Amazon Is Strictly Liable for Marketplace Items, Reinforcing That Online Marketplaces Are Doomed–Bolger v. Amazon

This is another one of my overlong angsty posts about the death of Section 230. Today’s angst is about the liability regime for online marketplace transactions, the inevitable demise of online marketplaces, and how we’ll all end up poorer when that happens.

The Bolger Decision

The impetus for this blog post is last month’s ruling Bolger v. Amazon by the California Court of Appeals. That case involves an exploding battery sold by an impossible-to-sue vendor via Amazon’s Marketplace. In the lower court, Amazon won on Section 230 grounds. On appeal, the court says that Amazon is strictly liable for the marketplace sale. The court provides this overview:

Amazon placed itself between Lenoge and Bolger in the chain of distribution of the product at issue here. Amazon accepted possession of the product from Lenoge, stored it in an Amazon warehouse, attracted Bolger to the Amazon website, provided her with a product listing for Lenoge’s product, received her payment for the product, and shipped the product in Amazon packaging to her. Amazon set the terms of its relationship with Lenoge, controlled the conditions of Lenoge’s offer for sale on Amazon, limited Lenoge’s access to Amazon’s customer information, forced Lenoge to communicate with customers through Amazon, and demanded indemnification as well as substantial fees on each purchase. Whatever term we use to describe Amazon’s role, be it “retailer,” “distributor,” or merely “facilitator,” it was pivotal in bringing the product here to the consumer…

Amazon is the only member of the enterprise reasonably available to an injured consumer in some cases, it plays a substantial part in ensuring the products listed on its website are safe, it can and does exert pressure on upstream distributors (like Lenoge) to enhance safety, and it has the ability to adjust the cost of liability between itself and its third-party sellers. Under established principles of strict liability, Amazon should be held liable if a product sold through its website turns out to be defective.

It’s fair to say that the opinion author, Judge Patricia Guerrero, was not interested in Amazon’s explanations. Each of Amazon’s arguments got a swift whack-down. The opinion is razor-sharp and 100% hostile to Amazon’s business model.

Strict Liability

The court provides even more detail about how Amazon’s conduct triggered strict liability:

Amazon created the environment (its website) that allowed Lenoge to offer the replacement battery for sale. Amazon attracted customers through its own activities, including its direct offers for sales and its Amazon Prime membership program, which includes benefits for some products offered by third-party sellers (including the Lenoge replacement battery at issue here). Amazon set the terms of Lenoge’s involvement, and it demanded fees in exchange for Lenoge’s participation. Amazon required Lenoge to indemnify it and, assuming Lenoge met the sales threshold, to obtain general commercial liability insurance listing Amazon as an additional named insured. Because Lenoge participated in the FBA program, Amazon accepted possession of Lenoge’s products, registered them in its inventory system, and stored them in an Amazon warehouse awaiting sale. Amazon created the format for Lenoge’s offer for sale and allowed Lenoge to use a fictitious name in its product listing. The listing itself conforms to requirements set by Amazon. Even setting aside the use of a fictitious name, the listing does not conspicuously inform the consumer of the identity of the third-party seller or the nature of Amazon’s relationship to the sale.

To purchase the product, the consumer adds it to her Amazon cart, not her Lenoge or E-Life cart. The consumer pays Amazon for the product, not Lenoge or E-Life. And, in the FBA program, Amazon personnel retrieve the product from its place in an Amazon warehouse and ship it to the consumer in Amazon-branded packaging. If convenient, Amazon will ship the product together with products sold by other third-party sellers or by Amazon itself.

Lenoge is not involved in the sales transaction. It does not approve the sale before it is made. It may not even know a sale has occurred until it receives a report from Amazon. It does not receive payment until Amazon chooses to remit the proceeds. Its use of any customer or transaction information, if it even receives any from Amazon, is strictly limited. But it accepts the burden of substantial fees for Amazon’s participation, approximately 40 percent here.

If a customer wishes to return the product, she ships it back to Amazon under the FBA program. Amazon personnel inspect the product, determine whether it can be resold, and if so return it to inventory in the Amazon warehouse. Third-party sellers like Lenoge are prohibited from communicating with Amazon customers except through the Amazon website, where such interactions are anonymized.

The court articulates the policy considerations supporting strict liability:

  • “Amazon, like conventional retailers, may be the only member of the distribution chain reasonably available to an injured plaintiff who purchases a product on its website”
  • “Amazon, again like conventional retailers, ‘may play a substantial part in insuring that the product is safe or may be in a position to exert pressure on the manufacturer to that end; the retailer’s strict liability thus serves as an added incentive to safety.'”
    • The court gave the example of how Amazon now requires UL certification for replacement batteries.
    • Amazon provides an implied representation of safety. The opinion quoted a deposed Amazon employee who said “Amazon does everything in its power and goes above and beyond to make sure that we’re providing the best customer experience, including safe products. And, you know, I want that for all of our customers and for myself when I buy from Amazon, so I hope people believe that.” This kind of stuff makes my head hurt, because the court seems to imply that Amazon could avoid liability by taking fewer steps to protect consumer safety.
  • “Amazon, like conventional retailers, has the capacity to adjust the cost of compensating injured plaintiffs between itself and the third-party sellers in the course of their ongoing relationship.”

Amazon cited dictionary definitions of “seller.” The court was nonplussed: “The doctrine of strict liability in California was intended to cut through such technicalities and compensate plaintiffs for injuries caused by defective products.” The court does acknowledge the extensive precedent in this area but says much of it is factually or legally distinguishable; in particular, the court says cases NOT involving FBA (Fulfilled by Amazon) are factually distinguishable.

Amazon argued that it didn’t choose to list the battery for sale. This argument also goes poorly:

Amazon did choose to offer the Lenoge replacement battery for sale. Amazon is no mere bystander to the vast digital and physical apparatus it designed and controls….Nothing aside from Amazon’s own choices required it to allow Lenoge to offer its product for sale, to store Lenoge’s product at its warehouse, to accept Bolger’s order, or to ship the product to her. It made these choices for its own commercial purposes. It should share in the consequences.

A variation of the “you make your bed, you lie in it” maxim, I guess.

Amazon argued that it was like an auctioneer, but the court says that would help only with respect to the sale of used goods, and auctioneers are less involved in the product sales than Amazon was.

The court has some discussion about who will bear the costs of liability. The court says that Amazon can vary its fees to merchants to allocate costs or “limit the sale of products that create a commercially unreasonable risk of injury.”

Section 230

The court says Section 230 doesn’t apply to strict liability claims. (Cites to Erie, State Farm). The court explains:

Bolger’s claims do not require a court to treat Amazon as the speaker or publisher of content provided by Lenoge. The content of the product listing is not determinative, and it need not be attributed to Amazon to support strict liability. Instead, Amazon’s own involvement in the distribution of an allegedly defective product supports strict liability for the reasons we have already discussed….The fact that some content provided by Lenoge was posted on the Amazon website does not automatically immunize Amazon for its own choices and activities unrelated to that content.

How Will Amazon Respond?

I assume Amazon will appeal the ruling to the California Supreme Court.

The trillion dollar question is what, if any, changes Amazon could make to change the court’s outcome. For example, the court repeatedly hammers on the Fulfilled by Amazon (FBA) feature and said the non-FBA caselaw was factually distinguishable. If Amazon eliminated FBA, would a court reach a different result? It’s hard to tell, because the opinion was hostile to so many other aspects of Amazon’s business.

Plus, is that really a good outcome? FBA has major benefits to consumers, including reduced shipping costs and speed. If Amazon turns off FBA and that avoids strict liability, Amazon restores the liability to the status quo but Amazon makes less money and consumers get worse service. (It’s not clear to me if vendors benefit in that circumstance or not). Yay?

Alternatively, Amazon could still offer FBA knowing that it would be strictly liable for those sales, but then limit FBA only to trusted vendors who can provide adequate liability insurance (or who sell items where the risk of strict liability is low). In this scenario, Amazon would force all other merchants out of FBA, or possibly out of the marketplace altogether. This policy decision would reward major incumbents and hurt startups and small business operators, and it would still reduce consumer choice and raise consumer prices. Yay?

Ending FBA may not be enough to eliminate strict liability. Amazon may have to do more radical surgery to its entire marketplace, to unknown effect. I suspect Amazon is wargaming scenarios as we speak. Furthermore, if the opinion governs more than just FBA, then it almost certainly picks up other rivals to Amazon’s marketplace, including Walmart’s competitive offering. Probably, this opinion will have repercussions across the entire online marketplace industry.

AB 3262

The Bolger opinion says: “Amazon first claims that the Legislature, not this court, is the appropriate forum to address whether those policies would be served in new contexts.” That was not idle speculation; as it happened, the California legislature was working on the topic at that very moment.

This year, the California Assembly and the California Senate Judiciary committee passed AB 3262, which would have mostly codified the Bolger ruling. The main bill provision said:

an electronic retail marketplace shall be strictly liable for all damages caused by defective products placed into the stream of commerce to the same extent that a retailer of that defective product would be liable and shall be deemed to be a retailer for purposes of California strict liability law

Exclusions included prominently labeled preowned goods, small-volume handmade goods, situations where the marketplace received no financial benefit (directly or indirectly) [this sounds like an oxymoron, no? how could a marketplace not make any money and still operate?], and auctioned goods–unless, in any of those cases, imposing strict liability would advance the policy goals of strict liability (in other words, plaintiffs would be encouraged to litigate the policy merits of the exclusions every. single. time.).

An electronic retail marketplace was defined as “an electronic place or internet website that is engaged in the business of placing or facilitating the placement of products into the stream of commerce in this state, regardless of whether the vendor, product, or the marketplace has a physical presence in the state or whether…the electronic retail marketplace ever takes physical possession of the product.” (The “facilitating” language was a late add). Obviously, this ensnares every online retailer, but what does it mean to “facilitate” the sale of products? Presumably that would cover Amazon’s marketplace, but it seems like it would cover all other marketplaces too, such as eBay and Etsy. Though not intended, I wonder if the “facilitating” language would reach other advertising venues–say, Craigslist? Or frankly, any site that accepted ads for goods? The bill had language saying that advertising wasn’t a direct benefit, but that language got struck. As a result, the undefined term “facilitating” was the bill’s linchpin, but I have no idea what it meant.

The “facilitating” word also raised Section 230 issues, including an interesting and vexing question of whether a state legislature can draft around Section 230 simply by creating a strict liability tort. (That also raises First Amendment issues, especially if it reaches advertising venues). Then again, the 9th Circuit’s HomeAway ruling essentially opened up the door for legislatures to do whatever they wanted, without worrying about Section 230, so long as they regulated consummated transactions. This means that state legislative floodgates are open for proposals like AB3262 to proceed without any inhibition due to Section 230, and the inevitable flood is guaranteed to swamp the industry.

Here’s the twist: Amazon initially opposed the bill, but it FLIPPED when the bill expanded to cover its competitors. OF COURSE IT DID. Amazon figured (1) the Bolger case already pinned it into an unwanted position, and (2) it could accept the financial consequences, but only if its competition bore the same costs even if they have less money. Hey legislators, when an incumbent backs your bill but only if the law applies to its competitors, there’s a pretty good chance you are advancing a law that hinders competition. YOU ARE BEING GAMED.

Despite the early affirmative votes, AB3262 died when the sponsor didn’t push the bill forward (I was told it was because of the lack of supporting votes) before the legislative session ended. However, I expect it will be reintroduced next year; and if Bolger remains good law, Amazon will re-embrace the proposal for all of the same reasons it flipped this year.

The Future of Online Marketplaces

The (temporary) demise of AB 3262 dos not change the prognosis for online marketplaces. The courts will continue to impose strict liability and say Section 230 doesn’t apply. State legislatures will revisit this topic. And Congress is interested in the topic, including the introduction of marketplace-regulation bills like the INFORM Consumers Act and SHOP SAFE Act. Even if Congress doesn’t pass any marketplace reforms this year, the issue will flare up in Congress again in 2021.

Most of these reform efforts in the courts and legislatures are succeeding without any changes to Section 230. In other words, even today, Section 230 doesn’t act as a barrier to liability. This is why I’m perplexed by proposals to amend Section 230 to exclude marketplace activity. We’re already there, and amending Section 230 for this purpose would take ZERO pressure off other Section 230 reforms.

As we transition to an Internet with strict liability for marketplace transactions (and a sidelined Section 230), I fear we will not like the consequences. First, in my WSJ editorial from earlier this year, I explained why strict liability for online marketplace transactions hurts everyone. Ironically, though it will surely reduce Amazon’s revenues and market cap, it’s likely that a strict liability will help Amazon consolidate even more marketplace control through its direct retailing function. So, is this really the outcome consumer advocates want?

Second, online marketplaces are one of the exceptionalist achievements of the Internet–there literally is no offline equivalent where complete strangers are comfortable enough with each other to blindly transact without doing any research on each other. That basic premise has unlocked hundreds of billions of dollars of wealth in our society (both producer surplus and consumer surplus). Nowadays, we just expect that we can find pretty much every product we could ever possibly want with minimal search costs, and that there are likely to be multiple vendors competing with each other to drive that price down. In a pandemic era where access to physical-space goods has gotten harder, not easier, and online marketplaces have been a primary channel for bridging that gap, THIS is what legislatures want to burn down right now???

Case citation: Bolger v Amazon.com, Inc., 2020 WL 4692387 (Cal. App. Ct. Aug. 13, 2020)