Bunge Global SA

source: coverage-next-full ticker: BG step: 01 title: Business Model (Overview) date: 2026-05-28

Step 01 — Business Model

Key Findings

• What BG does (one line): Bunge Global is the world's largest oilseed processor by capacity and, post-Viterra, one of the four dominant global grain merchants, monetizing the spread between farmer-origination prices and downstream food/feed/fuel demand [S1][S2].
• Operating model is dual-layer: (1) an asset-heavy industrial business — crush plants, port elevators, refineries — earning margin on physical throughput; (2) a commercial trading book — origination basis, freight, futures, and biofuel arbitrage — that can swing quarterly EBIT by hundreds of millions [S3].
• Four reporting segments post-Viterra (effective Q3 2025): Soybean Processing & Refining, Softseed Processing & Refining, Tropical Oils & Specialty Ingredients, Grain Merchandising & Milling [S4]. Q1 2026 segment Adj EBIT contribution was 57%/30%/7%/7% respectively — heavily soybean-led.
• Value-chain position: BG sits in layers 1-4 (origination, logistics, processing, refining); it does not operate downstream branded consumer foods (unlike ADM Nutrition or Cargill's specialty ingredient lines), nor upstream farm ownership [S5].
• Net read: MIXED-POSITIVE for thesis. Scale is now unmatched globally in oilseed processing; the strategic gap vs ADM is the lighter downstream specialty footprint. Bull case: synergies + biofuel pull-through. Bear case: commodity-cycle exposure with diluted equity base.

Implications for Thesis and Valuation

• BG is fundamentally a spread business, not a fee/recurring-revenue model. Valuation framework must accept cyclical earnings.
• The post-Viterra entity is more grain-merchandising-tilted than the standalone Bunge of 2019-2024 — closer in shape to ADM than to a pure crushing comp.
• The biggest valuation lever is the next-3-years average mid-cycle crush margin assumption (highly sensitive — modeled in Step 14 of /complete-coverage).
• BG's earnings are NOT linear / extrapolable from any single year; CY2023 was a high-margin year ($14.87 GAAP EPS), CY2024 was trough ($7.99), CY2025 reported ($4.91 GAAP / $7.57 adj) reflects integration drag.

Objective

Describe what Bunge does in plain language, map its value-chain position, identify revenue streams and economics, document the segment structure post-Viterra, and frame the operating-vs-trading split that drives quarterly variance.

Narrative Analysis

What BG actually does

In any given month, Bunge:

1. Buys raw oilseeds (mostly soybeans, canola, sunflower seed) and grains (wheat, corn) from farmers in Brazil, Argentina, US Midwest, Canada, EU, Black Sea, and Australia — increasingly so via Viterra's elevator network.
2. Moves them via owned-and-leased port terminals, river barges, rail, and trucks — the logistical optimization across origin/destination pairs being one of the trading book's main P&L levers.
3. Crushes oilseeds into vegetable oil + protein meal at ~75-80 plants worldwide.
4. Refines crude vegetable oil into branded/private-label cooking oils, bottled oils, margarine bases, biofuel feedstock.
5. Mills wheat and corn into flour and dry-mill products.
6. Sells to: food manufacturers (Nestle, Unilever, P&G's food brands, Procter & Gamble, Cargill itself for further processing), animal feed mills, foodservice distributors, biofuel producers (Chevron under the JV, plus arm's-length sales to Phillips 66, Marathon, Valero), and end-customers via private-label oils [S1].

Layered value-chain map

Layer 1: Origination — farmer-facing, basis-pricing
   ↓ (Bunge legacy: light; Viterra: heavy)
Layer 2: Logistics — port elevators, river/rail/truck, storage
   ↓ (Bunge + Viterra combined)
Layer 3: Processing — crush plants (oilseeds), mills (wheat/corn)
   ↓ (Bunge core competency, largest globally post-Viterra)
Layer 4: Refining & Distribution — refined oils, lecithin, specialty ingredients
   ↓ (Bunge: present; ADM: deeper)
Layer 5: Branded Consumer / Specialty — NOT operated by BG

[S1][S5]

Revenue streams

BG's segment structure post-Viterra (effective Q3 2025) [S4]:

[S4]

Operating-vs-trading split

A single Q1 reading suggests that BG's processing segments earn ~3-5% segment Adj EBIT margins from physical throughput, while the Grain Merchandising & Milling segment runs at ~0-1% margin most of the time because it's effectively a flow business: high revenue (port + elevator + trading throughput), low fixed margin per ton, with EBIT swinging when basis spreads or freight arbs open up. This is a fundamental difference from pure-play food companies — BG's trading book is a real risk and earnings line, not back-office hedging [S6].

The Chevron JV (Bunge Chevron Ag Renewables)

50/50 joint venture with Chevron. Bunge contributed soybean processing facilities in Destrehan, LA and Cairo, IL; Chevron contributed ~$600M cash. Capacity goal: expand from 7,000 tons/day (end 2024) to ~14,000 tons/day by end 2026. Sells low-CI soybean oil to Chevron's renewable diesel refineries. This is equity-method consolidation (not part of BG's reported revenue), but it is a strategic optionality on US biofuel mandates [S7]. Secondary track applies here: this asset is best valued via standard DCF / capacity multiples, not commodity-cycle logic.

What BG is NOT

• Not a branded consumer foods company (no Mazola/Wesson equivalent at scale within BG; some specialty oils are branded but small)
• Not an agronomy / seed company (no equivalent to Corteva or Bayer Crop Science)
• Not a fertilizer or crop protection player
• Not vertically integrated upstream into farm operations

This shapes the moat assessment: BG's moat — if any — is in scale, geographic optionality, and trading book skill, not in IP or brand. (Tested in Step 10.)

Evidence and Sources

• Segment structure and Q1 2026 segment Adj EBIT: company Q1 2026 press release [S4]
• Value-chain position vs ADM: peer comparison and industry analysis [S5]
• Chevron JV mechanics and capacity: 2023 biodieselmagazine + biofuels-news coverage [S7]
• General BG description and post-Viterra commentary [S1][S2][S3]

Assumption Register Updates

Tables and Calculations

Segment Mix — Q1 2026 (one-quarter sample)

Historic Segment Structure (pre-Q3 2025)

Pre-realignment, Bunge reported as: Agribusiness, Refined & Specialty Oils, Milling, Sugar & Bioenergy. The realignment in Q3 2025 was driven by the Viterra integration and management's combined-company operating view. Direct comparability for full-year segment series will require restating prior periods — pending in /complete-coverage Step 13 work.

Open Questions and Data Gaps

• Sugar & Bioenergy segment treatment under new 4-segment structure (likely embedded into Grain Merchandising or treated as held-for-sale separately) — to confirm from 10-K segment note
• Customer concentration explicit disclosure
• Geographic mix split by segment (especially South America vs North America)
• Full-year FY2025 segment data not transcribed (only summary press release fragments)

Next-Step Dependencies

• Step 02 will use segment structure here for industry/market positioning
• Step 03 will need segment-level margin data for margin tree
• Step 10 will test moat hypothesis using value-chain map

Source Index

source: coverage-next-full ticker: BG step: 04 title: Financial Quality (Snapshot + Adversarial Sweep) date: 2026-05-28

Step 04 — Financial Quality

Key Findings

• Earnings quality is structurally low — net income margins of 0.85-7.85% over CY2021-CY2025 with high cycle volatility; FY2025 reported $816M net income vs $2,243M in CY2023 [S1]. The headline EPS is heavily distorted by mark-to-market gains/losses on the trading book, FX, and Viterra purchase accounting.
• Adjusted EPS is the primary external metric — Adj EPS strips out non-cash adjustments (commodity hedge mark-to-market, integration charges, certain divestiture gains/losses). CY2025 adj EPS $7.57 vs GAAP $4.91 = 35% difference [S2]. This wedge is structurally large because of the trading book.
• Cash conversion is poor in the short run but normalizes over the cycle — CY2025 OCF $844M vs net income $816M (104% conversion looks OK in isolation, but heavy working-capital build offset trading gains); CY2024 OCF $1,900M vs NI $1,137M (168% — normal); CY2022 OCF NEGATIVE $5.5B vs NI $1.6B (negative working-capital absorption during commodity-price spike) [S1].
• Adversarial sweep finds NO material short reports, NO active SEC enforcement actions, NO material legal/regulatory overhangs. Bunge has historically been free from accounting-fraud allegations, unlike Glencore (its largest shareholder, with FCPA history). Bunge had a $112M settlement with the DOJ in 2013 for accounting issues at Argentine subsidiary but no recurrence since.
• Net read: NEUTRAL. Quality is cycle-typical for the agribusiness oligopoly — neither best-in-class (like the franchise-business comps) nor problematic. The Adj EPS wedge is large but understood by analysts and consistent across peers (ADM has a similar adjustment pattern).

Implications for Thesis and Valuation

• Use Adj EPS, not GAAP EPS, for forward valuation (consistent with how the Street prices the stock).
• Discount the trading-book P&L line in forward modeling — treat as zero-NPV mean-reverting, with quarterly variance built into the equity-cost-of-capital.
• The $14.3B net debt post-Viterra is the single largest balance-sheet risk; rating-agency response to be monitored.
• Working-capital cycle (especially South American harvest timing) creates seasonal OCF swings that are normal but real.

Objective

Assess earnings quality, accounting conservatism, cash conversion, balance-sheet integrity, and run an adversarial research sweep for short reports, lawsuits, SEC investigations, and other reputational/governance overhangs.

Narrative Analysis

Earnings quality — GAAP vs Adjusted

Bunge's GAAP earnings include several material items the Street strips out:

• Mark-to-market on hedge book — commodity derivatives held for risk-management purposes generate non-cash gains/losses that don't reflect underlying margin
• FX translation — large Brazilian, Argentine, EU footprint generates non-cash FX volatility
• Integration / restructuring charges — Viterra-related in 2025-2026
• Bargain purchase / goodwill items

CY2025 reconciliation:

• GAAP net income: $816M / GAAP diluted EPS: $4.91 [S1]
• Adj net income: ~$1.46B / Adj EPS: $7.57 [S2]
• Wedge: ~$643M ($2.66 per share) of non-cash / non-operating items added back

Historical adj-vs-GAAP wedges (per StockAnalysis-derived back-calcs):

• CY2024: GAAP $7.99 vs adj $9.19 (15% wedge)
• CY2023: GAAP $14.87 vs adj n/d
• CY2025: GAAP $4.91 vs adj $7.57 (54% wedge) — outsized because of Viterra purchase accounting

This pattern is typical of the agribusiness oligopoly. ADM, Wilmar, and LDC all show similar reconciliations. For valuation, use Adj EPS consistently.

Cash conversion analysis

OCF vs Net Income over the cycle:

[S1][S2]

The OCF series is dominated by working-capital swings tied to commodity prices, not earnings quality. When commodity prices spike (2021-2022), inventory and AR balloon → OCF turns deeply negative even as NI is positive. When prices fall, the unwind shows up as OCF release. This is structurally correct given BG's flow-business nature, but it makes any single-year OCF measurement misleading. Use multi-year OCF averages, not annual snapshots, for cash quality assessment.

Balance sheet integrity

Post-Viterra (Dec 31, 2025):

• Total assets: $44.5B (vs $24.9B pre-merger)
• Net debt: $14.3B (vs $3.8B pre-merger)
• Equity: $17.4B (vs $10.9B pre-merger)
• Net debt / equity: 0.82 (manageable, but materially elevated)
• Goodwill + intangibles: $3.4B (modest; not a write-down risk) [S2]

The doubling of assets reflects Viterra consolidation; the elevated leverage is the integration period's main rating-agency variable. S&P / Moody's actions are a monitor (likely BBB / Baa2 range; downgrade risk if deleveraging slips).

Adversarial Research Sweep

Methodology: searched for short-report distribution (Muddy Waters, Hindenburg, Citron, Quintessential, Wolfpack, etc.), SEC enforcement actions, class-action litigation, and FCPA/anti-corruption matters involving Bunge.

Glencore overhang note: Bunge's largest shareholder (~15% post-Viterra) is Glencore plc, which has a documented history of FCPA / bribery / regulatory issues across its mining + trading businesses. As far as I can determine, none of this exposure transfers to Bunge directly (the FCPA matters were settled by Glencore plc; no co-defendant designation for Viterra or Bunge in those actions). But Glencore's reputational/governance profile is something a long-only investor should be aware of, even if it has no direct contagion to BG.

Key financial-quality flags

1. Goodwill / intangibles low — only $3.4B combined vs $44.5B assets — minimal impairment risk
2. **No debt covenant disclosure suggests material headroom; new $XXBn Viterra-funding facility was syndicated 2025
3. No going-concern language ever issued
4. Audit firm — to confirm from latest 10-K (likely Deloitte or EY, no recent auditor change controversy)
5. Restatements — none material in past 5 years

Evidence and Sources

• Annual GAAP series [S1]; Adjusted reconciliations from press releases [S2]
• Adversarial sweep results from web search across short-report and litigation sources [S3]
• Glencore reference / non-contagion analysis [S4]

Assumption Register Updates

Tables and Calculations

Earnings Quality Summary

Adversarial Sweep — Detailed Searches Run

Open Questions and Data Gaps

• Auditor identity confirmation from latest 10-K
• 10-K's risk-factors detail (couldn't retrieve directly; relied on press summaries)
• Specific covenant details on Viterra-financing facility
• EUDR compliance certification confirmation

Next-Step Dependencies

• Step 05 will use the adjusted-vs-GAAP framework for quarterly KPI selection
• Step 06 will deepen the balance-sheet / leverage analysis
• Step 14 (in /complete-coverage) will use Adj EPS as the primary multiple input

Source Index