The global market is ready for 3G. However, there are still some constraints to large-scale deployment. Analysts from leading global consultant organisations discuss the future of 3G technologies (WCDMA, CDMA2000 1X, 1xEV-DO, etc.) – the demand for 3G services, its financial viability, and the main bottlenecks...
How financially viable is 3G?
Keith Mallinson: 3G is viable because many of the costs are sunk (spectrum) and because it lowers operating costs on core services including voice.
Manoj Menon: The business case for 3G needs to be practically demonstrated beyond Japan and South Korea. While a number of operators have launched 3G services, it is not straightforward to allocate revenue and costs to 3G networks alone as there is typically a mix of 2G and 2.5G networks in operation. At the same time, a number of 3G revenue contributors like premium content do not necessarily need 3G and can work on 2G as well. Nevertheless, over a longer period of time, countries with higher ARPUs and a higher proportion of non-messaging content will have a more defendable business case than others.
Jose F. Otero: I will confine my answers to Latin America and the Caribbean. At Signals, we consider only EV-DO and UMTS/WCDMA to be 3G technologies. I don't have a definite answer for the financial viability of 3G in either one of the two flavours being deployed or considered in Latin America and the Caribbean – CDMA2000 and UMTS/WCDMA. It all depends on the country's economy, operators' target market and financial situation, etc. For the moment, mobile 3G services will mostly serve the corporate and high-end consumer markets, which means that coverage will be limited to metropolitan and industrial areas. On the other hand, CDMA2000 presents itself as an attractive proposition for fixed line operators seeking to offer WLL services in areas not covered by their existing wired infrastructure.
Michael Thelander: When evaluating the financial viability of 3G, it is important to remember that CDMA2000 is an ITU-endorsed technology. Operators who have deployed CDMA2000 have seen its benefits; the technology essentially doubles an operator's voice capacity relative to 2G, and at essentially the same cost. Further, given the high volumes of 1X handsets that are shipped today (it is virtually impossible to buy an IS-95 phone), operators are not paying a 3G premium for a 1X handset. In addition to voice capacity gains, 1X has allowed CDMA operators to offer introductory data services that can help boost ARPUs for very little incremental requirements in their network.
Now, these very same operators are becoming more aggressive with their 3G plans by deploying 1xEV-DO, even though the technology has been commercially available since 2002. Although 1xEV-DO does require dedicated spectrum, operators only need to add a new channel card in their already deployed 1X base stations to enable the technology.Given the performance characteristics of 1xEV-DO and its ability to offer compelling data services, this is a very powerful attribute for operators who don't want to limit themselves to being providers of voice services. So, CDMA2000 1X and 1xEV-DO, from an expense perspective, are relatively modest, with operators able to roll out 1X and 1xEV-DO as needed.Incoming revenues are also an important part of the equation. Although they vary from market to market and by the type of service that the operator offers, there has been a measurable impact on ARPU for operators like SK Telecom, KDDI, Verizon Wireless and Sprint PCS – the last two benefiting more from 1X as they are in the early days of 1xEV-DO. Even Latin American operators such as VIVO are noticing the impact of launching data services in their networks and this means that 3G is suited to markets outside North America and the developed parts of Asia.
UMTS operators are also starting to benefit from deploying 3G. Certain Hutchison operators are closing in on profitability. Their road has been a bit more challenging from the cost side of the equation versus CDMA2000 operators while the relative differences on the revenue side are less discernible. In large part, these early challenges stem from two factors. First, some operators, particularly those in Europe, spent billions of dollars acquiring new spectrum to deploy 3G. This spectrum, which is at a higher frequency than their existing GSM networks, resulted in the need for additional base stations to fill in coverage gaps due to the reduced RF propagation characteristics of the higher frequency. Although this is not a direct attribute of UMTS, it was a necessary step for them to deploy 3G.
Second, UMTS operators have had to deploy an entirely new radio network versus leveraging existing base stations. Like CDMA operators, they were able to leverage their existing core networks, which in the case of these operators were their GPRS network elements. Now that they have deployed a new RAN, the incremental impact of introducing new technologies like HSDPA will be relatively modest, much like it has been from the start for CDMA operators.
Net net, 3G has gone from bubble to burst and now it is back to being considered a success story for those operators who have pursued it.
Alayne Wong: 3G services have garnered higher ARPUs on the back of a higher use of mobile data services, and this is a boon to operators. The financial viability of 3G is seen from the viewpoint of ARPUs, which have become substantially higher than before. For instance, in South Korea, the figure for 3G ARPU is more than double that for 2G ARPU, which means more revenue is being generated per user. In Australia, the figure was close to half a million subscribers at the end of 2004 and in Hong Kong, the figure stood at over 100,000 for the same period. On demand for 3G services, I can talk specifically about the AsiaPacific market, excluding Japan. There is an increasing demand for 3G in some of the advanced markets in Asia. For instance, South Korea has been a leading market in 3G. Some markets deployed 3G only two to three years ago. Take South Korea. By the end of 2004, there were 9.5 million 3G subscribers – more than a quarter of its total mobile subscriber base. In Australia, 3G subscribers stood at close to half a million at the end of 2004. The figure for Hong Kong was 200,000. Of course, the fastest growing areas for 3G worldwide will be in advanced countries, where consumers are latching on to mobile data. Countries in the Asia-Pacific region where demand will also be good are Australia, South Korea, Hong Kong, Singapore and Taiwan.
What are the key obstacles to the largescale deployment of 3G?
Keith Mallinson: Handset costs and lacklustre demand for data services beyond narrowband SMS are the key obstacles to large-scale 3G deployment.
Manoj Menon: There are seven:
Jose F. Otero: For mobile services, the main barrier is the high cost of terminals. In addition, there is no real value-added proposition (beyond real-time video telephony) that could not be offered with 2.5G solutions.
Michael Thelander: Again, it depends on the technology. UMTS is spectrum challenged in a lot of markets since it is being deployed at 2.1GHz – a spectrum that is not as well suited for RF as 900 MHz or even 1800 MHz. This obstacle is being addressed by "in-band" solutions and a relaxing of regulatory requirements, which allows UMTS operators to deploy UMTS in their existing GSM spectrum, subject to having sufficient spectrum available. The disadvantages associated with the higher spectrum should not be ignored since it will make it prohibitively expensive to deploy any wireless technology outside of large concentrations of population. As these in-band solutions become available – Cingular expects to deploy UMTS at 850 MHz in 2006 – UMTS deployments in markets such as Latin America are more likely. More affordable 3G (UMTS and 1xEV-DO) handsets will also help drive user adoption, in particular among the less affluent and in developing countries. Although higher volumes will be a big contributing factor, the elimination of handset functionality – such as video telephony and MP3 players – will also play a large role. UMTS handsets could fall to below $200 by the end of 2005, largely due to the elimination of this functionality and increased competition among component suppliers and handset manufacturers. To some extent, a price premium with 1xEV-DO handsets could develop due to the additional features on 1xEV-DO phones that can take advantage of the higher data rates. The upside is that these features can also enhance ARPUs, so operators may be more willing to subsidise them and then make up the difference during the subscription period. With Rev. A of the 1xEV-DO standard, these additional features on 1xEV-DO phones are not essential since Rev. A can also provide "cheap voice".
In developing markets, the focus is, and should be, on driving down the cost of the handset and infrastructure. In that regard, 3G beyond 1X will be difficult for the next few years. However, with the introduction of Rev. A of the 1xEV-DO standard, operators in these markets will be able to take advantage of the all-IP nature of the standard and offer real-time services such as VoIP in addition to ARPU-enhancing data services. As previously mentioned, "inband" UMTS solutions are critical for the adoption of UMTS in these markets as well as in non-urban areas.
Alayne Wong: Timing is crucial in weighing capital expenditure against revenue generation, to justify return on investment. Overall, the market is ready for 3G. However, the main obstacle to its deployment remains the rural areas where the population is by and large less sophisticated or less technology savvy. This results in slow service take-up and a drain on operators' cash flows.
What is your outlook on the different 3G technologies (WCDMA, CDMA2000 1X, etc.)?
Keith Mallinson: Demand will increase significantly for both. CDMA2000 will continue to be challenged by its much lower global market share.
Manoj Menon: Each technology will have its own market depending upon the regulatory scenario. The key is probably the business case/revenue models rather than the underlying technology.
Jose F. Otero: Deployment and adoption will be slower than in other markets. We consider EVDO growth to derive mostly from deployment by fixed line operators in Latin America and the Caribbean. UMTS/ WCDMA will be the dominant mobile 3G technology in the region. However, UMTS/WCDMA uptake will be slower than in other regions and during its initial deployment, it will only allow mobile operators to continue depressing voice rates while maintaining their current margins.
Michael Thelander: I have already addressed this question in large measure. UMTS and CDMA2000 both follow very similar evolutionary paths, so from a technology perspective they are comparable. That said, with 1xEV-DO being largely equivalent to HSDPA and Rev. A of 1xEV-DO being largely comparable to HSUPA, it appears that the CDMA2000 and 1xEV-DO migration path has a two-three-year advantage.
Beyond 3G, or Super 3G, is on the roadmap of both technologies with both standards bodies looking at similar technology enhancements, including OFDM, scalable bandwidths (1.25 MHz to 20 MHz) and smart antenna technologies. From a market opportunity perspective, it will be difficult for GSM operators to defect to a CDMA2000 evolutionary path or for a CDMA operator to switch to a UMTS evolutionary path. Thus, over the long term, the percentages between the two technologies should remain roughly the same as they are today. Interestingly, since both technologies are considering similar technology enhancements, there is the potential for compatibility between the two standards. This, however, will be difficult to achieve for political reasons, although it is technically possible.
TD-SCDMA is another technology that cannot be ignored. Although it is attractive for voice capacity, assuming that it can achieve what its vendors claim, the technology is not well suited for high-rate data applications, in particular in comparison with 1xEV-DO and HSDPA. TdSCDMA is also well behind the other two 3G technologies as dual-mode devices are not yet available. Ultimately, this technology will likely be deployed in China, but whether or not it will be deployed in other markets, in particular in the developed countries, is not clear.
Alayne Wong: In Asia, many markets have GSM as the main technology platform, therefore, WCDMA is the natural progression for operators in these countries. New operators have tried to differentiate themselves with CDMA2000 1X, or CDMA1x, which have roaming advantages. There, the absorption of technology will ride on the popularity of the technology platform on which the service provider operates. It will also depend on the mobile data usage in the country.