Good morning, everybody, and welcome to the MTN Group annual results presentation. My name is Thato Motlanthe, I'm the Head of Investor Relations for the MTN Group. Today, it's my pleasure to welcome members of the media, investors, analysts, members of the Group Exco team, and MTNers across our operations who've dialed in via various platforms and channels. Your time and attention is appreciated this morning. The results presentation will get underway shortly. Before it does, I'll just go through the order of events and a couple of housekeeping points for you to take note of. The running order of the day will be as follows. First of all, I'll welcome our Group President and CEO, Rob Shuter, to the stage. He's going to take us through a strategic overview, a strategic and operational overview, I beg your pardon, of 2019.
I'll then call on Ralph Mupita, who's our Group CFO. He'll give us an overview of the financial review of the past year. Rob will then return to the stage to give us his views on prospects and the guidance, along with important matters and priorities for the year. After that, we'll open up for questions and answers that you can put to Rob and Ralph. When the time comes, we just ask that you raise your hand. One of our support staff will bring you a microphone. Before you ask your question, if you could just state your name and the organization that you represent. At this point, I'd just like to ask that the media reserve their questions for the media briefing, which will happen afterwards at the Co-Lab. That will happen between 11:00 and 11:30, where you can ask your questions.
Once this presentation concludes, we'll invite you to the canteen, where you can enjoy some refreshments. For those of you who'll be tweeting during the session, it should be up on the screen. The hashtag is #MTNResults2019. Our Twitter handle is @MTNGroup. For those of us physically present, for your convenience, we've got a Wi-Fi code, which is ATYELLOEVENT, again, on the screen. The password for that is mtnresults, all in small caps. I'm not the safety officer, but I am obliged this morning to point out the safety exits, which are to my right and up the stairs at the top there. The bathrooms can be found on this floor, which is level -1, and upstairs on level 0. Finally, before we begin, I kindly request that you double-check that all your devices are set on silent.
It's at this point that I welcome our Group CEO, Rob Shuter, to the stage.
Good morning, everybody. A big welcome from my side. Thanks, everybody who's joined us in person here at the headquarters. It's great to see so many people. There are a couple of seats in the front for those that want to come down from the top. I guess also a big welcome to those that are joining us electronically, those that are conferencing elsewhere. It's a pleasure to have you here as well. To kick off, I thought just a quick snapshot of the financial results. Ralph will unpack these in more detail. I think overall, we're actually really pleased with the strong financial performance. Top left, you see service revenue growth in constant currency, 9.8% for the year. We've got guidance of double digits. We actually were at 10% for the second half. Improving momentum and I think pretty close to our guidance range.
I think you drop one level below that. EBITDA grew 13.6%. That is EBITDA on the same accounting basis, stripping out the one-off, that's really an underlying profit growth. Obviously, one to the right means that we opened up the EBITDA margin again in the year, 1.2% on a constant accounting basis. That's dropping through to headline earnings. Adjusted EPS up to short of 23%, EPS without adjustment around 62%. Really revenue dropping through to EBITDA, dropping through to EPS growth. We put ZAR 26.3 billion worth of investments into our networks in 2019, hitting all of our targets for rollout. Of course, because the absolute level is relatively flat, the revenue is growing. We are moderating also our CapEx intensity. Encouraging also, the holdco net debt has moderated in the period, down from ZAR 2.3 billion- ZAR 2.2 billion. We are absolutely maintaining our dividend policy.
That'll then be ZAR 5.50 for the full year, of which you have ZAR 3.55 for the second half. Pretty much I think everything we were aiming for in the year financially we've managed to deliver, which I think is good news. If we stack that up against our medium-term guidance, so there's no shortage of guidance at MTN. For sure, we were just a little bit short on group service revenue, but we've given it a half orange, half a green because the second half was strong. Very strong performance in MTN Nigeria. South Africa didn't quite get there, so that was 0.4% for the year. Obviously, a lot of pressures in the South African market. Fortunately, there's enough compensating growth in the group to deliver the group numbers. Everything down below, as I've said, EBITDA margin improved, CapEx intensity improved.
Our ARP, I'll cover a little bit later. That's going really well. All of the other metrics are going in the right direction. If we look at some of the underlying performance, I think one of the hallmarks of 2019 was really increased momentum in H2. We did discuss that at the half-year results. I picked out a few pieces of it already. H2 service revenue at 10%. What were the underlying drivers? Take a look at Nigeria block to the right. 12.2 H1, 13 H2. Where's the big recovery in Nigeria? Data growth in H2, up at just short of 53%. We secured the 800 spectrum, rolled out extra 4G, and that's really coming through in the numbers. MTN South Africa. Yes, there are challenges, but again, look at the two big metrics on the slide.
We've been working on our enterprise turnaround the last three years. We finally turned positive in H2. You see the 0.7. Well done to Godfrey, Wonder, and the team for delivering on that, and obviously expecting growth now in enterprise in 2020. Prepaid, which has been a real challenge, I think, across the industry. You see some of the pressures there, -5.5 in H1, actually moderating to -2.6 in H2 as we start to see elasticity and customer growth coming through, compensating for some of the challenges there. One of the other things we spoke about was the WACA portfolio. You'll remember that we had challenges, particularly in Cameroon and Ivory Coast. We were negative last year. We were just up +0.5 at the half, as you see, and actually in H2, 4.8 in the WACA portfolio.
Remember, average inflation in those markets is around 1%, actually that's almost 4%-5% of real growth in West Africa, which I think also a testament to the team for all the hard work. Across the bottom, a lot of this I'll cover in later slides, in general, you see ARP going well, good subscriber growth, good scale in fintech, and a lot of other of our key stakeholder reputation management issues have been addressed. Obviously, to have this improved momentum, you also need a lot of hard work over the years. We thought we'd just flash up for you a couple of pointers of what we see the progress over the last couple of years. Subscriber growth, so critical for MTN. Our markets are characterized still by population growth, relatively low levels of people penetration.
We are still succeeding in winning subscriber market share, as you see on the right. This is what gives us a lot of confidence also for movement and growth in voice revenues. You look at it on the left there, overall subscribers now 251 million, so we've added around 35 million over the last two years. You see a similar progress in fintech, similar progress in our MoMo customers. The market share is an interesting statistic. It is subscriber market share that we measure based on bidirectional traffic on our own networks, because we know who calls our customers and who our customers call. We build a market model across all the markets, and basically, you see there actually really strong growth in real subscriber market share over the last two years. The financial performance, you know what we'll be targeting.
We wanted to get service revenue up into the double digits. We started at 7.2, so we've now got two years, 10.7, 9.8, and of course, the improvement in EBITDA margin has come through now for the third year in a row. Not to forget about the life of the customer. One of our key metrics always has been to open up NPS leadership across the markets. When we started in 2017, we had three markets with NPS leads, and at the count of 2019, that had improved to 12. Obviously 12 is a lot better than three, but we've got 21 countries, so clearly there's still work to be done. Data, one of the reasons why we can harvest and monetize the appetite for data across the markets is that we invested ahead of the curve in terms of rolling out data coverage.
You see that again on the slide. Data coverage being combination 3G, 4G from 375 million covered across our markets in 2017. We're up now at 480. That's going to go well up into the five and the 600s in the next year or so. Of course, most importantly, to stress at the end of the slide that never you can make progress like this without committed, energized, enthusiastic people. One of the key ways we measure that is sustainable engagement in our GCA scores. You see from 75 up till 80 the last two years. That's the highest score that MTN Group has had in its history. It means that much as people are working really hard, I think that they believe in the story and some of them are smiling and nodding at me.
I think they're nodding at the working hard bit, but anyway. Let's have a quick look across the two largest markets in the three regions. Ralph's going to give you a lot of information on the financial performance. I'll cover a little bit more what's happening at a macro level. South Africa, I guess we all know that the macroeconomic environment is pretty weak. Recession in Q4, a lot of pressures going into 2020. We have a difficult macro. We also have some sector-specific issues. You see that at the layer below there. What we saw in 2019 was a change in how the end-user regulations worked for data consumption in prepaid. This put a lot of pressure on the results, which slowly starts to fade as customers get more used to buying bundles rather than running on at a rate.
That has been, I think, a big impact in the year. Of course, Cell C has been a really strong national roaming partner on the MTN network, but they've had their own challenges, and that's affected some of the accounting. Of course, we've had also the Competition Commission inquiry over this last while. With that, is market context. What the team have been working on and the results they've delivered to continue our efforts on price transformation, making data more affordable for our customers. Always recognizing, though, that one of the big limitations is to carry more data cost-effectively on a network, you really need more spectrum, and that remains still a challenge in the South African market. We've seen good growth in active data subscribers and also significant increase in data traffic.
A lot of work in the voice business, so that's customer value management, it's personalized pricing, it's distribution. As Ralph will show you a little bit later, we actually have a stronger voice performance in H2, which is encouraging. I spoke a little bit about enterprise. Complex business. You've got ICT services, you've got mobile services, you've got a complex distribution infrastructure, direct, indirect, working also through partners. It's not an easy thing to turn around a mobile-centric enterprise business, and hats off to the team for the work and firmly back into growth. From a CapEx perspective, I think you're well aware that our ambition is to have the best network in the country. That means that we can play a very strong challenger number two role. We put ZAR 7.6 billion into the network in 2019.
Every single benchmark score says that we have the best network in the country, some of them three times in the same year. Our internal metrics show that the gap actually between the number one and the number two is widening. Our network is getting even better. CapEx has been moderating, but leveraging a lot of the historic investment, smart CapEx. We've got a big modernization program running in South Africa, and I think that's really been a great result. If we turn to Nigeria. Nigeria, very strong performance in the year. Well done to Ferdi and the team. Ferdi is here. Where's Ferdi? Over there. He's here for the difficult questions. To celebrate the contribution of Nigeria to the group and the strong performance, we've taken our Nigerian outfits out. For those that haven't seen it before, it's not some coronavirus thing that's going on.
Market context in Nigeria, also a difficult macro. GDP growth, higher inflation. Sector-specific , we had a big reregistration of subscribers in the year. That has a couple of effects. One is a lot of the infrastructure you use to bring new customers on board gets used up in that reregistration process. Finally, if there's a base that you're not able to reregister, then they have to be disconnected at the end of the process. That had an effect, particularly, I think in H1. A lot of competition in the data space. Of course, we've had this very positive resolution of our dispute with the Attorney General. With that as backdrop, what have the teams been busy with? Firstly, rollout and 4G. It really is a characteristic now of MTN Nigeria. You see LTE really scaling in that market.
I remember looking at the split of data traffic a couple of years ago. It was 8% 4G, 92% 3G. What you've seen in the last 12, 18 months is the penetration of LTE devices is starting to move much more quickly through the market, and we're seeing a much bigger proportion of traffic moving on to LTE. We were delighted to secure the 800 spectrum. We've accelerated our rollout. You can see we've lifted our pop coverage now to 44%. A lot of work also on the device portfolio. A lot of the time in a market like Nigeria, we've got to replace customer SIMs because they've got old 2G SIMs that we need to replace for 4G. In the end, added almost 3 million active data customers in Q4 alone. I think you're really seeing momentum coming through there.
Still decent growth in subscribers, even after going through the SIM reg. Actually, across all the markets, we believe we've now completed our VAS optimization process, and we're expecting these revenues to move back into growth. In fact, MTN Nigeria had digital revenues and growth towards the back end of the year. Very quickly on the three regions. These are the results of our consolidated opcos. Obviously, there are some markets, for example, Irancell and Associate, Eswatini's and Associate, they don't form their way into these numbers. Very quickly, if you look at it, SEAGHA, very strong performance, 21.7% service revenue. Big driver there was strong results from MTN Ghana, which was up almost 23%. Really, hats off to Selorm and the team. On the MENA side, almost 20% as well, obviously under very difficult circumstances.
I touched already on WACA, 2.9 for the full year, just under five for the half. Very strong real revenue growth in the WACA portfolio. I think what you see in the row below there is that these regions are also very significant contributors to the subscriber growth across the market. You see those numbers reflected underneath there. I guess a quick segue then into the strategy of the group. We launched in 2019 the digital operator framework. What the digital operator framework is saying is that a key part of the MTN strategy is what we call the evolving telco on the left. That is where we take a business that was predominantly a voice-centric , consumer centric business. We repurpose our networks for mobile internet, so we can harvest the data move across the markets.
We also make sure that we've got very deliberate strategies to push our services and our capacity into the enterprise and wholesale markets. Voice to data to enterprise to wholesale. That evolving telco story basically allows us to access four very large pools of revenue to help secure our growth prospects going forward. MTN has always said that our ambition is not confined to the evolving telco, and we want to be a scale player in financial services at the bottom, and we want to be a scale player in digital services, which means that we want to build and own our own OTT services across our markets. MTN is really operating at the intersection of these three strategies.
What is critical for us is that our big advantage in rolling out that strategy is really recorded at the bottom there, that we have one network. We have owner economics for large networks across the markets, which can be leveraged for all of those areas. We have an extensive informal distribution system that is largely built historically for prepaid airtime distribution, but is being repurposed for fintech and digital. We are connecting and registering in scale customers every month. Nigeria alone is more than 2 million gross adds per month. We can register them for GSM services. We can also register them for fintech and digital services. We're very much still passionate about the strategy, and it is that strategy that then allows us to access the six revenue pools, four of evolving telco plus digital and fintech.
Let's have a quick look at how we're doing across those. The first slide is voice and data. Voice revenue it's ZAR 82 billion of the group's total. It's around 55% still contributing to the group. It's a very significant pool of revenue, which is a good thing. Of course, it also means that we have almost diversified now 50% of the group away from voice into the other curves, which I think is also positive. What is very much a key factor for MTN is we believe we can still deliver growth in voice revenue. You see it again in 2019, 4.2%. Why do we so much believe in the growth? Because we've got base growth, we've got subscriber growth. You see 18 million subscribers in one year alone. Remember what I said earlier. We've got growing populations across the markets.
We estimate that we will add 35 million extra people into our region over a three-year period. We've still got relatively low people penetration in some of the large markets. Of course, you saw in an earlier slide that we continue to win subscriber market share. If you put those three things together, we do believe that subscriber growth will be a factor of this business going forward. We still see usage per subscriber of voice minutes, 5% increase in minutes of use per subscriber across the portfolio. A lot of our hard work in CVM is driving that growth, and also a big focus going forward into the youth proposition. Remember, again, these markets are 60% of the population below the age of 24. A big proportion of that is in their teens.
It is very critical for an operator like ourselves that we position our strategy around the demographics of the market, and we have to compete strongly in the youth sector. On the right-hand side, you see data. Clearly a scale business, ZAR 35 billion worth of data revenues across the market. We've had 22% growth in that revenue in the year. Strong growth in active data subscribers, almost 17 million up to 95. As I said earlier, it is really a function of supply and demand. In the data business, supply is rolling out coverage 3G, 4G across the markets, and demand is smartphone adoption, the way your price plans work, the way your distribution works. It always needs to be very carefully managed.
You see that we continue to work on price transformation, so making the cost of data more affordable, and with a roughly 34% devolution in average price per megabyte over the year. Against that, you see 46% growth in traffic. Very importantly, that yes, we have 95 million active data customers today. It is still less than 40% of the 251 million total customers. We remain still in phase 1 of data adoption across these markets. Usage is relatively low. Pricing is already low. We still are really believers in our ability to monetize this data curve. Second is enterprise and wholesale. These are very significant businesses for the group. If you add the revenue contribution from the two together, it's almost ZAR 20 billion, roughly twice the size today of our fintech business. You see a 63% growth in wholesale.
It's where we leverage our infrastructure and sell services and capacity to other operators and OTT players. On the enterprise side, grew 8.4%. Sitting within that is still a shrinkage in South Africa for the full year. Actually, if you look at some of the large markets for enterprise like Nigeria, we had 22% growth in enterprise. We will see an improving trend in EBU as the South African recovery takes hold. A lot of what we need to do there, you see on the slide. For wholesale, we built interconnect hubs for services like A2P, which is an important service across the market. Enterprise partnerships, how we sell the products, signing the wholesale national roaming agreements. Very excited about the progress here. Finally, on digital and fintech. The digital business on the left, we've been very busy with VAS optimization these last years.
A lot of the pool of revenue there has been for what we would call legacy services, basic feature phone, caller ring back tone kind of services. In the optimization of that business, we've seen a significant decline of revenue. We sit now at ZAR 2.4 billion across the Group, so less than 2%. We are now, I think, in a situation where we can rebuild the right kind of revenues, the right kind of services. We expect that to go back into growth in the year ahead. Part of it also is going to be driven by our OTT strategy with our key apps like Ayoba and MusicTime. On the fintech side, 27% growth in revenue. Customers grew by 7.5 million, also around 27%. Just wanted to give you a feel for the scale of the business. 35 million customers, 16 countries.
We transacted ZAR 96 billion worth of transactions across the platform in one year. We processed 9,200 transactions a minute. Long as I've been boring you in the last 10 minutes, think of how many transactions went through the MTN mobile money platform. It is an industrial-strength scale business. I think what's super encouraging, though, is that, yes, we are going to scale the base and bring more customers on at these kind of ARPUs, but we are absolutely committed to expanding the product set. What you see at the bottom is about building a fintech ecosystem that is also going to increase over time the ARPU per customer at the same time that the customer numbers are growing. It's about insurance, it's about lending, it's about marketplace, e-commerce, deposits, and some really exciting things going on there.
Big shout-out to both Serigne and David for all the work they've done in 2019. I've given you a sense of some of the traction in the fintech business, but if we go back one year before to 2016, actually, that customer base has more than doubled in the last three years. Agent network's gone from 133,000 to more than half a million. That's really an important factor going forward. Of course, we're building a valuable financial services business on its own, but it also has a major benefit for our core GSM business. If you look at the chart on the right, what it's showing you is that in 2019, we saved almost $50 million of airtime commissions, because our customers are buying the airtime on the Mobile Money platform rather than through indirect distribution.
On the asset realization program, this has been a big feature of our strategy in the last year or so. We announced that we would target ZAR 15 billion of proceeds over 3 years, excluding the IHS investment. That was actually standing here a year ago. What have we done? We've delivered ZAR 14 billion actually, in the first 12 months. I think that's been a really successful program.
Of what's remaining, we now indicate for you on the slide that the assets that have been identified as either not long-term strategic, which is mainly e-commerce and the tower companies, or where we've committed ourselves to further broadening local ownership, like in Nigeria, our stated intention to move from 78% to 65%, that the aggregate value of those investments is around ZAR 45 billion fair value today, and we are targeting to realize at least ZAR 25 billion of that over the next medium term, three to five years. The ZAR 25 billion target is on top of the ZAR 14 billion that's been realized in the last 12 months. Why are we doing it? Very important. Simplify the portfolio, reduce risk, improve returns. Of course, also it's going to be very valuable to bring down our leverage ratios for the future.
I will talk to the material items impacting reported results. There are a couple that do create a difference between what you see as reported results and the analysis that we will provide a little bit later. I'll cover that. Secondly, I will cover the two largest markets in a bit more detail, building onto some of the points that Rob has raised, covering Nigeria and South Africa. I'll come back to the income statement, touch on a key few line items, service revenue and EBITDA. I'll go back to the balance sheet, cash flow, look at headline earnings analysis, and finish off with returns. In the commentary that I will provide, I will do mostly comparisons on an IAS 17 basis. As you well know that we had quite a big standard change in 2019, bringing onto IFRS 16.
For those of you who will be following the commentary, most of my points will be on an IAS 17 basis. Let us have a look at the material that impacted our reported results. There were three major impacts to our reported results, and I'll talk to the first one being currency movements. We saw a much weaker average rand and a stable naira in the period, and that led to actually reported results being higher than constant currency growth rates. We also had the negative impact coming through from the depreciation we saw in the Iranian rial. That meant the share of contribution from Iran in our associates and JVs was much lower than it could have been, given the currency depreciation.
We also saw a stronger closing ZAR at the end of the year against the US dollar. That impacted for sure the balance sheet items and the holdco leverage. We did see pleasingly that the share of US dollar to ZAR debt at the holdco level has reduced now to 50%. Many of you will remember several reporting periods ago when that number was much closer t 60. Now it's improved to 50. We did have foreign exchange losses incurred during the period. These were to the tune of ZAR 2.4 billion. The second material impact is the adoption of IFRS 16. On the income statement, we do see that reported EBITDA is ZAR 9.2 billion higher, as lease payments that would have previously gone through operating expenses are capitalized.
We now have a right-of-use asset, depreciation below EBITDA, as well as interest expense for the lease liabilities coming through in finance cost. Overall, we saw a 13.4% decrease in headline earnings as a result of the adoption of the new standard. Looking at the balance sheet effect, we recognize the right of use asset of ZAR 45 billion and lease liabilities of ZAR 46 billion. When you look at that composition, 43% of that total is Nigeria and 31% is South Africa. Lastly, the recognition of Mobile Money balances. Now, the group now recognize all Mobile Money balances held by the respective banks and the customers' rights to these balances as an obligation to repay the balances to the customer. The right to claim the corresponding amounts from the relevant banks is recorded as a current financial asset.
What you'll see in 2019 that we recognize across the markets totals amounting to ZAR 15 billion. Let us start by looking again at the selling points on the group income statement. I think you will be familiar with these views. On the right-hand side of the charts, we show the growth rates of some of the income statement lines on both an IAS 17 basis as well as on constant currency view. What do we see here? We had solid constant currency service revenue growth of 9.8% in the year, led by Nigeria, which grew 12.6%. Ghana was up a very pleasing 22.9%. South Africa was up 0.4% in a tough macro, basically impacted by the cash basis of accounting that we took on Cell C, as well as the tough macro, particularly in our consumer prepaid business.
The weaker average ZAR rate resulted in higher reported service revenue growth of 13.1%. When we exclude once-off items, reported EBITDA increased by 14.6% and increased by 13.6% on a constant currency basis, showing that overall we are getting operational leverage coming through. The increase you see in depreciation, amortization, and goodwill impairment was largely driven by increased prior period network CapEx additions. We also brought through an impairment of ZAR 342 million for MEIH, our Middle East e-commerce investment. As you move down the income statement, you can see noticeable changes on the JV and associates line. I'll provide a bit more detail of this later in my presentation. Further down, you can see a noticeable change on the income tax expense line. The group effective tax rate for the year was 39.3%.
The overall tax result achieved was largely as a function of the higher withholding taxes due to the increased cash upstreaming in the period. We had non-deductible expenses in Sudan. There were deferred tax assets not recognized, and we also had a non-taxable gain on the dilution of Jumia when that business was listed in April 2019. There was also a material movement in non-controlling interest that you'll see there. This was a result of the change in our shareholding in Ghana post the IPO in 2018. As you look further down, you'll see that adjusted headline earnings per share, which is our measure of the underlying operational momentum growth in our earnings, and that increased by 22.6% in the period. As Rob mentioned, the board has declared a final dividend of ZAR 3.55 in the year. This brings the total dividend for the year to ZAR 5.50.
I'll unpack the other elements of the income statement a little bit later. Let us have a look at the financial performance of the two major operations, starting with South Africa. The chart illustrates trends in service revenue, expenses, EBITDA, and CapEx in the period under review. Service revenue was up 0.4%, which was well below our medium-term guidance of mid-single digits. The performance of South Africa was mainly impacted by a decline in consumer prepaid revenues, which were down 4.1% and mostly impacted by the implementation of the ICASA new data rules as well as the reduction in out-of-bundle tariffs. The South African consumer is also under considerable financial pressures. In addition, South Africa did not recognize ZAR 283 million of Cell C revenues as we continued to apply the cash basis of accounting, given the liquidity challenges that that business faced in 2019.
If we look at the revenue bearer level, what do we see? Outgoing voice revenue was down 7.1% year-on-year. This was driven by the reductions in the effective tariff, which declined by 6.5% whilst the traffic remained stable. It was encouraging to see that prepaid voice revenues began to show trends of stabilization in the second half of the year. Data revenue grew by 5.2%, driven principally by the increase in the active data subscribers, which now total 14.1 million. The effective rate dropped by 16%, while usage was up 25%, where we now have 1.9 GB per active data as what we saw as usage in the year. The core digital revenue declined 31.4%, and as Rob said, we have now completed the VAS optimization work in the market. The fintech revenue, which comprised of airtime lending fees, were increased by 11.2% in the period.
In our wholesale business, we lost the Telkom national roaming contract at the end of the first half of last year. As mentioned earlier, given Cell C's liquidity issues, we continued the cash basis of accounting. We received total cash of two and a half billion ZAR from Cell C, of which ZAR 1.7 billion was received in the second half of the year. While still looking at service revenue, what you see here is a bridge view of the contribution of each of the segments within the South African business. What is encouraging is that the consumer prepaid business showed an improvement in momentum in the second half of the year. While the decline in service revenue reduced to -2.6%, compared to -5.6% in the first half of the year, this was due to seeing the price elasticity starting to come through in the data revenue growth.
The consumer postpaid business delivered a service revenue growth of 3.1%, which slowed due to deteriorating economic conditions, muted additions due to stricter vetting rules aimed at reducing the credit risk in the current challenging environment. The wholesale business grew 42.8%, and that was driven by the national roaming deals we have with Cell C and Telkom in the period. We're very pleased with the stabilization and improvement of the enterprise business, which was evidenced in a strong recovery in the fourth quarter, where year-on-year growth returned to growth, as Rob showed earlier. This helped to drive a reduction in the decline of the enterprise service revenue to -3.7% for the year. You will remember that at the end of 2018, that was a -11.3%. Well done to Godfrey and Wonder for the recovery we're seeing in that enterprise business. We saw a stabilization of churn.
We saw additions of new corporate customers and continued growth in the ICT business. Back to expenses, margins, and CapEx in South Africa. What do we see here? You can see that cost of sales were down by 0.9%. This was driven by good control of channel and SIM costs, lower commissions paid, and the benefit to interconnect costs of lower MTRs. Device cost of sales ticked up slightly, driven by the increased distribution of 4G devices. 4G devices sold were up by 28% year-on-year. OpEx was up 10.8% due to increased network operating expenses, staff costs, and bad debt provisions taken in the period. We had additional expenses for generator fuel and security as a result of the load-shedding experience in the period. These additional expenses added one percentage point to the total OpEx growth.
The EBITDA margin declined 1.5% to 33.6%, and these were driven predominantly by the new ICASA subscriber rules and our out-of-bundle tariff reductions. In combination, these impacted the EBITDA margin by 2.2 percentage points. CapEx intensity in the period improved to 16.6%. We completed our full-year rollout with a CapEx budget of 7.6%, and this was well supported by price book savings that we got and smart CapEx initiatives. For 2020, we have planned CapEx expenditure in South Africa of ZAR 8 billion. Moving on to Nigeria. We recorded a very solid 12.6% growth in service revenue. In the second half of the year, the growth was 13%, as Rob mentioned, and in the fourth quarter, it was up to 14.2%. Good acceleration coming through, and I think we've seen that's continued into the early part of this year. The core voice market performed well, growing at 8.3%.
The growth in the voice revenue was supported by base growth, relatively stable tariffs, and our CVM activities to stimulate voice usage. Solid data revenue grew by 42.7%. This was supported by the acceleration that you saw earlier in Rob's slides. This was driven by a strong increase in active data subscribers, which now in Nigeria total 25.1 million subscribers. The VAS optimization program, which again has now been concluded in Nigeria, impacted digital revenue growth. We recorded a 23% growth in digital revenue on a sequential quarter-on-quarter basis in the fourth quarter of 2019. We expect to see that growth coming through into 2020. The fintech revenue, which currently comprises of airtime lending and the bank-led mobile money business, had decent growth of 23.2%.
In line with our strategy, we focused on building the fintech agent network, surpassing the milestone of 100,000 registered agents by the end of 2019. In terms of expenses, these were well contained and grew below inflation. Cost of sales growth was driven by higher commissions and distribution costs, which were up 10.2%, and national interconnect, which was up 13%. OpEx increased 9.5%, driven predominantly by network expenses, and the digital revenue share was down just under 75% as we completed the VAS optimization work. We reported an expanded EBITDA margin of 44.8%, which was supported by a stable naira and operating leverage achieved. During the period, we made significant investments in our network to improve service quality and drive 4G expansion. The total CapEx for the period was ZAR 8 billion, and we achieved an improvement in CapEx intensity to 17.2%.
For 2020, we have planned for capital expenditure of approximately ZAR 8.7 billion for MTN Nigeria. Now coming back to the Group income statement. What do we see on service revenue? We see that voice, data, fintech, and wholesale revenues were the key drivers of growth in the period. Voice revenue, which still contributes approximately 58% of total service revenue, grew by 4.2%, driven predominantly by the base growth that Rob spoke about earlier. The Group added 18.2 million subscribers to reach 251 million. We continue to benefit from our CVM activities and our bundle and subscription offers. Data revenue grew 22.4% on the back of network rollout, base growth, increased usage, and managing pricing yields in very competitive markets. Our active data subscriber base grew by 16 million to close the year at 95 million.
The 39.6% decline in digital that you see there was the result of the VAS optimization, which as I've mentioned now twice, we have now concluded. It was encouraging to see that we had positive growth in digital revenues in the fourth quarter of 2019 on a sequential quarter-on-quarter basis. Fintech grew 27%, and that was driven by the base growth of 7.5 million mobile money users. The insurance business now has approximately 6.2 million policies and is expected to contribute to fintech growth in the future as we roll out a broader offering of financial services. The wholesale revenues you see there grew just over 63%, and those were driven predominantly by the national roaming deals in South Africa and the expansion of MTN GlobalConnect, where we signed up key accounts and grew our A2P messaging services through the Y'ello Connect platform. Moving on to EBITDA.
On this slide, you can see the drivers of group EBITDA in absolute terms as well as towards group margin. Overall, EBITDA was up 13.6% on a constant currency basis. As you can see on the chart, the growth was led by the performance of Nigeria, SEA, and the WECA regions, with particularly exceptional growth in Ghana. The turnaround in the WECA region during the period was very encouraging. We had a stronger second half from a service revenue perspective, where we had growth of 4.8% versus the growth that we've seen at the half year of 0.5%. This turnaround in WECA is supporting margin improvement and contribution to the group. The group EBITDA margin improved year-on-year to 35.5%. The gains in Nigeria, SEA, WECA, were offset by lower margins achieved in South Africa. Let us now have a look at finance costs and leverage, starting with finance cost.
The increase in net interest paid was a result of the additional bank debt taken on in Nigeria, which took on to its balance sheet an additional ZAR 8.9 billion. This was in line with its plans to optimize its capital structure. You can see also in the chart, interest expense for lease liabilities of ZAR 5.7 billion were recorded in the period. Net Forex losses were recorded at the head office, driven most notably by ZAR 424 million on the Iran receivable, ZAR 144 million on redemption of the Nigerian preference shares. Forex losses in South Sudan also increased due to a higher rate on settlement of foreign denominated creditor balances. As you can see, the average cost of debt marginally ticked up to 8.7%. This was largely driven by the shift in our debt mix to ZAR and NGN and less USD and EUR. Moving on to leverage.
At 1.3 times, the group leverage benchmarks well with emerging market peers. The increase in the group interest-bearing liabilities is largely driven by the new bank debt in Nigeria that I mentioned earlier on. The holdco leverage improved in the year to 2.2 times. On a pro forma basis where we include the ATC JV's disposal proceeds, holdco leverage would have been at 1.9 times. We had high upstreaming in the period as compared to 2018. Proceeds from our ARP supported the de-leveraging of the holdco, where we now have net debt at ZAR 55 billion, which is the lowest level we've had since 2016. As we look at our share of earnings from associates and JVs, you can see a big swing from 2018, and this was due in part to an improved contribution from Irancell relative to the prior period.
The underlying performance of Irancell was resilient considering the macroeconomic pressures that the market is facing. Service revenue was up 20.1%, supported by good voice growth. The EBITDA margins expanded in the period to 35.8% as data transmission costs were well contained. The e-commerce group results were also impacted by the fact that Jumia was no longer equity accounted as of April 2019, and lower losses were recorded for the period. The high share of results in the period were also impacted by the recommencement of equity accounting for Mascom, which had previously been held for sale on the group balance sheet. The tower companies were classified as non-current assets held for sale from October 2019. As of 10 March 2020, MTN had received in cash approximately ZAR 2.2 billion of proceeds related to Uganda.
The group awaits the finalization of the regulatory approvals for the Ghana transactions, from which proceeds of approximately ZAR 6.1 billion are expected. Let us now have a look at the statement of financial position and what do we see there. Across the group, we have recognized a right of use asset of ZAR 45 billion and ZAR 46 billion in lease liabilities on adoption of the new standard. Included in our other non-current assets is our investment for IHS, which we fair valued at ZAR 27 billion at the end of the period. As detailed in our annual financial statements, the fair value of IHS is calculated using an earnings multiple technique based on adjusted tower industry earnings, multiples of between 10 times to 14 times. This is less the estimated net debt of ZAR 20.2 billion.
In addition, a 10% liquidity discount was applied in determining this fair value. Mobile money balances are now recognized on the balance sheet in line with changes in the accounting policies that we have adopted and applicable for the 2019 year-end position. Total MoMo deposits and payables recognized at the year-end amounted to ZAR 15 billion. Looking further down the chart, there is a decrease in non-current assets held for sale as a result of the recommencement of equity accounting for Mascom. Interest-bearing liabilities increased by 10.9%, mainly as a result of the new bank debt in Nigeria. Other liabilities decreased by 14.8% after we made the final payments of the Nigeria SIM registration fine of ZAR 4.4 billion. Moving on to CapEx. This slide shows further improvement in our CapEx intensity, in line with our medium-term guidance.
We capitalized ZAR 26.3 billion in the year, which was lower than our guidance, still delivered our planned network rollout. We rolled out a total of 5,800 3G sites and 10,900 4G sites. As you can see from the charts, the RAN and transmission accounted for 74% of total CapEx, and we continue to invest in modernizing our IT systems. Although the focus of the CapEx rollout is on 3G and 4G deployments, we're ensuring that in our key markets such as South Africa, future RAN and transmission deployments will make our networks 5G ready. In South Africa, we are progressing with the multi-year modernization of our network, and that will deliver within the business as usual CapEx envelopes. Looking at our holdco net debt profile, there are a couple of points I would like to highlight.
We are seeing an improvement in our debt profile with a share of USD and EUR debt now at 50%. On the left-hand side of the chart, you can see the trend views of group and holdco leverage, which show an improving profile. The good progress we've made with the asset realization has supported the de-leveraging, where we now have the holdco net debt at ZAR 55 billion. As part of our capital management framework, where we have clear actions we are executing on to reduce holdco net debt further, and to ensure that our debt profile is appropriate and resilient to reasonable stress scenarios for currencies and upstreaming risks. Growth in earnings and cash flow remains the most important actions from a management perspective.
This will be supported by executing on this enhanced asset realization target of at least a further ZAR 25 billion over the medium term, as well as managing our CapEx envelopes within approximately ZAR 28 billion over the next couple of years. The rebased progressive dividend policy will support the stabilization of the holding company debt. We continue to expect dividends paid to be covered by cash flows remitted to the group by 2021. Given our enhanced ARP target of a further ZAR 25 billion, Progress we've seen in earnings and cash flow, we are changing our medium-term guidance for holdco leverage to less than 2 times from the current 2 times-2.5 times over the medium term. Moving on to cash flow, let me take you through some of the key drivers from EBITDA to free cash flow development.
In the year, we saw ZAR 5 billion working capital outflows driven by ZAR 922 million from Benin, which made payments to the authorities for frequency fees. The rest were mainly related to payments by Ivory Coast, South Sudan and Benin in relation to settlement of vendor liabilities. In the first half of 2018, the group had benefited from the repatriation of dividends from Iran amounting to ZAR 1.3 billion. No further amounts were repatriated from Iran and will remain with the receivable of ZAR 2.8 billion as of the end of 2019. We had cash CapEx of ZAR 27 billion as we built out our networks and delivered growth in operating cash flow of 60% on an IFRS 16 basis, which is 18% on an IAS 17 basis.
The Nigeria once-off payment included the SIM registration final payment of ZAR 4.4 billion, as well as the CBN, CCI resolution amount of ZAR 731 million. Financing activities of ZAR 6.5 billion were driven by the additional debts in Nigeria. This was also impacted by the capital portion of leases, which is now included in financing activities of ZAR 3.4 billion. On other investments of ZAR 2.4 billion, these were driven principally by proceeds from the disposal of Travelstart Amadeus and the sale of the shareholder loan in the ATC Ghana joint venture. Let us now have a look at the headline earnings analysis. The table provides a reconciliation of attributable earnings per share through to adjusted headline earnings per share that gives more visibility to operational performance. If we start at the top of the table, you'll see that earnings per share are ZAR 5.76 for the year.
Working down the table and adjusting for losses and gains accounted for in earnings, mainly the MEIH impairment and the gain on dilution for Jumia, we end up with a headline earnings per share of ZAR 0.45. On divestitures, we have normalized for asset disposals such as Cyprus and the Ghana reduction in shareholding. Working through all of that, we have an adjusted headline earnings per share of ZAR 6.77. As you see in the chart, adjusted headline earnings per share increased by 22.6%, showing positive operating earnings delivered in very challenging trading conditions. Let me conclude my presentation with a look at the progress we've made against our ROE targets. The chart shows a bridge view of the drivers of our return on equity.
On an IAS 17 basis, we see that return on equity increased from 11.5% to 14.3%, driven by operational earnings growth from the consolidated subsidiaries, with the most significant contributions from Nigeria and Ghana. The notable drag on ROE was a higher group effective tax rate and movements in non-controlling interests driven by the Ghana reduction in shareholding. In the period, there were no material impacts coming through from the asset realization program, which we anticipate in the medium term will contribute to the return profile. This is as we unlock value in the e-commerce portfolio and within our tower investments. On an IFRS 16 basis, the ROE was 12.8%. We remain comfortable with our medium-term guidance of achieving ROE of greater than 20% over the medium term. Ladies and gentlemen, that concludes my financial review. I'll hand you back over to Rob.
Thanks, Rob.
The music. I think just to conclude, a little bit of a look forward. We wanted to start just quickly a bit of context around coronavirus. On the left is more what we see on a macro basis. Obviously, it's going to affect the global economy, impact on the Chinese economy. You see increasingly a lot of noise, a lot of volatility. We've seen a lot of turbulence in capital markets, currency markets the last while. That's the general backdrop. Of course, the question is, well, what could the effect of this be on a multinational operator like MTN? I think we basically see three things. The first is, it will have a short-term commercial impact, I guess simply because people will travel less, which will affect things like roaming revenues.
Of course, also more telecommuting, more working from home, which would have a compensating effect on mobile data usage. In general, we would say moderate impact in the shorter term. I think the more significant impact is how does it affect our reliance on predominantly imported equipment? Because supply chain for us is about network equipment, it's about SIM cards, it's about handsets. There, we're putting a lot of processes in place to manage that. You'll remember that we had some issues with our Chinese vendors a year or two ago, not related to coronavirus, but more to supply chain disruptions. We put a lot of processes in place at that time to make sure we had an inventory of spares, et cetera. I think, in general, we are in good shape to weather this process.
Networks generally are quite resilient, so we think we'll be able to continue to service our customers. I guess the third thing, just longer term, with equity markets being volatile, debt markets being volatile, how could that affect us? I think we're very fortunate now, we don't really have a dependency on the equity markets. It's obviously challenging and disappointing that there's effect of this on our share price, but it's not really that we need that right now for our execution of the strategy. Debt markets, we've got a well-established debt structure. We've got long-term maturities. We believe that as MTN Group, we're well-placed to, in a resilient way, ride our way through this volatility going forward. Of course, also means we'll have more time at home, I guess, and not living on a plane so much anymore, at least for the short while.
Moving off Corona, just to set out again our investment case, and maybe to give a bit of context now with what you've seen over the last couple of years. We've always believed that what's most important in our industry, to have the right position in the right markets. We're in the largest economies in sub-Saharan Africa. We're in one of the two largest economies in the Middle East. Most importantly, we've got number one or number two positions in every one of those markets. It gives us scale, it gives us resilience. It gives us the ability to do well in the good times, and also to weather the more turbulent times. I think the second thing we've always really believed is that the demographic dividend is going to have a big impact on an operator like MTN.
The demographic dividend is recording the youthful populations, low level of data adoption, low level of financial services, still relatively embryonic wholesale and enterprise markets. We've always believed that we could harness that opportunity and generate double-digit service revenue growth in constant currency for the group. I think you've really seen that coming through in the last couple of years. Remember, we started at 7.2%, up above 10%, above 10% again in the second half. On the bottom left has been the philosophy that we can generate the revenue growth, but we can also harness our in-country and cross-country efficiencies to improve the margin. You've seen again, that over each of the last three years, we've had significant margin improvements, 1.2% in this year alone on an IAS basis. We still believe we can get operating leverage out of the group.
The third piece on the return profile, you remember a few years ago, we had group CapEx running around ZAR 31 billion-ZAR 32 billion. We brought it down to between ZAR 26 billion and ZAR 28 billion the last two years. We absolutely believe that with an envelope like that, we can invest in the networks to take advantage of all of the opportunities. Of course, if your revenue's growing at 10%, you've got efficiencies, which means your EBITDA grows higher than revenue, and you can stabilize absolute CapEx, then you get a big improvement also in operating cash flow. You've seen that again in the results for 2019, as you saw it in the results for 2018. Finally, in terms of being well positioned for the long term, we can optimize the portfolio. You see again, we realized ZAR 14 billion of proceeds in 2019.
We have another ZAR 45 billion of assets that are not long-term core strategic for the Group strategy. Of course, our ability to optimize the portfolio is what's enabling us to create sustainable leverage, and it's what's giving us confidence for our dividend policy, which was the progressive five and growing at 10%-20%. In the middle, of course, is the strategy we use for the operational execution. We still believe that the investment case is very much appropriate and coming to life in MTN. Key priorities for 2020. Really wanted to cover a couple of things on the slide. The first thing is the growth agenda. The growth agenda for MTN Group is really set out based on our belief in the digital operator strategy and the six revenue curves.
In the shorter term, in the 2020 horizon, in terms of evolving telco, monetizing the data adoption is a huge part of that. That is getting the right combination of 3G, 4G. It's pushing population coverage out. It's securing spectrum in the relative markets. It's seeing data revenues in South Africa optimized as we move through some of the pressures of the last while. A laser-like focus on that, of course, and particular, increasing the active data customers across the group. Secondly, pushing out on enterprise and wholesale, because we have relatively low market shares. We've got the best infrastructure on the continent. There's everything there to be won. In digital, number one priority is we want to see digital revenues growing in 2020. We do not want to put a slide up there again that's got a minus something in front of it. That's critical.
Second big priority is to scale Ayoba. We call Ayoba an instant messaging system for Africa, but it is much bigger than that. We are trying to build an ecosystem. An ecosystem with messaging, video, micro apps, mobile financial services. We're going to be embedding mobile money into Ayoba. We have big plans for Ayoba. David is here. Him and his team have got a laser-like focus on building the African super app, building the ecosystem for Africa. We are targeting this year 16 million active customers on Ayoba from the 2 million we finished at the end of last year. This is the year to scale Ayoba. For fintech, there, we need to scale the large markets. All the successes we've had in the group over the last while have been without really a scale operation in our two large markets, South Africa and Nigeria.
That's a big focus. As I said, of course, it's very important to round out the ecosystem. A lot of the revenues we generate in the fintech business now are generated from customers taking cash out of their wallets and from customers doing basic transfers. The way to build that ecosystem is basically to enable merchants to accept mobile money, then you build a payment system, and then to cluster your lending, deposits, and insurance activities around that. That's a big focus also for 2020. We did actually a lot of good work in building that system out, particularly towards the back end of last year. That's the growth agenda. Really wanted to get across as clear as we can this year, that there is an equal focus on efficiency.
The way to think about the MTN story is that we want to harness all the opportunities of an emerging market telco with demographics, low levels of adoption, et cetera, and that will drive the top line. At the same time, what you see in the developed markets is that the way they deliver profit growth and earnings growth is by focusing on efficiencies, because they have not so much top-line growth in the developed markets. For MTN Group, we have to harness the best of both worlds. We have to work as much on the efficiency agenda to improve returns as we are working on the growth agenda. What you see on the slide, firstly, on the right there, cost transformation. We're running an end-to-end efficiency review across the group. Global cost benchmarking. We brought Kearney in to do that exercise for us.
We're setting targets across the group, looking at every single element of running a telco business: radio, distribution, marketing, every support function as well. Cost transformation and zero-based budgeting is going to be a big feature of the MTN story. Of course, there's some big enablers of that, which is making sure we've got a modern digital infrastructure, which also will bring efficiencies. We are replacing our whole ERP system with a project called Project Boost. We are replacing the whole billing and BSS environment in South Africa with a project called Siyakhula. We have an Oxygen program, which is the overall IT roadmap that Charles has been running with in technology, and a lot of focus also on the internal processes within the organization. The fantastic thing about digitizing processes is the customer experience improves and you become more efficient, and you save money.
The third element is more around the corporate side. Obviously, we've got a big focus on the Realization Program, as we've discussed earlier. The portfolio in general, we are mostly focused on delivering on an organic strategy. Of course, we do look at potential opportunities across the continent. They need to be compelling, and they need to fit with the MTN investment case. Risk and regulatory. We've got a couple of license renewals to get through in 2020. We still have some work to do on Uganda. We've got a license renewal in Guinea, Conakry. Actually, the other larger license renewals are in 2021, so we need to make sure that we use 2020 to get really well prepared for that.
Of course, we continue to work on our risk management processes, operational risk, both in the GSM business and in fintech, and also managing some of the more complex matters like, for example, the class action we now have in Afghanistan. That's a complex matter that the group needs to manage. Finally, on sustainability. Firstly, a big focus on ESG, and I've got a slide to cover on that, which I'll do on the next slide. Then, of course, the other big focus is around management succession. What we announced this morning is firstly that our Group CTIO, Charles, who's sitting here in the front, will be moving on to the Group Executive Committee. He'll be reporting to me. Group Technology, Group Information Security, now becomes a full-blooded part of the Executive Team.
A big thank you also to Jens, who's had the stewardship of that function over the last while, and congratulations to Charles as well. I'm really looking forward to that. Second thing we announced is that our Group COO's contract has been extended; he will be with us until March 2022. We told our COOs last night that they have another two years of his budget stresses to look forward to. They didn't smile quite as much as we thought they would. Yeah, Jens has made such a contribution to the Group over the last few years, so I think we're very lucky to have him with us. I guess, of course, the final thing, as you've read this morning, is I was employed on a fixed four-year contract.
My contract finishes in March 2021, I will be leaving the group at the end of my contract. Four years at MTN is like eight years anywhere else. I'm hoping to walk out still in one piece, it'll be fantastic when I go home, the dogs don't bark at me anymore. That's where we are on management succession, I think the big focus is we've got 12 months to do an orderly process. The group board will be running with a succession process, seamless handover. Let's keep our commercial momentum. The executive team is going to be fully focused on implementing the strategy. I guess we really want to show that MTN can show how management succession should be managed. On ESG, this is really, really such an important thing going forward.
I mean, A, it's the right thing for businesses to focus on, their effect on the environment, contribution they pay in the social sense, how they think about governance. Of course, also, it's very important to recognize the contribution we make to society. Whilst this has been something that's been a focus of MTN in the last while, we really want to intensify our focus on ESG. We are doing well, we want to do better. We want to set bigger and more impressive targets going forward. We've got policy work to do, and we are going to really focus on this in 2020. Just a few things, though, that are on the slide. I think in terms of environment, we've done a lot of work about reducing our own carbon emissions, so scope one, scope two, our own infrastructure.
This site is itself in Emergent Stream's powered by a gas turbine. We're going to be increasing the capacity of that so we can run the whole of phase one and phase two. We've done similar work in Nigeria in some of the big data centers. E-waste being recycled, 784 tons. That was less than 300 tons in 2018, so a lot of progress being made there. Renewable energy sites. It is obviously always a challenge because the network is expanding. We're rolling out more sites, but there is a lot of work we can do to minimize our effect on the environment. In terms of society, you saw some of that on the video. We're the first operator to sign up with Internet Watch, blocking URLs for particularly content containing child sex abuse. This is a role that a network operator can play to protect our children.
I think it's super important. Strong score on our engagement. A lot of effort and focus going into CSI, and of course, very focused on the role we play in gender equality. We signed up last week to the Women's Principles with the United Nations. We have 37% of our workforce being females today, and I almost didn't want to put it up there because that's not a victory. That just shows that we have a lot of work to do, and we are very much focused on that. On the right-hand side, all about governance, how do we run our business in a responsible way? You see some of the metrics there, e-learning on ethics, the tip-off line for whistleblowers. That line alone is carrying 40% more traffic than it did the year before. A lot of work being done on corporate governance.
You see a much improved result bottom right, and also a lot of work in the vendor environment because a lot of our procurement is going through the large vendors. That's the ESG framework. I guess I just wanted to finally shine a light on the impact we have as an industry economically. It is a great privilege to work in an industry like this. We're not selling hamburgers, we're changing the world. At least that's how we like to think about it. If you look at the bottom there, 95 million people accessing the mobile internet on an MTN network. 35 million people connected to the economy through MTN financial services, creating jobs for more than half a million agents who are active in the mobile money space.
1,387 rural sites pushing coverage into rural areas, bringing the most vulnerable, the most disenfranchised customers into the digital world. It's also mostly, I think, what keeps us all working so hard is that we really also believe that we are making an impact. Final slide, just to pull that all together, is what are our financial guidance metrics now for the medium term. We stick with our service revenue guidance, double-digit constant currency for the group, driven largely by MTN Nigeria being in the double digits, and over time, MTN South Africa returning to mid-single digit. Now we've been off that for a couple of years. We believe in a three-to-five-year medium-term horizon, we can restore service revenue growth in MTN South Africa, but it's clearly going to be a little bit of a bumpy ride.
Fortunately, we've got enough outperformance in other parts of the group that we can stick with our group targets. Look at a market like MTN Ghana, now 23% service revenue growth. Strong contributions from a number of our tier two markets. We continue to focus on improving EBITDA margins, and that's a feature, again, of the results over the medium term. CapEx intensity is coming down. The two changes to guidance that really we've covered, both myself and Ralph, is we put another ZAR 25 billion of proceeds into the guidance over the next three to five years, and we are now targeting to run the group with a whole core net debt ratio below two times from our previous range of 2 to 2.5, and we're sticking with our guidance on return on equity improvements and with our dividend policy.
That's a whistle-stop tour through our results. I'll ask Ralph to come up and join me again, we've got time for Q&A.
Okay. There's a mic there on the right. Sorry, JP.
How's it, guys? It's Preshendran Odayar here from Nedbank CIB. I've got two questions. Firstly, on South Africa, if you include the Cell C roaming revenue in your numbers that you took out, it looks like service revenue is flat around 1%. How do you see that going into 2020? You've had investments into your network during the year, so I expected, especially on the prepaid side, but better numbers in 2019. Now you've also got this headwind of the Competition Commission, and an update on that. The second question is on Nigeria. Can you give us some color on when you can expect an IPO, and how are you seeing FX given the oil price? Thanks.
[crosstalk] If we look at the South African performance, as you quite rightly say, if we had added back the revenue we didn't recognize, ZAR 283, the performance would have been 1.2%. I think for you guys who look at comparatives, Vodacom's numbers would have been 1.3%, so similar kind of performance. I think the challenge for us going into particularly the first half of this year is to lap both the cost of data rule changes that came through the beginning of last year, as well as the out-of-bundle tariff reductions.
We are engaging with the Competition Commission right now. Godfrey and team are engaging with the Competition Commission to conclude our interactions. We anticipate that South Africa will still see H1 under pressure. A recovery that we'd have expected to see in Q2 probably is pushed out more towards Q3 from a South African perspective because of these pressures in consumer prepaid. On the question around Nigeria, we've put out this further localization as a target, obviously subject to market conditions. It's 14% additional shareholding we'd like to see in the hands, but obviously, it's always going to be subject to market conditions. Right now, we wouldn't give you a very specific timing, but when we see market conditions be conducive, that's when we would launch such a localization.
JP, you're second for once. The mics have been disinfected.
Not after me. Good morning to both of you. Two questions, please. The first one, Rob touched on it, the Afghan claims. I wonder between the headline and now what you guys have done as an exco and a board to just get comfortable that the claims are of no substance, or maybe to the contrary. The second one, this one for Ralph, you provide a very useful stress test scenario on your holdco net debt in the appendix. Just wondering if you could just talk through in the event of the naira having a big devaluation, is the only real number to worry about there the upstreaming of cash? There shouldn't be any impacts on the other bits and pieces, it's just the upstreaming of cash that would be the variable, or will there be a series of different consequences? Thank you.
Okay. Maybe let me tackle the first one. Yeah, clearly we've worked on that issue over the last few months. What we are saying is that, we believe that we conduct our operations in a professional way across the group, and in a way that complies with all the laws and regulations. We don't believe there's any merit in that case, and we will be defending it. To be frank, we don't really get drawn much beyond that, because we are in a litigation process, obviously now in the U.S., and I think we just need to be circumspect on communication.
Just on your point around the stress scenarios. I think just to give you guys a bit of color on Nigeria in particular. If you take the OpEx composition, because I know where you're going with this, J.P. , the OpEx composition of Ferdi's costs is probably more like 82% local currency, 18% foreign currency. We look at its CapEx, it's probably 45% local, 55% foreign. Obviously its debt, given what the local debt that's been raised is 92% is local. The business within itself is fairly resilient from any kind of external currency shocks.
The stress scenario that you see in the appendix, if you were to take it even further and work towards, let's say, a 30% devaluation of the NGN, 30% less upstreaming, broadly to the group, Nigeria and others, you take a ZAR at 18 to the USD, you would still be within the two and a half times range. You'd be at the top end of that. That's a much stronger stress scenario that would take you to two and a half. You'd need to see quite material impacts. That's why we feel in our modeling that the cash upstream to the group and the current macro conditions, we still feel comfortable that dividend will be covered for cash remitted to the group by 2021.
Yeah, Jonathan.
Yes. Jonathan Kennedy-Good from Standard Bank. Just on Cell C, where we are on the roaming agreement, I presume currently still under the first iteration of that agreement. I think there were conditions precedent to move to a broader agreement. Just trying to understand whether the recapitalization of Cell C is one of those conditions before you agree to expand that roaming agreement. I don't know if there's any news on whether you can talk about the potential revenue uplift if that does happen. Just on the sale of Nigerian, the sell-down in the asset there, I seem to recall that you were waiting for the mobile money license before thinking about that transaction. Any update on where we are on that score?
Yes, I think as you say, when we first brought Cell C on as a roaming partner, they were really just using some of our excess capacity outside of the metros. We call that colloquially phase I. The phase II agreement is obviously a much more comprehensive national roaming across the whole market. That has some conditions precedent, because once it kicks in, we'd have to put quite a significant investment in the network to enable it. Of course, on their side, to really harness the efficiencies of the broader roaming network, they would need to start rationalizing their own network. There's quite a lot that needs to happen before we make that move, and that's really on both sides. I think they are busy with their own refinancing recapitalization.
I think, provided we've got enough confidence that that is really well arranged, over time we could shift from phase I to phase II. I guess practically, as we sit here today, we're slightly in between because we have also reworked some of the phase I requirements to give more capacity at different rates. We've pulled down some of the out-of-bundle exposure there was historically. I think for now, it's actually working quite smoothly, and once they conclude their processes, then I think relatively quickly we could move from the two phases. That may still take some months. We're not expecting big shifts in 2020. The second question was?
Yeah. It's Nigeria, with the conditions. We had spoken previously that we'd like to see the AGF matter resolved, and ideally, the PSB. We always said ideally the PSB. The AGF matter has now been resolved, and the PSB hasn't been issued, but I think our view is that we've made a statement of intent to further localize, have more shareholding in local hands. Once market conditions are conducive, we will proceed with that. We won't make it conditional on the PSB.
Yeah, I think just to build on that, we did secure the super agent license in 2019, which enables us really to roll out the agent banking network.
We're very busy with that, and we're pretty much the only operator now in that market that is operationalizing fintech, because other operators got approval in principle, not yet formal and they've not really begun yet.
Okay.
Hi. Ziad.
Hi.
John Kim, UBS. Two questions, please. You mentioned that there were a series of license renewals for 2021. Can you give us a bit of a shape to that? Which markets are they? Do they tend to be lump sum payments, structured payments, service requirements? Secondly, on Mobile Money, I think, you had put a figure for a pretty aggressive agent rollout. Can you give us a sense of which countries this will be focused on? Thanks.
I think the license renewal is coming up in 2021. We have a 2G license in Nigeria. We have also a license renewal in MTN Irancell. I think those are probably the two most material. As I said, we'll work on those during the course of the year. The way licenses are renewed across Africa is really not harmonized. Some are taking revenue shares, some are lump sums, some are 10 years, some are 20 years. Some include spectrum, some don't. There's not really, I think, a clear set of principles. Each one has its own unique characteristics. For Mobile Money, there's a couple of pieces to it. There's the agent network that we use for cash in and cash out for the end consumer. That's where you see it up now at around 550 million.
To round out the ecosystem, we really need to increase the merchant ecosystem. That is sitting now at around 200,000 merchants, I think 220 or so. Both of these need work, but probably the one that needs to scale the quickest actually is the merchant ecosystem.
There we have a decent ecosystem in MTN Ghana, which is most of the existing merchants. We have a big program to roll that out across the large markets, particularly like Côte d'Ivoire, Uganda, Cameroon, Benin. That's where we have already a scale business.
Hi.
Yes.
Ziad from Nedbank CIB. Just two questions, please. The first one is on the sequential growth in network OpEx in South Africa. How much of that is structural? If you could maybe give us a bit more color why there was such a big uptick. Secondly, just on your tower exposure through IHS. If you look at the ratings or the performance of tower opcos, even through these tough times, they've done really well. How do you view that exposure? Is it something you'd still like to have significant exposure to, or would you look to exit tower exposure completely? Thank you.
Okay. Maybe I'll take the second one.
You start the second one.
You still have some figures out for the first one.
Try to remember.
You want to phone a friend?
My friend is right in front there.
I think the way we look at the tower portfolio is, over time, a significant portion of the towers were sold to tower companies in sale and leaseback transactions. Most of those sit with IHS. As a result of those and some reinvestment over the years, we ended up with a 29% shareholding in the IHS entity. Then two joint ventures in Ghana and Uganda with ATC. The balance of the tower portfolio is still owned and managed by MTN. What we've said is that we don't believe we need to have capital tied up in investments in tower companies. Obviously, we need to manage our operational impact with them because we have a dependency for rollout and quality, et cetera. That is the reason why we've been, over time, monetizing the tower company investments.
We've done the two ATCs and the loan. We would like ideally to monetize the IHS stake over the medium term. There, we're very much dependent on them moving forward with their own IPO processes, et cetera. It is an asset we've declared as not long-term strategic for the Group. It's a big number. It's fair valued in our accounts this year at ZAR 27 billion. The capital base of the Group is around ZAR 85 billion. We have roughly one-third of our capital base tied up in a non-strategic asset that is not generating any earnings.
Yeah.
It's a big part of actually what's been dragging our ROE over the years.
It's also important from a return perspective.
On SA, I think you see the big growth in there. There is a once-off vendor voucher in the base of 2018 that makes the growth much larger than what it is on a once-off basis. You'd have a growth probably more like 13%. That network has grown in size by 7% in the year. Obviously, just the sheer size of it and then the rental escalations coming through. Rob did mention earlier in the focus areas we have for 2020 around efficiencies across the group, and part of the efficiency focus in SA is seeking a lot more network efficiencies. Godfrey and his team, Giovanni, and the rest of the team will be very focused on that, given that the market is looking quite challenging for top-line growth.
Godfrey and team are pushing quite hard on getting that efficiencies, and some of it will come out of network.
I think it's fantastic for the Group that we can really optimize and manage both revenue growth and efficiencies at the same time.
Yeah.
I think that draws the Q&A to a close. We thank you very much for attending. Those of you for watching as well. We have refreshments outside. Please enjoy the great Motherland Coffee. Yeah, thanks for coming.
Thank you