BUSINESS OVERVIEW

The Coca-Cola Company

NYSE: KO · Atlanta, Georgia · SIC 2080, Beverages

7 September 2026

Evidence base: The Coca-Cola Company's SEC filings — Forms 10-K for FY2016 through FY2025 (the latter filed 20 February 2026), Forms 10-Q through the quarter ended 3 July 2026, and proxy statements — supplemented by company investor-relations releases, the filings of named competitors and bottlers, and independent government and peer-reviewed sources for industry structure.

This is an explanation of how the business works. It is not a valuation and not a recommendation.

1. Executive Snapshot

Sources: FY2025 Form 10-K; Form 10-Q for the quarter ended 3 July 2026; company Q4/FY2025 earnings release, 10 February 2026.

2. What the Company Does

The customer problem Coca-Cola solves is trivial to state and surprisingly hard to serve: a person wants a cold drink, now, within arm's reach, at a price they will pay without thinking. Satisfying that demand profitably has almost nothing to do with making the beverage and almost everything to do with getting it chilled, stocked and priced in several million separate retail locations every day. Coca-Cola resolved this by splitting the problem in two and keeping only half of it.

The unit, and the two ways the company earns on it

The unit of economics is the unit case, which the company defines as "a unit of measurement equal to 192 U.S. fluid ounces of finished beverage (24 eight-ounce servings)" (FY2025 10-K). Unit case volume counts cases sold by the company and its bottling partners to customers or consumers, so it measures the whole system, not only what Coca-Cola itself invoices. The Coca-Cola system sold 33.8 billion unit cases in 2025, against 33.7 billion in 2024 (FY2025 10-K).

Trace one case through the business and the model becomes visible. Coca-Cola manufactures concentrate and sells it to a bottler that holds an exclusive territory. The bottler adds sweetener, water and carbonation, packages the drink, loads it onto its own trucks, delivers it to retailers, and in many cases owns the cooler it sits in. The bottler collects the retail price and pays Coca-Cola for concentrate. Coca-Cola never touches the can, the truck or the shelf, and never funds them.

Where the company has not franchised — India, until recently Nigeria and the Philippines, and the other operations grouped in the Bottling Investments segment — it performs the whole chain itself and books the full retail-facing revenue. The two models therefore sit side by side inside one income statement, and the 10-K reports their proportions directly: concentrate operations generated 59% of net operating revenues on 85% of worldwide unit case volume in 2025, while finished product operations generated 41% of revenues on 15% of volume (FY2025 10-K). The company states the reason plainly: "Generally, finished product operations generate higher net operating revenues but lower gross profit margins than concentrate operations."

Dividing those disclosed proportions gives the arithmetic of the franchise decision (computed, and approximate because the geographic segments contain some finished-product activity):

Computed from FY2025 10-K disclosures: net operating revenues $47,941m; unit case volume 33.8bn; the 59/41 revenue and 85/15 volume splits; and segment operating income of $15,152m across the four geographic segments against $426m for Bottling Investments.

Company-owned bottling collects roughly four times the revenue per case and roughly one-sixth the operating profit. That single comparison explains the last decade of corporate behaviour better than any strategy statement, and it is why the Bottling Investments segment has fallen from 13.2% of net operating revenues in 2024 to 12.0% in 2025 (FY2025 10-K).

What the bottler contract actually says

The commercial power in this arrangement sits in the contract, and it is worth reading rather than summarising. The bottler receives an exclusive territory in which Coca-Cola "agrees to refrain from selling or distributing" its trademark beverages, and in exchange "is obligated to purchase its entire requirement of concentrates or syrups for the designated Company Trademark Beverages from the Company or Company-authorized suppliers" (FY2025 10-K). The bottler cannot dual-source, and cannot buy the input anywhere else.

Pricing follows an incidence model in most markets, under which "the concentrate price we charge is impacted by a number of factors, including, but not limited to, bottler pricing, the channels in which the finished products produced from the concentrates are sold, and package mix" (FY2025 10-K). Coca-Cola's concentrate revenue therefore moves with what the bottler charges the retailer rather than being a fixed fee per gallon. The consequence is that when a bottler raises prices to recover its own cost inflation, part of that increase flows back to Coca-Cola as revenue, without Coca-Cola having absorbed the aluminium, freight or labour cost that prompted it (inferred from the disclosed mechanism).

Duration differs by geography. Agreements outside the United States "generally are of stated duration, subject in some cases to possible extensions or renewals." Inside the United States, the Comprehensive Beverage Agreements run for ten years and are "renewable indefinitely for successive additional 10-year terms" at the bottler's election, and they embed mandatory incidence pricing and a binding national governance model (FY2025 10-K). The company is explicit that its theoretical pricing freedom has practical limits: its ability to set concentrate prices "is subject… to competitive market conditions."

Who pays, and how concentrated they are

Coca-Cola's direct customers are bottlers, not consumers, and there are not many that matter. Five bottling partners — Coca-Cola FEMSA, Coca-Cola Europacific Partners, Coca-Cola HBC, Arca Continental and Swire Coca-Cola — "combined represented 44% of our total worldwide unit case volume" in 2025, up from 40% in 2020 (FY2025 and FY2020 10-Ks). One bottler alone accounted for 10% of net operating revenues, reflected in the EMEA and Asia Pacific segments (FY2025 10-K). The counterparty base is small, sophisticated, and in several cases separately listed.

What is actually in the bottle

The portfolio is more concentrated than the company's "total beverage company" language implies. Sparkling soft drinks accounted for 69% of worldwide unit case volume in both 2025 and 2024, and Trademark Coca-Cola alone for 47% (FY2025 10-K). The remainder spans water, sports drinks, coffee and tea; juice, value-added dairy and plant-based beverages; and a small emerging-beverages category. A firewalled, wholly owned subsidiary now produces alcoholic ready-to-drink products including Jack Daniel's & Coca-Cola and Topo Chico Hard Seltzer — a category absent from the FY2016 description of the business entirely.

The reportable segments are EMEA, Latin America, North America, Asia Pacific, Bottling Investments and Corporate. Global Ventures, created after the 2019 Costa acquisition, was sunset effective 1 January 2025, with Costa's non-ready-to-drink business, innocent and doğadan folded into EMEA (FY2024 10-K). Revenue and profit are distributed very differently across what remains:

FY2025 10-K. Third-party net operating revenues. Percentages of segment profit exclude Corporate, which was a $1,816m cost in 2025.

Latin America is the point to hold on to. It produces 13.2% of revenue and 27.2% of segment operating profit at a 59.1% operating margin, because it is almost entirely franchised concentrate. North America produces three times the revenue at 25.9%, because it carries fountain, Costa, dairy and other finished-goods activity alongside concentrate. The margin difference is a difference in what the company chose to own, not in how well each region is run.

3. Industry, Competitive Position and Moat

The non-alcoholic ready-to-drink industry exists to solve a distribution problem dressed as a product problem. Making flavoured sugar water is not difficult and confers no advantage. Placing it, cold, in a hundred thousand outlets across a country every week, at a landed cost that leaves a margin, is extremely difficult and confers a durable one. The industry's value chain splits along exactly that line, and so does its profit.

Where the profit pool actually sits

The clearest evidence for how this industry works comes not from Coca-Cola's filings but from the filings of the companies that do the physical work.

Sources: KO Q4/FY2025 earnings release, 10 February 2026; KOF FY2025 results, Form 6-K; CCEP FY2025 results release, 17 February 2026 (margin computed from €2.8bn operating profit on €20.9bn revenue); COKE FY2025 results, $950.7m on $7,228.1m; PepsiCo Q4/FY2025 release, $11,498m on $93,925m; KDP Q4/FY2025 release, $3,575m on $16,603m.

Coca-Cola earns roughly twice the operating margin of the bottlers that carry its product, and it does so on a fraction of their fixed assets. This is not evidence that Coca-Cola is better managed than Coca-Cola FEMSA. It is evidence that the two firms occupy different positions in the same chain, and that the position matters more than the operator. The bottlers own the trucks, plants, warehouses and coolers, absorb aluminium and freight inflation, and clear 13% to 15%. Coca-Cola owns the trademark and the formula and clears 31%.

PepsiCo makes the same point from the opposite direction. Its total operating margin was 12.2% in FY2025 (computed from $11,498m on $93,925m), and its PepsiCo Beverages North America segment reported $1,089m of operating profit on $28,197m of revenue — though that figure included a $1,589m Rockstar brand impairment, so the underlying segment margin was closer to 9.5% (computed, PepsiCo Q4/FY2025 release). PepsiCo owns its own North American bottling. Its beverage business is roughly the same size as Coca-Cola's North America segment and earns a fraction of the margin, because it consolidates the part of the chain Coca-Cola gave away.

Keurig Dr Pepper is currently rearranging itself around the same logic. It closed the €15.7bn acquisition of JDE Peet's in April 2026 and intends to separate into a global coffee company and a North American beverage company, with operational readiness targeted for the end of 2026 (KDP press releases, 25 August 2025 and 1 April 2026). The strategic premise of that split — that a focused brand-and-route business is worth more than a conglomerate — is the same premise Coca-Cola acted on when it refranchised.

The market is local, whatever the share statistics say

Bottling territories are legally exclusive and geographically defined. Coca-Cola Europacific Partners serves a specific list of European countries plus Australia, New Zealand, Indonesia and Papua New Guinea; Coca-Cola FEMSA serves defined Latin American territories; Coca-Cola Consolidated serves a defined stretch of the American Southeast and Midwest. Within any one territory there is exactly one Coca-Cola bottler and one Pepsi bottler, and they compete for shelf space, cooler placement and delivery frequency at individual retailers.

This means a global market-share figure is an index rather than a competitive fact. The contest is won or lost store by store, on route density and execution, and the entity that wins it is the bottler. Coca-Cola's advantage is that it has, in most territories, the bottler with the denser route network — an asset accumulated over decades of capital spending by parties other than Coca-Cola.

Which barriers actually bind

• Route density and territory exclusivity — binds hardest. A new entrant must either build a national direct-store-delivery network at negative returns for years, or accept a distributor that already carries a competing portfolio. The bottler margin data above shows what the economics of that network look like even at scale: 13% to 15%. Nobody builds that to sell a new cola.
• Advertising scale — binds, though the comparison is incomplete. Coca-Cola spent $5.4bn on advertising in 2025, about 11.3% of net operating revenues, up from $5,146m in 2024 and $5,010m in 2023 (FY2025 and FY2024 10-Ks; percentage computed). Equivalent line items for PepsiCo and Keurig Dr Pepper were NOT FOUND in their FY2025 disclosures, so the relative comparison is unverified.
• Trademarks — binds, and is the one asset genuinely unavailable at any price. The company states that beverages bearing its trademarks account for 2.2 billion of an estimated 65 billion daily servings of all beverages consumed worldwide, against 1.9 billion of 62 billion in 2020 — a claim, but one whose numerator has grown faster than its denominator (FY2025 and FY2020 10-Ks).
• Concentrate manufacturing scale — sounds impressive, binds least. Concentrate production is comparatively low-capital. The protection here is the formula and the trademark, not the plant (inferred).

Regulation: what the evidence actually shows

Sugar taxation is the industry's most-cited regulatory threat, and enough jurisdictions have now run the experiment for the effect to be measured rather than assumed. The results are consistent in a way that is more reassuring for Coca-Cola than the headlines suggest: taxes reliably change what is in the bottle, and much less reliably change how many bottles are sold.

• Mexico, 2014, roughly one peso per litre: taxed-beverage purchases fell 4% in rural and 6.3% in urban households in year one, with the effect concentrated in mid-priced products (Health Affairs, 2017).
• United Kingdom Soft Drinks Industry Levy, 2018: around 45% of drinks originally in the highest-sugar tier were reformulated below the threshold and sugar purchased from soft drinks fell roughly 30%, but researchers found "no evidence of a reduction in consumer purchases of soft drinks by volume" (interrupted time-series analysis, 2017–2020).
• Chile, 2014: purchases of high-sugar beverages fell 3.4% by volume relative to trend (PLOS Medicine).
• South Africa Health Promotion Levy, 2018: the largest measured effect — a 29% fall in the volume of taxed beverages purchased per person per day, and a 51% fall in sugar intake from them (Lancet Planetary Health, via UNC Carolina Population Center).

The mechanism matters more than the average. Where a tax is tiered by sugar content, as in the United Kingdom and South Africa, manufacturers reformulate and volume largely survives. Where a tax is levied on the category regardless of content, volume takes more of the hit. Coca-Cola's zero-sugar portfolio is the direct commercial answer to the first design, and the company has been building it for a decade.

What kind of company wins here — and whether this is one

The evidence points to a single winning position: own the brand and the formula, franchise the trucks. Every margin figure in this section supports it, and the two large competitors are each moving toward it rather than away from it. Coca-Cola occupies that position more completely than anyone else in the industry, and has spent ten years and several billion dollars of divestiture proceeds getting closer to it.

The qualification is that the position is not the same thing as the operator. Coca-Cola's structural advantage is largely inherited; what remains genuinely contestable is whether it can keep growing the servings inside a portfolio that is still 69% sparkling, and whether the bottlers who carry it stay financially healthy enough to keep investing in the route networks that make the whole arrangement work.

4. Growth Engine

Coca-Cola grows revenue in a way that requires careful reading, because the reported number and the underlying number have diverged consistently and in the same direction. In FY2025 net operating revenues rose 2%, to $47,941m. Underneath that, organic revenue grew 5% (FY2025 10-K; Q4/FY2025 earnings release, 10 February 2026). The gap is not noise — it is the whole story of the last five years compressed into three percentage points.

FY2025 10-K; Form 10-Q for the quarter ended 3 July 2026; Q2 2026 earnings release, 28 July 2026. Components do not always sum to the total because of rounding. Organic revenue growth is a non-GAAP measure defined by the company as reported growth excluding currency and structural items.

The Latin America column is the one to look at twice. Price/mix contributed eleven percentage points and currency took away twelve, leaving reported revenue down 2%. That is not pricing power in the sense an investor usually means it. It is inflation being passed through in local currency and then translated away in dollars — the company recovering its position rather than improving it. Distinguishing that from the genuine pricing seen in North America, where price/mix added 5% in a low-inflation market, is the single most useful discipline when reading this company's revenue line.

The growth levers, ranked

• Price and mix — structural, with a cyclical component. Over FY2025, price/mix contributed four percentage points against one from volume. The structural part is the incidence pricing model, which mechanically links Coca-Cola's revenue to the bottler's realised price, plus package and channel mix shifting toward smaller, higher-margin formats. The cyclical part is inflation pass-through in high-inflation markets, which reverses when inflation does. Consequence: revenue growth is more durable than volume growth would suggest, but less durable than the headline organic rate implies in any year when Latin America is running double-digit price/mix.
• Emerging-market unit case volume — structural. In the quarter ended 3 July 2026, Asia Pacific unit case volume grew 8% and India specifically 13%; Latin America grew 3%, with Colombia up 20% and Peru up 16% (Form 10-Q, 29 July 2026). These are markets where per-capita consumption is a fraction of the developed-world level and where the constraint is distribution reach rather than demand. Consequence: the volume growth that exists is concentrated in the geographies with the lowest revenue per case, so system volume growth translates into dollar revenue growth at a discount.
• Portfolio shift beyond sparkling — management-driven. Water, sports, coffee and tea grew 2% in EMEA and 3% in Asia Pacific in FY2025, while juice, dairy and plant-based fell in three of four geographic segments (FY2025 10-K). The company reduced its brand count by roughly half, from about 400 master brands to about 200, in the 2020 portfolio review, discontinuing Odwalla among others (FY2020 10-K). Consequence: the portfolio is being narrowed toward brands that can carry advertising scale, which supports margin but limits how quickly the 69% sparkling concentration can fall.
• Currency — cyclical, and currently a tailwind. Currency subtracted two percentage points from FY2025 revenue and twelve from operating income, then added two points to revenue in the first half of 2026 as the dollar weakened against the Mexican peso, Brazilian real and euro (FY2025 10-K; Form 10-Q, 29 July 2026). The FY2026 guidance for comparable EPS growth of 9% to 10% assumes roughly three points of currency benefit (Q2 2026 earnings release). Consequence: nearly a third of guided earnings growth for 2026 is a translation effect, not an operating one.
• Calendar effects — temporary. The company disclosed that "the first quarter of 2026 had six additional days when compared to the first quarter of 2025" (Form 10-Q). Consequence: reported first-half 2026 concentrate volume growth is flattered by a shipping-day count that reverses later in the year, and the underlying demand trend is correspondingly weaker than the printed 6% organic growth suggests.

Management raised FY2026 guidance alongside the second-quarter results to organic revenue growth of approximately 5%, comparable currency-neutral EPS growth of 7% to 8%, and comparable EPS growth of 9% to 10% including the currency benefit and about one point of drag from divestitures (Q2 2026 earnings release, 28 July 2026). Stripped of currency, that is a company guiding to mid-single-digit revenue growth and high-single-digit earnings growth — which is roughly what the price/mix-plus-modest-volume engine has delivered through the cycle.

5. Margin, Cash and Capital Allocation

Coca-Cola converts revenue into profit through two mechanisms that are easy to confuse. The first is genuine: selling a concentrate whose cost of goods is a small fraction of its price, into a fixed advertising and overhead base. The second is compositional: over ten years the company sold the low-margin half of its own business, so the average margin of what remains rose without anything inside it improving. Both are real, but only one of them can happen twice.

FY2016, FY2020, FY2024 and FY2025 Forms 10-K; margins computed. These four years are NOT comparable on a like-for-like basis. FY2016 precedes the North American bottler refranchising and includes a large consolidated bottling base; FY2020 is a pandemic year in which revenue fell 11%; and FY2024's operating margin is depressed by $3,109m of non-cash fairlife contingent-consideration remeasurement and a $760m BodyArmor trademark impairment. On a comparable non-GAAP basis the company reported operating margins of 30.0% in FY2024 and 31.2% in FY2025 (Q4/FY2025 earnings release).

What actually moved the margin

Reported operating margin went from 20.6% in FY2016 to 28.7% in FY2025, which looks like eight points of operating improvement and is not. Between 2016 and 2018 net operating revenues fell from $41,863m to $34,300m, an 18% decline (computed), as the company refranchised its North American, Canadian and Chinese bottling operations. The FY2016 10-K attributes the decline explicitly: of that year's 5% revenue fall, structural changes — "acquisitions and divestitures of bottling, distribution or canning operations" — accounted for six percentage points, against a positive one point from volume and three from price and mix. Removing a business that ran near break-even raised the average margin of what was left, without changing the economics of the concentrate business at all.

The most recent step is a different kind of illusion. Operating income rose 38% in FY2025, and the company's own explanation leads with the reason: the increase was "primarily driven by lower other operating charges, due to the prior year remeasurement of our contingent consideration liability to fair value in conjunction with the fairlife acquisition" (FY2025 10-K). The FY2024 base carried a $3,109m non-cash charge. The honest comparison is the comparable margin — 30.0% to 31.2% — which is 120 basis points of improvement, not 750.

Gross margin tells the cleaner story, because it is largely undisturbed by these effects: 60.7% in FY2016, 59.3% in FY2020, 61.6% in FY2025. The underlying product economics have improved modestly and steadily over a decade. That is the real number.

Where the cash went

FY2024 and FY2025 Forms 10-K, cash flow statements. FY2021 is not covered by the filings in the evidence base, so the period is four years rather than five.

The ranking is unambiguous and has been stable for years: the dividend comes first, capital expenditure is deliberately small, buybacks are residual, and acquisitions have essentially stopped. The company's stated priorities match the realised behaviour — "investing wisely to support our business operations, continuing to grow our dividend payment, enhancing our beverage portfolio and capabilities through consumer-centric acquisitions, and using excess cash to repurchase shares over time" (FY2025 10-K). At $2.12 per share for 2026, a 64th consecutive annual increase, the dividend absorbs roughly 68% of comparable earnings (computed).

The share count does not shrink

This is the finding most likely to differ from a reader's prior. Coca-Cola's shares outstanding were 4,288m at the end of FY2016 and 4,302m at the end of FY2025 — fourteen million shares higher after a decade of continuous repurchases (computed from treasury share disclosures, FY2016 and FY2025 10-Ks). Gross buybacks were offset almost dollar for dollar by stock compensation issuance: $746m repurchased against $313m issued in FY2025, $1,795m against $747m in FY2024, and in FY2020 the company was a net issuer. Management now states the purpose without euphemism: "we expect to repurchase shares to offset dilution resulting from employee stock-based compensation plans" (FY2025 10-K).

The consequence for an owner is that per-share growth must come entirely from growth in earnings, with no arithmetic assistance from a shrinking denominator. Total return is dividend plus operating growth, and nothing else.

The two acquisitions that mattered, and what they showed

fairlife and BodyArmor were bought within two years of each other and have produced opposite results. Coca-Cola acquired the remaining 57.5% of fairlife in January 2020 for $979m in cash, booking an initial contingent-consideration liability of $270m against an uncapped earnout. The brand outperformed so far beyond the model that the liability was remeasured upward every year — $1,702m of charges in 2023, $3,109m in 2024 — and settled at $6,173m, paid in March 2025 (FY2020, FY2024 and FY2025 10-Ks). The company paid roughly six billion dollars more than it expected, because the asset was worth it.

BodyArmor went the other way. Coca-Cola paid $5.6bn for the remaining 85% in November 2021, allocating $4.2bn to the trademark. It then impaired that trademark by $760m in 2024, citing operating results "lower than expected," and by a further $960m in 2025 — $1.72bn of the original allocation written off within four years (computed, FY2024 and FY2025 10-Ks). The two outcomes together suggest a company whose brand-building machine works better on products it develops or nurtures than on positions it buys into an already-contested category.

Two cash-flow distortions, and a transaction that post-dates the filings

Reported free cash flow is uninterpretable in these two years without adjustment. The company paid $6.0bn to the IRS in September 2024 as a litigation deposit, and $6.1bn of the fairlife earnout ran through operating cash flow in March 2025. Reported free cash flow was $5.3bn in FY2025; excluding the fairlife payment, the company put it at $11.4bn (Q4/FY2025 earnings release). Guidance for FY2026 is approximately $12.4bn, being about $14.6bn of operating cash flow less about $2.2bn of capital expenditure (Q2 2026 earnings release). The underlying cash generation of the business did not move nearly as much as the reported line did.

Two matters post-date the FY2025 10-K and should be read as pending rather than done. In October 2025 the company agreed to sell 41.52 percentage points of its 66.52% interest in Coca-Cola Beverages Africa to Coca-Cola HBC, at a valuation of $3.4bn for 100% of the equity, retaining 25% with a six-year option for HBC to acquire the remainder. Closing is expected by the end of 2026 subject to regulatory approval, and a $1,274m charge has already been recorded on reclassification to held for sale (FY2025 10-K; company release, 21 October 2025). Separately, in June 2026 the company said it was exploring a public listing of Hindustan Coca-Cola Holdings, its largest India bottler, targeted for 2027 — exploratory, with no stake size or value disclosed (company release, 1 June 2026). Both continue the refranchising logic that has run since 2014.

Finally, the person executing that logic changed. Henrique Braun succeeded James Quincey as chief executive on 31 March 2026, with Quincey becoming executive chairman (company releases, 10 December 2025 and 14 January 2026). The transition is recent enough that no capital-allocation record exists to assess.

6. Cyclicality, Constraints and What to Monitor

Coca-Cola is about as close to non-cyclical as a consumer business gets, and the record contains one clean test of that. In FY2020, with restaurants, stadiums and cinemas closed across most of the world, net operating revenues fell 11% — from $37,266m to $33,014m — and operating margin held at 27.3%. Revenue exceeded the prior peak within two years. A demand shock of that severity producing an eleven-percent revenue decline and no margin damage is the strongest available evidence on cyclicality, and it is better evidence than any qualitative claim in the filings.

The mechanism behind that resilience is the channel mix. Away-from-home consumption — restaurants, entertainment, travel — collapsed, but at-home consumption partly replaced it, and the concentrate model meant the fixed costs of manufacturing and delivery sat on the bottlers' income statements rather than Coca-Cola's.

Where the business sits now

FY2025 was closer to a trough than a peak on volume and closer to a peak on margin. Worldwide unit case volume was flat for the year; North America fell 1% and Mexico fell 4%; and the company noted that certain markets "experienced high rates of inflation throughout 2025, which may continue in 2026" (FY2025 10-K). Comparable operating margin of 31.2% was the highest in the ten-year record.

The first half of 2026 shows a broad volume recovery: unit case volume up 5% in the second quarter, with North America returning to +3%, Mexico from -4% to flat, Asia Pacific +8% and India +13% (Form 10-Q, 29 July 2026). Two qualifications belong beside that. The six extra shipping days in the first quarter inflate the half-year figure, and currency swung from a two-point revenue headwind in FY2025 to a two-point tailwind. The volume recovery appears real; its printed magnitude does not.

Input costs, and who actually pays them

The industry's major inputs are moving against it. The United States Section 232 tariff on aluminium rose to 50% effective 4 June 2025 (White House fact sheet, June 2025), and LME aluminium traded around $3,292.50 per tonne on 4 September 2026, 26.3% higher than a year earlier. Sugar futures were roughly 16% higher year on year on the same date.

The important point is where those costs land. Cans, PET and freight sit on the bottlers' income statements, not Coca-Cola's, except inside the shrinking Bottling Investments segment. And because concentrate is priced on an incidence basis that moves with the bottler's realised selling price, a bottler raising prices to recover aluminium inflation increases Coca-Cola's concentrate revenue at the same time (inferred from the disclosed pricing mechanism). Coca-Cola participates in the price increase without funding the cost that caused it. The second-order risk — bottler margin compression eventually starving the route investment the whole system depends on — is real, and belongs in the next section rather than this one.

Durable versus borrowed

FY2025 10-K; Form 10-Q for the quarter ended 3 July 2026; Q4/FY2025 and Q2 2026 earnings releases.

What to monitor, and where it is published

• The split between concentrate volume and price/mix in each quarterly earnings release. Price/mix carrying more than three of every four points of organic growth means the volume engine has stalled.
• North America unit case volume, in the same release. It is the largest revenue segment and the one where price/mix cannot mask a volume problem for long.
• The divergence between concentrate sales volume and unit case volume, disclosed in the 10-Q MD&A. Concentrate persistently running ahead of unit cases means bottlers are building inventory, which reverses.
• Bottler operating margins, in the results releases of Coca-Cola FEMSA, Coca-Cola Europacific Partners and Coca-Cola Consolidated. Sustained compression there is the leading indicator for pressure on Coca-Cola's concentrate pricing.
• The Eleventh Circuit docket in Coca-Cola v. Commissioner, No. 24-13470, and the tax note in each 10-Q.
• Advertising expense as a percentage of revenue, in the 10-K selling and administrative note. A sustained fall would be the clearest early sign of the moat being harvested rather than maintained.
• Shares outstanding on the 10-Q cover page, against gross repurchases in the cash flow statement.

7. Risks, Unknowns and Questions for Deeper Work

The risks below are those that cyclicality does not capture. They are ordered by the degree to which they compound rather than by likelihood.

• The IRS transfer-pricing dispute is a recurring tax-rate risk, not a one-off payment. The Tax Court entered a decision in August 2024 requiring $2.7bn of additional federal income tax for 2007–2009, and Coca-Cola paid $6.0bn including interest in September 2024 while appealing. The company discloses that the potential aggregate remaining incremental tax and interest liability for 2010 through 2025 "could be approximately $14 billion" as of 31 December 2025, and carries a reserve of $512m against it (FY2025 10-K). The mechanism that matters is prospective: the disputed methodology reallocates income from foreign licensees to the US parent, so an adverse ruling would raise the effective tax rate permanently, not merely trigger a settlement. Oral argument was heard by the Eleventh Circuit on 25 June 2026 and no ruling had issued as of early September 2026.
• Bottler concentration transmits through three channels at once. Five bottlers carry 44% of worldwide unit case volume and one alone represents 10% of net operating revenues. Distress at a major bottler would reduce concentrate revenue, reduce equity income and potentially force impairment of the equity-method investment simultaneously. The company also notes that credit rating agencies "consider financial information of certain of our major bottling partners" in assessing Coca-Cola itself, so a bottler credit event raises Coca-Cola's own borrowing cost (FY2025 10-K). The same counterparties are the ones who negotiate incidence pricing, which means Coca-Cola's pricing mechanism and its largest credit exposure are the same set of relationships.
• Packaging regulation cannot be reformulated away, unlike sugar. Extended producer responsibility rules, deposit return schemes and restrictions on plastic formats act on the container and the distribution model rather than the liquid. The company frames the mechanism as one that "could affect our costs or require changes in our distribution model" (FY2025 10-K, Item 1A). Where a sugar tax can be answered by reformulating to zero sugar — a capability Coca-Cola has spent a decade building — a mandated reuse or deposit system changes the economics of the route network itself, which is precisely the asset the moat rests on. Quantified compliance costs were NOT FOUND in any primary source reviewed.
• Portfolio concentration is greater than the "total beverage company" framing implies. Sparkling soft drinks are 69% of worldwide unit case volume and Trademark Coca-Cola alone is 47% — proportions that have not moved in two years. Any structural break in carbonated consumption, whether from regulation, pharmacology or generational preference, hits close to half the volume base immediately, and the categories intended to replace it are the ones where the company's acquisition record is weakest.
• Acquisition execution in growth categories is unproven. BodyArmor cost $5.6bn in 2021 and $1.72bn of its trademark value has been impaired since. If organic volume growth in developed markets stays near zero, the company needs to buy its way into adjacent categories, and this is the evidence available on how well it does that.

What the sources could not answer

These are gaps in the evidence, not judgements. Each would need resolving before an investor could hold a firm view.

• The proportion of the portfolio that is low- or no-calorie. The company markets "numerous low- and no-calorie products" but discloses no percentage in any filing reviewed — a striking omission given that sugar regulation is the most-cited threat to the category.
• An independent, citable series for global or United States carbonated soft drink volume over the last decade. Government and trade-association sources do not publish one, and the only quantified series sit behind proprietary market-research paywalls, so the direction of the core category could not be verified outside the company's own disclosures.
• Competitor advertising spend. PepsiCo's and Keurig Dr Pepper's FY2025 advertising line items were not located in their primary filings, so the claim that Coca-Cola's $5.4bn constitutes a scale barrier is directionally reasonable but unverified.
• Any measured effect of GLP-1 medication on carbonated beverage purchase volumes. Projections exist; measured purchase data comparing users to non-users does not, in any primary source reviewed.
• The underlying demand trend in the first half of 2026, stripped of the six-day calendar effect and the currency tailwind. The company discloses both effects but does not isolate the residual.
• The water use ratio. Water scarcity is named as a risk in Item 1A, but no litres-per-litre efficiency figure appears in any of the four 10-Ks examined.
• Deal economics for the pending Coca-Cola Beverages Africa disposal beyond the headline $3.4bn equity valuation, and any terms at all for the exploratory India listing.
• Research and development expense, which is not broken out as a line item in any filing reviewed.

8. Investor Takeaways

• This is a trademark and pricing business wrapped around someone else's distribution network. Coca-Cola sells concentrate to bottlers who must buy it exclusively from the company, and lets those bottlers carry the trucks, plants and coolers — which is why it earns roughly 31% operating margins while they earn 13% to 15%.
• The economic engine is the unit case sold through a franchised bottler: about $0.98 of revenue and $0.53 of segment operating profit, against $3.88 and $0.08 where the company owns the bottling itself. Ten years of divestitures are that arithmetic being acted on.
• Growth comes from price and mix rather than volume — four points against one in FY2025 — with the volume that does exist concentrated in emerging markets that generate the least revenue per case.
• The story breaks if bottlers stop being able to fund their route networks, if the Eleventh Circuit affirms a transfer-pricing methodology that permanently raises the tax rate, or if packaging regulation forces a distribution model the franchise system was not built for.
• Watch the price/mix-versus-volume split each quarter, North America unit case volume, and the share count — which has not fallen in a decade, so per-share growth must come entirely from the business.