The People Powering Amazon's Trickle Down Monopoly

Audio of this conversation is available via your favorite podcast service.

Amazon is one of the world’s largest and most powerful companies. Yet one of the engines of its might is largely invisible to customers- its vast network of millions of third party sellers. In today’s episode we talk with Moira Weigel, an Assistant Professor of Communications Studies at Northeastern University and the author of a recent report for Data & Society, Amazon's Trickle Down Monopoly: Third Party Sellers and the Transformation of Small Businesses. For the report, Weigel spent a good amount of time trying to understand experience of the people operating the small businesses that power Amazon’s global expansion.

Below is a lightly edited transcript of the discussion.

Justin Hendrix:

Moira, I'm so excited to talk to you today about a report that you've done for Data & Society, Amazon's Trickle-Down Monopoly: Third-Party Sellers and The Transformation of Small Business. So you've taken a look at the role of third-party sellers on Amazon's platform and done a deep dive, talked to a lot of them as well. I want to ask you just about your methodology before we start.

Moira Weigel:

Yeah.

Justin Hendrix:

How did you go about doing this research?

Moira Weigel:

Yeah. It's a great question, and like the academic I am, it's always hard to know where to begin because then you go back through the whole process. I got fascinated by Amazon's marketplace at a fortuitous time. It was right before the pandemic and right at the beginning of the... or relatively early in the US-China trade war. So a lot of things were happening and changing, which I think is always an interesting time to look at a particular site. In terms of how I found the actual folks I spoke to, that was a real learning process and actually maybe is interesting to talk about for a moment.

Initially, I'd written a book in the past. I published a book in 2020 with Ben Tarnoff, who's a co-founder of Logic Magazine, which I edited just until this last January. We've just handed it to new editors, but we had done a lot of anonymous interviews with employees at tech companies, at the big ones, everything. In that book, we talked to everybody from a founder whose company was bought by one of the big companies to a Google engineer, to a cafeteria worker, and an in-house massage therapist is one of my favorites, actually.

In the past, I thought to learn about the tech companies, I would go find people from inside who were willing to talk to me anonymously. I did speak to a number of Amazonians and former Amazonians in doing this research, but after doing a couple engineer interviews, for a couple reasons, I decided that wasn't the way I wanted to go to understand the marketplace first. Of course, engineers are breaking their NDA to talk to me, and it's complicated to talk to people who currently work for Amazon. That makes it difficult to recruit a robust... any kind of random sample at all, really, but more importantly, engineers, brilliant engineers are very focused on one particular little piece of the marketplace. Right?

So, an engineer could explain to me how their widget that they developed to, let's say, determine what shows up in search in different countries, how that worked, but I quickly realized that to get a more global view, it actually might be better to go to these third-parties who, of course, are not NDA-ed and currently working for Amazon, and who also are hustling this really complicated sprawling system every day. I think there's a real sense that the single mom reselling stuff she buys for cheap in Target across North Carolina has a better understanding of certain global dynamics in the marketplace than the engineer in Seattle does. That, of course, is politically compelling or appealing to me, too.

I believe in telling the stories of technology through the people who use it, and so what was interesting then is I thought, "Okay. This will be easy. I'll go find some third-party sellers to talk to." I first tried a quantitative approach where with a friend, Francis Tseng, we scraped 5,000 or 10,000 sellers' contact information, tried to contact them at random. That didn't work at all for reasons I'd be happy to elaborate on. But then, ultimately, realized that the community is actually very, very secretive, and private, and hard to get into, and so it took some time.

Ultimately, what happened is I got to know a few gatekeepers in the community who then made personal introductions, and I think probably because sellers are so isolated and secretive in a lot of instances, once people trusted me, I found that, actually, they were very valuable, and willing and eager to talk. Some of them joked of me that I became their Amazon therapist, or someone just told me that his 2023 resolution was to complain less to me about Amazon over signal. But anyway, I could elaborate more on any of that if it's interesting, but I basically found a couple different kinds of gatekeepers, and then just tried to develop a rolling sample by asking for introductions. When I'd finished an interview, I'd ask if people knew other people in the community and built up the population that way, but it took months. It was much more challenging than I expected at first.

Justin Hendrix:

Well, you end up with a very compelling portrait of the lives of these individuals, and I suppose they're the people essentially living within the economic dynamics, the game dynamics that Amazon's created for them. You call them the ambivalent realities of building a small business under the conditions that Amazon has created and controls. One of the things I like about this report is that you give us a history of how those dynamics have evolved. There are three periods: the Old Days, the Wild West, the Jungle. Can you just give us that potted history? Take us through each period. What changed between those periods, and why are these periods, of course, reliant on language that, of course, is colonial?

Moira Weigel:

Yeah. Thanks for the question. So when I finally got people to talk to me, I started interviewing people. After you speak to a dozen, a couple dozen people, you start to realize or I started to realize that I was hearing certain stories or versions of certain stories again and again, what a social scientist might call saturation, what we call saturation. I realized, and it took some time. It took me a long time actually to just literally understand how all sorts of things about the marketplace work because it's very complicated. It's not written down anywhere, and I felt for a few months, really, beginning this research, that I was just learning all the lingo, and the different systems and programs, and so on.

Over time, as I interviewed more and more people, all of whom had been active had their own personal stories and had been active in different time periods and Amazon eras, I started to recognize that there were a few, still, narratives that recurred and even specific phrases that recurred, which you just alluded to, Justin. So I noticed at least among the English speakers, and I did interview a cohort of Mandarin speakers. I'm actually still conducting interviews with Chinese Amazon merchants now. But among the English speakers, I repeatedly heard these three phrases of the Old Times, the Wild West, and the Jungle.

To be honest, when I first heard subjects use these terms, especially the Wild West, I probably felt a little inwardly eye-rolly about it. It's a Silicon Valley cliché, a tech industry cliché to describe a quickly growing messy and unregulated space as the Wild West, but I came to believe two things. One, that each of these metaphors was describing something important and salient about the experience of being on Amazon at that time, and two, that, actually, the abrupt breaks between these periods... and I found that sellers who had very different biographies and lived in very different places generally concurred about when the breaks between these periods were. I came to think that that itself was a significant finding that Amazon marketplace shifts dramatically and abruptly at these particular moments in reaction to policy and other decisions that Amazon makes.

To put it concisely, the first era that relatively few people I interviewed had been active in because it was a long time ago, but the folks who called themselves old-timers, generally, were talking about selling between the year 2000 when Amazon opened its marketplace to third-parties. Basically, as someone said, they grafted an eBay onto a Walmart or something, but added this third-party platform dimension to their retail business. In that era, when Amazon offered sellers was basically just an online catalog, and subjects described to me faxing Amazon to apply to sell to it, giving Amazon their cut, and then every day, or couple of days or week, whatever it was, Amazon would send them orders with the names and addresses of the customers, and they would mail them off. So, in that era, people who got into it were mostly people with pre-existing small business experience. They had to have all their sourcing and their own logistics and so on.

The second era, the Wild West that people spoke about really started around the year 2010. Shortly after introducing the Prime program or subscription for customers in 2006, 2007, Amazon undertook this renovation of the third-party marketplace and importantly, opened their warehousing and fulfillment or logistics services to third-party sellers. What that meant was that whereas previously if I, Moira, had wanted to start a scissor company, I'm just looking at my desk, I would've had to get my scissors, store my scissors, ship them to my customers, and so on, even through Amazon.

Starting in the late 2010s and really, by 2010, they launched this program called Fulfillment by Amazon that sellers could use, paying Amazon to do those things for them as a service, and that also became obligatory in most cases to be eligible for Prime, for Prime shipping. So sellers were strongly incentivized, if not practically required to use that. So, in practice, what this meant was it really lowered the barrier to entry for starting one of these Amazon businesses, and all the people I spoke to who were active in that era described it as a period of chaotic growth.

The third phase, which again, people pretty uniformly identified starting around 2015, 2016, people described as the Jungle. It's funny. It was the Jungle. The use of this phrase, "The Jungle," that made me think a little bit harder because on its own, I thought, "Wild West. Whatever. Every tech guy says that." But I realized over time that this metaphor of the Jungle was really talking about two things. It was talking about the marketplace getting denser and more complicated, less the open terrain where you could scale in a very rapid way relatively easily as some people did in the Wild West period. Much more competitive, much trickier in terms of all sorts of policies, and regulations, and programs. I realized after spending some time doing these interviews that the main causal factor, the main thing that had changed was that Amazon opened their global, opened their marketplace in the US globally. I believe I should fact check this, but I think there are 182 countries that are currently eligible where people can sell...

Justin Hendrix:

Almost all of them.

Moira Weigel:

Yeah. I could be wrong about that. There are also 132 customs-free zones, so maybe I'm making up that number. Don't quote the 182, but many, many, many countries are eligible to sell in practice. In practice, only China matters outside the US. 50% of all sales to the US marketplace. So all third-party sales through Amazon.com come directly from China. Amazon has tried to recruit merchants, particularly in South Asia and increasingly in Latin America, especially in Mexico, to sell through the platform, especially in light of growing geopolitical tensions and instability between the US and China. But in practice, for better and worse, what opening the platform globally meant was a flood of Chinese manufacturers and small capitalists coming onto the marketplace, and that changed the dynamics dramatically in ways that everyone I spoke to pretty much recognized and saw as impacting their daily life. On the Chinese side too, of course.

Justin Hendrix:

Not that there's no danger in the Wild West, but I do associate it with wide open spaces. But in the jungle, there's a "killer be killed" mentality.

Moira Weigel:

Totally.

Justin Hendrix:

Danger around every bush, that kind of thing. You described how, essentially, the individuals who are in the Jungle are essentially taking on the mindset of Amazon itself. They're, at some extent, reading in Amazon's imperatives, and then building that logic into their businesses. Can you just explain how that works?

Moira Weigel:

Yeah, for sure. As you were talking, I remember the second thing I meant to say about the Jungle because I take these forms of lore and these metaphors people use to speak about their lives really... and I take them really seriously and try to think with them. I remember I was speaking with two Data & Society colleagues, Rigoberto Lara Guzmán and Ranjit Singh, who's an anthropologist, about these metaphors of the Wild West and the Jungle, and they said precisely what you did. They said, "How atmospherically is the Wild West different from the Jungle?" Well, it's darker. It's denser. Maybe you need a guide. So that really resonates.

The other thing I realized that this language of the Jungle reflects or that I believe it reflects is the huge growth in the mid to late 2010s of whole industry of consultants and firms providing software and services to Amazon sellers, specifically all of whom try to brand... or most of whom, I should say, try to brand themselves using the imagery of the Amazon and the jungle with Amazon's branding. So, in a way, I feel like it also bespeaks the rise of this influencer and service culture that promises whether or not, truthfully or accurately, to help sellers navigate this scary terrain.

In terms of businesses becoming like mini Amazons, this is one of the things that really interested me. Right? If I can answer in a slightly academic way, since this is an expert audience, one of the things that really fascinated me and fascinates me about the third-party sellers is that they are small businesses. They're registered as small businesses. A lot of the people who get into it do so because they want a kind of independence and autonomy that we associate with entrepreneurship in the US, and there's a long tradition in our political imaginarian life of thinking about entrepreneurship and small business that way. But in practice, particularly because Amazon is so dominant in e-commerce, most of these sellers are selling the vast majority of what they sell through Amazon primarily, if not exclusively through Amazon.

I think of the 40-something people I interviewed for this study, one person I interviewed sold only 60 something percent, did only 60 something percent of his revenues through Amazon, but most people were eight in the 90s percentages. So anyway, these are really Amazon native businesses, and that was another surprise to me, which was perhaps naïve. When I started the research, I thought I'd be meeting small business owners who used Amazon to "get online." Everyone. I'm trying to think if there's a single exception. I think every single person I spoke to who had a reasonably successful Amazon business over time had invented a Amazon native business, specifically designed to perform to Amazon's metrics and algorithm, incentives, and so on.

Justin Hendrix:

Wow.

Moira Weigel:

Sometimes, as you were suggesting in the question, the forms this took really made these small businesses look like mini Amazons, and that was a metaphor that people use jokingly, for instance, describing how their houses, and cars, and apartments, and garages got filled with stuff, how they've turned their living space into a mini fulfillment center. I always remember interviewing a Chinese seller from Shandong Province who is speaking about reading these books in Mandarin, many of which were published in the 2010s, about how to get rich selling on Amazon. All of which were saying to imitate Amazon's tactics of selling at a loss for the first month, for six months to gain search ranking and so on.

So sellers have to really imitate Amazon's behaviors in certain ways and certainly perform to Amazon's metrics and algorithms. In that sense, I think there's a big conversation about how folks ride hailing drivers and delivery workers have been misclassified as entrepreneurs when really, they're more like workers or employees. I think in a certain sense, I don't think the Amazon businesses are quite the same as driving an Uber or driving for Uber or Lyft, but I do think there are some similar dynamics in terms of gigification and forms of algorithmic control that significantly undermine the autonomy that we associate with entrepreneurship, and that's part of why we valorize entrepreneurship in the US. In China too, by the way, but yeah.

Justin Hendrix:

You say that these businesses are becoming less like mom-and-pops and maybe look a little more like day trading. Almost seems to me like that's the right metaphor, but maybe the other one would be that they're in a casino that does pay out, but to some extent, they're trying to game the different dynamics, trying to make a buck. It seems like the house always wins.

Moira Weigel:

Totally, and both the day trading... It's funny. The day trading metaphor is one that I think I highlighted in the report. At least two different people I interviewed explicitly used that metaphor and made that comparison saying... and one person who used it was someone who'd had a long history in retail before getting into Amazon selling, but two different people said to me, "Basically, it's less like running a corner store or a mom-and-pop than day trading. It's all about recognizing risk, and seizing short windows of opportunity on this global marketplace that you encounter through screens."

The casino metaphor was also one. It's funny I didn't highlight it in the report, I realize, but that came up a lot, and during COVID... This is a bit of a slant, and I don't think I talk about this very much in the report, but it's something I'm looking at in the longer project I'm doing on e-commerce. During COVID, this whole industry of what basically private equity funds or investment funds that call themselves aggregators came up out of nowhere raising capital to buy Amazon businesses. I was just looking at these figures today, but one of them here in Boston, which has since basically disappeared, raised a billion, nearly a billion dollars.

So it's these really big VCs for... They're not VCs. Yeah. Private equity firms for Amazon brands emerged out of nowhere and became a multi-billion-dollar industry during the pandemic, and that was... So that's in the time period when I'm conducting my interviews, and constantly, people I spoke to used the metaphor of wanting to take money off the table or get out of the casino and sell to one of these aggregators before their luck ran out because even very successful sellers, I think, have, for the most part, a pretty acute awareness that basically, anytime, Amazon could copy their product, change some rule, a manufacturer, some other manufacturer could figure out a way to make their product cheaper, and so on. So this sense of timeboundness and the house always wins definitely came out in this way of talking about risk and getting money off the table.

Justin Hendrix:

Can I ask you? Did you observe or hear about any defensive behaviors that people had to try to hide from Amazon somehow or keep Amazon from noticing their success? Were they at all aware of the possibility that they might get essentially pushed out of the casino at some point?

Moira Weigel:

Yeah. It's a great question. I don't know if this is quite what you're getting at, but one thing that I expected going into the study, again, not having a background in e-commerce or spoken to third-party sellers before, was that I would hear a lot about fears of Amazon copying products or stealing suppliers because that's something that I think comes up a lot in the public discourse around antitrust just because it's so egregious.

I actually don't think I ever once heard... once. I can think of one person who independently brought up that possibility or voluntarily brought it up. I sometimes would ask people about whether that was something they worried about, and almost everyone I can remember asking about that said something to the effect of, "Well, that can always happen. Well, of course, that's a given. Of course, that can happen." It's funny.

Another one of these recurring metaphors was the sandbox. Many people said like, "What's their sandbox?" You're playing box. So, interestingly, to me, people seem to fear that to an extent, but take it for granted, and then the things that they were more concerned about or defensive about were competition, either fair competition or abusive behavior by competitors, which Amazon, they often feared, would not effectively govern or prevent. I could offer a million examples of this kind of abuse tactics.

Then, the other thing people really worried about were unfair suspensions or getting their account or their product blacklisted or suspended in ways that I think we're used to thinking about for social media. We've all heard stories of Twitter accounts that are unfairly banned or a Facebook account gets brigaded and taken down unfairly, but these kinds of things happen on Amazon Marketplace too, these familiar platform dynamics. In a way, they feel lower stakes perhaps because it's not the alt-right or neo-Nazis brigading an activist. But on the other hand, it's people's entire livelihoods and many people's livelihoods in some case. I've spoken to multiple sellers who have lost over $1 million in revenue while trying to reach a human at Amazon who could explain to them what happened.

Justin Hendrix:

Wow.

Moira Weigel:

Keep in mind, while they're losing that money and waiting, their inventory is in Amazon warehouses. They can't access it. Their search rank, which is basically their brand, like if you think about what a brand is in the age of Amazon, it's a position in Amazon search results basically, is plummeting. So, again, maybe we're talking about dog collars and bath gels, so it seems less exciting than Nazi memes. But then, on the other hand, it's people's entire lives.

Justin Hendrix:

There is a broader intellectual project obviously going on right now that's trying to redefine monopoly harms, rethink, antitrust, rethink how to think about these mega platforms, the network effects, and other implications of platformization. You introduced this idea of trickle-down monopoly. I want to ask you two questions in one which is, how do you see your work fitting into this intellectual project, and what is trickle-down monopoly?

Moira Weigel:

I'll answer those questions in reverse order if it's okay. It's funny. Trickle-down monopoly was a phrase that just struck me talking to sellers, and then I tested it out with some of them, and they liked it and thought it was accurate, and I went with it. I used this concept of trickle-down monopoly to try to name this fundamental ambivalence of the seller position, and part of what interests me about them is their position is really ambivalent because in a way, they stand to gain a lot from Amazon's monopoly, but the promise that Amazon holds out to the seller is that they can grab their piece of Amazon's huge market share or Amazon's growth and to bring it back to the colonial metaphors.

That was part of what struck me about those two, right? It's like the promise of the Wild West or the Jungle is that the settler colonialists or conquistador, whoever gets to go grab their piece of this new terrain being opened up. At the same time, it struck me that many of the people I spoke to saw Amazon itself as the crown or the state that could arbitrarily take things away or change things in the background, so.

Justin Hendrix:

Yeah.

Moira Weigel:

Anyway, the trickle-down monopoly concept I used to refer to the way that Amazon, because of its huge market power, holds forth this promise of participating in its growth and wealth at the same time that it renders... People don't really mostly. It mostly doesn't really work, and it renders the small businesses dependent on it incredibly vulnerable to its changing whims and imperatives. In terms of the broader intellectual project, I think I'm broadly sympathetic. I mean, very sympathetic to it or agree with a lot of the criticisms that folks like FTC chair, Lina Khan, or scholars like Sabeel Rahman and Tim Wu have made about how post-Borkian definitions of monopoly that look at the problem exclusively through rise in consumer prices can't capture the kind of power that affirm like Amazon exercises.

I guess I probably tend to see the problem based on my work and my own orientation. Perhaps maybe it's the same conclusion, but from a slightly different angle in terms of just the level of power that firms exercise over people's lives and livelihoods. Maybe it is an exact agreement. I don't know, but that rather than focus so much on the benefits of competition, I'm probably coming from a place of thinking that it is a problem and a democracy to have individual firms exercise so much power over people's lives, livelihoods, and the political process.

One other point I want to make, I try to make a bit in the report, and I'm looking at in my ongoing research is also that entrepreneurship is a state project. We often think of these abuses to sellers as only consequential for those sellers or for those businesses, but small business is something the US invests in through low interest rate loans, and tax breaks, and subsidies in some places. It's been a central way that we've thought as a country in the past 50 or so years about...

Justin Hendrix:

Yeah, major part of our identity, right? You mean that's...

Moira Weigel:

Yeah. So I think that it's like... to the extent. There's a new study. Someone called Juozas Kaziukenas just published this week that got some attention where he, by looking at the P&L statements of all these businesses that shared their P&Ls with them because he's really respected in the space, found that the average Amazon seller pays upwards of 50% of every sale to Amazon, which to me at least was a shockingly high number. The most recent robust number I'd heard for that was from a few years ago and was more like 34%, 35%. But to the extent that small businesses or state project, public resources go towards supporting small businesses. I think it's important to recognize that if Amazon is increasingly taking 50%, 55% of a piece of every sale a small business makes, that's a matter of public interest and not just, not just to the sellers.

Justin Hendrix:

Well, I will look forward to the next result of your continued work on this subject.

Moira Weigel:

Oh, thank you.

Justin Hendrix:

I appreciate this report, and I appreciate you talking to me today.

Moira Weigel:

Thanks so much for having me on and for the interest, your interest in the work.