|
|
Lisa Malie , VP
Capital Markets Group
|
It’s hard to believe that just a few short years ago, much of the market approached the valuation of mortgage-backed financial instruments with a bit of a cavalier attitude. analysts had a set of truths they lived by regarding foreclosure rates, home prices, whole loan spreads, ability to refinance, and new origination programs. These truths formed the basis for the baseline assumptions used by analysts to value all different types of mortgage-backed financial instruments and while assumptions for a specific portfolio may vary from the standard based on portfolio-specific collateral characteristics, that variation often fell within a narrow band.
In today’s turbulent economic times, these “old” truths, not to mention the values they would produce, would seem almost laughable – that is if today’s economic truths didn’t have such extreme economic and personal consequences. The “new” truths are centered around a much more detailed set of collateral characteristics and a changing set of behaviors resulting from those characteristics. In today’s market, these “new” truths are as varied as the collateral characteristics upon which they are based.
At MIAC Analytics™ (MIAC), we have always believed that identifying and analyzing the detailed collateral characteristics of a portfolio to be a critical component in producing accurate, defendable valuation results. We use a variety of sources to create collateral behavior assumptions that reflect true market conditions including extensive discussions with our clients and other owners of mortgage-backed collateral, exhaustive review of Wall Street and other sources of published research, historical analysis, active interaction with buyers and sellers in the marketplace, and experienced managers and analysts who have practical experience in the mortgage market.
As is evidenced by our detailed approach to establishing collateral behavior assumptions, MIAC firmly believes in providing tools to our clients to access information from a variety of sources in the market. Through our Open Box Technology™, we give our clients the opportunity to interface with a variety of data and collateral assumption providers to allow them the opportunity to develop their own “truths” about how their portfolios, or portfolios they are evaluating for potential purchase, will perform in the current market environment.
Some of the key collateral attributes that impact collateral behavior assumptions include credit score and loan to value ratio. When evaluating portfolio risk, it is critical to understand borrowers’ current credit profiles as well as their equity position as these characteristics have a significant impact on projected losses and prepayment behavior. This data becomes stale very quickly as market and borrower circumstances change, particularly in the current market environment, so having an integrated solution for updating this data is important.
MIAC’s recent integration with Kroll Factual Data will give clients the ability to obtain updated borrower credit scores, property appraisal information, and other property-level credit risk attributes which could have a significant impact on their asset valuation. Not only will this data be available to the analysts in reviewing the collateral characteristics of their portfolios, it can be fed into other third-party prepayment and credit models to enhance the accuracy of the results of these models.
For example, MIAC and Standard & Poor’s (S&P) have agreed to integrate S&P’s LEVELS®, one of the most recognized mortgage risk models in the industry, with MIAC’s industry leading mortgage valuation and risk management solutions. MIAC’s DataRaptor™ data management tool allows users to pass standardized, validated loan-level information through the LEVELS engine. Inputs to LEVELS include such collateral characteristics as credit score, loan to value, property type, loan purpose, etc. Much of this information can be updated and/or validated through Kroll, which will enhance the validity of the LEVELS results.
In turn, LEVELS will pass back detailed, loan-level foreclosure frequency and loss expectations for each loan at a variety of credit levels. This information can then be passed to WinOAS™, which is MIAC’s cash flow engine, ultimately driving the cumulative default rate and loss exposure for each asset. This integration will allow our clients to take advantage of the most updated loss behavior assumptions available from S&P in valuing all of their mortgage-backed collateral.
From a valuation perspective, MIAC’s WinOAS™ cash flow engine has been integrated with prepayment models from a variety of sources, including Andrew Davidson & Co. (ADCo) and FIS (formerly Applied Financial Technology) for many years. We have recently upgraded our interface to the latest available versions of the prepayment models provided by these vendors, including ADCo’s Loan Dynamics model.
MIAC has further integrated with credit models from both of these sources, giving our clients a multitude of options for projecting prepayments, both voluntary and involuntary, along with loss severities. In many cases, these models take advantage of enhanced collateral characteristics, including credit score, loan to value, etc., in generating collateral behavior assumptions. Taking advantage of these enhanced data features in these models will give users the ability to evaluate the impact of leverage on borrowers’ ability to refinance, for example, even in a declining rate environment. Having the best information possible, from a source such as Kroll Factual Data, would produce the most accurate results from these models.
In the end, MIAC’s goal is give our clients access to a variety of collateral behavior models as possible, through a highly transparent framework. Clients can independently assess the results from these models and develop their own “truths” about the future performance of their portfolio. Given the transparency around the inputs and outputs to and from these integrated sources, clients have the ability to produce reliable, defendable valuation results for all of their mortgage-backed assets.