D1-3-01, Lender Post-Closing Quality Control Review Process (06/04/2025)

Loans must be selected for post-closing QC reviews on at least a monthly basis. The entire QC cycle (selection, review, rebuttal, and reporting) must be completed within 90 days from the month of the note date.

Note: If the lender's QC cycle is in arrears more than one 30-day cycle, the lender must provide written notice to its Fannie Mae customer account team or QC Specialist.

The lender must establish and document a process for identifying a representative sample of loans for QC file reviews. The post-closing QC process must include both random and discretionary file selections.

The random sample is a critical measure of loan quality. It provides representative insights into the overall quality of originations during a specific period and serves as the basis for calculating the lender's defect rate. The lender has the option to implement one of two methods of random sampling:

• a 10% sample of all monthly loan production, or
• a statistically valid sample of all monthly loan production.

Both sampling methods require the lender to randomly select loans. The lender must decide which sampling method is best for its organization to best allocate its QC resources.

With both options, the post-closing random sample must be completed using full-file reviews.

Random - 10% Sample

The lender must select for its post-closing QC review a minimum of 10% of the loans that it originates using a random selection methodology. 

Lenders who originate or acquire loans from a third-party must conduct a separate review for both its retail and third-party originations. Specifically, the lender must sample at least 10% of its retail originations and sample at least 10% of loans acquired through third-party originations.

If 10% is less than one loan, then at least one loan must be selected. The loans selected must be representative of the lender’s overall book of business, including:

• all of the different types of loans the lender offers,
• loans originated by each branch office,
• manually underwritten loans, and
• loans that were processed through automated underwriting system(s) utilized by the lender.

Random - Statistical Sample

If the lender uses statistical sampling for its selection process, at a minimum, the statistical sampling model (variables) must be calculated using a 95% confidence level with a 2% precision rate and a statistical statement of six months. (Fannie Mae recommends using a three-month statement.) 

Lenders who originate or acquire loans from a third-party must conduct a separate review for both retail and third-party originations. Specifically, the lender must have a statistical sample for its retail-originated loans and a separate statistical sample of loans acquired through third-party originators.

The lender must document its statistical sampling methodology and provide detailed written justification of the methodology to Fannie Mae, upon request.

Fannie Mae may require adjustments to the statistical methodology based upon its review.

Discretionary sample selections supplement, but do not replace a lender's random sample. The purpose of a discretionary sample is to look for or highlight areas that may pose unique or elevated levels of risk for the lender or to confirm that a particular control or process is working as intended. To be effective, the sampling methodology for discretionary review types must be flexible enough to target loans with higher potential for risk and adjust as these risks change over time.

The lender must establish a process for selecting loans for its discretionary post-closing QC. The process must take into account the lender’s assessment of the risks inherent in its origination processes, business sources, production channels, volume, and product mix, and must be reviewed regularly and, when necessary, adjusted to ensure that the sample selected, including sample size, is appropriate.

Loans selected for post-closing discretionary QC review must target areas the lender identifies as having a higher potential for errors, misrepresentation, or fraud. Targeted areas may include the following:

• loans with characteristics related to errors or defects identified in prior prefunding and post-closing review results;
• loans with complex income calculations (for example, rental income, self-employed, short history of receipt of income);
• loans requiring the use of non-standard processing or underwriting guidelines (for example, delayed financing, multiple financed properties, assets used as income, or manual reserve calculations);
• loans secured by properties located in areas with high delinquency rates or areas experiencing rapid increases or decreases in property values;
• loans with multiple layers of credit risk, such as high LTV ratios, low credit scores, or high DTI ratios;
• loans originated or processed through various business sources, a particular branch office, employee, contractor, third-party originator, or appraiser;
• loans originated by third-party originators where elevated risks are observed based on lender oversight and controls (for example, results from the stratified random sample, scorecard metrics, other investor results and third-party originators the lender identifies as having a higher potential for errors, misrepresentation, and fraud, etc.);
• loans originated or processed by newly hired loan officers, processors, appraisers, or other personnel or third parties involved in the loan origination process;
• loans that may be subject to concerns about delinquency rates or patterns identified in other reviews;
• loans for which the feedback or results from third-party tools indicates potential areas of concern; and
• loans that have a high risk for fraud, including loans that are early payment defaults (see A3-4-03, Preventing, Detecting, and Reporting Mortgage FraudA3-4-03, Preventing, Detecting, and Reporting Mortgage Fraud).