Tag Archives: y combinator

Y Combinator’s Basic Income Study, featuring Elizabeth Rhodes

The Basic Income Podcast
The Basic Income Podcast
Y Combinator's Basic Income Study, featuring Elizabeth Rhodes
Loading
/

What’s the latest with the basic income study piloted by Y Combinator Research? Jim and Owen sat down with Research Director Elizabeth Rhodes to find out. Rhodes shares insights from the initial pilot in Oakland and the much larger upcoming experiment. Rhodes details the goals and methods of these exciting, important studies.

——

Episode Transcript

Owen: Hello, and welcome to the Basic Income Podcast! I’m Owen Poindexter.

Jim: And I’m Jim Pugh. One of the biggest developments in the last few years in the basic income space was the announcement by Y Combinator Research that they were going to run a pilot on the topic of basic income, to actually better understand what it looks like for people to receive unconditional cash. This happened just about two years ago now, and it really helped to kick off the conversation around basic income in the United States.

Owen: So, today, we’re very fortunate to have the person leading that effort at Y Combinator Research. So, welcome to the podcast, Elizabeth Rhodes.

Elizabeth: Thanks. It’s great to be here.

Owen: So, why don’t you start off by just telling us a little bit about your background as a researcher and how you came to Y Combinator Research?

Elizabeth: When I first heard about the job posting I was finishing up my dissertation at the University of Michigan. My background is in social work and political science, and I have been involved in poverty research both as a social worker and in terms of policy on the political science side.

I have worked a lot with the existing social safety net, both domestically and abroad, and pretty early in my graduate school time I did an independent study with a Professor and a colleague and we were sort of thinking outside the box on poverty, anti-poverty strategies, and we read, even it was actually Milton Friedman’s first discussion of the negative income tax. And, we sort of reconsidered what that might look like now. And that was my introduction to the topic.

I had definitely not heard of the movement for universal basic income. It was more just a guaranteed minimum income in more of the negative income tax model. So, we had written a paper on that and calculated what it would cost to implement a negative income tax now at different levels and different marginal tax rates.

So, I was really interested in the topic and then when I heard, someone forwarded around the political science department in Michigan, actually Sam Altman’s announcement that YC Research was looking to do a study and I jumped at the chance. It was no one else. And people would talk about the idea but no one else was really willing to take the risk and invest in this idea that is pushing the envelope I guess in a lot of ways.

Jim: Now, Y Combinator Research and the basic income project in particular, is a pretty new thing for Y Combinator, which is a start-up accelerator. So, can you tell us a bit more? What was the motivation here? What were the intended outcomes? What were the goals of this effort?

Elizabeth: So, Sam Altman who is the President of Y Combinator had, YC was established to promote innovation and a lot of that can happen in for-profit through start-ups, but there’s a lot of questions and societal level problems that shouldn’t be addressed. Either they shouldn’t, because no one company should develop this certain particular idea or have control of it, or they’re too long, they’re not suited for profit, and so that’s what gave him the original idea for YC Research.

And, the first group within that was Open AI which is seeking to develop generalized artificial intelligence in an open source, not for profit environment. And I think he began to then think about what other types of questions like that do we want to tackle and with Open AI and with what automation might lead to got him thinking about basic income.

He also talked about the YC model in some ways as a basic income for founders at the beginning of launching their start-up. So that’s where this particular idea came up. Right now, there’s a group of independent projects, working under the banner of YC Research.

Owen: So, given this general interest in basic income, how did you go about designing the project that you’re now working on to achieve those goals?

Elizabeth: So, it’s been a difficult and a long journey in a lot of ways because there’s so many questions that we have, and there is so much interest in basic income. And so I think we started at the most general questions, these really broad, like what would universal basic income look like? How will it affect people? And then we pulled together large groups of academics and policymakers and had these day long workshops and conferences talking about “well if we’re going to operationalize this into a study, like a very concrete study where we can ask very specific questions.”

You know there were a lot of decisions that had to be made at the onset. Things like, do we saturate an entire community, do a more geographically dispersed randomized controlled trial. And we just met with a lot of different people and talked about, you know, there’s not only financial concerns with the cost of such a project, there is the ethical concerns about, you know, you’re giving people a lot of money and you’re potentially changing not only individuals’ lives but the social context within communities and there were just a lot of things. You know, how does it work with the existing safety net and existing policy?

There’s been so much to think about and so I think it was really kind of a, I think people expected us to launch really fast, but there’s just been this process of figuring out what can we do? What other questions we can answer and refining slowly into coming up with a study design that we feel like we’ve landed on.

We believe that its definitely a foundational study. It’s not gonna answer all the questions we have about basic income, but it is going to provide a foundation about how this affects individuals moving forward and we hope that there’ll be a lot of exploratory outcomes thinking about helping identify areas for future research and hope to continue pushing the research agenda forward.

Jim: So, speaking of doing upfront legwork to make sure you’re properly prepared, you’ve actually approached this in multiple phases, and you’ve already been running a pilot program in Oakland where you are giving people cash in advance of a larger experiment. So, can you tell us a bit more what that pilot program was? Who was involved? What was happening with them? And what that experience was like?

Elizabeth: Sure. So, before we even started anything we actually talked with some of the people that ran the negative income tax experiments, which is the most recent version of something similar to this. This was back in the sixties and seventies, and they were really criticized for a lot of reasons.

So, we actually talked with them and said “ok, how do we not repeat the mistakes of the past?” And one of them was to make sure to run pilots, to test these things out. There was a big push then to just get started and you’re not able to think through all of the consequences, everybody makes mistakes. So, we decided to do several phases of a pilot actually because we realized we could iterate and learn as we go.

So, I guess it was a year ago in August, we launched a small feasibility study. It ran for one year. It had only six people, which gave us a chance to work very closely, and it was just, at this point, when I started, it was just me and one colleague. So, we were kind of working on everything ourselves.

Half of them received $1500 a month for the full year, and the other half received a smaller sum, a nominal amount of $50 to thank them for participating, and it was really interesting to watch. Obviously, we randomly selected these people from Oakland. They were all lower income residents and spent a lot of time with them. We didn’t know in a lot of ways what all the potential pitfalls were and so we were able to learn, they were really great and willing to share their data, so I think we learned a couple things from that. I’ll talk a little bit more about some of their experiences, but I think we learned a couple things.

We learned that it’s possible to do this. That people were willing to provide their data and share their stories. We can only learn if people are willing to share that information and they were eager and willing. If we didn’t send out a survey, they would contact us and be like “hey can we do this?” So, they were willing to share a lot about their lives.

We also learned that even with a really small number of people, it was interesting to see what transpired over the course of the year. It’s certainly not a basic income. It’s very short-term defined cash. But just that security that it provided, did. It allowed someone who was kind of couch surfing between friends and family, working a part-time seasonal job to actually move out to Antioch, where she was able to get an apartment with a group of people and started working a full-time, like much more steady job. There was a student who was able to work fewer hours at like fewer part-time jobs and focus more on school and giving balance in their life and figuring out what they wanted to do next.

So, there’s just so many different ways that it can influence people, but I think we will just, in those six we’ve learned so much, and I’m really excited to see what happens with 3000 people. So those indivduals are actually still participating and receiving smaller amounts of money.

So, we’re launching now, this is our second phase of the pilot, where up to 100 people still in the Oakland area, they’ll be receiving smaller amounts of money. It’s less about what does a basic income do, but more about operationally, the process of distributing payments of collecting data of recruiting. So, we’re refining those. We’ve learned a lot with six and now we’re learning more. We’re working with a national survey research firm that’s helping us implement the entire study and so they’re getting involved now too.

So that’s our last pilot phase. That’ll probably run for another six months to a year. But hopefully we’re planning on later this year launching a larger 3,000-person randomized control trial across two states.

Owen: So, we want to get into that in a moment. I’m curious if there’s any pitfalls or things you learn not to do either from talking to the people who are studying this in the sixties and seventies or from your pilot studies in Oakland?

Elizabeth: From the sixties, I think one of the things that came up there was they were really looking at a macro level — they wanted to see if you give people money do they work less. That was the real concern, was the labor market response at that time.

But the study that they designed wasn’t really actually able to measure that because it could only look at a demand side response – do these people work less? But it’s not if everyone gets this and everyone has a floor, this injects money into the economy of people who are likely to spend it. Does it create more demand for jobs? Are people able to say, “well, I don’t wanna take that job because I have this floor” and wait till they find a different job and what does that do to wages. Also, it was a very short-term study, and, in some ways, it was like an opportunity for leisure as some have talked about.

So, one thing that we wanted to do first is say “ok, we wanna make sure that the questions we’re asking we can actually answer with our study.” And that’s why we really are focusing on what are the effects of this unconditional cash monthly over a three- or five-year period on individuals and their families and those in their network. We’re not going to be able to identify what happens to the entire community or what happens to prices or what happens to rent levels and things like that. But we are able to say “ok, how does this affect individuals?” “How does it affect the decisions they make? The opportunities they have? Their spending patterns? Their well-being?” All kinds of different things like that.

Jim: So, you mentioned that hopefully this year you’ll be launching this much larger experiment. Can you tell us more about that? Who is involved? What is the process going to look like? How will you actually run this?

Elizabeth: So, we have developed a proposed design that’s actually available on the website. We’ve shared it with, we’ve gotten feedback from probably hundreds of academics and policymakers. We’re still modifying as we continue to learn from the pilot, but we’ve partnered with universities. We have another group of academics, David Broockman from Stanford, Sarah Miller from the University of Michigan, and Eva Vivalt from Australian National University are working very closely with me. We’re the four PIs and we have a growing group of eight or nine senior academics who are advising and overseeing the project.

As I said, we have partnered with the Center on Poverty and Equality at Stanford. We’re partnering with Poverty Solutions at the University of Michigan. It’s not something we’re doing in isolation obviously. And we are working with a large national survey research firm that works on these kinds of experiments they’ve done, Moving to Opportunity 8 years ago and things like that. So definitely helping with the implementation.

So right now, we are actively fundraising and applying for grants and things like that and continuing with the pilot in preparation for, as I said, hopefully later this year, and we have a design that includes 3000 individuals, 1000 of whom will receive $1,000 a month. Most of them for three years and a smaller group for five years.

Owen: And can you speak to the specific data that you’ll collect?

Elizabeth: We are collecting a fairly wide-range of data, and there’s a couple different sources. One is administrative data which is data that the government collects on things like IRS earnings and that kind of reports. There’s some health data. There’s some use of existing benefits. For kids, there’s school attendance and all that kind of thing. So, with permission we’re actually, the project is overseen by the Institutional Review Board at Stanford, but with individuals’ consent we’re able to collect that data so that is sort of passively, they don’t have to stay in touch with us, we’ll be able to collect outcome long-term on some of these measures.

We also, we’re doing in-person surveys at enrollment and then midline and then end line. We’re also doing shorter web-based mobile surveys, maybe monthly, to collect data on some things that we need repeated measures or if there’s a lot of problems with recall.

And then a large group of people, probably about 200 of the recipients, will be followed much more closely for qualitative interviews. So, I’ll sit down with them open ended, really trying to understand their experiences. I’m trying actually not only like what are the effects of the cash but then why? What are the pathways? And what are the constraints people still face or how could the design be better or be more helpful.

And the outcomes themselves range, it’s a fairly holistic study. I think a lot of people as I said, in the past it’s been very much focused on “do people work or not work?” And that’s something we’re collecting data on, but it’s much more broad. I think, how do people spend their time in general? There are so many other productive uses of time just besides the paid labor or that they’re caring for a child or elderly relative, or they’re starting other businesses, or how do people spend their time?

We’re looking at financial health and measures of resilience. So not only, yes, they have a steady stream of income, but then if they face an unexpected expense would they be able to cope without filing for bankruptcy and things like that. We’re looking at lots of measures, health related measures. Not only outcomes but also service utilization and mental health outcomes which a lot of other studies, like lottery studies and the Earned Income Tax Credit, have all had very positive effects.

We’re looking at subjective well-being and measures of self-efficacy and locus of control and how does this affect well-being more broadly. Because I think those types of outcomes really have a long-term effect on all these other outcomes we’re looking at, and how does this minimal level of security, in what ways does this affect well-being and outlook?

Jim: I know that in some of the studies that have been done in developing nations around cash transfers, one of the aspects that’s been explored is context setting around giving people cash, and if it’s presented, it can be the same money and there could literally be no conditions, but if it’s described in a different way to people, and there’s been exploration as to how that description actually then has an impact in what ultimately happens to people. Is that something that will be either actively included in the experiment, or will there be some ways to do analysis around that?

Elizabeth: There will. So, in order to have, I don’t wanna go down the rabbit hole of statistical power, but in order to be able to really detect effects, we can’t have too many treatment arms. So, if we give people lots of different amounts of money or change the way it’s framed a lot of different ways, it’s a lot harder.

We want to focus on really more of what we call heterogenous affect. So, looking at how it affects people across different income levels and gender and things like that. So, we’re having a uniform across everyone framing. But, having said that and partly because we aren’t the government, we can’t, we’re a non-profit organization and researchers and there’s only so much that we can do. I mean we don’t want to deceive anyone in that way, but we want to understand how people view the money.

And, so, certainly part of the analysis is figuring out and there’s a lot of questions related to government, like really trying to dive into that because I think it does make a huge difference even in, I mean, as we talk about if this is viewed as a stigma, like just a welfare program versus this universal human right, it can have a very different effects. And so that’s something we want to explore through the analysis.

Owen: What are you personally curious about in terms of how this plays out?

Elizabeth: I think a lot of things. I think for me that level of security, having worked a lot in the existing system and knowing, like people just week to week, or even month to month, it’s a constant struggle and it’s inability to plan for the future or think about the future, because it’s just constantly “well, how am I going to pay my bills this week or this month?” And that stress just completely wears people down.

And so I think this would apply a lot across the income level, but I think specifically I’m really interested in seeing how that $1,000 a month is not, you know, a ton of money, but it does provide this level of security and how does that change peoples’ lives and their relationships? And I’m just very eager to learn, just to hear their stories and to, not just the data but just really on a very basis level, how does this affect so many different aspects of their lives?

Jim: So, you mentioned that the plan is to launch this experiment this year. I know at Universal Income Project we often have people curious, first curious about what’s going on but also curious about being involved in some way. Is there a way for members of the public to somehow be involved in this effort?

Elizabeth: Yes, so I mean one of the things, it’s taken us a while I think, the study was announced and then we kind of went underground, and it’s been a, you know, people are like “oh, it’s being secret” and I think that was never the goal. I think, we just wanted to figure out what we were doing before we start talking about it.

There are certain aspects of this. We’re gonna be extremely transparent with exactly what we’re measuring. How we’re measuring it. Every detail of the plan. You know, how we’re gonna do the analysis is all gonna be available except things like locations and protecting the privacy of participants is like our absolute first priority and so people cannot say you know “can I interview someone?” You know, protecting that privacy. But beyond that, you know, we’re gonna be very transparent.

So, in terms of following, we’re still trying to figure out framework. We have a website, whether it’s blogging or how we’re gonna be sharing that and sharing updates up on the research. Certainly, we are actively fundraising, so anyone who you know wants to contribute to the effort can do so through our website. This money is going, one to one, like 100% directly to individuals. Giving people $1,000 a month for three and five years is pretty expensive so it’s something we continue to raise money for. But we want people to be involved and share the information as much as we can.

Jim: Is there a website URL you can share for folks?

Elizabeth: Sure. It’s YCR.org/basicincome.

Owen: Thank you, Elizabeth Rhodes, for joining us on the podcast.

Elizabeth: Thanks so much for having me.

Owen: That was Elizabeth Rhodes on the basic income podcast. I’m very excited to see where this data goes. I think it becomes a lot more tangible to people, what cash transfers can do when you see things like kids stay in school longer, there’s some positive outcomes with health, and that’s a lot easier for people to grasp than they had more money, and it made their lives better.

Jim: Definitely. And I find it really encouraging to see them taking time to develop such a thoughtful approach to the experiment. They’re really thinking about what is the right way to execute this, taking the time to do the pilot in advance in order to assess logistics and really thinking through what it is they’re actually measuring here, what are the outcomes their going to be looking at.

That’s some pretty sharp contrast to the negative income tax experiments that were run in the sixties and seventies. Granted, these were amongst the first controlled experiments around policy that were done, but it was pretty muddled as far as what they actually were looking to assess through those, and I think as a result you saw people didn’t have clear takeaways. And we have our interpretations now, but it really wasn’t clear conclusions that could come out of that in a way that, to your point, could be really valuable to understanding how and if we move forward with the policy.

Owen: Yeah, and with the sixties and seventies ones there were at least some, it seems like some people had an agenda in how they wanted to interpret the findings. And it was easy enough to project an agenda.

Jim: I’m sure we’ll still have people with agendas now. Hopefully if it’s more clear from the start, it’ll be more difficult to derail that.

Owen: Right. And there’s a lot of quantitative data that they’re gonna collect which I’m excited to see. And I’m excited to see the qualitative data. I lean toward the quantitative because its immutable in some ways, but a lot of the real benefits come from reduced stress and just people knowing a little more about their future financially.

Jim: Alright. That’ll do it for this episode of the Basic Income Podcast. Thank you to our producer, Erick Davidson, and if you like what you hear, please do make sure to rate and review us on Apple Podcast or the podcast service of your choice and let your friends know about this. We’re always looking for new listeners. We’ll talk to you next time.