This conference call transcript was computer generated and almost certianly contains errors. This transcript is provided for information purposes only.EarningsCall, LLC makes no representation about the accuracy of the aforementioned transcript, and you are cautioned not to place undue reliance on the information provided by the transcript.
Operator
Good morning and welcome to the Crown Castle fourth quarter 2022 earnings conference call. All participants will be in listen-only mode. Should you need assistance, please signal a conference specialist by pressing the star key followed by zero. After today's presentation, there will be an opportunity to ask questions. To ask a question, you may press star then one on your touchtone phone. To withdraw from the question queue, please press star then two. Please note this event is being recorded. I would now like to turn the conference over to Ben Lowe, Senior Vice President. Please go ahead.
Ben Lowe
Thank you, Kate, and good morning, everyone. Thank you for joining us today as we discuss our fourth quarter 2022 results. With me on the call this morning are Jay Brown, Crown Castle's Chief Executive Officer, and Dan Schlanger, Crown Castle's Chief Financial Officer. To aid the discussion, we have posted supplemental materials in the investor section of our website at crowncastle.com that will be referenced throughout the call this morning. This conference call will contain forward-looking statements which are subject to certain risks, uncertainties, and assumptions, and the actual results may vary materially from those expected. Information about potential factors which could affect our results is available in the press release and the risk factor sections of the company's SEC filings. Our statements are made as of today, January 26, 2023, and we assume no obligations to update any forward-looking statements. In addition, today's call includes discussions of certain non-GAAP financial measures. Tables reconciling these non-GAAP financial measures are available in the supplemental information package in the investor section of the company's website at crowncastle.com. So with that, let me turn the call over to Jay.
Kate
Thanks, Ben, and thank you, everyone, for joining us on the call this morning. As you saw from our results, 2022 was another successful year for Crown Castle, and the positive trends across our business remain intact. With fourth quarter 2022 results coming in as we expected and no changes to our 2023 outlook, I plan to keep my prepared remarks brief before handing it over to Dan to talk through the numbers in a bit more detail. As I reflect on 2022, I'm proud of what our team accomplished. We led the industry again with nearly 6.5% organic tower revenue growth as our customers upgraded existing tower sites with additional spectrum and added equipment to thousands of tower sites they were not previously on to support nationwide deployment of 5G. And we deployed 5,000 small cells to support initial network densification efforts while growing our fiber solutions revenue by 2%. The positive operating trends in 2022 exceeded our initial expectations for the year and offset the impact of the rapid increase in interest rates. demonstrating the resilience of our business model and strategy. As a result, we were able to deliver strong bottom-line growth that supported more than 9% dividend per share growth. As we discussed when we initially provided guidance in October, we believe the positive operating momentum will carry into 2023, driving another year of expected strong growth with 5% organic growth in towers and a doubling of our small cell deployments to 10,000 nodes, With respect to tower leasing trends, the established national wireless operators are deploying mid-band spectrum in earnest as a part of the initial phase of their 5G build-out. To date, only about half of our sites across our top three customers have been upgraded with mid-band spectrum, providing a significant opportunity for additional revenue growth as additional sites are upgraded over time before their focus will likely shift to more infill with new co-locations. Adding to the substantial long-term growth opportunity, we continue to support DISH with their nationwide build-out of a new wireless network, and I believe we are in a great position to continue to capture an outsized share of that opportunity. Turning to small cells, we expect to double the rate of small cell deployments this year to 10,000 nodes, with over half co-located on existing fiber to meet the growing demand from our customers as 5G networks will require small cells at scale. With approximately 60,000 nodes on air and another 60,000 contracted in our backlog, I believe 2023 will represent the first year in a sustained acceleration of growth for our small cell business. We also continue to see opportunities to add to the returns we are generating from small cells by leveraging the same shared fiber assets to pursue profitable fiber solutions growth. And we expect to return to 3% growth as we exit 2023. Looking at the bigger picture beyond this year and why I'm so excited about our growth opportunity, we are still in the early innings with 5G as the industry is only a couple years into what we expect will be a decade-long growth opportunity. Our customers are seeing significantly higher levels of monthly data consumption as consumers upgrade to 5G, providing the need for significant network investment for years to come to keep pace with this persistent growth in mobile data demand. As we have seen in our industry throughout its history, generational upgrades to the wireless network occur in phases, with an initial push to provide nationwide coverage, followed by periods of continued network augmentation and densification that has led to long periods of sustained growth. We believe we are in the initial phase of the 5G build-out, with many phases to follow over the coming years. Consistent with their past practice, we believe our customers will first deploy their spectrum on the majority of their existing sites, as they are currently doing, before shifting their focus to cell site densification to get the most out of their spectrum assets by reusing it over shorter and shorter distances. The nature of wireless networks requires that cell site densification will continue as the density of data demand grows. and we expect 5G densification to require both towers and small cells at scale to fill in the network. With that view in mind, we've invested more than $40 billion of capital today in towers and more recently small cells and fiber that are mission critical for wireless networks to capture as much of this growth opportunity as possible. Importantly, we are already generating a 10% return on our total invested capital with the opportunity to increase that return over time as we add customers to our tower and fiber assets and grow our cash flow. As a result, I believe Crown Castle is an excellent investment that will generate compelling returns by providing investors with access to the most exposure to the development of next generation networks in the U.S. with our comprehensive offering of towers, small cells, and fiber, providing the opportunity to benefit from the best growth and lowest risk market, an attractive total return profile with a current yield of 4% and a long-term annual dividend per share growth target of 7% to 8%, and the development of attractive new assets that we believe will extend our runway of growth and create shareholder value. And with that, I'll turn the call over to Dan.
Ben
Thanks, Jay, and good morning, everyone. We generated another year of solid growth in 2022, and we expect the strong operating trends across our business to continue as we see a long runway of 5G investment in the U.S. The elevated leasing activity across our customers contributed to another year of industry-leading tower revenue growth in 2022 of nearly 6.5%, and to 9% growth in our annual dividends per share. Before discussing the 2022 results and 2023 outlook, I want to draw your attention to some enhancements we made this quarter to the disclosure in our supplemental information package. In response to feedback we've heard from our investors, we provided organic billings growth detail by line of business for towers, small cells, and fiber solutions to help investors better understand the composition of organic growth trends. This enhanced disclosure includes historical organic growth information going back to 2019. In addition to expanding our disclosure, We also reorganized the supplemental information package, in many cases by line of business, to make it easier for readers to follow. We hope you find this additional information in the new layout to be helpful. Now, turning to the full year 2022 financial results on slide four of our earnings presentation, site rental revenues increased 10 percent, adjusted EBITDA growth was 14 percent, and AFFO increased by 6 percent for the year. The 10 percent growth in site rental revenues included 5% growth in organic contribution to site rental buildings, consisting of nearly 6.5% growth from towers, more than 5% growth in small cells, and 2% growth in fiber solutions. Turning to page 5, our full-year 2023 outlook remains unchanged and includes site rental revenue growth of 4%, adjusted EBITDA growth of 3%, and AFFO growth of 4%. We also expect organic billings growth of approximately 4% when adjusted for the impact of the previously disclosed Sprint cancellations. The 4% consolidated organic growth consists of 5% growth in towers, 8% growth in small cells, and flat revenue in fiber solutions. As we discussed last quarter, we expect the rationalization of a portion of Sprint's legacy network to result in some movements in our financial results that are not typical for our business. Our expectations for non-renewals and accelerated payments associated with this network rationalization activity are unchanged, with approximately $30 million of new non-renewals and $160 to $170 million of accelerated payments during 2023. We expect the majority of the non-renewals to occur in the first quarter and therefore impact year-over-year billings growth in each quarter this year. We expect the accelerated payments associated with this decommissioning activity and related services work to be concentrated in the second quarter. As a result, we expect the second quarter to represent the high watermark for adjusted EBITDA and AFFL in 2023. Turning to financing activities, we finished 2022 with leverage in line with our target of approximately five times net debt to adjusted EBITDA. For full year 2023, our discretionary CapEx outlook is also unchanged, with gross CapEx of $1.4 to $1.5 billion, or approximately a billion dollars net of expected prepaid rent. Based on our current backlog of small sales that includes a significant mix of co-location notes, which have higher returns and require less capital relative to anchor bills, we expect to be able to finance our discretionary capital with debt while we're maintaining our investment grade credit profile. Earlier this month, we added to our strong balance sheet position when we issued a billion dollars in senior unsecured notes with a 5% coupon. to term out borrowings under our revolving credit facility. Following this financing transaction, we have more than 85% fixed rate debt, a weighted average maturity of over eight years, limited maturities through 2024, and approximately $5.5 billion in available liquidity under our revolving credit facility. So to wrap up, we are excited about the strength of our business and our ability to execute on our strategy to deliver the highest risk adjusted returns for our shareholders by growing our dividend over the long term and investing in assets that will help drive future growth. We have delivered 9% compound annual dividend per share growth since we established our 7% to 8% dividend per share growth target in 2017. And I believe that we are positioned well to return to 7% to 8% dividend per share growth as we move beyond the Sprint decommissioning impacts in 2025. With that, Kate, I'd like to open the call to questions.
Operator
We will now begin the question and answer session. To ask a question, you may press star, then 1 on your touch-tone phone. If you are using a speakerphone, please pick up your handset before pressing the key. To withdraw from the question queue, please press star, then 2. The first question is from Michael Rollins of Citi. Please go ahead.
Michael Rollins
Thanks, and good morning. I also just want to say thank you for the additional disclosures. Very helpful. Two topics, if I could. First, on small cells. If you could give a little more color on the small cell leasing that you experienced during the fourth quarter. And then when you look at the backlog and consider the typical 18 to 36-month cycle that you described to install a small cell for your customers, what's the opportunity to further accelerate that 10,000 deployment pace into 24 and 2025? Thanks. Thanks.
Kate
Good morning, Mike. Thanks for the comment. On the first question around leasing, as you noted, and I made a mention of this in my prepared remarks, we did increase the total number of nodes on air and under contract by 5,000 during the fourth quarter. So we didn't sign any large deals with customers, but this was just ongoing activity that represents additional commitment for nodes beyond the large commitments that we previously announced. And to my comments around cell site densification, we believe we're going to continue to see this throughout the 5G cycle of upgrades and deployments and beyond as the carriers move past touching the sites, the tower sites that they're on, and starting to look at densification of their network. And I think the activity that we saw in the fourth quarter is representative of exactly those longer-term plans, which ties really closely to your second question around the backlog and the timeline, I think we see and have visibility to what they're going to need in their network, particularly in small cells, 24 to 36 months in advance of when these nodes will actually be put on air. And as I mentioned in my comments, we think the acceleration that we're seeing in 2023, doubling the number of nodes that we expect to put on air from 2022 to 10,000, we think that's the start of an acceleration of growth and deployment of small cells. I'm not really ready to give guidance on how many we'll put on in 24 and 25, but given the backlog and the timing, we do think this is the start of an acceleration of growth in that business.
Mike
Thanks.
Operator
The next question is from David Barden of Bank of America. Please go ahead.
David Barden
Hey, guys. Good morning. Thanks so much for taking the questions. I guess along related lines, So two questions on this small-scale topic, I guess, Jay. One is, you know, you've got these large-scale relationships on small cells, and they say visibility on these next two to three years. There are some carriers that you don't have these relationships with, and I was wondering if you could elaborate a little bit on why you think that is. Is that because, you know, counterparty plans aren't as evolved? They might be less evolved in terms of the total size of their network bill. They're not ready for densification. Or is it the other way around, which is that they've chosen to go the self-perform route, and is that impacting the market opportunity that you foresee? I guess the second question would be related to, with respect to what you do have in the backlog, how do we think about how you're budgeting for the upfront CapEx contribution portion of that? Is it a constant drumbeat that's already known? Is it going to be on a case-by-case basis? Obviously, relevant to the cash flows and how we think about the runoff of prepaid rent amortization over the coming years. Thank you so much.
Kate
Good morning, Dave. On your first question around the agreements with the carriers, I think you're going to see over time in that business a combination of large-scale agreements with the carriers where whether they give us whole markets or a number of markets, we'll probably see large-scale agreements with carriers over time. We expect those large-scale agreements to be lumpy, so we wouldn't expect to see them every quarter or even every year, but it will be dependent upon the way that the carriers are thinking about the network and where the holes are and where the need is. I think there's value in some of those large scale agreements, particularly with respect to securing the resources when concentrated in a few markets. So I think you'll continue to see those. I also think you'll continue to see what we saw in the fourth quarter where carriers give us business and they're not as large scale and we do them in smaller chunks. So I think I think both are going to be there. I don't think one or the other is representative of an underlying trend or nature of the business. The carriers, I think, as I say, will contract with us, I think, on both basis. And I think over time, big picture, and we've talked about this for years, because of our disciplined and rigorous approach to capital allocation and our view of where small cells are going to be needed, everywhere that small cells are needed is not necessarily attractive for us to put capital to work. So we're going to pick and choose where we decide to put capital to work, which means the carriers are either going to need to find other providers who are willing to deploy the capital at lower return thresholds than what we're comfortable with, or alternatively, they'll self-perform. And today, we have seen the carriers self-perform for the most part in places where we were not willing to do it. And I think that will continue. You'll continue to see self-perform for the carriers. Again, I don't think that's indicative of the business. I think it's more indicative of the way that we think about return thresholds and our desire to both grow the dividend, elongate the timeline of returns, and be thoughtful about the risks that we underwrite. On the second question around the backlog and carrier CapEx contribution, The way the agreements are structured relates to the – pricing is related to the return based on the underlying cost to deploy nodes. So the CapEx will move – the CapEx contributions from the carriers will move in unison with the way that it's – the cost of actually deploying the networks that we're deploying in the locations that we're deploying them. In a market where they're more expensive to deploy, the capital contribution is going to be higher. In places where it's less expensive to deploy, those capital contributions will be less. I don't think at this point, you know, beyond the guidance that we've given around CapEx for calendar year 23, we're not going to provide guidance for 24 and 25. But as we give outlook for each individual year, we'll give you a view of what's the backlog, what do we think we're going to turn on air next, and then break out for you what we think our total CapEx for the year is going to be, and then what portion of that will be carrier CapEx, carrier contribution offsetting that outlay.
David Barden
And then, Jay, just – and maybe this is a question for Dan, but just to follow up real quick on how the mechanics work. So the CapEx comes in during this 24- to 36-month period. You recognize that contribution in CapEx, but – You don't start amortizing that contribution until after the lease begins. And so there's a window between where you've got the money and then where you start amortizing it through the income statement. Is that right or wrong?
Kate
You've articulated it correctly. So we will receive cash as we go through the process of deploying the nodes and incurring capital expenditures. And then we would start to run it through the income statement once we've completed the work of the operational work necessary to complete and deliver the node to the carrier. At that point, we would then amortize it over the term of the lease. One thing I would say, and I'm not sure exactly what you're trying to decipher in terms of this question. You've articulated it correctly. Maybe one additional piece of information that's helpful. The backlog in the timeline to build when we talk about 24 to 36 months to build is our average. Obviously, there are notes that take longer than that. The actual construction period of time is relatively short, and it occurs at the back half, the last portion of that long-dated period of time. the majority of the costs that we incur up until construction are soft costs, and they would be a smaller percentage of the overall CapEx. So you shouldn't expect, just to be extreme, you shouldn't expect on a 36-month timeline to construct a node that we would have significant CapEx in the first 12 months of that and then receive carrier contributions at that time. Most of the actual outlays would be towards the back half of that process, and the cash being received. So, the timeline between receipt of cash and booking the node is not 36 months, or not likely to be.
David Barden
Right. Helpful. Thank you so much, Jay.
Operator
The next question is from Simon Flannery of Morgan Stanley. Please go ahead.
spk00
Great. Thank you very much. Jay, thank you for the color on the phases of densification. That was helpful. Perhaps you could just help us with this transition as Verizon and AT Mobile wind down their 5G builds. Is it normal that we have like a pause and digest on the macro side, or do they go straight into cell site densification on the macro side? Have you got any color on what the carriers are starting to think about once they, particularly in the urban areas and suburban areas, have already put up the antennas? And then just a related point, any comments, any updated thoughts on M&A? Obviously, it's been a long time since you've done anything in organic of scale, but there's always market opportunities out there. So just love to get your latest thoughts on that. Sure.
Kate
On your first question, I don't want to comment specifically on Verizon and T-Mobile. I'll let them comment on their longer-term network plans. You know, as I mentioned in my comments, We think about half of our sites have been touched for the mid band spectrum at this point. They're obviously across the board. All of the carriers are working on touching the vast majority of the rest of those, and that will take some period of time in order for that to be accomplished. We're two to three years into the work that's been done to date, and it's taken about that long just to touch half the sites with just the mid band spectrum. So I think you're going to see other spectrum bands that are going to be deployed for 5G on existing sites. as well as the completion of the mid-band spectrum across the balance of the sites. Each of the carriers will think about how they deploy capital, how they budget that capital a little bit differently. But those offsets generally over a long period of time in the tower business have mostly offset each other to the point where we just haven't seen a lot of movements and a lot of movements up and down in terms of the overall capex. And I think we'll see in spending and focus on network deployment. I think we'll see a similar thing during 5G thus far, as I pointed to in my comments. Thus far, 5G has looked relatively similar in terms of its deployment activity as what we saw when 2.5G, 3G, 4G were deployed, where the carriers focused on upgrading the sites that they were already on, And then the discussions start to move towards the second half of it and start to think about infill sites. I think what's different about 5G that we're seeing, obviously, in our small cell businesses, as alluded to previously on the increase in the number of nodes that we signed, as well as the larger transactions that we announced previously, those infills are going to come from a combination of both tower sites and small cells. I think the unique thing about 5G, we saw a little bit of this at the end of 4G, but the unique thing about 5G is the necessity in those in-fill sites to use both towers and small cells. And we're starting to see the real beginnings of that as we start to accelerate. So similar to past and excited about our forward growth, excited about our 5% plus organic tower growth this year. We think there's a long runway of continuing at that level of north of 5% growth in tower. And on the M&A? Yeah, on the M&A side, no change to the comments that we've made historically. We're focused on making sure we deploy capital at very high returns that increase the dividend and elongate our opportunity for growth. We've chosen, based on the opportunities and price sets that have been in front of us and assets that we have looked at, we've made the decision that the best opportunity has been to invest in assets that we're building. But we think about acquisitions the same way we think about CapEx. We look at it as what's the best alternative for that use of capital. And thus far, we think the best opportunity at scale for the use of capital has been to deploy assets fiber and small cells. And so we'll continue to look and we would be open to it if we found an asset that met our return criteria for what I articulated previously of growing the dividend and elongating the growth rate. Could be interesting to us, but really excited about the opportunity for us to continue to invest capital to deploy small cells. Thanks a lot.
Operator
The next question is from Philip Cusick of JP Morgan. Please go ahead.
Philip Cusick
Hi, guys. Thanks. How are you? Morning. Thanks for your time. You know, just to follow up on some of the small cell stuff.
Brandon Nistel
1057 AM.
Philip Cusick
How many do you think will be upgraded versus about half are done today? And anything shifting in the small cell expansion? mix of overlay of 5G versus new locations, and when do you think infill should start to ramp? Thank you.
Kate
Yeah, on the tower side, I think we will see the second half of the other half of the towers be upgraded to mid-band spectrum Took about two to three years to do the first half, so I think that's probably a reasonable assumption that it will take that long to do the second half of the assets roughly on the tower side. I think the announcements that we've made previously with small cell commitments to be constructed are already a combination of overlaying on nodes that they were previously on with other technologies and upgrading those to technologies as well as infilling sites along the same fiber, increasing the number of nodes per mile, if you will, in a given geography, we're already seeing infill and densification on that front. We've talked about in the two large announcements that we made with the T-Mobile nodes, the committed T-Mobile nodes, that the vast majority of those were co-located on existing fibers, so those largely represent upgrades and densification, and then a mix of about 50-50 on the Verizon nodes, a combination of upgrades, co-location on existing fiber, and then the other component would be where we're building new sites and new locations. So I think we'll continue to see a mix. As we said in our comments, for 2023, we think the vast majority of the nodes that will turn on air will be on existing fiber.
Philip Cusick
Thanks, guys. And one more, if I can, on services. Talk about the current pace. Is this sort of a normal level, do you think, or are there particular projects driving the strength? Thank you.
Ben
No, it's a pretty normal pace, Bill. And the only thing I would point out is something I said in my prepared remarks, that the second quarter will likely be the high-water mark because we have some decommissioning work that comes with services activity that will hit in the second quarter, but a relatively normal pace where we are today.
Mike
That's helpful. Thank you.
Operator
The next question is from Rick Prentice of Raymond James. Please go ahead.
Rick Prentice
Thanks. Good morning, everybody. Good morning, Rick. Hey, first, echo Mike's comments. Thanks so much for the extra detail on the segments, but also hope you and the team are okay with all the weather issues in Houston.
Kate
We're doing well. Came through the storm well, but there were a lot of damage in the city. Great to hear you're okay.
Rick Prentice
I want to follow up on the small cell side, obviously a common theme today. I think previously you had said the $30 million Sprint cancellation churn item was maybe $20 million small cells, $10 million fiber. How many nodes should we expect that $20 million equates to? And then was it a total of 5,000 nodes they were going to turn off over a multiple-year period, and is that still the case?
Kate
Yes, thanks. You're correct. We did say there's about $40 million of churn that we expect in our fiber segment and split about equally between small cells and fiber solutions in calendar year 23. The churn expected, you correctly stated, of about 5,000, we expect about half of those to churn in calendar year 2023. the balance would be in 24 and beyond.
Rick Prentice
Okay. Now that we have the extra details, it seems like there's maybe a normal level of churn within small cell. What should we think that is? Is that kind of like a 1% to 2% normal level of churn in small cell that we should be baking into our long-term forecast?
Kate
Yeah, I think that's probably right, somewhere in the neighborhood of 1% to 2%. I mean, honestly, to date, we really have not seen hardly any churn in that business, except for the event of the consolidation of Sprint into T-Mobile. Churn has been near zero, very low, other than that event. But I think long-term, it probably plays itself out like towers. So if you're thinking about your long-term model, assuming churn of 1% to 2% is probably right.
Rick Prentice
The final question for me is, You mentioned small cell, look for profitable fiber solution items. T-Mobile has talked about for their high-speed Internet project, they might move beyond fixed wireless access and consider buying capacity of fiber from other people. Is that a type of profitable business in the areas where you've been deploying fiber and small cells at the T-Mobile might be an interesting return case?
Kate
It's possible. A good portion of our fiber business is leases that we have with the carriers where they use our fiber. So depending on the locations that T-Mobile were to desire, then our assets could be very attractive for that. But it's a case-by-case, location-by-location analysis that would have to be done. Okay. Very good. Thanks. Stay well. You too.
Operator
The next question is from Brandon Nistel of KeyBank Capital Markets. Please go ahead.
Brandon Nistel
Hey, guys. Thanks for taking the questions and appreciate the disclosures as well. I was hoping you could talk us through the run rate in terms of tower core leasing throughout 23, is first half any stronger than second half? And maybe the other way to ask the question is, As you look at the backlogs of new lease applications that you're receiving today, are those trending up or down at this point? Thanks.
Ben
Yeah, thanks, Brandon. The run rate tower leasing is relatively flat through the whole year. There may be a little bit of a skew towards the front half, but it's not anything I would say is going to impact the numbers very much at all. And that would imply that the applications are relatively flat as well. So if you're trying to figure out how to model it or how to think through the activity levels and the leasing in 2023, I'd say it's pretty even quarter to quarter.
Kate
Your second question on the trends we're seeing in the backlog, no change in what we're seeing from what we talked about in October, so seeing good demand across All three of our business lines, tower, small cells, and fiber solutions. The pipeline, you heard my comments and my prepared remarks, but we think by the back half of this year, we're going to exit 2023 with fiber solutions back at kind of a 3% growth area. And tower leasing, as Dan just mentioned, we think that's going to be – really similar across the year, so not back half loaded, but level loaded across the year. And then small cells, you know, we obviously had a good fourth quarter in 22, and we'll see what builds over the course of 23 and update you as we get orders on that front.
Mike
Thank you.
Operator
The next question is from Greg Williams of Collins. Please go ahead.
Greg Williams
Great, thanks. Just wanted to touch on the M&A comments you mentioned in your choose to build rather than buy per se. Does that imply that private fiber multiples just remain stubbornly high, and are they coming down to any degree, and do you foresee them coming down in the next few quarters? Second question is just actually on cable. We've been fairly dismissive that they're going to be touching the towers per se, but more on CBRS deployment on their own air strands. But, you know, there's possible conversations of cable getting more aggressive and a more macro facilities-based network. And just wanted to see here if you had any updates on conversations with cable. Thanks.
Kate
Sure. On your first question, I would say it's On some level, it's probably a function of price. It's more likely a function of our targeted approach to which assets we want to own. In order for a fiber asset acquisition to be attractive to us, it needs to be in dense urban areas, it needs to have high fiber strand count, and we need to have visibility that those areas are likely to have significant lease-up for small cells. And the reason why you haven't seen us do any fiber acquisitions in the last five years is much more related to the fact that we haven't seen anything meet those criteria than, frankly, it is price. We just haven't seen the opportunity to acquire assets that meet the criteria that's going to drive long-term growth from the wireless carriers, from the deployment of small cells. And we're going to remain disciplined on that front, continue to believe the vast majority of the fiber that we will accumulate over time will be as a result of building it rather than acquiring it because we just don't see a lot of assets in the market that meet our criteria for assets that we would want to own. On your second question, I believe that cable over the long term is a very attractive opportunity for us to increase our growth rates and think that we will see leasing from the cable operators. We'll see some of that on macro sites. Frankly, I think we'll probably see more of that in our small cell business, given the places that they tend to deploy infrastructure as they think about the density of population and users. So I think it's more likely that we will benefit from the deployment of network by the cable operators using their various spectrum bands. We're more likely, I think, to see that in small cells over a long period of time than we are in macro sites. but do think macro sites will benefit from cable. Great. Thank you.
Operator
The next question is from Nick Del Dio of SVB Moffett Nathanson. Please go ahead.
Nick Del Dio
Hey, good morning, guys. Jay, you noted that about half the sites on your towers have been upgraded with mid-band 5G spectrum. Very interesting statistics, so thanks for sharing that. I guess, do you see any meaningful differences in that percentage between more urban towers versus suburban towers versus more rural towers? Or is it reasonably consistent across the board? And I guess maybe to build on some of the previous comments you've made, what does your work suggest with respect to how high that number needs to get before a carrier starts to pivot more noticeably towards co-locations?
Kate
Sure. On the location point, I probably would not draw a lot of distinction between central business districts and urban, more suburban areas that are densely populated. The usage, as you think about it on a per-subscriber basis, while there is some differences, over time there has been less of a differentiation by the consumer in terms of usage. and therefore the network reflects that. So when we see the carriers deploy the first phase, there are a mix of dense suburban markets as well as what you would think of as the most dense part of markets down in the central business district. We'll see leasing in both of those kind of areas initially. Less likely for us to see that leasing in the early phases of deployment of a generational change less likely to see that in more rural applications or rural assets. So we have not seen that as much. To the second part of your question, how far do they need to go before they start to do infill, some of it is a matter of how they allocate their capital and the need to provide geographic coverage. The other part of the answer is where do they see traffic growth And where are there holes in the network that they need to do infill in order to improve the network? And there's not a big picture answer, really, frankly, to give to that question. It varies market by market. And so in some markets, two or three years ago, at the very beginning of the launch of 5G, they already knew where they were going to need infill sites. We started seeing small cells and demand for small cells as they started to think about infill. long before even devices were out and usage had started to increase. So they had a view based on population usage, their customer base, et cetera, that there was going to need to be an infill of sites in certain geographies. And so in other geographies, we haven't seen that yet. So it really is a – it's not a big-picture question that I could give you an answer that would be helpful. But it's driven by underlying data usage, and that – It's why so often in my prepared remarks, I talk about what we're seeing in terms of data usage, what the carriers are seeing in terms of data usage, because that long-term is the driver of the need of our infrastructure, and we believe that macro trend is very healthy and will continue, and we'll continue to see the need for both macro sites and small cells from an infill standpoint. Okay, okay, that's helpful.
Nick Del Dio
And then maybe one more if I can. Just thinking back, it's been a little over two years since you struck your deal with DISH. It was a new and unique structure on the tower side. It had a fiber component to it. Now that it's been enforced for some time, I was just wondering if you could comment on your satisfaction with that novel structure you chose and the degree to which it's accomplishing what you hope to accomplish.
Kate
Well, first, we're doing everything we can to help DISH get launched, and we've got teams of people that are very focused on that. DISH has been working hard to get their nationwide network deployed, and I hope that they would say about us we've been a good partner to help them get there. I know our teams are focused on it 24-7, 365, so it's been a good partnership with them, and we're happy to have them as a customer. It has accomplished what we expected. We expected that the fact that we got them locked up first would mean that our network would benefit from them designing their network around our existing assets. We've seen that play out. We believe we've gotten an outsized share of the overall opportunity as they deploy the network. And I think the nature of our agreement with them, we have the opportunity to continue to get an outsized share of their network deployment. So I think in that respect, it has accomplished exactly what we had hoped. And as we said at the time, and I think this has played out, our visibility to that network with the combination of providing fiber as a part of their backbone, as well as providing tower sites, has deepened that relationship and deepened our understanding of how they're thinking about the deployment of the network and probably led to us being able to capture more opportunities than we would have had if we were towers only. Okay. That's great. Thank you, Jay.
Operator
The next question is from Jonathan Atkin of RBC Capital Markets. Please go ahead.
Jonathan Atkin
Thanks. So you mentioned half of your tower sites have been upgraded with mid-band. And I think it's implicit in the answer to one of the earlier questions, but you expect that to get to 100%? Or is there an end state that's maybe a little less than 100% in terms of portion of sites that have 5G mid-band?
Kate
Good morning, John. I don't know that it'll get all the way to exactly 100%, but we would expect over time it will get pretty close to about 100% of the network will be upgraded, yes.
Jonathan Atkin
And then did you see on just the towers, did you give a split between co-loads versus amendments? And if not, can you tell us what that number was and then maybe what you anticipate that mix to be, again, just in your tower portfolio toward the end of this year?
Kate
Yeah. Consistent with my comments around the vast majority of the activity is on sites that they're already on, the vast majority of the total activity that we're seeing from the carriers is amendments to existing sites where they're adding additional equipment and therefore we're getting rent added to the leases that we already had. We are seeing some first-time installations on sites as a part of their desire to infill, but the vast majority of the activity that we're seeing would be from amendments.
Jonathan Atkin
So no pivot from 3Q into 4Q in terms of that mixed shift? I know it's majority amendments, but no noticeable change?
Kate
No, we haven't seen any change from kind of middle of last year. And certainly as we're sitting here today in early 2023, haven't seen any change from our expectations when we laid them out in October of last year.
Jonathan Atkin
And then lastly, in the fiber business, a lot of non-mobile tenants, a lot of products that you list around wavelengths, Ethernet, Spark Fiber, managed services. And just given the growth that you're seeing there and the mix of revenues that it represents, any way to give us a little bit of color as to what the different demand drivers that you're seeing are? which products are maybe getting more traction on, say, lit services versus others?
Kate
Yeah, the biggest driver there is the increased traffic, overall data traffic that's happening in the market. I mean, our business and our focus for customers is mostly large enterprise, universities, hospitals, and in those markets, the primary driver of what drives our revenue growth is data traffic and the movement of, in essence, the movement of data among their facilities, locations, offices, et cetera. And that's the biggest corollary.
Jonathan Atkin
Lastly, just any update on Edge, and it's been a number of years since you announced the Vapor.io investment, and any updates on your thoughts as it pertains to those sorts of opportunities and traction gained thus far?
Kate
Sure, we continue to be optimistic about the long-term opportunities around edge. I feel like our assets are really well positioned to capture that opportunity. In order for edge to work, you've got to have connectivity and you've got to have power. And our hub sites for small cells and our towers are both ideal locations for aggregating the traffic out of mobile networks at the edge and think that opportunity will develop as 5G ultimately develops. I think you've probably heard us say a number of times that the first benefit from the activities that will ultimately lead to benefit around the edge, we think we'll see in spades in the deployment of small cells and a lot of activity related to it with small cells. And then the follow-on will be the opportunity around the edge. So we certainly believe it's there and think we're really well positioned to capture that opportunity when it does materialize. as the applications that need increased data and compute power move to the very edge of mobile networks. When those applications are starting to be used both on an industrial level as well as a consumer level, I feel like our assets are really well positioned to capture that opportunity.
Mike
Thank you.
Operator
And the final question is from Jonathan Chaplin of New Street Research.
Jonathan Chaplin
Thanks for taking the question, guys. Just a quick housekeeping question first on small cells. So it's a backlog of 60,000 and it takes 24 to 36 months to go through the construction process. Does that suggest like a very material acceleration from the 10,000 that you're going to do this year and 24 and 25? Like you're going to get through 60,000 over the course of the next three years?
Kate
Now, we didn't provide specific guidance on when we'll get that done, but as I mentioned in my comments, we think that 23 is the start of an acceleration of growth in small cells. So it does imply that there will be an acceleration beyond the 10,000 nodes per year that we expect to do in calendar year 23.
Ben
But I would caution you not to expect that because we have those in our backlog right now, that 24 months from now or 36 months from now, all of them will be built. It is an average, and so I would not just make the leap that after 2023 there's 50,000 left, which means that we have to do 25,000 in each of 24 and 25. That's not the way the business rolls out. It's an average, and it takes some time in order even to get into that average as we go back and forth with our customers to cite the actual small cell nodes where they need to be built. So as Jay said, we believe there can be an acceleration, but I would caution you against expecting it's going to jump to 25,000 nodes each year of 24 and 25.
Jonathan Chaplin
Can you give me just a little bit more color on why it isn't sort of complete within three years? Is it that the carriers are really looking sort of four or five years out in terms of what they're contracting for small cells today?
Kate
Sure. When we did the two agreements with Verizon and T-Mobile, where they made large commitments Those were multi-year commitments. So the expectation was that they would identify the nodes while we had ideas on what markets they were going to use. The actual location of the nodes goes through an identification process over time. And so we did not expect in that backlog, while it's committed, contractually committed, and the rent will be there, it doesn't speak to, as Dan was saying, that's why you can't take the backlog and say, okay, all of that backlog will be completed in 24 or 36 months.
Jonathan Chaplin
Got it. And then is there a way to contextualize what DISH is contributing to growth at the moment? And have they reached a steady state with you at this point, or do you think that their contribution could still accelerate for you?
Kate
Well, I think we'll continue to see the benefit of DISH deploying their network, but being really specific with the number of sites and their percentage contribution, we've We do our very best to stay away from giving that level of specificity among any of our customers and their network and just let them speak to the number of sites and where they are in their deployment cycle.
Jonathan Chaplin
Okay. And then last one from me. You spoke about the sort of the three phases of network deployment. And, you know, the first phase is lots of sites of amendments. And then just sort of the next big phase is moving to, I would assume, fewer sites. bit with four or five times the revenue per site. As you look through the sort of the multi-year period that as you work through these phases, is the revenue growth you get similar in the first phase and the second phase and the third phase? Or is it heavily weighted towards the first phase just because of the number of sites they're touching?
Kate
You know, our experience has been that revenue growth over those various phases stays relatively stable and similar. As we talk about our long-term expectation around growth and organic growth in towers, we've said that we think that stays maybe a little bit above the 5% level. We think we can sustain that for a period of time. It also ties into our long-term target of being able to grow the dividend 7% to 8%. So we think there's an elongated runway of growth That's driven by that top line opportunity as the carriers go through the various phases of deployment We see good good opportunity to lease both towers and small cells and we think that extends the runway of growth and as we look at kind of the current environment that we're in really excited about where we are and excited about the top line growth that we're that we're seeing and the the consistency of the demand from our customers to need to improve their networks and ask for additional leases on our assets across all three of our businesses. Great. Thanks very much, guys. Well, thanks, everybody, for joining. Kate, thanks for your help on the call this morning. I do want to thank our team. As we wrap up 2022, I realize everyone is already really focused on what we have to deliver in 2023, but I did want to take the opportunity to congratulate our team for a job well done in 2022, navigating to a great outcome through some pretty difficult challenges over the course of the year. You all did a great job for our customers, and I know they appreciate it. So thank you to the team. I'm excited about what we'll do in 2023 and look forward to talking to everyone next quarter.
Operator
The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.
Disclaimer