Millicom International Cellular SA (TIGO) 2006 Q1 法說會逐字稿

Good day, ladies and gentlemen, and welcome to Millicom’s 2006 first quarter results conference call. For your information, this 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 a presentation summarizing the key features of the results.

I would now like to hand the call over to your hosts today, Mr. Marc Beuls, President and CEO, and Mr. David Sach, CFO of Millicom International Cellular. Please go ahead, sirs.

Thank you, operator, and welcome to everyone who has joined us today to discuss Millicom’s results for the first quarter of 2006. David will run you through the financials with you in detail in a few moments. You will need the slides in front of you for David’s part of the presentation. You will find them on www.millicom.com.

We will both be happy to answer any questions you have at the end, but first I would like to give you an overview of the quarter and run through the performance of each cluster. But before we look at the results, I would like to make a brief comment about the strategic review of the business.

In January the Board of Directors appointed Morgan Stanley to conduct a review of strategic options in response to the high number of unsolicited approaches received by the Company. The Board felt that it was in the interest of our shareholders to examine these approaches in detail and this review has been in progress during the quarter.

Since that date the Company has solicited and received non-binding offers for the acquisition of the entire share capital of the Company. However, there is no certainty as to whether any of these offers will lead to a transaction and, if any transaction is agreed, there is no certainty as of the terms of any transaction. We will make an announcement to the market when the strategic review is complete.

Millicom published record pro forma numbers today, which I will take you to in a moment. But what is driving this growth? The Company decided to take a more growth-oriented approach in late 2004, supported by an increased CapEx spend, improved contribution -- distribution and the launch of the new Tigo brand. This started in Latin America, and has delivered all of 2005 and 2006 exceptional revenue growth, combined with good profitability and strong cash flow generation for the region.

The customer-focused business model, based on affordable, accessible and available services, has since been taken into three countries in Africa, and will be launched late this year in three more countries in Africa and then our South Asian operations.

The first results in Africa are already visible as the region posted the highest percentage increase in total subscriber numbers within Millicom in the first quarter.

Let’s look at the results for the first quarter. As we indicated in our release issued two weeks ago, underlying year-on-year pro forma revenue growth continues to accelerate, and for Q1 2006 we recorded growth of 48% from Q1 2005, up from 40% in Q4 2005 and 33% in Q1 2005.

Pro forma numbers now also exclude Pakcom, our TDMA operation in Pakistan, which we agreed to sell, subject to regulatory approval, to the Arfeen Group for a nominal amount on March 26, 2006. As part of this conditional agreement, the Arfeen Group will also take a 10% [carry] in Paktel, our GSM operation in Pakistan. Millicom, however, will retain full management and operational control of Paktel.

The pro forma numbers we have given in our press release give you the real true underlying growth rate of the business, and let’s start with the subscriber numbers. In terms of subscribers, Millicom added 962,000 subscribers in the first quarter, bringing the total to 9.9m at the end of March, representing a 16% increase over the first quarter of 2005, or a 58% on a pro forma basis.

I’m also delighted to report that we exceeded the 10m total subscriber mark this month. Attributable cellular subscribers increased by 29% from the first quarter of 2005 to 8.5m at the end of March, or by 57% on a pro forma basis.

For those of you who are not yet familiar with the concept of attributable subscribers, these are calculated as 100% of subscribers in Millicom subsidiaries’ operation, and Millicom percentage ownership of subscribers in each joint venture operation.

We introduced this concept last quarter because of the buy-out of more minority partners as we moved to a need for proportional reporting, and our subscribers are reporting using the same methodology as the revenues, giving consistency across our key performance indicators.

With the year end, we have completed the buy-out of our minority partners in Tanzania, Sierra Leone, Ghana, Senegal and Paraguay. We now have 100% ownership of all of these operations.

Let’s turn to revenue growth. A strong subscriber growth translated into high -- the highest quarterly revenue ever for Millicom of $322m for Q1, an increase of 20%, or 48% on a pro forma basis, from the first quarter of 2005. The highest growth relative to the first quarter of 2005 was 142%, recorded in Honduras, or 82% on a pro forma basis, followed by 92% for Guatemala.

EBITDA for the three months ended March 31, 2006 was $142m, a 12% increase from the first quarter of 2005 but, more importantly, underlying EBITDA on a pro forma basis increased by 55%. Millicom’s EBITDA margin was 44%.

Over the last couple of quarters Millicom’s consolidated EBITDA margin has been impacted by Paktel investing in its new GSM business, and by the start-up costs for Chad and DRC.

Millicom’s priority today is growing its top line and taking advantage of the opportunities in its markets, whilst remaining focused on profitability. In 2006 we expect to see a CapEx spend of over $500m, again a 50% increase from last year.

Let’s turn to the clusters, starting with the Central American, which remains our largest cluster in terms of revenue and EBITDA. Attributable subscribers reached 2.2m, up 78%, or on a pro forma basis up 66%. Revenue grew by 77% from the first quarter of 2005, to $156m for the first quarter of 2006, or by 66% on a pro forma basis.

EBITDA for Central America increased by 78% to $79m for the first quarter of 2006, and by 66% on a pro forma basis. The EBITDA margin was 50%. The key driver for this growth has of course been the success of our Tigo brand, which has been rolled out on the back of our new extended GSM networks.

The effects of the Tigo brand can also be seen in South America -- in the South American cluster, where attributable subscribers grew by 54% to reach 1.5m at March 31, 2006. Revenue for Q1 grew by 43% to $44.7m, relative to Q1 2005, and Bolivia and Paraguay produced revenue increases of 31 and 51% respectively over the same period. EBITDA increased by 50% on a quarterly year-on-year basis, to $18.6m. EBITDA margin was 42%.

In Africa -- in our African cluster, attributable subscriber additions since the end of the first quarter of 2005 were over 945,000, including our new operations in Chad and the Democratic Republic of Congo. These additions resulted in a 78% increase in attributable subscribers, to 2.2m at the end of Q1 2006. The subscriber growth continued -- contributed to a year-on-year increase in quarterly revenue of 39%, to $66.7m, and 40% increase in EBITDA, to $29.7m. The EBITDA margin for Africa was 45%.

Attributable subscribers in South Asia increased by 20% from March 31, 2005 to over 2m at March 31, 2006. Revenue was $28.4m for the first quarter, and EBITDA was $5m. Revenues in South Asia have been under pressure as a result of the lower sales by Pakcom, following our decision to sell the business and focus solely on Paktel, our GSM operation.

Pakistan is very -- is a very competitive market. We are in the process of resolving network interference issues that occurred at the end of 2005, and our plan for Paktel is to continue to increase the quality of the network and focus on revenue-generating subscribers. Our target is for Paktel to relaunch the business in the third quarter this year on the back of a substantially improved network.

In February our operation in Sri Lanka had its license extended until 2018. This operation continues to perform well, producing an EBITDA margin of 58% for the first quarter. We are investing heavily in both of these South Asian operations, and expect to see the first results of this investment in Q3 this year.

Turning to South East Asia, subscribers for Cambodia and Laos increased by 31% from the first quarter of 2005 to almost 925,000 at March 31, 2006. Revenue for South East Asia was $25m for the first quarter, up 21% on a pro forma basis from Q1 2005, and EBITDA was $10.1m, up 23% on a pro forma basis.

Now I would like to hand over to David to talk you through the financials. David, please go ahead.

Thank you, Marc. Good afternoon to you all. I’m going to take you through some slides on the key financials. You will find the slides on our website, at www.millicom.com. It will help to have them in front of you.

Turning to slide two. The numbers on this slide speak for themselves, and show how growth has accelerated strongly, following on from what was a very positive final quarter in 2005. Pro forma attributable subscribers were up 57%, revenues up by 48%, and pro forma EBITDA up by 55%. In Q1 Millicom nearly doubled its CapEx to $95m, compared to Q1 last year, which will help sustain these high growth levels going forward.

Moving on to slide three. Attributable subscribers have grown by over 50% year-on-year for four consecutive quarters, which is a great achievement and reflects the fact that many of Millicom’s businesses are now well into the J-curve of emerging market growth, particularly in the more established markets in Latin America.

For Millicom, this higher rate of growth started in Latin America with the adoption of the AAA strategy of investing heavily in the networks to make them available on demand, investing heavily in distribution to make our products accessible when needed, and finding innovative ways to make our products affordable to as many customers as possible. When this strategy is firmly embedded in each of our operations, we are then ready to relaunch our services under the new Tigo brand. This is when we start to see the accelerating J-curve.

Slide four shows the pro forma revenue growth is accelerating on the back of the strong subscriber growth, as you would expect. In Central America, where penetration rates are higher and networks are more developed, pro forma revenue growth was 66%, reflecting the strength of the market over a sustained period. We are already seeing similar rates of acceleration in South America and expect our other markets to reflect this pattern upon the launch of Tigo.

On slide five you can see the pro forma EBITDA margins are rising, due to the economies of scale inherent in our business model. As we continue to invest in and expand our networks, a greater percentage of our revenue will come from calls made within our own networks, which is more profitable revenue because there are no interconnect charges associated with these calls. Furthermore, as we grow our businesses even more, we will be able to leverage our G&A costs, and possibly even our sales and marketing costs, across a bigger business, improving our margins even further.

Please turn to slide six. The recent step change in EBITDA growth to 55% for Q1 2006 versus the same period last year is a direct result of the investments that we are making in our businesses, particularly the high level of CapEx spend in the second half of 2005. As we accelerate our CapEx spending in 2006, we expect to see continued high EBITDA growth going forward.

Slide seven shows that in 2005 the average level of CapEx was $88m per quarter, almost as high as the full-year totals in 2002 and 2003. The $170m spend in Q4 2005 was exceptional, but unsustainable for us quarter-on-quarter. During Q1 2006 we needed to digest and embed this level of spending into our businesses, to ensure that these investments were properly implemented in order to receive the maximum benefits from this unusually high level of investment. However, our competitive markets and internal analyses are pushing us to continue this high level of CapEx spend, due to the results that it is producing.

As mentioned before, we nearly doubled the CapEx spend of Q1 last year and expect to invest at least 50% than the full year 2005 total. Our challenge is to continue to spend this money efficiently and effectively, which is our commitment to you. 2006 will be Millicom’s peak year of investment. After that we expect to see CapEx fall as a percentage of revenue, although 2007 will be another year of high CapEx spend.

In my year-end presentation I mentioned that there were several items that were only temporarily impacting profitability in 2005. As you can see on slide eight, these are no longer dragging down our profits, resulting in a significantly higher net profit of $33m in Q1 2006.

As anticipated, depreciation and amortization, write-downs and the overall Group tax rate have all fallen. Interest did not fall because we are still recording the interest charge on the Pakcom license payable that was applied in Q2 2005. But this will end when we get regulatory approval to dispose of this business.

Let me now focus on the three items that helped drive the significant increase in profitability below the EBITDA line - write-downs, other income, and taxes.

Slide nine shows that there were several significant write-downs in Q1 last year. But as we explained at year-end these are one-off in nature and similar items are unlikely to reoccur because they are associated with businesses that were disposed of. We are not anticipating any significant disposals requiring write-downs, going forward.

Turning to slide 10, there has been a $17m improvement in profitability from other income. This improvement is almost entirely due to the movements in the revaluations of the Tele2 shares, the related 5% exchangeable debt, and the embedded derivative associated with this debt. The impact of these revaluations appears on three lines of the profit and loss statement, including the exchange line, which is where the exchange rate movement on the debt is booked. Remember that the Tele2 shares and debt were denominated in Swedish kronor. These revaluations are a non-cash movement and will soon be gone when the Tele2 shares are exchanged for the debt in August of this year.

Slide 11 shows that the overall blended Group tax rate for the operating companies has fallen from 35% to 31% versus Q1 last year. This is due to the actions that we are taking to better manage our tax position. As more of our operations become profitable, tax-paying entities, we are able to increase the management fees and brand fees that the companies pay to corporate, creating additional tax deductions for the operating companies and creating more income at corporate. These changes help to bring down the overall tax rate.

As pointed out at year-end, the overall consolidated Group tax rate was extremely high because there is too little income of corporate to offset the costs, particularly the interest expense on the corporate debt. These non-deductible corporate costs drive up the overall consolidated tax rate. With the significantly higher pre-tax profits from the operating companies in Q1 2006, the impact of these non-deductible corporate costs is proportionately much less, resulting in a significantly lower overall blended tax rate.

In addition, as mentioned above, we are now able to move more income into corporate because of the increased profitability of the operating companies. As a result, the overall Group tax rate has fallen to 44%.

Slide 12 shows you how Millicom’s model of pre-paid services continues to benefit our cash flows. Under this pre-paid model, we get our money paid up-front and therefore we get a positive working capital movement when we grow our businesses. This is a fantastic model, as our cash flow from operations is higher than our EBITDA and, as you can see, growing as a percentage. This dynamic helped fund a portion of the much higher investments of $176m, comprised mainly of CapEx and buy-outs of minority partners in Senegal, Tanzania and Sierra Leone.

Finally, the bank closing balance of $526m puts us in an extremely strong position, since we can use this cash for investing in the business on an ongoing basis.

Thank you. I now hand you back to Marc for a summary.

Thank you, David. The Q1 performance was well ahead of expectations, with pro forma year-on-year revenues up by 48% and acceleration of growth from what we have -- had been a very strong Q4 quarter last year. The increasing growth momentum is the result of the heavy capital expenditure across the business in 2005, and we expect this growth pattern to continue, as our capital expenditure in 2006 will be more than 50% higher than in 2005.

Two important factors will impact the results of the Company, starting Q3 this year. First, the relaunch of Paktel in Pakistan, our most populated market, and second the launch of the business under the Tigo brand in the DRC or Congo, our second most populated market. These events will have a profound impact on the growth of the Company and therefore the prospects for 2006 and 2007 are excellent.

This completes my comments and we will now be happy to take your questions. So operator, may I please have the first question?

Thank you. [OPERATOR INSTRUCTIONS]. Our first question comes from Adrian Dawes from J M Hartwell. Please go ahead.

Marc, wonderful quarter. This is actually Bill Miller. Just calling on the CapEx cycle that we’re playing here, it seems to me that the efficiency with which you have spent your money in the fourth quarter and on into the first quarter really just indicates that your growth rate is going to accelerate well beyond.

Can we assume that the growth rate that we saw -- the pro forma growth rate that we saw in the first quarter is going to be typical of what we see this year and into 2007? Because you’ve said you’re going to spend about $530m and you’re going to spend it with the same effectiveness, and that implies something around 50%.

We expect to continue seeing a high growth rate, this -- continue to see it this year and also into next year, for two reasons. The first one is, as you said, we will continue to increase our CapEx, so the numbers for this quarter, Q2, Q3 and Q4 in 2006 will be higher than the numbers in Q1, and that will lead us to a number over $500m. So that by itself will generate a lot of growth.

The growth we expect going forward to come more out of our existing markets in Africa, but also markets like Sri Lanka where we are making important CapEx investments.

Secondly, as I said on the call, there are still two countries, we can more or less call them start-ups, Pakistan and Congo, that have -- really haven’t been delivering any growth over the last two quarters. Pakistan because of the interference issues and the network issues, Congo because we bought a skeleton and we are rebuilding it. We have to sign contracts with suppliers and we will launch that business in the third quarter. So, given the population of those countries, they will also be important contributors in this as revenue growth going forward. So we have a very bullish outlook for this year and next year.

Marc, just a follow-on in terms of growth. The ability to purchase minority interests is an additional factor as well as the possibility of Iran, which you didn’t mention on the call, but maybe you could give us an update on that.

And then finally, any other wild cards that may be on the horizon, or if you could give us an update on the status of Vietnam as well.

In terms of minority, we’ve done quite a lot in the first quarter. There might still be more to be bought out but we want to make sure that when we buy out minority partners we do this at the right price. And if that is not possible, then we will not do it and we will continue operating the business jointly.

In terms of Iran, there’s not much to report. The situation is status quo compared to the previous quarter. The business has not grown from the 450,000 subscribers we had at year-end 2005, and we’re still in talks with RIC and TCI and the regulator to resolve the interconnect issues. So there’s no conclusion there yet.

In terms of the wild cards, Vietnam - there also we continue to negotiate. We keep hearing warm words from decision takers and we expect that once the party congress is over - I think it’s ongoing as we speak - that Vietnam will start taking the necessary steps to move into -- move to a privatization of its telecommunications sector. So we remain hopeful that also in Vietnam we’re going to be offered the opportunity to buy into the existing business called VMS or MultiFon.

Has Vietnam continued to grow, even without your steady hand on the tiller?

Yes. Although we don’t have any clear insight of the numbers, but at least from what we read in the press it looks like the market continues to grow. But at the same time you read that there’s a need for further tariff reductions in order to further accelerate the growth. And with a fifth operator coming to the market any time now, we expect that it’s going to be a very competitive market but a very attractive market, because of the still relatively low mobile penetration. So there’s still a good outlook there.

Great. Thanks very much.

Thank you, Bill.

Thank you. Our next question comes from Lena Osterberg, with SEB Enskilda. Please go ahead.

Yes. I just had a question on the write-downs. Last quarter you took, I think it was $16m, on Pakcom, which related to Pakcom’s outstanding loans, and this quarter you reversed that by $4m. So I was wondering why, if you’re selling Pakcom now, you’re not reversing the whole $16m. So what’s that discrepancy related to, if we should expect more in the following quarters?

And then my second question would be, if you could give any indication of the launch costs for Congo?

And also potentially if you will incur further launch costs and margin erosion in Paktel, as you say you will, you will push that on GSM again?

Okay. I’ll answer the second part and then David can later on answer the first part of the question. In terms of the launch costs involved in DRC and Paktel, there will be a launch cost. We said that at the end of last year, that when we launched or relaunched those businesses there will be an impact on our EBITDA. We have not given a number as to what that launch cost is going to be, because we will be focusing on very specific distribution methods in Pakistan. And also we’ll be focusing on, to start with in Congo, the cities where we expect that, with our efficiencies in terms of sales and marketing spend, we can get a lot of revenues and subscribers against the money spent.

But yes, as we’ve said before, there will be an impact in the third quarter when we launch those businesses. I would like to point out, however, that when we launched Tigo in Latin America in 2004, for instance, there was hardly an impact on the sales and marketing level and on the EBITDA margin. So we are very effective and efficient operators when it comes to launching new brands and launching networks, so it will not be a bloodbath, I would think, from an EBITDA point of view.

So, David, could you comment on the write-downs for Pakcom?

Sure. Hi, Lena. At the end of last year we effectively locked in the sales price for Pakcom. And under the accounting rules, because we had two businesses in Pakistan, Paktel and Pakcom, because we’re getting rid of Pakcom it’s not a discontinued operation. Therefore, we still have to take the losses through the P&L, and it’s mainly the interest on the Pakcom license that I mentioned in my presentation that causes a loss. And therefore, because we‘ve locked in the sales price we effectively reversed that loss.

So going forward, if we continue to -- hopefully, Pakcom is approved by the regulator soon, and it’s disposed of. But if we continue to have to record some losses then we will have a further write-down in future, just to get it back to the locked-in purchase price.

But my understanding of the $16m that you took in Q4 when you explained that, that was the maximum liability that you would have to take on if you were to take on Pakcom’s loan and not sell it. And you have sold it. Shouldn’t that then be reversed?

Yes. But as part of the sale, we will have to repay those loans. We will look to get those loans into the Paktel business, but we do have to take those, so we have to -- that $16m had to go through the P&L, so we couldn’t reverse that $16m.

So, as part of this process, you’re taking on the external loans of Pakcom, plus you’re giving away 10% of Paktel. Is that correct?

Say again? We’re taking --

You’re both taking on Pakcom’s external loans and Arfeen gets 10% in Paktel. Is that correct?

Correct.

Okay. Then I understand. Thank you.

You’re welcome.

Thank you. We’ll now take a question from Peter Nielsen with Cheuvreux. Please go ahead.

Thank you. Peter Kurt Nielsen from Cheuvreux. A couple of my questions have been asked, so I have two quick ones. One relates to the African business, where you’ve shown an impressive margin, despite the launch in new markets. Could you tell us whether this is because we have most of the start-up costs still to come in coming quarters, or is this simply a level of the margin you expect to continue in Africa, going forward from here?

And then, related to -- it seems as if the corporate costs have gone up, from a low level admittedly, but gone up in the quarter. Is that a new level for corporate costs we should factor in going forward? Thank you.

In terms of the Africa EBITDA, because of the very successful launch in Chad, the business that was launched in October last year and today already has a market share close to 40%, we’ve had very, very small losses in the first quarter. If anything, probably already a small profit. This company already operates at an EBITDA margin of around 30%, as we speak. So the start-up of Chad was almost negligible in the first quarter.

DRC, we’re just running the business with the 60,000 or so subscribers on the network, and we’re not really doing much, so there’s no cost there. As I said earlier on the call, we expect there to be a cost in the third quarter, when we launch Tigo on the back of a substantial better and bigger network in that country.

David, maybe you can comment on the corporate costs?

Sure. The corporate costs are higher than usual for two reasons. One, the strategic review - we’re obviously incurring costs related to that strategic review. And then there was a one-off legal fee settlement that we had at corporate of $1m. So that’s temporarily causing our corporate costs to increase.

Alright. Thank you.

You’re welcome.

Our next question comes from [Anders Wenberg] with REM. Please go ahead.

Hello. Anders Wenberg here from REM. Looking at your ARPU trends for the last three or four quarters, you’ve had a very stable offer, driven by a mix shift towards Central America and also a little bit of slow down in the [indiscernible] markets, and I think Central America’s given up a little bit. What could you expect of ARPU going forward, particularly considering that Paktel is being divested? Will it actually increase a little bit? Do you have any guidance on that?

In terms of ARPU, we saw last year, on the back of the launch of Tigo in Central America, we saw an increased ARPU. The reason for that was that by building better and bigger networks, using up to date technology, GSM, that we attract better customers. So we were moving up in the market and we were able to sell services in market segments we hadn’t been able to sell services to in the past. So that continues in Central America.

On top of that, in South America we’ve had some very good experiences. In Paraguay, for instance, where we launched in November last year the per-second billing, as well as the e-pin with very low e-dominations. Today you can -- for $0.33 you can buy airtime in Paraguay. The combination of the two has allowed us, for those people who are using those e-pin services, we’ve seen the ARPU increase from $8 to $11. So this is the combination of a number of initiatives in Latin America.

We very much expect the same to happen in Africa and Asia, when we build better and bigger networks, that we will also be able to move up in the market and start selling to higher market segments. Don’t forget that in Africa, we completely ignored that continent for two to three years in the early part of this decade because of our restructuring. And so we were never the preferred operator. So now that we’re becoming a better operator, we’re becoming a preferred operator, and that should allow us to attract bigger -- better customers with higher ARPUs.

I very much expect to see the same happening in South Asia, once our Paktel network is relaunched, and also in Sri Lanka, where we are substantially increasing the size of our network. So we expect to see more of that.

So from a modeling perspective, we should look at stable ARPUs in Latin America for the next few quarters and a significant slowdown in the ARPU decline in Africa.

I would expect that to happen, yes.

Thank you.

Thank you. We now have a question from Bengt Mlleryd with Nordea. Please go ahead.

Thank you. Bengt Mlleryd from Nordea here. I just have a question around Pakistan, the dynamics and your view on the market. Have you underestimated the competition, or is your view on this kind of network, it’s necessary to really be competitive to watch competition change there? I mean, looking here at the first quarter, you’re taking growth by 5%, versus the market growing with 18%. So what kind of CapEx assumptions do you have going forward, or has your view on this market changed during the last couple of months?

No. I don’t think we underestimated competition. What did go wrong are two things. First of all, there’s the frequency interference. The fact that we are using the extended GSM band makes that we have interference into our network other operators don’t have. And we are in the process of resolving these issues by getting more 1800 spectrum so that we can move away from these interferences.

But more importantly, I think what we did not have was a solid network in Pakistan. Don’t forget that the GSM network in Pakistan was designed in early 2003 when we were still coming out of our restructuring, so it was really done in a very low-cost way by building an overlay on the existing PDMA network. Clearly, with the subscriber increase we’ve seen over the last twelve months, we did not have the grit, the structure to build this network on and that’s why we need to reconfigure the network and that’s what we’re doing as we speak.

Once those things are in place, once the interference has gone and once the -- we will have a bigger and better network, we will be able to compete with anybody in that market and we’ll be able to grow in that market at least the speed, and hopefully with a higher speed, than some of our competitors. Anyway, that has been the experience in most of the other markets we are operating in, in Africa and Latin America these days, so we expect to do the same in Pakistan.

May I have a follow up there? But isn’t that the case that you have six competitors, you are number six in the market, while you’re really successful in Central America. Do you still regard Pakistan as an interesting market, given here that fiercely competitive [TeleNova Read] or [Es Conferencion] both?

I don’t know any market that is not competitive within Millicom, so the days of monopolies are long over. And we are competing in Latin America with some of the biggest operators in this world, so we are operating in Africa with very big operators. So I don’t see a reason why we shouldn’t be able to do that in Pakistan. So we just need to sort our issues out. We were not prepared for the growth, we did not have the quality network in place that we should have, in order to be competitive in the market.

And as I said earlier today, it is when you have that quality network in place, it’s then that you see the growth coming in, because one of the factors is that you are able to move up in the market and to start selling services in market segments where people typically produce high ARPUs. And that’s what we expect to see also in Pakistan going forward.

So what kind of CapEx assumptions do you do for building that quality network in Pakistan?

We haven’t given any numbers, but I think the numbers we gave last year I think were $100m to $150m on an annual basis. And the focus will be on building the capacity in the places where we have our customers today, so as to make sure that we at least produce -- provide the quality service to them.

Thank you.

Thank you. [OPERATOR INSTRUCTIONS]. We now have a question from Sumit Saigal with Credit Suisse. Please go ahead.

Hi, good morning. Was the Pakcom transaction a gearing issue as far as the timing of the strategic review is concerned?

Unfortunately I won’t be able to give you any more than what I’ve said in this call -- or the beginning of this call. So, the review is ongoing, and we will report to the market as soon as there is news to be reported. So apologies but I can’t say more than that.

Okay. Well, do you feel more confident about the strategic review now, as to several months previously?

Again, I can’t answer that question. In terms of the strategic review, we will only report when we get to the end of it.

Alright. Thank you.

Thank you. We now take a follow up question from Peter Nielsen with Cheuvreux. Please go ahead.

Thank you. Just a quick one via the 10% stake in Paktel you’ve given up as part of this transaction. Does that come with a buyback option?

David, do you want to answer that one?

No, we cannot go out and get that back unless we sell all of our stake in Paktel. So if we wanted to fully get out of that business then the fees would have to come along. So there’s not a buyback per se, but there is a drag-along right. So if we wanted to get rid of all of it, we could drag them along.

Okay. Thank you.

You’re welcome.

Thank you. We now move to another follow up question from Adrian Dawes. Please go ahead.

Thanks. The possibility of further EBITDA margin improvement, Marc. Historically you’ve indicated that an overall rate of 50% was possible, obviously with Pakistan and [indiscernible] maybe that’s not feasible right now. But can you get further margin improvement in those countries like South America, and get back towards that 50% EBITDA rate?

And secondly, how low can the tax rate go?

And then finally, do you have a way of helping us measure your capital efficiency, or is there a correlation that we can make with CapEx and growth rate?

In terms of the EBITDA margin, we were at the 50% overall at the times when we were operating in Vietnam, where we were benefiting from a very high EBITDA margin given the structure of the BCC. Having said that, we showed that in a very competitive environment, like Central America, we can operate at a 50% EBITDA margin. We have been improving margins since last year in South America, and we expect that we can see further margin improvement there.

I think Africa, we should be able to improve the margins. Clearly, South Asia will substantially improve its margins over time. So I would expect, once we have the launches and relaunches in Pakistan and DRC behind us, we should be able to start moving in the direction of the 50%. Whether we’re going to get there will to a certain extent depend on regulatory and competitive environment. But we’ve not given up on getting back to 50%.

In terms of the capital efficiencies, you can use ratios like CapEx per subscriber or CapEx per revenue, but given that we are in this massive build out, I don’t think these ratios would make you any wiser. All we know is that we are extremely return-oriented. We also negotiate extremely well with suppliers and getting the best prices as a result of that. So I would think that we are very efficient in terms of our capital spend, because every spend that is made typically will have to be paid back within two years and produce IRRs of 20, 25%. So I think we are very efficient.

David, maybe you can comment on the tax rate?

Sure. You can see the tax rate from the operations was 31% this quarter. I think we can take that even lower. Part of that is that we’re incurring losses in the Paktel business and we have the start-ups in Congo that Marc mentioned before. I think when they become profitable and we can have more realistic tax rate there rather than losses that we don’t get any tax benefit for, I think we can drive the operations below 30%.

And then it becomes a matter of solving the corporate issue. That’s all about the -- more of our operations becoming profitable from a tax perspective. And then trying to shift income from the operations and get a tax deduction at the operations, which will drive that rate even lower. And move that income to corporate to offset some of the costs there, including the interest. And particularly when some of the debts, the 10% senior notes, when we’re able to pay that off in 2008 and possibly move that debt down to the operations, that will help even further. So over the long term I’m sure we can get it below 30%, all things being equal.

Great. Thanks very much.

You’re welcome.

We have another follow up question from Anders Wenberg with REM. Please go ahead.

Hello. Anders Wenberg from REM. On the same issue, could you help us once again understand how much the interest charge is going to fall after the Tele2 convertible is disappearing, and after the Pakcom license disappeared, due to the sale, and also in ’08 after the bond can be bought back? Thanks.

David, you want to answer that one?

Sure, okay. The interest, and I’ll give you some quarterly numbers here. The quarterly interest expense on the 10% notes is roughly $14m, so in 2008 that will disappear. The 4% convertible is roughly $4m a quarter, and the 5% exchangeable, that’s about $7m a quarter and that will disappear in August 2006. So in total there’s $25m a quarter from those three, but -- that will go away. And then it’s depending upon whether it gets replaced down at the local operating company level. So hopefully that answers the question.

Sorry, and the Pakcom license? I understand you have capitalized the future payments on this one, and then you -- that runs through the --

The Pakcom license, because it’s payable over 13 years and we present valued it, therefore we recognize an interest cost on that of about $4m a quarter.

Okay. And that would actually disappear --

That will disappear as soon as we get regulatory approval to sell Pakcom.

Okay. Thank you.

You’re welcome.

Operator, can we have the last question? Just one.

There are no further questions at this time. [OPERATOR INSTRUCTIONS]. No, it seems that there are no further questions at this time. I would therefore like to hand the call back to --

Once again I would like to thank you all for participating in today’s conference call. If there are any further issues you wish to discuss, we will be happy to deal with these on a one-to-one basis, if you wish to contact us directly, or alternatively you can call Shared Value at +44 207 321 5010. Thank you very much, and have a great day. Goodbye.

That will conclude today’s conference. Thank you for your participation. You may now disconnect.