Amazon preserves postal share

Amazon has spent years building its own delivery machine. It now moves most of its U.S. packages itself, through a dense mesh of warehouses, vans, contractors, and local hubs. That made this week’s news look almost backward at first glance: Amazon reached a new package-delivery agreement with the U.S. Postal Service, and the Postal Service kept roughly 80% of the parcel volume it already handles for the retailer. That still means a cut. But it is far smaller than the collapse Amazon had threatened only weeks earlier. Sources told Reuters the new deal preserves more than 1 billion packages a year for USPS, which makes Amazon the agency’s biggest single customer by a wide margin (finance.yahoo.com, channelnewsasia.com). That matters because this was never just another vendor contract. USPS has a budget of about $80 billion, and Amazon’s business was worth about $6 billion a year to the agency, according to Reuters. Postmaster General David Steiner said in December that USPS was delivering about 1.7 billion Amazon packages annually. Losing most of that volume would not have been a routine setback. It would have torn a hole in the Postal Service’s package economics at the exact moment the agency has warned it could run short of cash as soon as October and is pushing price increases to stem years of losses (channelnewsasia.com, money.usnews.com). The strange part is that Amazon did not actually want a clean break. In March, the company said it had negotiated with USPS for more than a year and had been trying to increase, not reduce, its postal volume until the agency “abruptly walked away” from talks in December. Amazon said it had spent more than $5 billion annually with USPS in recent years and needed a new long-term contract well before the old one expires on September 30, 2026, because delivery capacity cannot be rebuilt overnight. That is the real story here. Amazon was not trying to prove it could fire the Postal Service. It was trying to force certainty into a network that hates uncertainty (aboutamazon.com, distributionstrategy.com). That helps explain why Amazon’s rural expansion did not end in a postal divorce. In April 2025, the company announced a plan to spend more than $4 billion by the end of 2026 to triple its rural delivery network, expand to more than 200 delivery stations, and add capacity for more than 1 billion extra packages a year across 13,000 ZIP codes. The obvious reading was that Amazon was preparing to replace legacy carriers in the countryside, where last-mile delivery is slow and expensive. The new USPS agreement shows the limit of that reading. Amazon can build fast. It still cannot cheaply reproduce the Postal Service’s address-by-address reach across remote America, so it is keeping the public network in place while extending its own (aboutamazon.com, cnbc.com). That is why reports of a parallel arrangement with UPS make sense, even if the details remain thin. The point is not that Amazon picked a winner. The point is that it did not pick one at all. It is still assembling a hybrid system: its own network where density makes that economical, USPS where universal service is hard to beat, and outside carriers where they can plug gaps in rural coverage without forcing Amazon to overbuild. For all the talk of Amazon becoming its own national parcel carrier, the company just showed the opposite. In the hardest places to deliver, scale is not enough. You still need the old routes, the old handoffs, and the institutions that already know every road (reuters.com, news.futunn.com).