Linkedin    Twitter   Facebook

Get Started
Log In

Linkedin

Articles Tagged with: MSRs

RiskSpan’s April 2025 Models & Market Call: Credit Model v7, Prepay Volatility, and Credit Trends to Watch

Register here for our next monthly model update call: Thursday, May 15th at 1:00 ET.

Note: This post contains highlights from our April 2025 monthly modeling call, which delivered insights into the current economic climate, mortgage model enhancements, and borrower behavior trends. You can register here to watch a recording of the full 28-minute call.

Here’s what you missed:

Market Overview: A Climate of Volatility

With mortgage rates rebounding to 7%, the panel began by acknowledging the choppy waters ahead, flagging 2025 as a year likely to see persistent rate volatility. As recession risks grow and consumer stress indicators rise, modeling accuracy becomes more important than ever.

Notably, consumers are already strained:

  • Rising consumer debt burdens
  • Increased use of personal loans and second liens for debt consolidation
  • Spikes in HEL/HELOC originations and securitizations
  • Climbing Non-QM delinquencies, particularly among 2022–2023 vintages

Model Update: Credit Model v. 7.0

RiskSpan’s newly released Credit Model v7 marks a significant upgrade in loan performance modeling:

  • Delinquency Transition Matrix core structure
  • The model projects:
    • Monthly CDR, CPR, and delinquency balances (0 through REO)
    • Loss severities, liquidated balances, and P&I flows
  • Modular components include:
    • State Transition Model
    • Severity and Liquidation Timeline Modules
  • The model is fully integrated within RiskSpan’s platform, enabling custom inputs for whole loans and securities

This model empowers users with granular delinquency and cash flow forecasting, critical for managing portfolios amid market uncertainty.


Key findings here included:

  • Daily prepay data showing extreme volatility, but offering early trend visibility
  • Trend lines derived from daily data offering good proxies for future behavior
  • Notable discrepancies within MBS-level data, especially among higher-coupon pools

RiskSpan’s continued focus on benchmarking these data sources helps refine both near-term and long-term modeling strategies.


Prepayment Behavior of Top-Tier Borrowers

The panel spotlighted borrowers with FICO scores over 800, revealing some counterintuitive dynamics:

  • Initial refinance activity is higher in the 800+ cohort—”fastest out of the gate”
  • But post-seasoning, refinance rates fall below those of the 700–750 FICO group
  • This “crossover pattern” reflects a phenomenon the team called “Accelerated Burnout”
  • Assumed strategic behavior, like exploiting lender credits, may amplify early refinance intensity

These insights underscore the nonlinear and evolving nature of borrower behavior, especially under fluctuating rate environments.


Model Performance: Staying on Track

RiskSpan’s Prepayment Model continues to track closely with actuals, validating its calibration even in today’s turbulent landscape. Combined with Credit Model v7, clients now have powerful tools for capturing credit and prepayment risk with more accuracy than ever.

Be sure to register for next month’s model update call on Thursday, May 15th at 1:00 ET.

Want a deeper dive into the new Credit Model or Prepay insights? Contact me to schedule a session with our modeling experts.



Navigating the Bulk MSR Trading Market in 2025: Insights from Industry Experts

Earlier this week, RiskSpan hosted a webinar featuring a panel of experts who provided a comprehensive look at the current state of the mortgage market, with a particular focus on mortgage servicing rights (MSRs), market analytics, and risk management strategies. Featuring commentary from Michael Fratantoni, Chief Economist of the Mortgage Bankers Association, alongside Geoffrey Sharp of Eris Innovations and RiskSpan’s Dan Fleischman and Chris Kennedy, the event offered a timely and in-depth discussion on the evolving challenges and opportunities confronting bulk traders in the MSR space.

Register here to listen to the full webinar recording.

Economic Outlook: Slowdown in Sight

Mike Fratantoni’s introductory message was clear: the U.S. economy is showing signs of deceleration, and that slowdown is being felt acutely in the housing and mortgage sectors.

Fratantoni highlighted that inflation, while trending downward, remains a key concern. Mortgage rates, elevated through much of 2024, have shown some easing in recent months but remain a barrier to both home purchases and refinancing activity. He pointed to a growing recognition that the global and U.S. economies are slowing — and an increasing risk they could slow more than expected.

This economic climate has direct implications for mortgage originators and servicers. Origination volumes have been suppressed due to affordability challenges and low housing inventory. Meanwhile, servicers are navigating increased costs and evolving regulatory expectations, making effective risk management more important than ever.


Dan Fleischman and Chris Kennedy then dove more deeply into the MSR market. Despite market headwinds, investor appetite for MSRs remains robust, largely driven by the asset’s countercyclical appeal and attractive risk-adjusted returns.

Kennedy explained that bulk MSR trading is still quite active, with some notable dislocation between bid and ask prices depending on loan characteristics and servicing costs. He also emphasized the importance of data quality in navigating this market, especially given the divergence in GSE prepayment behavior and the wide range of models being used to value servicing portfolios.

Fleischman expanded on the analytics side, walking through how servicers are increasingly relying on machine learning and historical GSE data to refine valuation and hedging strategies. “There’s a clear shift towards more granular modeling,” he noted, “not just at the loan level, but factoring in behavioral differences by servicer, geography, and even sub-servicer.”


Interest Rate Risk: Zero Swaps as a Hedging Tool

Geoff Sharp of Eris Innovations focused on how MSR investors are using Eris SOFR swap futures to manage interest rate exposure. As interest rates remain volatile, the duration and convexity risk associated with MSRs has become harder to hedge using traditional instruments.

“Zero swaps give investors a cleaner, more precise hedge,” Sharp explained. Unlike standard interest rate swaps, which exchange floating for fixed payments, zero-coupon swaps strip out the coupon and focus purely on duration. This allows for tighter alignment with MSR portfolio sensitivities, especially in high-rate environments where convexity matters.

Sharp also emphasized that the adoption of these instruments is no longer limited to the largest institutional players. “We’re seeing more mid-sized servicers look into this,” he said, “because the volatility has made traditional hedges more expensive and less effective.”


GSE Behavior and Prepayment Models: The Devil in the Data

Panelists frequently came back to the complexity of modeling prepayments in today’s market. With refinance incentives mostly absent, borrower behavior is increasingly driven by non-rate factors like relocation, cash-out needs, and credit events.

Dan Fleischman noted a recent shift in GSE delivery data that is reshaping how investors think about prepay risk. “We’re seeing very different prepayment speeds by seller and servicer,” he said. “Some of that is a function of portfolio composition, but some of it is clearly behavioral or operational.”

RiskSpan has been at the forefront of efforts to normalize and benchmark this data, providing servicers with a clearer picture of how their MSR assets may perform relative to the market. The panel stressed that accurate, up-to-date GSE data is critical not just for pricing MSRs, but also for identifying outliers and opportunities in both acquisition and sale.


Regulatory and Operational Considerations

In the final portion of the webinar, panelists discussed the regulatory and operational realities facing servicers in 2025. Compliance costs continue to rise, driven by both federal scrutiny and investor expectations around data security, customer experience, and portfolio transparency.

Chris Kennedy underscored the importance of operational efficiency, especially as revenue margins tighten. “Servicers are having to do more with less,” he said, “which means automation, smart analytics, and scalable infrastructure are no longer optional — they’re table stakes.”

There was also discussion around how MSR buyers are performing increasingly detailed diligence, not only on loan-level characteristics but on the servicing platform itself. Buyers want to understand call center metrics, delinquency management strategies, and borrower retention initiatives before committing capital.


What we learned

Here are some of what we consider to be the webinar’s key takeaways:

  • Economic Softness: A slowing economy is constraining origination volume, but the MSR asset remains a bright spot due to stable cash flows and defensive qualities.
  • Evolving Analytics: Servicers and investors are leveraging advanced analytics and GSE data to improve pricing, risk assessment, and benchmarking.
  • Hedging Innovation: Tools like zero-coupon swaps are gaining traction as more precise instruments for managing rate risk.
  • Behavioral Complexity: Modeling prepayments is harder than ever, requiring sophisticated data approaches and continuous recalibration.
  • Operational Readiness: In a tighter margin environment, servicers must optimize platforms to remain competitive and compliant.

The mortgage servicing world is not immune to the broader economic uncertainty, but for those with the right tools, data, and discipline, the MSR space still presents compelling opportunities. Success in today’s market requires a mix of macro awareness, micro-level analytics, and a relentless focus on operational performance.

Contact us to discuss, learn more, or get a free demo or trial of RiskSpan’s award-winning MSR solution.


Mortgage Prepayment and Credit Trends to Watch

Register here for our next monthly model update call: Thursday, April 17th at 1:00 ET.

Note: This post contains highlights from our March 2025 monthly modeling call. You can register here to watch a recording of the full 28-minute call.

Mortgage and credit markets remain dynamic in early 2025, with macroeconomic conditions driving both volatility and opportunity. In yesterday’s monthly model call, my team and I shared key insights into current market trends, model performance, and what to expect in the coming months.

Market Snapshot: A Mixed Bag

After trending downward in February, mortgage rates ticked up slightly in early March. Despite the fluctuation, expectations are for rates to remain relatively stable until at least summer 2025. Most mortgage-backed securities (MBS) are still deeply out of the money, making housing turnover—not rate refinancing—the dominant prepayment driver.

Macroeconomic signals remain mixed. While unemployment is still low and wage growth continues, inflation shows signs of persistence. The Fed is expected to hold the Fed Funds Rate steady through mid-year, with a potential first cut projected for June. Credit usage is creeping higher—especially in second liens and credit cards—hinting at growing consumer debt stress.


Model Performance and Updates

Prepayment Model

RiskSpan’s prepayment model continues to track closely with actuals across Fannie Mae, Freddie Mac, and Ginnie Mae collateral. The model shows:

  • Prepayments rising slightly, particularly among 2023 vintage loans in response to rate moves.
  • Delinquent loan behavior providing rich insights: For “out of the money” (OTM) collateral, delinquent loans are showing higher turnover speeds than performing ones, as borrowers try to avoid foreclosure.
  • Turnover sensitivity to borrower FICO scores is especially pronounced for delinquent loans—highlighting the need for granular credit analytics.

These behavioral insights are informing the next version of our prepayment model, which will incorporate GSE data research to enhance forecast accuracy.

Credit Model v7: A Leap Forward

RiskSpan’s new Credit Model v7—now available—is a significant upgrade, built on a delinquency transition matrix framework. This state-transition approach enables monthly projections of:

  • Conditional Default Rates (CDR)
  • Conditional Prepayment Rates (CPR)
  • Loss severity and liquidated balances
  • Scheduled and total principal & interest (P&I)

The model’s core components include:

  • A vector-based severity model
  • A robust liquidation timeline module
  • Loan-level outputs by delinquency state (including foreclosure and REO)

By modeling the lifecycle of loans and MSRs more explicitly, Credit Model v7 delivers deeper insight into portfolio credit performance, even in volatile markets.


Emerging Risks and Opportunities

Consumer credit balances—especially HELs and HELOCs—have grown significantly, fueled in part by debt consolidation. Credit card utilization has jumped from 22% in 2020 to nearly 30% as of late 2024, indicating growing financial strain.

Meanwhile, delinquencies in the Non-QM space (2022-2023 vintages) are rising—suggesting that investors need enhanced tools to monitor and manage these risks. RiskSpan’s tools, including the enhanced credit model and daily prepay monitoring, help investors keep pace with these shifting dynamics.


Looking Ahead

RiskSpan’s modeling team remains focused on:

  • Continuing to improve prepayment modeling with newly available GSE data
  • Rolling out and enhancing Credit Model v7 for broader use cases
  • Providing clients with forward-looking analytics to anticipate credit stress and capitalize on market dislocations

Be sure to register for next month’s model update call on Thursday, April 17th at 1:00 ET.

Want a deeper dive into the new Credit Model or Prepay insights? Contact me to schedule a session with our modeling experts.



Webinar: MSR Trading Insights

ReGISTER for the recording

Webinar: Tuesday, March 25th | 1:00 ET 
MSR Bulk Trading Insights

Join us for an update from MBA’s Chief Economist, Michael Fratantoni, on the current state of the MSR market.

Then, stick around for actionable strategies from RiskSpan’s Chris Kennedy and Dan Fleishman on how to gain a competitive edge, including:

– How to effectively leverage strategic bidding to maximize outcomes.
– The importance of on-the-fly, ad hoc analysis in responding to market dynamics.
– Best practices for MSR valuations and trading analytics to ensure precise decision-making.

Whether you’re scaling your MSR portfolio or seeking to optimize your trading processes, this webinar will equip you with the tools and insights to stay ahead in a competitive landscape.

Panelists
Michael Fratantoni, Chief Economist, Mortgage Bankers Association

Chris Kennedy, Director, RiskSpan

Dan Fleishman, Head of Client Success, RiskSpan


MSR Documentation Form

MSR Documentation

Please register for this content


MSR Presentation Form

MSR Introductory Presentation

Please register for this content


February 2025 Model Update: Mortgage Prepayment and Credit Trends to Watch

Note: This post contains highlights from our February 2025 monthly modeling call. You can register here to watch a recording of the full call (approx. 25 mins).

As we move further into 2025, key trends are emerging in the mortgage and credit markets, shaping risk management strategies for lenders, investors, and policymakers alike. RiskSpan’s latest model update highlights critical developments in mortgage prepayments, credit performance, and consumer debt trends—offering valuable insights for investors, traders, and portfolio/risk managers in these spaces.

Prepayment speeds have continued to decline in Q1 2025, largely due to a lack of housing turnover and persistently high mortgage rates. While a drop in rates during Q3 2024 temporarily mitigated lock-in effects for borrowers with very low rates, MBS speeds remain low across most cohorts.

Key drivers of observed prepayment behavior include:

  • Mortgage rates are expected to stay high (~6.5%+) throughout 2025, keeping refinancing activity muted.
  • Turnover remains the primary driver of prepayments, with most MBS pools significantly out of the money.
  • RiskSpan’s Prepayment Model v3.7 effectively captures these dynamics, particularly the impact of deep out-of-the-money (OTM) speeds based on moneyness.

Growth in Non-QM and Second Lien Originations

The private credit market continues to expand, with increasing Non-QM and second lien originations. However, a concerning delinquency trend has emerged, with delinquencies among 2022-2023 Non-QM vintages now rising faster than among older vintages.

Consumer Debt Pressures Mounting

Consumer debt continues to rise rapidly, raising concerns about long-term credit performance:

  • Credit card balances have increased significantly, with utilization climbing from 22% in 2020 to 30% by late 2024.
  • More consumers are turning to personal loans for debt consolidation, a sign of financial strain.
  • Second liens (HEL/HELOCs) are being used to pay off high-interest debt, fueled by strong home equity growth since 2020.

Model Enhancements

To address these evolving market conditions, RiskSpan has rolled out key enhancements to its mortgage and credit models:

  • Prepayment Model v3.7 – Captures deep out-of-the-money lock-in effects with improved accuracy across Fannie, Freddie, and Ginnie collateral.
  • Credit Model v7 – Introduces a Delinquency Transition Matrix, providing more granular forecasting for loans and MSR valuation.
  • Non-QM Prepayment Model – Developed using CoreLogic data, offering improved prepayment insights for Non-QM loans.

Looking Ahead

  • Rates are likely to remain high, with no reductions expected before summer.
  • Home equity growth remains strong, driving continued second lien origination.
  • Debt servicing costs are beginning to strain consumers, as high interest rates persist.
  • Delinquency rates show strong correlation to credit quality, signaling potential risks ahead.

The evolving mortgage and credit landscape underscores the importance of robust modeling and risk assessment. With prepayments slowing, debt burdens rising, and consumer credit trends shifting, lenders and investors must adapt their strategies accordingly.


RiskSpan Launches Comprehensive MSR Analytics Solution

Arlington, VA – January 25, 2025 – RiskSpan, a leading technology provider of innovative risk management and data analytics for loans, securities and private credit, today announced the launch of its state-of-the-art MSR Analytics Solution, available through RiskSpan’s Edge Platform. This integrated, end-to-end data and analytics solution revolutionizes how mortgage servicing rights (MSRs) are analyzed, managed, and priced.

The solution is uniquely positioned to serve the needs of MSR traders and investors, offering capabilities tailored to agency, non-QM, and jumbo loans. It combines granular loan-level historical performance analysis, advanced machine learning models for tape cracking, and customizable scenario testing, all on a secure, fast, and scalable, cloud-native platform.

Key Features of the MSR Analytics Solution

  1. Loan-Level Analysis and Insights:
    Users can interactively query and filter loan data, create customized cohort stratifications, and access detailed historical performance metrics such as prepayment, default, and recapture rates. Visual reports and data queries are seamlessly integrated into Snowflake for enhanced accessibility and efficiency​.
  2. Streamlined Data Mapping and Consolidation:
    The platform’s Smart Mapper technology simplifies the process of loading and mapping portfolios from multiple servicers, saving hours of manual work. RiskSpan’s advanced QC rules and machine learning models further enhance data precision and reliability​.
  3. Robust MSR Pricing Models:
    RiskSpan’s loan-level MSR pricing models significantly reduce pricing errors by offering granular cash flow forecasts, option-adjusted valuations, and segmentation capabilities. The in-house modeling team continuously updates the tools to ensure accuracy and reliability​.
  4. Advanced Risk Analysis and Scenario Testing:
    Users can run multiple interest rate and pricing scenarios to explore a range of potential MSR valuations. The platform’s customizable interface supports automated overnight analytics, integrates with enterprise risk systems, and enhances decision-making confidence for buy/sell strategies​.

A Game-Changer for the MSR Market

“RiskSpan’s MSR Analytics Solution represents a significant step forward in delivering actionable insights to MSR portfolio managers,” said Chris Kennedy, Director of Sales at RiskSpan. “This new technology allows clients to navigate the complexities of the MSR market with precision and confidence. As the only commercial-grade MSR cash flow model that leverages GSE historical performance data, it offers unmatched transparency into market CPR speeds, delivering a comprehensive view of portfolio performance over time. I consider this to be the ‘secret sauce’ of our MSR Platform.” 

This solution empowers servicers, MSR sellers, MSR investors, and other stakeholders to make data-driven decisions, optimize portfolio performance, and meet critical deadlines with improved accuracy and speed.

For more information about RiskSpan’s Edge Platform and the new RTL functionality, please visit RiskSpan.com.


About RiskSpan

RiskSpan delivers a single analytics solution for structured finance and private credit investors of any size to confidently make faster, more precise trading and portfolio risk decisions and meet reporting requirements with fewer resources, and less time spent managing multiple vendors and internal solutions.   Learn more at www.riskspan.com.


Case Study: How a leading loan and MSR investor reduced costs with a loan-level approach

Learn more about how one whole loan and MSR investor (a large mortgage REIT) successfully overhauled its analytics computational processing with RiskSpan. The investor migrated from a daily pricing and risk process that relied on tens of thousands of rep lines to one capable of evaluating each of the portfolio’s more than three-and-a-half million loans individually (and how they actually saved money in the process). 

The Situation 

One of the industry’s largest mortgage REITs sought a more forward-thinking way of managing its extensive investment portfolio of mortgage servicing rights (MSR) assets, residential loans and securities. The REIT runs a battery of sophisticated risk management analytics that rely on stochastic modeling. Option-adjusted spread, duration, convexity, and key rate durations are calculated based on more than 200 interest rate simulations.

The investor used rep lines for one main reason: it needed a way to manage computational loads on the server and improve calculation speeds. Secondarily, organizing the loans in this way simplified the reporting and accounting requirements to a degree (loans financed by the same facility were grouped into the same rep line).  

This approach had some significant downsides. Pooling loans by finance facility was sometimes causing loans with different balances, LTVs, credit scores, etc., to get grouped into the same rep line. This resulted in prepayment and default assumptions getting applied to every loan in a rep line that differed from the assumptions that likely would have been applied if the loans were being evaluated individually. 

The Challenge 

The main challenge was the investor’s MSR portfolio—specifically, the volume of loans needing to be run. Having close to 4 million loans spread across nine different servicers presented two related but separate sets of challenges. 

The first set of challenges stemmed from needing to consume data from different servicers whose file formats not only differed from one another but also often lacked internal consistency. Even the file formats from a single given servicer tended to change from time to time. This required RiskSpan to continuously update its data mappings and (because the servicer reporting data is not always clean) modify QC rules to keep up with evolving file formats.  

The second challenge related to the sheer volume of compute power necessary to run stochastic paths of Monte Carlo rate simulations on 4 million individual loans and then discount the resulting cash flows based on option adjusted yield across multiple scenarios. 

And so there were 4 million loans times multiple paths times one basic cash flow, one basic option-adjusted case, one up case, and one down case—it’s evident how quickly the workload adds up. And all this needed to happen on a daily basis. 

To help minimize the computing workload, the innovative REIT had devised a way of running all these daily analytics at a rep-line level—stratifying and condensing everything down to between 70,000 and 75,000 rep lines. This alleviated the computing burden but at the cost of decreased accuracy because they could not look at the loans individually.

The Solution 

The analytics computational processing RiskSpan implemented ignores the rep line concept entirely and just runs the loans. The scalability of our cloud-native infrastructure enables us to take the nearly four million loans and bucket them equally for computation purposes. We run a hundred loans on each processor and get back loan-level cash flows and then generate the output separately, which brings the processing time down considerably. 

For each individual servicer, RiskSpan leveraged its Smart Mapper technology and Configurable QC feature in its Edge Platform to create a set of optimized loan files that can be read and rendered “analytics-ready” very quickly. This enables the loan-level data to be quickly consumed and immediately used for analytics without having to read all the loan tapes and convert them into a format that an analytics engine can understand. Because RiskSpan has “pre-processed” all this loan information, it is immediately available in a format that the engine can easily digest and run analytics on. 

What this means for you

An investor in any mortgage asset benefits from the ability to look at and evaluate loan characteristics individually. The results may need to be rolled up and grouped for reporting purposes. But being able to run the cash flows at the loan level ultimately makes the aggregated results vastly more meaningful and reliable. A loan-level framework also affords whole-loan and securities investors the ability to be sure they are capturing the most important loan characteristics and are staying on top of how the composition of the portfolio evolves with each day’s payoffs. 


MSR LP

MSR Buyers & Sellers

Eliminate Missed Opportunities and Costly Pricing Errors

  • Get complete coverage, including Agency, FHA/VA & Non-QM

  • Understand cohort-specific CPR trends relevant to your MSR sale or portfolio

  • Go beyond cohort-level valuation with loan-level, trading-quality prepayment and credit models

Get a free trial or demo

Product Summary

Introductory Presentation

Model Documentation

Built for Speed, Scale and Affordability

Cloud-Native for 15 Years

Get a Free Trial or Demo

Resources

view all

MSRs



Ready to elevate your analytics?

Contact Us
Get Started
Log in

Linkedin   

risktech2024