CHAKULA Issue #19: Innovating access – a special focus on CPRafrica


1. The need for competitive research for policy influence – e- interview with Alison Gillwald

2. Balancing convergence: using Constitutional rights as a framework for policy decisions – e-interview with Indra de Lanerolle

3. Levelling the playing field through benchmarking – e-interview with Christoph Stork

4. ‘Disruptive competition’: the case of One Network in East Africa – e-interview with Muriuki Mureithi

5. Innovation through competition: the budget telecom network model – e-interview with Rohan Samarajiva


What is CPRafrica?

Communication Policy Research (Africa) – or CPRafrica – is a new forum that seeks to strengthen the contribution of African academics and researchers in the global debates around information and communications technologies (ICTs). CPRafrica’s inaugural conference, entitled “Looking back at a decade of communications reform: Looking forward to 2020”, was held in Cape Town, South Africa, from 18-21 April. It was attended by 50 participants from 25 countries, with some 23 papers presented.

The forum intends to encourage intellectual endeavour and research in the area of ICT policy and regulation in Africa and allow African academics and researchers to engage and profile their research. The conference provided an opportunity for senior, junior and mid-career scholars to meet face-to-face and exchange ideas, network, and included a two-day session of tutorials for young scholars. Thirteen young scholars participated in these tutorials.

This issue of Chakula focuses on four papers from the conference. It also highlights a paper that was presented at an Organisation for European Economic Co-operation (OECD) and InfoDev conference in Paris, 10-11 September 2009, and references a 2008 policy paper on mobile banking (M-banking) prepared by Research ICT Africa (RIA). In different ways all of these papers deal with the issue of innovation in ICT access in Africa through policy and business model innovation.


The need for competitive research for policy influence
e-interview with Alison Gillwald


“[H]igh quality, rigorous research…is required to compete and complement with each other for policy influence… In mature economies researchers from multiple universities would be debating and refining the positions governments should be taking on everything from regulating next generation networks to demand stimulation for broadband.”


Alison Gillwald is Executive Director of RIA. She is also Adjunct Professor at the UCT Graduate School of Business, Management of Infrastructure Reform and Regulation, and a member of CPRafrica’s organisation and selection committee.

CHAKULA: You have just held the CPRafrica conference in Cape Town. What are you hoping to achieve through the conference?

ALISON GILLWALD [AG]: There is almost no scholarly research being undertaken in the field of ICT policy and regulation on the continent. A Google scholar search on the subjects throws up around five scholars on the continent who are published in peer reviewed or accredited journals. It is this kind of high quality, rigorous research that is required to compete and complement with each other for policy influence. In mature economies researchers from multiple universities would be debating and refining the positions governments should be taking on everything from regulating next generation networks to demand stimulation for broadband. Although there are pockets of applied research being undertaken there is no tradition of critical intellectual engagement in this area on the continent. The purpose of CPRafrica is to provide a forum for nurturing and showcasing research in the area of ICT policy and regulation on the continent and enhancing its quality through rigorous academic review and debate. The conference is complemented by a young scholars programme to expose young scholars who may be excluded from such peer-review, paper- acceptance-only style conferences without such a category. Some of these are part of the IDRC- [International Development Research Centre] funded PhD programme to encourage doctoral research in ICT policy and regulation. The idea here is to build a cadre of policy intellectuals on the continent able to critically engage government on the basis of relevant research and contribute meaningfully to research and policy excellence. This will further enhance Africa’s standing in international research and governance fora, in which its participation has historically been suboptimal.

CHAKULA: Reviewing some of the papers presented at the conference, it strikes me that there are a couple of threads that are emerging. Two in particular stand out: the notion of “innovation” in the telecommunications space, and the challenges around convergence and policy when two distinct sectors with different ways of doing things are brought into conflict with each other. I also went back to Research ICT Africa’s 2008 M-banking policy paper, which raises similar themes, and I would like to use that as a starting point.
First, on the issue of ‘innovation’. In the M-banking paper, the following assertion is made: “Policy-makers and regulators need to ensure that evolving systems serve the broader objectives of economic growth and development as well as protect consumer interests, while creating an environment that encourages and rewards innovation”. In what ways can policy inhibit or encourage innovation in the telecommunication’s sector?

AG: Indeed, providing certainty to investors and operators while retaining the levels of flexibility to enable innovation in a fast-changing environment is one of the most difficult balancing acts that policy-makers and regulators have to perform. I think the linkages and catalysts between technology, market and regulatory innovation are becoming clearer all the time. New technologies and service offerings have prized open markets and the entry into less policy and regulatory constrained markets has made taking certain technologies to market more viable. This has triggered further possibilities across historically distinct platforms, not only between broadcasting and telecommunications, but between fixed and mobile services and even entirely separate sectors such as telecommunications and banking. The challenges to the expansion of such services are really regulatory now rather than technological – and that is not to say that one does not want or need public interest regulation either in the telecommunications or banking sector, but it has to be done in new, innovative ways that enable to extension of these services to those who currently don’t enjoy them. Once these various forces are unleashed they are able to intersect and create new opportunities and innovative ways of doing things that have not been done before.

CHAKULA: Innovation here seems necessarily to be tied to market gain – the objective is to increase or capture market share. In both your M- banking paper, and the case study of the mobile operator One Network in Kenya, preconditions exists that facilitate innovation. With M-
banking there are low-income earners who are ‘unbanked’ and who could benefit from some kind of low-cost transactional instrument, and with One Network, there is a significant level of cross-border traffic that makes a seamless network attractive.

AG: It is true that innovation is often driven by market forces and pursuit of profits, and, traditionally, with new technologies have focused on high-end markets. But much of the ICT innovation we are witnessing in developing markets is focused on what has been referred to as the ‘gold at the bottom of the pyramid’ – very profitable turn-over of high volumes of sometimes minuscule margins on products that, by breaking them up or making them available at cost, the masses at the bottom of the economic and social pyramid can enjoy things like pre-paid phone vouchers, or transferable airtime vouchers. And many of these products have been commercialised innovative practices by the poor in order to access and affordably use communications services – such as missed calls, multiple sim card usage that allows for same net rates, or ‘plastic roaming’.

CHAKULA: If we consider Indra de Lanerolle’s fascinating case study on the South African convergence scenario, we see two sectors (broadcast and telecommunications) in conflict with each other because policy decisions are made according to different frameworks: simply put, economic versus public interest. In fact, Indra does seem to suggest that these are in competition with each other, and resolves this in an interesting way. It feels hard to believe that ‘consumer interest’ is the same as ‘public interest’?

AG: I think with the shift from public utilities to competitive markets many of the public interest objectives of delivery and service are met through serving the consumer interest. Nevertheless there is public interest regulation that is required to improve wider and collective consumer welfare – to provide access to ‘uneconomic areas’ for example – though with new more cost-effective, rapidly deployable wireless services, this concept in markets that enable competitive entry is regularly not proving to be the case. But as long as we have the large number of poor that we do, we will need some level of social regulation – even though a lot of the current pent-up demand could be met with greater market efficiency (more competitive markets offering better prices). And then there are the more traditional content regulation issues either to restrict certain ‘harmful’ content or activities or to enable it, such as local content regulation. That too may be found to be highly profitable, but may need either protection or encouragement.

CHAKULA: Indra’s paper, like your M-banking policy paper, shows that regulating convergence is tricky because of the ‘convergence’ of two or even more sectors; whether broadcast/telecommunications or telecommunications/banking etc. What are some of the key challenges that policy-makers can expect to face in Africa?

AG: The key challenge for African regulators is that they are still trying to deal with legacy regulation around first and second- generation infrastructure and access. At the same time, if they do not want the agenda to be set for them in international fora, they need to deal with next-generation issues, not only of converged IP [internet protocol] networks and services and the next-generation regulation issues of network and service-neutral regimes, but of cross-cutting issues of electronic commerce frameworks, intellectual copyright rights, security and privacy issues, and so on. And you have to do it all or be left behind…

CHAKULA: One frustration is that when one reads a good paper that seems to offer a solution to a problem, one is also met with the feeling that those with decision-making powers are probably not going to read that paper, or seriously consider its arguments. Do you feel the same? If so, how do you think CPRafrica picks up on this challenge? Is it just a case of repeating issues until policy-makers take them on board?

AG: No. CPRafrica is one of several strategic strands towards having evidence-based ICT policy on the continent. This is about organic and indigenous knowledge creation and contribution, at the national level, at the level of regional association and continentally, and also about global engagement and influence. For too long have the solutions come from the developed world. Of course, there are lessons to be learnt and we don’t need to reinvent the wheel, but we also have different challenges and Africa has demonstrated remarkably innovative responses to these when they are informed by sound policy, effective regulation or thorough and appropriate business plans. The indicator research done by RIA and its analysis in order to assess policy and regulatory outcomes is fed into several initiatives, globally and locally. RIA provides the only comprehensive public domain demand-side data on ICT access and usage on the continent. This is used in national, regional and continental meetings on ICTs, and in the database and reports of multilateral agencies such as the OECD and the International Telecommunication Union (ITU), to better inform their understanding of developments in Africa. It is true that sometime decision-makers do not like to hear of the widespread policy and institutional failure on the continent, but many do – especially those that are rapidly improving and beginning to see the rewards of their reforms. This research is also used to develop training curricula that address the needs of policy and regulators in a developing country context. So, for example, as part of the global research and training collaborative we conduct a professional development course on alternative regulatory strategies at the UCT Graduate School of Business Infrastructure Reform and Regulation Programme to build institutional capacity on the continent. So CPRafrica is just one arm of a multi- pronged strategy of research and education, institutional capacity building and technical assistance and dissemination and advocacy, through our website database, policy papers and workshop and public presentations.

CHAKULA: What is the way forward for the conference? Will there be more?

AG: Yes, in order to build and sustain this much-needed capacity we will have to find a way for CPRafrica to become an annual institution.

Related links:

M-Banking the Unbanked: RIA Policy Paper No. 4:

CPRafrica conference details:


Balancing convergence: using Constitutional rights as a framework for policy decisions
e-interview with Indra de Lanerolle

Paper link: Please e-mail the author at indra at


“[N]o country I know of has completed the project of aligning all regulation and law to [the possibilities of convergence]. But some countries and regions are moving far more swiftly than others. Among those that are moving more swiftly would be the EU countries. Among those that are moving slowly are many African countries.”


Indra de Lanerolle is an adjunct Lecturer at the University of Witwatersrand Journalism and Media Programme. An award-winning film and television producer, he is a regular speaker and commentator on the impact of the information society on organisations, media, social development and the economy.

CHAKULA: The South African government has committed to promoting convergence in the broadcasting and telecommunications sectors. So much is clear from the 2005 Electronic Communications Act, and the merging of the telecommunications and broadcasting regulators in 2000.
But you point out that there is a fundamental difference in approach between how the two sectors are seen at a policy level, which is standing in the way of ‘pure’ convergence of the sectors. In your paper, you argue that the telecommunications sector policy process has largely been justified on economic grounds, while broadcasting policy has been governed by political and social considerations. I am paraphrasing here, but can you say more about this distinction?

INDRA DE LANEROLLE [IL]: Most ‘utility’ regulation (applying to electricity, trains, telephones, for example) was originally justified on economic grounds – usually that these industries were monopolies and that as such, without regulation there would not be sufficient
competitive pressures to keep prices as low as they could be. Many economists went further and argued that these utilities were ‘natural monopolies’ so that the most efficient services would be provided by a single regulated provider rather than allowing many (less or unregulated) competing providers. Broadcasting regulation was sometimes justified on similar grounds but in my view these economic arguments were and are very weak when applied to broadcasting. The stronger argument (and the better explanation in historical terms of why broadcasting regulation has continued to the extent that it has) is that it is motivated and justified on social and political grounds.
These justifications include: a desire or need to preserve national cultures and languages (in the face of globalising forces in mass media, especially television); a desire or need to provide specific genres of programming (e.g. educational programming, children’s programming); and, less high-mindedly, a political desire on the part of governments to control information and news. The distinction in these justifications or motivations is important for convergence.
Convergence increasingly challenges economic justifications for regulation in telecommunications as monopolies fall away. But the social and political justifications for broadcasting regulation do not necessarily fall away in synchrony.

CHAKULA: You review various tests to see how well policy convergence has faired. First, there is the assessment of policy alignment – or the alignment of policy processes; then there is the test of technological neutrality; and finally, the ‘level playing field’ test.
South Africa has not faired well in any of these – can you elaborate?

IL: The first test I suggest (from Alison Gillwald) is whether policies in telecommunications and broadcasting are being developed together. In the face of convergence it is important that they are. As I show, in fact these policies are being developed autonomously. The second test is that of technological neutrality. Here I argue that while the Electronic Communications Act acknowledges convergence and attempts to develop ‘technologically neutral’ licensing criteria, it in fact maintains two separate regimes – one for telecommunications and the other for broadcasting. In applying content regulation, for example, it treats the two as entirely separate. So overall, South Africa remains far away from achieving the goal of technologically neutral regulation. The third test – of a level playing field for competition follows closely from the lack of technological neutrality.
One example of the lack of a level playing field can be seen in the treatment of voice – where VoIP [voice-over-internet protocol] and packet-switch telephony, for example, are treated entirely differently in law and regulation.

CHAKULA: Does ‘pure’ convergence exist anywhere – or is convergence in reality always a push and pull of many factors, economic, political social etc.?

IL: As I argue in my paper the best definition of convergence sees it as a set of technological possibilities. On this definition, the question could and maybe should be put rather differently. We might ask whether any country is fully adapting their regulatory frameworks to this set of possibilities. The answer would then be that no country I know of has completed the project of aligning all regulation and law to these possibilities. But some countries and regions are moving far more swiftly than others. Among those that are moving more swiftly would be the EU countries. Among those that are moving slowly are many African countries.

CHAKULA: You offer what appears to be a powerful guiding principle to make policy decisions based on a fluid and changing environment – one based on public interest, rather than sectoral interest. This involves using Section 16 of the South African Constitution to guide policy decisions, which states that “Everyone has the right to freedom of expression, which includes …freedom to receive or impart information or ideas…”. Can you say more about how this would work?

IL: We could say that the Freedom of Expression clause represents a test for measuring the benefits and dis-benefits of changing regulation and policy to promote convergence. Lets take the issue of the (limited) number of broadcast licences – especially television broadcast licences. As broadband take-up increases there is likely to come a point where there is the possibility of Video on Demand as broadcast video services become economically viable via the internet.
This is a classic case of convergence. So if convergence is to be promoted we might want to see a couple of things: firstly, we would certainly want these new services to be allowed and, indeed, encouraged; secondly we might want these services to be treated in a similar way to ‘traditional’ broadcasting. Unless we wanted to artificially restrict the number of new internet-based services, the only way to achieve neutrality would be to significantly lighten the regulation of broadcasters, or at least greatly increase the number of new licences issued for ‘traditional’ broadcast television services.
Given the social and political justifications for broadcast regulation though, this would probably be greatly resisted. To give one example as to why this would be the case we could see that increasing competition in ‘traditional’ television broadcasting would have a significant economic impact on the public broadcaster and thus diminish its ability to fulfill its public broadcasting mandate. My suggestion is that we could use the Constitution as a basis of balancing these issues. We would be able to compare, for example, the potential loss of people’s rights to receive information (because, say SABC might reduce its news services) with the potential gain of people’s rights to receive and impart information via the internet.
This would require judgements, but these judgements could be based on research of real benefits.

CHAKULA: It seems to me that as things stand you have enormous competing interests that stagnate the convergence policy process. On the one hand, the interests of telecom operators (including the state), and on the other, the interests of broadcasters (the state and private enterprises). How do you think a decision-making framework where the public/consumer would have the dominant interest would go down with these powerful stakeholders? What was the reaction to your paper from those gathered at CPRafrica?

IL: Most of the participants at CPRafrica were researchers and academics rather than the ‘powerful stakeholders’ that you describe. I’m not an objective witness but the paper appeared to be well received! Of course in all issues concerning regulation, there are always significant economic interests involved and there is always the danger of ‘regulatory capture’. Whether such ‘public interest’ arguments as the one I advance in the paper are pursued is largely a matter of political will – in government and in the independent regulator. When some of the ideas included here were presented on behalf of the Freedom of Expression Institute (FXI) and Association for Progressive Communications (APC) amongst others to the South African regulator ICASA, it appeared there was significant interest. I have also argued elsewhere that there is little chance of ‘public interest’ arguments holding sway in the development of policy and regulation if there is no interest shown by the public! The Save our SABC (SOS) coalition that has brought together a range of organisations in South Africa, including the major union federation, to campaign on public interest questions on broadcasting is a helpful


Levelling the playing field through benchmarking
e-interview with Christoph Stork

Paper link: (or e-mail the author at: christoph.stork at


“Operators are not accountable to the public beyond the audited financial statements that do not include detailed cost data. That could be changed given that telecom operators have the privilege of operating in a sector with restricted market entry.”


Christoph Stork is a Senior Researcher at Research ICT Africa (RIA), and also a consultant to LIRNEasia’s Teleuse at BOP3 project. His responsibilities at RIA include designing and conducting quantitative and qualitative research and interacting with policy-makers and regulators.

CHAKULA: You were commissioned by the Namibian Communications Commission (NCC) to conduct a benchmarking study in Namibia in 2009, following a dispute about interconnection charges. You have used the case of Namibia to illustrate and argue for setting interconnection benchmarks. As you argue, benchmarking is possibly a more efficient way of determining a fair termination rate (based on an efficient operator). For the layperson, some of this terminology is tricky. The standard way of determining termination rates is to look at the “forward-looking long-run incremental cost” (LRIC). The problem with termination rates – or the cost one provider charges another to carry its calls – is that operators have been burying inflated profits at this link in the communications chain; they have been charging competitors more than they needed to, with the knock-on effect to the consumer. Can you explain what LRIC is exactly? And then how does
benchmarking model differ from this?

CHRISTOPH STORK [CS]: LIRC is a way of calculating the cost of termination. There are many different approaches to it. In short, it is a method to calculate the cost of terminating a call based on current technologies and only including costs that arise from terminating a call. Dominant operators often argue that access infrastructure such as base stations need to be part of termination costs. However, base stations are being built to provide services to their own subscribers. The incremental cost of termination is the cost difference between an operator that does not offer termination services and one that does. Implementing LRIC is challenging, expensive, time-consuming, and the required information is often not available in developing countries. The benchmarking methodology benchmarks termination rates, termination costs and regulatory best practice.

CHAKULA: In your experience in Namibia are operators open to cost- based termination rates to level the competitive playing field? Or do they want to milk the termination of calls for as much money as they can get? For instance, you point out that operators have typically been quite fickle, arguing against cost-based termination when they have a market advantage, but arguing for it when they are a market entrant…

CS: Operators that are net receivers of termination payments will always try to maintain the status quo, anywhere in the world. New entrants and small operators are often the net payers and are keen to get to cost-based termination rates quickly. In Namibia, South Africa and most other countries, cost-based termination rates are required by law or licences or both. In Namibia, reducing termination rates increased competition and led to lower prices for consumers and an expansion of the market. The result for the incumbent operator MTC has been: more subscribers, more traffic and higher EBITDA [Earnings Before Interest, Taxes, Depreciation and Amortisation] margins.

CHAKULA: What is the link between cost-based termination rates and IP- based Next Generation Networks (NGNs)? You say that leveling the playing field when it comes to termination rates will help with the transition to NGNs?

CS: NGNs are technological and service neutral. Billing will no longer be based on minutes but on MBs [Megabytes]. Cost-based termination rates help implement service neutrality and are a stepping stone towards NGNs. NGNs might however require a different billing system altogether that would not require termination rates. In a NGN environment receiving and sending data would be billable. In Africa and Europe, the receiver of a call does not have to pay.

CHAKULA: You point out that one of the challenges is getting good data. What are the difficulties around getting this data from operators (or regulators, for that matter) in Africa, and elsewhere?

CS: Most countries in Africa have not conducted detailed cost studies yet. Those that did use information from operators that are considered confidential. ICASA [the South African regulator] collects very detailed information from operators and would probably be sued by them if it were to pass it on to third parties. Operators are not accountable to the public beyond the audited financial statements that do not include detailed cost data. That could be changed given that telecom operators have the privilege of operating in a sector with restricted market entry. Regulators could prescribe cost reporting for annual statements that need to be made public.


‘Disruptive competition’: the case of One Network in East Africa
e-interview with Muriuki Mureithi

Paper link: (or e-mail the author at: mureithi at


“One advantage Africa has is the policy and the regulatory framework are not yet cast in stone; in some cases it is not there and with few administrative nods some uncooked products can launch in the market.”


Muriuki Mureithi is an independent ICT consultant based in Nairobi Kenya with a passion to help exploit the benefits of ICTs for Africa with a career now spanning 27 years. His interest is policy, regulatory , strategy evolution and competitive issues on ICTs in Africa with special focus on the rural and disadvantaged communities and strategies to empower the communities using ICTs. He is currently completing his PhD.

CHAKULA: Your paper, which you co-wrote with Alison Gillwald, offers a case study of how Zain’s One Network disrupted the mobile market monopolies in East Africa (and beyond). They did this by removing roaming charges and letting their customers access a seamless and comparatively cheap service across its networks in Kenya, Tanzania, and Uganda. This then forced the other operators to offer similar services. What do you mean by ‘disruptive’ competition – you also go on to suggest that One was not entirely disruptive, because it failed to keep its competitive advantage?

Muriuki Mureithi [MM]: Disruptive competition in our context and conventional language is a game changer. Operators had perfected the game of increasing numbers and profits by charging for all services, all trudging alone on a strategy path set by the leader. Risk reduction was a key factor. The Zain One network was a game changer and the big operators had to look back and go back to the drawing board. Unfortunately for Zain, the big boys came back with ‘hammer and tongs’. Their solution covered more networks (Safaricom, MTN Uganda, MTN Rwanda, Vodacom Tanzania and UTL [Uganda Telecom Limited]), more countries in East Africa (Rwanda, not covered by Zain) and more customers (the competition are market leaders in all the markets apart from UTL). Zain briefly increased numbers in all the markets but soon fizzled under the pressure of the bigger operators. Its gamble that the other operators could not work together failed. It had never been tested before and the risk it took did not pay off. The customers were, however, the beneficiaries in the long run.

CHAKULA: The point you are making in your paper is that given the right regulatory environment, market forces can quite quickly result in lower prices and better access options for the consumer. The regulatory ‘kick’ here was the granting of an international gateway license to One in all three countries. What were some of the other enablers that allowed One to be so successful? For instance, you mention historical factors, and the cross-border movement of people that pre-dated its market offering…

MM: One is an innovation without doubt. However this is founded on some features or drivers that pre-existed, and Zain took advantage of the institutional memory of some of the out-of-the-box initiatives that it had exploited successfully. The first case is direct cross border connectivity between Brazzavile and Kinshasa by Celtel, instead of through Belgium. Celtel, later bought by Zain, had operations in both countries. Tariffs fell by 80% overnight, and this was to be extended to all border towns between the two countries with similar dramatic effect. It is noteworthy that the private sector could see an opportunity, but that opportunity could only be realised when the regulatory environment changed. That gave Celtel a head start as an innovator in both countries. In East Africa, the conditions were there: highly integrated communities travelling across the porous borders, formal and informal trade, and a legacy backbone linking the three countries. Zain also had licences in all the three countries, but no gateway in Kenya. Their performance was failing in all the three markets, and was a catalyst. So it was time to do something.
When the international gateway was liberalised in Kenya, it was time to jump and they did. With similar branding and ease of use across the borders, the product was a scare to the bigger operators. They had to act – and they did by launching competing products. An interview with
Safaricom indicated that the traffic is not a major consideration, but the marketing advantage that Celtel stood to gain as an innovator is.

CHAKULA: One of the main points of your case study is to offer a comparison to the European Union where they have attempted to force operators to lower their international roaming charges, with great resistance, it seems?

MM: European consumers were keenly aware of the high cost with no justification for roaming charges. The tariff setting was and still is very opaque and the consumer had no choice. The European Commission (EC) carried out studies confirming the same. Unfortunately, despite the overwhelming evidence, operators refused to play ball opting to continue reaping off the consumer. The EC came back and set the tariffs to bring the tariffs down over a three-year period and committed to review another downward reduction if it still found it warranted. What is surprising is that the competition did not trigger such dramatic options as it did in Africa. Perhaps the positioning of the operators is fixed and each is comfortable with their space. What Celtel (now Zain) showed is that operators have a wide range of options to compete, and it takes tough going to think out of the box and take a daring option. They did and changed the game forever. All Zain countries in Africa and Middle East are on the One network. It is growing: one operator in Egypt joined the network this year though they are not a Zain operator, and Cell C in South Africa was joining the One network in time for the World Cup. Not to be left behind, MTN has been implementing a similar package. What was even more innovative is that very different operators Safaricom (Kenya) MTN (Uganda and Rwanda), UTL (Uganda) and Vodacom (Tanzania) worked together and eliminated roaming. Today, an East African travels without roaming, apart from Burundi. Regulators have some work to do where the operators do not see a competitive advantage. In the case of Burundi this is forgotten.

CHAKULA: You suggest that it is not always fruitful to look to developed countries for regulatory and requirements in order to stimulate market growth – in fact One offers a useful case that other regions could follow?

MM: The legacy networks and conformist approach sometimes is a disadvantage. The raw battle for the customer is bruising and those who pass the post spell doom to those behind. In Kenya, one operator has an 80% market share. Its profit before tax in 2009 is equal to the revenues of the other three competitors, and they have mounting losses. The future is tough. Of the top 15 major cellular markets in Africa 10 markets have a dominant operator with over 80% market share.
With those market dynamics, Africa provides a cooking pot for out-of-the-box solutions, because the option is dire. At present One network, the runaway M-PESA [M-PESA is a Safaricom service allowing you to transfer money using a mobile phone] were born in Africa and are being exported to the rest of the world. I believe more will come. One advantage Africa has is the policy and the regulatory framework are not yet cast in stone; in some cases it is not there and with few administrative nods some uncooked products can launch in the market. This may not happen in developed countries.

CHAKULA: In your recommendations, you talk about actions to expand the “innovation space”. What are some of the key actions?

MM: The operators have control of most of the resources they need in search of the customer. The policy and regulatory space is out of their control – this is space that governments can expand as illustrated by Zain’s One. MPESA would not have succeeded if the government did not do the same in the banking regulatory space.
Governments need to regularly review the regulatory space with active engagement of a multi-stakeholder process.


Innovation through competition: the budget telecom network model
e-interview with Rohan Samarajiva

Paper link:


“The status quo must be unbearable.”


Rohan Samarajiva is the Chair and CEO of Lirnasia. His paper, “How the developing world may participate in the global Internet Economy: Innovation driven by competition” was presented at a workshop organised by the OECD and InfoDev in Paris, 10-11 September 2009.

CHAKULA: In your paper, you talk about the Budget Telecom Network Model (BTNM), which is brought about by competition allowing operators to reduce the transaction costs of low-end clients. This, as you point out, is different to the standard Average Revenue Per User (ARPU) model. How does it make the ARPU model redundant?

Rohan Samarajiva [RS]: ARPU is a short-hand that outside observers use to see if the firm is doing well, whether its prospects are good, etc. It is, like any indicator, imperfect. You get it by taking total revenue (preferably without extras like roaming) and dividing by number of subscribers. Of course no one really knows what a subscriber is any more, with even poor people holding up to five SIMs, foreigners having SIMs, no agreement on what an active SIM is and so on. You can get better results by looking at revenue per minute.
Take total revenue (less roaming and other stuff) and divide by Average Minutes of Usage per User per Month (MOU). This is a better indicator. But investment analysts are still not used to this and it would require disclosing MOUs to calculate.

CHAKULA: Can ARPU be used as a business model?

RS: Operators do not actually do much with the ARPU. It is not a business model as such, just an indicator. But getting more from each subscriber (if this is known) is not a bad idea. Just that it does not predict whether the company will make money or not. The best indicator for that is EBITDA [Earnings Before Interest, Taxes, Depreciation and Amortization] margin. Sri Lanka in 2007 had an operator with LKR311 (approximately USD3 at the time) ARPU making close to 50% EBITDA margin. In the end, the success of a business model lies in whether it generates profit.

CHAKULA: What is your understanding of ‘innovation’ in the telecommunications space? You talk of “business innovation”, rather than, say, technological innovation?

RS: Tech innovation is important, but it is not the only thing. Pure tech innovation is done by manufacturers of network equipment and handsets. That is good. Business process innovations (e.g. lowering the costs of base stations through software) are done by operators.
These include technical aspects, but are not limited to them.
Shifting from one business model to another (discovering the latter)is also innovation, but it may or may not not have a tech aspect at all.

CHAKULA: What are the preconditions for innovation, do you think?

RS: The status quo must be unbearable. The BTNM innovation occurred when competition got so intense that there was no way to gain market share or even survive without doing something new.

CHAKULA: Does BTNM have implications for increased access to broadband internet for the majority of people on a continent like Africa?

RS: Yes. The latter part of the paper is entirely on the extension of BTNM to broadband. Some headlines are that operators must have enough money from voice that can be invested in the 3G plus networks.
Once the overlay network is built out the operators have to offer low prices. Prepaid sachet pricing is best, where one buys packages of connectivity in minutes or in capacity. Here, because of lower transaction costs and prices there should be an influx of new customers. This is already on offer in Asia. Africa has to lower prices. Access will be over mobile networks, using dongles or built in modems, for laptops and other devices, including phones. ADSL will be a niche product. Wireless access is the future.


CHAKULA is a newsletter produced by the Association for Progressive Communications (APC). It aims to mobilise African civil society around ICT policy for sustainable development and social justice issues.

We welcome your opinions about this newsletter. Send your comments, feedback or contributions to chakula at To subscribe or unsubscribe e-mail chakula at or go to:

Sign in to