Linkedin    Twitter   Facebook

Get Started
Log In

Linkedin

Articles Tagged with: RS Edge

GenAI Applications for Loans and Mapping Data

RiskSpan is actively furthering the advancement of several GenAI applications aimed at transforming how mortgage loan and private credit investors work and maximizing their efficiency and performance. They include:

1. Tape-Cracking 3.0: Making RiskSpan’s Smart Mapper Even Smarter

RiskSpan’s Edge Platform currently uses machine learning techniques as part of its Smart Mapper ETL Tool. When a new portfolio is loaded in a new format, the fuzzy logic that powers the Platform’s recommended mappings gets continually refined based on user activity.

In the coming months, the Platform’s existing ML-driven ETL process will be further refined to leverage the latest GenAI technology.

GenAI lends additional context to the automated mapping process by incorporating an understanding not only of the data in an individual column, but also of surrounding data as well as learned characteristics of the asset class in question. The resulting evolution from simply trying to ensure the headers match up a more holistic understanding of what the data actually is and the meaning it seeks to convey will be a game changer for downstream analysts seeking to make reliable data-driven investment decisions.

RiskSpan made several updates in 2023 to help users automate the end-to-end workflow for loan valuation and surveillance. AI-based data loading combined with the Platform’s loan risk assumptions and flexible data model will enable users to obtain valuation and risk metrics simply by dragging and dropping a loan file into the application.

2. Modeling Private Credit Transactions

Many financial institutions and legal advisors still spend an extraordinary amount of time reading and extracting relevant information from legal documents that accompany structured private credit transactions.

RiskSpan has partnered with clients to develop a solution to extract key terms from private credit and funding transactions. Trained multimodal AI models are further extended to generate executable code valuations. This code will be fully integrated into RiskSpan’s risk and pricing platform.

The application solves a heretofore intractable problem in which the information necessary to generate accurate cash flows for private credit transactions is spread across multiple documents (a frequent occurrence when terms for individual classes can only be obtained from deal amendments).

Execution code for cash flow generation and valuation utilizes RiskSpan’s validated analytics routines, such as day count handling, payment calculations, discounting, etc.

3. “Insight Support”

Tech support is one of today’s most widely known (and widely experienced) GenAI use cases. Seemingly all-knowing chatbots immediately answer users’ questions, sparing them the inconvenience of having to wait for the next available human agent. Like every other company, RiskSpan is enhancing its traditional tech support processes with GenAI to answer questions faster and and embed user-facing AI help within the Platform itself. But RiskSpan is taking things a step further by also exploring how GenAI can upend and augment its clients’ workflows.

RiskSpan refers to this workflow augmentation as “Insight Support.”

With Insight Support, GenAI evaluates an individual user’s data, dynamically serves up key insights, and automatically completes routine analysis steps without prompting. The resulting application can understand an individual user’s data and recognize what is most important to identify and highlight as part of a loan data analysis workflow.

Insight Support, for example, can leverage insights obtained by the AI-driven “Smarter Mapping” process to identify what specific type of collateral reporting is necessary. It can produce reports that highlight outliers, recognize the typical analytical/valuation run settings a user would want to apply, and then execute the analytical run and summarize the results in management-ready reporting. All in the name of shortening the analysis time needed to evaluate new investment opportunities.

Conclusion

Considered collectively, these three applications are building toward having RiskSpan’s SaaS platform function as a “virtual junior analyst” capable of handling much of the tedious work involved in analyzing loan and structured product investments and freeing up human analysts for higher-order tasks and decision making.

GenAI is the future of data and analytics and is therefore the future of RiskSpan’s Edge Platform. By revolutionizing the way data is analyzed, AI-created and -validated models, dashboards, and sorted data are already allowing experts to redirect their attention away from time-consuming data wrangling tasks and toward more strategic critical thinking. The more complete adoption of fully optimized AI solutions throughout the industry, made possible by a rising generation of “AI-native” data scientists will only accelerate this phenomenon.

RiskSpan’s commitment to pushing the boundaries of innovation in the Loan and Structured Product Space is underscored by its strategic approach to GenAI. While acknowledging the challenges posed by GenAI, RiskSpan remains poised for the future, leveraging its expertise to navigate the evolving landscape. As the industry anticipates the promised benefits of GenAI, RiskSpan’s vision and applications stand as a testament to its role as a thought leader in shaping the future of data analytics.

Stay tuned for more updates on RiskSpan’s innovative solutions, as we continue to lead the way in harnessing the power of GenAI for the benefit of our clients and the industry at large.


Enriching Pre-Issue Intex CDI Files with [Actual, Good] Loan-Level Data

The way RMBS dealers communicate loan-level details to prospective investors today leaves a lot to be desired.

Any investor who has ever had to work with pre-issue Intex CDI files can attest to the problematic nature of the loan data they contain. Some are better than others, but virtually all of them lack information about any number of important loan features.

Investors can typically glean enough basic information about balances and average note rates from preliminary CDI files to run simple, static CPR/CDR scenarios. But information needed to run complex models — FICO scores, property characteristics and geography, and LTV ratios to name a few — is typically lacking. MBS investors who want to run to run more sophisticated prepayment and credit models – models that rely on more comprehensive loan-level datasets to run deeper analytics and scenarios – can be left holding the bag when these details are missing from the CDI file.

The loan-level detail exists – it’s just not in the CDI file. Loan-level detail often accompanies the CDI file in a separate spreadsheet (still quaintly referred to in the 21st Century as a “loan tape”). Having this data separate from the CDI file requires investors to run the loan tape through their various credit and prepayment models and then manually feed those results back into the Intex CDI file to fully visualize the deal structure and expected cash flows.

This convoluted, multi-step workaround adds both time and the potential for error to the pre-trade analytics process.

A Better Way

Investors using RiskSpan’s Edge Platform can streamline the process of evaluating a deal’s structure alongside the expected performance of its underlying mortgage loans into a single step.

EDGEPLATFORM

Here is how it works.

As illustrated above, when investors set up their analytical runs on Edge, RiskSpan’s proprietary credit and prepayment models automatically extract all the required loan-level data from the tape and then connect the modeling results to the appropriate corresponding deal tranche in the CDI file. This seamlessness reduces all the elements of the pre-trade analytics process down to a matter of just a few clicks.

Making all this possible is the Edge Platform’s Smart Mapper ETL solution, which allows it to read and process loan tapes in virtually any format. Using AI, the Platform recognizes every data element it needs to run the underlying analytics regardless of the order in which the data elements are arranged and irrespective of how (or even whether) column headers are used.

Contact us to learn more about how RMBS investors are reaping the benefits of consolidating all of their data analytics on a single cloud-native platform.


The future of analytics pricing is RiskSpan’s Usage-based delivery model

Usage-based pricing model brings big benefits to clients of RiskSpan’s Edge Platform

Analytic solutions for loans, MSRs and structured products are typically offered as software-as-a-service (SaaS) or “on-prem” products, where clients pay a monthly or annual fee to access the software and its features. The compute needed to run analytic workloads is typically purchased in advance and is fixed regardless of the need or use case.  

However, this traditional pricing model is not always the best fit for the dynamic and diverse needs of analytics users. It is technologically outdated and does not meet users where they are – with varying data volumes, usage patterns, and analytical complexity requirements that fluctuate with the markets. It is simply wasteful for companies to pay for unused, fixed-fee compute capacity, year-after-year in long-term, set price contracts, when their needs don’t require it. 

Usage-based pricing is a trend that reflects the evolving nature of analytics and the increasing demand for more flexible, transparent, and value-driven pricing models.

RiskSpan has just announced the release of industry-innovating usage-based pricing that allows clients to scale up or down, based on their needs. Further, clients of the RiskSpan platform will now benefit from access to the full Edge Platform, including data, models and analytics – eliminating the need to license individual product modules. The Platform supports loans, MSRs and securities, with growing capabilities around private credit. Analyzing these assets can be compute- and data-intensive because of the need for collateral (loan-level) data and models to price, value, and calculate risk metrics.

A Single Platform
Integrated Data | Trade Analytics | Risk Management

Core Engine

Usage-based pricing is an innovative alternative approach based on user-configured workloads. It enables RiskSpan to invoice its clients according to how much compute they actually need and use, rather than a fixed fee based on the modules they purchased during the last budget cycle.  

Usage-based pricing benefits RiskSpan clients in several ways, including: 

    • Lower Costs: Clients pay only for what they need, rather than being locked into an expensive contract that may not suit their current or future situation.

    • Cost-Sharing Across the Enterprise: Clients can share costs across the enterprise and better manage expense based on usage by internal functions and business units.

    • Transparency: Clients can monitor their usage and directly link their analytics configuration and usage to their results and goals. They can also better control their spending, as they can track their usage and see how it affects their bill.

    • Flexibility: Clients can experiment with different features and options of the Platform, as they are not restricted by a predefined package or plan.

Usage-based pricing is not a one-size-fits-all solution, and it may not be suitable for every organization. Based on needs, large enterprise workloads will require specific, customized licensing and may benefit from locked in compute that comes with volume discounts.

Bottom Line on RiskSpan’s Usage-based Pricing Model

CONS of Traditional Fixed Fee Pricing PROS of Usage-Based Pricing
Flat-fee pricing models force customers to pay for unused capacity​. Lower Costs — Pay only for what you use, not the wasted capacity of a dedicated cluster
Unused capacity cannot be shared across the enterprise, which translates into wasted resources and higher costs. Cost Sharing — Costs can be shared across the enterprise to better manage expense based on usage by your internal functions and business units
Fixed pricing models make it difficult for customers to scale up or down as needed. Transparency — Transparent pricing that fits your specific analytics workload (size, complexity, performance)
Traditional “product module-based” purchasing runs the risk of over-buying on features that will not be used. Flexibility — Scale up and scale down your use as new and in-place features become useful to you under different market conditions

With the introduction of usage-based pricing, RiskSpan is adding core value to its Edge Platform and a low-cost entry point to bring its solution to a wider base of clients. Its industry-leading capabilities solve challenges facing various users in the loans, MSR, and structured portfolio domains. For example:

    1. Loan/MSR Trader seeks analytics to support bidding on pools of loans and/or MSRs. Their usage is ad-hoc and will benefit from usage-based pricing. Traders and investors can analyze prepay and credit performance trends by leveraging RiskSpan’s 20+ years of historical performance datasets.

    1. Securities Trader (Agency or Non-Agency) wants more flexibility to set their prepay or credit model assumptions to run ad-hoc scenario analysis not easily handled by their current vendor.

    1. Risk Manager wants another source of valuation for periodic MSR and loan portfolios to enhance decision making and compare against the marks from their third-party valuation firm. 

    1. Private Credit Risk Manager needs a built-for-purpose private credit analytics system to properly run risk metrics. Users can run separate and run ad hoc analysis on these holdings.

For more specific information about how RiskSpan will structure pricing with various commitment levels, click below to tell us about your needs, and a representative will be in touch with you shortly. 


What Do 2023 Origination Trends Mean for MSRs?

When it comes to forecasting MSR performance and valuations, much is made of the interest rate environment, and rightly so. But other loan characteristics also play a role, particularly when it comes to predicting involuntary prepayments.

So let’s take a look at what 2023 mortgage originations might be telling us.

Average credit scores, which were markedly higher than normal during the pandemic years, have returned during the first part of 2023 to averages observed during the latter half of the 2010s.

FICO

The most credible explanation for this most recent reversion to the mean is the fact that the Covid years were accompanied by an historically strong refinance market. Refis traditionally have higher FICO scores than purchase mortgages, and this is apparent in the recent trend.

Purchase markets are also associated with higher average LTV ratios than are refi markets, which accounts for their sharp rise during the same period

LTV

Consequently, in 2023, with high home prices persisting despite extremely high interest rates, new first-time homebuyers with good credit continue to be approved for loans, but with higher LTV and DTI ratios.

DTI

Between rates and home prices,​​borrowers simply need to borrow more now than they would have just a few years ago to buy a comparable house. This is reflected not just in the average DTI and LTV, but also the average loan size (below) which, unsurprisingly, is trending higher as well.

Recent large increases to the conforming loan limit are clearly also contributing to the higher average loan size.

WOLS

What, then, do these origination trends mean for the MSR market?

The very high rates associated with newer originations clearly translate to higher risk of prepayments. We have seen significant spikes in actual speeds when rates have taken a leg down — even though the loans are still very new. FICO/LTV/DTI trends also potentially portend higher delinquencies down the line, which would negatively impact MSR valuations.

Nevertheless, today’s MSR trading market remains healthy, and demand is starting to catch up with the high supply as more money is being raised and put to work by investors in this space. Supply remains high due to the need for mortgage originators to monetize the value of MSR to balance out the impact from declining originations.

However, the nature of the MSR trade has evolved from the investor’s perspective. When rates were at historic lows for an extended period, the MSR trade was relatively straightforward as there was a broader secular rate play in motion. Now, however, bidders are scrutinizing available deals more closely — evaluating how speeds may differ from historical trends or from what the models would typically forecast.

These more granular reviews are necessarily beginning to focus on how much lower today’s already very low turnover speeds can actually go and the extent of lock-in effects for out-of-the-money loans at differing levels of negative refi incentive. Investors’ differing views on prepays across various pools in the market will often be the determining factor on who wins the bid.

Investor preference may also be driven by the diversity of an investor’s other holdings. Some investors are looking for steady yield on low-WAC MSRs that have very small prepayment risk while other investors are seeking the higher negative convexity risk of higher-WAC MSRs — for example, if their broader portfolio has very limited negative convexity risk.

In sum, investors have remained patient and selective — seeking opportunities that best fit their needs and preferences.

So what else do MSR holders need to focus on that may may impact MSR valuations going forward? 

The impact from changes in HPI is one key area of focus.

While year-over-year HPI remains positive nationally, servicers and other investors really need to look at housing values region by region. The real risk comes in the tails of local home price moves that are often divorced from national trends. 

For example, HPIs in Phoenix, Austin, and Boise (to name three particularly volatile MSAs) behaved quite differently from the nation as a whole as HPIs in these three areas in particular first got a boost from mass in-migration during the pandemic and have since come down to earth.

Geographic concentrations within MSR books will be a key driver of credit events. To that end, we are seeing clients beginning to examine their portfolio concentration as granularly as zipcode level. 

Declining home values will impact most MSR valuation models in two offsetting ways: slower refi speeds will result in higher MSR values, while the increase in defaults will push MSRs back downward. Of these two factors, the slower speeds typically take precedence. In today’s environment of slow speeds driven primarily by turnover, however, lower home prices are going to blunt the impact of speeds, leaving MSR values more exposed to the impact of higher defaults.


Edge: Zombie Banks

At the market highs, banks gorged themselves on assets, lending and loading their balance sheets in an era of cheap money and robust valuations. As asset prices drop, these same companies find their balance sheets functionally impaired and in some cases insolvent. They are able to stay alive with substantial help from the central bank but require ongoing support. This support and an unhealthy balance sheet preclude them from fulfilling their role in the economy.

We are describing, of course, the situation in Japan in the late 1980s and early 1990s, when banks lent freely, and companies purchased both real estate and equity at the market highs. When the central bank tightened monetary policy and the stock market tanked, many firms became distressed and had to rely on support from the central bank to stay afloat. But with sclerotic balance sheets, they were unable to thrive, leading to the “lost decade” (or two or three) of anemic growth.

While there are substantial parallels between the U.S. today and Japan of three decades ago, there are differences as well. Firstly, the U.S. has a dynamic non-bank sector that can fill typical roles of lending and financial intermediation. And second, much of the bank impairment comes from Agency MBS, which slowly, but surely, will prepay and relieve pressure on their HTM assets.

Chart
Source: The Wall Street Journal

How fast will these passthroughs pay off? It will vary greatly from bank to bank and depends on their mix of passthroughs and their loan rates relative to current market rates, what MBS traders call “refi incentive” or “moniness.” It is helpful to remember that incentive also matters to housing turnover, which is a form of mortgage prepayment. For example, a borrower with a note rate that is 100bp below prevailing rates is much more likely to move to a new house than a borrower with a note rate that is 200bp out of the money, a trait that mortgage practitioners call “lock-in”.

Chart
Source: RiskSpan’s Edge Platform

As a proxy for the aggregate bank’s balance sheet, we look at the universe of conventional and GNMA passthroughs and remove the MBS held by the Federal Reserve.

1

The Fed’s most substantial purchases flowed from their balance sheet expansion during COVID, when mortgage rates were at all-time lows. Consequently, the Fed owns a skew of the MBS market. Two-thirds of the Fed’s position of 30yr MBS have a note rate of 3.25% or lower. In contrast, the market ex Fed has just under 50% of the same note rates.

Chart
Source: RiskSpan’s Edge Platform

From here, we can estimate prepayments on the remaining universe. Prepay estimates from dealers and analytics providers like RiskSpan vary, but generally fall in the 4 to 6 CPR range for out-of-the-money coupons. This, coupled with scheduled principal amortization of roughly 2-3% per annum means that for this level in rates, runoff in HTM MBS should occur around 8% per annum — slow, but not zero. After five years, approximately 1/3 of the MBS should pay off. Naturally, the pace of runoff can change as both mortgage rates and home sales change.

While the current crisis contains echoes of the Japanese zombie bank crisis of the 1990s, there are notable differences. U.S. banks may be hamstrung over the next few years, with reduced capacity to make new loans as MBS in their HTM balance sheets run off over the next few years. But they will run off — slowly but surely.


Edge Platform Adds Fannie and Freddie Social Index Data

ARLINGTON, Va., January 18, 2023 — RiskSpan, a leading technology company and the most comprehensive source for data management and analytics for residential mortgage and structured products, has announced the incorporation of Fannie Mae’s and Freddie Mac’s Single-Family Social Index data into its award-winning Edge Platform.

Fannie and Freddie rolled out their social index disclosures in November 2022. Consisting of two measures, the Social Criteria Score and the Social Density Score, the social index discloses the share of loans in a given pool that are made to low-income, minority, and first-time homebuyers, as well as mortgages on homes in low-income areas, minority tracts, high-needs rural areas, and designated disaster areas. Manufactured housing loans also contribute to the score.

Rather than classifying each individual bond as “social” or “not social,” the new Agency data available on the Edge Platform assigns every pool two fully transparent scores – one indicating the percentage of loans in a pool that satisfy any of the defined social criteria, the other reflecting how many criteria a pool’s average loan satisfies.

Taken together, these enable Agency traders and investors to view and understand each pool along a full continuum of the social index, as opposed to simply assigning a binary social designation. Because borrowers behave differently at various places along this continuum, traders and investors fine-tune their analytics in ways never before possible to isolate pools with potentially slower prepayment speeds in a way that transcends what has traditionally been available using so-called “spec. pool” stories alone.

Comprehensive details of this and other new capabilities are available by requesting a no-obligation live demo at riskspan.com.

This new functionality is the latest in a series of enhancements that further the Edge Platform’s objective of providing frictionless insight to Agency MBS traders and investors, knocking down barriers to efficient, clear and data-driven valuation and risk assessment.

### 

About RiskSpan, Inc. 

RiskSpan offers cloud-native SaaS analytics for on-demand market risk, credit risk, pricing and trading. With our data science experts and technologists, we are the leader in data as a service and end-to-end solutions for loan-level data management and analytics.

Our mission is to be the most trusted and comprehensive source of data and analytics for loans and structured finance investments.

Rethink loan and structured finance data. Rethink your analytics. Learn more at www.riskspan.com.

Get a Demo

About RiskSpan, Inc. 

RiskSpan offers cloud-native SaaS analytics for on-demand market risk, credit risk, pricing and trading. With our data science experts and technologists, we are the leader in data as a service and end-to-end solutions for loan-level data management and analytics. 

Our mission is to be the most trusted and comprehensive source of data and analytics for loans and structured finance investments. 

Rethink loan and structured finance data. Rethink your analytics. Learn more at www.riskspan.com. 

Media contact: Timothy Willis 


HECM Loan Data, Smart Assumptions, and Cross-Sector Trade Impact Headline New Edge Platform Functionality

ARLINGTON, Va., December 8, 2022RiskSpan, a leading technology company and the most comprehensive source for data management and analytics for residential mortgage and structured products, has announced a flurry of new functionality on its award-winning Edge Platform.

GNMA HECM Datasets and Involuntary Prepayment Breakdown: The GNMA HECM dataset is now available to subscribers in Edge’s Historical Performance module, allowing market participants to find performance differentials within FHA reverse mortgage data. As with conventional datasets available on Edge, users slice and dice by any loan attribute to create S-curves, aging curves, time series and other decision-useful analytics.

Edge users also can now parse GNMA buyout metrics by reason, based on whether individual loans were in delinquency, loss mitigation, or foreclosure when they were removed from the security.

Smart Assumptions: Rather than relying on static assumptions to back-fill missing credit scores, DTIs, LTVs and other data on loan acquisition tapes, the Edge Platform has begun employing a smart, dynamic approach to creating more educated estimates of missing assumptions based on other loan characteristics. Users have the option of accepting these assumptions or substituting their own.

Cross-Sector Trade Impact: As a provider of loan and securities analytics, RiskSpan is making it easier to forecast the combined performance of loan and securities portfolios together in a single view. This allows traders and analysts tools to evaluate the risk and return impact of not only different loan selections or bond selections but also cross-sector reallocation.

These new enhancements all further the Edge Platform’s purpose of providing frictionless insight, knocking down barriers to efficient, clear and data-driven valuation and risk assessment.

Comprehensive details of this and other new capabilities are available by requesting a no-obligation live demo at riskspan.com.

This new functionality is the latest in a series of enhancements that is making the Edge Platform increasingly indispensable for Agency MBS traders and investors.

Get a Demo

About RiskSpan, Inc. 

RiskSpan offers cloud-native SaaS analytics for on-demand market risk, credit risk, pricing and trading. With our data science experts and technologists, we are the leader in data as a service and end-to-end solutions for loan-level data management and analytics. 

Our mission is to be the most trusted and comprehensive source of data and analytics for loans and structured finance investments. 

Rethink loan and structured finance data. Rethink your analytics. Learn more at www.riskspan.com. 

Media contact: Timothy Willis 


RiskSpan Wins Risk as a Service Category for Third Consecutive Year, Rises 6 Places in RiskTech100® 2023 Ranking

ARLINGTON, Va., December 6, 2022RiskSpan’s Edge Platform, the only single solution to include data management, models, and analytics on fully scalable, cloud-native architecture, wins “Risk as a Service” category for a third consecutive year in Chartis Research’s vaunted RiskTech100® ranking of the world’s 100 top risk technology companies.

RiskSpan was also called out as a most significant mover, climbing 6 places in the overall ranking and improving its position for the fourth year in a row.

Chartis_RiskTech100 “RiskSpan’s strong innovation in data management helped drive its six-place rise in the rankings this year,’ said Sid Dash, Research Director at Chartis. ‘The company has won the RaaS award for three consecutive years, reflecting its tech-centric and pragmatic approach in a key area of the risk management space.” 

Licensed by some of the largest asset managers, broker/dealers, hedge funds, mortgage REITs and insurance companies in the U.S., the Edge Platform is a fully managed risk solution across all asset classes with specialization in residential mortgage and structured products.  

 This year’s award reflects the Edge Platform’s unique ability to help users find alpha, execute transactions with ease, and effectively manage portfolio risks,” noted Bernadette Kogler, RiskSpan’s co-founder and CEO. It is satisfying to be recognized for our continued efforts to help clients transform their business with modern workflows and operations to optimize productivity, cost, and resilience.” 

CONTACT US

About RiskSpan, Inc.  

RiskSpan offers cloud-native SaaS analytics for on-demand market risk, credit risk, pricing and trading. With our data science experts and technologists, we are the leader in data as a service and end-to-end solutions for loan-level data management and analytics. 

Our mission is to be the most trusted and comprehensive source of data and analytics for loans and structured finance investments. 

Rethink loan and structured finance data. Rethink your analytics. Learn more at www.riskspan.com. 

 About Chartis Research:  

Chartis Research is the leading provider of research and analysis on the global market for risk technology. It is part of Infopro Digital, which owns market-leading brands such as Risk and WatersTechnology. Chartis’ goal is to support enterprises as they drive business performance through improved risk management, corporate governance and compliance, and to help clients make informed technology and business decisions by providing in-depth analysis and actionable advice on virtually all aspects of risk technology.  

 Media contact:  Timothy Willis 


Incorporating Covid-Era Mortgage Data Without Skewing Your Models

What we observed during Covid represents a radical departure from what we observed pre-Covid. To what extent do these observations impact long-term trends observed for mortgage performance? Should these data fundamentally impact the way in which we think about the effects borrower, loan and macroeconomic characteristics have on mortgage performance? Or do we need to simply account for them as a short-term blip?


The process of modeling mortgage defaults and prepayments typically begins with identifying long-term trends and reference values. These aid in creating the baseline forecasts that undergird the model in its most simplistic form. Modelers then begin looking for deviations from this baseline created by specific loan, borrower, and property characteristics, as well as by key macroeconomic variables.

Identifying these relationships enables modelers to begin quantifying the extent to which micro factors like income, credit score, and loan-to-value ratios interact with macro indicators like the unemployment rate to cause prepayments and defaults to depart from their baseline. Data observations aggregated over extended periods give a comprehensive picture possible of these relationships.

In practice, the human behavior underlying these and virtually all economic models tends to change over time. Modelers account for this by making short-term corrections based on observations from the most recent time periods. This approach of tweaking long-term trends based on recent performance works reasonably well under most circumstances. One could reasonably argue, however, that tweaking existing models using performance data collected during the Covid-19 era presents a unique set of challenges.

What was observed during Covid represents a radical departure from what was observed pre-Covid. To what extent do these observations impact long-term trends and reference values. Should these data fundamentally impact the way in which we think about the effects borrower, loan and macroeconomic characteristics have on mortgage performance? Or do we need to simply account for them as a short-term blip?

SPEAK TO AN EXPERT

How Covid-era mortgage data differs

When it comes to modeling mortgage performance, we generally think of three sets of factors: 1) macroeconomic conditions, 2) loan and borrower characteristics, and 3) property characteristics. In determining how to account for Covid-era data in our modeling, we first must attempt to evaluate its impact on these factors. Three macroeconomic factors have played an especially significant role recently. First, as reflected in the chart below, we experienced a significant home-price decline during the 2008 financial crisis but a steady increase since then. Covid Era

Second, mortgage rates continued to decline for the most part during the crisis and beyond. There were brief periods when they increased, but they remained low by and large. Covid Era

The third piece is the unemployment rate. Unemployment spiked to around 10 percent during the financial crisis and then slowly declined. Covid Era

When home prices declined in the past, we typically saw the government attempt to respond to it by reducing interest rates. This created something of a correlation between home prices and mortgage rates. Looking at this from a purely statistical viewpoint, the only thing the historical data shows is that falling home prices bring about a decline in mortgage rates. (And rising home prices bring about higher interest rates, though to a far lesser degree.) We see something similar with unemployment. Falling unemployment is correlated with rising home prices.

But then Covid arrives and with it some things we had not observed previously. All the “known” correlations among these macroeconomic variables broke down. For example, the unemployment rate spikes to 15 percent within just a couple of months and yet has no negative impact at all on home prices. Home prices, in fact, continue to rise, supported by the very generous unemployment benefits provided during Covid pandemic.

This greatly complicates the modeling. Here we had these variable relationships that appeared steady over a period of decades, and all of our modeling was being done (knowingly or unknowingly) relying on these correlations, and suddenly all these correlations are breaking down.

What does this mean for forecasting prepayments? The following chart shows prepayments over time by vintage. We see extremely high prepayment rates between early 2020 (the start of the pandemic) and early 2022 (when rates started rising). This makes sense.

Covid Era

Look at what happens to our forecasts, however, when rates begin to increase. The following chart reflects the models predicting a much steeper drop-off in prepayments than what was actually observed for a July 2021 issuance Fannie Mae major of coupon 2.0. These mortgage loans with no refinance incentive are prepaying faster than what would be expected based on the historical data.

Covid Era

What is causing this departure?

The most plausible explanation relates to an observed increase in cash-out refinances caused by the recent run-up in home prices and resulting in many homeowners suddenly finding themselves with a lot of home equity to tap into.  Pre-Covid , cash-outs accounted for between a third and a quarter of refinances. Now, with virtually no one in the money for a rate-and-term refinance, cash-outs are accounting for over 80 percent of them.

We learn from this that we need to incorporate the amount of home equity gained by borrowers into our prepayment modeling.

 Modeling Credit Performance

Of course, Covid’s impacts were felt even more acutely in delinquency rates than in prepays. As the following chart shows, a borrower that was 1-month delinquent during Covid had a 75 percent probability of being 2-months delinquent the following month.

Covid Era

This is clearly way outside the norm of what was observed historically and compels us to ask some hard questions when attempting to fit a model to this data.

The long-term average of “two to worse” transitions (the percentage of 60-day delinquencies that become 90-day delinquencies (or worse) the following month) is around 40 percent. But we’re now observing something closer to 50 percent. Do we expect this to continue in the future, or do we expect it to revert back to the longer-term average. We observe a similar issue in other transitions, as illustrated below. The rates appear to be stabilizing at higher levels now relative to where they were pre-Covid. This is especially true of more serious delinquencies.

Covid Era

How do we respond to this? What is the best way to go about combining this pre-Covid and post-Covid data?

Principles for handling Covid-era mortgage data

One approach would be to think about Covid data as outliers that should be ignored. At the other extreme, we could simply accept the observed data and incorporate it without any special considerations. A split-the-difference third approach would have us incorporate the new data with some sort of weighting factor for use in future stress scenarios without completely casting aside the long-term reference values that had stood the test of time prior to the pandemic.

This third approach requires us to apply the following guiding principles:

  1. Assess assumed correlations between driving macro variables: For example, don’t allow the model to assume that increasing unemployment will lead to higher home prices just because it happened once during a pandemic.
  2. Choose short-term calibrations carefully. Do not allow models to be unduly influenced by blindly giving too much weight to what has happened in the past two years.
  3. Determine whether the new data in fact reflects a regime shift. How long will the new regime last?
  4. Avoid creating a model that will break down during future unusual periods.
  1. Prepare for other extremes. Incorporate what was learned into future stress testing
  1. Build models that allow sensitivity analyses and are easy to change/tune. Models need to be sufficiently flexible that they can be tuned in response to macroeconomic events in a matter of weeks, rather than taking months or years to design and build an entirely new model.

Covid-era mortgage data presents modelers with a unique challenge. How to appropriately consider it without overweighting it. These general guidelines are a good place to start. For ideas specific to your portfolio, contact a RiskSpan representative.

SPEAK TO AN EXPERT


RiskSpan Unveils New “Reverse ETL” Mortgage Data Mapping and Extract Functionality

ARLINGTON, Va., October 19, 2022 – Subscribers to RiskSpan’s Mortgage Data Management product can now not only leverage machine learning to streamline the intake of loan data from any format, but also define any target format for data extraction and sharing.

A recent enhancement to RiskSpan’s award-winning Edge Platform enables users to take in unformatted datasets from mortgage servicers, sellers and other counterparties and convert them into their preferred data format on the fly for sharing with accounting, client, and other downstream systems.

Analysts, traders, and portfolio managers have long used Edge to take in and store datasets, enabling them to analyze historical performance of custom cohorts using limitless combinations of mortgage loan characteristics and run predictive analytics on segments defined on the fly. With Edge’s novel “Reverse ETL” data extract functionality, these Platform users can now also easily and fully design an export format for exporting their data, creating the functional equivalent of a full integration node for sharing data with literally any system on or off the Edge Platform.   

Market participants tout the revolutionary technology as the end of having to share cumbersome and unformatted CSV files with counterparties. Now, the same smart mapping technology that for years has facilitated the ingestion of mortgage data onto the Edge Platform makes extracting and sharing mortgage data with downstream users just as easy.   

Comprehensive details of this and other new capabilities using RiskSpan’s Edge Platform are available by requesting a no-obligation live demo at riskspan.com.

SCHEDULE A FREE DEMO

This new functionality is the latest in a series of enhancements that is making the Edge Platform’s Data as a Service increasingly indispensable for mortgage loan and MSR traders and investors.

### 

About RiskSpan, Inc. 

RiskSpan is a leading technology company and the most comprehensive source for data management and analytics for residential mortgage and structured products. The company offers cloud-native SaaS analytics for on-demand market risk, credit risk, pricing and trading. With our data science experts and technologists, we are the leader in data as a service and end-to-end solutions for loan-level data management and analytics.

Our mission is to be the most trusted and comprehensive source of data and analytics for loans and structured finance investments.

Rethink loan and structured finance data. Rethink your analytics. Learn more at www.riskspan.com.

Media contact: Timothy Willis

CONTACT US


Get Started
Log in

Linkedin   

risktech2024