Skip to main contentSkip to navigationSkip to footer

Building an Underwriting Quality Control Program

10 min
5/6

Key Takeaways

  • Pre-funding QC should review 10-15% of pipeline loans; post-closing QC must review at least 10% of closed loans within 90 days.
  • Average defect rates range from 5-15%, with targets of below 2% for critical defects and below 8% for total defects.
  • QC reviewers must be independent from production staff to maintain program credibility.
  • Trend analysis by originator, underwriter, product, and branch identifies systemic issues before they produce losses.

Quality control is the safety net that catches underwriting errors before they become defaults, repurchase demands, or regulatory violations. Every lending company is required to maintain a quality control program by its investors, agencies, and regulators. This case study follows the design and implementation of a QC program for a growing lending company.

Quality Control Program Requirements

Fannie Mae, Freddie Mac, FHA, and VA all require lending companies to maintain written quality control programs with specific minimum standards. Pre-funding QC reviews a sample of loans before closing to catch errors while they can still be corrected—best practice is to review 10-15% of loans in the pipeline. Post-closing QC reviews a random sample of closed loans within 90 days of closing—the minimum requirement is 10% of closed loans. Targeted QC reviews 100% of loans with specific risk characteristics: early payment defaults (loans delinquent within the first 6 payments), loans originated by newly hired loan officers, loans with documented exceptions or compensating factors, and any loan type that has produced prior defects. The QC review covers all aspects of the origination process: application accuracy, income and asset verification, appraisal quality, compliance with regulations, and adherence to investor guidelines.

Defect Tracking and Remediation

QC findings must be categorized by severity and tracked for trends. Critical defects (material misrepresentation, compliance violations, ineligible loans) require immediate remediation, potential loan repurchase, and personnel action. Significant defects (documentation gaps, calculation errors, guideline exceptions) require correction, retraining, and process improvement. Minor defects (formatting issues, non-material omissions) require coaching and monitoring. Industry data shows that average defect rates for lending companies range from 5-15% of reviewed loans, with income calculation errors, missing documentation, and compliance timing violations as the most common findings. Trend analysis—tracking defect rates by originator, underwriter, product type, and branch—identifies systemic issues before they produce losses. A well-run QC program targets a critical defect rate below 2% and a total defect rate below 8%.

QC Program Implementation

Implementation of a QC program involves several key decisions. In-house vs. outsourced: companies originating fewer than 50 loans per month often outsource QC reviews to third-party firms ($100-$250 per review) for objectivity and cost efficiency. Larger companies build in-house QC departments for control and responsiveness. Technology: QC management software tracks review assignments, findings, remediation, and trend reporting. Popular platforms include ACES Quality Management, QC Ally, and MetaSource. Staffing: QC reviewers should be experienced underwriters or processors with no reporting relationship to production staff (independence is essential for credibility). Reporting: monthly QC reports should be distributed to senior management and include defect rates, trend analysis, remediation status, and recommended process improvements. The QC program should be reviewed and updated annually based on findings, regulatory changes, and investor guideline updates.

Go / No-Go Decision Framework

Go Indicators

  • Pre-funding QC should review 10-15% of pipeline loans; post-closing QC must review at least 10% of closed loans within 90 days.
  • Average defect rates range from 5-15%, with targets of below 2% for critical defects and below 8% for total defects.

No-Go Indicators

  • Treating QC as a check-the-box exercise rather than using findings to improve underwriting quality: The same deficiencies recur month after month, the deficiency rate remains elevated, and investor repurchase risk accumulates.
  • Selecting only random samples for QC review without risk-based targeted sampling: Higher-risk loans (exceptions, new originators, complex income) may not be reviewed at a rate that detects systemic issues.

Scenario: Designing Pacific Lending’s Quality Control Program

Pacific Lending is a 14-month-old mortgage banker originating 35 loans per month across 3 states. After receiving a repurchase demand from an investor on a loan with overstated income, the company needs to build a comprehensive QC program. Current staffing includes 4 loan officers, 2 processors, and 1 underwriter.

Outcome

Pacific Lending’s QC program costs $1,500/month ($700 outsourced reviews + $800 allocated internal time) but produces measurable results within 6 months: critical defect rate drops from 8% to 2.5%, total defect rate drops from 18% to 9%, no additional repurchase demands, and the company uses QC trend data to identify that one loan officer consistently miscalculates self-employed income—targeted retraining resolves the issue. The investor relationship is preserved, and the QC program becomes a selling point when applying for additional investor approvals.

Common Mistakes to Avoid

Treating QC as a check-the-box exercise rather than using findings to improve underwriting quality

Consequence: The same deficiencies recur month after month, the deficiency rate remains elevated, and investor repurchase risk accumulates.

Correction: Implement a feedback loop from QC findings to underwriting training, process improvements, and originator performance management.

Selecting only random samples for QC review without risk-based targeted sampling

Consequence: Higher-risk loans (exceptions, new originators, complex income) may not be reviewed at a rate that detects systemic issues.

Correction: Supplement the 10% random sample with targeted selection criteria: all exceptions, new originator loans, complex income, and high-LTV loans.

Test Your Knowledge

1.What is the primary purpose of a post-closing quality control (QC) program?

2.What percentage of loans must be reviewed in a standard post-closing QC program?

3.What is a "deficiency rate" in underwriting quality control?