Tero - I don't see how infra sales could have dragged down the handset/chipset sales growth from 50% to 24%.
As you say, the company did not break down communications systems revenues, but I would guess that in the year ago Q infrastructure and Omnitracs made up more than 50% of comm systems revenues (remember that Omnitracs was, for a long time, a big cash cow for Qualcomm that funded them during their CDMA rollout). If you assume the combined Omnitracs and Infras revenues did not grow over the last year, but handset and ASIC sales grew by 50% then you get total rev growth of 25%. (in reality I would guess that handset/ASIC revenues grew >60% Y/Y, but infras lost revs.) See my guestimates:
exchange2000.com
Looking at the overall CDMA subscriber growth and adding the replacement sales to that I don't see how the global CDMA handset/chipset growth could have been lower than 100% during last quarter.
Their unit growth was >100, but we have to guess at revenues. As for price erosion of Qualcomm products (ASICs and QPE phones), the gross margin for phones is substantially above what it was last year, as is the comm systems GM. Both of these indicate that price erosion wasn't that steep.
Didn't Qualcomm say that network sales are picking up before they sold the division?
Actually I do not remember such a statement in regards to this last quarter. In the last quarter of last calendar year they shipped a lot of product to Mexico, but this last quarter was probably pretty weak. Also, note that in the year ago quarter Qualcomm shipped a $10M(?) G* gateway.
wouldn't you say that the Qualcomm sales growth is somewhere between 30-50% of the overall CDMA handset/chipset sales growth? I'm not saying that this has to be lethal - but it is weird and something that should be balanced against the projected licensing revenue growth. Qualcomm's position as the creator of IS-95 should have been strong enough to help the company at least match the industry growth.
I think the best way to measure Qualcomm's product revenue growth relative to the total CDMAOne growth is to measure the growth in royalties vs the growth in product revenues. In the last year the royalty rate grew at perhaps 100%. I would be very surprised if Q! handsets were not losing market share (I would guess, based on ASP erosion rates and units sold, that handset revs only grew at 50-60%), but ASICs are a different matter. Again, based on ASIC units shipped and ASP erosion I would guess that they kept up with royalties. But I agree that this is based on estimates.
Qualcomm does not seem to have much production capacity in handsets and they are extremely cautious in bulding more. They seem to be saying that the phone production capacity will go up just moderately, to 1 million units per month at the year's end. This does not display much confidence in their ability to match the market growth in the future.
Much as I would like to disagree with this, I can't. I started 'bemoaning' the same thing back in January ('bemoaning' is in scare quotes because it is hard to bemoan near 100% unit growth). The only comments are that I suspect that they are beng conservative in their predictions (i.e. by the end of the year I'll bet they are doing 1.1M or more per month which is still >100% unit growth in a year.), and they are using the handset sales largely as a method of validating their ASICs et al.
Qualcomm is riskier than Nokia. I think that's at the core of this debate - when you look at the P/E ratios of Qualcomm, Cisco, Lucent, etc. you see that the markets are assuming that American IT companies are far less risky and/or far more likely to grow faster than their European counterparts. I don't think that this is supported by any evidence. Markets are already pricing in falling margins for Nokia and rising margins for Qualcomm. What exactly is the rationale behind that?
Well, I agree that Nokia is oddly priced, as is Lucent in the opposite direction. My guess as to the reason that Nokia is so low priced is that it is for the same reason that I don't buy it. It is very dependent on a 'commodity' market and I have trouble understanding what its competitive advantage is. What is to say that they aren't going to fumble and lose massive market share next year. (My point is not that Nokia doesn't have a competitive advantage, but it isn't clear to me what it is.) Lucent is over priced for the inverse reason - patents are easy to understand in their competitive advantage. Finally, note that Qualcomm is priced at maybe 30 times reasonable estimates of their earnings in calendar 00. Hardly overpriced especially in a company which has a competitive advantage that I can understand (having watched powerhouses try to produce their own ASICs and phones).
Clark |
