Trouble for Big LEOs
telecoms-mag.com
Having spent years becoming a reality, but now embroiled in financing and
marketing problems, what future is there for what can be argued are the world?s
most ambitious personal communication systems?
Chris Bulloch
The ?Big LEO?s, otherwise known as the global mobile personal
communications (GMPCS) systems, had problems enough last
autumn. The two pieces of bad news then were, initially that Iridium,
the first and most complex system due to come to market, was
slipping its in-service date from 23 September last year to 1
November. The prime cause was unavailability of tested user
handsets by the earlier date, largely through unavailability of planned
supplies from Japanese manufacturer, Kyocera. The same month
(September 1998) saw the loss of 12 Globalstar satellites in one go,
on a failed launch by Ukraine?s Zenit rocket, and the announced
likelihood that Globalstar might begin service with a reduced number
of satellites, at correspondingly lower performance levels.
But at the time of writing in mid-August, the miserably sparse up-take
could no longer be blamed on equipment unavailability. A partial
excuse was that operating licences for some territories took longer to
secure than predicted. There have also been suggestions that dealers
have seen little merit in the products and accordingly have not pushed
them.
Testing times
The most dramatic news item affecting Iridium sees it under very
heavy pressure, with some commentators accurately predicting its
Chapter 11 bankruptcy (or worse) by the time this issue appears. The
former has just happened: at the time of writing it had just entered into
a voluntary Chapter 11 bankruptcy filing under Delaware law,
following the end of its third 30-day extension -- terminating 11
August -- of its obligation to meet the requirements of its US$ 800
million senior secured credit facility. These included demonstration of
at least 52,000 subscribers -- of which 27,000 must use satellite
voice service -- plus cumulative revenues of at least US$ 30 million.
Although no figures are yet available, clearly this demonstration could
not be made.
The nub of the problem has been severe shortfall in Iridium
penetration figures (subscriber growth). At end-April the company
reportedly had a mere 10,230 voice customers and just over 2,000
pager users, out of 200,000 projected for end-1998, revised sharply
downwards to 27,000 by the end of March 1999. Since then Iridium
has been shy of issuing new subscriber figures, but now it is likely to
be forced to come clean.
What happens next is anybody?s guess. It should be remembered that
a large and vital segment of the overall Iridium system lies in its 12
gateway earth stations, which are owned neither by Iridium nor its
prime contractor Motorola, but by telco service providers who are
also equity holders. The current state of Chapter 11 bankruptcy will
protect it from its creditors while allowing it to stay in business under
existing management, but some other form of restructuring would
clearly have been preferred. But this approach will allow Iridium to
practice its business for four months. Iridium is now promising a
restructuring in 30 days.
The smart money has been suggesting a wholesale takeover by
Motorola. But the latter might find it cheaper to let Iridium fail
completely (i.e. go into liquidation) and then try to make something of
the residual assets later. The question as to whether Motorola can
afford a near-term bailout of Iridium and its debts is also becoming
pertinent, but so far its support continues. So does that of Nippon
Iridium Corp, the second-largest investor.
ICO?s story
ICO Global Communications is now frankly admitting that it is up
against the wall financially: it failed to attract the minimum US$ 500
million of its needed final US$ 1 billion, by way of a once-extended
discounted share rights offer to existing shareholders. Unexpectedly at
the end of August, ICO filed for reorganisation under Chapter 11 in
the US Bankruptcy Court, Delaware. The application was made in
the name of three subsidiaries which can make use of this protection;
the holding company, registered in Bermuda, cannot.
ICO is the only one of the original three systems which has yet to
launch. Strictly speaking it is a MEO (medium earth orbit) system at
around 10,000 km altitude. It is also the only one without massive
backing from a large quasi-parent company, given Inmarsat?s attitude
of detachment from the company.
ICO is proceeding well with preparations for system rollout in 2000,
including training, where Iridium was very remiss. Its latest coup is
arranging its first US roaming agreement with AT&T, which has over
10 million subscribers and 250 markets in the US. However, ICO still
does not yet have a solution to the S-band frequency conflict in the
US, where it may yet have to pay large sums to get existing users
(chiefly satellite news-gathering operators) to relocate to other
frequencies. This matter is still under study.
The main technical flies in the ointment consist of launcher problems.
Proton, booked to fly the first and three other launches, has not only
been shut-down for investigation of its second stage, which failed
(with a Russian satellite) this summer, but more critically, the
Kazakhstan government has closed the Baikonur base for launches of
any vehicle. Reason: the last failure landed on its territory. The RL-10
upper-stage investigation has grounded not only the Boeing Delta III
(booked for five launches; it has yet to perform successfully), but also
the Atlas IIAS (two launches). The only ray of light is the successful
first (dummy) mission of Sea Launch, previously booked for the 12th
and final flight; it now looks as though more may be needed.
The Globalstar experience
Following its Zenit disaster last September, all has gone well with the
other LEO system -- Globalstar -- which has always made a virtue of
its simplicity compared with Iridium, as well as its lower end user cost
(but that was before Iridium?s recent cuts). Globalstar is of course still
installing its satellites. Its big find was Starsem, the Franco-Russian
venture blending western launch service standards (and dispenser
technology) with the demonstrably good record of the veteran Soyuz
booster -- some 1,500 successful missions since Sputnik 1.
Globalstar has already performed three launches (four satellites at a
time) on Soyuz and has three more booked for later this year,
provided Baikonur reopens (see above). Boeing?s Delta II is handling
the rest. Globalstar has not yet reported a single post-launch in-orbit
failure, unlike Iridium, which in March had admitted to 12.
However, not everything is quite so rosy on Globalstar?s financial
front. After announcing last year that the system was funded to
?completion?, the partnership now discovers that some US$ 565
million more is needed before beginning service in September. Part of
this is related to additional launches to replace those lost on Zenit --
these were insured -- but other items (ground spares, for example)
were foreseeable. Despite the jittery state of the market, Globalstar
has succeeded in raising a further US$ 500 million credit facility,
underwritten by the Bank of America, which is guaranteed by
subsidiaries of Loral Space & Communications, owner of 42 per cent
of Globalstar.
The company now expects to begin a ?regional rollout? of commercial
service this month, with 32 satellites and nine gateways in operation.
This is expected to provide ample coverage to get revenues flowing;
we shall see. By the end of this year, a total of 48 active satellites plus
four spares should be on station, and 16 gateways operating. The
remaining 22 gateways should come on stream in the first half of
2000. ?A sufficient number? of handsets from Ericsson, Qualcomm
and Telital to meet demand are promised to be ?in the distribution
pipeline? by year-end; production is set to reach 40,000 per month
by early next year.
[djane note: article apparently is missing a section here on Iridium]
The company has meantime been undergoing a top-level revolution,
with both its CEO, Ed Staiano, and CFO, Roy Grant, departing their
respective hot seats in April. Staiano has been replaced (it is said
provisionally) by John Richardson from Iridium Africa, and Grant by
Leo Mondale, previously in charge of business development.
What went wrong?
The real nub of the problem has to be that too many of the
repondents to the innumerable surveys conducted throughout the
1990s -- and earlier -- by reputable market research organisations
were simply lying through their teeth. With no commitment likely to be
invoked, nothing is easier to exaggerate than the frequency and length
of calls you ?would probably make? to your broker in London or
New York from your Jeep in the middle of the Gobi Desert or
wherever. Too much of this sort of thing undoubtedly skewed the
international business traveller (IBT) predictions of many surveys,
despite the experience of the researchers. After all, they were hired to
present a survey which would confirm the promoters? pre-existing
wish to push ahead with ?their? system. And so the capital, recurrent
and utilisation costs of GMPCS were assumed to be ?acceptable?.
Which obviously they were not.
Another distortion came from the undoubted success of terrestrial
cellular phone systems. Millions of ?ordinary people? use cellphones
for often inconsequential local chats -- over 130 million in western
Europe, according to the ARC Group. They do this because of the
increased convenience of mobile phones over searching out a public
payphone, despite a price premium which does not usually involve a
substantial charge for the cellphone handset itself: something not even
partially swallowed by Iridium until now. Consequently terrestrial
cellular use does not translate into even roughly proportional IBT use
of international satellite phones. This is despite the fact that for the
IBT the payphone alternative often does not exist. Too often, the calls
just never get made.
This highlights the unsuspected fact that the IBT market is price
sensitive after all. It does not consist solely of people with abundant
private means or unlimited expense accounts. This is the inevitable
conclusion drawn from Iridium?s chosen strategy to remedy the
situation: massive price cuts.
And massive is the word: the cost of Iridium handsets is being cut by
two-thirds, to under US$ 1,000, while cost of pagers is being halved,
to US$ 350. Retail airtime costs are going down by up to 65 per
cent, with a flat rate for international calls. These changes are being
implemented by arrangement with manufacturers and service
providers, which may imply subsidies, payable down the line.
Iridium is also redirecting its marketing strategy. Increasingly now
institutional users rather than the free-wheeling IBT will be the targets.
This means primarily government users (including the military), and
industrial users. Phones will be aimed at companies rather than
individuals. There have already been some successes: the US General
Services Administration is spending US$ 52 million on Iridium
services, with the Defense Department due to spend US$ 219 million.
Recently the State Department bought 1,000 Iridium phones worth
US$ 1.4 million. But the recent decision of AlliedSignal to withdraw
from developing an in-airline sevice must be a disappointment.
It is probably unfair to blame Iridium for getting so much wrong, but
then they were ?first to market?, something that at the time was seen
as a considerable advantage. But it also involves making the initial
mistakes, from which its followers will inevitably profit. Doubtless they
will also find a few mistakes of their own to make.
Little attention seems to be being given, however, to the generic faults
of all GMPCS systems, among which should loom large the negative
effect of lower user convenience than the familiar cellphone
(particularly in terms of worse in-building penetration). It would be
interesting to obtain data on potential users? reluctance to invest a
premium price for equipment with lower utility. Here again, were the
pre-launch surveys strictly honest? Most potential users claimed they
would be willing to ?cooperate? with the satellites, as by going to a
window, or even outdoors and walking a short distance. Is this
proving to be true?
Attention is also being drawn to the possibility that GMPCS systems,
with handsets oriented solely to handling voice traffic (via
circuit-switching), are already obsolescent. What people really want,
it seems, is packet-switched mobile data -- specifically Internet
access and also email, which overcomes time-zone differentials. The
third generation cellular system (UMTS/IMT-2000) is being heavily
slanted towards data traffic in response to a perceived demand. If the
satellite systems of the near future are to act as a complement to the
3G terrestrial systems, accessing those parts of the world the
cellphones cannot reach, then the satellites will have to provide
identical services. Whether they can do this by re-equipment of the
user ground segment-only is an interesting question.
The fourth contender
To conclude with the fourth GMPCS system to have advanced as far
as making a launch booking (not necessarily a guarantee of success,
or for that matter of even achieving the launch), there is Ellipso. This is
largely the brainchild of CEO, David Castiel, who has attracted
various backers and collaborators over the long time (at least a
decade in the public eye) that his system has been brewing.
Ellipso has an unusual architecture, which must by now be regarded
as fairly frozen. As it is less familiar than most, it is spending time
describing. It too has a two-stage implementation scenario. First,
seven satellites (according to current plans) will be placed in medium
8,050 km equatorial orbits called Concordia. The same name also
covers another set of five satellites in a slightly elliptical (8,050 x
6,149 km), slightly inclined orbit, for additional capacity. Together,
these will provide continuous coverage of the earth between 40
degrees north and south, which according to Ellipso contains 85 per
cent of the world?s population. The date mentioned for these launches
is 2002. Total system cost has been put at US$ 1.5 billion; it will be
interesting to see if this relatively modest price-tag is maintained. The
cost to users of the Ellipso system is claimed to be the lowest of all.
Launch capacity has been booked at end-June on four Ariane 5s,
each of which will carry up to four Ellipso satellites, for a total of 16
(the complete Concordia system requires 12 satellites). At an
undisclosed later date, 10 further satellites will be placed in two
elliptical sun-synchronous orbits, called Borealis, to cover high
latitudes, specifically those in the northern temperate latitudes. Boeing
has been widely predicted to provide Delta launches for Borealis; it
has also been nominated as satellite prime contractor; an investment is
clearly expected (US$ 300-500 million has been mentioned), but
Boeing is so far sitting on its money while waiting to see what Ellipso
manages to raise from other sources. Some say Boeing is aiming at a
total takeover of Ellipso.
Until recently Ellipso?s assets were believed to include Israel Aircraft
Industries, Venture First II LP and HarbourVest Partners. Marketing
and distribution partners are said to include Spectrum Network
Systems (Australia) and Vula Communications Holdings (South
Africa). But simultaneously with the launch order, Arianespace agreed
to put up vendor financing and an equity investment in Ellipso.
Understood to be worth around US$ 200 million, it was thought to
include up to US$ 50 million in cash.
Chris Bulloch is a freelance journalist specialising in satellite
communications. |
