In StreetSmart Episode7, Streetsblog California editor Damien Newton and Sam Speroni from UCLA discussed the unequal distribution of car ownership costs in Los Angeles, using data from 2021 to 2023.
Speroni and his fellow researchers found that while wealthy neighborhoods had high debt per borrower, lower-income areas like South LA had higher debt burdens. Black and Latino neighborhoods faced higher debt per borrower and debt burdens, and higher delinquency rates. The study highlighted the impact of the COVID-19 pandemic on automobile borrowing and the need for better data collection to address predatory lending.
UCLA has published two articles on the research. The first covers who has debt and the second where debt is. The policy briefs that the podcast and articles are based on can be found here.
A full transcript can be found below.:
Transcript:
Damien Newton
I'm here with Sam Speroni, with UCLA, and we are discussing the costs of car ownership and how they are distributed, in some cases unequally, across Los Angeles.
Welcome! You and your team studied this issue from 2021 through 2023. Why don't you start out with just a brief explanation of what your research was and why it was important?
Sam Speroni
Sure. Thanks Damien for having me. Great to be here.
We were interested in changes in automobile debt as part of both the overall profile of debt for people, but also in geographies, both in Southern California and across California as a whole. This was enabled through access to a unique data set called the University of California Consumer Credit Panel. That panel tracks credit lines for people living in California, and we can see how that changes over time. It also gives us an approximate location of where those borrowers live.
So we're able to look at different neighborhood aspects of the borrowers, and see where we may see some spatial patterns in this borrowing. And so ultimately, we were interested in two research questions. One was, "How the landscape of automobile borrowing, both loans and leases, changed from before to during the COVID 19 pandemic." And then secondly, we were interested in how the landscape of automobile borrowing might differ across different types of neighborhoods in California.
Damien
This entire study happened during different phases of the pandemic. How much of that do you think may have impacted the final results, which we'll get into in just a just a second?
Sam
It's a really fair question. I certainly think yes. Some of that we see in our results. For example, we look at quarterly borrowing patterns in one of our papers, and there's an enormous drop in quarter two of 2020 for obvious reasons, and then an enormous spike in automobile purchasing in quarter three. There's all these people that would have bought a car in quarter two and/or needed a car as a result of the pandemic, who might have otherwise relied on other means of transportation. But one of the things that we want to highlight is that generally, not quite to the same degree as a house, but a car is a long term purchasing decision, particularly if you're financing the car.
Typically you're committing to several years of payments when you finance a car. And so, on the one hand, there's certainly some changes in the pandemic, and we can delve into what we found there, on the other hand, the people who bought their cars in 2021 are largely still paying them off.
Damien
One of the things that stuck out to me, is the loan size was also increasing. Not just the number of loans in an area, but the size of the loan was increasing. Most car loans are in the five to 10 year range...I don't think there's very many people who take a 15 year car loan. Most don't last 15 years, but that five to 10 year range...people that bought a car in 2020 are definitely still paying it off.
You also noticed that there was at least some correlation in the racial makeup of a neighborhood and the car loans that the population was getting. Can we go a little bit into that?
Sam
So the premise that I think is worth mentioning is that while there are lots of different means of transportation, and many of them are more environmentally sustainable than automobiles, particularly for most of California, the automobile dramatically increases access to opportunities for people of all races, ethnicity, and income brackets. And the reality is that it takes an enormous upfront cost to acquire a car, even an older, somewhat broken down, unreliable car is still an enormous expense, particularly for low income families.
And so cars require this large upfront cost.
And while public transit can step in and provide mobility for those without automobiles at a per ride unit cost. It comes nowhere near providing the same access to opportunities. So we're interested in essentially this group of people that are likely to need a car to increase their access to opportunities, but live in a place where it's basically required to have a car.They may not be facing the same additional obstacles compared to others.
Damien
I just wanted to interrupt for a second, because I think a lot of people, when they see a podcast like this on Streetsblog, assume that we're going to be going, "Oh, and the cars are bad, and everyone's bad for having them." And that's very clearly not what we're talking about. And I should have said that up front, and I've actually made a note to say it in the intro, but I wanted to say it now too. We're talking about the world that we live in, not the world we want to live in.
I would prefer to live in a world that wasn't so car dependent, but in a world that is car dependent. People need to get to work to have access to opportunities. And so we're talking here about the barrier of cost and what that means for people's access to opportunities. Yeah, 100% the long term or midterm, depending how you want to define debt, means to the people that are having to put out money to have access to the car and have access to the educational, cultural, right, work opportunities.
We're talking about the world we live in, not the world that we want to live in. I think we're talking here about, how can we make the world we live in a little bit better for people who are getting left behind in the near term?
Sam
So okay... you asked, "Where are we seeing these patterns?" We looked at three measures for debt outcomes related to automobile debt in California. The first is the amount of debt per borrower in a neighborhood. And so that's essentially we add up all of the car loans outstanding balances, and count up the number of people who could borrow, not just who do, but anyone with a credit line in that neighborhood, and try to get a sense of how much is this neighborhood spending on cars relative to the number of people that could buy them. The second measure we look at, we call the automobile debt burden.
What we do is take that same measure of all the car loans and leases in the neighborhood, but instead of dividing it by the person, we divide it by the amount of annual income in that neighborhood. The American Community Survey from the US Census Bureau has a measure of aggregate income in a spatial unit.
We created a measure that looks at how much money people are spending on cars relative to the amount of money they make as a neighborhood. The third is more straightforward, and that's the rate of delinquencies, the rate of loans that are at least 60 days behind schedule in a given neighborhood based on the number of loans that are there. What we find is that there's higher debt per borrower in a variety of different places. We see that in wealthy neighborhoods as well as lower income neighborhoods.
And part of that is that wealthy people buy expensive cars. I don't want to focus on that measure, because what it reveals is just how much money do people spend on cars, and did they finance them? Where I want to focus is on the higher debt burden. And so remember, that's debt for cars related to overall income in the neighborhood.
There's places where there's high debt per borrower that typically signals wealthy neighborhoods that have really low debt burden. Some examples of that are Silicon Valley or Palos Verdes or the Hollywood Hills. Really high debt for cars, but also a lot of money to pay for them. Conversely, there are places with high debt per borrower and high debt burdens. A lot of those are the rural areas of southern and central California.
In Imperial County, the agricultural areas of the Central Valley: those places tend to both spend a lot of money on cars because they are relatively rural and cars are an absolute necessity, but people also have less money to pay for them. Then there are places that tend to have higher debt burden, but maybe not high debt per borrower, and those are typically the lower income areas of Los Angeles, like South LA. We find on the flip side that highly urbanized areas control for all of these other things, like race and income and credit score. Highly urbanized areas tend to have less automobile debt. And so you could think there about a family that maybe has two parents, two children and one car, because maybe one parent is walking distance to work or bikes to work. And we find that in highly urbanized areas, and especially in urban areas around rail stations.
Damien
I was actually late for this podcast recording because I had to pick my kid up at school. But until this year, we were a family with one car because we worked from home and had kids that could walk to school. Now our kids go to different schools that aren't walkable, and it changed. I mean, I live in West LA, and I would say an upper middle class neighborhood for Los Angeles. No people have all their money in their houses, so, like, looks good on paper, but not a lot of disposable income to disposable income.
It's definitely changed our lifestyle, having a second car and the cost of it. We're not even we're not talking about the cost of gasoline, oil changes, insurance, etc. We're only here talking about the loan which is significant.
Sam
And so when you think about that burden and how much of the percent of money that people are spending on their car, just to finance the car. You add on all of those expenditures on top of it that aren't part of what we're looking at, but are a little more uniform in their nature than maybe the price and financing cost of the car (to get the true cost).
We found in all of those metrics that I mentioned, we found that the outcomes are more adverse for black neighborhoods. So black neighborhoods compared with white and Asian neighborhoods tend to have higher debt per borrower, tend to have higher debt burdens, and, importantly, have higher delinquency rates.
I'll return to that in a second. We also found for debt per borrower and debt burden, in particular, debt burden much higher rates of debt burden in Latino neighborhoods. And so when I refer to black neighborhoods or Latino neighborhoods here, I mean neighborhoods with at least 50% of its population is of that race or race or ethnicity group.
Damien
I was going to ask about that, because one of the findings that I remember is that the increase in the size of the loans was the largest in Latino neighborhoods. But then you were talking about the debt burden being highest in African American and black neighborhoods. And I was wondering about predatory lending and how much of that is happening.Why don't you briefly explain what predatory lending is, rather than me do it, and what impact that's had.
Sam
There's two issues that we're concerned with, sort of indirectly, here. One is predatory lending. That's when lenders target specific demographics in providing loans, and may do so in a way that is into the next part: discriminatory practices.
Predatory would be a situation where, for example, an auto dealer would offer a higher interest rate to a woman than a man or to a black person than to a white person. Discriminatory practices may, may exclude those people overall. You may see situations where, where somebody who is a white man, might have an option of six different lenders, and they can pick the best of those options.Somebody on the on the the wrong end of discriminatory practices, which typically is black, Latino or women in automobile lending, somebody may only have two choices in their choice set for loans, and they may be the two worst interest rates. In our study, when we look particularly at debt per borrower, which counts the interest rate and the debt burden, we're controlling for income in our statistical models, and we're controlling for credit score. If we were just thinking that this is a function of the ability to pay or a function of a recorded history of credit, then we would expect race and income, race and ethnicity, to not have any effect on what we're seeing in our metrics.
Almost the opposite of what we see. We see strong correlations, particularly with Latino neighborhoods, but also with black neighborhoods for debt per borrower and debt burden. And we see strong correlations with black neighborhoods and delinquency rates. And so that getting to your predatory practice: if someone who isn't being discriminated against has all of these choices, and they pick the best loan for them, then they're less likely to default on that. They may also be in a position where they've been given a bad loan that maybe the bank wanted them to default on.
Damien
What is there to do about this? New laws to try and address this? I mean, there are already laws that say you can't discriminate, so what do we do? The Streetsblogger in me is hoping that we'll go back to those bike lanes, and talk about making communities that are more walkable and bikeable, and communities that have workplaces near job centers and all that sort of stuff. But I'm going to guess that you've got some other things to say.
Sam
It's an interesting time to talk about data and research in our current national environment, but one of the things that became clear to us is the advantages that we had in this project is that we had access to a really comprehensive administrative data set. We were able to see different financial patterns for a variety of different people, but we knew very little about the people themselves. We knew stuff about where they live. And so we can draw some conclusions about where where these loans are, but it was much more difficult to understand who the borrowers are themselves, outside of their neighborhood.
One of the things that we found as a gap that could be addressed...that could help address those laws that do already exist... is a standardized data collection about borrowers for automobile loans. This exists in some capacity for home mortgages. And there's, there's some work from the UCLA Lewis Center and one of our colleagues here, Jose Loya about what, what we know about borrowers that help us to understand how we can make that process more fair. We have made laws to say that you can't discriminate in lending, and that includes automobile lending, but there's very little that we can do about that if we aren't able to recognize there's a problem.
And so while we're able to say we have some problems at the neighborhood level, we can't even say definitively that they're individual based problems neighborhoods that are 50% or more of a given race or ethnicity group. We can be pretty confident in that, but we're able to get a lot more confident and work toward resolutions if we have those data attached to the loans themselves. That's a important first step: developing a way that we can evaluate whether discrimination is occurring or not in automobile.
Damien
Do you think that's the next step? Taking the data to like the next level, or is there, is there something else that you think needs to be done.
Sam
In the short run, Damien, yeah, the sort of the low hanging fruit here is to have a to be able to give in a value, a metric that can be evaluated for the laws that already exist. In the long run...to what we talked about earlier, we looked at the world we live in now, but there are things that we can do in the long run. (And yes, we can get back to not bike lanes and public transit and so in some cases.)
One of the things that we found is that automobile debt burden. And remember, that's all the debt in the neighborhood compared to all the income. And so automobile debt burden is lower in places that are closer to rail transit stations. And some of this is a little bit of chicken and egg. On the one hand, a rail station may enable families to have lower car ownership. On the other hand, those families that wanted lower car ownership may have moved near a rail station and so it is a little difficult to tell exactly how that goes. What does become clear from this is that whether there's self selection there or not, households living closer to rail stations tend to have lower automobile debt. We used rail stations, thinking that that proxies a higher quality of transit service than the state the bus typically offers when we're looking at a statewide data set.
The other thing to highlight, though, is what works in parts of Los Angeles, San Diego, Sacramento and San Francisco. But I don't think that the public transit, increasing public transit supply as much addresses the issues in, say, the agricultural region of the Central Valley and Imperial County. Where I think the state resources can be directed to help those who live in rural communities that are 100% car reliant to fairly and affordably afford an automobile, and to help with that upfront cost in a way that can be less discriminatory and predatory.
Damien
All right, so I always joke that we're approaching our artificially created time limit, but we are getting close to a half hour, so this is when we start to wrap up. Is there anything you would like to add or have a closing thought here?
Sam
We talked earlier about how we're focusing on our study here and in our discussion today about the world we live in. In planning, we often think about the future in a good way. What a lot of us in transportation planning are focused on is, how can we make things better in the long run? And in a lot of cases, I think we sometimes shortchange the medium, and in particular, the short run. I'll give you an example. My first career was as a high school English teacher. It was really tempting to think about how we could move the needle of educational inequality and improve outcomes for kids.
A lot of that requires a long runway. It was an interesting challenge in that role, and a difficult challenge to understand that the students in front of me in a given year did not have the luxury of that long runway. That they're there now.
What this research highlights is there are people now, and these data are already a few years old, there are people now who face these challenges. We can improve that by building out a more robust, more frequent transit system.
There are important ways that we need to focus on helping people now. And so while I agree public transit can be part of this solution. I really encourage people who are listening to this to also think about what are the immediate solutions that we can offer for people who are in financial precariousness because of their need for an automobile now.
Damien
That's a great city council speech. I think you're ready to run for office now. There are city council elections in your city, I think next year. So you know, work on pulling papers. And thank you so much for being here today. And again, I wasn't kidding earlier when I made a joke that I was late to the podcast and Sam was sitting here waiting for me. So I want to call out his extra patience and my appreciation for that
Sam
Damien is worth the wait. Thank you.
Damien
Links to his research and summaries and all of that stuff is all available in the text that accompanies this podcast if you're listening somewhere that is not Streetsblog California, you can find it at c, a, l, dot, streets blog, dot, O, R, G, you.
