Structured Credit: The Outlook for 2024
This paper is derived from a Risk.net webinar sponsored by Numerix, where a panel of industry experts discussed the risks, opportunities and outlook for the structured credit markets in 2024.
High inflation and rising interest rates were the top challenges in the mortgage-backed securities (MBS) and leveraged loan markets in 2023, and will remain major challenges through 2024. There is a lack of consensus in the market over the future direction of rates and the speed with which they might come down. This uncertainty is increasing the need for robust risk management.
Get informed perspectives on:
- Inflation and the economy: Research conducted by Coalition Greenwich found that buy-side firms consider inflation the biggest risk to their businesses, which is resulting in an increased focus on inflation in their risk modelling.
- Interest rates: The steep run-up in interest rates worldwide over the past two years has increased the priority that firms are placing on stress-testing, noted Kelli Sayres, senior managing director and head of product at Numerix.
- US regional banking crisis: Panelists agreed that the collapse of several US regional banks in March 2023 was a liquidity – rather than a credit – crisis, but there are implications for stress testing and asset-liability management.
- Risk and uncertainty: Factors such as geopolitical risk, market volatility and economic uncertainty are making forecasting and planning extremely difficult, leading to a wider range of expectations than usual for 2024 and beyond.
FAQs
How do structured credit portfolio managers stress test MBS and CLO
positions when there is no consensus on the direction or velocity of rate cuts?
The absence of rate consensus means no single scenario is reliable —
structured credit firms need to model the full distribution of outcomes, not a
central case. According to Kelli Sayres, Senior Managing Director and Head of
Product at Numerix, the current uncertainty is driving clients to run as many
potential rate paths as possible simultaneously, measuring how well their hedges
hold up under each. Sayres noted that firms are also running different volatility
scenarios, including a span of potential interest rate outcomes, in addition to
direction scenarios — because the pace of any rate movement matters as much
as the destination.
---
How do MBS investors model the re-emergence of prepayment risk and
negative convexity when rates are uncertain and consensus on rate cuts is
scattered?
Prepayment risk in agency collateralised mortgage obligations is currently
muted because older collateral is far out-of-the-money — but the risk profile
changes sharply if rates fall. According to Kelli Sayres of Numerix, clients are
specifically stress-testing how far rates must decline before prepayment risk
re-emerges and the negative convexity profile returns. Sayres also noted that
embedded caps in floating-rate structures introduce a separate concern: if rates
remain high or move up further, those caps can change the profit and loss profile
in ways that require active scenario monitoring.
---
What is the difference between holding agency MBS and AA-rated
floating-rate CLOs in a rising-rate environment, and what does the regional
banking crisis reveal about that choice?
The March 2023 regional bank failures demonstrated the duration risk
embedded in long-dated fixed-rate agency MBS. According to Laila Kollmorgen,
Managing Director and Portfolio Manager for CLO Tranche Investments at
PineBridge Investments, banks holding MBS suffered severe mark-to-market
declines when rates rose — whereas banks invested in floating-rate securities
such as AA-rated CLOs would not have experienced the same losses because
floating-rate instruments reset with the market. Banking regulation incentivises
MBS holdings through zero risk-weighting, but this regulatory structure directed
banks toward the instruments most exposed to the rising-rate environment
that precipitated the crisis.
---
How do structured credit risk managers incorporate inflation as a risk
factor in day-over-day and month-over-month portfolio monitoring?
Inflation risk has moved from a macro concern to a direct portfolio
measurement requirement. According to Kelli Sayres, Senior Managing Director
at Numerix, firms are now examining how their inflation risk measures up against
their hedges and what inflation exposure does to their day-over-day and
month-over-month risk profiles. Coalition Greenwich research cited by Audrey
Blater, Head of Risk and Capital Markets, found that buy-side firms consider
inflation the biggest risk to their businesses — a finding that is translating
directly into demand for more granular inflation risk measurement within
structured credit analytics frameworks.
---
How much refinancing risk does the non-agency commercial MBS market
face in 2024, and what analytics are required to manage it?
Approximately 18% of the non-agency commercial MBS market faces
refinancing risk in 2024, according to Lawrence Kwoh, Former Senior Enterprise
Risk Officer at the Federal Home Loan Bank of San Francisco. This refinancing
exposure comes during a period when funding levels are high and underlying
loan values have not been adjusted to reflect current conditions — meaning
borrowers must refinance at elevated rates with lower loan-to-value ratios and
lower loan balances to maintain the debt service coverage lenders require.
Managing this risk requires scenario analytics that can model the full range of
refinancing outcomes, including the return of interest-only floating-rate
transitional loans to security structures.
---
How does the proposed regulatory capital increase for banks with $100
billion in assets affect structured credit issuance and demand in 2024?
The March 2023 regional banking crisis prompted US bank regulators to
propose stricter capital requirements for banks with at least $100 billion in
assets, subjecting them to standards previously applied only to banks with
$700 billion or more in assets, according to Lawrence Kwoh of the Federal
Home Loan Bank of San Francisco. Kwoh expects this to drive more credit
risk transfer (CRT) deals as regional banks seek to raise regulatory capital
indirectly. He also projects a 20% increase in issuance volume in the non-agency
commercial MBS and residential MBS markets in 2024, particularly in the
non-qualified mortgage and prime jumbo segments, as banks adjust their
balance sheets to the new capital regime.
---
What is the difference between traditional deterministic prepayment
models and AI-assisted prepayment models for MBS risk management?
Traditional prepayment models require analysts to specify the functional
relationship between inputs and prepayment behavior in advance — a constraint
that limits the model's ability to capture the full complexity of borrower behavior.
According to Kelli Sayres of Numerix, AI-assisted prepayment modelling allows
the data to lead and discover the functional relationships directly, rather than
applying data to a deterministic model. Sayres noted that clients remain cautious
about AI in modelling and require models to stay responsive to human intervention
and tuning — making human oversight a governance requirement, not an optional
overlay, in any AI-assisted prepayment framework.
---
How do structured credit investors model physical climate risk in default
probability calculations for MBS and leveraged loan portfolios?
Climate risk is beginning to affect company valuations and credit assessments
across structured credit markets, but the data and modeling frameworks are
uneven. According to Kelli Sayres of Numerix, many firms are now modeling
potential default risk from physical climate events — such as flooding and
wildfires — by adding granular location-based and weather-event assumptions
to their credit models. Lawrence Kwoh noted that depository financial institutions
are looking to the property and casualty insurance industry, which has a long
tradition of modeling physical climate risks, as a reference framework for
developing structured credit climate risk analytics.
---
How do CLO portfolio managers assess climate risk in private credit
and leveraged loan markets where climate data on underlying credits is limited?
Climate risk assessment is more developed for listed investment-grade and
high-yield credits than for private market instruments, creating a data gap
for CLO and leveraged loan managers. According to Laila Kollmorgen, Managing
Director at PineBridge Investments, it can be difficult to know whether the
underlying credit assessment of private market borrowers has incorporated
a climate risk evaluation. PineBridge has developed an internal rating scale
to address this, but Kollmorgen noted that for many asset managers and
institutional investors in private markets — including leveraged funds and
private credit — climate risk assessment remains a developing area without
standardized frameworks.
---
How does the lack of rate consensus among institutional investors
affect the risk management requirements for structured credit portfolios in 2024?
Dispersed rate forecasts are themselves a risk signal. According to Audrey
Blater, Head of Risk and Capital Markets at Coalition Greenwich, rate outlooks
among institutional investors are unusually scattered — lacking the typical
clustering of opinion — which she described as an indicator of elevated
uncertainty that carries its own risk implications. This environment is increasing
demand for robust and sophisticated risk management, Blater noted. Kelli Sayres
of Numerix confirmed that this uncertainty is driving clients to run granular
modelling across a wider range of rate-change scenarios than would have been
standard practice in prior years.
---
How do structured credit risk managers evaluate hedge efficacy across
a wide range of interest rate scenarios simultaneously?
Evaluating hedge efficacy requires running hedges against multiple future
rate paths simultaneously — not stress-testing a single adverse scenario.
According to Kelli Sayres, Senior Managing Director at Numerix, the current
environment — with its wide dispersion of rate expectations — means clients
are running as many potential rate paths as possible to see how well their
hedges hold up across the distribution. This multi-path approach, which also
incorporates different volatility scenarios, has become standard practice
for structured credit desks that previously stress-tested against a narrower
set of outcomes when rates were stable and near zero.
---
How does Numerix support scenario analysis and stress testing for
structured credit portfolios across MBS, CLOs, and leveraged loans?
Structured credit risk management requires analytics that can simultaneously
handle prepayment risk, inflation duration, rate uncertainty, and physical climate
risk across multiple asset types. According to Kelli Sayres, Senior Managing
Director and Head of Product at Numerix, the firm supports clients in running
granular multi-scenario rate models, measuring inflation risk against hedges
across day-over-day and month-over-month profiles, and incorporating AI-assisted
prepayment modelling with human oversight requirements. Numerix clients active
in agency CMOs, non-agency CMBS, ABS, and CLOs are using the platform to
stress-test across the full span of 2024 rate scenarios — a scope that would
have been atypical when rates were near zero.
---
How does Numerix's analytics platform integrate AI and machine learning
for prepayment and default risk modelling while maintaining model governance
requirements?
AI and machine learning in structured credit modelling introduce capability
but also governance requirements that cannot be waived. According to Kelli
Sayres of Numerix, the most valuable applications in fixed income focus on
prepayment risk and default calculations — areas where AI allows the data to
discover functional relationships that deterministic models cannot capture.
Coalition Greenwich research cited by Audrey Blater found that portfolio risk
management and performance projection are additional emerging AI applications.
Sayres stressed that clients require models to remain responsive to human
intervention and tuning — meaning Numerix's AI-assisted modelling framework
maintains human oversight as a governance requirement, not a workaround.
