Setup

Albertsons (ACI, $16.57) just filed its FY2025 10-K. The headline picture is consensus: leverage rising to 2.24x EBITDA, $1.49B FY2025 buybacks plus a 13% dividend hike, three-year organic Adjusted EBITDA decline ($4,318M → $4,005M → $3,833M ex-53rd week). What's not consensus is the asymmetry against Kroger (KR, $65.90) on Inflation Reduction Act pharmacy exposure — a 110bps quantification gap with no rebate-offset disclosure.

What the filing says

ACI guides FY2026 to a 150bps headwind to identical sales from IRA Medicare Drug Price Negotiation. Pharmacy is now 13.7% of revenue ($11.4B, +19% YoY) — the marginal driver of comp growth and the marginal source of compression. The 10-K does not quantify EBITDA dollar impact. It does not disclose any rebate-offset mechanism. It adds new risk-factor language acknowledging that union contracts may limit AI/automation deployment.

Capital return is structurally debt-financed. FY2025 OpCF $2,367M minus capex $1,834M = $533M maintenance FCF. After $323M dividends, true discretionary FCF is ≈$210M. The $1,492M FY2025 buyback required ABL drawdowns plus $2.1B in February 2026 note issuance (5.625% / 5.750% to 2032/2034). FY2026 capex steps to $2.0-2.2B. Opioid escrow begins April 30, 2026 ($136.6M of $774M settlement). Washington State excluded from the settlement; DOJ subpoenas active on Controlled Substances Act dispensing.

What Kroger says

KR Q3 FY2025 call (Dec 4, 2025): IRA was "approximately 30 to 40 basis points" headwind to Q4 identical sales fuel, with explicit "no impact on earnings." Mechanism: KR purchases drugs at current acquisition cost; manufacturers fully reimburse the difference via rebates that offset COGS. Q4 FY2025 actual confirmed ≈40 bps drag and no earnings hit. KR's FY2025 10-K body has zero substantive IRA risk-factor disclosure — silence consistent with earnings-neutral framing.

ACI's 150 bps full-year guide is roughly 3.75x KR's quantified Q4 number. Even after annualizing KR's Q4 (negotiation effective Jan 2026, half the quarter affected) to a ≈60-80 bps run rate, ACI's drag is still ≈2x KR's at similar regulatory exposure.

What the market thinks

ACI: $16.57, ≈7.3x forward P/E ($2.27 mid-point EPS guide), 4.1% dividend yield, 12.2% short interest, RSI 30.9. KR: $65.90, ≈12x forward P/E, 2.1% dividend, RSI 25.6. Multiple discount KR-vs-ACI ≈64% — already prices ACI as inferior. Pharmacy IRA risk is priced as a sector-wide headwind, not as a 110bps differential between two operators with similar pharmacy footprints (KR ≈2,231 pharmacies, ACI 1,713).

The second derivative — that ACI's earnings will diverge from KR's specifically on IRA mechanics over the next 2-4 quarters — is not in either name's consensus.

Why the gap exists

Three reasons, in order of importance:

  1. KR's earnings-neutralization disclosure was buried in a Q3 transcript exchange, not in the 10-K. Sell-side and retail rarely synthesize transcript color into structural margin views.
  2. ACI's silence on rebate plumbing is conspicuous only by comparison. Reading the two filings in isolation, 150bps and 40bps look like different companies, not the same factor at different operational sophistication.
  3. ACI standalone idio variance is ≈28% — well below Paleologo's 75% target. Outright short ACI bleeds factor noise. The trade only survives in pair structure (long KR / short ACI), and few desks build pair theses on grocery.

Risks (ranked)

  1. ACI Q1 FY2026 print absorbs IRA cleanly (May/June 2026). Thesis dies on the first data point if pharmacy GM holds.
  2. Defensive bid lifts both names. ACI's lower starting valuation makes it the larger beneficiary of staples rotation, hurting any pair structure.
  3. ACI wins Kroger trial October 19, 2026. $1.08/share contingent realizes; pair takes ≈7% hit on the binary (estimated 30% probability ACI prevails).
  4. KR posts a worse IRA print than ACI. Flips the differential. Unexpected given disclosed mechanics, but not impossible.
  5. ACI throttles buyback. Leverage trajectory leg dies; multiple compression thesis weakens.

Catalysts

  • April 30, 2026: Opioid first escrow $136.6M (priced)
  • May/June 2026: ACI Q1 FY2026 — first IRA dollar test
  • August/September 2026: ACI Q2 FY2026 — quantification update
  • October 19, 2026: Kroger v. Albertsons trial begins (Delaware Chancery)
  • April 2027: ACI FY2026 full-year results
  • June 2027: leverage and Kroger settlement prediction deadlines

What would change our mind

  • ACI Q1 call discloses a manufacturer rebate pass-through mechanism — disclosure asymmetry collapses
  • KR Q1 FY2026 quantifies IRA at 100bps+ with material earnings impact — flips the relative position
  • ACI announces buyback pause or material throttle — leverage trajectory leg dies
  • ACI receives strategic or PE bid — multiple compression thesis terminates
  • Activist publicly pushes for asset sales (Media Collective spinout, banner divestiture) — multiple expansion offsets EBITDA decline

Evidence

EvidenceSourceCredibilityLR
ACI guides 150bps FY2026 IS headwind from IRA; pharmacy 13.7% of revenue (+19% YoY); no EBITDA dollar impact disclosedACI 10-K FY2025, MD&A and Risk Factors0.950.8
KR Q4 FY2025: ≈40bps IRA drag, "no impact on earnings" via manufacturer rebate offset to COGSKroger Q3 + Q4 FY2025 earnings calls0.951.15
ACI maintenance FCF ≈$533M FY2025; $1,492.5M buybacks entirely debt-fundedACI 10-K FY2025, Cash Flow + Capital Structure0.950.8
ACI new risk factor: union contracts may limit AI/automation. KR with comparable union exposure has no equivalent languageACI 10-K FY2025 vs KR FY2025 10-K Risk Factors0.950.85
Three-year organic Adj EBITDA decline: $4,318M → $4,005M → $3,833M (ex-53rd week)ACI 10-K FY2025, MD&A reconciliations0.950.9
Net leverage 2.24x EBITDA (up from 1.88x); $2B remaining buyback authorizationACI 10-K FY2025, Capital Structure0.950.7
Opioid $774M settlement; WA State excluded; DOJ Controlled Substances subpoenas activeACI 10-K FY2025, Note 120.951.1
Kroger $600M trial scheduled October 19, 2026 (Delaware Chancery); ACI plaintiffACI 10-K FY2025, Note 120.951.1
ACI standalone idio variance ≈28% — well below Paleologo 75% target; pair structure requiredyfinance realized factor decomposition0.850.9