Intro. [Recording date: August 23, 2021.]
Russ Roberts: Today is August 23, 2021, and my guest is Glen Weyl of Microsoft, where he is the Office of the Chief Technology Officer Political Economist & Social Technologist--a mouthful. But, it does create a great acronym, which is OCTOPEST. Glen is the OCTOPEST of Microsoft. More simply, he advises Microsoft's senior leaders on macroeconomics, geopolitics, and future technology. His mission statement there is: I help a corporate octopus reimagine itself as democratic infrastructure rigorously accountable to those it holds power over.
Glen was here in May of 2018 talking about radical markets.
Our topic for today is antitrust. And, this draws on a conversation Glen and I had with the American Bar Association's antitrust sections magazine, Antitrust, on what policy objectives antitrust law ought to advance and how it can achieve this. A transcript of that conversation will appear in the published issue of Antitrust Magazine. And, we'll post a link to that when that issue appears.
Glen, welcome back to EconTalk.
Glen Weyl: Thanks, Russ. It's great to be back with you, and such a great conversation a few years back, and I look forward to talking to you again.
Russ Roberts: Me, too.
Russ Roberts: Now, traditionally, antitrust has been focused on making sure that firms didn't gain monopoly power or market power more generally, which would let them raise prices on consumers in the name of higher profits. How should we think about this model in a world where so many firms charge nothing out of pocket to their consumers--Google, Facebook, social media--generally? What role does antitrust have in this world? And, it's a world that a lot of people are scared about, and, one has got to do something. But, the traditional tools of antitrust are not so obviously applicable. What's your take on that?
Glen Weyl: So, first, you said, traditionally, antitrust is focused on X. And, the question is, what would it means by tradition? So, traditions have changed in antitrust quite a bit over time. The origins of the antitrust law are quite different, I think, than the way that they were, say, applied in the early part of the 20th century, and then from the New Deal, until maybe the 1970s. And, then, again--so, there's been a lot of evolution in the way the doctrine has been applied over time.
But, I think its origin is best summarized by a quote that I don't directly have in front of me, but was given by John Sherman, who wrote--after whom the original of antitrust law was named on the house of the Senate. And, he said that, 'Just as we would not tolerate a king as a political power, we should not tolerate a[?] autocrat over the sale of the--or distribution or transport of any of the necessities of life.' And, as such I really think antitrust is fundamentally about power, and in particular, about what it means to have a democratic society.
And, the way that it has often been interpreted, even in those early periods, is that the role of antitrust is to block, or break up, or stop accumulations of power.
But, I think the thing that's confusing about that is that if you think of the 18th century, there were two great revolutions--the French and the American. And, the approach of the American Revolution to concentrated power was to make it accountable to the people. And, the approach of the French Revolution was to sort of kill or break up those concentrations of power.
And, I think that the former worked much better than the latter.
And, you know, the British, sort of gradual revolution of the 19th century was similar. It didn't end the monarchy. It made it gradually accountable to the people.
And, I think that we need to get much more back to that spirit, back to the spirit of using antitrust as a means to make corporations accountable to their stakeholders democratically, rather than using it as a means to just block or break up agglomerations that really results fundamentally from what you could call economies of scale, increasing returns, etc.
Russ Roberts: So, that sounds really beautiful. I could hear the music swelling in the background when you use the word 'democracy' and 'accountability.' But, that's a tough--it's not an easy thing to do.
So, let's talk about what you mean by that, to start with. We'll get to how you might achieve those goals. But, I don't even understand yet what you mean by that.
So, let's start with an even simpler, more basic idea. Let's talk about power, because that's at the root of how I think you think about this. When you say a corporation is powerful, what does that mean?
Glen Weyl: So, I think the reality is, power is one of these many concepts in Social Science that we can model in a variety of ways, but we've never managed to fully grasp. I think that antitrust has come. And, I think it's reasonable for it to focus on market power--that is, the ability of firms to--the limits on the options that a consumer effectively has, or a worker effectively has, or any counterparty to trade effectively has in an interaction with the firm.
Now, the existence of market power doesn't necessitate charging a higher price. It's a condition. It's not a choice.
So, you can have power, but [?] exercise it in a variety of ways. And, so, focusing on the existence of the power rather than on it's--a particular way in which it's exercised, I think allows you to get around the concerns about these companies not choosing to charge prices and instead finding other ways to exploit their market power.
Russ Roberts: So, let's talk about that in more detail on the power side. So, again, I'll use the word 'traditionally,' recognize that tradition changes over time, as you pointed out, nicely. But, traditionally, the argument would have been, say, a monopolist or an oligopolist--meaning, a market where there's only a few sellers--one example would be from 50, 60 years ago, the U.S. auto market. Before, there was a large influx of imports. There were only three major suppliers of cars. Their competitiveness--sometimes three firms could compete quite aggressively.
But, the argument would be maybe they didn't try so hard--that it's either an implicit cartel or form of implicit collusion. Explicit collusion is against the law in the United States. But, it was not that competitive a market. And, when the Japanese and others came in with competitive products, the price fell, quality improved, and consumers were better off. That's sort of the traditional model of market power: They had some competition from outside; reduced it; and, it was an improvement in the wellbeing of consumers.
In today's world, how would we think about that?
In the car market, I only had three options in the United States before there were a lot of imports. And, that restraint on my choice set was hard on me, because it allowed the firm, sometimes, perhaps, to charge more than they otherwise would, offer fewer options that I might want. What's going on in today's world that mirrors that kind of power?
Glen Weyl: I would say that the most interesting examples of accumulating market power are ones that go along with tremendous gains and changes to the types of opportunities we have.
So, if you ask, why did those U.S. firms end up with so much power? Well, fundamentally, it was because the internal combustion engine and the car was one of the most important revolutions of the 20th century. And, along with it, the mass production techniques--assembly line techniques pioneered by Ford Corporation--were one of the most important social and organizational innovations of the 20th century.
And, it was those combined revolutions that allowed for that market power to accumulate in the first place.
And, eventually, there was a mutation; those become standardized; we sort of ran into the limits of what those could achieve. And, competition set in, and we were able to fully take advantage of the benefit that was brought by those disruptions to the nature of the market. It took a long time, though. And, fundamentally, the question they face is, 'How can we accelerate the process of fully taking advantage of new transformational technologies?' And, today, I think we see very similar dynamics.
The Internet, certain forms of what's often called artificial intelligence, personal computing have transformed so much of how we live. And, yet, we've been slow to be able to harness them fully. The productivity statistics don't reflect there being these fundamental transformations going on. Something is standing in the way of that general-purpose technology, working itself out to the extent that things like air conditioning, sewage, electricity, the automobile did.
And, the question is: how do we accelerate that process more effectively? How do we bring the full benefits that these new technologies can allow to the people who they serve?
And, I think you see that with Amazon today. You see that with Facebook, Google, in the cloud computing market that my employer Microsoft participates in. I think there are many examples here of limits to fully taking advantage of what is offered by these new technologies.
Russ Roberts: And, you're suggesting--just to add a footnote to what you said about the original story that I told. This massive opportunity that was created by this technological innovation of the internal combustion engine created a giant prize for the firms that could turn that invention into a product. And, they did it quite well, you could argue--or they didn't. Who knows? It's not even an answerable question to me. But, obviously, there was progress. It took a while.
The early cars were not as good as the later cars, and they weren't as good as the cars we have today. And, that inevitable process of innovation, tinkering, trial and error, consumers changing, landscape changing, so that what people want in their car is totally different--all of that is going on at the same time. And, there's this big prize called profits that are motivating firms to try different stuff. And, no one knows in advance what's going to work. And, in the early days of cars, there were dozens and dozens of producers. When we get to the three largest ones, that took a very long time. And, no one knew in advance which three would be--if it would be three--that it would be three.
And similarly, you're suggesting that in today's world, we have this explosion of innovation as people try to use the Internet in creative and interesting ways. But, they're limited by the fact that you do have to cover your costs. And, there's a prize or profit out there. And that, you're suggesting, has slowed the ability to use that tool as effectively as possible. Is that a fair summary of what you said?
Glen Weyl: Yeah, absolutely. So, profit stands in two ways, right? And, it's really a result of a fundamental economic idea, which is that when you have an increasing returns process--
Russ Roberts: Explain what that is.
Glen Weyl: Increasing returns is a case where all of us together can achieve more than we can, the sum of us separately. The whole is greater than the sum of the parts. That's the general way in which I think about it. So, cities are a great example, networks are a great example, etc.
Whenever you have a process like that, the basic idea of economics--classical economics--is you're supposed to pay out to everyone their marginal product--how much, on the margin, they contribute.
Now, with a decrease in a decreasing returns context, you can make a profit and still pay everyone their marginal product. Because, if you have decreasing returns, then the marginal product of everyone in total is less than the amount that you produce. And, so, you can take the rest as profit. But, in an increasing returns process, that's not true. If you try to pay everyone out their marginal product, you'll consume more than everything that's created.
And so, you have to run a loss in order to pay everyone out their marginal product in increasing returns process.
Now, if you try to make profits on an increasing returns process, what you do is either you make some profits, which gives you an incentive to create the thing in the first place. But, then, you need to pay everyone way less than their marginal product or charge them way more than the marginal cost for you to produce it. And, therefore, you exclude participation.
But, more importantly, you reduce the benefits that it has. Because it's an increasing returns process, it wants to be used by lots of people. But, you limit that as you try to take out profit.
Russ Roberts: So, some would say, 'Now,'--
Russ Roberts: I know some listeners are going to be lost at this point because of the complexity of marginal this and marginal that. But, to make it simple--and I can't believe I just said this, 'marginal this and marginal that.' It's the lifeblood of our field. But, the point is that, if I can make it even a really brutal simplification--
Glen Weyl: Please, I'd appreciate it--
Russ Roberts: In these kind of products--the network kind in particular--where the more people that are playing in the sandbox, the more pleasant is to be in the sandbox--as opposed to the traditional world or the decreasing returns world where as more people come in the sandbox, there's less room for other people, and it's no fun to play there, or it gets less and less fun--here, it's more fun. And, we understand that from the virtual world that many of us spend a lot of time in.
But, if you want to have lots of people in it, that goes against what is usually what a profit-maximizing firm is going to sometimes need to do, which is to constrain how many people have access by price. You're going to charge people a certain discouraging amount in order to capture some of those benefits. And, in doing so, you'll have lost some of the potential social gain. That's, I think, a variation of what you're trying to say.
Glen Weyl: And by the way, of course, as you mentioned earlier, it's not necessary that you do it with price. You can do with another [?], right?
Russ Roberts: Correct.
Glen Weyl: And we can come back to those later.
Russ Roberts: Yeah, we'll come back to that.
Russ Roberts: But, the point I want to--
Glen Weyl: But, that's exactly right.
Russ Roberts: But, the point I want challenge is, is that increasing returns to scale--and to get away from the jargon--products with network benefits, products where, markets where service is better, where you want to be around more people rather than fewer--that's just a reality of what you're just describing.
So, one counterpoint to your claim that this is a problem, one way to say it is, 'Well, just this is the way the world is.'
Glen Weyl: Well, I don't think it's a problem. I think it's great--
Russ Roberts: Okay, carry on--
Glen Weyl: I actually think it's very far from being a problem: It's the source of all value.
I mean, if there were no network effects, if there were no increase in returns, none of us could live better than we could live in isolated huts without even our families.
So, increasing returns are the source of all civilization and the source of everything we want. It's not bad. It's great. It's the best thing that there is.
The problem is not with increasing returns. The problem is with the standard capitalist logic, which is optimized for setting of decreasing returns and not optimized for setting of increasing returns.
All the theorems that we just read about market efficiency, all of that is true only under the assumption of decreasing returns. And, it actually is a really interesting historical relationship to the Marginal Revolution and so forth that we could go into, but it may bore[?]--
Russ Roberts: We're definitely not getting into that--
Glen Weyl: There was less focus on that. Yeah--
Russ Roberts: We're not going to get into that.
Russ Roberts: But, let me take an example that we've talked about on the program before and see if I understand what you're saying correctly, and then we'll see if your conclusion follows, for me. I don't see it yet.
So, Adam Smith said that the division of labor is limited by the extent of the market. And, so, what he meant by that was, is that if you can sell over a wider and wider range of customers, as your customer base expands, you can take advantage of economies of scale. And, this has nothing to do with comparative advantage. It has nothing to do with David Ricardo's insight.
It's just that pure size, in theory, up to a point at least, has these incredible productivity benefits--because you can rearrange how people work. You can use capital with those human beings to produce things you would never be able to produce--and, the obvious example is, if you're selling 10 or 100 million cars a year, you're going to use a different technology than if you're selling one car a year. If you have a market of one, you might hand-make a car. It would never make sense to create a giant production line--the capital that makes cars so inexpensive today compared to the past.
So, as the market for cars expands, such as the market, but your market, your ability as a producer to meet customer demand, you can use all kinds of improved and better ways and cheaper ways to produce your product.
The fact that the firm has to make a profit in doing that is what motivates them to learn about those techniques, to discover those things. You can't just--and I don't understand where you're going to--how's your story of capitalism going to deal with that?
I mean, I love that--I love capitalism. Because capitalism, profit, and loss, gives firms the incentives to discover those new techniques and new products and so on, and it works pretty well. What we need that's new? What do we need that's different?
Glen Weyl: Yeah. So, one question is: what can we do with this better? And, let's get into that at some point.
But, the first question is whether the usual justifications that are used to say that capitalism is the best system, the most efficient or whatever, have any relevance in this context. And, my claim is they don't. Like, there might be other reasons why you'd like capitalism, you know; and we can talk about those, right? And, then, we can compare those to other systems.
But, the usual justification that standard economics would give, which is that competition with decreasing returns leads to efficient allocation, is just not applicable in contexts like that. And, so, we should expect--
Russ Roberts: But, those 300 years--but the 300 years of civilization that you invoked implicitly, right, which were the exploitation of the increasing returns to scale of trade as it went from neighborhoods to countries to globe--we managed to pull that off with that old-fashioned model of people just trying to do the best they could.
Glen Weyl: Well, that's not my interpretation of what happened during those 300 years. Like, I think obviously we moved beyond the forms of social organization that pre-existed capitalism. But, most of what was going on during the most productive periods there, I don't think it's well-understood as capitalism in some simplistic sense of what capitalism means.
I think you take the most productive period in human history, which was the 1920s to the 1950s, the greatest rate of productivity growth, was the period of progressive reform, the period of the New Deal, the height of precisely the set of institutions that were aiming to complicate, democratize, move beyond some of the limitations of 19th-century institutions.
And, even those 19th-century institutions themselves were built on a process precisely of expanding the market by democratizing the institutions within which the market lived[?], because the sort of feudal, ownership-based structure didn't serve the new economies of scale well.
So actually, the way I read the last 300 years, is a gradual process of building greater and greater economies of scale and building democratic institutions to govern those to allow their full benefits to be taken advantage of.
Russ Roberts: I love talking to you, Glen. It's the greatest. That's so interesting. And I can't disagree more. So, let's see if we can get into why, and let you defend your view.
You're suggesting that the productivity of the mid part of the 20th century, in which we saw, by the way, the greatest depression in American history, and maybe world history for the 300 years that we know of, that that was--that productivity came from policies like the Social Security Act of 1936, the price controls of the Roosevelt Administration, the slaughtering of hogs to foolishly tried to keep prices high. Like, really? What are you thinking about there?
What in particular, besides post hoc ergo propter hoc--after this, therefore because of this--what do you have in mind when you talk about these democratic changes?
Glen Weyl: Yeah, great question. So, some of the ones that I would point to as being most successful include the antitrust and utility regulatory policies of 1930s through the 1950s, 1960s; the dramatic investment, the public investment in research and development, which helped create the Internet and so many of the advances, the fundamental technological advances, really, for the second half of the 20th century that were fueled by the investment in dealing with the Second World War.
And, I do believe, by the way, that a lot of the social insurance institutions also played a critical role in allowing entrepreneurial activity to flourish in the years following the World War: the G.I. Bill [Servicemen's Readjustment Act of 1944--Econlib Ed.], and the way that it enabled the dramatic accumulation of capabilities and broadly-spread innovation; the rise of labor unions, and the way that they allowed the basic democratization of the workplace and a more inclusive labor environment that created incentives for productivity enhancement. So, I think all of those things were relevant.
I also think many of the examples you gave, I would agree, were not terribly productive. And, I would argue that if you look at the entrepreneurial markets, which I think you would champion, you'd find that 90% of the things that are tried don't work there--
Russ Roberts: For sure--
Glen Weyl: Ten percent do.
Russ Roberts: Yeah, I cherry-picked them unfairly--
Russ Roberts: I can see that. But, the problem I have with that story--and let's not spend too much time on it because--well, it's entertaining; I enjoy it.
The problem is that the Social Security Administration, Security Act of 1936, was a trivial--just like the income tax; it was trivial at first. They were very small amounts.
The kind of democratization and social programs you're talking about really hit their stride in the mid-1960s--right?--with the Great Society, the magnitudes that really were important.
Similarly, the unionization from 1950s on is declining. It's declining during a time of the post-war era--which was very productive, 1950 to, say, to 1975. Unions are getting less and less important every year in the American workplace over that period. The later part of that--the late 1960s, mid-1960s to late 1960s to early 1970s--is when welfare programs are starting to really grow and get significant. And, that's when we start to get not such a healthy economy.
So, I think it's going to be hard--you're going to have a hard time convincing me on that story. I'll give you one more chance, and then we're going to move on to some other issues in the modern era.
Glen Weyl: That's fine, Russ. I mean, we can go through timing in many different ways. And, I think it's going to be tricky to completely settle things. But, I do think it's quite striking that the period of greatest productivity growth in U.S. history, in world history, really anywhere, was also a period that I think most people would not characterize as a height[?] for, you know, a simplistic notion of capitalism. Like, I think it's [crosstalk 00:26:41].
Russ Roberts: Or laissez-faire--
Russ Roberts: It was not. I would agree with you. It's not the golden era of laissez-faire. You know, people would point to the late 19th century as sort of the closest thing America has had to realizing laissez-faire capitalism--which did have quite a bit of innovation in economic growth. But, let's put that to the side.
Russ Roberts: And, I'd love to spar somewhere at another time.
Russ Roberts: But, for today: In what sense should we think about democratization, which is a--that doesn't excite me. A lot of people find that phrase, I'd say, emotionally resonant. It doesn't resonate with me. So, make the case for why we want our economy to have a democratic partici[?]--whatever phrase you want to use--role. What do you mean by that? What does that mean?
I don't think you mean stockholders voting. I don't think you mean majority rule on referenda within a corporation. You mean something bolder than that for how society I think interacts with these firms, correct?
Glen Weyl: Yeah. So, the first thing I want to be clear about is that sometimes when people talk about democratization, they have in mind a nation-state. They think the United States of America should somehow take hold of all the companies within America, that the same process that governs our elected representatives should govern the corporate governance. And, that's not what I mean.
Russ Roberts: Okay.
Glen Weyl: Why is that not what I mean? Because there's a fundamental problem. Like, the fundamental problem at the heart of democracy is what I would call the misalignment of actual and natural polities[?].
So, what do I mean by that? We know that in the United States, we have a long history of disenfranchising people--whether it was women, African Americans, etc.--not treating people who had just as much of a stake in the country as part of the democratic process.
And, we know that as much as we might respect our democratic ideals, that without those people being included, we weren't really living up to those ideals. Right?
Russ Roberts: 100%.
Glen Weyl: And, in the same way, I think that having the United States control Twitter or Facebook or whatever, would be just as much of a betrayal of the ideals of democracy as would having the white men of property control the United States. Because, the reality is, the user base of Twitter is, like, I don't know, 15%, 20% in the United States, or something like that? It just makes no sense if you were going to have a democratic governance of Twitter for it to be based around U.S. government.
And, the reality is that we have these nation-states that were set in stone at some historical moment as a result of certain accidents.
And, the set of people who are there are now the ones who get democratically represented. The only structures we really have for democratic representation, approximately, are those. And, that is not an appealing basis for building democracy in an era like this.
So, what we need is democratic institutions that are actually adapted to the structures of power and interaction and economic activity, social connection, that govern our world today.
So, that's what I mean by democracy. And then that's, in some sense is much more radical than what's most people think of.
Russ Roberts: It's even crazier--it's even crazier than I thought. Okay. Not crazier--less attractive to me. That's what I mean by crazy. I don't mean it's weird. I like weird.
So, how would that--so when I think of why I like markets, for example, or sometimes why I like markets more than, say, political governance, it's because of the feedback loops and the skin in the game that profit and loss give and that are often missing in the case of political markets.
And, you want to take it a step further.
You want to say, 'Let's water those down even more and spread them across the entire globe.' How are we going to get invited[?], invest in the wisdom and the decision-making process and the risk-taking that's necessary if they don't have much of a stake in it? We're going to say there's a stake, but their voice is going to be irrelevant--just as is in voting, by the way, mostly.
Glen Weyl: Well, I think that the genius of large-scale democracies is how thin they are. So, I think we need like--the United States, I think, was forced to agree on a set of very minimal national government principles, which many people respect because of the diversity of the nation that it sought to govern.
Russ Roberts: Absolutely.
Glen Weyl: And, I think that we should aim for the same thing with the corporations. Like, a lot of people talk about interoperability as an answer to market power. I'm a big believer in that. And, I think that--
Russ Roberts: Explain what that is.
Glen Weyl: Interoperability means that you basically--the things that are most broadly shared are also the thinnest. They're the things with the least substance to them. They just are basic protocols in which people can interact.
Like, the Internet is a great example of this. Like, you hardly even perceive the Internet, right? The Internet is like a background on which other things happen. Whereas, many people feel that things like Facebook and Twitter and Google have a lot more substantive control over our experiences. And, I think that the broader, the more global that things are, the more that they should be that thin and neutral layer.
And, as we get more local--and local doesn't need to be physically local. It can be communities of interest. It can be people who listen to Russ's podcast, etc. Then, we want much more substantive control in this area.
So, I actually think--contra the usual way of thinking about things--that as we go towards those broader and more flexible systems of democratic control, that that's precisely the way of controlling the power at those levels, and making it--actually creating the space for decentralization and localism precisely through that democratization process. Because, the only thing you can get such diverse people to reach consensus on is things that allow a lot of flexibility.
And, I can give you some very concrete examples of this from Taiwan. So, in Taiwan, they've done an incredible set of experiments in participatory democracy using digital tools. And, in the process, I think they've come to a number of judgments that would, at least in a certain form, appeal to you very much. So, let me give you an example of this.
So, they have this participatory governance platform. Rather than having these divisive Twitter-like things, they have these things that actually have dynamics that lead people towards consensus. In the process though, they're forced to reach compromises that leaves space for everyone to do things in the way that they want to do them. And, one example of this was on the issue of gay marriage.
So, Taiwan is a very traditional, Confucian society, where extended families play a very fundamental role. And so, when a couple weds, there's an automatic union of their extended families that comes with that marriage. And, this was a source of great opposition to gay marriage within the Taiwanese society, because extended families didn't want to be connected by something they didn't believe in. Right? So, they had this polis[?] process, which we can talk about if you want, but it's a process that uses modern machine learning-type technology to form clusters and to allow people to do democratic deliberation scale.
And, the conclusion that they ended up coming to which was built into law was that they would create two separate stages to the marriage process for all couples: one in which the individuals who were marrying marry, and one in which the extended families signed a contract with each other in order to form a union.
And, this ended up basically honoring the values of both sides more, because the younger generation wanted the freedom to marry as they choose. But, they actually believe very strongly in the individualistic contractual list values, right?
And, on the other hand, the families wanted a direct recognition of their role in the wedding. And, they got that as well. So, that was a way in which democratic deliberation at scale actually gave more space to everyone to reach their desired outcomes.
Russ Roberts: So, you're suggesting--but I can't grasp it. You're suggesting there was a virtual town hall and an internet-mediated conversation that took place that allowed a set of options to emerge rather than a one size fits all? Would that be a fair summary?
Glen Weyl: Exactly, yeah. And, actually, they've done this for hundreds of issues. And, you can find that many of the solutions, most of the solutions, were about this sort. There were, like, creative ways of finding new mechanisms that allowed sort of changing the space to allow greater freedom to everyone. And, they arrived at that through this sort of virtual town hall. And, I'm happy to talk about how it works. It's actually quite ingenious because it enables really a deliberative democratic conversation at scale, which is something we haven't been able to do in the past.
Russ Roberts: Yeah, one of the things I don't like about the political process--there are many things I do like, by the way. I always like to make it clear. I'm not an anarchist. People like to caricature, sometimes, people's views. I'm a classical liberal. It means I want a restrained role for government and an expansive role for individual responsibility and choice, and voluntary associations of people coming together for profit and for nonprofit and nonprofit ways to work on things.
One of things I don't like about government is usually it's blunt. There's a law. There's a piece of legislation that applies to everyone in a certain way. For just to take an example, everybody on welfare gets the same amount of money based on a handful of characteristics that are all very clearly observable and would have to hold up in court: how many children you have, whether you work or not, and so on. Whereas, in theory--in theory--a private charity, while it might struggle to raise money because of the free rider problem could tailor its solutions in ways that a government agency would struggle to do. I'm not going to romanticize private charity that way, it's all [?]. It has other problems.
But, what you're suggesting is that this underlying technology, which I do want to [?] in a second, but I want to make clear what I think I hear from you: What it's allowing people to do is, even though it's using the top-down government legislative solution, it's allowing for some multiplicity that often does not emerge in, say, U.S. institutions where something is either legal or it's not. It's allowing for a lot more nuance and subtlety. Is that true?
Glen Weyl: Yes. And, I would even go further than that, and say--I actually agree with everything you said about private charity and love private charity, in many ways. I think that's the spirit we should aim for.
But, as you pointed out, it struggles because of the free-rider problem.
So, like, the goal of all these institutions, in my mind, is to find ways to actually empower that third sector. I want that third sector to grow. I agree with all your critiques [crosstalk 00:38:36].
Russ Roberts: Civil society, I would call it.
Glen Weyl: Yeah, yeah. Yeah. Exactly.
Russ Roberts: So, at one extreme, we have the American system, which is--it has evolved a little bit, but it's a lot like it was in 1789. Obviously, there's more inclusiveness. There are different rules about how we pick Senators, exactly. There's been some change. But it's fundamentally: We vote for people who have power over us. And, the only way we restrain them is through the ballot box every few years.
We have this radical impeachment thing occasionally--never actually been used in an effective way except as a threat. And, it may deter some misbehavior. But, it's a very blunt strategy. Explain what a Taiwanese set of institutions or what an American set of institutions would look like that would be different than what we have now.
Glen Weyl: Let me describe a very radical future and then we can come back to what they're actually doing in Taiwan. But, I think you can imagine a world in which most of what is going on is emergent civil society institutions funded based on public goods principles. We can come back to what those are exactly, but something about sort of the general value that's being thrown off by the organization rather than what it can extract through the price mechanism.
And, those institutions, sort of in a bottom-up, democratically-assembling way, getting funding: that funding sort of decaying over time in some kind of way that's a little bit like an election, but not an abrupt--like, every four years, you have a re-election--but rather, that over time, that sort of capital that it has gets drawn down into it: unless it sort of renews its mandate, it's going to die off. And, that being how both most of what we consider private enterprise, and most of what we consider the public sector are governed. And, so, there being very limited role for some, like, fixed-geographic, fixed-polity nation state, but also very limited role for private profit--simply profit-seeking enterprises. People seeking things like profit instead do it through these charitable or public-goods-type institutions.
Russ Roberts: So, what might that mean in the real world for Google--to pick a nonemployer of yours? I mean, I don't want to put you on the spot in talking about Microsoft, obviously.
Glen Weyl: Yeah. So there's a short-term question and there's a long-term question.
So, in the long-term, what you could imagine is that there wouldn't be a Google. Like, none of these things would exist. Actually, what instead would happen is that when Sergey or Larry [Sergey Brin, Larry Page, Google founders--Econlib Ed.] had this idea for a search engine, and they knew that advertising would make it awful--which they wrote in their original paper. I don't know if you've ever seen the original Google paper, but it says, 'It's very important this never be become based on an advertising model. Because given the power of AI [artificial intelligence] and whatever, it would totally distort the environment.' And, suppose that they had that vision, instead of going and seeking venture capital, instead, they would go on sort of like a crowdfunding site.
Now, crowdfunding sites, the problem is, it's hard to raise much money there because of the free-rider problem. Right? But, imagine that there was a big pool that--
Russ Roberts: But, they do pretty well.
Glen Weyl: They do okay. They do okay. They do--
Russ Roberts: But, maybe not enough to fund Google. Okay, go ahead.
Glen Weyl: Yeah. Yeah. I mean, it's not enough to--I think there's very few really-scaled enterprises that have been funded through that mechanism. But, I agree, it's an exciting dimension. Like, I like those things. Yeah. But, imagine there was a big pool of public matching funds that were raised through taxes or philanthropy or whatever that was there to match some of those contributions.
And, so, we got this private, decentralized signal to direct these public funds to build an enterprise like this. And, it could either be initially at that funding stage, or it could be that there was even a venture capitalist who gave them funding. And, then, afterwards, there was some, you know, crowd-funding voting thing, if people really liked the service. You know. So, it could either be private profit anticipation of sort of a prize through this, you know, quasi-civil mechanism, or it could be initial funding directly through a civil mechanism. Both seem useful to me.
And, that's the way that they were funded. And so, then, rather than putting advertising whenever, they did everything to try to maximize their likely reward in this type of a mechanism that combines elements of, you know, democracy and markets.
Russ Roberts: So, you know, that's fascinating. I mean, the part that I've thought about--maybe I don't think about it correctly, and you can tell me--but, when a firm doesn't make it in the current world--and I'm going to pick an example that is close to my heart, which is Evernote. I really like Evernote. I understand there's alternatives, but I really like Evernote. And, I'm invested quite a bit in it. And, it might not make it. I assume at some point, it won't. I assume at some point, like most firms, it'll be beaten by some competitor. And, then, the question will be, 'Well, what about me?' And, the answer in the world usually is, 'Well, too bad.' The--whatever product you bought, the private--there's usually a market for the parts among aficionados and so on.
But, we could imagine that these firms would turn over their technology to a public--not a public trust; that's not quite the right word--but to a Wikipedia-like collection of people who care about it, either because they use it, or they just think it's beautiful, or they think it's interesting and fun. And, it wouldn't improve in the way, at the rate it has, perhaps, in the past. There might not be any more innovation in it. It would just be, quote, "static, stuck." Which is great. I love the way it is now. I don't need any more innovation.
You know, people say, 'It doesn't have a very good calendar.' I don't care. Or a to-do list. I don't care. It's not what I use it for. And, I'd like it to persist and be available.
But, all of these products that have created this network of users, as long as they're communicating with each other, obviously, something like Twitter, couldn't they be sustained in some sort of public/private/whatever-you-want-to-call it, democratically governed product that would--the way a nonprofit would run it. Right? That seems to be a possibility.
Glen Weyl: I'm all in favor of that. And, I'm even more excited about it for the ones that are successful. So, I don't think that that needs--
Russ Roberts: Yeah, it makes that case.
Glen Weyl: needs to kill innovation. I actually think that can be the source of innovation.
So, I think that--there's something called exit community, this notion that we should aim for a world where, as you grow, as you scale, you actually become the collective property of those who you govern effectively--
Russ Roberts: Who you serve. You mean, who use the product--
Glen Weyl: Yeah, exactly. Exactly. And, serve and govern. I mean, that's the thing: There's always that two-way relationship. And, you know, with anything that we have reason to value, we also have reason, in some sense, to fear, because we have to fear the loss of what it is that we value. You know?
Russ Roberts: Yeah. But, I'm thinking about--I think about someone like Mark Zuckerberg, who I don't know anything about personally and I'm just using as an example. If you built a product that is phenomenally successful as that--meaning, it gives an immense amount of joy to many, many people--and you don't like what it's become, you realize--and there was this interesting set of ads that Facebook ran about a year ago. I think they ran during the Super Bowl last year about how, 'Yeah, Facebook's gonna be what it used to be,' or whatever--kind of a nostalgic thing.
But, suppose you're the founder of a firm like that--you don't like what it's become. Kind of hard to get off that treadmill, because you've sold stock to all these people who have priced into that stock, this future flow of ad revenue of manipulating people through what they see in their feed. And, I have mixed feelings about that whole phenomenon. But, let's say you don't like it. And, if you don't like it, this equity market kind of says, 'Too bad. This is the way it's going to be.'
Glen Weyl: Right. You, in the process of creating something--you know, like, to me, the entrepreneur--what an entrepreneur does is sees a new community that didn't exist that could exist. In that sense, I don't think a social entrepreneur is so different from a market entrepreneur--
Russ Roberts: Yeah--
Glen Weyl: And, if you believe in that community--if you believe that's what you were put on this earth to create--then you should want to empower that community, and not some random other set of people who happen to give capital to the project, right?
And, that option isn't available, really, today. It isn't available to sustain yourself and to empower a new community to come into existence.
Russ Roberts: But, what's stopping it? Right? Why couldn't we have the world that you're talking about? And, some people I think would choose that world, just like people choose to make open source software and give it away--that aren't trying to make a profit, either because they can afford to do that, or they just prefer it aesthetically, emotionally, philosophically, theologically, spiritually. We can do that now. Struggling to compete against the world that you're trying to change. So, what's stopping it?
Glen Weyl: Because of the--we don't have the capital to support investment at scale in that type of thing.
I think, like, relative to the capital that is invested, the return--actually, I think that's the best way to think about it. You think about the, like, funds that actually go into open source software or the funds that go into a lot of things like mutual aid--these civil sector quasi-charitable organizations, or I don't know if you know of Hilary Cottam's work, the incredible work she's done on sort of horizontal, social support.
The return on--the ROI [Return on Investment]--on those projects is just, like, infinitely higher, than is the ROI on private sector software development or the ROI on government, you know, administrative social support. And, so, I think--
Russ Roberts: ROI meaning Return on Investment.
Glen Weyl: Return on investment. Yeah, exactly.
So, we should want--when ROI is high somewhere, we should want to invest more in that area. And, we should want to find ways to empower that type of activity more fully.
Russ Roberts: So, your argument is, is that: The free-rider problem is stopping that to some extent. And, the way we can overcome that is by pledging tax revenue and other funds like that, other taxation of various kinds, and use the signals of the bottom-up crowd-source story to funnel the funds from the coercive state so that they're not just spent on, say, special interests. That [?], right?
Glen Weyl: Yeah. I don't necessarily like the term 'coercive state,' because--
Russ Roberts: No, I knew you wouldn't. I thought politics[?]. I thought--
Glen Weyl: I think the--yeah, there's a lot of things to be sanctioned[?]. Yeah.
Russ Roberts: But, you do go to jail if you don't pay your taxes.
Russ Roberts: So, there's a coercive part to it.
Glen Weyl: Yeah. And, you do go to jail if you, you know, walk onto someone else's property and set up a house there, or something like that.
So, like, I think most coercion out there is done through the state by private entities, not by the state on private entities. So, I have a different perspective on it than you might.
Russ Roberts: Okay. But, how do we get there from here? And, by the way, as we're talking, I actually think I wrote a paper in 1987 in the JPE [Journal of Political Economy] on this. So, I'm going to dig that up and send it to you, Glen. But, you'll be able to understand it because I won't.
Russ Roberts: It was too long ago for me.
But, the idea was that we could radically change. And, I think this is kind of what you're saying. The charitable deduction is an attempt--it is a very tiny step in this direction. Because, the charitable deduction says, 'If you're willing to give some of your money to a cause, the state will match it in some dimension in the form of a reduced tax burden.' I think people fundamentally misunderstand tax deductibility. I think they think that you make money, giving your money away. You don't. You just don't lose as much.
And, the idea would be--perhaps--that the welfare state could be funded by a radically larger--instead of your marginal tax rate, it could be just a very, very enormously, many-many-fold tax subsidy to things that people give money to, not just related to your own.
Glen Weyl: Yeah. So, that's exactly right.
Russ Roberts: So, that's kind of what you're saying, right?
Glen Weyl: That's exactly right. And, we need to structure it in such a way that really accounts for the free-rider effect.
So, what is the free-rider effect? If you're a small part of a large good, the free-rider effect is the strongest. Right? If you're, like, giving most of the money--in fact, turns out that there's zero, essentially zero, almost zero free-rider effect for the single person who values the thing most. The free rider effect is all for the smaller players, basically. I mean, that's an overstatement.
Russ Roberts: Well, it's Saudi Arabia and OPEC [Organization of the Petroleum Exporting Countries]. And, OPEC, right?
Russ Roberts: Everybody's free riding; but that's Saudi Arabia because they have a huge stake in it.
Russ Roberts: And--go ahead.
Glen Weyl: Yeah. So, I was just going to say, so the right way to structure this is you want to give the biggest matches to the smallest players.
In fact, one way to think about it is the Kantian, kind of Categorical Imperative: You want everyone to act as if they are sort of controlling everyone's behavior. It's something like that, right?
So, if you're one-1000th of the total thing, you should get 1,000 for one match.
If you're 50%, you should only get a two-for-one match. Right?
And, it turns out, you can write that thing down as a differential equation, and you can just solve for what the solution it is. And, it takes a particular form, which is this quadratic thing, which is being used to fund lots of open source software right now on the Ethereum blockchain and in other environments. And, I think it's a very exciting direction. I think there's a lot of limitations to it. And, I can tell you about some of the limitations. But, I think it's a very exciting direction to think about.
Russ Roberts: But, this idea of funneling public funds based on private votes of money, is not really your most radical idea, I'm thinking, because that's really working within the current institutional framework. Now, maybe the current institutional framework couldn't innovate that kind of change. But, I think, don't you want to do something a little more radical to the whole governance thing?
Glen Weyl: Well, I think that the best innovations are ones which can be applied at a very small scale within the existing framework so they can grow and blossom, but that when they do grow and fully blossom, they transform the whole environment. And, I think that that is what this is a great example of.
So, yes, it can be done in small steps and local ways, it's being done that way to show its merit. But, if it really shows its merit, eventually, I think basically, the whole concept of the state and the whole concept of capitalism both sort of melt away. And, all of the institutions we have supplying things to us come through this kind of participatory democratic mechanism--at many different scales, in many different ways, in this very flexible form.
Now, there's even more radical things that I could show you as well, Russ, but I think that they would go--they would all have this property that they can be built from within the existing system in bottom-up ways, because I think trying to impose something like that top-down is the way to create a lot of chaos and destruction and danger.
I think only when we can show that it can build within the existing system it is likely to be beneficial and gain legitimacy.
Russ Roberts: I want to go back to the word 'democratic' in that last lovely peroration you just gave. I think that's the right word. Usually, we think of democracy as votes. One person, one vote. You're suggesting--I mean, implicitly, I think you're answering my objection I raised earlier, which is the skin in the game. You're saying, 'It's not so much democracy as the voting, electoral process that we associate with democracy: It's the participatory part and the spreading out of power over the organization or the product.' Is that a fair point?
Glen Weyl: Yeah, I mean, I used to believe that democracy represented a particular mechanism. That's how I used to think about it, as well.
But, I've changed my perspective on that. To me, democracy has come to mean much more as a set of values. And, those values include things like symmetry of power. If entity A has power over entity B, entity B should have equal and opposite power of entity A. You could also call that self-government, or something like that.
Another fundamental value to me in democracy is what I would call epistemic comparative advantage--the notion that, rather than saying, like, 'This person is smarter than that person,' instead, we look for the differences in the way that different people think. And, look at them as complementary to each other, and look for gains from trade from exchanging perspectives. So, those are some of the values that I see as essential to democracy.
And, we don't really have a set of institutions that instantiate those values well. We have a set of institutions that are somewhat motivated by those concepts, and maybe at the time that they were created did a reasonable job of trying to achieve some of them, but don't really adapt to the ways in which society has changed very completely.
Russ Roberts: So, let me give you an example from, again, from the tech world. I'm going to try to invoke kind of arguments that my libertarian friends would make in response to your viewpoint. I'm not totally convinced by those maybe the way I might have been in the past. So, I want to try to evoke those; I want to let you respond to it. Now, you could argue: Well, you know, I don't have to use Google.
First of all, they don't charge me anything directly. So, when I do use them, it's really kind of a great deal for me, as you were suggesting earlier. I've got all this fabulous search capability; I get all these--I could even argue I get all these great ads that come to me. Where's the power there? What power do they have over me? They don't take my money. They give me a bunch of stuff without charging me for it. I really like it.
Where's this asymmetry of power you're talking about? And, if you think there is one, say, the classical liberal viewpoint--I'm going to call it the Schumpeterian viewpoint that I associate with Book Two of Capitalism, Socialism, and Democracy--then, there's going to be overturn. It's going to change. You know, 'Google has got the upper hand now, but those large profits they're making through advertising, those are a prize for someone else to do it better. They'll do that better by luring me away. I don't have to be afraid of them. Everything is fine.' What's wrong with that argument?
Glen Weyl: So, I think the first thing to recognize is that the option of exit is not, in most circumstances, a sufficient way to allow people to deal with the problem of power. So, I mean, the United States has no prohibitions on leaving the United States. In fact, you can even take most of your property with you when you leave the United States. We have very few export capital controls, right? And, yet--
Russ Roberts: Yeah, I'm doing great here in Israel, so far.
Glen Weyl: Yeah. So--but, I think most people wouldn't say that that's sufficient to give them a capacity to control their lives, to self-govern, because there's a lot of things that tie them to the United States that they can't effectively take with them--
Russ Roberts: Well said--
Glen Weyl: whether it be their communities, whether it be knowledge and value that they have within particular industry structure. There's many connections that make them rooted, placed somewhere. And, just going off to a Seastead or whatever, it's not really a terribly appealing option for them.
And--Seastead, by the way, for those who don't know about it--it's like an imaginary notion of building some land in the middle of the ocean or something like this.
Russ Roberts: We have an episode on it with Patri Friedman. It's in the archives. We'll link to it.
Glen Weyl: And, so, if something ties someone to an environment, they need to have voice in that environment in order to exercise their, you know, right to self-government.
Now, of course, things will and can and should evolve over time to undermine the power of existing institutions. But, in my mind, that's all the more reason why we should focus on accelerating forces that allow that to happen. So, yes, it's true, those institutions will be disrupted, the new things will come along. And it should.
But, what determines the rate of economic growth is the rate of technological progress. What determines the rate of industrial progress is the rate at which we're able to bring new institutions into existence to respond to new social realities. And, if anything is slowing those down, it should be a primary goal, I think, of social policy to counteract that and to, instead, accelerate that process of change and diversification and so forth.
Russ Roberts: So, that's a really interesting set of ideas. I think most people when they hear these ideas, go, 'Well, that's just goofy and weird.' And, you know, 'Ehhhh.' But, I don't like to think that. There's a lot of things I like about what you said. What are you doing to get there from here? Or, another question might be: What's the steps, the smaller steps locally that might help?
I mean, Paul Romer has been on the program talking about charter cities. Paul has tried very, very hard to get one city to just take a chance, try it. And, that if it really is a good idea, you might need more than one. But, the idea is you don't need to have every city in the world become a charter city, to innovate in the way that he described in that episode, if you want listeners [?].
But, my point is that, a lot of times people hear radical ideas that are new, and they go, 'Ehhh, those are radical. I don't want to deal with them. I don't want to think about it. Too scary.' And, maybe they're right. Conservativism, I think has something wise to say about radical change.
But, you are suggesting I think, from what you said before, that it would be nice to let some things change. Instead of letting a thousand flowers bloom, maybe 20. Is Taiwan that example that will be transformative, that people will go, like, 'Oh, I want some of that'? How do you think about this process of institutional innovation that you're making a case for?
Glen Weyl: Probably my favorite book on this topic is Hannah Arendt's On Revolution. And, what Arendt argues in that book is that there's two types of, quote, "revolution." There's the American and the French model. The American model builds legitimacy first and authority second. And, the French builds authority first and legitimacy second.
What does that mean? So, legitimacy comes from people having lived in a particular way and knowing that it's a better way to live. And, just wanting that way of life, not to be sort of crowded out or overrun by other things.
And, authority first means you don't like the way things are, so you blow things up, and then hope something better comes in its place.
And, I'm a big believer in that distinction. And, I'm a big believer that we need legitimacy first and authority second.
And, so, my view is that we need to create conditions that allow for widespread experiments within the existing system in cooperation with existing institutions to build a future like this.
And, that's the reason why I work at a big corporation like Microsoft, and work with governments. I think that those possibilities come not from just chaos or destroying things, but rather from, sort of, checking and balancing existing centers of power to open space for new things to emerge.
And, checking and balancing isn't the only thing we need to do to get there. You need to actually foster that entrepreneurship. You need artists to help people imagine different futures. You need to tell stories about the best examples out there.
And, one of those best examples is Taiwan. And, I do strongly encourage everyone to look into what the Digital Minister there, Audrey Tang, has done. She refers to herself as a conservative anarchist. And, much of what I'm talking about sort of comes from the inspiration that she's offered. She says she works with the government, not for the government, to bring forward these types of civil collaborations to serve what, you know, people in the government are trying to do better, and eventually offer these civil alternatives to the administrative state.
Russ Roberts: My guest today has been Glen Weyl. Glen, thanks for being part of EconTalk.
Glen Weyl: Thanks, Russ. My pleasure.