Shenandoah Telecommunications Co (SHEN) 2016 Q4 法說會逐字稿

Good morning, everyone, and welcome to the Shenandoah Telecommunications fourth-quarter and year-end 2016 earnings conference call. Today's conference is being recorded. At this time, I would like to turn the conference over to Ms. Adele Skolits, CFO. Please go ahead, ma'am.

Good morning and thank you for joining us. The purpose of today's call is to review Shentel's results for the quarter and year ended December 31, 2016. Our results were announced in a press release distributed this morning, and the presentation we'll be reviewing is included on the investor page of our website at www.shentel.com.

Please note that an audio replay of the call will be made available later today. The details were set forth in a press release announcing this call.

With us on the call today are Christopher French, our President and Chief Executive Officer, and Earle MacKenzie, our Executive Vice President and Chief Operating Officer. After our prepared remarks, we will conduct a question-and-answer session.

As always, let me refer you to slide two of the presentation, which contains our Safe Harbor disclaimer, and remind you that this conference call may include forward-looking statements subject to certain risks, uncertainties, and other factors. These may cause our actual results to differ materially from any future results expressed or implied by such forward-looking statements. Shentel provides a detailed discussion of various risk factors in our SEC filings, which you are strongly encouraged to review. You are cautioned not to place undue reliance on these forward-looking statements. Except as required by law, we undertake no obligation to publicly update or revise any forward-looking statement.

Also, in an effort to provide useful information to investors, we note on slide three that our comments today include non-GAAP financial measures. Details on these measures, including why we use them, and reconciliations to the most comparable GAAP measures are included in our SEC filings. These reconciliations are also provided in the appendix to today's slide presentation.

I will turn the call over to Chris now.

Thank you, Adele. We appreciate everyone joining us this morning.

As a preliminary matter, I want to note that, as we reported earlier, the filing of our 10-K for 2016 was delayed. The delay was a result of the time required to finalize the purchase accounting associated with the acquisition of nTelos and the complexity of the three-way nature of the transaction, including the exchange with Sprint.

Although we previously expected to be able to file our 10-K with the SEC by March 16, we were not able to do so by that date. KPMG, our independent auditors, indicated they required additional time to complete their audit fieldwork and complete their internal quality assurance reviews. The Company and the audit committee of the Board of Directors continue to work diligently with KPMG to complete the audit process to enable us to file our 10-K as soon as possible. We currently expect we will be able to file by the end of this week.

While not happy with the delay, it is important that we ensure the complexities of our three-way transaction acquiring nTelos be properly recorded. We are correcting the material weaknesses we identified in our internal controls. Going forward, we expect our control environment to be more robust and we have recommitted to ensuring we have a solid control framework to support our continued growth.

Moving to our results, we had a strong fourth quarter, capping off a very important year in the evolution of our Company, and I am pleased to provide details about the Company's continued growth. During the year, we delivered an expanded customer base and significant revenue growth. We are ahead of our expectations on the very significant projects we undertook with the acquisition to enhance the acquired network and to migrate customers to the Sprint back office.

On slide five, you will see the fourth-quarter 2016 net income decreased to a net loss of $0.2 million, compared to a $12.1 million net income in the prior year, primarily due to the acquisition, integration, and migration costs associated with the nTelos acquisition. For the quarter, adjusted operating income before depreciation and amortization, or OIBDA, increased approximately 89% to $76 million.

Revenues were $155.6 million in the fourth quarter, an increase of about 78% over the prior-year period. Revenues increased chiefly as a result of the wireless acquisition, but also from solid growth in our cable segment. Cable revenues improved as a result of an increase in the number of revenue generating units, or RGUs, and higher average revenue per customer due to the pass-through of video programming cost increases and customers increasingly selecting higher-speed data packages.

On an annual basis, as you see on slide six, consolidated 2016 revenue grew 56% to $535.3 million and adjusted OIBDA increased 63% to $246.1 million for the year. These large increases were primarily due to the nTelos acquisition.

Our wireless highlights start on slide seven. Postpaid customers are up 131% over the last 12 months, while prepaid customers are up 65%. Prepaid's percentage increase was less than previous quarters, due to a change in Sprint's criteria for how long a prepaid customer is inactive before being dropped from our customer rolls. For the fourth quarter 2016, wireless adjusted OIBDA is up $63.6 million, or 133%.

Turning to slide eight, our cable segment also delivered outstanding growth in the quarter, as demand for our high-speed Internet and voice services outpaced the anticipated decrease in video subscribers. Operating revenues increased 10% to $28.3 million, while cable-adjusted OIBDA grew 13.7% to $9.3 million. Customer demand for higher bandwidth data services and growth in RGUs of 5.5% drove the revenue and OIBDA increases. Our steady margin improvement continued, reaching 29% in 2016.

As many of you know, we have diverse revenue streams, and slide nine profiles two additional assets. Our fiber lease revenues increased more than 15% to $11.3 million and our 196 towers generated $1.6 million in OIBDA from lease revenue, down slightly compared to last year's fourth quarter. Our extensive fiber network serves both as a source of revenue, but also helps control our operating costs by supporting both our wireless and cable segment transport needs.

Our long-term success has enabled us to consistently pay dividends since 1960, as you can see on slide 10. In December 2016, we continued that commitment, paying a dividend of $0.25 per share, up 4.2% from the prior year.

With the nTelos acquisition, we committed to a significant investment in a wireless network in the acquired footprint. Consistent with the major wireless and cable upgrades we made in past years, we are confident in our ability to grow our customer base as a result of these enhancements. Our success in doing so when we upgraded our existing wireless markets from 3G to 4G and upgraded the acquired cable markets to offer a fully robust triple play are a testament to our ability to deliver value to our shareholders from making the investments needed to provide a quality customer experience. As customers continue to expect more from the telecommunications providers, our enhanced service capabilities and product offerings position us well to become the provider of choice in the markets in which we operate.

I will now turn the call back to Adele to review the details of our financial results.

Thank you, Chris. I'll begin with slide 12, which shows our changes in profitability.

In the top line, you can see that we had a $9.5 million or 44% decrease in operating income for 4Q 2016 over 4Q 2015. Over the same period, net income is down $12.3 million and earnings per share is down $0.24.

Slide 13 begins with the $9.5 million drop in operating income and then adjusts for certain non-cash or nonroutine items. As you can see, for 4Q 2016, adjusted OIBDA is up nearly $35.9 million, or 90% over 4Q 2015. For the year ended December 31, 2016, adjusted OIBDA is up $95.2 million, or 63%. We also provide continuing OIBDA to reflect the amount of adjusted OIBDA the Company would generate without the benefit of Sprint's waiver of the management fees. This measure is up $26.9 million or 67% for the quarter and up 47% for the full year.

As a reminder, when we acquired nTelos, Sprint committed to waving the 8% postpaid management fee and 6% prepaid management fee up to $4.2 million a month until the total waive reaches $252 million. We provide continuing OIBDA to ensure investors are aware that we will reach this threshold at some point.

In addition, for the first six months after the May 6 closing date of the nTelos acquisition, Sprint did not charge or waive any management fees relative to the nTelos customer billings prior to the customer being migrated to Sprint's back office.

These improvements are consistent with the longer-term trends, as you can see on slide 14, where I have provided the long-term view of adjusted OIBDA. Since 2010, adjusted OIBDA has grown by $162 million. This represents a compound annual growth rate of 19.6%.

To better understand the factors driving the consolidated results, I provided the fourth-quarter OIBDA results by segment on slide 15. Adjusted wireless OIBDA has increased by $36.3 million, or 133%, while cable results have improved by $1.1 million, or 13%. Wireline results have increased by $300,000, or by 4%.

On slide 16, I've analyzed the changes in adjusted wireless OIBDA results between 4Q 2015 and 4Q 2016. Postpaid revenues have grown by $46 million, due to a 134% increase in the average number of those customers in 4Q 2016 over 4Q 2015. Prepaid revenues have grown by $10.4 million as a result of an 80% increase in those average customers. The waived management fees I described earlier contributed $9 million to the growth in adjusted OIBDA.

The amendment to the affiliate contract with Sprint, effective January 1, 2016, specified that we separately receive or pay certain revenues and expenses that were previously included in the postpaid net service fee. As a result, we began receiving travel revenues, which amounted to $5.8 million in the fourth quarter of 2016, and reimbursing Sprint for handset subsidies and commissions related to sale through national channels of $9.6 million. Handset subsidies incurred on customers acquired through Shentel stores and other acquisition costs rose $2.5 million in 4Q 2016 over 4Q 2015. Network costs increased $22.2 million as a result of a 163% increase in the average number of cell sites from 550 in 4Q 2015 to 1,446 in 4Q 2016.

On slide 17, I have shown the components of the changes in adjusted cable segment OIBDA. The positive changes include significant growth in high-speed data revenues of $1.9 million as a result of a 10.3% increase in average high-speed data, or HSD, customers. Also, the HSD customers are choosing higher speeds that carrier -- carry higher service charges.

Video revenues were up by $500,000 as a result of a slight increase in video subscribers, combined with increases in video rates driven by higher programming costs, which also grew by $500,000. Voice equipment and other revenues grew by $300,000 as a result of the growth in customers, primarily a 6.7% increase in average voice customers. Selling, general, and administrative costs have increased by $200,000, primarily as a result of increases in bad debt expense.

Finally, network costs have increased by $800,000, primarily as a result of increases in [poll] rentals and maintenance.

As Chris mentioned, we have concluded that we have a material weakness -- or material weaknesses in internal control. The weaknesses primarily relate to the management review controls over the work performed by third-party experts we engaged to perform the valuation of the assets acquired from nTelos in exchange with Sprint and the tax accounting for these assets. These valuations impact the allocation of the purchase price between intangible assets and tangible assets on our balance sheet. There is also a weakness regarding the statement of cash flow [presentation] of the buyout of nTelos's minority shareholders with shares of Shentel stock.

Management is expected to perform and document a complete and precise review of the assumptions provided to these experts in the results of their work. These reviews were not adequate or precise enough to satisfy our control objectives. The results of having subsequently refined these assumptions is included in the financial statements, where we show the initial estimates of the purchase price allocation in the revised estimate. The shift in the purchase price allocation also affected related depreciation and amortization of those assets. These deficiencies did not result in any material uncorrected misstatements in the financial statements as of December 31, 2016.

I fully support and reiterate Chris's commitment to ensuring that we have adequate resources on staff to ensure that we can adequately support the financial reporting going forward.

At this time, I will turn the call over to Earle to go into greater depth on some of the operating factors driving our results.

Thank you, Adele. Good morning, everyone. We released customer numbers in early February so I won't spend much time rehashing all those stats, but I will highlight some of the key Q4 results.

As shown on slide 19, we ended 2016 with 722,562 postpaid and 236,138 prepaid subs for a total of almost 959,000. Netted from that number are 24,348 prepaid customers that Sprint purged in December as a result of shortening the time that nonactive customers are deemed to have churned. Since these were nonrevenue producing subs, the reduction had no impact on total prepaid revenue, but will cause the billed revenue per customer to increase.

Moving to slide 20, in Q4 we had 47,988 gross prepaid additions, with the gross additions in our legacy area similar to last year. The South and West regions, the former nTelos areas, actually had more gross adds than our legacy area with mostly the Sprint national advertising, which bodes well for the prospect of gross adds once the network is upgraded and we layer in network and other local advertising as well.

We had net additions of 3,777, with 7,014 net additions coming from our legacy area. Phones represented 95% of the legacy area net adds and 74% overall. Our Q4 porting ratio in the legacy area was 1.7 to 1, taking share from all carriers. Overall churn was 2.1%, with legacy area churn at 1.6%.

You see on slide 21 that our gross billed revenue per postpaid user has leveled off the past few quarters and actually increased in Q4 over Q3. We expect that the revenue per user will continue to fall with promotional pricing and as more of our base migrates from subsidized service plans to phone leasing and installment sales. In Q4, subs with subsidized plans dropped from 43% to 39%.

Slide 22 shows that, similar to Sprint nationally, we continue to lose prepaid subs. The 39,000 loss was primarily due to the 24,000 purge I discussed before, with the remainder coming from the South and West regions. Bucking the trend, we had a net loss of only 43 in the legacy area, with a net increase in higher-margin Boost customers and the loss of Virgin Mobile and Assurance subs.

On slide 23, we see that prepaid Q4 legacy churn was down from both Q4 2015 and Q3 2016. The higher combined churn in Q4 was the result of the final cleanup and shutdown of the nTelos prepaid platform. As of year-end, we had migrated a total of 128,934 nTelos subs, far exceeding our target of 93,000. We expect to finish the remaining postpaid subs by Q3 2017.

On the right side of the slide, you see that the average gross prepaid billed revenue has rebounded to almost $31, primarily due to continued growth of the Boost base.

As shared in the customer release in early February and shown on slide 24, we exceeded our network upgrade and expansion goals for 2016. Let me recap where we are on the project and review our plans for the next two years. We will complete the upgrade of the existing nTelos network by the fourth quarter this year. We will build approximately 90 new coverage sites in 2017 and 100 in 2018. As stated when we announced the deal, the cost of the upgrade and expansion was set at $240 million to be spent over a 24-month period.

After 2016, we will still -- we still believe that that is a good number. We spent $104 million in 2016, we will spend $86 million in 2017, and $50 million in 2018 for a total of $240 million, although it will likely be the end of 2018 before we finish all the expansion sites.

On slide 25, we are providing you the network stats we have provided in the past. Our legacy network continues to run well and we expect dramatic improvement in the quality of our new area once the upgrade and optimization is completed, likely by the end of the third quarter. You see the continued growth in data usage, with our average user in the legacy area using 6.6 gigabits and 5.9 gigabits in the new area. I've not seen any other carrier showing these levels of usage, which I believe is an indication of the amount of capacity we have built into our network versus other carriers.

Moving to cable on slide 26, we continue to see the same trend of steady growth in our high-margin, high-speed Internet and phone customers and a decrease in lower-margin video customers. We continue to see not only growth in high-speed Internet customers, but existing customers continuing to upgrade high-speed plans.

That migration to higher speeds is reflected on slide 27. Our high-speed Internet ARPU, including equipment revenue, increased $6 in 2016 to $69 in December without any change in our pricing. In January, we have passed on the 2017 increase in programming costs with a video rate increase and dropped the regional sports channel, Root Sports Pittsburgh, due to the cost without any measurable loss of customers due to the dropping of the channel. By the end of the contract, it would have been $9 per month.

Slide 28 shows the changes in homes passed, RGUs, and penetration percentages between year-end 2015 and 2016. The addition of the heavily video-only customer base of Colane Cable, which we purchased on January 1, 2016, distorts the actual penetration trends we are experiencing.

Speaking of Colane, the upgrade of that network is coming to an end and we've started selling new services. As you recall, we had launched Internet in part of that area, but could only offer 6 megabits. We are seeing great results from our initial door-to-door sales campaign selling, in many cases, all three services and adding phone and Internet to our video-only customers.

We continue to see a decrease in regulated access lines, on slide 29, but service revenues have not decreased, due to the growth of high-speed Internet customers and, similar to the cable segment, existing customers upgrading their speeds.

Another factor in the continued growth of our wireline and cable segments is the growth of fiber revenue. On slide 30, you see the growth in affiliated and nonaffiliated revenue. The growth in affiliated revenue is a result of us continuing to build fiber to our wireless cell sites and pay ourselves, rather than outside parties. We had another record year in fiber sales, with new third-party contracts totaling $27.1 million.

I want to point out that in the last two years we have taken advantage of selling capacity on the significant fiber in our cable footprint and we see the future of fiber revenue in the cable segment to be material.

Finally, on slide 31, actual CapEx for 2016 was $183.2 million. Spending was slightly above the revised budget we provided in Q3, primarily due to the acceleration of expenditures for the nTelos network upgraded to 2016. We are projecting 2017 at $152.3 million, with 57% going towards completion of the upgrade of the nTelos network and the expansion of the coverage. The biggest item in both the cable and wireline budgets is for the expansion of our fiber network.

I would like to discuss the release we put out last week on expanding our relationship with Sprint. We've been discussing expansion with Sprint for some time. These specific areas shown on slide 32 were targeted for expansion first because they are areas where nTelos had assets but were not part of the wholesale agreement with Sprint. These areas add approximately 0.5 million additional POPs to our current 5.5 million POPs.

The agreement gives Shentel the Cumberland, Maryland, BTA, which has been a hole in our community of interest. Having Cumberland will allow us to expand coverage between Hagerstown, Maryland, and Morgantown, West Virginia. We also picked up the city of Parkersburg, West Virginia, and the areas in southeastern Ohio, with strong community of interest with Huntington, West Virginia.

We will pay Sprint $6 million at closing, which is expected to be April 1. Upon closing, approximately 20,000 customers will become Shentel affiliated customers. We plan to invest $32 million through 2020 by upgrading existing sites and adding new sites. The plan is to select the best of the Sprint and nTelos sites and then upgrade them to support 4G LTE.

Since the Sprint sites have Samsung network in those areas, we will install Nokia base stations and Sprint will redeploy the Samsung equipment. Once we have upgraded the network and have made progress in expanding coverage to be competitive, we will open Sprint stores throughout the service area. We see the expansion with Sprint as providing Shentel the opportunity to continue to have above-average growth into the future.

I will now turn the call back to Adele.

This concludes our prepared remarks. Vince, would you now review the instructions for posing a question?

(Operator Instructions). Rick Prentiss, Raymond James.

Glad to hear the updated thoughts on the 10-Q currently expected end of this week, but obviously the Board felt comfortable enough with 4Q, I guess, to do the call even before the 10-Q was out there -- or 10-K was out there.

Correct.

Okay. As we look at the numbers, the margins were definitely better than we were looking for, particularly on the wireless side, but across the board. Can you help us understand, first, the acquisition, integration, and migration costs within the quarter? And just to reaffirm, those costs are then excluded from the EBITDA, then, as an adjustment?

They are. On slide 15, you can see us adding back integration, acquisition, and migration expense of $6 million for the fourth quarter.

And so, that would be things like the handsets that we give customers to migrate them to the back office, the reconditioned handsets. It would be the shutdown of costs of the duplicate cell sites. Recall that we had 150 cell sites that overlapped with nTelos, so we are shutting those down. The cost of severance is also included there. Then we've got another $4.2 million, Rick, of temporary back-office costs. That is the billing and IT costs and customer care costs that we incur for the nTelos back office that will drop off at the point that we have all the customers migrated.

Okay. And so as we think about the margins, then, that you report with the wireless EBITDA, how should we think about those margins going forward as you add more cell sites, but then also you start selling into the new territory?

Certainly we will have additional network costs. The additional costs of selling predominantly advertising, and then whatever handsets are being subsidized, and then commission costs would be incurred.

From a macro sampling, any thoughts of where margins could go from here? Are we talking about going down noticeably as you finish the integration, and then add the sites and the sales effort? I'm just trying to think -- a lot of moving parts here to your story. So I'm just trying to think. Good results in the quarter; as we look out into the future, where might margins head? Maybe look at legacy versus the new territories.

Rick, let me take that. This is Earle. I think in the legacy area we will continue to see business as usual. There's really no change in what we are doing there. We are continuing to advertise; we are continuing to put on good gross adds. So I don't think you'll see any significant change in the margins there.

I think in the new area we will see a little bit of a step back towards the end of the year -- second half of the year as we ramp up advertising, as we ramp up distribution. You will see some impact on the margins. We'll also be bringing on the additional cell sites.

But as we continue to add customers, you will see that margin start to go back up because, as you are aware in this business, there's significant fixed costs, not as high a variable cost, and as we add customers, we will be able to move the margins up.

I think we have said, though, before that the margins in the new area are going to be lower because we have twice as many cell sites. And so, the cost of maintaining that network at the current level of customers is going to give us a lower overall margin, but as we continue to add customers over the next couple of years, those margins should move towards the legacy area margins.

Okay. And then, one of the things with the K being late and year-end reporting season anyway is we are almost done with 1Q. What can you help us understand as far as trends that you've seen in the marketplace? It is competitive out there, but any update on maybe porting ratios or I think you mentioned ARPU trends, but just kind of your thoughts on -- now that we are about 10 days away from 1Q ending, what you've seen in the quarter?

Basically, we are continuing to focus in the new area on the migrations and we have seen -- we have had a good pickup there as far as the number of customers migrating over. We have seen some elevated churn because of, as we get closer to people having to make a decision, we've seen a little bit of turn uptick there. There has been more competition, but we haven't been dramatically impacted by the -- by both Verizon and AT&T coming up with an unlimited plan. I won't say it has had no impact, but it hasn't been a dramatic impact. So I think first quarter has been a good quarter for us.

Great. I will come back if there's time at the end. Thanks.

Barry Sine, Drexel Hamilton.

Wanted to talk a little bit about prepaid. Obviously, there's a little bit of weakness going on there. And I thought it was interesting, I think you said that in your legacy areas, you actually added a small number of prepaid. If you can talk about some ongoing trends there, I think you said that the Boost brand seems to be pretty strong. Virgin hasn't been relaunched. I don't know what's going on with Assurance. And then, what are your -- what is your outlook for those prepaid brands as you get the new markets upgraded with network?

Barry, this is Earle. We actually had a net loss of 43 prepaid in the legacy areas, so it was virtually a zero.

For the last several years in the legacy area and the same strategy that we are now deploying in the new area is we have been working with Sprint and expanding the Boost distribution. Actually, even some of the nTelos stores will be rebranded as Boost stores. So we have really been pushing the Boost brand because it is the highest ARPU revenue, and we will continue to do that in both the legacy and the new areas.

Sprint has talked about rejuvenating the Virgin Mobile brand. At this point, that hasn't happened, and so we are seeing a steady deterioration of that customer base, but have been able to, in the legacy area, offset that with increases in Boost.

As we have talked about before, Sprint is getting ready to spin off the Assurance customers into a joint venture. When that happens, those customers -- we'll no longer count those customers, but we will be getting the wholesale revenue from those customers, which from a bottom line net makes that neutral.

Going forward, we are just going to continue to support the brand. What we have announced as part of the -- actually, it was lost in the shuffle, but we are actually taking over effective April 1 not only the new area in the western part of our service area, but we're also taking over the servicing of the prepaid customers -- distribution in our area. Historically, Sprint has done that nationwide. We are now going to be doing that ourselves. They are going to actually give us a rebate back of approximately $55,000 a month for us to do that.

We already have people in the market working on that today. So we will have an incremental number of folks that we'll add in order to support that, but I'm optimistic that with us taking complete control of supporting the prepaid distribution in our area we are going to be able to continue to accelerate prepaid growth.

Okay, and then a related question, still on prepaid. I think you said that you were able to shut down the nTelos prepaid third-party billing platform in the quarter. Could you give us the financial impact of that? How much did you pay, and presumably that is a pretty significant delta between fourth-quarter and third-quarter OpEx?

Yes, the contract was $0.25 million a month and we did close that down in December. And so, that is an expense that went away effective January 1.

Okay. And my last question, on the network enhancements that you are doing on the nTelos footprint. The longest timeline is obviously the new coverage cell sites and the slide -- your goal is adding 220. One of the first milestones is obviously getting zoning approval. So could you talk about -- give us a little more granularity in terms of where you are on those 220? Have you located them all and you've applied for zoning on them all or are you still in the early stages on some of these?

No, the good news, we actually started the zoning work prior to closing. So, we are coming up to close to a one-year anniversary on doing the zoning.

So the search rings have all been released. In many cases, we have identified the location. As we have always said in the past, our first objective is always to go on an existing tower, rather than to build one ourselves. But it looks like of the 220 we may end up having to build up to as many as 50, because we are getting into areas where there isn't a lot of existing coverage.

So, we'll continue to work on those and those will be the ones that will primarily get done in the 2018 timeframe. But the areas where there is the most competition are where we are focusing on building towers -- building cell sites first. And so, the 90 that we get done this year and the ones that will get done early in 2018 will be where we are getting the biggest competitive parity in our buildout.

Okay, yes. Thank you very much.

Hamed Khorsand, BWS Financial.

Hi. First off, I just wanted to start off with, are you seeing any kind of data yet from the user base as you are upgrading the network from the acquired nTelos to 4G? Are you seeing any sub changes at all as far as usage and the plans and so forth?

Well, you see how dramatic their usage has been. They are now almost using as much data as our existing customers, which I think is really kind of an indication of the capacity we added, because that was the hidden cost in this upgrade that we did is that nTelos had started their 4G upgrade, but they had only put one carrier in those sites for LTE. And so, with one carrier in a site, you're not going to get a lot of LTE usage.

And so, what we have seen is the dramatic shift from primarily data being carried on 3G in the nTelos footprint to now it's being carried on 4G. And so, as part of that, you are seeing the significant increase in data usage.

I can tell you that, anecdotally, when we look at the churn rate of the various parts of our nTelos area, the customers with the lowest churn are those who have migrated over from the nTelos back office to the Sprint back office. And I think the primary reason for that is we've had the opportunity to sit down as part of that migration process and explain to them exactly where we are and what our plans are for upgrading the network, and I think that when you look at it from that customer standpoint, they have gone from a network that was truly inferior to one that is catching up fast and they see the prospects of that getting better over the next six to nine months. And so, they have been a patient customer group and one who I expect we'll have for a long time.

Okay. And then, I just wanted to try to understand what your gains are here as far as the new deal with Sprint. You are essentially paying $38 million for basically 20,000 subscribers that, at your current ARPU rate, generate about $9 million a year in revenue and that's just on a revenue basis. So I'm just trying to understand the mathematics and the strategy as far as going out -- obligating yourself to $38 million.

Well, we are obligating ourselves to that, but we are not spending that today. We are spending $6 million at closing. And then, we will spend the $32 million over the next three years upgrading the customer base.

And when you look at our ability, these are areas where we believe we can make a significant difference. We will have a -- once we finish spending the $32 million, we will have a very competitive network and we will be able to grow the penetration to something similar to what we have in our other areas.

So, yes, if you look at it from a very short-term view, it is a fairly significant investment. But our profile has always been one where we make the investment and then we harvest the returns for our shareholder. And so, really what we are doing here is we are investing in growth for 2019, 2020, and 2021 today. And we are going to have to start spending some dollars today, but the good news is we do have those customers today that can pay the operating costs of running the network that is there today, so this is not a significant impact on the bottom line near term. And we will stage out the distribution costs and start adding customers when we have a network that is competitive.

So we are not going to start spending a lot of advertising or build a lot of stores or have a lot of distribution costs. Initially, we will maintain the existing customers. We will add some nominal number of customers in the short term, but in a year and a half from now or so when we start having a competitive -- more competitive network, that's when we will ramp up the distribution and the advertising.

And Earle, it is fair to say that we have in excess of 20% penetration in our existing markets. And we're talking about buying a market that has 500,000 POPs. So a better way to think of it might be in terms of applying our current penetration to 500,000 POPs.

No, I got that. That is exactly what I was trying to understand, so, yes, I got that. And Adele, my last question is just as far as integration expenses go, how much will that taper off in Q1 and where do you think it will zero out?

We expect that it will zero out, when all is said and done, below $100 million.

Okay, and will it just trend down from here?

It will, yes.

All right, great. Thank you.

Amy Yong, Macquarie.

-- for Amy. On this $32 million of incremental CapEx, when do you think we will see the majority of that investment? And then, secondly, I think you had just mentioned that you expect penetration to be pretty similar. Do you expect that from the get-go or do you think it will lag a little bit?

This is Earle. As far as CapEx, there won't be a significant amount spent this year, simply because we are already through the first quarter and we will primarily use this year for planning and for putting in purchase orders and looking for cell sites.

You're going to see the CapEx kick in significantly starting in 2018. I won't say it's going to be one-third, one-third, one-third. I would assume that it will be heavier in 2018 and 2019 than it will be in 2020, simply just from a CapEx standpoint.

But from a customer acquisition, you are not going to see significant customer acquisition in 2017 or 2018. It really will be 2019 before we start adding any significant number of customers because, as we've said, we really don't want to start adding customers to a network that is inferior, because they will be disappointed and they will not stick around. So we want to wait until the network is not necessarily complete. We are not going to wait until 2020, but it will be 2019 before we start seeing any significant number of customers. And it will take a number of years to get to 20% penetration. But I think you will see a nice ramp starting in 2019.

Okay, thank you.

Rick Prentiss, Raymond James.

Some follow-on on the new market areas. When you looked at those market areas, obviously you guys just have, I guess, 4% share that you are taking on there, hoping to get it up to 20%. Help us understand who the existing competitors in that marketplace -- I would assume it is fairly heavily skewed to the cellular operators, but just want to know have you looked at the market share of the competitors and who they are.

In the areas, it really kind of depends. In the Cumberland area, actually US Cellular is very strong there. They have built a significant amount of network in the northeast part of West Virginia and into Maryland. And you have the -- Verizon and AT&T are primarily focused on the interstate. If you look in the western part of the area, similar to what we see in West Virginia generally is AT&T is the strongest player in the Parkersburg and Huntington area, so they will be the primary competitor we will see going forward.

So if you think about the market share there, it's really not really T-Mobile. Sprint didn't have much, Verizon had the interstates, but AT&T and USM were your biggest competitors that you will be looking to take share from?

They are the biggest -- they are the biggest competitors there. No, T-Mobile has no network at all in the entire state of West Virginia, so -- and Sprint's network was fairly limited.

Okay. And then as you think about the margin impact on bringing on this area, how should we think about the margins at 4% penetration versus what kind of margins you could achieve when you double it to 8% and maybe get up to 15%? I'm just trying to think of the progress on the margins. Obviously, it will take a couple years to get the network built and then to start ramping, but I'm just trying to think of the margin progression.

I think, as I said earlier, day one basically we are collecting enough revenue to pay for the cost of running the network. So there's really not a lot of margin there. So for the next couple of years, the margins will be basically zero, could be even a little negative as we start bringing on sites without the corresponding customers.

But I think that's when you see the turnaround starting in 2019 where we can -- the good news, it won't take a lot to double it as far as the penetration, but that's when you are going to see the uptick.

And the fact is that because -- unlike nTelos, I think that the big difference between this and the nTelos is this is a very controlled environment. It is a limited number of customers, an area where we can go in and we are not having to re-change the brand; we are not having to move anybody to a new back office. These are Sprint customers that are already on the Sprint back office. And so, I think we can control the process without a lot of excess expenses and we can migrate over to a smooth operating environment without taking a significant hit on the bottom line.

As far as the improved margins, this is not a big enough area that it's going to have a huge impact on the consolidated. But I think you will start seeing some positive impact, probably in -- probably it will take until 2020 before you will see any real significant impact.

Okay. And did you say how many cell sites are going into the area?

It will be roughly 100, either ones that we have to upgrade or additional sites. So it's a fairly significant build over a three-year period.

Okay, last one for me, then. You mentioned that this was an area that was targeted first because it was nTelos territory, but it wasn't part of the affiliate or maybe even the wholesale agreement. That suggests that there's more in the wings.

What is your appetite? How do you pace it? How do you get Sprint's attention as we go into the post-broadcast auction timeline? So just trying to think of how many of these could we see over what point in time or is this kind of a similar size that we might expect?

As we've said before, from our standpoint, the closer we can get our network to Sprint's major metropolitan areas, the better it is for both parties because probably the weakest coverage areas that we have right today are the areas just outside of our service areas. So, we are looking to have discussions with Sprint to expand our coverage really in both East and North, primarily. And I think that you -- there is a very good prospect you will see some additional areas like this.

As far as the size of them, the good news is that that is kind of in our control. We can decide, and have had this discussion with Sprint, that we do have an appetite to continue to grow our footprint, but we want to do it in a way that we know we can manage it well. We can finance it without putting any burden on the Company, and also to be able to utilize our resources without burdening the resources.

So I think you can see a steady stream of these over the next couple of years. And we will, as we have in this one, pace them so that we will be doing the construction over a couple of year period. Obviously, the quicker we can get it done, there is some advantages. But on the other hand, if you look at it from Sprint's point of view, these are not areas they probably would have spent any capital. So whatever capital we spend on whatever timeframe is positive for both sides.

Great. That helps a lot. Thanks, Earle.

That concludes our Q&A. Thank you for participating. I would like to invite you to let me know if there are additional details you would like us to provide in future calls. My contact information was provided in the press release.

And, again, ladies and gentlemen, that does conclude today's conference. We thank you for your participation. You may now disconnect.