Spark New Zealand Limited

Good morning, everyone, and welcome to this briefing on Spark New Zealand's results for the 6 months ended December 2018. Simon Moutter, Managing Director speaking, and alongside me is David Chalmers, our CFO.

Look, I'm genuinely pleased to be presenting another set of solid results, because I took over tough industry, and it takes a lot of hard work across a lot of improvement initiatives to consistently deliver strong results. And at the headline level, Spark is holding on to its revenue, as the old world of landline and voice fall away and the new world of mobile and cloud IT take its place.

We're holding our overall market share by delivering great value and good service to our customers. And we're grinding our costs through our simplification, digitization and automation programs, now supported by Agile ways of working.

And all of those things have helped us deliver a 4.3% lift in adjusted EBITDA, which is a pretty solid performance for a telco anywhere in the world. Spark can now also confirm and I'm very proud to say that we're the first telco in the world to adopt Agile across our entire business, the benefits of flowing, and we're getting better at it every day.

And we've already had visits from four other international telcos that have come to understand how we've achieved that and are considering doing the same. So it's a genuine international interest what we've achieved here.

In particular, you would have noticed the shifts into the wireless -- our wireless business. The market forces have seen a slight slowing in market revenue growth.

We're still holding our own, and we're in positive territory. Right now, there's a couple of our competitors that's now moving into negative.

But we've moved strongly, and I think the industry is working very hard on margin improvement through the focus on customer value and profitability. We're also seeing significant changes in the handset buyer behavior.

These are generally beneficial to us. We sell through handsets as a part of creating a service for customers, that's not a business that we make any money on.

It's part of providing a complete solution for customers. It's very evident now that we've seen volumes pulling away, particularly in the mid- and higher-range devices.

I think they would anecdotally indicate that customers are hanging on to the devices a bit longer, which make sense to us. We've also seen the vigorous moves by our Lean with some of the pricing practices having shifted a group of customers up above $1,500 price point.

That's holding the average price, but it is really through pricing initiatives of the OEMs. It's nothing that Spark contributes to.

And we're now definitely seeing resistence at the very top end of the device range. Once you get above $2,000, we're seeing material slowing in the volume.

So you might have noticed that we saw an inventory impact of that in the half year endpoint, where our normal volumes and orders at that level did not move through. So it's quite interesting to observe those changes.

We're also seeing a range of pricing and margin-focused initiatives in broadband contributing to an uptick in broadband margin for the first time in a long while. And we're very positive about seeing some margin growth here and hopeful that we're back on the path to order more sustainable levels of broadband margins in the longer run.

And a little bit of slowing in our overall cloud security and service management revenue growth trajectory. Actually, cloud underneath is still very strong and in double digits.

The flatter proportion of it is really the service management element. And again, that's a lower-margin business that we provide support for transition work into the cloud.

And we're not too worried about where the revenue trade goes on service management, but a slight moderating from the competitive intensity around cloud and security business. Moving to the key areas of focus, I'm referring to Slide 3 of our presentation deck.

These areas we signaled a couple of years ago, sort of the main instances for the business to continue on our strategic path. We're very, very focused in this company on wireless.

We believe strongly in the wireless story and that it offers many benefits to both customers and to our business over elements of the fixed business. So strong progress now, gone through the 58% of our broadband base now, lifted off the company.

We've been very pleased with that performance. Wireless broadband continues to grow and had another little lift up to 129,000 at the half year and continuing good adoption of Unlimited mobile now moving into the shareable concept on those plans with good uptake.

And I think it was very notable in the half that Spark, for at least a decade, has had more prepaid customers than postpaid. At the end of the half, we tipped back through that point where our postpaid connection base exceeds prepaid, which is not a good indicator of focusing on the higher value customers.

And the Spark doing well in that zone. On the price-sensitive end, our Skinny players continue to work very well.

The sub-brand is doing a good job and is playing a much bigger role now in the broadband market. And we've just introduced the new SD-WAN product for more managed data service, which is attuned to the more pricey sort of do-it-yourself customer base in the managed market.

So continuing to work well there. On our lowest cost operator scenarios, we met our indicated target for our net labor costs, as we continue to drive simplification, digitization and automation through the business.

And it's notable that we tracked through that $470 million net labor run rate at the -- during the half. And I'd just remind you that only in March 2017, that number was $580 million.

It's a very, very strong performance in terms of improving the productivity of our product organization. We have also seen those initiatives deliver tremendous changes in the service experience for customers.

Back in the mid-2000s, when I was the Chief Operating Officer in old Telecom, we took a million calls, a million voice calls a month from customers. That was our sort of sustained number.

By the half ended December 2017, we were down to averaging 360,000 calls. And in the half just gone, we're down to only 250,000 calls per month.

So literally running at 1/4 of the inbound calling volumes that we had 10 years ago on a much bigger customer base. So that's proof point that digitization and self-service initiatives are working.

And it was also notable that in December -- November and December of 2018, over 10% of all of our inbound communications with customers were handled by virtual assistants. So that also is a significant milestone.

Swinging through now to some of the strategic issues. I want to refer quickly to 5G on Slide 10 of our deck.

We want to reemphasize strongly that Spark is a believer in 5G. We strongly believe in the technology and the difference that makes to our economics and to the capacity and performance of wireless as a genuine alternative to fixed networking.

And we continue to get ourselves prepared for that. So we've been working hard on 5G capability.

We've got a lab operating. We've got test sites up.

We've completed most of our tests and evaluation programs. We're starting to densify the network in the key areas.

And we're -- we simply await now the auction of 5G spectrum. We'd like to be up and running by July 2020 at the absolute latest, and the sooner the auction can be completed, the sooner we'll come to market.

We're ready to go. We do have multi-vendor capability in the network.

So while the decision by the GCSB is a big pain for us in terms of [indiscernible] to that as terrific technology and it being a very strong performer, it does not alter our plans or the timeframes for our intended launch and participation in 5G. Moving forward to Slide 14, just quickly on Spark Sport.

We have made some announcements today on Spark Sport. So just to reaffirm, we're coming to market with this over the next few weeks.

We're very on track with our platform build. We've got a world-class platform being driven at the back end of that.

We're in internal beta at the moment. So we have live users on the platform, and we'll go into a public beta in March.

And we're set and ready to go for that. It's a -- it'll be a terrific news as it will change the game.

It will be much better than the sports offering available in New Zealand today, and it will be available on -- over the next few months will become available on at least 5 million screens in New Zealand for those who wish to access it. I wanted to emphasize it will be a premium offer.

So we've talked about the packet price today of $19.99 per month for the sport content that is, of course, the premium, the live sport and the premium content. We will also have a live re of free content so that all New Zealanders can take the service, become a member and access free content that's available in the Spark Sport service.

Moving to Slide 15. I was trying to think of a great way to sort of convey the benefits of Agile in a not such a generic manner.

And so we chose to offer you a taste of its success through a case study. This is a very impressive result.

A very substantial and important customer of Spark is the Network for Learning in New Zealand. They provide the connectivity service and managed Internet services to 2,800 schools in New Zealand, all fiber-enabled.

And we've always been the back-end provider to their service. Pre-Agile, in the days when we built this network, the left-hand side of this chart gives you a flavor of how we went about that program and sort of seeing that pilot taking 9 months migration over 40 months to get those schools migrated from their previous broadband technologies on to the new platform.

Comparing with that complete network and technology refresh, we just executed an Agile, where the customer joined our squads, the collaboration was amazing and the design and build. We had piloted in 5 weeks.

The migration was underway within 6 months after those pilots were fully tested. And we're now on schedule to complete the entire rollout in 14 months.

So this is a terrific example, and we have many, many other examples of that type of -- it was a great way to convey what the new world of Agile can do. And the tribe that runs at -- our Managed Data tribe that's currently running on an employee Net Promoter Score level of over 70%, which is an absolutely world-class level, up from something like 30% for the same folks in involved in the area pre-Agile.

So the benefits are certainly evident. And if I could just close on Slide 20, just a quick comment on our indicators of success.

As you know, we do like to make it clear what we think drives long-term value in this business each year and keep our business updated on those. We're on track, ahead or completed on 13 of the 17 measures that we laid out at the commencement of the year.

Let me just comment quickly on the 4 that don't -- that aren't quite there. The first, of course, is clear pathway to 5G.

And I'm just really noting that at this point, the risk here is if we don't have an auction, we don't have a 5G launch story. And that now is firmly in the hands of government.

So we do await that and hoping for a timely move forward on the auction. Our voice-only copper connection substituting wireless, so we did outline a plan and we do have an excellent new technology alternative for voice-only customers.

This is your customer who has no broadband and they're still on a landline-only service. We're substituting that to a better product that's on wireless that provides better value.

It's -- selling the sales process is proving difficult. This is a customer base who are wary of new technology and wary of sales, people banging on the door or ringing them, they're worried about scanners.

And so we haven't quite cracked the sales process. But certainly, we're in the game of technology and it's very well received and the value of the new offers is good.

So we're determined to continue. I think the target is looking very tough, but we're going to do our best to get us close to that number as we can at the end of the year.

And in just -- on mobile service revenue and cloud security and service management revenue, we have pulled those targets back just pragmatically, given the shift in market in both of those business areas where we're seeing more focus on margins and execution and a little more product competition around the revenue track. So I think just pragmatically, we should reseat those until it's clear what we think the next few months can deliver.

So those are my opening comments. David, can I hand you to run through some of the key financials?

Thanks, Simon, and good morning, everyone. I wanted to start with a few comments around the accounting changes, given that FY '19 is the first reporting period following three significant updates to our external reporting.

The first of this is the adoption of IFRS 15, which is revenue from contracts with customers. The second, the early adoption of IFRS 16, which covers leases.

And thirdly, trying to, again, disclosure relating to our long-term investments, which principally impacts the reporting of dividends with Southern Cross. This last change has also resulted in updating our definition of EBITDA, which you'll see in the presentation is now EBITDAI, noting that in addition to the normal definition of EBITDA, that number now excludes investment income.

A full explanation of these changes and reconciliations were detailed for the market in our update on 4th of December. I won't go back over these in detail.

But in summary, the three changes that impacted on our FY '19 guidance, where revenue guidance decreasing by $70 million. That's principally driven by IFRS 15 having a negative $75 million impact on revenue.

EBITDAI guidance increased by $40 million, largely driven by $30 million decrease in -- from IFRS 15 and a $70 million benefit from IFRS 16. And then finally, as we noted earlier, the EBITDAI guidance no longer includes the Southern Cross dividend.

The impact of both these also changes impact, and that was why on December 4th, we updated our full year guidance to lower EPS by $0.01, which is simply a translation of those accounting changes. Despite those changes, which are complex and difficult to work through, there's no associated change in cash flows.

And so as a result of these changes, one of the things we've updated in the deck today is a new underlying free cash flow measure. And I'll come to talk about that later on as to what that is and how it is that we're using it.

Turning now to an overview of the key financials results for the first half of FY '19. Spark recorded operating revenue of $1.754 billion, a decline of $7 million or 0.4% over the previous period.

Despite this revenue decline, reported EBITDA was $489 million, an increase of 7.2% over the first half on a reported basis. And if you compare that back to the adjusted number that we gave for the first half 18, that is inclusive of the cost at the time for the Quantum restructuring, EBITDA growth is $20 million or 4.3% higher on an underlying basis.

It's also worth noting that within this first half, EBITDA results and something that we've not adjusted, in keeping with our normal practice, is a write-down of $5.3 million in relation to an investment we've made into global IoT startup. Reported net profit aftertax for the period was up $153 million corresponding to EPS of $0.083, which is down 5.6% against reported first half '18.

I just want to make take a moment to step through 5 drivers of these financial results. The first one of the revenue lines, while overall revenue fell by $7 million, we actually delivered gross revenue growth of $45 million.

This growth was driven across our three principal areas, being mobile, also broadband and cloud. Starting with mobile.

What was pleasing was the majority of the $9 million of mobile revenue growth was driven through service revenue. Now the service revenue part is typically the higher-margin part of our mobile business, and that came through both from an increase in mobile connections of just over 27,000 and also growth in several of our operating lines, in particular from Skinny, coming through.

What we do see in the market is while it's been a challenging market, as Simon outlined, we're very pleased with the way that we're performing. And so if you look at the percent of growth that we're taking in, in the pay monthly market, we're kind of 43.5% of total market growth within that period of time and more than offsetting the decline that we see in prepaid as we continue our focus on exiting some of the heavily discounted segments of the market.

Overall ARPU in mobile was down $0.02 to $27.56. And that was made up of prepaid ARPU increasing, the overall pay monthly ARPU down.

Again, in pay monthly ARPU, as we've seen across previous periods, we see a divergence in terms of what's happening at the consumer level where we have increasing ARPU and the business level where ARPU continues to be under pressure. The remaining $1 million of mobile growth comes from our other -- our nonservice revenue part, which principally is in relation to CPE.

And as Simon mentioned earlier on, while that number has typically been -- to get back to FY '18, that was about $43 million of revenue increase over the period. It's typically very low single-digit margin.

And what we continue to see in that market is a moderation in terms of handset prices. So handset -- average handset prices over the period increased by only 2% relative to 16% in the previous period.

And the total volume of handset sold continues to decline as people tend to hold on to those handsets for longer. The second driver of the result related to our broadband, which pleasingly returned to revenue growth following a review of copper pricing and the launch of our unplanned proposition.

And these pricing initiatives more than offset a modest decline of around about 4,000 in terms of the overall broadband base compared to the second half of FY '18. Finally, the largest driver of revenue growth was our cloud security and service management business, which grew revenue about $16 million, a pleasing result of almost 9%, albeit a moderated level of growth compared to what we've seen historically.

So those are the growth drivers. But what we also found this year was, again, the continuing pressure from our legacy business is declining, the most significant of which was voice and collaboration, with voice revenues declining by $43 million year-on-year.

Around about the same in terms of dollar terms as we've seen historically, but a larger percent, given the declining connection base there. Not a surprise to us that this number came through.

And the reason for that is, as we called out at the full year '18 results announcement, if you go back to the second half of FY '18, we had quite a large decline in the number of voice connections, principally in the wholesale area. And so what we're really now seeing is the full -- is the half play through of that lower connection base.

The second key driver to call out is gross margin. So while we have revenue going down, we've been able to grow gross margin by $8 million over that time.

Mobile, talked about already. And again, the benefit of that higher service margin making up a greater proportion of our overall growth.

We also saw the benefits within margin of the buyback of about 26 deal operated stores that we made in the first half of FY '18, lowering our commission cost within the mobile business. Our broadband margins continued their strong recent performance, growing about 10 million.

Again, we see the benefit of wireless broadband, slightly higher growth than we had in the previous half, again showing the sustainability of growth in that area. And a reminder that wireless broadband now generates a gross annualized cost reduction of $62 million for Spark.

Final area, of course, contributing to that was cloud service and security management, which delivered with revenue growth translating into $11 million of gross margin benefit. So again, those gross margin increases offset some of the declines we saw in areas like voice and helped to drive the overall EBITDA performance, along with our cost reductions in labor.

So you might recall that at the full year '18 results announcement, we signaled an intention to achieve mid labor cost of $470 million, that's an annualized rate, within the first half of FY '19. That result has been delivered.

And we see that coming through a 5.3% reduction in FTE and a $26 million saving in terms of our net labor costs through the period. If we go back to the start of the Quantum program in the second half of FY '17, that program has delivered net benefit of $110 million in annualized benefits with our net labor rate, which at that time was annualized at about $580 million, falling now to around about $470 million.

The next key driver I wanted to comment on was Southern Cross. Again, late in December, on December '18, we updated the market.

Southern Cross announced that shareholders entered into at great terms with Telstra to acquire 25% of Southern Cross including capacity. And while that transaction is not yet complete, as part of funding the estimated project cost, Southern Cross decided to not declare any dividend payment for the first half, instead retaining that accumulated cash.

Whether this cash is eventually retained to lower the equity contribution required from shareholders for the new cable or distribute it to shareholders and then reinvested will depend on the completion of the transaction and restructuring advice we receive at the time. Now within our FY '19 guidance, we indicated we expected a full year Southern Cross dividend of between $10 million and $20 million.

And while it's too early yet to know whether there will be any full year dividend, the fact that no dividend was declared in the first half impacts our first half NPAT with Southern Cross contribution to investment income falling from $28 million in the first half a year ago to 0 in the first half of this period. Ultimately, Southern Cross Next represents what we think is an excellent long-term opportunity for Spark shareholders.

And so while dividends being retained would ultimately lower full year '19 EPS, it ultimately makes no substantive difference between those dividends being paid and then reinvested, because this is being retained and offsetting the required capital investment for this project. The next driver, which Southern Cross also has an impact, is tax and depreciation.

So Southern Cross is usually the largest driver of our effective tax rate across the period. You'd note that our effective tax rate for this half was 31%, which is quite a lot higher than we've had in the past.

And this is because our New Zealand tax rules, Spark pays tax on its shares of Southern Cross earnings, while dividends are treated as tax exempt. But effectively, it's the offset between those 2 that determines whether or not the effective tax rate is increased or lowered because of Southern Cross.

So in the first half of '19, there was no dividend to offset our share of those taxable earnings, meaning a net increase to Spark's effective tax rate of approximately 2%. Depreciation is another one we're commenting on.

While -- we've talked for a while about the trending of depreciation hitting towards CapEx, and that trend has been interrupted, if you like, by the changes to the accounting standards, because under IFRS 16, we've had a one-off uplift in depreciation and amortization. It's increased by around about sort of $47 million, as we now bring in amortization as leases, meaning we no longer expect CapEx and depreciation and amortization to converge because of that accounting change.

Then when you look at it on a like-for-like IFRS 16 basis, in the period, we had depreciation and amortization increase by about $8 million due to some shortening asset lives as we see a greater proportion of our CapEx moving into more software-like assets. Moving on.

I just want to make a quick comment on cash flow and debt. Given the changes that we've seen in this accounting period from IFRS 15 and 16 as well as treatment of Southern Cross, we've decided to start introducing new cash flow measure to look beyond the IFRS exchanges and highlight the underlying cash flow performance of the business.

We currently use it to measure underlying free cash flow, and we've set a detailed calculation of that in Appendix 3. We've also opted to exclude changes in working capital from this definition, given the depending on the year, our largest supplier can be paid as few as 11x, whereas many as 13x, which gives a maximum net working capital variance just for that one supplier over that $80 million.

And that's the rationale why we've also included a working capital definition, but it is not inside our calculation of underlying free cash flow. Pleasingly, in the first half of FY '19, underlying free cash flow was $146 million, which is $36 million higher than we had in the first half of FY '18, with the Quantum benefits coming through more than offsetting the loss of those Southern Cross dividends.

Turning now to net debt. For the period, we saw net debt increase by $132 million with two key drivers, the first being dividend top-up.

It's important to note that we're now giving a dividend top-up over and above that underlying free cash flow measure. In the past, we've typically only commented on the difference between EPS and DPS.

So we have changed the basis for how we now talk around that increase. We also saw an increase in the handset receivable over the period.

But as Simon indicated earlier, that is slowing as we start to see some of the heat coming out of handset sales with an increase of around about $23 million in handset receivables, meaning the total handset receivable now sits at just over $230 million. Moving on to first half '19 dividend guidance.

Directors have declared a first half '19 dividend of $0.125 a share, made up of an ordinary dividend of $0.11, 75% imputed and a special dividend per share of $0.015, also 75% imputed. Our FY '19 guidance remains unchanged.

However, if Southern Cross dividends continue to be withheld as prefunding for the build of Southern Cross Next, and FY '19 earnings and associated guidance will be reduced by around about $0.01 per share. And we'll provide a market update in the event that Southern Cross dividends are withheld for the remainder of FY '19.

So that ends the key financial highlights for FY '19. And as usual, we have all the further information in the KPI workbook and the interim financial statements.

Over to you, Simon.

Thanks, David. So operator, we'll turn the call over now to questions and answers.

Thank you.

Just three from me, please. Just firstly on that full year guidance, you obviously haven't changed the revenue target, but you haven't pulled down your growth expectations for mobile service and cloud security.

Can you just comment on whether there have been any positive offsets in that guidance? Or whether we should now be thinking more on the bottom end of that range?

Secondly, just your annual labor costs. And that all $70 million annualized labor cost you achieved in the first half, do you think that should be trending down into a -- could trend down into the full year?

Or should we be thinking about some incremental areas of investment heading into '19? Then finally, just more data point.

Can you just comment on what the growth overall in your average data usage on your mobile base and how that compares to the fixed wireless subscribers, please?

Okay. Look, I might take the last one first.

So mobile data usage increased by about 50% on our pay monthly basis. So it's sitting up around 2.8 gig.

That's a larger increase than we've seen in terms of fixed wireless. Fixed wireless tends to be reasonably static.

We don't see wireless broadband average growth changing really all that much. It tends to -- it's been reasonably static, in fact, since we started off bringing that product.

So certainly growth weighted on the mobile data side. Coming back to revenue, we do normally see -- so your revenue question about the second half split, we do normally see a widening of both revenues and more in particular cash flow and earnings for the second half.

So it's important, I guess, to take that into account when you look at what we're seeing on the earnings side. The other way to think about guidance is if you just look back, we were expecting EBITDA to uplift within our guidance range of between $35 million to $65 million.

We're sitting at about $20 million for the half inclusive of a one-off $5 million write-down. So if you play that forward and then look at the seasonality, I think that gives you a good indication for where we sort of see the full year and why we're comfortable on the guidance number.

Finally, in relation to the labor, yes, so that's the annualized rates. I think it's one that we always sort of look to.

We're not going to continue -- we started giving comments on expected annualized net labor rates as a result of the Quantum initiatives that we put in place. It's not something we did previously to that.

But certainly, you'd expect the full year benefit, because if you look at our first half labor cost and annualize that, that takes you to about $500 million. So we certainly do expect to see the full year number trending down.

Yes. Just on that revenue comment, so the guidance that you alluded to 2% revenue growth that you pulled back mobile service revenue from 5% to 3% and cloud security 15% to 10%.

Just interested if there has been an offset there in terms of when you set that guidance initially? Or whether maybe it's more on the 0% to 1% range from here?

Look, I wouldn't comment within that range. I mean, I think between 0 to 2%, we're still comfortable within that range.

Just wondering if you could give us a little bit more detail -- like just noting that cloud and security revenues and gross margins up at the very high gross margins, I think roughly about 85% gross margin. Can you give us a sense on about when you think forward in terms of the prospects you've got for that product set, but whether you're expecting any change in the gross margin?

And just how the economics work? I'm assuming that below that, it's a pretty good fixed cost leverage product set to be sort of growing into.

So Adrian, cloud, security and service management, we bundle them together, because it's often sold as a set. But we even think of cloud is the infrastructural service.

What we're primarily providing behind that is Infrastructure as a Service, which is why the gross margins are strong, effectively infrastructure-related return. Security, we do have the biggest cyber security practice.

That's a human-delivered service, supported by some technology platforms. It's a heavy, highly expert people-based service, and it's very focused on the largest end of town today.

I mean, the service management element of that is also a big customer service proposition, which is partly to do with transition work that also in an operate mode. So the flat bit of the story this year is service management.

And I emphasized before, that would be the low-margin sort of labor -- on charging of labor. So we're relatively unconsumed about what -- that would -- we would always run that business just to meet the need of security and cloud, which is a business where we really -- our profession and have a higher margin service.

Security grew, not as well as we would have liked. And the battleground there is solving for another small-, mid-sized corporate customer.

We haven't quite found a solution there yet. And -- but we're confident that we will, that we're very strong at the top end of [indiscernible].

And in cloud, actually continued strong double-digit growth. And we do -- we are confident about the runway.

There's a lot of work to do still in New Zealand, at least 3/4 of workloads haven't got across to the cloud yet in New Zealand. So there's plenty of room.

We'd always spot a double-digit growth performance in cloud. We've had one -- we have one large customer in budget here, very large deal that's just taken a little longer to transition in, which has slightly tinted the performance there than we would have liked.

But it's a good story. When we renew, yes, we have some price down work, but we're also winning new customers with a lot of upside too.

So we still feel very good about the cloud business, and you shouldn't read anything into the current number around that forward view on that.

Okay. And just a separate question, which is sort of not a million miles away from what Kane was asking But in terms of Agile, should we assume that, that is now in full swing?

I noticed that the cost, the change bucket, which was obviously had some numbers significantly for '18, there's nothing in there for first half '19. Is the cost of change done now and Agile is in full swing?

Is that what we should read from...

Agile is in full swing. As I've said before, we've rolled the operating model out both across what we call Agile heavy, the big cool product, network, segment marketing areas of the business and the big technology units that underpin them.

And then we've taken an Agile-light model through the rest of the organization, which would include all the corporate functions and the operating units. So we have done rolling the model out.

What we have still is, I'd still describe as enthusiastic amateurs. We're still learning to [indiscernible] the benefits.

And we have some areas giving -- delivering incredible results and another areas still finding their feet with Agile. So -- and we'll be tweaking the model as we go along just to make sure we keep on focused on delivering and getting to the fully professional level of Agile over the coming year.

So that's still in play. As we indicated at the beginning of the year, we always have some cost of change, they're in beta than they are labor numbers.

And we expect that this year's cost of change numbers to sit within those normal parameters. And at the moment, there is no reason to believe anything different from that.

Okay. And just final question.

Can you just provide us -- I know this is kind of a little open, but some sort of indication on 5G CapEx timing? But what your base case is now for spectrum, I mean, whatever other sort of CapEx is required post that?

Timing, well, we -- as was indicated, we're already using part of our mobile CapEx envelope for densification work. All sites we're building today are what we call 5G-ready.

And so part of the mobile program is headed towards that. We would like to believe that within this calendar year, we can get to the point where we spend on 4G and move all -- switch all mobile infrastructure spend toward a 5G future.

Again, without any indication of that needing any additional or overall increase to the Spark CapEx envelopes. And spectrum, we're willing and able to buy it.

And I just have no ability at this point to predict when government will give clearance for that process to proceed. From our perspective, the sooner the better.

And so I had -- this is my final question. But just -- can you just give us a quick recap on the heavy IT reengineering spend?

I noticed in IT systems were still reasonable in the first half. But when does that draw to a close?

Or is that already drawn to a close?

We're starting -- we had a runoff -- reengineering completed [indiscernible], but we have a lot of capability there, which we've been able to redeploy into some of the digitization and automation agenda that is delivering these incredible results. So it makes -- these -- continuing that higher spend has been an enabler of those tremendous cost reductions and service improvements.

But you can reasonably expect that will start to wind back from here on.

Just firstly, the $14 million growth in other operating expenses, is that -- does that incorporate the $5 million write-off from the IoT?

Yes, that's right. So that sits in our other costs, and it's in the -- we also had some increased electricity costs over the period as well.

All right. So it's the $5 million and the other?

The $5 million is in the $220 million other expenses line.

Yes, and the other component of that other expenses category?

It's under -- you're right, there are two there. It's in the $220 million line.

That is a -- which subcomponent line of that is just -- sorry, it's a bit confusing. Can you clarify?

It's in the KPI workbook. It's actually a dedicated line we call out in [indiscernible].

All right, cool. Just -- was that -- I mean, obviously, that's a decent chunk of the $14 million.

But those overheads have sort of been increasing a little bit. I guess, just sort of thinking ahead to sort of the launch and the investment that is going into Spark Sport now, is that going to drive further growth over the next year or two?

On what sort of Quantum should we expect? Or are you taking some of the investment on the white box and channeling it towards back sport instead?

Yes. Look -- so Arie, I think if you look at core operating expenditure, the 2 areas, which are pumped back a bit, are computer costs, which is reflective of the amount of infrastructure we have, and we're starting to have another look at that line and it's jumped a bit on us recently.

And the other one is electricity. We've taken the view over the last few years that we should be exposed to the spot prices, a significant buyer, that's proven to be not our smartest discussion.

So we've been injured a little bit in the first half by that. And so they are the main 2 drivers other than cost growth associated with new initiatives, a lot of which will appear in cost of sales in due course, things like accounting costs and more clearly.

But clearly, Spark Sport will have an overheat function, just like Lightbox does, and that will come into the numbers over from this half on really as we're starting to -- again, we start seeing that come to the OpEx lines.

Okay. But the Quantum is going to be amortized in one of the cost of sales categories, is it?

It will go through cost of sales. At the moment, it's in prepayments.

Okay. And what should -- where -- what's cost of sales category will it go in?

Well -- so we'll sit down as we get towards the first half -- sorry, the end of the second half. We'd like to go into other, which is where Lightbox goes, but we're still just working through the best way of reporting Sport, Lightbox and how we do that.

As you'd appreciate, there's only minimal cost of sitting there within the first half, even within the second half is likely to be minimal, because much of those costs are really coming in FY '20.

Sure. And then just in terms of the platform, if I understand it right, why you sort of positioned that, are you saying that you'll be launching Formula 1 on the platform with the Grand Prix, while it's still in beta phase?

Or are you out of that phase when you launch Formula?

To be completely clear, when we took the -- the beta we're using in the document here implies the public detail. So it means -- we're an Agile organization.

And agile organizations launch a minimum viable product. Usually, turn the beta phase where we put real customers on it and use those real customers to test and find the bugs, prove it out.

So we expect in the order of 4 weeks' worth of that process -- a few weeks as a typical arrangement for it to be taken beta, and during which time, we would -- we'll make it free. so it's not a -- no one is paying for that.

So that's helpful to us, and it's good for customers to start to get a feel for it. And yes, so the Formula 1 -- the first round of Formula 1, so we have bought the rights to the entire Formula 1 series.

The first round -- the first Grand Prix is the Australian Grand Prix in mid-March. And given the beta -- during the beta period, we would not be foolish enough to only run it on the platform.

So we are simulcasting it on TVNZ, just to give all New Zealanders an opportunity to view it. It's a good promotion for Formula 1 and the fact that we have [indiscernible] In the Spark Sport.

But, of course, that would just mitigate if we have any technical problems during the release. But this doesn't deny any one the opportunity to view it.

But we're going to encourage as any many people as possible during the beta to come on to the service and watch it that way and give us their feedback on how it performed.

That's great. That answers my follow-up questions on that too.

Just in terms of the CapEx envelope, is there a point where -- like do you see what's, I guess, Southern Cross Next that there'll be a need to kind of call out and separate that out in the CapEx guidance? Or is that unlikely to be the order of magnitude where either capacity or rather sort of investment for next $1 billion of magnitude that will be required, because at the moment that's included in your guidance, isn't it?

Yes. I think if you look through, Arie, for the Southern Cross press release talked about an overall project build in the order of USD 300 million, that's obviously all debt gets applied to that.

You then take the equity proportion. Should Tesltra come in, that will dilute our equity proportion down.

And then you offset against -- in the first half prepayment. So that does mean that the overall Southern Cross number is more manageable, certainly if you relate back to the dinner Southern Cross build, I think it was 2 or 3x that amount.

And we're funding 50% of that in an environment where leverage -- while on that top of assets. So I think it is going to be a smaller amount that goes through as an investment that we make into the Southern Cross Next.

And we had envisaged that the ongoing capacity commitments we make the purchase and capacity will continue to just show up under the normal -- within the normal CapEx. And we don't see any reason why that would change, Arie.

Sure. And if it is a smaller investment outside of that, then not necessarily sort of CapEx any way, that sort of an equity investment of the smaller magnitude?

Sure.

Now that's helpful. Just on guidance also, if you do come back on EPS, because after the withholding of that dividend -- and you sort of moved to looking at the order of top-up in relation to underlying free cash flow.

Are you essentially saying that if you do need to come back on EPS, certainly no intention to come back on the $0.25 per share guidance for this year?

Correct. So we think about it.

I mean, as I said, whether or not we get paid the dividend and then reinvest it or just gets held on to, it's the same cash effect. So there's no difference in terms of our ability to pay the $0.25 dividend.

Sure. And then final question just in relation to, I guess, labor costs, which -- I mean -- and clearly, there have been cost savings as you point out and cost of sales as well, which are pretty substantial.

But the labor cost saving has been a big driver of EBITDA performance in the last. But you're kind of in full swing on Agile now, I guess, as the business you're in, there's sort of ongoing pressures and there are areas where you're wanting to reinvest.

Are you starting to turn your mind to the next sort of round of kind of cost out on labor? And if you are, when do you think you might sort of share that with market?

Actually, Arie, we're just coming into our planning process, so you can imagine we're sort of early days. We commenced planning in some of those strategic course during -- over the next 2 or 3 months.

Bust just quite clearly, we've given labor a very hard run for the last a couple of -- three years. And actually, it's -- while productivity, especially in the commoditizing [indiscernible] business may have stopped being a key value driver.

So you're always going to be looking for that. With our strongest right now that it's time to have a look across -- a hard look across some of the other big cost lines, which have had an easier run, while we've had a very [indiscernible] story on automation, digitization, simplification, which is teams to go the productivity issue.

It's time to have another think about some of the other big cost lines, be they cost of sales or other core OpEx lines. And a tight [indiscernible] how we're using our capital.

So we're into that discussion internally now, and we'll have plenty to update at the full year on that.

Congratulations on an excellent cost outcome. That was really good.

I had two questions. One is just on the Huawei band.

What do you think it's going to be? Do you think it strengthens your relative hand in the market, because I presume competitors face that pressure as well?

And in that context, you also mentioned that Spark is multi-vendor capable. Could you sort of elaborate on that?

That's kind of question one. And my second one is just on gearing.

Gearing has nudged up to 0.12, thresholds that by 1.4. How do you think you'll sort of manage over the next couple of years?

Yes. Let me take the first and then David can pick up on the other.

Look, so just on how Huawei, the peers in New Zealand market, we have had Huawei in the RAN, and not only the RAN, they do a lot of CPE for us, modems, et cetera as they do for most. So we find Huawei very embedded in the CPE market in New Zealand, be it on mobile devices and both fixed and wireless broadband modems.

And they're a big provider of our RAN infrastructure and my understanding of the 2degrees is pretty much an entirely Huawei shop, so they both RAN and core for 2degrees, I'm very unsure about, Vodafone's NetLoop, but I know Vodafone sales and utilizes Huawei CPE in most of the market in New Zealand; Chorus has a heavy exposure to Huawei. So Huawei is actually very, very embedded in the infrastructure of New Zealand and communications and big enough I've said before, an outstanding, being their technology is excellent, we've never had any cause to suspect of anything inappropriate and that being very pleasant part of our being the mix.

And so obviously we're disappointing to lose them. That said, we have Cisco Ericsson core in our mobile network and that core has been designed, given that we've always -- has been brought strategically been able to tracked by single window use for commercial purposes, and obviously, the risk of -- has been heightening over the last 1.5 years around whether Huawei would be able to have a roll on it.

So it would've been foolish for us to not had prepared for that, if we needed to. So we came today put other vendors into our RAN, and as I said, that does so, therefore we have no real impact on our ability to slow down.

With regards to how it impacts the competitive arena, I think it would not really be right for me to comment on that. I think you would hit -- get the main insight around who has Huawei in the market.

I think you need to form your own view on whether it makes any difference to the other parties. David?

So the second question, so you're right. In terms of gearing, it picked up from 1.17 to 1.2.

We've been clear I think for a while that we're absolutely committed to that A- rating. So we've still got some good headroom to that.

In addition the heavy treatment finance on the handset receivables means that as we see those handset receivables grow while it may increase our gearing, it doesn't actually increase our headroom in relation to the S&P rating. In terms of first half, second half, we do normally see higher first half.

And I think we commented at the full year announcements, full year '18, that we do expect the growth in debt to moderate across FY '19. So up in the first in the first half, but certainly, in second half we would expect to see that come back and that's mainly because we tend to see improved cash flows in the second half.

So still certainly focused on that 1 point -- on that 1.4. But we do expect to see second half be slowing of overall growth in debt.

Just a couple of questions. In terms of the cost line with the electricity prices are you able to give us any feel for kind of, like spot prices going up, just much of an impact that might have had in the first half?

And then the other question I just have was on the mobile ARPU pressure on the business. I'm assuming that the reason why they were under pressure, are you seeing some of the moves to the unlimited plan, is that the main reason why you were seeing kind of ARPU pressure on the business, ARPUs in mobile?

So taking those in order, so well firstly, let me talk about the ARPU side. So not really on unlimited.

It tends to be more that when we sign up a series of contracts with business as enterprise, as those kind pass through the time and as you come out of contract, there tends to be some pricing pressure pushing those down at the renewals. And we're probably that seeing now for the best part of say 2 years in terms of those [indiscernible].

So less to do with unlimited, but it's certainly one that we're keeping an eye on particularly in the sort of main side of the market. In terms of your first question about electricity prices, high single-digit millions in the first half and we're lookings at ways of delineating that risk in the second half.

Simon, David, the investment case on 5G, did the investment merit stand on their own in terms of lowering OpEx and CapEx intensity? Or do you need a decent increase in customers bypassing fixed line onto your 5G or even ARPU increase for the retenants to be acceptable?

We'd look to both. We believe it supports an improved mobile economics.

It's really the capacity uplift to the CapEx dollar that keeps us excited, which will enable continuing greater value to fewer mobile handset users. But it also enhances the prospect of wireless broadband continuing to grow its share of the broadband market in New Zealand.

As is very typically the case with those parties chasing 5G around the world, wireless broadband alternatives are an anchor of 5G investment case globally.

Your aspiration for perhaps ARPU increase is probably not that done -- not that great with 5G?

We're very doubtful that at 5G, anymore than 4G just because we change a technology, suddenly people pay more. That's not likely.

But what you do see over time is increasing usage and substitution away from other devices. So the shift from 3G to 4G has grown our mobile revenues.

That's plainly obviously on our results over the years, but it's not repricing -- pricing going up per unit, actually it reverses too, prices have come down over time, but people will get more and more value product, so are they willing to pay a bit more for all the value, and we think 5G is just another one of those; it just makes wireless services more and more attractive.

Okay. And in your cloud and IT business, can you please name some of those local and global players that are making it more competitive?

And how are you sort of recruiting good tech people in the face of this competitive environment?

Look, I think Brian, demands are obvious. So yes, who the global competitors are, and we go head-to-head with 2 or 3 well-known IT names in New Zealand who also operate in the cloud and the security and service management business.

We're the top-telling shop in New Zealand in terms of recruiting. You'd want to look to Spark before you're going to work anywhere else.

Today, I mean, we do the exciting stuff and we have no trouble recruiting very talented people.

Yes, look quickly, I think I had very similar question to Brian, but on that 5G business case, as you said, a lot of it's about lowering unit cost. But there is some revenue opportunity as you win more people to wireless-based services.

I just want to get what's going on in model at the moment with flatter service revenue guide outlook and a focus on the margin expansion, does that change the business case somewhat to 5G? And in terms of -- the second one, in terms of 5G CapEx, you said, you can do it within the envelope of 11% to 12%.

That's very commendable compared to network operator benchmarks. And I wondered to what extent you marked down the leverage, the relationship with Chorus and where that decision point lies between where you might use your own fiber -- your own fiber density versus using Chorus for some of that backhaul?

Ian, I mean, one of the reasons we commented around our CapEx profile on 5G is that New Zealand has lucky in having the USB fiber program, which has enabled all of urban New Zealand. And it's obvious that our first preference will be to leverage Chorus and the LSCs fiber infrastructure.

So they does mitigate a lot of what we've seen other operators CapEx assumptions build around, so [indiscernible] meeting a whole lot of new fiber. They took -- their problem doesn't exist in New Zealand.

So we have a faster and more economic pathway, which is excellent. And that's why, I think the long-term growth for Chorus and LSCs is a direct provider of fiber to the New Zealanders, it really need a great connection and to the backhaul for the cellular networks that provide an untethered broadband service for those customers who don't need a direct physical connection.

I think it's a very good future for them, and been a great partner to us in delivering that. And I'll just sort of reemphasize on the peers, just restating Brian, in the end you've got believe increasing value for the use of untethered broadband capable devices just pays away at other sources of spend that you're forecasting.

For example, if you believe in a pervasive 5G future you will not need Wi-Fi, Wi-Fi distribution capability in your home or your office. So these are substitution opportunity.

You would not need multiple devices. You can concentrate and, why wouldn't your mobile become your hotspot for your other devices, et cetera.

So you don't need -- you pair away for reasons as well to head to these separately connected in some way and shift that value to wireless over time. So that's why we are very big on this.

We believe in the wireless future we're concentrating our assets and investments and our effort toward wireless, because we genuinely believe customers pursue it. That shows in every Net Promoter Score we've got and that our economics and our business will be better up for it.

So hopefully, that's the answer. I'm sorry, we're going to have to close the call off now.

We didn't got time for any further questions, but being in our Investor Relations team, we'll certainly be valuable for anything that we haven't covered off. So thank you all for taking the time this morning to join the call.

Cheers.