Credit Scores Decoded: FICO's 1958 Formula, VantageScore and 60 Versions

SPEAKER_1 (0:00)

You know, it is a truly mind-bending contradiction when you actually stop to think about it. I mean, if you ask the average person what their credit score is, they would probably picture just a single definitive three-digit number.

SPEAKER_2 (0:12)

Right, like a halo hovering over their financial life, or, you know, a storm cloud.

SPEAKER_1 (0:16)

Exactly. But the reality, you do not have just one credit score, you actually have over 60 of them floating around in different proprietary databases, and they're constantly shifting in real time.

SPEAKER_2 (0:28)

Which is wild. And while you have dozens of these invisible mathematical grades dictating what you can buy and where you can live, approximately 45 million American adults are completely credit invisible.

SPEAKER_1 (0:42)

Yeah, they are fundamentally unscorable.

SPEAKER_2 (0:44)

They have absolute zero. It's a massive systemic paradox.

We are simultaneously overscored and completely underscored, depending entirely on how you interact with traditional banking institutions.

SPEAKER_1 (0:58)

So how did we get here? Today we are going to do a deep dive and unpack the architecture of the United States credit scoring system.

SPEAKER_2 (1:05)

We'll be using the source materials to explore the secret math behind that all-important number.

SPEAKER_1 (1:10)

And investigating how a tool originally built just for back office banking evolved into the ultimate gatekeeper of modern life. Okay, let's unpack this because to understand how a rigid mathematical algorithm came to dominate our lives, we have to look at what it replaced.

SPEAKER_2 (1:25)

Right, and before the 1950s, the concept of credit was entirely based on human judgment. It was all about localized relationships.

SPEAKER_1 (1:32)

Like you would just walk into a local bank or a neighborhood credit bureau.

SPEAKER_2 (1:36)

Yeah, exactly. And a loan officer would make a highly subjective call about your personal character. I mean, sure, they were looking at your job, but they were also looking at your suit.

SPEAKER_1 (1:45)

Wow, really?

SPEAKER_2 (1:46)

Oh yeah. They'd ask about your family background and basically decide based on a gut feeling if you were good for it.

SPEAKER_1 (1:52)

Which sounds completely ripe for bias. It feels very much like an exclusive handshake deal, where if the banker simply didn't like your vibe or your neighborhood, you were walking out empty-handed.

SPEAKER_2 (2:03)

That subjectivity was the defining feature of the era, and it created an environment where systemic discrimination was just standard operating procedure.

SPEAKER_1 (2:12)

Which is terrible.

SPEAKER_2 (2:13)

It is, but that environment is actually what paved the way for the algorithmic approach. In 1958, an engineer named Bill Fair and a mathematician named Earl Isaac created the first credit-scoring algorithms.

SPEAKER_1 (2:27)

Okay, so that's the FICO guys, right? Fair Isaac Corporation? You got it.

SPEAKER_2 (2:31)

They explicitly wanted to redefine creditworthiness away from personal character assessments and turn it into abstract statistical risk.

SPEAKER_1 (2:39)

But looking at the notes here, it seems like this mathematical system just hovered in the background for a couple of decades. The sources point to the 1970s as the moment it really started taking over.

SPEAKER_2 (2:49)

Right.

SPEAKER_1 (2:50)

And surprisingly, it was driven by civil rights legislation. What happened there?

SPEAKER_2 (2:54)

Well, the Equal Credit Opportunity Act of 1974 and its expansion in 1976 served as a massive catalyst. That legislation strictly banned lenders from denying credit based on race, gender, marital status, religion, and age.

SPEAKER_1 (3:10)

So the banks were suddenly staring down enormous legal liabilities.

SPEAKER_2 (3:13)

Exactly. If a human loan officer made a subjective, biased decision, the entire institution could be sued.

So the adoption of computerized credit scoring rapidly accelerated because banks needed a mathematical shield.

SPEAKER_1 (3:27)

Ah, so they could just point to the computer.

SPEAKER_2 (3:29)

Right. They needed to be able to sit in a courtroom and say, hey, we aren't biased. We are just following the objective output of this algorithm.

SPEAKER_1 (3:37)

But, I mean, an algorithm is only as good as the data you feed it. Did the formula actually succeed in removing human bias, or did it just launder that bias through a computer?

SPEAKER_2 (3:48)

That's the big question.

SPEAKER_1 (3:50)

Because if the historical data you are building the model on is inherently skewed by decades of redlining and discrimination, the algorithm is just going to replicate that exact same bias under the guise of math.

SPEAKER_2 (4:03)

If we connect this to the bigger picture, you have hit on the crux of the historical debate surrounding these models. The intention was standardizing risk.

And in doing so, it certainly removed the blatant face-to-face discrimination of the local loan officer. But inadvertently, it shifted the systemic focus.

SPEAKER_1 (4:20)

How so?