in_beta Podcast Transcript
Transcript of in_beta podcast episode, “Should the Tech Giants of Silicon Valley Be Broken Up?”
Charles: Welcome to in_beta where we discuss the big questions around human rights in the digital age. I'm Charles Bradley, GPD's Executive Director. It's a relatively little-known fact that for most of the twentieth century communications to the U.S. was almost entirely controlled by one entity, the Bell Telephone Company.
Charles: If you had a telephone call, whether local or long distance, it would've been on a telephone manufactured by Bell through Bell telephone lines speaking to bell operators. The Bell System, or Ma Bell as it was affectionately called, was a government sanctioned monopoly. Its motto was quite frank about this: One system, one policy, universal service. By the 1970s, it was the largest and most powerful company in the world. By all accounts, it seems to have delivered a mostly reliable, efficient service to its customers. Monopolies after all aren't without their advantages and its experimental research division, Bell Labs fostered some of the century’s most important scientific breakthroughs, from the development of the transistor and the laser, to Claude Shannon's landmark work on information theory and cryptography. It was an institution, part of the fabric of American life, and like many powerful institutions, it had its dark side.
Charles: In its early days, it would simply cut the telephone lines of rival telecoms operators or send people to break their machines. Later, when it was dominant, it forced what competitors were left to pay heavy fees to access its network, and for all its contributions to science, Bell wasn't shy of holding back progress when its own commercial interests were at risk. We now know from archival research that technologies which emerged in the 1990s, including the answering machine, had in fact been invented by Bell Labs in the '30s, but suppressed as a fear they would mean people spending less time on their phone networks.
Charles: Bell's dominance was eventually to end. In 1982, following an antitrust lawsuit supported by the U.S. Justice Department, it was broken up into seven smaller companies. In his book, “The Master Switch,” law professor Tim Wu sees this moment as a kind of big bang, unleashing a huge wave of innovation and competition in the U.S. telecoms market, even suggesting that without this lawsuit, the development of the digital environment might not have even been possible. In a way, it was remarkable, looking back on this history from 2018, is not how much has changed, but how little. AT&T, which emerged out of the broken-up Bell Company, is the world's biggest telecommunications company. At the same time, companies like Facebook and Alphabet, once upstarts, are beginning to take on some of the qualities of monopoly, in their aggressive purchases, increasing tendencies towards vertical integration, and their unprecedented influence over our attention and time. In this episode, I want to explore this idea and the questions which come out of it: How are these companies the same, and different? Are centralizing tendencies in communications inevitable? And if breaking up Bell was necessary, then does it follow that these companies should be broken up too? And to discuss it with me, I'm pleased to welcome Gene Kimmelman, President of Public Knowledge and GPD advisory board member. Gene, welcome to the show.
Gene: Thanks for having me.
Charles: So, Gene, from a layman's perspective, there seems to be clear similarities between Bell and the sort of present crop of tech companies. What are your thoughts on this comparison? How are they the same and how do they differ?
Gene: Well, there are some very important differences. I think the way we need to think about it is, for the discussion we want to have, you always need to start with what are the fundamental underlying economics of the market that you're talking about. For the Bell Monopoly it was the wired infrastructure, which actually originally had started as separate companies competing against each other, but failing to work out interconnection arrangements, it became a monopoly through an agreement with the government. But the fundamental economics were that there were multiple networks and it did not seem workable. And then as it became a monopoly under government oversight, the economics unfolded to where portions of what was happening in the telephone network were seemingly more competitive. You could start buying equipment that could connect, other than the phone that was offered by the Bell Company. There were new companies developing long distance services so that you could actually connect with different companies to make calls across the country and globally. Then there was a piece of it that was a remaining monopoly, which was the local connection, which no one seemed able to duplicate.
Gene: And so out of that economic environment with government oversight, there were a variety of behaviors that were anticompetitive, like blocking connections to the network for new electronics companies that could offer various kinds of phones, fax machines, and new kinds of equipment that the monopoly did not want to have connected. Similarly, different long distance companies were being blocked. And so there was a clear need based on the economics of the underlying monopoly portion, the behavior of the company controlling it, to break it up. So then let's take that kind of thinking and put it into the tech sector, where this is not one simple communications network. There are layers of applications and operating systems and a variety of new digital services that do connect to an underlying communications network, but those tech companies don't control that network.
Gene: We are seeing the growth of these digital platforms. Facebook for communicating in a particular way. Google for search, and then a lot of different products that connect to it. Amazon in the U.S. growing probably more and more globally offering delivery services and connecting equipment, and new opportunities for people to actually shop online. These companies are growing and what we're seeing is that there are enormous, what economists would call economies of scale and scope. There are cost savings caused by growing more broadly and offering a deeper set of services that keep offering consumers better options and lower prices. At a certain point that can become something like a communications network where everyone is connected and everyone needs to be connected. We haven't really gotten quite there yet, but we're seeing signs of dominance. Google in search, Facebook in social networking. These things are beginning to look a bit more like what the Bell monopoly was, but there are clear differences both in how they've developed and how they're behaving. What we have not seen yet is pervasive blocking of new services from these distribution platforms. We see some specific allegations of problems with Google on shopping and Android that are being adjudicated. But we don't have the pervasive abuses that we had with a monopoly that was broken up. So I think there are some similarities but some important distinctions and differences.
Gene: And also the final thing I'll say is that monopoly for the phone network was always regulated at the same time as there were questions of antitrust abuse. The tech sector has not been regulated. And I think besides looking at economics, when we see something that we find problematic and illogical for consumers — why aren't they interconnecting certain services or why can't I use the same device on different kinds of services — I think we have to examine the problem for the consumer,for the citizen, and for the society, and look at the harm that is occurring, and what would be the best tool to address the harm. Many of the tech discussions turn immediately to antitrust when in reality a lot of the problems may be better addressed by other policy tools like opening up APIs, interconnection to differing platforms, making equipment operable across multiple platforms, or not allowing totally closed down networks. These may be better policy initiatives that can truly address specific problems as opposed to some generic antitrust “just break up companies.”
Charles: It's really interesting to get those comparisons there. I think again from a layman's perspective looking at these different industries, it seems like the outlines of these industries, or sort of sectors within the tech industry, are still being drawn. So there are many claims to say that obviously Google is a monopoly of search in the percentage of search that is run through the Google platform, and is search different to a social network like Facebook. So is the monopoly model then not a useful way of thinking about tech companies and the power that they represent in the 21st century?
Gene: I think the monopoly model is one piece of the analysis. But again I think we should start with looking at the economics that are pushing the market to develop in a certain way. Let's use search for example. Google has grown enormously and is dominating more and more in search — more in Europe than in the U.S., but becoming increasingly dominant everywhere. If we broke Google into three search companies, what would happen? Would you somehow see new innovations from one versus the other? Would one be really competing against the other? A second line of questions might be if you split them into three, would you have a problem with an oil company buying one of them, or would you have a problem with a major bank buying one of them, or Walmart buying one of them? Or turn it around, would you be comfortable if Apple bought a third of Google search or a telecom company bought a third of Google search? So the monopoly analysis is complicated and difficult because it can deal with a simple problem of a monopoly being abused, but then you still have to step back and say, what is the remedy? What am I trying to accomplish?
Gene: And invariably we're talking about areas where we want consumers to have more choices. We don't want platforms to dictate our expression, our choices, our ability to speak, our ability to receive information and engage in commercial activities. But if size is the problem, but breaking them up could lead to some other combination that could be a similar problem, you have to look at it and say, is the better way to deal with this to prevent discrimination in that function? Should we prevent a Google favoring itself in search as opposed to saying it shouldn't be allowed to have this much search? So these are just different policy tools that align with different goals that we would have and I would say fundamentally need to align with the underlying economics. Let me go back to the Bell example. We broke up the monopoly and we had a explosion of innovation in new companies offering both electronics and new services connecting to a monopoly network. In both Europe and in the U.S., we maintained a fair amount of regulation over the telecom sector. In the U.S., the power of the underlying monopoly became so prevalent that they wore down the regulations and were able to reintegrate.
Gene: Now, it's not just that they were powerful. The economics supported their drive to grow again, even though they had been broken up. And now we really have only a few firms dominating, once again, not one monopoly, but still not a vibrant explosion. In Europe, you maintain a much more thorough regulatory platform around the underlying telecom monopoly and you've had much more vibrant competition sustained in add-on services, mobile, other long distance type services, other equipment. So there again, I think it has been a regulatory set of tools that have preserved the goals for consumers — the innovation, the choice, and the broader freedom of expression — that has been a big struggle in the U.S. since the reintegration of many of these companies.
Charles: Another way of looking at these platforms — not just their role in a particular sort of search or social networking — but the control or collection of data and the way in which they're using that to sort of horizontally integrate in other sectors or innovate off of that. Do you see there to be an antitrust argument or some sort of a competition analysis argument in how they use their position of power or dominance in their existing market to enter new markets or deliver new products and services off of their existing dominance in their industry?
Gene: So control of data and the ability to gather data I think is definitely an important part of an antitrust analysis, has been under-assessed in the past, and I think people are becoming increasingly aware that it should be a significant portion of the analysis. But then again, it's not just the gathering of the data. You have to look at what is being done with it. Is it being used to draw advertising revenue? Clearly some of that is. Is it the fuel for the power in the marketplace? Clearly there is quite a bit of that. But then you have to look at an advertising market and say, not just Google controls this percentage of online advertising and Facebook controls the rest of it, and therefore they are two almost monopolistic players. You have to ask where else people are advertising in a similar fashion. Invariably it's still is on television and other video distribution products, so one needs to be careful what one is trying to challenge.
Gene: And now Amazon is rising in the online advertising world. So the fact that there's a lot of data amassed is important and a key input to an analysis, but it's not just an antitrust analysis. It should be a personal data protection analysis of individual control and autonomy. Even if there's not a competitive harm, there still may be an imbalance in intruding on the individual's privacy rights and right to control their own information. So that would have nothing necessarily to do with competition, but a different policy goal. And both the European directors and some of the U.S. discussion are trying to address those concerns. I think that we don't know fully how data can be used to steer or harm the competitive process. And that needs to be evaluated. In other words, it's not clear that by knowing a bit more about what you or I do that the company that has a unique advantage over every other company. There might be other ways in which others can compete equally and we need to understand that.
Charles: I think the point there around the personal control of data is quite interesting, and when the internet was invented and proliferated around the world it was going to be this disruptor, sparking all this innovation, distributed network around the world. And what we've actually seen is a doubling down on concentration of the number of platforms and corporate power. What do you see as the relationship between the concentration of corporate power on those wider human rights questions?
Gene: I think there are very significant human rights questions and very significant questions about the future of democracy and equal rights under law. If too few companies are dominating, even if they're not necessarily violating competition principles, if too few companies are dominating markets, invariably they tend to dominate political processes as well. And I think there's a separate argument based on democracy, equal rights, and therefore human rights framework to be cautious about allowing too few companies to dominate in a society. Again, it can be that there is a strong antitrust framework that keeps a number of large players at bay and prevents domination. And yet if they have parallel interests politically, that's dangerous to democratic institutions. And so again, I would say it's not an antitrust tool, it's more of a broader policy perspective about the key cornerstones and building blocks for a vibrant democratic society that need to be the tools assessed against dominance in politics.
Gene: So there could be a reason why, even if a Facebook, a Google, an Amazon, an Apple are competing quite effectively against each other that they're preventing smaller players from really emerging. They're preventing diversity of viewpoints and voices in a marketplace and therefore in a political framework. And they are too powerful over citizens' ability to assess their own self-interest and their own political views. So that could be a very separate policy framework for wanting diversity of ownership, more small distributors and suppliers and chains of distribution in a society. I think those are extremely relevant parallels to antitrust.
Charles: Looking at where these discussions are taking place, obviously a lot of the companies we're talking about today are American companies based out of the West Coast, but obviously have global reach and have global impact. Where do you see in the future the key policy discussions taking place and where should we as human rights defenders be engaging to strategically shape this conversation to ensure that what you've described actually realizes in the future?
Gene: I think that's very different. The geopolitics around this are enormously complicated. Every one of the major policy concerns resides at a local, regional, national level. There is no doubt that the nation state plays a critical role in each one of these policy analyses, and there needs to be a substantial focus on how those local policies are developed, but they are not divorced from global trade and global geopolitical power disputes. And so I think trade agreements, I think in this space of technology and telecommunications, the International Telecommunications Union is critical. Wherever there are norms set, wherever there are standards invoked and pressure applied from nation states, from the large multinational corporations, we need to have a broad open policy debate about how those norms, how those standards, to the extent that there are trade agreements or treaties, how those binding obligations are established. Civil society needs to be at the table. All stakeholders need to be at the table. So unfortunately it's complicated. One needs to be engaged everywhere from locally through the major, multilateral discussions that affect policy.
Charles: And how would you say the antitrust or the more economic analysis and competitive analysis community could be working more with the digital rights or human rights online community to help shape that policy landscape?
Gene: There is a moment in time now when people are starting to realize with the explosion of the digital economy and the enormity of the growth of the major tech companies that there is a power and a dominance concern. And so antitrust needs to be looked at carefully, the tools need to be resharpened, refocused. There needs to be greater emphasis and attention paid, resources applied to that form of analysis, but it should never be divorced from other nation state, oversight, regulatory, legal frameworks, and other global norm setting tools. And so this needs to be a comprehensive approach to all policy tools on the table because each one of them both contributes to or can be harmful to fundamental human rights. And so I think we're seeing a moment in time when antitrust rises to a more significant level as the tech sector explodes. We also see the underlying telecoms sector remaining quite monopolistic in transmission functionality, so that becomes a more important piece of the puzzle. But again, just one piece of the puzzle. Anyone who is just focusing on antitrust and thinking it will solve the problems I think is going to miss quite a bit of what we truly need and policy protections for human rights, for supporting democracy, and for making markets function to benefit the freedom of expression.
Charles: So I'm going to put you on the spot and I think I know what you're going to say to this question, but the big question about this podcast is, do the big tech companies need to be broken up?
Gene: I think the big tech companies need to be scrutinized carefully and we need to start by then analyzing what about them is causing a harm. What is that harm? How does it relate to the fundamental economics of what those companies are doing, what the tech platforms are doing? Is it the natural outgrowth of the economic forces or is it some nefarious behavior? Once we've defined what a harm is that we want to address, protection of personal data for our own privacy, protection against domination and controlling viewpoints or controlling markets, whatever that is, I think we then need to look and say, what is the tool that truly will address this harm, and that we have confidence is sustainable? If you break up companies and they reintegrate, have you accomplished something? If you just change the name of who got one-third of that company or one-fourth of that company and someone else starts doing a similar thing, have you accomplished something?
Gene: Sometimes the breakup tool is not the correct tool if it runs counter to the underlying economics of what drove dominance. So I caution against just immediately looking for a breakup. But having said that, I do believe there are significant concerns about the acquisitions of some of the large tech companies of smaller players who no one really knew whether they were going to compete against a Facebook or a Google in the future, but who clearly are potential rivals. I think there ought to be a much more rigorous antitrust analysis and crackdown against those types of acquisitions because as we see these companies gain dominance, the question will become who can be the new small Google or the new small Apple or Facebook to change the competitive landscape? And unless you allow those smaller players to survive, you have no way of ascertaining whether competition is even possible. So I think it's less of an immediate look at breakup, but more of a look at what are the harms to be addressed and is there a stronger need to prevent greater growth of these companies through acquisition.
Charles: That's a great call to action and a great way to end this podcast. Thanks Gene so much for joining us for this very interesting conversation, and we'll be putting up a few links and the link to the Public Knowledge website on GPD's website, and we look forward to welcome you back on the show at some point soon. Thanks again.
Gene: Thanks for having me.