On June 18, the Federal Housing Administration provided lenders with a new Defect Taxonomy for Single Family Housing Loans, or Loan Quality Assessment Methodology. It has been working on the new taxonomy for almost a year, and the final product is very similar to what it presented for comment in September 2014. The Department of Housing and Urban Development began this process in May 2014 with a document titled Blueprint for Access, aimed at improving access and education for borrowers and providing clearer rules for lenders. The taxonomy targets lenders only, and attempts to provide a better understanding of FHA’s expectations, how it plans to “grade” lenders and address defects, and to simplify the the methodology as a whole. Whether it is successful depends on how many lenders agree that the new system is easier to follow and provides them with more certainty on how defect loans will affect them than the previous system.
The new Quality Assurance Defect Taxonomy does the following:
- Identifies nine fundamental characteristics of loan insurability;
- Uses those nine characteristics to create nine fundamental defects;
- Provides numerous sources and causes to attach to those defects;
- Gives lenders four different tiers of severity for each defect; and
- “addresses FHA’s plans to identify and capture information about defects and severities revealed through an individual loan-level review.”
The nine fundamental characteristics are separated into four categories–Underwriting (Borrower Income, Borrower credit/liabilities, LTV and Max Mortgage Amount, and Borrower Assets), Valuation Process (Property Eligibility and Property Appraisal), Eligibility (Borrower Eligibility and Qualification and Mortgage Eligibility), and Operations (Lender Operations). Each characteristic is also a defect category encompassing various reasons such defects occurred. For example, Borrower Income is a characteristic of loan insurability that would become a defect if it was inappropriately calculated by the underwriter or fraudulently provided by the borrower. The reason the defect exists would be the “source” and “cause” of the defect, and the Defect Taxonomy provides between 13 and 15 source and cause codes for each defect.
The defect will then be assigned a Tier depending on the whether it is minor or major, with Tier 1 being the most severe, and Tier 4 being the least.
Tier 1 Includes loans submitted “with information which the lender knew (or should have known) was misrepresented;” data used in TOTAL that is untrustworthy; and loans with a material violation of a statutory requirement.
Tier 2 Includes “errors in the loan file that even if identified and corrected, lead the loan to be unapprovable” either because the loan exceeds approval limits by “a large margin” or breach a loan guideline by “a large degree.”
Tier 3 Includes “errors present in the loan file that, even if identified and corrected, would lead the loan to be unapprovable” because the loan exceeds approval limits by “a small margin” or breach a loan guideline by “a small degree.”
Tier 4 Includes “errors … present in the loan file that impact key calculations or inputs, but which would not lead the loan to be unapprovable” (emphasis in original).
While none of the Tier levels changed between September 2014 and June 2015, the new Defect Taxonomy removes the terms “Unacceptable” and “Deficient” from the ratings system whereas the September draft stated that Tier 1 & 2 errors would lead to an “Unacceptable” rating and Tier 3 & 4 errors would lead to a “Deficient” rating. Instead, the FHA states that it “will use the tiers to communicate the severity of the defect rather than relying on the terms Unacceptable or Deficient.”
The consequences of various Tiers are also explained, in addition to how lenders can appeal any findings. First, “the defect code with the highest tier severity determines the overall loan rating.” Second, any instance of fraud or material misrepresentation will receive a Tier 1 rating, irrespective of a lender’s culpability, and a code will be assigned to note if a lender was not culpable in the fraud or misrepresentation. Third, the FHA outlined four possible outcomes after a defect has been identified: (1) Indemnification (lender indemnified FHA for any future issues), (2) Remediation (lender reimburses FHA for borrower fees, principal reductions, etc.), (3) Mitigation (no action necessary by FHA because lender resolves defect), or (4) Reversal (FHA applies incorrect defect code).
Importantly, the Defect Taxonomy notes that “lenders will not be held accountable for those defects about which FHA determines the lender did not know or could not have known.” And, defects that a lender successfully rebuts or reverses are not included in an lender’s defect rate.
While the above new Defect Taxonomy is an attempt to make loan-level reviews by the FHA more transparent, there are key questions lenders should consider in determining how useful the new protocol will actually be. In particular, the new Quality Assurance Defect Taxonomy DOES NOT:
- Address how FHA plans to sample loans for review (risk-based or random);
- Include standards for how other agencies such as the Department of Justice would review loan defects; or
- Establish “standards for administrative or civil enforcement action, which are set forth in separate law. The Taxonomy does not address FHA’s response to patterns and practice of loan-level defects, regardless of severity, or FHA’s plans to address fraud or misrepresentation in connection with any FHA-insured loan.”
Thus, lenders can still be targeted for large False Claims Act lawsuits or audits by various federal agencies working off an entirely different set of criteria and have no way to protect themselves. Lenders who want certainty that they can avoid lawsuits by government agencies if they agree to lend to more vulnerable borrowers simply do not have that certainty based on the FHA’s most recent attempt to be transparent. While the new Defect Taxonomy may be better than what was in place before (depending on your view), lenders must wonder if they are getting enough transparency from the government.