The Kenyan government has recently put forward a plan to
simultaneously promote cost-effective use of the 2.6GHz band
and save operators from having to spend money in an auction.
The plan is to offer the management of the band (up to
190MHz of spectrum, which is suitable for high-speed mobile
data services) to an independent company in order to create
an open access wholesale network. Operators would purchase
capacity from the company, and bundle it into packages and
products that they would sell in a competitive market to
retail customers. The end result would be a single, highly
utilised network with low unit costs.
The plan sounds good, but the country’s government could be
‘jumping from the frying pan into the fire’.
The frying pan may not have been that bad, after all
Why does the Kenyan government feel wary of assigning
spectrum to individual operators? The reasons suggested by
Bitange Ndemo, Permanent Secretary of the Ministry of
Information and Communications, include:
- high prices paid in an auction lead to high prices
charged to consumers
- operators failing to honour roll-out commitments
- an auction would have 19 potential buyers and room for
only three winners, so would discriminate against those
that could not afford it.
Do high auction prices lead to high prices for consumers?
Actually, it is the other way around: if operators can
charge high prices, they will be willing to place high bids
at auction. Service providers can charge high prices if
there is limited supply of their product or of reasonable
substitutes – and low auction prices do not guarantee low
service prices. Furthermore, it is possible to design award
processes to achieve ends other than extracting maximum
revenue.
The 2.6GHz spectrum is unlikely to attract 19 serious
bidders, given that it is relatively poor for providing
coverage outside urban areas. A new entrant operator that
aims to launch services based on this spectrum may be aiming
to target a niche, such as medium-sized and large
enterprises in urban areas. Giving them a chance to do this
may be useful, but not as useful in expanding access to
broadband as, for example, raising money in an auction to
fund rural access.
Think carefully before jumping into the fire
Kenyan policymakers have correctly identified that greater
supply (and more competition) is the key to reducing mobile
broadband prices. They have also commendably declined to
squeeze auction revenue out of operators. However, their
solution turns away from infrastructure-based competition,
which has revolutionised the telecoms industry during the
past few decades. Will their alternative approach, which
focuses on retail competition, work?
The bad news is that pure wholesale business models for
mobile data have not done well in the past. Reasons include
difficulty in developing ‘one-size-fits-all’ wholesale
tariffs, absence of profitable voice services, and
dependence on the strategy and performance of other
operators.
The better news is that network-sharing deals can work.
Operators around the world are signing deals to share radio
access networks.1 In all likelihood the critical
feature is that the retail operators have a strong interest
in the entity that runs the network: financial interest to
ensure that it makes a success of the services, and
management interest to allow flexibility and tailored
‘tariff plans’.
Tantalisingly, the arrival of Indian-style telecoms in
Africa (led by Airtel, among others) may naturally lead to
greater sharing (and price cuts), without the need for
government intervention.
It is clear that making this 4G plan work in Kenya will be
a challenge for the selected network manager (a “group from
the USA” has been mentioned as a candidate). It will be even
more of a challenge for those charged with negotiating an
agreement and monitoring its performance.
This Insight article is based on an interview conducted
by Russell Southwood for Balancing Act Africa. Please click here to view the
full article.
Analysys Mason has experience in developing concrete,
coherent national
broadband plans and in spectrum
policy and auctions.
1 For further details, see
Analysys Mason's Insight article Wireless
infrastructure sharing saves operators 30% in capex and
15% in opex.