Russ Roberts

Matt Stoller on Modern Monopolies

EconTalk Episode with Matt Stoller
Hosted by Russ Roberts
Ideologically Convenient...Eco... The Mighty Amazon (and Faceboo...

internetmonopoly.jpg Matt Stoller of the Open Market Institute talks with EconTalk host Russ Roberts about the growing influence of Google, Facebook, and Amazon on commercial and political life. Stoller argues that these large firms have too much power over our options as consumers and creators as well as having a large impact on our access to information.

Size:33.7 MB
Right-click or Option-click, and select "Save Link/Target As MP3.

Readings and Links related to this podcast episode

Related Readings
This week's guest: This week's focus: Additional ideas and people mentioned in this podcast episode: A few more readings and background resources: A few more EconTalk podcast episodes:


Podcast Episode Highlights

Intro. [Recording date: December 14, 2017.]

Russ Roberts: Matt Stoller... is also working on a book on the history of monopoly in the 20th century.... Now, our main topic for today is what we might call the Big Three--Google, Facebook, and Amazon--but I'm sure we'll get into other things, as well. But I'm going to start with them, as you do in some of the pieces you've been writing. What's alarming about them? They seem to be very popular. People like using the services the companies provide. What's alarming?

Matt Stoller: So, they are, Facebook, Amazon, and Google are effectively becoming so powerful, with the ability to manipulate and control the way that citizens interact with each other, information and ideas and goods and services that we can buy and sell from each other, that they are in some ways replacing democratic government. So, they've gone way beyond sort of controlling markets. Now they are controlling in some ways our political system. And I'll just give you a quick example. So, I guess it was last month, Amazon put out a bid for it to locate 50,000 jobs in a second headquarters outside of its first one, which is Seattle; and they asked cities from all over the country to bid on it. And you had cities offering to allow, to collect taxes from Amazon employees and just hand that money over to Amazon. You had some cities offering to name their cities after Amazon. You had some cities that were saying, 'We'll basically allow you to run this city and determine where your tax money goes, because we think it's ridiculous to waste money on things like fire stations in the edge of town.' You had this really weird outcry from mayors all over the country saying 'Democracy doesn't really work and we can't generate our own commercial activity. We have to beg Jeff Bezos for that, and so we are going to do whatever we can to attract him.' And that is incredibly disturbing. You see the power everywhere, in lots of different crevices of society. But I think the political power is most alarming.

Russ Roberts: Yeah, well, that particular example is a little bit--not as alarming to me as it is to you. There are other things that I think are alarming, and more alarming than that. The fact that mayors do stupid things and are desperate for something to wave about as an economic development success has been a problem for a while. They do lots of bad things to attract large corporations to their towns from giving up tax revenue, for example, not as dramatically as you've described it with Amazon, at least in the past. But, I'm much more concerned--and I'm not sure how powerful Amazon is. I'm more concerned about Google and Facebook and their control over the information we receive. Are you worried about that?

Matt Stoller: Oh, absolutely.

Russ Roberts: In what way?

Matt Stoller: Well, I mean, you know, I guess there are sort of two different ways to think about it. The first is: You have institutions that are so large that they simply can't be managed. So, Facebook has two billion users on its various networks. Google has 7 products with more than a billion users. And what you find is all sorts of areas where the people that are managing these networks are just not paying attention to parts of the networks that are damaging people. So, as an example, Google organizes the world's information; they've engaged in a whole range of conduct to make sure that specialized search engines don't emerge and potentially challenge them in niche areas. And so, one of the results is: You have a health crisis. That exists already, the opioid crisis. But, when people who are addicted to opioids do a search for rehab clinics, what they will find is, first, about a year ago they would find a whole bunch of different ads from rehab clinics that weren't particularly good but were from out of state, would rip them off, and most importantly, wouldn't help them off opioids. And when Google found about this, they got rid of the advertising. So, they weren't--they were making a bunch of money off of this but they stopped making money on it. And then afterwards, Google's search engine was actually manipulated and gamed by some of these kind of scummy rehab clinics. And that's just an example of--the net result is that a bunch of people who are addicted to opioids and want to get off them, can't. And that's incredibly harmful. And that's not because anybody at Google was sort of a bad person. It's just because the institution, there's just--the network is too big for them to actually manage. And you see that in all sorts of ways with institutions like Facebook, where you have lots of things just coming in the back door. So, that's one problem. That's a problem of absentee ownership--no one's minding the store. The other problem is that the algorithms themselves drive extremist behavior. So, these are based, the algorithms that Facebook uses or that Google uses to sort of attract you and keep you using their technologies, keep your attention so they can sell you more advertising--they have specific biases built in that are not good for human beings. So, as an example, if you are, say, a conspiracy theorist around vaccinations, they will say--the recommendation engine will tell you--'Well, maybe you are interested in this thing called PizzaGate.' Or, maybe you are interested in this thing called, you know, 'Sovereign Citizen.' If you are--an doesn't matter if it's a Right or a Left thing. It's an extremist-generating engine. So, if you are interested in being a vegetarian, maybe you should try becoming a vegan. Or, whatever. And it keeps you and pulls you into a more and more extremist, sort of socially dysfunctional position, because essentially it's manipulating your brain in a way that is very similar to the desire to see a fight going on, where you just can't look away. And the algorithms that these guys use, because they are so attention-based, actually incentivize sort of the worst socially dysfunctional behavior. So, those are two basic problems.

Russ Roberts: You left off the third. Which is the ability, of, say, a foreign country to encourage people to look at nasty things.

Matt Stoller: Well, that's true. That actually falls under the absentee ownership problem. So, we have a huge national security issue, with the ability of foreign actors or disruptive social elements to come in and just manipulate people, because there is nobody managing the store in a lot of these places. So, Russia, could, for example, come in and organize people who are pro-Black-Lives-Matter, or against-Black-Lives-Matter, to all come to a, you know, to a protest together outside a mosque and encourage people to engage in open carry. You know, this is really dangerous stuff. And it's because there are no actual community leaders in Facebook. Facebook is just a giant kind of platform where there are no rules and there are no real community leaders. So, that's a subset of the absentee ownership problem. But it is a massive problem. Then you have all the other problems, too--I mean, the monopolization issue it's just chewing apart--local media in particular but also just funding streams, financing streams for media; Amazon is crushing the book market. You know, there's just, there's just--it's just crazy what we're doing.


Russ Roberts: Well, I want to come back to that. Because I disagree with that part of the critique, but I'm very interested in the first parts. And we'll talk about all of them, I hope, if I remember. But, on the case of the extremist argument, which I think is really interesting. Now, it's not working on me. As it turns out. Because I don't use Facebook. I do use Google a lot; and I use Twitter. And I think Twitter is a little bit different: it's sort of a self-generated extremist problem--that people who want to get attention on Twitter, the louder, the better. And so, I do tend to see a lot of angry things on Twitter. But it hasn't--it's pushed me, I'd say, it hasn't pushed me at all, for other reasons. I've avoided those encouragements that you are worrying about. And, I think most people have. But, many haven't, obviously. I don't know how much of that extremism we can put at the feet of Google and Facebook for sort of that the algorithms are giving people really creepy things to look at. There's no way of--I don't know if that's true. Right? That's a supposition. Right? That to get--let me say it differently. There are a lot of things they do to keep my attention. To keep me using the product. They are not all sinister. But some of them might be.

Matt Stoller: Most of them are part-sinister. But let's start with the first premise: that it doesn't work on you. Because, I think it's absolutely the case that it doesn't drive everyone to become a Pizzagate extremist. I think fundamentally we are people who have some free will. We have some cognitive ability to control our own environment. However, let's take Twitter. Because one of the things that--these guys draw from gaming, from gaming--I don't what to call it, but basically from casinos and from video games--

Russ Roberts: Addictive things.

Matt Stoller: Addictive social, psychological tricks that work on everyone. So, as an example, when you see on your Twitter notification or on your Twitter screen, it says, 'You have 8 notifications that have mentioned you,' or 'You have 12 notifications,' that number is there to make you click on that.

Russ Roberts: I get so excited! I can't help it. That part does get me--I confess.

Matt Stoller: I mean, it does. So, that I know that when I sit down to Twitter, it may not be changing my opinions, although it puts a lot of things in front of me. And so, I'm just exposed to information that I wouldn't have necessarily been exposed to, anyway.

Russ Roberts: For sure.

Matt Stoller: And, yet, I get sucked in because of these notifications. Right? So, I might sit down and I might say, 'Oh, I wonder--I'll just take a quick look at Twitter.' And then half an hour has passed. And I didn't mean to spend half an hour doing this. It's just, I sort of got sucked in. And I can then pull myself out. But, it's very, very hard to actually--you are just kind of controlled by these things in some ways. There's a Pavlovian response that you have, and it's not going to turn you into a vegan. It might turn someone else into a vegan. But it's going to waste that half an hour. It's going to waste that hour. And you have no ability to say, 'I'd like a Twitter without those addictive properties.' You have no ability to say, 'I'd like a Facebook that doesn't generate these, you know, these types of algorithms.' So, just as a metaphor, for what these guys do--I mean, I think Facebook more so than Twitter. Twitter has a different set of problems. But, Facebook, interacting on Facebook, because they want you to--or Instagram--they want you to engage and they want you to stay addicted or stay engaging--it's kind of like having somebody in a bar, where you are talking to a friend, and there's a guy in the corner just constantly saying, 'Fight. Fight. Come on. Come on. I heard what he said about you. I heard what she said about you.' And it's like--it's not necessarily going to make you fight, right? But, like, having one person, a couple of people--and these algorithms improve. They get better. Having those people sort of in the bar, kind of encouraging this kind of conflict--it's not going to make fighting less likely.

Russ Roberts: That's for sure.

Matt Stoller: It can only make it more likely. So it's like--even, and it will work on the most peaceful person out there. They will be--they will still be incredibly peaceful but they will just be a little less peaceful. So it moves, it moves, kind of everyone. And so, that's where we are right now. We're in, just psychologically when we are dealing with these institutions, they are having this kind of massive impact on the things we see. And I was talking about the opioid sort of rehab problem; but Google governs a lot of the information that I get. Right? And Google is just choosing what information it thinks I will find useful. I don't have any--I don't know how they are doing that. They are governing what I see. And Facebook is governing the social interactions that I have with my friends on Facebook. Mark Zuckerberg is mediating that; these algorithms are mediating that. And I have no power over how that is mediated. And that is crazy. Because fundamentally those are not Mark Zuckerberg's friends. These are not Twitter's relationships. And the fact that they can choose how I have to interact with these people and I have no power over that--I can't build competitive systems on top of it--that's, I think, really worrisome for our society, for our markets, and for our democracy.


Russ Roberts: Well, I find it a little bit worrisome. And we'll talk about how worrisome it is. Maybe in a minute. But, let's ask the question which, I've said many times on this program, and I know you don't agree with it, which is: You don't have to use it. So, what's wrong with that argument? If you don't like the way Facebook connects you to your friends or the way it adds things to your timeline or your feed or whatever it's called--I'm not a good Facebook user: I'm a Twitter guy. I don't have to use it. And, in fact, those sort of compulsive--it's like, the guy in the bar; I can stop going to the bar, in real life. And I go less to Twitter now than I did 3 months ago because I don't like what it does to me. I don't like that guy on the corner saying, 'Fight. Fight. Fight.' So, I agree with you: there's some damaging things about it. Aren't we free to just say no to them?

Matt Stoller: You'd think you can--you don't have to use these products; you can just sort of just log off. You don't like it, just log off. And, in fact, that's not true. In fact, these institutions are mediating our society. You live in that society, they are tracking you on websites and services that you are not a part of. They are organizing the financing of the media systems that you read. You are not able to escape them. If you don't use Google, everyone you interact with does. If you don't use Facebook, you know, Facebook is still taking money from the newspapers that you read, so you have worse news. You are still living in a society where conversations about our elections are run on Facebook, so you are a part of these institutions--these institutions are governing you whether you use them or not--

Russ Roberts: Well, I'm a [?] by them--

Matt Stoller: Hang on, let me--what?

Russ Roberts: I'm affected by them. I certainly understand that.

Matt Stoller: Right, but let me give you a more granular story, which is the one about Mike Turk--but this is something I think a lot of parents are experiencing, and it is that, Mike's kid--he has a couple of them. One of them's a teenager who is using YouTube and kind of, their recommendation engine on YouTube is very good, and so his kid will just watch it all day. And YouTube has a bunch of creepy stuff, like really weird, sexualized, violent type of videos that are, just millions of kids are watching. So, Mike is like, 'Look, my kid watches video games; he watches YouTube. He hangs out in these online hangouts talking about video games; that is a problem.' And Mike just cut it off. He said, 'I'm going to block all of these sites so that--you are a normal kid.' And he did that, and his normal kid, who just has this addictive, who is addicted to the Internet, but when he's not allowed to use it, he's fine. Okay. So, Mike is like, he blocks the video games; he blocks the video game hangouts; but he can't block YouTube, for some reason. And he tries. But, it turns out that you can't block YouTube if someone is using Chrome. Chrome is a browser that's made by Google. So he's like, 'That's kind of crazy.' And Mike is from the telecom world: he's very good at tech, he's like an IT [Information Technology] guy; but he couldn't figure out how to actually block YouTube. So, he's like, 'I'm just going to delete Chrome off of my son's computer, and I'm going to put a different browser on; and then, I won't let him go to any Google sites. Because if he can go to, he can just download chrome; and he knows how to do this: he'll download Chrome and then he'll be able to watch YouTube again. So, he tried to do that. So, he deletes Chrome and he puts a different browser on. But then he finds out that his son has to use Google, because Google has to deal with his son's school system; and the homework is actually done on Google products. And this is true across--there are a lot of kids who--Google is handing out Chromebooks, and Chromebooks are really cheap. They are good computers but you can't actually block YouTube. So, what Google is doing is, they have vertically integrated not just the browser and YouTube and their search engine and Maps and a bunch of other products. They've actually vertically integrated into the school system. And they have more power over what Mike's kid sees than Mike does. Mike's a conservative Republican. He's not a Democrat; he's on your side of the fence.

Russ Roberts: I'm not a conservative Republican, Matt. Just for the record. But go ahead.

Matt Stoller: Oh, you're not. Okay. Then, you're a--okay, well, whatever. He's a conservative Republican and I'm not.

Russ Roberts: That's fine. Keep going.

Matt Stoller: But the point here is that, Mike can't actually structure the environment that he wants for his kid because of the power that Google has. So, that's--if you think that you don't have to use these, it might be true for you. But it's certainly not true for millions of parents and millions of kids who are being raised on this stuff and have no choice about it.


Russ Roberts: So, that's a great example. And I read that article, where you told that story. It's a great article; it's really interesting, and I wanted you to tell that story, so I'm glad you did. The question is: Let's stick with that for a minute. What's the lesson there? What should we do? So, that's outrageous, right? So, what should you do about that? What should we favor? What policy response do you think is appropriate in that situation?

Matt Stoller: Well, I think the first thing you have to do is you have to just admit that we have a really serious problem with the power of Google. That's the very first thing that we have to do. And, once we've admitted that, and that it's a political problem--it can only be addressed through politics--then there's a whole bunch of--

Russ Roberts: Wait a minute. Why is that? I mean, couldn't I--my first thought is the school doesn't have to require that relationship. That might not be a healthy thing to have a corporation and a school in cahoots like that. Might be[?] a bad idea.

Matt Stoller: But that's a political--that's a political statement, that you're making.

Russ Roberts: Yeah, sort of.

Matt Stoller: The school is a political institution.

Russ Roberts: Yeah. It is. It's a shame. But that's a truth. And we have limited control over what our schools do. And we have limited control over what Google does. So we're kind of at the mercy of both of them. So we have a kind of a dilemma here. Neither of them are responding potentially to what we want. What's the best way to get closer to what we want without some unintended consequences? And I would just add--and I think this is the real challenge: You said you have to first recognize that Google's too powerful or harmful. Most of us love it. And again--I'm very worried about some of the things you are raising: I think they are very concerning. But at the same time, I've got this great thing going where I buy a plane ticket and I get a confirmation that I bought the ticket via Gmail. And Google just is so smart. It's not really smart, but the algorithms are so well done--that it immediately puts that trip into my calendar. I don't have to deal with it. And I really like that. I'm not alone, obviously. There are many, many things Google does that don't do anything for me. And I think a lot of other things they do people like a lot that don't excite me positively. But, part of the problem of this is that a lot of people like what they do. They love it, actually. They don't just like it. They enjoy using it every day. And that Mike Turk's kid is not protected from some gross stuff on YouTube--they'd say, 'Well, that's Mike Turk's problem.' I think it might be everybody's problem, or somebody else's besides Mike, or his school's problem. There's a lot of stuff going on there. But, isn't that a big problem with your starting point, to get people worried about? I don't see what the worry is. They like all this stuff.

Matt Stoller: Well, I think people are--I think you are not giving people enough credit. People can both enjoy the ability to fly across the country while at the same time recognizing that, you know, they don't particularly love baggage fees, or they don't particularly think that American Airlines is doing a particularly good job with that technology. So, they can love--I love the ability to use Google or Amazon to find out all sorts of great things; and I use these products constantly. But I can separate out the technology from the political power. And I think you--I think people can say, 'Yeah, I really like using the search engine. But I also think that Google as a political institution has too much power.' People are mature enough to handle that. So, you know, schools are political institutions. And, thank God they are. I mean, school boards are elected in a lot of places. And, you know, you don't have to--you can say to Google, you know, 'You want to be in this school? You are not allowed to sell Chromebooks that prevent the blocking of YouTube.' Or, you can say, 'We're going to investigate you and figure out how the company works, figure out how these algorithms work. And then we'll make a bunch of decisions about the best way to fit your system into a democracy.' And I think we have to do that, because these are not simple questions. It's very important to not sort of shoot from the hip and say, you know, we need to, you know, chop you up this way, or chop you up that way. I mean, these are multinational companies. They have billions of users. They are very complicated institutions. And what we need to do first is we need to actually investigate and understand how they work. That's kind of the first thing that I would sort of call for. So, I think you have to acknowledge they are political problems. And I think people are ready to acknowledge that they are political problems. Not that we should eliminate them. Not that they are not great--that they don't have great technologies. But, I would also note that, the things you are talking about that people like about Google--you know, Google is very well liked. And Amazon is trusted. But Facebook is not so much. But Google and Amazon are really well-liked. But these are institutions--you have to disaggregate the technology and what the engineers have built--the search engine, the video service, the various, you know, Alexa, those kinds of things--from the financial holding company structure. So, Google is a financial--it's a search engine but it's also a financial holding company called Alphabet. And it has a bunch of different products in that holding company. And those products are largely the result of mergers and acquisitions. So, nobody would care if you say, 'Split off, say, YouTube and the Google Search Engines and Google Maps,' because those are all-- or DoubleClick. Because those are all actual--Google Search Engine was developed internally. But the others were all just sort of bought, and cobbled together, and then used in ways that are potentially anti-competitive. So, it's not that you are actually destroying the technology, destroying the things that people loved. In fact, you are liberating them. And that's kind of the way that I these things--is, you have to think about it in the context of what is the financial concentration, versus what is the necessary concentration to do the really great things that we all love? But, you also have to give credit to people to people who are mature enough to see that they really like, you know, they really like being able to do these cool things, whether it's flying or whether it's driving a car or whether it's using a search engine. And the political institutions, the financial structures that are actually managing those industrial arts.


Russ Roberts: Some of this, of course, it is surely under our choice, in the following sense. Let's take the search engine part. If I think that, or if a news report comes out, that Google is corrupting search findings to benefit, say, people who pay more, in certain ways, or disrupting our political system, and I think there's some evidence that's true--let's move to a different search engine. There are other search engines. There's DuckDuckGo; there's Bing. We could use those, right? I mean, we could have a campaign to stop using Google's search engine to reduce Google's power. Do you think that's a good idea?

Matt Stoller: Well, look. I mean, we've had a campaign to get people to stop using Google's search engine. It's called Bing. I mean, Microsoft has spent enormous amounts of money to try to get people to move from[?to?] Bing, and it hasn't worked at all. So, there's sort of this notion that competition is just a click away. But the actual evidence is that it's not. It just isn't.

Russ Roberts: Well, that's because people are just content with Google. Now, maybe they shouldn't be, because they don't know about these effects that you are talking about. Which I think is a very [?]

Matt Stoller: Well, we don't actually know that. I mean, we know that people are using Google. But, beyond that we can't make a supposition that people are just content with it and if there was a sort of, you know, different and better products, they wouldn't, you know, go to that. As just an example: You know, there was this big European Union case that Google, you know, was fined about $2.7 billion dollars for. And the origin of the case is with a competitive search engine called Foundem. And Foundem, which, I don't think it exists any more, but it used to be a shopping comparison engine, which means that--it was basically a search engine, but for prices. So, if you were to type in, say, a certain brand of bicycle, you would find, you know, if you went into Foundem and typed it out, it would give you prices from a bunch of different vendors, and you could pick the one that you wanted. And there were a bunch of these shopping-by-comparison engines. In, I guess it was around 2009, 2010, Google decided that this was spam. Or they decided they didn't like it. One of the two. And so what they did is, first they created their own shopping engine. So, you could compare on the Google shopping engine as well as Foundem. And a bunch of these others. But, then, they actually downgraded all of these shopping engines in their natural search algorithm into spam. So, you'd have to go through multiple pages when you searched to even find them. And this, of course, killed all of their traffic. And naturally Google didn't actually change, do that to their own shopping comparison engine. So, people would got to Google's shopping comparison engine. And they killed all their rivals. And all these guys went out of business. Now, here's what's interesting. What happened next is that Google then killed their own shopping comparison engine, and replaced it with an ad-driven engine. So, you would type in that bicycle; and now, you don't get different prices from across the web. You get a list of people of ads for people who have paid Google to show you whatever product they want to show you at whatever price they want to show you, when someone types in that, say, bicycle. And that is a very dangerous move. And it's something that--you know, people have no idea that it even happened. But I imagine that people, when they are looking for prices, for the best prices for things, would prefer to have a comparison engine that they can look at as opposed to things that are put in front of them by merchants that are paying to get there.


Russ Roberts: And, of course, I agree with that. Except, the question is: what do you want to do about that? And, there's two issues there. One is: What level of intervention do you think is appropriate to reduce the probability of that? The real puzzle, though, to me, as an economist, is: So, these things are happening. There's a lot of things like that, I think. I'm going to assume you are right about that story. I didn't know it. But it's plausible to me. And let's say it's true. So, why wouldn't an alternative search engine that exists, and that's really good, by the way--I don't use Bing. The reason I don't use Bing is I'm not in the habit of using it. But if I did use it, I bet it's pretty good. And DuckDuckGo is, I think, pretty good--maybe very good. Maybe they are both very good. I don't know. Hard to measure. But, let's say they are. Wouldn't they want to take out some ads to remind people that they don't do that? Now, of course, you don't know if that's true. But there is actual competition. It just doesn't seem to be manifesting itself in the way that it normally would in a more competitive market. I don't think it's because there's only 3 players. I'm not sure what the reason is.

Matt Stoller: Well, I have used DuckDuckGo. And it is--so, okay. Let's just talk about the search engine, how to build a search engine 10 years ago, 15 years ago. Or, when Google started, right? There were a bunch of search engines when Google started--Lycos and all these other things.

Russ Roberts: AltaVista, Yahoo, etc.

Matt Stoller: Yeah. And what these did was they mapped the web; and then they chose different ways of splitting it to you. And Google did the same thing. And so you'd think--well, it's a web; it's the web. You can map it today if you want to compete with Google. But, in effect, Google's search engine no longer works by mapping the web. Google is so powerful that it in fact in many ways structures the web based on the data that it has about everyone else. So, Google knows a lot about you. More about your web habits than you do. Or, more about my web habits than I do. And it knows--it has a lot of other information about me. And it uses that to improve its search engine that I then access. And Google has that ability because they have access to that data, because of--and this is what I'm told; I'm not a, I don't know anything about (AI) artificial intelligence, but I'm told that these big data sets are increasingly important to actually be able to produce relevant search queries, to be able to do a whole bunch of things like, for example, Google Translate, or--there's all of these services, and search is just one of them, that can only be perfected if you have mass quantities of data. DuckDuckGo just doesn't have that data. And, you know, it's not clear that Bing has that data. And so the quality of the result is bad. So, there effectively aren't competitors.

Russ Roberts: Well, I don't know, you say 'bad'--

Matt Stoller: Well, I've used DuckDuckGo; it just doesn't work for me. And I wanted to use it. But it just doesn't work. And people will tell you, if you want to create a search engine in today's Web, you need enormous quantities of data. And there are three companies--maybe a few more than that, but basically just three--that have that quantity of data; and it's Facebook, Google, and Amazon. If you don't have that data, you just can't play. Maybe Microsoft; maybe Apple.


Russ Roberts: So, before we get to the--I just have to make an historical remark. Which, you can respond to if you like. Of course, many, many, many times in the past we've been told that this company or that is going to have too much power over us. It used to be IBM [International Business Machines]. They're not important any more. Used to be Microsoft. They're not important any more. It used to Apple, with iTunes: 'They're going to control the music market.' And Spotify crushed--seems to be crushing--them; I don't know if they really are, but to my eye, ear, and the rest of my life, Spotify is dominating my iTunes experiences. So, why is this going to be different? Why is it that, in the past, when companies got too big or too powerful, they were destroyed by upstarts? Is that not going to happen this time, do you think?

Matt Stoller: Well, let's talk about IBM and Microsoft. And before that, I think you could point out a bunch of them: AT&T [American Telephone and Telegraph]. You know, IBM was a dominant market player really until the late 1960s and early 1970s. And it controlled--there was no software market. Because, you bought IBM hardware; IBM gave you software for free; and IBM pursued a range of tactics to prevent people to prevent people from actually using other companies' computers. And, what happened is, the Department of Justice, I guess it was the last day of the Lyndon Johnson Administration, filed an antitrust suit against them. It was actually, I think, the third antitrust suit: There was one in the 1930s; there was one in the 1950s; and then there was one in 1969-1982. And, that suit caused IBM--I mean, it was a crazy long suit [?]--

Russ Roberts: Yeah--

Matt Stoller: but it caused--but one of the things that it did was it made IBM act on its best behavior. So that IBM was worried. It was a very capable company, and they were able to crush rivals very quickly. But once the suit started, they got really worried that they were engaging in anti-competitive behavior. And so, one of the things that they did was they unbundled their hardware and their software. So, they said, 'You can buy the software separately.' And what this did, is it opened up the market for other software players to sell into IBM Machines. It created the American software industry. Before that, in the 1950s, the antitrust suit against IBM said--it did a number of things, but one of the things that it did is it said, 'You have to share your patents. And you have to open up what you are doing, the know-how, to other institutions, other companies.' This was true for IBM and AT&T and RCA [Radio Corporation of America]. And as a result, we had an open electronics industry; and in that decade, you had companies like Motorola, Texas Instruments, and a variety of others forming, using the patents that had been unlocked through these DOJ [Department of Justice] antitrust strategies. In the 1970s, the software industry was created. Also, another thing that happened was the first computer language--I think it was--not the first, but UNIX [not an acronym, but may have been a pun on Multics (Multiplexed Information and Computing System), contrasting a unique-user to a multi-user operating system--Econlib Ed.] was a computer language that was created, actually, by AT&T, but because they were constrained into the telephone market and they were not allowed to sell computer languages, so they just gave it away to universities. And that's what created sort of open-source software. Microsoft [MS] is another interesting example. Because, Microsoft had been planning to dominate the Internet, in 1995. I mean, Bill Gates had a famous all-company memo about the importance of the Internet. And so, one of the things that they did is they first tried to buy Intuit it, which made banking software and personal finance software, because that's--

Russ Roberts: That's a company name--Intuit.

Matt Stoller: Yeah. It's Quicken, Quicken Lending, Quicken Software. Because they thought that was going to be the backbone of the Internet. But they would have continued to buy companies if they had been able to--they would have would have continued to buy companies but the DOJ blocked them from buying into it, because they said 'We want the Internet to actually be open.' And that was actually a precursor to the antitrust suit. Which had the same effect on Microsoft that it had on IBM. And I just want to say one, actual--go back to IBM for a second. In 1980, I think it was in the early 1980s, must have been 1982 or 1981 or something like that; maybe it was the late 1970s. But, when IBM decided to put together their personal computer, they did it because Apple had come out with a personal computer. And one of the reasons Apple was able to do what they did with the Graphical User Interface [GUI] is because Xerox was under an antitrust investigation. So, they were willing to share a bunch of stuff with younger companies. But, IBM was under an antitrust investigation. So, they, when they built the Personal Computer, instead of controlling all the suppliers themselves, which they used to do, they said, 'Well, we will license, we will treat you differently. We will treat you well.' And they gave this contract for the operating system to Microsoft. They gave a contract for the microchip to Intel. And it created this kind of open ecosystem for the personal computer, which allowed all these companies like Dell and--I forget, but there were a lot of personal computer manufacturers. So, you had this open ecosystem, which was an explicit creation of DOJ policy. And then, in the late 1990s after Microsoft--and Microsoft was formed after that, and that's the reason it was able to be a massive company instead of just a vendor to IBM. And then, in the late 1990s, after the Intuit acquisition was blocked by the DOJ, after Microsoft tried to buy that, then there was the antitrust suit against Microsoft. And Microsoft stopped engaging in anticompetitve behavior. They were able to sort of strangle Netflix--not Netflix--Netscape. Although Netscape wasn't particularly managed well. But they weren't able to do the same thing to Google, and to Facebook, and to the sharing, you know, the sort of web--to point out. Because they had been, their culture had been changed by the DOJ suit, not to strangle anyone in the crib[?] And so--

Russ Roberts: Yeah... I find that story, Matt--some of that I [?]--

Matt Stoller: Well, what you find is that the revolutionary who becomes the king doesn't want the next revolutionaries to come and overthrow them. And so they will do everything they can to prevent that. And the only thing you can do--that American antitrust law--was written to stop that from happening. To make sure that the King couldn't keep other people out of the market. And it worked. Companies don't just fall because of a natural cycle in the market. They fall because of political choices to open up and oxygenate the market for other people. And let's just be clear about something. Microsoft is still a very profitable company. It still exists. And IBM still exists. These companies didn't die. They just had to make way for other innovators in the ecosystem.

Russ Roberts: Well, I, part of that story I think is true. I think it's true that they may have been on their best behavior. But, I think their failures to innovate were not due to the DOJ oversight. You can argue that their culture became less innovative. I find that a little bit of a stretch, given that monopolists tend to be less innovative to start with. So, that's a debatable--I don't want to debate it, but I'm just saying, I'm not convinced. It's an interesting argument. I think most of the failure of IBM and Microsoft were that--they had more nimble, smaller companies that came along and did a better job, new opportunities. There's more, perhaps, to the story.


Russ Roberts: The question is: I'm agnostic about whether the DOJ, the Department of Justice, has been important in that kind of thing. I'm open-minded about your story. The question is: What do they do now? Right? Now, we have this very strange thing, where these companies are large in a way--the part that's alarming to me is they are large in a way that isn't the way a company was large before. If a company was--like, General Motors [GM], had a big share of the auto market. And yes, you could still buy Ford. You could still buy Chrysler. Or many imports at the time. Now that market is much more open, and all those American companies are smaller. They are more nimble. They make much better cars. It's been a great thing, as that market's been opened up, mostly by foreign competition and some domestic attempts at doing better. But, it seems to me, that what makes Facebook different, or Google different, is: First of all, the network effects--that they need lots of people, and it's no fun to be on a network of friends that only has 7 people on it. And the second is that it's become an enormously important piece of the fabric of our lives, way beyond, say, my car. Or, the, say my--some item of clothing that might have some--in theory it doesn't but it could, monopoly power. This is--it's much harder to start a new Facebook. I'm open to that fact. Now what? And I would ask you--I'm going to give you two challenges. One: Tell me what policies we should pursue to make this problem that you are identifying better. And, second: Are there policies in place now that are making it harder for new startups to innovate in these areas?

Matt Stoller: Okay, so let's just start with--

Russ Roberts: That we should get rid of?

Matt Stoller: So, the first one: What should we do now? And, I--it's a big question. I think we need to do a lot of investigating. But, that's kind of--I don't want to make that a dodge. Because it sounds like a dodge, to be like: 'We need to know more.' But, we really do need to know more--

Russ Roberts: For sure--

Matt Stoller: We don't understand how these companies work. But I think a very simple thing that we could do is to block all acquisitions by Facebook, Amazon, and Google. I mean, just stop them from acquiring companies. Then you would see that they couldn't, they wouldn't be able to block the emergence of new competitors into the market. It would be much harder for them to do that. Because, right now, what Facebook does is they have--they have this--I think it was about 2011, they bought a company called Onavo[?], which is--it doesn't really matter what it is. It's just a bunch of phones. And people--it's like a malware tracker and a bunch of people downloaded it. And it essentially, enough people have downloaded it that it gives--it gives--Facebook can--Onavo can like see what you are doing on your phone. See how you use your phone. Track you in real time. And there's a statistically large enough people who have Onavo so that if you own the company you can see how the Web is being used. And so, what Facebook does with the data that it gets from Onavo is it basically looks and sees which apps are becoming popular. And how people are using them. So, it can spy on any new potential competitor. And what it does, is it keeps a database of new companies that are growing very quickly, and will buy them. Or copy them. Very quickly. And I'm not talking about within a year. I'm talking about, within a month. And, will see a new product that's come on the market, that's gained a bunch of users. And they will begin incorporating those products, design choices, into their existing products. And they will actually buy that company, if it's growing fast enough. So they've just bought one called TBH [To Be Honest], which is a social network for teens that was growing really quickly. They bought it after 3 months. They--so, first, you just got to say, 'No more of that. No more Onavo and no more buying companies. Because that way, people will actually start businesses and they will try to compete with these guys instead of just trying to sell out to them. And the other thing you can do is you can undo some of the recent acquisitions. So, you can undo the--maybe the Instagram acquisition or the WhatsApp acquisition. You can look back and undo the Amazon WholeFoods acquisition. Those are some things you could do that wouldn't be particularly difficult, and are not, like, you are not going to like mock up any really deep-seated sort of relationships. There's a bunch of other things you can do. There's a bunch of tests that are being performed. There are some regs[?] that are happening in Europe which are going to get implemented next year, they are privacy rules, we'll see how they work out. They might do good things. They might do bad things. But, basically, we need to be really studying these companies. We need to be comparing different political choices. And we need to be stopping acquisitions. And that's the law. The law exists to do that. It's called the Clayton Act. The FTC [Federal Trade Commission] and DOJ could just take a different enforcement approach and there we go. So, that's an answer to your first question. What was your second question? Or do you want to--

Russ Roberts: Let's stick with that. My second question was, are there are in things that are in place now that are making it harder for competitors. I've read that it's hard for people to share their Facebook information with a would-be Facebook competitor that makes it harder for--and that government enforces that. And that not a good thing.


Russ Roberts: But, let's stick with the first thing. Let's stick with this first point. And I want to ask you about you about one other thing, too. Which is--I want to come back to the media issue and the freedom of the press, 1st Amendment issue, because I think it's very important. But I'm really kind of glad--aren't you kind of glad Amazon bought Whole Foods? Does that bother you? Does that make you nervous? Or is it, just sort of, you think, they are big enough; let's not let them get bigger? I mean, I'm not even sure Amazon's going to make money. I know this is a crazy claim. We've talked about it here on the program before. Almost all their profit comes from their cloud services. Their retail business is limping along, barely making a profit.

Matt Stoller: Well, no. That's not exactly true. They aren't making a GAAP [Generally Accepted Accounting Principles]--profit based on accounting metrics. But they generate massive free cash flow. They spend, something like, whatever, something like $20 billion dollars a year on research. They could just turn on the spigot whenever they want. Amazon is a massively profitable company. It's why--you know, Jeff Bezos is not worth $100 billion dollars, you know--you are not worth $100 billion dollars because Amazon's a charity. And that's not a--

Russ Roberts: No, no; but I don't think they make--I don't think the retail side of their activity, which is the books, the clothes, the gardening tools, and everything else that they now have on their site--I don't think they make much money on that. If any. Barely. They are barely profitable.

Matt Stoller: Of course they do. I mean, of course they do.

Russ Roberts: Why is that?

Matt Stoller: I mean, enormous amounts of money.

Russ Roberts: I don't think so.

Matt Stoller: Well, I mean. Okay. Well, then, you know, then Amazon--then Jeff Bezos's hundred billion fortune isn't real. I mean, they make a lot of money--

Russ Roberts: It's not. Well, it's based on the stock that he holds. And it's based on the projection that people think it's going to make a lot of money at some point. Or, that their, other parts of their business will make a lot of money.

Matt Stoller: All right. All right. Well, then it's a charity. I mean, what do you want me to say? I--Like, they spend, they make huge capital?--

Russ Roberts: What I want you to say is I'm not sure their retail position is a real threat to American retailing outside--it looks like it is, because it might not make it.

Matt Stoller: That's not true. I mean, they are making a lot of money. Like, they are not, they are not showing accounting profits because they don't want to. But, they, they--you know, Jeff Bezos is very clear. He says, 'Anything we make, we invest back in.' They are spending a ton of money on expanding operations. And at any point they could just choose to spend a little bit less and show a profit. And if they want to just show a big profit, they could. But, Amazon--you know, Jeff Bezos understands that he's building a long-term monopoly, and that's his play. And this idea--he can generate cash whenever he wants to. He just has chosen not to.

Russ Roberts: But that's--my claim is that's coming out of his--cloud services--which are extremely successful. And we use them all the time, indirectly and directly--that's a separate issue to be worried about, too. But I'm just thinking about Whole Foods or bookstores, or--this is one of the greatest times in human history--not 'one of'--this is the greatest time in human history to be a consumer. You could argue that's a waste of money. Many times it is. Or, that we get seduced by the material world. I think that's also true, if you are not careful. But, our opportunity to buy clothing, books, everything, at very low prices is gloriously good. I don't see anything alarming about Amazon as a retailer. So, what are you worried about?

Matt Stoller: But that's because you are thinking about--you are not thinking about us, as consumers. But you can't think about people as just consumers. You know, we are citizens, and we make things. And if you look at, say--if you look at authors, right, you think it is a great time to have books, because you can go on Amazon and buy a ton of books. Right? And you can buy them, and on a Kindle, and download instantly; and 'Wow! That's amazing.'

Russ Roberts: It is.

Matt Stoller: Well, you know what? The average income of authors has dropped by 25% since Amazon came out with the Kindle. And probably more by this time. That's a few years old. Now, you might not care, because you might not be an author--

Russ Roberts: But I am an author, Matt. I'm an author. And I love it.

Matt Stoller: But hang on a second. Let me finish. If you're an author--and there are people that people used to make their living writing books, mid-level books. Or you are a band[?], right? Bands, you know, are getting just savaged, it's more by YouTube more than by any of the others. What's happened is this whole mid-tier of the artistic, creative community whose livelihoods have just been shattered. And there are large numbers of books that have not been written because there's no money in it any more. And that's a massive loss to the free flow of ideas in America. Beyond that, Amazon can choose, and does choose which books to put in front of you. So, they had a big fight with Hachette, where they just pulled Hachette's books off of their shelves. And, if you look for certain books, they will choose to promote other books in front of those books, through their recommendation engine. And so, they are manipulating the flow of information to you. They are manipulating the flow of ideas from author to reader in a way that we have never seen before. And that is incredibly dangerous. So, sure: If you are just looking at low consumer prices and your ability to just acquire books for cheap prices, you might say, 'What a great time to be alive!' Even though I don't think that Amazon's prices are necessarily that good. I think it's undeniable that they, as a consumer, it's an amazing platform. There are all sorts of aspects about who we are as a people, as a free people, as a creative people, as a people that have ideas, as a people that bring crops to market, that are incredibly disturbing. Amazon is not something that you want to see if you want to have a free society, if you want to have a democracy, if you want to have citizens who have any dignity. Sure--

Russ Roberts: Whoa, whoa, whoa. Hang on, hang on, hang on. I disagree with almost everything you just said. That's okay. That last piece, I don't understand at all. My dignity as a citizen is being impaired by the fact that Amazon has lots of books for sale that are relatively cheap?

Matt Stoller: No, it's being impaired by the fact that Amazon now controls, and manipulates, the flow of information between the book-reading public and the book-writing community. And that's just undeniable. That's what they do. They choose who gets paid. They choose who doesn't get paid. They choose who succeeds. They choose who don't succeed. And they choose what gets read. And yeah, it's not total control. But it's a vast amount of control that we've never seen before. And Amazon, by the way, is not only the biggest book seller in the country, by far--and they sell all, basically all e-books go through Amazon, and most of the non-e-books go through Amazon--but they are now also the biggest publisher in the country. Which means, if you want to--they can manipulate their platform to move books on their own publishing platform above and beyond books that are published by other publishers. So, there's a whole set of conflicts of interests that are baked into here. And there's an enormous amount of power that Amazon can use to sell what they want, and promote the ideas that they want, and not promote the ideas that they don't want to promote.


Russ Roberts: So, this is happening in a lot of other places, of course. Not just at Amazon. It's happening at Netflix, where Netflix makes their own movies that compete with the movies that subscribe to on their service. And yet--and this is where I disagree with you, and you can respond to it--and yet, I think it's the greatest time in human history to be either a creator or a consumer. That mid-tier, hollowing out you are worrying about--I don't get it. I bet there are more--I don't even have [?] the data; so I'm open to being wrong. I challenge--I don't expect you to have it at your fingertips. Perhaps some of our listeners can find it. More books get published; there's more music than ever before; the quality of entertainment--television and movies--is off-the-charts better than it was 20 years ago, and 50 years ago. It's mind-bogglingly better. There's an enormous amount of opportunity to explore things. And I have many other sources besides Amazon where I find out about books I want to read. I can go to Goodreads. I can read the New York Times Book Review. And, compared to 25 years ago when people were stuck with a bookstore that had maybe 3000--and that would be a lot--copies, 3000 different books in stock; and then we finally got the Barnes and Noble that came along, and Borders that expanded that to a much bigger number. But it's dwarfed, dwarfed by Amazon. So, the amount of opportunity people have to learn and explore, it's just unparalleled. So, I don't see this sort of desert or drying up of citizen access to information. Now, I am worried about the media news. I'll come to that. But, do you want to respond to that point about creation, creative folks, in music and the arts?

Matt Stoller: Well, yeah. I'll just say--the data, I got it, it was that data on income is from the Authors Guild. So, there's an author, T. J. Stiles, who has won a National Book Award, a Pulitzer Prize--actually multiple Pulitzer Prizes. He's, you know, written books on--he mostly writes about history. So, he's written--he used to be an editor--so, he wrote about Jesse James--Jesse James: Last Rebel of the Civil War; The First Tycoon: The Epic Life of Cornelius Vanderbilt; a book about Custer. And he kind of gave a speech; and he just pointed out that to actually write--he spends about 4-5 years writing each of his books. And his books are just great; and they change the way that we understand our culture. And he just said, he's like, 'Look. Look. These are expensive books to write. It takes 4-5 years of research and writing to actually do this.' And, the way that Amazon is structuring the book market makes it much, much harder for people to actually write books like this. Now, he's fine, because his books are very popular; but he's noted that if were just starting out, it would be much, much, much harder for him to actually do that. What you find in the book market itself--I mean, the 25% drop in income is for the median book writer--if you are trying to get into the book market now, it's just much harder if you are an author. And you may not care because you might think, 'Well, I have plenty of choices for books.' But, a book isn't--having, writing a book about the life of Cornelius Vanderbilt and the construction of corporate America that's fun to read and compelling and well-sourced--the amount of time and effort that goes into making a book like that is very different than, you know, a book that, you know, George Bush used to read which was called the Fart Book--it was just like a book of fart jokes. It's like--those are both books, and you can say, 'Well, now we have access to more books than we've ever had before.' But, the question isn't whether we have access to more books. The question is whether we have access to a good set of diverse ideas that are compelling and effective--essential for democracy. And I don't think that we actually do. I think the number of ideas that we have access to is drying up. And it's masked by this illusion of choice, which is really what you're pointing to: Look at all this choice.

Russ Roberts: Why is that an illusion? Anybody can go to Medium [] right now and write fantastic, long essays; share them with lots of people. Anybody can start a blog.

Matt Stoller: But, again, it's like--what you are saying is that the Fart Book is a book, so it's the same as, you know, Shakespeare. Right? They are not the same product. Like, I can write--

Russ Roberts: I don't think that's what I am saying--

Matt Stoller: But it is. You are saying you can go to Medium and write a wonderful essay. What about going to Iraq and spending a month there and trying to understand the culture and what the conflict has done and actually putting something in writing around that? Those are both very wonderful--

Russ Roberts: You think that's harder to do now than it was 25 years ago?

Matt Stoller: Well, I'm just saying that they're different--I just think that they are different products.

Russ Roberts: Well, of course they are.

Matt Stoller: And one of them takes massive amounts of financing, and one of them takes a smart person sitting, like writing for an afternoon or two afternoons. And you can look at them and say, 'Wow. There's all this great choice.' But, in fact: No. You don't have a great choice. You used to have a whole bunch of foreign bureaus, and they would do a lot of different reporting. And now you have a bunch of different TV channels [television channels], but no foreign bureaus. So, are you really getting more and better foreign coverage just because you can use your TV clicker or look on the Internet and find lots of different people talking about foreign stuff? Of course not. You are getting worse information. It just looks like you have more of a choice. It's like the beer market, right? Where people say, 'Well, look at all of these choices of beers,' when in fact most of all the beer that you buy, or most of the toothpaste you buy, or most of the consumer products that you buy, even though you think you have this great illusion, all these great choices, are in fact made by a small number of conglomerates. You are in effect buying, when you drink a beer--

Russ Roberts: Matt--

Matt Stoller: when you drink a beer you are effectively buying from ABMBev [?Anheuser-Busch/Miller beverages?], even if you think you are buying from some sort of small--

Russ Roberts: Matt--that's a tough case to make. Yes, it's true that Anheuser-Busch and others own a lot of the brands. But, again, this is the greatest time in America to drink craft beer made by hundreds of independent producers around the country.

Matt Stoller: I mean, look, you can say that. But I talk--

Russ Roberts: It's true. I can say it--

Matt Stoller: I talk to independent craft brewers, and what is happening now is you've got this--this massive private equity and conglomerate, beer conglomerate, buy-out media going on. You have ABMBev, which has also kind of concentrated its power in the production of hops, choosing to deny hops to companies that it wants to deny hops to. They have enormous control over distribution. And so you might way, 'Well, this is the greatest time to be alive,' because there's all of these sort of choices. In fact, you don't really have as many choices as you think. And what about the rights of all of those people that want to brew beer, or all those people that want to write books or want to create or want to produce things or want to bring their crops to market? This matters. This is the essence of America, the ability to make things, to tinker, to create--

Russ Roberts: Yeah! I agree--

Matt Stoller: And you are saying that that doesn't matter.

Russ Roberts: No, I--

Matt Stoller: As long as I get cheap, free stuff--

Russ Roberts: What?--

Matt Stoller: And I can pretend like I have this big set of choices--

Russ Roberts: No, I think it's--

Matt Stoller: I think it's fine. But that matters. And you can't discount the fact that those, the rights of those people are being violated. They are--

Russ Roberts: Their rights are being violated?

Matt Stoller: in open and fair markets.


Russ Roberts: Well, I disagree with you totally. But, the way I disagree with you is, I happen to agree with what you said a minute ago, that you said I didn't agree. I agree with you that the ability to bring your talents to the marketplace is what makes America great. I just think those opportunities are extraordinary right now. I think you can write a great book, you can build a great beer, you can build an incredible new line of clothing--they are all over the place. There are all these incredible small companies creating shoes and shirts, and all kinds of things are going on.

Matt Stoller: The rate of small business formation is at a 50-year low.

Russ Roberts: I know. I know. But that doesn't disprove my point.

Matt Stoller: Well, no, but you are not actually--you are just saying these things like it's your impression that these things are true, but in fact there's no data.

Russ Roberts: Oh yeah? Are you telling me, that if I want to, say, learn about--oh, let's say, Chinese history, Russian history, American history, the Civil War--that I don't have more access? That a schoolchild today doesn't have more access than they had, say, 25 years ago, to all kinds of new and fantastic books?

Matt Stoller: Well, wait a second. You are just, you are just running in circles. Right? So, first you said that people have more access to beer and to a whole bunch of--

Russ Roberts: They do--

Matt Stoller: But that's not true! That's just not true. There is less access to--like there are many places you go where you think you are buying independent craft brew and you're not. You think--you are saying--you support open and competitive markets. But Amazon's not a market. Right? Go try to sell on Amazon. Like, see what happens. I mean, I can tell you that Amazon--if your products are selling on Amazon and they are doing well, then Amazon will simply downgrade--like, they will simply start selling it themselves. And when people search for it, Amazon will place their own product ahead of yours.

Russ Roberts: Well, when your book comes out, a monopoly power, on the history of monopoly--

Matt Stoller: No, but I'm saying that's--

Russ Roberts: They're not going to write another book next to it and underprice it. Your book is going to stand on its merit. It's going to get reviewed in good places, presumably, if it's well done. Which I presume it will be. And you will sell either a lot, if it's really good. Or maybe not so many And nothing is going to stop you from doing it.

Matt Stoller: In fact, that's not--in fact that's not true.

Russ Roberts: No?

Matt Stoller: Amazon has enormous power over--first of all, they can decide whether they want to show--you know, whether they want to promote the Kindle version, the hardback version, the paperback version. And they can also decide whether they want to promote used books. So, the used-book version of mine or not. And how the interface looks has a lot of difference. It depends heavily on what flow of money actually moves to the publisher. So, they have a lot of choices. Design choices. They also have choices over copyright and a whole series of other things. So, they can structure--not just whether people just look for my book, but they can actually structure how my book is sold--I mean, I don't have a book out. But, eventually--

Russ Roberts: When you do--

Matt Stoller: Eventually I'll have a book out. But they can make a lot of choices about how those, how books are sold. And also make a lot of choices about how easy it is to find that book, if you are browsing? And as you know--

Russ Roberts: Don't they want me to find it?

Matt Stoller: No!

Russ Roberts: They don't want me to find your book?

Matt Stoller: No.

Russ Roberts: They don't want to sell me your book?

Matt Stoller: No. They don't care if you find my book or not. They just want you to find a book.

Russ Roberts: Yeah, that's true--

Matt Stoller: Or, they want you to watch, you know, Amazon Prime, or whatever. They don't care or whether you are finding the right book for you, or the relationship that you and I have. Just like Facebook doesn't care whether clicking on the thing you're clicking on is good for you, makes you angry, or confuses you, or is from a Russian bot. As long as you are clicking, you are fine. Right? So, these are not markets. These are not well-structured, open and fair institutions--

Russ Roberts: Yeah, I agree--

Matt Stoller: These are controlled intermediaries who have placed themselves between American citizens and our ability to actually interact with each other freely and fairly. We need to[?] rebel against them the way we rebelled against the British East Indies Company. There we go.


Russ Roberts: Mmmm--I'll just say one more thing and then I'll let you have the last word. Which is, this part of what we're talking about is a political question of how best to deal with the fact that some companies have gotten quite large. But, we got off onto a different question--in ways, maybe more interesting--which is: How has this transformation of, say, retailing affected our actual choice, as opposed to our apparent choice? And, what has it done to the livelihood and opportunities for people to be creative? And, I want to make the claim that I think a musician today, an author today, a farmer today, a brewer today, a baker today has enormous opportunities that they didn't have 25 and 50 years ago. For all kinds of reasons, not just because of the Internet, not just because of Google, and not just because of Amazon. But I just think your--the glass is half full for me. Maybe a little more, a lot more than half full, as both a consumer and a creator. And I think other institutions will come along, and have come along, to help deal with this. We see all kinds of ways that--for example, it is expensive to spend 5 years writing a book; and we have more think tanks now--like, you and I are both at a think tank--that gives us the luxury of having, not having to make a living working in a boiler room, as William Faulkner did. And we can work on the books we want to work on. And I think it's a great time for that. But, maybe I'm overly optimistic. I'll give you the last word.

Matt Stoller: I'll say this; I'll finish off with this. I think it could be an amazing time for entrepreneurship. And these technologies are just unbelievable. And they can allow us to create the most wonderful, creative, free society where we can compete and rise and fall on our own merits. But, because of the way that--and we can make that choice to make that happen. But right now, because of the political choices that we've made, the way we structure antitrust, among other things but, the way we see political economy, we have allowed these unbelievable technologies that are kind of, we were gifted, these arts that we were gifted--we have turned them into basically mechanisms to get us to look at ads. And so, somebody at Facebook said that the greatest minds of our generation are trying really hard to figure out how to get people to click on more ads. And that is such a tragedy. Because, imagine what we could do with this amazing technology if the incentives were oriented around political liberty and freedom instead of capturing power over other people.

Comments and Sharing

TWITTER: Follow Russ Roberts @EconTalker

COMMENTS (91 to date)
Billy writes:

[Comment removed. Please consult our comment policies and check your email for explanation.--Econlib Ed.]

Nicholas Georgakopoulos writes:

[Comment removed. Please consult our comment policies and check your email for explanation.--Econlib Ed.]

Robert L Kennedy writes:

I agree with Nicholas about IBM. I've been around IBM for over 40 years. When IBM saw Apple doing well with the Apple II, they felt that they needed to respond and respond quickly. They also understood their limitations in succeeding with consumer products and explicitly decided to engage with Microsoft & Intel & others to get a product to market quickly Nothing to do with reacting to the DoJ. Further market dynamics: With the mainframe in the early years, IBM explicitly stayed away from the software side of the business in order to build an infrastructure of other suppliers. Eventually, some of those suppliers concluded that they could charge premium prices, particularly with ongoing support fees. Customers started to migrate away from the mainframe, for lots of reasons, including the premium support fees of the "ISVs", independent software vendors. IBM then responded by entering into the software business with solutions at lower support fees in order to protect their hardware revenue stream. just regular market dynamics.

Years later, IBM decided that they were not very good at selling solutions to consumers, particularly those that had become commodities like PCs. so they sold that business to Lenovo. again, normal market dynamics.

And the guest is wrong about Amazon, as Russ pointed out. Amazon makes most of their profits from their web services business and their retail business is mostly a money loser. Amazon has the edge on web services right now but there a lot of others vendors trying hard to catchup: IBM, Microsoft, Google, Oracle, etc. if Amazon loses their dominance in web services, lots of things will inherently change in their retail business. All normal market dynamics. I would liked for Russ to have pushed that point further. The guest revealed his own tunnel vision during that discussion

My gut tells me that Facebook will not inherently dominate going forward. My daughter tells me that neither her or any of her friends use Facebook. "That is for adults". yeah, they use FB properties like Instagram. Not sure if the primary FB platform will inherently always have the network advantages as it does now.

Finally, I suspect that Google is a different situation. They have their tentacles in a lot of things that are hard to avoid: Search, Youtube, Gmail, Google Docs, Android, etc. That said, i'm personally confident that if they abuse their position too much, other firms will come in and consumers will have other ways to manage their online lives. i would have liked the discussion to have spent more time on this topic.

All that said, it has been a long time since Russ has had a guest that he inherently had a different position. My sense was that his debating chops were a little rusty. But I greatly enjoyed the provocative discussion and would like more of the same.

Darren writes:

[Comment removed. Please consult our comment policies and check your email for explanation.--Econlib Ed.]

Hudson writes:

[Comment removed. Please consult our comment policies and check your email for explanation.--Econlib Ed.]

Ryan Waggoner writes:

never comment but I was compelled to after this episode.

I often find the more progressively-minded guests to have interesting and challenging arguments that I enjoy grappling with. But I found Matt’s arguments in favor of his cause to be completely unconvincing. The logic was poor, but worse, I felt that he tried to make his arguments stick by proclaiming them forcefully, instead of by unassailable logic. For example, two specific arguments that he was most upset about that I felt were very weak:

1. The big beer companies own lots of craft beer brands (but not all), therefore we actually have less choice than 25 years ago?

2. Median author income is down (a stat helpfully provided by an author advocacy group, but let’s take it as provided), therefor author opportunities are drying up. Sort of compelling, unless you stop to ask whether the number of authors has grown dramatically over the past few decades (it has).

There are much better arguments to be made for the anti-monopolist perspective, but they weren’t provided here.

Michae Byrnes writes:

Wow, this was a strange one...

I don't think the guest was adequately prepared for some of Russ' challenges, which hae [the guest] handled poorly, detracting from the discussion and causing interesting topics to be missed. I usually like it better when Russ interviews guests that aren't aligned with his views than when he has a guest he's closely aligned with, because it usually makes for excellent discussion, but this particular inteview seemed to go off the rails midwat through.

Just a couple of points:

1. At one point, Russ unintentionally made the Stoller's point, but he failed to pick up on it. In discussing alternative sources of book recommendations, Russ suggested that he wasn't that dependent on Amazon recommendations because, among other reasons, he could get recommendations from Goodreads instead. But Goodreads is (now) an Amazon subsidiary.

2. The discussion on craft beer had the potential to be interesting, but the guest missed the point. Russ was unquestionally right here - now is absolutely the very best time to be a craft beer lover. Where I live (northern suburn of Boston), there are 4 small independent breweries within a 7 mile radius of my house. I should say, four that I know about, there are probably more. They aren't (and almost certainly won't ever be) owned by InBev. We're actually living in a craft beer rennaisance. In the past year, I've toured craft breweries around Boston, Portsmouth, NH, Portland, ME, and even Vancouver, BC. The number of craft breweries (and the varieties of craft beer) available today is unprecedented.

The guest did have a point, in that InBev and other big producers have bought and do own various microbreweries. But these are usually bigger scale microbreweries. I feel like the small local scene is where craft brewing thrives - my guess is that the proliferation of tiny local breweries is putting more of a sqeeze on the more traditional microbreweries than is industry consolidation, but I don't really know.

Robert Kennedy writes:

Michea made the main point about craft beer that i intended to make earlier Small independent craft brewers are absolutely thriving right now in many part of the states. One big reason is that the states have loosened up their post depression 3 tier rules that prevented any company from doing more than brewing or distributing or selling.

That said, the guest is correct that some brands are presented as craft beers but are either the creation of a larger brewery (Blue Moon) or an acquisition by a large brewery (Red Hook, Kona, Goose Island, etc) or heavily invested in by a large brewery (Ballast Point) but that is more the exception than the rule,

Large breweries (or any large company as far as that goes) will always try squelch competition from small suppliers however they can. They try to make harder for smaller breweries to distribute or get tap space in bars or whatever. And they have made it harder, for sure. But so what? Consumers have proven that they want the product so the breweries find a way to get it to them Neither InBev nor SABMiller nor Heineken have the market power to prevent consumers & independent brewers from finding each other.

Dave Westburg writes:

Matt is way wrong in talking about the brewery business. More breweries and more independent brewers now than ever. See this Jan 2016 Washington Post Article.

Attila Balázs writes:

Thank you, this was a great episode. A very insightful debate, with points well made on both sides (though my bias leads me to believe that Russ doesn't really appreciate the degree to which the "internet" is centralized).

I really hope that Matt will be back and they will hash the argument out a little more.

Merry Christmas and Happy new year everyone!

Nicholas Georgakopoulos writes:

Despite the removal for rudeness of my original comment with my substantiated and very negative reactions to this guest as being far below the standards of this exceptionally good podcast, let me add some thoughts about the proper role of government.

Granted, offshoots of political theory or philosophy exist that argue that a role of the state is to induce more public participation (civic republicanism; have discourse and vote, not rational apathy to voting) or the cultivation of the citizentry (perfectionism; have the state subsidize libraries rather than monster truck rallies). Those can be sensible counters to unfettered free-marketeerism.

This guest's approach to the role of government is much broader than that. Effectively, on mere suspicion that any market player, who has the largest fraction of a market, engages in what the guest sees as manipulation of citizens' choices that are even very remotely related to political choices, the antitrust enforcers should prevent this competitor from growing by acquisitions and force this competitor to open their processes to government inspection. (E.g., Google/Amazon's recommendations influence our choices of browsing/books, our choices of browsing/books influence our political thinking, therefore prohibit Google/Amazon from acquisitions and investigate their algorithms.) Russ repeatedly tried to indicate that market competition should address any disapproval by the citizens; that Google's dominance is a result of people's approval of Google's algorithms compared to the ones of its competitors. Allow me to add, moreover, that the role of government is not to influence public discourse. Public discourse and its oversight belongs to the citizenry, not government. The opposite view leads to totalitarianism. Granted, various factions complain that the right or the left seek to influence discourse unfairly but that is the language and craft of politics. Returning to a Lockean view of what would exist in an organized primitive society, that is not that every citizen should have equal ability to influence every other one. The citizens who have broader networks, who are seen as wiser, will naturally have greater influence than others (perhaps that is why primitive societies tend not to be democracies). The undesirable type of populism is that which enables the lower-denominator crowd to override the influence of the wise. Democracy can be subverted. Its oversight should guard not against the able becoming too influential but against the less able coalescing to become more influential than the able. I am less concerned about political donations by the wealthy and more concerned about the political influence of the underclass.

That having been said, opaque algorithms can have undesirable effects. I agree to some degree with Cathy O'Neil that some algorithms need to be audited. I am very far, however, from feeling that algorithms that are neither used by the government nor for the government should be audited. The algorithms of Google and Amazon are audited by their competitors, who can advertise. Algorithms used by parole boards or for government procurement should be subject to audit.

Jeff writes:

I have been listening to Russ from the start of his podcast career and I think this is the first time I was almost speechless after the interview.... Bad time to start a business? You can't find good local beer? A well writen book is going to fail and not get to market as well as it might have 25 years ago? We should have the government investigate the companies so we can figure out how and what they are doing? " I tried the other search engine but it was not any good because they don't have google's data"?.... For the record my son's school does all of their work on an Ipad, they use google Docs, YouTube is blocked on the Ipad and it is the best learning experience I could imagine. Additionally, if your child has grown up spending all day on the internet and on YouTube that is a parenting decision you have made. If over time that decision is proven to be a bad one that is not google's fault....

Nathan writes:

[Comment removed for supplying false email address. Email the to request restoring your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Econlib Editor Note to our commenters:

We have had to reject an unusually large number of comments in this thread for rules violations.

Please keep in mind when you submit comments our longstanding comment civility policies, which explicitly state:

Comments are required to address the substance and content of the original posts or podcast episodes

and also restate for emphasis:
Comments that focus on personal characteristics of individuals... are usually not focused on the substantive economic points of the discussion, and thus are generally disallowed. A good rule of thumb is: Discuss or critique the argument, not the person making the argument.

If you submit a comment to EconTalk that focuses on talking about an individual rather than on discussing the substance and content of the podcast episode, likely your comment will be rejected. We expect comments to focus first and foremost on the substance and content of the podcast episode.

If you would like to suggest to Russ a different guest, you can email him at
If you think a guest had an incomprehensible accent or was rude to Russ, you can also email Russ there. The EconTalk comment section, though, is not a good place for you to air your views or frustrations about meta-issues or gripes about a guest.

Note also that if you are rude or crude toward one of our guests--so, if you badmouth one of our guests, or Russ, or one of our other commenters--your comment will not be published. That usually goes without saying--but perhaps it occasionally bears reiterating. It's fine to disagree with Russ or the guest speakers in a civil manner. Many commenters do that. What's not acceptable is is to badmouth or denigrate others in the process.

It is never our goal to reject comments on EconTalk. We know you spend time composing your comments. However, when commenters write many comments in a single thread, some citing the others, none of which accord with even our basic published civility policies, we can't even consider publishing those comments. So: If someone submits a comment that says, "Hey, that guy was weird," or "She was irrational," or "That guy's argument was stupid," or "What a jerk!" or "He's just another typical left-winger," or "He's just another typical right-winger;" or worse, a characterization of the guest that uses crude language or four-letter words, or comments on the person's appearance or demeanor--or, if someone writes comments that are fundamentally rude or ad hominem or badmouthing of others--and then someone else chimes in saying, "Yeah! I agree! That person pissed me off, too," those comments will not be considered or published on EconTalk. We expect you--our commenters--to focus on the content, not the individual.

We have had an unusually large number of comments submitted to this thread that have violated multiple comment civility policies. Most of those violations involve the commenter's talking about a person--the guest or someone else--rather than the actual substance and content of the argument discussed. Some comments have been submitted that are overtly rude, crass, inappropriate, irrelevant, or multiply in violation of our published comment rules.

Most--95%--of our commenters or more have no problem getting their comments published on EconTalk and EconLog. All you have to do is talk about the substance and content. Are you in the 95% whose comments are always accepted? Or are your comments in the often-rejected 5%?

Lauren Landsburg
Econlib Editor

szymon moldenhawer writes:

[Comment removed. Please consult our comment policies and check your email for explanation.--Econlib Ed.]

stuart writes:

[Comment removed. Please consult our comment policies and check your email for explanation.--Econlib Ed.]

Blake McDonald writes:

We have far more choices today than in the past and this is why there isn't as much profit in these things for most people. This is due to far more competition and choices of books, music, and beer. You can still make money writing a good book but since there are so many more choices unless you market the book well you would probably think the things are worse. Our problem is actually the opposite of what Matt says, we have too many choices so good things can get lost in the chaos.

Since Amazon, Google, and Facebook market products you must also market your products if you wish to be competitive and successful. It seems like Matt is against companies marketing other people's products.

As for search engines, Bing is nearly as good as Google however most people are used to Google and have Android phones. DuckDuckGo shouldn't be compared to Bing or Google since it doesn't track users activity. This is why it tends to return the best matching results but not the most common results that users visit. There are some things you can only find on DuckDuckGo.

Jeremy writes:

Below are a few comments.

For the Youtube blocking issue, there are options to restrict websites / domains at the router or OS level via proxy config. This is a fairly easy (and technical) solution which doesn’t require government intervention.

What about improving competition by eliminating the patent office? The established firms are stacking up patents on all types of abstract ideas in what is basically a legal ‘arms race’ against innovation and new entry.

Facebook has spent a huge amount of resources integrating Instagram. It would be very destructive to just mandate a separation after all these years. By the time the separation was complete, the market will (most likely) have moved to a new photo/social/etc platform.

On Amazon or any firm profitability, the best way to understand the business is reading the annual report and listening to the quarterly earnings calls with analyst Q&A. There are plenty of technology companies with little or no actual profit (Tesla, Amazon, Twitter for almost a decade) who are surviving on investor expectations of future growth.

The dynamic technology marketplace is moving too fast for regulation. Amazon is facing stiff competition from Walmart/Overstock/and others on the retail side. On the enterprise cloud side, Microsoft, IBM, and Google have dramatically improved offerings to the point where some companies are migrating off Amazon Web Services. Microsoft Office 365 has taken a good swipe at Google Docs. IBM Watson Cloud APIs have really challanged the speech-to-text, vision recognition, and natural language processing (NLP) players. Open source data solutions have taken on Oracle and Teradata for years.

Social media will see more change going forward given the age of the industry. In addition, the past has plenty of failures to reference including Friendster, Myspace, Tumblr, and even Google+. Its easy to imagine someone building a new social media platform (say zero ads, no data marketingj, whatever is demanded) using a simple 1-click import of all your Facebook information.

Kevin English writes:

The case against Amazon as a monopoly in retail is not that convincing at all as Russ correctly points out. I live in a semi-rural area and we frequently use Amazon because there aren’t many other retailers around. However, every week, we drive 40 minutes to the nearest Costco where there is also a Wal-mart and Trader Joes. If I add up my purchases for the month, Costco usually takes most of it. Amazon is in 2nd or 3rd. Whole Foods has 500ish locations, Costco has 700 or 800, Walmart has like 6,000.

Also, the guest says that authors make 25% more but there are sooo many more authors and books now. Also, how many serious book readers do you know that buy books based on Amazon’s recommendation engine? Most of the book I read are recommended from podcasts, blogs, friends, family and I use Goodreads (which is owned by amazon) to find out what my friends are reading.

Austin E writes:

I had to laugh out loud when the guest broached the topic of craft brewing. Because I have an astonishing range of choices from giant conglomerates and independent brewers, I'm somehow worse off than I was a couple decades ago? In 1980 there were fewer than 100 breweries in the US; now there are over 3,000. I can hardly think of an example that more thoroughly challenges his argument, and he's the one who brought it up!

Nathan I Harris writes:

Russ would first like to compliment you you are very gracious and a congenial interviewer when a guest has a very different point of view as yours and you always give the guest the last word.

IMO talk of craft beer cars etc is a differnt argument and a distraction from matts overall point about information business how we consume it how it effects us and how it has been monopolized to what matt was trying to make.

The big differnce between companies Matt is warning us about and companies of the past is a few things.

1. These companies becuase of smart phones and 4g internet connection are by are side 24/7/365 all the time. Even with a newspaper you don't take it to work or to a restaurant you put it down and leave it behind at some point. With Google and Facebook they are always with us constantly vying for and getting our attention wherever we are at especially millennials which i am one and post millennials who are on Social Media and Google everywhere. Whatever effect this has on a whole generation of people we have yet to see.

2. Google and Facebook and to a certain extant Amazon (but definitely the latter 2 a lot more) are in the business and defenitly control the information that we receive these companies are primarily in the information business and providing info to us. IBM Microsoft ATT Chevy provided traditional goods and services where google and facebook primarily work as gatekeepers to information in some ways thathistorically Newspapers and news organizations did and they also provide information in other things ie friend updates etc but we use their product for info.

3. Google and Facebook do not openly disclose how they filter and give a user information (ie search algorithms or facebooks news feed) and they also do not allow users to customize their own news feeds. I find this the most dangerous aspect and where real legislation can change this. In traditional media IE a newspaper and Editor not only had ethical standards that he was held to but if I as a user did not like the way my local or national newspaper chose to cover or display info I could write a letter to the editor and petition for change. At Google and Facebook algorithms have no moral or ethical standards and since their algorithms are kept in secret there is no way for me to petition an algoritm for change or hold an algorithm accountable for improper or immoral news feeds

I have encounter several people who have bought into crazy conspiracy theories about fiat money becuase of youtube and what google has allowed to be posted there. I have also seen wrong and misleeding info shared and reshared on facebook.

Without transparent and public algorithms we as user can not decide how we want our info displayed and are many times at the whims of these algorithms to make important decisions about politics and life if I was a policy maker I would call for the complete disclosure of any Social Media news feed algorithm and any Search Algorithm so that user can understand how news and info is curated to influnce them.

On the other subject yes the gatekeepers have fallen and anyone can make a medium article,but twitter has in less than decade shown us that this new openness has not lead to a renaissance of thought but it has lead to extremism and to a place of misinformation.

(Matts point on books was valid we may have more books at a lower price but the opportunity cost of this is that human capital has moved away from writing quality books becuase quality cost money and Amazons cartel is not about quality it is about quantity. The money for things we cherished in this country are gone newspapers are folding left and right book publishers are diminishing and musical artist are dying, and what we have left is the ability for anyone to write anything yet most are uniformed.)

Michael Byrnes writes:

Austin E wrote:

I had to laugh out loud when the guest broached the topic of craft brewing. Because I have an astonishing range of choices from giant conglomerates and independent brewers, I'm somehow worse off than I was a couple decades ago? In 1980 there were fewer than 100 breweries in the US; now there are over 3,000. I can hardly think of an example that more thoroughly challenges his argument, and he's the one who brought it up!

Yes, he got way out over his skis there. I'd venture to guess he's not much of a beer drinker... the evolution of the beer industry over the past 30 years or so has got to make for a fascinating story.

Nathan Harris wrote:

Russ would first like to compliment you you are very gracious and a congenial interviewer when a guest has a very different point of view as yours and you always give the guest the last word.

Seconded. At the very end of the episode, it actually felt as if Russ managed to gently steer him back on track, which was quite an interviewing feat.

Google and Facebook do not openly disclose how they filter and give a user information (ie search algorithms or facebooks news feed) and they also do not allow users to customize their own news feeds. I find this the most dangerous aspect and where real legislation can change this.

I agree that this is a real problem, for a couple of reasons. One, it's almost certainly true that, say, Google's algorithms are biased in ways that favor Google. Not because Google has evil intent, but simply because they are probably consistently tweaking their systems, testing different approaches to things, and the ones that happen to work well for them commercially are likely to stick around in their ecosystem. Two, because, as an end user, it's impossible to know what relevant hits you DON'T get when you do a Google search.

Matts point on books was valid we may have more books at a lower price but the opportunity cost of this is that human capital has moved away from writing quality books becuase quality cost money and Amazons cartel is not about quality it is about quantity. The money for things we cherished in this country are gone newspapers are folding left and right book publishers are diminishing and musical artist are dying, and what we have left is the ability for anyone to write anything yet most are uniformed.

Here I'm skeptical. Industries are certainly changing as a result of this technology, but it is far from clear to me that the changes are bad as opposed to just different. "There are no good books being written anymore" just seems hyperbolic.

The changes in the music industry seem to be leaning away from the large major producers. Before the wide availability of recorded music, there were lots of local musicians who weren't as good as the best but they were available. As distribution of high quality recorded music became commonplace, local musicians were put out of work. Nowadays, production and distribution of music is almost too easy. Just about any musician can produce studio quality recorded music from his own home. The flip side is copyright protection of music is a lot less successful and there is robust pirating. Bands seem to rely more heavily on touring now (as opposed to record sales) - and this sort of keeps with the idea that in the "on demand" media world we live in now, live events such as concerts or sporting events are some of the only activities that bring people together in the physical world.

I can see where some of thesechanges could be bad, and I have skepticism about the big information-controlling entities like Google and Facebook, but there's still a lot of interesting stuff out there, arguably more than ever before.

No Free Brunch writes:

One thing that I didn't hear brought up are the incentive effects of a possible buyout for a newcomer to the market (eg the beer market). I would guess that, at the margin, this possibility encourages some new entrants to the market who wouldn't otherwise enter. Thus we don't really know the counterfactual where the suspected monopolists aren't there.

As for the author salary issue...I think a prospective author would prefer a large chance of making 0.75X dollars to a much smaller chance of making X dollars. And we, as consumers, have access to knowledge from authors who would never have had the capital or connections before.

Russ, you did a fantastic job of keeping this conversation civil. Even when the guest became strident, you kept your cool and were an incredibly kind, charitable, and respectful interlocutor. Your civility, both as podcast host and on Twitter, are so needed in this day and age.

Blackthorne writes:

It's a shame the discussion got a bit heated near the end, since I think what Mr. Stroller is worried about (or at least my interpretation of his worry) is a valid concern.

I interpreted his concern with Google and Amazon being that these companies essentially run markets in which they have a vested interest. Thus when the success of your company/book/song/film depends on being a success on Amazon/Google/YouTube/Netflix then you are at the whims of these companies, and if they decide they don't like your product, there really isn't much you can do about it.

A pretty clear example of this is YouTube/Twitch/other streaming services. Content creators on these platforms get paid by the platforms based on the number of views they bring in. However, as the platforms themselves make money by selling ad-space, they are often more worried about the interests of the advertisers than they area about the interests of the content creators. As a result situations can occur where companies change their Terms of Service in ways that restrict the freedoms of content creators, which can seriously hinder the livelihoods/incomes of certain individuals. I believe it is these individuals that Mr. Stoller is worried about. Of course on YouTube these individuals are the content creators, but on Amazon these are the authors/booksellers/merchants and on Google it's really any company/website that has an interest in receiving traffic.

Of course these large companies also have an interest in making sure their viewers/customers can find the "good" content they are seeking, or else they'd risk losing customers to competition, but as we don't really know how Google's search engine works or how Amazon's recommendations work we cannot be sure they aren't gaming the system. This lack of transparency, coupled with the fact that Google/Amazon are growing to the point where they are the only "market" that matters, poses what I believe to be a credible risk to the idea of free/open markets.

All that being said, I wholeheartedly disagree with Mr. Stoller on the difficulties of publishing/surviving as an artist/author today. I think the barrier-to-entry is much lower today than it previously was, and I think that it is probably easier for today's average individual to make a living as an artist/author than it previously was. Arguing that "good" authors have it harder today is meaningless as I'm sure all of us have different ideas of what a "good" author is. Plus let's not forget that many great authors of the past were also not commercially successful.

Lastly, I also wanted to express kudos to Russ for keeping his cool throughout the entire episode. I truly believe EconTalk is one of the few places where individuals with differing views on certain topics end up having reasonably level-headed discussions, and I think that is mostly the result of Russ's efforts.

Todd D. Mora writes:

After I listened to Mr. Stoller's arguments, I went to and searched "Amazon Monopoly" and found in the first 10 results a book by Mr. Johnathan Talpin, Move Fast and Break Things: How Facebook, Google, and Amazon Cornered Culture and Undermined Democracy This appears to be a book that is critical of Amazon, however, I was able to find it quite easily.

I think Mr. Stoller brings up some very good points, however, his emotional responses to Dr. Roberts undermines his credibility in my opinion.
Dr. Roberts does a great job of remaining calm and productive during even the most intense exchanges.

Thank you for a very thought provoking podcast.

Eric johnson writes:

The guest was at his strongest when he was talking about how google and social media companies can shape our public discourse and do so without our knowledge or input.

His retelling of the history of the computer industry and the DOJ was just factually inaccurate. Other commenters have covered the IBM misconceptions. His telling of the Intuit/Microsoft merger block is a head scratcher. I can't make sense of his counterfactual where he claims that if Microsoft was allowed to purchase Intuit it would have dominated the internet.

Kyle Norton writes:

I am not an author and know very little about the publishing industry, but I think Mr. Stoller's concern about Big Tech's control over publishing is not new phenomenon. Publishers were also "gate keepers" 30 years ago. If they didn't like your book it didn't get published in the traditional/main stream way. Same thing goes for Amazon today, but now there are more/better ways to publish in "non-traditional" ways.

Dr. Roberts does a great job with these podcasts. Keep up the good work!

Shane Delamore writes:

Did anyone catch when Matt Stoller claimed Facebook bought a company called "Innovo" (spelling?) ? He claimed that Facebook uses this software to spy on users' phones to target, acquire and/or clone new startups before they can become a threat.

I cannot find any information about this company or the acquisition on Google, Bing or DuckDuckGo. Wikipedia has a list of Facebook acquisitions, but I don't see it there either.

Can anyone find any information to back up Mr. Stoller's claims?

John Mininger writes:

[Comment removed. Please consult our comment policies and check your email for explanation.--Econlib Ed.]

Bardstown writes:

Shane, I think the company in reference was Onavo. Per wikipedia:

Onavo, a mobile analytics company based in Israel, is the developer of Onavo Insights and Onavo Mobility apps. The app provides information for users on data usage and compression of data, as well as mobile data protection. Onavo Insights provides data to companies about [3] app usage and consumer trends in online gaming, social networking, finance, and entertainment.

In October 2013, Facebook bought Onavo for an undisclosed sum. Onavo is now headquartered in Palo Alto, CA, but also operates from Israel’s first Facebook office in Tel Aviv.

Bardstown writes:

Matt's core problem with the FAANGs (Facebook, Apple, Amazon, Netflix, Google) seems to be the idea that algorithms are playing a new and unfair role as arbiters of what content people should be eligible to consume. However, this is not a new role! Industry gatekeepers like publishers, newsroom editors, movie and music producers, etc. have historically performed the exact same types of filtering and prioritization that the FAANG algorithms are being criticized for now.

While we have lost certain desirable characteristics of human gatekeeping, I think it's worth pointing out that with algorithms, at least you know that no one else can get an inside track by belonging to the same country club as the Facebook feed algorithm or feel pressured to sleep with Google PageRank to get better search results.

Similarly, you don't have to worry about dogma or preconceived notions about what "works" in the mind of a human gatekeeper keeping you from the market. If we're going to constructively debate the responsibility of tech companies to best serve their customers in a democracy, I think this is the point at which to start:

FAANG algorithms are inarguably less human and have key areas where they are weaker than the humans that they've largely replaced. Some themes of these weaknesses:

1. Opaque decisionmaking - You can't see how optimization decisions are being made on your behalf in real time, and have few ways to react to this lack of transparency beyond not using the product altogether.

2. Knowing what to do with "new" stuff - Algorithms aren't designed to read for comprehension or have editorial opinions, so they are typically dumb when it comes to imputing true value of unstructured content *prima facie*. Until humans start to establish relationships to content by consuming it or linking to it, everything looks basically the same, which makes the initial discovery process something of a crapshoot.

3. Uncontrolled feedback loops - There is inarguably a problem with solely showing people more of the content that they are most likely to engage with without layering any value judgement about content quality on top. These unconstrained short-term feedback loops are biased towards promoting salacious content, fake products, clickbait and scams. Matt called attention to this, in a way, when he talked about absentee ownership and extremism. However, he's wrong to insinuate that Google et al *want* that type of interaction from their users. They know, probably better than anyone else, that users' impressions of their products will crater over time if the algorithm continuously pushes junk at them.

4. Finding an "algorithmic Mittelstand" - The issues listed above, combined with the extreme scale that these companies operate at—often optimizing at national or global scale—create a distribution of algorithmic winners and losers that tends to be push results to the extreme edges of the distribution, either becoming extremely popular (going viral) or remaining completely undiscovered. Figuring out how to create a more equitable distribution of winners and losers seems to be a reasonable topic for further discussion.

Overall, do these weaknesses represent fundamental misalignment with democracy and human dignity, as Matt says? How might alignment be gained by baking more competition and human guidance into these systems? From the perspective of someone in tech, I was disappointed that this set problems wasn't fully engaged with during the interview, as, in my experience, these types of issues that are discussed heavily by FAANG employees internally and are topics that keep all of us up at night as well.
Graham writes:

Russ, congratulations on inviting a guest with such different views. It's a shame that some of Matt's arguments were weak when one of his chief concerns, the strengths/weakness of the networking effect, is still being worked out. I would have much preferred to hear a more nuanced discussion about how we respond to the 'winner takes all' phenomena beyond regulatory responses.

Justin writes:

[Comment removed. Please consult our comment policies and check your email for explanation.--Econlib Ed.]

Madeleine writes:

I just want to object to Mr. Matt Stoller's argument that the internet publishing industry (and Amazon in particular) has decreased the quality of books or degraded the writing profession.

Perhaps Mr. Stoller is romanticizing what the pre-Internet publishing industry was like? Recall all of the books that were blackballed and not published for being "too controversial" or written by women . Recall all of the nepotism. All of the censorship.

Here's a great site
about all of the amazing books the corrupt and useless publishing industry nearly failed to bring to market.

Keep in mind that the only alternative, self-publishing, used to be out of reach for all but extremely wealthy authors.

I don't understand how anyone on the Left can romanticize the publishing industry of yesterday. It's hard for me to see anything except a true incarnation of the Left's classic Bogeyman industry that favors the wealthy and privileged at the expense of the underdog and the consumer. Good riddance to bad rubbish.

John Aiton writes:

Russ is not "woke" or he would realize how benevolent and much love government has for individuals.

I worry that FB , Google , Amazon are already agents of the state .

Gern Blandersong writes:

I've been a listener for many years, but never commented. This was an exciting podcast to listen too. I disagreed with most of Matt Stoller's opinions, especially when it comes to his views on the the spread of ideas and "books". First of all, books are old media. All of the listeners of the econtalk podcast are here because of the internet and podcasts are a new media method of spreading ideas and is only possible in a wide scale with the internet. Second, I have read some of the lesser known books by the authors that Russ has featured on econtalk and these are books that I would most likely never have known about before the internet.

It did get very interesting when Matt talked about the "illusion of choice". When Matt talks about Google and Amazon being the new gatekeepers, he appears to be romanticizing a media past however, there were gatekeepers back then. Maybe there will always be gatekeepers among the media to be the major filters?

I would like Russ at some point in the future to have guests that could explore the topic more about media gatekeepers as it relates to this new economy of internet distribution and the ideas of consumer choice in this new distribution model.

Brandon writes:

[Comment removed. Please consult our comment policies and check your email for explanation.--Econlib Ed.]

John Kranz writes:

[Comment removed. Please consult our comment policies and check your email for explanation.--Econlib Ed.]

Peter J Schueth writes:

[Comment removed. Please consult our comment policies and check your email for explanation.--Econlib Ed.]

Daniel Nunez writes:

I want to thank Russ and his guest Matt, and everyone on EconTalk. I have been listening to Econtalk for maybe 3- 4 years now and this is my first time ever commenting on an episode. But the main thing I got out of the episode is the large gap that exists between both arguments in their view of the glass, Russ reflecting the glass half full and Matt reflecting the glass half empty. Russ arguments and points definitely reflect a view of positivity and that all is well with the big 3 mentioned in the episode, while Matt points reflect a negative view on the underlying actions of the big 3. Like most things, I believe the answer is in the middle, which leaves me with the point that Matt makes and Russ agrees on, that we just don't know enough of what is going on with those companies and what the net effects are as a whole in our lives. I agree that the solution to any of the bad problems presented in the podcast probably lie within both market forces and government intervention, but the extreme solutions presented in the market forces and on government intervention, I disagree completely.

Gabriel writes:

In his book "Fifty things that made the modern economy" Tim Harford explains that before the invention of gramophone, there were plenty of artits making a living just by singing or playing in different venues. After the gramophone, the number went down and the income of the best artists sky-rocketed. The podcast version of that chapter is here:

People were better with the gramophone, just like we're better off with YouTube. New technologies bring creative destruction and increase the competition between the very best and the not so good. Before the gramophone, a mediocre artist could make a living. Today you have to compete with virtually every artist in the world. No surprise then that only a very small fraction can make a living.

I just want to express my admiration for Russ. His kindness and patience should be an inspiration to us all in this age of outrage.

Cynthia writes:

It was only a week ago that I said to my husband how crazy it is that Google and Amazon have so much influence on my business. With the vastness and openness of the internet, how did it come to be that two companies controlled who saw my products and how much I sold? I am an internet entrepreneur who has been in the e-commerce industry for over 15 years. For the first 10 years of my career, I relied on Google to send customers to my e-commerce site. For most of those years I was lucky enough to have my products rank high in the organic search results. But when a product line didn’t rank well, I ended up spending a good amount of money on Google’s ad platform. I couldn’t help but think how easy it would be for Google to generate revenue by demoting my organic listings, which would force me to pay for those keywords. I am not at all saying that Google does this and believe they specifically state they do not manipulate organic search results this way, but I have always thought it would be challenging for a profit seeking company to turn down that type of opportunity.

For the past few years I have been selling my own branded product line on the internet with most of my B2B business coming from Google organic search traffic and most of my consumer business coming from Amazon. I am grateful for the traffic I receive from Amazon but product sales are all over the place so I have no idea what Amazon’s algorithm does. Some weeks I will sell a particular product really well and then a couple of weeks later it will barely move. I never know what “mood” Amazon will be in but have learned to be very flexible with my inventory projections and not to have any particular expectations. Tomorrow Amazon can decide that my products aren’t worth showing and that would be the end of it. I think that is what the guest is expressing concern about – Amazon can decide which businesses survive and which don’t. We hope that Amazon’s desire for profits will lead to showing the highest quality products, which is how many of us expect the free market to work. I think the big different between Russ and the guest is that Russ does believe this to be true and the guest believes Amazon is using this power in other ways.

I have also heard the story the guest told about Amazon batteries. I know Amazon could destroy me pretty quickly if they decided to make my product themselves. They have all of the sales and traffic data and definitely know which products are the most profitable. I know none of us want a world where Amazon predominantly sells Amazon branded products and I would be interested in hearing a lively discussion about the likeliness of that path. With all I hear about Jeff Bezos’ ambitions, I imagine that vision is something that he could reasonably have for Amazon.

In regards to what the guest said about Google product search, I had not heard the part about how Google demoted listings from competing price comparison sites in order to destroy them. But I clearly remember when Google switched from free product listings to paid listings. I had been submitting my products to what was called Google Product Search for 3-4 years, with positive results, when I was notified by Google that they were changing to a pay per click model. I still remember the email saying that this would allow them to improve the experience and provide better results, which didn’t resonate with me as the true reason. Over the years I have participated in their current program off and on with mixed results. Currently I am paying to list my products with Google but mainly because Google is showing my branded products in the paid results from unauthorized sellers that are selling counterfeit items, selling items that “fell off the truck” when they were being processed at Amazon fulfillment centers, or are listing my items at double the retail price (according to an amazon rep I spoke with, these sellers hope for a fluke purchase and would fulfill an order by purchasing the item from my site).

I too am concerned about the how powerful Amazon and Google are but the guest lost me with his extreme statements. Personally, it is unnerving to own businesses that rely on the whim of these companies. I have experienced the sudden loss of revenue when one of them decides not to show my listing as often. But maybe another product is better than mine and deserves to be shown instead? Maybe that means I need to spend some time improving my product.

Unlike the guest I do not have a lot of faith in government regulation to address issues such as these. As a small business owner, government regulations usually hit us small people the hardest, so I am hoping for some free market solutions such as competition to naturally evolve. I am also hoping for some technological solutions that could allow for a better way to show listings. Maybe someday Google will have an interface that shows more than 10 sites per page, which would give users more options to decide which results appeal to them.

A.G.McDowell writes:

I finally switched off when I heard Unix described as a programming language, but prior to that I think points were made that could have been challenged.

1) To the extent that Google (or anybody else) is pushing ad-supported products instead of best search results, it is not benefiting from any lead it may have in search technology or knowledge of the user. It is therefore not implausible for a competitor with merely good search technology to provide better results, especially in queries with a limited domain, such as finding cheap goods.

2) While I am sure the IBM anti-trust litigation closed off some possible routes for IBM, the IBM PC was created at a time when many companies, some very small, had demonstrated that they could produce Personal and Business Computers mostly using commercial chips and chip support packages from other sources, such as Intel, Motorola, Zilog, National Semiconductor, MOS Technology, and others. As an example of the uncertainty, note that MS-DOS was only one of three operating systems offered for the original PC (CP/M-86 and the UCSD p-System were the others). I think the competitors to the IBM PC, including the machines before it and the machines after it that were not clones, show that the huge change in technology rendered many of IBM's previous advantages and investments in older technology irrelevant - and these technology changes were at least as responsible as the anti-trust suits in stripping IBM of its lead.

Bob Meyer writes:

I always find it odd that so many people think that the answer to commercial monopolies is the government, the ultimate monopoly.

Xerox didn't give away the mouse and GUI interface because of antitrust litigation. Their sales force had no idea what to do with these things. Having a government enforced patent monopoly on copy machines, their sale force was little more than order takers. In the 1970's I worked for Xerox fixing copy machines and observed up close how panicked the company became when their patents expired and competition started to drive them out of the business. The sales force was clueless about how to sell against competition.

When Apple attempted to sue Microsoft for their Windows GUI interface looking like the Macintosh, Xerox threatened to sue them both, which is odd for a company that "gave away" their ideas from fear of antitrust actions.

Xerox's Palo Alto Research Center was an extraordinarily inventive group of engineers who consistently produced products that the order taking sales force was woefully unequipped to sell.

Market dominating companies like Xerox, IBM and Microsoft all eventually suffer from a kind of arteriosclerosis where inventiveness and novelty get blocked by intellectual plaque and are replaced by programs to preserve their self destructing "corporate culture". The rigid PC enforcement of Google and Facebook presage their impending senility. These companies are the IBM of the future.

Aside: I never trusted a company whose motto was "Don't be evil". Who gets up in the morning, looks into the mirror and reminds himself "Don't be evil"? Seriously who does this besides Hannibal Lecter?

Cal writes:

Regarding the Amazon and earnings discussion, there's a very well designed info graphic that provides a fascinating look at Amazon's "river" of revenue here:

Dallas Weaver Ph.D. writes:

While I strongly support some restrictions on our "crony monopolies" and oligopolies feeding on the citizen, it seems that "steel", "aluminum", solar panels, sugar, ethanol, and the rest from raisins, to catfish obtaining special tariffs, quotas, subsidies, and protections are a far bigger "monopoly" problem than Google, Facebook, and Amazon. The guest seemed oblivious to anyone other than his claims of harm from his crusade targets.

Being a lot older, when his statements drifted into anti-trust history in the era of IBM being the 900 lb gorilla of the computer world, it seems like his history "search" focused on confirming his biases, not on what actually happened (a good example of the use of the web to obtain the answers you want). Upon IBM becoming an old bureaucracy with lifetime employment, their new hardware development schedule had stretched too almost a decade while the basic chips were doubling every 18 months. IBM knew they couldn't compete so they started a new PC shop without any of the bureaucratic controls and they fouled up and actually made a poor decision on the "operating system" by allowing Microsoft to have control, and the rest was history. This had nothing to do with the anti-trust actions, just a bad business decision in the view of history.

Even the Justice Department forcing IBM to open up the hardware interfaces between pieces of equipment did allow some competition in hard drives, etc., but that also forced IBM to freeze standard between devices that may or may not have been optimal. They couldn't utilize a new protocol for new system designs which can prevent innovation of far better or faster protocols. Standards can freeze technology and prevent innovation, even when they make the Justice Department happy. In hindsight, according to my Son, even the fundamental internet protocols would be different and much more robust in a world where internet security is a real issue.

As someone who had a business that used Google Adds for a specialized niche market producing zebrafish, embryos, staged frozen embryos for scientific researchers around the world. I could reach this niche market that ranged from Max Plank Institute in Germany to the University of Singapore and every major University (from Harvard, MIT to UC Berkeley, Stanford) or Biological research Institute in between for just a few dollars per year. I did get my $0.15 check for the class action settlement from the legal parasites. Without those adds, perhaps some 200+ peer-reviewed publications wouldn't have been done or would have cost a lot more (they could all do the same thing, but I could do it a lot cheaper as I was selling 10 million zebrafish per year into the ornamental pet trade).

When talking about authors and books, he was conveniently forgetting that in his "good old days" before Amazon, the gate-keepers were the publishers who would support their favorites and block those they don't like while keeping out newcomers. In scientific publishing, the publishers had even a better deal by paying the authors nothing and having peer reviewers and free editing while they kept all the copyrights. From what I heard, I am not sure that the guest would contribute anything beyond rhetoric to the issues, creating nothing worth reading independent of the distribution mechanisms.

My conclusion was that he had no valid arguments beyond his desire for some "god-like" all wise bureaucrat from the Justice Department to specify what algorithms can be used at a time in history when actually seeing the weighting function in an AI neural network leaves even the most expert software engineers not knowing how the system actually makes it decisions.

Jacob writes:

There has been a lot written about Matt's apparent lack of nuance in discussing the craft beer industry, but I want to give him more credit than other commenters.

I am a brewer at a small (5000 barrels/year), independent brewery and, in fact, was brewing while listening to this episode! Matt's argument regarding craft beer is similar to his gatekeeper argument. While I don't know enough about the way tech companies do what they do to discuss it intelligently, I do know that craft beer has a massive gatekeeper - and it is named AB-InBev.

Beer is an agricultural product that must stand on the shoulders of giant agricultural companies. Barley producers, malting houses, hop growers, hop sellers, yeast manufacturers - all are private entities who - rightly so - seek to maximize their own profits. When Matt talks about AB-InBev putting the squeeze on small craft breweries, it is at this agricultural level. As breweries get absorbed under the AB-InBev umbrella, another contract for hops gets absorbed, and the purchasing power for one company grows relative to the industry.

Count this comment among those that disagree with Matt's assertion that there is less choice for beer consumers. With that said, this is an extremely young industry that is vulnerable to the influence of a very small number of companies dominating market access. I hope this aligns the beer topic more with Matt's argument about the way tech-giants dominate the digital world.

Tom Murin writes:

This interview was painful at times, but certainly interesting, as expected.

I can't get over the "Who we are as a people" and "Illusion of choice" cliches. It is amazing how two people can have such a divergent view of things. Russ sees choice/free will where the guest sees us essentially as dupes or victims.

This is the "we know best for you" mentality at work - which scares me more than Amazon, etc.

Don Crawford writes:

This guest's argument was that some corporations are acquiring undue influence as they get large. This fear of the "power" of large and successful corporations is at the heart of the left's critique of capitalism. They are constantly harping about the power of corporations saying it must be curtailed. Who can do that? Well, governmental agencies and bureaucrats should be empowered to do that. And what happens then? The agencies are captured by the corporations and the regulations end up hurting, not the big corporations, but the new entrants into the field, the ones who we are relying upon to provide the competition that checks the power of the big corporations. Witness how well Dodd-Frank has helped provide ordinary folks options in terms of borrowing.
Ultimately, we cannot "guarantee" how start-ups will innovate and bring down the giants of today, but we do know it has happened over and over. And we also know that the alternative--giving power to government bureaucracies has never worked to our advantage. Unfortunately, those on the left believe a different story--one in which the government has always saved us from these terrible corporations. The guest told a story of how the anti-monopoly crusades against Microsoft made everything better for consumers.
It would be nice to be able to have a calm and civil discussion about the relative merits of these two competing visions of reality.

Dr Golabki writes:

@Jacob and other

I agree that there's CLEARLY more choice in beer now. Probably an order of magnitude more than there was 20... maybe even 10 years ago.

Matt is right that a couple companies control a huge amount of the distrabution, and those large companies noe own a lot of the micro brewers, but (A) just because Inbev owns Goose Island doesn't mean that Matilda is the same as Bud Light, and (B) even though the big companies own a lot more brands than you might think there has still been an explosion of very small professional brewers that make locally distributed, or only available on-site craft beer.

What's bizarre is that Matt seemed to bring this up for no reason and it's a terrible example for him. This despite the fact that Russ was begging Matt to talk about news media... where Matt clearly has a stronger case and is in more agreement with Russ and most of the listeners.

Dr Golabki writes:

It felt like Matt came in with the liberal bias discussed in the previous week's podcast... 'big companies are inherently bad, therefore everything they do is bad, everything bad can ultimately be attributed to them'. I know that's an unfair characterization, but I think that tone came through and it undermined some very valid points.

I'll raise a few here, which I wish had gotten more air time in the podcast:

1. The advantage of scale is fundamentally different in the information industry. While there are advantages to scale in almost every industry, the advantages are much larger here. it's not just network effects on the user base. As Matt said, the underlying products themselves are dependent on the massive amounts of user generated information. And it's not just the information side. Amazon is, I think pretty clearly, investing to be a near monopoly retailer. Russ poo pooed this because they aren't making money from that part of the business... yet. And Matt said they can turn on the tap whenever they want... both are right, so why aren't they turning on the tap? The thing that's concerning is that as powerful Amazon is in retail, Amazon and it's investors think it can get much much more powerful. The barriers to entry are extraordinarily large and they are growing. I wish Russ had enough market skepticism to see this as a big potential problem... and not something that must nessasiarly be fixed by emergent market forces. Sure it'll be fixed in the long term .. when we're all dead.

2. The danger of scale is fundamentally different in the information industry. As was discussed when you're as big as these companies you can push around city and state governments pretty easily. To Russ, I think this is a problem of too much government discretion, to Matt it's a problem of too much private power... seems to me that both have a point. But that's a problem with every large company. Beyond that the control of information is central to everything we do. It's like if Ford not only was a huge car company in the mid 19th century, but also printed all the maps in the country. If they wanted to hide a street, or a city or even a whole state they could effectively hide it from 99% of the population. That sounds implausible of course, but that's because the scope of the US or the globe is so much smaller than the scope of the internet. We're dealing with a new type of thing. Now, we might believe that Google is self-constrained because it's most important asset is it's user base, or because Facebook/Amazon could enter search, and I think that's true. But at the same time Google has the power to disappear large swaths of information, and the search engine makes choices about what to hide and what to find all the time, and we're unable to see how those choices are made, let alone control them. That amount of concentrated, unaccountable power is worrisome even if you think Google is a great company with great people and great incentives.

Lastly, Russ asked "what's the solution". To me the obvious solution is to change how data ownership is determined. I'm not a data engineer, so I won't claim to know the "correct" allocation of ownership, but it seems pretty clear that companies have fair to much control over the data we generate, while we have essentially none... its clear to me we need to rebalance.

Steve Shank writes:

So, if I understand at least one major argument he makes, Google has vast stores of information about us so Google can tailor search results to us giving us just what we really like to see, conservative news for conservatives, liberal news for liberals, etc. This hurts democracy. This makes us like them better so no competitive search engine like duckduckgo can compete. He uses himself as an example, saying he has tried duckduckgo but it didn't satisfy him. So apparently, he doesn't want to see different opinions.

There are other people, who do want to see different opinions and find DuckDuckgo wonderful. NO ads, no tracking, Great results. Furthermore, startpage let's you use google, but strips out the personal information and advertising.

But better than all those is the open source Searx search engine. I use that and have it use: Qwant (a privacy respecting French engine) Swisscows (a privacy respecting swiss engine) duckduckgo, bing and google. It has no ads, no tracking and lives off donations like wikipedia. I get results showing a range of opinions.

We have access to more stuff than ever before and do not need to live in echo chambers, unless we want to.

He also made the point that we were being limited in thoughtful books now. Much more than before, by Amazon's monopoly. So I did a search (using searx and amazon) for Sankhya Karika - This is an ancient philosophical text from India explaining the Samkhya system of philosophy, which was the philosophical foundation of Yoga and has largely disappeared, being supplanted by Yoga and Vedanta. This would be the kind of deep philosophical book which we shouldn't have anymore, according to him. Furthermore, I needed this book in English, not Tamil, or sanskrit, or hindi.

I got pages and pages of results from Searx, so I could find out lots about it, including lots of places to purchase books. Checking just Amazon, I found 41 books, 29 in stock. Then I switched spelling to an alternate way to spell sankhya and got 19 more results including 3 kindle books I could immediately download.

Never before have we had so many options from so many places and so much freedom.

If the median income for authors and/or musicians has fallen, another of his assertions, the most likely explanation is that hundreds of thousands of people are writing books and making music and self-publishing with nearly a zero following and little or no income from it. Their books tend to have the "Be the first to review this book". But they are available now. There are many philosophical and religious organizations who are publishing their books including old classics on Amazon as ebooks. They have few purchasers, but those people who want them have them.

Don't forget all the free classics available from the Guttenberg project and similar projects, making many classics from the Greeks and Romans available inexpensively for us.

Dr Golabki writes:

@steve s

I think Matt had two decent points on publishing and book retail that can be true even if there is generally much more choice for readers and freedom for most writers.

1. There is a point at which amazon controls x% of all book retail, where we should be worried that it's just too much control of an important industry for one company. Exactally what "x" is might be a point of debate. But I think even most libertarians would agree there's a number less than 100%.

2. Even though there is more publishing freedom now, there is a subset of important books that are declining, well reported nonfiction books that take lots of up front time and money to research and write. I don't know for a fact that that's true but Matt said it was and it strikes me is fairly plausible.

Mike Diaz writes:

If Matt had his way, it would be illegal (ultimately criminal) for me to sell my technology business to Amazon, Facebook, or Google if I so choose. That would be too bad for me, my family, and the families of all my employees who would benefit from selling the result of their hard work. I honestly don't care if authors are supposedly getting paid 25% less than they used to. Blacksmiths don't get paid what they used to either.

Seems like I can either have my choices influenced by Google, Amazon, Facebook, or by the government. I choose the Google, Amazon, and Facebook because last time I checked they don't have a police force to ensure that I do what they say. And, as long as there is competition (which is principal disagreement with Matt) then Google, Amazon, and Facebook will be ultimately checked. It is already happening with Facebook, which is facing significant competition from other social media (my Facebook feed is much, much less than in the past).

Steve Shank writes:

Dr. Golabki,

I think Matt had two decent points

I disagree. First point:

There is a point at which amazon controls x% of all book retail, where we should be worried

How would this occur? I do use Amazon, but I buy three times as many books from Kobo. The only way Amazon could put all the other companies (, Kobo, booksamillion, Google books, iTunes etc.) out of business would be to offer so much better deal that people choose them overwhelmingly, including being forced to use their proprietary book format instead of Epub. Is the solution for other people to authorize the state to take this freedom away from us? If people don't care, should a minority of people, or people who don't buy books, really control those of use who do? I don't find this a compelling argument, nor one that is likely to occur because Amazon just isn't closing off other outlets.

In fact, software is making it easier and easier for authors and publishers to make their books compatible with the Epub format which is used by essentially everyone else. So, when you make your book available as an ebook for instance, it is simple to also make it available to Google, itunes, kobo, etc.

For print books, Amazon to keep its customers, must provide a wide selection at a good price, they cannot offer inferior books and keep its customers. Something Matt doesn't seem to understand.

point 2:

there is a subset of important books that are declining, well reported nonfiction books that take lots of up front time and money to research and write.

This was refuted in the podcast. Both Matt and Russ work for Think tanks. So they can research books and take a long time to write them, and more think tanks are available than ever before. Furthermore, established authors get an advance from their publisher.

So the only people who can't take 4 years to research a book, and not have other gainful employment, are people not employed by think tanks and not established authors. When has it ever been any better? Probably never.

John Rosendahl writes:

That feeling when “something is wrong on the Internet” can be unbalancing, I never really expected to feel that from an Econ Talk episode. I have always loved the way econ talk respectfully allows diverse opinions to be be discussed in their strongest possible representation. Additionally, I share the concern that large Internet based media companies may act in ways that are detrimental to society as a whole.

However, this interview stands out in that was so riddled with unsupported assertions, misrepresentations, and falsehoods that it undermined any valid point that Mike was attempting to make.

One area where Mike misunderstands is displayed in his assertion that Bing and Duck Duck Go do not and can not provide useful search results because Microsoft and Duck Duck Go lack the “Big Data” required to provide useful search results.

I will gloss over the fact that he simply assumes that because he prefers Google results to other engines all other users would have the same preference. Additionally, I will not argue that by their existence these alternative search engines demonstrate that they must have some competitive advantages (video for Bing, privacy for Duck Duck Go) that some users value.

Instead let’s take a somewhat simplified look at the data that is used to create search results and how this data is employed to provide the best possible results.

The primary objective of a search engine it to determine the most relevant and highest quality source of of information in response to a users query. This can be broken into several independent questions. 1) What resources are of high quality, 2) what is a specific resource about, 3) what is the intention of the user’s query, 4) based on the previous answers what are the most relevant resources for that user.

The first source of data a search engine uses are the webpages/resources themselves. This is the most critical source of data and by definition is available to all users. Based on the contents of a page and interconnections between pages assumptions can be made about subject matter and quality. This analysis can be very simple, based on the number of times a word appears on a page (original Yahoo), more complex using information about what pages link to other pages (original Google) or more sophisticated AI analysis methods (like word2vec).

The second source of data a search engine uses is the actions a user takes when they perform a search. To gather this data, you must already have a search engine and user activity, however there are ways of getting around this ( For instance, more analytical resources should be employed to more commonly searched for terms. Additionally, by monitoring user activity you can improve the understanding of the results you provide. For instance if 80% of all users click on the second result on a page, this is a signal that the first result is less relevant or of lower quality according to actual users than expected.

The third source of information used by search engines is information about users themselves. This is how highly relevant results are found. An example is an environmentalist or chemical engineer searching for TCP probably wants to know about Trichloropropane while a systems administration who enters the same search is probably looking for information about Transmission Control Protocol. Additionally, a user would be more interested in results that are geographically proximate, or a bi-lingual user may be interested in results in either language.

I assume this is the data that Mike claims only Google, Facebook, Amazon, (and for some reason) Apple(?) have. While it is true that these companies do have some of the largest and most valuable data sets of user information. There are several issues with this assertion. First of all these three companies have access to and use different data. Amazon has purchase history and product interest information, Facebook specializes in interests and social connections, Google has the best information about browsing history. Each of these data sets provide a lot of insight about a user, but each one build their data set independently. There is no reason to assume these three methods are the only possible ways to understand users.

However, even if users can only be understood through purchase history, browsing history, and social connections, there are still other sources of this information or example, while Amazon knows a lot about my purchasing history, my credit card company knows more. Google knows a lot about what web pages I have visited so does my ISP. While Facebook, knows about my social network, my phone company knows who I call on a regular basis.

In fact there is an entire industry devoted to suppling user information information to interested parties (

(TLDR) - There are three major sources of data that are used in a search and a competing search engine could potentially access all of these sources. Yes Google has an advantage, they have been doing this for longer than anyone else. But they do not have anything that can not be recreated by someone else.
The assertion that only Goggle can provide useful search results because of their data advantage one of the many false assertions made through the interview.

Robert Swan writes:

Could have been a more thought provoking conversation if the guest had made it beyond indignation. Most of my thoughts on the talk (beer, authors, big government) have already been covered by other commenters (several times over).

One new point on authors' incomes. In days gone by, a budding author would write his masterpiece and then try to find someone willing to publish it. If he couldn't, he would be left out of any count of authors, despite having written a book. If you were to count all such authors I think it would bring the historical average income down rather a lot.

I'd suggest commenter A.G.McDowell should listen on beyond Unix being described as a "programming language". It's not such a grossly inaccurate statement (Unix provides a "vocabulary" for processes, files and memory which programmers use) but, even if it were, I don't see how that makes the rest of the talk unlistenable.

Far from being more powerful than historical monopolists, I'd say that Facebook and Twitter, in particular, are in a precarious situation. Along with the likes of Apple and Tesla, they seem to be in the fashion business. I am yet to hear why the MySpace collapse couldn't happen again for Facebook or Twitter. And while Google and Amazon are in a less tenuous position (I suspect because they have clearer objectives), the Internet is still a pretty wide-open place. Have a look at what's happening in Internet shopping in China and then tell me again that Amazon has a monopoly.

The guest uses "we" a lot, without clearly identifying who "we" are. To paraphrase Margaret Thatcher: there is no such thing as "we". I suppose the nearest approximation to "we" might be a representative government. Makes all those "we" statements just a tad sinister to the likes of "me".

And on that point, it was good that Russ pressed several times for exactly what Matt Stoller wanted the government to do. While I might be won over that there's a problem, I'd take some convincing that any solution wouldn't entail bigger problems.

Michael Byrnes writes:

Steve Shank wrote:

I do use Amazon, but I buy three times as many books from Kobo. The only way Amazon could put all the other companies (, Kobo, booksamillion, Google books, iTunes etc.) out of business would be to offer so much better deal that people choose them overwhelmingly, including being forced to use their proprietary book format instead of Epub.

For print books, Amazon to keep its customers, must provide a wide selection at a good price, they cannot offer inferior books and keep its customers. Something Matt doesn't seem to understand.

I think this is basically correct, and the logic can be extended beyond book sales to Amazon's other retail business. As essentially a middle man (although there are some blurred lines here as Amazon also produces some of its own content), Amazon can gobble up more and more business only by being better than its competitors, which would be a net good for consumers. But the idea that Amazon will put all of its competitors out of business and then "jack up prices" seems unrealistic - they will just lose business. Much of what can be bought via Amazon as middleman can also be obtained directly from the producer - less convenient, more costly, but still there as a fall back plan if Amazon becomes less good of a deal.

To the extent that there is a concern with an Amazon or a Google, I think it is just that you can only evaluate them by what you see when you search on their sites - not necessarily by what hits you don't get (assuming a basic level of competence). Whether and how much of a problem that is is debatable, but that and regulatory strategies are where the concern should be, to the extent there is a concern.

Steve Labinski writes:

I agree with everyone that this was a fascinating episode. Although I disagree with Matt's solutions to the problems discussed, the fact remains that it's extremely difficult to opt-out of Google's products. As an owner of several two-bit websites, I fully share his anxiety with this company.

Something that was not discussed, which I think would inject a "property rights solution" into this situation would be removing the DMCA safe harbor provision which makes Google YouTube essentially unaccountable for all the copyrighted songs and videos posted to their website, which Google makes money off of using their ad network. The safe harbor rule has significantly harmed small companies who produce music and movies because it's a ton of work. This is an example of federal law that protects a big company at the expense of small guys.
Also - was Net Neutrality never brought up? Maybe I missed it. In my opinion, repealing it adds competitive pressures to Google and Facebook. Yay!

Andrew Swantak writes:

Econ Editor posted:

"We have had to reject an unusually large number of comments in this thread for rules violations."

Having listened to this podcast I am not in the least surprised.

My answer to Matt's concerns: More Google's, more Facebook's, and (certainly) more Amazon's.

Ryan writes:

Could anyone find evidence for Matt's claim that you can't block YouTube on Chrome? I did a DuckDuckGo (and Google) search for "can you block youtube on chrome" and got multiple pages describing how you can easily block YouTube on Chrome (including some YouTube videos). I didn't see any results claiming that it couldn't be done.

It doesn't invalidate his entire argument, but if he is going to cite it as an example of how hard it is to avoid Google products, then he should make sure it is accurate. It does speak to his credibility and the care he puts in to verifying claims that support his opinion.

Vincent Passanisi writes:

Lively and thought-provoking. I haven't had a chance to read all the comments yet, so I apologize if it's already been mentioned, but the Wall Street Journal ran an article this morning about Alibaba's UC Browser which is challenging Google's Chrome in India and Indonesia. And if the article is correct, I also found it interesting that it is actually Apple's Safari browser that has a majority of the US market at 52%.

Competition often comes from places we least suspect, and in forms we never anticipated.

I am less concerned with Google's and Facebook's monopoly power than what another commentator once pointed out: that whenever a product is provided for free, we must keep in mind that we then become the product.

Matt W. writes:

A government solution to the problem of monopolies is internally contradictory. If monopolies are bad because of their dyscivic properties then the same should apply to the government itself because it holds the monopoly of force.

One might argue that the government is a special case because citizens can vote, but this is not exactly true as anyone familiar with politics can attest. Just as you as an individual are powerless to stop Amazon, Google, Facebook, etc. by withholding your dollars/time from them, the same is true in politics. Very few elections are swayed by just one vote.

In my opinion government power is therefore more insidious. At least by foregoing using Amazon, Google, Facebook, etc. I can avoid their controlling what information I have access to (even if I can't always avoid their affects on society), but with government, I am forced to participate & contribute regardless of how I voted.

Furthermore, the government cannot be trusted to use their power to break up/regulate monopolies any more than those monopolies can be trusted with their market power. Bureaucrats and politicians that run the government have their own incentives that do not necessarily correspond to the general populous. Doubtless their choice of prosecution will have more to do with the political affiliation and donor status of the companies than their danger to the public good.

Dr Golabki writes:

@ Steve S

There is a point at which amazon controls x% of all book retail, where we should be worried

How would this occur? I do use Amazon, but I buy three times as many books from Kobo. The only way Amazon could put all the other companies (, Kobo, booksamillion, Google books, iTunes etc.) out of business would be to offer so much better deal that people choose them overwhelmingly, including being forced to use their proprietary book format instead of Epub. Is the solution for other people to authorize the state to take this freedom away from us? If people don't care, should a minority of people, or people who don't buy books, really control those of use who do? I don't find this a compelling argument, nor one that is likely to occur because Amazon just isn't closing off other outlets.
The claim I'm trying make is pretty narrow, just that there is some point at which would start to worry about it. Once we agree on that we can address if we are likely to ever reach that point and what the potential solutions might be if we did... I generally agree with you're point above that we probably aren't there now and policy cures may be worse than the disease.

In fact, software is making it easier and easier for authors and publishers to make their books compatible with the Epub format which is used by essentially everyone else. So, when you make your book available as an ebook for instance, it is simple to also make it available to Google, itunes, kobo, etc.
I don't think anyone would disagree with this.

For print books, Amazon to keep its customers, must provide a wide selection at a good price, they cannot offer inferior books and keep its customers. Something Matt doesn't seem to understand.
In fairness to Matt, he did say several times things are currently much better most consumers.

there is a subset of important books that are declining, well reported nonfiction books that take lots of up front time and money to research and write.

This was refuted in the podcast. Both Matt and Russ work for Think tanks. So they can research books and take a long time to write them, and more think tanks are available than ever before. Furthermore, established authors get an advance from their publisher.
So the only people who can't take 4 years to research a book, and not have other gainful employment, are people not employed by think tanks and not established authors. When has it ever been any better? Probably never.

I think "refuted" is pretty strong. Matt sited a comment by an author in field. Russ disagreed with Matt and pointed out think tanks exist. Both are pretty colloquial, I wouldn't say I was convinced either was empirically correct. I would also question whether most work coming out of think tanks is really the same thing that Matt was tanking about. I think Matt would have been better served talking about local news reporting and foreign reporting in the US.
Alex J. Philips writes:

I normally listen to learn something new, but this episode was a disappointment as I felt that the entire argument boiled down to we as consumers can't be trusted in our choices and it has caused disruption/destruction which is bad. Each argument seemed to imply government was the only solution. Just because authors are paid less isn't bad. Supply and demand exists. Why do we need foreign news bureaus to travel the world to learn about a culture when we can simply have someone from that culture report for us from that location since we are digitally connected. It is the removal of these inefficiencies that leads to progress.

Dr Golabki writes:

@ Michael B

As essentially a middle man (although there are some blurred lines here as Amazon also produces some of its own content), Amazon can gobble up more and more business only by being better than its competitors, which would be a net good for consumers. But the idea that Amazon will put all of its competitors out of business and then "jack up prices" seems unrealistic - they will just lose business. Much of what can be bought via Amazon as middleman can also be obtained directly from the producer - less convenient, more costly, but still there as a fall back plan if Amazon becomes less good of a deal.

So Amazon's plan is to dominate retail in books and numerous other industries, in order to give consumers a better experience, and, according to Russ, make no money?

The "jack up prices" view of the world isn't really correct, but it's also not simply that Amazon is serving consumers. They are trying to innovate and get scale to get competitive advantages, not all of which are necessarily good for everyone. With scale they can obviously reduce costs, and maybe starting marginally increasing prices at some point, but they can also pressure suppliers to deliver the type of product they want at the time and in the format they want under the terms they want (in this case publishers, which was discussed, and authors, but also all kinds of business), they can pressure governments (which we've seen with state sales tax, and the HQ auction), and they can act to block potentially competitive innovation.

Seth writes:

I enjoyed this podcast.

Substitutes like blogs and podcasts may have also contributed to declining author earnings. Sample size of one: I read fewer books than I did 20 years ago, but I read more.

Big companies du jour have power. Russ frequently says that being pro-capitalist isn't the same as being pro-business. Of course, companies will act in monopolistic and unsavory ways.

I didn't see Matt's example of Amazon copying good-selling products as unsavory or monopolistic. That's a good market-based innovation strategy that retailers have used for a long time with their store brands. It expands choice for consumers, rather than restricts it.

Every generation has had those who criticize the power big companies du jour have.

In 'most?' (maybe all) cases, competition has proven to be a better antidote than regulation.

Jay writes:

Beer ye, Beer ye,

If all the beer in the world were made by a single company and tasted as good (and had the current variety) would a beer drinker care?

It seems like a lot of the arguments here could use more flesh. the average wage of a book writer has decreased? This could have just as easily been because more people are writing books tailored to a specific audience.

Matt mentions we're getting fart books instead of Shakespeare. But why not have both? 'Ole Bill the bard could write a good fart joke in Iambic pentameter. as I have linked in the URL.

David Zetland writes:

Stoller is right. Markets are distorted and THAT'S a problem.

Kyle writes:

I was looking forward to this podcast because I am very sympathetic to the guests view about google and Facebook. There are peer-reviewed studies showing that google can influence close elections. This is a problem because google as a corporation has helped different politicians and thus is not a neutral party but most people treat the search results as if they are neutral. So, as I said I started the podcast eager to hear about these potential problems.

Unfortunately, the arguments presented were often not only unpersuasive but poorly framed. Right out of the gate the first objection to Amazon was that cities were trying to get Amazon to build in their town. People may object to this - I don't - but it is so common in the US that Amazon doing it is boring and would be shocking if they didn't do it. The discussion continued in that way often focusing on things that did not even register to me as concerning.

Some data would have made the Amazon discussion more useful:
Number of fiction and non-fiction books per year published in the US
Number of authors being paid for writing per year

The little data I could find seemed to suggest that publishing continues to increase

The argument that we have more books but they are all fart joke books (the gratuitous swipe at George Bush was beneath the entire discussion, especially odd since Bush is known to be a prodigious reader of mostly intellectual books) is an argument of elites vs common people. You can write all the books you want about high brow history and the majority of people will be more interested in the latest offering on Netflix. The reality is that there is relatively little demand for high brow literature because our current population when forced to choose rejects them. That is a change that is far beyond Amazon and Amazon could dedicate its life to pushing intellectual history and probably more people would be watching the Housewives of Miami. Lament all we want, but the blame is far beyond one new player in the publishing industry.

We have more of everything today, but is everyone getting paid less? I don't know, but yesterday a coworker suggested I listen to a new band - I went to Apple Music (Spotify isn't completely winning yet) and was able to listen to the music and they got paid some money for my doing that. This band is obscure and previously they could only have gotten money from me if I saw them live. That seems better for them and me.

In college I took History of Language, which was world history but viewing all history from the lens of language. It was incredible, but despite our focus on language I knew that there were 30 other classes (history of technology, revolution, war, warfare, agriculture, money) that were giving a parallel view. The argument in the discussion seemed at times to be that the one true history of the recent US technology is the history of anti-trust law, but that is just one current among a river of changes that led us to today. Corporations and technology are no more solely driven by anti-trust law then they are by just language.

Blocking Youtube is fairly easy, and I am surprised someone published an article about how they could not do it when DuckDuckGo can easily show you many ways.

Many of the arguments probably made sense if you shared a world view of fearing corporate power while not fearing government power. But if you feared both or feared only government they were not convincing.

Google's anticompetitive activities regarding pricing were rightfully punished in Europe and as they do more of that they need to be punished.

Finally, thanks to Dr Roberts for maintaining exceptional decorum amid a somewhat heated discussion.

David Haglund writes:

This Interview was unbelievable. Very engaging and heated at times. Russ showed extreme patience with Mark. While they never made it to expressing exact solutions to these problems I would guess that the guest would surely want the "government" to step in and solve things. While I wouldn't agree with that, I did learn some things that were not good. I can see where some of the EU's problems with Google originated. Thanks.

Mark writes:

I don't think there's any other way to put it: Stoller made poor arguments through and through. Many of his claims he backed up only with anecdotes, and he often seems not to understand basic economic principles.

1) A company making little profit is not the same as a charity; it's a for-profit company in a competitive market. His remarks about Amazon being a charity were preposterous.

2) Big companies buying up small start-ups that they worry will cause competition should *incentivize* new start ups, not deter them; a lot of people in tech these days create apps or tools with the main intention of selling to a bigger firm; it's almost regarded as a form of insurance. It's less risky to do so when there's not a ready buyer for your business.

3) But Stoller's most eggregious error, imo: his claims about publishing and music are 100% backward. Youtube and Amazon have driven down the cost of publishing and marketing books/music, making it so almost anyone can publish or broadcast to a prospective audience. It's precisely this heightened competition that drove down average income of authors and musicians: there are more of them than ever and competition has never been fiercer. There use to be oligopolies under the old publishing companies and record companies; Amazon, Youtube, et al. broke them. It's quite ironically that someone so fiercely anti-monopoly fails to recognize how much these platforms have done to destroy monopolies in other industries.

I applaud Russ's patience. I think if anything he went to easy.

I mean, we need to rebel against Amazon like against the British? Wow.

I will note that I like guests who disagree with Russ's (and my) free market perspective. I'd like for them to have Paul Krugman on even; I enjoyed his interview with David Beckworth's podcast. And there is an interesting discussion to be had about the potential conflict between natural monopolies and the value of competition (economies of scale weren't mentioned once! That's seems like a pretty important thing to mention). But this guest, in my opinion, did not offer a challenging discussion of this issue.

[comment edited with commenter's permission--Econlib Ed.]

Karl writes:

Long time listener, first time poster. I just heard the podcast today and I agree with the sentiments of some of the other posters that Russ was quite patient with the guest.

A couple of points that led to this post:
- I live in Portland, OR and I think it's a great time for microbreweries. I lived in Boston in the early 90s when Sam Adams got started and I think there are so many more options now than before. Could most of the microbrew labels be owned by InBev? Quite possible, but that doesn't mean consumers don't have a lot of options.
- The guest mentioned some statistic about how the average author makes X% less than Y years ago. Russ mentioned how consumers now have much more choices (books, TV, beer). I wonder if the two are related in that there is many more books out there ("long tail") but most of them have very little sales. I recall reading somewhere else (Tyler Cowen?) that the return to winners is now much greater.

Sam writes:

I am very sympathetic to Stoller's cause, I think the U.S. government should pursue vigorous antitrust action against many of the companies he talks about but he really wasn't persuasive in this conversation both in the manner in which he argued and his command of the facts.

Some points Stoller did not make:

(1) if there has been an explosion in consumer choice it is not clear that it would not have been greater with antitrust enforcement.

(2) suppose there is great choice and monopoly practices have made consumers better off. What is the guarantee that this will continue? Commenters here say Stoller is wrong about the craft beer market. Suppose he is, why should we allow the craft breweries to be bought up by large beer producers? What is the benefit? Why let them have a future option of eliminating competition? The fact that they are not exercising that option now doesn't mean they will never do it.

(3) finally a rhetorical point on Amazon, which is not original to me but I find powerful: we are at a point when one can manage one's life by buying from Amazon only. That is a phenomenal amount of power concentrated in a private entity and it has happened without any political control.

Dan writes:

I would to hear more of Mr Stoller on the history of monopolies, perhaps he could return after his book is published.

His views on Google, Amazon and Facebook are clearly a work in progress, but it couldn't be otherwise, given how recent and unique these businesses are.

It was disappointing that so much time was taken up discussing "Craft Beer" during this talk. Quite a few of the comment posts concern this tiresome topic, which may mean that the audience is more interested with beer choice than choice as to how they are informed and educated by Google et al.

David writes:

This was a great discussion. I definitely see the merits in Matt's argument that there is a concentration of power, both economically and politically, for the "big 3" tech companies. But at the same time, there is a great amount of choice available to Americans. I can see that Amazon may disrupt that consumer to product connection to favor themselves, but I think Matt overstated that point. The beer example illustrates this.

While AB-InBev and MillerCoors have certainly bought numerous "craft breweries" it does not get anywhere near the number of truly independent breweries, which now totals over 4,000 in the U.S.. This is the most at any point in our country. I do see concern on the distribution aspect of Matt's point though. The big beer players can disrupt that connection between the consumer and the product (i.e. beer in bars, ballparks, stores, etc.). But I do think there is considerable power for consumers and independent breweries. Craft breweries have changed a long list of policies at the local and state level and have been successful in favorable tax cuts at the federal level.

andy elan writes:

what a fresh wind.
as much as I appreciate hayeks idea of freedom with these 5,6,7 500lbs gorillas in the room it is naiive to believe they will not drop:
-consumer choice
-citizen rights
-cultural life.
not because they are evil or so - its the incentive structure in our economy.
decartelization to support competition and keeping power balanced would be the remedy.
I love the beer thing.

Butch Dorian writes:

I think that Matt makes several good arguments on Facebook, Google and Amazon. However, his political rhetoric got in the way with the Beer industry. Given his political emotional attachments he puts his other arguments in jeopardy of being taken seriously.

George writes:

@Matt W

"A government solution to the problem of monopolies is internally contradictory. If monopolies are bad because of their dyscivic properties then the same should apply to the government itself because it holds the monopoly of force."

This is a great example where logic cannot be substituted with reality. Competition on the monopoly of force (legitimate use of violence) is called civil war and more generally war. Competition is wonderful when the losers don't actually die. That type of competition should be clearly avoided. The losers in modern market economies are typically corporations. Individuals are protected by corporate liability and bankruptcy laws. (Rule zero of business: Pay yourself first.)

George writes:

I think Russ and Matt are talking two sides of the same coin here.

Russ is absolutely correct there's no doubt the information age lowered the bar for writers and authors with the rise of blogs etc to share their thoughts. For a consumer of information this is amazing we are in a sea of ideas. Matt is correct to point out that not all of these ideas are equal and there may be a tradeoff in quality for volume (I'm skeptical but open to data on this). The issue for Matt as I understand it is that with so much information out there now we face a signal/noise problem. How to efficiently sort through that sea of information? My understanding is that network effects are really important in this environment as are initial conditions. This paper talks about this exactly:

Starting conditions are important in popularity contests. In this study they asked students to rate unknown songs, allowing them to download the ones they liked the most at the end of the experiment. To show the effect of social buzz, they let one group of raters see how many times a song had been downloaded before they could rate it; the other group was not allowed to do so. Both the ratings and the market share of downloads were more volatile when raters could see the record of previous downloads. Early winners greatly benefited from the initial buzz and gained a disproportionate market share. This social-snowball effect creates an outcome that is not exactly winner-take-all (there is still a bit of room for the losers), but it is awfully close: the winners enjoy a disproportionate share of the market and the profits, and the rest get the crumbs. What the authors showed that is particularly interesting is that the really good songs make it anyway and the really bad ones lose anyway. It is the ones in the middle that get lucky or unlucky. But this is exactly where most of us are.

In this context the big information monopolies Amazon, Google, Facebook can unfairly set the initial conditions through preferential ranking, displaying or not displaying features and relevant information etc. The issue comes when these factors dominate and drive the market instead of consumer preferences. My understanding of Matt when he says "Amazon is not a market" is that consumers and producers don't meet in an open market they have to go through Amazon but Amazon is becoming more than a traditional gatekeeper to an otherwise competitive & open market. Amazon can use it's monopoly position in personal data in one market to compete with "upstarts" in other markets. In order to compete with them you actually have to compete with them on the data front. One market is a barrier to another market. In principle there is an argument that this is anti-competitive behavior.

The data supports Matt's thesis that monopoly power is on the rise.

Also Matt is not alone on this front. People like Luigi Zingales also put forth convincing arguments to reform data, and use anti-trust law to break up the monopolies.

Tom writes:

This was a good discussion with valid points on all sides. However, while I agree with Stoller's concerns, and much of his thesis, his view is myopic in that he sees Google, Facebook, etc. as somehow commandeering their power by brute force when in fact, their power is a result of the demands we place on them. In other words, we demand the convenience and services they provide, and in order for them to provide what we demand, they must have the power they do. I agree that we should be concerned about their power. I agree that their power can be abused. But any discussion of restricting their power should include the acknowledgement that we will likely be restricting, in some way, their ability to provide what we demand of them.

BTW, Stoller is flat wrong on the profitability of Amazon's retail operations. The bear case against Amazon's stock has always been that Amazon cannot bring it's retail operations to profitability (e.g. by raising prices) without simultaneously diminishing it's power and ceding ground to competitors.

Nicolas Petit writes:

Thank you for a revealing episode. Theory without evidence is ideology. And this got very clear here.

The excessiveness of certain statements is spectacular. Russ and listeners spotted the most glaring. Let me add a few:

- "Companies don't just fall because of a natural cycle in the market". What about Yahoo, Blackberry or even Kodak to name a few? And more generally, should we want to set in stone a policy rule that companies should exit, even when they are efficient? Obviously not.
- On startups like Onavo and TBH selling out and a policy of prohibition of big tech M&A: those "people will actually start businesses and they will try to compete with these guys instead of just trying to sell out to them": most acquired startups work on complementary applications which are not designed - neither today nor tomorrow - to compete head to head with the platform. A policy of this kind would simply deny functionality and innovation to users, full stop.
- But this last one really left me voiceless: "they are not showing accounting profits because they don't want to. But, they, they--you know, Jeff Bezos is very clear. He says, 'Anything we make, we invest back in.' They are spending a ton of money on expanding operations. And at any point they could just choose to spend a little bit less and show a profit": taken on the word, the logic here is that any large firm that invests - say in R&D - is in fact trying to hide fat monopoly profits, and should thus be dismembered. I cant help but think of the perverse incentives that would result from this idea.

MH writes:

[Comment removed. Please consult our comment policies and check your email for explanation.--Econlib Ed.]

Kevin Smith writes:

Two links supporting that Google and Facebook are a potential threat to our political order and that at some point we may need to reign them in:

Andy Wagner writes:

Stoller seems to greatly underappreciate the roll that publishers and agents have played as gatekeepers in the publishing world for the past century or so. New authors have succeeded only through years of persistence to escape the publishing "slush" pile.
Today, a book can be self-published for a nominal fee and a become a bestseller. To Stoller's point, some of these are not great works, but some are and the removal of the filtering layer--the New York agent/publishing oligarchy enables this. (Some examples: Fifty Shades of Grey, Andy Weir's The Martian, Mike Duncan's Storm Before the Storm. Weir started blogging his book serially, getting fan feedback and enhancing the story cooperatively. He self-published on a whim and became a best seller without ever getting an actual publisher until late in the game. Duncan created The History of Rome podcast and only years later was asked to write the book based on it's proven track record.
So while "average author" pay maybe down, the barriers to entry for new authors are way down and the secondary and supplementary income for people who would not otherwise have been authors needs to be considered.

JRo writes:

[Comment removed. Please consult our comment policies and check your email for explanation.--Econlib Ed.]

Andy Wagner writes:

Podcast from a writer/agent/editor perspective:

Short version: Major publishing--the top five publishing houses--have become an oligarchy, not a market. Meanwhile, Amazon/CreateSpace/Kindle books have made it possible to go around the major publishing houses, and if you work hard on marketing, and are lucky, you can gain success.

Seano writes:

Matt's comments about beer really don't make much sense and I think are actually the opposite of what is happening in real life. I'm not sure if he actually looked at the beer stats or is just basing his arguments off anecdotes. Here is an article from the Atlantic basically saying the exact opposite of what he claimed.

Plus here are the actual stats which are totally contrary to what Matt claims:

It seems he makes quite a lot of claims that can't actually be backed up by any evidence. His way of arguing seemed to be getting angrier/louder instead of providing good facts which in general isn't a very good way to convince people of an opinion.

Libertarian Heretic writes:

[Comment removed. Please consult our comment policies and check your email for explanation.--Econlib Ed.]

Rob Wiblin writes:

I found the arguments about Google and Facebook's excess power over what information we are exposed to to be quite persuasive. They will be very hard to displace due to network externalities and information economies of scale. I'd love to hear more suggestions for what we can do about that.

It's a shame the conversation ended up being dominated by discussion of Amazon and effects on producers, where the case for harm is much weaker. Amazon may end up having monopoly power over consumers at some point, but not yet. And like Russ I find it hard to believe it's harder to break into writing or music production than it used to be.

Charlie T writes:

I like to add a comment on a previous podcast on Modern Monopolies with Matt Stoller because the comment section is closed. I understand that that many listeners as well as Russ were critical as well as skeptical about Stoller's claim. But I feel compelled to give my opinion about the issue.
I'm getting empathetic for Stoller because I, too am feeling a sense that the modern technology companies are somehow disrupting our society. In a good way? In many aspects yes, but I can't say for sure about all of it. Otherwise, a reasonably knowledgeable and pretty articulate person like Stoller wouldn't be so infuriated about Facebook and Google, granted he may have used some bad examples.
Needless to repeat your conservatives' side of the story but since I'm Chinese and now live in China, I'm going to use the examples of the Chinese tech giants to support my argument. The situation may be better in the US and you're not feeling it as strongly because 1. You're old (no disrespect but older generation are much less impacted by these tech companies and usually you experience all the perks and little of the down sides), 2. The monopolistic phenomena is just as bad or worse in China. Of the BAT, Alibaba dominates the ecommerce market, they use extremely predatory measures to maintain its status as China's number one in consumer online retail. They work with small start-ups in the name of partnership then steal their ideas and kick them to the sidecurb; Vendors, especially large and well-known brands who sign with them are barred from signing with other ecommerce platforms such as; and like stoller said they just keep buying smaller startups if they can't kill them and will keep on growing bigger and bigger until everyone relies entirely on them. Alibaba creates an ecosystem that makes it hard for any other players to exist. It's like a green desert where it looks like there's prosperity and a lot of economic activities but there's a lack of variety. I think what Stoller says that we have less choices really means that we have fewer marketplaces. The upstream suppliers are getting better off (arguably), but we have fewer middleman now. Is that necessarily a good thing? Alibaba is now in the supermarket and logistics business too, like Amazon. They may be revolutionizing the next generation of logistics and food supply chains but I'm highly skeptical of how they're going to help all the small farmers and ranchers, fisherman, etc. should all of them bond together to become a large efficient agriculture supplier? Signalling the demise of small businesses? Even George Soros criticized google and facebook, companies that he owns shares that they are too big, and too have become too dangerous.
Next of the BAT, Tencent is the largest by far on social media with Wechat and QQ, they're just the same. Tencent is notorious for stealing others' ideas. They can copy and produce a game with a whole system of marketing and app development in a matter of days. Baidu on search engine (I personally prefer Bing cause baidu is too focused on making money with ads that they have neglected many of the social responsibilities they have, and have caused grave problems because people took false medical advice made by fraudulent companies that paid the most for rank bidding, which Baidu cleverly dodged all liabilities). These tech giants are too good at making money, not to mention the amount of personal information given WILLINGLY to them. In America, Netflix consumes people's time that they're so easily forget and just keep on binge watching shows after shows. Same thing is happening in China. All young people do is to let themselves be addicted to the games and social media tools. The fact that these companies' presence are everywhere is somewhat akin to those sci-fi films where one corporation dominates everything but the future isn't that bright as we hope it too be and somewhat bleak. I must say our reliance on these technologies are on a different level than our reliance on automobiles. Because when I subjectively know and feel completely immersed in those experiences I feel like I don't have a choice at all! I couldn't pull myself out if I want to!! I think many young people would agree with me on this. So who's the matrix or the Evil Genius, if you know what I mean?
Lastly, I'm not here to offer any solutions because I don't know enough about antitrust, may be there's an argument for natural monopolies here, or what the future is going to be like? Or what we really need as a society.

[entered by Econlib Ed.]

Comments for this podcast episode have been closed
Return to top