Reliance's gigabit gamble

It's a move that has left competitors divided. Is it a masterstroke that will catapult the Reliance Group into the glamorous, fast-moving world of international telephony? Or is it an overpriced acquisition in a field plagued by overcapacity that should have been left well alone?

Flag Telecom Group Ltd, which the Reliance Group has offered to buy for $207 million, isn't a giant of international telephony.

The UK-based company, headquartered in Bermuda, is a mid-sized player that controls a 50,000-km global undersea cable network linking Asia, Europe, USA and south east Asia. Its customer list includes 180 of the world's telecom companies.

If shareholder clearance goes through in the coming weeks, followed by regulatory approvals, Flag will be swallowed up and become part of the Reliance Group.

Under the agreement a new company floated by Reliance will acquire 100 per cent in the submarine cable company at a share price of $95.61 -- that's a premium of 50 per cent on the current stock price.

A bandwidth lexicon

Bandwidth: The information carrying capacity of a communication channel. It can either be an undersea cable or a transponder in a satellite. It is a medium through which voice and data travel from one part of the world to another. Telecom companies use bandwidth to connect calls from, say, India to other parts of the world, call centres use it to answer and receive queries from all over the world. IT companies use it to transfer software to their clients abroad and ISPs use it to connect you to websites which have servers in the US. Bit: The smallest unit of data in a computer Bits per second: A measure of the amount of data that can flow in a given time which is in this case one second One kilobit per second: 1,000 bits of data or voice travelling in a second One gigabit per second: A billion bits of data and voice in a second One terabit: The movement of one trillion bits of data or voice in a second STM: One gigabit per second is equivalent to STM 8

Is it a corporate coup or not? Flag has been on the block for over a year but hasn't had many suitors. Insiders say VSNL looked over the company and was ready to offer around $100 million for it.

Other well-known names in the international telecom world like Cable & Wireless also put Flag under the scanner but didn't get closer to striking a deal.

Reliance clearly looked at it differently. For the group the deal has several pluses.

If it goes through -- Flag's shareholders still have to clear the deal -- Reliance will be catapulted onto the international telecom scene and it will become a key player in the international bandwidth market.

Also, Reliance believes that the telephonic data market in India will boom. If that happens it will require huge bandwidth and connectivity to international markets which Flag will provide.

In a short press release, Reliance pointed out that it "believes the acquisition price of $207 million is very attractive and offers great value to Flag's shareholders."

It also said that Flag's global fibre optic network will complement Reliance's next generation digital network in India and enable Reliance to serve its customers worldwide more effectively.

Will the deal make international long-distance calls from India cheaper?

For most people who make voice calls out of India there will be only a minimal difference. But people who send huge quantities of data -- like IT and BPO companies, ISP providers and even corporates -- could find prices dropping steeply.

Competitors are cautious about the new deal.

Says Siddharth Ray, chairman of Data Access, one of the leading international long-distance operators: "As we do not own capacity in cables but buy them from others this will offer us an excellent opportunity. Instead of two players (VSNL and Bharti) now we will have three players to choose from. We expect data prices to fall. But voice tariffs might fall only marginally."

A source close to VSNL (which is controlled by the Tatas) -- says the deal won't affect the company at all.

"Corporates look at redundancy-alternative cable routes in case something goes wrong. That is what VSNL offers (it has capacity in other cables which come to India), so we don't see BPO and other companies depending on a company with only one cable."

They also point out that VSNL will be adding new capacity from India, so it isn't threatened by the deal.

So what made Reliance decide to buy Flag? After all, the cable company has been in serious trouble, facing a swelling debt burden which forced it to file for bankruptcy in the US.

It's now out of Chapter 11 in the US. But it's still piling up losses; for the six months ending June 30, 2003 the company ran up losses of $41 million on a $63.1 million turnover.

Reliance executives aren't talking about strategies for the future because they are in a 'quiet period' until the deal is approved by Flag's shareholders.

But close watchers of the deal say it makes a perfect strategic fit for the company. Reliance Infocomm is banking heavily on a revolution in the data market.

Currently, data constitutes only 5 per cent of the ILD business, and the rest is voice traffic. But Reliance reckons that in the next five years 40 per cent of its revenues will come from the data business.

Cable crisis

It was one of the great excesses of the '90s. As the telecom industry boomed it seemed obvious that there would be a great future for anyone who built cables to meet the anticipated jump in demand. What went wrong? In short, capacity was way in excess of demand. The result: companies like Global Crossing with 150,000 km of cables filed for bankruptcy protection in the US last year. It had run up staggering debts of $12.4 billion. Flag, which was one of the first cable companies built essentially with non-telecom capital (it's not a consortium cable of telecom companies) ran up debts of over $2.6 billion and also went into Chapter 11. Other companies like 360 Networks and Carrier 1 ran into similar trouble. Looking back, their problems were predictable. The cable creating spree resulted in capacity climbing by 23 times between 1999 to 2002. But the recession in the global economy took its toll. One estimate is that almost 100 terabits of total bandwidth capacity will be available in 2004 globally. But demand will probably be barely 20 per cent of that. Also, the data market hasn't grown as swiftly as expected. Globally, data accounts for about 10 per cent of total traffic. Even in advanced countries like the US data accounts for 30 per cent of revenue. The only exception is Korea and Japan where it has zoomed to 40 per cent of traffic. Says Data Access's Ray: "Data will not be a big revenue generator in the next 10 years." Predictably, bandwidth prices have plummeted -- by around 50 per cent on major routes every year on a compounded basis. And as a result cable companies can't finance their bulging debts. The good news of course is that Flag and Global Crossing are slowly coming out of the sea of red. Both have been restructured and have found new buyers. But they'll be moving into the future with extreme caution.

Why does it expect data demand to go up dramatically? One, Reliance watchers say the company believes that bandwidth prices controlled by VSNL are currently too high.

Therefore, if prices fall, demand will increase. Also, it believes that capacity has been virtually rationed out and once it is released, demand will follow.

Is Reliance right? It costs about $4.5 million to buy STM 1 capacity (that's a very large unit of capacity) between India and the United States. Between Japan and the US similar capacity is available for 10 per cent and 20 per cent of that price.

Says a Reliance watcher: " Reliance is expected to drop prices dramatically. We see prices falling by 70 per cent to 80 per cent."

Secondly, Reliance watchers say it expects large demand from its own domestic fibre network.

Says a Reliance watcher: "They have already lit six ducts each with large bandwidth capacities. So, obviously, they need similar connectivity out of India. They cannot build a highway in India and then connect it by a village road to other countries."

Also, ILD operators need to have hubs in other countries like in London or the US from where they can redistribute data and voice to other countries. That reduces costs dramatically. And Flag will provide that facility for Reliance.

But is the gamble worth taking? And has Reliance paid the right price for Flag? Opinion is divided. Some analysts think it has paid a very high price.

Says a telecom analyst: "I don't think it is worth more than $100 million to $120 million."

He points out that S T Teleport (a Singapore government company) paid $250 million for a 61 per cent stake in Global Crossing the sub-sea cable company, which is three times the size of Flag.

But some disagree. Says a Delhi-based equity analyst: "Following the court proceedings, the company's assets were written down from $1.7 billion to $300 million to $500 million. So the Reliance price is fair."

Telecom experts say Reliance might be overestimating the market. Says a senior executive in an undersea cable company: "There's no way that demand will suddenly growfrom 5 gigabit a year by three and four times to justify owning bandwidth instead of leasing it."

Also, new capacity is being added on the lucrative Indo-Europe route. For instance, in SeaMe We 3 -- the consortium cable in which VSNL is a partner -- capacity is being enhanced by 38 per cent.

Then, there will be SeaMe We 4 (again VSNL is a partner) which should come into operation in 2005. And Bharti-Singtel has 8 tera bits of capacity on the Chennai-Singapore route, which can also be used to route traffic to the US.

There are other serious issues for Reliance. One is the tie-up that Flag has with VSNL. Sources close to VSNL say Flag has a deal under which it will use a VSNL landing station (where the cable lands) for the life of the cable (another seven to eight years).

Says an ILD operator: "Effectively, it means VSNL will control the capacity on the fibre. (Under the new policy, VSNL cannot deny access to any ILD operator in India but there are fears it could squeeze business.) It also means Reliance can't build its own landing station to pick up Flag traffic."

Could VSNL make it difficult for other private ILD operators who want to use Flag? That could be a worry. In fact, Flag's key weakness could be that it doesn't own the landing stations in most countries.

Says an analyst: "Whether it is VSNL in India or in UAE the usage of bandwidth is rationed or determined by the ILD companies. That is a sticky situation to be in."

But Flag also has problems in south east Asia and on the Atlantic routes point out experts. For instance, it faces a large oversupply situation in the Europe-US route where supply is 14 times demand.

Worse, its north Asia loop bypasses Singapore, which is becoming the hub in this region for business.

But Reliance's entire gameplan is based on its calculation that the data market will grow in leaps and bounds. The company has already started wiring up over 500,000 commercial buildings across the country to offer end-to-end communication solutions to companies.

It is also making forays into business process outsourcing and call centre business. That's not all -- it will offer customers a range of services including broadband, video on demand and IPTV (it has tied up with Microsoft).

Some ILD experts believe Reliance can leverage its strengths to ensure that Flag doesn't drown.

For instance, it can upgrade capacity on the India-Europe route by at least four times.

Secondly, it could reallocate the company's network operations business from London to India and save around $10 million.

Thirdly, it could use its muscle to reduce maintenance bills and save another $5 million to $10 million.

Says a sub-sea cable expert: "Through their management abilities I think they can start making operational profits within two years".

But the entire strategy will depend on whether Reliance has read the data market right. There are immense growth possibilities but no guarantees about how much business India will get. Take a recent case.

A US-based banking and financial powerhouse has been considering reallocating its data disaster recovery centre from Singapore to either India or Hong Kong for which it needs 2 STM of bandwidth capacity.

Indian ILD companies have quoted $20 million dollars for 15 years, but Hong Kong is ready to do it for only a few million dollars. The choice is obvious. But can Reliance drop prices dramatically to woo such big users to set shop in the country?