A few years ago a friend of mine, I’ll call him Tom, cosigned a mortgage for a friend of his named John.
They’d known each other for years. John had just come out of a divorce, and it had taken most of what he had. Assets divided, cash gone, household income cut roughly in half overnight. He needed a place of his own, and the financial hit had done enough damage to his credit that he couldn’t get approved by himself.
So Tom signed.
You know where this is going…John fell behind, and every missed payment landed on Tom’s report too, because that’s what cosigning is, and Tom hadn’t done anything except be a good friend.
I think about those two whenever I look at what’s happening to credit in this country right now, because between them they’re the whole story. Neither man was reckless. One got divorced. The other helped. And both ended up on the wrong side of a line that’s getting harder to cross back over than it used to be.
The Average Is Hiding Something
The average American credit score is 714.
That number sounds unremarkable, and it may be the most misleading statistic in personal finance right now, because the average is hiding something.
In March 2026, FICO published its Spring Credit Insights report. The national average score had fallen two points, to 714, continuing a slow decline that began in 2023 and ending an eleven-year run of steady improvement.
At the same time, a record 48.1% of Americans now have scores of 750 or higher. That’s up from 43.3% in 2019. Nearly half the country sits in the top tier, the highest share ever recorded.
Both things are true at once. FICO found that the high-score and lower-score groups both expanded while the middle continued to shrink.
FICO’s own analysts described the result as a credit market consistent with a K-shaped economy. That’s their language, not mine. One arm going up, one going down, and fewer people in between each year.
What’s Pushing People Down
FICO named two drivers, and neither is about a lack of responsibility:
Student loan delinquency reporting resumed- After the pandemic pause, payments restarted and so did the reporting of missed ones. FICO identified this as the primary factor behind the decline.
Mortgage delinquencies kept climbing- Coming back toward pre-pandemic levels. Meanwhile auto, credit card, and personal loan delinquencies leveled off or improved, which makes housing the outlier rather than the rule.
There’s a third pressure that never shows up as a delinquency at all. When balances rise, utilization rises with them, and higher utilization can pull your score down even when you’re making every payment on time. People in this position often can’t work out why their score is falling when they haven’t missed anything. That’s usually why.
What’s Pushing People Up
The other arm of the K is real too, and some of it may have already happened to you without your knowledge.
Medical debt came off a lot of reports- The three bureaus voluntarily stopped reporting paid medical collections, medical debts under $500 don’t appear at all, and unpaid medical debt gets a full year before it shows up. If you had old medical collections dragging you down, they may simply be gone.
One clarification, because there was a lot of coverage about this in early 2025. A federal rule that would have banned medical debt from credit reports outright was struck down in court in July 2025. What survives is voluntary bureau policy, not federal law. It still helps you. It’s just a policy choice rather than a protection.
FICO Score 10T reads trends instead of snapshots- This particular model uses trended credit bureau data covering roughly the previous two years of balances and payments, rather than judging you on where you stand today. If you’ve been steadily paying down balances, it can see the direction of travel. Note that this is one specific model, not a change to every newer FICO score. And it cuts both ways, which is only fair to say: two years of slipping shows up as a trend too.
There’s another change that could help.For decades, mortgages sold to Fannie Mae and Freddie Mac were underwritten using versions of Classic FICO. In April, an alternative model called VantageScore 4.0 was cleared for a limited rollout, capped at fifty lenders.
On September 3rd, the Federal Housing Finance Agency directed Fannie Mae and Freddie Mac to accept VantageScore 4.0 from all mortgage origination lenders, effective immediately.
Why this could matter to you: VantageScore 4.0 can incorporate reported rent payments and certain utility and telecom data. Classic FICO generally can’t see any of that. If you’ve paid rent reliably for fifteen years but never carried much traditional credit, older models had almost nothing to work with. VantageScore says its model scores tens of millions of adults that legacy models can’t score at all.
Two caveats, and they matter.
Your rent only helps if it’s actually being reported- The model can use rent data. It can’t conjure it. Rent reporting is still far from universal, and if your landlord doesn’t report to the bureaus, paying on time for a decade is invisible either way. There are rent-reporting services that will do it for you, and that’s worth investigating if you have a thin file.
This doesn’t mean your lender will use it- or that you get to pick. Lenders now have the option. Whether any particular one exercises it is up to them, and no scoring model guarantees an approval.
Still, if you’ve been told your credit file was too thin to work with, the ground under that conversation shifted this week.
Which Arm of the K You’re On Isn’t Entirely Up to You
I want to be careful here, because there’s a version of this post that turns into a lecture, and the data doesn’t support the lecture.
A substantial share of the people whose scores fell did nothing wrong in any meaningful sense. Their student loan payments restarted. Their costs went up and the gap went onto a card. They’re paying everything on time and watching their utilization climb anyway.
And a substantial share of the people whose scores rose didn’t do anything clever. They had stable income and established credit through a period when both were valuable.
Which brings me back to John. He didn’t make a series of poor decisions. He got divorced, which happens to a great many people, and a divorce is an income shock and an asset shock arriving in the same month. His credit report recorded that faithfully. It just had no way to record why, and neither does any lender reading it afterward.
Tom is the other half of the lesson, and the harder one. He was solidly on the upper arm. Stable income, long history, nothing wrong anywhere. And he attached himself to somebody else’s file with a signature, which is the one way to move arms without changing a single thing about your own finances.
I’d hold onto that when anyone asks you to cosign. It isn’t a character judgment about the person asking. John was, and is, a decent person having a terrible year. It’s simply that a cosignature merges two credit histories into one outcome, and you only find out which direction it went afterward.
What’s happening more broadly is that the systems that lift people and the systems that sink them have both gotten stronger, and the middle where most people used to sit has thinned out. If you’re in that middle, you’re standing on a slope now rather than a plateau, and it’s worth knowing which way it tilts.
What a Better Score Actually Buys You
You’ll find plenty of articles listing the exact score you need for a mortgage, a car, an apartment. I’d treat all of those numbers carefully, because credit scores rarely work as one universal cutoff. Every lender and every product has its own underwriting rules, and they change.
What stays consistent is this: stronger credit improves your odds of approval and lowers the price of borrowing. And the pricing tiers go higher than people assume. Fannie Mae’s mortgage pricing grid runs credit-score bands all the way through 780 and above, so the old wisdom that everything maxes out at 740 isn’t quite right anymore.
The goal was never 850. The goal is landing in the strongest pricing tier available for the specific thing you’re planning to buy.
Understand the System Before You Try to Game the Number
That’s really the lesson across both the last post and this one.
The last one was about what moves your score, and your behavior genuinely matters enormously. This one is about everything else that moves it: policy changes, reporting changes, economic pressure, and which model a given lender happens to pull.
Your score is not a morality tale. It’s an output of a system, and the system has been changing underneath you for two years.
So go find out where you stand. Pull all three reports at AnnualCreditReport.com. Check a score through your bank or a free service. Look for surprises in both directions, medical collections that vanished as well as errors you don’t recognize.
If your score is lower than you expected, don’t pay anyone to fix it. Next time I’ll walk through exactly what to do yourself, for free, and why the industry that sells credit repair is currently having a very bad year in federal court.