Take Five: Amazon’s growth tests us in profound ways
Amazon has built the most dominant e-commerce delivery system in the U.S., and the difficulty that competitors have in matching either its price or delivery speed is the reason the $280 billion, multisector giant is on track to be America’s largest company. Dr. Robin Gaster, George Washington University public policy scholar and president of Incumetrics, an innovation consulting company, has been tracking this remarkable business story and is about to release a new book titled Behemoth — Amazon Rising: Power and Seduction in the Age of Amazon.
The book does a deep dive into the factors that make Jeff Bezos’ secretive empire so formidable. The London-born Gaster refers to these factors as the five interlocking defensive moats that stifle the competition: logistics, Amazon Prime, the Amazon Marketplace, low pricing, and constant innovation. In this Take Five interview, he explains why it’s possible that Amazon might actually get out of the retail business within the next decade and why it’s dabbling in areas like cloud services, publishing, and groceries.
As I’ve interviewed independent retailers since the pandemic hit last March, they fear that Amazon has only grown stronger during the pandemic. What’s your take on that?
“It seems like Amazon was the retailer that solved the logistics problems best. Let’s put it this way: There is no sign at all that independent retailers are better able to avoid Amazon, right? We’ve basically had a year since the pandemic started and I don’t see any real traction for Walmart Plus. Shopify is interesting but they don’t have any delivery network. I mean, they are just using FedEx as far as I can figure out, and they don’t have any entertainment value. They are an alternative if you have an independent store already. It seems to me they are good for brick-and-mortar companies to try to get into the web without being ‘Amazoned.’ Turn it around, really. If you were an independent retailer and you were using Amazon in March, is there anything that would say, ‘Oh, I can leave now?’”
No, not really.
“No, I don’t see it. Amazon is also squeezing them a bit harder. It’s becoming a little tougher in there. So, the answer is no, Amazon is absolutely still dominant on the platform.”
One of the points you make pertains to the “dangerous power” Amazon has unleashed. Explain that and also why you believe transparency and sunlight are the antidote to that?
“Amazon is a fantastic company. I really do think it’s unique and it’s uniquely brilliant. This probably comes down to their capacity to run things through small teams, pretty semi-autonomous small teams. That’s a very difficult thing to do when you’ve got a huge enterprise and somehow you sort of Heath Robinson together these teams so that they all work. It’s very, very impressive. I benefit from Amazon every day. They’re brilliant, but it seems to me the world is tending toward platforms. The retail platform is only one. There will be others in health care, probably finance, maybe in education. Who knows? Amazon is unlike most other companies in that it appears to have no natural boundaries.
“Bezos has been quite clear. Your margin is my opportunity. There is margin everywhere. There’s opportunity everywhere. Satellites are next. It’s not a normal company that works in its core and then maybe puts something else next door. That’s not Amazon. So, they continue to grow. They continue to expand. They continue to cross subsidize, which is clearly going on. I am sure — as sure as I can be, given how secretive they are — that they lose a lot of money in retail. That means their pricing is predatory, almost by definition. Somehow, they manage to lose money in retail and have predatory pricing and still lose money. There is a puzzle to explain there and I have some ideas about what it is, but nonetheless, there is this enterprise with cross subsidization going on, tremendous ambition, a process that says ‘OK, if you have a good idea, we’ll give you some running room and if it kind of starts to work, we’ll put real resources in and you can build a team and you can go do it.’ And so, it sort of metastasizes. It’s designed to metastasize.
“Amazon famously has a growth flywheel where all these different pieces fit together, and everything gets faster and faster and better and bigger. But flywheels typically come with a brake so that they don’t destroy themselves. Amazon has no brake, as far as I can see. There is nothing there to stop them from expanding indefinitely.”
If they are losing money on retail, where are they making it up?
“This is an interesting point. People point to AWS, [Amazon Web Services], their internet service, which makes about $9 billion a year. They look at the $14 billion in total profit and they say, ‘Well, AWS is 60% of Amazon’s profit. If we just push that aside or broke it up or did something, that will solve the problem. It would just be flat.’ But that’s not really true. Amazon makes $14 billion in profits. I think they lose $30 billion in retail. So, the rest of the segments, which are AWS, advertising, Prime, and Marketplace, together make $45 billion in profit. So, AWS is not as important as it appears, and these other semi-hidden segments are much more important. You can’t see them because basically they are busy filling in the hole caused by Amazon retail. So, they are not very visible, but I think Marketplace makes twice as much money as AWS. That’s the money flow and it looks a bit different then. They have plenty of segments that are doing well, and there are others obviously coming. Amazon B2B is here already. Nobody knows how big it is because it doesn’t report at all, but the estimates I’ve seen say this is an $18 billion enterprise.”
Is that the primary reason that you believe Amazon will leave the retail business in the next decade?
“Whether they fully leave is an open question. Imagine that you run Amazon and you have two huge revenue drivers. You have your own retail business, and you have Marketplace. The thing about Marketplace is that it’s infinitely scalable. You can continue to add vendors and add segments and add areas to your Marketplace business and there is zero risk for Amazon. They get 15% profit in commissions plus somewhere around 7.5% on average in advertising. So, 22% of revenue is a very nice margin for a zero-risk business, and their catalog could extend indefinitely. It’s wonderful for Amazon.
“On the other hand, you go to Amazon retail and it’s tough. Margins at the best of times are low. Amazon is taking tremendous risks. They have inventory risks all the time. Now, they are pushing into areas that make no sense at all. I mean, Amazon as kind of a high-couture fashion plate — through Luxury Stores — is ludicrous. It’s hubris absolutely gone mad. It’s not scalable. It’s not a business that’s associated with Amazon in any way. Some team had an idea, and they are off and doing it. Either it will work, or it won’t, but it’s certainly not going to move the needle for a $120 billion enterprise. It’s just not big enough.
“So, if you’re Bezos and you’re looking at it systematically, you have to say that Marketplace is a much better business. Historically, Amazon has been shifting the share of revenue from retail to Marketplace by about two percentage points every year. It’s been happening, and it will continue to happen. If it goes on for another 10 years at the same rate, and it might accelerate, Marketplace would be 80% and Amazon would hold on to just the [retail] nuggets that it is really dominant in, like books.”
Is part of this because the retail face is so much in the public consciousness that the average consumer doesn’t really realize that Amazon is much more than that?
“People think about Amazon as delivering shoes and now, in an interesting and I think disastrous twist, delivering groceries. There are three levels, right? The people who know about Amazon retail, and then there are people who think AWS is really the big dog in this, and then there are people who know all the various segments and understand what’s going on. Most people don’t. They don’t follow Amazon. Why would they care? But Amazon is also a model. It’s quite likely that if it keeps going at the current rate, it will overtake Walmart in a couple of years. It will be our biggest employer. Its employment practices demand pretty good scrutiny. They are not terrible. In some ways, they are better than Walmart’s. They pay better. But some of this stuff, and the sort of disposability of both blue-collar and white-collar workers, is really worrying. That’s not an economy any of us want to live in.”
Well, they are viewed as heroes by some because of the way they came through for displaced workers during the pandemic. They filled a lot of gaps.
“Oh, they were fabulous. Amazon is absolutely unbelievable for customers. It’s better than the best thing since sliced bread. It’s incredible. I can voice a request to Alexa sitting here and say, ‘Buy me this pair of shoes,’ and they show up on my doorstep the next day. But, and there is a but, we all know people in the warehouses work in a different way. I feel that Amazon is just basically an amoral company that just does what it’s allowed to do. It’s up to us to set the rules so that it does so in a way we all accept.”
If they are so immoral, what can’t we stop using them?
“Well, because they are brilliant. They have become indispensable. Well, they are not actually indispensable. I have friends who don’t use them. The parallel is with the 1840s. You would not have wanted to stop using cotton and woolen because of the conditions in the mill. What you want to do is change the conditions in the mills. It took 100 years of struggle, but eventually we got a kind of social contract that said, ‘OK, the work day is eight hours. You can’t employ kids. You pay overtime. It has to be safe. There should be some security network behind you.’ All of that took tremendous struggle, and I feel like we are now facing a new, I mean people are happy to say oh, the digital revolution. Well, it is. It’s the new industrial revolution, and all of us who don’t control it will have to figure out how to put new rules in place. We’re already seeing that with the social networks.”
You note that Amazon has built defensive moats. Well, not only has Amazon built defensive moats, it appears to have populated them with alligators. If you were an independent retailer located in, say, Madison, Wisconsin, how would you try to compete?
“The reality is that if you want to be on the internet in the U.S., I believe you have very little choice about using Amazon because it’s simply too expensive to acquire customers. People can’t find you and Amazon solves that problem, and it solves the delivery problem for you. Yes, it takes half of your revenues but to be honest, how much did Walmart take? How much is Walmart’s markup? It’s between 30–40% and then you still have to ship the stuff there and you have to do the market, and you are dangerously dependent on Walmart. I have friends who sell to Costco and in order to sell to Costco, you have to commit to huge volumes, and Costco could send you off the next day — huge risk. So, Amazon is risky, yes, but is it worse than that? I don’t know.
“What I would do is say, ‘OK, I’m going to go into it with eyes open. I’m going to make sure that I have a team to ensure that I am compliant with Amazon and to watch my listing like a hawk in case black hats get after me.’ That’s the cost of doing business on Amazon. You have to protect yourself from black hats and from Amazon’s justice system and from Amazon’s private labels. The private label thing, overall, is not a big deal. Amazon says it’s less than 1% of its retail, which is a pinprick. They could drop it tomorrow. I think they lose money on it, too. If they ship you a small packet of AA batteries, they are losing money. There is absolutely no doubt the shipping cost is more than you’re paying. They must be losing money. So, do they make money on their private labels? Probably not, but it could be enough to crush you. The final publication [of the book] contains a long interview with a guy called Jason Boyce, who testified before Congress, and Amazon absolutely came after him. They came after him four times in different iterations and eventually gave up. It’s possible that you could find yourself in Amazon’s crosshairs.
“I guess the other thing to do is to make sure you’re somewhat diversified. If all you make is one thing, or all you sell is one thing, it’s pretty dangerous.”
One of those defensive moats is the way they innovate. Explain Amazon’s approach to innovation.
“They are the most innovative big company in the world. In the book, I have a list of innovations and it stretches for a page, and they are major innovations. Not necessarily technical things but business innovations or the kind of things nobody really thought of like one-click purchases, which was a huge edge for them for about 10 years. They do this through a combination of culture — they have a drive to innovate that permeates the company in an unusual way. There’s structure in the team-based notion that you can form a small team of seven to 10 people and take on a big project and do it kind of outside the current lines of hierarchy.
“That’s very powerful and it’s supported by a decision that Jeff Bezos made 20 years ago to insist that all teams open themselves up to electronic information exchange. When you think about it, they have a lot of small teams. How would you exchange information between them? How would any of them know what’s going on? It’s an impossible problem unless there are electronic ports built into each of them to connect to everybody else. So, they have to fully publish their assets and their processes and how you connect to them and how you ask for things. That’s a very important part of Amazon that’s pretty much hidden. Nobody really knows about it.
“Then, there are the processes, and these are quite famous. They have an unusual innovation process where they work back from the end point and famously start by designing the press release for the product. You know in particular who is the customer and what they are getting out of it, and then there is an interesting, highly iterative process to get the idea fully fleshed out as well as you can before you start to actually build anything. Then, they are pretty good at scaling things. They have ideas like the Prime Now project, which was to do two-hour delivery in a city area. They put that together and put it on the street in six months. That’s ridiculous. How aggressive is that kind of schedule? It’s hard to keep up.
“So, you have structure, you have process, you have culture, and you have resources. Amazon can afford to let people go off and do their own thing because they’ve got a lot of money and they’ve got a lot of technical resources.”
When it comes to the big tech companies, we keep hearing the lament of why doesn’t the government take them on in an antitrust fashion? The politicians always talk a good game, but they never do anything. What’s going on there?
“Well, they’re going to. They are going to do it, but I don’t think Amazon is particularly vulnerable to be honest. Warren (U.S. Senator Elizabeth Warren) has bitten on this private label thing. She’s been going on about this private label thing, but it’s so unimportant that in my more paranoid moments, I think that Amazon is actually doing it to bait the political class into looking in the wrong direction. It’s like bait in the water for antitrust and they will fight it and push back for five years and say, ‘OK, you got me, well give it up.’ It doesn’t matter. In any case, they know they can rely on Walmart to block it because it really matters to Walmart.
“Could they break off AWS? That’s currently a pretty hot topic. Why? Two things. One, it’s absolutely not possible under current antitrust law because it wasn’t an acquisition. It was organic growth. So, there is nothing to unwind. Secondly, let’s assume they change antitrust law, and they spin off AWS. What does this accomplish? It gives Amazon an enormous pile of cash. If there was an AWS initial public offering, it may be the most valuable company in the world, and Amazon would take its cash and do whatever it wants with it. It wouldn’t give it back to shareholders. It doesn’t do that. And would it change the relationship between AWS and Amazon? No, why would it? Amazon would buy services from AWS and AWS would be happy to have Amazon as its biggest client. No problem.
“The third thing about the antitrust issue is that people know that antitrust disputes take a long time. The past big antitrust cases, six to 10 years. Well, all right, but in 10 years, is Amazon going to be the same company? Hardly. We don’t know what it’s going to be, but it’s going to be very different. So, you risk irrelevance.
“The other problem is subtle. Antitrust is fundamentally a one-off project. You charge them with certain antitrust violations, and it’s a binary thing. They either have them, or they don’t. And then you go through this discovery period. You ask Amazon for documents. Well, there’s a primitive idea right there — documents. But you ask them for documents, and they give you millions of documents. And then you search through them and try to put together a case. I mean, this worked really well for Standard Oil about 100 years ago, but the idea that you could somehow prevail over Amazon by doing this, that it would make any difference, that it could make a permanent change, is exceptionally naïve. What you need is real-time monitoring of real-time rules and regulations. It’s not enough to say to Amazon, ‘We’re going to give you a slap on the wrist for what you did seven years ago.’ When you get your head around the whole thing, doesn’t that seem like a 19th century approach to a 21st century problem?”
Maybe it’s all they have at the moment.
“But they could change that. What they need to change is the assumption of corporate privacy. Once these platforms get big enough and dominant enough in their niche to begin with — I mean, I have a wider agenda. I think corporate privacy is a scourge and we should be thinking about it more in general. In their platform, don’t they act more like utilities? It seems to me they look like utilities, and we have different information rules for utilities. If utilities want to ask for more money, if they want to ask for a rate increase or they want to get into a different business, or they want to buy some upstream supplier or something — any significant business decision — they have to reveal everything. They have to explain what they are spending money on, how much they are spending, and what their labor rates are. There is all kinds detailed information that they have to reveal.
“I don’t want to put price controls on Amazon and I don’t want to put innovation controls on Amazon. I’m not interested in those things. I’m interested in making them reveal everything about their platform, and that includes their own retail activities on it. They should be telling us everything. Insofar as their logistics network makes it possible to have such a platform, they should reveal everything about that too. Personally, I believe we’ve allowed business in general to be much more secretive about what they do than is good for the community. The balance has shifted tremendously and corporate is much more powerful than it used to be. This is one way to rebalance things. If we start with Amazon and a platform, we need to know what they are doing. We need to know how their algorithms work. If we don’t, it’s a kind of public utility that is run entirely privately. I don’t think that’s a good thing.”
You don’t consider this proprietary or perhaps even intellectual property?
“Intellectual property is a function of state power. We define what intellectual property is and what rights it has. So, if we insist, for example, that Amazon publish its algorithm, that’s fine. There is nothing immutable about privacy. It’s the result of law and Supreme Court decisions and custom. I just believe it’s time for a change. After all, you and I won’t make it across the digital plain. We are entirely naked. I don’t care how much privacy, security you personally operate with because in the end, I don’t believe it makes much difference. I think I’m completely naked but somehow, we have come to believe that corporations are allowed to be entirely private. That’s ridiculous.
“We need to rebalance, and Amazon is a good place to start because what they are doing is within the area of a monopoly. They are the marketplace monopoly. They are not the retail monopoly. They are not the e-commerce monopoly. But in the platform world of retail in the U.S., yes, they are a monopoly.”
On this topic, do you have anything else to leave our readers with?
“It’s important for us to think about the consequences of allowing Amazon to acquire more and more authority with no corresponding responsibilities. I just don’t think that’s tenable in the end. We can’t give them everything. In the end, they are prepared to say, ‘That’s collateral damage. It’s not our problem.’ Well, it is their problem.”
Click here to sign up for the free IB Ezine — your twice-weekly resource for local business news, analysis, voices, and the names you need to know. If you are not already a subscriber to In Business magazine, be sure to sign up for our monthly print edition here.