Hermes Europe Railtel is an emerging carrier in the
pan-European market. The company provides cross-border
communication transmission services for other carriers,
including incumbent PTOs. It has one of the largest
trans-European networks. By the end of 1999 it expects to be in
18 countries and 44 cities with a network of more than
20,000km. The network is based on DWDM (Dense Wave Division
Multiplexing) technology. It is a high-capacity platform and
enables Hermes to deliver high-end services at high speeds.
Hermes Europe Railtel began operations in 1996. It currently
serves more than 50 customers. The company has been leveraging
this first to market advantage, as Trent Spiridellis, a
telecoms equity analyst at NationsBanc Montgomery Securities
explains: "At the beginning of 1998 Hermes along with WorldCom
and Ulysses were the only players that actually had broadband
seamless communications networks."
As new entrants target the European market, prices may drop
dramatically. Spiridellis notes: "I would say that pricing
pressure is the greatest challenge that all these companies are
going to face. If you look at the price of STM-1 capacity, it
is falling dramatically. I think that only those companies that
are deploying broadband networks, broadband-owned capacity will
be able to compete in this declining price environment. We
ultimately think that the European cross-border and
long-distance communications markets will resemble that of the
US. So with Hermes, the greatest challenge is not necessarily
on the cost side of the equation, but really in terms of what
prices it can charge for carrier customers, who are now going
to have numerous different alternatives. The problem is that
KPN/Qwest, Viatel, Global Crossing are also targeting carriers
The company benefits from operating under the GTS umbrella.
GTS, which holds a majority stake in Hermes, recently announced
its acquisition of Esprit Telecom and set up a joint venture
with FLAG Telecom to build a $1 billion trans-oceanic cable
system to carry voice, high-speed data and video traffic at
speeds up to 1.28 terabit/s. The recent bout of activity
underlines GTS' ambitions to be one of the largest independent
carriers in Europe.
In an exclusive interview with Global Telecoms Business, the
managing director of Hermes Europe Railtel Jan Loeber talks
about the changing nature of the pan-European carriers' carrier
market and explains how Hermes aims to keep one step ahead of
the new entrants.
Hermes Europe Railtel was one of the first
pan-European operators to market. How have you been leveraging
that competitive first-to-market advantage? How many new
customers do you expect to gain this year?
Loeber: The first-to-market advantage is a
substantial one here, because it is so difficult to create a
network in the European environment, where you cannot count on
any single kind of infrastructure to stretch from one country
to the next. You are forced to work with multiple
infrastructures that don't necessarily meet at borders and
don't necessarily run uniformly through cities to the points
where you need the nodes. Consequently it is a gargantuan task
to create a seamless network that is managed under a single
operating centre that allows you to provide high quality
service and feature transparency throughout the entire network.
First-to-market also means that we have been able to sign up
over 50 customers, premier names in the telecoms industry in
Europe, who have been using our network now for quite some
We actually initiated service in November 1996 and have been
rolling out the network ever since up until the 17 city
position that we are in today. Our customers have had "soak
time", in other words they have seen that when we say that we
will roll out a new city by a certain date, we meet this
deadline. They see that we deliver the level of availability or
quality on the network that we promised. So they are extremely
comfortable with our level of service. They like our pricing
today, compared to the alternatives available on the market.
This means that anyone coming to the market after us will find
it extremely difficult to attract the attention of customers
that we have already signed.
Our customers also know that our pricing is going to evolve
with the market and that we have a cost base in our network
that allows us to be competitive with anybody who comes along.
So our pricing is going to remain competitive. First-to-market
involves us first in a major engineering/integration activity
with each customer.
The first mover advantage has this subtle aspect that it
starts a close and hopefully long-lasting relationship between
the customers we signed and ourselves. As to new customers this
year, our growth will come from a combination of increased
business from current customers as well as from additional
customers - as many as 50 more.
Hermes Europe Railtel has one of the largest
trans-European high capacity fibre optic networks. How many
cities does it currently cover? What plans do you have for the
build-out of the network this year?
Loeber: Right now we are actively providing several
high-capacity SDH transport service offerings to 17 cities
throughout Europe. Most recently we added Copenhagen, Stockholm
and Berlin. Before that we added Milan. We are in the heart of
Europe, covering all the major cities. Next we plan to expand
into Spain, where we have just received a licence. We will be
adding Madrid in February, Barcelona and Valencia in March. Our
original business plan called for reaching 34 cities across
eastern and central Europe. But we recently discovered that our
cash generation, coupled with our ability to support debt and
build human resources, enable us to expand that roll-out. So we
now aim to reach 44 cities this year in about 19 countries,
including access to the US.
The country list includes three new countries beyond the
original plan: Portugal, Norway and Finland. By the time we
have added those countries we will cover 20,000km of fibre
route. We will expand to 25,000km in the year 2000 reaching
over 50 cities in about 22 countries.
Why have you targeted exclusively the wholesale
market? How does your arrangement with Carrier1 to supply SDH
transmission work? Do you have certain non-compete
Loeber: Our focus on the trans-border wholesale
market was a strategic choice. It has served us very well. By
selecting this course, we have avoided attacking our customers'
end-user base and his home market.
Therefore, even without any non-compete agreements, every
carrier that we work with has a sense of comfort that we are
facilitating their trans-European operations, instead of
competing with them.
Our network is specifically designed to meet carriers needs.
For example, carrier locations are very much inner city
locations, whereas end users tend to be distributed all over
industrial parks. Aware that our customers are carriers and
service providers, it is very clear where we need to deploy our
network to serve our carrier market. We can be very focused in
putting the resources into the best possible connectivity to
those locations. Secondly our pricing is aimed at somebody who
is using our network as an architectural element and sells
services to end users. So it involves special pricing that
allows our customers to generate a good margin, as they sell on
to end users.
Our arrangement with Carrier1 illustrates the fact that the
Hermes Europe Railtel (HER) network is a facilitator of
competition in Europe. Carrier1 could be in business from a
cold start over a five-month period actually offering
end-to-end services for the following reason. In addition to
their aggressive management of the elements they needed to
assemble, HER was able to provide inter-city links. We were
also ready to provide local access connectivity from our nodes
to their nodes. Without a network like HER in place, this would
absolutely be impossible for any new European competitor to do.
This is why we call ourselves a facilitator of competition. I
think it is safe to say that, just as INTEL is inside most
computer devices, almost everyone who is offering
trans-European inter-city services has the HER network inside.
Why did you decide to use DWDM technologies?
Loeber: The choice of DWDM is driven by the
favourable economics of these technologies compared to the use
of a lot of fibres. If you are running a long-distance network
and can choose between building and lighting a 96-fibre bundle
or acquiring two fibre pairs on 40 wavelength DWDM, the
economics always, and by an absolutely massive margin, favour
the creation of capacity on few fibre pairs.
When you perform the technical analysis, the cross-over comes
very quickly: it occurs whenever a network segment is longer
than about 100km and you are lighting two wavelengths or more,
which is 5 gigabits. We reach that point pretty quickly on all
our network segments. And the economics works ten times to one
better using DWDM, than lighting multiple fibres. Also the
quality is higher, because you are actually deploying far fewer
hardware devices, thereby reducing exposure to device failure.
For example, if you used 40 600km fibre pairs and lit each of
them, every 60km you would have regenerators/fibres and you
would have terminating equipment at the end of every fibre. So
you have all this equipment (880 boxes). On one fibre pair,
with DWDM, we support 40 channels (22 boxes) providing the same
capacity as 40 pairs.
Why did you choose Ciena as your supplier?
Loeber: There is always the right vendor at the right
time. When we made the vendor decision we had certain criteria.
We wanted, without fail, a delivery promise by June 1998. We
made the decision sometime in February. We put out a tender to
the top five producers of DWDM. Only one of them could deliver
us a working system for our labs, so that we could take it
through its paces. So that gave us confidence. At the time they
also had the highest capacity system supporting 40 wavelengths
on 2.5 gigabits fibre. Their European support team has provided
excellent service and support.
Similarly, what are the benefits of Alcatel as
supplier of SDH transmission equipment?
Loeber: We decided back in 1995 to work with Alcatel.
They competed with the other big guys, Siemens, AT&T (now
Lucent), Ericsson and others. At the end of the day they were
among the top three contenders, who were fully compliant on a
technical basis. As we were a new pan-European company, Alcatel
was attractive, since they had the most widely and uniformly
deployed resources throughout Europe. In other words, they had
substantial support companies in every country in which we
would operate. So we felt comfortable about their on-the ground
resources to support us. In addition, their commercial proposal
was very competitive. Their equipment has been performing
How much did the network cost to build? When will it
be completed? Do you have any plans to extend the network reach
outside Europe or will the company remain Europe-focused for
the foreseeable future?
Loeber: We are witnessing an expansion of our
European network to new countries and cities. We don't
necessarily see the construction of terrestrial capabilities on
other continents. We have decided to integrate trans-oceanic
capacity into our service offerings. You are probably aware
that the HER parent company, GTS, has completed a 50/50 joint
venture with FLAG to form FLAG Atlantic-1. This will be the
world's first terabit cable across the Atlantic. And this will
have about 20-30 times the capacity that all transatlantic
cables today have cumulatively.
So this represents a huge advance and implies an extremely
aggressive and low cost base. GTS is doing this to enable HER
to be in a better position to serve US-based customers that
have an interest in moving traffic in and around Europe. We can
pick them up in their back yard, namely at twin New York FA-1
nodes. We can then deliver the traffic to any of our European
cities through the Hermes network.
It also provides us with an opportunity to contemplate
interesting alliances or commercial arrangements with major US
carriers to provide city-to-city services, looking at the range
of US cities across to the range of European cities.
For such trans-oceanic city-to-city services we would focus on
building an IP highway, an Internet protocol-based environment.
This is a highly efficient protocol in terms of optimizing the
utilization of this trans-oceanic capacity to support the
fastest growing segment of the telecoms market, the Internet
Building the network through zero will have required almost $1
billion in investment. Maintaining, growing and upgrading will
require $200 million annually. We will continually keep pace
with emerging customer requirements and emerging market
What is your projected EBITDA target for 1999?
Loeber: As we are associated with a public company,
GTS, listed on the Nasdaq and Easdaq exchanges as GTSG, we
cannot offer projections on such issues. Let me just say that
from a performance level we have been EBITDA positive since May
1998, which is extraordinarily fast for an infrastructure-based
company such as ourselves. We have been meeting or exceeding
our financial goals steadily. We are extremely heartened and
are in fact very surprised by the robustness of demand.
We have installed by now well over three times the number of
circuits that we had initially projected in our business plan.
We have accelerated our DWDM deployment by almost two years to
be able to deliver in line with requested demand from our
customers. So our business is enjoying considerable growth.
Owing to this growth and the very strong revenue uptake and
cash inflow, we have been able to take the decision to
substantially expand our roll-out. From a financial standpoint
we have enjoyed an excellent environment of strong customer
demand, which we expect to continue.
Do you expect to return to the markets for additional
Loeber: Not at the HER level. As you may be aware, we
recently closed a second high-yield bond issue for almost $200
million, plus almost 100 million euro ($112 million) to
accommodate the planned acceleration that we are able to
sustain. So we are now in an excellent position cash-wise to
execute the increased roll-out objectives that we have, plus
some. DLJ has been very effective at leading both this and the
previous $265 million financing.
If we can now turn to your acquisition strategy. How
has the acquisition of Ebone improved your strategic position?
Do you believe that HER will make further acquisitions in
Loeber: HER is part of a larger GTS Group, called
Carrier Services. Carrier Services will make further
acquisitions. HER may also do so. Ebone was a very ingenious
HER deal. It was a win-win situation for both seller and buyer.
It actually was a deal where the controlling interest that we
achieved, namely 75% of Ebone, was purchased in exchange for
HER service. HER service replaces systematically the entire
backbone network of Ebone and, in fact, reduces the cost base
of Ebone by more than 50%. This means that Ebone will be able
to translate those benefits, in terms of pricing and the
quality of the HER network, to the benefit of their ISP
customers. Ebone has about 85 major top tier ISP customers.
As you probably know, Ebone is a transit provider. They are
therefore a carriers' carrier as we are ourselves and they
serve ISPs. Their strategic value to HER is that we instantly
acquire a knowledge and experience base in what we call the
"Internet space". And that is very valuable, because the
Internet is a focus for us. Much of the future acquisition
activity that you mentioned will occur in the Internet space.
There is a third element of the Carrier Services Group in
addition to trans-oceanic systems and HER. The third is what we
call the IP Services companies: that is, Internet Provider
Services. We are pursuing three objectives which we believe to
be avenues in the Internet space which will generate and draw
traffic into our network structures. One of them relates to the
development of advanced Internet services.
Another is web hosting and the third is voice-over-IP
capabilities. We have just announced at HER the controlled
introduction of a new IP transport highway, which operates
directly on optical interfaces. These three new IP Services
areas will function aver and alongside this new highway.
As a pan-European operator, how do you view the
regulatory environment in Europe? Do you think that the
European Commission has done an effective job in liberalizing
the telecoms markets in Europe?
Loeber: The development of a regulatory environment
supporting liberalization in Europe has been progressing more
or less as anticipated and mandated by the EU. As we developed
our network across Europe, the licensing regimes have been
materializing more or less on time for our deployment. I
believe that the EU has been instrumental in causing that to be
possible. The EU enabling function has been very helpful to new
competitors across Europe.
The huge remaining problem area for every new competitor in
the European space is at the local access level. I am sure that
you have heard this a thousand times from everybody else that
you have talked to. Our local access issue is private circuit
connectivity between our node and where our customers are
located. The pricing of those connections and the lead time
over which circuits can be made available are totally
unsatisfactory in Europe in most cases. In any case we see an
extremely mixed bag of the ability or willingness of the
incumbents to be responsive on timely delivery and the
willingness to be reasonable on price. Increased competition is
the only answer.
As you probably know, we have an arrangement with COLT to
provide local access in the cities that we both serve. That has
been working very well. Wherever COLT and other competitive
access providers are, we see competition causing the incumbent
to react and be more responsive on availability and price.
Which service offerings will enable HER to distinguish
itself from the competition?
Loeber: Let me just start this discussion with my
view of how the market is structured. If you look at the range
of possible carriers' carrier transport capabilities, at the
low end of the value chain is what I call the real estate
business. In other words, people who build fibre and sell dark
fibre or sell cheap point-to-point IRUs (indefeasible rights of
use). We are not in that business. Dark fibre is the raw
material that we build or lease and integrate into our network.
If you look at the other end of the value spectrum, you see
the wholesale minutes business, which is also a wholesale
business, but on a message level. We are not in that business
either. We are in the mid-range business of the wholesale
transport services continuum.
We differentiate ourselves in this industry by packaging
value-added services in the transport environment. For example,
among the SDH services that we offer, one of them is clearly
structured and priced for the requirements of a new carrier
which needs very high-quality services, with full redundancy.
It is called "Point-to-Point" service. This is point-to-point
in the context of our total network environment and provides
100% redundancy, so that the same traffic travels on two
separate routes. If one route is damaged, the customer will
never notice it.
It is not a trivial task to be able to provide that. It takes
a lot of engineering and human resources to support that
service. The second service, named the "Ring" service, is aimed
at existing incumbents or operators which have their own
networks and network architecture and engineers. This service
covers three or four cities on a ring or sets of inter-locking
rings that a PTT, for example, can use as a network element and
integrate into their own network, providing for their own
redundancies. It is very aggressively priced and is very
Thirdly we provide a service for the Internet Service Provider
that involves the capabilities of Ebone and its peering
relationships with other European Internet carriers and
providers, as well as US Internet providers. We assemble the
use of that peering relationship, Ebone router capability, our
core network and trans-atlantic capacity and we are able to
offer an end-to-end IP service over SDH that spans from any
Ebone or HER city in Europe through to a New York Internet
provider outside Ebone's Internet exchange pop.
These three structures of service are focused on specific
customer needs. In terms of the economics, the provision of
value-added transport solutions to the fibre and the equipment
that lights and manages it, represents only about 10-15% of the
company's total resources. The rest is invested in the
engineering resources, back office resources, customer services
and the marketing and service development resources required to
deliver these services. We do not consider companies selling
IRUs as being in our competitive space. Selling an IRU is
selling an asset. We are leasing advanced services with all the
long-term support and development commitment that accompanies a
long-term commitment to a customer relationship.
With more competitors coming to market, how is HER
going to react to pricing pressure from its rivals? Do you
think that prices for cross-border services will significantly
decline over the next two years?
Loeber: Absolutely. They will continue to decline.
Users will want larger and larger tranches of bandwidth to run
their services on, as they are becoming more and more inventive
and less bandwidth sensitive about the services that they
provide. Prices will go down with the leverage of technology
and increased competition. We are more than prepared to respond
to that. We have probably the lowest cost base in the industry.
Anybody new that comes into this game at this late date faces
a huge challenge to be cost competitive for the following
reasons. It is not just the efficiency of the technology that
causes costs to be low. Everyone can have new technology. It is
how well you fill your network pipe. If you take a fibre pair,
for example, and you light it up with DWDM and make a two and a
half gigabit pipe into a 100 gigabit pipe, with 40 wavelength
DWDM, you have not gained any economic efficiency if you fail
to fill the pipe.
So HER has loaded 40-50 mega-carriers operating across Europe
onto its network, filling these pipes and only lighting them
fast enough, four wavelengths at a time, to keep the size of
the pipe bigger than demand, but not that much bigger. We are
actually achieving phenomenal fill factors with these huge
traffic streams. We are achieving extremely low equivalent E-1
costs. We expect to be able to keep up with everything and
anything that happens in terms of price movement in the
European market in our value-added services space.
Viatel has won considerable plaudits in the industry
owing to its willingness to post prices on its web site. Will
HER follow suit? If not, why not?
Loeber: It is possible to post prices on a web site,
if you are selling, for example, a point-to-point IRU. Then you
can say what you get and how much it costs. In our value-added
service space we are in a different business. We do not and
will not sell dark fibre or virtually unmanaged service. For
example, when we do a 40 million Euro deal with a carrier, we
are discussing whether it should be a one-year contract or a
ten-year pay in advance, lease, or a five-year contract. Should
the payment be quarterly, monthly, annually or in advance?
Exactly which cities and which kinds of provisions for growth
in the number of cities or in the capacity over time will they
want? Which service package or combination of packages do they
want? Do they want local access integrated in the near-end, the
far-end or both (75% do want it)? This is not something that
you can post on the Internet. Ours is a highly tailored
response that is customized to every customer's different
Do you plan to compete on price with companies such as
Viatel or quality of services? For example will you undercut
Viatel for the price of STM-1 capacity?
Loeber: We would not expect to lose a deal to a
competitor on price where we have an equivalent service
offering. An STM-1 can be provisioned in many ways (for example
with or without integrated local access). Some of our
competitors try to get mileage from citing our price for a
one-year contract on a fully redundant E-1 circuit with bundled
local access both ends and compare it with their single
node-to-node equivalent E-1 price provisioned as an STM-1 on an
annualized basis, assuming 20 years IRU pricing! We are selling
the customer a proposition that the customer can see has value
that he is willing to pay for and that is available now, not
six months from now.
There will clearly be instances where somebody wants to buy
simple point-to-point solutions. I would say that this leaves
room for commodity suppliers to have a business. I would add
that it is just not where we intend to do business, it is not
how we want to position ourselves. We are in the solutions
business and the value-added services business. We will
constantly be climbing the value-added chain. We will provide
typically multi-city solutions, rather than linked
point-to-point solutions. In other words, we provide virtual
network spaces that have various service packages.
How do you benefit from being under the GTS umbrella?
How does its acquisition of Esprit Telecom affect your
operations? Will the new acquisition lead to conflicts of
interests in terms of clients?
Loeber: I don't think so for the following reason:
the avoidance of a conflict of interest is under our control. I
have agreed with the GTS leadership that we will regard each
other across the various lines of business, as operating at
arms' length. In other words, whether GTS' new local access
business, called Access Services, or their Business Services
business, where Esprit is resident along with Netsource, we at
Carrier Services will relate to these companies just as we do
to any of our customers, such as COLT or Demon. The pricing
that we will offer them and they will offer us will reflect how
we would behave towards any other customer in the market.
Across those lines of business we will not have pricing that
differs in any way for the tenor of the contract, the volume of
the commitment that is made and the service that is offered
from the same contract that we offer to our other customers.
In your view how many carriers' carriers can survive
in the European market? How do you see the European telecoms
landscape changing over the next two-three years?
Loeber: First of all, the landscape is going to be
like the surface of the earth in the first million years of its
formation: a boiling pot of bubbling lava. Some will harden
into mountains, while others will just burst and some will
swallow up others. When you have 16-20 countries, which are
almost simultaneously liberalizing telecoms - emerging from an
environment where there was one incumbent to where there might
be, as in the UK, hundreds of licensed carriers - it is indeed
a boiling pot of competitive activity. We will see new
companies forming every day. The whole Internet environment is
a key driver of that activity, as well as traffic growth.
Voice also continues to be very important in this evolution.
While new enterprises continually pop up, we will also see
consolidation. Some companies will get into trouble because
they won't be properly financed. Or there may be a hiccup in
the economy from time to time and they will become ripe for
acquisition. So we will see a tremendously changing environment
due to a variety of factors.
How many will survive? There is an old adage in business that
states: all markets go to three. I think that we have seen it
in many environments. In the US we have had a fairly
competitive long-distance market for a number of years. We saw
AT&T, MCI and Sprint on top and a whole bunch of others who
were less prominent. We saw the same thing in the computer and
automotive industries in the US. I think that in an equilibrium
situation, which may be a number of years away in Europe, maybe
beyond our lifetimes, we may see three or maybe four
significant global players.