Selling the stream: Starbucks's retreat from running its own shops

Starbucks is weighing the sale of a majority stake in its Japanese business, a deal that could value it at around $3bn. Japan is the chain's largest wholly-owned market outside America, with 1,883 stores—nearly a tenth of the global total. That such a business is on the block at all is a small confession. The world's most recognisable coffee brand is concluding, one overseas market at a time, that it does not particularly want to run coffee shops.

The confession has been a while coming. Eighteen months into Brian Niccol's "back to Starbucks" turnaround, the company is closing underperforming American stores, cutting corporate jobs and selling claims on its fastest-growing foreign markets. Japan follows China, where last year Starbucks sold 60% of its retail operations to Boyu Capital, a buyout firm, in a deal valuing the business at $4bn; counting the retained stake and future royalties, the company puts the total value of China at more than $13bn. Japan's review is at an earlier stage—advisers have been invited to pitch, a sale process could begin in the fourth quarter and the exact stake is undecided—but the shape is familiar.

The term of art is "asset-light," and it describes a genuine transformation. A company-operated store books a store's full revenue and its operating profit, along with all the rent, labour and risk that go with them. A licensed or franchised store books only a fee—smaller, but with no capital attached and no baristas to manage. If a Japan sale goes ahead, Niccol would be converting the former into the latter, edging Starbucks from a global shopkeeper towards a retailer whose brand rents itself out abroad.

The move looks backward, but for a company in his position it has a logic worth taking seriously. The home turnaround is expensive. In the quarter that ended in September, North American operating margin fell to 4.5% from 18.7% a year earlier, crushed by restructuring charges tied to store closures. Selling Japan, were it to happen, would monetise three decades of brand-building in a mature market at a moment when the balance sheet could use the cash.

The trouble is that the arithmetic treats Japan as a saleable asset, and on that test Japan does not flatter itself. The business was valued at roughly $1.5bn in 2014, when Starbucks bought out its Japanese partner. Twelve years and nearly twice as many stores later—about 1,050 then, 1,883 now—it is worth about $3bn. In other words, all that expansion bought almost no appreciation in value per store. The market has never paid up for Starbucks Japan as an operating business, which is precisely why a patient buyer and a reluctant seller can agree on a price.

That is the investor's real question, and it is not whether Japan is a good business. It plainly is: even now it is helping to drive international comparable-store sales up almost 6% in the most recent quarter. The question is whether a retailer that would license its brand abroad is worth more than the global operator it is dismantling. Niccol is betting that the residual value of the brand—the royalty stream that needs no rent, no labour, no risk—exceeds what Starbucks could earn by running the shops itself. Watch whether the fee income that replaces Japan's consolidated revenue can match the operating profit given up, and at what multiple. If the stream is worth more than the stream bed, the deal is the right one; if not, the pile of proceeds will be a poor monument to a company that gave away the thing it was best at.