Capitalisn’t: Universal basic income—why and how?

Aug 02, 2019

Anyone who’s followed Andrew Yang’s Democratic presidential campaign is probably familiar with the concept of universal basic income. On this episode of the Capitalisn’t podcast, hosts Kate Waldock and Luigi Zingales give the economic outlook on how a UBI might work, or not work, and investigate how automation and techno-anxiety are driving the conversation.

Kate: OK. Luigi, I have a question for you. What do Mark Zuckerberg, Elon Musk, Gary Johnson, and Milton Friedman have in common? 

Luigi: I think the first three have in common universal basic income, and— 

Kate: You’re not allowed to say that. 

Luigi: Why? 

Kate: Because that’s the answer. 

Luigi: I don’t think that actually Milton Friedman was in favor of universal basic income, more of an earned-income tax credit, but anyway.

Kate: All right. What if I added Andrew Yang to that list? 

Luigi: I will stick with my answer. 

Andrew Yang: If you’ve heard anything about me, you’ve heard that there was an Asian man running for president who wants to give everyone $1,000 a month. 

Speaker 4: Universal basic income, $1,000 a month for everyone. 

Andrew Yang: Yes, that’s exactly right, and we need to make this move because we’re in the midst of the greatest economic and technological transformation in the history of our country. 

Luigi: You probably have heard a lot of debate around this concept of universal basic income recently. Just a crash course for the few of you who may have missed it: The concept is that the government would periodically give cash payments to citizens without any means tests and without any work requirements, helicopter cash coming to you without any strings attached. 

Kate: As you can imagine, this concept is highly contentious. There’s been a lot of talk around Alaska’s UBI, the Permanent Fund, randomized experiments in Africa, and Finland. I think YCombinator is running a UBI test right now. But there are still a lot of important questions about what the effects of the UBI would be. 

Luigi: From the University of Chicago, I’m Luigi Zingales

Kate: And I’m Kate Waldock from Georgetown University. You’re listening to Capitalisn’t, a podcast about what’s working in capitalism today.

Luigi: And, most importantly, what isn’t.

Kate: On today’s episode, we’re going to get into the economic outlook on how UBI might work or not work. But before we do that, we’re going to explore what conditions would necessitate a UBI. And do those conditions actually exist today? Why is this conversation happening now? What is it in response to, and why are people so excited about it?

So, there are questions of, how do we fund a UBI? Will it lead to disincentives to work? Does it actually fix the problem of poverty? But before we get into that, I want to pose a slightly different question, which is, why now? And I think the answer really stems from techno-anxiety, from this idea that robots, AI, machine learning, it’s all becoming more advanced, and it’s eventually going to displace jobs, to the point where there just won’t be enough jobs for everyone in the economy. And in order to make sure that there’s sufficient aggregate demand for goods and services, we need to pay people the UBI so that they can meet their basic needs. 

Speaker 5: There will be fewer and fewer jobs that a robot cannot do better. What to do about mass unemployment? I think, ultimately, we will have to have some kind of universal basic income. 

Andrew Yang: The reason why Donald Trump is our president today is that we automated away 4 million manufacturing jobs in Michigan, Ohio, Pennsylvania, Wisconsin. And my friends in Silicon Valley know full well that we’re about to do the same thing to millions of retail workers, call center workers, fast food workers, truck drivers, and on and on through the economy. 

Kate: There has been massive job displacement as a result of automation that’s been going on for centuries. Between 1900 and 2000, the fraction of US workers employed in agriculture fell from 41 percent to 2 percent. So, today everyone’s talking about how automation is going to reduce the need for truck drivers and people who were involved in transportation. But imagine living in 1899, when 40 percent of jobs are going to be destroyed by these technologies that we’re inventing, and yet unemployment now today is at historic lows. What we’ve seen over and over again is that there is new technology that’s created that creates new jobs.

Luigi: I think you’re absolutely right. It’s important to distinguish between the technological obsolescence of the existing workforce and the employability of future generations. The Luddites, when they complained against the introduction of textile machines, were right in the sense that those machines were displacing them, making them less valuable, they were destroying that human capital. There was increasing productivity for the economy. This was the beginning of the Industrial Revolution as we know it, but it was the end of most craftsmen as we used to know them. I think that those people were heavily penalized by the introduction of those machines, and they fought very aggressively against that introduction. 

But it is not that that introduction destroyed the number of jobs overall in the economy. It actually created more jobs and led to an expansion of the economy. The question is, how easy is it to absorb those people that get displaced, and also at what wages can they be absorbed?

Kate: I think from at least an empirical perspective, the interesting question is, why has it been that other jobs have arisen at roughly the same rate, or at a rate that has been able to make current labor conditions OK in the past? And will that continue to happen in the future? Which is to say that, is there something fundamental about innovation and automation that necessarily creates jobs at the same time, so that we’re on a stable job-growth path? Or were we just lucky in the past hundred years, and now the nature of automation is such that we’re going to really start hurting?

Andrew Yang: Here in the 21st century, a lot of the economic rules we take for granted have stopped applying. You can build a very successful, profitable company that does not employ many people, and if it does employ people, you don’t have to treat them well. You can make them all independent contractors and gig workers. So, the things we take for granted have broken down, and we have to start solving this problem by evolving our entire system.

Luigi: I think it is highly possible. I think that it is not guaranteed that the future will look like the past. However, the problem is not so much whether there will be jobs in the future. The problem, in my view, is whether there will be jobs that are worth adding, jobs that give you a decent amount of income. If the price of labor goes to a very low number, many of us will hire people at very low wages to do menial tasks. When I was in the Italian army, I had to serve in the military because it was mandatory. I used to carry up bedsheets for three flights of stairs, up and down for an entire day. And why? Because the cost of the elevator was higher than the cost of my time, because they were paying me nothing.

Kate: But they had an elevator and they made you do that anyway?

Luigi: No, they did not invest in the elevator—  

Kate: Oh, I see, OK. 

Luigi: It was not worth investing in an elevator if you have free labor. What I’m saying is that if the price of labor is low enough, then you substitute. We know as economists that you substitute, so instead of adding capital, you have labor. But I can tell you, I was not very highly productive there.

Kate: Right. I think that this ties into an interesting paper by David Autor from MIT: “Why Are There Still So Many Jobs? The History and Future of Workplace Automation.” And the point that he makes in this paper is that he thinks most people are overly concerned with this idea about automation and labor substitution. But actually, there are reasons to be positive about automation. It leads to labor complementarities, what we were talking about before, which is that there can be jobs that are created as a result of automation. 

But something that has happened is that automation has been linked to polarization of employment, which is to say that over the past hundred years or so, as a result of increasingly advanced technology, we’ve seen more jobs at the high end, which is to say professionals, lawyers, doctors, things like that, and an increase in jobs at the very low end, house cleaning, hair and makeup.

But there have been a pretty large swath of jobs in the middle that have disappeared as a result of automation. And his hypothesis for this is that, all right, it’s pretty difficult to automate creativity. It’s pretty difficult to automate really advanced tasks, and so we still need people for those things, and we pay those people a lot. At the low end, let’s say when it comes to cleaning a house, you probably could create a robot, right? We have Roombas and stuff, but we could create more advanced robots that clean houses, but because it already is pretty cheap to pay people to do these things, we don’t really need to create those robots. And so those are the driving forces between this disappearance of the middle jobs, the jobs in the middle end of the income spectrum have been replaced, and so that’s led to further polarization.

Luigi: But I think that the future is going to bring more of the past, more jobs becoming obsolete and more new jobs that are created. What I think is more challenging is, to what extent are there limits on the human side? When people moved massively from the countryside to factory jobs, two things took place. Number one, there was a massive movement for universal education. In fact, the states in the United States that pushed for mandatory high schools were exactly the states that were more agricultural, that saw these needs for training their workforce. So, that allowed people to have the skills to do more-complex jobs. And, two, we were lucky that at the same time manufacturing exploded, creating jobs that people leaving agriculture could take. 

My fear is not that there will not be jobs in the future. My fear is that the jobs that will be available for people that lose their job in manufacturing or in services are either very low-end jobs that do not pay very well, or jobs that are above the skills of the people that are losing the jobs. 

So, if you lose a job as a truck driver, it is not obvious that you can become a webmaster, and it’s not obvious that even after some training you can become a webmaster. We need to think about whether technological innovation is outpacing our ability to train people or even the intrinsic ability of a large fraction of the population. 

Kate: I think I agree with you, and I’m actually even more pessimistic than you. I think that the nature of technology is changing in a way that’s not good for us. Even though, historically, automation and innovation have gone hand in hand with complementary technologies that have led to new types of jobs, I think that that’s slowing down. The types of jobs that are displaced in the future by AI and machine learning and neural networks, I don’t know that they’re necessarily going to be associated with an equivalent rise in other types of jobs. What types of innovation are taking place now are focused more in the area of pure replacement. 

For example, McDonald’s introducing new panels so that you can order on a screen, rather than having to order from a person. That’s something that just directly replaces jobs but doesn’t necessarily go hand in hand with other types of new-job creation. I think that, more and more, innovation is along those lines, and less of total, new disruption that creates a ton of jobs, which we’ve seen in the past century. 

Luigi: But, sorry, the McDonald’s innovation is no different than the telepass that allows you to go through a highway tollbooth without the person asking for money. That kind of innovation has always taken place. I think it is very positive, because it’s not a particularly exciting job to be at a tollbooth or to be at the counter of a McDonald’s. Even if they don’t directly generate complementary jobs, I think that the higher productivity and the higher wages they can afford will create other jobs in other sectors. We have many more jobs in entertainment today than we had in the past. Why? Because people spend more of their income on these other activities. If I don’t spend my income at McDonald’s, maybe I spend some of my income at the gym, and I want a trainer to help me train in the gym. So, I think that there is a natural substitution.

Kate: I think the question of natural substitution is an empirical one, right? As some jobs are replaced by job-replacing technologies, does that necessarily lead to a one-for-one increase in other forms of jobs? So, for example, the one that you gave is in entertainment. I think that that natural rate of replacement is less than one. And the reason that we haven’t seen really low unemployment in the past is just because we’ve gotten lucky with other completely different, new forms of disruptive technology. Let’s say people spend less on McDonald’s, McDonald’s becomes cheaper, it’s also quicker to go there, so they have more time and more money. Let’s say they spend that on playing games, they play phone games. That’s not necessarily going to lead to an equivalent increase in jobs.  

And I think the fact that we have seen roughly equivalent increases in the past is just a matter of luck, not an economic law. At the end of the day, I think it’s sort of a matter of faith, right? We don’t know whether the future will look like the past. We don’t know what the sources of innovation will be, which is why I simply think that I’m more pessimistic than you.

Luigi: Yes. But we can do things to be prepared. Even if we don’t know for a fact what the outcome will be, there is a clear need to better train younger generations, and in particular better train the lower end of the income distribution. In America, particularly, we’re leaving behind a significant fraction of the population in training that should be addressed, because there are a lot of job postings that don’t get filled for a lack of people with skills. So, I think there is a lot that can be done on that margin.

Kate: OK. But bringing this all back to the UBI, and why have we been talking about jobs and automation and techno-anxiety? Because, for the most part, that has been what’s been animating the debate on whether or not we should have a UBI here. I think this conversation has pointed in the direction of a more-nuanced version of that. So, I’m going to call this Marxism. Marxism is obviously a broad term, but Marx proposed this idea that robots are going to become so powerful and so advanced that they will displace jobs. 

I would like to amend that and say that there’s a strong form of that theory, which I sort of actually agree with a little bit, which is that automation will just eliminate jobs, but then there’s a weak form of that, which is that, OK, maybe automation doesn’t eliminate employment, but maybe it polarizes employment, and most of the job opportunities that are out there are at the lower end of the income distribution. They barely allow you to earn enough to survive. So, it sounds to me like you don’t agree with the strong form idea, Luigi, but that you do sort of agree with the weak-form, automation-anxiety idea.

Luigi: I think it is enough to necessitate some form of support at the lower end of the income distribution. If a job I can take with all my efforts is not a job that allows me to survive, I think that I should have some form of subsidy, and whether the UBI is the better way to do it or whether unemployment insurance is the better way to do it, this is a matter for discussion.

Kate: Yeah, I agree. I do think that this has played out historically, this polarization of employment opportunities. I don’t think that that on its own necessitates a UBI. I will agree with what you said that that necessitates welfare programs that are aimed at people at the lower end of the income distribution, but why that money should necessarily go to the people at the top of the income distribution is not clear to me. And if that’s the case, I think it’s a much stronger argument for a better social safety net, not necessarily in the form of a UBI. But I do think that there are other factors that make a UBI more compelling.

Luigi: So, what are these other factors, Kate?

Kate: Before we talk about specifics, I think it’s worth having a discussion about how we pay for all of this. Andrew Yang’s proposal for a UBI is something on the order of $1,000 per adult per month, and I think that would amount to about three-and-a-half trillion dollars, at least in gross terms. Would we be able to afford that as a country?

Luigi: I think that the idea of paying $3.6 trillion seems crazy to me, because this is almost twice what the income tax collects today. So, we would need to triple the income tax in order to pay for that. I think that this would generate a revolt. However, as many commentators have recognized, we are talking here about what is called the gross cost of UBI, because, for a large fraction of the population, people will be both receiving and paying the UBI. So, the net cost for that fraction would be zero. 

The best estimate I found is actually from a colleague of yours, Kate, Karl Widerquist, who tries to estimate the actual net cost of UBI. His idea is, if you’re making zero today, of course you’re going to receive $12,000 a year. But if you are making a bit more than that, you are going to be phasing out that $12,000 with some taxes you pay on the other side.

When you do that, what you find out is that the number of people who will receive some support will be on the order of roughly 100 million adults and 30 million kids, with a household income that is less than $55,000 a year. If you do that calculation, the net cost of UBI is roughly half a trillion a year, which still means increasing your income tax by 30 percent. But I think that is more sustainable, though it clearly is going to be a gigantic social program.

Kate: What confuses me a little bit when we start getting into net costs is that it necessarily assumes some form of payment for the UBI, right? What if we don’t increase taxes on individuals at all, but we increase taxes on corporations or, I don’t know, on fuel? I think the discussion of net costs depends heavily on how you’re financing the UBI. And I think it’s worth mentioning that Andrew Yang’s big proposal is that we introduce a VAT tax, a value-added tax, which is sort of like a sales tax, except with the sales tax that we’re used to in the United States, you’re taxed at the very end of a sequence of transactions. So, when you as a consumer buy some good, you have to pay a sales tax.

But everyone who was involved in the production of that good, maybe there were 15 steps along the way, where each person made a component and then added that, and then it was ... it all became one big good that was sold at the end of the day, a VAT tax would tax each of those producers at all the intermediate stages on whatever value they added in the production process. This isn’t something that we have in the United States today, but it is something that they have in Europe, and a lot of economists argue it’s a more efficient tax. And so, I think one of Andrew Yang’s statements is that we would raise more revenue with a VAT tax rather than our current sales-tax system.

Luigi: Yeah. But at the end of the day, we’re still paying for that amount, and this is a 10 percent value-added tax, I mean, roughly 10 percent higher prices for everything we buy. And so, we are all taxed at 10 percent on our consumption. Now, as a European, I can tell you a lot about the experience of the VAT in Europe. First of all, people have thought about the regressivity, and what they have done is they’ve introduced different tax rates depending on the goods. So, if you are buying bread, generally VAT is at a very low rate. If you buy diamonds, it’s at a very high rate. So that is a way to compensate for the regressivity. 

But it is a very efficient tax. That’s part of the reason why conservatives don’t like to introduce it in the United States, because once you start to introduce it, you’re going to go up and up. Just to give you a sense, in Germany, the VAT is 26 percent. If you start to put a 10 percent VAT on top of the sales tax of most states that are between 8 and 9 percent, you’re going up pretty close to 20 percent. Some people will consume more, some people will consume less. So, I think some people will be taxed more, some people will be taxed less, and some people will net gain, probably the people at the end of the income distribution, and some people will net lose. But whether you finance it with the income tax, with a corporate tax, with a VAT tax, there are people who are taxed, and there are people who are net beneficiaries.

Kate: Another question about the funding of the UBI is whether it would replace any other social-welfare programs, right? If we got rid of Social Security, if we got rid of Medicare and Medicaid and food stamps and everything, and we replaced it with a UBI, would that end up helping or hurting poor people? And this is an important part of the discussion, one that’s often pushed aside, especially if you’re running for president. I’m not sure that Andrew Yang has been too explicit about this, but his proposal for funding the UBI does involve some cutbacks on certain social-welfare programs. I think, in particular, there would be an option if you’re on food stamps, and if you’re receiving housing subsidies, then you would either be allowed to continue using those food stamps and receiving the housing subsidies, or receive the UBI, but not both. So, there is some sort of substitution there. Obviously, there are some progressives who think that we shouldn’t cut back on any social-safety programs at all. 

And then I think the reason that libertarians, some at least, like Gary Johnson, support the UBI, is because they think that we should just do away with all welfare programs and replace it with the UBI, because that would reduce the reach of the government. A lot of, I think, how you view UBI should depend on what’s in the background, right? How are you paying for it, and what are you cutting?

Luigi: I agree. But we have not tackled the fundamental question, which is, why is UBI better than targeted welfare programs?

Kate: I think that’s a great question. To some extent, the answer is that it might just be more efficient to implement. For example, if you think about cash assistance directly to poor people in the form of temporary assistance to needy families, or TANF, I think that back in 1996, if you thought about all people who were eligible for cash assistance, something like over two-thirds of them were actually receiving it through this program, whereas today, that number has fallen to less than a quarter. A lot of our welfare programs just aren’t very good at getting cash to people who need it. So, if we had a UBI, maybe that would just be a more efficient means of redistribution, because money wouldn’t get siphoned off and lost along the way.

Luigi: Yeah, but if you really have a UBI for everybody, then you are back to the $3.6 trillion cost, which would be very large. And imagine a program that needs to raise twice as much as the income tax raises today, it would go through an enormous amount of administrative cost. So, it seems to me that at that level it is a pretty crazy idea.

Kate: I don’t care about the administrative costs as much when it comes to chasing after people’s money, though, right? If we need to spend more in making sure that billionaires can’t dodge taxes, then fine. But if what we’re doing right now is we’re spending a lot administratively on making sure that poor people can’t get the money that they’re entitled to, then that worries me.

Luigi: No, but Kate, you cannot support a Universal Basic Income only by taxing billionaires. There are not enough of them. They need to tax you. They need to triple your income tax. 

Kate: OK. I know.

Luigi: How do you feel about tripling your income tax?

Kate: How would I feel about tripling my income tax? In percentage terms, that would be more than a hundred percent, so that wouldn’t make any sense. But yeah, I wouldn’t be a huge fan of paying a 95 percent effective tax rate, but if it were something more like 50 to 60 percent, I think I would be OK with that.

Luigi: Yeah, but that’s not enough to pay for the Universal Basic Income.

Kate: Maybe it is. It all depends on what else you’re cutting.

Luigi: I think that the idea of having better-targeted welfare programs is good. The idea of making it easier for people who need the money to get that money, I think that’s good. I think that the universality of the program, unless you have an enormous amount of natural resources like Saudi Arabia, doesn’t strike me as a good policy, and even in Saudi Arabia, I’m not so sure that this is the way to go, because what we don’t know is, what are the long-term effects of a generalized Universal Basic Income? We have a lot of experiments for a short a period of time, with relatively small amounts, because even the Alaska project is, what, $1,000 per year, roughly? It is not some life-changing amount. But if you start to have massive programs like in Saudi Arabia, where half of the population lives on dole by the government, does this really promote a society that wants to educate itself, become better and work? Probably not.

Kate: That’s a fair point, but I think it’s worth mentioning that the little empirical evidence that we do have works against your point. The main concern with the UBI is that it will discourage labor-force participation and cause inflation. Historically, empirically, that isn’t what happened. The main countries we can look to are Iran and Saudi Arabia, which have pretty sizeable UBIs. Alaska, as you mentioned, has had the Permanent Fund, which ... Was it $1,000 a year? And in none of those instances is there evidence that it led to an increase in inflation. In fact, I think inflation went down after Alaska introduced a UBI. Same in Kuwait, actually. 

And in terms of labor-force participation, the evidence is that in Alaska there was no significant effect. In Iran, there was no significant effect except for the youth population. I think young people were, to some extent, discouraged from entering the labor force for the first time after they instituted a UBI there. And I’m not sure that we know enough from Saudi Arabia about how it directly affected its labor force. So, I don’t think that there’s a smoking gun that a UBI would necessarily lead to inflation or people never working. But your point earlier was that, how much can we really infer about the external validity of these experiments, because they haven’t existed for that long? And I do agree with you, especially to that point about youth labor-force participation in Iran, there is concern that in the long run, this would lead to a decrease in people getting jobs.

Luigi: Yeah. I think that we tend to underestimate the long-term effects of welfare. There is actually a Swedish economist, Assar Lindbeck, who first pointed out the fact that when you introduce a welfare program, you are piggybacking on the preexisting social norms. And, to some extent, you have the best of both worlds. You have the benefit of the transfer without the immediate undermining of social norms. But as the welfare becomes part of the social construct, then people develop different social norms, and then this might have long-term negative consequences. So, the result in Iran, where the young people are less willing to work, I think is pretty scary in my view, because if you don’t enter the labor force early on in your life, you are unlikely to enter later in your life. This is a permanent effect that could easily last for 40 years. Before we do any massive program like this, I think we need to have better evidence, because the irreversible effect can be devastating.

Kate: But Luigi, I hate to say it, you’re sounding a little bit like you’re repeating the conservative refrain, which is that we shouldn’t have direct cash assistance to the needy because it will make them lazy.

Luigi: Look, there is a concern about that. In particular, the essential point about the Universal Basic Income, and the essential difference vis-a-vis targeted welfare programs, is that Universal Basic Income is a right to everybody, unconditional of whatever you do. And that, in a way, is a change in mentality. One thing is assistance for people who are down, who have been hit by a disease, who have been hit by a shock, that I am all in favor of as part of some social safety net. A right, regardless of what you do, I think does create an entitlement that is dangerous in the long term, because we’re not sure that we can pay for it, and we’ll create a different attitude vis-a-vis your participation in society. Not to mention, I think most people want some dignity in their job, not just the income, and so, by creating this, you are reducing the pressure to create jobs that might also give them some dignity.

Kate: I think that that dignity argument works against you, though. I think almost everyone is smart enough to know that there is dignity in work and being a constructive member of society, and that if you receive $12,000 a year, that you’re not going to be so distracted by that that you’re just going to give up on that dignity altogether. I don’t think that people at the lower end of the income distribution would really be lazy, as some people put it. I think that they would still look for work. I think that a lot of the people who can’t find work are trying to get work, and they are looking for work. It’s just that in some cases, it’s difficult to find. And I think that there is some amount to be learned from the empirical experience so far, which is that there isn’t overwhelming evidence that it discourages labor-force participation. I think it’s because people do fundamentally like working.

Luigi: I think that the at the level that is provided in Alaska, for sure, if we’re talking about $1,000, it is not going to dramatically change anybody’s incentive. When you introduce a marginal tax rate of 50 percent, like many of this Universal Basic Income de facto will introduce for people in dual-income families, I think that the incentive for one of the two not to participate is pretty strong. You receive the money unconditionally, you pay the money conditionally. So, if you work, you pay more in taxes, but you still receive the money on the $12,000. On the margin, the incentives to not work will be increased, because people say, why should I pay more? That’s a big difference with Alaska, if you pay this money out of a fund, there’s no taxation. So, there’s nothing that is discouraged. You don’t discourage people from working. 

In a system where you increase the taxes on everybody, whether this is VAT or income tax to pay for the Universal Basic Income, it’s true that most people receive money with one hand and pay with the other, but it’s also true that they receive unconditionally, while they pay conditionally. They pay conditionally on consuming or working. And so, whatever is taxed, they’re going to do less. So, either consumption or work, they’re going to do less. I think all these stories about Alaska not having any effect on labor income is only valid on the ethical consideration that people don’t change their ethics to work when they receive some money, which is completely moot when you receive $1,000, because it’s hard to imagine $1,000 would change your ethics. 

If you start to receive, with a wife and kids, $40,000 a year, for most of the United States, $40,000 a year is a pretty decent income to live on. And then the question is, are you still motivated to work, or does the ethic of working change when you receive $40,000 a year? I think that that’s a more likely scenario. I think that that’s where my concern is vis-a-vis the UBI. 

Kate: Luigi, I think that one thing I definitely agree with in what you just said is that a lot of the incentives or distortions that arise from a UBI depend on how it’s financed. In that regard, we can’t necessarily infer that much from experiments that we’ve seen in Alaska, Saudi Arabia, et cetera, because those experiments in UBI were largely funded by pools of wealth generated by natural resources in those areas, and that wouldn’t really be the case if we had a UBI in all of the United States. So, instead, we would have to finance it. Perhaps through higher taxation on your income, which might actually discourage incentives to work, or higher taxes on sales, things that are sold. The VAT tax, as Andrew Yang proposed, which might discourage people from consuming. And so, I agree with you in that regard. 

To be fair, there are other sources of financing that Yang is proposing. He’s saying, let’s remove the Social Security cap. Let’s introduce a financial transactions tax. Let’s remove favorable treatment of capital gains. He even wants to introduce a carbon tax, I’m not sure. There are other areas, as well as cutbacks on certain welfare programs that would all, in combination, pay for his proposed UBI.

Luigi: Yeah, but I still don’t understand why this, vis-a-vis not a negative income tax or a targeted welfare program. I still don’t see the benefit, given the enormous administrative costs and burden of such a proposal, I don’t understand why it’s so popular vis-a-vis some more traditional form of welfare.

Kate: I think that’s completely fair, and I agree with you on that one. At the end of the day, if what robots have done is polarize our employment but not necessarily eliminate jobs, then it seems like the objective should be helping poorer people, not necessarily giving everybody money. I think that the more compelling argument for UBI is if welfare is so administratively costly, and introduces so many potential loopholes that it’s completely ineffective. I think some, not all, but some of our welfare programs are pretty poorly administered, and when it comes to administration costs, I think the government is slightly better at getting money that it wants than dispersing money that it doesn’t want to disperse. So, to the extent that you think that that’s true, that is one argument for the UBI, but I’m not sure that completely justifies the UBI.

Luigi: You may actually argue that some of the disincentive effect to work might be useful if you want to keep up the wages of people who are working. So, part of the story is, if you are desperate, you start to compete to work for any amount of money, and that will decrease the wages of everybody at the low end of the income distribution. If you have a minimum income, then people will not be working below that minimum income, and that will force employers to offer higher wages. That is not a trivial benefit in a world in which we have this polarization. So, going back to the point that Kate was making earlier about polarization, if you want to protect the lower end of the income distribution, I think that that is a good way to do it. So, in general, I’m all for simplicity. But not if it implies raising $3.6 trillion. That, to me, does not seem to be a very simple program.