The first set of unattributed #'s looks a lot like Chase H&Q. The Asia/Pacific numbers look like 100% Korea, implying zero for the PRC, zero for New Zealand (I know Maurice bought at least one), etc. Even doubling them for an above-average Korean 50% replacement rate.
The second set (Roberts, a guess) does look more realistic.
Now, I'd tend to buy the 70MM number if Unicom spends 2001 installing the network, and doesn't deploy terminals en masse until early 2001. This would be a normal expectation for a "normal" new network with one or two infra vendors, where securing base station locations was a time-consuming process. Unicom has a multiplicity of infra contracts, as well as local terminal mfgrs. who are buying MSM chips. As I recall, Roberts has now factored *some* amount of China sales into his model. So there could be some low millions of embroyonic Unicom handsets that at least make it to the MSM stage in 2001, albeit not into end-users' hands.
More important to the stock market than aggregate stats for 2001 (half of which is already over) is going to be a view of how the likely numbers translates into 2H01 results, and how those trends map to 1H02. We all know 1H01 results for vendor handset sales will suck, and color me naive, but I suspect that in both GSM and CDMA lands the percentage of 2001 vendor handset sales occuring in the second half will be significantly greater than historical norms. GSM vendors will benefit from inventory work-off and (hopefully) a measurable amount of GPRS sales; the CDMA'ers have worked off inventory (MOT was probably the largest offender) and will benefit from SK meeting the market share reduction, and all Korean carriers marketing 1X.
In the U.S., the carriers seem to be doing reasonably well, although Verizon is lagging on a % basis. Sprint has been running a promotion in July where the normal deposit requirements are reduced or waived, and the volume of new adds, based on anecdotal reports, has been rather striking.
L.A. and S.A. probably do need to be reduced somewhat; however, again anecdotally, a mobile phone is increasingly viewed as a necessity, not a luxury, because the wireline alternates are so pathetic. So I'm not sure the demand curve is as elastic as some might surmise. This is definitely a low ASP market, however.
Judging from AWE's recent results, TDMA handset sales aren't falling off a cliff either. Don't know about Cingular - they generated some replacement sales switching the balance of the old GTE and Ameritech bases to TDMA this year, that seems to be over now.
All in all, 2H01 may turn out to be a lot less unpleasant than some might think, but nothing to write home about. (Hold the yacht order.) The real issues are more 2002ish:
- Will GPRS take the GSM world by storm, or will new improved WAP still be viewed by users as crap?
- How many CDMA users will Unicom put on line, and will plain vanilla GSM continue its strong growth in the PRC?
- How well will Sprint and Verizon do with 1X?
- Will Nextel finally quit fence-straddling?
- Will we see a secular wave of TDMA to GSM replacement sales?
- How are U.S. spectrum issues resolved?
- What carrier data billing models work, and what don't work, and what are the implications for the size of the overall wireless goods & services market (i.e. value potential)?
I think these issues, far more than pure 3G stuff, will dominate NOK's and QCOM's next 12 months' stock price action (but not, of course, message board postings).
To me, the item of greatest interest in 2002 will be consumer acceptance, or lack thereof, of GPRS and 1X - not just the technology, but the interface, content availability, pricing - the whole damn value chain, which has huge implications for 3G. If I do not see this taking off in a meaningful way*, my investment dollars (assuming there's still some left) will be taking off outta this sector.
* Eric L. might call this "crossing the chasm" |
