Modeling Securitized Credit
Dave Goodson
Head of Securitized
Brett Cornwell, CFA
Co-Head of Client Portfolio Management, Fixed Income
Investors looking to assess the potential effects of a securitized credit allocation have limited options given the lack of a widely accepted benchmark or peer universe. Our modeling framework offers an interim solution for optimizing exposure within a higher-yielding multi-asset credit mandate.
Introduction: Why securitized credit needs a custom framework
The challenge
Securitized credit has come a long way since the global financial crisis. Regulation has improved, the technology for structuring deals is observably better, and more types of securities have emerged. As the market has grown and evolved, the opportunity set has become a diverse range of economic sectors, security structures, coupon types and credit ratings, from AAA to below investment grade and nonrated securities.
Yet this very diversity—a.k.a., fragmentation—has posed a significant challenge to index providers, resulting in indexes that have fallen short of fully capturing this market beta. For example, the Bloomberg U.S. Aggregate Index excludes floating-rate securities and any issues not scored by the Big 3 ratings agencies, and size minimums can prohibit entry into the index, leaving the Agg holding mostly agency RMBS. Meanwhile, the J.P. Morgan CLOIE Index and the Vista Credit Risk Transfer indexes only represent single sectors. Furthermore, Morningstar has yet to create a dedicated category for securitized credit funds, lumping them together with dissimilar strategies in either Multisector or Nontraditional bond categories.
We believe the lack of a reliable and reasonably indicative dataset to use in portfolio optimizations has caused investors to consistently underallocate to this dynamic opportunity set.
So we decided to build one. In 2020, we developed an analytical framework that has served as a focal point for countless client conversations, providing an intuitive model in lieu of a standardized benchmark to support allocations to this asset class. Our 2024 report extends the analysis through 2023.
Our approach
- Define “securitized credit” in the context of a dedicated allocation, focusing on the four major securitized food groups: commercial mortgage-backed securities, asset-backed securities, non-agency residential mortgage-backed securities (including credit risk transfers) and collateralized loan obligations.
- Identify appropriate constituents, starting with securitized credit weighting constraints, then filtering out tourists and category hoppers that might misrepresent the asset class, and finally confirming broad availability. The outcome is a list of 10 constituent funds we can use to create a performance series.
- Build a performance series covering 2013 to 2023, using equal-weighted allocations that are rebalanced and reconstituted annually.
- Define a base portfolio for optimization based on correlations and risk/return characteristics. This analysis leads us to a 50/50 portfolio of high yield and bank loans as our starting point, against which we optimize securitized credit weights. This approach reflects clear parallels among the three asset classes, often referred to as “plus” sectors and frequently combined as part of a higher-yielding multi-asset (or multi-sector) credit mandate.
- Determine strategic allocation bands that maximize the Sharpe ratio, using (1) constrained analysis that optimizes securitized credit against a static base portfolio and (2) unconstrained analysis that allows weightings to vary freely. Both approaches show improved risk-adjusted returns with the addition of securitized credit, owing primarily to lower volatility.
The resulting optimization shows that a 40-55% weight to securitized credit within an allocation to higher-yielding fixed income “enhancers” is a good starting point. An analysis of other fixed income allocations, overlaid with goals and objectives, may influence allocations when trying to build a better fixed income framework for clients.