One Battle After Another: The War for the American Stomach

Every American eats at these places, has an opinion about these places, and — if they own an index fund — owns a piece of these places. That familiarity is exactly why restaurant stocks are where retail investors get overconfident: knowing the menu is not the same as knowing the business. So this is going to be a walk through a food court of names you’re all familiar with — names that don’t all compete with each other for market share ($SBUX and $MCD overlap on some products but serve different primary customer bases) — pairing up direct competitors in fast food, fast coffee, and fast casual. The one thing all six DO compete for: the same finite dollars in the American consumer’s pocket.

Here’s the problem, and it’s the spine of everything that follows: America’s restaurant traffic has stopped growing. Circana projects industry traffic growth below 1% this year, after traffic actually fell about 0.8% in 2025. The pie is fixed. Every incremental customer at one counter walked away from another counter. That turns the entire restaurant sector into a share war, and it means the only honest way to read these companies is to ask one question the earnings releases are all built to obscure: is anyone actually walking in the door, or is the “growth” just the register charging more?

Comparable sales — the industry’s favorite number — is price times traffic. Price is a decision. Traffic is truth. And when you decompose the latest quarter into price and traffic, these companies stop looking like one sector and start looking like six different bets — which is why I’ve paired them into three battles: the value war (McDonald’s vs. Yum), the coffee war (Starbucks vs. Dutch Bros), and the bowl war (Chipotle vs. Cava). In each one, an established incumbent defends its turf against a faster-growing insurgent. No single winner this time; a horse race has one finish line, and these six aren’t even running the same race. By the end you’ll know which war you’re actually betting on with each name, and which one fits the investor you are. (And whether the answer is “none of them” — that’s a legitimate row in the table.)

The Pie Stopped Growing

Before the three battles, three industry facts that sit under everything.

First, the traffic freeze. Sub-1% industry traffic growth (Circana) means the sector’s aggregate volume is a rounding error from flat. Fast casual — the growth segment — decelerated from 3.3% traffic growth in December 2024 to 1.7% by October 2025 (Technomic), as customers started asking hard questions about $15 entrees. When the growth segment slows to a walk, everyone below it is fighting over scraps.

Second, the bottom fell out of the bottom. The low-income consumer — the traffic backbone of legacy fast food — has been eating at home more for over a year. McDonald’s own management says low-income traffic is still declining, just “not as pronounced as they were maybe 6 or 12 months ago when we were talking about high single digit,” and warned that rising gas and grocery prices land hardest on exactly the cohort its value menu is built to retain. When the company whose founding premise is cheap food for everyone says its customers can’t afford it, the problem is bigger than McDonald’s.

Third, the quiet revenge of the little guy. Independent and specialty coffee keeps taking share — the segment is growing while the big chains stall — and the youngest coffee drinkers keep telling surveys they prefer independents. The chains spent two decades teaching Americans to pay $6 for coffee; that generation is the first to question the price. Starbucks’ turnaround — and it is turning, we’ll get there — is happening into this headwind, not in the absence of it.

And hovering over the entire sector, the thing nobody’s income statement shows cleanly yet: GLP-1s. Roughly one in eight American adults now uses an appetite-suppression drug, and nearly one in five have tried one (KFF, late 2025) — and every packaged-food and restaurant CEO gets asked about it every quarter. For a long time the honest answer was that you couldn’t see it in restaurant traffic — it was one candidate explanation inside the low-income decline, tangled up with gas prices and grocery inflation. That’s starting to change: a Cornell study published in late 2025 found households on GLP-1s cutting their fast-food, coffee, and limited-service spending by around 8% (grocery by ~5%), though survey evidence is still mixed and some users spend more. If you are underwriting any of these stocks on a 10-year horizon, you are making a call on American appetite itself, whether you admit it or not.

One last piece of scaffolding, because it decides who actually carries the risk in everything above: this is a US story, and a frozen American pie is far more dangerous to some of these names than others. McDonald’s, Starbucks, and Yum are global. McDonald’s now earns a majority of its operating income outside the US — 53% international versus 47% domestic in FY2025 (per its 10-K); Yum runs roughly 72% of its restaurants abroad, with KFC about 90% non-US; and even Starbucks, the most US-weighted of the three, still books about a quarter of its store-level profit internationally. The three insurgents have no such hedge: Chipotle is ~97% US, and Cava and Dutch Bros are 100% domestic. So the frozen-US-pie thesis this whole piece is built on lands hardest on exactly the names carrying the most growth in their valuation — the incumbents can lean on a foreign business the insurgents simply don’t have. Put another way: the insurgents are winning the frozen pie right now — but they’re also the only ones with nowhere else to fight.

Before the pairings, one chart: each name’s latest quarter split into traffic and price.

Battle I — Value: $MCD vs $YUM

The value war is the fight for the trade-down dollar — the customer who used to order without checking prices and now checks. It’s the war the macro pushed to the front, and its two combatants run nearly identical business models to very different results.

$MCD: The Landlord

Start with the name everyone thinks they understand, because almost nobody does. McDonald’s is not a burger company. Roughly 95% of its restaurants are owned by franchisees; the corporation’s actual business is collecting royalties on system sales and rent on the land underneath — a tollbooth on more than 45,000 locations’ revenue, not their profits. Harry Sonneborn, McDonald’s first president and the man who built its finances, said it out loud sixty years ago: “We are not technically in the food business. We are in the real estate business. The only reason we sell fifteen-cent hamburgers is because they are the greatest producer of revenue, from which our tenants can pay us our rent.” Royalties on sales is a beautiful place to sit in an inflation: when franchisees raise menu prices to survive, corporate’s cut rises automatically, and corporate never buys the beef.

The latest quarter shows the machine working. Revenue up 9.4%, its strongest in several quarters; global comps up 3.8%, US comps up 3.9%; EPS beat. And here’s the part that surprised me, because it cuts against the easy story: US traffic was positive. Management didn’t ride price alone — guest counts grew, and McDonald’s took share from its near-in competitors. But underneath that positive headline, the low-income guest kept shrinking, “still declining,” just more slowly than the high-single-digit drops of a year ago. So the real McDonald’s picture isn’t “declining cars on the toll road” — it’s subtler and, if anything, more interesting: the tollbooth is gaining cars overall by pulling in trade-down traffic from pricier competitors, even as its own founding customer, the person the Dollar Menu was invented for, quietly thins out. That works — genuinely, for years at a time — right up until either the trade-down reverses or franchisee margins crack, and franchisee patience is the load-bearing wall of the whole model. (Hold the thought about where the low-income guest actually went — that’s the other half of this war.)

What you’re buying at $269.43 and about 22x trailing, down 12% on the year and a hair off its 52-week low, with a dividend raised every year for 49 straight: a bond with a brand attached. The bet is that value-menu economics plus the royalty structure make the earnings nearly recession-proof even if American fast-food traffic never grows again. The risk is that “49 years of dividend growth” is exactly the phrase people used about other aristocrats right before their moats turned out to be demographic. McDonald’s is the pick for the investor who wants food-sector ballast and will trade away the upside to get it. It is nobody’s growth story anymore, and the market finally priced it like that this year, which, for the ballast buyer, is the point of entry, not the warning.

$YUM: The Counter-Landlord

Now the army on the other side of the value war, and the reason the McDonald’s story can’t be blamed on the consumer alone. Yum Brands is three franchised chains stapled together — KFC, Pizza Hut, and Taco Bell — and the composition is the whole point. Taco Bell is the thesis: 8% US same-store sales growth last quarter, its eighth consecutive quarter leading the US industry, and the mix is the tell — three of those points were transactions. Actual humans, choosing the value menu, at the exact moment McDonald’s is telling us the low-income consumer is disappearing. They’re not disappearing. They’re at Taco Bell. The Luxe Value menu is winning the trade-down dollar the Extra Value Meals relaunch was built to defend, Taco Bell’s operating profit grew 16% to $281 million, and Yum’s digital mix hit a record 63%. Same franchised-royalty model as McDonald’s — Yum is even more of a pure franchisor, 97% of its restaurants in franchisee hands versus McDonald’s ~95% — same macro, same customer, opposite traffic result.

What you’re buying at $161.59, about 24x forward and up ~7% on the year: the same landlord economics as McDonald’s, with one crucial difference — the flagship tenant is gaining share of the value war instead of defending it — and you pay a few turns more than McDonald’s ~20x forward for the privilege. The catch is the portfolio. Taco Bell is roughly 44% of Yum’s operating profit but travels with two passengers: KFC’s US system sales fell 2% last quarter and Pizza Hut’s US comps fell 4%, dragging Pizza Hut’s division profit down 14% — a business Yum has now put under strategic review and reports around, quoting its own numbers “excluding Pizza Hut.” You’re buying the best brand in fast food inside a holding company that waters it down.

Battlefield report — the value war: the insurgent is winning on traffic, and taking more price on top of it; the incumbent is defending with price alone. If you believe the trade-down era persists, the uncomfortable conclusion is that the value war’s best asset isn’t under the arches — it’s a Cravings Box. But you can’t buy it clean; you buy it wrapped in a declining pizza chain.

Battle II — Coffee: $SBUX vs $BROS

The coffee war is the fight over a daily, habitual purchase — the most valuable kind in restaurants, and the most contestable. Two decades of chains trained Americans to treat a $6 coffee as a fixed cost; that spend is now under pressure from faster, cheaper drive-thru formats and a consumer who has started trading down on it. Starbucks is repairing traffic it spent years losing to its own missteps; the insurgent behind it never had that problem to fix.

$SBUX: The Turnaround That Already Turned

Here’s where I have to argue with my own premise. I came into this piece ready to write “Starbucks is losing to local cafes” — the third-place-turned-mobile-order-warehouse story, the unionized-and-demoralized story. That story was true. It’s why the stock needed Brian Niccol at all. But the current numbers say the turnaround is no longer a hope — it’s printing. Global comps up 6.2% last quarter, and the composition is the part that matters: transactions up 3.8%, ticket up 2.3%. Actual humans, actually returning. Guidance raised to comps of at least 5% and non-GAAP EPS of $2.25–2.45. And the China problem — a decade of capital and attention for a business that just comped +0.5% — got structurally answered: the Boyu joint venture closed, Starbucks keeps 40% and the brand economics, someone else fights the local price war.

The market noticed. Up about 23% this year, one of the three names in this piece the market has genuinely re-rated — which means the question isn’t “is Niccol fixing it,” it’s “what are you paying for the fix now that everyone agrees it’s working.” The secular pressure I planned to build the section on didn’t disappear because the comps turned: the independents keep growing, the youngest coffee drinkers keep telling surveys they prefer local, and Starbucks’ answer to that — being faster, cleaner, friendlier, more like a coffee shop again — is a treatment, not a cure. The bet here is momentum with a competent operator: the fastest earnings revisions in the group, bought at a price that already includes the applause. You’re not early. You’re paying for certainty, and certainty at up-23%-YTD prices has a way of being expensive precisely when the easy half of the turnaround is done. (Worth knowing on the multiple: Starbucks’ trailing GAAP P/E looks absurd — near 80x — because reported earnings are still depressed by the turnaround spend; the honest lens is the roughly 44x you pay on the company’s own forward guidance.) Starbucks is the pick for the investor who wants the operator, trusts execution stories, and accepts that the discount window closed months ago.

$BROS: The Insurgent

And here’s who Starbucks is actually fighting, besides the independent down the street. Dutch Bros is the Cava of coffee: a drive-thru beverage machine founded in 1992 in Grants Pass, Oregon by two brothers with a pushcart and an espresso machine, now 1,177 shops across 25 states and growing revenue 31% a year. Last quarter system same-shop sales rose 8.3% — and transactions rose 5.1%. Set that number next to Starbucks’, because it’s the entire coffee war in one comparison: Starbucks’ turnaround, the one the market paid a 23% re-rating for, brought 4.3% more humans through its US doors. Dutch Bros grew humans 5.1% — while opening 185-plus new shops a year, without a turnaround, without a celebrity CEO, without a single “Back to Starbucks” campaign. One of these companies is fixing its traffic. The other one never broke it.

What you’re buying at $73.31, up ~20% on the year and about 73x forward earnings: the beverage version of the Cava trade — real traffic growth, a long unit runway (management now pegs the US opportunity at more than 7,000 shops, up from a prior 4,000, against those 1,177 today), guidance just raised across the board. And the same problem as Cava: a multiple that assumes the runway gets built. Same duration risk, same no-margin-for-a-bad-year setup, same psychological question — can you hold it through the stumble.

Battlefield report — the coffee war: the incumbent is executing a genuinely impressive recovery into a category where the insurgents — chains and independents both — are still growing faster than it is. Paying up for the fixed incumbent or paying more for the unbroken insurgent is a legitimate coin flip; what’s not legitimate is calling either one cheap.

That scoreboard is missing the second-biggest name in American coffee, because it doesn’t trade: Dunkin’, the largest US coffee chain by store count after Starbucks — roughly ten thousand shops, private since Inspire took it off the market in 2020 — grew systemwide sales only about 3.5% in FY2024, while a pack of private drive-thru upstarts behind Dutch Bros (7 Brew, Scooter’s) compounded several times faster. The private board near the end of the piece has those numbers.

Battle III — Bowls: $CMG vs $CAVA

The bowl war is the cleanest fight of the three: the same format — assembly-line fast casual, a rail of proteins and toppings, roughly $3-million unit volumes — one company running the playbook the other one wrote fifteen years ago. Incumbent versus insurgent with almost none of the confounders. Just one is cheap and stalled, and the other is expensive and sprinting.

$CMG: The Broken Compounder

Chipotle is the LULU of this piece, and I mean that precisely: a genuinely elite operating model that the market has re-priced as broken because the momentum stopped.

The fall first, honestly: four consecutive quarters of declining traffic before last quarter’s +0.6%; earnings down 18% year over year; restaurant-level margins compressed about 250 basis points to a still-excellent 23.7% (adjusted); the stock down about 5% this year — though it’s clawed back a good chunk of a much deeper drawdown — and still roughly half off its split-era high. The Brian Niccol departure tax, in effect — he built the machine, left for Starbucks, and the machine promptly stalled under price increases that finally found the customer’s ceiling.

Now the other side. The machine itself — the thing that made Chipotle a hundred-bagger — is a unit-economics engine: roughly $3.1 million in average unit volume, throughput per square foot that the rest of fast casual has spent fifteen years failing to copy, no franchisees to negotiate with, no debt, and a new-store paradigm that still earns returns most retailers would kill for. Traffic just went positive again — barely, +0.6%, but the four-quarter losing streak is broken. And here’s the piece of history that reframes the whole name: McDonald’s incubated Chipotle — invested starting in 1998, built its stake to roughly 90%, then spun it out in a January 2006 IPO at $22 a share. Chipotle has since returned roughly 80-fold — about +7,850% — against roughly +670% (about 7.7x) for McDonald’s over the same span, both price-only from that $22 IPO (a split-adjusted basis of 44 cents after the 2024 fifty-for-one split). The legacy giant’s single best investment of the century was the fast-casual insurgent that would help freeze its own pie. That’s the industry’s whole share-shift story in one corporate family tree.

What you’re buying: quality at a bruise, mid-repair, with the specific risk that the price increases which broke the traffic did permanent brand damage in a value-conscious tape. This is the contrarian’s pick — the same shape as my Lululemon argument — and it demands the same honesty: cheap-for-a-reason stays cheap until traffic proves otherwise. Watch transactions, not comps. Two more positive traffic quarters and this is the best risk/reward in the group; two negative ones and the compounder story is a memory.

$CAVA: The Rotation

And then the one your growth-investor friend already owns. Cava is running the exact playbook Chipotle ran fifteen years ago — category-defining fast casual, Mediterranean instead of Mexican — and the numbers are the real thing: revenue up 32%, same-restaurant sales up 9.7%, and the crown jewel, traffic up 6.8% — the strongest growth in actual humans in this entire piece. Restaurant-level margins of 25.1% — higher than Chipotle’s — with $403 million in cash, zero debt, and guidance just raised. Roughly 459 restaurants against Chipotle’s ~4,090 is the argument in one ratio: if the concept has even half of Chipotle’s runway, the unit count can grow nearly ten-fold, and unit growth times traffic growth is how hundred-baggers happen.

The price of admission is the entire problem: about 150x trailing earnings, in a sector where the growth segment is decelerating and the $15-bowl question is being asked at every counter. The stock is actually up about 36% this year — still the best performer of the six — but it’s roughly half off its 2024 all-time high, and it’s still the most expensive name in the group by a mile. The June run that took it there had a name attached — a June 10 UBS upgrade to Buy with a $90 target, on exactly the same-store-sales scarcity this piece is about — and it has since given some of that pop back, which is the Cava experience in miniature: real business, violent tape. At that multiple you are not buying this year’s earnings or next year’s; you are buying a decade of flawless expansion, in advance, with no margin for a single bad year of the kind Chipotle just had — and Chipotle’s bad year is the base rate, not the tail. Every great fast-casual story eventually prints one.

What you’re buying: duration risk on execution. The Cava bet isn’t about whether the bowls are good (they are) or whether the unit economics work (they demonstrably do); it’s about whether you can hold a 100-plus multiple through the inevitable stumble without selling the bottom. That’s a psychological question more than a financial one, which is exactly why this piece doesn’t end in a verdict.

Battlefield report — the bowl war: the insurgent isn’t just matching the incumbent, it’s out-earning it at the store level — 25.1% versus 23.7% — on a ninth of the footprint. Chipotle is the value play only if you believe the damage is temporary; Cava is the growth play only if you can stomach the multiple. The bowl war is the one with a real winner on the field and no cheap way to own it.

There’s one more field in this war, and it’s the one you can’t buy. Some of the fastest-growing chains in American food are private — no ticker, no comps line, just franchise-disclosure documents and Technomic estimates. Dave’s Hot Chicken grew systemwide sales roughly 57% in a single year before Roark bought control at a ~$1 billion valuation; 7 Brew, a Blackstone-backed drive-thru coffee upstart, nearly tripled its. That’s where a lot of the appetite that never shows up in the public tape actually went — and it spans the spectrum, from those drive-thru wildfires down to mature giants like Chick-fil-A and Dunkin’ still grinding out low-to-mid single digits.

The Table

Strip the sector tag off and line up what each dollar actually buys — two names per battle, plus the honest seventh row:

• McDonald’s (value): royalty income, gaining trade-down traffic while its low-income core thins. A yield instrument wearing a brand. Fails if the trade-down reverses and franchisees crack.
• Yum (value): the winning tenant — Taco Bell — inside a diluting portfolio. The best brand in fast food, sold to you wrapped in a shrinking pizza chain. Fails if the KFC/Pizza Hut drag outweighs Taco Bell’s engine.
• Starbucks (coffee): an execution turnaround, fully noticed, priced after a 23% run. Fails if the momentum was the easy half and the local-coffee grind takes back the gains.
• Dutch Bros (coffee): the unbroken insurgent at a price that assumes it can’t break — real traffic, a 7,000-shop runway, a triple-digit trailing multiple (~116x; ~73x forward). Fails the first time traffic prints below +3.
• Chipotle (bowls): an elite machine at a discount, guilty until traffic proves innocent. Fails if the pricing damage is permanent.
• Cava (bowls): the only true growth asset in the group, at a multiple that prices perfection through 2035. Fails if it has one normal year too early.
• None of the above: the honest seventh row. A sector with frozen traffic, a squeezed core customer, rising labor costs, and an unquantified pharmacological overhang is a share-transfer machine, not a growth industry. If you don’t have a specific view on one of these six machines, the rotation out of the industry entirely — the thing this piece was half-built to test — is not cowardice. It’s the base case.

What Could Go Wrong (With All Of It)

The sector-wide bear is bigger than any single name’s. If GLP-1 penetration keeps compounding, every terminal value in this piece quietly shrinks — the landlord’s rent, the counter-landlord’s Cravings Boxes, the turnaround’s transactions, the insurgent’s drive-thru lines, the machine’s throughput, the rotation’s decade of assumed expansion, all of it, at once, slowly. The Cornell data is the first real signal that it’s already begun at the margin. No quarterly print will announce it. It will just show up as an industry that stops beating, then stops guiding, then stops growing per-unit volumes entirely, and everyone will argue about gas prices the whole way down. I don’t know if that’s the world we’re in. Nobody does yet. But it’s the one risk all six names share a boat with, and it’s the reason the “none of the above” row exists.

Conclusion

Six names, three battles, one frozen pie. In value, McDonald’s collects the toll on trade-down traffic while its poorest customer disappears, and Taco Bell out-traffics it. In coffee, Starbucks fixed its traffic and Dutch Bros never broke it. In bowls, Chipotle is cheap while it proves the damage isn’t permanent, and Cava is magnificent and priced for a decade without weather. “Which is best” is the wrong question — because these six aren’t running the same race. The right one is which failure you can actually sit through: a decade of nothing (McDonald’s), a great brand taxed by a bad one (Yum), full price after the fix (Starbucks), a triple-digit trailing multiple (~116x) riding on a thousand more drive-thrus (Dutch Bros), two more ugly traffic prints (Chipotle), or a 40% air pocket on a normal year (Cava). Pick your poison, size it honestly — or take the seventh row and let someone hungrier fight over a pie that stopped growing.

view all sources here

Suggested Citation: Rettke, Sterling. “One Battle After Another: The War for the American Stomach.” sterlingrettke.com, July 2, 2026.

Disclosure: I do not currently hold positions in any of the companies discussed ($MCD, $YUM, $SBUX, $BROS, $CMG, $CAVA). This piece is for informational and educational purposes only and is not investment advice. Do your own research.