Our industry outlooks across the U.S. telecom and cable sector are again a mixed bag, although ratings trends are increasingly negative. Cable operators continue to perform well despite the accelerating erosion of the pay-TV ecosystem, while competitive intensity in the wireless market has abated since T-Mobile and Sprint announced their plan to merge. Conversely, wireline companies--which are facing secular industry pressures and heightened competition--continue to lose voice lines to wireless substitution and broadband and commercial customers to cable providers while smaller infrastructure companies have struggled with integrations from previous mergers and acquisitions (M&A).

Against this backdrop, there are several risks as we head into 2020. Despite a more benign competitive environment in U.S. wireless, service revenue trends and customer growth should moderate in 2020 due to mature industry conditions and increasing competition from cable, especially as Altice enters the market. Additionally, Comcast and Charter may enter into wholesale arrangements with AT&T at more favorable rates than Verizon, which could make them more competitive. Wireless operators will also look to differentiate their 5G networks, resulting in higher carrier churn. The outcome of the T-Mobile and Sprint merger is still unknown but should be finalized in the first half of this year. Assuming the deal is approved, we believe the combined company will focus its efforts on integration, although the impact of DISH's potential entrance is uncertain and will likely depend on third-party financial support. If the deal is rejected, we believe more competitive promotional offerings led by T-Mobile could resume. Financial leverage for the sector remains elevated and spending in upcoming spectrum auctions could pose risk for credit quality over the next couple of years.

For the U.S. cable sector, our outlook is stable, bolstered by continued growth from broadband and commercial services despite fewer traditional pay-TV customers and potential competition from 5G fixed wireless. We also have a stable outlook for the telecom infrastructure segment, which comprises data center, fiber, and tower operators, although operating trends have been somewhat mixed. While larger infrastructure companies continue to perform well, smaller issuers are showing evidence of slowing growth while execution missteps from recent M&A have caused deteriorating credit metrics and several downgrades. Still, these industries should benefit longer term from greater data and video consumption by residential, commercial, and mobile customers.

U.S. Economic Outlook: Telecom And Cable Issuers Are At Greater Risk This Time Around

For our telecom and cable ratings, we focus on economic indicators that we believe most correlate with consumer demand, including real GDP, unemployment, personal consumption expenditures, and housing starts (see table 1).

Table 1

Under our baseline assumption, S&P Global Ratings' economists see a 25%-30% risk of a recession starting in the U.S. over the next 12 months. Consumer spending has remained robust despite trade-related worries. We expect real GDP growth of 2.3% in 2019 as protectionist policies and weaker private investment slowed growth. However, recent softness in business fixed investment is expected to be replaced with a modest upturn in 2020, although we still expect GDP growth of 1.9% in 2020.

Despite interest rate cuts during 2019, we believe credit risks are rising as the cycle continues to age--particularly in the corporate debt market, where debt and leverage have built up substantially. Although the U.S. telecom and cable sectors held up well during the 2008 financial crisis, we believe they could be less resilient under a more stressed scenario this time around given more mature industry conditions, changes in consumer preferences, and technology shifts (e.g. migration to the cloud from legacy networking solutions and online video streaming services from linear video). Furthermore, these issuers carry more leverage than they did in 2008 because of M&A, driven by a decade of historically low borrowing costs, although interest coverage and free operating cash flow (FOCF) to debt have not deteriorated as much. While healthy credit markets enabled many telecom and cable providers to push out debt maturities, there are still a number of vulnerable issuers that face substantial debt maturities over the next couple of years. If credit market conditions deteriorate, companies at the lower end of the rating spectrum such as Frontier Communications Corp., Intelsat S.A., GTT Communications, Inc., Gogo Inc., Uniti Group, Inc., and Sprint Corp. (on a stand-alone basis), could struggle to refinance.

U.S. Telecom And Cable Ratings Trends Are Negative

Rating trends in the U.S. were largely negative in 2019, with 22 downgrades and only five upgrades. Downgrades were largely concentrated in the 'B' rating category, where we lowered multiple ratings to 'B-' or to the 'CCC' category from 'B'. We expect more negative rating actions in 2020, especially if capital markets conditions weaken, because about 30% of rated issuers either have a negative outlook or are on CreditWatch negative. About 88% of U.S. telecom and cable issuers are speculative grade, including two-thirds in the 'B' or 'CCC'/'CC' categories. Ratings in these categories largely reflect secular industry declines, intense competition, debt-financed M&A, and, in some cases, refinancing risk.

In 2020, we expect rating trends among U.S. cable providers to be relatively stable as broadband growth offsets the loss of lower-margin video customers. Conversely, we expect ratings pressure in other subsectors to continue, stemming from secular industry declines in wireline and linear video, intense competition, slowing growth, and weaker credit quality as a result of integration missteps from acquisitions among data center and fiber providers. More importantly, with the threat of recession and prospect of weaker capital markets conditions, this decline in credit quality among many lower-rated issuers could result in greater refinancing risk, which could trigger further downgrades and more defaults over the next few years.

Chart 1

Chart 2

Chart 3

Key Themes For 2020

Balancing shareholder returns and debt deduction

U.S. telecom providers are at a crossroads, trying to balance the need for investment with debt repayment and shareholder returns. With pressure from activist shareholder Elliot Management, AT&T decided to allocate 50%-70% of its discretionary cash flow to share repurchases. In aggregate, it is looking to retire 70% of the shares issued for the Time Warner acquisition, totaling about $30 billion over a three-year period (or $10 billion per year). A further negative for near-term deleveraging, the company said it would front-end these stock buybacks. In 2020, for example, assuming the company achieves its guidance of $28 billion of FOCF, about $15 billion would go to dividends, leaving around $13 billion for debt reduction and share repurchases. However, if management decides to front-end these stock buybacks, we do not believe there would be any material FOCF left for debt reduction outside of its proposed $5 billion-$10 billion of asset sales. Furthermore, the company announced a public leverage target of 2.00x-2.25x by 2022, which implies only about a quarter-turn of leverage reduction at the high end of its range over the next three years.

Verizon also came out with a stated net unsecured leverage target of 1.75x-2.00x, which was a contributing factor to our positive outlook revision in April 2019, signaling the potential for an upgrade to 'A-' from 'BBB+'. That said, we expect Verizon to be aggressive in upcoming auctions, especially the C-Band auction, which could ultimately curtail leverage improvement, depending on how much it spends.

In the wireline sector, we view CenturyLink's lower leverage target and dividend cut favorably from a credit perspective. The company reduced its dividend to about $1.1 billion annually from $2.3 billion and revised its leverage target to 2.75x-3.25x from 3x-4x. As a result, in February 2019 we revised the outlook to stable from negative and affirmed the 'BB' issuer-credit rating. Similarly, Consolidated Communications announced that it would eliminate its $100 million annual dividend and use the proceeds for debt reduction to achieve leverage below 4x by 2021, which management believes will enable it to refinance its unsecured bonds at more attractive levels. Still, Consolidated's top line continues to decline about 4% annually and its cost synergies are running out. Unless it can improve top-line performance, we believe that EBITDA declines could accelerate and constrain leverage improvement.

In cable, we believe further consolidation of small and midsize operators is possible in an attempt to combat rising programming costs and achieve other cost synergies. Absent M&A, we expect Charter to continue to repurchase shares to reach the high end of its 4.0x-4.5x leverage target. We expect Comcast to limit share repurchase activity and continue allocating excess cash flow to debt repayment in 2020 as it aims to restore leverage to the low- to mid-2x area following the leveraging Sky transaction, although the company has not provided public guidance on its share repurchase intentions.

Table 2

5G deployments could be more risk than reward for U.S. telcos over the near term

U.S. wireless carriers are aggressively pushing their 5G deployments. T-Mobile announced that it would cover 200 million points of presence (POPs) with 5G using its 600 megahertz (MHz) spectrum licenses by year-end 2020. While using low-band spectrum will limit data speeds, if it is successful in acquiring Sprint, the company will have access to an abundance of mid-band spectrum in the 2.5 gigahertz (GHz) band, which should enable a more robust 5G experience for customers. AT&T deployed 5G in 21 cities throughout the U.S. using millimeter Wave (mmWave) spectrum but similarly, will use low-band spectrum for nationwide coverage this year. Verizon reached about 30 cities with 5G in 2019, but has not outlined its buildout plans for 2020.

Overall, we have a cautious view on 5G, based on our expectation for accelerated deployments. We believe that the acquisition of spectrum licenses and higher capital expenditures (capex) to support 5G network deployments, which includes small cell and fiber builds, could constrain leverage improvement at a time when the carriers are trying to reduce debt and improve credit metrics. We also expect that the demand for mid-band spectrum such as the C-band will be strong, since in our view these licenses will likely serve as the foundation for 5G deployments.

At the same time, we believe that revenue opportunities associated with 5G will be slow to materialize in the near term since we question consumers' propensity to spend more for faster data speeds on their devices, and most of the internet of things (IoT) and enterprise opportunities are likely several years away. While 5G fixed wireless offers some potential to monetize investments at an early stage, the technology is still unproven.

Table 3

5G fixed wireless could pressure U.S. cable operators longer term

We do not believe new competition from 5G fixed wireless will significantly pressure cable earnings in 2020 since buildouts are still in the early stages. Verizon has had mixed results in its initial markets, raising questions around its ability to cover its targeted 30 million homes (approximately 25% of the U.S.) using mmWave spectrum, which the company now hopes to achieve within five to eight years. We believe T-Mobile's in-home broadband strategy hinges on acquiring spectrum from Sprint, which is still uncertain. Even if the acquisition is completed in 2020, we believe reaching its goal of 9.5 million customers by 2024 would be more back-end-loaded since the spectrum still needs to be deployed.

Still, longer term we believe 5G fixed wireless could put some pressure on cable earnings in four to five years if wireless carriers successfully execute communicated strategies. We recognize that cable will have a speed and reliability advantage over new competition, but we believe there could be wireless pricing discounts given cheaper last-mile deployment costs, particularly if wireless operators are aiming to help subsidize necessary investments to support 5G mobile. Therefore, we believe the introduction of new wireless competition into selected markets could result in some market share erosion if the technology can be demonstrated at scale. On top of that, and of greater impact to the cable industry, more competition could make it harder to monetize increasing data usage, especially if there is a cheaper option available with service speeds that are good enough for many consumers.

Strong demand at upcoming auctions could delay credit quality improvement

Notwithstanding the desire to reduce leverage, the extent of deleveraging will partly depend on how much spectrum wireless and cable operators purchase in upcoming auctions including:

• Auction 103 for mmWave spectrum, which is currently underway. This spectrum can deliver very fast data speeds with low latency, but has poor propagation characteristics with limitations on the distance it can travel and its ability to penetrate buildings and other objects, such as trees. So far, bidding has been healthy, with $6 billion of potential gross proceeds as of Jan. 6, 2020. While this is above our initial estimate of up to $5 billion, spending on recently completed mmWave auctions (about 1,500 MHz in the 24- and 28-GHz band) were at the low-end of our range (totaling about $2.7 billion compared with our estimate of up to $8 billion), leaving more room for operators to spend in Auction 103. As a result, we continue to expect total mmWave auction proceeds of less than $13 billion, mainly in dense urban markets.
• C-Band: The Federal Communications Commission (FCC) also aims to auction 280 MHz of spectrum in the 3.7-GHz and 4.2-GHz band, which is currently used by fixed satellite service providers to provide video signals to cable head-ends. We believe this mid-band spectrum could prove valuable for 5G as it offers more data capacity than low-band spectrum and better coverage than mmWave spectrum. The government plans to complete a public auction by the end of 2020, although this could be delayed if satellite operators and the government cannot agree on the terms and conditions around the auction because the FCC lacks the authority to confiscate in-use spectrum. Given the complexity involved, we believe this could delay timing to 2021.
• Citizens Broadband Radio Service (CBRS): The FCC plans to auction 70 MHz of mid-band spectrum in June 2020. CBRS represents 150 MHz in the 3.55 GHz-3.7 GHz range, viewed as prime spectrum for 4G and 5G deployments. The 150 MHz of CBRS spectrum will be managed under a three-tier access plan that includes priority use by the incumbents (Department of Defense and satellite providers), priority access licenses (PAL) determined by the 2020 auction, and general authorized access (GAA) reserved for unlicensed use. Most carriers anticipate using carrier aggregation with other licensed bands to use both GAA and PAL spectrum together.
• The upper 6-GHz (5.925 GHz-7.125 GHz) band could also serve as important mid-band spectrum for 5G, although an auction is not yet planned. The wireless industry has proposed to free up the upper 600 MHz of the band for licensed use in an auction, which would involve relocating incumbents to higher frequencies. The next step in the process will be an FCC order, likely in the first half of 2020, with a plan for the 6-GHz spectrum (likely after the C-Band order). Given the federal government's focus on 5G leadership, we believe it is possible that it adopts the wireless industry's plan to make more mid-band spectrum available, particularly considering that other markets, such as China, Japan, and South Korea, have freed up a lot of mid-band spectrum. If so, we believe gross proceeds could be similar to the C-Band; however, moving higher in frequency increases network buildout costs, which could somewhat hamper demand.

Estimating gross proceeds and MHz/POP is difficult because it is based on a variety of factors, including a carrier's spending ability, visibility into clearing the spectrum, the timing of other auctions, and the number of players participating (which will be heavily affected by the outcome of the T-Mobile/Sprint merger). We have lowered our estimate for the upper end of the range for C-Band to $.30 MHz/POP (from $.40) because the amount of spectrum to be made available has increased for the second time to 280 MHz from the initially planned 100 MHz. We also believe that if the 6 GHz auction comes to fruition as wireless operators propose, valuation on a MHz/POP basis will likely be lower than C-Band because roughly twice the amount of spectrum could be auctioned (600 MHz versus 280 MHz).

Overall, we expect that Verizon will be the most aggressive bidder in mid-band auctions since its current spectrum holdings are the lowest on a per-subscriber basis compared with other wireless carriers. In contrast, we believe that AT&T will be less aggressive since it is working through the integration of Warner Media while trying to navigate secular industry pressures in several business lines and manage its massive debt burden. That said, AT&T could be more aggressive than Verizon in the current mmWave auction since Verizon outbid AT&T for StraightPath, which held a significant amount of mmWave licenses. Demand from the third-largest carrier, T-Mobile, will partially depend on whether or not it successfully acquires Sprint, which would give it an enviable spectrum position. Sprint owns a vast amount of mid-band spectrum in the 2.5-GHz band.

We also believe cable operators could purchase mid-band spectrum, as Comcast, Charter, and Altice USA all recently launched wireless offerings based on wholesale agreements (known as mobile virtual network operators [MVNOs]) with wireless carriers. It could be beneficial to gain owners economics in certain markets as data consumption increases, and owning spectrum could also be leveraged in future MVNO negotiations.

We believe Verizon could spend about $10 billion in spectrum purchases and still achieve our upgrade target of leverage below 2.5x by year end, although the timing of the C-Band auction is still uncertain. We believe AT&T has some flexibility in the 'BBB' rating for spectrum purchases, although we do not factor in material spending given its strong spectrum position and elevated debt burden. We also believe that cable operators have flexibility within the current ratings to purchase spectrum. We do not explicitly factor in spectrum purchases for Charter or Altice USA, although both could pull back on share repurchases to maintain existing ratings if necessary. While we expect Comcast to continue to use free cash flow to pay down debt in 2020, we have incorporated $1 billion-$2 billion in investments into our base-case, which could include spectrum purchases.

Table 4

Capital spending will be modestly lower in wireless, stable in wireline and cable

Capex from U.S. wireless carriers increased significantly in 2019 in order to deliver faster 4G LTE speeds and to pave the way for 5G deployments over the next few years. T-Mobile continued to build out the 600-MHz spectrum band acquired in the broadcast incentive auction, while AT&T aggressively deployed its public safety network for first responders (FirstNet) and is now about 75% complete with its buildout. At the same time, Verizon is building out fiber infrastructure and densifying its network with small cells, partly to support its 5G fixed wireless build. Sprint has been deploying its 2.5-GHz band and upgrading its network with massive multiple-input and multiple-output (MIMO) in anticipation of its 5G launch

In 2020, we expect modestly lower capex in wireless than 2019, reflecting AT&T's planned $3 billion reduction in capital spending over the next year, a portion of which is allocated to wireless. However, aggregate capex for the sector will likely still remain elevated as carriers densify their networks and deploy fiber for backhaul, but the lack of new mid-band spectrum could push substantial capex growth into 2021 or 2022. That said, if the T-Mobile-Sprint deal is approved, it could accelerate capital spending by the combined company and push DISH to deploy its spectrum aggressively next year. Overall, we project a modest 1%-2% decrease in capital spending while capex to revenue remains in the 14%-15% area in 2020.

For other telcos, capital spending also increased significantly in 2019. CenturyLink used some of the savings from its dividend cut to invest in its network, including expanding its fiber network to new buildings to better serve its small to medium business (SMB) and enterprise customers. We believe that its network investment will also position it well to provide wireless backhaul services to mobile operators as they build out 5G. TDS increased its network investment in 2019 to support its deployment of 5G and extend fiber to offer faster broadband speeds to its customers. For 2020, we expect overall telco capex to be relatively flat due to ongoing fiber investment for wireless backhaul and enterprises, especially among the larger and financially stronger issuers.

Chart 4

Chart 5

For U.S. cable operators, capex declined precipitously in 2019, mainly driven by Charter. We expect similar capex levels in 2020 and for capex to revenue to decline modestly, driven primarily by lower spending on customer premise equipment (CPE) and scalable infrastructure, such as headend equipment, which together typically account for at least half of cable capex. We believe longer-term trends are also favorable, as capital-intensive video customers will likely continue to decline. While cable operators will need to continue network investments to sustain their competitive edge in broadband, we believe this can be accomplished affordably. For example:

• Charter Communications has indicated it can operate with capex to revenue in the low-double-digits excluding mobile capex (currently around 14.5% year to date).
• Altice USA indicated capex/revenue of 11% once fiber-to-the-home rollout is complete in about four years, (from 14.1% year to date).
• Comcast is guiding to an improvement in 2019 cable capex/sales of 150 basis points (bps) from 13.8% in 2018, with trends beyond that expected to continue.

Chart 6

Breaking down the major components of capital spending, we expect the following trends:

CPE (typically about 30%-40% of total capex):  Video cord-cutting trends will keep CPE spending down as fewer video customers result in fewer set-top-boxes (STB), the cost of equipment itself declines, and more customers opt for app-based boxless video options.

Scalable infrastructure (20%-30%):   Spending on headend equipment can be lumpy and we project spending in 2020 to remain paused following the recent completion of DOCSIS 3.1, which enabled 1-gigabyte speeds virtually nationwide. However, we believe cable operators can achieve 10 gigs through Full Duplex DOCSIS with a relatively moderate amount of spending; Charter Communications indicated that it spent only $9 per passing (or about $450 million) to upgrade to DOCSIS 3.1.

Line extensions/upgrade (20%-30%):   We expect this amount to remain relatively flat as operators continue to drive fiber deeper into the network and edge out their footprint, offset by networks becoming more software-defined, which is cheaper to maintain.

Support (10%-20%):   Likely to spend less on vehicles and buildings over time with fewer truck rolls, service calls, and network activity.

Shifting technology can make capital spending difficult to predict longer term. Right now, the biggest wild card is wireless investment. Comcast, Charter, and Altice all currently employ a capital-light approach to wireless through wholesale agreements with nationwide wireless carriers. We project that mobile capex will be around 5%-10% of total cable capex in 2020, as operators continue to spend on building out retail stores. However, we believe profitability could be challenged longer term by rising data consumption on unlimited plans, where cable operators currently pay a variable-use fee to wireless operators. Therefore, it is possible that cable operators purchase spectrum in upcoming auctions to build out their own wireless networks to offload traffic in more densely populated regions, relying on MVNO for blanket nationwide coverage. If so, this could make it difficult for cable operators to achieve low-double-digit capex to revenue longer term because it is costly to build and maintain a wireless network.

Impact of T-Mobile and Sprint merger

While both the FCC and Department of Justice have approved T-Mobile's proposed $26 billion merger with Sprint, 14 states and the District of Columbia are suing to block the deal. If approved, we believe the two companies would close by the first half of 2020.

Notwithstanding the proposed concessions, which are designed to maintain competitive intensity within the wireless market by facilitating the creation of a fourth nationwide provider, a combination of T-Mobile and Sprint would likely have significant industry ramifications over the next three years. A successful combination would dramatically improve T-Mobile's scale and spectrum position, enabling it to better compete with industry heavyweights AT&T and Verizon. T-Mobile's postpaid market share would increase to 29% from 17%, which is comparable to AT&T's postpaid share but still behind Verizon. Despite having to sell 14 MHz of nationwide spectrum in the 800-MHz band to DISH for $3.6 billion in three years, T-Mobile would still be able to retain Sprint's valuable spectrum in the 2.5-GHz band, which could give it a competitive advantage as it deploys 5G. T-Mobile would also have to sell Sprint's prepaid business to DISH and provide it with an MVNO agreement, although we do not believe Sprint's prepaid business is profitable and in fact, the wholesale agreement with DISH is high margin and would be modestly accretive to T-Mobile's EBITDA.

If the transaction is not approved, we believe that T-Mobile would likely resume it share repurchase program and will be more aggressive in upcoming auctions, particularly the C-Band, since it lacks mid-band spectrum. Sprint faces significant operational hurdles, including postpaid subscriber losses and elevated churn. Additionally, its adjusted leverage is elevated at close to 5x, FOCF is negative, and it has significant debt maturities over the next three years of around $15 billion. Our 'B' issuer credit rating is on CreditWatch developing. If the deal is not approved, we would expect to lower the rating by at least one notch depending on our view of likely support by its higher-rated, 85% owner SoftBank Corp.

Chart 7

Chart 8

The impact on AT&T and Verizon is difficult to assess. On the one hand, we believe they would have a window of about three to four years to capture market share as T-Mobile goes through a massive and complex integration and DISH builds out its network. We do not view DISH as a big threat to the industry in the near term given that its small customer base (acquired from Sprint) and MVNO agreement would constrain cash flow. Over the longer term, competitive intensity will depend on DISH's ability to find a partner and profitably expand its base of business to better compete with the incumbents. At the same time, T-Mobile would have a robust and balanced portfolio of spectrum licenses. We believe it could look to capture share by monetizing this spectrum by aggressively pricing its service for customers.

For U.S. cable operators, we believe a successful merger could be negative, mainly because it will strengthen T-Mobile's spectrum position, potentially enabling more competition from 5G fixed wireless. We believe T-Mobile's fixed wireless plan could be disruptive because of its spectrum position, including access to Sprint's 160 MHz of 2.5 GHz mid-band spectrum, which provides a good coverage layer and better throughput than low-band spectrum. Perhaps more importantly, we believe having a cheaper alternative to cable in the marketplace could make it more difficult for cable to monetize increasing demand for data in five years.

For DISH, we believe the conditions it agreed to increase the likelihood it will build out a network, which could be a credit negative depending on financing and partnerships. That said, DISH has capacity at the current rating to absorb the combined $5 billion purchase of spectrum and Sprint's prepaid business. We estimate pro forma adjusted debt to EBITDA will remain around 5x for the $1.4 billion purchase of the prepaid business and could rise to 6.0x-6.5x to fund the $3.6 billion purchase of the 800-MHz spectrum in three years, below our downgrade trigger of 6.5x. Still, this will partly depend on operating trends at DISH DBS over the next few years, as well as unforeseen developments.

However, the network buildout is not factored into these leverage calculations. Furthermore, we believe DISH's lack of experience in wireless could make it challenging to compete against established players such as AT&T and Verizon. Adding more spectrum to DISH's portfolio does little to increase long-term revenue visibility. Therefore, we maintain the negative outlook on our rating on DISH Network given the uncertainty around how it plans to finance the wireless network buildout

Cord-cutting is worsening, but not all is lost

With video cord-cutting accelerating in 2019, traditional pay-TV providers are taking different approaches to adapt to the shifting environment. Overall, the tone from cable management increasingly indicates that video subscriber losses are a lower priority than in the past. Still, large cable and satellite providers continue to invest in improving the customer experience, targeting high-value customers with less promotional activity, and offering lower-value customers alternatives. Comcast has taken the lead by building off its X1 platform and introducing technology to better integrate over-the-top (OTT) apps, dubbed "Flex." We believe this positions the company to provide aggregation, discovery, and billing in an increasingly fragmented video landscape, thereby increasing the value and stickiness of its core broadband offering. We expect other cable operators to follow suit with similar offerings, or by licensing the Flex platform from Comcast.

However, smaller cable operators do not have the scale to negotiate with programmers or the financial resources to invest in innovative technology. Therefore, many of these operators have reached an inflection point whereby video gross margins have evaporated, forcing them to pass along the full amount programming costs, resulting in a significant erosion of video customers.

Unlike cable providers, satellite companies do not have the luxury of a true broadband product to deliver streaming alternatives, which is often discounted when bundled with video. Therefore, the cost savings for satellite subscribers to switch to an online alternative are more significant than for many cable defectors. As a result, we expect this segment to continue to account for most satellite subscriber losses in 2020. However, we expect the mix of losses between DISH DBS and DirecTV to begin to normalize following a year in which DirecTV reduced promotional activity, likely leading to some temporary market share gains by DISH in 2019. Still, we expect DirecTV losses to outpace losses at DISH because AT&T is launching a streaming multichannel product that will become the focus of future gross adds, dubbed "AT&T TV."

Overall, we expect that traditional pay-TV distributors will lose video subscribers at approximately 5% in 2020, which is higher than the 3%-3.5% rate we projected a year ago for 2020 as satellite remains under significant pressure and cable operators increasingly shift their focus from asset utilization to growth in EBITDA per subscriber, with broadband as the key service. While there is uncertainty around our forecast, we believe there is greater downside risk than upside potential, as consumers will have more video alternatives from a flood of new subscription video on demand (SVOD) services available mid-year. Still, we are projecting that 2020 won't be quite as bad as 2019, mainly because of some moderation in satellite subscriber losses as the base shrinks away from customers on promotions.

Chart 9

The rate of cord-cutting could be higher if declining viewership combined with higher prices puts more stress on the ecosystem, resulting in a reinforcing loop. This would make SVOD services more attractive to consumers on a relative basis, attracting even more content investment and potentially more viewership. Declining traditional viewership also makes content investments less attractive for traditional video providers, potentially putting even more pressure on viewership.

We believe there are two wildcards to the equation: the value of sports and the rate of piracy, both of which are difficult to quantify. The rate of cord-cutting could be lower if the remaining traditional pay-TV base is unwilling to drop costly live sports (which cheaper SVOD alternatives do not have) and if password sharing for virtual multichannel video programming distributors (vMVPDs, which do carry live sports) is curbed. We believe that the combination of rising prices for vMVPDs (which were introduced at unprofitable rates) and fewer people sharing the cost of an account could aid the traditional pay-TV universe.

While industry executives widely agree password sharing is a problem, there's no consensus on where to draw the line. Programmers and distributors blame each other for being too lenient in how many people can simultaneously stream from one account and online TV services vary widely in how generous they are about password sharing. Still, there appears to be momentum for slowing password sharing: the Alliance for Creativity and Entertainment has been formed and is devoted to reducing online piracy. Participants include Netflix, Amazon.com Inc., The Walt Disney Co., ViacomCBS Inc., AT&T Inc.'s HBO, Comcast Corp., and Charter Communications. While details were not disclosed, recent carriage agreements with Charter and DISH have also included language around piracy. While this is an issue for SVOD, we believe it is more of a threat from vMVPDs because they are more of a direct substitute to traditional live TV and are more expensive than SVOD, magnifying the cost savings from sharing an account.

The ratings impact from an acceleration in cord-cutting is manageable for cable companies, in our view. Cable's broadband service remains the key differentiator and the basis for our credit ratings on the sector. Ratings could be pressured over time if the rate of video cord-cutting materially worsens and broadband revenue growth stalls or even starts to decline, which we do not expect to occur in 2020. However, there could be negative ratings implications for DISH DBS if ongoing subscriber losses result in lower earnings and cash flow such that it cannot reduce leverage from debt paydowns.

Regulation could come back into focus

We believe regulation could come back into focus with the upcoming presidential election. While we do not expect any near-term impact on cable earnings, the biggest long-term threat to the industry remains pricing regulation, in our view. It is possible that under a Democratic administration broadband could be reclassified under Title II, opening the door for future federal-level price caps. We do not believe the recent appellate court decision to uphold the FCC roll-back of net neutrality rules reduces this risk because the FCC could reinstate these rules under different leadership.

Separately, it is unclear if, and when, the FCC can block state-level net neutrality efforts. For example, New York recently followed California's lead to propose legislation that will restore--statewide--the net neutrality rules eliminated in the 2017 Restoring Internet Freedom order adopted under current Republican FCC Chairman Ajit Pai. While this order preempted any state efforts to reinstate net neutrality rules, the DC Circuit has said that the FCC exceeded its authority and could only preempt state efforts on a case-by-case basis. We believe it is unclear whether or not the court's decision to wipe out that blanket preemption opened the door to a potential 50-state internet regulatory regime, in which each state could adopt its own rules.

Sector Outlooks

Wireless Industry: Negative Outlook

Competitive intensity in the U.S. wireless market has remained relatively stable since T-Mobile and Sprint announced their proposed merger in April 2018. Since then, service revenue trends have remained positive while margins have expanded due to low device upgrade rates. That said, we expect competition to increase in 2020 and therefore, we maintain our negative industry outlook. We believe that the deployment of 5G networks and availability of new handsets (including the likely launch of a 5G-capable iPhone in the second half of 2020) could intensify competition and lead to higher churn and upgrade rates. With the entrance of Altice into the marketplace with aggressive pricing, competitive pressures from cable will likely ramp up in 2020. Furthermore, leverage for the sector remains elevated and the potential for spending in upcoming auctions, as well as higher capex, pose risks to credit quality over the next couple of years.

That said, a lot will depend on the outcome of T-Mobile and Sprint. Resolution on the outcome of the transaction would provide greater clarity around the broader wireless market, whether it's approved or not. If the deal is permitted, we would expect T-Mobile to largely focus on the integration and building out Sprint's 2.5-GHz spectrum, and we believe that a wireless market with three nationwide carriers would lessen competitive intensity in the near term. However, this positive development would be partly constrained by carriers' aggressive marketing of 5G capabilities, mid-band spectrum license acquisitions for 5G deployments, and higher capex over the next few years, which could constrain overall credit quality improvement. If the deal is rejected, we would expect T-Mobile to resume its aggressive behavior with lower pricing and more "uncarrier" initiatives. We would also expect T-Mobile to be aggressive in upcoming auctions, particularly the C-Band, which could push up valuations. Sprint's operating and financial challenges on a stand-alone basis should enable AT&T and Verizon (along with T-Mobile) to take share.

We expect service revenue growth for the industry to moderate but remain positive at around 1%-2% in 2020 from 2%-3% in 2019 as competition from cable increases and because of mature market conditions. The launch of 5G could also prompt increased switching activity, resulting in higher churn. At the same time, we believe improved spectral efficiency from 5G technology will be largely offset by modestly higher upgrade rates, resulting in limited margin improvement over the next year. Longer term, we have a favorable view of the industry assuming that wireless companies can monetize their investments in 5G from growth in IoT, which comprises internet-connected devices, vehicles, buildings, and "smart" devices.

Chart 10

Chart 11

Wireline Industry: Negative Outlook

Our outlook for the U.S. wireline industry remains negative. Revenue continues to fall because of secular industry declines and aggressive competition. Given the stress on the sector, in our view wireline operators have limited access to the capital markets and less financial flexibility to address longer-term debt obligations, as evidenced by significantly wider bond spreads, unless they can dramatically improve operating and financial performance. That said, we believe that near-term maturities are manageable for the sector overall.

Chart 12

Weak fundamentals and limited financial flexibility contributed to downgrades for Frontier Communications Corp. and Consolidated Communications Inc., as well as the Chapter 11 bankruptcy filing for Windstream in 2019. We lowered the rating on Frontier twice in 2019, ultimately to 'CCC-', following the drawdown of the remaining $499 million available under its $850 million revolving credit facility. We believe this buildup of cash signals a high likelihood that the company will initiate a distressed exchange or file for bankruptcy over the next six months.

In the consumer segment, wireline operators continue to lose broadband customers to cable operators, which have a superior broadband product. Similarly, in the SMB segment, cable operators are taking market share from the incumbent phone companies and are expanding their capabilities through network upgrades. While cable is targeting larger business customers, we still believe wireline companies face less risk in the enterprise segment. That's because this category is less likely to churn than smaller business customers and isn't subject to significant competition from cable. Still, we believe that new cloud-based technologies such as software-defined wide area networks (SD-WAN), which are used to connect enterprise networks, pose a threat to the industry since they are more flexible, open, and cheaper than traditional WAN technologies.

In order to conserve cash and focus on debt reduction, both CenturyLink and Consolidated cut their dividends. CenturyLink, the largest pure wireline company in the U.S., is benefitting from cost savings from its acquisition of Level 3, which has enabled it to maintain stable EBITDA and expand margins despite declining revenue. Coupled with its $1.2 billion dividend reduction, the company reduced S&P Global Ratings-adjusted leverage to 4.1x as of Sept. 30, 2019, from 4.5x at year-end 2018. That said, while CenturyLink has two more years remaining in its transformation program to reduce expenses and maintain current EBITDA levels, there is heightened risk longer term that EBITDA will decline unless it can improve top-line trends.

Similarly, revenues at Consolidated Communications are declining around 4%-5% and the company has fully achieved its cost synergies from the acquisition of Fairpoint. Despite eliminating its dividend in April 2019, adjusted debt to EBITDA has not improved significantly and remains above 5x. We believe it will need to find additional cost-saving opportunities to offset revenue declines in order to improve leverage.

Notwithstanding ongoing revenue declines in 2019, wireline issuers improved margins through cost reduction initiatives and synergies from previous acquisitions. In 2019, we expect mid-single-digit revenue declines but overall margin improvement to around 39% from 36% in 2018. For 2020, cost-saving opportunities will diminish and, as a result, we do not expect any meaningful margin improvement while revenues decline around 4%-6% on average.

Overall, wireline issuers face a difficult dilemma. They need to allocate capital to fiber deployments to better compete with cable, but at the same time they have limited financial flexibility and need to use FOCF for debt reduction in the face declining earnings.

Chart 13

Cable Industry: Stable Outlook

The outlook for the U.S. cable industry remains stable, reflecting our view that continued growth in high-margin broadband, commercial services and strong political advertising revenue will more than offset the earnings impact of video subscriber declines in 2020. We believe that the favorable mix shift toward high-speed data should result in modestly higher EBITDA margins, partially offset by wireless startup losses and shrinking video profitability. Overall, we expect top-line growth of about 5%, with weighted average margins improving by about 50 bps to approach 39%.

Chart 14

We believe cable companies can continue to increase high-margin broadband revenue for the next two years through a combination of subscriber growth and higher prices as subscribers demand faster internet speeds with rising data consumption.

• We expect that the average number of broadband subscribers will increase by about 4% in 2020 from a combination of new home builds and increasing penetration. We expect new home builds to increase by about 1.3 million (about 1%) in 2020, while higher penetration should come from both market share gains from slower copper-based services and greater overall broadband adoption. Currently, only about 82% of U.S. households pay for fixed broadband connections and we expect this number to increase gradually over time and eventually surpass the peak of 88% that pay-TV reached in 2010 given the importance of the internet and the potential for future government support.
• We expect average revenue per user will increase by a weighted average of about 5% in 2020. We expect the trend of migration toward higher-priced, faster tiers will continue as data consumption increases with more devices attaching to the network, more IP video consumed, growth in online gaming, and new applications emerge. For reference, Charter now offers a minimum speed of 200 megabits per second (Mbps) across most of its footprint, Altice USA's average speed is now four times higher than it was just three years ago at over 200 Mbps, and Comcast has indicated that average usage has doubled on its network over the past three years.

Chart 15

Chart 16

Telecom Infrastructure Industry: Stable Outlook

We continue to have a stable industry view for U.S. telecom infrastructure providers, which includes towers, data centers, and fiber providers. That said, growth trends, credit quality, and rating trends are largely bifurcated. While larger, more established issuers such as Equinix and Crown Castle have continued to perform well, the smaller and more highly leveraged providers that are typically rated in the 'B' category experienced both execution missteps from acquisitions and slowing organic growth that resulted in weaker top-line performance and rating downgrades in 2019. We believe these companies are at risk if economic or capital market conditions deteriorate.

Data centers:   While we still have a favorable view of fundamentals in the data center industry, supported by increased IT outsourcing, data growth, and greater application complexity, rating trends have been negative and sector performance has become increasingly bifurcated. From a demand perspective, we believe enterprise growth will continue as roughly 60% of applications still reside on-premises, leaving substantial runway for growth in third-party data centers because businesses' efficiency, reliability, and physical security can improve by outsourcing to a dedicated facility. That said, 2019 was a year of underperformance among many data center issuers that resulted in several downgrades, especially among the smaller providers. Ratings pressure has stemmed from a combination of integration missteps for some operators and slowing colocation growth as certain applications migrate to the cloud and supply increases in certain markets, resulting in pricing pressure..

Data center operators represent a diverse group of companies whose value proposition, strategy, and outlooks vary significantly. Undoubtedly, long-term growth in data traffic will continue to drive greater need for communications infrastructure housed in these facilities. However, shifting technology and the ever-increasing influence of cloud service providers will continue to distinguish between interconnected facilities that provide an attractive ecosystem, in our view, and various other business models that will find it more difficult to differentiate themselves.

We believe that interconnected data center facilities with carrier-neutral ecosystems will continue to have greater demand among cloud and network providers, as well as enterprise customers, enabling greater pricing power than data centers that do not have carrier-neutral hotels. Since 2018, we have taken one positive and seven negative rating actions among the eight data center operators that we rate (excluding wholesale providers that have more real estate-line characteristics), with most negative rating actions on providers that do not have carrier-neutral facilities. Chart 17 displays an average of our forecast credit metrics segmented out by providers that currently do or do not generate significant (greater than 15%) revenue from interconnection. We expect providers with carrier-neutral interconnection facilities will continue to outperform those that lack interconnection capabilities, since we believe that interconnection provides a key competitive advantage. Operators that lack interconnection are forced to primarily compete on price or upselling their customers with managed services, which tend to have lower margins, shorter contract lengths, and higher churn than colocation.

Chart 17

Fiber providers:  Similar to the data center industry, the largest independent U.S. fiber providers underperformed our expectations in 2019 due to aftershocks from acquisition stumbles and softer fiber installations. Results for GTT Communications, which has grown aggressively through acquisitions in recent years, worsened this year as a result of large restructuring expenses coupled with higher churn and lagging install growth due to sales force limitations. Further, its earlier acquisitions of fiber-heavy Hibernia Networks and Interoute clashed with its asset lite model and the company is now looking to dispose of some of the fiber and data center assets it acquired. Meanwhile, Zayo's fiber business missed its sales growth targets while overall revenue continued to be weighed down by legacy wireline operations acquired from the purchases of Integra and Allstream.

While Zayo's pending acquisition by EQT Infrastructure and Digital Colony could help it streamline operations, we believe the take-private transaction would be credit negative. The deal could result in adjusted debt to EBITDA rising above 6.5x, our trigger for a lower rating, which would constrain its financial flexibility if its operational focus and sales execution do not improve. The transaction is expected to close in the first half of calendar year 2020.

Despite these company-specific challenges, fiber industry fundamentals remain sound, in our view. Our favorable long-term view is based on rising demand for bandwidth, which benefits all fiber providers. Demand continues to be driven by mobile data traffic, enterprise connectivity and IT outsourcing to data centers and the cloud, wireless backhaul and carrier network densification, migration of media content to OTT, and connected devices. That said, we believe the growth in tier 1 markets may slow due to competition and maturing conditions. In contrast, operators in tier 2 and 3 markets may experience better growth trends because of less competition and penetration. For 2020, we expect overall revenue growth for the industry to be in the low- to mid-single-digit percent area.

Chart 18

Towers operators:  U.S. tower operators benefited from a healthy demand environment in 2019 as the wireless carriers continued to upgrade their networks and deploy spectrum. Despite the favorable environment, American Tower's results were hurt by choppier conditions in India, where it has been working through an extended period of carrier consolidation caused by disruptive competition in India's wireless market, which has resulted in higher churn. Furthermore, delays around the outcome of the T-Mobile/ Sprint merger have resulted in slower U.S. leasing trends recently, which could continue into the first half of 2020.

As T-Mobile and AT&T work toward their completion of spectrum deployments (AT&T FirstNet and T-Mobile 600 MHz licenses), we believe that revenue growth trends could slow in 2020, in line with our assumptions for modestly lower wireless capital spending. Still, we expect domestic tower leasing revenue growth to be solid, in the 5%-6% range, in 2020 although modestly lower than 2019 levels. Investment in 5G will likely be constrained until new mid-band spectrum is made available to the carriers, limiting upside in 2020.

Our longer-term view for the tower operators is favorable, as they are poised to benefit from healthy demand conditions in U.S. and international wireless markets, reflecting continued network investment from wireless operators to accommodate greater demand for mobile data and video. While demand fundamentals remain intact in international markets, we believe risks to performance are heightened due to industry consolidation, currency fluctuations, and political uncertainty. In India, American Tower was signaling a return to normalized growth by the end of 2020 following a long cycle of carrier consolidation, but the recent Supreme Court ruling on the definition of adjusted gross revenue (allowing the government to collect $13 billion in unpaid license and spectrum fees from carriers) has made 2020 growth prospects less certain. Further, the judgement could cause financial distress for some carriers, spur additional consolidation, and prolong the period of heightened churn for tower operators in the region. At the same time, weakness in the Brazilian real has constrained growth for U.S. tower companies operating in Brazil.

Crown Castle's domestic focus and emphasis on small cell/fiber buildouts position it best to benefit from initial 5G deployments and 4G network densification, particularly in urban and suburban markets. Zoning and other regulatory approvals have slowed deployments and in some cases wireless carriers are self-deploying, which reduces opportunities for colocation and margin enhancement. Still, we expect small cells to make up an increasing share of carriers' network investment plans. As a whole, the small cell industry is expected to grow at a compound annual growth rate of 23% through 2024, compared to 2% growth for traditional towers according to S&P Global Market Intelligence.

We believe the impact of a merger between T-Mobile and Sprint would be manageable for all three rated U.S. tower operators given that only about 4%-6% of their revenue is at risk to site decommissions from overlapping sites. Additionally, given staggered contracts with average remaining terms of two to six years, any loss of revenue would be phased over a multiyear period. Further, DISH has the option to take over up to 20,000 of the 35,000 sites that T-Mobile plans to decommission, which would help the satellite provider satisfy its network build-out requirements. From an issuer perspective, American Tower is least exposed to a potential merger, in our view, with less than 20% of its revenue coming from the two wireless companies. In contrast, Crown Castle and SBA each derive about one-third of their revenue from T-Mobile and Sprint, in aggregate.

Chart 19

Chart 20