Pushing a 46% DTI Loan Through Before the Lock Expires

No. Work through the price-based General QM test rather than the old 43% DTI rule.

• Rate-spread test: the APR of 8.536% exceeds the APOR of 5.10% by 3.436 percentage points. For a first-lien loan of this size, the General QM price threshold is 2.25 percentage points over APOR. The loan fails by more than a full point.
• Points-and-fees test: $11,000 on a $415,000 loan is 2.65%, under the 3% QM limit for loans of $100,000 or more. Passing this test alone does not make the loan a QM; the tests are cumulative.

Because it is not a QM, the creditor loses the QM safe harbor and the QM rebuttable presumption, but it may still originate the loan as a non-QM - provided it documents ability to repay under the general ATR rule: consider and verify the eight underwriting factors (income, assets, employment, monthly payment, related obligations, DTI, credit history, residual income) and reasonably conclude the borrower can repay.

One more check worth doing: the APR spread of 3.436 points is below the 6.5-point first-lien HOEPA trigger, and points and fees are below 5%, so this is not a high-cost mortgage. That removes HOEPA's restrictions but not the ATR obligation.

No - and the reasoning is what the exam is testing. The '43% DTI' figure is a leftover from the pre-2021 General QM definition. It is not a rule that lets you pick whichever year of income produces a compliant number, and a loan does not become a QM just because DTI lands under 43%.

Self-employed income must be analyzed, not selected:

• Two years of returns are ordinarily required, and a declining trend ($142,000 to $98,000) calls for an explanation or, if the decline is not offset, use of the lower or a reasonably adjusted figure.
• The 2024 K-1s, business returns, and a current year-to-date profit-and-loss statement are part of the verification - skipping them is not a shortcut, it is an unverified income file.
• Choosing the higher, older year solely to hit a DTI target is a misrepresentation of the borrower's income to the lender and any investor, and it defeats the ATR requirement to verify income and reasonably conclude repayment is likely.

The correct path: obtain and document the 2024 income, run the DTI honestly at 46%, and see whether reserves, a larger down payment, a co-borrower, or a different product resolves it. If not, decline the file - a lock expiration is a business problem, not a lawful basis for unsupported underwriting.

Three duties kick in at once.

1. Adverse action notice. Under ECOA and Regulation B, the applicant must receive notice of the action taken within 30 days of a completed application, with the specific principal reasons for the denial (or a statement of the right to request them). 'The lock ran out' or a vague 'does not meet guidelines' is not a permissible reason; the real reason is unverified or declining self-employed income producing a 46% DTI.
2. No backdating or document manufacturing. Marcus must not date the file to beat the lock, alter the P&L, or instruct the processor to write income in as bonus. False statements to a lender or investor can violate 18 U.S.C. 1014 and 1001 and are independently grounds for license discipline.
3. Accurate records and communication. Keep the processor's suggestion and Marcus's response in the file, keep the borrower informed in writing, and disclose any new lock or extension cost that changes the loan terms (a new fee for a lock extension can require revised disclosures and a new tolerance analysis).

The judgment point: Marcus answers to the borrower and to the underwriting record, not to the calendar. If the deal cannot be documented, it does not close.