Thursday 2 April 2009

Femtocells in India: No thank you.

So many people ask me if Femtocells will be big in India but I am not sure if I know the answer to that. Honestly I will be surprised if any Indian operators have any plan of Femtocells and even if they are rolled out people might not be tempted.

In a post last month, David Chambers gave an interesting overview of facts and statistics of mobile and internet users in India:

First some details on the current situation in India:
- Population: 1.1 Billion
- 80% live in rural areas and survive on agriculture
- 39% are illiterate
- 27% live below the poverty line
- 77% live on less than US$0.50 per day
- The economy is growing at around 8-9% annually (and has done so for a few years), similar to China, but is still far behind in infrastructure – traffic congestion is throttling and there is not yet a metro/underground in the main cities, although some are being built.
- 2% PC penetration

And from a mobile phone perspective:
- It’s all 2G. No 3G licences have yet been awarded.
- It’s GSM. Both the CDMA technology operators (Reliance and Tata) who have a combined share of around 20% are said to be planning GSM technology rollouts, because the technology is cheaper and there is a wider range of handsets.
- It’s growing faster than anywhere else. Over 9 million new subscribers every month, with around 180million subscribers reported today. That’s still a huge growth to catch up with China, which has almost 500M subscribers, increasing by some 4M/month.
- Around 200,000 towers, with around 20% of towers hosting more than one operator’s basestation equipment.
- ARPU figures vary, with a lot of the newer subscribers probably in the US$ 2-4/month bracket, although overall the industry makes around $8/month
- Call rates are in the order of US 2 cent/minute.
- Yankee reports operator share of Bharti 30%, BSNL 18%, Reliance 17%. Vodafone (formerly Hutch) is growing quickly.
- 2G cellular data cards, using fixed price data plans, are becoming popular in a similar way to 3G cards/USB dongles promoted in developed countries. The data rates and capacity are more limited.

Yet from a wired broadband internet position:
- 3 million broadband DSL subscriptions
- 8 million copper loops capable of delivering broadband.
- 10 million dialup internet users.- Fibre is being laid across the country, but fibre-cuts remain a regular fault
So if we look at the potential for 3G femtocells, we find there is little comfort:
- No 3G licences available yet – this is probably some years off, whilst the country focuses on basic 2G voice/text rollout.
- Virtually no wired broadband to support significant volumes.
- Any 2G femtocells would likely conflict with the tight spectrum reuse and frequency planning of the macro network, so would be discouraged except where deployed and managed directly by the operator in extremely high capacity locations.
- The business case for coverage fill-in doesn’t stack up – broadband is unavailable in areas which don’t have cellular coverage.
- The business case for data in the home is even more difficult, with relatively low levels of domestic computer penetration. There is more likely demand for mobile broadband access via macrocells, using the new HSPA, HSPA+ and LTE technologies when 3G spectrum is made available.

So it’s not a question of deployment of 3G femtocells in the Indian market coming a few years after the developed world. The complete lack of copper loops, and the changing technology that makes it more feasible to deploy wireless broadband than dig up and lay new copper loops. The services that femtocells would offer are therefore more likely to be delivered over the macrocellular network (with microcellular support) in the medium and long term for any developing country. There is some interest in 2G picocells which would be deployed by the operator in enterprise situations to handle high traffic concentrations, but this is a different application, technology and market segment from the 3G proposition.

A similar view is reflected in this article here:

Bharti Airtel, Reliance Communications and Vodafone, all became members of Femtoforum about two years back but don't have any deployment of femtocell in the country. Femtoforum is a not-for-profit membership organization founded in 2007 to promote femtocell deployment worldwide. According to some media reports Bharti Airtel is conducting pilot projects for femtocells. The same is true for Reliance Communications and Vodafone, both are members of Femtoforum but are doing nothing about femtocell deployment in the country. Despite repeated attempts none of the service provider is forthcoming about femtocells deployment status.

In my recent visit to India I found that people have a different perspective of mobiles. For instance people couldnt understand why we use SMS so frequently and as a medium for communication (between friends, couples). India may probably have lowest tariff for voice and that is one of the reasons people use mobiles for. Many people have removed their landlines and use mobiles only, for their calls. Another most common use of the mobiles is to be reachable wherever you are. People havent learnt to switch their phones of silent and hence many places of worship in India are installing Jammers to stop mobiles working while you are thinking about God.

Similarly, people are not too bothered about the internet. They would generally use it on the weekends to write to their, friends, etc. If there is free net available in the office then its a different thing. The net speeds are also not very good and the link is not too reliable. One of the most popular application is Skype follwed by chatting applications.

I met many people who had Iphones or latest Nokia's/Samsung's but when I asked them if they did any data usage on their phones they all drew blank. I found one guy very actively using net on his E71 but he was connected via WiFi.

In this kind of situation, Femtocells may not be of much use to people. Femtocells would be useful as voice boosters but would that justify its cost. I dont think so. The main reason for surge in Mobile takeup is because its very cheap to make calls. You get some very good call bundles at really low cost. There are off peak rates which is 1/5th the normal rate. If the reception of a network is not good in somebody's house, he would change to a different network. In fact even now Mobile Number Portability is not available in India. As a result some people change mobile numbers every year.

With all these things in mind, Femtocells would be hard sell in India.


David Chambers said...

Hi Zahid

Thanks for validating my thoughts.

The article you reference on thinkfemtocell was actually written a year ago (March 2008). I still believe that domestic femtocells would not be relevant in these markets because of low wireline broadband penetration.

Since then I've written an updated view (March 2009), suggesting that enterprise femtocells do have a market in India and elsewhere. Many medium to large businesses are connected via internet today, and femtocells may be a cost effective and efficient way of providing good capacity and coverage for businesses. See

Zahid Ghadialy said...

Thanks David, I keep on forgetting that its already 2009 ;)

Vikas said...

Hi Zahid,

Your article seems impressive, but many of the facts/figures are either pretty obsolete or ..
Anyway, being a telecom professional in India, I would like to point and correct at least some of them here:
1. Reliance has already launched nation-wide GSM service in Jan 2009.
2. India has around or more than 350 million mobile subscribers today. with addition of 15 million subscribers in the month of Jan 2009 only.
3. And myself and many many of my friends use data/internet services/surfing over EDGE on a $15 USD (INR 700/-) per month package that gives us unlimited GPRS/EDGE access over mobile.

Zahid Ghadialy said...

Hi Vikas,

Thanks for the updated figures. As Daid pointed out, the article I took was over a year old. Nevertheless I am glad that you and your friends are using the mobile data packages. You still have to remember that for a very high percentage of people spending Rs 700 on something like this is a luxury.

Nadeem Akhtar said...

Deploying femto-cells in India doesn't make much sense given the poor state of broadband connectivity. This is why in-band relays provide a better way to extend coverage and capacity. Both LTE-Advanced and 802.16m standards are going for such relays in a big way.