When the Bank Manager Knew Your Father's Name — and That Was Your Credit Score
Imagine walking into your local bank to ask for a mortgage and having the loan officer say, "Oh, you're Bill Henderson's son — good family, good people. Let's talk." No credit report pulled. No debt-to-income ratio calculated. No algorithm running in the background. Just a man behind a desk making a judgment call based on what he knew about you and yours.
That was American lending for most of the country's history. And depending on who you were, it was either the most reassuring system imaginable or an invisible wall you could never climb over.
The Town Knew If You Were Good for It
Through the late 19th and well into the 20th century, American financial life was intensely local. Community banks and savings and loan associations were the primary lenders for ordinary people, and they operated on personal knowledge rather than standardized data.
A banker in a small Ohio town likely knew his borrowers personally — their employment history, their family reputation, whether their father had paid his debts, whether they were known as reliable people in the community. Merchants extended credit on similar terms. The local hardware store might let a farmer run a tab through a rough season because they'd been doing business together for fifteen years and trust had been built the old-fashioned way: slowly, through repeated contact.
This was the "character" component of what lenders called the three C's — character, capacity, and collateral. Of those three, character often carried the most weight. If the community vouched for you, the numbers were almost secondary.
For people who fit the community's definition of trustworthy, this system felt natural and even generous. Loans were negotiated with flexibility. Terms could be adjusted based on circumstances the banker actually understood. There was a human conversation at the center of every financial decision.
The System Had a Darker Side Nobody Likes to Mention
Here's where the nostalgia has to stop.
The character-based lending system wasn't just informal — it was exclusionary by design. If the banker didn't know your family, or didn't like your family, or simply didn't see people who looked like you as part of the community he served, you were out. No appeal. No alternative metric. No recourse.
Black Americans were systematically denied access to credit in ways that compounded across generations. The practice of redlining — where federally backed maps literally colored Black neighborhoods red to signal them as bad lending risks — meant that entire communities were locked out of homeownership and business loans regardless of individual financial behavior. The Federal Housing Administration, from its creation in 1934, actively encouraged discriminatory lending practices for decades.
Women fared barely better. Until the Equal Credit Opportunity Act passed in 1974, lenders could legally refuse to issue credit to a woman without her husband's signature — even if she was the primary earner. Divorced women often found themselves with no credit history at all, because every account they'd shared with a spouse had been recorded in his name only.
The personal, relationship-based system that felt so warm and community-minded for white men with the right backgrounds was, for everyone else, a locked door with no key.
Enter the Algorithm
The modern credit scoring system emerged gradually through the mid-20th century, with the FICO score — developed by Fair Isaac Corporation — arriving in 1956 and becoming the dominant standard by the 1980s and 90s. The pitch was simple: replace subjective human judgment with objective data. A three-digit number that captured your financial behavior without caring who your father was or what neighborhood you grew up in.
In theory, this was a genuine civil rights advance. The algorithm didn't know your race, your gender, or your last name. It looked at payment history, credit utilization, length of credit history, and a handful of other factors. Anyone who played by those rules could build a score. The playing field, at least in principle, was level.
And to a real extent, it worked. Credit became accessible to millions of Americans who would have been quietly turned away under the old system. Women could build independent credit histories. People who moved to new cities weren't starting from scratch in terms of their financial reputation — their score traveled with them.
What the Algorithm Can't See
But the algorithmic system brought its own blind spots and frustrations.
For starters, the data it relies on isn't neutral. If you grew up in a household that operated primarily in cash — as many lower-income and immigrant families do — you may have an excellent real-world track record of managing money responsibly while having a thin or nonexistent credit file. The algorithm simply can't see you. You're invisible to the system until you participate in it on its terms.
The system also has no mechanism for context. Lost your job during the 2008 financial crisis and missed three months of payments? The algorithm records the missed payments. It doesn't record the recession. A medical emergency that wiped out your savings looks identical to reckless spending in the data. The nuance that a human banker might have understood — and weighted accordingly — doesn't exist in a credit report.
And for all its supposed objectivity, research has consistently shown that credit scores still correlate with race, largely because they reflect the accumulated effects of historical discrimination. If your parents couldn't build wealth through homeownership because of redlining, the generational financial disadvantage that created doesn't disappear just because the scoring model is colorblind.
A Fair System in an Unfair History
The shift from handshakes to algorithms is one of those evolutions where both the old and the new are genuinely flawed in different ways. The personal system was human and flexible and also deeply prejudiced. The algorithmic system is consistent and scalable and also blind to context and historically burdened.
What's striking is how little public conversation there's been about what we actually want from a lending system — what fairness really means when you're trying to predict financial behavior across millions of people with vastly different starting points.
The banker who knew your father's name could be your greatest advocate or your biggest obstacle, depending entirely on who you were. The algorithm that doesn't know your name at all might be more consistent — but consistent application of an imperfect system is still an imperfect system.
We traded one set of problems for another. The question is whether we're paying enough attention to the ones we've got.
