Archer Daniels Midland Through a Long-Term Owner's Lens

Understanding the Business and Industry Structure

How the business actually makes money. [Fact] ADM organizes itself into three reportable segments: Ag Services and Oilseeds, Carbohydrate Solutions and Nutrition. Ag Services and Oilseeds covers agricultural procurement, trading, transportation, storage, oilseed crushing and downstream refining; Carbohydrate Solutions is mainly corn and wheat wet and dry milling; Nutrition produces higher-value-added human and animal nutrition ingredients. The company defines itself as a "global agricultural supply-chain manager and processor," connecting origin to end demand.

Who the customers are and how it charges them. ADM does not earn money by selling a single product at a brand premium; it earns money through a full suite of supply-chain services and processing capabilities: selling grains and oils, protein meal, sweeteners, starches, ingredients and nutrition solutions to food, feed, fuel and industrial customers. Its revenue includes both physical processing sales and the spreads and service income generated by trading and distribution activity. The company states clearly in its annual report that revenue and product costs in its agricultural processing and trading businesses are highly correlated with underlying commodity prices and volumes, so one should look at gross profit / spreads rather than revenue alone.

Whether revenue is recurring, stable and predictable. [Fact] On the demand side, global consumption of grains and oils shows no sign of decline. The OECD-FAO projects that, through 2034, global consumption of grains for food use will still grow about 1.1% a year, continuing to be supported by population growth, feed demand and some industrial and biofuel uses; the FAO also notes that, by 2034, roughly 40% of cereal use will still be direct food consumption, one-third for animal feed, and the rest for biofuels and industrial uses. [Inference] ADM's demand floor is stable, but ADM's profit is not, because profit depends on crush and processing spreads, trade flows, inventories, freight, policy and weather. [View] So this is a business with long-term demand but meaningful short-to-medium-term earnings volatility.

Cost structure and dependencies. ADM's costs center on raw agricultural inputs, energy, logistics, storage, labor and capital tied up in the business. For ADM, raw-material procurement is not a "stably lockable cost" in the ordinary manufacturing sense; it is part of the business model itself. The company also needs enormous working capital to hold inventory, receivables and customer margin, and the valuation of inventory and derivatives is itself complex—in 2025 the auditors identified the fair-value estimation of certain agricultural inventories and related contracts as a critical audit matter. [Inference] ADM is not a company that "runs cash off a single production line"; it makes money through a global network plus risk management plus capital deployment plus execution.

Whether it depends on a few customers, suppliers, channels, policies or key people. This business does not depend on individual core technical staff the way a software company does, nor on a few large customers the way a luxury brand does; but it clearly depends on the global trading order, agricultural policy, biofuel policy, freight, weather, geopolitics and credit markets. In 2026 the EPA finalized the RFS rules for 2026 and 2027, with the conventional renewable-fuel implied requirement held at 15 billion gallons, showing that U.S. biofuel policy remains an important variable for crush demand. [View] ADM's biggest external dependency is not any single customer, but policy and the commodity chain itself.

What stage the industry is in. Looked at purely as "food and feed" demand, the industry is mature and stable; looked at through the profitability of the crush, trading and biofuel chains, the industry is clearly cyclical. Among strong public comparables, Bunge significantly strengthened its scale and global network after completing the Viterra deal; in the high-value-added ingredients line, Ingredion represents a more "ingredient-focused, higher-margin" model. [View] ADM is not a "bad industry," but it is not an inherently high-return industry either; more precisely, it is a key infrastructure-type company within a middling industry.

Whether this is a business I would hold even if the stock market closed for five years. [View] If the entry price is low enough, I would hold it; if the entry price is merely "roughly fair," I would not be especially excited. The reason is simple: stable demand does not equal stable shareholder returns, and ADM still requires investors able to ride through crush-spread and working-capital swings.

Business understandability score: 4/5. The business itself is not mysterious, but its earnings drivers are more complex than they appear, especially in commodity inventory, fair value, working capital and cross-regional trade flows.

Industry attractiveness score: 3/5. Demand is stable over the long run, but margins and returns on capital are heavily affected by the cycle, policy and competition.

Moat and Management

Assessing the moat item by item. ADM's most obvious moat is not its brand or its patents, but its global scale, asset footprint, logistics network, origination capability, risk management and depth of customer coverage. In Ag Services and Oilseeds the company covers procurement, transportation, storage, crushing, refining and an export network; this kind of infrastructure cannot be replicated in a year or two. Bunge completing Viterra in 2025 and quickly realizing initial synergies, in reverse, also proves that the truly important barriers in this industry are network density and a global supply-chain position.

If I break the moat apart, my assessment is as follows: brand advantage: weak to moderate; cost advantage: moderate; scale advantage: moderate to fairly strong; network effects: weak; switching costs: weak; channel and logistics advantage: fairly strong; patent / license barrier: weak; data advantage: weak to moderate; corporate culture / operating capability: moderate; capital-allocation capability: moderate to weak. The key point here is that ADM's advantage shows up more in "it's hard for anyone else to roll out the same amount of storage, logistics, crushing and customer relationships across the globe at the same time" than in "customers can't do without it" or "it can easily raise prices."

Whether the moat is widening, stable or narrowing. My judgment is: stable overall, but with no clear widening at the margin. On one hand, scale and the global network still have value; on the other, the integration of Bunge and Viterra makes another heavyweight in this industry stronger, and in its Nutrition business ADM has not proven it possesses the higher-quality moat of a consumer brand or specialty-ingredient company. Nutrition took an animal-nutrition goodwill impairment in 2023, and its goodwill still stood at roughly $1 billion in 2025, with the auditors continuing to list the animal-nutrition goodwill impairment judgment as a critical audit matter.

Can it raise prices in an inflationary environment. It can pass through prices, but not necessarily expand margins. This is the biggest difference between ADM and a genuinely strong branded consumer company. The company itself stresses that, in its agricultural processing and trading businesses, revenue and costs tend to move together with commodity prices, so the focus should be on gross profit and spreads rather than nominal revenue. [View] ADM has inflation pass-through ability, but not durable inflation pricing power.

Whether it can stay profitable in a downturn. Historically, generally yes, but earnings elasticity is large. 2021 and 2022 were very strong years, 2023 was still solid, and 2024 and 2025 fell back clearly; nonetheless, the company kept an investment-grade rating into the end of 2025, with about $12.3 billion in credit facilities, of which roughly $9.4 billion was unused. [View] This is a company that "can survive," not one that "stays highly profitable even in a downturn."

Whether management is trustworthy. This has to be looked at separately. [Fact] CEO Juan Luciano has served as a director since 2014 and is now Chairman and CEO; as of March 13, 2026, he beneficially owned roughly 1.921 million ADM shares; executive long-term incentives are tied to average adjusted ROIC and cumulative adjusted EPS, with a small ESG modifier on top. [Positive] This at least shows the company is not entirely indifferent to capital efficiency. [Negative] The 2024 segment-disclosure issue, the material internal-control weakness, the subsequent SEC settlement, and the gap between Nutrition's long-term promises and its returns have genuinely weakened management's credibility score. [View] I would not place ADM's management in the "clearly untrustworthy" bucket, but I would not give a high score either.

Whether capital allocation is rational. On the good side, ADM has paid dividends consistently over the past few years, made sizable repurchases at undervalued levels, and kept an investment-grade balance sheet. On the bad side, the Nutrition strategic line has not delivered a convincing record on returns on capital, taking an animal-nutrition goodwill impairment in 2023 and still carrying a large Nutrition-related goodwill balance in 2025; in 2025 the company did not repurchase stock, indicating that management chose caution in a low-profit period and prioritized stabilizing liquidity and credit. [View] Capital allocation cannot be called bad, but it is hard to call it excellent either, especially given that the M&A and integration results in Nutrition have not yet been fully proven.

Management and capital-allocation score: 2/5. Without the 2024–2026 accounting and internal-control episode I would give a 3; given the trust discount and the uncertainty over Nutrition's returns, I can only give a 2.

Moat strength score: 3/5. There are global-network and scale barriers, but no strong customer lock-in or pricing power, so the moat is real but of average quality.

Financial Quality and Owner Earnings

Look first at the long-term financial profile, not a single year's P/E. [Fact] ADM's verified data shows: 2021 revenue of roughly $85.2 billion, a cyclical peak in 2022 of roughly $101.6 billion, roughly $85.5 billion in 2024, and roughly $80.3 billion in 2025; 2025 net income attributable to shareholders was only $1.078 billion, a clear decline from $3.483 billion in 2023 and $4.340 billion in 2022. [Inference] This shows ADM is not a "steady compounder," but a company with large revenue, fast turnover, thin margins, and a big difference between good years and bad years.

Note: figures in the table are in billions of U.S. dollars; 2021–2022 data come from the 2021/2022 annual reports, and 2023–2025 data come from the three-year statements and cash-flow statement in the 2025 10-K; diluted share counts for 2023–2025 can be cross-checked against the company's disclosed net income and EPS.

Gross margin, profit margin and cash-flow quality. 2021 gross profit was roughly $5.987 billion, a gross margin of about 7.0%; 2022 gross profit was roughly $7.570 billion, a gross margin of about 7.5%; in 2025 the company disclosed that gross profit fell roughly $745 million year over year to about $5 billion, a gross margin of about 6.2%. Net margin fell from about 4.3% in 2022 to about 1.3% in 2025. [Inference] ADM's high profits over the past two or three years look more like a cyclical dividend and a high in spreads than a durable high margin spilling out naturally from a moat.

Whether cash flow and accounting earnings match. The answer is: not unreasonable over the long run, but with large single-year swings. 2025 operating cash flow of $5.452 billion was far above the $1.078 billion in net income; but the company itself explains that the sharp year-over-year improvement in 2025 cash flow was driven mainly by changes in working capital. 2024 free cash flow was only $1.227 billion, and 2025 jumped to $4.204 billion; a swing like this is itself a warning that you cannot capitalize any single year's free cash flow directly.

Capital-expenditure intensity and the "the more it grows, the more cash it needs" problem. ADM's annual capital expenditure over the past five years ran roughly between $1.169 billion and $1.563 billion—by no means an asset-light model; operating cash also has to absorb huge swings in inventory, receivables, margin and customer-related funding. [View] ADM is not the classic high-quality model of "the more it grows, the more it earns"; rather, growth can bring scale effects but often comes with greater working-capital and supply-chain funding pressure.

Whether the balance sheet is sound. At the end of 2025, ADM held roughly $1.015 billion in cash and cash equivalents, total long-term debt including the current portion of roughly $7.865 billion, plus short-term financing including commercial paper to support operations; the company still held an investment-grade rating but with a negative outlook, and had about $12.3 billion in credit facilities, of which roughly $9.4 billion was unused. [Inference] ADM's leverage is not dangerous, but it is not "loosely indulgent" either; it is sound by cyclical-stock standards, and not light by purely defensive standards.

Changes in share count, dividends and repurchases. ADM has not been stingy with shareholder returns over the past few years. In 2023 it paid roughly $977 million in cash dividends and repurchased roughly $2.697 billion; in 2024 it paid roughly $985 million in dividends and repurchased roughly $2.347 billion; in 2025 it paid roughly $987 million in dividends but no longer repurchased. At the end of 2025 it still had 115 million shares of repurchase authorization remaining. [Inference] Management at least knows repurchases are a capital-allocation tool, but pausing repurchases in 2025 also shows it placed more weight on credit and liquidity during a low-profit, repair period.

Whether there is any sign of fraud, aggressive accounting or earnings manipulation. I have to be honest: there is the objective fact of internal-control and segment-disclosure problems, but no evidence that historical consolidated income and cash flows were restated or overturned. In its 2026 disclosures the company stated clearly that the corrections and restatement concerned historical segment reporting and did not affect the consolidated balance sheet, income statement and cash-flow statement of the periods presented; at the same time, the 2025 auditors gave an unqualified opinion on internal controls. [View] This is not a minor flaw to be waved away, but a real event sufficient to make a conservative investor raise both the discount rate and the margin of safety.

Owner Earnings analysis. If you computed "owner earnings" strictly from 2025 operating cash flow, you would reach an overly optimistic result, because 2025 cash flow was significantly lifted by working-capital release. A more reasonable approach is to look at 2023–2025 free cash flow together with the adjusted EPS in management's long-term incentives. The proxy statement shows cumulative adjusted EPS of about $15.15 over the three years 2023–2025, or roughly $5.05 per share annually; on the current share base of about 484 million shares, that corresponds to annual adjusted earnings power roughly in the $2.4 billion to $2.5 billion range. Considering further that ADM is a capital-intensive supply-chain company, that maintenance capital expenditure is not low, and that working-capital release cannot be assumed to persist over the long run, I prefer to place conservative normalized Owner Earnings at $2.1 billion to $2.3 billion, with a midpoint of about $2.2 billion.

[Assumption] My treatment of Owner Earnings is this: rather than taking the $5.452 billion of 2025 operating cash flow directly, I take a compromise value for "mid-cycle free cash flow / adjusted earnings power." [Inference] Doing so sacrifices some of the "undervalued" conclusion, but it better fits a long-term owner's lens. Estimating on $2.2 billion of normalized Owner Earnings, the current market capitalization of about $37.5 billion corresponds to roughly 17x normalized Owner Earnings; on an enterprise-value basis (adding net debt), the multiple is even higher. For a fairly cyclical agricultural-processing business with average pricing power, that cannot be called cheap.

Valuation, Margin of Safety and Opportunity Cost

The conclusion first: the current price is not cheap, but it is not absurd either; it is closer to "fair to slightly high" than to "deeply undervalued." ADM's current trailing P/E of about 34.6x looks expensive; but that is mainly because 2025 sits at an earnings trough. The problem is that, once you swap the P/E for the normalized Owner Earnings I care more about, the valuation is still only neutral to expensive, and does not automatically become a high-margin-of-safety opportunity.

Discounted owner-earnings method. I use normalized Owner Earnings as the starting point, not 2025 net income and not 2025 free cash flow. The core reason: 2025 net income is on the low side, while 2025 operating cash flow was lifted by working-capital release, and neither is suitable as a single anchor point. Below are three scenarios:

[Assumption] These growth rates are not high, but they already imply that ADM can restore its 2024–2025 trough earnings to a more normal mid-cycle level. [View] A more conservative investor should focus mainly on the conservative and neutral ranges, rather than embracing the optimistic range to make the conclusion look better.

On this basis, my range judgment is: Conservative intrinsic-value range: $45–55; Fair intrinsic-value range: $60–72; Optimistic intrinsic-value range: $88–100. At the current price of $77.52, the stock trades at a clear premium to the conservative range, roughly a slight to moderate premium to the fair range, and only at a discount to the optimistic range. [View] This is precisely what "no sufficient margin of safety" means.

Relative-valuation method. Among public comparables, ADM's current P/E is about 34.6x; Bunge's is about 31.9x; Ingredion's is about 9.85x. On a book basis, ADM's shareholders' equity at the end of 2025 was roughly $22.74 billion, a current P/B of about 1.65x; Bunge, on shareholders' equity of roughly $15.90 billion at the end of 2025, trades at a P/B of about 1.47x; Ingredion, on equity of roughly $4.366 billion at the end of 2025, trades at a P/B of about 1.50x. On operating quality, Ingredion's 2025 operating profit was roughly $1.016 billion, cash flow $944 million, and net debt / adjusted EBITDA only 0.6x, clearly closer to the picture of a "high-value-added ingredients company"; Bunge, because of the Viterra acquisition, has larger debt and integration variables in 2025. [View] ADM is not cheap enough to clearly outweigh these two public comparables.

Asset or liquidation-value method. ADM's shareholders' equity at the end of 2025 was roughly $22.74 billion, implying net assets per share of about $47 on a book basis; but this includes substantial goodwill and intangibles, and in 2025 the auditors specifically listed the animal-nutrition goodwill impairment judgment as a critical audit matter. At the same time, ADM's equity-method investment in Wilmar had a book value of roughly $4 billion at the end of 2025 against a market value of roughly $3.4 billion, so the book is not obviously conservative. [Inference] ADM's book value can serve as a "floor reference," but cannot be taken directly as strong evidence of undervaluation. My asset-method conclusion is: a more lenient asset value sits in the low-to-mid $40s, and stripping out goodwill and some intangibles more strictly provides even weaker support for the share price.

Margin of safety. I tend to require a 20%–30% discount to the neutral intrinsic value before I feel comfortable with a company like this. That implies an ideal entry point roughly in the $48–60 range; $60–78 counts as an "acceptable to hold, but not exciting enough to add" range; and if the stock reaches above $90, it enters a clearly overvalued zone in my framework. [View] At today's price, ADM looks more like "keep tracking, wait for better odds" than "deploy heavily right now."

Opportunity-cost comparison. The U.S. 10-year Treasury yield was about 4.56% on May 22, 2026. For a single stock whose profits are heavily affected by the cycle, policy and working-capital swings, if the current price corresponds to only about 4%–7% of neutral long-term return expectation, then the excess return it offers over the risk-free rate is not generous. Compared with a broad index, ADM's advantage is not certainty but the possibility of offering higher odds at a worse price; but at the current price, those odds are not outstanding.

Risks, Checklist and Final Judgment

The most important risks. First, competition and industry-structure risk: after Bunge completed the Viterra acquisition, the competitive concentration and the strength of rivals in the global grain-and-oil chain have changed, and ADM's network advantage is not unique. Second, technology and product-substitution risk is relatively limited, but structural-upgrade risk exists: truly high-quality profit lies more in ingredients, brands and specialty solutions, and if ADM remains at the "basic processing plus trading" level over the long run, its valuation center will not be high. Third, regulatory and policy risk: trade policy, tariffs, sustainability regulation and biofuel policy all directly affect the direction of demand and processing spreads. Fourth, financial and working-capital risk: profit is thin while inventory and customer funding are huge, so cash flow swings sharply with working-capital changes. Fifth, management and accounting risk: although the 2024 internal-control event has been resolved for now, trust repair is not automatic.

The strongest bear case. If I had to summarize the most powerful argument for the bears, it would run roughly like this: ADM looks cheap because many investors treat the 2025 trough profit as a temporary phenomenon; but if the high profits of 2022–2023 were themselves abnormal and 2024–2025 are simply a return to the true center, then the current share price is not cheap. Going further, if the improvement in 2025 free cash flow came mainly from the release of inventory and other working capital rather than a recovery in business quality, then the market may be paying for an unsustainable cash improvement. Finally, the Nutrition strategy has consumed substantial capital and management attention for years yet still has not proven it can deliver steady high returns, instead producing goodwill impairment and disclosure problems, which casts doubt on management's ability to turn a low-return business into a high-return one.

What facts would overturn the current judgment. If the following facts emerge over the next two to three years, I would admit my judgment was wrong, and might even shift from "Watch" to something more positive: first, ADM can stably restore normalized Owner Earnings to above $2.4 billion to $2.6 billion without relying on abnormal working-capital release; second, the Nutrition business can keep improving its margins and returns on capital without another large impairment; third, management directs more capital toward high-return projects rather than continuing to chase scale, without additional internal-control or disclosure incidents. Conversely, if there is another material internal-control problem, a further impairment, a credit-rating downgrade, or profit that stays near the 2025 level for the long run, then the "undervaluation-repair" thesis can easily fail.

Comparison with other opportunities. Compared with Bunge, ADM lacks the synergy story of a large integration deal but also lacks some of the integration risk; compared with Ingredion, ADM looks more like a bulk-agriculture network company while Ingredion looks more like a high-value-added ingredients company—the latter currently trades cheaper on public valuation and has a lighter balance sheet. Compared with SPY, ADM does not currently offer a discount large enough to compensate for single-name and cyclical risk. Compared with the 10-year Treasury, ADM's currently implied neutral-return advantage is not wide. [View] If your capital can only go into five positions, I do not think ADM yet qualifies to displace a higher-certainty asset.

Investment Checklist

These conclusions do not mean ADM is a "bad company"; they show it is more of a name that demands very high price discipline.

[Final Rating] Watch

[One-Line Investment Thesis] ADM is a real, important, globalized agricultural supply-chain and processing company, but its profit quality is heavily affected by the cycle and working capital, its moat comes more from scale and network than from pricing power, and the margin of safety is insufficient at the current price.

[Core Bull Points]

• Long-term global demand for food, feed and some biofuel still has a floor, and the chain ADM occupies will not easily disappear.

• The global origination, transportation, storage, crushing and customer network constitutes a real scale barrier.

• The balance sheet is still investment-grade, with ample liquidity and strong survivability.

• Internal-control repair was completed in 2025, and the most serious regulatory uncertainty has been resolved for now.

• If future profits recover to a more normal center, the current valuation is not entirely without upside.

[Core Bear Points]

• Moat quality is average, lacking durable pricing power.

• 2025 cash flow was significantly lifted by working-capital release, and trailing free cash flow may mislead.

• The Nutrition business has a poor historical capital-allocation record, accompanied by goodwill impairment and disclosure problems.

• The normalized Owner Earnings multiple implied by the current price is not low.

• Compared with public comparables such as Ingredion, ADM does not show a clearly superior quality/valuation combination.

[Key Assumptions]

• Normalized Owner Earnings can recover to $2.1 billion to $2.3 billion rather than staying at the 2025 trough over the long run.

• Nutrition will not produce another large impairment or strategic misstep.

• Management will not make low-return acquisitions in pursuit of scale.

• The investment-grade rating can be maintained, and the cost of capital will not deteriorate markedly.

• Global trade and biofuel policy will not undergo a structural reversal that is extremely unfavorable to ADM. Most of the above are assumptions and inferences, not facts. Their support comes from the company's historical financials, the incentive framework in the proxy statement, and policy and industry materials.

[Fair Buy Price] The range I give is $48–60 per share. The basis is conservative-to-neutral DCF, the asset method after book-value adjustment, and the 20%–30% margin of safety I require for a cyclical asset with an average moat.

[Target Holding Period] If bought, it should be held with a horizon of 5 years or more, preferably 10 years or more, because your return comes mainly from normalized earnings, improvements in capital allocation and dividends, not from short-term valuation swings.

[Expected Annualized Return] Estimating from the current price:

• Conservative scenario: 0%–3%

• Neutral scenario: 4%–7%

• Optimistic scenario: 8%–11% This already includes a common-sense allowance for dividend reinvestment, but it is not a precise forecast.

[Maximum Loss Risk] If normalized profit fails to recover for a long time and the market reprices ADM at a lower Owner Earnings multiple and a stricter asset value, a pullback toward the $35–45 range is not unimaginable, implying roughly a -40% to -55% risk of permanent capital loss relative to the current price. The worst case usually comes from three things happening at once: a downward shift in the earnings center, a compression of the valuation multiple, and another credit/governance blow.

[Tracking Metrics] I recommend continuing to track the following: normalized Owner Earnings, Ag Services & Oilseeds spreads and gross profit, Nutrition margins and signs of impairment, working-capital occupation, capital expenditure, repurchase and dividend discipline, the credit rating, changes in the value of the Wilmar investment, changes in U.S. biofuel policy, and whether management produces disclosure/control problems again.

[Signals That Trigger Reassessment] If any of the following occur, I would reassess immediately:

• Another disclosure of a material internal-control weakness or a new regulatory investigation.

• Another large Nutrition impairment.

• A downgrade of the investment-grade rating.

• Normalized Owner Earnings below about $2 billion for two or three consecutive years.

• A large acquisition with murky return logic.

• Commodity-price pass-through remains, but gross profit / pre-tax profit cannot recover.

[Reasons Not to Buy] The three most important reasons not to buy are: first, this is not a high-quality, high-certainty compounder; second, the management-trust discount has not fully disappeared; third, the current price is not cheap enough to effectively cover the industry, policy and execution risks.

[Final Recommendation] Put plainly, ADM is worth studying, but not worth owning at any price. If you are a conservative investor with a 10-year-plus horizon, I think ADM fits better on a "wait for mispricing before acting" list than as a name to buy heavily today. The truly attractive entry point should appear when the market is excessively pessimistic about the cycle, governance or short-term profit while the company's long-term survivability remains intact; at the current price, I would rather keep watching than rush to buy.

[Open questions / limitations] This report is based on verified 10-K filings, annual reports, proxy statements and authoritative public data; but because it deliberately prioritizes the company's original disclosures, some 2023 individual revenue/asset line items are not laid out one by one in the current excerpts, so certain ratios are expressed as conservative estimates or ranges. For a capital-intensive company with large working-capital swings like ADM, this conservative treatment is more reliable than a valuation that is "precise to the decimal point but built on shaky premises."

This report is based on public information and does not constitute investment advice. Markets carry risk; invest with caution.