4 2004, and Strategy Analytics estimated 197 million total handsets sold-in, so calculating share from these estimates yields's 24.1% technology market share for 'CDMA' [CDMA + WCDMA] handset sell-in.

FUD

The acronym FUD which stands for any conscious attempt to create Fear, Uncertainty and Doubt is often used on financial message boards, and sometimes takes on a life of its own there. Let it be known that I freely and consciously acknowledge that what I have posted above is pure, unadulterated, FUD. What I have posted, although 100% accurate (if I did not make any transposition or calculation errors in which case I request that anyone point those out for me), unfortunately creates fear, uncertainty, and doubt, about QUALCOMM's capability or willingness to reasonably or accurately give a credible global view of handset sales, past, present, and future.

That is doubt, that as a long-time, long-term QCOM investor I continue to live with, even if I don't approve of it, or accept it. Qualcomm has, IMO, demonstrated considerable and consistent maturation in the last 3 or 4 years, in terms of how they publicly present themselves, their financials, and their technology. This is an area, however, where I believe they have room to mature -- considerably.

I have taken Nokia to task in the recent past for giving an unrealistic global view of only 640 million units sell-through in 2004 (v. Gartner's 660 million), as though they could magically make their dramatic loss of market share in Q1 & Q2 disappear, but Qualcomm's view of global handset sell-in for 2004, is not just unrealistic, it is surrealistic, and it strains credibility to the breaking point.

Conversely I have given Ed Zander of Motorola an attaboy for calculating their Q4 2004, and CY2004 market share reasonably in-line with consensus estimates, and more realistically than any other handset vendor. I am sure that the analyst community gives him high marks for doing so, and welcomes that industry newcomer to the wireless industry, and I suspect they would like to see others fall his example.

- IC -


Become a Complete Fool
Join the best community on the web! Becoming a full member of the Fool Community is easy, takes just a minute, and is very inexpensive.