STZ › analyze
STZ · Analyze
0. Knowledge check
Knowledge/kb.py find STZandfind "Constellation Brands"returned no live note on this name — first coverage of STZ in the knowledge base.Themes/defensive-income-fields.md(2026-07-29) covers "Healthcare, energy, industrials/materials, staples & telecom" broadly and was rejected on business-model grounds, but that sweep did not touch beverages/alcohol specifically, so it doesn't answer this question — treated as fresh research.- Playbook pitfalls checked before trusting vendor numbers:
- ADR pitfalls (
pitfall-adr-*) do not apply. STZ is a US domestic filer (Rochester, NY), not an ADR — no depositary ratio, no dual-currency EV, no FPI share-count issue. pitfall-vendor-forward-eps-is-the-wrong-fiscal-yearfired. Yahoo'sforwardPE(10.61x) pricesforwardEps$12.39 — cross-checkingepsCurrentYear/priceEpsCurrentYear($11.80 / 11.14x) shows the current fiscal year (FY2027, ending Feb-2027) is actually priced at ~11.1x, one year cheaper-looking than reality. The company's own guidance (comparable EPS $11.20–$11.90) confirms $11.80 is the right FY2027 number — Yahoo'sforwardPEis quietly FY2028. All valuation work below uses the correct current-year multiple, and the watchlist-style trim is set on ttm, not fwd, per that note's guard.principle-down-a-lot-is-not-cheap(with its RBLX companion clause) governs the whole valuation section: STZ is 22% off its 52-week high and trading at less than half its pre-2025 historical multiple — the question the note forces is whether the earnings base reset along with the price. Addressed in §3.pitfall-dyt-inverts-when-price-caused-the-yieldfired on the dividend-yield read — addressed explicitly in §4.pattern-margin-expansion-is-a-finite-growth-leverandpattern-buyback-manufactured-eps-screens-as-growth— checked against STZ's growth algorithm in §2 (Moat) and §1 (Fundamentals); neither is the dominant driver here, but the buyback contribution to per-share value is large enough to name explicitly.
1. Fundamentals
Data source: python .mcp/fin.py STZ --news (4 fiscal years, FY ends last day of February)
cross-validated against roic.ai profitability/credit/cash-flow/per-share endpoints (roic's free
tier returned only 2 annual periods per the known pitfall-roicai-free-plan-caps-history-at-two-years
cap, so it served as a cross-check on FY2025–FY2026, not additional history) and Yahoo quarterly
statements for the just-reported Q1 FY2027.
Snapshot
| Metric | Value | Metric | Value |
|---|---|---|---|
| Price (2026-08-27 close) | $131.43 | Market cap | $22.4B |
| 52-week range | $126.45 – $168.60 | All-time high (2023) | $274.87 |
| TTM P/E | 12.81x | Current-FY (FY27) P/E — corrected | ~11.1–11.4x |
| EV/EBITDA | 9.60x | P/B | 2.73x |
| Dividend yield | 3.06% | 5yr avg div yield | 1.78% |
| Payout ratio (GAAP) | 39% | Beta | 0.40 |
| Debt/Assets (FY26) | 51.2% | Net Debt/EBITDA (FY26) | 3.33x |
| ROE (FY26) | 22.5–23.7% | ROIC (FY26) | ~10.3% |
Free cash flow — the backbone
| FY (end Feb) | Revenue | FCF | FCF/share | Capex | Buybacks | Dividends paid |
|---|---|---|---|---|---|---|
| 2023 | $9.45B | $1.72B | — | -$1.04B | -$3.20B | -$0.587B |
| 2024 | $9.96B | $1.51B | — | -$1.27B | -$0.250B | -$0.654B |
| 2025 | $10.21B | $1.94B | $10.68 | -$1.21B | -$1.12B | -$0.732B |
| 2026 | $9.14B | $1.79B | $10.23 | -$0.875B | -$0.924B | -$0.716B |
FCF CAGR: only ~1.4% over the 3 usable years (FY23→FY26) — flagged as a genuine data-depth
gap against the framework's 5–8yr standard: fin.py and roic's free tier both cap out around
3–4 annual periods for this name. FCF is essentially flat-to-choppy over the period, not a
compounder — but the composition changed for the better: capex fell as the Mexico brewery
build-out matured and the low-margin wine estate was pruned, so FCF held up on a shrinking capex
base even as revenue fell.
Why revenue fell 10.5% FY25→FY26 despite flat-to-up EPS: Constellation closed the sale of its "mainstream" wine portfolio (Woodbridge, Meiomi, Cook's, SIMI, J, Robert Mondavi Private Selection, plus related vineyards) to The Wine Group on 2025-06-02, retaining only wine/spirits brands priced predominantly $15+. This was a deliberate quality upgrade of the revenue base, not organic decline — the divested brands were the low-margin, low-growth tail of the portfolio. Diluted EPS actually improved over the same window (FY24 $9.39 → FY26 $9.61) despite the ~$1B of revenue leaving, which is the clearest evidence the earnings base was curated rather than eroded — the key test the "down a lot is not cheap" principle demands.
Capital allocation
Aggressive, buyback-first allocator: -$3.2B buyback in FY23 alone, tapering since (management
appears to be pacing repurchases to leverage, not funding them with new debt — DebtIssued in
FY23 was $3.34B against $3.2B of buybacks that year, which is the one year worth a flag: the FY23
buyback was partly debt-funded, not pure FCF). FY25–FY26 buybacks ($1.12B, $0.92B) ran within FCF
generated in those years — the "buyback-funded-by-balance-sheet" pattern does not describe the
current posture.
Share count: diluted shares 192.5M (FY23) → 175.6M (FY26), a -3.0% 3yr CAGR — a genuine, sustained per-share tailwind, not a manufactured one: operating income and FCF did not shrink over the period the way the buyback pattern requires to be a red flag (op income $2.84B→$2.86B FY23→FY26, roughly flat-to-up even through the divestiture).
Debt-to-Assets & leverage
Debt/Assets sits at ~51% (FY26), consistent with ~51–56% over the last three years — elevated for a "quality compounder" screen, but this is a highly cash-generative staple, not a cyclical, and interest coverage is solid: EBITDA/interest 8.68x, operating income/interest 7.9x (FY26).
Caution on FY25 credit ratios: roic.ai's FY25 credit-ratio pull shows net-debt/EBITDA of 14.3x and EBITDA/interest of 1.9x — both look alarming in isolation, but they are artifacts of the ~$3.28B non-cash impairment charge that year (wine-and-spirits goodwill plus the Canopy Growth equity-stake write-down, detailed in §2) crushing reported EBITDA to $801M. Operating income that year was still $3.36B and interest coverage on that basis was a healthy 8.0x — the credit-ratio distortion is a one-off accounting artifact, not a real solvency signal, and it reversed cleanly in FY26 (net debt/EBITDA back to 3.33x). Flagging this explicitly so the FY25 numbers aren't mistaken for a leverage crisis.
Per-share metrics
Revenue/share $52.05, FCF/share $10.22-10.68 (last two years), BVPS $46.77-$48.15. FCF/share comfortably covers the $4.08-4.12/share annual dividend — FCF payout ratio ~38-40%, ample room.
Dividend Grower overlay (applies — 3.06% yield, 51 straight years of increases per company
disclosure)
| FY | Div/share (annualized) | YoY growth |
|---|---|---|
| 2020 | $3.00 | — |
| 2023 | $3.56 | ~5.8%/yr (3yr) |
| 2024 | $3.96 | +11% |
| 2025 | $4.04 | +2% |
| 2027 (current run-rate) | $4.12 | +2% |
Dividend growth accelerated into FY24-25 (double-digit) then decelerated sharply to ~1-2%/yr over the last two annual raises — coinciding exactly with the earnings-impairment year and the tariff/demand-softness period. Read this as prudence, not distress: FCF payout is still only ~39-40%, well inside a safe band, so the deceleration looks like management choosing to protect the balance sheet and the buyback rather than a coverage problem. Not a golden flag, but not a red one either — worth watching at the next raise (typically announced with Q4 results, ~April).
2. Moat & Competitive Advantage
ROIC ~10.3% (FY26), gross margin 51.6-53% (mild compression from ~53% historically, partly tariff-driven), operating margin a strong 31-36%. The moat is real but not expanding — a "stable, high-quality, not accelerating" read, which is the correct one for a mature staple.
Revenue-stream map: Beer (~80% of revenue — Corona Extra, Modelo Especial, Modelo Chelada, Pacifico, Victoria) is the growth engine and franchise; Wine & Spirits (~20%, now premium-only post-divestiture — Robert Mondavi Winery, Kim Crawford, Ruffino, The Prisoner, Casa Noble, High West) is a smaller, higher-margin but currently-shrinking complement.
Sources of moat, named explicitly: - Brand/intangibles — Corona and Modelo are the #1 and #2 imported beers in the US by a wide margin, built over decades of marketing and distribution investment. Modelo Especial overtook Bud Light as the #1 beer overall in the US in 2023. - Exclusive US distribution rights to the Mexican-brewed Corona/Modelo/Pacifico portfolio (via the original Grupo Modelo agreement, now wholly integrated) — a structural, contractual moat a domestic rival cannot simply replicate. - Efficient scale in Mexican brewing capacity (Nava, Obregón facilities) built specifically for the US import category. - Weak network effects / switching costs — beer is a repeat-purchase consumer good; loyalty is brand-driven, not structurally locked in, which is the honest limit of this moat.
Adversarial stress-test — "how would a well-funded rival attack this?" AB InBev (which sold the US rights to Constellation but retains global brewing scale) and Molson Coors could both compete on price/promotion and shelf space; craft and RTD/hard-seltzer entrants fragment the "premium" category from below. But replicating Corona/Modelo's decades of Hispanic-American and mainstream-US brand equity and exclusive distribution economics is not something a rival can do on a normal competitive timeline — this is a real, durable moat against direct imitation. The more dangerous attack vector is not a rival brand, it's category-level substitution (below).
Disruption vectors, 5-10yr: 1. Alcohol-moderation / GLP-1 substitution — the sentiment-driving fear across the entire sector in 2026 (see §4). This is a genuine secular question, not company-specific execution failure, and no brewer/vintner has a clean answer to it yet. 2. Trade-policy/tariff risk — ~80% of revenue rides on Mexican-brewed beer, ~41% of which ships in aluminum cans now subject to a 50% US tariff. This is a real, ongoing structural cost headwind unique to STZ's geographic concentration, not shared by a domestically-brewed competitor. 3. RTD/hard-seltzer and spirits-based cocktail fragmentation continuing to pull share from traditional beer occasions. 4. Cannabis legalization as a long-horizon substitute good — STZ's own (now-arm's-length, largely written-off) Canopy Growth stake is the company hedging this exact risk, and having walked away from it (see below) is arguably the right read of how that bet played out.
The Canopy Growth legacy investment — moat-adjacent but not moat-relevant today. Constellation took a $1.1B writedown on its ~40% Canopy stake in 2022, then converted to non-voting exchangeable shares and gave up its board seats in 2024 — effectively exiting active involvement while retaining residual (~26.3M exchangeable shares) upside optionality. This explains much of the "Gain/Loss on Sale of Security" and "Earnings from Equity Interest" noise in the income statement and the bulk of the ~$3.28B non-cash impairment charge in FY2025 that produced the GAAP net loss that year (alongside a wine-and-spirits goodwill impairment tied to the divestiture). This is now a closed chapter, not an ongoing risk — flag it so it isn't mistaken for a live overhang.
Evergreen assessment: Beer/alcohol is a durable, multi-generational consumer category and
STZ's beer franchise specifically is a real, cash-generative moat. But the company's growth
algorithm for the last decade — premiumize the beer portfolio + prune the wine tail + shrink
share count — is now running into two simultaneous limits: (1) premiumization mix-shift has
largely already happened (gross margin is flat-to-compressing, not expanding, so
pattern-margin-expansion-is-a-finite-growth-lever is not yet the dominant story here, but
worth re-checking if margin guidance resumes an "expand X bps/yr forever" framing), and (2)
organic beer volume is now shrinking (depletions -0.3%, shipments -3.3% in Q1 FY27) for the
first time in years. The moat is intact; the growth engine that used to sit on top of it has
stalled.
3. Valuation
Applying models conditionally, per framework: STZ is a profitable, dividend-paying mature compounder — Graham, DDM, DYT, and Bogle all apply; weight Graham/DDM as the conservative floor and Bogle for the total-return picture that credits multiple normalization.
| Model | Inputs | Output |
|---|---|---|
Graham IV √(22.5·EPS·BVPS) |
EPS(ttm) $10.26, BVPS $48.15 | $105.43 |
| DDM (Gordon growth) | D1 ≈ $4.33 (5% growth), r 8-9%, g 4-5% | ~$105-115 |
| DYT | Yield 3.06% vs 5yr avg 1.78% | See caveat below — do not read this as a clean buy signal |
| Bogle expected return | Div yield 3.06% + earnings growth 4-6% ± P/E normalization | ~8%/yr (flat multiple) to ~14-16%/yr (multiple reverts toward historical band) |
Graham and DDM both land in the $105-115 zone — meaningfully below the current $131 price. This is the conservative, no-re-rating-credit floor these models are built to produce for a name with real book value and a stable-not-growing earnings profile; the framework explicitly notes Graham tends to understate fair value for names with a real brand moat and growth optionality, so treat $105-115 as a bear-case floor rather than the target.
DYT caveat — pitfall-dyt-inverts-when-price-caused-the-yield applies directly. The yield
expanded from 1.78% (5yr avg) to 3.06% almost entirely because the price fell (~$230-270 in
2022-24 down to ~$126-155 through 2025-26), not because the dividend grew into a higher yield —
the dividend itself grew only ~2%/yr over the same window (see §1). A price-driven yield spike is,
per that note, "a question, not a signal" — it says the market has gotten more pessimistic,
not that the stock is automatically cheap. Read alongside the fundamentals (dividend still
comfortably covered at ~39% FCF payout), the elevated yield is consistent with value but does
not independently prove it.
Multiple-normalization view (the load-bearing case for the verdict): roic.ai's own FY26
closing-price band shows STZ traded 13.2x-20.5x P/E within the fiscal year just ended — the
current TTM multiple of 12.8x and corrected current-FY-forward multiple of ~11.1-11.4x sit
below STZ's own most recent annual band, and far below the ~20-27x multiple the stock
carried through 2022-2024 (at $230-270 against $9-13 diluted EPS). Per
principle-down-a-lot-is-not-cheap's companion clause, the pass is only meaningful if the
earnings base hasn't reset — and per §1, EPS held up (even improved) through the revenue
decline because the divested revenue was low-margin. This reads as a genuine below-band
value case, not a shrinking-denominator trap — with the caveat that organic beer volume
decline is a live, ongoing risk that could still force a real earnings reset if it continues
(see §4).
Fair value range: $135-165, built on 12-14x a normalized $11.20-11.90 FY27 comparable EPS guide (company's own number, corrected per the knowledge-check above) — below STZ's own 20-27x historical peak (acknowledging the growth deceleration is real and durable enough that a full re-rating to peak multiples isn't the base case), but above the Graham/DDM no-credit floor and above where the stock trades today.
Entry zone: $125-140 — the current price ($131.43) sits inside this zone already.
Trim: 17x ttm (not fwd — per the knowledge-check, STZ's vendor forward field is one fiscal
year mislabeled, so the trim is set on the trailing multiple to avoid inheriting that defect).
At the current TTM EPS of $10.26, 17x ttm ≈ $174 — a multiple that would represent real
normalization without assuming a return to euphoric premiumization-era pricing. This dollar
level moves automatically as trailing EPS changes — re-set the multiple, not the dollar figure,
at the next /analyze or /value.
4. Sentiment — why STZ is near its 52-week low right now
The move is demand-driven and tariff-driven, not company-specific execution failure, and it is not new news as of this week — it is a slow bleed that has been running since roughly September 2025.
Price trajectory: STZ traded $220-270 through 2022-2024, cracked to the $160-175 range in Jan-Feb 2025 on the first weak-demand/tariff signals, held $150-190 through mid-2025, then dropped hard to a 52-week low of $126.45 in the Sept-Nov 2025 window on a "sobering" earnings forecast — beer segment net sales guidance was cut from flat-to-+3% growth to -4% to -2% at that time, driven by CEO commentary (leadership had just transitioned to Nicholas Fink) that high-end beer demand had slowed, with fewer shopping trips and lower spend per visit. The stock partially recovered to the mid-$150s by Jan-Feb 2026, then round-tripped back down to $130-138 through May-August 2026 on continued soft prints and tariff cost pressure.
The specific, current drivers, as of the Q1 FY2027 print (2026-06-30) and since: 1. Tariffs — real and quantified. A 50% US tariff on imported aluminum (raised from 25% in March 2025) hits STZ because ~41% of its Mexican beer packaging is aluminum cans, and Mexican beer imports are not exempt the way some other alcohol categories are. Company-quantified impact: ~$20M over three quarters, plus an operating-margin drag of roughly 150bps in the quarter it first hit. This is a real, structural cost specific to STZ's Mexico-concentrated supply chain — a domestic-only competitor doesn't carry it. 2. Weakening core demand. Q1 FY27 (ended 2026-05-31): revenue -3.3% YoY to $2.43B, beer shipment volumes -3.3%, beer depletions (distributor sell-through) -0.3%. Adjusted EPS $3.43 beat consensus ($3.19), and management reaffirmed full-year comparable EPS guidance of $11.20-$11.90 and beer net-sales guidance of -1% to +1% — i.e. a beat-and-hold quarter, not a fresh deterioration, but the underlying volume trend is still negative. 3. Analyst split, not consensus bearishness. Of the 16 recent actions pulled: Citigroup and TD Cowen both upgraded to Buy in Mar/Apr 2026 (targets $175, $190); Roth ($209), Needham ($185), BMO ($190), RBC ($185), UBS ($175), Wells Fargo ($170, Overweight) all carry Buy/ Outperform ratings with targets 30-60% above spot. Against that: BofA (Underperform, $145), Barclays (Equal-Weight, cut to $132 — essentially at spot), JPMorgan (Neutral, $165), Morgan Stanley (cut hard from $183 to $158, Equal-Weight), Deutsche Bank (Hold, $150). Mean target $170.83 implies ~30% upside from here — the sell-side is not capitulating on the name, but the price-target cuts across nearly every firm in mid-2026 track the same demand-softness story, not a single shock. 4. No insider-buying signal at these levels. Insider transaction history shows mostly RSU-vesting "D" entries (non-signal) plus a handful of genuine open-market sales in the $143-169 range over the last year (Bourdeau $143 May-26, Hernandez $154 Apr-26, Glaetzer $154 Feb-26) — modest, not alarming, but not a golden flag either. The one real open-market purchase on record (Chairman Baldwin, 2,000 shares at ~$170) was July 2025, well above today's price — management has not stepped up to buy at the current 52-week-low band. Institutional ownership (88%) is stable, led by BlackRock, Capital World, Vanguard.
Is this sentiment/cyclical (opportunity) or structural (trap)? Genuinely mixed, and this is the real tension in the name: - Cyclical/sentiment case: consumer alcohol spending is famously tied to discretionary sentiment; a "fewer shopping trips, lower spend per visit" description sounds like a macro squeeze on the marginal buyer, not a brand losing relevance — Modelo Especial is still the #1 or #2 beer in the US by volume. Tariffs are a policy cost that could ease or be partially passed through over time. The earnings base (EPS) has not actually cracked. - Structural risk case: the alcohol-moderation/GLP-1 narrative is a real, first-innings secular headwind hitting the entire category, not just STZ, and it has no clean historical precedent to size against. If depletions keep sliding rather than stabilizing, the "premiumize + buy back shares" algorithm that has carried EPS for a decade runs out of an organic base to sit on top of.
5. Synthesis — weighted verdict
Per the framework, for a mature dividend-paying staple, Fundamentals + Valuation carry the most weight, with Moat providing the durability check and Sentiment identifying whether the market's pessimism is proportionate.
- Fundamentals: Constructive. FCF resilient and well-covering the dividend (≈39% payout), share count shrinking on genuine (not debt-funded) buybacks, EPS held up through a deliberate portfolio quality upgrade rather than eroding. The debt load (51% D/A, 3.3x net debt/EBITDA) is elevated but comfortably serviced. Weakness: FCF CAGR is flat over the only history available (3yr, a real data-depth gap against the framework's 5-8yr standard).
- Valuation: Favorable, with the DYT caveat named. Trading below its own most recent annual P/E band and far below its 2022-24 multiple, on an earnings base that has not reset — the clean pass the "down a lot is not cheap" companion clause asks for. Graham/DDM floor ($105-115) and the multiple-normalization case ($135-165) bracket a fair-value range that sits at-to-above today's price.
- Moat: Real but not expanding. Corona/Modelo's brand and distribution moat is durable against direct rivalry; the risk is category-level (moderation, tariffs), which no brand moat neutralizes.
- Sentiment: Proportionate to a real but not (yet) thesis-breaking risk. The drop tracks quantified tariff costs and a genuine, admitted volume slowdown — not a scandal, an accounting problem, or a single bad quarter. Sell-side remains split Buy/Hold rather than capitulating, and the earnings base is intact. The absence of insider buying at the lows is the one soft spot in an otherwise "priced-for-pessimism, not priced-for-disaster" picture.
Named tension: this is a name where a real, durable moat is priced cheaply because the market is genuinely uncertain whether beer-category volume decline is cyclical (buy the dip) or the start of a secular moderation trend (a slow-motion value trap in disguise) — and that uncertainty will not resolve on this analysis, only on a few more quarters of depletion data.
Verdict: ACCUMULATE, conviction 6.5/10. The valuation and balance-sheet-quality case is strong enough to build a position into the current 52-week-low band; conviction is capped below 7 because the core demand question is genuinely open and the next print (~Oct 5, 2026) is the first real read on whether depletions are stabilizing or still sliding.
Key risks that would break this thesis: 1. Beer depletions turn materially negative (beyond the current -0.3%) for two consecutive quarters — would signal the moderation/GLP-1 story is structural, not cyclical, and would force a re-think of the whole "premiumize + buy back" algorithm this stock has run on for a decade. 2. Further US tariff escalation specifically targeting Mexican beer (not just aluminum cans) — STZ's ~80% Mexican-beer revenue concentration is a real, geography-specific exposure a domestic competitor doesn't carry.