Credit scores may be more sinister than we realized. Liberty and Scott are finding out just how antagonistic they are and what alternatives could revolutionize the current system.
We face many overwhelming challenges in America today: systemic racism, data privacy, and political misinformation. These are big problems, and there are a lot of opinions and ideas on how to fix them. Scholars and industry experts often disagree on how to find solutions. So, how can we find the right way to move forward? We let the data speak for itself. Join hosts Liberty Vittert and Scott Tranter as they gather data and get the facts about today’s most pressing problems to find out: are solutions even possible? They’ll investigate with MIT professors dedicated to researching these issues, and talk with the people on the ground encountering these problems every day so that we can find the best solutions that triumph over these challenges and solve America’s biggest problems.
Data Nation is a production of MIT's Institute for Data, Systems, and Society, with Voxtopica.
Munther Dahleh (00:04):
Welcome to Data Nation. I'm Munther Dahleh, and I'm the director of the MIT's Institute of Data, Systems, and Society. Today on Data Nation, Liberty and Scott are investigating credit scores and how this data affects certain people's access to fair housing loans.
Liberty, one of the things we've got to do in this is we've got to establish kind of why we're here with credit scores and things like that. We looked during the 20th century around redlining, which was very common in America. Redlining goes all the way back to the New Deal as African Americans were purposely left out of the suburban housing offered to Whites during the New Deal. In every city, lenders and real estate agents had access to the color-coded maps, and each color corresponded to the riskiness of lending in that neighborhood. If an area was red, it meant that it was high risk for no repayment on a loan. These areas were concentrated in the inner cities unless held most of the African American population.
Well, yeah, I mean, until the Fair Housing Act of 1968, it was also perfectly legal in the United States not to loan to African Americans under this false premise that it was higher risk. And in addition, if an African American could get a loan, the Federal Housing Administration made it illegal for homes that they subsidized to be sold to African Americans, sort of unfathomable racism. And the Federal Housing Administration believed that the property value of homes would go down if any African Americans moved into the neighborhood. And what's interesting is that their racism led to making less money because the home values actually went up when an African American moved into the neighborhood.
Say someone gets through all that red tape, homeowners themselves still resisted having African Americans in their neighborhood despite it increasing their home values, at least that's what data showed afterwards, despite claiming all these false beliefs of the contrary. It's interesting. If you look at Linda Gartz' memoir, Redlining Remembered, she reflects on how her own parents did not want to rent to anybody that was African American as they thought it would make their property value go down. They even banned her from inviting her friend who was African American to her birthday party for fear that someone would see that and make it harder to sell their home. It's this kind of bad data or bad analysis that you can see kind of just percolates or snowballs and perpetuates this type of stuff.
Well, a lot of people see redlining as a thing of the past. Surely, everyone has access to equal opportunity to good properties, but that's not actually the case, because today in America, it really could be argued that redlining is still happening with credit scores. And we wanted to know exactly how, in 2022, this could still be happening. We talked to Yuri Beckelman, who has spent over 14 years working in Congress as a staff director, deputy chief of staff, and senior advisor to members of congress. Yuri's legislative expertise can be seen in dozens of bills that he's worked in the past; from veterans issues to consumer protection and everything in between, even pending bills that would ban credit scores. Yuri, we've heard why credit scores are bad. Why is that the wrong way to do it, and why is that still being done when, I think, according to you, it's not legal?
Yuri Beckelman (03:30):
Yeah. Well, here are the indicators of why someone is a good tenant or not. It is about their level of income. It has to do with the payment on rental properties in the past and their eviction history. And none of those things show up on a credit score. That's not a good indicator of why someone should be eligible to rent an apartment, and it's pretty frustrating. But what you see is landlords who falsely equate your ability to pay a credit card on time and regularly to whether you'll be a good tenant or not. And I understand that argument. I hear it, but I think it's relatively unfair because your willingness or your need to prioritize what gets paid back first is not your credit card. It's your house. It's things that can be taken away from you.
Your unsecured debt is the last thing that you are going to pay back if you are going through a financial challenge. However, that's what you've been graded on, whether you would be a good tenant or not. Additionally, and I think that's best case scenario. But worst case scenario, you see landlords who are setting up a flag that they don't want to deal with a certain type of tenant that might not have a good credit score. And that good credit score, bad credit score can be created by a whole lot of things. I think you've heard a lot about this: There's 45 million Americans that have no credit score or are so credit thin that they don't even show up with a score. That's just a huge... And that number goes up to, I think it's 28% of African American consumers aren't scorable. They would look at a rental listing and say, "I'm just not even eligible for this so I'm not even going to apply."
So what you're saying is that credit scores are basically allowing landlords to be racially biased in some sense without actually using race as a factor? So, it's like a pseudo factor that's being used.
Yuri Beckelman (05:22):
Yeah. Here's the thing. In DC and many cities, you are not allowed to make a final determination based on a credit score. However, you are allowed to include a required credit score, which sounds crazy, right? And they're definitely flirting with the border of that law and it's really frustrating. And you would say, "Well, why would someone do that?" Well, there's implications for having bad credit, and being poor is very expensive. And why would not low income individual, seeing that they're being taken advantage, move somewhere else? Because they are terrified that they don't have credit or they don't have good credit and they don't feel comfortable applying for some other place, and so they get stuck where they are.
So, when I threaten my landlord, "If you don't fix my faucet, I'm moving out," because I know I can move to another apartment, I have the ability to do that because I know I can move to another apartment? But somebody else wouldn't?
Yuri Beckelman (06:12):
Absolutely. A people don't have that. They are terrified of the prospect of moving, because if you're homeless, if you move out, and you don't have credit so you can't move somewhere, that is a terrifying feeling.
Are you saying that the concept of redlining, basically the active way of dividing and segregating out people of different economic and ethnic backgrounds... Is the credit score kind of the modern tool in which that happens? Redlining was a purposeful thing 30, 40 years ago. It's illegal now. People would say they're against it, et cetera, et cetera, but the 21st century of redlining, credit scores are essentially recreating that inequality and that problem.
Yuri Beckelman (06:49):
I don't think that there's an active movement by banks or even by landlords to actively red line. However, there's a new caveat to this. If you Google: "Do I need to have a good credit score to rent somewhere," there are lots of places that give you great ideas, how to find landlords who are going to rent to you even if you have bad credit. Talk to that landlord, offer to pay up front, all of these things. However, as we see less and less single family owners of properties, because more and more single family homes are being purchased by large private equity firms like Blackstone and turned into massive kind of operations that are managed by their corporations, by an entity underneath the corporation, they don't follow that. You can't go to Blackstone and be like, "Hey, I had a medical problem that knocked down my credit score. You should talk to my past landlord. He would say I'm a great tenant." You can't do that.
It's one thing to say things need to be fair and equitable, but you also have to look at the business side of things. The people who are doing the lending, the landlords, they need to get paid. Are any lenders or landlords trying alternative ways of checking potential tenants? And if so, what is the incentive for the landlord besides just being fair and equitable?
Yuri Beckelman (08:08):
There's a service by TransUnion, which is, in the credits field, called SmartMove. And what they do for a landlord is they do a full check. They look at credit history, which is different than an arbitrary credit score. They look at criminal history, they talk to past landlords, they make the claim that they have a 15% higher return on investment for people who use a system that looks at all of these other key determinants rather than just your credit score. So, yes, absolutely, a landlord is going to do better if they take the time to figure out who the renter is going to be.
We know it's easier to collect data on people today than it was 30, 40 years ago when credit scores were being created. We have the tools to collect the data, but how does it really get fixed? Specifically, how does government and industry work together to fix this?
Yuri Beckelman (08:58):
You can do that partially by industry, stepping up and doing this, but you can also do it through some forms of legislation and turning and passing a bill that says into law. That's something that says that you're not allowed to include credit scores as a rental requirement. You're not allowed to use it as a determining factor of someone's eligibility. And I think you would see more and more of the credit reporting industries offering a tool for landlords that would still allow them to access a credit history, which would show a lot more detail about whether someone actually paid their debts or went through a divorce or had to medical history problem-
Forgot they had to pay the gas.
Yuri Beckelman (09:33):
Forgot they had to pay the gas. Or here's an interesting one. Often times, you have abusive relationships where the spouses doesn't have a credit history because it's all been under one person's name, and they leave and now they're stuck. And that's a scary proposition. I think that as we start having more and more access to data, that is one of the major contributor and a major factor in how we make decisions on a ledger of whether this will work or won't work. It can't be the only one. All of this is an art and not a science.
Liberty, it's pretty clear to me that credit scores are intentionally designed to be antagonistic or purposely pushing people down in underserved communities, but they're also not helping them build wealth or move up the social ladder. In fact, they seem to be a big reason why people can't build wealth or buy property in these areas they'd rather live in.
And it's not just hurting a small proportion of Americans, but just this sort of shocking amount. Almost 75 million people are being kept from, really, the American Dream because of how their credit score is used against them.
The question is, at least to me, what can we do to solve this? It's obvious that the system of looking at one's credit score to determine their housing needs needs to be adjusted. But this is a complicated issue that involves banks, governments, lenders. And again, the banks and lenders and industry, they need to be in a place where they can recoup their investment and make money, but also be encouraged to fix the system. And so, we're going to bring in an expert, Professor Munther Dahleh who is the William Coolidge Professor of Electrical Engineering and Computer Science, and director of the MIT Institute for Data, Systems, and Society, to talk to us about it today.
Professor, we're trying to figure out how to solve this credit score and housing issue. It seems like the biggest pushback is creating a more "fair system" that lenders and landlords still allow them to make money and manage their properties fairly. Their concern seems to be that if credit scores aren't used to determine if a borrower or tenant will make the payments, then how can lenders know whether or not they'll be paid? How do they reduce their risk? In your opinion, what's the best new system or adjustment to the current system that will allow the banks and lenders to loan, make back their money, coup their investment, but also create an equitable system for people looking for housing?
Munther Dahleh (12:00):
Let me just say that there are short-term and long-term things. How do we do a short term fix? Well, first of all, credit scores could actually focus on things that are more predictive, bring in more features into the score than just the history of credit cards. You could have 20 years of having paid your mortgage and you have a $30 dispute with your bank that has not been paid for 60 days. You may lose a hundred points on your score. They're weighing it in a way that is more advantageous to the lending companies than it is to the consumer, and I think we need to change that a little bit. 20 years of paying your mortgage is a hell of an indication that you are going to pay your mortgage, and that $30 dispute with your bank is not an indication that I'm not going to pay my mortgage in the future. But it's really the people that don't have the money are affected the most by these things.
Notice the coupling with these things. People that don't have a lot of money also don't have great banking services. They end up running into these problems a lot more than people that have a lot more money, that have a better banking service where, if there is a little problem of your bank credit card not clearing, you get a phone call and then they say, "Oh, we have this problem." And then you resolve it on the phone and it's over. Some other people don't get the phone call and they just get a letter from a collection company saying you have not paid your debt.
All of these things are coupled, but the third technological component that is added to all of this is the appraisal. Appraisal is also another instrument, technological instrument, that has segregated what is considered sort of good property versus not good property. Right now, if you think about suburbs, they're essentially synonymous with White affluent people, and then inner city is essentially associated with non-Whites. And if you look at Zillow or Redfin and so forth, you would see that the appraisal of these areas is reinforcing the segregation. So, then you qualify for a mortgage that is low enough that you can only buy in an area that is appraised law, which over time does not appreciate, which over times does not help you create wealth, which is exactly what we're trying to do with this home buying is to create wealth for the people that don't have wealth.
Okay, Professor. I got to ask a little bit more directly. Do we keep the same system and make adjustments to it, or do you think we need an entirely new system and a method to establish as someone can buy or rent a home?
Munther Dahleh (14:33):
In the short-term, I think fixing the credit score is important. In the longer term, I think abolishing the system where it's all these technological instruments that are playing in a concerted way and going to another system of looking for other data that is indicative of your ability to pay your mortgage and pay your rent. I have to say that there are certain banks who have a history of giving mortgages that don't sell the mortgage. They keep the mortgage. Those banks do look at the portfolio. They look at you as a whole. They look at your education, your income, your actual history of paying rent and paying your mortgage and so forth, and they give you a reasonable mortgage. I think we want to do more of that than the sort of pure numerical approach of just looking at a few numbers and making a decision on you.
It seems like in the long term, the best way is to create a new system. Where would we start? How would you start a new system from scratch? What's your vision?
Munther Dahleh (15:30):
That's a great question because I would build a system that is not entirely based on numbers, but one that has a social fabric, social support in it. For example, I think we are in a system right now where it's merciless. You have a hard time falling behind, so there's no robustness in it. You're working and you're paying your mortgage, you're paying your rent, you lose your job, you could lose everything. And I think that doesn't mean that you couldn't dip a little bit and then come back. There should be a lot more support for you to not pay in another system that can help you out sort get out of that hole and then help you sort of build more capital and then pay the debts that you have.
I would do that first, start thinking about a system that has social support, potentially a state support to some of these things, like some of the programs that we do have for housing, that do allow the state to help people in terms of the down payment or kind of adjustable interest rates or co-ownership programs and so forth that allow people to come in and not be able to afford the house. Co-own the house with them and allow them to buy out that house over time. Again, allowing you to be in the market at the level that you can afford in an area that you would like to live, but then I come in as a state and co-owned that with you, and then allow you to buy out for me. We need to evaluate some of these systems and see if they work. And of course the state takes a little bit of the brunt of the defaults, but over time, I think we're actually helping a large population becoming independent than owning a lot more wealth.
China implemented their social credit system in 2014. And it honestly seems straight out of a dystopian novel, where both the citizens and the businesses are scored on their behavior. Citizens can be marked down for things like bad driving or smoking in a non-smoking area, which frankly, I think people should be marked down for. Just kidding. Buying too many video games or even just posting on Instagram too much. And punishments can be minor like slowing down your internet or keeping you out of high end hotels, but they can also be really severe punishments that restrict domestic and international travel, bar you or your children from continuing education or even a job. And while this is, of course, an extreme example, it goes to show that having one score that ranks people's behavior can be a very slippery slope. And some people might look at China and be concerned that we could be heading in the same direction. What is that fine line, the boundary to prevent this from becoming a slippery slope into this problematic system that China has implemented?
Munther Dahleh (18:17):
I think that there is no slippery slope. I mean, the system in China is very different than the system in the United States. And China was never a democracy and it's a top down institution and it's a different system altogether. Some things work and some things don't work. And I don't think that we are going to see a trajectory moving in that direction. Setting up the correct mechanisms is very important, ones that incentivize people to do the right thing.
That's interesting. It seems that we don't need to be too worried about ending up like China and their overreaching social credit system. That's good. We don't want to be there. But Professor, you mentioned earlier something that might concern some people that the state would take a little bit of the brunt of the defaults, should we move to this new system. And by state, that's really taxpayers paying for this, right? Is there evidence to convince the taxpayer that this new credit system actually creates more wealth and brings the whole country up and is better for everyone in long term?
Munther Dahleh (19:19):
I think that there is very strong evidence that depletion of the middle class is affecting the overall economy in any country. The system we're at right now is depleting the middle class. What it's doing is actually allowing rich people to build more and more wealth, and middle class people and lower middle class people, having it harder and harder to build wealth. That is increasing that gap. There is a lot of evidence of that. While in fact it may appear that the taxpayer is taking on some of the responsibility, actually in the overall benefit of the society, the healthiness of the society with a very healthy middle class, better education, better schooling, better professionals, all that comes from the middle class. Having a healthy middle class is an overall positive to the society.
It's clear that establishing a new method of credit scoring specifically for home renting and buying is crucial for America and that there are alternative methods to creating a more fair credit scoring system.
Access to housing, rents, all those things have been something political leaders have been trying to figure out the best way forward. President Biden, during his presidency, wanted to introduce a tax force that would attempt to correct the inequitable way we score people's credit. Under his administration, federal lenders would be required to use this new system established by the Consumer Federal Financial Protection Bureau. The agency was designed to force borrowers to use a fair credit scoring system, and one that takes into account non-traditional, as Yuri talked about, to more equitable metrics like paying rent or paying utility bills. Important aspects so we can continue to adjust the system.
Big business is making this almost impossible and, frankly, shooting themselves in the foot at the same time. We learned today that landlords and lenders can actually make more money if they make a more equitable system, which like... How often does that happen? By creating a more equitable system, businesses would be doing the right thing and help their bank accounts. From the single apartment landlords to the BlackRock of the world, to the political leaders signing legislature, it's clearly time to both do the right thing, stop modern day red lining, and make more money in the process.
Thanks so much for listening to this episode of Data Nation. This podcast is brought to you by MIT's Institute for Data, Systems, and society. And if you want to learn more about what IDSS does, please follow us at MIT IDSS on Twitter or visit our website at idss.mit.edu.