© 2024 Alpha Spread Limited. All Rights Reserved.
www.alphaspread.com is not operated by a broker, a dealer, or a registered investment adviser. Under no circumstances does any information posted on www.alphaspread.com represent a recommendation to buy or sell a security. The information on this site, and in its related newsletters, is not intended to be, nor does it constitute, investment advice or recommendations. The information on this site is in no way guaranteed for completeness, accuracy or in any other way. In no event shall Alpha Spread Limited be liable to any member, guest or third party for any damages of any kind arising out of the use of any content or other material published or available on www.alphaspread.com, or relating to the use of, or inability to use, www.alphaspread.com or any content, including, without limitation, any investment losses, lost profits, lost opportunity, special, incidental, indirect, consequential or punitive damages. Past performance is a poor indicator of future performance.
You don't have any saved screeners yet
Intrinsic Value is all-important and is the only logical way to evaluate the relative attractiveness of investments and businesses.
Intrinsic Value is all-important and is the only logical way to evaluate the relative attractiveness of investments and businesses.
Ladies and gentlemen, thank you for standing by. And welcome to Charter’s Fourth Quarter 2019 Investor Call. At this time, all participants are in a listen-only mode. After the speakers’ presentation, there will be a question-and-answer session. [Operator Instructions]. Please be advised that today’s call is being recorded. [Operator Instructions].
I'd now like to turn the conference over to Stefan Anninger.
Good morning and welcome to Charter's fourth quarter 2019 investor call. The presentation that accompanies this call can be found on our website, ir.charter.com, under the Financial Information section.
Before we proceed, I would like to remind you that there are a number of risk factors and other cautionary statements contained in our SEC filings, including our most recent 10-K filed this morning. We will not review those risk factors and other cautionary statements on this call. However, we encourage you to read them carefully. Various remarks that we make on this call concerning expectations, predictions, plans and prospects constitute forward-looking statements.
These forward-looking statements are subject to risks and uncertainties that may cause actual results to differ from historical or anticipated results. Any forward-looking statements reflect management's current view only and Charter undertakes no obligation to revise or update such statements or to make additional forward-looking statements in the future.
During the course of today's call, we will be referring to non-GAAP measures as defined and reconciled in our earnings materials. These non-GAAP measures, as defined by Charter, may not be comparable to measures with similar titles used by other companies. Please also note that all growth rates noted on this call and in the presentation are calculated on a year-over-year basis unless otherwise specified.
Joining me on today's call are Tom Rutledge, our Chairman and CEO; and Chris Winfrey, our CFO.
With that, let's turn the call over to Tom.
Thank you, Stefan. Our operating strategy continues to deliver strong results and we remain focused on driving customer growth by delivering superior services and value to our customers, improving the quality of our operations by reducing unnecessary service transactions and truck rolls per customer, delivering sustainable free cash flow growth by driving EBITDA growth while reducing capital intensity; and finally, positioning our company for long-term customer relationship growth with current and new products by continuing to evolve a fully converged network that delivers high speed, low latency, seamless connectivity services inside or outside customers’ homes and sanitary environment around the move.
In 2019 we created over 1.1 million new customer relationships, substantially more than 2018. And we added over 1.4 million new Internet customers, also more than 2018. We grew our full year cable adjusted EBITDA by 6.6% in a non-political advertising year. And combined with our lower cable capital expenditures, our 2019 cable free cash flow grew by over 100% year-over-year.
We expect that our cable EBITDA growth combined with our declining capital intensity, and disciplined capital deployment strategy will drive continued strong free cash flow growth. Our mobile business is a strategic extension of our core connectivity capabilities today and in the future and it continues to grow. We now have well over 1 million mobile lines in service with over 900,000 of those added in 2019 at an accelerating pace.
In 2019, we saw a substantial reduction in service transactions per customer relationships. And those improvements were reflected in lower per relationship [repair] calls, trouble call, truck rolls, and billing calls. We also performed fewer physical installation truck rolls giving growing levels self-installation activity.
Looking forward to 2020, we're well positioned to continue to reduce service activity per customer, given a higher level of customers in Spectrum pricing and packaging, the completion of our in-sourcing program and increasingly experienced in-source call center and field operation workforces, the overall improving quality, reliability and maintenance of our network, greater levels of online service activity, self-service activity and growing levels of self-installation activity as I mentioned, which in the fourth quarter exceeded 50% of sales.
So in 2020, these lower levels of overall service activity per customer relationship should continue improvements in customer satisfaction, lower churn and reduction in cost to service customer relationship. Our in-home connectivity product set also continues to improve with 85% of our residential Internet customers now receiving minimum Internet speed of 100 megabits or more and nearly half receiving minimum speed of 200 megabits or more.
We continue to outperform in megabits, our ultra product and our gigabit speed tier across our entire footprint. We made our network capable of providing gigabit service everywhere we operate using DOCSIS 3.1 technology over the course of 13 months for $450 million planned capital. Through our 10G plan, we also have a cost efficient pathway to offer multi-gigabit speeds and lower latency to consumers and businesses that continue to attach more devices to our network, use more and more data on a daily basis, demanding greater network responsiveness and reliability.
During the quarter, we also continued to deploy our advanced home Wi-Fi capabilities to new markets beyond Austin, Texas, including Dallas, San Antonio, and closes of Southern California. Our advanced Wi-Fi capability provides enhanced security, privacy and app-based control of all IP devices in our consumers’ homes, while simultaneously delivering a superior customer experience to better in-home Wi-Fi coverage, and managed Wi-Fi solutions through dynamic band switching and channel optimization within the bands. So far, our deployment has gone well and we plan to rollout this capability throughout our entire footprint.
Our Spectrum Mobile business continues to grow quickly, and we added over 280,000 lines in the fourth quarter more than we added in the third quarter. While it's still early, we believe that our results so far indicate our mobile product drives core connectivity, customer satisfaction and will generate standalone profitability at scale.
We also continue to test tools and technology using CBRS spectrum by small cells mounted on our own high capacity two-way network. Our tests continue to go very well. We continue to add features and functionality to our Spectrum Mobile product. And in the first quarter, we plan to offer 5G mobile service.
Now, I'll turn the call over to Chris.
Thanks Tom. Before covering our results, a quick reminder that we closed the sale of Navisite managed cloud services business within the Spectrum Enterprise in September. We have not prepared pro forma financials of recorded results to include Navisite in fourth quarter of 2018, but not in fourth quarter 2019. For the next few quarters, I will discuss Enterprise revenue growth including and excluding Navisite for comparability. On an annual basis Navisite generated roughly $150 million in revenue and its impact on our EBITDA and CapEx was not material.
Turning to our results on Slide 5. Over the last 12 months, we grew total residential and SMB customer relationships by over 1.1 million, or 4% and by 268,000 in the fourth quarter. Including residential and SMB, we grew our Internet customers by 339,000 in the quarter and by 1.4 million or 5.6% over the last 12 months. Video declined by 101,000 in the quarter, wireline voice declined by 128,000 and we added 288,000 higher ARPU mobile lines. At the end of the fourth quarter 85% of our residential customers were in Spectrum pricing and packaging. And residential customer relationship growth accelerated to 3.8% year-over-year.
As we look at 2020 our goal is to accelerate our full year customer growth rate as we deliver highly competitive products with better service driving connects and reducing churn.
In residential Internet, we added a total of 313,000 customers in the quarter, resulting in residential Internet customer growth of 5.4% year-over-year, driven by continuing churn improvements.
In residential video, we lost 105,000 customers in the quarter, it’s primarily driven by lower video gross additions year-over-year. And in wireline voice, we lost 152,000 residential customers in the quarter driven by lower sell-in following our transition to selling mobile in the bundle, and continued fixed and mobile substitution in the market generally.
Turning to mobile, we added 288,000 total mobile lines in the quarter driven by the value of our mobile product offers, growing brand and product awareness and increased sales effectiveness. As of December 31st, we had nearly 1.1 million lines with a healthy mix of both Unlimited and By the Gig lines.
Spectrum Mobile continues to scale with less EBITDA loss per line even at an accelerating net addition rate. And that does not include any benefits to our traditional cable connectivity business.
Over the last year, we grew total residential customers by just over 1 million or 3.8%. Residential revenue per residential customer relationship grew by 1.8% year-over-year given a lower rate of SPP migration and promotional campaign roll-off and rate adjustments.
Those ARPU benefits were partly offset by a higher mix of non-video customers, a higher mix of streaming and lighter video packages within our video base and $29 million of lower pay-per-view revenue year-over-year in the quarter.
Our cable ARPU does not reflect any mobile revenue, but Q4 especially benefited from the timing of the rate adjustments this year.
The Slide 6 shows our cable customer growth combined with that elevated Q4 ARPU growth resulted in year-over-year residential revenue growth of 5.7%.
Turning to commercial, SMB revenue grew by 6.3% faster than last quarter as the revenue effects from the repricing of our SMB products and Legacy TWC and Bright House continues as well. SMB customer relationships grew by 6.8% year-over-year.
Enterprise revenue declined by 4.5% year-over-year driven by the sales of Navisite and the one-time cell tower backhaul fees that we mentioned as a benefit in fourth quarter of 2018. Excluding Navisite from the fourth quarter of 2018, Enterprise revenue grew by 1.3% in the fourth quarter of 2019. And excluding both cell tower backhaul and Navisite, Enterprise grew by 8.5%. I expect that retail portion of Enterprise to continue to grow nicely as the wholesale piece including cell tower backhaul will continue to be challenging.
Fourth quarter advertising revenue declined by 23% due to less political revenue in 2019. Our non-political advertising grew -- revenue grew by 4.6% year-over-year primarily due to our advanced advertising capabilities, and recently deployed products that efficiently sell highly viewed long tailed inventory, using our own anonymized much more detailed viewing data.
As we look to the full year 2020, we expect to continue that growth driven by our advanced advertising business and a healthy year of political revenue.
Other revenues declined by 6.6% year-over-year driven by lower processing fees and lower home shopping revenues related to video subscriber declines partly offset by higher levels of videos CPE sold to customers. Mobile revenue totaled $236 million with $138 million of that being device revenue. In total, consolidated fourth quarter revenue was up 4.7% year-over-year. Cable revenue growth excluding mobile was 3.4% or 5.2%, when excluding advertising in both years and Navisite in 2018.
Moving to operating expenses on Slide 7. In the fourth quarter total operating expenses grew by $165 million or 2.3% year-over-year. Cable operating expenses excluding mobile were essentially flat or up 0.6% when excluding Navisite. And that's despite strong relationship and revenue growth. Programming increased to 0.6% year-over-year, due to higher rates that was offset by video customer decline of 2.8% year-over-year, a higher mix of streaming and lighter video packages such as Choice and Stream and lower pay-per-view expenses year-over-year tied to the $29 million of lower pay-per-view revenue I mentioned.
Looking at full year 2020, we expect programming costs per video customer to grow in the mid-single-digit percentage range. Regulatory, connectivity and produced content grew by 4.3% and that was driven by higher costs of video CPE sold to customers and original programming costs. Cost to service customers declined by 2.3% year-over-year, compared to 4% customer relationship growth.
So we're meaningfully lowering our per relationship service costs through a number operating, quality and efficiency improvements, which is core to our strategy. Key metrics like calls per customer, truck rolls per customer and churn, self-install rates all continue to move in the right direction as Tom mentioned.
Looking ahead, we expect further decline in cost to service customers on a per customer relationship basis but this quarter’s level of operational productivity was exceptional and it will be replicated every single quarter. Cable marketing expenses increased by 2.1% year-over-year, and other cable expenses were down 1.4% driven by lower ad sales costs, which reverses in a political year, and lower costs related to the sale of Navisite, which will carry through the third quarter of this year.
Mobile expenses totaled $372 million and they were comprised of mobile device cost tied to device revenue, customer acquisition and usage costs and operating costs to stand up and operate the business. Our 2020 mobile EBITDA probably looks similar to 2019 due to growth and the scale of costs. But looking at it further, our current expectation is that in 2021, our mobile service revenue will exceed all regular operating costs, excluding acquisitions and growth related mobile costs.
Turning to EBITDA, we grew total adjusted EBITDA by 8.8% in the quarter when including the mobile EBITDA startup loss of $136 million. Cable adjusted EBITDA grew by 8.9% in the fourth quarter, including a roughly 3% negative growth rate impact from advertising revenue net of its associated expense in both periods.
Turning to net income on Slide 8, we generated $714 million in net income attributable to Charter shareholders in the fourth quarter versus $296 million last year. The year-over-year increase was primarily driven by higher adjusted EBITDA and the loss on financial instruments in the prior year period partly offset by higher income tax expense.
Turning to Slide 9. Capital expenditures totaled $2.3 billion in fourth quarter with the cable CapEx declining about $200 million year-over-year, driven by the same CPE trends, DOCSIS 3.1 benefits and lower in-sourcing and integration costs I’ve mentioned throughout the year. We spent $151 million in mobile related CapEx this quarter, which is mostly accounted for in-support capital, it was driven by software and development costs and retail footprint upgrades for mobile.
In 2020, we expect our mobile capital expenditures to be similar to 2019 and then decline meaningfully in 2021 as that work will be behind us. And mobile CapEx outlook excludes any potential mid-band Spectrum acquisition build out, which would be based on a separate ROI evaluation.
In 2020, we expect our cable CapEx intensity to continue to decline from the 15% in 2019. And on an absolute dollar basis we don't expect our cable capital expenditures to be meaningfully different from 2019 levels. Similar to what I just said about our mobile capital expenditures, if we find new high ROI projects during the course of the year with that accelerated spend on existing projects to drive faster growth, we’d still do so.
The Slide 10 shows, we generated $1.6 billion consolidated free cash flow in this quarter and excluding our investment in mobile, we generated $1.9 billion of cable free cash flow, up $700 million versus last year's fourth quarter.
For the full year 2019, we generated $5.8 billion of cable free cash flow, up $3 billion versus 2018, despite a cable working capital headwind of $800 million this past year.
For the full year 2020, we would expect cable working capital to improve significantly with a neutral to slightly negative impact on our cash flow. We still have typical seasonal swings including the first quarter, in which our working capital is almost always a use of cash. With respect to mobile working capital, we continue to add mobile customers at an elevated pace, which will continue to drive handset related working capital needs in 2020.
In any event our free cash flow profile improved significantly last year. We’re positioned to continue to couple our free cash flow growth with our return focused investment and capital strategy.
We finished the year with $78.4 billion in debt principal and $74.9 billion in net debt. Our current run rate annualized cash interest is $4 billion. And as of the end of the fourth quarter, our net debt to last 12 months adjusted EBITDA was 4.45 times at the high end of our 4 and 4.5 times leverage range. And while calculating our leverage, we include the upfront investment in mobile to be more conservative than looking at cable-only leverage, which was now 4.31 times at the end of the fourth quarter.
During the quarter, we repurchased 5.6 million Charter shares and Charter Holdings common units totaling about $2.6 billion at an average price of $459 per share. For the full year 2019, we repurchased over $7.6 billion of our equity. And since September 2016, we've repurchased over $27 billion, or a bit more than 25% of Charter's equity at an average price of $346 per share.
Turning to taxes, on Slide 2, our tax assets are primarily composed of our NOL and our tax receivables arrangement with Advance/Newhouse, we now don't anticipate becoming a meaningful federal cash taxpayer until 2022 with some modest Federal cash tax payings beginning in late 2021, as we expect the bulk of our existing NOL to be utilized by the end of that year.
For the years 2022 through 2024 we expect our Federal and state cash taxes to approximate our consolidated EBITDA, [less] [ph] our capital expenditures, and less our cash interest expense multiplied by 21% to 23%. That estimate would include part of our tax distributions to Advance/Newhouse captured separately in cash flows from financing in the financial statements. There are multiple factors that impact what I just described, and we're always looking for ways to improve our cash tax profile but it's a good baseline for today.
So we're pleased with our results and we believe in our operating strategy, our network capabilities and the balance sheet strategy which all work in concert to create value over a long period of time. Charter has infrastructure assets with strong growth characteristics and cash flow yield. We have a lot of ancillary products to use for, and so on top of core connectivity services and combined with good value and service will drive cash flow growth with tax advantaged levered equity returns.
Operator, we're now ready for Q&A.
[Operator Instructions]. And our first question comes from the line of John Hodulik from UBS. Go ahead please. Your line is open.
Chris, you gave a lot of good detail on some of the cost metrics, as we look out into 2020. And the 200 basis points you guys saw in cable margin improvement this quarter, especially with the advertising headwind I think was really the highlight. But putting it all together in terms of the programming and cost to serve continuing to come down, how should we think of cable margin trends heading into 2020?
Sure. On one hand, you had a headwind, as you mentioned, political advertising not being in the fourth quarter of 2019. On the other hand, we had some fairly exceptional pieces that were taking place as well as the timing of our rate increases earlier inside of Q4 2019 than it has been in prior years. So I think that results in higher ARPU for customer relationship growth in what would be a sustainable either prior to that quarter or inside of 2020. So you need to take that into account. I think that growth rate given the subscriber mix could look a lot more like what we saw for the full year of 2019 and that with what we saw in Q4.
The second thing I'd mention is that programming has done very well on a gross and per customer relationship basis. I think it'll still do well relative to other years. But at least like a mid-single-digit per customer relationship growth in programming cost in 2020. And then finally, as I mentioned, the cost to serve was exceptional in the fourth quarter. Our cost to serve for customer relationship has been declining. It's going to continue to decline is our expectation. But to have an actual gross decline, significant one inside the quarter was a big step down. And I just would encourage people to replicate that every single quarter.
So then working against that, again, I am not giving guidance but ways to help you to build a model, particularly later in the year, we expect to see a pretty good year for political advertising. So all-in, I think it's going to move around and it's not something that we actively manage inside the business. What we’re managing is can we grow our customer relationships faster? And that's our goal, which will then drive better revenue growth, continues to be our goal to grow faster.
And it has been our experience and continues to be our goal to grow EBITDA at a faster rate than our revenue growth rate with or without a political advertising year. So that's I think the right way to think about that and not to get strategically hung-up in a particular quarter of year-over-year comparison on a margin rate to what’s happening with the underlying trend.
Our next question comes from the line of Philip Cusick with JPMorgan. Go ahead please. Your line is open.
Following up there, Chris, you talked about lower video gross ads year-over-year. Is that a result of fewer promotions and changed incentives? And you said the goal is to accelerate customer growth rate. How do you expect that to go from, in terms of gross ads and churn going forward? And then quickly a follow-up if I can on the in-sourcing program, you talked about it in your commentary. Where are we on the completion of that, and how much in terms of costs are still being duplicated as you in-source labor and everything else?
Sorry, Phil, what was your second question? So first one is lower video gross ads, third one is status of in-sourcing program, and any double -- or what was the second one?
That was the gist of it?
Alright, I thought there was a second. But anyway, lower video gross ads up -- I think it's more function of the marketplace more than anything else. We still believe in video as an our attractive piece to user connectivity package that we offer. It’s an important attribute and it’s investment in growth of the traditional setup box as well as all of our IP platforms. And I don't know if Tom wants to add any more on that.
Well, yes. The lower gross ads I think are a function of the marketplace. It's not material to what drives our economic model, but it is a nice small addition to our broadband connectivity business.
And then on the in-sourcing program, the in-sourcing is essentially complete. We have well over 90% of our service calls are handled in-house and onshore. And then in the field service side, you need to have some contract labor available for PC guys that were well over 75% of our labor in in-source there. And that's been the case for both of those for some time now. There's very little in the way to talk about costs anymore that's in the system.
I do think that it's going to continue to get better for all the reasons that Tom mentioned. The tenure of our employees gets longer, which means they get more experienced and better and do higher quality transactions with our own employees who have a career path at Charter. The amount of self-installation is going up. The number of calls and truck rolls are going down. The churn is going down. And the amount of digital transactions as Tom talked about is also going up which means less labor intensive service infrastructure. And I think that the trend is going to be with us for a long period of time.
Our next question comes from the line of Brett Feldman with Goldman Sachs.
Tom, you just mentioned how video still can be very complementary to your core broadband product. There are a number of new direct-to-consumer video products coming soon, whether it's HBO Max or Peacock and those providers have all said they're looking for distribution, including through MVPDs. I was hoping you could give us your updated thoughts around being a partner for some or all of them. And maybe some of the factors that you have to think through as you decide whether it makes sense, whether it's the economics, if there is technical issues, or just some strategic considerations that are weighing on whether or not you're intending to do this?
Alright. Well, Brett, yes, there's an opportunity enough marketing direct-to-consumer products in our relationship with programmers and those relationships in many cases are also operating under the old model too, which is the bundled cable model. And we can hold both thoughts in our head at the same time. And so bundled products which I think will be selling for years to come, but also selling direct-to-consumer products and because of our customer relationship, because of the way we can package those products into our overall product mix, and the user interfaces that we developed with products like Flex are designed to help enhance.
We have an opportunity to create and help programmers sell their content. And do that in a way that's mutually beneficial to both of us. And so we're working through those models with the various companies. We have already placed direct-to-consumer products like Netflix on our user interface and customers can purchase products directly from our user interface, other card product services, which we've been doing for a long time by the way.
In many ways I look at these products like I look at pay TV. There are opportunities to enhance the video experience and part of the customer relationship. So we have ongoing discussions with all of the entities out there. And fundamentally, I think while there's a lot of dislocation going on in the video business, there's an opportunity in there for us.
Our next question comes from the line of Ben Swinburne with Morgan Stanley. Go ahead please. Your line is open.
Two questions for either of you or both of you. Tom, any change in how you're thinking about pricing the video business in particular? I don't know if you would characterize your rate adjustments in the fourth quarter as a change from prior philosophy or not. But at least it seemed to us like a more aggressive increase in your video pricing than you've done in the past. I was wondering if that reflects a change in your mind about how you manage the business?
And then Chris mentioned in his prepared remarks sort of mid-band Spectrum and any acquisition build out there. I was just curious where your guide collective head was at on that opportunity, if that’s something that's becoming more likely, less likely or just how we should be thinking about that as we go through this year and think about some options coming down the road?
Ben, I think fundamentally, we didn't or haven’t changed anything with regard to our pricing strategy, which is price isn’t the major component of how we drive our revenue growth, it's subscriber growth and we have accelerating subscriber growth. And we expect to continue to have accelerating subscriber growth, because we have the price strategy, we do with the product strategy, we do. And the bulk of our revenue growth comes from that.
So, we have been passing-through things like retransmission consent fees and other things in the video business and looking at the margins in the video business, and the competitive environment in the video business and how we're priced, but our fundamental strategy is not different.
And wireless?
Mid-band spectrum? We're interested in mid-band spectrum, there's an auction coming up for CBRS. We're likely to participate in that auction. I think there’s an opportunity for us. The SEC has been helpful in positioning that spectrum in a way that's an opportunity for us. And so we're carefully considering our auction there.
We'll be Ben disciplined as you'd expect about ROI for us to -- in spectrum that’s required; costs, what are the costs to build and what’s the financial return for doing so, and where would you do it. So, I think people should expect that we will be disciplined around that first what we think is pretty attractive and then clearly the more mobile lines that we have, the more attractive that ROI is.
Our next question comes from the line of Craig Moffett with MoffettNathanson. Go ahead please. Your line is open.
Two questions if I could. First on the wireless business, if I could just continue down that line of thinking. How much -- just given what you know today, how much traffic do you think you will eventually be able to offload with sort of thinking about as a percentage of total maybe, that you'd be able to offload under your own facilities if you think about where it might make sense to build where you have aerial infrastructure to be able to build? And is there any scenario where you would ever want to go to a full facilities based strategy where you would own your entire network?
And then just back to the videos question for a second. How do you think differently about programming renewals just given the change in tone around your video strategy?
So look on the wireless situation it's really a math question of what's a cost to pay for your MVNO rate and what's the cost to build, where's the traffic and what's -- how does all that work and that's the discipline that Chris was talking about. So, as we think about it, there's -- it depends on what you're paying for mobile traffic and what the economic traffic zone is in inside of a particular area and how you would switch that traffic. So I don't -- there's no immediate plan to change our fundamental relationship with our carriers. And -- but over time as the market evolves and speeds go up, and capacity goes up, the economics may change and we'll take advantage of those through time as the marketplace unfolds.
With regard to programming, obviously, the biggest issue in the bundled package is price and a lot of people have been priced out of the market and we continue to negotiate contracts. And as penetrations in the overall distribution change, the relative value of the content changes and it changes the relationship and changes how much programming is worth to you as an operator. And so I don't expect the general circumstance of distribution could precipitously change over the next couple of years. I think we'll still have a big bundled product, but the relative pieces of that are changing in value and we will act appropriately in the marketplace.
Craig, you had a sub question on the wireless side that asked if there is any scenario where we want to evolve in the networking. And so far we haven't seen anything that really demonstrates that need. We have a very good partner Verizon. It's going to very well. And they have a very strong microCell powered network. And so I don't think that there's any economic case that we’re seeing today that says we need to -- now we need partners within the interim partners for the years to come.
Our next question comes from the line of Jonathan Chaplin with New Street. Go ahead please. Your line is open.
Just following on the question around cost of service. I just want a clarification. You said we shouldn't impute the same improvement in cost of service that we saw year-over-year this year or we shouldn't assume the same cost of service. That is something that would drive cost of service from here assuming what kind of quarter you passed there. And then related to that, I'm wondering if you can give us an update on where the Legacy Time Warner churn is relative to Legacy Charter churn and whether Legacy Charter churn is finally bottomed down or there’s still opportunities for that to come down? Because it seems like the churn improvement was a big component of the improvement in cost of service.
So you little gobbled there, Jonathan. But on the cost to serve, the cost to serve trend has continued to go down on a per customer basis. There are all sorts of reasons, that's true. I think Chris is trying to say that what the pace that is and how that will manifest itself quarterly isn't something that you should straight line and extrapolate. But the fundamental part that we're on in terms of customer self-serving using high quality service operations with well skilled people doing the work is reducing overall transaction volume. And this is the digitization and the definition of the network-as-a-software defined network and many of the operations becoming software defined. All of those things are taking fundamental operating costs out of the business and capital expenditures that we made over the last three years as a result of the integration puts us in a position to realize that upside. So that's the fundamental notion that he was trying to express. And Legacy TWC churn continues to be both elevated relative to Legacy Charter, but continues to be declining at a faster rate than Legacy Charter continues to decline.
And it should be a gap now, it used to be ....
Hey, Jonathan, you've got really bad cell signal and so this time around we couldn't hear anything on that question.
Thanks Jonathan. Operator, we'll take our next question, please.
Our next question comes from the line of Jessica Reif Ehrlich from Bank of America. Go ahead please. Your line is open.
I have a follow-up. So I think it was Tom who said that you will accelerate or you expect to accelerate customer growth rate this year. And I was just wondering, if you could elaborate on that? Do you mean high speed data only or are you including video in here? And then you also Tom mentioned, Flex and I don’t know if it was just in the context of direct-to-consumer services, but do you have plans to offer a service similar to Comcast Flex and/or can you talk about how you're thinking about the evolution of your broadband product that would differentiate -- make it more different besides speed, but that else is a differentiator?
Well, when I spoke if accelerating growth, I was talking about high speed data and customer relationship growth accelerating. So that's what I meant by that. And in terms of Flex or similar products, yes, the answer is yes. There are a variety of IP relationships we have. We have one with Apple, selling IP devices through Apple. We have Roku devices that are product design. We've got millions of customers who subscribed to us directly through an app based product. And so our video product on the Internet or IP delivered cable TV and IP delivered over-the-top products are all being delivered through a variety of new technology platforms. Flex is one of those. And so the answer is, we're pursuing all the various opportunities in video that are available to us and including those in our broadband strategy.
In addition to our -- we've enhanced our broadband through speed differentiation and taking our minimum speeds up. More than half of our customer base now gets a minimum speeds of 200 megabits as the slowest speed we saw. And we've enhanced that with our advanced Wi-Fi products, which deliver high quality service throughout the house, allow you to manage all the devices in your house, see what the service quality of those devices is and how they're connected to the network, as well as allow you to manage the privacy or the utilization of any of those devices throughout your house, so that you're secure, private and getting a high quality service everywhere inside your property and on the go. And so we continue to invest in the broadband platform to make it a better platform. And we continue to invest in legacy video and we continue to invest in the way legacy video transforms itself into a IP platform.
Our next question comes from the line of Vijay Jayant With Evercore. Go ahead please. Your line is open.
Hi, it's James Ratcliffe for Vijay. Two if I could. First of all Tom mentioned 10G is sort of the next phase and that with the 3.1 rollout done. Can you give us any idea of the timing or a magnitude of that sort of deployment? I assume it's differing in kind just degree to the $9 for home-pass so that 3.1 was? And secondly, in a slide you mentioned that the strategy in the business is not dependent on M&A. Can you talk about what you see the M&A landscape looking like both any opportunities for horizontal or if there are any incremental vertical acquisitions that would make sense?
Sure. So with regard to 10G, we just upgraded the whole network to 1 gigabit everywhere. That was the $450 million capital project I mentioned over a 13 month timeframe. 10G is a set of technology specifications that the industry has developed that allows us to ultimately get to 10 gig symmetrical services, which are provided at very low latency, delivery specifications and allows you to put high compute capabilities throughout your network. And there is no immediate need to deploy a new upgrade in the marketplace today, but it's an evolution that we can invest in as we go forward. And it allows us to do that incrementally in a very cost efficient way, a lot less cost than it would cost to build a brand new network. And so we have no immediate capital deployment associated with it. But it's a variety of tools to incrementally get you to at least a 10 gig symmetrical. And there's -- we have already surpassed that capability in our laboratory testing. And so that's just a stake in the ground in terms of what the potential of our existing infrastructure is.
With regard to M&A, first of all, we don't have anything that we're about to talk about. But the cable business is owned by controlled shareholders mostly through the United States. And I think we have a great business and lots of opportunity, and with the right combination of assets there's always value in scale and market approach. And -- but there's nothing for us to say at the moment.
Our next question comes from the line of Bryan Kraft with Deutsche Bank. Go ahead please. Your line is open.
I wanted to ask you a couple questions on CapEx. Can you talk about some of the puts and takes in scalable infrastructure capital? It's down quite a bit in 2019 from the prior couple of years. Just wondering how we should think about that spending category in the coming years relative to history. Is there a sort of normal level if you will?
And then secondly on CapEx. As you reach the end of the SEC commitment for new homes passed, what should we expect in terms of a normal homes passed growth rate? Will it kind of be the -- just the rate of home growth in your footprint? The way it's been maybe historically over the years or do you see opportunities now to extend the network to existing premises that are currently off-net? Thanks.
So with regard to scalable CapEx, Bryan, the -- we are getting advantages from the 3.1 deployment. And I think that those advantages, which are in with -- what the advantages is that the growth in Internet utilization on a per customer basis is a continuous investment required by us and the 3.1 so expanded the capacity of the plan that some of that scalable infrastructure capital that might have been in prior periods, isn't required right now. But that opportunity continues for a while yet, because once it's a function of the data usage per unit per customer.
And interestingly our Internet-only customers now are using over 500 gigabytes per month, half terabytes of data usage. So it is something that continues to climb. And that's relative to a wireless average customer of [8], to put that in proportion. But the capital expenditure, I think will stay in that space similar to what it is now for some period of time, clearly 2020, because of the capacity that’s been built by 3.1 and the evolution of traditional video towards DOCSIS, any of that opportunity.
Homes passed, look, we build everything that we can build from a homes passed perspective without regard to the requirements of our consent decree. So, we've accelerated and actually built more passings than the consent decree required, and there's nothing going forward that would change the rate at which we're building other than the rate of household formation.
And Bryan just on the CapEx, Tom mentioned we're temporarily depressed in 2019 and probably in 2020 for the continued DOCSIS 3.1 benefits. But if you think about the different line items for CapEx, our CPE which includes traditional installed, which is now our growing higher self installed, our CPE is declining and already the majority of that mix of CPE is more tied to Internet-related products as opposed to video products.
As Tom mentioned, we've been building at an accelerated pace on the line extension. So I think that'll continue to be elevated with positive ROI attached to that build out. And the scalable while lower last year, probably this year as well, over time that's an area that we will continue to invest like we have in the past. And then in support, we continue to have not the same magnitude that we've had the past few years, but there's still a fair amount of integration -- [back into] the integration capital that's going on and you'll see that in the support category, whether that's through real estate or through IT systems. And so, whereas scalable may be temporarily lower, I think support is temporarily still higher. And that's the way the different pieces of CapEx will move over the short to medium term.
And on-balance, continued declines in capital intensity as you look beyond 2020 for the most part?
We don't see anything that changes the arch that we're on. And mobile will have its own trajectory, which I talked about in the prepared remarks, which we expected to step down significantly in 2021 but for any ROI based investment in spectrum and/or build out.
Thanks, Bryan. James we'll take our last question, please.
And our last question comes from the line and Doug Mitchelson with Credit Suisse. Go ahead please. Your line is open.
I wanted to dig in more specifics on a couple of topics. Tom, I think to ask the question directly that I think folks were trying to, will you need to subsidize OTT video services to compete in broadband in the future against the likes of AT&T that's going to include HBO Max for certain customers. So, one, the competitive outlook going forward with broadband? Two, on the spectrum, it's super interesting that the conversations you've been having with us around wireless and the lack of the need for a physical network, I guess my question is, how much spectrum do you need because if you're just going after high capacity areas, I think you could do that with small amount of CBRS or CBM spectrum, 10 or 20 megahertz. But if you're going to build out 10 or 20 megahertz, you might as well build out more because the cost to build is sort of the same regardless of how many megahertz. So I'm just curious how you think about quantity of spectrum relative to that return on invested capital dynamic? Thank you.
Well, on the subsidization of OTT, that's really just price, what you're selling broadband for. And I think that we have a relatively good competitive posture from a price perspective in the marketplace. And so I don't see us changing that. But there we will do promotional offers that have a price effect in them, the cost in them. I won’t say never, but it's just a marketing technique. It doesn't mean you have changed your fundamental pricing strategy. And with regard to wireless, you're into the math of what's the value of the physical network pieces and that's a function of a lot of things. It's not -- it is the cost of bills, the cost of what your rented network is and how those two things interact with each other. And so I'd say your thinking is correct, but we haven't decided exactly what we will do.
If I could just follow up on that, Tom. Do you need updated MVNO with Verizon or a new MVNO to take advantage of mid -- maximize taking advantage of building out an advanced spectrum in high traffic areas? Does there need to be any evolution in that relationship for you to be able to switch to your spectrum?
We have a good relationship and we're happy with our MVNO relationship with Verizon. And I'm sure it will evolve through time.
Thanks Doug. And thanks to everyone who listened to our call today. That concludes our call.
Thank you very much.
And it does conclude today's conference call. We do thank you for your participation. You may now disconnect.