Features

Opinion: Committed, but not involved: AT&T raises 3G cash

Don't be fooled. It's probably true that AT&T needs $3bn - every telco in the world is short of cash. But "committed to 3G by 2004" is a meaningless phrase, if not positively misleading.

It's a lot of money, but less than half what the telco giant raised a year ago. According to Cellular News, "AT&T Wireless announced a bond sale totaling US$3 billion yesterday. The offering consisted of three tranches: US$250 million of 3-year notes were sold with a coupon of 6.875%; US$750 million of 5-year notes were sold with a coupon of 7.500%; and US$2 billion of 10-year notes were sold with a coupon of 8.125%. AT&T Wireless also sold US$6.5 billion of bonds in March last year."

The story suggests that this huge money-raising exercise is a sign that third-generation universal mobile telecoms services (UMTS) are just around the corner.

They aren't.

Oh, they're on the way. As we've discussed here on this site several times, the straws are blowing in the wind, at last. The people who make test gear are starting to get orders. There are "trial service demonstrations" at exhibitions. There are exercises like KNP's i-mode launch - which isn't 3G, and thank the powers, they aren't pretending it is any more - which are market research exercises to see what people will spend money on if they have fast data links. And there are plans, tentatively announced, to see 3G services in small, local areas, for 2004 "or later" from several European telcos.

About the only organisation which is doing anything serious is Hutchison Telecom, and it's ordered a couple of thousand handsets.

Let's take a realistic view, here. Remember when the very first mobile phone services started? And how only stockbrokers with strong wrists could use one in public? And now, try to remember how long it was between those irritating days, and the days when everybody had a phone.

At a generous guess, if UMTS goes like a train, it could have as many subscribers as GSM does in five years' time from when it starts. At a realistic guess, it will take far longer; because nobody is planning to switch off GSM and force people to move over to UMTS.

What is going to happen is something like this:

2003: a few trial masts are built, and a few handsets are given to test customers, mostly internal. They take them to parties, where they are stolen.

2004: one or two reckless adventurers actually announce a commercial service, rather like the GPRS services which were announced at the beginning of last year, and are only just working correctly now. Six months later, at least one of them withdraws the service, without telling anybody. Both users complain, and are ignored.

2005: one or two more reckless adventurers, not discouraged by the failure of the pioneers to make any money at all, follow suit, and start "urban, metropolitan" services for areas like the City of London. Handset prices are announced, and cries of pain are heard. Handset prices are hastily subsidised. Two or three hundred users join the network, and discover that they need dual-standard handsets, because the UMTS service is down more often than up.

2005: the number of Bluetooth phones reaches "several million" in Europe. The number of other devices needing Bluetooth to talk to them starts to climb exponentially. Bluetooth access points to the Internet become commonplace in offices.

2006: roll-out starts. Masts cover 50% of London, Birmingham, Munich, Berlin, Paris, Lyon, Geneva, Milan, Stockholm, Copehagen, and Madrid. Brussels coverage reaches 80%, a figure which later turns out to include the i-mode users.

2006: Bluetooth access to the Internet from PDAs becomes standard. Voice over IP becomes standard. Traffic over GSM networks becomes stable, as more and more phone users place calls over the Internet instead, using Bluetooth.

2007: By this stage, at least one, probably two major UMTS licence holders have gone bust, and their licences are squabbled over in Government. Should the new licence holders be able to buy the licence with the bankrupt comms firm? It goes to Appeal. But three major networks now offer pretty good metro area coverage, and handset costs are coming down. A typical 3G phone costs only twice what a top-of range GSM GPRS Bluetooth device does; and the Bluetooth traffic becomes large enough that Vodafone and mmO2 both sign deals with Bluetooth syndication companies like Boingo.

2008: 3G rollout reaches most suburbs of most big cities. Countryside services are minimal. High-speed Bluetooth access on trains and aircraft becomes standard. Demand for rural UMTS is high, but the cost of coverage prohibits further expansion, (exactly as demand for rural cable is high today, but the cable companies refuse to provide it).

2009: Hutchison claims that its UMTS service is "cash-generative" but refuses to reveal when it thinks it will get into profit. Vodafone reveals that its Bluetooth franchise is making 10% of the revenue of its GSM network, at twice the profit margin.

2010: The Isle of Man announces a 4G network. Shares in 3G firms crash.

We can be pretty sure that this prediction, like any prediction, is wrong. Predictions are never right; nobody can read the future. However, I'd be prepared to bet the odd fiver that however badly wrong it is, it's closer to the truth than "AT&T will have substantial 3G phone networks running by the end of 2005."