Chorus Limited (CNU) Earnings Call Transcript

Cheers, everyone. Good morning, and welcome to Chorus' half year's -- half year results announcement. I am JB Rousselot. I'm the new CEO of Chorus, following Kate McKenzie's return home to Australia in November. I'm joined today by David Collins, our CFO. I'll start today's briefing with a quick introduction on some key numbers and trends before handing over to David, who will take you through the details of our financial results, including an update on our FY '20 guidance. And then I'll come back and talk about some of the key initiatives that we have facing us for the next few months and how they fit within our strategic focus. As Kate said in the full year results, FY '19 was really a turning point for the company, and you started seeing that playing out in the numbers that we're releasing today for the first half of the year. EBITDA in particular has seen a very strong growth from $318 million for the first 6 months of FY '19 up to $332 million for this half. This very strong result has been delivered as a mix of reduction in operating expenses but also strong growth in broadband connection and in particular fiber broadband connections and also growth in ARPU. These strong results allow us to declare a half year interim dividend of $0.10 per share. As the chart says, shows, you also see that our net profit after tax has grown just by $1 million to $31 million. And that's primarily due to expense -- financing expenses that have grown a little bit due to the bond raising that we did in 2018 and 2019. So it's been a very busy 6 months, and we've had a number of milestone impacting pretty much every areas of our business. In addition to the CEO transition, we've seen significant events in pretty much every areas. On the regulatory front, we've seen the Commerce Commission releasing its draft determination on input methodology. That's a big milestone for us on the regulatory front. In operations, we've hit the massive results of completing the initial build of the UFB1 program. That's a huge milestone for the team here internally. We're now into a UFB2 build. We've added 83,000 fiber connections to our network; and all of that while we were merging 2 of our largest operational pieces of our organization to create a unique customer and network operations area, reflecting our shifts from a build focus to much more of an operating focus. In product and sales, we've launched our Hyperfibre services, and those really take fiber broadband to the next level. We've also won a contract to provide fiber backhaul to cell sites of the rural connectivity group. And then finally, on the financing area, we've seen S&P increase our credit threshold from 4x to 4.25x. And as I've mentioned, we've also completed the 7-year EUR 300 million bond raising. So a very busy period. And as you've seen from our Q2 connections update, which we released last year, also a really strong performance on connections. We've seen a significant slowdown in the disconnections from our network. And if you look at the last 6 months, the disconnections have gone down by just 18,000 compared to 40,000 for the previous period last year. And then when you look at the mix of those connections, it's also very positive. What we've seen is a growth in broadband connection. Broadband connections have grown by 10,000 over the last 6 months, and that compares to a loss of about 1,000 for the same period last year. A lot of that broadband connection growth has taken place in our largest area, the UFB area, where we have our own fiber. And now fiber connections represent almost 70% of our total connection, so really strong trends. Now over the last few weeks, we've seen a little bit of an uptick in disconnects. That's something that we typically see happening in the December, January time frame. We've also seen a bit of an uptick in fixed wireless activity, but again the trend that we see early in January is a continuation of those strong volumes of fiber connections. Another very good trend for us over the last 6 months has been the uptake of our 1 gigabit product. Through a combination of incentives but also reducing the line costs from $65 to $60, we've seen the uptake of 1 gigabit product grow by 50% and reaching 87,000 connections at the end of the 6 months. We've seen a number of retailers with very sharp offers for the 1 gig product. We've also seen some of our retailers actively migrating their customers from the 200 megabits per second product to 1 gigabit. And as a result of all this activity, 1 gigabit now represents 13%, up from 10% 6 months ago, of our growing fiber connections. To give you a bit more of an idea: If you look at the orders that have come since the beginning of January, about 20% of those were for the 1 gigabit product. So it's very exciting for us to see that 1 gigabit product become a bit of a go-to product for people who wants to connect to fiber. At the other end of the spectrum, we've also seen a number of retailers stop selling the 15 megabits per second product. And as a result, the number of connection that we have on that speed tier is declining, but that's not a bad trend for us. The next slide is one that you would have seen also on our Q2 connection updates, and it talks about how things are faring across our various geographies. The majority of the disconnections are occurring in what we call the local fiber company areas or LFC areas but at a slightly lower rate than last year. In rural or area without any UFB fiber, things have been fairly stable despite some promotion by fixed wireless operators in those areas. And in fact, our broadband connections in those area have actually increased on the back of strong performance in VDSL and also further fiber rollout in those areas. Today, we account for about 19,000 fiber connection in those geographies. And then finally, on our largest area, the UFB area, we've seen really strong broadband growth. We've gone up by 23,000 for the half on the back of increased broadband penetration and growth in the number of premises in those area. And that's more than offset the increase of 19,000 in terms of copper lines without any broadband. I've said earlier that the completion of UFB1 was a key milestone. And really for us at Chorus, for our service companies, for Crown Infrastructure, the wider government and a number of significant stakeholders in that project, it was a massive milestone to be able to deliver. Today, we think about UFB1 as just 28,000 kilometers of fiber rollout, hundreds of thousands of premises passed, hundreds of thousands of connections onto the network, but it's taken a lot of work and a lot of effort to get there. And it's a tribute that we've been able to deliver this big project up -- in time. In fact, we are a month early in terms of completing UFB1 rollout and well within the $1.8 billion guidance that had been given. We're now in UFB2 rollout, and the good news is the trend continues. We're now more than 1/3 done in the UFB2 rollout. And you can see on this graph that we now have a very clear steady downhill trajectory towards the completion of UFB2 by December 2022. The UFB rollout has also been a big success in terms of takeup. Initially, the contractual target for UFB1 was to reach 20% takeout (sic) [ takeup ] by 2020. On UFB1 today, we're already at 58% takeup rate. And when you look at the UFB2 areas, the areas that we've already completed, we also see a very healthy takeup rate. We're already at 34% in those UFB2 areas. So when you combine UFB1 and UFB2 completed area, we're in an average of 56% uptake of fiber connections in area where we've rolled out fiber. So really, really strong. And if you take the example of Auckland specifically, it's even higher. We're at 68 -- 63% fiber uptake in that area. And almost 3/4 of our connection in that area are fiber connections, when you look at our total broadband connections, so again very strong growth here. Another thing that we've done over the last 6 months is to continue fiber installations. We've delivered almost 100,000 fiber installation during the last 6 months. And you can see on the graph that the monthly rate at which we've done those pretty much mirror the rate that we were doing those last year. We probably had a bit of a tailwind in August and September on the back of the Rugby World Cup but fairly much a same trend as the one that we saw last year. Another great development for us has been the improvement that we've done in customer experience. We're very happy to say that our connect CX, the ratio that we measure customer satisfaction for connection, has now reached 7.8 in December. And in fact, the number for January was even higher at 7.9. So very happy to see that trend. Our lead time for connections has also decreased. In December, it was at 7 days. It's gone up a little bit in January. Again, that's a seasonal trend that we regularly see with people coming back from holidays and big volume of orders coming back, but we're confident that we'll bring back to that number of December very soon. And as far as field crews are concerned, we've kept fairly steady on those. We've got about 670 field crews out there helping us with connections. And that helps us keep track of what we refer to as our WIP, our work in progress, and making sure that those volumes continue. Speaking of the field workforce, we thought it was important to update you on the work that we've done continuing a lot of the initiatives that were part of the follow-up on the independent review that we did and published in April. The bulk of the work that we've done in this area is to help our service companies develop their own business process so that they meet the administrative requirements of employment standards. We now have people within our own team that monitor the activities of the service delivery partners in that field, and we give regular update to our Board on their activity. To give you an indication: We have about 400 service companies that help us deliver connections throughout our network. We've actually terminated relationship with 28 companies and their directors on the based of the investigations that we've done so far. And we're also very aware that, as we complete the build of the UFB, as copper volumes decrease and as we move more into a stable maintenance and assurance mode, the volume of field activity will reduce. And so we're already working very closely with our service companies, specifically Visionstream and UCG, to make sure that they have a very clear plan to manage this transition. On that front, our contracts for connection field workforce expire in September. We're already in market to extend that connection capacity beyond September and well into 2022. We'll review that connection workforce requirement together with field workforce and maintenance workforce in a global review of our workforce requirements when all of those contracts expire at the end of 2022. So overall, very busy 6 months, very good results and solid trends that we're very happy about, but I'm now going to hand over to David to take you through a lot more details in our financial returns and also through our revised guidance. Over to you, David.

Thank you, JB. And good morning, everybody. Starting with our income statement. We're reporting EBITDA today for the half of $332 million, up some $14 million on the prior comparative period. Revenue is down $6 million, and expenses are down some $20 million. I'll talk about the components of both of those in the ensuing slides. On the interest line, interest is higher, reflecting the growth in our gearing levels driven by the UFB program and also higher drawdowns of Crown funding over the period. So turning first to revenue. Revenue was down $6 million over the period. There are a couple of components to this. Firstly, broadband revenue is higher versus the prior comparative period, driven by higher connections, which is a very pleasing performance for us. On the other hand, voice connections, copper voice connections, continue to fall. And we've also seen a reduction in third-party relocation services which flows through the field services revenue line for the business. So overall $6 million lower for revenue. Moving to expenses. We've been putting a lot of focus on reducing our cost base as a business, and we're very pleased with the progress that we've made in the current half. Starting with labor. Labor is actually slightly higher than the prior corresponding period, driven by some one-off restructuring costs of some $1.5 million which relates to the CNO or customer network operations restructure, which JB mentioned earlier on. We do also, or however, have reductions across a number of our expense element lines. Maintenance is down some $4 million against the prior comparative period. I'll talk some more about the key drivers of that on the next slide. Our IT costs are down $3 million, driven by moving off shared systems and continuing investment in our IT platforms. And regulatory levies are also down, driven by the reduction in the telecommunications development levy. You'll also note on the other network costs line the corresponding impact of the reduction in third-party demand for network relocation services. Moving to maintenance specifically, we provide here the usual dissection of our maintenance costs for reactive maintenance both by spend type on the top right-hand corner and by area on the bottom right-hand side. The key takeaways from this slide are that reactive or variable maintenance continues to fall across our business. It's primarily driven by reducing copper connections which flow through to our expense line, particularly in LFC and bronze areas as customers move off copper connections, but also in our UFB areas as customers migrate from copper broadband to fiber where we have a newer fiber asset and better levels of reliability. It's also true that we've had some favorable weather impacts over the first half, which has seen a reduction in our variable or reactive maintenance. Specifically with regard to fixed costs, we've talked previously about an opportunity for us as a business, as we migrate customers off copper, to realize savings in our fixed costs base. We estimate that at approximately $10 million per annum. And to realize that, we need to move our customers off copper area by area or, if you think about it, street by street, exchange by exchange. So that's an opportunity for us in the future which we've spoken about previously. Moving to CapEx and looking at fiber initially. We've spent some $300 million in the current 6 months on fiber CapEx. This reflects the continuing downward trajectory reflecting the completion of the UFB1 build and moving now well through UFB2. There's the $100 million we spent on communal spend, some $26 million related to UFB1 and $74 million for UFB2. So clearly a change in trajectory for us. As JB mentioned, we completed some 99,000 installations over the period. And it's also worth calling out that within the greenfields and growth line there was some $20 million of CapEx on greenfield spend during the period. Looking at connections, CapEx specifically. We spent $155 million on connections during the period. Our cost per premises connected was $995, which very pleasingly was below the bottom end of the range that we've provided or the guidance that we've provided. The UFB2 areas, our cost per premises connected has averaged $1,179. This is at the bottom end of the guidance range that we've given, once you allow for layer 2, backbone and service desk costs. Moving on to copper and common CapEx, fairly stable in both cases. With regard to copper, we are seeing a continuing reduction in sustaining CapEx on our copper asset, which is in line with the reducing number of connections, but also our pole replacement program is now mature and starting to taper down or slow down. In terms of guidance update for our business, we are today announcing an upgrade in our EBITDA guidance for full year '20 from a previous level of $625 million to $645 million up to a revised guidance level of $640 million to $655 million for the current year. The key drivers of the uplift in our guidance are, firstly, on the revenue front continuing growth in broadband connection and broadband revenue, which is very encouraging for us; and on the costs front, as I discussed earlier on, continuing strong cost saving trends in maintenance, IT and regulatory levies. When we think about half 2, we have made an allowance for a couple of issues, which I wanted to mention. Firstly, if you look historically, half 2 is normally lower than half 1 due to copper voice revenue disconnections. And we expect that to continue. We also have made some specific allowance for investment in accelerating our UFB uptake levels, which we believe is strategically the right thing for us to do. And lastly, we've made a smaller allowance for weather, just on the -- if it was to be that the weather favorable conditions were not to continue. So EBITDA guidance lifted to $640 million to $655 million. In terms of CapEx guidance, we've left that unchanged at $660 million to $700 million, although within that total envelope we have increased the connection CapEx guidance to $295 million to $315 million, which reflects 180,000 to 200,000 installations over the full year. Given that we've reached the end of UFB1 and are now some 86% complete across the whole UFB program, we thought it would be useful to do a little bit of housekeeping in terms of our guidance for the overall program. As JB mentioned earlier on, we have completed the UFB1 program within guidance levels. And we do expect to claim for Crown financing some 827,000 premises passed or $924 million for UFB1. Given that we've reached the end of UFB1, we think that the program guidance which we've had in place since 2011 has reached its natural end, and hence going forward we will provide annual guidance on cost per premise passed for UFB2 -- for UFB1. And then for UFB2, the guidance is unchanged and is noted on the slide. Moving on now to dividends. As JB mentioned earlier on, we are announcing an interim dividend today of $0.10 per share fully imputed. The change that we have noted is that we -- whilst we've kept the dividend reinvestment plan or DRP in place, given the strong performance in the first half, we have decided to remove the discount from the DRP plan, so there will be 0 discount on the DRP. As with prior periods, we will continue to reassess the overall DRP plan at each of the reporting periods looking forward. I'd also like to briefly mention imputation. I spoke to this at the full year '19 annual results release. We are currently not in a taxpaying position due to the accelerated deductions we received for depreciation on our fiber asset. That means that, over time as we impute our dividends, the imputation balance does run down. Broadly, we would expect, by approximately full year '22, we will have run our imputation balance down to 0. Moving on to dividend policy. Firstly, this slide reiterates our existing dividend policy which runs through until June of 2021, which we announced at the full year '19 results release. The slide talks to the key milestones which underpin that policy time. And also, in particular I'd call out the Moody's commentary that they released reasonably recently whereby they said that they would reconsider our credit rating downgrade thresholds or down driver thresholds at June of 2021, which is when the final price-quality decision is due from the Commerce Commission. So that's our existing policy to June of 2021. As we start to look forward in terms of future dividend policy, I wanted to make a few comments around the free cash flow profile of our business. As we reach the end of the UFB build and as connections start to taper off looking forward, we do have substantial free cash flow growth in front of us as a business. This slide describes the key components of our CapEx spend, and I'd like to call out just a few of those. As I've already mentioned, UFB1 communal CapEx is complete. UFB2 communal runs until December of 2022 and tapers down over that period of time. In terms of connections CapEx, these are demand driven, and we've given an illustration on the chart on the right-hand side of what we see that profile looking like going forward. The key message from the slide for our business is we have substantial free cash flow growth in front of us as we look forward and as we contemplate what our future dividend policy might look like. So coming on to our future dividend policy. We are approaching the end of a long build period. We're very conscious that investors have been patient and have had a constrained level of returns over that period of time. As we approach growing free cash flow from our business, investors have been seeking more clarity on future dividends and on our future dividend policy, so we therefore thought it would be appropriate to provide some guidance today, also noting that we have had some significant milestones over the last 6 months with the completion of UFB1 and also with the Commerce Commission's draft decision. Specifically, from full year '22, our expectation is that we will transition to a free cash flow based dividend policy with a set payout range in place. Whilst we believe it's a little too early to specify the exact payout range, we do expect to be consistent with Australasian utility-like and infrastructure businesses who operate free cash flow based dividend policies and who pay out a majority of their free cash flow as dividends. Just to be a little specific, when we talk about free cash flow, we define that on the slide, and it is net operating cash flows -- or net cash flows from operating activities less sustaining CapEx. That's the definition for dividend purposes of free cash flow. And specifically with sustaining CapEx, we define it as simply as we can, which is our total CapEx spend across fiber, copper and common excluding UFB communal and future footprint spend, excluding connections and greenfields and excluding customer retention spend. So that's how we define sustaining capital in the context of our future dividend policy. It is important to note that we deliberately use the term transition from full year '22. The reason for that is we do still have ongoing communal spend for UFB2. And we have an ongoing, albeit reducing, connections CapEx which we do need to fund. The impact of this spend on our gearing and credit rating metrics will therefore, for a short period of time, restrict our ability to move immediately to full implementation of the policy. There will therefore be a short-term transition to the new free cash flow based policy. I would add that we would expect that transition period to be no longer than 2 years and that we would expect a higher proportion of the UFB communal and connection spend to be in the first year of the transition, which is full year '22. I would also add that we would expect the dividend to grow over the transition period. The last comment I would make on this slide is around capital management. We are very conscious that Moody's have said publicly that they will reconsider our credit rating threshold from June of full year '21. If we were to have an increase in the down driver metric from both Moody's and S&P, we would consider appropriate capital management activities to return funds to shareholders, noting that we have an ascribed capital balance at circa $225 million and also noting the imputation profile I mentioned before, both of which would guide our future decisions on capital management. Moving now to our funding and our debt profile. Our net debt-to-EBITDA is just over 4x, which is under the S&P metric and under the financial covenant metric per our debt facilities. We had some $678 million of cash on hand at December, which is in place to refinance the GBP bond maturity which is due in April of this year. Our overall weighted average cost of debt has reduced to 5.47% from 5.75%, and we would expect that reduction to continue looking forward as that GBP bond is refinanced. Moving on to our Crown financing and our debt profile. We've now drawn just over $1 billion from our Crown financing facilities. And we have circa $16 million left for UFB1 to claim. The chart on the right shows the debt maturity for our existing external debt and also the debt component of the Crown funding. With the Crown funding, you can see the drawn and undrawn levels on quite a long repayment profile. As part of our treasury management, we also have undrawn revolving credit facility of $550 million, which is available as a contingency source of funding for our business. Moving on to regulation. We thought it would be useful to provide a recap on where we stand at the moment with the Commerce Commission's recently handed-down draft decision in November, which JB referenced earlier on. In summary, we believe that good progress has been made, from the emerging views paper in May through until the draft decision, in developing the fundamentals of the regime, in particular key principles such as real financial capital maintenance which refer to the need for investors to receive an adequate and fair return on and of capital over the period of the asset's life, which we think is very important. The table summarizes some of the key parameters within the draft decision; and we're very pleased to see positive progress in areas such as Crown financing, taxation and market risk premium. We're very pleased to note that the commission has been actively engaging with stakeholders, has been taking onboard feedback and has been listening. And we commend the commission for this work. There are, however, a couple of areas where we would like to continue engagement and discussion with the commission; and that is in particular with regard to the cost of capital, which we believe is understated in the draft decision. So moving on to the cost of capital and one of the key components, which is the asset beta. In the draft decision, the commission has proposed an increase in the asset beta from 0.46 to 0.49. Whilst we are pleased with the fact that there has been an increase, we do believe that the comparator group on which that 0.49 was based is a little bit too narrow. We have therefore proposed an alternative, broader comparator pool, which is the Damodaran telco comparator group, which provides a larger range and we believe a more representative sample. We've also noted on the slide that the U.K. regulator, which is Ofcom, has provided some useful guidance recently with regard to U.K. regulation and fiber assets in particular. We note that Ofcom has had regard to decisions of other U.K. regulators in gas and water in setting the asset beta for fiber, and those other regulators are quite consistent with the way the Commerce Commission has looked at those industries in New Zealand. Specifically, Ofcom has proposed an asset beta of 0.57 for BT's Openreach business. I would note that Openreach is a legacy business to BT, so if we were to think about what a fiber business asset beta might be, the Ofcom view is that 0.65 is where they have set the asset beta for a fiber business, reflecting higher demand risk. We would also note that Ofcom's views are consistent with the views espoused by the Crown back in 2011, when the contracts were signed, and also Australia's nbn around the asset beta for that business. Lastly in terms of regulation, moving on to a couple of comments on a fair return. Investors have been talking with us around what is a fair return for the investment that has been undertaken over a long period of time, for the risks that have been taken in the past and the risks that continue looking forward. There's a couple of aspects to this. There's a backward-looking view of what the WACC should be over the build period and then a forward view of what an appropriate WACC would be for the first regulatory period from 2022. With regard to the build period, from our perspective, we took on risks from 2011 over a 10-year period. Risks were set at that point in time. Pricing was put in place. And we did not have the luxury of being able to reconsider each year what our investment profile looks like, what our risk profile or refinancing profile looks like. What the commission has proposed over the 10 years is an annual reset of the risk-free rate and therefore the weighted average cost of capital, which we believe doesn't represent the contractual process that actually occurred for us. In terms of setting the risk-free rate, we believe that a 10-year rate is a more appropriate approach. And we would note that, that would be consistent with methodology the commission has recently used in the fuel market study that was undertaken quite recently. In terms of the look-forward weighted average cost of capital, the draft decision implies a WACC of 4.88% approximately, post-tax. This number is barely above the WACC for the electricity industry in this country, which we don't think is an acknowledgment of the industry risks that we face as a business. As a comparator, Ofcom in terms of thinking about a WACC for a fiber business has a number of around 6.5%. So to summarize, we are pleased with the progress that has been made on the regulatory front, but we are looking to engage further in terms of the weighted average cost of capital. I'll hand back to JB.

Thank you, David. And yes, great results, and as an incoming CEO, it's a privilege to be able to announce those results. And I'd like to pay tribute to the great work that the entire team, the executive team and Kate specifically have done to get us there. Now going forward, as somebody who's followed the rollout of fiber networks around the world and who've lived in Australia for the last 25 years, I can't say enough how good Kiwis have it in terms of broadband connectivity with almost 1.2 million homes and businesses that have fiber right to their doorstep. And it's really something that's amazing. Fiber, there is no debate that fiber is the best and most future-proof broadband technology. And it comes down to delivering high-speed, high-capacity, reliable, uncongested, unlimited broadband. Nothing come close to a fiber connection directly into your home. And the people who already have done this here in this country are definitely taking advantage and getting the full benefits of that capacity. We've seen the demand for data grow in our network. On the fiber network, it's reached for the month of January an average of 372 gigabytes of data consumed in a month on average for consumers on our fiber network. That trend is also going across our entire network. Overall, we've seen data consumption go to 302 gigabytes in the month of January, average per customer. And on copper the trend also increases. Copper itself was growing to reach 205 gigabytes consumed per month. So great trend; more and more people, regardless of the network that they're on, consuming a lot more data. One more thing I'd say about those numbers, and they're all January numbers. We typically see a small dip in January as people take advantage of being on holidays, good weather and long daylight hours, but the growth is still massive. We are growing exponentially and the numbers are really big. If you want to compare: In 2014 -- in January 2014, the total amount of data that had trafficked over our network was 30 million gigabytes. Fast forward to last month, so January 2020, that same number over our network was 365 -- 362 million, sorry, gigabytes traveling over our network. So more than 10x increase over that period. And this is why fixed line capacity is at the heart of high-speed, reliable broadband. We think that, at this stage, about 90% of Internet traffic in New Zealand carries across a fixed-line connection. And when you see that data growth, you understand why that's the case. Further to my point in terms of reliability. The Commerce Commission published in December their independent report on broadband performance. What it showed is that fiber has extremely low latency. What it means is that you will see absolutely very little, almost no instances of lag or jitter or buffering when you are using applications such as video streaming or gaming. What it also says is that fiber, closely followed by VDSLs, are the only 2 technologies that you can reliably deliver ultra-high-definition or 4K video streaming to multiple users simultaneously. And again that goes to my point about the benefits of fixed network connections into your home to provide reliable, uncongested and unlimited data to your home. We expect this trend to become even more important as we see the volume of 4K content grow. A recent example, in Korea, South Korea Telecom is seeing about 65% of their IPTV customers already consume video streaming on ultra-high-definition or 4K devices. And then closer to home, here in New Zealand you see new service providers, like Disney+, actually making 4K content available at no extra cost. So we believe that there will be a continuing growth of data, and that's a really good trend for us to see there. Now I know that there is always a lot of interest in what a new CEO means for a company's strategic direction. I can tell you that my focus for Chorus is very much on keeping it simple and on focusing on our key core strengths and requirement going forward. So you should not be surprised to see those 4 key strategic priorities on the screen today: #1 and foremost, winning in our core fiber business; then also optimizing our nonfiber asset; and growing new revenue; and then developing the shape and the capability of our company to ensure its long-term future. Now let me give you a little bit more detail on each of these. I said that our #1 priority is to win in our core fiber business. Now what does that mean? First, it means winning in fiber penetration and market share, making sure that we bring more homes and businesses onto our fiber network and that they take on a higher business plan. We've had a very good first half of the year in terms of broadband connections and ARPU growth, and that's been combined with a slowing down of the disconnection on our network. As I said earlier, we've brought fiber to almost 1.2 million homes and businesses in this country. 56% of them have already taken a connection on our fiber network. Our biggest opportunity, our challenge is to make sure that the 44% remaining premises or homes and businesses actually come to our network and get to enjoy fiber. Now how do we do this? Previously, we've taken a bit of a active wholesaler approach, and you're going to see us actually take that to the next level in the next 6 months. By that, we mean that we can't just rely on the retailers to explain to consumer how great fiber connection is. And we want to make sure that consumers have full information or very well informed when it comes down to making a decision on how broadband is going to be delivered to their homes. So what you're going to see is us doing a lot more in helping us Kiwis understand that not all broadband connection is created equal; and that when it comes to high-speed, reliable, uncongested, unlimited broadband, nothing come close to a fiber connection to your home. We're also going to tell people that it can pay to compare and shop around. We believe that websites such as the broadbandcompare.co.nz can play a similar role to what the whatsmynumber website did for New Zealand electricity consumers in the past. We'll continue to visit various community groups, prior to rolling UFB areas, to explain the benefit of fiber, but we'll also ramp up teams visiting selected areas post the rollout in a program that we refer to as managed migration programs. It's a program that's been very successful for the first 6 months. We believe that it's helped us win back almost 3,000 off-net connections. And the bulk of that is through Chorus-led door-knocking activities in area where we've already rolled out UFB networks; and convincing people to let us connect a network termination and -- terminal -- sorry, an optical network terminal, ONT as we call them, in their premises as we are in the area. And together, that initiative, with some joint retailers initiative, have allowed us to install 20,000 new ONTs in people's premises. And we believe that we can step this up even further in the second half of the year. To give you an idea: Once we do door-knocking activity, it generates a 15% lift in uptake on average for the 12 months compared to an area if we don't do that type of activity. And in UFB2 areas specifically, we've seen a really big appetite for people to actually take on those installations. 35% of the homes that we visited have agreed to us installing that ONT in their premises within a week of our campaigns, and that in turn turns into about 50% of these taking up a service within 6 months. And the -- as we've said about the uptake in our gigabit offer, we can also drive increasing uptake and increasing speeds through retailer initiatives. So we'll continue to target specific offers to our retailers that are targeted and will help us grow the number of connections to fiber and also the uptake of higher speed tiers. And then finally, you'll see us going directly to consumers. A recent example, we've sent about 15,000 gift cards to consumers that are currently off our network with an incentive that, if they then can place a connection order for their premise in a reasonable time frame, they can then convert that prezzy card and go redeem it into a number of retailers. So these are all the initiatives that we'll have in place to grow market share and grow the penetration of fiber as we go through. Here is a quick example of how those can actually deliver good results. In the Wellington area, we've completed the UFB rollout. And in that geography, we compete with the existing Vodafone cable network. So what the graph shows you here is 2 Wellington suburbs where we've completed the UFB area and where we've seen, following that rollout, a significant uptake in fiber connections. And we believe that there is even further opportunities for us to do off-net win=back in that area. Now another big part of winning in fiber is also making sure that it is as easy as possible for people to connect to fiber and then, once they are connected, to make sure that we deliver a flawless service for the customers that are on it. The great thing about the managed migration effort that I've described is that it helps us deliver a better customer experience. I've already mentioned that the standard CX rating or the customer satisfaction rating that we measure on a standard installation, so when a customer contact their RSP to have a fiber connection. We brought to 7.8 in December and 7.9. And this is already a good result, but when you look at the managed migration program, we're able to take that number to 8 and even 8.1. It's because the fact of us installing that ONT device prior to the order being placed makes it that much faster and that much easier for a customer to connect to fiber. Another area we're focusing in terms of improving the CX experience is in reconnecting homes that have been connected to fiber when a new owner or a new tenant moves into that home. We're doing a lot of efforts to streamline the process of how we bring back those premises onto the fiber network. And we also think that there are opportunities in leveraging the new WiFi capability of our newest ONT. We've now already deployed about 130,000 of these new ONTs in our network. And we think that it could deliver a significant cost and speed improvement in bringing these type of premise back onto the network when a new consumer moves in. And we'll be doing a test of that with a selected number of retailers over the next 6 months. And in terms of CX, we've also launched a number of new services for small businesses, and those basically deliver targeted response times to them that typically were reserved to larger businesses. We've seen really good uptake of that new service. We have about 1,500 customers that have taken it up, and we think that there is some growth also left in that area. New premises growth is also a big area of work for us. It triggers a lot of pipeline of activations for us, especially in the Auckland area where the city is growing very rapidly. We've done a lot of work with developers to improve the way a customer will be able to order a new connection when they move into those new developments; and in particular in areas like the [ land market ], Westfield and Commercial Bay new development. We believe that we'll be able to deliver a significant improvement in CX when people order a fiber service. The last thing I'd like to make on CX is, if you look at the slide that's here in front of you, the strong correlation that exists between the speed tier that customers are on and their satisfaction. You can see that people that are on the highest speed tier, 1 gigabit speed tier, have a Net Promoter Score or NPS of 50. That's about a bit more than 4x the NPS of people on our lower-entry price. And it shows how important it is for us to get more and more people taking on those higher speed tiers. Now of course, in order to win in fiber, we need to understand the rules where we're operating under. David already mentioned a lot of the work that's happening in terms of the regulatory framework. At a high level, there are still about 18 months to go before we have the full commission's final determination on our price and on the quality requirements. And those will be guiding the starting RAB value and also the MAR or the maximum allowable revenue value that will be part of our regulatory framework. The good news is we're getting greater clarity as the process goes on, and the final decision on the input methodology of what we call the rule books is due before mid this year. So some progress there. And we have also reasonable degree of clarity already in some areas, specifically in pricing, with the copper and the fiber prices paths already spelled out throughout the first regulatory period that we'll be facing under that new regulatory regime. Still in regulatory. The commission has now published its draft view on the areas where copper may be withdrawn. And we also expect them to release a copper withdrawal code before midyear. And this will specify the conditions that we have to meet if we want to reach -- withdraw copper services should we decide to do so in the future. Finally, still on the regulatory front, I'd like to mention the fact that, after a lot of work and effort last year, we've launched our unbundled fiber product in the UFB1 areas. So beyond winning in fiber, I think it's important that we run the ruler over our asset portfolio, with an eye to not only the future but also the trends in connections both in fiber and in copper. And that's why our second priority is to optimize our nonfiber assets. As we said previously, that involves applying a network horizon lens across the 3 geographies that we operate under: our largest, UFB area; the LFC or local fiber company areas; and then in rural. As we have less customers in our copper network, particularly in LFC, we're thinking more about our property requirements in those areas. And we've already started divesting our less-valuable assets in those area. And so this will in turn reduce our ongoing costs for things like maintenance, for rent and rates. So we'll continue to pursue this. In the rural area, in areas where there is no or very little fiber, we've got a great-performing VDSL network that delivers unlimited broadband at some urban prices. It's doing very well. And it's seeing some growth in areas even when fixed wireless was trying to push in those areas but were facing difficulties trying to deliver the significant growth in data that I've described earlier. So we'll continue to optimize that network in those areas. And then more broadly, we'll continue to work closely with government to see how we can leverage that asset to bridge geographic and socioeconomic digital device where we can do so. Now we have a great network, and so our third priority is to make sure that we leverage that network to grow revenues both in our core business and also in ancillary areas. In our core business, you would think that 1-gigabit speed is already a great product, but we were very happy to connect our first customer on our Hyperfibre services. And those are our 2 gigabits and 4 gigabit services that use the latest technology in optical fiber and ultimately will allow us to provide up to 10 gigabit services in terms of speed. We're rolling these services on a progressive regional rollout. And those services are so fast that the team has had to build a new meter to be able to test the speed on them. We believe that those services will bring great values to customers, in particular customers that have really big amount of datas to traffic over their network, whether it's upstream or downstream but also to customers that need really instantaneous communication. Another area that we're paying close attention is the growth of WiFi standard and in particular the WiFi 6 standard. We're now starting to see WiFi 6 devices being available; and they deliver significant improvement in terms of speed, in terms of latency and also in terms of the number of devices that can be connected to a router. The second speed test that we've put on this slide shows you a test that we've done on a WiFi 6 device connected at the end of one of our Hyperfibre service. And what we were able to deliver was 1.5 gig and just 2 milliseconds in latency on that trial. So expect significant developments in terms of WiFi improvement in the home as those WiFi 6 devices become more and more common. Another thing about WiFi 6 is we've talked to network vendors. And we understand that the rollout of WiFi 6 technology and especially in premises that already have fibers coming to them can be very cost effective especially in areas that requires connection to a big number of devices. So think stadium. Think concert hall, convention centers. And we really believe that WiFi will continue to play a very important role in wireless connection. It's not all about 4G and 5G. WiFi will continue to play a role, and we think that WiFi 6 in particular will play an increasing role in that space. Now in more ancillary areas, we've talked to you in the past about our EdgeCentre product, EdgeCentre Colocation product. This is a service that's showing some really good promise. In fact, we've now filled our initial Mount Eden rack space capacity. So we're now looking at introducing further space in more exchanges going forward. Now when we look at ancillary areas and innovation, we were also quite clear then that, if it turned out that some of the things we were looking at did not have the potential that we initially thought they had, we'd move on quite quickly. This was the case with fiber to the desktop, which was an initiative that we talked to you in the past. When we did further research, we discovered that it was probably a bit early for a product of that nature. And we've now redeployed the resources that we're looking at that project onto more exciting and promising areas. And finally, the thing that I want to say in terms of our fourth priority is the fact that, as we shift from a build to an operate mode and as the new regulatory framework start being implemented, clearly there is a lot to be done in terms of the shape and the capabilities of the organization in the long term. I've already mentioned earlier that we did a big restructure of our operations, about half of our organization was impacted by this, to create a single customer network operation team. And we're now building the technology platforms and tools that will enable them to deliver a seamless end-to-end high-quality service to customers. And as we move to a RAB framework, asset management also becomes a very important thing, and so that's another capability that we're developing. And then I've also already mentioned that, as the UFB rollout completes, as copper volume reduces and as we move into a more regular operating mode, then our requirements in terms of field workforce would evolve. So there are a lot of things that are happening in our business in terms of shaping the organization and developing new capabilities. The good news is that our team is very resilient, and in fact, they seem to be thriving in all this change. We measure on a regular basis the engagement of our workforce. We do that on a quarterly basis, and it's been very encouraging to see that engagements continue to grow. And we'll be closing our February review, our latest review, at the end of today, but already we see some really strong numbers, reaching 8.1 in terms of employee engagement as of this morning. And also, what we call the employee NPS or employee Net Promoter Score has significantly increased over that same period. Conditional upon the final number at the end of today, we were at 48 this morning. So again, it shows that our team is ready and up to the challenge, engaged and look forward to capturing those opportunities. So I'm going to stop it here, and I'm now going to open to questions from the call. I believe that the operator will now take on your questions if you can put them to them.

[Operator Instructions] Your first question is from the line of Arie Dekker from Jarden.

Just first question just relates to how we should sort of think about the dividend and some of the high-level guidance you've given. Firstly, just an update on what the right range is to think about in terms of sustainable CapEx from FY '22, FY '23; and what some of the swing factors in that range could be.

Sure, Arie, no problem, happy to answer that. We've given a definition in the -- on the slide of how we come to sustaining CapEx in terms of what that number might look like in the future. Of course, I need to caveat that, that there'll be some various factors that will influence that, but if you wanted to take an estimate, I would say $200 million will be a reasonable midpoint within a range. The components of sustaining CapEx, there's 3 of them. There's copper. There's common and there's fiber. Copper will continue to transition down. Common, I would think, over time would be broadly consistent, albeit the copper component will also have an impact. And fiber, we'd expect to be steady over the medium term. So that will be my best estimate, Arie, based on where we sit today.

Sure. And then I guess there's a reasonably big gap in FY '22 and FY '23 but less in FY '22 in particular between your underlying free cash flow and your sustainable free cash flow. I guess, up until this point and through the next 18 months, the Board's approach has been for modest growth in the dividend through that period, where the underlying free cash flow has been something that's constrained a higher dividend. How should we sort of think about the approach you're looking to take given what could still conceivably be very little, if any, free -- underlying free cash flow in FY '22 in particular with regard to step-up in the dividend?

Sure. No problem, Arie. We very deliberately use the word transition to the new policy. I've used the word "short term," which if you look at the slide around the key drivers that's connection CapEx and UFB2, short term means 2 years. And if you dissect that a little bit further, most of the impact of the completion of UFB2 and a higher proportion of connections is in full year '22. So that would be where most of the gap will be that you're referring to. The way that we'll look at the dividend for that year as we transition to the new policy is that the credit rating down driver threshold will be the key constraint for us, and we will optimize our dividend within that constraint. Now I would also reference, as I talk about, that Moody's have indicated that they will be reconsidering our credit rating threshold in full year '21. And if they were to move, and S&P, of course, then that would be another factor that we would need to take into account. So the dividend will grow through the transition period. It will be a 2-year approximate period; and the majority of the impact will be in the first year, which is full year '22.

The majority of the CapEx, you mean, as opposed to the majority of the growth in dividend.

Correct. Correct. That's right.

Yes, sure. And then just in terms of the imputation credits. You made a reference to your expectation that you'll have exhausted the balance around FY '22.

Yes.

Can you just sort of talk to what the likely approach is going to be with regards the dividend and imputation credits? I mean I guess you're basically signaling that you'll be happy to pay an unimputed dividend given the outline of the approach being taken and the focus on a high payout of free cash flow. Is that right?

I think that's reasonable, Arie. Just to give a little bit of color to that: We expect the imputation balance, as best -- based on current forecasts, will run down by full year '22. It's then not a long period again before we would start paying tax, so yes. If we didn't have an imputation balance, our dividend would likely be unimputed, but I wouldn't like investors to think that, that will continue for a long period of time. It is a shorter period of time from that point.

Similar to the transition definition of short period, a couple years.

I wouldn't want to be too specific, Arie, but I will say it wouldn't be a long period of time. Thank you.

Yes.

Yes, sure. And just the last one on this dividend question line, just the DRP. I mean you've obviously come out with a pretty confident view on where the dividends go. And why have you decided to keep the DRP in place?

Sure, yes, absolutely. We consider, as management and as a Board, at each reporting date what to do with the DRP program and also the discount itself. Given the strong performance in the first half of the year, given that broadband connections and revenue are growing and the strong performance on the cost line, we were confident and comfortable within our credit rating metric threshold that we could remove the discount from the DRP. In terms of leaving the DRP on, we do believe that it's important that we continue to demonstrate our commitment to our credit ratings, in particular thinking about the messaging to the agencies. We are very committed to BBB and Baa2, so we thought that it was more appropriate for us to have, to take one step forward, which is to remove the discount. And then again, at year-end for full year '20, we'll reconsider the overall program as well.

And just a couple of quick questions probably. Just the incentives on copper seemed quite high. Can you just talk about where this has been targeted and whether we should expect that to come down? I mean it's being sustained at what looks like a reasonably high level against the ongoing decline in the number of connections you have in that space.