Last Wednesday marked Jeff Bezos’s debut appearance before Congress, testifying alongside the CEOs of Apple, Facebook, and Google. To be blunt, it did not go well. But that’s probably to be expected, given things were already off to a rocky start when the Committee threatened to force Bezos’s appearance (by subpoena), after several months of protracted but unproductive negotiations trying to get him to testify.

To help give some baseline context about Amazon’s recent, huge numbers prior to Bezos’s appearance before Congress last Wednesday:

CEO: Jeff Bezos

Market Cap: $1.585 trillion

During Covid: Increased in value roughly $600 billion since 17 March 2020


Biggest Issues

The single-biggest specific takeaway from Wednesday’s hearing was in Amazon’s fluid definition of data “aggregation”. Bezos was pushed hard by North Dakota Republican, Kelly Armstrong, and eventually conceded that while Amazon does aggregate between many different, similar products to extract data on what, why, and how something sells best, “aggregation” could mean as few as 2 different sellers’ listings – meaning it could be possible for Amazon employees to infer which anonymous data was whose – allowing them to better position Amazon’s own products more advantageously.


Which done at scale, raises questions of unfair competition regarding how Amazon uses private data it’s aggregated, to position its own branded goods more favourably vs. the same competitors it’s drawing the data from.

Amazon in two stats

If you were trying to pick a single stat that embodied Congress’s regulatory questions about Amazon, the obvious candidate is probably revenue: $75.5 billion for Q1, 2020 alone.

But a huge number like that doesn’t necessarily tell you a lot. Any discussion of Amazon’s possible monopoly (or simply its de facto retail position) can’t be accurately told just through revenue.

A more telling stat could be the share of products sold by third-party sellers on Amazon, which had been steadily increasing for over a decade, from 26% (Q2, 2007), to a peak of 54%. During which time, Amazon was presumably hard at work, aggregating data to understand and streamline what sells best. And during that time, the sheer volume and range of different products they offered increasingly from third parties, brought more and more traffic to the site, and built more and more base for their brand, resulting in an amazing 89% of online shoppers saying they trust/prefer Amazon over almost any other retail site. Seemingly a win-win for Amazon and its third party sellers.

But that percentage has now been steadily decreasing over the last 12 months, which wouldn’t seem abnormal on a case-by-case basis – ie. Amazon stops selling one item, and starts selling another (which it happens to make its own branded version of, as it knows what people want). When taken at scale though and filtered through different definitions of the meaning of aggregation, it raises very obvious questions about data collection, competition, monopoly, etc.

Elephant in the Room

Amazon can easily claim it competes with Cosco, Target, Walmart and others in retailing physical goods. It competes with IBM, Microsoft, Oracle and others in web services. It competes with Apple, Disney, Netflix and others in streaming services.

What’s notable about that list? Most obviously, that there’s no repeats – in other words Amazon can claim it’s simultaneously competing with many of the world’s largest, most innovative companies, in every area of its business. That its dominance in so many areas is simply a natural evolution of its being thrifty, reinvesting religiously, and making the same strategic moves any other company would make (M&A, cost-cutting et al). And in doing so, inevitably growing into a dominant force in so many parallel industries.

In essence, Amazon can claim it’s simply fighting (and generally winning) individual business battles on multiple fronts in a way that, individually Netflix, Walmart, or Oracle simply aren’t. From one point of view, it can almost look like Amazon is engaged in a sort of economic Jus Ad Bellum. An assertion backed up by the fact that Bezos repeatedly cited examples of others, implying that Amazon is not necessarily the organisation Congress should be worried about at the moment. 

Conversely, if Amazon’s detractors take that same view to its logical conclusion (ie. Amazon’s building economy-wide momentum by winning battles in every key area of its super-diverse business), the takeaway is: yes, Amazon’s fighting and winning on many fronts, which individually may not necessarily be monopolistic, but after enough victories on all fronts it may inevitably win a war of economic and technological attrition. Where a victory by any one company in a war for the economy, would be a loss for all consumers.


Amazon’s position in a quote

“I can’t answer that question yes or no. I can tell you we have a policy against using seller-specific data to aid our private label business but I can’t guarantee you that that policy has never been violated. “ in response to Parmila Jayapal’s question: “Does Amazon have access and use third-party seller data when making business decisions?”


Reading between the lines

Maybe the most interesting, underlying part of Bezos’s testimony was what he didn’t say. He spent far less time than any of the other CEOs, on trying to argue that his company (and its growth), were somehow broadly “good for America” – by comparison an assertion that Pichai made more than ten times during his testimony, and Tim Cook made at least a half-dozen times during his much shorter testimony. It’s impossible to know Bezos’s mind, but it could arguably be that Amazon is simply so obviously huge, that some points become moot.

Bezos was also the least evasive of the 4 CEOs (i.e. in claiming he either didn’t know, or would need to get back to Congress with further information), especially compared to Zuckerberg or Pichai who regularly defaulted to more evasive positions.

Congress’s position in a quote

In response, Pramila Jayapal asked “You have access to the entirety of sellers pricing and inventory information, past, present and future, and you dictate the participation of third-party sellers on your platform. So you can set the rules of the game for your competitors but not actually follow those rules yourself. Do you think that’s fair to the mom and pop third-party businesses that are trying to sell on your platform?”