Good morning. My name is Ross Moffat, Telstra's Head of Investor Relations. Welcome to our 2020 half-year results presentation. Before we commence, on behalf of Telstra, I'd like to acknowledge and pay my respects to the Wurundjeri and Boonwurrung people of the Kulin Nation, the traditional custodians of the land we're meeting on, and I pay my respects to elders past and present. After presentations from our CEO, Andy Penn, and CFO, Vicki Brady, we'll be taking questions from investors and analysts, and I'll now hand over to Andy.
Thanks very much, Ross, and good morning, everybody, and welcome to Telstra's results announcement for the half-year ended the 31st of December 2019. Our results for the half-year were in line with our full-year guidance and continue to reflect the significant changes that are underway across the business. While they reflect the headwinds that we continue to face, particularly in relation to the NBN, they also show that we continue to build positive underlying financial momentum as a result of our T22 strategy. Before we get into the numbers, however, I wanted to just give a brief update on our bushfire recovery work underway in many parts of the country. Thank you, Ross. Just excuse us for one moment. We'll be back on in a minute.
Sorry. Sorry, Ross. As I said, Vicki will take you through the details of this in a moment, but before we get into the results, I wanted to reflect on some of the comments I made last week in a speech that I gave to the American Chamber of Commerce. The speech highlighted my view that as we enter the 2020s, there has never been a more important time for business to think deeply about the role it plays in society. Contemporary thinking argues that a greater level of empathy and responsibility is required today from companies in understanding their customers' needs and meeting them fairly.
Society is rightly holding us more accountable than ever before, and the expectation for all of our stakeholders today is for businesses to act responsibly, and this is happening at a time of great change, great change in technology, change in the climate, change in our customers' expectations of us. You have heard me talk before about the Fourth Industrial Revolution, where a convergence of new technologies, including 5G, is enabling a rapid evolution of automation and robotics. While this will bring tremendous new opportunities, it also poses challenging questions on issues such as the impact on the workforce and ethics of AI. But it's not just technology that is changing our lives today. More than ever, businesses must consider their impact on the environment and climate. The 2020s have started with devastating bushfires.
Lives have been lost, homes have been burned, wildlife destroyed, and whole ecosystems that have been changed forever. We estimate the impact of the bushfires on Telstra will be in the order of AUD 50 million, including AUD 10 million in assistance packages. Large teams of Telstra staff and technicians continue to work incredibly hard to support recovery and rebuild efforts in affected communities. I'm very proud of their work and have visited Corryong, Jingellic, Sarsfield, Clifton Creek, and Bairnsdale to see firsthand how important this work is. Climate change will be the defining challenge of the 2020s, and the biggest single risk that we face is thinking that it is somebody else's problem to solve.
While it is popular to have an opinion on what others should be doing about climate, including the government, what I am focused on is what we are doing as a company and personally as an individual. The 2020s will also see new attitudes toward how businesses engage with customers. In my speech last week, I made the point that this is more important in a world where the rate and pace of change continues to accelerate and where customers are becoming ever more dependent on the services that we provide. I shared how a deep commitment to our purpose and values drives Telstra's efforts to do the right thing for our customers by delivering the best technology on the best network so our customers can thrive.
However, I know we do not always get it right, and last week I shared the details of a difficult issue we are currently addressing, which is the subject of an ACCC investigation regarding possible unconscionable conduct. In 2017, we became aware of an issue where a small number of our partners had sold mobile devices and plans to customers that they ultimately could not afford and also may not have been appropriate for their needs. This includes sales to Indigenous Australians, some living in remote communities. In some cases, the checks and balances we had in place to ensure this did not happen were not followed by the staff in those stores.
For us, the challenges have arisen while we have been expanding our network to provide connectivity to communities that had not previously had it during a time of dramatic increases in data usage, together with the escalating cost of smartphones. We also did not fully understand the customers in those areas, and we have let both of them and us down, something that we regret. We will obviously respond appropriately to the ACCC. However, what is equally important is where our conduct did not deliver the right outcomes for our customers, even if it did comply with the law.
We have put in place a significant number of measures to address the core issues, including providing support to customers who have found themselves in financial hardship by proactively buying back debt, actively reaching out to customers to ensure that they are on the right plan, additional training, strengthened external credit assessments, and enhanced performance monitoring tools. I have also traveled personally to communities in the Northern Territory and Western Australia to apologize in person and to hear from affected customers. The matter is the subject of a contingent liability note in the accounts released today, to which I draw your attention. Let me now then turn to our half-year results. We finished the half with good momentum and a strong focus on delivering our T22 strategy, which, as you know, had been designed to meet Telstra's challenges of today while putting us in the best possible position for tomorrow.
Indeed, one of the key objectives of our T22 strategy is to radically simplify our products and services to remove customer pain points and minimize exactly the risk highlighted by the matter I have just described. There are many good indicators of our progress on T22. Firstly, our leadership in 5G. I spoke earlier about my pride in our response to the bushfires. We cannot help also to be proud of our position with 5G. We're not just leading Australia; we're a global leader as well. We also set ourselves a very ambitious target to get down to 20 in-market plans from the more than 1,800 we had less than two years ago, and we have done that. In consumer and small business, we now have 2.4 million services on those plans.
That ambition to offer fewer better products and services extends to things like our smart modems, and more than 1.7 million people are now, sorry, more than 1.7 million of those are now in homes around Australia, and to our loyalty program, Telstra Plus, now with more than 1.2 million members. Finally, I want to call out Telstra InfraCo, which is now well established. It was a pleasure to be able to talk through the detail of this and of the next steps and demonstrate how our optionality is beginning to emerge when we were last together at our investor day in November of 2019. In terms of financial highlights, total income for the half decreased 2.8% to AUD 13.4 billion on a reported and guidance basis.
EBITDA increased 12.1% to AUD 4.8 billion on a reported basis. After adjusting for lease accounting on a like-for-like basis, EBITDA decreased 0.1% to AUD 4.5 billion. Underlying EBITDA on a guidance basis, which excludes one-off income and restructuring costs, decreased 6.6% to AUD 3.9 billion. Excluding the end-year NBN headwind, underlying EBITDA grew by approximately AUD 90 million as a result of an ongoing focus on simplification and cost reduction. Net profit after tax decreased 6.4% to AUD 1.2 billion on a reported basis.
The board has resolved to pay a fully franked interim dividend of AUD 0.08 per share in line with the prior year and comprising of an interim ordinary dividend of AUD 0.05 per share and an interim special dividend of AUD 0.03 per share. Drilling down a little further into the operational drivers this half, we saw mobile EBITDA stabilizing, which is a good leading indicator, and also our transacting minimum monthly commitment increasing AUD 2-AUD 3. Postpaid handheld ARPU declined by 7.4%, which included the impact from competition in prior periods.
In fixed, we remained focused on differentiated customer experiences, including through the Telstra Smart Modem. Across fixed and Data and IP, we continue to face headwinds from the migration of customers to the NBN and price competition. Just on the NBN, we were pleased to see NBN recently acknowledge the industry's view that it should not be contracting directly with end customers as a wholesaler. We will review the detail of these changes as they are implemented. Returning to the results, though, we were particularly pleased with our NAS and Global Connectivity results for the half. NAS EBITDA grew almost AUD 200 million, with an EBITDA margin of 15.6%, driven by improved product mix and productivity. Global Connectivity EBITDA grew almost AUD 60 million from improved product mix, productivity, and some one-off benefits, including foreign exchange.
On costs, underlying fixed costs were down AUD 422 million, or 12.1%, bringing our annualized cost reductions achieved since the program began in 2016 financial year to AUD 1.6 billion. Vicki will provide more details on our productivity program. However, I am pleased with the progress that we have been making. Turning now to customer experience, where the investments we have been making and the initiatives under our T22 strategy continue to have a positive impact for our customers.
Episode NPS, which is the measure of individual customer experience with us for particular processes such as their sales experience, assurance, billing, and other, improved six points compared to December 2018 and two points in the half. Strategic NPS, a more general measure of both Telstra's and non-Telstra's customers' attitudes towards us, was unchanged over the 12 months to December 2018 compared to December 2018, but down one point in the half.
Our strategic NPS results included strong performance from mass-market customers, but disappointing enterprise results. However, after restructuring our enterprise business, we have pleasingly seen enterprise results stabilise at the end of 2019. An important consideration for any business is its ability to continue to attract new customers, and we continue to see a good response to our simplified product offerings and focused on improved customer experience. Through our multi-brand strategy, we added 137,000 net retail postpaid mobile services in the half, including 46,000 branded and 91,000 from Belong. We added 135,000 net retail prepaid unique users and 173,000 wholesale MVNO mobile services in the half, bringing total wholesale services to now over 1.3 million. In fixed, we added 27,000 net new retail bundle and data services, including growth of 44,000 from Belong. Belong now has over 600,000 services, with almost 340,000 mobile services and about 300,000 fixed.
One of the features of the half was increased activity in the price-sensitive end of the market, with strong performance in Belong and wholesale compared to Telstra branding. This bifurcation in the market strongly supports our multi-brand strategy. We also added 359,000 new NBN connections, with an estimated cumulative NBN market share, excluding satellite, of 47%. Total NBN connections from Telstra are now almost 3 million. Turning to our T22 strategy, where we are approaching the halfway point on its implementation. T22 is about radically simplifying our operations and products, improving customer experience, and reducing our cost base. T22 is built around four key pillars and two critical enablers: building the networks for the future and digitisation. In pillar one, the focus is to radically simplify our product offerings, eliminate customer pain points, and create all-new digital experiences.
I mentioned earlier the positive customer response to our new in-market consumer and small business plans and our smart modem. We're also seeing strong engagement with our new loyalty program, Telstra Plus, where reward redemption rates doubled in the second quarter compared to the first. I also wanted to call out that since 2018 financial year, we have seen a 32% reduction in the volume of service calls to our call centers from our consumer and small business customers. This improved 10 percentage points in the half, or another 2 million less calls on prior corresponding period. In fact, since 2016, we have almost halved the annualized number of calls into our call center. That's a reduction of more than 19 million calls. Digital channels now account for more than 57% of service transactions, including account management, prepaid product and billing-related inquiries, up almost 4 percentage points in the half.
At the same time, digital sales interactions for consumer and small business customers were up 9 percentage points in the half to 26%. Online gaming continues to be a major opportunity for us, particularly in relation to 5G, and in October 2019, we commenced an exclusive Australian telco partnership with Microsoft for its Xbox All Access gaming package. We also continue to see great opportunities to support small business customers, and in July 2019, we launched Telstra's Mobile Works uite providing data, software, devices, and 24/7 support. We also launched Telstra Business Services as a month-to-month subscription service, providing access to expertise in tech support, cybersecurity, and digital marketing. In the enterprise space, we launched Telstra Connect, a single digital channel for business-to-business customer interactions, including tracking the status of orders, raising and tracking incidents, and checking on network performance.
Also for enterprise customers is Telstra Purple, which has more than 1,500 people working on more than 8,000 projects in professional and technology-managed services, including network, data center, security, cloud, augmented reality, workplace and mobility, data and analytics, and design services. Simplification also remained a focus, with the number of enterprise products cut by 23% as we continue to unravel complexity. In pillar two, Telstra InfraCo is fully operational. InfraCo's revised organizational structure and operating model, which we outlined for you at the November Investment Day, has subsequently been announced internally and is being implemented. In pillar three, the focus is on simplifying our structure and ways of working to better empower people to serve our customers. We have continued to focus on removing hierarchies and SIOs and redesigning our organization from the ground up.
As at the 31st of December, we had announced 6,900 of the net 8,000 direct workforce role reductions, of which 6,300 had left the company, and we have also reduced our indirect headcount by approximately 8,000 in the last 18 months. At the same time, we continue to build new capabilities for the future, creating 1,500 new roles in areas including cybersecurity, software engineering, data analytics, and AI. 1,300 of these have already been appointed. We have also opened our state-of-the-art innovation and capability center in Bangalore, India, where we have over 600 Telstra and partner staff working closely with a range of local and international partners and vendors. Pillar four is focused on delivering an industry-leading cost reduction program and portfolio management.
We have increased the cost reduction program in the first half of FY 2020, and we're on track to reach our target of reducing annual underlying fixed costs by AUD 2.5 billion by FY 2022. We have already delivered AUD 1.6 billion, and we're on track to achieve a reduction in underlying fixed costs this year of AUD 630 million. The other aspect of pillar four is focused on actively managing our portfolio to monetize up to AUD 2 billion of assets. The total value of assets monetized as part of T22 so far is around AUD 1 billion. T22 is built on the foundation provided by our investment in digitizing the business, building the networks for the future, continuing to reinforce our network security, and launching 5G.
The 5G rollout is ongoing, and we have now reached 32 cities nationally and remain on track to reach at least 35 Australian cities and major towns by the end of the financial year. Over 100,000 5G-capable devices are now connected, and around one in four Android mobile handsets that we have sold since July of 2019 have been 5G-compatible. We have six 5G devices available today for our customers, including the mid-priced Samsung Galaxy A90. We will also be launching second-generation 5G handsets over the coming weeks. In addition to leading in 5G, we're also continuing to reinforce our network superiority. Nationally, our mobile footprint is more than 2.5 million km2, at least 1 million km2 more than any other mobile network in Australia. This half, we've built more than 237 new mobile sites and upgraded a further 267.
In the half, we also proved the technology to expand our Narrowband IoT coverage to nearly 4 million km2 from over 3.5 million. Our Cat M1 IoT coverage is around 3 million km2. Service reliability and resilience obviously remain critical factors for our mobile customers and a key network differentiator for Telstra. Since 2016, customer impact hours from outages have reduced by nearly two-thirds as a result of our ongoing network improvements. Telstra also continues to lead the market in key industry network performance benchmarks. Independent third-party recognition for the speed and quality of our network this year included winning Best in Test in the renamed P3, now known as Umlaut, a 2019 mobile network benchmark for Australia. We also ranked number one for the Netflix ISP index over the last 24 months in a row. Turning now to digitization.
Upgrading and digitizing our CRM, provisioning, billing, HR, and many other systems is a key enabler of many improved customer experience initiatives that we are delivering. We are making strong progress on the customer and agent-facing components of this transformation. Firstly, we expect to launch an entirely redesigned customer-facing mobile app by Telstra in the second half of this financial year. This will replace the Telstra 24x7 app. The new app will offer a range of customer benefits, including personalized service and promotional offers, and will deeply integrate our Telstra Plus loyalty program. Secondly, we have progressed replacing multiple agent-facing systems with a single Salesforce console. We have over 5,000 agents using the console today for a range of customer management transaction types. The single system will allow us to more rapidly process customer transactions and provide personalized management of customer needs.
This and other process automation improvement have given us more than 257,000 labor hours back to the business. Before I close, I would like to take you through our T22 scorecard, which we use to track our progress. Importantly, the scorecard represents a very transparent view, and it is reviewed by our auditors. Consistent with our full year 2019 results, six key 2022 milestones are complete, including the key measures to simplify to 20 consumer and small business plans, to be 5G-ready, and to ensure Telstra InfraCo is fully operational. Of the remaining 28 measures, 17 are on track, and we have made good early progress on a further two, which we show in gray as they are too early yet to measure. Nine measures are rated either amber or red, and I want to take a moment to comment on those briefly. Firstly, NPS.
We are on track with enterprise NPS, as I mentioned before. However, given that our strategic NPS results are flat year-on-year, we have identified this NPS metric as amber. Secondly, our new technology stacks. We have very well progressed. For enterprise, the stack is live, with our customers experiencing the benefits. Consumer and small business, as we discussed earlier, we are strongly progressed with customer and agent-facing components, which will largely reach scale in the second half of the financial year through the launch of My Telstra and the continued rollout of Agent Console. In terms of new product build and customer migration, we have launched several beta trials of products on the new stack, which has proved out the benefits of faster activation, easier digital self-service, and a greatly simplified charging experience for customers.
We are now taking the time to expand the breadth of product capability before commencing migration at scale, which we expect to do in the first half of 2021 financial year. While we are confident we will deliver the remaining functionality, given we are behind where we would like to be on the build, we are therefore behind where we'd like to be on the migration of customers to the new stack. Thirdly, the Telstra 24x7 App. The number of active users of the Telstra 24x7 App has been steady at 4 million. While we are targeting an increase in active users, the good news behind the result is that our customers have had fewer reasons to visit the app to check their data consumption due to our simplified mass market plans.
We're expecting to increase customer engagement through the launch of the new My Telstra app that I mentioned before. For average services per customer, the number has been steady but below where we wanted it to be. We're targeting increasing the multi-product holdings of our customers, including through entertainment, mobile assurance, and gaming add-ons. For key industry network surveys, we continue to lead in all of them, with the exception of the ACCC Measuring Broadband Australia report. This report includes lines that are not capable of achieving the NBN tier speeds, and this is due to NBN's constraints, not the RSPs. And it's a matter we've raised with the ACCC because this obviously creates a misleading impression of individual RSP performance in this report. Moving to Agile at level three maturity. We've now introduced Agile at scale to the business and trained over 16,000 employees.
However, in the second quarter, we were slightly below our maturity target, and consequently, we have identified this metric as amber. On costs, although we did not reduce total costs in 2019, we have reduced total costs in this half, and we have therefore marked this measure on track. Lastly, on underlying ROIC. While we expect growth in underlying ROIC from FY 2020 to FY 2022, we have identified this metric as amber because we do not expect growth in underlying ROIC in FY 2020, as we have previously advised. Let me summarize then before I hand over to Vicki. Our results this half were in line with guidance, reflect the significant changes underway across the sector, and show we are starting to build positive underlying financial momentum as a result of our T22 strategy.
As expected and as previously flagged, our results also reflect current market realities, including the impact of NBN and competition. At the same time, we are sitting at an incredibly exciting inflection point, the dawn of the 2020s, the dawn of 5G, and we see significant opportunities in technology, in telecommunications, and for Telstra. Society is rightly holding business more accountable than ever before for our actions, and the expectation from all stakeholders today is for responsible business at a time of great change.
Great change in technology, changing the climate, changing our customers' expectations of us. T22 is about positioning us for this new world as a simpler, more digitally enabled business with the best network, the right economic model, optionality, a strong balance sheet, and the skills, capabilities, culture, and ways of working that we need to be successful. Thank you, and with that, I will hand over to Vicki to take you through the detailed financials before we open for Q&A.
Thanks, Andy. Today's half-year 2020 financial results reflect increased momentum from the execution of our T22 strategy. The benefits from the radical simplification of our business and products and continued focus on reducing our costs are evident. We are especially pleased with the scale of cost reductions we are achieving. This is important given product margins will take time to turn. First half 2020 shows a significant turnaround from the level of decline in recent years. As we have stated previously, the clearest view of the future financial performance of our business is provided by underlying EBITDA, excluding the NBN headwind. In first half 2020, we achieved growth in this figure of around AUD 90 million, the first time we have achieved growth since FY 2016.
We estimate that we are now around 60% of the way through the recurring financial impact of the NBN. We expect FY 2021 to be the peak year of NBN headwind. As we look forward, second half 2020 and into FY 2021, we expect a stronger contribution from improved product margin trajectory, with mobile gross margin turning positive versus PCP around the end of calendar year 2020.
I will discuss the performance of our mobile business in more detail shortly. Let me now turn to details of our financial performance. Our half-year results were in line with guidance and market expectations. The numbers on the left of the slide are our reported statutory results. First half 2020 reported income was AUD 13.4 billion, down 2.8%, and reported NPAT was AUD 1.2 billion, down 6.4% on first half 2019. First half 2020 is the first results following the implementation of AASB 16.
The new accounting standard meant that from 1 July 2019, operational lease costs moved onto the balance sheet and below EBITDA in the P&L. Given different accounting treatment of leases in first half 2020 and first half 2019, and our exit of mobile lease plans, the reported lease adjusted columns provide a like-for-like view of our results and reflect the view we use when managing the business. Reported lease adjusted includes all mobile handset leases as operating expenses and all rent and other leases below EBITDA. AASB 16 also impacts the representation of our cash flow and net debt. Details of AASB 16 impacts and adjustments can be found in the slide appendix. The remainder of this presentation will focus on the reported lease adjusted results.
On a reported lease adjusted basis, EBITDA was broadly flat at AUD 4.5 billion, including around AUD 200 million of restructuring costs and around AUD 800 million of net NBN one-off receipts. Underlying EBITDA, which we provide guidance on, was AUD 3.9 billion, down 6.6% on PCP. The largest reason for this decline was the NBN, where we absorbed around AUD 360 million of in-year recurring headwind in the period. A breakdown of the component parts of this headwind is outlined in an appendix slide. Underlying EBITDA, excluding the NBN headwind, grew around AUD 90 million in first half 2020 versus PCP. Included in underlying EBITDA is reduction of fixed costs of AUD 422 million, or 12% on PCP. I will speak more to the drivers of this strong performance shortly.
Depreciation and amortization on a reported lease adjusted basis increased 2.6%, and we expect FY 2020 D&A to increase modestly on FY 2019 due to mix shift to shorter asset lives. We then expect D&A to decline materially in both FY 2021 and FY 2022, predominantly due to fixed access assets associated with NBN completion and legacy IT assets fully writing down. Net finance costs increased marginally. However, our real cash borrowing costs declined AUD 30 million. Tax declined on lower profit before tax, with first half 2020 reported lease adjusted NPAT down 4.6% on PCP. Looking now at income by product, excluding one-offs, underlying income for the period declined AUD 445 million, or 3.3% year-over-year. There is significant detail in the appendix on each of our products, but I will touch on the highlights. Mobile revenue grew AUD 17 million in first half 2020.
This was due to hardware growth of $202 million on higher unit device costs, offsetting services decline. In postpaid handheld, our lead indicators show positive momentum. Transacting MMC was up $2-$3 across consumer and small business in first half 2020 versus second half of FY 2019, and we expect continued improvement into second half 2020. Transacting MMC represents the average minimum monthly commitment, excluding hardware, of new and existing customers that have taken up new plans in the period. We are pleased with our SIO performance, driven by our multi-brand strategy in a slowing market. As outlined at our FY 2019 results and in line with expectations, reported ARPU decline accelerated to 7.4% in first half 2020, with all segments seeing similar levels of ARPU decline. This was caused by several factors.
Firstly, we saw the impact from a period of intense price competition in FY 2019 washing through our customer base. Secondly, $60 million lower out-of-bundle revenue as we removed excess data fees consistent with our T22 strategy. Thirdly, dilution from a higher mix of Belong customers. And finally, accounting for new plans which allocate more revenue to hardware. Of the 7.4% decline in ARPU, around 40% is due to the first factor. In second half 2020, remaining wash-through of FY 2019 competition is expected to be offset by initial benefits from improvement in transacting MMC. We expect transacting MMC improvement to have a more material impact on ARPU in FY 2021. The other three factors will continue to impact ARPU. In the second half, we expect out-of-bundle declines to be similar to first half 2020. We also expect a growing accounting impact from increased take-up of our new plans.
As a result, ARPU declines are expected to moderate in second half 2020 versus PCP. Turning to other mobile categories, in prepaid handheld, unique users were up 135,000, with less migration to postpaid, and we have stabilised both average voucher volumes and revenue sequentially, a pleasing result. Likewise, mobile broadband is seeing stabilisation in revenue sequentially after several years of decline. Our wholesale MVNO business, a crucial part of our multi-brand strategy, achieved revenue growth of 5%. We remained focused on growth and building value in mobile and have a clear strategy, which includes improving customer experience through T22, growing our network differentiation, executing our multi-brand strategy, and leading in 5G. Now let me turn to our fixed line business, where revenue continued to be impacted by NBN migration, competition, and legacy decline. We continue to add SIOs in a slowing market thanks to our multi-brand strategy.
In first half 2020, Belong contributed 44,000 net adds; however, our branded business was weaker, including due to continued competition at the entry-level end of the market. In our branded fixed business, we are still experiencing ARPU dilution as more customers move to our in-market plans. This is expected to be less negative in the second half of 2020. Our branded fixed business remains focused on differentiated experiences. We extended the number of customers using a Telstra Smart Modem to 62% of consumer customers, delivering greater reliability, and continue to take a leading position on in-home speed and reliability. Turning to Data and IP revenue, revenue was down 12.6%. ISDN decline accelerated in the period to 32.5% as expected due to factors including copper switch-off and customer migrations to equivalent voice products within the NAS portfolio. IP VPN in total declined 7.4%.
Within IP VPN, fibre-based services represent around 90% of the revenue, and revenue declined approximately 1%. Copper-based services are a smaller part of the portfolio, but very significant declines have had a large impact on reported results. We expect IP VPN decline to moderate over the next 12 months- 18 months, but it remains a challenging period. Reported NAS revenue declined modestly due to the expected reduction in NBN commercial works and a focus on driving profitability rather than revenue opportunities. NAS revenue, excluding commercial works, grew 2% and shifted to a higher quality mix of annuity revenue. We expect these trends to continue in second half 2020. Turning to our operating expenses for first half 2020, in line with our T22 commitment, total operating expenses reduced. First half 2020 underlying costs declined 2%.
An increase in NBN payments of AUD 204 million was more than offset by our productivity program, which reduced underlying fixed costs by AUD 422 million in the period. We are delivering very strongly against our AUD 2.5 billion net productivity target, having now achieved a AUD 1.6 billion net reduction in underlying fixed costs since 2016. We split our fixed costs into three broad categories: direct labor, indirect labor and service contracts, and non-labor. Around two-thirds of our net cost reduction in first half 2020 was delivered through direct labor, which included the full impact of FY 2019 reductions and the impact of a further 1,500 direct FTE reductions in second half 2020. The other third of our first half 2020 net cost reduction was from indirect labor and service contracts, which included reductions from customers transitioning onto the NBN network, associated contract optimisation, and reduction of project-related activity.
This has also resulted in a reduction of indirect headcount by approximately 8,000 in the last 18 months. We continue to expect to achieve our full-year underlying fixed cost reduction target of AUD 630 million. Second half 2020 reduction will be smaller than first half 2020 due to timing and a much tougher comparative period. Moving to EBITDA, where our results were in line with expectations, underlying EBITDA declined AUD 274 million, with growth in NAS, recurring NBN, global, and other offset by declines in fixed, Data and IP, and mobile. This includes the around AUD 360 million in-year NBN headwind, which is included in fixed, NBN recurring, and Data and IP. Underlying EBITDA, excluding the in-year NBN headwind, grew around AUD 90 million. Looking at the major products, mobile is now close to 50% of our underlying EBITDA.
Mobile EBITDA is stabilizing, down AUD 47 million in first half 2020 versus PCP after declining AUD 342 million in FY 2019 and AUD 274 million in FY 2018. The decline in first half 2020 is due to lower service revenue, partially offset by productivity and improved hardware margin. Fixed EBITDA, excluding cost to connect, declined by AUD 443 million in first half 2020. This includes an over AUD 290 million revenue reduction and an over AUD 200 million increase in network payments to NBN Co, partially offset by fixed cost reduction.
We continue to focus on items in our control, such as our cost to serve and productivity, and advocate for lower NBN wholesale costs. Turning to Data and IP, accelerated revenue pressure, particularly in high-margin legacy products, especially ISDN, flowed through to EBITDA. Looking forward, EBITDA is expected to continue to be impacted by legacy decline, competition, and resale of NBN at lower margins.
However, we still have opportunities in productivity. This is among our next focus areas in our productivity program. NAS performed very strongly against a weak PCP due to productivity, improved revenue mix, and the migration of PSTN and ISDN services to unified communications. NAS growth offset Data and IP declines at EBITDA in first half 2020. NAS EBITDA margins of 15.7% in first half 2020 are consistent with our mid-teens margin outlook. We expect second half 2020 NAS margins to be broadly consistent with first half 2020. In Global Connectivity, EBITDA improved by AUD 59 million. The underlying business has delivered continued growth of circa AUD 25 million. Reported results were also supported by one-offs and foreign exchange benefits. Other growth includes media software losses dropping out after the sale of Ooyala and a small improvement in health. Turning to free cash flow.
Free cash flow after operating lease on a guidance basis in first half 2020 increased 36% versus the PCP. Improvement was primarily due to our success in lowering CapEx, which was high in the PCP due to timing and our strategic investment program. In second half 2020, we will continue to accelerate 5G deployment, and we confirm our FY 2020 CapEx guidance in the range of AUD 2.9 billion-AUD 3.3 billion. The working capital movement of negative AUD 1.5 billion in first half 2020 was largely due to the exit of mobile lease plans, inventory, and restructuring. Our FY 2020 free cash guidance included an expected approximate AUD 1 billion increase in working capital, which remains unchanged. Therefore, we expect an improvement in the second half.
We recently announced that prior to the end of this financial year, we would move to 20-day payment terms for suppliers who invoice us up to AUD 2 million annually. The one-off cash flow impact of this change to supplier payment terms is estimated to be AUD 50 million and will be incurred in FY 2021. We are on track for our FY 2023 free cash flow after operating lease payments guidance of AUD 3.3 billion-AUD 3.8 billion.
Also, a reminder that second half 2020 will have an outflow of AUD 386 million for 5G spectrum, which is excluded from guidance. Moving to dividends, the board has resolved to pay an interim dividend for first half 2020 of AUD 0.08 per share, fully franked, including an ordinary dividend of AUD 0.05 per share and a special dividend of AUD 0.03 per share. The ordinary dividend represents an 83% payout ratio of underlying earnings.
A full reconciliation of reported to underlying earnings on a lease-adjusted basis is available in the appendix of this presentation. Turning to our capital position, net debt in first half 2020 includes AUD 3.6 billion of lease liabilities. Excluding recognition of operating lease commitments in our debt under AASB 16, net debt declined. We remain within our comfort ranges for all our credit metrics. The debt servicing comfort zone has been recalibrated for FY 2020 under the new AASB 16 reporting framework from 1.3x-1.8x previously to 1.5x-2x . In second half 2020, we also received AUD 698 million from the sale of a 49% stake in a property trust. This is included in financing cash flow. We have achieved around AUD 1 billion in asset monetisation since the announcement of our T22 strategy. Our reported and underlying return on invested capital were 8.5% and 6% respectively.
This includes an around one percentage point impact from the implementation of AASB 16. Turning now to guidance, we reconfirm our FY 2020 guidance. Our FY 2020 guidance ranges, along with the assumptions and conditions upon which we have provided them, are shown on the slide. Our FY 2020 guidance remains for up to AUD 500 million growth in underlying EBITDA, excluding the in-year NBN headwind. Let me summarize. Today's first half 2020 financial results reflect the momentum of T22 execution. Underlying trajectory has improved, and we are navigating NBN and other headwinds.
We anticipate mobile gross margin turning positive versus PCP around the end of this calendar year. We are delivering our T22 strategy, building value, improving productivity while giving customers better experiences and creating opportunities for growth. Finally, I would like to take this opportunity to recognize and thank our dedicated teams right across Telstra. I also look forward to meeting many of our investors over the coming weeks. I will now hand back to Ross to moderate the Q&A.
Thank you, Vicki, and thank you, Andy. We'll now start to move to the question section of our presentation. Can we please limit initial questions to two questions each, just to allow all callers an opportunity? I believe if we move to questions, the first question is from Kane at Goldman Sachs.
From me then. Firstly, just on the mobile ARPU outlook, it's a 7.4% decline in the first half. I think you're saying, Vicki, you're expecting that 3% competition impact to dissipate, the 2% out-of-bundle to continue, and I suppose the 2.4% balance to marginally increase from here. So it sounds like roughly 4%-5% declines in the second half as a base case.
Then could you just comment on what you're expecting in terms of that AUD 15 5G price point into FY 2021 and I suppose what impact that could have? And secondly, just the NAS business, can you comment on whether there are any one-off or timing benefits in this result and whether we should expect the second half margins to trend higher from here as they typically do? Cheers. Thanks.
Thanks, Kane. Thanks, Kane. Andy, I might just pass over to Vicki. I mean, just on the NAS one, maybe just to comment briefly. I mean, I think what I would say is that obviously, as you know, our long-term aspiration has been to get our NAS business to EBITDA margins mid-teens. So really pleased with the great work of the team. There's not really any particular one-off thing that's driving the performance in the first half. It's more just the cumulative effect of the productivity initiatives and a bit of mix in there. We would expect the second half to be at least at the same level of EBITDA margin. So I think that's pretty encouraging. But Vicki, maybe hand over to you on the ARPU questions.
Yeah. Thank you, Kane. And I think the way you're thinking about the ARPU is the right way to think about it. So you're right on the first one, the 40%, we expect that to dissipate in the second half. And then you're right, out-of-bundle the same as first half. We do expect the Belong mix and the increased accounting treatment to both increase in the second half. So the way you're thinking about that is correct. Your second question was around monetization of 5G. We remain committed to monetizing 5G. We've obviously previously announced that our customers signing up to 5G get 5G included through to the end of this financial year, and that still remains our intention. So there's no change on our commitment to monetizing 5G.
Thanks, Vicki. Is it possible to get a sense of how many of your customers are on the small and medium plans that will face that $15 price point?
Kane, Andy, I mean, as you can imagine, the customers who are sort of the early adopters of 5G tend to bia s towards the top two-tier plans. The devices in the first rounds are relatively expensive, so they tend to be sort of those customers are sort of aiming towards the premium plans. I don't actually have the number in front of me in relation to what the percentage mix is, but I would say overwhelmingly they're sort of biased towards the top-end plans.
Cheers.
Okay. Our next question is from Eric at UBS.
Hey, team. First question's a bit of a mouthful, so I apologize. But I just wanted to reconcile some comments you made, Vicki, at the investor day where we sort of said second half underlying EBITDA growth, ex-NBN headwinds, would be stronger in the second half than the first half. I'm just wondering if we're still sticking to that guidance because for that to be the case, under my math, I think second half EBITDA needs to be sort of AUD 3.9 billion versus the AUD 3.9 billion you did in the first half, which then suggests sort of full-year EBITDA of at least AUD 7.8 billion for the full year, which sort of means it's very hard to hit the lower end of your AUD 7.4 billion-AUD 7.9 billion EBITDA guidance range.
So just wondering if we're still sticking to that commentary from the investor day. Then just a second question around, I guess, mobile hardware margins. I guess given we've moved to no lock-in contracts, can you just comment on, I guess, how much loss we're making on handset devices this period versus PCP? Thanks.
Okay. Eric, let me answer your question. So firstly, in terms of our outlook on guidance, we're not changing. Obviously, we're reaffirming our guidance, and that includes our underlying EBITDA. Again, on the underlying EBITDA, excluding the NBN headwind, we've obviously achieved growth of around AUD 90 million in the first half. Again, I see no change in our outlook in terms of our guidance, so there are no changes on those. It is tracking as anticipated in the first half. Just on mobile hardware margin, just to be clear, under our new plans, when a customer elects to pay for their hardware over time, either 24 months or 36 months, we don't make a loss on that hardware. The customer does pay us back the recommended retail price over that period. So I might just clarify and check if I've understood your question correctly on that one.
You have, Vicki, and that makes sense for the current period. I guess you still would have been making mobile hardware losses in the PCP, though. So just wondering how much of the mobile EBITDA improvement in this period we can sort of attribute to better hardware margins.
Well, I guess in the prior period, even when we had lease in place, again, you had lease revenue and lease cost, so you had a matching of that. So at an EBITDA level, that wouldn't have an impact when you look PCP.
Okay. Thanks.
Great. Thank you. Eric, now it is Eric at JPMorgan.
Good morning, guys. Thanks for taking my questions. First one, you mentioned that your online gaming partnership with Microsoft Xbox late last year. What does the economics look like there? Is that simply meant to drive broadband uptake, or are there revenue benefits? And then the second one is you also mentioned that the company needs to focus less on the P&L and more on responsible corporate citizenship. What does that mean for costs in the medium and long term?
Thanks, Eric. On the Xbox partnership, I think it's reflective of a very deep partnership we historically have with Microsoft. So we're really excited by bringing their all-games package exclusively to our t elco customers. I won't go into the commercials because they're confidential, but I mean, essentially, it's a broadband at the moment. It's a broadband opportunity, and effectively, a customer can, in addition to the broadband plan that they have with Telstra, access those games on a subscription basis.
But also, I think what it does is it's a pathway through to us seeing how we can really start to draw out the benefits of 5G, as well as many gamers are keen to be able to do so on 5G as well. And so needless to say, we have a roadmap ahead of us in terms of continuing to build on what I think is a quite an exciting part of the market and one where we can demonstrate the benefits of 5G. I didn't say I was going to focus less on shareholder value. I think my point would be that as we sort of sit here and reflect at the beginning of the 2020s, I think never has there been a more important time for businesses to think deeply about what responsible business means.
In fact, I think that ultimately we won't deliver long-term shareholder value if we don't give real deep thought to some of the other aspects and some of the other stakeholders that I mentioned. In fact, the shareholder community, I think, has been, and investment funds and investment managers have been some of the leading thought provokers in relation to responsible business. And many of the investors that we see that invest in Telstra are very interested to learn about our corporate social responsibility programs, our thoughts around environment and responsible business.
And so I want shareholders to know that there is no less focus of me on delivering shareholder value. It's just recognizing, I think, the changing world in which we live in, what are the things that we need to. I think it's placed to one of the strengths of Telstra. We always have been a company that thinks deeply about our role in society, and no more has that ever been demonstrated in our response in the bushfires.
Great. Thank you.
Okay. Our next question is from Sameer at Bank of America.
Hi there. Just two questions. One, how should we think about the cadence of second half cost outs? Kind of what initiatives do you have that you think will sort of deliver on the various sort of cost line items, and how will this sort of manifest itself? And the second one was just on asset monetization. How should we think about the use of cash? Do you think you're going to use the cash to de-gear, or do you think some of that finds its way back to shareholders? Thank you.
Thanks, Sameer. I might pass to Vicki. Vicki, take this.
Okay. Thank you. Thanks, Sameer, for those questions. So firstly, on cost out in the second half. So we're absolutely committed to delivering our full-year cost out at AUD 630 million. In the first half, we've achieved AUD 422 million. So we expect second half cost out on a PCP basis to be around that AUD 208 million mark. Now, it is lower on a PCP, as I mentioned in my speech. It is a tougher comparative because in second half 2019, we did have more cost out than in the first half of 2019.
So still committed to the AUD 630 million. On a PCP basis, it is lower, but we are pleased with how we're tracking from a cost out point of view. And then your question was related to our monetizations of assets under our T22 strategy. As you've seen, we're about AUD 1 billion of the way through that, and that has been used largely. You can see our overall debt coming down slightly. So that was absolutely a focus just to further strengthen our balance sheet.
Vicki, just on that AUD 200 million in the second half, are we going to see this in labor lines, other cost lines? I'm just trying to figure out what is it that should be moving in the models and what's going on within Telstra that kind of brings the cost down as the service contracts?
Yeah. So there's two big drivers. They're fairly consistent with the first half. We will continue to have benefit on direct labor due to the announcements we have made and the full-year impact of that. And the second piece is absolutely a focus on indirect labor and service contracts. So both of those components continue to play a part. As we look to FY 2021, we are already doing work to make sure we start to shift the non-labour lines. And so we would expect to see that emerge in 2021.
Perfect. Thank you.
Thanks, Sameer. We now move to Entcho from Credit Suisse.
Hi, guys. So my two questions. The first one, out-of-bundle revenue decline looks to be tracking at about AUD 120 million for the full year, which I think is down from your prior guidance of AUD 200 million for FY 2020. Just interested in whether that changes your expectation of the AUD 500 million total decline in out-of-bundle revenues or whether it just means you're pushing a further decline out into FY 2021. And just secondly, on fixed, I think you spoke at the investor day about transacting TMMC in the first quarter having been up about a dollar as a good leading indicator. Are you able to tell us what the trend was for the rest of the half and if that's continued?
Thanks, Andrew, for that. So let me take your first question, the out-of-bundle revenue decline. You are correct. In our full-year outlook, we had expected a AUD 200 million decline. First half, we've tracked to AUD 60 million, and we expect second half to be at a similar level. So you're correct. Our outlook is a AUD 120 million decline. That is purely timing in our estimates at the moment. So it's just a shift in timing, no change to the overall longer-run outlook. And then on fixed TMMC, absolutely, it remains a focus for us. We did speak to it up about a dollar. That was the case for the half. As we're looking to the second half, just with the migration of customers, we probably expect TMMC to broadly hold in the second half.
Okay. Got it. And sorry, just going back to the first question, so does that mean that you obviously have some additional headwinds in FY 2021 that you've got to deal with when it comes well, if you're looking for out-of-bundle growth?
Yes. It will be a shift in timing of that out-of-bundle revenue. That's correct.
Okay. Got it. Thank you.
Thanks, Eric. Our next question comes from Andrew at Macquarie.
Thanks for the question. My first question is just on the use of the Boost and Belong brands, given some of the sub-trends. I just wanted to get a high-level view on how Telstra is seeing the implementation of those brands and the pricing of those brands in market. And I guess my question is, does competition at the bottom end inevitably push up to the main brands, and how do you isolate that? That's the first one. And the second one is we're obviously expecting an announcement on the TPG Vodafone merger fairly soon. So I was just wondering, Andy, if you could give some comments on what you think of the implications for Telstra if TPG Vodafone merger does go ahead later this year. Thanks.
Hey, thanks very much, Andrew. But first, on the first point, as we, I think, sort of talked about before, I mean, our focus on the multi-brand strategy is very much to ensure that we focus the Telstra brand at the premium end of the market and the Belong brand at the price-sensitive end of the market and ensuring that we differentiate between those propositions quite clearly and robustly so that we don't drag the Telstra brand and compete at that end of the market because we clearly put in a lot of extra premium into the Telstra brand, both in the network and also from a service perspective, and then also some of the other customer benefits as well. And so the way that manifests itself, as an example, is Belong doesn't necessarily have the same network coverage as the Telstra main brand does.
Belong is not available in the stores in the same way that the Telstra brand is. It's more of a digital offering. And then, of course, the pricing is different as well, albeit we have also been refocusing well, not so refocusing, but moving pricing on Belong up a bit as well. So that's really the key points around differentiation. And as I said before, I mean, we launched Belong, I think it was five years ago, maybe six now, initially in fixed, and we actually built from scratch. And we launched Belong into mobile a couple of years ago when we felt that there was a possibility that the market at the price-sensitive end could sort of develop further, which has transpired to be a little bit the case. So that was, I think, a good strategic move for us.
In terms of, look, the TPG Vodafone merger, I understand it's being announced today, or rather, the judge's decision is being announced today, albeit I also understand that the reasons are not published for another couple of weeks yet. So, look, I think you can see all sorts of different impacts on the market, either from the merger going ahead or from the merger not going ahead. So, candidly, I think what we're focused on is our continued T22 strategy, continuing to differentiate the Telstra brand, continuing to lead in network, continuing to lead in 5G. And I think that's very much demonstrated in our postpaid handheld performance today and the continued improvement in our mobiles business. I can't control the regulators or those decisions. So that's what we're focusing on. But look, I'm very comfortable with how we're positioned.
Okay. Thank you.
Thanks, Andrew. Now we move to Roger from Jefferies.
Hi. Good morning. My first question is, you mentioned about the fact that FY 2021 will be the peak year of NBN headwind. So do you expect that the net one-off NBN DA to increase as well in FY 2021? The second question is on the mobile growth margin, which is going to turn positive versus PCP. And I'm just wondering how much of this was due to the change in plan for hardware versus your improvement in service margins.
I'd like to get Vicki to comment on the second point. Look, in relation to the first, I mean, obviously, the receipt of the NBN one-off payments follows essentially the rate and pace at which people connect to the NBN. And so, therefore, to the extent that those numbers are different next year, that will flow through, albeit subject to a little bit of timing difference in terms of when the payments are made, etc. On the second point, Vicki, maybe you want to comment?
Yeah, sure. So the question on mobile growth margin turning, there's a number of factors in that. Obviously, the biggest part is ARPU and what's happening there. So there'll be a number of factors in that, but a major driver is really, as I talked to, we're very focused on transacting minimum monthly commitment as our lead indicator. And that translating into ARPU is undoubtedly the biggest driver of turning mobile margin.
Right. So not much from that hardware.
There will be some, but as I said, it's not much. The big driver is absolutely at the ARPU line.
All right. Thank you.
Thanks, Roger. We now move to Ian from New Street.
Yes. Good morning. Look, I've got 3 kind of related questions, particularly around the fixed product retail standalone voice, where revenues are about AUD 135 million lower than previously and 500,000 fewer customers over 12 months. And so another 2 years of migration suggests that you're looking at a rump of 100,000 standalone voices-200,000 standalone voice in 2 years' time. Is that a fair assessment? And secondly, some of those are migrating to bundles, but a lot of them seem to be migrating to mobile when you look at the big increase in retail prepaid and Belong and so on on a somewhat lower ARPU, but presumably a bit of margin and staying on the NBN.
Is that a fair reflection of the pattern between that fixed voice and what's going on in mobile? And thirdly, how did that pattern vary between the first quarter when NBN had their previous price structure and we saw quite a big change in the entry-level bundle wholesale price at the end of September? So was there a change in that pattern in the second quarter?
Thanks, Ian. Let me have a go at trying to respond on those. I think, look, the first point is, yes, there's no doubt that as customers migrate to the NBN, an increasing amount, or rather, a good number of them are choosing not to have a separate standalone voice service. And that's because, basically, there's a voice service bundled into the NBN service for the 92% of the population that are covered with a sort of a fixed service. In the last 8%, however, we continue to have a basic telephony obligation to provide basic telephony services to people in that footprint. And that's where ultimately you will see, once the NBN is fully rolled out, we'll continue to have standalone voice customers.
I think you sort of mentioned a particular number. I'm not sure I could confirm the number, but I mean, that's basically how I would sort of respond to the question. So we will definitely see a good number of standalone voice customers in that 8%. In terms of whether customers are migrating to mobile only rather than voice, I mean, I'm not sure we got the evidence to determine whether that is the case or isn't. I mean, I think there's definitely and historically been a trend that the number of homes in Australia that are connecting to broadband, or rather just going mobile only, has definitely been a trend.
I think last time I looked, it was somewhere in the order of 1%-2% of homes per annum were choosing not to connect to broadband, particularly new homes and multi-dwelling units, apartments. That's definitely been a dynamic. Whether we saw a variance in that pattern through the pricing changes, I don't know. I'm not sure. But one thing I would say on this point is one of the things we have advocated for is a standalone voice product from NBN so that customers who do want a voice-only service are not having to pay a higher price for it, where they don't actually have a data demand. That's something that we feel is still quite important and something we've put into our submissions previously on pricing.
And the bulk of the 1,000 sorry, 1.6 million standalone voice, they're still on copper, or I think there's a couple of hundred thousand on NBN that are voice-only, isn't it?
Yeah. I mean, there would be some, but I mean, not just on copper. I mean, again, in the last 8%, some will be on the CAN radio and some of other satellite technologies and otherwise. And the last 8% is obviously something that we support through the Universal Service Obligation. And something we'll continue to work with government on in terms of thinking about how do we continue to provide the best connectivity for customers with changing as technologies evolve and develop their new opportunities to be able to support those customers.
Great. Thanks, Andy.
Thank you, Ian.
Next, we have Craig from CLSA.
M orning. First question, just on Global Connectivity, very strong margin performance there. Just wondered if that margin you think is sustainable or if there's any sort of particular one-off items that boosted the margins there. And then second question, just on the mobile postpaid subscribers, I noted that for this period, a very high proportion of Belong subscribers compared to your branded, whereas in the past, during FY 2019, it was a more even split. Just wondered if you can kind of think of why that could be a lower proportion this year, and do you expect that to change? Thanks.
Just I'll have a go on a couple of them and check with Vicki that I haven't missed anything. But look, on the Global Connectivity thing, there was a little bit of one-off period to period, and maybe Vicki can provide some color on that. There was also some exchange help, but that shouldn't impact the margin, I don't think. Or maybe it does, does it, because of the difference between okay, so how that flows through. But no, underlying that is there's no reason why it shouldn't be sustainable. There's some good productivity benefits and product and exchanges which the global team have been implementing. So do you want to comment then on the pieces which are perhaps having a one-off impact?
Yeah. So you're right. Global Connectivity had a good half. Its EBITDA growth was AUD 59 million. Of that, about AUD 25 million was growth in the underlying business. And our outlook on that is for it to be consistent. We would expect to see continued growth. It was supported by some one-offs and FX benefits. So the key number to think about there is the underlying growth of AUD 25 million in the half.
And then on your second question, which is really around just the proportion of, sorry, Belong postpaid compared to Telstra branded, I think that there's certainly we saw more activity at the price-sensitive end of the market. And you can see that by that proportion on Belong. But you also see that given the strong performance in prepaid and also in wholesale, whether that continues. To be honest, I'm not sure that I can predict. I think the more important thing is that for us, our multi-brand strategy is working because it enables us to compete in all sectors of the market and compete strongly. But as I say, that definitely was a dynamic in that half of the year. As I say, whether that continues or to the extent to which that continues, hard to judge at this stage, I think.
Okay. Just one follow-up to that point then. Did you see any more cannibalization than you have in the past of your branded, or is it more that you just saw more intense competition at that price end, but you didn't see more branded customers swinging down to Belong?
Craig, it's a good question. We do monitor closely cannibalization. And I can tell you that the cannibalization between Belong and our branded business stayed inside our parameters. So there was no issue with that.
Okay. Thank you.
Thanks, Craig. Our next inquiry comes from Brian at Morningstar.
Andy, at the board level, has there been any discussion on what intermediate goals or mileposts need to be achieved before a dividend increase will be even considered? Or is Telstra happy to keep its powder dry until T22 is done? And also, I wonder if you can help me out here. On the mobile subscriber front, do you have a sense of your market share movement in the half in both the postpaid and prepaid space? Thanks.
Thanks, Brian. Look, on the dividend one, obviously, as you would appreciate, it would be appropriate for me to divulge the detail of board discussions. I mean, other than to say, obviously, the board, firstly, as I am, I'm acutely aware that the dividend is a very important part of the overall shareholder proposition or Telstra investment proposition for our shareholders. So we take any decisions incredibly seriously. We did, as you might recall, update our capital management framework some time ago, a couple of years ago, and reset our dividend policy, which is around 70%-90% of underlying earnings as an ordinary dividend and up to 75% of the one-off payments over time received from the NBN.
We continue to be committed to the dividend policy. That's really essentially what drives the considerations around dividend at a board level. On the market share question, the short answer is I don't know for certain because I know what my numbers are, but I don't necessarily know what the total market is. But my strong suspicion is, yes, we will have gained a bit of market share in the half. But I don't know that for certain until the rest of the industry's numbers come through. But I think other people may be reporting today, so you might want to compare our numbers with this.
Thank you.
Thanks, Brian. Thanks. Next, we have Nick from Morgans.
Thanks for the question. Just one on InfraCo. In the first half, it obviously delivered some low double-digit decline in revenue and EBIT. Just trying to get a feel for the steady state of that business. I guess given you 80%-90% through the NBN and that NBN's created some one-offs in InfraCo, is this sort of current level broadly steady state, or should we expect a couple of years of still some one-offs in that division? Thanks.
So, I mean, one of the revenue streams in InfraCo is essentially our wholesale revenues associated with the proportion of the country that's not connected to the NBN in relation to those customers, broadband customers of our peers in the industry where we're providing that wholesale service. So obviously, as the NBN continues to roll out, that will continue to decline. And I think we have previously disclosed the proportion overall, which I'll point to, I'll ask Vicki maybe to comment on in a second, which sort of is partly associated with rolling off as the NBN rolls out and therefore that which is sort of longer term, more sustainable. So that's, I think, the main answer. But Vicki, do you want to give us the update on the proportion?
Yeah. So Nick, I would have a look at its Page 13. It's one of the appendix to the CFO slides. You'll see there the breakdown of InfraCo as a segment. As Andy just referenced, inside that is our wholesale business. There's obviously a fixed legacy business in there impacted by the NBN rollout. So you can see fixed, excluding NBN cost to connect, you can see an EBITDA in the first half there of AUD 326 million declining at 34%. That's the legacy business there. And obviously, that is linked to the NBN rollout.
That's great. Thank you.
Thank you, Nick. We're cycling through now. So from Kane at Goldman Sachs, please.
Thanks, Brian. Some follow-ups. Just two for me. Just fixed EBITDA margins, given next year's meant to be the peak NBN headwind, how should we be thinking about the fixed EBITDA margin trajectory from here and whether that could go negative? And then secondly, just on your media strategy, obviously seeing the $170 million shareholder loan to Foxtel or the Kayo discounts that press around the AFL rights extension. Just remind us how you think about your media strategy and the rationale for that shareholder loan. Just.
Yeah. No, thanks very much. Just let me try and make a couple of comments on both and see if Vicki wants to add anything. I think on the first point in terms of the trajectory of the fixed mobile margins, I mean, firstly, as you've heard me previously say, given where wholesale prices are, which from NBN, we still continue to be of the view are too high ultimately for going to lead to higher prices for end customers.
We've previously said, I think, as others in the industry have, that essentially fixed reselling fixed retail services, NBN services, basically is pretty much at a zero margin. Bear in mind, I think we're around about just over 70% of homes are now connected to the NBN. And so therefore, within our EBITDA for the half, you've got a mix of services which are still sort of on legacy and still on NBN. So you need to sort of think about that. And then the other point is that it does include the cost to connect.
And so as we're migrating customers over to the NBN, there's a sort of a one-off cost of that migration, which is the cost of disconnecting and reconnecting customers to the NBN. And I think in the half, that was about AUD 200 million. So if you sort of actually added that back in, you would sort of see that the blended EBITDA margin for the half, which is around, as I say, 70% NBN and 30% legacy, was around about 15%. So you could backwards engineer sort of the math and the timing and all the rest of it to sort of work through that. I mean, does that broadly cover it, Vicki?
Yeah. That's absolutely right. And as we've talked about before, our focus in NBN, as I talk to, is absolutely addressing the costs we can, getting our costs to serve down and driving productivity. But as Andy, you've spoken about many times, it is difficult to see that business delivering sustainable margins with the wholesale cost where it stands today. That's right.
You'll also see from our report, our results, and also from our NBN's report the other day, our ARPU was down a little bit. Vicki's already commented on where we're endeavoring to sort of increase transacting MMC. That's more challenging in fixed at the moment than it has been in mobile. We were getting a bit more movement. But also, you'll note that NBN's own ARPU, which is not quite the same as their price, but it's obviously a flow through essentially of the implications of their pricing, was up again another AUD 2 to AUD 45 in the half. So the wholesale prices are in flow through that is only going up. And so we've got to continue to advocate our point of view on that. On the media strategy, so look, we see media as very much a very important part of our overall proposition for our customers.
People love their media and being able to offer some exclusive or sort of unique, let me put it that way, experiences through Telstra to our customers, whether it's through our partnership with Foxtel or whether it's through our ownership of mobile digital rights directly or whether it's through Telstra TV is definitely something that drives a great customer experience and greater usage of our network and also improved NPS. And so that's fantastic. So it works really well. We don't need to necessarily own all of those media assets or every media asset, but being able to provide them on a unique basis is absolutely helpful. And we're able to do that by virtue of the partnership that we have with Foxtel.
I mean, Foxtel, obviously, itself is going through, as indeed many media companies or pretty much all media companies, traditional media companies are going through a significant transition as streaming and digital, obviously, is changing the behavior of customers. So we're obviously supporting Foxtel in that transition. As you probably also followed, there's been quite a lot of work led by the News Corporation team around the restructuring of Foxtel.
We've supported that through the AUD 176 million loan, which is effectively us agreeing to translate some of our HFC payments, which we're due to receive, into a loan for them as well to support that overall restructuring package and support Foxtel through that transition. We're just about to go into a new footy season. We just put out some promotional offers on Kayo. We're pretty excited about how that's going. It's great product, great service, and a number of the other initiatives that the Foxtel team are undertaking.
Cheers, guys.
Thanks, Kane. Now, Eric from UBS again.
Thanks, guys. First follow-up, just on your retail channels such as JB Hi-Fi and The Good Guys. I was just wondering if you can comment if that's becoming a more material portion of your subgrowth. And given Telstra funds at least a portion of their promotional offers, whether that channel's becoming a more material drag on your mobile COGS and ARPU. And then just the second follow-up, can I just get your view on what NBN's recent concessions in enterprise mean for Telstra? I mean, obviously, they haven't lowered their AUD 1.2 billion business revenue aspiration. But I guess, can we assume at the very least the margin outlook in enterprise has improved for you guys? Thanks.
Thanks, Eric. Did you want to take the brief for one and I'll comment on the NBN one?
Yeah. Absolutely. Look, Eric, there's no doubt JB Hi-Fi and Good Guys, which sits under their portfolio, are important in our multi-brand strategy. They obviously address a different segment in the market. And as a result, we do different types of offers in those channels. There have been some shifts in those offers over the last little while, you may have noticed in those channels. It is an important part of our multi-brand strategy and helps support our overall net SIO growth. But in terms of it being a material drag on our cost of goods sold in our mobile business, no, it wouldn't be having a material impact.
Thanks, Vicki. Look, so on the NBN concessions, I mean, I think firstly, I should say, look, we were pleased to see NBN effectively acknowledge that they shouldn't be contracting directly with end customers. It is a bit of a weird thing to be doing if you're meant to be a wholesaler contracting directly with end customers. So I think that's the first point. We felt that that was both against the spirit and also their constitution where they were effectively applying or providing rather an eligible service. They've obviously put out a consultation how they'll continue to engage in the industry. And whilst they've said they're intending to stop that activity, we'd like to see the detail of when they're going to stop and how that's going to look and the results of the consultation. And so we'll continue to ensure we follow that very closely.
In terms of what it means for our outlook, I mean, I'm not sure we've changed anything in that regard. Yeah, I can't comment on their own sort of forecast. But we've obviously got some continuing to develop our offerings in the market. And we do offer certain NBN services to our enterprise customers as well, which we'll continue to do. But I think the main point is that a wholesaler is meant to be a wholesaler, not contracting directly with end customers.
Thanks, guys.
Okay. Thank you, Eric. And now we move back to Andrew at Macquarie.
Thanks a lot. I've just followed up with two questions. Firstly, and they might sort of point to Vicki, but I'm happy for anyone to take them. Where did the AUD 422 million fixed costs out that you achieved in the half land? I was just wondering if you could give some color between the segments as to where that is. It doesn't look like that much was received and fixed, but maybe it was.
And related to that, if you could just follow up with the comment before that you're going to pursue costs out in Data and IP, if you could just give some clarity on sort of what the strategies and opportunities are in that segment to protect margins over the next few years. And the second question, Vicki, is just on the D&A drop you alluded to in FY 2021 and 2022, are you able to give us some kind of quantum for how that might roll through and if it's an even step down across those two years? I assume that's the main driver of ROIC for 2021 and 2022, or especially 2021, given the NBN headwinds peak, or should I think about the operating ends of the business also driving ROIC in 2021? Thanks.
Great. Thank you. Thanks, Andrew, for the question. Just starting with the costs out question, there is an appendix to the CFO slides that gives you a breakdown of operating expenses across the various products. It's on page, I think, 20 in the pack. So it's a good one to have a look at. In terms of my comment in relation to Data and IP, when you look at it, it hasn't benefited in the half from productivity benefits. You can see its cost goes up slightly. And that is related to Data and IPs feeling the effects, as I talked to, of legacy products coming off and transition to NBN.
So actually, the level of activity transitioning customers goes up a little bit through this period. However, we're still confident there are productivity improvements in Data and IP. And our focus area is really some of the backend processes and handoffs within our business and looking at automation and efficiencies in those. So that's where we're certainly focused from a productivity point of view for Data and IP. If I just turn to my comments in relation to D&A, yes, we do expect material step down in FY 2021 and FY 2022 in terms of our overall D&A. It is linked to two broad things. So firstly, you may recall we announced an impairment to IT legacy assets last year. That was a AUD 499 million impairment. And that was actually shortening the life of those legacy assets by two years.
And so that does come into play across 2021 and 2022. And in addition, linked to the NBN rollout, we have fixed access assets that both some transition across to NBN and some retire. And that's the other component. So we expect at this stage, our outlook would be that D&A in 2021 and 2022 in each year would step down somewhere around AUD 200 million-AUD 300 million in each year. And as I said, it is linked to that impairment and linked to the impact of NBN rollout on our fixed access assets.
That's helpful. Thanks a lot.
Thanks, Andrew. Now we have Entcho from Credit Suisse.
Thanks, Ross. Just two quick ones from me. One is just the clarification around the transacting MMC increase in postpaid. Just wondering if for clarity, does that apply to both the Telstra and Belong plans, particularly given that you've had greater subs growth in Belong? And then just secondly, the bushfire impact, Andy, that you mentioned, was there any split between OpEx and CapEx, or was that sitting or within OpEx? And as you're thinking about longer term, is there any need to upgrade the network to deal with disasters like that, or do you feel like you've sufficiently provisioned?
Yeah. No, thanks very much, Entcho. I'll comment on the bushfire, and then I'll ask Vicki to address the transacting MMC question. So firstly, the AUD 50 million is across CapEx and OpEx. I mean, it essentially sort of arises in three areas. Firstly, we estimate that the assistance packages we put in place will cost around AUD 10 million to support the firefighters and customers, and we've been matching giving our staff donations. Secondly, there is some physical damage to our infrastructure, which will obviously arise predominantly as capital expenditure.
And then thirdly, there is increased sort of field activity where we've had to obviously literally on the ground through the period of the bushfires, but also subsequently in checking infrastructure and maintaining and making fixes, etc. So the most obvious damage, obviously, is when it's to a tower or to a telephone exchange, and you can observe it. I was in Jingellic the other day, which is in the southern part of New South Wales, where we completely lost a tower, which we have now completely rebuilt and re-erected.
But also, obviously, all of our towers, or 95% of our towers, are actually connected by fiber. And of course, we have an incredible amount of fiber and copper, which is distributed throughout the country and throughout communities. And not all of it you can necessarily see because a lot of it is obviously underground in pits, or some of it is aerial. So we are going through and checking all of that because the damage, or not, as the case may be, is slightly less obvious. And so undoubtedly, we will experience the impacts of some of that damage over time as it becomes more apparent or something sort of triggers it. So that's something that we're on top of as well.
But in terms of the overall impact, I mean, I think our network actually responded and performed remarkably well because to the extent any of our connectivity was impacted, it was overwhelmingly. And by overwhelmingly, I mean more than 90% was a consequence of the loss of power, which occurred. And of course, what happens when power goes out, we switch over to battery backup predominantly in relation to mobile towers. And then if the batteries deplete, then we need to get out there to replenish them and/or to put in temporary infrastructure such as generators, which we do. During this period, actually getting into those sites after the fires had come through was hampered by virtue of many of those areas being declared red zones by the emergency services and then not being able to get into them for health and safety and concerns.
But having said all of that, I mean, in terms of ultimately, I think we've lost 5 sites basically out of hundreds that could have been impacted. We still have 20-odd that we've continued to support with temporary sort of generator type of power. And we've already rebuilt 1 of those sites. And so we're in a, I think, in a very good shape. I mean, there'll always be obviously things that we'll take away. We'll do a major incident review, but the things we'll always take away, which particularly as technology changes over time as well, gives us more options to continue to improve resiliency and redundancy. But no, I think we performed remarkably well through that period. But Vicki, transacting MMC.
Yeah. Thanks, Andy. So Entcho, yes, the transacting MMC I referenced was our consumer and small business branded TMMC for postpaid mobile, and that's up AUD 2-AUD 3. Our Belong business actually is also focused on mix. So I didn't reference it, but they are also focused on mix and seeing improvements also in their TMMC. But the AUD 2-AUD 3 relates to branded consumer and small business.
Great. Thank you.
Thank you, Entcho. And now coming to our last question, which is Eric from JPMorgan.
Thanks. Just two more from me as well. Can you just comment on the strength of the prepaid net adds in the period versus the weakness in mobile broadband? Is one migrating to the other? And then secondly, Optus launched their long-anticipated fixed wireless broadband product last November. Now that you've seen their offering, do you have any plans to roll out your own?
Thanks, Eric. I don't know that we see there's a connection between mobile broadband and prepaid. We've done a lot of work in terms of our prepaid offering and improved that offering following a period of decline. So I think it was more to do with the initiatives that we've taken around marketing and pricing, etc. On fixed wireless, as we've sort of commented on before, I mean, I think fixed wireless definitely is an appropriate solution for some customers, but not for all customers. We have our own trials on fixed wireless. And obviously, the technology continues to evolve and continues to develop. I don't have any further things I'm in a position to sort of disclose on that today. But also, of course, we continue to have a large number of customers that take advantage of mobile through mobile hotspots. And we have a good number of those.
In fact, one of our 5G offerings, which is the HTC 5G Hub, basically provides that facility for customers who choose to have a mobile-only service. But I mean, our focus on 5G is really about bringing 5G to mainstream mobile plans and mainstream customers in Australia. We're able to do that because we have a contiguous network. In other words, our network isn't scattered such that customers don't get a contiguous coverage. That's really, really important. And that's why we were unable to do that. And in fact, some of our 5G experience has been pretty exciting. Average download speeds have increased from 200 Mbps when we launched to 340 Mbps now. We also recorded about 1.9 Gbps in the live network as well, which is the peak speed which we've done. We've seen data usage on 5Gs increase more than 28 times in the first half.
We've now got, as I say, 5 handheld devices available, 1 mobile broadband device available, which I've already mentioned. We expect that device range to more than double in the second half of the financial year. We're seeing a 4-point higher net ports versus 4G, which shows that our leadership in 5G is really helping our momentum in the market.
Andy, I might just jump in. Eric, I'm not sure if your question regarding prepaid and mobile broadband, but just to reinforce our prepaid handheld growth of 135,000 unique users, that is completely independent to we have noted on the product performance for mobile slide in the appendix. You may have picked up in footnote 4 that we did make an opening adjustment to our mobile broadband base of 365,000. That was simply $0 mobile broadband silos that we took out of the base. It is completely unrelated. The prepaid UU growth comes from sales to prepaid customers, handheld customers. So those two are not related if that was your question.
Great. Thank you.
Thank you very much, Eric. That concludes our investor Q&A session. I'm now going to hand over to my colleague, Nicole McKechnie, and after a short break, we'll moderate questions for the media. Thank you.