Throughline - What's Your Worth?

Episode Date: May 4, 2023

The credit score: even if you don't think much about it, that three-digit number can change your life. A high score can mean the keys to a new apartment or a new car, while a low score can keep you locked out of the American Dream. Around 40% of people in the U.S. have a low credit score or no credit score at all. So what happened? Today on the show, we talk with media historian Josh Lauer about credit's origins as a moral judgment, and how a tool intended to level the playing field has instead created haves and have-nots.Learn more about sponsor message choices: podcastchoices.com/adchoicesNPR Privacy Policy

Support for this podcast and the following message come from the NPR Wine Club, which has generated over $1.75 million to support NPR programming. Whether buying a few bottles or joining the club, you can learn more at nprwineclub.org slash podcast. Must be 21 or older to purchase. Why are you checking your credit score? You don't want to live with mom and dad forever, do you? You know, your credit score is the key to getting a credit card, a mortgage. I feel like the world is my oyster, frankly,

because I have a great credit rating. Credit invisible. You were credit invisible when you got to the I was credit invisible, but I'm not alone. Credit card balances especially have hit a new record at the end of last year, nearly $1 trillion. 26 million Americans are credit invisible. It's terrifying. It means you have no, you're nowhere on the credit score. I went from 768 to 753 because this is where I bought my new car. Gee, Mr. Money, do girls have to learn all this about credit too? Repairing your credit? It's a bit like losing weight. It takes discipline and it takes patience. But most people have no

real idea what goes into a credit score. How about giving me a little credit? Nobody gives you credit, John. It's something you have to earn. So the credit score is uh it's kind of this like mysterious ephemeral thing it's something that changes that defines us in a lot of ways but many of us don't really understand it could could you break it down a little bit like exactly exactly what is a credit score, who creates it, and why it's really important right now, particularly living in the United States? Well, a credit score is essentially a numerical ranking of your likelihood of paying back a debt. That's really what it boils down to. That's Josh Lauer, author of the book Creditworthy, a history of consumer surveillance and financial identity in America. Well, and there's something really, I think, interesting and telling about the title of your book, Creditworthy, not the full after the colon title, just the first word, creditworthy.

Because especially today, you know, that word worth, it's come to be almost a value judgment of a person's worth. And I wonder if you can just walk us through to what extent you see it that way. So it's really only fairly recently that credit worthiness, credit scoring has been redefined in terms of risk and metrics. For most of the history of American credit, creditworthiness was explicitly a moral judgment. And that judgment was comprised of what was known as the three C's, capital, capacity, and character. Capital, as in the cash you have available to pay back your debts. Capacity, meaning your work experience and ability to earn a living. And as for character... Your personal behavior, your personal morality was always judged to be the most important variable in the three Cs.

So it's really only fairly recently that lenders have begun to try to detach some of the moral language from credit risk assessment through the use of scoring. And this ranking is based on data that's collected by credit bureaus. And this credit bureau data is then processed by an algorithm. Processed into a number. Your credit score. That three-digit number is like a passport. It determines your access to things like mortgages, credit cards, loans. It can mean the keys to a new apartment or a new car. But if your score doesn't measure up, it can mean you're out of luck. But okay, the question of how your credit score is calculated is one thing. But let's

zoom out first. What do these scores represent? And why do we even have them in the first place? Here's what surprised us. Talking to Josh, we learned that initially credit scores were supposed to make things more fair, to make it possible for people to get credit and invest in themselves and their businesses, even if they didn't have the quote unquote right credentials. But today, a credit score can shut doors as well as open them. So what happened? In this episode, we're going to answer that question. We'll look back centuries to the roots of the credit system to understand how one small number has come to define our lives and who really benefits from its existence. This is another episode in our new series we're calling Past is Prologue.

Ramtin and I sit down with journalists, historians, and big picture thinkers to give big topics the through-line treatment, go deep on why things are the way they are, and start imagining the future. I'm Ramtin Arablui. I'm Rand Abdel-Fattah. And you're listening to ThruLine from NPR, where we go back in time to understand the present. Coming up, how a tool intended to

level the playing field has ended up dividing us into haves and have-nots. This is Tony in Almbaville, Tennessee, and we are all listening to ThruLine on NPR. This message comes from WISE, the app for doing things in other currencies. Send, spend, or receive money internationally, and always get the real-time mid-market exchange rate with no hidden fees. Download the WISE app today or visit WISE.com. T's and C's apply. What do you think people misunderstand about the credit score? I think what people don't understand about the credit score is the way that it's produced. It is mysterious to people how the credit score goes up and down.

And that's partially by design. So the developers of credit scores don't want to share with the public exactly how their algorithms work. And the reason is that they fear that consumers will try to game the system if they know somehow that if they do X behavior, it will change their credit score. So there are efforts to try to reverse engineer it. But really, the algorithm is obscure on purpose. Which is a little bit ironic, because originally, credit scores were supposed to make things more transparent. Before scoring, credit was based more on feelings than on facts. Before, the use of credit scores were

common, and this is really only in the 1970s and 1980s, the way that you were approved for credit was that you went to a bank or you went to a department store and you sat down with another human being who took the application and made notes. And part of the assessment among professional credit managers was to look at the person and to make a judgment about their morality, their responsibility through the interaction itself. So you can imagine there are all kinds of opportunities for bias and discrimination. So in many ways, the argument for the use of a credit score is that it's much fairer than having an individual sit down with another human

being and have them judge them. And like I said, the judgment was in part their moral disposition, whether or not they could be trusted. That's so subjective and so context-based, where you judge someone's kind of moral character to be. So if we just take a step back, credit is a very old thing. The idea of someone lending someone else credit or lending someone else goods based on like whether they're going to pay them back or believe they're going to pay them back. But at least in the United States, when does this idea of judging people in this way with the three C's, as you described it, when does that start? The credit economy used to be based on interpersonal relationships. You knew the person you were lending to,

or you knew their family, or you knew somebody you could write to to ask about them. And again, these are, of course, all subjective interpretations, but it's a basis of trust that is personal. People often did not have cash. They would go to a lender or go to a manufacturer and promise to pay them when their crops came in and they sold them. So credit was completely commonplace in 19th century America. It's not something that's new that has only arrived with credit cards. And borrowing was often either, you know, in a consumer level, it might just be food or domestic items. But at a commercial level, it might just be food or domestic items. But at a commercial level, it might be a farmer coming into town trying to get equipment or a shoemaker trying to get leather

in advance from a big commercial center and agreeing to pay later when the goods were sold to farmers in the local area who were waiting on their crops. So these long chains of credit. In the 1830s and 1840s in the United States, you begin to have the expansion of markets, the growth of populations, the changes in transportation so the goods and people can move around. And you have an increasing interaction of strangers with other strangers who want to do business with each other. But as the country got bigger and more complicated, it was much more likely that you would not

know who wanted to borrow something from you. And you gradually begin to have credit reporting organizations that emerge whose sole function is to collect information about the character and capital and collateral of people around the country, centralize this information, and then to provide credit reports, which are narrative documents, they're not scores or numbers, to subscribers who request this information. Wow. And that, I mean, that's fascinating, because obviously, yeah, the world is changing so fast, as you said, right. And so it's a lot more strangers interacting with strangers, especially in the post Civil War era. I imagine that, you know, what used to be mainly probably white

landowning men who were qualifying for bank loans and things like that was being challenged. And so I guess I wonder, what is this desire to standardize the process responding to in terms of kind of the social forces and the demographics that you mentioned? So you mentioned the Civil War, and that's sort of a turning point for the development of consumer credit reporting. After the Civil War in the 1870s and 1880s, you began to have organizations devoted just to collecting information about local consumers. So the people that lived in a town or a city,

and they would centralize information about whether they were viewed to be creditworthy or not, whether they paid their debts. What kind of information would they be gathering? What kinds of things would they have been writing down? They wanted to know things like your race, your age that might have been circulating in the community, whether somebody had been arrested, whether they were known to be a drinker, whether they're known to be a gambler, whether they're known to be a philanderer. Because these are all things that end up being drains on a person's income, and not to mention showing up for work and other problems which affect income. But it was information about the total person and especially bearing on their personal character. Were they the kind of person who was stable, who was conformist, who was normal? By normal, being a white person in particular, being heterosexual, and being a stable member of the community, a property owner, somebody who is regularly employed. These are sort of the marks of somebody who is creditworthy. And these ideas are embedded directly into the credit reports. The files themselves collect this kind of information,

but eventually those pieces of information are converted into credit scores or credit ratings, as they're called in the late 19th century, which are not statistical, but they are a hierarchical set of codes to indicate whether a person is creditworthy or whether somebody should be asked to pay cash or whether they should be refused altogether. So when I'm saying credit reports, I'm talking about credit bureaus, which are local. They're in a town or a city. Some cities have more than one credit bureau, and each of them is collecting information about all the credit bureau and ask for a report on this particular person. That credit bureau gives them that information over the telephone, and then they can make a decision whether they want to lend to the new customer or not. So there are reference books

that are published by major commercial credit reporting firms in the 19th century that are basically like a telephone book filled up with people's names and their profession, and then a little code next to it that indicates their creditworthiness. And it's usually from C to A plus being the top. So an A plus or an A1 rating is top credit. And a lower ranking indicates declining creditworthiness. You think about what happens after the Civil War, obviously with the Reconstruction, and suddenly these formally enslaved people who have the ability to start businesses and try to get money, etc.

Was there any effort there to try to make the score a way to, like, equalize that or make that more fair? Or does it actually just make it worse or easier to discriminate against people? But there's really no effort. And in fact, the only evidence that Black Americans were included at all in the white credit economy was in some credit rating books that you can find that still exist. Sometimes there's a code next to their name, a letter C for colored, for example, and they're rare. So, I mean, it tells you something about how uncommon it was for white people, or at least in the official white credit economy, to lend or to sell to non-white people. So it was completely

unfair in terms of how credit was allocated. And in fact, people of color, they were more likely to buy things on the installment plan, which required you to put money down. And it was a durable item that could be repossessed if you didn't make your payments. So a sewing machine or some other domestic item, furniture, for example. It really was a narrowly conceived idea of what creditworthiness was, and it was primarily in terms of white people. Obviously, this is pre kind of algorithms as we know it. But is it fair to say that, albeit primarily in terms of white people at that time, it was an attempt to make what would have been before purely based on kind of human subjective assessment into something at least ostensibly more scientific or more calculated? Is that fair to say?

Yes. So one of the discussion points in terms of the development of credit reporting was that all Americans had an equal chance. So a credit report, if you collect a credit report or credit information about a mechanic and you collected the same information about a banker, it's going to work out the same, that if you were a person of lower economic status, you should be able to be trustworthy just by virtue of paying on time and having that information reported to credit bureaus. So credit reporting was absolutely promoted in terms of democratization. But it was really only democratization for white Americans. And this, like I said, this continued well into the 20th century. And in fact, it wasn't just

white Americans, it was mostly white men. And women, white women who had accounts at department stores, for example, their information was collected in terms under their husband's name, or under a father's name, a male guardian. So women did not have their own credit accounts, really, in the early part of the 20th century. And again, this is something that happened much later in the 1960s and 1970s and was part of the civil rights movement and also part of the women's movement to make credit more equitable, to gain credit access for women and people of color. Coming up, credit for all at a cost. Hello, my name is Lydia Witherspoon Threep, and I am calling from Hinesville, Georgia, and I just wanted to say I love ThruLine. Thank you, ThruLine. for their inherent craft, each hotel tells its own unique story through distinctive design and immersive experiences, from medieval falconry to volcanic wine tasting. Autograph Collection is part of the Marriott Bonvoy portfolio of over 30 hotel brands around the world.

Find the unforgettable at AutographCollection.com. If you're listening as a subscriber to ThruLine+, we just want to say thank you. And if you're not yet a subscriber and want to learn more about how to listen to the show without any sponsor breaks, head over to plus.npr.org slash ThruLine. Becoming a plus subscriber helps support all of our work at ThruLine. So we hope you'll join. Now, back to the show. Today, your credit score determines your access to loans, mortgages, apartments, the list goes on.

It's a way for lenders to estimate your ability to pay back a debt. The scores are relatively new themselves, but credit is a concept that has existed for thousands of years. In the 19th century United States, cash was in much shorter supply, and people didn't necessarily have steady income. So they'd buy what they needed on credit, then pay merchants back after, say, the harvest. And that could be months later. Back then, local credit bureaus compiled information on consumers' wealth and reputation into reports that judged their credit worthiness. It was a laborious process. And as the 20th century began, more businesses saw the potential for credit scores to save them a whole lot of headaches and make them a whole lot of money.

We pick up the conversation with media historian Josh Lauer. Going through the 20th century, another thing really changes in society, right, as far as I understand, which is that, you know that consumerism increases. More people have access to more products, things like refrigerators and then later televisions and radios. And so it would seem to me more individuals are trying to access credit in order to buy items that they may not necessarily have the cash for. How does that change the dynamics and need for a credit score? It's mostly department stores and other mass retailers who want to increase their profits by selling to as many people as possible. And often by offering a charge card or a charge account, they can increase the volume of their sales and then maximize their profits. So banks really don't get involved in lending to individual consumers very heavily until after the 1930s. The reason that they don't get involved is because they have a real bias against lending for consumption. It's hard for us to think about this today, but there was a time

where even though credit was widespread, even though in the 19th century people borrowed for consumer needs all the time, for food, domestic goods, for furniture, there was a bias against credit used for consumption. The privileged form of credit was for productive purposes. If you were borrowing to build a business, that's fine. I mean that's a legitimate use of credit. But if you're borrowing because you want to buy new furniture or you want to buy clothes, this is something that seems illegitimate. It's still happening. But from the banker's perspective, they are against this kind of consumer borrowing or they're skeptical of it. But the other thing I would say about this, and this may be a little bit inside baseball, is one of the reasons why banks don't get involved is because they don't really like to work with the credit bureaus.

The credit bureaus, they think, are these illegitimate businesses that are on the margin. Bankers saw themselves as sort of the epigee of the business community. They are the keys to the kingdom, and they imagine themselves as being the most prestigious members of the community in terms of commerce. And they look down on credit bureaus as being sort of beneath them in terms of the hierarchy of the business class in their towns. But soon the bank's position on lending would start to shift.

Technology was starting to develop at a rapid pace. And that's what laid the groundwork for what we now know as the credit score. In the late 1930s and 1940s, there is a study of consumer credit scoring that is done by a statistician. And he basically applies some new statistical techniques and says it's entirely possible to develop a statistical model to try to predict credit worthiness, to predict credit risk of the individual. At the time, this is before computers are commercialized and available, the guy says, it can be done, but it's so complicated and so time consuming, it's probably not practical. It's expensive to keep all the files. It's expensive to have people working in a credit bureau and answering the phone and doing all this bureaucratic labor. So this is one of these things that we take for granted now because we use computers and information flows much more easily. But it was a serious paperworking endeavor to have a credit

bureau and collect that kind of information. So it's shelved in 1940. But after World War II, you have the introduction of computing and you have its commercialization during the 1950s. And you have a company that begins to experiment with applying statistical principles to computerized data to try to develop a credit score. And this is Fair Isaac in California. So they began to develop the first credit scoring models using this kind of data and, importantly, using computers, which actually makes it practical. And that's sort of the origin of modern statistical credit scoring. Fair Isaac & Company would go on to create the FICO score. You probably recognize that name. It's become the industry standard. But long before FICO, in the late 1950s,

a finance company hired them to develop its first credit scoring system. And one of the things that sort of propels the computerization of credit bureaus during the 1960s is you have banks who want to have this information more quickly. And the first computer credit bureaus start to emerge in the mid-1960s, and the entire industry begins to computerize. And it's partly in response to these credit card companies that, after experimenting with giving credit cards to people without doing credit checks and getting burned badly and losing a lot of money, they decide to start using credit bureaus and to try to find the kind of information to make better decisions about who to offer a credit card to.

So walk us through that development, because this seems like a really significant shift. Can you describe the intersection of the rise of credit cards and also the rise of the civil rights era and how those two collide? So one of the things that sort of inspired changes to the way credit reporting and the way that credit scoring was done was the fact that many women and people of color could not get credit access. They couldn't get credit cards. Credit is absolutely allocated by mainly white men who are credit managers and who are meeting with credit applicants and making these kinds of decisions. For white middle-class women, they were angry because they could only get a card if it was in a man's name, in their husband's name. And you had women who were into

the workforce, they could support themselves. They didn't need a guy to have a credit card and they wanted to be able to have that kind of access. And the same sort of dynamic played out with people of color. They wanted to be able to have credit access, and they were denied for a whole host of reasons, some of which were just explicit racism in a credit interview with a white credit manager, but also with the development of new credit scoring technologies that had variables that were proxy variables for race that ended up discriminating against them. So the development of credit scoring in the 1960s and 1970s is really seen as a technology that has the potential to democratize credit in ways that are not discriminatory. And this gets to some of the debates over credit scoring, because the idea is that

if you take the variables out that are discriminatory, if you take out gender, if you take out race, if you take out marital status, then the computer program will automatically, you know, make these decisions without seeing these people or having the opportunity to be prejudiced or biased. And as it turns out, it's more complicated than that, because as we learned today with algorithms, there are all kinds of ways that variables can become proxies for protected characteristics or protected classes and can discriminate unwittingly in the process. Initially, credit scoring was very closely tied to discrimination because they used variables such as zip code.

And zip code, of course, can be used as a proxy for race. So just having a variable like zip code in a credit score or a credit scoring algorithm could reproduce these kinds of discrimination. The idea of credit scoring is that it will be democratizing and that it will do away with all these other problems that come up when you have individuals judging other people. You know, you talked earlier about the FICO score, but I want to sort of understand when that is introduced. And for what reason do civil rights leaders see that as an opportunity to kind of create more economic equity, basically. In terms of the credit score and the FICO score, early credit scores are done using credit applications. So if you are Macy's and there's another department store, the credit

scoring company, if it's Fair Isaac, would come to your store and would look at your customers over a period of time and would develop a model that predicted credit risk for your customers. Each credit score was a custom product for each client, whether it was a bank or some other retailer. If you were Bank of America, you weren't supposed to use Citibank's credit model because you had different populations. However, because credit reporting, credit bureaus were beginning to computerize and they were beginning to consolidate, by the 1980s, you had a small number of credit bureaus that use data from a national sample and begin to make predictions about risk that could be used by anyone in whatever context. And this is really what the Fair Isaac FICO score is. It's a credit bureau score. It's a generic credit score. But most importantly, it's based on credit bureau data. It's not based just on the client's own customer

data. And so this is a completely different way to score consumers and a completely different way to imagine risk as a national pool rather than as just your own local customers. So on the question of democratizing and making credit access more fair, the credit score is supposed to be able to do that. And in many ways, it does. I mean, credit scoring makes it possible for lenders to make credit decisions much faster and more cheaply. Again, remember, before the development of computerized statistical credit scoring, each

credit decision was made by an individual, either taking an application and meeting with another person or using some rudimentary scoring system. So it's really hard to imagine living in a modern credit society like we do without having some sort of technology that can actually process applications this quickly. So the FICO score is introduced in the early 80s. How do we know that it started to have that effect of democratizing credit? Were more, you know, houses being sold to, you know, Black homeowners, for example? Were women, single women able to access more credit? How do we know that it was making a difference?

That's a really good question. I don't know that I have a, you know have a concrete answer to that. But I can say that in theory, these scores should not be discriminating against people of color and they should not be discriminating against women. In terms of homeownership and especially how much people pay for mortgages, we know that that's not fair. And there are still problems with that. There have been studies that have done that suggest that black families pay more for interest. And it's hard to know exactly how that's happening, but it's not equally distributed in terms of how interest rates are calculated. But the idea is that it's not discriminatory because it's a program that screens out the categories that should be prohibited. In credit scoring,

the case is maybe more clear than other kinds of algorithms because credit scoring is very highly regulated. But I'm certainly not arguing that credit scores are not discriminatory or they don't have problems. But compared to other algorithms, they're actually kind of the gold standard. Coming up, we take a hard look at that gold standard, if it's still doing its job, and you're listening to ThruLine, one of the best podcasts out there. Thank you for everything you do. The three-digit credit score is hugely powerful. It can change your life for better or for much worse. So we asked media historian Josh Lauer, is it fair? When a lender looks at a credit score, it doesn't mean that they're automatically going to follow what that credit score risk,

what it's telling them, right? So what I mean is they could then decide based on other factors, whether to give you the loan or not, or what interest rate to give you. And it seems to me that in that space, there's lots of discrimination that can happen. But this is a tool rather than a mandate, right? Yes. And I think that's especially the case with mortgages. And that's probably why there are problems with mortgages, because the credit score does not dictate the entire decision. So what I'm really trying to get at here is, you know, this idea that in theory, credit scores should reduce some of those extraneous factors that often lead to discrimination and decision making. I'm trying to understand whether it's kind of hard for that to always be the case, given the fact that it's ultimately human beings making the decision to lend someone money or at what interest rate to lend them that money. So the other way that credit scoring can really hurt people is because it's not just a decision whether to lend or not. The credit score allows lenders to not only decide who they're going to

lend to, but to price that loan or that service. So this is where you begin to have sort of the tiering of interest rates. And it also opens the door to things like subprime interest rates. So in the past, it was sort of, well, we'll lend you the money or we won't, or we'll let you take these goods up to a certain credit limit or we won't. But when it comes to the use of the credit score, you can say, well, yeah, we'll let you have a credit line of $1,000, but you're going to pay 18% interest on that. Whereas somebody else could apply for it and they'll say, oh, we'll give you $1,000 credit limit and you can pay 13% interest. So credit scoring really opens the door to this stratification based on interest rates. And so

the poor or the persecuted elements of society still pay more. It's just that they pay more in a way that appears more democratized or the access may still be there, but there's still an unequal interest rate, which, you know, obviously impacts some people more than others. I think this gets at something that, you know, I think throughout this conversation, we've sort of been touching on in some way, which is that whether it was the early credit reporting, then the early kind of the rudimentary credit scoring, and then the actual FICO score that we continue to have to this day,

there is this sort of like through line right throughout this story of essentially trying to find a way, a scientific mathematical way to solve for human bias, the judgment that we see as being a part of human bias. And, you know, we recently did an episode about artificial intelligence, and this is something that I think was driving that field, which was coming about, you know, around this time as well in the 50s, 60s, 70s was starting to develop. And that broader idea that we can solve for human discrimination and misjudgment, it creates, I would say, an illusion that with a score, with the algorithm, you have now gotten rid of these long legacies of discrimination. And of course, as you're describing and talking about, that clearly is not the case. To this day, Black consumers are affected disproportionately by credit scoring and mortgage lending and all of that. That is underpinning the story of wanting to basically solve for X, X being our humanness,

our capacity for discrimination and bias. So during the 1960s, when credit scoring was being introduced, credit managers whose job was to sit down with applicants and take their application and make credit decisions, many of them said credit scoring will never work. It'll never replace the sort of expertise and intuition that the credit manager has when they meet with a person. And of course, that intuition could work both ways. It could work in someone's favor or could work as a bias or prejudice that ends up having someone denied access to credit. But one of the things that having an interview or a personal

interaction with someone would do was allow the applicant to contextualize information to say, yeah, you know, I lost my job last month, but I got another interview and things are turning around, or I had a health emergency, someone in my family was sick, or I got divorced and it really set me back. But, you know, things are changing now. That information is completely eliminated and maybe for a good reason from credit decisions. But it's a human interaction where you can actually try to contextualize this kind of information. So the credit score, when it's developed and begins to diffuse through society and becomes more popular, you have Congress questioning whether it's fair or not. And you have banking executives

brought before Congress in the 1970s to defend credit scoring. And the bankers say, listen, we know that credit scores don't always work for everyone. We know that some people are less likely to be scored high than others. But you can't hold us responsible for the problems of American society. We're just using information as we find it. We're not dictating whether or not society is fair or discriminatory. We're just basically using this tool, this technology, to try to process credit applications as quickly and as fairly as we can. And in some ways, they're actually making an important point.

The bankers weren't responsible for systemic racism or gender discrimination in the United States. I mean, they may have been partly responsible for policies, but they weren't wholly responsible. It wasn't as if the banking community had installed all of these problems in America. They were part of American society. They were built into the fabric of the society. And that's one of the things that scholars have learned as we've studied algorithms, is that the algorithms reflect reality in some ways. And that reality is often quite ugly. It's misogynist and racist. And these kinds of biases are reproduced in the functions that algorithms are designed to process. So in some ways, the question is whether technology and algorithm can actually make the world more fair, make it more just.

At the moment, credit scores are designed to really only measure things that have to, or to use variables that have to do with financial performance. So things like your payment history, the kinds of loans you have, your debt ratios and debt loads. These are directly related to financial performance. There might be characteristics in here that are proxies for gender or for race, but generally it's pretty much about financial behavior. But if you compare that to other kinds of algorithms that are being used, where any kind of data can be used to make decisions, to make inferences about people, there's no regulation.

There are no limits on the kinds of information that can be used. By contrast, it makes credit scoring look like it's the gold standard of a protected, regulated algorithm. You make a convincing argument that in a lot of ways, because of its regulation, it is the gold standard. But I think the underlying problem that perhaps lenders are not responsible for, but is real, is that it's hard to build credit when you're already poor. When you start from a place where you don't have much money, there's obviously things are changing and there's more outlets and more businesses emerging to try to give you that. Like there's this, I saw this commercial for an

app recently where paying your rent can build towards building credit. But I think when you start from that place, it then becomes a tool which automatically creates a disadvantage for someone who's starting with less money. So you're absolutely right. And the formal credit system is designed for people who are part of the mainstream economy. It's designed for people that have bank accounts, that have credit cards, that have mortgages, that participate in this formal institutional system. If you do not, then you have what's called a thin file.

There's not very much information about you. And that can lead to low scores or no scores. The paradox, though, is that if you want to build a credit scoring system that's more fair and inclusive, it means that each person has to surrender more information about themselves. It means that you have to provide information about rental payments. It means that you have to provide information about cash transactions and bank account information. Very specific information has to be provided if you want to be registered in that system. So really, it comes down to a question of surveillance. I mean, if you want to expand participation, you actually have to expand surveillance of individuals to participate in that system. And I'm not suggesting that that isn't a good idea or it has to be a bad idea,

but it does matter who's running the system. And currently the system is run by private corporations whose interests are not really with consumers. Their main concern is protecting businesses from consumers who might get into trouble and not pay debts. So you can imagine an alternative credit system where the organization is responsible for collecting the data and managing the data are not private corporations

who don't care about losing your data or making mistakes with your data. That's fascinating. Yeah. That trade-off, is that a lose-lose trade-off. Yeah. That trade-off is, is, is that a lose, lose trade-off, you know, like, sort of building on, on, on what Ramti was asking. There's a statistic that, that I came across that I think is kind of mind boggling, which was that nearly 40% of the country is considered either credit invisible or has a low credit score. And if you're in that

position, it's, it's like this kind of a system where data and with more surveillance, even more data becomes your destiny. How do you get out of that hole? The only way out is to somehow find a way to get into the mainstream financial system by opening bank accounts and paying for some services that are reported to credit bureaus. That is the main way of doing it now. So one thing that I would be concerned about is that there are lots of firms that are experimenting with alternative data and fintech companies that are trying to develop models that can use alternative data that is not part of a credit bureau.

So for example, information that might be collected online, like social media data, other kinds of personal information that might be collected through the internet. And while in some ways this looks like, hey, a great idea, we cut the credit bureaus out and we just use big data to solve the problem of credit invisibles. But the problem is that that data is absolutely going to have proxies that are going to unwittingly discriminate against women, people of color, other categories of consumers that are not part of the financial mainstream. Well, here's an obvious question. So if the point of it is to predict whether someone's going to pay back the loan and follow their payments, et cetera. Is there any studies to indicate whether it actually works? So, you know, that kind of information would be a trade secret, really,

to know exactly how well a credit model is working. But I will say that most credit scoring models are designed to have less than a 1% default rate. So they're expecting to lose very little money. So in real numbers, it might be that there are a lot of people that don't fit into this pattern. But overall, it's actually quite efficient for the way it works. So the other thing I would say just about using statistical information, population information to make credit decisions is that it's depersonalized. So each of us, we imagine ourselves as individuals who are unique and we have agency and we make our own decisions.

But in reality, we're being treated as if we're just a member of a population that behaves in certain ways that make us look like other people that behave in that way. And it essentially treats individuals as an artifact of a statistical population rather than as individuals. And I think that works against this idea of risk assessment as being democratizing, because, you know, we're not really being treated as individuals, we're being treated as representations of a statistical group based on data, which we hope is not tainted or bad or discriminatory, but could be. And we have no control over that.

You know, this is the way that algorithms make decisions based on past behavior of groups that are believed to be like us. That's how algorithms reproduce discrimination and reproduce inequality. And that's the argument for even looking at a credit score. Even though credit scores might be fairly well regulated and fair in the grand scheme of things, they're using data that reflects past behavior and reproducing predictions about the future based on this past data. And back to the question of systemic discrimination and systemic problems that the bankers were objecting to being responsible for. We know that on average,

whites have higher credit scores than blacks and Hispanics. Why is that? It's not because a personality trait or a racial trait. it's because of a legacy of historical systemic discrimination. And it's reflected in the credit scores. It doesn't mean that it's not possible for a Black person or a Hispanic person to have a high credit score. It just means that on average, people that are put into that group have lower credit scores. And it follows with Asian Americans too, because they do better economically and then they have higher credit scores. Exactly. So it's like what you start with really dictates where your score is. And that's where you see generational wealth reflected in

credit scores indirectly. But the credit scores themselves do not measure generational wealth. That's not what they're trying to do. But of course, it's reflected in the big picture. So then the question is, looking at the future and thinking about how do you get to a more democratized actual, like a more democratized system, economic system, when it comes to the credit score, is it about reforming the credit score? Or is it about reforming our society more broadly, our economic system more broadly, or is it both? It's definitely both things. And I would say that the credit bureaus that collect the data that's used in credit scores are far from perfect. They have errors, as I mentioned. They're not designed

to serve the consumer. Historically, they've always been organized to protect lenders. So consumers really are not their priority. You can imagine a different kind of credit reporting system that is not in the hands of a for-profit company and that the interests of the consumer are put forefront rather than having the lenders be the only real concern. You can imagine credit scoring systems that are developed by nonprofits or even the federal government that are regulated in a much more transparent way where trade secrets are not a concern. And you can actually be more transparent in terms of what's going on, report data about how the scores operate, whether there are problems with discrimination or inequity. These are all questions that are difficult to even think about in the context of the United States because we have a 150-year history of private credit reporting firms

and private efforts to sell credit scores. It's an industry. It's a big business, a billion-dollar business. One of the ways that you could imagine a more just and equitable world would be to have these kinds of decisions and data collection processes in the hands of an organization that's more trustworthy. It's hard to imagine a modern world where you are continuously interacting with strangers,

businesses that you don't know, and are using credit without some sort of system that's behind the scenes that actually vets people. It confirms their identities, confirms their past payment behaviors, confirms that they had some kind of income to make the payments. The alternative is you return to a world where every transaction is personal and it's time consuming. And it also means you have to share information about yourself. I mean, one of the nice things about credit reporting and credit scoring is that it allows us to be anonymous around other people.

When you go to a store or go to a restaurant, you don't have to explain your whole life story to somebody when you're making the payment. You just make the payment and leave. But that, of course, also means that you have to share a tremendous amount of information with someone else. And that someone else is the credit reporting infrastructure. And, you know, like I said, the credit reporting system is imperfect because it's really not designed for us. It's designed for lenders. Josh Lauer is a professor of communications at the University of New Hampshire and author of the book Creditworthy,

a history of consumer surveillance Financial Identity in America. That's it for this week's show. I'm Randa Abdel-Fattah. I'm Ramteen Adablui. And you've been listening to ThruLine from NPR. This episode was produced by me. And me and... Lawrence Wu.

Julie Kane. Anya Steinberg. Yolanda Sanguin. Casey Minor. Christina Kim. Devin Katayama. Yordanos Tisfazion. Fact-checking for this episode was done by Kevin Vogel.

Also, thanks to Micah Ratner, Johannes Dergi, and Anya Grunman. This episode was mixed by Robert Rodriguez and Maggie Luthar. Music for this episode was composed by Ramtin and his band, Drop Electric, which includes... Anya Mizani. Naveed Marvi. Cho Fujiwara. And one more thing. We're working on an episode about the ways motherhood is and isn't compensated.

And we want to hear from you. What does motherhood cost? Call us and share your thoughts and experiences at 872-588-8805. And as always, if you have an idea or like something you heard on the show, please write us at ThruLine at NPR.org. Thanks for listening. This message comes from NPR sponsor Grammarly. What if everyone at work were an expert communicator? Inbox numbers would drop, customer satisfaction scores would rise,

and everyone would be more productive. That's what happens when you give Grammarly to your entire team. Grammarly is a secure AI writing partner that understands your business and can transform it through better communication. Join 70,000 teams who trust Grammarly with their words and their data. Learn more at Grammarly.com. Grammarly. Easier said, done. Support for NPR and the following message come from the Kauffman Foundation,

providing access to opportunities that help people achieve financial stability, upward mobility, and economic prosperity, regardless of race, gender, or geography. Kauffman.org.