Discussion: (0 comments)
There are no comments available.
A public policy blog from AEI
More options: Share,
Are large tech firms like Google, Apple, Amazon, and Facebook unassailable monopolies? Is there anything new about the way these companies try and maintain their market dominance? And have antitrust activists ever successfully predicted which big businesses will be forever companies? On this episode, Cato’s Ryan Bourne discusses his recent paper, “Is This Time Different? Schumpeter, the Tech Giants, and Monopoly Fatalism.”
Ryan Bourne occupies the R. Evan Scharf Chair for the Public Understanding of Economics at the Cato Institute. Before joining Cato, Ryan was the Head of Public Policy at the Institute of Economic Affairs and Head of Economic Research at the Centre for Policy Studies in the UK. What follows is a lightly edited transcript of our conversation. You can download the episode by clicking the link above, and don’t forget to subscribe to my podcast on iTunes or Stitcher. Tell your friends, leave a review.
It deeply saddens me that some of our some of our listeners may not know who Joseph Schumpeter was and why we would be talking about him in relation to the Big Tech giants everybody is talking about these days.
Yeah, it’s a shame. He’s somebody we in free market circles talk about a fair bit. His main claim to fame is coining the term creative destruction to describe the process by which capitalism delivers new products and services that arise as a result of robust competition, which leads to the collapse or change of whole industries and structures over time.
He even went as far as to say this was the essential fact of capitalism: the birth and death of companies from competition.
Yeah, I think that’s right. And there are two reasons why he’s particularly important to this debate about tech. The first is that there are several economists — even people who subscribe to the consumer welfare standard and believe that antitrust policy should be focused on trying to deliver as much consumer welfare as possible (such as Jason Furman, who used to work for the President Obama’s administration) — seem to believe that there’s something fundamentally different about tech companies. Whether it’s the fact that these companies benefit from network effects, or that they’re very large and have a massive economies of scale, or that many of them are platforms and therefore differ from companies delivering traditional retail goods, that’s what they believe.
I think that’s important in a Schumpeterian sense, because if you go back and look over the past hundred years or so, there were numerous other companies that operated in similar industries to today’s tech giants. People thought they were unassailable monopolies for the exact same reasons as people now think the tech firms are fundamentally different. If you take that as given, the way Schumpeter really adds value is that he recognized that antitrust policy — to the extent that it does work — often misses a very important margin of competition, which is this competition from the fundamentally new product or new innovation delivered in such a way that marks a decisive quality or cost break from previous industrial output.
Traditionally, if you look through some of those firms that we believed were unassailable monopolies, they didn’t get blown away by similar competitors competing with them on price. They got blown away by new forms of competition that one couldn’t even have imagined.
So it wasn’t some other company doing something 10% better with a little better advertising or maybe their product was branded slightly different — it was something very different. So you would really be competing in a fundamentally different way or else you probably couldn’t beat a big established competitor if you proceeded otherwise?
Yeah, that’s exactly right.
Just to be clear about what Schumpeter is saying about “Monopoly fatalism,” is that just the idea that these companies can’t be assailed without intervention?
Yeah, I think that’s right. That’s what we hear in a lot of these debates about the tech companies. People talk about them as if there’s just no way that a competitor could overcome the economic advantages that they seem to have. For example, people just say the network effects associated with Facebook are just so large that it would be impossible for another entrant.
What’s an easy way to explain network effects to the audience?
Essentially, it’s the way that people get more value from the product the more users there are. Then there’s also a secondary term which is called indirect network effects, which basically means that once a product becomes overwhelmingly popular, other people making complementary products might start designing them to work most effectively with this popular product (such as a single internet platform). As such, it becomes much more difficult for a new platform to come in because they’d have to attract all these secondary markets in such a way that they could actually have the supply of goods and services that people want.
When I think of this, I often think of video games. There are only so many platforms that developers can write games for, so an ecology builds up around these platforms. Let’s go through these historical examples you mentioned, starting with the A&P grocery store, which I’m just old enough to remember. What can we learn from that company that applies to technology companies today?
I describe this as the Amazon of its day. It was a company that fundamentally disrupted the retail sector. Prior to A&P and other chain stores coming around, people used to go to small independent stores. Goods were sold to them on credit, and often then delivered to customers.
You didn’t walk in these stores and start taking things off the shelves and putting them in a basket.
No. Often, goods were behind the counter, as you’ve seen in some pharmacies today. Other times you went in and placed your order and then the goods were delivered to you at home. These stores also tended to buy their goods wholesale from middlemen who were often corrupt. They were often buying and selling their products as middlemen and that made the industry as a whole quite high cost and inefficient.
What A&P did is: it came along and offered standardized stores; it vertically integrated in some ways; it started producing its own bread, started producing its own canned goods, some dairy products, too. It founded its own distribution network in the same way that we’re talking about now with Amazon. For products that it didn’t produce, it built so many stores that it actually benefited from being able to buy in bulk from wholesalers and was able to drive down the prices that it had to pay for those goods.
So you combine all those things together and those efficiency gains actually led to consumers benefiting up to 15% in terms of price cuts. Now, the reason that this is so similar to the story you hear about Amazon today is that all the same complaints were made by industries disrupted by this change of model, alongside some concerns at the consumer level too. It was said that a A&P harmed local economies by usurping those traditional retailers, and it was said that they were engaging in predatory price cutting.
That’s a very common complaint about Amazon, that you’re cutting prices to drive someone else out of business, and you’re going to crank up the prices as soon as they get out of business.
Exactly. That’s the broad story. They also said that A&P was giving preference to their own goods within the stores. So, Amazon did not invent private label goods. This is something that’s been around in the retail market since chain stores came into the system. So this is nothing new under the sun here.
I think the thing that really did have an impact on policy is that people said these chain stores were able to drive prices down from wholesalers in a way that was fundamentally damaging. You hear about monopsony power now — big buyer powers. Monopoly is when you have a single seller in the market; monopsony is when you have a single buyer in a market. This stuff led to a bunch of pushback from wholesalers and other associated retailers, such that by 1936, we saw the introduction of the Robinson-Patman Act, which essentially banned wholesalers from offering different prices to different retailers. That led to prices going up for shoppers.
So when we’re thinking about Amazon, we have to be aware that there are lots of vested interests — industries that have been disrupted as a result of Amazon’s business model — that are currently wailing about the effects of Amazon, but that’s something very different from consumer welfare. In the A&P case, it’s clear that A&P was good for consumers — they drove down prices and increased choice substantially. My fear is that we’re returning the debate to an antitrust policy predicated on protecting competitors rather than protecting the dynamic process of competition.
As you wrote, a big complaint is that A&P was driving the small mom-and-pop retailers out of business. Decades later, those complaints also were launched against Walmart, and now those same arguments have been applied to Amazon. So we can follow this storyline over a century with very similar arguments being made.
Yeah, that’s entirely right. Whenever you seem to have a period of immense disruption to a sector, you get these sorts of complaints. It’s also worth noting that in the A&P case, people thought that the chain store model was the end of the line for retailing, but chain stores themselves then had to compete with the arrival of big-box supermarkets. We then had supermarkets integrated into shopping malls so that people could do all of their shopping in one place. And we had the rise of national television, which established brand products in the consumer consciousness and made it much more difficult for firms like A&P with their own branded products to compete and maintain their model.
To what extent was the decline of A&P because of these new competitors versus government action hurting their business model either directly or distracting them with regulations and lawsuits and the need to lobby?
Overall, it seems to have been a slow decline, which suggests that it wasn’t purely about the antitrust action. But evidently the antitrust case didn’t help: in 1949 A&P launched a nationwide advertising campaign just making the case that their activity was beneficial to consumers. They took out ads in 2000 newspapers across the country. So clearly, they were having to devote both time and resources here. But it’s difficult to ascertain exactly how much that affected the fortunes of the company, not least because we saw the rise of national brands as a result of national television at the same time they were overcoming this antitrust case. Antitrust action did have an effect in terms of the distraction, but ultimately they themselves were undone by the exact same changing dynamics of a market that they themselves had once inflicted.
It’s a really classic case of an incumbent unable to adopt a different business model in the face of new competitors, because their original model had worked so well for them.
So, there’s your Amazon analog. Another great one — it always comes up — is MySpace. In fact, you lead off your section on MySpace with a quote from a technology writer in 2007: “Will MySpace ever lose its monopoly?” It was unimaginable that as powerful a social media platform as MySpace could ever be beaten, but now it’s a punchline. So what happened to MySpace?
Commentators often dismiss this example and say you’re attacking a straw man because not many people thought that MySpace was an unassailable monopoly. But when you actually go back, there were people writing exactly these types of articles, and there were antitrust cases brought against MySpace. LiveUniverse, for example, said that MySpace was engaging in anti-competitive conduct because the company wouldn’t deal with them on a contractual basis.
News Corp didn’t buy it for half a billion dollars because they thought they were they were as you know, I a fly-by-night company that was going to easily be disrupted. This was a huge bet that MySpace would have sustainable dominance.
Yeah, Rupert Murdoch’s News Corp saw the potential value in social networks, but it just so happened that MySpace just didn’t endure for very long. People claimed MySpace wasn’t just a monopoly, but a natural monopoly. The broad idea was that because people had invested time in uploading content, coupled with those network effects, meant that their position was pretty unassailable.
So a competitor wasn’t a click away because it was sticky, because you already you already invested this work by designing your page and uploading your content. Theoretically, you could go to a competitor, but it was so difficult. It was a very high barrier.
Yeah, that’s right, and there’s a very funny quote in one article by John Barrett in TechNewsWorld where he essentially says that social network websites like MySpace are much stickier than search engines like Google, where with just one click you would take someone to a competitor’s site. So Barrett foresaw that MySpace would endure as a dominant Monopoly, but Google would never be able to change their positioning in such way.
So MySpace was founded in 2003 and saw a rapid expansion of users. They were bought out by Rupert Murdoch’s news Corp in 2005 for $580 million. For a brief time, their revenues exploded by about 50 times. In 2006, they were the most visited website in the US over Google. So we’re talking about a massive company here, but then Facebook came along.
Facebook was a different type of social network with a different balance between the user interface and the extent of adverts on a page. It was more user-friendly and allowed more space for innovation and the eventual development of the apps that we saw. Importantly, Facebook was able to overcome those network effects in part because it adopted what might be described as an email address importer. So, when you joined the network, you could agree to give access to your email account to Facebook. It could then send emails to certain other people, friends or whatever, and it would try to build a network to overcome the incumbent network effects that MySpace had.
MySpace’s market share, as a result of this aggressive move from Facebook, declined from over 70% of the social networking market in the US to less than 30% within a year. So, this was an incredibly rapid decline, and I think what this shows is that network effects definitely yield a cost of entry which can be difficult to overcome, but they’re not insurmountable. Their existence doesn’t entrench monopolies for the long term, and the reason for that is that competition can still exist for the whole market. Yes, we might not want to have the type of social network in the form of Facebook on four or five different websites. There’s value in one network being dominant to the extent that it has a large network of friends that we’re able to connect with, because it’s a one-stop shop for contact storage, for messaging, and for all of these other things. But that doesn’t preclude competition for the whole market. So I think what the MySpace example shows is that if Facebook doesn’t keep up its game of constant renewal and seeking measures to try to keep people interested in the site, then network effects can lead to a rapid deceleration once you start losing active users.
A criticism of the MySpace example is that Facebook’s dominance is much more prolonged and sturdy. Granted, you’ve had these competitors come up, whether it’s Snapchat or Tik Tok. But do you think that changes your thesis in any way — that Facebook is now even a kind of holding company for social networks like Instagram and WhatsApp? Can you have challengers to that position?
I think one of the really difficult things here is to define the contours of a market. Is Facebook primarily a social network? Is it a digital advertising space? Is it part of the broader advertising ecosystem?
My old friend Nick Clegg, former Deputy Prime Minister of the UK, who is now working for Facebook, had an interesting article in the New York Times where he actually outlined one way of thinking about Facebook: it’s competing in all sorts of different sub-markets. It’s competing in the instant messaging market with a range of different firms, including Snapchat and iMessage; it’s competing in the uploading of photos with a bunch of different companies; it’s competing in contact storage; it’s competing in the digital advertising space. So defining the market is incredibly difficult, and from an antitrust perspective, that’s the first step. We have to define the relevant market, then we have to assess whether the firm has monopolistic power and is acting in anti-competitive ways.
So, the point I’m getting at is this: there are many people — even people who agree we shouldn’t fundamentally overhaul antitrust laws — that believe that there’s just something different about firms with significant network effects. And I think the MySpace example just serves to show that network effects don’t preclude a new competitor coming in and taking some subsections of the market or acting as competition for the whole lot. That’s what Facebook did to MySpace. Yes, Facebook has endured for longer. Clearly, it’s doing something right that’s maintaining consumers’ attention, but that doesn’t mean that something else can’t come along in the future.
The whole point of the paper is to hearken Schumpeter’s point that we can only really assess this stuff by looking retrospectively, because none of us have a crystal ball able to predict the technological and company changes that will come.
Yet the historical record should give us at least a degree of confidence that even though we can’t predict what that future competitors will look like, they do arrive. I wasn’t even going to go into this example, but I love this Forbes headline about Nokia. It read, “Nokia: one billion customers — can anyone catch the cell phone king?” Obviously, someone did catch the cell phone king. Interestingly, this was in 2007, so this Forbes article pinpointed Nokia’s ‘unassailability’ at the very moment that its competitor was emerging.
That’s right. It’s almost too good to be true. There’s one quotation within that article where the Forbes writer explicitly says that as a result of the number of handsets that they were selling and the reinvestment through R&D, Nokia was able to create economies of scale. The article said no mobile company will ever know more about how people use phones than Nokia.
And supposedly they had so much data extracted from their customers that it would be impossible to beat them.
I think people talk about this data issue in the wrong way. People talk as if companies like Facebook and Apple are a blank page where we go on and write everything about us, and then from that they extract value and sell it to advertisers. Actually, much of the data that we give Apple or Facebook is created as a result of us using their platforms in ways that are beneficial to us. There are things that we click or engage with or watch that inform the company about our desires, but the company has obtained that data by providing a service that we enjoy engaging with. So it’s not purely our data. People talk about the need for portability to other social network platforms, but that data is in part a creation of the networks and platforms that Apple, Facebook, and others have themselves created. So I think it’s wrong to think about this stuff in terms of “our data.”
But you’re right. Nokia was thought to have so much data and have such big economies of scale from their large global market share that they were going to be able to dominate for a long time. Nokia also primarily specialized in hardware, and what they didn’t foresee was how important app-based technologies were going to be for smartphones.
That’s where all these stories lead—to a “what they didn’t foresee” moment. The companies don’t foresee them, the Wall Street analysts don’t foresee them, the journalists whose beats are these companies don’t foresee them, and yet continually people assume that they can foresee.
Sometimes they do foresee certain possibilities, but they don’t foresee the potential. Nokia actually designed and produced some of the first smartphones, but it didn’t see how important app-based technologies within those smartphones future could be in the future. Kodak was overwhelmingly dominant in photography, and they actually produced the first digital camera — Steve Sasson in 1975. But they didn’t foresee the potential for that to be a large-scale consumer product for the mass market. They had that technology and only sold it as a high-end niche product.
So, yeah, it’s not always a case of firms not being able to conceptualize a different way of doing things or a different type of product. Sometimes they do, but they don’t think that they can produce them on the right scale to benefit from their creation or they don’t foresee the potential latent demand there for the products. Internally, there’s also sometimes resistance to change in a model that is perceived to be working. You know, if it isn’t broke, don’t fix it. So it’s sometimes very difficult for incumbents to make that transition from producing one type of product to another.
Yes, especially if you may cannibalize your existing business line, which is already producing revenue, for something that may or may not produce revenue in the future.
What you think about the argument that companies with a lot of data and money can go out and see which start-ups may be potential competitors, buy these companies up and therefore stop them from growing up to become real competitors? To what extent do you think that’s happening, and are you concerned at all?
Well, horizontal mergers of that variety already fall under the purview of antitrust law. The main thing I’m worried about is—
The classic example is the Facebook, Instagram, and WhatsApp mergers, the claim that Facebook should not have been allowed to buy those companies. So instead of us having to choose between two or three competing companies that may have very different privacy policies, data policies, or some other differences, they’re just all Facebook One, Facebook Two, and Facebook Three.
Yeah, and you see that with Google and YouTube as well, but clearly there are potential consumer benefits here. Google bought YouTube, and by integrating it with search technology, Google has been able to fundamentally make YouTube a better product for consumers. It’s much easier to find things on YouTube now, and it’s much cleaner to upload videos to YouTube. And when people are worried about some of these issues to do with child protection, having a firm the size of Google behind YouTube, one would imagine they will be able to handle these problems more effectively, though obviously there are difficulties at the moment. We’ve seen a similar thing with Facebook and Instagram as well. Instagram is still a fundamentally decent product. The consumer still has the choice to use it and opt for it, but now you can also share your photos from Instagram on to Facebook as well, which is a new addition to the product that consumers are able to enjoy.
I think there’s something people miss here. Yes, there’s a potential problem with companies that are in dominant market positions buying up potential competitors and quashing competition. We can look at these problems through existing antitrust law and cases. But it’s also important to note that the potential for a start-up to be bought by one of these huge companies is massive.
That is often hand-waved away in these discussions. I’m not entirely sure how to balance the problem of a company just getting big with the benefit of providing an off-ramp for these companies. Because the latter allows the founders to create another company, and those investors could put this money back to work, and maybe that product makes the acquiring company better. This would seem to be a significant factor you need to consider if you’re going to suddenly become much tougher on letting big companies buy these small startups.
One thing I’d just say in conclusion is, if you look at these tech companies, they’re spending vast amounts of their revenue on research and development. They’re constantly looking to innovate and serve the low ends of markets that other tech companies aren’t currently serving. And if you read their annual reports, they tend to be overwhelmingly worried about things that they can’t even foresee for some of the same reasons we’ve outlined here.
It isn’t that they’re not aware of this issue — they’ve read the Clayton Christensen book, they’re wary of disruption, and they’re spending so much on R&D. So they don’t look like they’re comfortable, that they’ve figured it out and will never be challenged.
No, the Big Tech companies’ behavior do not look like companies that think they’re entrenched monopolies with a walled garden and barriers to entry so high that it would be impossible for potential new companies to come along and steal some of their business.
My guest today has been Ryan Bourne. Ryan, thanks for coming on the podcast.
Thank you very much.
There are no comments available.
1789 Massachusetts Avenue, NW, Washington, DC 20036
© 2019 American Enterprise Institute