Case study — Food & beverage · 12 min read · 5 questions
Why Starbucks is fighting its own baristas
Published · Updated
The thesis
Starbucks is the poster child of corporate social responsibility, taught in business schools as the model of branding, vertical integration and doing right by employees. It is also running the ugliest anti-union campaign in American retail. Apple, REI, Walmart, Target and Amazon all face organizing drives, and all of them hide behind jargon and bureaucracy. Starbucks chose to be as subtle as napalm. For a company selling $5 cups of syrup and sugar made in seconds by minimum-wage labor, the obvious question is whether it really cannot afford a few dollars more an hour.
In cash, of course it can — it takes home $3 billion a year. What it cannot afford is the margin. Run the union's $20 demand through the company's own wage disclosures and total wages rise 16%, adding $1.38 billion of operating expense and dropping the operating margin from 14% to 10%. That is worse than Chipotle and level with a full-service restaurant. Even a smaller raise pushes it toward single digits.
That number is the whole case. Kevin Johnson sold the CPG business to Nestlé for $7 billion, so Starbucks is now purely a retailer. Europe has never worked and China is at the mercy of a government that changes its mind. The image of a fast-growing, forward-thinking, socially responsible company rests on margins that do not look like food service — and unions would make labor a fixed cost and expose the fundamentals underneath. Business is as much optics as product.
How do you think about this? 5 strategy questions this case raises and does not answer.
Read the comments, or add yours
The statistics
By the numbers — swipe or use arrows
Revenue by segment and geography, operating margins, store counts by ownership type, retail wage totals, CPG revenue and the Nestlé transaction from Starbucks Corporation annual reports and 10-K filings, 2008 through 2022; acquisition prices for Le Boulange, Evolution Fresh and Teavana as announced; customer mix and cold beverage share from company earnings calls; union demands as published by Starbucks Workers United. The wage calculation is derived, and its assumptions are stated in the takeaways
Key takeaways
Beverages are a better business than food and that is why everyone keeps entering. Water, milk, syrups and beans arrive stable, ready to use and slow to spoil, preparation is choreographed into recipe cards anyone can learn in days, and the product sells for $5 against about $1 of cost. The catch is transaction size: the margin per cup is high, so the model needs volume.
Howard Schultz built that volume, and the credibility he spends today. He left in 2000, returned in 2008 to a company his successors had broken, left again in 2017, and forced his way back in 2022 after a failed presidential campaign.
What broke it the first time was ordinary MBA expansion. Chasing Wall Street targets, Starbucks opened stores sometimes within a mile of each other without proper site selection, spread staff thin, and switched to pre-packaged ground coffee, powdered mixes and automated machines — until customers asked why they paid $4 for a cup from a bag when McDonald's charged $1.
The 2008 turnaround was operational, not strategic: freshly ground coffee, mandated in-store aroma, shorter machines so customers could watch drinks being made, brewing every 30 minutes in small batches, hundreds of underperforming stores closed and baristas retrained nationwide. Domino's copied the public-self-criticism playbook a year later.
The genuinely durable wins were the unglamorous ones. Starbucks Rewards remains one of the best loyalty programs in retail; gift cards sold through big-box retailers leave $1.3 billion of unused balances sitting on the balance sheet; and free Wi-Fi with no purchase necessary still has no equivalent at any American chain a decade later.
Schultz bought his way beyond coffee. Le Boulange bakery for $100M purely to make food worthy of the brand, Evolution Fresh for $30M to enter cold-pressed juice, and Teavana for $620M to reach a $90 billion global tea market.
The expansion model is franchising with the franchisees screened out. Rather than licensing to individuals, Starbucks partners only with established companies — grocery chains, colleges, hospitals, hotels, airports — who have no incentive to cut corners because Starbucks is supplemental income, not their business. It is why some stores refuse gift cards or rewards.
It also removes cannibalization by design. Licensed stores sit in airport terminals, supermarkets, train stations and campuses — captive venues the public cannot otherwise reach, which do not compete with corporate stores for the same customer.
The second Schultz era worked on the top line: revenue went from $10.3B in 2008 to $22.3B by 2017, and store count grew 50% from 16,000 to 27,000 — company-operated up 40% to 13,000, licensed up 60% to over 14,000.
But the CPG business was the margin story, and it was always small. VIA, K-Cups, bottled Frappuccinos, Evolution Fresh and Teavana grew from $674M in 2010 to $2B by 2017 at 40%-plus operating margins — and still never exceeded 11% of total revenue. Licensed stores contributed about 9%. Company-operated stores were 80%.
Geography hid a problem. America runs low-20% operating margins on over 60% of revenue, and Asia matched it — $400M in 2011 to $3.2B by 2017 as store counts went from 3,000 to over 7,000. Europe went backwards: revenue fell from $1.046B in 2011 to under a billion by 2017 despite growing from 1,869 to nearly 3,000 licensed stores, at a 9.8% margin.
Then Kevin Johnson sold the margin. He signed away global rights to market and distribute every Starbucks CPG product to Nestlé for $7 billion plus a profit share — undoing the one segment Schultz had built specifically to lift the company above retail economics. Teavana went to Unilever at a loss.
The product has drifted past coffee entirely. Millennials and Gen Z are 51% of US customers, over 76% of American beverages sold are cold, and over 60% of all drinks are customized — leaving the largest coffee chain in the world citing its collection of syrups and modifiers as its competitive advantage.
The union's demands are narrow: protection from no-notice termination, guaranteed hours, and $20 an hour against roughly $15 today. Starbucks employs 248,000 retail workers in its US corporate-operated stores.
The arithmetic is the case. Of 18,000 corporate-operated stores worldwide, 10,000 are in North America — about 56% — so roughly 56% of the $8.1 billion paid in retail wages in 2022, or $4.5 billion, went to US workers. At 25 hours a week across 48 weeks that reproduces the known ~$15 an hour, which is what validates the method.
Run it again at $20 and total wages rise 16% to $9.5 billion — $1.38 billion of additional operating expense, dropping the operating margin from 14% to 10%. Even a smaller 6-9% rise costs 1-2 points. Strip away the branding and the economics are a restaurant's: ingredients 30%, labor and rent 50% — which is exactly the thing the image cannot survive being seen.
Common questions
Why is Starbucks against unions?
Because of what unionization would do to its operating margin, not to its cash. Starbucks takes home about $3 billion a year and could pay the raise. But the union's $20 an hour demand would raise total wages roughly 16%, adding about $1.38 billion in operating expense and dropping the operating margin from 14% to 10% — worse than Chipotle and level with a full-service restaurant. It would also convert labor into a fixed cost. The company's valuation and reputation rest on looking like a fast-growing innovator rather than a food service business, and margins in the low teens make that story hard to tell.
How much do Starbucks baristas make?
About $15 an hour on average in the US as of 2022, against a union demand of $20. The figure can be derived from the company's own disclosures: of 18,000 corporate-operated stores worldwide, roughly 10,000 are in North America, so about 56% of the $8.1 billion paid in retail wages — around $4.5 billion — went to the 248,000 US retail workers. At 25 hours a week over 48 weeks, that works out to roughly $15 an hour, which matches the known figure.
Is Starbucks actually profitable?
Yes, but far less impressively than its image suggests. It generates around $3 billion a year with stable cash flow, and the American business runs low-20% store-level operating margins. Company-wide operating margin sits around 14%. Once you strip out the branding, the cost structure is a restaurant's — ingredients around 30%, labor and rent around 50%, the rest absorbed by overhead — despite Starbucks cooking almost nothing and selling a product that costs about $1 to make.
Why did Starbucks sell its CPG business to Nestlé?
Kevin Johnson signed away the global rights to market, sell and distribute Starbucks CPG products — K-Cups, VIA instant coffee, bottled drinks — to Nestlé for $7 billion plus a profit share. It was counter to everything Howard Schultz had built the segment for: CPG ran 40%-plus operating margins and was the one part of the company whose economics were better than retail. Selling it left Starbucks entirely dependent on its stores for both income and growth, which is precisely what makes labor costs existential now.
What does the Starbucks union actually want?
Three things, and they are not extravagant: protection against no-notice terminations, guaranteed hours, and a $20 an hour starting wage in all states, up from roughly $15 today. The modesty of the list is part of the case — the fight is not about the specific demands but about what conceding any of them would do to the company's margin structure and, by extension, to how investors read the business.
How did Howard Schultz save Starbucks in 2008?
By reversing his successors' cost-cutting rather than by innovating. They had over-expanded — opening stores within a mile of each other to hit Wall Street targets — and switched to pre-packaged ground coffee, powdered mixes and automated machines, until customers noticed they were paying $4 for what McDonald's sold for $1. Schultz closed hundreds of stores, mandated freshly ground coffee and in-store aroma, installed shorter machines so customers could watch drinks being made, required brewing every 30 minutes, and retrained baristas nationwide. The public self-criticism was rare enough that Domino's copied the playbook a year later.
Why does Starbucks use licensed stores instead of franchises?
To keep control. Instead of licensing to individuals the way McDonald's or KFC do, Starbucks partners only with established companies — grocery chains, colleges, hospitals, hotels, airports and resorts — that have exclusive space and want more income from land they already hold. Those partners have little incentive to cut corners, because Starbucks is supplemental revenue rather than their livelihood. It also avoids cannibalization, since licensed stores sit in captive venues corporate stores could never reach. The trade-off shows up when you visit one: some don't accept gift cards, honor rewards, or match prices.
Is Starbucks still a coffee company?
Increasingly not, by its own numbers. Over 76% of beverages sold in the US are cold, more than 60% of all drinks are customized, and millennials and Gen Z make up 51% of the customer base. The company now describes its ability to customize — its range of syrups, flavors, modifiers and foams — as its main competitive advantage. Most of the menu sits closer to dessert than to coffee, which is a strange position for the largest coffee chain in the world.
Discussion
Starbucks can afford the raise in cash and cannot afford it in margin. When a company resists a cost it can pay, what is it actually protecting — and who is that protection for?
No answers yet — be the firstApple, REI, Walmart, Target and Amazon face the same organizing and none of them fight like this. What makes Starbucks behave differently, given it has the most to lose reputationally?
No answers yet — be the firstKevin Johnson sold the CPG business to Nestlé for $7 billion — the one segment carrying 40% margins and the company's claim to being more than a retailer. Reconstruct the argument for that sale. Does any version of it work?
No answers yet — be the firstThe company now cites customization and syrup range as its competitive advantage. Is that a real moat, or the sound of a business that has run out of ones?
No answers yet — be the firstEurope has lost money for a decade while Asia covered for it. At what point does a segment stop being an investment and start being a decision nobody wants to make?
No answers yet — be the first
Related case studies
Food & beverageHow much a Texas BBQ joint actually makesOne in three fail, brisket yields 40%, and the biggest variable in whether you survive is one magazine editor's palate.16 min read
Food & beverageWhy nobody wants to own ice cream anymoreUnilever, Nestlé and Kroger all owned the biggest brands in ice cream. All three decided it wasn't worth keeping.16 min read
Food & beverageWhy donut shops out-earn Krispy KremeThe average Krispy Kreme grosses $119K a year. A 650 sq ft Portland shop grosses $600K at a 40% margin.12 min read