# Amazon Squeezing Sellers: Two Federal Lawsuits and a Decade of Fees

**Author:** Robert Parr
**Date:** 2026-09-06
**Description:** Amazon faces two federal cases, a 2023 antitrust trial and a 2026 suit alleging rigged ad auctions. What a decade of rising fees means for sellers.
**URL:** https://thrive3pl.com/blog/amazon-squeezing-sellers-federal-lawsuits-decade-of-fees

---

> **TL;DR:** The FTC and a large number of US states have multiple suits pending against Amazon. A 2023 antitrust case covers punishing sellers who discount elsewhere and conditioning the Prime badge on Fulfillment by Amazon, and its bench trial starts February 9, 2027. On August 31, 2026, the FTC and 22 states filed a second case alleging that Amazon sold advertisers a second-price auction while charging them their own full bid close to 80% of the time. Between the filings Amazon raised fees, delayed payouts, and priced relief from its own new inbound charges through AWD.

On August 31, 2026, the Federal Trade Commission and 22 state attorneys general sued Amazon for deceiving the businesses that advertise on its store. Amazon told advertisers that a competitive auction set their price. According to the complaint, Amazon had quietly inserted a charge of its own into that auction and collected it for seven years. New York's attorney general calls that fraudulently overcharging Amazon's advertising customers, and the complaint counts more than a million of them, including over 500,000 small and mid-size businesses. Per Marketplace Pulse, third-party sellers account for 69% of the dollar value of everything sold on Amazon, so the conduct alleged here ran against the merchants who supply better than two thirds of the store.

That filing is the second major federal case against Amazon. A major antitrust suit brought in 2023 over how the company treats the sellers on its platform is still pending, and its trial begins in February 2027.

The 2023 case turns on fulfillment. In 2019, Amazon executives held a meeting to decide the future of Seller Fulfilled Prime (SFP), a program that let a brand earn the Prime badge while shipping from its own warehouse or a third-party logistics provider. According to the FTC's amended complaint in that antitrust case, the meeting was titled "3PL impact mitigation." Amazon shut down new enrollment in SFP a few months later, which only reinforced their apparent desire to limit fulfillment competition from other providers.

These two complaints charge separate conduct under separate statutes, but the underlying concerns appear to be the same. Amazon sets the fees, the fees rise every year, Amazon withholds some of the data required to verify those charges were fair, and the sellers have nowhere to go to claw back their shrinking margins.

My first reaction to the 2023 complaint was that it read thin. At the time, it seemed more like the FTC just didn't understand an innovative business model. Three years on, though, it seems that Amazon has spent the intervening time flaunting their market power and supplying an increasing amount of evidence that supports the claims against them.

## The August 2026 Amazon Ad Auction Lawsuit

Digital advertising markets rely heavily on trust. Sellers can't observe PPC ad auctions in real time, so they find it difficult to challenge the costs they pay for each click. A seller sets a bid, the auction settles, and the price that comes back looks like the market's answer. According to the complaint, part of that price came from a bidder Amazon fabricated.

*FTC v. Amazon.com, Inc.*, No. 2:26-cv-03097, filed in the Western District of Washington, is a consumer protection case under the FTC Act rather than an antitrust case. The attorneys general of 22 states joined as plaintiffs.

For years, Amazon told advertisers it ran a generalized second-price auction. This is the accepted standard in digital advertising, in which the winning bidder pays one cent more than the next-highest bid. That format changes how a rational advertiser bids. A seller who knows they will pay only enough to beat the runner-up bids close to the true value of the placement, while a seller who expects to pay their full bid shades that number down to protect against overpaying.

The [complaint](https://www.ftc.gov/news-events/news/press-releases/2026/08/ftc-states-sue-amazon-over-secret-ad-surcharge-scheme) alleges that beginning in 2019, Amazon added an undisclosed surcharge to that auction and referred to it internally as a "soft reserve price." The executive in charge of Amazon Ads explained internally that the price an advertiser pays "isn't set by an actual bidder" but is instead a "proxy 2nd price that we calculate." Another internal document describes an "invented auction participant." The FTC characterizes those bids as shill bids.

According to the FTC, this fake bid compounded over the years, and the numbers provided in the filing appear to back them up. Sponsored Products advertisers paid their own full bid between 30% and 40% of the time in 2021, 70% of the time in 2022, and roughly 80% of the time by 2024. The filing also alleges that Amazon raised the surcharge for Prime Day and Black Friday, the days a seller has the least room to pull back.

Amazon's own documents, as quoted in the filing, explain the secrecy. Disclosure would cause "irrevocable damage to advertiser trust" and send bids into a "downward spiral." Notes from a 2024 discussion among senior executives, including the head of Amazon Ads, describe the practice as "a clever non-transparent way to charge first price" and "an incredibly effective way to drive revenue." Amazon has called the suit misguided and says the FTC cherry-picked its documents. Notably, though, its public response did not dispute how the auction works.

No seller could have found this independently. Amazon sells the placement, runs the auction, and reports the result, so a brand checking its advertising spend is measuring Amazon's numbers against Amazon's own reporting of them. Advertising already runs up to 15% of revenue for a typical seller, and the FTC now alleges that a hidden surcharge sat inside that number for seven years, amounting to tens of billions of dollars in overcharges.

## The 2023 Antitrust Case

An earlier case, No. 2:23-cv-01495-JHC, is before Judge John Chun in the same court. The Federal Trade Commission, 18 states, and Puerto Rico brought it. Their fullest public filing, the amended complaint of March 14, 2024, runs 179 pages. The plaintiffs filed a second amended complaint on October 31, 2024, after Judge Chun let them replead some of the state claims he had previously dismissed. Across both, the case focuses on two things Amazon does.

**Anti-discounting.** Amazon allegedly punishes sellers who offer a lower price on their own website or a rival marketplace, most often by removing the seller from the Buy Box. The complaint states that Amazon internally recognizes that losing the Buy Box drops a seller's sales to nearly zero. The FTC argues that this creates an artificial price floor across the internet, so Amazon's fee load raises prices everywhere rather than only on Amazon.

Amazon takes a larger share of each sale than the marketplaces it competes with. A brand pricing to one margin would therefore charge more on Amazon and less on other platforms where the fees are lower, which is how a cheaper venue would take customers from an expensive one. Punishing the lower price off Amazon ends that. The brand lifts its prices everywhere instead, a rival marketplace's lower fees stop reaching shoppers as lower prices, and the one structural advantage it had over Amazon disappears. Amazon still has two-day delivery and a Prime membership the shopper has already paid for, so the shopper loses the main reason to purchase anywhere else.

**Fulfillment coercion.** Amazon allegedly conditions Prime eligibility on the purchase of FBA, which prevents sellers from using independent fulfillment providers and raises costs for anyone trying to compete with Amazon's logistics network. Fulfillment is the largest controllable cost a brand carries after the goods themselves, and the tie removes its ability to shop that cost. A seller who could fulfill more cheaply, or better, still has to buy Amazon's service to keep the badge that makes the listing sell.

Those two practices carry the entire federal case. The government charges them as illegal monopoly maintenance, once for the FTC and again on behalf of the states, and adds a separate claim about "Project Nessie." Nessie was an algorithm that predicted which products rival retailers would follow Amazon upward on, because their own pricing software tracked Amazon's. Amazon raised the price on those products, kept the increase when the rivals matched it, dropped back when they did not, and paused the system during Prime Day and the holidays when shoppers and reporters were paying attention. The complaint puts the total taken from American households above $1 billion.

Amazon asked the court to throw the case out. Judge Chun refused on nearly all of it in September 2024, and the federal claims are still standing.

## Amazon's Own Data on Third-Party Fulfillment

The antitrust complaint reports internal Amazon analysis of how SFP merchants performed, broken out by who did the shipping. In the last quarter before Amazon suspended enrollment, sellers using independent fulfillment providers met Amazon's delivery requirement 98.4% of the time, where the average across all SFP sellers was 96%. On the shipping requirement the same group scored 99.8% against a program average of 96.8%.

Amazon's own numbers showed that third-party logistics providers outperformed sellers shipping for themselves and outperformed the program average. The FTC uses those figures to argue that tying Prime to FBA never protected the customer promise. Some Amazon employees argued to keep the program, warning that shutting it would cost Prime customers faster delivery on 115 million items, and the complaint says a senior executive vetoed the idea.

Demand for an alternative showed up immediately. Six weeks after Amazon closed enrollment, more than 8,000 sellers had joined the waitlist.

## What the Antitrust Complaint Said About Advertising

The 2023 filing reached advertising as well, on a different theory: not the price of an ad, but how many of them a shopper has to scroll past. The complaint quotes Amazon executives acknowledging that the advertising load creates "harm to consumers" by making it "almost impossible for high quality, helpful organic content to win over barely relevant sponsored content." It quotes Jeff Bezos instructing his team to "accept more defects," the internal term for irrelevant ads, because the advertising revenue exceeded the cost of the worse experience.

The complaint also states that Amazon takes close to half of every dollar from a typical seller using its fulfillment service, and it quotes Amazon's own recognition that sellers find "it has become more difficult over time to be profitable on Amazon" because of "increasing fees and costs." One seller, quoted in the filing, put it plainly: "we have nowhere else to go and Amazon knows it."

Read the two complaints together and the picture sharpens. The 2023 case says Amazon degraded organic search results to sell more advertising. The 2026 case says Amazon then charged more than the auction produced for every one of those placements.

## Where the Antitrust Case Got Stronger Since 2023

**The fee trajectory continued through the litigation.** Marketplace Pulse now puts Amazon's total take at roughly 50% of a typical seller's revenue, up from about 40% five years ago, composed of a referral fee near 15%, FBA fulfillment and storage between 20% and 35%, and advertising up to 15%. Amazon raised those fees while defending a federal complaint that named them.

**Amazon raised the cost of getting into FBA, then sold the way around it.** An inbound placement service fee took effect on March 1, 2024, averaging $0.27 per unit on standard-size products and $1.58 on large bulky ones, charged when a seller ships to one location instead of spreading a shipment across the network. Across the accounts we watched, the least-expensive month-over-month result was a 40% increase in inbound shipping cost. Some doubled.

A low-inventory-level fee followed on April 1, charging up to $1.11 a unit whenever a standard-size product drops below 28 days of cover. The stated way to avoid that one is to send more units into FBA and keep them there. Amazon's remedy for holding too little inventory at Amazon is to hold more inventory at Amazon.

There is one other way out of both. Inventory routed through Amazon Warehousing and Distribution (AWD), the company's own upstream storage service, pays no inbound placement fee, because AWD pricing absorbs it. Amazon then discounts Amazon Global Logistics (AGL), its freight forwarding arm, by up to 25% on cross-border shipments bound for AWD, and discounts domestic trucking into AWD by up to 25% through its Partnered Carrier Program. It amounts to "pay us, or pay us, but you are going to pay us."

Neither complaint addresses the AWD and AGL issues directly, but the pattern is clear. Amazon invents a new fee, then offers to waive it, provided the seller uses Amazon's other services in adjacent industries. They aren't competing on level terms with independent warehouses, but rather are leveraging their dominance in one industry (online marketplaces) to gain an unfair advantage in other ones (freight forwarding and fulfillment warehousing). Setting a fee and then selling relief from it is the structure of a protection racket, not a business competing solely on its merits. It is the shape of the fulfillment count all the same.

**Payment timing became a new lever.** In spring 2026, Amazon changed when sellers get paid, moving from payment tied to the purchase to payment at least seven days after the customer takes delivery. Sellers have reported holds running considerably longer. On Amazon's own forums they note that Amazon confirms delivery faster for FBA shipments than for merchant-fulfilled ones, so the policy rewards sellers who move fulfillment into FBA. The complaint alleges that Amazon coerces sellers into its fulfillment service, and this policy produces that result through the payment system rather than the Prime badge.

**Advertising became harder to finance.** In April 2026, Amazon began taking pay-per-click spend directly out of sales proceeds rather than allowing sellers to pay by credit card. Sellers lost the float and the card rewards at the same moment the payment hold delayed the underlying revenue. A sellers' group organized a one-day advertising boycott. In response, Amazon merely delayed the rollout before proceeding with it anyway.

**Sellers started raising prices.** The Million Dollar Sellers group polled roughly 200 Amazon sellers this summer and found 60% had raised prices in response to Amazon's fees and policy changes. The FTC's central consumer-harm theory holds that Amazon's seller costs reach shoppers as higher prices. A survey of 200 sellers will not be enough to carry that theory at trial, and the plaintiffs will need economic evidence, but the direction favors the government.

**The government filed again.** The August 2026 advertising complaint adds no count to the antitrust case and will be tried separately under a different statute, so Judge Chun will not rule on the ad auction. It still changes the backdrop. The antitrust case argues that Amazon's fee load reaches consumers as higher prices, and Chairman Andrew Ferguson said of the advertising surcharges that "these higher costs were largely passed on to American consumers." The FTC's account of how Amazon treats the businesses on its platform now rests on two filings instead of one.

## Where the Antitrust Case Got Weaker

There are some areas where the antitrust claims may have lost ground. Amazon surely will lean into these in their defense.

**Amazon reopened Seller Fulfilled Prime.** In October 2023, one month after the FTC filed, Amazon reopened enrollment to new sellers. Closing the program was the most concrete act in the fulfillment count, and it no longer describes the present. Amazon will argue that a company shutting out rivals does not reopen the door. The government will answer that Amazon tightened performance and enrollment requirements in 2025 and raised delivery-speed thresholds again in July 2026, so the reopened program admits far fewer sellers than the original. Judge Chun will weigh both accounts.

**Project Nessie has been dormant since 2019.** Amazon says it built the tool to stop its own price matching from driving prices to unsustainable lows, that it did not work as intended, and that the company scrapped it. The complaint acknowledges the pause and pleads that Amazon "retains the ability" to switch it back on. The claim therefore asks a court to act against a system that has not run in seven years, which is a harder case to make than one about conduct happening today.

**The market definitions likely are too narrow.** The FTC built its argument on an "online superstore" market and an "online marketplace services" market. Critics across the antitrust bar have called these definitions gerrymandered, drawn narrowly enough to make Amazon look like a monopolist by excluding the retailers and platforms competing with it every day. Walmart's online business keeps growing, and eBay quietly remains a major marketplace in the US. TikTok Shop has arrived, and Temu and Shein reshaped the low-price end of the category. Judge Chun allowed those definitions through at the earliest stage of the case, which is a low bar. Proving them at trial requires economic evidence the FTC has not yet shown publicly.

**Consumer harm remains genuinely hard to show.** Amazon's defense is straightforward and not unreasonable: prices are low, selection is enormous, delivery keeps getting faster. Ryder's 2026 consumer study found expectations for one-to-two-day delivery at a four-year high, a standard Amazon largely created. A court can find Amazon's conduct toward sellers indefensible and still decline to call it illegal under a consumer-welfare standard.

My read is that the anti-discounting claim is the strongest of the antitrust group; this should be a clear win by the government. The fulfillment claim lost ground when Amazon reopened SFP, but was strengthened by the introduction of fees and policies designed to penalize sellers unless they use all of Amazon's logistics stack. Project Nessie is the weakest claim in the complaint, and may be found moot by the court. The advertising case is a different animal, because deception cases turn on what a company said and what it knew, and the FTC has quoted Amazon's executives describing the surcharge in their own words.

## What This Means for Your Business

The antitrust trial begins February 9, 2027, and the court will decide whether Amazon broke the law before it decides what to do about it, so even a government win means a second round before anything changes. The advertising case was filed last week and has no trial date. Any structural remedy takes years and then survives appeal or does not. No brand should build a plan around either timeline.

The pressure on sellers runs independently of both verdicts. Amazon sets the fees, sets the payment timing, and — according to the FTC — sets the advertising price with a bidder that does not exist. Each of those moved against sellers while the company was under active federal litigation, which tells you what to expect when it is not.

Amazon's market power is real, and a brand that abandons the channel gives up the largest pool of shopping demand in the country. Selling on Amazon remains a reasonable decision. Depending on Amazon for your margin and your enterprise value has stopped working.

Paying half your revenue to your largest channel rarely kills a company in a single quarter. Your margin runs thinner each year until a soft season flips you upside-down. I have watched this happen to many good Amazon-only brands with real demand behind them. Ask yourself this: if Amazon went dark tomorrow, would the rest of your business still clear its own costs?

You can answer that in an afternoon with your P&L, and the answer tends to reorganize the following year's priorities. If it comes back no, the work ahead is building channels that stand on their own, which means a direct storefront, a wholesale or retail motion, and fulfillment that Amazon does not control.

We covered the economics of that shift in [Amazon Isn't the Easy Button Anymore](/blog/amazon-isnt-the-easy-button-anymore), and the case for splitting inventory between FBA and a 3PL in [Amazon Storage Fees vs 3PL Storage Costs](/blog/amazon-storage-fees-vs-3pl-storage-costs-2026). The courts may eventually change how Amazon behaves. Your fulfillment strategy is the part you can change this quarter.

---

*Published by Thrive 3PL — Houston-based fulfillment for e-commerce brands. Learn more at [thrive3pl.com](https://thrive3pl.com).*