Fannie Mae Will Approve a 50% Debt-to-Income Mortgage. Should You Take It?

Run a home affordability calculator and you'll usually get three numbers: a conservative one, a "with strong credit" one, and a much bigger one that represents the absolute ceiling a lender will approve. It's tempting to anchor on the biggest number — it's the one that gets you the most house. But that ceiling has a specific, fairly recent history, and even the two government-sponsored agencies that set it don't fully agree it's safe.

Where 36%, 45%, and 50% actually come from

These aren't abstract rules of thumb — they're the literal thresholds in Fannie Mae's and Freddie Mac's underwriting guides today, and the 50% ceiling specifically traces to a 2017 policy change. Before 2008, both agencies bought mortgages with debt-to-income (DTI) ratios up to 65%. The 2008 crash changed that fast: by early 2010, both had cut their limit to 50%, and Fannie Mae's automated underwriting system added an extra internal checkpoint (an "overlay") that, in practice, made anything over 45% hard to get approved.

That overlay held for seven years. Then in April 2017, the Federal Housing Finance Agency — the regulator that oversees both companies — directed them to stop letting DTI alone disqualify a loan up to 50%, specifically to widen access to mortgage credit. Fannie Mae rolled that out in its underwriting system that July. That single directive is why "50% DTI" is a real, approvable number today, and not a hypothetical.

A rule that briefly said 43% instead

You may also see 43% cited as a debt-to-income cutoff, and that number has a different origin: the Consumer Financial Protection Bureau's 2014 Qualified Mortgage (QM) rule, which required lenders to reasonably verify a borrower's ability to repay, and set 43% as one path to that safe harbor. Loans eligible for purchase by Fannie Mae or Freddie Mac were temporarily exempted from that specific cap (nicknamed the "GSE Patch"). In 2020 the CFPB replaced the 43% bright line entirely, for loans going through this path, with a price-based test instead — a loan qualifies if its rate doesn't run too far above the average prime rate, regardless of the borrower's exact DTI — effective for applications from mid-2021 on. So a hard 43% cap did exist, briefly, as the GSE Patch's scheduled expiration date, but it was replaced before it ever took effect for agency loans. DTI is still weighed in underwriting; it just isn't the single bright line it briefly threatened to become.

What happened once the 50% ceiling opened up

The FHFA's own internal watchdog studied exactly this in a 2019 white paper. After the 2017 change, both agencies' share of max-DTI mortgages climbed fast — Fannie Mae's peaked around 26% of its acquisitions by mid-2018, Freddie Mac's around 18%. More of those loans also stacked other risk factors: a higher share had a loan-to-value ratio above 95% or a credit score under 680 than before the directive — what the industry calls "risk layering."

The market noticed before the regulators finished studying it. In early 2018, five of the six major private mortgage insurers announced they'd stop insuring loans with a DTI above 45% and a credit score under 700, starting that March. Practically, that means the 50% figure a calculator shows you as "approvable" may not be insurable at all if your credit isn't strong — a real gap between what the GSE guideline allows and what you can actually close on.

Fannie Mae and Freddie Mac don't even agree it worked out

This is the part a calculator can't show you: as of the OIG's reporting in January 2019, Fannie Mae said its max-DTI loans were performing at or slightly better than expected on early delinquency. Freddie Mac said the opposite about its own 2017-2018 max-DTI loans — that on average, they were performing worse than loans with lower DTI. Same policy change, same two-year window, two different risk managers looking at their own portfolios and reaching different conclusions. That's a genuinely unsettled question, not a solved one — which is a good reason to treat "a lender will approve it" and "it's a good idea" as two separate questions.

So which number should you actually plan around?

• The 36% figure is the one worth budgeting to. It's the baseline both agencies document before a lender needs any special justification — the number that assumes nothing has to go right for you to stay comfortable.
• The 45% and 50% figures describe what's approvable, not what's wise. Fannie Mae's own compensating-factor list — higher credit score, larger cash reserves, a bigger down payment — is really a checklist for whether you can absorb an income shock at that payment level, not just whether a computer will say yes.
• Predictable income growth is a legitimate reason to stretch; income uncertainty is a reason not to. A household with one income about to step up on a known schedule is in a different position than one relying on variable commission or a single earner in an at-risk industry, even at an identical DTI ratio today.

None of this is a verdict on any individual loan — DTI is one risk factor among several, and FHFA's own reporting notes that both agencies consider it a less reliable predictor than credit score or loan-to-value on its own. But "a lender will approve this" and "the two agencies who'd own the risk agree this is safe" turned out, in the years right after this ceiling opened, to not be the same claim.