Amazon Charged School Districts Up to Three Times More for the Same Supplies on the Same Day
What happened
When a Colorado city paid $8.99 for a 12-pack of Sharpie markers and a nearby school district paid $28.63 for the exact same product on the exact same day, no negotiation drove the gap, no logistics justified it, and no human at Amazon consciously decided one public buyer deserved to pay three times more. An algorithm made that call, automatically, drawing on internal data about inventory, demand, and estimated willingness to pay. The school district buying at the inflated price had no visibility into the lower price existing at all.
The Institute for Local Self-Reliance analyzed tens of thousands of purchases by 128 U.S. school districts and local governments on Amazon Business and found this pattern repeated at scale. Across roughly 2,500 commonly ordered items, including paper, glue, pens, and cleaning supplies, districts collectively spent around $3 million on purchases that would have cost approximately $2.5 million if they had consistently received the lowest price Amazon made available at the time of purchase. That implied overpayment of about 17 percent, with individual transactions sometimes exceeding double the lowest available price. In one documented case, the same district paid more than four times as much for an identical stapler within a few days, buying from different Amazon sellers on the same platform.
The structure that made this possible is not an edge case in how Amazon Business contracts are written; it is the default. Many school districts and local governments moved away from traditional competitive bidding and fixed-price contracts with local suppliers in favor of broad purchasing agreements with Amazon Business. Those agreements rely on dynamic pricing rather than locked rates, meaning the price shown at the moment of purchase reflects Amazon's algorithmic assessment of the transaction rather than any negotiated ceiling. Public procurement rules designed to protect taxpayer money typically assume that price comparison is possible. On a platform where the same item can cost $8.99 and $28.63 simultaneously depending on which account is buying, that assumption breaks down.
Amazon disputed the ILSR findings, calling the report flawed and arguing its prices compare favorably to other major retailers overall. That response does not address the specific question at the center of the analysis: whether public buyers systematically received higher prices than lower prices available on the same platform for the same goods on the same day. Senator Elizabeth Warren wrote to Amazon CEO Andy Jassy in March 2026 pressing the company for answers about its pricing practices with schools and local governments, citing the overcharges directly. Amazon has not provided the transaction-level data that would allow independent verification of its rebuttal.
That gap is where the accountability failure sits. Public procurement requires not just favorable average pricing but a verifiable record showing that each transaction was conducted at a fair and defensible rate. A dynamic pricing system that adjusts in real time based on proprietary signals produces outcomes that cannot be audited after the fact without access to the platform's internal state at the precise moment of each purchase. Without a provable record of what a system did, at what price, and why, there is no mechanism for the public to confirm that its money was spent appropriately, and no basis for regulators or legislators to determine whether the system behaved within acceptable bounds.
Reported impact
- Affected parties
- Not publicly disclosed
- Harm type
- Not publicly disclosed
- Scale
- Not publicly disclosed
- Financial impact
- Not publicly disclosed
- Regulatory action
- Not publicly disclosed
Classification
Relevant governance controls
Governance control mapping is not available for this record.
- No controls mapped
Not publicly disclosed
Control mapping is analytical. It does not state that any control would have prevented the incident.
Sources and evidence
This record was researched and written by the Index. The event is also catalogued in the following database, which is listed for cross-reference.