The problem
A credit score reads everyone who does not fit as risky. People stop fitting for very different reasons. Treating those stories as the same risk is the mistake.
Snapshot failures
The invisible
Often the young, new arrivals, or people who pay in cash. The system has no story to read, so it says no by default.
~32 million U.S. adults cannot be scored or have files too thin to read. Federal Reserve.
The recovering
One setback — a missed payment, a medical bill, a layoff — freezes into the picture.
The New York Fed found a single missed student-loan payment cut strong borrowers’ scores by 171 points. New York Fed.
The misjudged thin file
A recent graduate or a wealthy buyer with little borrowing history looks thin, but the life behind it supports the loan.
Lenders lose the revenue — not because they don’t want the customer, but because they cannot defend the approval.
The fake
It looks identical to a recovering real person until you check the world outside the file.
Modeled annual synthetic-identity loss estimates run from $6–$35 billion. Boston Fed.
The common thread
Three real people and one fake can fail the same frozen test. The full research record lives on The Evidence; this page is the diagnosis.
Each needs a different answer. That requires looking outside the file.