A California state judge has refused to order Amazon to stop the retail price-fixing conduct that the state attorney general has spent nearly four years alleging, leaving the case intact but stripping California of the one remedy that would have changed anything before trial.
Judge Ethan Schulman of San Francisco Superior Court denied Attorney General Rob Bonta’s request for preliminary injunctive relief, finding that the state had not shown an injunction was necessary to prevent impending injury, according to reporting by MLex and Law360. Reuters reported that Schulman signalled the denial at a hearing on Thursday, August 27, telling the parties he had tentatively ruled against the motion.
The underlying case survives. Trial is scheduled for January 19, 2027.
In short
- What happened: A San Francisco Superior Court judge denied California’s bid for a preliminary injunction against Amazon in the state’s 2022 price-fixing suit.
- Why it was denied: The court found the state failed to show impending injury, and questioned why a late-stage motion targeted conduct not shown to be ongoing.
- What survives: The Cartwright Act and Unfair Competition Law claims are untouched, and Amazon already lost a summary judgment bid in April 2026.
- What is at stake: California alleges Amazon pushed vendors including Levi Strauss, Hanes and Allergan to raise prices at Walmart, Target, Chewy and others.
- What comes next: A January 19, 2027 trial in San Francisco, running ahead of the separate FTC monopolization case in Washington state.
What the San Francisco court actually decided
The ruling is procedural, not merits-based, and the distinction matters. A preliminary injunction is emergency relief. It asks a court to change a defendant’s behavior before any trial has established liability, and it is granted only when the moving party can show that waiting for judgment would cause harm the court cannot later undo.
According to MLex, Schulman’s order concluded that the California Department of Justice failed to make that showing. The state did not persuade the court that an injunction was needed to prevent impending injury, which is the threshold question. Everything else in the motion, including the strength of the underlying evidence, becomes secondary once that threshold is missed.
Law360 reported that the judge criticised the timing as a “last-minute request” aimed at conduct the state had not demonstrated was still happening. That framing is important for anyone reading the headline as a verdict on the case. It is not.
Bonta’s office signalled it will keep going. Reuters quoted a spokesperson calling the ruling disappointing while adding: “We continue to believe that Amazon’s ongoing price fixing behavior is harming California consumers.” The same spokesperson said the office looks forward to trial.
What a preliminary injunction would have required
Had the motion succeeded, the practical effect would have been immediate. California asked the court to bar Amazon from fixing prices, from communicating with vendors about competitors’ pricing, and from applying the coercive vendor practices described in the complaint.
Each of those is an operational constraint on how Amazon’s retail organisation negotiates with first-party suppliers. An injunction would have reshaped those conversations from the day it issued, roughly 17 months before the case reaches a courtroom.
Instead, nothing changes operationally. Amazon continues to run its pricing and vendor programs as it does today, subject only to the ordinary risk that a January 2027 judgment goes against it.
Why the timing question hurt the state
California filed the injunction motion in February 2026, more than three years after the original complaint. That gap was the crux of Amazon’s opposition and, by the reported account of the hearing, a live concern for the judge.
Reuters reported that Schulman expressed skepticism toward the request, noting that the state relied on evidence from 2023. Separately, the tranche of material Bonta’s office unsealed in April 2026 was described in coverage by Forbes as spanning roughly 2019–2021.
Either way, the evidentiary record placed in front of the court on this motion was several years old. Courts are reluctant to treat historical conduct as an emergency, and that reluctance appears to have decided the motion.
What California says Amazon did
The substance of the case has not narrowed. Bonta sued Amazon on September 15, 2022, alleging that the company suppresses price competition across online retail rather than winning on price, and that consumers pay more as a result both on and off Amazon.
The claims are brought under California’s Cartwright Act, the state’s principal antitrust statute, and its Unfair Competition Law. Both are broader in some respects than federal antitrust doctrine, which is one reason the state case is proceeding on its own track.
The alleged mechanism runs through Amazon’s leverage over first-party vendors rather than through direct agreements with rival retailers. That structure is what makes the case unusual, and legally contested.
The three alleged schemes
Bonta’s office has described the conduct as falling into three patterns. In the first, Amazon and a competing retailer effectively land on higher retail prices, each able to match the other’s increase. In the second, a competitor raises prices at Amazon’s prompting through a shared vendor, after which Amazon matches and raises its own price.
In the third, a vendor simply removes the lower-priced listing from a competing site, which eliminates the comparison entirely. All three routes, the state argues, converge on the same outcome: the cheapest visible price online rises.
The alleged enforcement tool is Amazon’s platform placement. The complaint describes threats to suppress listings, restrict advertising, impose financial demands, or remove products for vendors that did not cooperate.
The named brands and the named rivals
The April 2026 unsealing put specific companies on the record. Named vendors included Levi Strauss, Hanes, Allergan, Newell Brands, SkullCandy, Westinghouse, GlobalOne, Agrothrive, BabyVision, Maxi-Matic and Songmic. Named rival retailers included Walmart, Target, Chewy, Best Buy, Home Depot and Wayfair.
The examples in the state’s materials are granular. Per the attorney general’s office, Levi’s coordinated with Walmart on khaki trousers priced at $25.47–$26.99, which then moved to $29.99. Allergan eyedrop pricing was described as coordinated at $16.99.
In one exchange cited by the office, Amazon sent Hanes links to Target.com and Walmart.com showing lower prices, and Hanes confirmed it “reached out to Target and Walmart to have the prices increased.” Other cited episodes involve pet treats sold through Chewy, fertilizer pricing at Home Depot, and pressure on Wayfair to stop discounting a trash can.
Internal Amazon language quoted by the state includes offers to “help you hunt the disrupters in the market” and a statement that the company was “very determined to help you hunt disrupters.” Amazon disputes the characterisation, and none of this has been tested at trial.
How this differs from the FTC’s federal case
Two large antitrust actions are running against Amazon at once, and they are frequently conflated. They allege different things, sit in different courts, and will be decided on different timelines.
The California case is about price levels: whether Amazon’s vendor pressure raised the prices consumers pay. The federal case, brought by the Federal Trade Commission with 18 state attorneys general and Puerto Rico, is about market structure: whether Amazon monopolises online marketplace services by degrading the shopping experience and overcharging sellers.
| Dimension | California v. Amazon | FTC v. Amazon |
|---|---|---|
| Filed | September 15, 2022 | September 2023 |
| Court | San Francisco Superior Court | US District Court, Western District of Washington |
| Judge | Ethan Schulman | John Chun |
| Legal basis | Cartwright Act, Unfair Competition Law | Federal monopolization claims |
| Core theory | Price fixing via vendor coercion | Monopoly maintenance in marketplace services |
| Plaintiffs | California attorney general | FTC plus 18 states and Puerto Rico |
| Trial format | State court trial | Bench trial, liability phase first |
| Trial date | January 19, 2027 | Proposed March 29, 2027 (moved from February 9) |
Judge Chun granted Amazon a partial dismissal in the federal case and agreed to try liability first, with remedies deferred. That sequencing means the federal matter is unlikely to produce structural consequences quickly even if the FTC prevails.
The California trial therefore arrives first. It will be the earliest full public airing of the vendor-pressure theory, and its evidentiary record will be read closely by every enforcer working on retail pricing. Our earlier analysis of why US surveillance-pricing rules are emerging from states rather than Washington describes the same dynamic in a different corner of the same enforcement map.
What the ruling changes for marketplace sellers and brands
For third-party sellers, the immediate answer is very little. No court order now constrains how Amazon manages the Featured Offer, formerly and still widely called the Buy Box, or how it treats listings priced higher on Amazon than elsewhere.
The practical consequence is that the pricing discipline sellers already experience continues unchanged into 2027. Sellers who have built operating models around maintaining price parity across channels have no new reason to alter them this quarter.
For first-party vendors, the calculus is subtler. The unsealed record has made vendor-side pricing conversations legally salient in a way they were not before, regardless of how the injunction motion turned out.
Documentation risk is now the live issue
The most quotable material in California’s filings is not Amazon’s, it is the vendors’. Emails in which a brand confirms it contacted Target or Walmart to have prices raised are, on their face, evidence of coordination among competitors mediated by a supplier.
That exposure sits with the brand as much as with the platform. Compliance teams at consumer brands with meaningful Amazon first-party volume have a concrete reason to review how retail pricing requests are recorded and escalated.
None of this depends on who wins in January. The discovery record already exists and is public.
Why parity policies are not automatically unlawful
It is worth stating the counter-position plainly, because the headlines rarely do. A retailer declining to promote a product it can buy more cheaply elsewhere is ordinary commercial behavior, and Amazon has consistently argued that its pricing policies are consumer-protective rather than collusive.
The legal question is narrower: whether Amazon’s conduct crossed from unilateral pricing preference into facilitating agreements among competing retailers. That line is genuinely contested, which is part of why the case has taken four years to reach trial.
The wider pricing-enforcement map in 2026
The Amazon ruling lands in an unusually active year for retail pricing regulation, and the centre of gravity has moved to states and to Europe rather than to federal agencies.
| Measure or action | Jurisdiction | Status in 2026 | What it targets |
|---|---|---|---|
| Fair Price Protection Act | New Jersey | Signed July 23, 2026 | Personalized algorithmic pricing at food retailers and grocery delivery platforms |
| Honest Pricing / all-in pricing rules | California | In force | Mandatory fees hidden from the advertised price |
| DMCCA drip-pricing guidance | United Kingdom (CMA) | Guidance issued November 2025, first cases pending | Drip pricing, hidden fees, false urgency |
| California v. Amazon | California | Injunction denied August 2026, trial January 2027 | Vendor-mediated retail price coordination |
| FTC v. Amazon | United States (federal) | Liability trial proposed for March 2027 | Marketplace monopolisation |
| Digital Services Act marketplace enforcement | European Union | Active, including a EUR 200 million Temu fine in May 2026 | Illegal products, dark patterns, recommender transparency |
Two patterns stand out. The first is that price-presentation rules (fees, drip pricing, all-in display) are advancing faster than price-level theories, because they are far easier to prove.
The second is that algorithmic and personalized pricing is being addressed by statute rather than by litigation. New Jersey’s approach, which we covered when surveillance pricing bans began spreading across states, requires no proof of an agreement at all. It simply prohibits the practice within a defined retail category.
The contrast is instructive. California spent four years assembling a coordination case and could not clear the bar for emergency relief. New Jersey legislated in a single session.
What Amazon has said, and what it has not
Amazon has not issued a detailed public statement on the August order beyond confirming the outcome, based on the coverage available. Its position on the injunction motion, however, is on the record from earlier in the year.
When California filed the motion, an Amazon spokesperson said: “The Attorney General’s motion is a transparent attempt to distract from the weakness of its case, coming more than three years after filing its complaint and based on supposedly ‘new’ evidence it has had for years.”
The same statement added: “Amazon is consistently identified as America’s lowest-priced online retailer, and we’re proud of the low prices customers find when shopping in our store. Amazon looks forward to responding in court at the appropriate time.”
That argument, that the state sat on its evidence and then sought emergency relief, is precisely the one the court appears to have credited. Amazon has not conceded any factual allegation, and the merits remain fully disputed.
The April 2026 loss Amazon does not talk about
The company’s record in this case is mixed, and the August win should be read against an April setback. In April 2026, the same court denied Amazon summary judgment, rejecting its argument that California’s antitrust laws did not reach the conduct at issue.
The court found unresolved factual disputes and held that Amazon had not conclusively demonstrated it avoided entering prohibited pricing agreements with sellers in exchange for visibility. That is what sent the case to trial rather than ending it.
Amazon’s regulatory and operational posture in other markets has not slowed either: the company confirmed a EUR 10 billion European robotics and logistics commitment earlier this year even as antitrust pressure built on both sides of the Atlantic.
The road to the January 19, 2027 trial
With the injunction denied, the case reverts to a conventional pre-trial track. Discovery is substantially complete, which is why the state was able to unseal specific vendor communications in April.
The next meaningful decisions are evidentiary. Motions in limine will determine how much of the unsealed material reaches the trier of fact, and Amazon can be expected to contest the admissibility and the framing of internal communications that read badly in isolation.
There is also the question of what California ultimately asks for. A trial judgment can reach further than a preliminary injunction, including permanent injunctive relief and civil penalties under the Unfair Competition Law.
What the denial does and does not signal
Denial of preliminary relief is a weak predictor of trial outcomes. The standards differ: the injunction motion turned on urgency and irreparable harm, while the trial will turn on whether the alleged agreements existed and whether they restrained trade.
That said, the ruling removes California’s leverage. Settlement pressure in antitrust cases often comes from the prospect of an interim order that constrains the business before judgment. Without one, Amazon has little reason to negotiate before seeing how the evidence performs in court.
The trial date itself is the variable most worth tracking. January 2027 trial dates in complex antitrust matters slip frequently, and any continuance would push the state case closer to, or past, the federal proceeding in Washington.
Where enforcement attention may go instead
State enforcers watching this outcome have a reasonable read: coordination cases are slow and evidentiary, while conduct-specific statutes are fast and self-executing. Expect more of the latter.
The same logic is visible in adjacent areas of retail regulation, from subscription cancellation flows to product recommendation systems. Our earlier reporting on how US subscription-commerce enforcement is sharpening traces a near-identical shift from litigation toward prescriptive rules.
Why the Cartwright Act made this case possible at all
California’s antitrust statute is older than the Sherman Act’s modern interpretation and, in several respects, reaches further. That is the structural reason this case exists as a state action rather than a federal one.
The Cartwright Act prohibits combinations that restrain trade, and California courts have historically applied it to vertical arrangements that federal courts have grown more reluctant to condemn. The state has argued that Amazon’s vendor relationships form exactly such a combination, with the supplier acting as the conduit between two competing retailers.
Federal law would frame the same facts as a hub-and-spoke conspiracy, requiring proof that the rival retailers knowingly joined a common scheme rather than each responding independently to a supplier request. That is a demanding standard, and it is where cases of this type usually fail.
The rim problem
In hub-and-spoke terms, Amazon would be the hub, the vendors the spokes, and the agreements among competing retailers the rim. Without a rim, what remains is a series of independent vertical relationships, which are generally lawful.
California’s unsealed evidence is aimed squarely at that gap. Communications showing a brand confirming it asked Target and Walmart to raise prices are, in the state’s framing, the rim made visible.
Amazon’s answer is that a retailer expressing a pricing preference to its own supplier is unremarkable, and that what the supplier then does with other customers is not Amazon’s agreement. The trial will test which reading the evidence supports.
Why the Unfair Competition Law claim matters separately
Running alongside the antitrust claim is a claim under California’s Unfair Competition Law, which prohibits unlawful, unfair or fraudulent business practices. Its “unfair” prong does not require proof of an agreement.
That gives the state a second path. Even if the coordination theory fails at trial, conduct found to be unfair under the statute can still support injunctive relief and civil penalties.
It also explains why Amazon’s April 2026 attempt to argue the statutes did not reach its conduct was significant. Losing that motion left both theories in play.
How large is the alleged consumer harm
California has not published a single damages figure, and the public record does not support one. What it contains instead are individual price movements, which is a different kind of evidence.
The Levi’s example described by the attorney general’s office involves khaki trousers moving from $25.47–$26.99 to $29.99, an increase of roughly 11% to 18% at a single retailer on a single item. Allergan eyedrops were described as settling at $16.99.
Those are illustrative rather than aggregate. Scaling them into a statewide consumer-harm number requires an economic model that has not been made public, and any figure circulating without one should be treated as speculative.
Why the absence of a headline number is not a weakness
Public enforcers in state antitrust actions frequently seek injunctive relief and civil penalties rather than classwide damages, which changes what they need to prove. The state must show restraint of trade, not quantify every dollar.
Civil penalties under the Unfair Competition Law accrue per violation, which in a case involving many products and many transactions can reach large totals without a damages model. That is a meaningful exposure independent of any consumer-harm estimate.
It also means the trial will be argued largely over conduct and intent rather than econometrics, which tends to favour whichever side has the more quotable documents.
What this means for price-parity policies across the industry
Amazon is not the only platform that ties visibility to competitive pricing. Comparable mechanics operate across marketplaces, travel platforms and app stores, usually expressed as fairness or best-price provisions rather than as explicit parity clauses.
European regulators moved against explicit parity clauses years ago, particularly in hotel booking, on the theory that they dampen competition between distribution channels. US enforcement has been slower, and the California case is the most advanced US test of the softer, policy-based version.
A California judgment against Amazon would not automatically invalidate similar policies elsewhere, because state court rulings do not bind other jurisdictions. It would, however, give every other state enforcer a tested template and a public evidentiary record to work from.
The practical read for multichannel sellers
Sellers running the same catalogue across several marketplaces have been managing parity constraints for years, and the operational answer has generally been differentiation rather than confrontation: bundle composition, pack size, or channel-specific variants.
Those tactics remain available and are unaffected by the ruling. What changes, if anything, is the documentary discipline around them, because a paper trail that reads as coordination is now demonstrably discoverable.
For brands with both first-party and third-party relationships on the same platform, the two channels should be treated as separate compliance surfaces. Pricing conversations in the vendor relationship carry risk that seller-side repricing automation does not.
What retailers should actually do now
The honest answer is that no compliance program needs to change this week because of this ruling. What has changed is the visibility of the underlying practices, and visibility is what usually precedes regulation.
Three areas deserve review by anyone selling through large platforms. First, how retail pricing requests from platform partners are received, documented and escalated internally.
Second, whether any internal process results in a brand relaying pricing expectations between competing retailers, which is the conduct at the centre of California’s case. Third, how automated repricing tools are configured, since the same tooling that maintains parity can also produce coordinated-looking outputs.
That last point connects to a live regulatory question about how far automated systems can go before their outputs are treated as deceptive or collusive, a question we examined in the context of AI shopping recommendations facing a US deception case.
What to watch next
Four dates and events will define the next phase. The first is any written elaboration of Schulman’s reasoning, which will indicate how the court views the strength of the coordination theory even though the motion turned on urgency.
The second is the pre-trial motion schedule in San Francisco, which will show whether January 19, 2027 holds. The third is the final scheduling order in the FTC case, where the parties proposed moving the bench trial to March 29, 2027.
The fourth is legislative. If more states follow New Jersey in prohibiting specific pricing practices outright, the strategic importance of the Amazon trial diminishes regardless of who wins it, because the conduct will be governed by statute rather than by case law.
For now, the position is straightforward. California’s case against Amazon is alive, fully contested, and scheduled. What it no longer has is any ability to change Amazon’s behavior before a judge rules.
Frequently asked questions
Did the judge dismiss California’s case against Amazon?
No. The court denied a request for a preliminary injunction, which is emergency relief sought before trial. The underlying claims under the Cartwright Act and the Unfair Competition Law remain intact and are scheduled for trial on January 19, 2027.
Why was the injunction denied?
According to MLex and Law360, the court found California failed to show an injunction was necessary to prevent impending injury, and questioned a late-stage motion aimed at conduct the state had not shown was ongoing. Reuters reported the judge noted the state relied on evidence from 2023.
What exactly does California accuse Amazon of doing?
The state alleges Amazon used its leverage over first-party vendors to get competing retailers to raise prices, or to have lower-priced listings removed from rival sites, so that Amazon would not be undercut. The alleged enforcement tools include listing suppression, advertising restrictions and financial penalties.
Which brands and retailers are named?
Vendors named in the unsealed April 2026 materials include Levi Strauss, Hanes, Allergan, Newell Brands, SkullCandy, Westinghouse, GlobalOne, Agrothrive, BabyVision, Maxi-Matic and Songmic. Rival retailers named include Walmart, Target, Chewy, Best Buy, Home Depot and Wayfair.
Is this the same as the FTC case against Amazon?
No. The FTC case, filed in September 2023 with 18 state attorneys general and Puerto Rico, alleges monopolisation of online marketplace services in federal court in Washington state. The California case is a state antitrust action about price coordination, and it goes to trial first.
Does the ruling change anything for Amazon sellers today?
No. No court order now restricts how Amazon manages the Featured Offer, applies pricing policies, or negotiates with vendors. Sellers face the same commercial conditions as before the ruling.
What has Amazon said about the allegations?
Amazon called the injunction motion “a transparent attempt to distract from the weakness of its case,” noting it came more than three years after the complaint. The company says it is consistently identified as America’s lowest-priced online retailer and that it looks forward to responding in court.
Has Amazon won every ruling in this case?
No. In April 2026 the same court denied Amazon’s summary judgment bid, rejecting the argument that California’s antitrust laws did not reach the conduct and finding unresolved factual disputes about whether Amazon entered prohibited pricing agreements in exchange for platform visibility.
Could the January 2027 trial date move?
It is possible. Trial dates in complex antitrust litigation slip frequently as pre-trial motions and evidentiary disputes are resolved. Any continuance would push the California trial closer to the federal FTC proceeding, for which the parties proposed a March 29, 2027 bench trial.
Public access to the underlying evidence in the case was secured through the California attorney general’s office, which has published the relevant case materials on its official press release page.