Ethiopia Focus Group report: what we know about opportunities in the liberalising market so far

ethiopia-feature.jpg

By the end of Q2 2020, three MNOs should be active and towercos will be on their way to adding 5,800 towers over the next three years

Is Ethiopia about to host the next ‘land grab’ for the tower industry? That was the view at the TowerXchange Meetup Africa’s Focus Group on Ethiopia. After considerable growth from 2013 to 2016, telecoms network investment slowed, leading to a new push by the government to part-privatise the incumbent telecoms operator and issue two new licences to kick start a telecoms boom.

Investment by towercos is likely to be central to enabling rapid network expansion and densification in the country, and thousands of new sites will be built over the next few years. Updating our last editorial on Ethiopia with insights from our Focus Group, TowerXchange examines the opportunity.

Why now?

Ethiopia has averaged over 10% annual GDP growth over the last decade. Thanks to a more stable government and a series of relatively successful Five Year Plans that have been aimed to improve agriculture, infrastructure, logistics and global connectivity, Ethiopia’s economic growth has outperformed almost every other economy in Africa, and much of the rest of the world.

Now home to over 100mn people and Africa’s last major telecoms monopoly, the stage is set for telecoms liberalisation and a resultant surge in network investment. In April 2018 Abiy Ahmed became Prime Minister on a reform ticket and rapidly set about transforming society, freeing journalists and opening up the economy to outside investment. These efforts have seen Abiy Ahmed awarded the Nobel Peace Prize.

Worried Ethiopia’s growth would fade, Ahmed has moved to reinforce Ethiopia’s economy by privatising major state owned enterprises, beginning with Ethio Telecom. Ethio Telecom’s privatisation is not only a test case, which gives it more credibility than other privatisations which have failed in Africa, it is also a key enabler of Ethiopia’s plan to boost its ICT sector, which gives it additional political support.

Despite some political instability and currency depreciation (the Birr has lost over 60% of its value versus the U.S. dollar in the last decade), Ethiopia remains the largest unliberalised telecoms market in the world, a sure sign of pent up demand for network investment that is on the brink of being unlocked.

Famous for the ancient empire of Aksum, Emperor and King of Kings Haile Selassie, long distance running and a series of disastrous 1980s famines, Ethiopia has a long history of being in the news, often for the wrong reason. So it is nice to write about the positive changes taking place, and they are positive changes to be built on a solid foundation of infrastructure – starting with telecommunications.

Ethiopia – TowerXchange central projections for cell site and power, year end

ethiopia-projection.png

On the ground in Ethiopia

The most reliable information we have on the level of coverage in Ethiopia comes from Ethio Telecom’s own reports. As of July 23rd 2019 Ethio Telecom had 41.9mn mobile subscribers – 19.6mn 2G only, fewer than half a million 4G subscribers, and the remainder on 3G – totalling 38.8% of the 108mn population. 2G coverage has reached 85% geographic coverage, while Ethio has achieved 66% 3G population coverage. There are 332 4G sites, all in Addis Ababa. It is clear there is substantial room for further investment to expand 2G coverage, overlay 3G and rollout 4G beyond the capital.

We are yet to find a precise and reliable count for towers in Ethiopia, but the range of credible counts range between 7,500 and 8,500. In fact, we believe one of the delays in the privatisation process of Ethio Telecom is the conducting of a proper count and audit of their existing sites and business through KPMG.

Many of these sites, particularly the rooftops sites common in Addis and other major cities, are often unshareable. The questionable structural capacity for sharing of Ethio Telecom’s existing towers is one of the major unknowns facing future telecoms rollouts and network design. Few of the existing sites can take upgrades, and most of those that have space for extra antennae cannot accommodate three operators as currently configured.

Another source of uncertainty, revealed at TowerXchange’s Ethiopia Focus Group, is around ground leases, especially on rooftop sites. At the moment, Ethio Telecom has a soft form of eminent domain, defined as the right of a state owned entity to expropriate private property for public use, which gives them access to most sites they want. This is not a viable long-term process for international mobile operators and towercos who dislike expropriating land and prefer legally sound leases for their cell sites.

While Ethio Telecom does pay rent on some sites, on others it does not. These sites do have signed leases but landlords have not thought to – or have not been able to – monetise many leases. Landlords are not thinking about site acquisition or lease negotiations at the moment, but they will start to once international operators and towercos enter the market. For example, as revealed at TowerXchange’s Ethiopia Focus Group, in Myanmar towercos asked the country’s MNOs to not come on site visits because “their logos look like dollar signs to locals”!

The network of Ethio Telecom has been divided into 12 regions for network development. ZTE provide the network equipment in two regions, Ericsson in two more, and Huawei takes the lion’s share of the business, supplying equipment in eight regions. Each region saw considerable growth, from a total of around 3,000 towers in 2013 booming to around 8,000 in 2019. Over a thousand towers a year were added in the key years of 2013-2016, with a decelerating pace of build ever since.

To reach comparable number of subscribers per tower as Nigeria currently, Ethiopia would require an additional 7,000 towers. However, because towers in Nigeria are better suited to sharing than towers in Ethiopia, that figure of 7,000 may have to be augmented by the strengthening or rebuilding of many existing towers, so the total number of new site builds could be over 10,000. To provide effective 3G and 4G coverage, Ethiopia will require a total inventory of over 30,000 towers by 2029, it was suggested at TowerXchange’s Ethiopia Focus Group.

More conservatively, TowerXchange still predicts 5,800 new sites will be built by 2023, based on each of the three licence holders adding 500-1,000 towers per year, inclusive of Ethio Telecom undertaking a substantial programme of rebuilding and strengthening. This is a conservative forecast as Ethiopia added over 1,000 towers in several years between 2013-16, and that was without outside capital, towerco and international mobile operator expertise.


A very short history of Ethio Telecom

Ethio Telecom was formed in 2010, and was previously known as the Ethiopian Telecommunications Corporation, a division of the Ministry of Post, Telephone and Telegraph. It is one of the last monopoly telecoms providers in the world, and the largest. It is one of the ‘Big Five’ group of state owned corporations in Ethiopia, along with Ethiopian Airlines, the Commercial Bank of Ethiopia, Ethio-Insurance, and Ethiopian Shipping Lines.

The proposed privatisation and removal of barriers to entry in telecoms is the spearhead of a reform effort which will open up the whole Ethiopian economy. Its position in the broader strategy of the Prime Minister, H.E Abiy Ahmed (PhD), which lends credibility to the proposed privatisation. From 2010 until 2013, Ethio Telecom was managed by France Télécom, Orange’s predecessor. This was the beginning of a professionalisation of the company. Local control was reintroduced in January 2013. From 2013 onwards, Ethio Telecom embarked on a large-scale state-backed expansion programme of investment which saw coverage expand to around 85% of the population and 41.9mn mobile customers. In its July release of its annual results, the company announced it made 36.3bn Ethiopian Birr (US$1.27bn) in revenue and a 67% EBITDA margin of 24.5bn Ethiopian Birr (US$860mn). However a national shortage of hard currency, required for modern telephone network technology, means the state-backed strategy has likely achieved all it can, and Ethiopia requires outside investment.


Backhaul

Ethio Telecom has over 20,000km of fibre in the country and three fibre links to the rest of the world through Kenya, Djibouti and Sudan, including a new link along the Dijbouti Railway to the Djibouti landing point. Once again, substantial new build is required, and Dark Fibre Africa and Liquid Telecom are both rumoured to be interested in investing following liberalisation.

In 2010, Ethio Telecom selected Gilat to provide satellite backhaul on remote sites, and other satellite backhaul providers are keen to enter the market to support rollout of network extension programmes and achieve whatever coverage targets the regulator sets.

Power in Ethiopia

It is not clear what proportion of Ethiopia’s cell sites are on good grid, on unreliable grid connections and off-grid, but grid stability is no better than in much of Sub-Saharan Africa. According to the CIA World Factbook, only 85% of the country’s urban population has access to electricity, and that dips to 27% in rural areas. Poor access to power in rural areas suggests that the initial, urban-centric phases of new tower rollout could to some extent rely on grid connections, but later stages will require many sites which are off-grid.

A number of large renewables projects are underway, including solar farms and The Grand Ethiopian Renaissance Dam. This megadam will more than double Ethiopia’s annual generation from 11 GWh to 27 GWh, however political wrangling by Nile states, a delayed completion date, and an unreliable transmission network will mean telcos will be unable to rely on this new power source for many years to come.

Power outages began to increase in length and frequency in 2019, but the two state run companies that generate and distribute energy have not explained the cause, nor any remedial action plans.

To manage its power issues, Ethio Telecom issued a 1,300-2,500 site ESCO RFP in 2019. From TowerXchange’s understanding it is not only management of power that is the problem, but the financing of new energy equipment and professionalisation of site management which is required. Firstly, due to exchange rate controls, Ethio Telecom has difficulty accessing hard currency, and is actively seeking ESCO partners that can help it finance its energy capex. Secondly, sites have been erected and maintained to extremely variable standards, with old generators and air-conditioning units simply disconnected and left on site. It was also suggested that past dangerous power installs would require remedial work.

We also learned that availability is very low in Ethiopia, with Ethio Telecom’s RFP for an ESCO partner proposing a Service Level Agreement (SLA) availability of just 96%, which would be a significant improvement on current standards. Some sources suggested uptime was as low as 82% on many Ethiopian sites. At the moment, in-house teams at Ethio Telecom manage its cell site power, but that will soon change. It is likely more than one ESCO partner will be selected to work with Ethio Telecom due to the size of the contract.

Ethiopia Birr versus the US Dollar

ethiopia-currency.png

Forex risk

The risk of the continued devaluation of Ethiopia’s currency is the elephant in the room in terms of investment in Ethiopian communications infrastructure. The Ethiopian Central Banks runs the Ethiopian Birr on a managed float, gradually devaluing it. Although Ethiopia’s broader economy has grown strongly over the past decade, exports have not been as strong, causing a gradual devaluation and a depletion of foreign currency reserves.

However, thanks to improved infrastructure, a booming agricultural sector, and strategic investments in the garment and automotive industries, the situation may improve in the medium term. But to start with it will be easier to put money into Ethiopia than get it out, and only those with a long-term perspective will reap the rewards of their patience.

Local partners

Key local partners in the banking system will be required to access hard currency. Those requiring to move money into and out of Ethiopia will be required to find a local bank to manage their account, and apply for access to foreign currency. Telecoms is a privileged sector, but there is no free convertibility as in developed markets.

Similarly there are restrictions on shipping to Ethiopia, which has to be managed by the Ethiopian Shipping Line and paid for by a local customer. This reduces the flexibility of importing power equipment and other passive infrastructure. Internal logistics can be managed in a more conventional manner, but importation will require some changes to business as usual for towercos and MNOs seeking to invest in Ethiopia.

There are also advantages to operating in Ethiopia. The country hosts two established tower manufacturers, one factory which is brand new and capable of producing 1,000s of towers a year at full capacity. Similarly, Ethiopia is producing a large number of engineering graduates and has a sizeable existing civil engineering industry developed over the last decade of infrastructure development.

Ethiopia mobile market timeline

timeline-ethiopia.png

What is the liberalisation timeline?

Announced in drip-fed statements that revealed the gradual formation of policy, a firmer view of the timeline for the liberalisation of the Ethiopian telecom market has now formed. While Prime Minister Abiy Ahmed hoped to have the liberalisation of telecommunications complete by the end of 2019, reality dictates it will take longer. But one step at a time the government has moved towards a market opening, and the initiative has now achieved ‘launch velocity’ such that it is unlikely to be reversed.

At the start of July 2019, the government proclamation on telecoms regulation became law. In September, we learned that regulator-designate Belcha Reba had been officially appointed as Director General of the Ethiopian Communications Authority. This is a crucial step towards issuing new licences in a country which previously lacked an independent telecoms regulator to issue such licences.

It was originally mooted that just one licence would be issued, but it has now been confirmed that two greenfield unified licences for mobile network operators will be issued. The final timeline remains unclear, with the regulator preferring a more rapid timeline, and the Prime Minister recently suggesting timelines may be pushed back until July. With elections due in May, it is possible that the competition will be complete by the end of March, or be subject to a longer delay

A consultation on the liberalisation was launched in October to conclude in November, which should resolve timings. Roland Berger, an advisory firm which ran the licence process in Myanmar, is doing the same in Ethiopia, and so a similar two part process is likely. TowerXchange understands that Nicholas Teisseyre who ran the process for Myanmar is running the project for Ethiopia. This again points towards the licence process finding a consensus on the need to reach an advanced stage by March or April next year.

Likewise, the privatisation of Ethio Telecom has been agreed, although we learned in our Focus Group that the final form of the privatisation remains unclear. We still expect to see 49% of Ethio Telecom sold to an international mobile operator, but a complete sale or majority sale can no longer be ruled out. Ethio Telecom remains a major employer and conduit for development policy, it employs 22,000 Ethiopians and enjoys privileged access to hard currency, so the final privatisation plan will only emerge after much wrangling.

Following advice taken from the World Bank and some industry insiders, it is likely that Ethio Telelcom will be split into a Serviceco and an Infraco to encourage the sharing of towers and other infrastructure. Were this to take place international towercos would show considerable interest in acquiring Ethio Telecoms’s assets. However, final plans for the assets remain unclear, as does to what degree all of Ethio Telecom’s infrastructure will be rendered shareable. TowerXchange understands that draft regulations have marked Ethio Telecom as a “dominant player” with duties to share its passive and active infrastructure, but a final determination has yet to be made.

Who are the major players?

The names of those mobile network operators most interested in Ethiopia is well known. Who takes the licences and who wins the ownership of Ethio Telecom remains to be seen.

MTN would be on anyone’s list for prospective Ethiopia licence holder; active in Sudan and South Sudan across the border already, proficient in both large and operationally challenging markets and well established across Africa.

A Vodacom/Safaricom Special Purpose Vehicle is also in the running. Abiy Ahmed is said to be particularly enamoured with Safaricom’s M-Pesa and the two have had an SPV established for such joint bids.

Other bidders could include Orange, which is the only international MNO to have experience in Ethiopia thanks to France Télécom helping establish Ethio Telecom from 2010 to 2013; and Etisalat, whose Emirati owners have had a long economic interest in Ethiopia and other states of the Horn of Africa.

Both MTN and Vodacom have made it clear they would prefer to work with towercos to rollout their networks rather than deploy their own capital into passive infrastructure, while Ethio Telecom’s ESCO RFP shows their appetite for outsource some or all of their network operations management, perhaps extending to towers. TowerXchange also understands that Helios Towers, IHS Towers and American Tower have all expressed interest in Ethiopia. CREI, an ESCO owned by managed service provider i-eng, has responded to Ethio Telecoms ESCO RFP, alongside IPT PowerTech, Voltalia and Energy Vision.

Conclusion

While predictions are always dangerous, TowerXchange feels confident Ethiopia represents the most compelling greenfield opportunity for the tower industry (rivalled only by the Philippines). Every African country has unique culture and so there will be no “cut and paste” from an existing market, but participants at the TowerXchange Ethiopia Focus Group felt confident that the lessons learned elsewhere in Africa meant that the supply chain would be able to deliver in Ethiopia too.

There is always the chance of political unrest or upheaval upending the table in the developing world, but the direction of travel is clear. Ethiopia may well be the best thing to happen to the tower industry since Myanmar’s liberalisation five years ago.

TowerXchange would conservatively estimate the opportunity for passive telecom infrastructure investment in Ethiopia in 2020 at a quarter of a billion dollars. With ongoing investment at a similar level for a number of years.


What Myanmar can teach you about Ethiopia

Parallels There are obvious parallels between Ethiopia and Myanmar, which was the last monopolistic telecom market to be liberalised, enabling two international MNOs to leverage towerco partnerships to deploy networks and transform the economy. Phil Cooper, ex-CFO of Apollo Towers in Myanmar shared his thoughts in Johannesburg on what lessons Myanmar has for those seeking to enter Ethiopia. 

Liberalising Myanmar’s telecom liberalisation story began in 2012 when a foreign investment law eliminated many restrictions and paved the way for international companies to enter the country. A new Telecommunications Act followed in 2013, heralding a fiercely competitive process where 91 companies bid for two MNO licenses, Telenor and Ooredoo emerging victorious, with their licenses awarded in 2014. In the subsequent five years, the number of mobile subscribers increased by 743% (from a notably lower base than Ethiopia), and teledensity rose from 13% to 102.6%. The number of towers in Myanmar increased from ~1,500 in 2014 to 17,144 today, 11,244 (65.6%) of which have been built by towercos, with deployed fibre increasing from 7,600km to 68,000km.

Timings Ethiopia is almost twice the size, and the population is almost double that of Myanmar, but the similarities in process mean it is a worthwhile comparator. For example, advisory firm Roland Berger are fulfilling the same role in Ethiopia advising on the licence process as they did in Myanmar. Similarly, as in Myanmar, timetables have slipped. A licence process which was originally mooted to be complete by September 2019, will likely be complete by July 2020. Phil Cooper suggested taking all anticipated deadlines, and adding 50% to them. 

Education Although Ethiopia’s telecoms sector is much more advanced than Myanmar’s was, education will remain key. Establishing a trade association for the towercos should be an early priority. Just because the liberalisation will be big news to the industry and is important to the Prime Minister, it does not follow that information or processes will cascade down to ground level. A trade association can help smooth the permitting process and educate local bureaucrats – turning them into enablers, not blockers.

Partnerships At TowerXchange’s Focus Group, Phil Cooper also emphasised the need to be selective with vendors, and to work closely with those you do select. You are going to be in this together facing a number of challenges which will require quick thinking, flexibility and follow through. Similarly, working with multi-lateral agencies like the IFC, the World Bank or regional development bodies helped manage political risk in Myanmar as it was keen to burnish its global reputation, and the same will be true in Ethiopia. Finding contacts in central government that have bought in to the project will also help.

Land Site acquisition in cities and later in rural areas will also be an essential part of the process and difficult. Land ownership and permitting processes were murky in Myanmar, and will be little clearer in Ethiopia where the government retains final ownership of the land even if leases will be signed with local landlords.

Locals Engaging local talent was a huge help in Myanmar, as was engaging with expats returning home, well trained and educated from their time in the West. A similar, if larger, expatriate community is available to draw on for Ethiopia. Unlike Myanmar though, Ethiopia has a domestic manufacturing base, which can supply steel structures, some quality civil engineering and domestic universities providing engineering graduates to work with – all of which will make Ethiopian telecom pioneers’ lives easier than it was in Myanmar.


 

Gift this article