The Death of CDMA in India
CDMA, which stands for Code Divison Multiple Access, was a second generation telecommunications standard pioneered by chip maker Qualcomm. GSM, thanks to its wide adoption in all the European countries had gained critical mass much before CDMA and became the dominant standard for 2G telecommunications all over the globe. However, CDMA still managed to find a firm footing in America and Japan.
CDMA in India
CDMA as a technology had managed to find a use case in India by acting as substitute for wired landline connections. Wireless Local Loop (or WLL) was widely used by BSNL and Tata to provide landlines that could be used anywhere. Remember Tata Indicom’s Walky? BSNL being a landline operator had lower interconnect charges and WLL operators tried to leverage the same but had failed. After their inability to have the same interconnect charges as that of BSNL, WLL operators tried to convince TRAI that since CDMA required lesser spectrum than GSM, they should be allowed to provide full fledged CDMA services on the spectrum they got for WLL. With the migration of CMTS licenses to UASL based licenses, spectrum allocated for WLL i.e the 850 MHz band could now be used to provide full fledged CDMA services including calls and SMSes.
Migration of CMTS licenses to UASL had given life to CDMA telephony services, but the real kickstarter for CDMA came from Reliance and its Monsoon Hungama offer where Reliance had partnered with manufacturers like ZTE and flooded India with cheap CDMA handsets with voice and SMS offers bundled. Reliance through its cutthroat prices and CDMA network had managed to drop call prices significantly in India which really helped spread basic telecommunication services to the masses.
But why did CDMA fail?
However, since CDMA was pioneered by Qualcomm, almost all CDMA handset manufacturers had to pay a significant royalty to Qualcomm for its patents. As far as GSM was concerned, the patents were held by a consortium of key manufacturers such as Nokia, Motorola etc and they cross-licensed each other’s patents which didn’t really make the royalty rates apparent, but this cross-licensing also meant that for a long time other than the manufacturers who held the patents, no one else was able to make GSM handsets. In case of CDMA, Qualcomm which was the major holder of patents didn’t divulge into making handsets itself, rather licensed it to other interested in making handsets. However the upfront royalties really discouraged GSM manufacturers from making CDMA handsets. Apart from that, both Tata Tele and Reliance applied and got approval for dual license where they also got GSM spectrum allocated to them in areas where they previously only had CDMA spectrum.
This had now led to situation where almost all operators in India were having a GSM based network with some like Tata Tele and Reliance operating a CDMA based network as well. As GSM was the omni-present network and royalty rates for CDMA were high, GSM based handsets became the norm and CDMA based handsets became the exception. In fact India’s most dominant manufacturer during those periods – Nokia – had completely stopped production of CDMA based handsets.
Another issue in India was that handset sales were completely de-linked from the process of getting a mobile connection. Consumers in India would first buy the handset of their choice and then select the network operator they would want to use their handset on. This incetivized handset manufacturers to produce GSM handsets as a GSM handset would work with all carriers whereas a CDMA handset had a limited market. The situation was opposite in American and Japan where consumers first chose which carrier they would go with and then chose a handset provided by the carrier.
As far as handsets were concerned, GSM had pretty much won the game in India. The last CDMA operator in India i.e MTS had entered during the 2G spectrum scam and was the lone CDMA-only operator in India. However, MTS never allowed people to use any of its services on third-party devices and allowed their services only on devices MTS provided in its stores. MTS got all its licenses barring Rajasthan (cancelled in the Supreme Court judgement regarding the 2G spectrum scam). In the subsequent auctions, MTS had managed to bag spectrum in a few selective circles of India.
Splurge in data card market
Even though CDMA operators had lost the battle of handsets to GSM, they had managed to gain a solid foothold in the data card market. In its initial years, the Indian telecom industry was mostly voice led with SMS being a small part, but by 2010, data had started gaining significance and was going to be major factor in future. As far as data was concerned, 3G networks were going to be its real enabler.
GSM operators had to bid huge sums of money for 3G spectrum in the 2010 3G spectrum auctions. Also since the spectrum used by GSM operators for 3G operations was 2100 MHz band, its coverage was weak in the initial years.
By comparison, in case of CDMA, its 3G was EVDO and it could be enabled by using the 850 MHz spectrum already allocated to telecom operators. The 850 MHz band, being a low band spectrum, provided better coverage and in building penetration than 2100 MHz spectrum. Also, CDMA operators already had the network architecture in place with 850 MHz spectrum and just needed to upgrade their cell sites to support EVDO equipment. GSM operators on the other hand had to build a completely new network architecture for 3G over 2100 MHz band.
Data was of great significance in India, but wired broadband networks were limited either because of last mile coverage or the minimum plans being too expensive for some people. When it came to wireless broadband networks, data cards were the closest replacement to wired broadband networks. Now as far as the data card market was concerned, Tata, Reliance and MTS had cornered it with their data card offerings such as Tata Photon+, Reliance Netconnect Broadband+, MTS MBlaze. State run BSNL also provided unlimited EVDO for a flat fee of around Rs 600-700.
The data card market was what kept CDMA alive for several years despite its loss in the handset market. However, recently, even this no longer seems to be the case. CDMA as a whole is sooner or later going to die in India and several incidents point to its demise.
Tata Docomo surrendered all extra CDMA spectrum beyond 2.5 MHz in all telecom circles of India barring Delhi and Mumbai. Tata Docomo has also recently decided to close its CDMA network in Andhra Pradesh. MTNL has surrendered all its CDMA spwctrum in Delhi and Mumbai.
What Tata Docomo and MTNL did doesn’t matter much, but the biggest blow to CDMA came from Reliance Communications and MTS. Reliance Communications has already begun a process to merge MTS with itself.
Apart from merging MTS with itself, Reliance Communications is in the process of liberalizing all its CDMA spectrum holdings in India so that they can be used for 4G. After liberalisation of its CDMA spectrum holdings, Reliance Communications plans to share or trade them with Reliance Jio at selective circles and the duo plan on using the airwaves for 4G. As a result of this, users all over India have started receiving messages and email from Reliance Communications to either upgrade their CDMA sim to 4G or move on to GSM. In a phased manner, Reliance Communications plans to decommission its entire CDMA network all over India and refarm the CDMA airwaves for 4G.
With Reliance Communications, MTS and MTNL completely shutting down their CDMA networks in a few months, the number of CDMA users would come down drastically as Reliance and MTS combined owned a vast majority of CDMA subscribers in India.
The only CDMA operators now left in India are BSNL and Tata Docomo but both of them are in a weak financial position. Their CDMA network is currently being used by data card users for the most part, but once Reliance Jio enters the market with such excess capacity in the form of 20 MHz pan India 2300 MHz spectrum, it’s difficult to see how the current Tata Photon and BSNL EVDO users wouldn’t jump ship and buy a Reliance Jio Mifi device especially since the tariffs are going to be attractive as Reliance Jio is a new operator and needs to fill in capacity.
Whatever be the case, CDMA in India is gonna die sooner, rather than later.
Was this inevitable?
Spectrum is a scarce resource and the problem is more significant in India as telecom operators on average have much more limited spectrum than their global peers. Once a particular technology becomes too old, the logical path forward is to disband that technology and use spectrum devoted towards that technology for newer technologies.
Spectrum refarming is the process of using spectrum allocated for an old technology like say 2G and repurposing it for a newer one such as 3G or 4G. GSM operators in India have started refarming 900 MHz spectrum that was traditionally used for 2G and are now using it for 3G. Idea in Delhi / Maharashtra, Vodafone in Mumbai and Airtel in Andhra Pradesh / Karnataka have started using 900 MHz band for 3G.
The 850 MHz band had first been used for providing WLL services, then used for providing CDMA and EVDO as well. EVDO is now an old technology and it makes total sense for telecom operators to refarm spectrum used for EVDO to provide LTE services considering that the penetration of EVDO compatible handsets is very low and that almost all EVDO operators provide services on data cards locked to their networks.
However, apart from the natural evolution of disbanding older networks and using the spectrum freed for newer technologies, market forces also have a role to play. In this case, the market force was Reliance Jio. Reliance Jio already had pan India 2300 MHz spectrum and had also selectively bid for 1800 MHz spectrum in circles. However to ensure great in-building penetration and great coverage, Reliance needed low band spectrum as well.
The 700 MHz as has been seen by Telecom Commission’s final pricing was prohibitively expensive. As far as the 900 MHz spectrum was concerned, the current incumbent telecom operators would not let go of it since they needed 900 MHz for preserving their 2G networks as well rolling out 3G on it. This left only one low band spectrum i.e the 850 MHz spectrum. Reliance Jio was quick to buy 850 MHz spectrum in certain selective circles at the 2015 spectrum auction but to get pan-India 850 MHz, Reliance Jio had to go through the trading and sharing route and this only led to Reliance Communications merging MTS with itself and liberalizing all its 850 MHz holdings on a pan-India basis to be used for 4G.
Is this a good thing?
Technically, YES, as it expands the ecosystem significantly. Penetration of EVDO handsets in India is very low compared to that of HSPA compatible handsets. This means that anyone who wants high speed mobile data on their smartphones is pretty much stuck with HSPA. However, LTE smartphone shipments have been on the rise recently, repurposing 850 MHz spectrum for LTE would give smartphone users another avenue to access high speed data that has great coverage.