The SAT3 fibre - a monopoly that stands in the way of cheaper international bandwidth
Africa has some of the highest costing international bandwidth on the planet. Building fibre connections should lower bandwidth costs. But despite the building of the SAT3/WASCSAFE fibre, the price of E1s on this route to the rest of the world remain at levels that are often above their satellite equivalents. As Africa's national monopolies slowly begin to disappear, attention will now begin to focus on the new strategic heights of international competition. For there will be little point in achieving competition at a national level if there remains a large obstacle between Africa's competitive markets and the rest of the world. Russell Southwood looks at why SAT3 has not yet produced cheaper bandwidth prices and whether it is yet in profit.
The building of SAT3/WASC was a considerable achievement. Its promotors - notably Telkom South Africa - seized the last moments before the end of the telecom investment boom to raise finance for it. It combined those wanting to transport African traffic with those needing traffic to transit from other destinations.
It is in effect two cables: SAT/WASC which travels down the west coast of Africa to Cape Town and SAFE which goes from Cape Town to Asia. The latter connects Mauritius and Reunion (a French "department") to the rest of the world and accounts for the former's ambitious "cyber-island" strategy.
The consortium that built these two connected stretches of cable has 36 shareholders. The SAT3/WASC section has 11 African shareholders: Angola Telecom, Camtel, Cote d'Ivoire Telecom, Ghana Telecom, Maroc Telecom, Nitel, OPT Benin, OPT Gabon, Sonatel, Telecom Namibia and Telkom South Africa. For reasons that will become clearer, two other things need to be noted about the shareholders. Two do not have landing points (Maroc Telecom and Telecom Namibia) and among the full list of shareholders is the Indian carrier VSNL that is owned by Tata that has bought into the soon to-be-set-up South African SNO.
Until the building of private fibre capacity across the North Atlantic, all fibre capacity was either built directly by a single operator or through shareholder consortia. This was the way things were done so there should be no sense of surprise or disappointment that it was done this way. But that was then and this is now.
What does the "club consortium" mean in practice? It is very difficult to say precisely as its basis is a shareholder agreement that is "commercially confidential". However the broad outlines of how it works are fairly clear. The consortia's shareholders appoint a managing agent who runs it on their behalf, taking care of day-to-day performance and maintenance issues. In the case of SAT3/WASC this is Telkom South Africa which also has the largest amount of traffic among consortium members. In terms of shareholdings, these can only be sold with the permission of all of the other 35 shareholders in the consortium.
Consortium members have a monopoly on selling fibre in their own country. There are rumours that this monopoly is time-limited but no proof of this has ever emerged from behind Sat3's closed doors. If the consortium builds more capacity than its members can take up (on which they have first right of refusal), then the "pool" capacity will be sold off as what are known in the jargon as IRUs (Indefeasible Rights of Use). There has been a considerable debate about their status in accounting terms but they might best be described as contractual permissions to use a certain level of capacity.
Being the only fibre cable available puts the consortium's owners in a fairly unassailable position until recently. However a number of disquieting pieces of news have begun to emerge which taken together illustrate the very negative effects that this monopoly is exerting on the African market.
There's a landlocked African incumbent that needs access to SAT3 capacity. It wants to provide cheaper international bandwidth to its customers. Let's call it country A's incumbent because it cannot speak publicly about what has happened. Country A's incumbent approached one of the consortium's international members who expressed a willingness to sell it cheaper bandwidth. The existing African consortium member selling them bandwidth objected but lost the argument based on the consortium's contractual agreement. But Country A's incumbent found that its problems had only just begun. The existing African consortium member who had lost the business was now charging it as much money to get to gets its traffic to the landing station as it was being charged to get from that landing station to Portugal. In other words, it is being held to ransom by the landing station owner.
Country B's incumbent invested in the consortium but did not put up enough money for a landing station. So it has to ship its traffic through a neighbouring country to get to a landing station. You may be surprised to know that the cost of transiting the traffic and the international leg make it cheaper for Country B's incumbent to send their traffic directly by satellite.
Under pressure from Ghana's ISP association, GISPA, the local incumbent with the monopoly on the landing station, Ghana Telecom lowered its prices by about a third to USD8050 a month for a leased circuit from Ghana to Portugal. Previously E1s had been quoted at between USD12-15,000 a month. The agreement is again covered by commercial confidentiality but is understood to say that the backhaul prices will be set by Ghana Telecom.
An STM1 was costing USD50,000 a month in Nigeria, Ghana and Benin but costs UAS225,000 in Angola and Cameroon and USD300,000 in South Africa, according to Michael Blair of GS Telecom at the ACT 2004 conference. But dealing with monopoly incumbent providers on SAT3 fibre is a problem:”Nitel restricts the use of SAT3. It won’t give us (more capacity) unless we use their facilities and who would want to do that?”
Nigeria's SNO Globacom realising that it would need access to its own international fibre capacity to be competitive tried to buy a shareholding in the consortium. It was told that it could not do so, presumably because Nitel would be threatened by its access to capacity. In the meantime, Globacom has announced that Alcatel will build it a fibre between Lagos and London. Some parts of the Nigerian Government have talked about hiving off Nitel's SAT3 capacity as a separate operation but this has not become publicly accepted policy.
There are three recurring issues: the impact of the monopoly on landing stations; the monopoly on the sale of capacity; and the fact that shares in the consortium are not tradeable. In Europe, European Competition policy talks about fair access for all users. Unfortunately in Africa there is no continental competition institution and at present there is only competition legislation in South Africa. In other words except for South Africa it would be difficult to mount a competition challenge because the legislation does not currently exist.
So how do large customers in South Africa, SAT3's major market feel about the situation. According to MTN CEO Mike Brierley they feel as if they're caught 'between a rock and a hard place':"You have to use SAT3 and we’re all victims of this issue. As MTN, we are working as part of the EASSy consortium (through our Ugandan subsidiary) so that we have landing rights on an alternative route".
Like a number of people in South Africa we have spoken to, he feels that there is a good case for an anti-competition referral:"There is good cause for an anti-competition case. The licensing of the SNO will heighten this. It will need access to SAT3 at a reasonable price". So it would make sense to mount a challenge through the Competition Commission to start to address this issue.
"Satellite bandwidth prices are substantially lower than SAT3 prices. Telkom started reducing prices this year and this will probably be an on-going process". However he pointed out that that those currently making large profits from what can really only be described as over-priced international bandwidth:" Large ISPs in SA get 60% of their revenues are from selling high cost international bandwidth. What will happen when the prices fall?".
Mike van den Bergh, Gateway Communications sees a pretty similar picture on prices:"SAT3 prices are coming down but there’s a lot of room (for that to continue). We’re seeing price erosion but it’s got a long way to go". He wonders whether Tata's involvement in SAT3 and the SNO will provide a way of tackling the issue. VSNL is a consortium member and Tata has bought Tyco's international fibre assets so is seen as having greater negotiating power with Telkom South Africa. But as van den Bergh told us:"But it will probably take pressure from the Government and the regulator to resolve the issue".
We asked someone who has been involved in international fibre cabling projects to look at SAT3's likely profitability over its first five years based on published figures. His calculations are as follows:
Sat3 started in April 2001.
Annual running costs
Maintenance USD10m Linked to the World, Financial Mail, 7 June 2002
Investment in SAT3 USD300m.
USD60m over five years
Interest USD15m @ 5%. Interest declines over the period so probably slightly less.
Total running costs USD115
X 5 years = USD575m divided by 60 months = USD9.58m income needed
E1 = USD6000 This is is probably lower than the prices that have been achieved so far.
1mb = USD3000
USD9.58m divided by USD3000 = 3 gigs
On this basis, the whole operation would be amortised in five years (by April of next year) and then running costs would drop to USD30m a year. SAT3's first upgrade cost USD33m. The proposed upgrade might therefore be slightly higher but the decision to upgrade indicates that there is confidence that the breakeven point is not far away.
(If any of these figures are wrong, we would be happy to publish the corrected version of these calculations).
According to the expert we spoke to: ”On these rough calculations, the total cost of running SAT3 is slightly over USD100m a year. On this basis, breakeven would be reached in five years or under. From there on, the cost base would be maintenance and administration, roughly USD30m a year if no upgrade is done.”
“With the new upgrade costing somewhere between USD30-50m, you could get a doubling of capacity to 40 gigabits per second. The previous upgrade (according to the Alcatel announcement) cost USD33m. So again breakeven would be achieved in a relatively short space of time. On this slightly expanded cost base, rates could come down to something closer to those found on the North Atlantic.”