What the Debt-to-Income Ratio Sees When a High Earner Applies for a Mortgage

A mortgage loan officer told MarketWatch that they had turned down wealthy couples because of overspending, and warned that “people in the U.S. need to wake up.” It is a striking line, and it is also slightly misleading about how a denial actually happens. An underwriter does not read your restaurant tabs and render a verdict on your character. The machinery is narrower, more literal, and in some ways more punishing than a judgment call: it reads a single ratio, and that ratio only sees the part of your lifestyle you financed.

That distinction is the whole story of why a household earning $300,000 can be declined while a household earning $95,000 closes without friction. The debt-to-income ratio is not a measure of how much you spend. It is a measure of how much of your gross monthly income is already promised to someone else in writing. Below is the mechanism, step by step, with the places it goes quiet.

The denial that looked like a lecture

Start with what the loan officer in that account was almost certainly looking at. Under the Ability-to-Repay rule the Consumer Financial Protection Bureau issued to implement the Dodd-Frank Act — codified at 12 CFR 1026.43 and effective January 2014 — a lender must make a reasonable, good-faith determination that a borrower can repay the loan. The rule names eight underwriting factors the lender must consider, and two of them are current debt obligations and the resulting debt-to-income ratio.

So the “overspending” conversation is real, but it reaches the file through the debt column, not the discretionary one. A couple with a $1,400 SUV lease, a $900 second-vehicle note, a $650 minimum on a credit line used to furnish the last house, and $500 in student loan payments is carrying $3,450 in monthly obligations before they have paid for a single thing they enjoy. At a 45% ceiling, that $3,450 has consumed the entire debt capacity of a $7,667-per-month gross income. High earners rarely fail because the number is small. They fail because they built a payment structure to match it.

1️⃣ Which income the formula will accept

The numerator is the easy half. The denominator is where high earners lose first, because the income that makes them high earners is often the income underwriting discounts.

Agency rules want stability and history, not magnitude. Fannie Mae’s Selling Guide treats variable income — bonus, overtime, commission — as usable only with a documented history, generally two years, and requires the lender to analyze whether it is trending down. Self-employment income is averaged and reduced by business expenses; a profitable S-corp owner who pays themselves modestly and leaves cash in the company has, on paper, a modest income. Restricted stock, carried interest, and partnership distributions all face documentation tests that a salary does not.

This is the quiet asymmetry at the top of the income distribution. A W-2 employee’s $200,000 base salary is fully countable on a pay stub. A consultant’s $280,000 of lumpy 1099 revenue may underwrite to less, and a founder’s equity is usually worth nothing at all to the ratio. The denominator shrinks exactly where income becomes interesting.

2️⃣ The debts that get picked up, and the spending that never does

The numerator counts recurring obligations that appear on a credit report or a court order. It does not attempt to model your cost of living. The result is a ratio that can look pristine while a household is running a cash-flow deficit.

Read the right-hand column again. A family spending $4,000 a month on tuition, premiums, and travel is invisible to the ratio. A family leasing two cars at $1,000 each is not. Two households with identical bank balances and identical net savings can receive opposite answers, and the difference is the financing structure, not the restraint.

There is a practical consequence here that cuts against instinct. Paying cash for a car — or keeping an older one — improves mortgage capacity more than cutting discretionary spending by the same dollar amount, because only one of those changes is legible to the underwriter. At a 6.5% 30-year rate, roughly $650 of monthly payment supports somewhere near $100,000 of loan principal before taxes and insurance. Retiring a $650 car note is, mechanically, a six-figure increase in what the formula will let you borrow.

3️⃣ Where the 43% rule went

The number most borrowers have heard is 43%. It is worth knowing that this is no longer the regulatory line, and the change explains why loan officers now sound more discretionary than they used to.

The original General Qualified Mortgage definition did contain a hard 43% DTI limit. In December 2020 the CFPB replaced it with a price-based test, keyed to the loan’s annual percentage rate relative to the average prime offer rate. The 43% ceiling is gone from the General QM; the obligation to consider DTI under the Ability-to-Repay rule is not. Underwriting judgment moved from a bright line to a priced one, and lenders filled the gap with their own thresholds.

The VA row deserves attention even from borrowers who will never use the program, because it is the one official framework that does what the MarketWatch loan officer was reaching for. Under 38 CFR 36.4340(c), VA underwriting applies residual income tables — dollars left after housing, debts, taxes, and estimated maintenance and utilities, scaled by family size and region. It asks whether money remains, not merely whether a ratio clears. No agency program applies that test to conventional lending.

Jumbo loans and the overlay problem

High earners disproportionately borrow above the conforming loan limits, which means they are disproportionately exposed to loans that no GSE will buy. Those loans are priced and underwritten against the appetite of private investors and bank balance sheets, and the terms are set by the lender rather than a published guide.

In practice that typically means a lower DTI tolerance than the 50% agency ceiling, reserve requirements measured in months of housing payments, and tighter scrutiny of income volatility — exactly the three dimensions on which a lifestyle-inflated high earner is weakest. Because overlays are proprietary, there is no single figure to cite, and any article quoting one as a market-wide standard is guessing. The honest version is that the agency ceilings in the table above are the loosest numbers in the market, not the typical ones, and jumbo borrowers should assume the bar sits below them.

⚠️ Where this explanation stops holding

Several traps sit outside the clean mechanics above.

• A lease ending next month still counts. Agency guidance treats auto leases as ongoing obligations regardless of remaining term, on the reasoning that the vehicle will be replaced. “It’s almost paid off” is not an argument the file accepts.
• Deferred student loans are not free. Where no payment is reported, agency rules impute one from the balance. Income-driven repayment plans showing $0 can still be underwritten at a calculated figure depending on program and documentation.
• Business debt can follow you personally. A loan reporting on your credit report counts even if the business services it, unless you can document twelve months of payment from business funds and that the obligation is excluded from your cash-flow analysis.
• Compensating factors are discretion, not entitlement. Large reserves, a long credit history, or a low loan-to-value ratio can push a file through at a higher ratio — but the lender decides, and a borrower cannot plan on it.
• Credit utilisation moves the numerator. Minimum payments are read as reported. Carrying heavy revolving balances raises those minimums, which raises your DTI without your debt load having changed in any economically meaningful way.

One more limit worth stating plainly: how often excessive DTI is the stated cause of denial is reported annually through Home Mortgage Disclosure Act filings, where it has long ranked among the most-cited reasons, but the current share is not a figure I will quote here without the release in hand. Treat the ranking as context and the mechanism as the fact.

When the ratio stops deciding

The debt-to-income ratio is decisive for a borrower who needs a conforming or government-backed loan at a standard rate. It becomes progressively less decisive as the borrower’s balance sheet does more of the work: a larger down payment lowers the housing payment inside the numerator, substantial documented reserves become a compensating factor, and in asset-depletion or portfolio programs the lender underwrites holdings instead of cash flow. At the far end, a buyer paying cash has no ratio at all.

Which means the loan officer’s warning lands differently than it reads. The system did not reject those couples for enjoying their money. It rejected them for having converted their income into contractual monthly obligations, and it would have approved the same spending funded from cash. What would change the answer for a household in that position is not a budgeting epiphany but a change in the shape of its liabilities — fewer financed commitments, more documentable income history, or enough assets to be underwritten on something other than a ratio. If the CFPB or the GSEs ever import something like VA’s residual-income test into mainstream conventional underwriting, the gap between spending and financed spending would narrow, and the loan officer’s instinct would finally have a form to fill out. Until then, the formula sees only what you signed for.