Millicom International Cellular SA (TIGO) 2010 Q1 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Millicom Q1 2010 Results Conference Call. For your information, today's conference is being recorded. May I also remind you that this call is being audio streamed over the web and is accessible at www.millicom.com together with the presentation, summarizing the key features of the results.

I would now like to hand the conference over to your host today, Mr. Mikael Grahne, President and CEO, and Francois-Xavier Roger, CFO. Please go ahead, sir -- gentlemen.

Thank you, operator, and welcome to you all. As usual, you can find the slides for this call on our website. Please turn to slide number four. We have made an encouraging start to the year. Total revenues were up 16% with an acceleration in constant currency revenue growth to 11.5% and a positive currency impact. Our EBITDA margin reached a record level of 46.8% with a strong performance in all regions and earnings per share was up 11%.

Cash generation continued to be good with nearly 26% of revenues converted to operating free cash flow. With a strong balance sheet and following the Asian disposals, we have decided to return $800 million to shareholders as you will have seen from last week's announcement. Together with strong EPS growth, we are demonstrating our clear commitment to shareholder returns.

Slide five. Customer numbers were up 21%, despite the uncertainty caused by the SIM registration process in four of our markets. Total revenue growth of 16% mainly reflects continued good customer momentum with an increasingly stable ARPU performance as well as some currency benefits. The 20% increase in EBITDA is the fastest rate of growth for several quarters and demonstrates good cost control as well as the benefits of increasing scale.

CapEx was low in the quarter at $99 million, as a result of timing issues, but the working capital outflow reflects the payment of CapEx books in the previous quarter. Taken together, these movements create a fair reflection of our investment activity during the period and produce an operating free cash flow at nearly 26% of revenues.

Slide six. I know that many of you focus on our customer intake from one quarter to the next. While it is clearly important, there are any number of factors that can make it quite volatile. So, we do not see that as a reliable indicator of our overall performance. We are much more focused on revenue growth where our performance has been more consistent and where growth has accelerated this quarter.

This chart shows our quarterly intake over the last few quarters as well as the rolling average over the last four quarters, which gives a better picture of the underlying trend. As you can see, although quarterly data is very volatile, the moving average is really very stable. While we expect this average to decline over time as markets become more penetrated, it does demonstrate that our underlying performance remains very consistent.

Slide seven. Looking at the sequential performance quarter-on-quarter, revenues declined marginally as expected after the holiday season, but EBITDA margins made further progress despite Q1 usually being seasonally weaker. Cash generation continued to be very strong despite the negative swing in working capital as a result of timing of CapEx payments.

Slide eight. As you know, ARPU has been a major focus for us over the last years. By stimulating usage, not only through value-added services and data, but also by customer segmentation and carefully targeted voice promotions, we have really begun to stem the decline in ARPU as this slide demonstrates.

In South America and Africa in particular, local currency ARPU is now very robust year-on-year, while Central America has been affected mainly by external factors such as tax increases that are outside our control. We are also up against a negative mix effect with our lower ARPU Africa markets delivering higher growth than Latin America. If we remove the effects of the country mix and the impact of taxes, the actual decrease in ARPU is around half of the reported figure.

Slide nine. Breaking down revenue growth into its component parts, we were able to accelerate constant currency revenue growth to about 11% as a result of good ARPU management. We are now also benefiting from currency movements, particularly in South America.

Slide 10. VAS continues to be a major driver of our top-line growth with a 40% local currency growth rate on top of last year's high growth. Voice revenue growth is steady, but continues to be affected by interconnect regulation and taxes.

Slide 11. VAS represents over 21% of recurring revenues, delivering a more dependable and higher margin revenue stream that generates greater customer interest and loyalty. We maintained this rate in Q1 despite the holiday season typically being a time of increased usage of non-voice services. Within VAS, 3G/data is becoming an increasingly material element and will be a major focus for us over the coming quarters.

Just over a year after its launch, data is already delivering over 3% of recurring revenues in Latin America from around 2% of customers. Broadband, whether wireless or fixed, has attractive growth prospects and our total broadband revenues were up 93% year-on-year in the quarter.

Slide 12. Our market share increased by 0.5 percentage point in the quarter, with gains in all three geographical regions. We were particularly pleased to continue to grow share in Africa, despite a more significant competitive response in recent months after our strong showing in 2009. Since the beginning of 2009, we have grown our (inaudible) market share by 2 percentage points.

Slide 13. Churn in percentage terms continues to be relatively stable. However, given our very strong customer intake in Q4, we did experience a higher absolute number of disconnections in Q1.

Please jump to slide 15. Now, we will review the regional clusters in more detail. In Central America, customer numbers continue to grow at 15% annually, in line with Q4 growth. In local currency, revenues were flat. Given the impact of taxes and interconnect costs, this represents an encouraging performance and an improvement on the negative trends of 2009. EBITDA margins have so far held up better than anticipated, reflecting good cost control and strong growth from higher margin value-added services.

Slide 16. In South America, revenues increased by 17% in local currency; the continued strength of the Colombian peso to dollar revenue growth to 32%. In revenue terms, South America is now almost on par with Central America. EBITDA for Q1 was up 41% and EBITDA margin was 42.4%, an improvement of 2.9 percentage points year-on-year, driven mainly by improvements in Colombia. As with Central America, CapEx was unusually low due to timing differences, helping to deliver a very strong cash conversion.

Slide 17. In Africa, customer growth continued to be strong, despite the uncertainty created by SIM registration in three markets and a tougher competitive environment, particularly in Tanzania. At the end of March, we launched our TZS0.50 per second tariff in Tanzania, just as our competitors had matched our previous TZS1.00 offer and this has resulted in stronger customer intake with no impact on ARPU.

We expect further volatility in customer growth while the registration process continues over the next few quarters. Revenue growth at 26% in local currency remains strong and is now much closer to the rate of customer growth, thanks to our focus on stabilizing ARPU.

EBITDA for Q1 was $83 million, up 41% year-on-year with the margin increasing by 4 percentage points with our growing scale. Excluding Rwanda, the margin was over 40%. We generated $17 million of operating free cash flow in Africa, despite continuing to make significant levels of investment, thanks to our improving margins.

Slide 18. Turning now to our cable operations, we saw a consistent rate of growth in Amnet, with revenues up 9% year-on-year. EBITDA margins have improved and cash generation was good. Amnet was cash flow positive and we expect it to continue to be so.

Now, I would like to hand over to Francois-Xavier, who will talk you briefly through the financials. Francois-Xavier?

Thank you, Mikael. Slide 20. During the quarter, strong EBITDA growth translated into very good progress in earnings per share at plus 11%. Depreciation continued to increase quite sharply year-on-year, reflecting a continued investment on a negative effect from currency movements, particularly in Colombia.

However, depreciation is down 7% on Q4 2009 as we are starting to benefit from the outsourcing of stores in Ghana and we have revised the useful lines of all of our stores from 10 years to 15 years to align with industry standards.

Slide 21. Our tax rate was higher than in Q1 last year as a result of a very strong stream of dividends paid from our operations which attracts withholding tax. We still anticipate a full year of tax rate below 30%.

Slide 22. Cash generation has continued to be a feature of our financial performance. Combined with an already strong balance sheet and most of the proceeds from the Asian disposals, we are proposing the return of $800 million to shareholders. This will still leave Millicom with net debt to EBITDA of less than 1 after the completion of the buyback program as well as excess cash for external growth.

Slide 23. All our clusters were cash flow positive in Q1. In Africa, this position may continue to vary from quarter-to-quarter, depending on the timing of CapEx bookings. We confirm our guidance for $700 million of CapEx full year and the low level book in Q1 is just a timing issue.

Slide 24. Group free cash flow was $200 million during the quarter off 22% of revenues. While we benefited from unusually low CapEx booked as a result of timing differences, we also saw a negative cash impact in Q1 from higher CapEx payments than in previous quarters.

Slide 25. Turning to our debt maturity, we now see the average maturity of our growth debt at under three years. Our funding strategy is based on four elements -- to extend maturities, to lower the net cost of debt, to secure sufficient liquidity, and to push down debt to operating level for tax efficiency and to mitigate country risk. We expect to make further progress on these fronts in the coming months.

Slide 26. At this early stage in the year, we are happy to confirm the guidance given in February. Although we have outperformed so far, we continue to monitor changes to tax on regulation that could affect our performance. We are not aware of any other additional risk and any past or future or known exposure has obviously been fully provided for in our returns and in our guidance.

Slide 27. Finally, as you know, we plan to return $800 million to shareholders in addition to the annual dividend already announced. While we continue to look for external growth opportunities, we have no intention of over paying for assets and in the meantime, we wanted to reduce our cash balances that were generating almost no return for shareholders.

This move still leaves us in a very strong and flexible financial position should new opportunities arise. As I mentioned before, our net debt to EBITDA will still be less than 1 after these shareholder returns while our cash flow generation remains strong. I would now like to hand over to Mikael for his final comments.

Thank you, Francois-Xavier. We are pleased with the start we made to 2010. Economies are showing some early signs of stabilization, although we remain cautious of external factors. Although competition remains a fact of life, we continue to demonstrate that we are equal to it with a good form of accelerating top-line growth, strengthening margins, and improving market share. We aim to continue to deliver a similar mix through the rest of 2010.

That concludes my comments and we will now be happy to take your questions. Operator, may we have the first question, please?

(Operator Instructions). Stefan Pettersson from Nordea Bank.

Yes. Hello. I have a few questions. First of all, can you give some info on the taxes that might be introduced in Africa? And then, if you could give some information on the magnitude and timing of introduction of these taxes.

Secondly, I would like to have some information regarding the competitive situation in Colombia after the regulatory changes. Has America Movil cut prices to defend position and also what was the EBITDA margin in Colombia this quarter? Finally, how does the changed life of towers change the average life of assets?

Okay. I will take the first question. On the taxes in Africa, so we have two taxes that are currently under review and under discussion with the governments in Ghana via international incoming calls. And in DRC, a numbering tax as well as a spectrum tax.

I mean, I can't give you any amounts or any information on timing because this is still under discussion between the industry mainly and the authorities. But anyway, as I said earlier, I mean, everything that -- any liability from the past regarding these taxes or other taxes has already been provided for as far as Q1 is concerned. And anything that we are aware of has been provided for in our guidance.

Okay. On Colombia, the regulatory body declared America Movil as having a significant competitive position. And as of December, they were asked to reduce their across rate from their on-net tariff plus the interconnect that meant basically on-net tariff plus around $0.05 compared to a situation before when the across-net tariff might have been the on-net tariff plus $0.25.

We have not seen any significant acceleration of incoming calls from that operator as a result of the change. Usually, this takes a long time because it's something that our competitor is not very actively marketing to their customers. In terms of EBITDA performance, we had continued growth in the EBITDA margin in Colombia in Q1.

Regarding the change in depreciation policy for the towers, so we move from 10 to 15 years to a line of practice with -- I mean, industry practices because we were really on the low side. Even -- understand some operators even depreciate them over 25 years. The impact on the quarter on Q1 was around $8 million.

Okay. Thank you.

You're welcome.

Ric Prentiss from Raymond James.

Hi. It's Ric Prentiss. Good morning, good afternoon. Two questions for you guys, if I may. First, on the Central American region, margins continue to do nice there, even in a competitive environment and with the changing of the taxes and the interconnect rate. Can you share with us a little bit about where you think margins might be headed in that region and what it might mean for the group margins?

And then, the second question is on the stock buyback. I noticed you were going to be doing the buyback on Nasdaq by year end. What form will that buyback take and, as you look out, how much cash do you want to keep on hand long-term in the business?

Okay. I'll start with the Central America margins. As we said before, we don't see, basically, from an operating point or from an industry point of view, significant threats to our margins because we have very strong positions, high level of on-net calls which really drives the margin, and very strong advance on [west] that continue to strengthen it.

As we always said, there could be some regulatory or fiscal changes that would have an impact on those margins. But from an operating point of view, we don't see that and we are quite confident that, from an operating point of view, we cannot keep on holding onto those margins.

As far as the share buyback is concerned, so we will do it through an open market per shares with a broker. On Nasdaq only, we will -- the intention is to complete it within the next nine months or before the end of 2010. We will obviously not do any share buyback during blackout periods. We have the approval to start now, but we will need further approval to complete it because the approval that we have now is valid 'til the next GM that is taking place in May.

As far as the cash that we intend to maintain, as we said in the past, I mean, we don't need more than $500 million in cash for liquidity purposes. So, any cash over $500 million will be excess cash. We have no intention, frankly speaking, to keep any excess cash because we are destroying value somewhat. But we may keep some if needed for a short period of time or if we think that we may use it for external growth at a later stage, as we did over the last couple of months.

Right. And the external growth, I think you mentioned you don't want to overpay for assets, which I applaud. Some of the pricing has gotten high out there. How do you view the marketplace? Are you a buyer? Are you a seller? Are you a patient watcher? Just how do you view the external M&A process?

First of all, I would like to start by that we concluded that mostly, as you are aware, the most important scale is really the in market scale being number one or number two in the markets where we compete. That drives a sustainable long-term profitability growth. So, as we said before, any acquisition would have to be purely opportunistic and complementary rather than strategic. So, we would only buy assets if they meet -- hit that criteria.

In addition, we are looking at the number one or number two position in the market, something that we can reach within a reasonable period of time. I think, clearly, we would not have bought the assets from Zain at the kind of a price Bharti reportedly has paid for that. It's a little bit difficult to really fully assess that pricing because Bharti had -- Zain has not publicized fully their 2009 results for Africa. But if the prices paid is as per reported, we would not have done those acquisitions at those pricing levels.

Great. Thank you.

James Rivett from Citi.

Good morning, guys. I just have one question. As you're talking about the African competition, we know that it's heated up in Tanzania. Can you just give us an update on the position there and whether you're seeing things get tougher in any of the other markets since the turn of the quarter? Thank you.

Yes, I'd just like to comment on Tanzania. I mean, it's been quite competitive in Q1. Vodacom reduced their pricing significantly. We looked at it and we actually followed in March. But just to quote you some numbers, in Q4 we had a local currency growth in Tanzania at 47%. And in Q -- this quarter we were at 45%. So, I would believe that we clearly are outperforming our competitors in Tanzania at this moment, even in those very competitive situations.

And there's been no change since that period -- the tariffs, the stock, and everything's the same?

Well, we announced, as I said, in the opening remarks, we have moved -- come up with a new tariff, which is basically TZS0.50. And we've seen an increased customer intake as a result of that. But this is nothing new. I mean, we go through -- I mean, we operate in competitive situations, period. And we just have created a model and an implementation capability to be the winners in it.

Great. And this hasn't spilled through into any of the other markets at this stage?

I think all the markets continue to be competitive. I mean, that's -- there's no big change from that perspective.

Okay. Great. Thank you.

Lena Osterberg from Carnegie.

Yes. Hello. I was wondering, when you decided to do the extraordinary dividend on the share buyback program, why didn't you opt to also address the high-yield bond at the same time? That's my first question. The other issue, I was wondering if you could maybe say something about where the margin in Colombia is currently? And also, I was wondering, Rwanda, the expected future negative effect on EBITDA -- if you could give us some guidance on that.

Okay. As far as the high yield bond is concerned, we did indeed consider redeeming the high yield bond and we said in the past that this was one of the options that we were looking at. I mean, the Board and the management have decided not to redeem it at that stage. We are aware of the fact that this is a fairly expensive financing.

And so, there is a cost attached as well to the fact of early redeeming it. So, for the time being, we have decided to maintain it because it provides some liquidity as well. But as I said earlier, I mean, we are seriously looking at different options in order to extend maturities of our debt -- I mean, reduce the cost of our debt, secure sufficient liquidity, and push down debt. So, we'll -- we are looking at it and I mean, this decision is not entirely final at this stage.

Again, I ask you -- my understanding is that the cost of redeeming the bond is 3.35%. You pay an annual interest of, I think, $45 million and on top of that, you don't have a tax shield. So, I'm -- I just can't figure out why you're not doing it.

Yes, but if we really meet -- first of all, so we have to pay this penalty of 3.3%, and if we issue, for example, a new bond we have to pay probably around 2% up-front fee. So, in terms of payback, I mean, it's quite a lot of money to pay up front -- the 3.3% plus the 2%. So which make that the payback is not always as attractive as it may look at first glance.

But once again, I mean this is something that we permanently look at and much depends on the other option that we have to secure long-term financing under other configuration. And indeed, as you said, the ideal situation would be to be able to secure long-term financing at operating level and we are working on it.

Is it very difficult to do that currently?

It depends -- I mean, we are working on certain different markets, so it is indeed difficult in some markets, might be easier in some other markets. And I mean, there are different options as well because we can go for our private financing. We can go for financing through direct financial institution or we can go through public financing as well. So, there are different options and we are looking at different options.

Colombia margin in Q1 between 25% and 30%, continued margin improvement. There are two factors driving that -- scale and focus on cost. But the primary factor is basically continued growth on value-added services that come with the higher margin. So, we are very busy establishing Tigo as the real young urban cool choice in Colombia. That strategy is working well for us.

Rwanda -- of course, it's a greenfield project and we are just, I think, in month four or since the introduction. So, we are right now focusing on building enough scale on the subscriber base so that we can start working on the financial numbers once we have enough momentum and revenue. So, I would expect that Rwanda would cost us 1% to 2% of EBITDA margin in the next few quarters.

Okay. Thank you.

David Kestenbaum from Morgan Joseph.

Okay, thanks. Mikael, could you just update us on what's going on with the process in Laos and what exactly happened with the Vimpelcom? And then, also the process in Costa Rica. Thanks.

Yes, we had an agreed -- share purchasing agreement with Vimpelcom. All the conditions were satisfied. Vimpelcom elected not to complete because they wanted, basically, full approval from the government, which wasn't a condition on the sale. And in order to protect our positions, we came out with the release that you might have seen. We are still confident that we will complete the disposal of Laos in given time.

So --

Yes, go ahead.

Sorry. So, if they do get full approval of the government before, in the near term, then they may actually buy the asset, is what you're implying, right?

Yes, that's speculated. But we are still -- we still believe that we are in a strong position to sell this asset.

Okay.

In Costa Rica, we're still looking at the opportunity, the license -- exact license rules are not fully known. But it's something that we are studying.

What's -- sorry.

I'm not updated on the time, but I think it's -- the final rule should be quite imminent.

Okay. Thank you.

Martin Mabbutt from Nomura.

Well, thanks very much. Just a quick question on CapEx, if I may. You talked about the benefits of the 3G networks that you've got in one or two places. I wondered how that's going to impact on CapEx maybe beyond this year. Do you expect CapEx to rise from the $700 million level or do you see that as an ongoing fair level for future years?

Well, we haven't given any guidance beyond 2010 on CapEx. We continue to invest in 3G coverage in our markets, primarily in urban areas when it comes to Central America and also in some smaller cities in Colombia where 3G actually is performing very strongly because of lack of alternatives. So, I know, we are comfortable with the $700 million guidance for this year and we haven't yet done any plans for 2011.

Okay. Thanks.

Jan Dworsky from Handelsbanken.

Thank you. A question on Senegal. Any update on the process in Senegal?

Not any material update. The arbitration committee has gone through a few hearings and so the process is ongoing. And as I said before, we do not expect a very rapid resolution. In the meanwhile, the business in Senegal is performing relatively strongly. We had revenues up 11% in the quarter. And basically, subscribers were up 21%. So, steady progress there. I think we have strong support among the ordinary people there in the country. So, slow progress, as we said before.

When it comes to Ghana and I'm not sure if Globacom has launched yet or if they are just about to launch. Do you see that as a sort of potential threat in Ghana or just another --?

Yes, we are used to competition. They have not published a launch date in Ghana. They are quite busy doing free advertising. We are slightly uncertain when the launch date itself is going to be -- clearly, Ghana is too crowded with six plus operators. So over time, there has to be some consolidation. We are the number two operator with a very strong number two position. We feel that if there's any consolidation, we are in a good position to take advantage of that.

Then, I just want to come to back to your answer in relation to the margins in Central America. Is it -- I think, previously, you said that in terms of the Group margin reaching mid -- sort of being maintained at mid-40s that you saw some pressure in Central America and margins in Africa reaching Group levels. But now based on your answer, it sounds like you don't see any operational issues that will really take down the margin in Central America. Is that a correct interpretation?

Yes, I think we always -- yes. I think we always emphasize that, from an operating point of view, we felt the margins were quite dependable. The threat was always some either regulatory or fiscal change that would impact that. And because of that, we still maintain the stance that if Africa EBITDA margins over time go up to Group average, probably we have a decline in the Central American markets to counter that.

Then, just finally on the operating cash flow target of mid-teens, with CapEx at the same level as last year and based on the organic growth tendon and the margins, why shouldn't operating cash flow of central revenues be in line with 2009?

Yes, I mean, if we look at it --at the cash flow level that we generated in the last two quarters, indeed, I mean, our guidance may look a little bit conservative. We are still in Q1 and I mean, cash flow can be so volatile due to working capital adjustments and maybe CapEx payments.

So, we are still early in the year. I tend to agree with you that might be a little bit conservative. But for the time being, we maintain that guidance and we will review that carefully -- we keep on reviewing it carefully regularly. And if needed, I mean, we'll review it accordingly in the next quarter.

Okay. Thank you.

Kevin Roe from Roe Equity Research.

Thank you. Mikael, I wanted to follow up on your Zain/Bharti commentary. They compete with you in four key African markets. Do you see this ownership transition as a potential market share opportunity or do you see it as a competitive threat with these assets now under Bharti's control?

I think -- I mean, I have full respect for Bharti as an operator. They're probably going to do a better job than Zain did at the end of the period, given that they were quote unquote for sale for quite a long period. I think we have very strong positions in Africa and, as you can see from our numbers, we have consistently over the last year and even in this quarter, probably outperformed our competitors.

So, to get that kind of a momentum going, it normally takes a few years because we have to get all the right people into the right position with the right know how. So, I expect that it's going to continue to be as competitive as ever, but it's going to take maybe some time for Bharti to get up to speed.

So, potentially a near-term opportunity, given the transition?

Yes, potentially. But we have full respect for them as a competitor, so.

Great. Thanks.

Sven Skold from Swedbank.

Hi. Actually, most of my questions have been answered, but I have a few of them left here. First of all, when you say that you have seen a stabilization of the macroeconomic environment, is that an improvement in minutes of use or was it -- what is it that you have seen? And I'm wondering specifically about Central America.

And second, when you have cut the termination rates in El Salvador, have you seen -- I mean, as the year is picking up, as a consequence of lower prices?

Let me answer the El Salvador first. Interconnect rates, typically across -- outgoing across rate falls, have less elasticity to it than if you cut the rate on on-net calls. And so, we haven't seen a big increase there in shifting of minutes.

In addition also, since we have such a dominant position there, we don't -- we have a relatively limited incoming call compared to our on-net call. So, we haven't seen a big difference in incoming calls either.

In terms of macroeconomic environment, that's probably more sort of spots here and there. You can look at things like the Ghana currency that was declining very rapidly in 2009, now seems to have stabilized. So it's more signs like this than any hard macroeconomic data that lends us to say that we've seen some stabilization. I think in general, particularly in Africa, it's very difficult to get reliable up-to-date macroeconomic data.

And how is minutes of yours doing? I mean, excluding the value-added services, which is obviously going up. But I mean, the call minutes, how are they doing?

Yes, I mean, you have -- I mean, here you have a lot of volatility. The reason we don't report the minutes of use because we have a lot of volatility quarter-to-quarter. It depends on promotional activities. We don't find that necessarily a very good data to share. So, there hasn't been any material difference in the calling patterns compared to previous quarter.

In general, as we said in 2009, there was sort of a tendency for more on net calls, primarily driven by promotional activities and less incoming international calls, primarily in Central America because of taxes on those international calls.

Okay. Thanks.

Okay.

Stephen Mead from Anchor Capital Advisors.

Yes, hi. Can you talk a little bit more about the value-added services in terms of what level of ARPU those services are at in different parts of your operations? And are any of those on sort of a contract basis? And as you look forward, what does the growth of broadband or value-added services mean as far as a CapEx number?

Yes, in general, our value-added services, perhaps, are only unique for a period of time because we tend to be first. It's a little bit more difficult to copy value-added services in comparison to a price change because you have a small IT machinery working in a complex environment. And we focus a lot in terms of education and execution. So, our value-added services over time get copied by competition. The reason we stayed ahead is that we tend to be first and we do it better.

In terms of what the growth could be, our best performing marketing in terms of value-added services is Paraguay, where value-added services represent 39% of recurring revenue versus a Millicom average of 21%. There's no reason why, over time, many other markets could be at those levels.

In terms of 3G CapEx, it's actually on voice minutes. It's more productive or more cost efficient than 2G. And so far, also the margins we generate on data tends to be slightly higher than the margins on voice. So, we have good returns on those investments.

But do you see the impact on ARPUs?

Well, I think one of the key reasons for ARPU, as having relatively defensible ARPU, is the fact that we have that strong growth in value-added services and that's now going to be complemented with strong growth on data. Basically, we are attracting the good users from the market and from our competitors, people to whom the number is important and also higher users. So, we think the focus on VAS truly has helped us in protecting our ARPUs.

And where are you in sort of the banking services and some of the other value-added services actually kind of get you into a different functionality as far as the use of the phone?

Very early on, we basically are going to open up a few test markets in 2010, primarily focused stage one on pure in market transfer, where there seems to be a big need, basically, transferring money from one city to the other. So, we will have a few markets in Latin America and in Africa up and running in 2010, based on in market transfers. And then, that's a platform that we will build other services over time on.

And do you expect the improvement in Colombia, in terms of margins, to continue as the year progresses?

Yes, we said -- yes, we believe maybe there could be some variation quarter-to-quarter, but the guidance we've given is that in two years' time, we believe it can be up in the mid-30s as a margin target.

All right. Thanks.

You're welcome.

Andreas Joelsson from SEB.

Hi. It's Andreas Joelsson from SEB Enskilda. You mentioned good cost control in the quarter. Is there any areas where you have been sort of, say, cautious on costs, any specific area or what is this cost control consisting of?

Well, I guess I should have said continuous good cost control because it's really the kind of a mantra we have. Part of our philosophy always has been that we spend the money as it would be our own. So, just a continued focus on trying to improve processes, look at smarter way of taking the product to the market or energy usage at the site. I mean, it's a continuous challenge to figure out how we can do -- what we do better at the lower cost. So, it should be really continued cost control.

Perfect. Thanks.

(Operator Instructions). Bill Miller from J.M. Hartwell.

Hi. Good morning. Good evening to you.

Good morning.

Just -- are there any other tower deals or anything else like that that can decrease your capital intensity and also increase your margins? And secondly, historically it's been noted that the African margins would progress at about 1% a quarter. Now, taking out Rwanda as the negative, do you still see that as a possibility and there a target of getting to 45%, which is your -- the margin that was mentioned at that time. Is that a possibility for this year or is that something that is going to be further out? Again, ex-Rwanda.

Yes. As far as African margin is concerned, we didn't say 1 point a quarter, we said 3 points per year, which is a little bit less. I mean, a little bit more than a year ago, we were at 35% and we said that we would reach within three years about -- close to the average margin of Millicom, which is 45% -- which was 45% at that time.

So, we are already at 40%, which means that we did half of the way in about one year. So, we'll get there eventually. I mean, we confirm the fact that we will be certainly within less than two years at the average margin of Millicom, which is close to 45%. Sorry, I didn't hear your (multiple speakers).

Tower deals. Any more?

Tower deals, yes. Sorry. So, we are -- as we said in the past, actively working on similar deals to the one that we did Ghana. Maybe under the strategy from configuration because each market has its own specificities. But we are certainly working very actively on similar transaction, both in Latin America and in Africa because we believe that the time when -- I mean, the coverage was a differentiating factor -- is really not -- it's not really the case anymore.

So, we don't see a real competitive advantage except maybe in one or two markets in controlling and owing the towers. So, we expect to announce similar deals by the end of the year.

Good. Thank you.

You're welcome.

Jean-Charles Lemardeley from JPMorgan.

Yes, hello. Just first of all, what's to confirm that you took a one-off charge in Ghana of about $4 million in reduced revenues in EBITDA for the quarter? And then, my second question is on what -- the interest of Bharti. I mean, what are your expectations of Bharti?

And particularly, if you could discuss what you think the ability for them to implement their minute factory model in those African markets. How easy would it be for them to replicate what they've done in India and to be disruptive on pricings? If you would comment on that, it would be great.

Regarding Ghana, I confirm that we took a provision in Q1 for any liability that we are aware of regarding mainly income, VIT and taxes on international incoming calls. But we don't disclose the amount because we are still negotiating with the authorities. So, I can't tell you more about it at this stage.

But that impacted your reported --

Yes, it did. It did in Q1, which is the reason why you may see that actually our top-line growth was a little bit lower. But we gave in our release the top-line growth excluding this provision.

Yes.

Yes, on Bharti, we think Bharti could be a good competitor for us in Africa since they passionately believe in sharing the passive infrastructure given their tower call venture, an outsourcing model they have implemented in India. So, we are also strong believers in sharing the passive infrastructure and they could be a good partner for that.

In terms of the minute factory, that's nothing new. We have already been working on it for many years in Africa. The only difference, I would say, between India and Africa is that Africa really is a high cost environment when it comes to energy, which there is not enough for or even salary costs and so on.

So, I think it will be difficult for Bharti to get down to similar very low level in terms of rates per minute as they would have in India and still make a return. But we already have quite attractive on-net recurring minute prices in our key markets, which are substantially below what you would see on the tariff or rate cut itself because of promotional minutes and other packages we sell.

Okay. And just a follow-up on the tariff sharing, maybe end of tower deals. Can you guide us again on the impact? I mean, the Ghana deal, the impact it has on your financials. So, you -- obviously, your balance sheet, you get an active -- and asset off the balance sheet. The license of the balance sheet -- but has it done -- does it do in terms of OpEx -- of cash OpEx on an ongoing basis?

The benefit that we get from such a deal, at least the way we structure it in Ghana because I'm not sure that all the switch-out deals will be similar. But what actually happened is that we sold afterwards to a tower company. So, I mean, we have a reduction of assets. We have a reduction of depreciation of the consequence, which already happened in Q1.

It's -- the impact for Ghana was about $1 million in the first quarter for Ghana only. We will have no CapEx going forward in the future for towers because the tower company is going to do it on our behalf. And we will reduce our OpEx as well, due to the fact that the leasing agreement that we have with the tower company, the cost of the leasing will be lower than our current operating costs.

So, both -- we are winning both in the balance sheet and the P&L, which is really an indication that this is a good deal. It's not a tradeoff between the balance sheet and the P&L. It's a win-win situation. On the top of it, as far as the leading guy is concerned, we have a minority stake -- a strong minority stake in the tower company, which means that we will get the upside of the fact of having more than one tenant. We're already at close to 1.2 tenants on our towers with this joint venture with Helios in Ghana.

Thank you very much.

Bill Miller from J.M. Hartwell.

Yes, one other minor point. You were talking about the macroenvironment. Is there anything that you can point to in the way of remittance trends that would help the macro [which originally,] obviously, [wasn't macro]?

Yes. We had a -- for the first time I don't know in how many quarters, we had an increase in the remittances into Central America for March. But that's a small number and there's been a little bit -- quite a lot of volatility in those numbers. But that could be one small encouraging sign. I think for the first time in something like six or seven quarters, there was an increase in the actual remittance number from US into Central America.

Great, thanks.

You're welcome.

(Operator Instructions). As there's no further questions, I'd like to hand the call back over to Mr. Grahne and Mr. Roger for any additional or closing remarks.

Thank you, operator. I would just like to thank you for joining the call today and we look forward to seeing you soon. We have now fixed the dates for our market with Tanzania, which will be on the 8th and 9th of September and we will be in touch with logistic details in the due course. Thank you very much and good-bye.

Thank you. That will conclude today's conference call. Thank you for your participation, ladies and gentlemen. You may now disconnect.