Selected for the Terner Labs Housing Venture Lab

Chapter 5

Debt, Risk, and the Brutal Math of DTI

Published by Series Homes. 4 minute read.

Underwriting collapses every borrower’s financial life into one fraction: total required monthly debt payments divided by stable monthly income. That fraction is the debt-to-income ratio. It expresses two layers of risk. The front-end ratio measures only the proposed housing payment. The back-end ratio measures every recurring obligation the credit report, pay stubs, or public records reveal. A residential loan closes only if both layers fall inside program tolerances after the automated system accounts for every dollar of verified income and every listed debt.

The ratio limits for FHA, conventional and VA loans

FHA calls its model 31 over 43. Thirty-one percent of gross monthly income may go to principal, interest, taxes, insurance, and mortgage insurance. Forty-three percent may go to all debts combined. Desktop Underwriter and Loan Product Advisor apply a dynamic grid. Conventional approvals usually clear at forty-five percent back-end if the file is clean and up to fifty percent when compensating factors appear. VA replaces front-end limits with a residual income test but still caps the back-end near fifty unless longstanding active-duty allowances push residual well above the benchmark.

What counts as a monthly debt

Every ratio counts only the minimum required payment on each debt, not the outstanding balance. A credit card with a five-hundred-dollar balance that reports a twenty-five-dollar minimum hurts more than a charge card that reports zero. Installment loans in deferment still register if the terms dictate future payments. Buy-now-pay-later plans count once they appear on the credit report. Co-signed auto notes, student loans, or mortgages remain in the denominator unless the primary obligor shows twelve months of unbroken payments from an account the borrower does not control. Alimony and separate maintenance enter exactly as recorded on the court order because they survive bankruptcy and lien stripping.

How student loans are counted

Student debt exposes the difference between program handbooks. FHA assumes every deferred or income-driven loan will cure to repayment and charges the ratio with one-half of one percent of the outstanding balance per month when the statement shows less. Fannie Mae uses whatever payment appears on the credit report as long as it fully amortizes the debt or the servicer confirms the income-driven number will continue for at least twelve months after closing. If the report shows zero, the lender must document a calculated payment or use one percent of the balance. The stricter treatment under FHA often pushes younger borrowers above the limit even when their real cash flow seems comfortable.

Compensating factors and manual underwriting

Compensating factors do not erase math. They allow the automated engine to stretch the back-end ceiling when certain objective strengths exist. Significant verified cash reserves, a thirty-six-month history of on-time housing payments at or above the proposed mortgage, an excellent credit profile, or documented residual income far beyond program floors each improve the statistical odds of repayment and therefore justify a higher ratio. The engine decides whether the combination merits an approve-eligible finding. An underwriter cannot override the system without moving the file into manual processing where the ratio caps snap back to handbook minimums.

Manual underwriting occurs when the automated system returns a refer or caution. The lender must then validate income, assets, credit, and property conditions line by line against written standards. Housing and total ratios typically revert to thirty one and forty three for FHA and to the residual matrix for VA. Conventional loans rarely proceed without automated approval because secondary markets add pricing penalties that erase consumer value. Manual review lengthens closing timelines, invites more conditions, and elevates repurchase risk for the lender. Most originators avoid it unless the borrower presents extraordinary strengths elsewhere in the file.

How rent from the other units changes the ratio

Rental income from the subject property changes the numerator and the denominator simultaneously. 75% of verifiable market rent from each non-owner unit is added to gross income. The entire proposed mortgage payment still loads into the housing ratio. On a three or four unit FHA purchase the file must also pass the self-sufficiency test that requires net rent after a vacancy haircut to meet or exceed the full mortgage payment. Failing that test or missing the global back-end cap leads to denial even if the borrower’s spreadsheet shows cash flow comfort. Appraisal-based rent shortfalls and inflated projections are frequent causes of last-minute collapse.

Three ways to lower your ratio that survive an audit

Borrowers near the limit have only three levers that withstand audit. First, pay off installment debts and obtain a recorded satisfaction so the monthly payment drops to zero. Second, refinance short-term or variable-rate obligations into longer-term structures that post lower required installments, bearing in mind that any new trade line must season for twelve months before the lower payment is fully credited on FHA files. Third, increase documented income with additional verifiable earnings that have seasoned for at least two tax years, such as consistent overtime or a formally reported side business. Tactics like moving balances between credit cards, deferring student loans to exploit lower income-driven payments, or closing seasoned accounts can trigger new inquiries, score drops, or handbook resets that injure the file more than they help.

When the answer is no

Lenders decline applications when the back-end ratio exceeds the program’s statistical tolerance, when compensating factors are weak or unverified, or when the automated engine refuses to believe rental projections will materialize. The system is not interested in narratives about eventual cash flow. It calculates only what is provable on the day of underwriting. If the resulting fraction sits to the right of the guideline slash the answer is no. Understanding that simple boundary—and the limited, documented ways to move it—is the borrower’s final filter before committing time, earnest money, and inspection fees to a transaction that underwriting math will never approve.

All chapters

  1. IntroductionThis Is Not for Most PeopleWhy house hacking is a test, not a shortcut, and why thinking like an underwriter is the only way to pass it.
  2. Chapter 1The Ownership Illusion: Why the American Dream Skips RentersHow institutional capital redefined single-family housing, and the one lane federal rules left open to owner-occupants.
  3. Chapter 2The Federal Loan Rule Nobody Taught YouThe occupancy rule that lets a 2-4 unit building be financed as a home, and the FHA, VA, and conventional terms that follow from it.
  4. Chapter 3You Are the Underwriter NowYour loan is a file scored by a machine. What that file has to contain before a listing is worth your time.
  5. Chapter 4Income, Employment, and What Lenders Actually TrustW-2s, self-employment, commission, and rent: how each income type is documented, averaged, or thrown out.
  6. Chapter 5Debt, Risk, and the Brutal Math of DTIFront-end and back-end ratios, program caps, student debt, and the three levers that actually move DTI.
  7. Chapter 6Credit Scores, Derogatories, and Why You’re Not Being JudgedScore floors, overlays, collections, and manual underwriting, and how to shape the report before a lender sees it.
  8. Chapter 7The Down Payment Myth: What Entry Actually CostsClosing costs, reserves, inspections, and mortgage insurance: the real cash it takes to close and operate.
  9. Chapter 8Rental-Coverage Gates Across Major Loan ProgramsThe FHA self-sufficiency test, conventional rent treatment, VA reserves, and why the appraiser's rent schedule rules.
  10. Chapter 9The False Security of Cash FlowWhy gross rent minus mortgage is marketing, and what real, fully loaded cash flow has to survive.
  11. Chapter 10Finding Properties That Qualify, Not Just Ones That Look GoodLegal unit counts, safety repairs, rent haircuts, and the filters that remove most portal listings before an offer.
  12. Chapter 11You Just Bought a Small Business with TenantsYou acquired an operating entity with tenants, revenue concentration risk, and one point of failure: you.
  13. Chapter 12Vacancy, Turnover, and the Cost of Being Too OptimisticRealistic vacancy, turnover cost, and time, and why a flat three percent vacancy line is a bet against reality.
  14. Chapter 13You Live Where You Work Now: What Nobody Tells You About Sharing WallsProximity to tenants is an edge only when it is paired with written rules and consistent enforcement.
  15. Chapter 14Refinancing, Holding, or Getting Out: What Comes NextWhen to refinance out of FHA, when holding still pays, and the three compliant ways to exit.
  16. Chapter 15This Is Only for a Few: Why That’s GoodThe friction is the filter. Why this model stays mispriced, and who it is actually for.