T-Mobile US Inc (TMUS) 2017 Q2 法說會逐字稿

Good afternoon.

Welcome to the T-Mobile US Second Quarter 2017 Earnings Call.

(Operator Instructions)

I would now like to turn the conference over to Mr. Nils Paellmann, Head of Investor Relations for T-Mobile US.

Please go ahead, sir.

Thank you very much.

Welcome to T-Mobile's Second Quarter 2017 Earnings Call.

With me today are John Legere, our President and CEO; Braxton Carter, our CFO; and other members of the senior leadership team.

Let me read the disclaimer.

During this call, we will make projections and statements about the future performance of the company, which are based on current expectations and assumptions.

Our Form 10-K includes risk factors that could cause our actual results to differ materially from the forward-looking statements.

Reconciliations between GAAP and the non-GAAP results we discuss on this call can be found on the Investor Relations page of our website.

With that, let me turn it over to John Legere.

Okay.

Good afternoon, everyone.

Welcome to T-Mobile's Second Quarter 2017 Un-carrier Earnings Call and Twitter conference, and we're coming to you live from our headquarters in Bellevue, Washington.

We're mixing things up this quarter and we're breaking with tradition a little bit by getting our results out first and ahead of the pack.

And why not?

We've spent the last 4.5 years breaking industry rules and dashing the hopes and dreams of our competitors, so why not change the order this quarter and give you the biggest piece of the wireless industry puzzle first?

Our Q2 story is simply outstanding, so buckle up and let's get started.

You have the release and the Factbook but, as always, I'd like to hit the highlights of the quarter.

If you're looking for the big news from Q2, you could simply say that T-Mobile's business continues to perform at peak levels across the board.

Our incredible customer growth, that I'm sure the other guys won't mention next week, combined with an all-time low churn, yielded a record quarter for the Un-carrier.

In fact, Q2 marks 17 quarters in a row where we've added more than 1 million customers.

Now think about that for a minute.

17 quarters.

Now we added 1.3 million total net adds this quarter, so it's safe to say, our traction with customers continues.

We added 817,000 branded postpaid customers, so it's looking like Q2 will be the sixth quarter in a row that we led the pack on the total postpaid side.

By the way, our business channel at work contributed its highest share of postpaid customers ever, so we're making great progress towards breaking down the duopoly stranglehold on business customers too.

In fact, 181 new logos in the enterprise, federal and public sector were added in Q2, and over 40% of Fortune 1000 companies are now T-Mobile customers.

And we added 786,000 branded postpaid phone customers.

That's 14 quarters in a row that we've led the entire industry in postpaid phone net adds.

We expect to capture all of the industry postpaid phone growth this quarter again.

Our competitors have yet to report, but based on analyst forecast, it seems like AT&T is focused on big-time M&A and fine with losing a couple hundred thousand postpaid customers this quarter, and frankly, every quarter.

Verizon's massive marketing blitz on its unlimited plan looks like it might disappoint.

I bet it also cost them a real pretty penny.

Sprint has been giving away phone service for free, like literally giving it away.

What would their results be without a free year of service?

Our record low churn certainly contributed to our exceptional postpaid phone net adds.

Branded postpaid phone churn in Q2 was down 17 basis points year-over-year and 8 basis points sequentially to a new record low of 1.10%.

Okay.

So how did we do in porting ratios?

Well, this quarter makes more than 4 years in a row every quarter that we've ported positive versus the industry overall and more than 3 years every quarter that we've ported positive against each and every major carrier.

This was a competitive quarter.

It was the first full quarter with all the unlimited plans in the market.

It was also a quarter in which one desperate company gave away service for free.

And yes, we even had a new entrant from cable but I'll give them a hall pass for now.

In our prepaid business, we added 94,000 new customers.

MetroPCS continues to win customers at a healthy pace, but we also made it clear that we chose not to respond to irrational offers from some of our competitors.

Oh, and I should note that prepaid ARPU reached a new record of $38.65, up 2.1% year-over-year.

Last, but for sure not least in our list of highlights, our financial results are fantastic too.

In Q2, we delivered our highest absolute service revenue ever with 8% year-over-year growth and 10% in total revenues, where we expect to lead the industry for the 16th time in the last 17 quarters.

We generated strong net income, which was up more than 2.5x compared to last year, while free cash flow grew by 15% year-over-year.

Adjusted EBITDA of $3 billion reached a record high, up 19% year-over-year with a 40% margin, up 300 basis points.

These amazing results across the board are in a large part due to investments we've made and will continue to make in our network.

No magic tricks here, just good old-fashioned focus and execution from Neville Ray and the engineering rock stars.

The team has already started network deployment on 600 megahertz spectrum that we acquired in the recent auction.

In typical T-Mobile fashion, we're not wasting any time, and we plan to light up the first 600 megahertz site in August.

We expect spectrum covering more than 1.2 million square miles to be clear in 2017, with actual deployments in many areas by year-end.

We expect to have several compatible devices by the holiday season so our customers can take advantage of this right away.

And we'll use a portion of our 600 megahertz spectrum holdings to deploy America's first nationwide 5G network in the 2019, 2020 time frame.

We continue to grow our 4G LTE network, which covers 315 million people today, and we have 321 million in our sights by year-end 2017.

We remain the fastest network in America.

We've been the fastest network in America for 14 quarters in a row and the gap is getting even wider.

Amazingly, Verizon fell behind AT&T in terms of download speed.

Both Verizon and AT&T are completely choking in the wake of their unlimited launches and have seen significant network slowdowns.

Their networks just can't take it.

Meanwhile, T-Mobile's network has actually become even faster in download speeds, and America's best unlimited network just keeps getting better.

As usual, we're just getting started.

Our network expansion enabled us to compete in every inch of the country now and in every segment of the market.

We're making incredible progress opening new T-Mobile stores to go from covering 2/3 of the country to 3/3 of the country.

Last week, we opened our 1,000th new T-Mobile store this year, with 500 more planned by year-end.

T-Mobile has opened stores now in over 400 new cities and towns this year alone.

And that's in addition to the 1,500 MetroPCS stores planned this year, 1,100 of which have been opened to date.

By year-end, we will have nearly 17,000 branded locations across the country where customers can buy T-Mobile or MetroPCS, and that's just incredible.

So putting all this together, the new spectrum and the work we're doing with our current spectrum sets the stage for continued momentum and future growth of T-Mobile.

We will bring the Un-carrier to every inch of the United States, bringing real choice and competition to all wireless customers.

This is also a great story for rural America, much of which is seeing or will see real wireless competition for the first time.

Places like Wyoming and Montana can now have more options, thanks to T-Mobile.

I really have never been more confident about the future of T-Mobile as we look to the second half of 2017 and beyond.

Braxton will update you on our detailed 2017 guidance here in a minute.

But since we're reporting first, I thought I'd share just a few predictions about Q2 results.

Are you ready?

I predict that T-Mobile will be the only provider to grow total wireless revenues by double digits year-over-year.

I predict that we'll be the only carrier to grow wireless service revenues at all year-over-year.

And I predict that we will take all of the industry's postpaid phone growth again, at the same time achieving record profitability and strong free cash flow growth.

Okay.

Let me hand it over to our CFO, Braxton Carter, for more financial highlights.

Braxton?

Yes, thanks, John.

Let me give you a quick snapshot of our very strong financial results and the details of our 2017 guidance.

Let's start with the financial results that we simply crushed for the second quarter.

Our customer growth continues to translate into strong financial growth as we delivered industry-leading metrics once again.

Service revenues grew by 8% year-over-year, net income grew by 158% and adjusted EBITDA grew by 19%.

The adjusted EBITDA margin expanded to 40%, up from 37% a year ago.

EBITDA benefited from strong cost discipline.

SG&A as a percent of service revenues declined by 100 basis points year-over-year, while the cost of equipment sales declined to 114% of equipment revenues, down from 131% in the first quarter and 120% in the second quarter of 2016.

Cost of service increased in absolute dollars due to our network expansion and the timing of expenses associated with that expansion but still came down 30 basis points year-over-year in percent of service revenues.

Free cash flows improved 15% year-over-year to $482 million in the second quarter of 2017.

Recall that the second quarter of 2016 benefited from net proceeds from the sale of receivables of $371 million related to an upsizing of the EIP securitization facility.

Net cash from operating activities increased by 3.5% year-over-year to just over $1.8 billion, while cash CapEx was essentially flat at $1.35 billion.

As predicted last quarter, the second quarter was also impacted by higher cash interest payments of $727 million, $640 million of which was impacted free cash flow compared to $399 million in the second quarter of 2016.

This resulted from higher net debt post the auction and the payment of call premiums amounting to $238 million in the second quarter.

Earnings per share came in at $0.67 in the second quarter compared to $0.25 in the second quarter of 2016.

The effective tax rate amounted to 38%, returning to the normal rate following the net tax benefits recorded in the first quarter.

This also explains the sequential decline in reported EPS.

Recall, the first quarter EPS, excluding the after-tax spectrum gains and the net tax benefits, amounted to $0.48.

Therefore, EPS adjusted for these factors actually increased sequentially.

Branded postpaid phone ARPU was $47.01 in the second quarter, essentially flat compared to $47.11 in the second quarter of 2016.

ARPU was impacted by the continued migration to T-Mobile ONE, including taxes and fees, the impacts of promotions and the successful launch of DIGITS.

We continue to expect ARPU to be generally stable from full year 2016 to full year 2017, with some quarterly variation driven primarily by the actual migration ramp to T-Mobile ONE, including taxes and fees.

In terms of customer quality, we had excellent results in the second quarter.

Total bad debt expense and losses from sale of receivables were $162 million or a record low of 1.59% of total revenues compared to $165 million or 1.78% in the second quarter of 2016, even with a significantly higher customer base.

Let me now come to our 2017 guidance.

We expect branded postpaid net additions to be between 3 million and 3.6 million, increased and narrowed from the previous guidance range between 2.8 million and 3.5 million.

Based on our strong Q2 results, we are increasing our guidance range for adjusted EBITDA to $10.5 billion to $10.9 billion, up from the prior guidance range of $10.4 billion to $10.8 billion.

Our adjusted EBITDA target includes expected leasing revenues of $850 million to $950 million, increased from the prior guidance range of $800 million to $900 million.

Second quarter leasing revenues were $234 million.

We target cash CapEx of $4.8 billion to $5.1 billion in 2017, excluding capitalized interest, unchanged from the prior guidance range.

We do, however, expect to come in at the very high end of the guidance range as a result of our initial 600-megahertz rollout.

Finally, we expect free cash flow, defined as net cash provided by operating activities minus cash CapEx, to increase at a 3-year CAGR of 45% to 48% from full year 2016 to full year 2019, again, unchanged from prior guidance.

During the same period, we expect the underlying net cash provided by operating activities to increase at a CAGR of 15% to 18%.

Now let's get to your questions.

You can ask questions via phone, text message or via Twitter or Facebook.

We'll start with a question on the phone.

Operator, first question, please?

(Operator Instructions) And we'll take our first question today from Brett Feldman with Goldman Sachs.

Just 2, if you don't mind.

I want to quickly follow up on your ARPU commentary.

It sounds like you continue to feel like you have visibility on stable postpaid phone ARPU over the course of the year.

Based on what we saw in the second quarter, it would imply you'd have a stronger back half, so I was hoping you can maybe just walk us through what gets you comfortable with that.

And then, you gave a lot of color on store openings.

I was wondering if you can help us understand the extent to which some of the new stores are making a contribution to your gross additions?

Okay.

Yes, Brett.

I think the ARPU is actually an outstanding story when you compare it to the rest of the national players in the industry.

We have continued once again to reaffirm generally stable ARPU from full year 2017 to full year 2016.

And we've been very clear that the actual migration ramp associated with customers moving into T-Mobile ONE taxes and fees can create a little bit of a quarterly variation here.

But we have a significant pricing umbrella.

You probably saw that we just increased pricing on our most feature-rich T-Mobile ONE Plus offering.

We have great visibility.

We have a significant pricing umbrella compared to the other parties, and we're highly confident of achieving this guidance that we've continued to repeat.

Mike, do you want to talk about the stores?

Absolutely.

Brett, most of the opportunity is still in front of us.

We have -- we're excited to have passed the 1,000th store mark.

I think, as we've talked about in the past, the ramp-up period happens over a period of about a year.

We're ahead of all of our ramp targets.

Our team is doing an amazing job, but most of the upside in terms of our performance is geared for 2018 and beyond.

I will say you're going to see it come in, in a number of ways.

One will be obviously in gross add and net add performance.

Another may be in the SG&A line because as we grow our business, in the absence of expansion geographically or expansion into segments such as business and prime customers, you would be faced with ongoing promotional intensity in order to be generating the kind of growth that we're targeting.

This is an alternate way to generate that growth and it may be more cost-effective.

So you may see it in the nominal growth numbers or you may see it in the SG&A lines in the out years or both.

And I think, Brett, appreciate that question because this year's activity would be driven by the stores that we added last year and our base.

So one of the things that I think is really something for people to pause on is this is -- we're going to end this year with 17,000 branded doors, which is really, when you think about a lot of the discussion of people entering wireless, for example, having 17,000 doors of places where people can go get what is, by far, the most desired customer experience in the industry is really, it's a significant, significant barrier for people to overcome.

So -- and we're not done.

And I think that's the most important stuff.

But I would acknowledge, the team's ability in this year to put 3,000 new doors in service and our ability to attract and retain, which is what you realize is tens of thousands of employees, is really a great story.

We could possibly be the fastest-growing retailer in America.

I'm not sure about that but I can't -- I'm not sure there's many other stories like that but thanks, Brett.

We'll take our next question from Simon Flannery with Morgan Stanley.

John, there's been a lot written in the media about potential combinations, M&A, and we've gone 3 months since the end of the anti-collusion period.

Can you just update us on your thoughts about opportunities for T-Mobile from consolidation or other partnerships?

And then, I think you started to go down the porting route, but maybe you could just give us some specific numbers about Q2 or -- and how you're doing so far in Q3?

Sure, I'd be glad to.

In fact, I'll do the second one first and then we'll have a little fun with the first question.

I think your understated statement about there being noise in the system, I'd be glad to comment on.

But porting for us, again, has been a long-term story.

Quarters that are seen as competitive, not as competitive, we've been very, very consistent with our ability to port positively against the industry as well as each carrier.

So Q2 was just under 1.4, so 1.38-ish, and that kind of broke around to 1.2 with Verizon.

And certainly, the trajectory with Verizon was getting stronger as the quarter went along.

But 1.2 against Verizon in their biggest swing of the bat ever, I'll take that.

We're about 1.3 with Sprint, and we were about 1.63 with AT&T.

Now that trend pretty much continues into this quarter.

We're over 1.3.

And I would categorize it as mostly flat with AT&T versus last quarter, improved versus Verizon even further, and a little bit down with Sprint as they go through some of their promotions, but still, solidly over 1 and 1.1 even with Sprint.

So that's the trend.

The second item is a great question.

I tell you, the anonymous source fax machine to Bloomberg.com has been very busy in the last quarter.

And if Mr. unnamed sources is on this call, maybe he can speak up as well.

I've never seen a group have to respond from an analyst, so I appreciate how hard your job is because somebody meets and says hello, and all of a sudden, you have to model the whole item.

What I would say is this.

First of all, very important as you can imagine, what I want to make sure we walk away with, this is the 2 weeks where all of the wireless carriers have to stop for a minute and then tell us how their existing business is going, not their hopes and dreams and aspirations and side meetings but how is the business going.

And our operating momentum has delivered record results and it continues and we feel good about it, which clearly shows that although everybody would like to say they've got a stand-alone business, we do.

Now secondly, this quarter, I would say we have the same but maybe more opportunities from an inorganic or an expansion standpoint than we had last quarter.

I feel equally as strong about all the same things I did last time.

And I think the comment I would make is what's happening now in rumorville?

First, there was a full year where people couldn't talk to each other.

By the way, we decided to take that year to do the smartest thing you can do, which is win the auction and leave with 45% of all the spectrum.

There's been a couple of months now since the auction, and what we're doing is we're being very thoughtful, very methodical, very judicious about how we decide how to expand our business and act in a fashion that's consistent what's best for our shareholders and customers.

What's happening though, Simon, is that there are rumors out there, and the reason they catch fire is that the concepts have a shred of the philosophical, the academic things that we've been talking about.

We know let's -- what do we know?

We know Comcast and Charter made the gigantic move of deciding that they would sit together and not do anything for a year.

Pretty sure they could have done that separately together without having that announcement.

Sort of sounds like restrain of trade to me but we'll let them pass on that.

And then, together, while they're not doing anything for a year, they decided to have an exclusive discussion for 60 days with Sprint.

And that's all we know.

We know that billionaires, Bezos and Ergen, are meeting and having a nice time but that's it.

The noise-to-action ratio is 0. And all of those things are because we know cable has a (expletive) MVNO relationship with Verizon that they want to do something different with.

We know that Sprint has a need to do something.

And we are interested in all of these options, but we're interested in focusing on our business and doing things in a methodical way, at our pace and at our schedule.

So sorry for going on.

I'm very confident about where we are and I love all the same things I did 90 days ago.

And, John, the official winner of the office pool is the second question was the consolidation question.

So for those that had a second question, congratulations, see us later.

Any more clarity from Washington on how they think about all of this?

How Washington thinks about the fact that there's a nonstop rumor mill going at Bloomberg?

Well, that depends on quarter 3 consolidation.

First of all, I think one of the things that I think we all have to at least accept and acknowledge, all of the activity and the noise that's taking place makes it very clear that there is not 4. There's far more than 4, and there's -- outside of the 4, there's a whole new group leaning against the window trying to get in.

So my -- and I certainly don't have any inside track on how Washington would view the kind of horizontal mergers that you're talking about.

My assessment is that the current administration maybe look more favorably upon this, but I also think, if carriers were to go and propose a merger, if that were to happen, it's up to them to make the story as to why that makes sense for the consumers, why it makes sense for the country, why competition would get greater instead of less, why the 5G opportunity is huge and needs the capital that comes with scale, and also clearly make the case that not only now are we're hearing noise from Amazon but Comcast and Charter, ultimately, Altice, are all going to be playing in the wireless space, and that's the environment under which I think an administration would look at this.

And I would -- certainly, just my own opinion, I think they'd look favorably upon it.

Our next question today comes from John Hodulik with UBS.

Two if I could.

First, a follow-up to Simon's question, maybe for John.

You mentioned that Sprint is potentially in negotiations for a MVNO with cable.

Would that -- how would that change the sort of attractiveness or -- of a combination between T-Mobile and Sprint, in your mind?

That's number one.

And then, number two, you mentioned strength in the business market in terms of the results.

I don't know if you shared specific numbers for this but maybe you can give us a sense of an order of magnitude in terms of share of -- or percent of the gross add that come from the business market, where it was, where it is now and sort of what the opportunity is in -- on that side of things?

Yes, I'll do the first one, Mike, and then I -- but really do want to make sure Mike (inaudible) talked about the business opportunity because that's the place, if you want to look at a pocketful of growth opportunity for us there, I don't think you've all been counting on, that's the one.

Now let's also be clear, John.

I don't have any insight as to what Sprint is negotiating with who.

I read what you read and I'm very patient watching it.

It would seem logical that one of the things that a cable company would want to talk to Sprint about is an MVNO.

And I think one of the things that would possibly be thought about is their network needs investment, so how are they going to provide an MVNO alternative unless they get a significant investment in their network?

These are all academic.

They're not as simple as they sound.

And again, I think -- and I've seen a lot of what you've written as well as others, nothing has changed to the logic stream of carriers like T-Mobile and Sprint considering coming together.

We have different views about MVNO.

I certainly wouldn't -- it wouldn't be my first choice to arm the cable players with a tool that could help them be more competitive.

However, I've been very clear that I believe, over time, these industries are all coming together anyway.

Those are the kind of things that may -- even if you did attempt a transaction, they may happen in a regulatory approval environment.

So these are all the cart before the horse, what's going to happen.

Over time, if you look 3 to 5 years, you know how all this is coming.

I've been very clear.

Customers are going to drive a ubiquitous look across all of technologies for you to get together and provide them a seamless experience having their content available to them wherever they are.

So never say never.

I think we're going to have probably and equally, if not more, interesting environment in the next 3 months.

I think, what we all need to do is pause, catch our breath, understand underneath all these Sun Valley meetings and Bloomberg articles.

How are the businesses doing?

How are the networks performing?

How are the investments taking place?

And what are the real companies' assets, brands and people behind these conversations?

I couldn't be more confident that T-Mobile is a very strong kind of player in whatever ultimately happens to bring things together for customers.

And on the Business Markets, we're really, really excited about the development we're seeing here.

We refer to the group as the At Work group.

I think we mentioned in our remarks that 181 new logos across large enterprise and public sector in the last quarter.

Now 40% of the Fortune 1000.

What we didn't mention, and this is really interesting, is that our share of requirements in that Fortune 1000 and public sector is still very low.

Our market share is still very low.

What we've established now are successful relationships from which we can grow.

And we're now starting to mine those relationships and deliver value to the customers.

For example, this was -- this quarter represented the highest-ever percentage of our postpaid net adds that were represented by the At Work group.

And that's across all those larger sectors like enterprise and government but also business customers interacting with us through our fleet of retail stores.

So that's a terrific accomplishment.

The growth rate year-over-year is more than twice the rate of consumer.

And I think it reflects the opportunity that we have, the acceptance that we now have from businesses, thanks to the big network push, and it also represents the upside because this is an area where we have much lower historical market share than we have on the consumer side.

So when we talk about our push from 2/3 to 3/3 of the country as a strategy, we're talking about 3 things: geographic expansion with all that retail expansion that John talked about; segment expansion in 2 flavors: one, prime consumers and suburban families; and secondly, business customers of all sizes.

And what's driving all this, the catalyst for all of it, is the fantastic network expansion that we've realized over the past 2 years.

Anybody see one on the board they want to grab?

Any suggestions?

Point me out.

Okay.

Kyle Romanoff.

As phone upgrade season approaches, what steps will T-Mobile take to attract switchers?

I'll jump in.

What is question.

Question for self.

Kyle, this is really interesting.

I mean, these phone launch moments that happen a couple times a year from the big phone manufacturers are huge share switching opportunities for us as competitors.

I will point out a couple of things.

One is that the share switching opportunity happens over time.

It doesn't happen all at once, and that has to do with supply constraints that we've seen in past years.

This launch cycle happens over 2, sometimes even 3 quarters as people reassess whether they've got the right phone and then use that as an opportunity to assess whether they've got the right carrier.

I'll point out that we've been seeing real success with people picking T-Mobile, catalyzed by a new phone choice.

But it's not always because of something we do around the phone value proposition.

In fact, people are coming to T-Mobile because of T-Mobile ONE.

They're picking T-Mobile because we have America's best unlimited and T-Mobile ONE is a differentiated offer.

That's certainly the way we approach the Galaxy launches this year.

We took a lot of share through that Galaxy launch.

As you know, we took all of the postpaid phone net adds in the last quarter and we expect to again this quarter.

But we did it by focusing on our service value proposition and our differentiation, T-Mobile ONE, not necessarily through doing something differentiated on the phone offer itself.

And that's a formula we're very comfortable with.

By the way, for those on the call who don't know who Kyle Romanoff is, a few quarters ago, he's a young man that came in and acted as an interview for us.

He just had his birthday.

How old is he, Dave, approximately?

16.

Yes.

Okay.

I'm just sitting and thinking, when I was 15, I probably wasn't sending a note to the CEO of T-Mobile about what we're going to do to attract switchers with the decks.

Certainly not having an answer.

No.

No, I'm pretty sure you can dial forward 15 years and know who's going to be the CEO of whatever company they want.

Let's see.

He has a great question.

Does anyone else think Braxton won the Q2 earnings results wardrobe award with that hat?

Yes.

I agree.

I think they meant that hat.

That hat, by the way, I was in Guatemala last week with one of our big care centers, and there was a performance by a team from our Mexico care team and they gave us that hat for Braxton.

But as you can see, to be careful, Braxton went straight back to the hat that's brought him the best luck over time.

Let's -- well, let's see.

Shocking, there's about 25 questions from Walt Piecyk.

Let's see, which is -- is Walt also on the list over here?

Wow, you get no respect anymore, Walt.

It's a good thing.

Pick a Walt question here before he gets around to asking Neville.

Let's see.

I think, Braxton already answered the postpaid ARPU question, so let's go to what type of Sprint promotion would you react to if you didn't respond to 1 year for free?

Look, we don't react to people's promotions.

I think what we've got is a market where other people react to our game plan.

That's been going on for years in this marketplace.

We take moves and actions, bring innovation to the market.

We drop it on the industry for the benefit of customers and the rest of the industry scrambles to react.

That's the way it's been for years.

We intend to keep it that way.

John, we had a really interesting Twitter question on would we consider a dividend.

I think that's one of the most exciting things about our story, and that's the free cash flow generation and development.

And this year, for the first time, we put out a CAGR on free cash flow between 45% and 48% between '17 and '19.

And what that's going to translate to is over $10 billion of net cash generation after paying all expenses, after paying all interest expense.

And that's quite exciting.

And we've commented in the past that we have, in fact, been looking at options to return capital to shareholders.

We just got past the auction.

We're going to significantly delever organically.

And of course, we will always look at ways to invest in the business for the highest return.

But with that amount of cash, we're actually starting to have conversations about instituting a small quarterly dividend that we can grow in the future.

And I think that's going to be a wonderful thing for all of T-Mobile shareholders.

And we could obviously supplement it with buybacks as our cash progression develops.

So huge item there.

Perfect.

By the way, those of you who are watching on Twitter just saw that Mike Sievert didn't realize that the table here is electronic.

He levitated it and he had a complete explosion of the coffee here.

I blame it 100% on (inaudible)

But we didn't lose our focus for a second.

And, Walt, I -- Walt, I was going to piggyback on the back of the question.

I think it's fair to say that a lot of what Sprint is doing, this is very aggressive ways to attempt to offset churn and to get some subscriber growth.

I think even Sprint would probably not believe that some of their promos are sustainable.

But we've been in that stage where you've got to get very aggressive and turn things around.

So I don't think those are programs that are meant to be responded to.

And I think as you watch what we did with prepaid this quarter, very, very aggressive things that we decided not to overly respond to, but we still were able to keep our prepaid ARPU at a record high and protect our base and grow.

And I think that's probably the logical way.

Our next question comes from Michael Rollins with Citi Research.

Two if I could.

First, you're talking about the porting ratios, that they were a little lower than they were historically, but the gross adds on phones looked like they were up.

And so I was wondering if you'd talk a little bit more about where the incremental gross adds may be coming from between those that port and the end result of what you achieved in the quarter.

And then, secondly, I was wondering if you could unpack more specifically what happened with the margin improvement in the quarter.

And it looks like you've already captured over 50% of your core or cash OIBDA goal for the year.

So how should we think about opportunities to improve EBITDA even further?

Okay.

And again, Mike, if you want to start, or Braxton.

I would say, we were -- first, let's be clear.

We're very pleased with porting in that range.

And ultimately, if the porting -- as an example, last quarter and this quarter, if this goes on in perpetuity, we eventually control the entire industry.

We've been -- we're porting churns at a record low, so we're quite comfortable, for example, that we would port 1.2:1 with Verizon during their most aggressive quarter, and now, as we go back in, being even higher.

And to port positive with Sprint no matter what they do and to keep AT&T at its very, very high levels, I think I'm comfortable with that.

But, Mike, do you want to go through the components?

Yes.

Well, as we run the business, there's 2 things we look to on customer growth.

One is are we winning over more families and switchers on the consumer side, and, of course, businesses on the business side.

Porting's a great way to kind of measure that.

You can see that in our net add numbers, in our share of total growth.

You can see it in our record low churn and our continued porting positive against every carrier.

The second thing we look for though is existing customers deepening their relationships with us.

And we saw that to a record extent this quarter as well.

And that's why you can see a quarter where porting, while continuing very positive, could soften some, and yet, phone net adds continue apace.

Our customers double down with us.

Not only are they staying longer with record low churn, but they're deepening their relationship with us with additional lines.

And I think you saw in the numbers that we disclosed, we achieved an all-time high this quarter of lines per customer on the postpaid side and that helps explain the difference.

And, Mike, the very good question on the EBITDA or OIBDA.

First of all, we give a range and that range has our full growth expectations for the year embedded in it.

And you know the game plan.

Quarter-after-quarter, as we demonstrate actual results, we adjust that growth accordingly, as warranted.

You're seeing tremendous business momentum.

You're seeing tremendous momentum that's just starting to develop with the distribution expansion that will really pay dividends next year.

But we give a range on EBITDA for a reason, and that reason is there can be growth variability for the balance of the year.

I will point out, for the first time, we raised EBITDA guidance midyear.

That shows our confidence in what we're doing, and it shows the confidence that we have on execution for the balance of the year.

Operator, I am so excited for you to introduce the next question from one of the biggest staunch, long-term believers and supporters in T-Mobile US' business.

I'll let you introduce the next question.

Our next question today comes from Mr. David Barden with Bank of America Merrill Lynch.

This is a very Un-carrier.

I wanted to maybe talk a little bit about the recent price change that you guys instituted in the -- I think it was a $5 increase in the kind of T-Mobile ONE higher-level plan.

Could you talk about what the kind of reasons for that was?

And I guess there's been a conversation about what that might mean.

Does it mean that you want to take advantage of your position in the market to try to focus on more profitability?

Or do you want to set yourself up to kind of lower prices again when we get back to the iPhone?

And I've got a prediction of my own, which I think you'll be seeing Verizon do some interesting moves to kind of split the difference on SD and HD pricing as well.

So I look forward to that in the next couple of weeks as well.

Okay.

First of all, thanks for all your work.

I look forward to seeing Verizon make any move.

They haven't really been breathing lately.

But I think your question was a good one because there was a lot of confusion about whether there was a promo expiring or a price increase and what we're really doing around that.

So Mike, why don't you talk about that?

Yes, Dave, I think what you're referring to is T-Mobile ONE Plus.

This is a really incredibly popular add-on to T-Mobile ONE, and it provides customers with a number of extra benefits.

That's something that was available promotionally for $5.

We've put it at its standard pricing of $10, and we're really excited about the value that this represents both for customers as well as for the owners of T-Mobile.

As you I think know, our strategy is always about providing more: more value for our customers.

It's not necessarily about charging less.

It's about providing more for what you pay and overall, better value proposition.

T-Mobile ONE Plus provides not only high-definition video but 10 gigs of tethering; double the global data roaming, which is already an incredible benefit, 256 kilobit per second global data roaming; and an extra line through DIGITS, essentially an add-a-line.

So all that value is packaged together for $10 and what we're finding is at that price, it remains an incredibly popular add-on.

So it's just an example of what I was talking about on the last question, which is customers are demonstrating that they appreciate the value that we're bringing and they're doubling down on their relationships with us, and T-Mobile ONE Plus is one of the best examples.

We were asked when we moved to unlimited over and over again: What's your upside going to be?

Are you going to be able to attract people to deepen your relationship now that you've given away all the data?

And I think this is the latest installment that says we're far from out of ideas on how to entice our customers to deepen their relationships with us.

And I just -- you brought it up, so I'll pig pile it as well.

Kidding aside, obviously Verizon has to do something as it relates to how they not just price unlimited but how they deliver it because their network -- this isn't just an attack.

It's an attack, but it's a data-oriented attack.

Their network is significantly slowing down.

It's choking on this.

And clearly, the only explanation as to why they haven't offered a Binge On-type capability is because they don't know how.

And once they figure it out, it will be a safety net for them to figure out how to get their customers onto it.

And again, Binge On is one of the most incredible technological things that we've done and customers love it.

And so from a standpoint of how it's positioned, also allowing customers who want to have full HD, that's not just a pricing game.

That's a technological game, and we are -- pretty much if you look at AT&T and Verizon, this last quarter, both of their networks slowed down.

So I think what they're learning is unlimited is not the easiest thing in the world.

And as Neville's deployment continues and continues, our speeds are only getting better and better, and we expect that to continue.

Should we talk about prepaid?

We're getting a lot of questions on prepaid.

And -- before we go back to the phones.

We have a lot of fans of MetroPCS out there.

This was just another fantastic quarter for us on prepaid, and what our team has accomplished across both of our brands and particularly on MetroPCS in recent quarters is unbelievable.

This quarter, again, as the largest prepaid provider in the industry with 20 million subscribers, we were able to again post growth in a very interesting and tough competitive situation, while also achieving the highest ARPUs in our company's history on prepaid.

So we're really proud of that.

One big milestone that's interesting, we believe that next quarter, we will surpass the point where we will now have double the customers on MetroPCS versus the 8.9 million customers that came in on May 1, 2013, when we all came together; double the 8.9 million customers just on MetroPCS.

When companies merge, the big question that's asked usually is how many will they be able to hang on to?

How many will they be able to entice to stay?

Instead, our MetroPCS team led by Tom Keys has engineered a doubling of that business.

We believe we'll hit that milestone during fourth quarter.

Mike, I'm glad you brought that up because we don't report MetroPCS numbers individually.

So let's just suffice it to say that we did 94,000 prepaid nets, and Metro did significantly more than that.

And so when we say we chose to kind of ignore certain attacks in the prepaid market, that was mostly around Magenta prepaid, and you can see over time the evolution of where we're going.

We have over 11,000 doors by the end of the year of MetroPCS-dedicated doors, and that's becoming where we anchor our business.

And that brand is stronger than ever, the distribution is great and their ARPU has been tremendously strong.

And as Mike says, you can figure out the math in the time that they've been here if they'd doubled their size.

We have a real winning hand with MetroPCS.

Our next question comes from Jonathan Chaplin with New Street Research.

In comments that the guys at Deutsche Telekom and you guys have made in the past, it seems like in a potential tie-up with Sprint, for you guys -- for T-Mobile to have control and Deutsche Telekom to have control of the combined entity through T-Mobile was really important.

If we're looking at a situation now where there's potentially a much bigger deal involving Sprint and T-Mobile and Cable, is it still critical for Deutsche Telekom and T-Mobile to have control in that situation?

And now that -- sorry, go ahead.

Yes, you go ahead.

I was just going to throw on one more question, which I'll be very happy if I get answered.

Now that the discussions with Sprint are sort of officially on hold, can you give any commentary around where the sticking points were in the prior set of discussions?

Yes, no.

This is a long game.

And first of all, I want to just make sure -- I'm not attacking kind of some of the things that are being rumored.

Let's remember that Masayoshi Son is one of the richest, biggest dealmakers in the world and his moves are significantly tracked.

And I dare any of you to dissect when he's working on Vision Fund and when he's working on -- the guy is one of the biggest players in the world, and what he's been doing makes sense.

That's Masa.

Sprint's very lucky to have him as an owner.

We -- as I said, every option that we had before we have and then some other ones.

And I don't need to point out to you some of the newer things that are coming.

You watch who's playing in the market.

And from a standpoint of talks being officially on hold, I'm not going to comment on any of that.

It's still -- any option still looks the same.

There have been some comments, I believe, made by Deutsche Telekom about what their preferences are.

These are point-in-time preferences.

You can talk to them about what it -- what is their current state.

Remember, in the last year or 2, we've gone from Sprint's buying T-Mobile, T-Mobile's buying Sprint, nobody's buying everybody.

And now we think that Charter and Comcast and maybe Altice and DISH and T-Mobile are all going to get together in some big amorphous pile and come running out.

And it's just -- and by the way, the amount of things you had to write, Jonathan, are incredible.

I mean, I think you as much as anybody would like some clarity.

I think the same kinds of things that made sense 3 months ago make sense now.

These aren't linear.

We're going to be very patient and methodical.

Things that make sense for our customers and shareholders are very apparent, and we have nothing negative to say about those opportunities.

And when things are tangible and when we can come out of the Sun Valley meetings and into the light of the day, I think our shareholders would be very pleased to see the things that we're working on.

So I'm not going to -- it's just I'm not commenting on talks.

I'm not going to comment on if talks ended and if they ended, why and when they're going to start.

But it was a good try.

We'll take our next question from Phil Cusick with JP Morgan.

I'm going to stay away from M&A.

But, Braxton, I've got a follow-up on your dividend comment.

With so much volatility in the industry and a number of carriers tied down by their probably way too big dividends, why would you do a dividend rather than buying back stock?

Is there like a share liquidity issue or a DT capital return component that you're thinking about?

There really isn't, and in my comments I said starting off with a small dividend.

And, Phil, I think one of the benefits to this would be to all shareholders of T-Mobile.

We would open up a whole another class of investors that require yield versus growth.

And it also shows the confidence that we have in our cash flow generation.

By all means, we would start out with a modest dividend, but even a modest dividend would accomplish that.

And then we would look at using buybacks as a supplemental, and what better place to put our money than buying back T-Mobile stock.

I would assume you wouldn't make a decision on this until probably year-end?

It's something that we're starting to discuss now.

There's no imminent decision.

Okay.

And then if I can, one more.

Mike, what's happening with online sales?

It seems like there should be a bigger share of sales at every carrier but seems to be pretty slow.

How are you doing there?

It's coming along great.

We've put in some terrific new technology which has really changed the game in terms of how our customers are interacting with us, not just online but also through partnerships like Facebook Messenger, for example; 2-way asynchronous messaging through our own app, which has been downloaded 20 million times; and many other breakthroughs.

A thing I want to point out though is that there's work to be done in this entire industry before consumers shift wholesale into the online world, and our view is that the industry needs to be radically simplified.

That's why we're leading the way with our Un-carrier moves.

That's the inspiration behind T-Mobile ONE.

It was the inspiration behind Un-carrier Next with taxes and fees included, creating a monthly subscription to the Internet that's so simple that customers really are inspired to take on more and more of it on their own.

We also are rapidly building stores, and you might say, "Well, isn't that intention with an online future?" And we think absolutely not.

Our view is that in the next few years, what we're going to see as customers move their way towards a digital-centric relationship is that they'll have a digitally accelerated retail relationship with us.

And what we mean by that is making retail better, making it more interesting, turning it into a better showroom for our brand, turning it into a fulfillment center for online-first transactions.

And having people come to us for what we do best, which is provide our knowledge and expertise, because no matter how good you may get at it as a consumer, you do this transaction once every 2 years.

We do it several times a day.

So we're better at it, and people like to be able to hold us accountable to provide something we do better than any other carrier in the marketplace, which is provide great customer service.

So we're building stores like crazy, we're putting in the tools to create a great digitally accelerated relationship at retail and we're radically simplifying everything about this company to create a product that deserves a digital relationship in the first place.

Can you give us an idea of where those digital sales are today?

Are there handsets or gross adds?

No, we haven't disclosed the percentage of sales that are purely digital.

I will tell you that an increasing number of our total transactions are purely digital.

So there's 2 things, and eventually we'll give you some more color on this.

There's pure digital, there's retail that's been touched by digital and there's also transactions.

Every single time you call care or change a rate plan or change something about your device is an opportunity for a transaction to be conducted either partly or totally digital.

All those things are growing is all I think we can disclose today.

Some of them are growing really by leaps and bounds, but we aren't in a position to lay out a lot of details for you quite yet.

But I think the preface with which you asked the question, we completely agree with, which is everybody has room to grow in digital utilization at the sales capability.

And we probably have more to grow than any one, which puts it in the giant opportunity basket for us and one that we're significantly planning about.

Now I just want to comment because I'm -- if you take the last several questions together, and this is before we even unleash Neville for one of his very short network updates that would take us well into the evening, my -- you can see the reasons for our excitement and enthusiasm.

Like, one is -- again, I hope I'm very clear.

Value chain migration, merger and acquisition to get much better scale and capability is something we're completely supportive of and very methodically looking at a way to do in the best interest of our shareholders, and I'm optimistic some of that will continue to be on the front burner.

At the same time, the expansion of our retail capabilities, the expansion of our network, the tools that we've been creating and have far room to grow on that, along with the guidance that we've given about the cash profile of the company over the next couple of years, you can see that the hands that we have to play to support shareholders, we even found ourselves here thinking out loud about things like dividends and share repurchase programs.

So I think we've got a basket of tools that are built on top of this great business momentum we have.

That's a hand that I'm very, very excited to play, and I think you've been probing on each of those angles very well, so thank you for your question.

Let's go -- do you see a question up here that's a good launching pad for us to unleash Neville Ray?

Well, let's go to Walt Piecyk.

Our next question today comes from Walt Piecyk with BTIG.

That's a great segue because that's exactly I want to ask about.

Can you hear me?

Yes, you are one of our favorites, and I think one of the biggest changes in the industry in the past 3 months is you finally have stopped Periscope-ing, which is really...

You're going to follow me on Instagram now.

By the way, did you think because you let Barden ask a question on your call that I'm going to get a question on Sprint's quarterly call for the first time in however long?

I'll see if I can help you out.

Come on, Marcelo.

You can take it.

So Braxton's talked about this treasure trove of spectrum at 2.5.

So I'm just curious.

First of all, what's the minimum amount, since you guys were doing these like fat channel LTEs to get your very fast speeds, what's the minimum amount of 2.5 that you would need, that you would deploy in your network?

Do you have to change the number of cell sites that you have today?

Or is your current link budget for that 2.5 spectrum enough to use it?

And then lastly, this is all part of the same question, you paid $0.80 for the 600.

I think if you look at DISH and where the stock is trading right now, it's discounting spectrum at $1.

Where would you think about the relative value of the 2.5, let's say, compared to DISH, which is effectively, their stock is discounting about $1 megahertz [POP].

Is that worth less, more, the same?

So anything on the 2.5 that Neville can fill in would be great.

Let's do this, Neville.

I want to make one comment about the treasure trove because it keeps coming back to haunt us, and then maybe you can go through Walt's question but also give an update and we will -- The Gong Show has come back to television, by the way, so we'll just keep the gong ready if you go to [Fox].

Walt, I wanted to comment on the treasure trove.

Very important to note that if you -- we've spoken about 2.5 from a standpoint of, hey, you're walking down the street and you look up on the right and you see a gigantic pile of 2.5 spectrum and then you ask yourself, "Hey, what do you think of that pile of spectrum?

Do you like it?" And we say, "(expletive), yes, of course, we do.

It's a big pile of 2.5.

It's a treasure trove." It wasn't a commentary on our requirement for it, our desire to transform our company to get our hands on it.

It's simply a commentary on is there value in a gigantic pile of 2.5 spectrum.

And if you took it other than that, you took it the wrong way.

So we don't withdraw our treasure trove comment because I don't think there is a pile of almost any kind of spectrum that Neville wouldn't run over and bear hug.

Neville, do your best to give a brief answer and update on network.

You said brief, no questions.

Well, clearly, Walt, never seen a megahertz we don't like, right?

That's -- no network guy is going to say anything different.

I mean, I think for us, the thing we keep forgetting is that we've just closed on a massive investment in spectrum for this company.

I mean, John referenced it earlier on.

We came out with 45% of the proceeds of the broadcast incentive auction.

That increased our spectrum assets by just shy of 40%, 4-0 percent.

So we have a lot of spectrum to grow into as we continue to grow this business.

We talked earlier on about the speed at which we're deploying 600 megahertz.

So this isn't spectrum that's going to see the light of day 3 or 4 years from now, as many would like to talk about or talked about pre the auction.

This is spectrum that will be in customers' hands with handsets before the end of this year.

So we have a lot of growth and capability in that spectrum.

And then I look at all of the other things and sources of growth and capacity that we're driving in and onto our network today.

We blogged earlier this week and we showed what's happening with speed performance across the major networks.

And I think we might even be able to put up a quick visual for you on the webcast of what that looks like to remind you.

John referenced earlier on, I mean, the Verizon and AT&T guys have been tanking in terms of their overall speed performance.

Neither network was ready for unlimited.

They moved too fast, too quick.

AT&T ended up above Verizon in terms of speed performance for last quarter, well below us but only because of Verizon's god-awful performance.

AT&T ended up in second place.

They declined themselves, but Verizon fared far worse.

So at a point in time in the industry when the other big guys are running around trying to scramble as to how to deal with unlimited, we're actually rocking in terms of our network performance.

We have the best performance in our network in our company's history: the highest speeds, the lowest congestion if you can find it.

And that's because we've done so much to maximize the asset base we have, an all-LTE network, driving VoLTE, driving the feature set with LTE Advanced at a pace which has the other guys' heads spinning.

I mean, you compound all those factors, we're in a tremendous place with the assets that we have.

We're full bore on exploring 5 gigahertz and unlicensed LTE and LAA.

We recently conducted testing on LAA.

We were putting down just shy of 800 megabits per second on LAA testing.

That's not going to be seen everywhere, but the opportunities in terms of how to grow capacity and capability on this network are endless for us.

And so then you come back to this question about how important is it for us to have spectrum from other sources at this point in time.

I mean, my sights are on really I don't think about 2.5 that much.

I don't think about DISH spectrum that much.

It's tough spectrum to come after and to secure.

But we will make sure we have a great future with the assets we have, we've just purchased and we're moving into with unlicensed.

And then I think about the 5G story that's coming.

We're the first carrier that has the capability to talk about launching a nationwide 5G network with the 600-megahertz spectrum we acquired.

Really, really difficult and tough for anybody else to talk that way.

Then you think about millimeter wave.

Industry needs more millimeter wave spectrum for urban hotspot-like deployment.

We have some.

We'd like to see more, and we'll push on the regulatory front to make sure that, that happens.

And then in the middle, Walt, is this thing called 3.5, and there was a question up earlier on, I didn't go at it, but what's our interest in 3.5-gigahertz spectrum.

And I have huge interest in the 3.5-gigahertz block.

If you want to talk about where is the most formative block of spectrum emerging globally for 5G, it's in the 3.5- to kind of 4-gig range.

And so we're now seeing the FCC and the administration start to look at not just CBRS, but really opening up a powerful block of spectrum in that 3.5- to 4-gigahertz range, 500 megahertz on top of CBRS NOI from the FCC just this week.

So think about all the alternatives that are coming that need to be pushed on and need to be worked on, and then what is that 2.5 worth?

What is that DISH spectrum worth?

Well, there's a lot of alternatives is all I can say at this point in time to us.

Neville, I've just...

How about the minimum amount that you could use though, Neville, aside from the valuation?

Like, what's -- is it a 40-block minimum?

20 block?

Like, what would you want to have as a minimum just to even endorse that band in your own network?

Yes, I mean, you're always going to want to look at something of about 40 megahertz-plus, Walt.

I mean, ideally 60.

I mean, I look at kind of the unused spectrum in the Sprint network today, it's primarily the EBS asset.

The BRS is pretty much consumed with their existing operations.

There's an 80-20 rule there.

That's about 80% of the POPs, but probably 20% of the geography of the U.S. So it's far from a national band.

It's got urban deployment use, but yes, you'd need 40 to 60 megahertz.

Our next question today comes from Craig Moffett with MoffettNathanson.

I'm almost afraid to tee up Neville one more time, but can we just add to that very helpful discussion, Neville, about small cells and talk about where you are with small cells, what you're seeing in terms of the pace of getting them zoned and online?

And can you put some numbers around what you're expecting for where you are today and then what you're expecting for this year and next?

Maybe, Neville, add in the evolution of wireless industries in the world as a kind of tailwind onto that.

Go ahead.

Well, I think I covered most of the bases, Craig, but you caught me.

I didn't cover off on densification on the network, and that was one of the pieces Walter implied in his question.

Obviously, we have the most dense network in the U.S. today, and the low band build networks from Verizon, AT&T continue to scramble to come close to matching -- anything close to matching the density we have.

But we're not sitting on our hands.

We've been very active on the small cell front.

As you know, we picked up about 13,000 kind of DAS nodes small cell equivalents through the Metro transaction, which have been moved across to support LTE in all its glory.

In addition to that, traditional small cell -- new small cell build, well over 2,000 in the ground, that number will hit about 8,000 by the end of this year.

And more importantly, we've built a pipeline of over 25,000.

These are fiber-fed, small cell locations that we can run at.

So we're very busy, very active there.

We spent many months making sure we had a very cost-effective and modular model that we could run out and move out at pace.

So we've built a very strong pipeline that we can call down on over the next couple of years.

Performance is really, really good.

We just had a major milestone in L.A., so we've got kind of formative mass in L.A. now, over 1,000 small cells turned up in the Los Angeles market.

And we're actually now starting to see what are the full network benefits: macro network offloads, speed enhancements, really good numbers coming through when you get a material volume of small cells on the ground in a key urban and metro environment.

So we're very active there.

The last thing I'd add is that our goal and what we're working towards is that every small cell come fourth quarter of this year that we deploy would be leveraging unlicensed LTE in the 5-gigahertz band.

And so back to this question about spectrum scarcity, think about small cells proliferating that can leverage primarily underutilized 5-gig spectrum in these outdoor environments is a very powerful tool in the toolkit.

A round of applause for Neville's brevity.

Our next question today comes from Amir Rozwadowski with Barclays.

A couple of questions if I may.

First and foremost, if I think about sort of the churn trajectory of the business, if I look at the churn performance this quarter, both down sequentially and on a year-over-year basis, you folks have made -- sort of continue to make marked improvements there.

What's the long-term goal in terms of the churn and where you think you can get to on a postpaid basis?

And unfortunately, I will turn it back to Neville on a question in terms of size and scope of 5G investment.

There's been a lot of discussions in the marketplace about the size and scope of what will be needed for 5G investments just on an industry-wide basis.

Given sort of the cash expectations and growth expectation, do you feel as though you've got sufficient capital to make those investments on a going-forward basis?

Or does scale, via any sort of inorganic activity, help accelerate that and drive sort of improved technology augmentation by the overall industry?

Amir, I think your question was longer than Neville's answer.

I'm just saying.

I'll start on the churn piece, and then we can turn to the investment piece.

Yes, Amir, we see no reason why T-Mobile can't achieve industry best churn benchmarks.

And if you look around the world, here in the U.S., Verizon has had some of the best churn performance.

We see no reason why we can't meet or beat the churn that any provider in the industry is able to provide.

We have the best customer satisfaction, the highest NPS scores, a rapidly growing network, a rapidly growing network reputation, which is lagging the network and an Un-carrier differentiation that customers really love.

And what we're doing is building the customer base that over time will be able to contribute to long-term churn reduction.

It's a journey.

It's going to have its ups and downs.

It has a seasonal curve to it.

We do expect the second half of the year to be higher than the first half of the year, again, this year as we've seen the last several years, but the long-term trajectory is for continual improvement.

Okay.

Neville, take us home.

Yes, thanks, Amir.

So I'm trying to lay this out the right way.

I mean, I look at 5G and what's happening not just in the U.S. but globally, and there's kind of 3 flavors, right.

And we've got ourselves pretty much wrapped up around the axle in the U.S. on just millimeter wave.

Millimeter wave is important, and it's important for very high throughput capabilities and services in -- primarily in urban environments because the propagation on millimeter wave is going to be very taxed and very limited.

But for us, when I think about capital intensity deploying millimeter wave, for example, a lot of the small cells that we're going out securing space on now, we are looking to secure space to provide a 5G box as part of that rollout.

So the actual cost to come back on the small cell and add millimeter wave capability will be actually very small.

The question is how much of that do you need to do and if there's a massive, massive volume because you just depend on millimeter wave, yes, it could cost you an incredible amount of cash and investment and time and money to make that happen.

But that's not what 5G is all about.

If you swing the pendulum to the other end of the spectrum and you think about the low band spectrum and the need for massive coverage, IoT capabilities and technologies, that's where you need the low band spectrum, similar to what we've now secured in 600 megahertz, to go make that happen.

That's the complete -- it's almost a polar opposite from the millimeter wave story, but it's a very, very important part of the 5G story.

Now for us, how are we looking at that?

So as we start to roll out 600 megahertz, we are very close to securing 600-megahertz radio product already that will be 5G new radio-capable, not fully from a software perspective.

But as we put those radios on the tower tops and spend all that money, it's a radio product that we can upgrade to 5G with software when the standards and the software is complete.

So again, this is another great opportunity for us to minimize this kind of cash burn and effort and time and resources when it comes to low-band 5G rollout.

Then in the middle, you need something in this kind of mid-band space, and that's where 3.5-gig will probably dominate from a global perspective.

Work to be done there, but again, for us, what's that cost?

It's actually rolling out a mid-band radio on a very dense grid, a very dense macro grid that we already have.

And so I think I differ from some of my peers in our competitors about we're going to need 1 million small cells dotted around the U.S. hung on every street pole.

There are a lot of ways to go crack the 5G nut with the right spectrum resources, and that's the piece we're focused on.

Now then you say does an inorganic transaction on spectrum or scale help you?

Sure, it does.

But my job for the company is to make sure that we have an incredibly strong organic path not just for LTE and what happens in this marketplace the next 2 to 3 years, but come the 5G story that will emerge and evolve around us in the '19, '20 time frame.

With this, I appreciate everybody.

Before I give Braxton the final word here, this was a long call.

We covered a lot of ground.

I hope, amongst other things, that you note our energy and enthusiasm not only for where the company is, but for the environment we're in and the opportunities for our shareholders in many different paths down based on and built on a momentum of a business that I couldn't be prouder or happier about.

And I appreciate you hanging in there.

There was a lot of great questions.

Sure, we could have gone for another hour.

Braxton, the final word?

Yes, thanks, everyone, for tuning in and we'll look forward to speaking to you again next quarter.

Thank you very much.

Ladies and gentlemen, this concludes the T-Mobile US Second Quarter 2017 Earnings Conference.

If you have any further questions, you may contact the Investor Relations or media department.

Thank you for your participation.

You may now disconnect, and have a pleasant day.