High Hopes for Towers in 2020 Fade in Second Half [U.S.]


Originally Published by AGL Media Group
Author: Nick Del Deo

ACCELERATION IN TOWER GROWTH MAY BE DELAYED UNTIL 2022

As we approach the New Year, we think it is helpful to look back over the past year, and to look out over the year coming, to refresh our perspectives and revisit the research that remains most relevant. It would be an understatement to say a lot has happened in 2020. The economy was shaken by the coronavirus pandemic, the way we live and work was upended, and a bright light was shined on the critical importance of communications infrastructure.

For the towers, 2020 was originally expected to be a segue to an acceleration in growth in 2021. The T-Mobile/Sprint deal was approved, which would unlock a flood of transition and integration investment ahead of any churn. Dish Network had committed to build a 5G network and agreed to fairly strict deployment timeframes. 5G was around the corner.

The pandemic didn’t change that narrative but did benefit the towers’ relative share price performance given their safe haven status. The collapse in interest rates and associated effect on [weighted average cost of capital] WACC didn’t hurt, either. They meaningfully outperformed the S&P during the first half of the year.

In recent months, however, the group has lagged the market, with Crown Castle and SBA likely to have an in-line year overall and American Tower set to underperform by about 20 percentage points. There’s been a broader rotation away from defensive names toward cyclical names, and the towers clearly fall in the former camp. And expectations for growth, at least in 2021, simultaneously ticked down. The cessation of leasing activity by T-Mobile in advance of its merger with Sprint being approved hit growth harder and for longer than many appreciated. Dish Network has indicated it won’t start construction of its network until H2 2021 due to equipment considerations. American Tower’s master lease agreement (MLA) with T-Mobile was less remunerative than expected. Crown Castle’s 2021 guidance called for little growth in gross tower leasing. Clearing the C-band spectrum that is currently being auctioned will take some time, suggesting its impact on tower leasing won’t be felt until later in 2021.

Still, this points to an acceleration in growth in 2022. Absolute tower multiples remain toward the upper end of where they’ve historically traded, but relative multiples (vs. the S&P) are consistent with their long-term average, offering a more constructive starting point. We continue to prefer SBA, which has the most appealing asset portfolio and mix-adjusted valuation. We worry about the risk that American Tower won’t monetize Verizon C-band deployments to the same degree as its peers due to capacity rights granted as part of the 2015 Verizon portfolio acquisition and/or residual effects of its 2017 MLA with Verizon. The outlook for Crown Castle’s small cell leasing over the medium term seems soft.

Heading into 2020, Crown Castle’s shares had meaningfully underperformed those of its peers, prompting us to examine whether this might present an investment opportunity. It was generally understood that the underperformance was, in part, a function of the company’s aggressive push into fiber, something Elliott Management would highlight several months later. Instead, two other considerations prevented us from getting more enthusiastic about the stock’s prospects. First, lagging domestic tower growth was partially driven by capacity rights Crown Castle granted to T-Mobile and AT&T in conjunction with the portfolios it acquired in 2012 and 2013, which we expected to persist to varying degrees. And second, the quality of Crown Castle’s EBITDA and AFFO remained inferior to its peers’ and weighed on the company’s warranted valuation.

In the early months of the COVID crisis, we observed that while tower valuations were near all-time highs relative to the S&P 500, the premium was not as high as it appeared after adjusting for expected T-Mobile and Dish Network leasing and a likely dip in S&P 500 profits. We concluded that SBA had the most attractive valuation and upgraded its shares to Buy. Our view was underpinned by several key attributes, which are as relevant today as they were in April:

  • SBA had the highest quality bottom line metrics, meaning its multiples relative to peers were more attractive than they first appeared.
  • SBA had the most attractive asset mix and highest warranted multiple given its domestic tower skew, the makeup of its international portfolio, and lack of exposure to fiber.
  • The company’s growth outlook was the cleanest, without any latent churn like American Tower and without any outstanding capacity rights or escalator variances stemming from large carrier portfolio acquisitions like its peers.
  • SBA’s capital allocation policies were the most favorable, since they skewed toward attractive tower acquisitions and share repurchases.

In light of tower equities’ strong performance in the first half of the year and macroeconomic uncertainty, we thought at that point it would be appropriate to identify and examine key macroeconomic risks:

  • Towers are exposed to changes in interest rates. However, the spread between the 10-Year yield and tower unlevered cash flow yields was at the upper end of the historical range, providing some cushion.
  • An increase in inflation would be a negative for towers, since their domestic escalators tend to be fixed and their contracts are long term. The ability to price new business at market and ground leases generally being subject to fixed escalators help to mitigate that risk, however.
  • Towers don’t pay domestic income taxes due to their REIT status, so an increase in corporate tax rates would enhance the EBITDA multiple at which towers should trade relative to C corps.
  • Dish Network’s ability to secure financing and successfully build its network is reliant on the macroeconomic environment.

While the broader macroeconomic outlook has since improved, these risks are all still relevant. In particular, we view interest rate exposure as the most pertinent: a 100 basis point increase in unlevered cash flow yields would likely drive tower enterprise values down by ~25 percent and their stock prices down by about one third.

In July, Elliott Management launched a campaign to, among other things, compel Crown Castle to cull what it viewed as low-return small cell investments. While we haven’t been supportive of Crown Castle’s push into fiber, we fell somewhere in between Elliott’s argument and the company’s position. We don’t think current yields for small cells are indicative of their long term prospects, even if we don’t think long term returns will prove to be as good as management suggests. Hard evidence, to the extent it exists, yields mixed signals.

We then took a closer look at returns on Crown Castle’s fiber unit. The fiber unit in aggregate has generated incremental EBITDA yields comparable to those of Zayo, which suggests the company’s segment is not underperforming its closest peer. Crown Castle’s overall ROIC lags that of its tower peers, but that had been more a function of its towers, not fiber. Crown Castle granted concessions to T-Mobile and AT&T as part of the 2012 and 2013 acquisitions that have dampened its relative ROIC trajectory. Management has made significant acquisitions and investments in fiber, but they were not material enough to move the needle until 2019. We judged a sale of the fiber unit or a JV as unlikely to unlock value, since the business was not trading at a significant discount.

Two large tower MLAs have been inked over the past few months: American Tower’s with T-Mobile in September, and Crown Castle’s with Dish Network in November.

Enough information was disclosed about the American Tower/T-Mobile MLA to get a sense of what the aggregate financial impact will be relative to expectations, which had assumed a meaningful level of transition/integration activity from T-Mobile. It appears as though T-Mobile will ultimately pay American Tower about 10% less in total than we had expected, when all is said and done. In addition to lowering our American Tower forecasts, we also trimmed our estimates for Crown Castle and SBA, since the American Tower deal suggested that some component of our industry wide assumptions was off.

Details regarding the Crown Castle/Dish Network MLA were sparser, but we could draw some high level conclusions. Crown Castle appears positioned to win something more than its “fair share” of Dish Network’s deployments, at least initially, given its somewhat more urban tower skew and the additional flexibility Dish Network has when choosing locations given its focus on coverage rather than capacity. If Dish Network’s entry into wireless is successful, however, it will need to densify and expand its network, which will erode that site selection flexibility and result in more normalized levels of new business for American Tower and SBA. We’re skeptical of the notion that combining fiber with towers played an important role, however. Crown Castle’s fiber solutions unit only has 7K to 8K towers on-net, of which some fraction are Crown Castle towers, of which some portion fall under the 20K towers covered by the MLA. Towers and fiber are severable and there’s no clear engineering or operating advantage to leasing both from a single vendor. Crown Castle likely offered volume discounts for fiber to avoid losing Dish Network’s fiber business to competitors.

Tower leasing was supposed to accelerate in 2020, but the delayed closing of the T-Mobile/Sprint deal and associated pause in network spending pushed that expectation to 2021. Surveying the landscape for 2021 suggests that step up is likely a 2022 event, with T-Mobile, Dish Network, and C-band the most important drivers.

  • Leasing activity from T-Mobile should increase in 2021 from 2020 levels, but appears likely to underwhelm. Crown Castle’s initial 2021 guidance for gross tower leasing was marginally higher than in 2020. American Tower’s MLA with T-Mobile suggests limited upside. SBA’s commentary indicates T-Mobile’s leasing is ramping, but is still below year-ago levels. In a broader sense, more of this activity seems likely to substitute for prior work rather than drive more incremental work than originally expected.
  • Dish Network appears set to begin installing gear in earnest in the second half of 2021, meaning its revenue impact to the Towers will be more pronounced in 2022. The small arrays it intends to initially deploy should garner relatively modest initial rents.
  • The C-band auction, currently underway, will likely start to drive leasing later in 2021. American Tower may not monetize C-band deployments from Verizon to the same degree as its peers, however. Capacity rights granted to Verizon as part of the 2015 acquisition of its towers could be used to support C-band gear (these towers host perhaps 35 percent of Verizon sites on American Tower’s portfolio), and the 2017 MLA with Verizon may also come into play.

Taken together, our forecasts suggest 2021 growth will look a lot like 2020 overall, but with growth improving over the year rather than decelerating, and with that upswing carrying through into 2022. We modestly reduced target prices to account for these updates to our forecasts.

Read original article at aglmediagroup.com.