Chorus Limited (CNU) Earnings Call Transcript & Summary

Ladies and gentlemen, thank you for standing by, and welcome to the Chorus Regulatory Process Update Call. [Operator Instructions] I must advise that this conference is being recorded today, Thursday, the 17th of December 2020. I would now like to hand the conference over to your speaker today, Mr. David Collins, Chief Financial Officer. Thank you, sir, and please go ahead.

Thank you, James. Good morning, colleagues. Thank you for joining me. It's a pleasure to be with you today. I'll be, this morning, talking through the presentation which we've released to the exchange this morning. I'll step through each of the slides at a reasonably high level and touch on the key points, and there will be an opportunity to ask questions at the end of the call. When I've completed the presentation, I will hand back to James, and he will moderate the questions for the call. So kicking off, and I'll call out the slide numbers as I go through, starting with Slide 2. Fiber, we have made a substantial investment in fiber over the last 10 years or so. We've put New Zealand in a leading position globally in terms of the quality of our fiber asset. We've seen over the last few years and particularly most recently with our experience with COVID-19, fiber is an essential service for New Zealand. And we certainly expect the demand for data usage and for our product will continue to grow. Our price quality proposal seeks to leverage this investment. We seek to drive better outcomes for consumers through the expenditure that we propose, both in terms of OpEx and CapEx. More specifically, our proposal includes the completion of the UFB build and the continuing growth of our fiber uptake or connection levels across our assets. Our proposal looks to maximize consumer value, both now and into the future through efficient costs, but also through promoting fiber and investing in new products and technologies to leverage our assets. We also look to smoothly transition through the completion of the build phase and into our operate phase as a regulated utility. Lastly, we do plan to continue our active wholesale position in the market. Hence, our proposal does include the continuation of the use of in-market incentives, the retailers which we believe ultimately benefits consumers through better retail offers and choice as more and more consumers connect to fiber. Looking at Slide 3, an overview of our price-quality expenditure proposal. We've submitted yesterday to the Commerce Commission an operating and CapEx proposal for the first regulatory period, which runs from January 1, 2022, to December 31, 2024. We've elected today to release this proposal to investors and analysts to give you an opportunity to get a better understanding of our proposal and particularly to do so ahead of the financial reporting season in February. The Commerce Commission, we expect will release more details of the proposal in quarter 3 for feedback from stakeholders, and that will provide more detail around what we will talk about today. The reason there is a delay in time is that there is a confidential information review process, which needs to unfold in the coming weeks before the commission will release details in the new year. Importantly, the proposal builds on our business as usual processes. Our 5-year business plan underpins our regulatory expenditure proposal. And of course, we have made updates for developments that have occurred since the end of full year '20. This does reflect Board and management's best view of the operation of Chorus and our most current plans for our business in a market that is very dynamic. We do make judgments around allocation of costs, firstly, between costs that are directly attributable to fiber or directly attributable to copper; and then secondly, where costs are shared what the appropriate allocation approach for those costs will be. The Commerce Commission will be making the final decision on allocation principles as part of the price-quality process in calendar 2021. Importantly, the regulatory period is based on calendar years rather than financial years. Hence, you'll see that flavor through the document. Lastly, we have subjected the submission to an independent verification process from an external party. That report will be public as part of the proposal when the commission publishes this early in the new year. The purpose of that report was to demonstrate that our costs are efficient, and they reflect good telecommunications industry practice. Moving on to Slide 4, translating our regulatory proposal into, call it, financial reporting speech. And there's a couple of important differences, which I've called out on this slide. Firstly, the proposal itself covers fixed fiber line access services only. And I'll call those or refer to those as FFLAS going forward. It therefore excludes other fiber and copper-related costs that will be incurred in future years and also FFLAS costs in local fiber company areas. So when we talk about the proposal, we refer to it as price-quality FFLAS. So it excludes those 2 areas that I've just mentioned. In terms of the cost categories, we'll step through those a little later in the presentation. For regulatory purposes, the categories are based on functional groupings. So they do look different to the financial reporting categories that we give to market as part of our statutory results each period. The functional groupings are set by the commission and they're defined in the input methodologies process, which is now final. They are broadly consistent with how other regulated entities report, and we provide a view through the presentation of how they track back to our financial reporting categories. Importantly, as I mentioned before, the regulatory forecast is a calendar year basis, so there is a difference with our financial reporting. And then there are a couple of nuances within both the OpEx and CapEx proposal which we'll call out through the presentation, but to call them out briefly now or at least the major ones. IFRS 16, so right-of-use assets. As our listeners would be aware for our financial reporting purpose, they are included as assets and liabilities on our balance sheet. For the purpose of the regulatory proposal, and this is consistent with the way other regulated entities with the commission work and is a requirement that the commission has set is that we give a view for presentational purposes of our IFRS 16 lease cash flows within the OpEx proposal. Hence, you'll see that there is a call-out for IFRS 16 cash flows when we go through OpEx. The reason for that is to allow the commission on a cash flow basis to compare our cash flows with the pre-IFRS 16 period. So when we think about what our OpEx building block will ultimately be, it will not include IFRS 16 cash flows in terms of the flow-through to the MAR. It's included for the purpose of a cash flow or comparison with prior periods. Secondly, capital contributions. The most common example of this where we do new property developments, and we receive a developer contribution towards that cost. In our statutory reports, you normally see that as part of field services revenue. The way that the commission through the input methodologies has specified that we treat capital contributions is that we net them off the CapEx or we net them off the RAB. So what that means, for example, new property developments is they are effectively not in the regulated asset base because they've been netted down. And hence, those revenue lines within field services revenue, when you look at our accounts going forward, will be excluded from the MAR or separate from the MAR. And thirdly, pass-through costs. We have a couple of costs, specifically local body rates and regulatory levies, which, again, the commission has specified through the input methodologies process will be treated as pass-through. What this means is that we exclude these costs from our OpEx building block. Hence, you'll see these are called out through the presentation, but these will be separately reimbursed through a separate regulatory process with the commission. So when you think about these costs in our profit and loss account for statutory reporting, they will be reimbursed separate to the MAR. So hence, the OpEx building block excludes pass-through costs. Last call-out on this slide, regulatory inflation allowance. The reason we call that out is in our existing business plan forecasts, as you'd expect, we have an allowance for CPI. Undoubtedly, the forecast that the listeners on this call have to do the same thing. For regulatory purposes, that is also the case. It's just that there are different indices and allowances that the regulator uses. So it's consistent with what other regulated entities are subjected to. And ultimately, when you see our regulatory templates, you'll see both constant and nominal values. Throughout this presentation, we quote nominal because we think that's the most relevant in terms of our listeners. Moving on to Slide 5 and looking at cost allocation parameters. The Commerce Commission has given guidance on cost allocation parameters as part of the input methodologies process. This slide that you're looking at is consistent with previous disclosures that we've made. So there's nothing new on this slide. We have followed the approach in the IMs in determining our OpEx and CapEx building blocks. Commission will determine the final approach to cost allocations as part of setting the initial asset value or RAB process in the first quarter of calendar 2021. The assumptions that we have used to underpin this submission are consistent with what our proposal will be for the initial asset value. We will be making a separate submission to the Commerce Commission in quarter 1 next year to justify our approach to cost allocations, both in terms of what underpins this submission and also what will underpin our initial asset value proposal in the new year. Moving on to Slide 6, our OpEx building block. Our proposal for OpEx is nominal FFLAS OpEx of $625 million for regulatory period 1. When you think about what that means related to our statutory numbers, you need to be aware that it includes IFRS 16 lease costs of $41 million. Per my earlier comments, that's the presentational and comparative purposes for the commission. It excludes pass-through costs of $45 million. These are in 2 main buckets: firstly, rates or FFLAS rates; and secondly, regulatory levies. And lastly, my comment about regulatory inflation. So $625 million nominal. Moving on to Slide 7. The categories that we have used for OpEx. As a reminder, these categories have been set by the Commerce commission as part of the input methodologies process. We have given a functional view rather than a cost element view as required by the commission. What we've done on this slide is give you a view or to put it differently, we've mapped how these categories, both in terms of customer, network and support and in the various subcategories, what they actually mean. And then secondly, how they map to our statutory reporting categories. You'll note that there are some categories, which appear in a number of cost elements. And likewise, some cost elements, such as labor, appear in a number of subcategories. So we try to map those 2 to give you a view of how it relates to our statutory reports. A couple of specific call-outs on this slide. My comments about IFRS 16 costs. For property leases, they are in terms of our leases with Spark included within network operating costs; for our office property leases, which are IFRS 16 costs, they are included within the corporate line within support. In terms of pass-through costs, for rates, they are included within the network operating costs line. In terms of pass-through costs for reg levies, they are included within the corporate line within support. So that's a view of -- an overview of our mapping from reg categories to financial reporting. Moving on to Slide 8. This is the regulatory template and gives the values for each of the categories in nominal terms over the 3 years. As I mentioned before, when you see the reg templates themselves, you'll note that there's a constant dollar view, then there's an adjustment for regulatory inflation and then there is a nominal view. Again, we're showing you the nominal view today. The call-outs on the right, I've made comments on already. So that's probably the key call-outs in terms of our OpEx submission for our building block. Looking at Slide 9, we thought it would be useful for our investors and our analysts to give a financial reporting view on an indicative basis of what these allocators would mean in terms of our cost element view or statutory reporting view. We've given this for numbers that are already public, which is our full year '20 view. We've elected not to give a forward view over '22 to '24 by statutory reporting cost element at this point as I feel that would be a little too granular and a little too specific at this early stage of the process, but we did want to give you an idea of how this is likely to flow through our statutory reporting cost elements. The headlines here is for full year '20, we estimate FFLAS OpEx was around 55% of our total OpEx. So $170 million out of $311 million. Critical point to call out on this slide is that as we look forward, we would expect that FFLAS proportion to increase significantly as our fiber asset grows and as our copper asset or network shrinks. So as we look forward, whilst we haven't given a FFLAS percentage as it applies to our regulatory submission or our OpEx building block, you should expect that number to be higher than what it was in full year '20. A couple of other call-outs when you look at this slide. When you're thinking about relating this to our OpEx submission, bear in mind that the slide, our full year '20 results include pass-through costs of $11 million, which are excluded for reg. And of course, because this is our statutory P&L, it does not have finance leases flowing through it. A few specific call-outs in the numbers themselves. Looking at network maintenance, that is quite a low percentage of FFLAS, which reflects the fact that our copper network is older and aging and has a higher level of fault and reactive maintenance. And you'll note that we've given disclosure on our maintenance cost as part of our statutory reports previously. As we look forward, we expect fiber maintenance will grow and copper maintenance will fall. Looking at other network costs, that includes a higher proportion of copper-only costs and that's -- examples would include some warehousing costs, radio license fees and a number of other costs that are copper specific. And secondly, this line in our P&L includes examples where we are able to recoup the cost of copper expenditure in our network. So for example, chargeable copper network maintenance and also network relocations. So hence, the remaining FFLAS portion is quite low for other network costs. Lastly, looking at the rent rates and property line, just calling out, that's quite a low FFLAS proportion. For us, a significant portion of our property assets are in non-UFB areas. And secondly, copper does take up more space in a number of our network properties. So at the moment, it has a larger share of our -- particularly our property maintenance line. So that's a view of full year '20 in terms of our financial reporting. The key takeaway from the slide is if we apply the cost allocators to full year '20, we get 55% as the FFLAS proportion. When we look forward, we expect that percentage will grow. Changing gears now to CapEx on to Slide 10. We estimate our total FFLAS CapEx will be $1.29 billion for regulatory period 1. A couple of important call-outs. As I mentioned earlier on, the way that capital contributions will be treated as determined by the Commerce Commission with the most common example being greenfield developments is that the capital expenditure will be netted down for that capital contribution. The total of that over the 3 years is about $56 million. As I mentioned earlier on, expenditure in local fiber company areas on fiber is excluded from our submission. This is not a material number, but we do have a small element of fiber activity in LFC areas. Moving on to Slide 11, the detail about our regulatory CapEx categories. What we've tried to do on this slide is to take the categories that the commission has set because as a reminder, these are determined by the commission in consultation with us and in the subcategories that attach to each of those to describe what it means and then to give you a view of where you have seen that previously and going forward in our financial reporting categories. As with OpEx, it's a functional view rather than a more detailed expense element view. To call out the key things from the slide. If you look at the first 2 CapEx categories, which is extending the network and installation, broadly, the total of those 2 is our nonsustaining CapEx going forward. So that includes things like the completion of the UFB build, which is scheduled for December of 2022. It includes growth CapEx for greenfields, albeit netted down for the capital contribution. That's both within extending the network and then within installations, we have our connection CapEx. So that's how we have tagged it previously in our statutory reporting and the connection CapEx or the installations CapEx, I should say, also includes customer retention costs. If you then look at the remainder of the categories, IT and support, network capacity and network sustain and enhance, broadly, that is equivalent to our sustaining CapEx going forward. We just have more detail around the components within that. That's probably enough on the categories themselves. You'll see a lot more detail on what is in and what is out for both fiber or both CapEx and OpEx in terms of our categories when the submission is made public by the commission in quarter 1 of next year. Moving on to Slide 12. This is the CapEx regulatory template for the 3 years. Once again, this is in nominal terms, and when you see the template from the commission, you'll note there's both constant and nominal dollars. Key thing that we've called out on the slide is we've separated out what are the nonsustaining CapEx components, which are extending the network and installations versus the sustaining CapEx component, which are IT and support, network capacity and network sustain and enhance. There are a few nuances in terms of where particular items are recorded for the regulatory categories. As an example, the nonsustaining CapEx does include some related customer retention costs which we would actually treat as sustaining CapEx. To give you an idea of quantum, that was about $7 million in full year '20. So you just need to be aware of that. We've also given a view here of what the underlying assumptions are on installation levels. So 165,000 standard installations over the 3 years, 20,000 backbone and 10,000 complex. Importantly, as you will see when our submission is published, we are proposing a wash-up mechanism for volumes within installation CapEx. It's very difficult, of course, to forecast what installation levels will be, what demand will be. We have a robust and detailed process to build that up. But of course, forecasts will always differ to what actually occurs. So we have proposed a wash-up mechanism for volume-related risk, underlying installation CapEx within our proposal. The key messages from this slide, if I was to summarize. Firstly, on sustaining CapEx, we've previously given guidance to market that we expect a reasonable midpoint for the total Chorus sustaining CapEx to be $200 million per annum going forward. That is consistent with this regulatory submission and that is unchanged. In terms of the assumptions underpinning extending the network, so I think UFB, in particular, but also new property developments and installations or otherwise known as connections, those numbers are consistent with forecasts we have previously provided to the market, in particular, in our year-end presentation on Slide 24, where we gave a view looking forward of what we believe our connection CapEx will be. So that is consistent with existing forecasts and guidance that we have provided. I think that's probably enough in terms of the CapEx regulatory template. I did want to make a few comments on revenue. We have been working through finalizing the definition of what of our revenue lines are included within FFLAS revenue and what is not. We are very close to getting a view on that. Our intention is at our half year results in February, we will give a view, firstly, for full year '20 for our actual revenue of what the FFLAS component will be, and we will also give you a view for our half year results of what the FFLAS component of that will be. To give you a round number or to give you an estimate, if you were to look at full year '20, we expect that our FFLAS revenue will be approaching $500 million, so a little bit under $500 million for full year '20. Importantly, and I emphasize this point, that will grow materially looking forward into RP1 as our uptake levels grow and as our connection mix changes. So as we think about the appropriate comparator for the MAR in RP1, it will be the FFLAS revenue, and that number will grow significantly from full year '20 into the regulatory period 1. A little bit more flavor on revenue. I mentioned that we're working through some of the details around what is included and what is not in FFLAS revenue. To give you some headlines about what will be outside the FFLAS revenue net. Firstly, capital contributions, as I mentioned earlier on, for greenfield developments, the CapEx and therefore, RAB will be netted down for capital contributions. That means within the MAR, there will not be a return on or return of capital on those assets. Therefore, in our statutory revenue lines, field services revenue for greenfields will be excluded. Secondly, fiber revenue in LFC areas will be outside the price-quality FFLAS revenue net. That's not a significant number for us, but we do have a level of revenue in LFC areas for fiber. Thirdly, we have some legacy backhaul services, which will be outside the FFLAS revenue net. We will give you specifics on that at the half. And lastly, within our infrastructure revenue line in our stat accounts, there is a portion of copper colo revenue. So revenue we received for the rental of our exchange sites where that relates to copper. So they are some of the key exclusions from FFLAS revenue. We will give you detail on that at the half. And to give you a round number for full year '20, it will be -- we expect a little bit under $500 million, but growing forward into RP1. Lastly, in the appendices, we thought it would be relevant to give an update on the copper withdrawal code that was recently finalized by the commission, and we've given a few details within the appendix. I won't step through that for now, but I'm happy to take questions, if need be. On Page 14, in Appendix B, we've given just a reminder of what the look-forward position is for the commission and ourselves in terms of what's next. And on the right-hand side, this is a different depiction than what we have shown previously, but this has come from one of the Commerce Commission's recent papers. And it just talks about how, firstly, the RAB roll forward works, in the left-hand column. In the middle column, the return on capital or WACC. And in the third column, the actual MAR or building blocks allowable revenue. And in red, you can see the parts that are being determined through the price-quality process. Lastly, appendix C, you've seen this one before, the key components of the RAB or MAR calculation, in particular for WACC, through the input methodologies. So I hope that's been useful. Thank you for joining us. We're going to move to questions now. So I might hand back to James, if you could moderate for us. Thank you, James.

[Operator Instructions] Our first question comes from the line of Arie Dekker.

Yes, just a -- first question just on the labor costs for FFLAS FY '20. I guess the thing that just sort of surprised me is the $7 million of nonregulated against, I guess, what is still very substantial nonregulatory revenues in that period, including $370-odd million of copper revenues. What's sort of driving such a low apportionment of labor?

Yes. Sure, sure, Arie. I can make some general comments on that, and there'll be a lot more detail on this in the submission we provide to the commission in early Feb. Probably the key headline here is the -- our corporate-type costs, which are a significant component of our labor bill. When you think about Chorus, we were established to be a fiber business. That's why we were established. So a lot of our corporate functions, if you think about Board, as an example, the company secretary, et cetera, they were established specifically for a fiber business, or to describe it differently, they were an incremental cost, which occurred because we were established as a fiber business. So we've looked at across a number of our corporate costs, and we believe they are attributed directly to fiber. There's a lot more detail to this, but that's the concept that sits behind the higher fiber proportion for labor. And again, there's a lot of detail on this, Arie, I just can't go into it yet, but we will be making a submission on this in early Feb.

Just in terms of that concept, is that a concept that was well covered in the IMs and accepted through the IMs or guidance provided that, that would be acceptable?

The IMs have given us a broad framework to operating in terms of cost allocations. The details, the commission hasn't given a view yet because we haven't given a proposal in detail on this. So it wouldn't be fair for me to say, yes, that's been agreed because it hasn't. What has been agreed is what's on Slide 5, which are the principles. And then there's judgment that needs to be applied. So that has not -- none of the cost allocation approach that underpins this submission has yet been agreed by the commission, what has been given guidelines on Slide 5, and we have interpreted those and we'll be providing our support for that, including, as you'd expect, independent experts to support in early Feb.

Sure. The OpEx is reasonably stable over the '22 to '24 period for the regulatory building block. Your view on just total OpEx over that period, including the portion that isn't.

Yes. Sure. Sure, Arie. So I -- firstly, in terms of Chorus level OpEx, so total for the company would reiterate some of the points that we made over our year-end reporting process. I would expect our total operating cost base looking forward to decline on a consistent, but gradual basis. When you look at our full year '20 results, we did have a significant cost reduction from $334 million to $311 million. That was some 6% or 7%. I don't expect that level of reduction to continue. But I do expect it to be a gradual, but consistent reduction in our total cost base looking forward. Into the other element, of course, that feeds the OpEx proposal is the FFLAS percentage because there's 2 elements here. There's the -- what's the total cost base and then there's what's the total FFLAS percentage. So whilst the total cost base will gradually decline, I expect the FFLAS percentage will grow over time. So they're the 2 kind of competing levers. What I haven't given here is the FFLAS percentage assumption in RP1 because to do that, would have given a forecast for our total cost base, wasn't quite ready to go there out to 2025, but I can reiterate those broad comments that I've made previously about how I see our cost base looking forward.

Sure. The connection costs that you've submitted or provided a snapshot of with regards to submission for '22 to '24, they looked reasonably high against the number of connections you're assuming, presumably some additional costs around more complex, maybe UFB2 a bit higher. Can you just sort of talk about that? And also what level of incentives costs you've factored into the building blocks over '22 to '24?

Right. Yes. Sure, Arie. In terms of doing the maths, and I did agonize over this because I knew you'd do the math on the number of installations. The key thing to call out is that the regulatory category for installations, it includes customer retention costs. So when you're trying to do a calculation of like cost per premises connected, the numerator, which is the dollars here have got customer retention costs in them. So that's the first point. In terms of how much is included in the submission, we will be seeking to claim or we have sought to claim confidentiality over a component of the customer retention costs in our forecast because they are market sensitive. So what I can't do is give a view of what those costs are in our submission. And specifically, the parts that are confidential are the incentive components because we wouldn't want that to be available. We think that's commercially sensitive. So that's the first comment. And you can get a reasonable view if you look at our full year '20 spend, that's a reasonable starting point, but I just can't be any more specific when I look forward within RP1. The other component is the backbones themselves. So there's 20,000 backbone installations in RP1. In our full year '20 results, we gave a view of what a backbone cost looks like and they are expensive. So that's the other component that will be within installation costs. Most importantly, from my perspective here, if you were to look back at the view that we've given the market previously on what we believe connections CapEx will be going forward, which is on Slide 24 of our full year results deck, and it was the purple dotted line if you happen to have it nearby. These are broadly consistent with what's implied for full year '22 and '23 on that chart. So that's I just wanted to ...

What chart is that, sorry, David?

So on Slide 24, within our year-end results pack. We give a forward illustrative view of our connection spend. So you've got to allow for calendar versus financial year, et cetera, et cetera, but in the round, they are similar.

And then just last one for me. I mean I guess you've given a view there on FY '20 of revenue of just under $500 million or $500 million give or take. You're pretty well progressed now on various things, including these 2 building blocks. In this spot interest rate environment that we're in, what are you sort of seeing in terms of the MAR versus that $500 million in FY '20?

Sure, Arie. Firstly, the comparison for the MAR will be against a significantly higher number than $500 million because by the time we get into RP1 we'll be much further connected. So that math is important. At a macro level, we expect to be under our MAR. The reality for us is that we won't be fully connected across our network. So we wouldn't expect to be able to earn our MAR. And secondly, our starting RAB will have a significant financial loss asset in it, which I wouldn't expect we would be able to recoup immediately. So our expectation is still that our FFLAS revenue will be under our MAR for the first regulatory period at least. And yes, I just can't give actual numbers just yet, but we are getting closer to being able to do that. But conceptually, for a network that's fully built, but only 2/3 is connected, that has a large financial loss asset at inception. We would not expect to be able to recoup all of the MAR immediately on that. So we expect to be under our MAR.

Sure. And when will that additional -- when are you expecting to provide that additional detail?

Sure. So the -- if we look forward at what's going to happen in calendar '21, the commission's time line for a draft decision on RAB and MAR is June or end of quarter 2. So that will be a commission draft view. But ahead of that, around March, late March, we'll be providing our view of the regulated asset base or starting view, so the models that are attached to that which the commission will then put through its processes. So that's -- they're the broad time lines. And then a final decision later in the year from the Commerce Commission.

Will you do something similar at half year reporting as you have done here in terms of -- you've already signaled the revenue, but on those models that you might be providing the commission around March, will you be doing something similar there in terms of high level sort of overview of what's coming?

I think I'm not sure yet, Arie, until we get a little closer. I think it's less likely that I would give detail on views on RAB and MAR at that point. I think that's too early. We'll certainly be at that point at year-end. I will give, as I say, some color on revenue at the half. But I think it will be a little early for a RAB or MAR view at that point.

Our next question comes from the line of Ian Martin.

Look, I'm interested in asking questions about how the -- how your revenue shapes up against the MAR, so I understand the constraints you're at, at the moment. Can you tell me just in terms of what's in the FFLAS revenue? Does it also -- does it include all of the 5 products, including gigabit products? And there was a question at the full year results where JB talked about another round of fiber investment out to devices in the community, streetlights and traffic lights and so on. Whether that might be included or excluded in the regulated asset base? Has that been working out?

Right. Sure. Yes. So in terms of gigabit fiber investment, yes, I would expect that would form part of the regulated asset base and therefore, the MAR. And secondly, things like -- I'll refer to them as the Internet of Things. So fiber expenditure to whether it's poles in the street for bus stops or traffic lights or CCTV-type connections, I would also expect that would be within the regulated net. So we've got an appropriate level of allowance for those sorts of expenditure within our submission, and that would be -- our understanding is that would be within the RAB, and it would be within the MAR.

All right. And your comment about that you're likely to be well below the MAR with the revenue expectations you've got. Does that hold up even at these low interest rates we've got contributing to that MAR boundary? How sensitive is that outcome to low interest rates?

Sure. Sure. Look, there's no question that interest rates are low. We have the same problem as all other regulated entities on that front. I mean the key factor here is the starting RAB value. So our view based on the likely range for that is that we will be under our MAR. So it won't be as far under as it would have been if rates were higher, absolutely. But I don't think they're at such a level that, that would bring that into question.

Our next question is from the line of Phil Campbell.

Just a few questions from me. I noticed in the release, you do have a regulatory inflation adjustment built into the numbers. I'm assuming from like an analyst point of view, particularly on the RAB, obviously, we do include revaluations within our numbers. So have you effectively included the revaluations into your CapEx estimates so to compare to our numbers, we need to like deflate them? Is that correct or adjust for the revaluations?

Sure. So the way to think about that is as -- it kind of depends how your model works, but if I think about our business plan, as we forecast our CapEx out 5 years, we allow for CPI in that. So our CapEx assumptions are inclusive of CPI. So -- and then how that feeds into your RAB model, I guess, it will depend how you're doing it, but we can't get it twice, so that the RAB will be indexed each year. And if that's how that model is working, then you would feed in constant values. And as I mentioned earlier on, we've got, within the templates, you'll be able to see both constant and nominal values. But to reiterate the way indexation works within the RAB, and I know you're aware of this, but it's worth just to note it. The WACC itself is a nominal number. The RAB is indexed annually by actual CPI and then the return on capital is deflated by a forecast allowance for inflation. So you don't get inflation twice within the model. So I hope that's clear, Phil, but it's -- that's how the -- that's how inflation is treated within the model.

Yes. Okay. So I think I understand it. So basically, if I'm comparing the CapEx numbers, and I've got indexation on my RAB, I just should be effectively inflating my FFLAS CapEx by inflation to kind of compare it to yours, would that be right?

If you've got constant dollars in there, yes.

Yes. Okay. That sounds good. The other one was just, obviously, in the OpEx and CapEx, you quantify the adjustments. And like, for example, in CapEx, you've got $56 million of the capital contributions. I'm assuming that's over the 3-year period, not per annum, is that correct?

Correct. Yes, definitely. Yes.

And the same would be IFRS adjustments and...

Yes. Yes, definitely.

Okay. That sounds good. Yes, I suppose just following on from Arie's question here. The CapEx for connections does seem a lot higher than what I've got and what consensus has got. And it does seem as though is the 20,000 of these backbone connections -- because my model I've assumed that those are pretty much all done by about FY '21. So I was wondering if you could just give us a bit more color around what that's possibly including? And then obviously, the cost of those, is there any -- you have given costs historically on what those connections are, but I'm just wondering if you can give us any more color on how much of that kind of connection CapEx would be due to the 20,000?

Right. Okay. I guess at a macro level, just reiterating my earlier comment, if you compare the installations CapEx in the reg proposal with the guidance we've given market previously for '22 and '23, it's in line, which is on Slide 24 of our year-end pack. There are nuances within the installation line which you're right to call out. Backbone is more expensive, there's 20,000 of those that we've assumed. And there's a view in our full year '20 results of the average cost of a backbone build, which I'll come back to in a moment when I can see what it is. And thirdly, the installations CapEx includes customer retention costs for reg. So that will distort the cost per premises connected account, if you're trying to divide the installations to this number. But the key takeaway is that, that is consistent with what I've advised market of previously. So I wouldn't expect that to be a significant increase. The other comment I would make is, as I mentioned earlier on, we are proposing a wash-up mechanism for the volume component of our installation forecast. So there is no volume risk attached to installation CapEx.

Yes. Just a couple of follow-ups on that, I suppose. One, could you just maybe explain a little bit more detail how the wash-up would work? And then I suppose, if I go back to the full year result presentation, Slide 26, we talked about the dividend kind of assumptions. Obviously, it looks as though from this presentation, the sustaining FFLAS CapEx is pretty similar to what you've guided to, but the nonsustaining is quite a bit above. So I suppose the question was, what's the implications for the DPS given that it's quite a bit above?

Yes. So Phil, again, I don't agree that it is above what we have advised previously. But even if you were to assume that, what it would impact would be the transition period. So our dividend framework and policy talks about a credit rating constraint over a transition period, whilst communal build finishes and whilst connection CapEx tapers off. And then you have a more BAU period with low levels of connection CapEx and the communal build finishing. So if you were to agree, which I don't, if you were to agree that this is higher than what has been assumed, what I would suggest is that would impact the early transition period. So full year '22, '23 which we had already been pretty clear on would be constrained by the credit rating, but I would not expect that to affect full year '24 or '25. The other component that's in here, Phil, is our credit rating down-driver levels. So both -- Moody's have said publicly that they will reconsider our down-driver thresholds when they have certainty over the regulatory process, which we would expect to be next calendar year. And S&P have also, in their most recent release, talked about getting greater comfort with the credit risk around Chorus and took a first step towards re-rating us in December of last year, which was a lift up from 4x to 4.25x net debt to EBITDA. There's a few things in that, but I would not expect this to impact the market view of dividend forecasts looking forward.

Okay. That's awesome. Yes, I suppose just going back to a Slide 24, the result. I suppose maybe I should have paid a bit more attention to it, but it does have illustrative only written on it. That was the only, I suppose, it was more -- I'm just kind of looking at it more directional. So maybe just a wash-up mechanism, if you could just maybe give us a little bit more color on how you expect that to work will be good?

Yes. No, no worries, Phil. And yes, you're absolutely right. It's says illustrative, but yes, it's a good illustration, if I can describe it that way to you. So the wash-up mechanism, the nuances or the detail of this hasn't yet been agreed with the commission that will wash -- I won't say wash-through, that will be decided through the price-quality process next year. But the way that we've proposed that it would work is that, as you're aware, with our connections process, we've got an outsourced model. So we have a couple of service companies who provide the services to us. We have unit rates agreed with those companies. So what we've proposed to the commission is that we would take unit rate risk on our installation or connection spend going forward. But then if actual installation levels were different to what we had forecast, that would not be a risk that we think it would be appropriate to take. So there would, therefore, be a process to correct the MAR for that in a subsequent period. What we can't yet describe to you in detail is how that would actually work because that hasn't yet been thought through by the commission, but at the most macro levels over the course of the asset's life, the amount that would get into the RAB and the MAR for installation CapEx volumes would equal actual volumes. So that's the concept over the life of the MAR and then it's around timing of how and when those wash-ups would work. So unfortunately, I can't be more specific than that right now, but that's how we've proposed it.

Do you think that would be annual? Or is it going to be more in between regulatory periods?

Not sure yet, Phil. I probably couldn't comment yet on that.

David, we appear to have no further questions at this time.

Okay. That's great. Thank you, James. Well, thank you, listeners, for joining us today. I know there was a little bit in that. I appreciate your support and queries. Brett and I are happy to talk with any of you later on in the day, or if you have other queries, please don't hesitate to reach out. But otherwise, I'll sign off now, and we'll be in touch soon. Thank you very much.

Thank you. Ladies and gentlemen, that does conclude today's conference call. Thank you all for attending, and you may disconnect your lines.