This case is not about whether Amazon broke antitrust law — it is about how a platform that controls the largest online marketplace shaped prices across the entire retail internet without ever directly agreeing with competitors. The internal emails, presentations, and employee interviews that form the backbone of California’s price-fixing lawsuit depict a system in which Amazon used its gatekeeper position as leverage to coerce suppliers into raising prices on other retailers. That supplier-coercion channel, the evidence shows, was complemented by a separate algorithmic mechanism that automatically matched competitors’ price increases, creating a two-sided pressure system that neither mechanism alone would fully explain.

How supplier coercion worked

A mapping of the relationships among the parties reveals a hub-and-spoke topology in which Amazon sits at the center, exerting pressure outward on suppliers, who then transmit that pressure to competing retailers. The suppliers occupy a middle tier: dependent on Amazon for marketplace access, they become instruments for raising prices on platforms that compete with Amazon directly.

In one of the three documented instances, an electric ice-cream maker manufactured by Maxi-Matic was listed at $17.99 on both Amazon and Best Buy. According to the California attorney general’s filings, Amazon temporarily removed Maxi-Matic’s inventory from its platform. Shortly after, Maxi-Matic pulled the product from Best Buy, making it unavailable there. By the end of the same day, Amazon restored the listing and increased the price to $59.99 — a 233 percent increase.

A leather table lamp from the supplier All The Rages was selling at Walmart for $24.99. After Amazon contacted the supplier, an employee, Joseph Martin, emailed Amazon saying his company had reached out to unnamed retailers “to fix the retail.” The lamp’s price at Walmart rose to $39. Martin told the Guardian the change reflected “the correction of a pricing error” and said his company believes the attorney general’s claims “are without merit.”

Ryan Turano, chief technology officer of the fertilizer company AgroThrive, told the Guardian that an Amazon vendor manager would list specific retailers — including Home Depot — for him to contact. In one email, Turano informed the manager that he had “just got out of a meeting with the Home Depot manager” who had “agreed to raise the prices this time.” Turano said, “We were at their mercy,” adding that the company lacked the resources to push back.

A former Amazon “customer success” manager said that higher-ups instructed employees to have certain conversations with suppliers over the phone to avoid a digital trail. “The only rule was, ‘Do not have this in writing,’” he recalled. An in-house Amazon presentation from 2019 instructed staff to send screenshots of rivals’ product listings to suppliers when making compensation requests but to avoid mentioning the companies’ names. A 2022 presentation reminded staff to “not use email” for certain conversations.

California Attorney General Rob Bonta said in an interview that such euphemisms would not be an adequate defense. “Whatever euphemism it is, whether it be ‘update pricing’ or ‘resolve an issue’ or ‘address a concern’ in the market, all that means is: ‘Fix prices. Get your lower price up to Amazon’s higher price or suffer the consequences,’ and that’s illegal.”

Amazon has characterized the California AG’s case as a distortion of a handful of emails from nearly one million vendor communications — a defense that the coded-language presentations, the direct supplier testimony from Turano, and the documented price changes appear to undercut. The company has also said the practices are “common in the retail industry and part of the pro-competitive give-and-take between Amazon and its vendors.”

The algorithmic parallel

The supplier-coercion channel is not the only route the evidence shows. The Federal Trade Commission’s separate lawsuit adds a second pathway: an algorithm that, the FTC alleges, “rapidly copied rivals’ price changes to the penny.” According to the complaint, the algorithm was designed to teach competitors that they could not compete with Amazon by offering lower prices, so they might as well raise prices — increases that Amazon’s algorithm would then follow. The FTC case is scheduled for trial on March 29, 2027.

Amazon did not directly respond to questions about the algorithm but pointed to academic literature noting how common it is for retailers to track prices.

The convergence of the two pathways is the most analytically consequential structural feature of the evidence. The supplier-coercion network and the algorithmic price-following mechanism both push in the same direction: higher prices across retail platforms. The implication is that the competitive-harm outcome does not depend on any single mechanism. If the supplier-pressure network were dismantled through a legal injunction, the algorithmic pathway would preserve the price-elevation effect. Enforcers cannot address the harm by removing one channel alone; they must target both. The legal question is not simply whether each mechanism individually constitutes illegal conduct, but whether their combined effect produces a market outcome that would not be corrected by removing either one.

Who stands where — a stakeholder analysis

The legal landscape positions four sets of active enforcement actions against Amazon: the California attorney general’s lawsuit, the FTC’s parallel case, a suit by Washington DC’s attorney general, and a private class action in federal court in Seattle. The California attorney general has said his case is set for January 2027; the FTC’s is set for March 29.

A stakeholder classification places California AG Bonta and the FTC as definitive stakeholders — high power, high legitimacy, and high urgency given the approaching trial dates. Amazon is also definitive: it controls the marketplace that suppliers depend on and faces acute legal exposure across four fronts.

The suppliers caught in the middle are high in legitimacy and urgency but low in individual power — a dependent stakeholder group. Their structural dependency on Amazon is precisely what enables the coercion scheme, and their fragmented responses — Turano cooperating with investigators, Martin defending the price change as an error correction — prevent them from forming a collective counterweight. The legal system’s reliance on these same dependent suppliers as witnesses creates evidentiary fragility: Amazon’s defense can exploit the inconsistency between Turano’s “we were at their mercy” and Martin’s “without merit.” The dependent position that makes them coercion targets also makes them unreliable witnesses.

The competing retailers — Best Buy, Home Depot, Walmart, Target — are affected parties with a conflicted position. In the short term, price increases pass through to their consumers and may improve their margins. Over the longer term, Amazon’s ability to dictate pricing erodes their pricing autonomy. The evidence does not include any public statement from these retailers on whether they view Amazon’s coercion as a net long-term competitive threat, leaving their classification ambiguous between discretionary stakeholders and dormant ones. If retailers privately benefit from the price increases, they may not cooperate with enforcers — a strategic ambiguity that could be determinative for the cases’ outcomes. Amazon Prime members face a particular bind: benefiting from the ecosystem’s convenience while paying higher prices on competing platforms that were allegedly pressured to match Amazon’s rates.

Consumers are the end harmed party — they pay the inflated prices — but they have no procedural standing in any of the four lawsuits aside from the class-action device in Seattle. The litigation map reproduces existing power asymmetries: the parties who pay inflated prices have no procedural voice in the proceedings that will determine whether the practice continues. Consumer interests are represented only through government enforcers, making them a dependent stakeholder whose voice depends on the litigation’s success. No consumer advocacy groups or labor unions appear in the record as amici or intervenors. This structural absence of the end harmed party is the most striking feature of the stakeholder landscape.

What to watch as the cases advance

Three questions will determine how this story unfolds. The most consequential is whether the courts will treat the supplier-coercion evidence and the algorithmic allegations as separate theories of harm or consider their combined effect — because only the latter framing would allow a remedy that addresses both pathways. The second question is the communications-concealment policy that instructed staff to avoid email and to omit competitor names from screenshots: it cuts two ways, suggesting awareness of legal boundaries while also generating self-incriminating evidence that survives discovery. The third is the suppliers’ dual position as victims and instruments — whether cooperating witnesses such as Turano will be treated as whistleblowers or as participants in a price-fixing conspiracy. The answer in each case will ripple through the other lawsuits.

Analytical techniques used in this piece

This analysis applies the methods below. Each links to a short, plain-English explainer you can read and reuse.

Red-Team Assessment
Models a capable adversary probing a plan for the seams they would exploit.
Relationship Mapping
Extracts the network of ties among people, institutions, and entities.
Stakeholder Mapping
Charts the parties to a situation — their interests, power, and alignments.