Initiated 2026-10-23 · Price $27.51 (as of 2026-10-23 intraday, stockanalysis) · Mkt cap $9.27bn · Communication services / subscription audio · Model: verified

Rating: Sell — Conviction: Low

SiriusXM is a cash-rich but shrinking subscription business wearing about $9.5bn of debt, and at $27.51 the equity already prices something close to a stabilized subscriber base. Q2's record-low churn is real, but revenue has fallen three years running, the car dashboard is being taken over by phone-projected apps, and the satellite royalty rate resets after 2027. Once share-based pay is counted as the cost it is, the Base case sits below the price, and the downside if the base keeps eroding is far larger than the upside if it holds.

Conviction tests (3a-v-c): T1 fail (Sell → Hold at −1pp: Base midpoint $27.07, −1.6%) · T2 pass (Gordon $22.34 and exit-multiple $23.08 both below price) · T3 pass (historicals from EDGAR XBRL and the 10-Ks; no est inputs) · T4 pass (the 2028+ royalty reset is inside the bear case; Q3 results on 2026-10-29 are not a >25% event)

The call leans on the discount rate and on counting ~$200mm a year of stock compensation as a cost; the ~15% yield on company-defined FCF, which adds it back, is the main argument against.

Model value range vs price
Bear $3.76Base $22.71Bull $44.48Price $27.51

Business overview

SiriusXM sells subscription audio: satellite and streaming radio in ~33mm US accounts (31.26mm self-pay at Q2 2026), mostly activated through factory-installed radios in new and used cars. The SiriusXM segment was $6.42bn of FY2025's $8.56bn revenue (75%). Pandora and Off-platform added $2.14bn, which is mostly advertising, including a growing podcast network. Earnings turn on three things: (1) self-pay subscriber count × ARPU ($15.11 FY2025, +1% in Q2), (2) the cost of content and the statutory sound-recording royalty (15.5% of gross satellite revenue through 2027), and (3) Pandora/podcast ad pricing. Satellite and content costs are largely fixed, so lost subscribers fall almost straight to EBITDA.

Competition

The thesis rests on the car dashboard, and the competition there comes from Apple CarPlay and Android Auto projecting Spotify, YouTube, Apple Music and podcasts. Management itself says most in-car listening is still AM/FM (iHeartMedia's turf) and is launching cheaper music-only ($9.99) and ad-supported tiers to compete for it (Radio Ink, Feb 2026). SiriusXM's own Pandora shows how the streaming fight is going: monthly active users fell 7% year on year to 39.8mm in Q2 2026. Meanwhile Spotify is forecast to grow revenue ~13.9% a year over three years (stockanalysis consensus, 2026-10-23). Pressure would show up first in trial-to-paid conversion and self-pay net adds, not in price. Peers trade far apart: Spotify at 36.7x TTM EV/EBITDA (lease-inclusive, adjusted), the levered legacy names iHeartMedia and Nexstar at 10.6x and 10.0x, and SIRI at 7.4x.

Bull case

  1. Subscriber base stabilizes — record-low 1.4% self-pay churn and Q2's +22k self-pay net adds (first positive Q2 in four years) persist, ARPU keeps rising ~1%, and podcasts carry Pandora advertising (+5%), so revenue grows ~1–1.5%. Plays out if the cheaper tiers convert the AM/FM audience and new-car penetration holds. Model: rev_growth, terminal_growth
  2. Cost cuts and a lighter capex cycle — the restructuring program and the satellite build roll-off (capex 5.5% of revenue in 6M 2026 vs 7.6% in FY2025) push margin up ~2pp and capex toward 6%. Plays out if FY2026's raised guide (FCF ~$1.375bn) is the new run-rate, not a timing benefit. Model: ebitda_margin, capex_pct_rev
  3. The multiple stops pricing decline — with leverage at target, buybacks resume and a proven stable base re-rates toward 8.5x. Plays out if two or three more quarters of positive self-pay adds arrive. Model: exit_ev_ebitda

Bear case

  1. The dashboard is lost slowly — CarPlay/Android Auto and free streaming apps erode trials and conversion. Self-pay losses return to the ~300k a year of 2024–25 and widen, and revenue falls 3–3.5% a year. Plays out if Q2's improvement proves seasonal (6M 2026 is still −89k). Model: rev_growth, ebitda_margin, terminal_growth
  2. Royalty reset after 2027 — the 15.5% satellite sound-recording rate expires at end-2027. Each +1pp on ~$6.4bn of satellite revenue is ~$60mm of EBITDA (est.), and fixed content and satellite costs deleverage as revenue falls. Plays out if the CRB sets a higher 2028+ rate. The status of that proceeding was n/a (unverified) this session. Model: ebitda_margin
  3. Leverage on a shrinking base — $1.25bn matures in 2027 and $2.6bn in 2028 (incl. the convertible), refinanced well above 4–5% coupons, and the equity is ~50% of capital. Plays out if HY spreads widen or rates stay above 5%. Model: wacc, exit_ev_ebitda

Valuation & balance sheet

Metric (definition) Current Own history (range or 5y avg) Peers Source, as-of
EV/EBITDA, model basis (mkt cap − cash + debt carrying value) ÷ FY2025 EBITDA after SBC, ex-restructuring 7.56x n/a (unverified) n/a (different basis) 10-K, 10-Q, model, 2026-10-23
EV/EBITDA TTM (aggregator, lease-inclusive, adjusted) 7.42x n/a (unverified) SPOT 36.72x · IHRT 10.61x · NXST 10.04x stockanalysis, 2026-10-23
FCF yield TTM (aggregator, company FCF) 16.7% FCF $1,182mm / $1,015mm / $1,256mm FY2023–25 (10-K) SPOT 3.05% · IHRT 13.30% · NXST 14.29% stockanalysis, 2026-10-23
Forward P/E (consensus) 8.73x n/a (unverified) SPOT 36.34x · NXST 5.32x (looks distorted vs 31.3x TTM) · IHRT n/a (losses) stockanalysis, 2026-10-23

Model-implied value range (from model-summary.json; generic module, Gordon-growth DCF and exit-EV/EBITDA DCF, midpoints): Bear $3.76 · Base $22.71 · Bull $44.48 per share, i.e. implied returns of −86.3% / −17.4% / +61.7% vs $27.51. These ranges show how the bull and bear drivers translate into value; they are not price targets. The price sits between Base and Bull, so the market is already paying for a stabilized base and a modestly higher cash conversion than the Base case delivers. Neither method embeds a re-rating: Base exit 7.0x is a ~7% de-rating against today's 7.56x same-basis multiple, and the two methods agree within 4% ($22.34 vs $23.08). Tail sensitivity (SiriusXM segment is 75% of revenue): revenue −5% a year from Y2, margin falling to 22%, capex 8%, NWC −1.5%, exit 4.5x, WACC 9.5%, terminal −3% gives a $0.00 midpoint. At ~50% equity, a permanent decline leaves the equity little.

Balance sheet: net leverage 3.79x FY2025 model-basis EBITDA (company: 3.4x net debt/Adjusted EBITDA at Q2 2026), EBITDA/interest 5.7x. Liquidity: $174mm cash plus ~$1.9bn of revolver availability (est., from the $1,980mm available at 2025 year-end, less the $80mm drawn since). Nearest material maturity is $1.25bn of 5.00% notes in August 2027, then $2.0bn of 4.00% notes and the $575mm convertible in 2028. Ratings: n/a (unverified).

Model note: built and verified (LibreOffice recalculation matched all 3,207 formula cells). No unverified inputs; no assumptions without basis; no scenario CHECKs. Model EBITDA is operating income + D&A + impairment/restructuring, i.e. after share-based compensation. That is why Base Y1 levered FCF ($1,066mm) sits ~22% below the company's ~$1,375mm FCF guide: the company's figure excludes ~$200mm of SBC, and cash tax and working-capital timing explain the rest. The company gives no EPS guidance. Base Y1 EPS is $3.12. Narrow NWC excludes payables. The Sirius XM Canada stake (70%, equity method) is not counted, a small conservatism.

Relative value vs peers
MetricSIRISPOTIHRTNXSTPeer medianvs median
P/E (TTM)11.1x32.1xn/a31.3x31.7x−65%
P/E (forward)8.7x36.3xn/a5.3x20.8x−58%
PEGn/a1.4xn/an/an/a
EV/EBITDA (TTM)7.4x36.7x10.6x10.0x10.6x−30%
EV/Sales (TTM)2.2x5.4x1.5x2.9x2.9x−25%
P/B0.8x12.5xn/a2.3x7.4x−89%
FCF yield16.7%3.0%13.3%14.3%13.3%+3.4pp
P/E (TTM)SPOT32.1xNXST31.3xSIRI11.1x
P/E (forward)SPOT36.3xSIRI8.7xNXST5.3x
PEGSPOT1.4x
EV/EBITDA (TTM)SPOT36.7xIHRT10.6xNXST10.0xSIRI7.4x
EV/Sales (TTM)SPOT5.4xNXST2.9xSIRI2.2xIHRT1.5x
P/BSPOT12.5xNXST2.3xSIRI0.8x
FCF yieldSIRI16.7%NXST14.3%IHRT13.3%SPOT3.0%

As of 2026-10-23. One source and one definition for every company: aggregator TTM and consensus-forward multiples, lease-inclusive EV on the source's adjusted EBITDA. Not the model's own EV basis (see the report's valuation table). Quotes taken 09:52-10:05 ET with the market open, so multiples reflect intraday prices. Dashed line = peer median. Source: stockanalysis.com /stocks/<ticker>/statistics/, fetched 2026-10-23.

Notes. SIRI's page shows a PEG of 0.41, which implies ~21% EPS growth against the same page's 3-year EPS growth estimate of 3.55%; it does not rest on a credible growth estimate, so it is null. IHRT and NXST: n/a on the source. IHRT is loss-making (EPS -$1.83), so P/E is null. IHRT has negative book value (-$12.69/share), so P/B is null. NXST's forward P/E (5.32x) versus 31.27x TTM looks distorted by one-offs in TTM earnings; the source does not explain the gap.

Scenario stress test

Reasoned from the bull/bear drivers above. The model column comes from the scenario overlays (Base case + shock).

Scenario Effect Mechanism Magnitude Model Δ value vs Base ($/sh)
S1 Fast equity crash − Beta ~1 levered equity de-rates; fixed-rate notes, no funding channel Med −$2.67
S2 Slow bear / recession − Weaker auto sales cut the trial funnel, Pandora ads fall, churn rises; fixed-cost deleverage. Milder than Base − Bear ($18.95) because it is a two-year cycle, not permanent dashboard loss High −$6.55
S3 Rapid rate shock − Higher discount rate on a ~50%-debt structure; 2027–28 maturities reprice High −$4.62
S4 Slow rate grind − Gradual refinancing at higher coupons and a higher discount rate Low −$1.88
S5 Soft-landing cuts + Lower rates lift a levered, high-FCF-yield equity; steady auto sales and ad market Med +$2.58
S6 Recession-driven cuts − Same auto/ad/churn hit as S2; lower rates only partly offset High −$4.12
S7a Credit liquidity shock − HY issuer with $3.9bn due 2027–28; spread gap raises refinancing cost Med −$2.67
S7b Slow default cycle − Wider HY spreads, small-advertiser cuts, more involuntary churn Med −$2.60
S8 Stagflation − Talent, labor and memory/hardware costs inflate while price rises meet churn; rates rise High −$3.91
S9a Dollar spike 0 No material effect, not modeled (domestic revenue; Canada unconsolidated) Low $0.00
S9b Dollar slide 0 No material effect, not modeled Low $0.00
S10 Melt-up + Low-multiple cash yielder re-rates modestly; lags momentum Low +$1.67
S11 Energy supply shock − Gasoline spike cuts driving and auto sales (fewer trials), squeezes ad budgets; rates rise Med −$2.09
S12 Mega-cap/AI derating + Rotation toward low-multiple cash yielders; no fundamental channel Low +$1.00

Currently active/on watch per the playbook: S3 partially active (the macro log for 2026-10-22 shows the 10Y back below 5%); S8, S10, S11 on watch. S11 is the live one for SIRI: Brent near $113 intraday on 2026-10-22.

Model value change vs Base, by scenario
S2 Slow bear / recession−$6.55S3 Rapid rate shock−$4.62S6 Recession-driven cuts−$4.12S8 Stagflation−$3.91S1 Fast equity crash−$2.67S7a Credit liquidity shock−$2.67S7b Slow default cycle−$2.60S11 Energy supply shock−$2.09S4 Slow rate grind−$1.88S9a Dollar spike$0.00S9b Dollar slide$0.00S12 Mega-cap/AI derating+$1.00S10 Melt-up+$1.67S5 Soft-landing cuts+$2.58

What would change the call

Upgrades if: self-pay net adds are positive for full-year 2026 and again in H1 2027, revenue returns to growth, and a 2028+ royalty rate is set at or near 15.5%. Or the price falls toward the low $20s without a deterioration in subscribers. Downgrades if: n/a in rating terms (already Sell). Conviction rises if Q3/Q4 self-pay losses resume at the 2024–25 pace, or if FY2026 FCF lands below the ~$1.375bn guide.

Watch items

Sources