Everyone read the loss. USPS reported a net loss of $2.5 billion for the quarter ended June 30. It led every story.
The number that actually bills you was three lines further down.
USPS revenue per package rose 11.5 percent in the quarter while package volume fell 3.4 percent. Shipping and Packages revenue reached $8.250 billion on 1.554 billion pieces, against $7.662 billion on 1.609 billion a year earlier. That is $5.31 a package against $4.76.
Fewer packages. Considerably more money per package.
The loss is a mail problem. The bill lands on parcels.
In March the Postal Regulatory Commission limited market-dominant mail to one price increase per year, and held that limit in place through September 30, 2030. In the same decision the Commission said USPS’s problems “cannot be resolved by using pricing authority alone.” It estimated the eliminated second annual increase was worth at most $700 million a year, which it characterized as roughly one to three days of USPS operating costs.
Competitive products carry no equivalent frequency cap. Packages can be repriced as often as the Governors approve and the filing clears.
So one side of the house has a regulated cadence through 2030 and the other does not. Whatever the intent, the effect is that the side without a cadence limit carries the recovery.
The number that tells you it is a floor
Break the quarter apart by product and the interesting thing is not the average. It is the spread.
Ground Advantage revenue rose 14.6 percent to $4.645 billion on 785 million pieces, and volume grew 2.3 percent. Yield per piece went from $5.28 to $5.92, up 12.0 percent.
Priority Mail revenue fell 7.4 percent on 101 million pieces, and volume fell 20.5 percent. Yield per piece went from $11.48 to $13.37, up 16.4 percent.
Both products got materially more expensive in the same quarter under the same surcharge. One lost a fifth of its volume. The other grew.
Where that Priority volume went is worth being careful about. Priority lost 26 million pieces. Ground Advantage gained 18 million. Those are close enough that a meaningful share of the decline may have moved down into Ground Advantage rather than off USPS entirely, and the filing does not separate the two.
Both readings land in the same place. Priority Mail has somewhere cheaper to go, whether that somewhere is a competitor or the product one tier down. Ground Advantage does not. When the price moved, volume left the product with an alternative and stayed on the one without.
That is what a floor looks like when you measure it rather than assert it. And Ground Advantage is not a corner of the business. It is 50.5 percent of USPS package volume and 56.3 percent of package revenue. It is the reference price for lightweight residential in this country.
One caution on the volume figures. The Amazon volume reduction announced in April sits somewhere inside this quarter, and USPS does not break out where. Read the elasticity as directional rather than clean.
Two different floors, and one causes the other
I have written before about the floor at the bottom of last-mile economics: roughly two minutes per stop that route density does not compress, which constrains every carrier chasing margin to raising rates, cutting driver pay, or shifting cost to the consignee.
USPS sits outside that constraint. The stop is already scheduled, the wage is already funded by the mail product, and the mailbox is already there. Adding a package to a route that mail already pays for is a marginal cost question, not a stop cost question.
That cost position is why Ground Advantage can be the price floor. It is not why the floor is moving.
The floor is moving because of a $2.5 billion quarterly loss on the mail side and a regulatory cadence limit that runs to 2030. Those are different drivers, and holding them apart is the whole analysis. USPS holds a structural cost advantage in lightweight residential and is repricing anyway, because the pressure is not coming from the cost of carrying the package.
What sits above a floor that moves
When the cheapest published option in a market gets 12 percent more expensive, everything priced against it gains headroom. Not through coordination. Through the ordinary mechanics of competitive positioning: a product that had to stay under Ground Advantage to win lightweight residential has more room to work with than it did a year ago.
You can already see it in published schedules. UPS’s 2026 general rate increase carried a 5.9 percent headline, but 1 to 5 pound parcels and Zone 7 and beyond ran well above that line, with several surcharges up 11 to 14 percent.
That is the part worth sitting with. Lightweight residential repriced across the market, not at one carrier.
What the year actually did to a published rate
Three dated changes landed on Ground Advantage Commercial in twelve months: 7.8 percent in January, an 8 percent temporary surcharge from April 26 running through January 17, and an 11.8 percent average increase on July 12.
Compounded, those three published averages come to roughly 30 percent across one calendar year. Treat that as what the published schedule did, not as what any particular shipper paid. Your realized number depends on mix, weight, and zone.
Two mechanics inside the July change did more damage to lightweight than the average suggests. The sub-pound ounce tiers were collapsed into the 12 to 15.99 ounce rate, so a 4 ounce parcel takes between $1.43 and $2.04 more depending on zone. And the dimensional divisor moved from 166 to 139, which is about 19 percent more billable dimensional weight on any dim-rated package, before the new rule rounding fractional dimensions up to the whole inch.
The line most shippers missed
USPS stated that the July restructure “will not impact customers that have negotiated commercial rates.”
Published Commercial and a negotiated agreement diverged by the full weight of that change on the same day. If you are shipping sub-pound volume on published rates, the sub-pound collapse alone prices the gap for you.
Whether the April surcharge reaches negotiated rates is a genuinely open question. No source states it either way, and it is a better question for your account team than an assumption for either of us.
And no, this is not a failing carrier
Package Services on-time performance ran 97.2 percent in the first fiscal quarter and 95.9 percent in the second, against 95.7 and 92.0 a year earlier. Third quarter performance is not published yet, and USPS does not break service out by individual competitive product.
Service improved while price rose hard. That is a real trade, and it is not the story most coverage told.
What you can still control
Benchmark your other carriers against published Ground Advantage. This applies whether or not you ship a single USPS package. The gap between what you pay for lightweight residential and what the cheapest published option costs is the distance your carrier can move before switching becomes rational for you. That gap widened this year, because the floor rose faster than most negotiated rates did. Nothing on your invoice reports it, and it is the number that decides what the next negotiation can take.
Re-baseline on invoice data, not the published increase. The headline percentages understate effective change badly once the surcharge, the sub-pound collapse, and the divisor interact.
Price the published-versus-negotiated gap. July made that gap explicit rather than theoretical.
Re-run packaging ROI. A divisor move from 166 to 139 plus fractional round-up changes the return on right-sizing enough to invalidate a 2025 analysis.
Check that your concessions are still in force. Effective periods lapse quietly and the tariff underneath them does not.
Budget for the window, not the year. The mail cadence limit runs to September 2030. That is four more years in which packages remain the flexible side of the house.
The bottom line
The cheapest published option for a lightweight residential parcel in America absorbed a 12 percent yield increase in one quarter and grew volume anyway. That is not a USPS story. That is the floor under every residential rate in the market moving up, with four years of runway left on the structure that is pushing it.
Takeaway: When the floor rises, nothing above it has to be renegotiated for your cost to go up. Check the floor, not just your contract.
Did your 2026 budget carry a USPS number that looked like the published increase, or one that looked like your invoices?
