HScreen

rendered 2026-09-01 14:15 UTC STALE inputs (317h)

PM · Philip Morris International Inc. 15d old

conviction3
coverage tierT3-PRIMARY-COMPLETE
horizon12 months

Return distribution probability-weighted scenarios, never collapsed to a point

-15%22%
+12%53%
+25%25%
EV +9.31%
why these probabilities: All three legs are TOTAL return (price + ~3.1% yield on the 5.88 annualised dividend) over 12 months from 187.58 (2026-08-17), and all three are defined against what the market expects. BASE = consensus delivered: FMP consensus 2026 revenue +6.8% / EPS +11.2% and 2027 revenue +5.9% / EPS +9.3% (fmp:/stable/analyst-estimates?symbol=PM, as_of 2026-08-17T16:13:24Z, grade EST), consistent with the company's own guide of organic revenue +5-7%, organic OI +7-9%, adjusted EPS $8.36-8.51 (PM/PM_2026Q2.json:30, PM/PM_2026Q1.json:16). Deliver that, hold ~22x on rolling-forward EPS of ~$9.2 → ~202 price, +8-9% price plus 3.1% yield ≈ +12%. This is descriptive, deliberately not an independent forecast. BULL = growth materially above the market: organic OI at or above the +9% ceiling with smoke-free scaling faster than modelled — ZYN US and IQOS ILUMA if FDA authorises (PM/PM_2026Q1.json:14), off-take share gains already visible in key cities (PM/PM_2025Q2.json:20) — implying ~13-14% EPS growth against consensus 11.2%, EPS ~$9.5-9.6 in 2027, and the multiple earning 24x on accelerating high-margin mix → ~228, +21% price plus yield ≈ +25%. BEAR = a slower PACE, not a collapse, and it respects what bounds the downside: pricing plus low-single-digit volume decline makes a revenue decline implausible on this evidence, and the ~$13.5bn operating cash flow guide (PM/PM_2026Q2.json:32) plus the ~2
not built — options-implied distribution on the same axis, with the KL number and where the disagreement sits
needs: desk-options implied bands (v1.4 Part 4a) — activates at 60% coverage; currently 33%

Visibility how far ahead this name can be seen

not built — visibility score
needs: visibility_score.py

Trust how much weight this name's own record can carry (credibility Z)

not built — credibility
needs: credibility.py

Sizing survival-first size (half-Kelly) vs the enforced ladder

kelly not computable — no data

Payout durability does the distribution survive a downturn?

PASS
DPS trajectoryprogressive
interest cover now9.41x
interest cover at FORWARD rates6.55x

Falsifiers what would prove this wrong

Organic EPS growth stays at or above the guided +7.5-9.5% currency-neutral range, keeping dividend growth fundable from earnings rather than payout expansion
Gross margin expansion continues as smoke-free mix scales (H1 2026 +150bp to 67.7%)
Category share is held (~68 vs competitor 69) rather than eroding to competitors two and three
Deleveraging to ~2x net debt/EBITDA by end-2026 is achieved without cutting investment or slowing the dividend
The US flavour-ban and Japan excise drag is transitory — growth trajectory re-establishes within roughly two quarters, as it has in prior flavour-ban markets
The FY26 dollar EPS guide of $8.36-8.51 is not carried by FX — a currency reversal does not push reported EPS below the currency-neutral trajectory

Conflicts surfaced, never netted out (T4)


Street consensus target 210.5 (median 215, range 182-225) implies +12.2% price / ~15.4% total return, with 17 of 25 at buy and only one sell. My probability-weighted expectation is +9.3% total. I do not disagree with the
high — this is the entire 6.1pp gap in our_vs_street_gap_pct and the reason the name may not clear an ALPHA bar despite scoring well on quality and delivery

p_bear 22% sits 15.9pp above the node-pooled base rate of 6.1% for a terminal -15% close. Defended in deviation_defence on the starting multiple, the FX component of guided EPS, the named excise/flavour-ban mechanism, an
high — moving p_bear to the anchor would lift the expectation from 9.3% to roughly 14% and flip the sign of the street gap

estimates_query drift returns n_up 0 / n_down 0 / breadth 0.0 / median_move null over 365 days and no outliers, while FMP simultaneously reports 25 covering analysts with dated 2023-2030 estimates. The tape is missing, n
high — eps_revision_breadth is a scored ALPHA input and it reads 0.0 for infrastructure reasons, not analytical ones; any score derived from it is measuring a data gap

2026 revenue dispersion is tight at 1.3% (n=10) but 2027 widens to 9.8% (n=11) and 2028 to 5.8% (n=11). The covering universe agrees about the next four quarters and disagrees materially about the year after — precisely
medium — supports a wider scenario spread than the tight 2026 dispersion alone would suggest, and is part of why I do not narrow around the consensus mean

No options chain entry for PM, so the strongest anchor — a dated daily market price for exactly the drawdown question — is unavailable on both legs. Both bear and bull probabilities rest on pooled staples history with ef
medium — the anchoring control is running at reduced strength; a real chain would likely move p_bear materially in one direction and it cannot be known which

Reported ROE of -1.135 is negative because book equity is negative, while ROIC is 25.6%. Any screen or factor build reading ROE will classify a high-return franchise as impaired.
medium — no impact on this dossier's scores, but it will corrupt downstream quality/quantitative factor loadings unless PM is flagged

The case against independent adversary, different model family (P6) weakened

Version history from the research record

archived versions2
archivedhashbytes
2026-08-17T10:34:1788661a4acf9939,619
2026-08-19T05:54:53e560c1246fbd30,512

What changed in the latest version

-coverage_tier: T3-DATA-ONLY
-analyst: analyst-income
+coverage_tier: T3-PRIMARY-COMPLETE
+analyst: analyst-aisemi
-provenance:
- producer: analyst-income
- generated: 2026-08-17
- inputs:
- - evidence_pack PM 2026-08-17T04:21:09Z (build_evidence_pack.py)
- - state-files/PM.json (state_file_build.py 2026-08-16T07:00:23Z)
- - payout_durability.json PM as_of 2026-08-16
- - dividend_metrics PM as_of 2026-08-16T15:20:44Z
- - estimates_query.py drift/outliers as_of 2026-08-17T04:21:33Z
- scenario_returns_frozen: true
diff truncated
Rendered from canonical artifacts only (T1) — no terminal database exists. Every figure is read from the system's own files at render time; where an artifact is missing the panel names the dependency rather than showing an empty box. No execution path exists anywhere in this interface (T5), not even a link.