KBRA releases its prime RMBS default study, which analyzes over 455,000 loans representing $292.3 billion in original balance from nearly 640 prime transactions issued between 2010 and 2025. This report examines performance dynamics across key loan attributes—including vintage, combined loan-to-value (CLTV) ratio, credit score, occupancy, loan purpose, product type, and borrower reserves—and identifies how layered risk factors impact credit outcomes.
Key Takeaways
- Prime RMBS credit performance has remained exceptionally strong throughout the RMBS 2.0 era. Across more than 455,000 loans from over 640 prime RMBS transactions issued since 2010, cumulative default rates have reached approximately 1%, with negligible realized credit losses. Observed stress has been concentrated in discrete periods rather than reflecting sustained deterioration in collateral performance.
- Default activity was largely driven by a single macroeconomic shock. Defaults peaked during the COVID disruption in 2020, when annual default rates reached approximately 1.5%. Outside of this period, annual defaults have generally remained below 0.2%, underscoring the resilience of prime RMBS 2.0 underwriting standards.
- Realized losses are rare and concentrated in a small number of liquidation events. Of the 3,008 loans with reported losses—most of which are associated with active or prepaid collateral—average loss severities remain low, at approximately 2.8% for active loans and 0.4% for prepaid loans. Involuntary liquidations, totaling just 40 loans, exhibit substantially higher loss severities, averaging approximately 22.4%.
- Vintage-level default variation is driven primarily by seasoning and economic timing. Earlier post-global financial crisis (GFC) vintages (2010-17) exhibit low cumulative defaults, generally ranging from approximately 0.0% to 1.6%, supported by strong borrower profiles, higher prepayment rates, and macroeconomic tailwinds. Defaults increased for the 2017-19 vintages, which exhibited modestly higher leverage and risk layering and were outstanding during the 2020 COVID shock, resulting in cumulative defaults of roughly 1.8%-2.0% for the 2018 and 2019 cohorts. More recent vintages show limited defaults to date, reflecting their reduced seasoning.
- Default rates show clear variation by key mortgage default drivers like CLTV ratio and credit score. Defaults decline from roughly 3.3% in the sub-680 FICO bucket to approximately 0.3% for loans with scores above 800. Loans with CLTV below 80% generally exhibit default rates between approximately 0.3% and 0.8%, compared with 1.6% for 85%-90% CLTV and 2.7% for 90%+. Across the distribution, default rates remain contained, reflecting the overall underwriting quality of prime RMBS 2.0 pools, particularly when compared with NQM collateral of similar leverage and credit.
- Affordability and liquidity metrics show generally bounded performance differentiation. Default rates rise from approximately 0.2%-0.4% in the 10%-25% DTI range to roughly 0.9%-1.2% in the 35%-45% range, indicating a gradual but bounded increase in risk at higher affordability levels. Borrower liquidity, measured by months of reserves, demonstrates a generally inverse relationship, with loans reporting zero reserves exhibiting default rates of 1.3%, compared with 0.2%-0.6% for borrowers with more than 60 months of reserves.
- Loan purpose and product features also show marked differentiation in performance. Cash-out refinance loans (1.2%) exhibit higher default rates than purchase (0.7%) and rate/term refinance (0.6%) loans, despite broadly similar borrower credit characteristics. Across occupancy type, origination channel, property type, and borrower characteristics, default rate differences are marked but remain modest. For example, occupancy defaults range from 0.3% for second homes to 0.9% for owner-occupied loans, origination channel performance clusters between 0.7% and 1.1%, and property-type defaults generally remain below 1.5%.
- Risk layering, rather than any single attribute, drives meaningful performance deterioration. Default rates increase as multiple elevated risk characteristics accumulate within a loan, rising from approximately 0.6% for loans with no identified risk layers to roughly 0.9%-1.2% for loans with one to three risk layers, and to approximately 4.9% for loans with four risk layers.
Click here to view the report.
Recent Publications
- KBRA Non-QM RMBS Default Study: A Decade of Insights
- KBRA Non-QM RMBS Default Study: Credit Attribute Insights
- U.S. RMBS Credit Indices
- The Evolving Landscape of Noncitizen Borrowers in U.S. RMBS
- 2026 U.S. RMBS Sector Outlook: Normalizing Credit and Growing Issuance
About KBRA
KBRA, one of the major credit rating agencies, is registered in the U.S., EU, and the UK. KBRA is recognized as a Qualified Rating Agency in Taiwan, and is also a Designated Rating Organization for structured finance ratings in Canada. As a full-service credit rating agency, investors can use KBRA ratings for regulatory capital purposes in multiple jurisdictions.
Doc ID: 1013577
View source version on businesswire.com: https://www.businesswire.com/news/home/20260220383070/en/
Contacts
Armine Karajyan, Senior Director
+1 646-731-1210
armine.karajyan@kbra.com
Jack Kahan, Senior Managing Director, Global Head of ABS & RMBS
+1 646-731-2486
jack.kahan@kbra.com
Yee Cent Wong, Senior Managing Director, Lead Analytical Manager, Structured Finance Ratings
+1 646-731-2374
yee.cent.wong@kbra.com
Media Contact
Adam Tempkin, Senior Director of Communications
+1 646-731-1347
adam.tempkin@kbra.com
Business Development Contact
Daniel Stallone, Managing Director
+1 646-731-1308
daniel.stallone@kbra.com
