Morgan Stanley Names T-Mobile Top Pick Despite Starlink Competition Concerns

Morgan Stanley keeps T-Mobile as its top telecom pick (Overweight) but cuts key forecasts amid Starlink-led broadband disruption, projecting Starlink could reach ~16M US consumer/small-business broadband subscribers by 2030. It lowered Q2 2026 prepaid service revenue by ~$100M to $2.45B and wholesale service revenue by ~$55M to $635M, while expecting postpaid ARPA growth to partially offset. Longer-term, it trims 2028–2030 fixed wireless access and fiber net additions by ~600k and ~100k subscribers annually, and revised its TMUS price target from $260 to $230 (to mid-year 2027). T-Mobile also declared a $1.02 per-share quarterly cash dividend and saw new coverage from Wells Fargo at Equal Weight.

Analysis

The market is likely underestimating how asymmetric this is for cable: even a slow bleed in broadband adds can matter more than the absolute subscriber loss because the best remaining cable households are the highest-ARPU, lowest-churn cohort. That creates a negative mix loop over 1-3 years: weaker growth, lower pricing power, and eventually higher promotional spend to defend the base, which can pressure free cash flow and multiples even if top-line declines look manageable.

TMUS is the cleaner relative winner because its capital-light broadband substitute offers optionality without the stranded-fiber problem. The key second-order effect is that satellite competition may actually validate wireless as the preferred “good enough” alternative for price-sensitive households, supporting FWA penetration and limiting cable’s ability to reprice aggressively. The near-term risk is that the street extrapolates Starlink too quickly into 2026-27 numbers while execution remains constrained by capacity, install friction, and customer service quality.

The contrarian view is that the move in cable names may be less about Starlink itself and more about a secular demand plateau that was already in motion. If so, the real trade is not a binary satellite winner but a persistent valuation haircut on levered distribution assets with limited organic growth. Falsifiers: cable net adds stabilizing for two consecutive quarters, TMUS FWA growth decelerating sharply, or evidence that Starlink adoption is materially below a low-single-digit share of the broadband market by 2027.

Key Decisions for Investors

• Long TMUS / short CHTR as a 6-12 month pair: TMUS has better pricing power, lower stranded-asset risk, and cleaner capital returns; use strength in cable to build the short. Falsify if CHTR broadband losses reaccelerate less than expected or TMUS FWA margins compress.
• Long TMUS / short CMCSA as a slower-moving version of the same thesis for investors who want less balance-sheet leverage risk. Best entry is on any post-earnings relief rally in CMCSA; thesis weakens if CMCSA guides to stable broadband adds.
• Buy TMUS on pullbacks for 12-18 months, targeting relative multiple support from capital-light growth plus dividend support. Risk/reward improves if the stock de-rates on any short-term worry that is more relevant to cable than to wireless.
• Use CHTR or CMCSA put spreads 6-9 months out rather than outright shorts if borrow or carry is a concern. This expresses the structural downside while defining risk if Starlink adoption proves slower than consensus expects.
• Set an alert on Starlink adoption milestones and cable quarterly broadband net adds; if cable losses remain near historical levels without a worse trend, trim shorts because the thesis is becoming crowded rather than improving.