Pat - An excellent article Fibre pipes and prisoners
Sorry if this format is screwed up, no time, have to run to the cape.
JW@KSC
Fibre pipes and prisoners
July 96
Sorry if this format is screwed up, no time, have to run to the cape.
The question of replacing copper with fibre in the local loop is not one of straightforward technical substitution for Europe's telephone
companies. It is likely to be one of the most important strategic
decisions they will have to make. A bit of game theory may help.
Children play 'games'. Most adults do not play 'games', on the grounds that
they are not serious enough. The exception to this rule are economists,
who have invented game theory to enable them to play 'games' and solve
problems at the same time.
Game theory analyses the way that two
or more players choose actions or
strategies that jointly affect each other.
It provides a model for those situations
where it is impossible to make decisions
in isolation, but which necessitate
actions and reactions based upon what
an opponent or partner decides. While
game theory's terminology may sound
frivolous, its implications are anything
but childish.
Take the case of investment strategies for local loop infrastructure in
telecomms. Observers may be surprised to learn how many network
planning professionals in telecomms are playing a game called 'The
Prisoner's Dilemma' -- whether they know it or not.
The Prisoner's Dilemma is based on a hypothetical situation in which two
people are arrested on suspicion of having committed a crime together.
However, the police do not have enough evidence to convict them for the
maximum sentence. The two prisoners are isolated and each is offered a
deal: if he gives evidence against his partner, he will be freed and the
partner will be punished. If neither accepts the offer, both of them will
receive only a small punishment because of lack of evidence. If each
betrays the other, both will be punished severely. The dilemma resides in
the fact that each prisoner can escape punishment but only if he knows
what the other prisoner is going to say.
By replacing some elements of the prisoner's dilemma with choices facing
network operators in the local loop, certain insights into the complexity
facing infrastructure builders become apparent.
Handcuffed
If a telco invests heavily in local fibre -- to the curb, building or home --
and there is little demand for bandwidth-intensive applications or services,
it would be economically disastrous. If the PTO does not build out local
fibre and someone else does and finds demand for fibre-based services, it
would also be economically disastrous. If it does not build out fibre and
neither do its rivals, it may squeak through but the rewards of such a
strategy are slight. And if the PTO and its rivals build out fibre, then it
could find itself competing in a commodity business.
Add to this the changing technical, commercial and regulatory
environments in which telcos are formulating these strategies and taking
these decisions and the 'game' starts to look deadly serious. When it
comes to the question of local loop investments, there is more than one
right answer and a multitude of wrong answers.
The dilemma for all parties lies in predicting the nature of demand for
information and entertainment services for a market which does not yet
exist and which is unlikely to behave as historical markets have done.
Local loop choices...
For network operators, the local loop or access network is the area
between each subscriber and the first switching centre. The configuration
of the local loop across much of Europe is constrained by how far an
unamplified electric signal can be carried over a copper wire. The local
loop can account for more than 50% of the total cost of an end-to-end
public network in many cases.
As a result, the local loop has long been regarded as a natural monopoly
for incumbent operators. The existing copper network provided PTOs
with a huge advantage over any competitor, who would have to incur
massive construction costs to build their own local infrastructure, fibre or
not.
Even monopolies face difficult choices. According to a report by Analysys
called 'Prospects for Competition in the Local Loop', the costs of civil
works required to physically replace copper with fibre account for over
half of the total cost of the new infrastructure. Moreover, new technologies
and protocols such as asymmetric digital subscriber loop (ADSL) throw
confusion into the equation by increasing the usefulness of copper
infrastructure even as fibre improves its price/performance ratio.
Therefore, the PTO's case for upgrading to fibre depends on two
important factors: an assessment of when market demand for new services
will make fibre cost-effective; a willingness to concede that existing copper
infrastructure is over-valued, now that competitors are in the market using
cheaper technologies or building all-fibre networks.
The Analysys report states that fibre will become attractive only when
operators are confident that a market for higher-bandwidth services exists
and that pre-providing this capacity will enable them to exploit this
opportunity and reap other benefits such as rapid deployment of new
services.
Yet today's telcos cannot make the copper vs fibre deployment decision in
isolation; they must predict what their rivals in the marketplace -- cable
operators and wireless companies -- will do. One of the biggest players in
the game is the cable industry. Cable operators are bound to be the chief
challengers to PTO loop dominance in the near to medium term.
Cable will become a major competitor to incumbent local telcos for the
provision of local telecomms services as markets are deregulated. In doing
so it could provide much of the local broadband infrastructure associated
with the 'information superhighway'.
According to a new report by Ovum, 'Cable: the Emerging Force in
Telecoms and Interactive Markets', the cable business is evolving from a
broadcast entertainment services industry into a telecomms and interactive
services industry. This is being accelerated by deregulation and technical
advances in transmission systems. Cable access lines by definition have
greater transmission capacity than their POTS counterparts and the cable
industry as a whole has more experience with tailoring entertainment and
information service packages for local markets.
However, even for cable operators, infrastructure investment is not
straightforward nor can it be based solely on technical considerations.
Indeed technically there is no competition -- fibre is clearly superior. The
crux of the investment decision -- and hence the dilemma facing operators
whether telco or cable -- is what type of demand for information or
multimedia services will they find in the local loop.
The Analysys study on local loop competition concludes that for
competition -- and therefore lower prices -- to spread effectively at the
local level, several conditions must be met: operators must be able to
create a sufficient revenue stream, for example by providing both
entertainment and telephony services; tariff structures must be rebalanced
to make it worthwhile to offer local telephony; investment must be
encouraged by offering a low upgrade cost or a 'first entrant' advantage.
Services and infrastructure
Whether investment decisions are proactive or preventative depends on an
operator's perception of the 'services vs infrastructure' argument. Services
and infrastructure are engaged in a dance which is reminiscent of the
debates over demand for the first VCRs or fax machines.
Innovative services will not be commercially viable unless there is sufficient
bandwidth and there is no rational reason to build high performance
infrastructure unless demand for broadband services exists. This argument
is circular because it rests on the idea that the services to be offered
tomorrow will be similar to those offered today.
It will also be overcome in much the same fashion as the VCR or the fax
machine. "The important thing to remember is that these infrastructures will
be paid for through new ways of working, new forms of entertainment,
models and organisation which are not considered part of the industry
today but will become part of it", says Peter Kreisky, Head of the Media
and Entertainment Practice at Mercer Management Consulting.
Commenting on a recent Mercer report, 'Colliding Worlds -- Separating
the Virtual from the Reality', Kreisky did not question the long term
potential of the new media products such as video on demand (VOD),
interactive television, multimedia products and information services. He
believes, however, that the short term models for success were
dangerously dependent upon the status quo. "They [telcos] are assuming
that demand will be for the digital equivalent of what they are carrying
today", says Kreisky; "I think they are assuming that it will be services for
the masses."
To test what the actual market is today for information and entertainment
services over different infrastructures, Mercer designed a research project
which took 850 North American consumers representing a national cross
section. Mercer used a computer simulation to show how the
communications, information and entertainment services would appear
based on different technology choices in the local loop.
Mercer simulated everything from the present copper plant through
intermediate technologies such as ADSL, all the way to all-fibre access.
For example, with video phones, the researchers simulated the level of
jerkiness at all levels. For entertainment, Mercer simulated the level of
VOD ranging from a fixed number of movies starting at five minute
intervals all the way to thousands of choices available at anytime.
One of the striking aspects of the survey was that of the 850 people who
took part, only two did not complete what was nearly a two hour exercise.
Moreover, many of the participants had never sat in front of a computer
before. "So the inherent appeal of these services is evident through the
ease with which we were able to shepherd people through what was a
very lengthy survey", remarks Kreisky.
That said, the survey indicated that for information services or simple video
phone service to experience an uptake in demand, the operator need not
go all the way with fibre. Large shares of demand could be captured at
lower levels of bandwidth. However, in the case of VOD, the opposite
result held. The Mercer survey showed that respondents held quite
negative impressions of VOD over copper infrastructure yet not for
reasons most had assumed. Beyond the picture quality of fibre, more
important for the survey participants were choice and interactivity, where
copper can not match the performance of fibre.
One of the most important aspects of the VOD trial was that
self-scheduling of television programming seemed to be more desirable for
the participants than typical models for delivering on-demand movies. The
ability to self-schedule television programmes to perhaps retrace some
soap opera star's love life may be a sad commentary on technological
civilisation -- but it was found to be attractive all the same.
Perhaps the most important result from the Mercer study involved the
segmented nature of demand for services delivered over electronic media.
The 'Colliding Worlds' study identified more than seven distinct patterns
of demand, not only with regards to content preferences but also to the
likelihood of early adoption as opposed to users waiting for certain
services to mature.
"What they want is very different in terms of content, in terms of packages,
how it should be delivered and over what terminal, how it should be priced
and what would be their brand preferences", says Kreisky. "In my opinion,
the winners of this will be the organisations who figure out which sections
are going to be the most profitable to serve and offer bundles which satisfy
that -- service, content, package, price -- and use those margins to
underwrite services for the broader population", he adds.
Go to jail
If telecomms operators base their strategy upon the past, they know they
are doomed. The copper-based PSTN simply cannot handle the demands
of a market based on interactive information services. Moreover, many are
assuming that consumers will act as a 'mass market' even though research
strongly suggests the opposite. "Our research showed that even with heavy
promotions, discounts, sampling and branding, consumers won't want to
buy everything from a single vendor", according to Mercer's Peter
Kreisky.
Pilot trials, market surveys and all of these 'step one' exercises have been
valuable to make the industry and its customers aware of the possibilities
offered by the convergence of computers, communication and content. The
problem comes at 'step two', when the fanfare has died down and
operators must invest serious money in a market which they do not fully
understand.
This is made doubly difficult by the fact that the performance of these
investments will be heavily influenced by other converging industries which
may or may not have an incentive to cooperate with the infrastructure
builder. In the meantime, government regulators are opening competition in
the local loop, while politicians are braying for the construction of a local
loop fit for the 'information age'.
For incumbents, the decision to replace their copper plant not only involves
when to replace copper with fibre but to what extent, how close to the
user, and to meet what demand. Moreover, for this to work, operators will
be forced to choose based on a market which does not yet exist for a
future payback they cannot predict while working with industries with
whom they have no real experience.
All of this implies that, whether 'prisoners' by choice or accident, in the
local loop the one certainty is that the line between investing and gambling
has become incredibly thin -- a dilemna indeed. |
