A Zenit-2 booster carrying 12 satellites for the Globalstar global cellular phone system crashed on the Siberian mountains shortly after its launch from Kazakhstan in September 1998. This event reinforced the fears of project financiers that new telecom technologies projects pose a series of complex risks which are often difficult to understand and even harder to incorporate into financing documents. Since the Globalstar crash, Iridium's highly publicized satellite project financing has hit problems. The company is having difficulties servicing its debt when forecast phone sales figures failed to live up to expectations. Iridium is now remarketing its products and its debt structure is being analysed.

With two high profile deals already facing difficulties and other projects in the pipeline, the question is who really understands the risk involved in financing such telecom projects?

More importantly some wonder whether satellite projects really represent the future of telecom technology. Fibre-optic cable networks and private undersea cables are fast outpacing satellite technology and it is a sector in which banks are establishing their own niche.

But the message from the Iridium and Globalstar financings is that sponsors and bankers have to be quicker than their competitors and better at structuring the financing. New technology is coming on to the market all the time.

An example of new telecom technology outstripping older designs is the Flag Atlantic-1 Cable System project. It involves the construction of a transatlantic cable system, able to carry voice, high speed data and video traffic at speeds of up to 1.28 terabits a second. The new network will be able to accommodate multimedia, business imaging and videoconference applications. It is financed by a combination of sponsor equity, customer sales and non-recourse bank debt, arranged by Barclays Capital. The existing Flag cable system entered commercial services in November 1997 and stretches over 27,000km from the UK to Japan. According to Scott Gottdiener, associate at law firm Simpson, Thacher & Bartlett in New York, this deal ?set the template for private cable deals?.

In the first transactions of this kind, banks have taken very little risk on their shoulders. Sales contracts were arranged and half of the debt was provided by pre-sales contract. More recently, the amount of risk taken on by the banks has increased, as confidence in these types of deals has increased. Among the most recent deals to be financed are the Pacific and Mid-Atlantic crossing deals which are entirely based on market risks. An exception is the Southern Cross project, which still has half of the debt guaranteed by pre-sales contracts. In the first tranche of the transaction, $640 million worth of pre-sales contract supported the $480 million tranche. Tranche B is the only part to take full market risk.

Technology is improving on a day-to-day basis for fibre-optic cable networks and banks are eager to cash in on what is perceived as a growing market. Says Louis Kenna, principal at Bank of America's telecoms project finance group in New York ?We're following the development of these projects very closely and we are hoping to get more involved in them.?

Truly global ? but terribly local
Satellite projects are ultimately telecoms companies delivering a telecom service to a niche market that has not previously been covered. As Bank of America's Kenna points out, the fact that they use satellites is incidental to what they are really doing. ?They are the first truly international telecom companies,? he says.

At the same time, their success depends on local factors and retail issues. For satellite mobile phones, distribution is a key factor.

This is where Iridium has encountered some problems. Handsets need to be distributed around the world, in places as diverse as France, Cameroon and New Zealand. The project company has to rely heavily on the strength of local providers and distributors, such as local telecom companies. One of the possible strategies is to include local telecom companies as sponsors. In addition, issues such as billing are very complex and need to be taken into consideration.

Fibre-optic cable projects also face similar considerations. As David Gordon, partner at Latham & Watkins in New York points out, fibre-optic cable network projects are multi-jurisdictional and when a private cable goes under the ocean, one cannot apply, for example, a specific law of security interest.

In addition, a cable crosses several different countries and each section needs a set of permits from each country of pertinence. But the problem is that it is not always possible to make sure that all the permits are in place before the financing is put together, due to the speed required in structuring the deal.

Just like merchant power
When structuring a project finance deal for a fibre-optic cable, banks usually want to make sure that the likelihood of getting the permits in different countries is high. But it can be unpractical to wait for all the permits to be in place. Gordon compares these kind of deals to the merchant power plants which do not have fixed long-term power-purchase agreements and where deals carry a far greater market risk.

Says Gordon: ?The key issue is the timing, and not running out of money. These structures are less contractually tight than in the power sector. It's a risk ? it's a due diligence exercise, but without the certainties of power projects. That also explains some of the higher pricing of these deals. And it is important to have credible sponsors willing to prevent delays and to fix cost-overruns. And if the sponsors are strong, the appetite for these kind of deals is fairly high.?

With new risks emerging all the time, lenders have to appreciate how the market is evolving. ?It is evolving as fast as you can imagine,? says Gordon. ?And as the demand for new capacity will be fulfilled, you can expect a downward pressure on pricing of these deals.

The first one wins ? speed is the key,? he adds.

Wiring the world
So far the Atlantic region has been the most popular for fibre-optic cable project developers, due to the extremely high demand for data transmission. But there are opportunities in emerging markets as well.

Teleglobe Communications is among the corporates expanding their geographic reach and capacity. The company has established a 28,800km cable system between Europe, Africa and Asia. Gian Franco Bucelli, vice-president and general manager at Teleglobe Communications expects the investment will help Teleglobe meet the ever-increasing demand for internet capacity. Once completed, the cable will provide the first undersea fibre-optic service route between Portugal and Malaysia with landing points in Senegal, Côte d'Ivoire, Ghana, Benin, Nigeria, Cameroon, Gabon, Angola and South Africa, with further connections to southeast Asia.

The Commonwealth Development Corporation (CDC) is also among those involved in financing innovative mobile and satellite phone projects in emerging markets. It has financed such projects in Bangladesh ? now the first country in the world where cellular phone usage outstrips fixed-line telephony ? India, Africa and Latin America.

But perhaps the most intriguing project to sign in 1999 is Thuraya, a satellite project covering the Middle East, Africa and South Asia (see Deals and Developments). It has given a positive signal to the market for satellite finance deals. According to Peter Pontidas, assistant director in ANZ's global media and telecoms group in London, it is a deal that scores many firsts for the industry.

For example, it is the first permanent loan of its kind in the mobile satellite industry. It involves repayment directly from revenues generated from the project rather than through subsequent refinancing.

Says Thuraya's chairman Mohamman Hassan Omran: ?It is a big financing deal for a pioneering telecom initiative of tremendous scope. The deal signals the trust and confidence that has been placed on Thuraya by the international financing community at a critical time for the mobile satellite industry.?

Nick Avery, partner at law firm Ashurst Morris Crisp in London adds: ?The fact that the banks were able to underwrite the deal at this stage shows confidence in the technology for satellites, given the fact that other satellite projects have had problems and there has been bad press for satellite financing.

This might help deals with similar risk profiles coming back to the market.?

Iridium, an ambitious $5 billion project for the launch of 66 satellites, has suffered from project delays and poor subscriber numbers and is now attempting to reschedule its debts. The Motorola-backed company announced in June 1999 that it had received a waiver extending to August 11 1999 of the financial covenants relating to customers and revenues under an $800 million senior secured credit facility.

The company has also launched a new marketing strategy aimed at increasing subscriber numbers, but many observers doubt that the company will be able to face its new competitors. Iridium launched its services at a time when mobile phone networks were limited in their geographic coverage, but recent improvements in mobile technology have stolen some of the strengths of its offerings.

Among its competitors is the Globalstar project, for which Bank of America has been mandated to arrange financing. ICO Global Communications, is also launching satellite phone communications services in 2000 and hopes to take up a quarter of the market by 2005. ICO will provide mobile telecommunications by satellite, including digital voice, data, fax, and messaging. The London-based project company will rely on 12 medium-orbit earth satellites. According to company officials, there is growing demand for satellite phones to cover areas where setting up fixed-line infrastructure is not cost-effective. They expect the key markets for the company to be China, Japan, South Korea and Australia.

However, at the end of June 1999 Moody's downgraded ICO's senior unsecured debt ratings to Caa2 from B3, claiming that the downgrade was due to the company's difficulties in raising funds to implement its business plans.

The company has secured only $3 billion of the $4.7 billion needed so far to deploy its satellite network services.

According to Kenna, ICO's biggest advantage over other projects is that it is a very simple system, with only 10 satellite in the sky, 10 ground stations and computers located in ground stations. The call goes to a ground station, which bounces the signal to the sky, then to the ground station where it is then routed from one country to another. This is a very different system from Iridium's, where calls have to go from one satellite to the other and if there is a fault in a computer which is located inside the satellites, repairs can be fairly complex. ?The signal has to bounce around 60 satellites and you can imagine how complex that is,? says another source.

Satellite financing is a relatively new sector that banks are just beginning to understand, says Kenna. There are a lot of new risks to understand which need to be incorporated into the debt structure.

Says Kenna: ?We are watching the satellite sector with great interest. But the big phrase in the market is that the jury is still out. And it is not yet clear who will win.

Everybody's watching and everybody is waiting, nobody on the bank side is absolutely convinced, but we are supportive of the sector and we are hoping that there is a big market for these projects.?

Satellite projects ? a different beast
One thing that makes satellite project very different from more traditional types of project financing is that there is a different balance of power between satellite providers and the purchaser, the project company, compared to the one between a developer and a contractor in a project finance deal. Normally in non-recourse finance deals the project company can insist on having their scheme developed on a turnkey basis, for example. In turn, the contractor has to comply to a lot more conditions set up by the project company. In satellite projects the conditions are are different. Satellite providers are in a stronger position compared to the project company. Says a banker: ?There is a situation of uncertainty, and a lot more construction risk, because you can't dictate the terms in the same way you can do with contractors. If satellite providers are six months late, there is not much you can do about that. They just reply ?look, we have the world's best scientists working on it but we need more time'. You can't just tell them to get it finished and shoot it up in the air, like you would do with a turbine. You need to budget that into the financing structure.?

Undersea cables are also very different from other more traditional types of project financing. David Gordon, partner at Latham & Watkins in New York says that while in the power sector the security interest lies mainly in the power contract, private cable financings are very often done without any hard-asset security and ?at the end of the day the structure presents more risks?, he says.

Lenders are also concerned whether there is a turnkey agreement in the project. The Flag project has been developed on a club basis, where local telecom operators were responsible for building the infrastructure. But, according to Gottdiener there is an increasing trend towards the project companies owning the cable, especially when it comes to building them undersea. ?It was a logical step,? he says.

What's the hurry? Is project finance the best way forward?

No matter how many risks there are, a number of financial institutions are looking closely at cable and satellite deals says Latham & Watkins' Gordon. And since the groundbreaking hybrid project finance Flag deal, the ?market has become a lot more educated to the system,? he adds. Gottdiener adds that with Atlantic Crossing, for example, banks took 100% market risk. Says Gottdiener: ?That's a big change.?

But one of the problems is that there is a potential conflict between the borrower's goal and the banks' need for security. Lenders might have structures in mind which do not satisfy the developers' long-term growth strategy. Or deals have been project financed on a regional basis, whereas sponsors would like more flexibility. In the end, what private cable sponsors really want is to have a ring of cable-network to go all over the world. But raising financing for this kind of project can be difficult.

So if the non-recourse finance model ? with its constraints on the covenant package ? falls short of fulfilling the developers' goal of building out global networks it could not be long before the magic moment for successful private cable projects will be over. A lawyer who has advised sponsor on private cable projects, points out that ?project finance is very helpful in the development stage of a project. However, with time, the developers don't seem to distinguish between the systems that have been built in different geographical areas and look at them as a global inventory.?

But if the long-term goals of a developer are to have a worldwide network, ?it is difficult to imagine that they can keep developing multiple systems on a project finance basis,? he adds.

This results is potential friction between the project financiers trying to obtain the highest possible value and profit out of one system, with the developer trying to find the most value out of the global system. One source suggest that this might mean a move away from traditional project financing and towards the corporate model for future financing needs, with more flexibility in the covenant package.

As Gottdiener says on his experience with Global Crossing: ?It has been a great ride.? But how much time is left for new projects? The frenetic competition for fibre-optic cable development in certain parts of the world means that developers need to make sure nobody is laying cables in the same area as they are.

And there is an increasing competition on the satellite finance side as well, with an ever-growing number of private satellites crossing our skies. So those who wait might find that they have missed valuable opportunities. Speed is of the essence.

---

Which technology will win the race?
Interest in high-tech telecom deals is growing. But according to Louis Kenna, principal at Bank of America's telecoms project finance group in New York the ?the jury is still out? on which technology is going to win. Once you have the cable into the country, the question is what is the best way to get the data from providers to users? Kenna calls this the ?last-mile competition?.

This is where other, newer technologies such as wireless local loops come into the picture.

A project known as Ionica involved plans for the build-out of wireless local loop technology developed by Nortel to deploy a network quickly and cheaply that would compete with British Telecom. The project eventually failed because it relied on a heavily conditional credit facility that was never drawn down before the company had to give up its plans. According to Kenna, though, developers are still looking into these kind of projects, because they could avoid telecom companies having to dig up roads in front of homes in order to put fibre-optic networks. He thinks that there is a chance that other projects like that one might be developed in the near future and there is a strong chance they might be project financed.

Motorola, the wireless telecoms and semiconductor company which is also the main shareholder in Iridium, has decided to invest its resources in an ambitious programme called Teledesic. The project involves a satellite-based ?internet-in-the-sky project? to be developed in the next 10 years. It relies more on data transmission than voice transmission and is one of several proposed satellite-based net access services that would also offer additional services such as videoconferencing and other broadband services. The project is supported by mobile phone head Craig Mcgraw and Microsoft's chief executive Bill Gates.

Among other technologies project financiers are keeping an eye on is the Universal Mobile Telecom Systems, considered the technology of the future for mobile phone providers and expected to take over from GSM systems around 2005. The most innovative aspect of this technology is that it enables higher data rates ? a maximum of 2Mbps indoor, 150kbps for moving terminals ? multimedia applications and the convergence of fixed and mobile communications.

In a couple of years, telephones could have full functions for diaries, telephone and full access to internet, enabling users not only to make phone calls all over the world, but also to book trains or have a videoconference meeting while travelling. The idea behind the word universal is that one system will serve different areas.

But one of the big issues that could hinder the development of this system, is the licensing system. The UK government , for example, is starting to talk of selling the licence for these systems, but prices are considered too high. To buy a licence for this system would cost at the moment around £500 million ($800 million). On top of that, a provider would have to add about £1.5 billion to run the system. ?People need to start thinking where this kind of money will come from,? says Kenna.