The entire mortgage industry is at a crossroads. As the market traverses through the chaos caused by the housing industry collapse, companies are faced with huge inventories of defaulted loans, new mandates from the Dodd-Frank Act, and real pressure to reduce costs.
This, combined with multiple mergers and acquisitions (M&As), has resulted in fragmented data and software platforms that cannot keep up with the changes and demands in this new world order. Throughout all of this, the Mortgage Industry Standards Maintenance Organization (MISMO) has been working to provide key data standards for the entire mortgage process. Moving to a MISMO-based infrastructure may require an up-front investment today, but in the intermediate and long run, making the transition will result in the ability to be more agile and manage costs more effectively.
MISMO is an all-volunteer organization, organized as working committees of industry experts from across the country, that promotes data consistency and transparency through standardization in the single-family, commercial and multifamily real estate finance sectors. Over the years, MISMO has created standards for underwriting, mortgage insurance application, credit reporting, flood and title insurance, property appraisal, loan delivery, product and pricing, loan servicing, and secondary mortgage market investor reporting.
MISMO provides regular updates to its XML-based data architecture and data dictionary and is currently on Version 3.0. With this latest update, all of these standards have been combined into one comprehensive specification, providing a single standard that defines and organizes all data required in the mortgage process.
As beneficial as MISMO can be for many companies, there has been little reason historically to convert. Converting to MISMO requires an up-front investment that can be fairly small for community and regional lenders or sizable (i.e., millions of dollars) for national lenders. With all of the pressures to reduce spending, converting to MISMO has been a tough business case to make, especially when originations are down and defaults are up. However, in spite of the market turmoil and the political and regulatory uncertainties that accompany it, now may be the best time to consider investing in a MISMO-oriented architecture.
It is inevitable that everyone doing business with Freddie Mac or Fannie Mae will eventually need to create MISMO-compliant interfaces to these entities just to conduct business with them. The Federal Housing Finance Agency has orchestrated cooperation among the players that is resulting in standardized data sets and formats for appraisal and delivery.
For example, MISMO Version 3.0 serves as the basis for the government-sponsored enterprises' (GSEs) Uniform Loan Delivery Dataset, while the MISMO Property Valuation Response Version 2.6 Schema Errata 1 was the foundation for the GSEs' Uniform Appraisal Dataset.
The Uniform Mortgage Data Program (UMDP) is here to stay. In a Sept. 9 document titled ‘On the Right Track with UMDP,’ Freddie Mac illustrated all of the specific milestones related to the UMDP rollout and required compliance by vendors and sellers. Fannie Mae provides a similar information set on its monthly ‘UMDP Yardstick’ document, which enables companies to gauge the success of their UMDP implementation efforts.
Due to the standardization of data sets across the industry, the major service and information providers will be forced to move to these new standards in order to remain competitive. Although backwards compatibility will likely be offered for a time, it will eventually become a cost burden for providers to maintain the legacy standards. This is also true of software firms offering loan origination, underwriting, servicing and secondary marketing tools.
Expect to see additional data sets announced as a result of the Servicer Alignment Initiative and other coordinated efforts by the GSEs.
For small companies that are in a spreadsheet environment or that use a combination of end-user and off-the-shelf servicing packages, a MISMO-oriented approach will allow for rapid change coupled with standardization required for controls. Nearly all off-the-shelf packages will move toward a MISMO-compliant standard, meaning servicers will not need to be bound to one software or another due to the high costs of switching systems. Creating spreadsheets, uploads and reports based on MISMO will allow for the use of new packages or multiple packages without the need for shops to constantly create end-user applications from scratch.
For bigger companies with large databases, warehouses and proprietary models, the move to MISMO may be quite expensive. It requires a concerted effort, time and money to create a MISMO-oriented architecture, but the payoffs can be significant.
For the most part, companies have ignored the need to integrate purchase and servicing platforms as a result of M&A activities. Many companies are running multiple versions of the same servicing platform across the country, but because they were implemented in a proprietary manner, the data sets are not consistent and the platforms cannot be integrated without significant costs. In some instances, it is virtually impossible to integrate the data fast enough to provide executive-level information in a timely manner. Making incorrect business or operational decisions due to poor data quality can be devastating from both a cost and controls perspective.
A specific example can be found in the user-defined codes contained in one of the market's leading servicing platforms. Many of the top servicers that use the platform have introduced codes to define data elements or combinations of data elements. When information needs to be extracted from these platforms, the user or programmer must first decompose the codes into the true data elements using substitution or other logic. If they had been in a MISMO environment, each data element would be defined the same and the need to create another layer of logic would be irrelevant. Moving to MISMO takes the complexity out of integration and makes it a data-movement activity, which is much simpler and less risky.
Utilizing MISMO as the backbone of an integration effort means that these companies can eliminate the cost of defining their own proprietary data models – models that will likely be inferior because they replicate existing shortcomings or short-sightedly address a subset (the in-focus hot spots) of what is needed. Additionally, a MISMO-based approach will position these companies for easier integration of commercially available tools and platforms, as these platforms themselves will be moving to MISMO. Similarly, a MISMO-based approach will provide for seamless interfacing across the mortgage processing chain, from originators to subservicers to data providers to secondary market participants.
In the haste to shore up internal operations, some companies are putting a layer of data in place between their servicing platforms and their reporting and analytics tools. Though it may provide some short-term efficiency and increase controls, it is only putting off the inevitable need to perform large-scale corporate data integration and a system-rationalization effort.
Ultimately, the move to MISMO will become a competitive advantage: The companies that move to MISMO first will have a huge head start on the rest of the field.
Having a nimble data set will allow the rapid launch of new products and will provide the agility to react to regulatory changes and mandates. Companies with a flexible data set will quickly be able to make changes to interfaces, models and reports. Companies with fragile infrastructure will need to react under pressure by throwing bodies at the problem and may run the risk of noncompliance – at the very least.
Matthew J. Seu is a principal and Tim McLuckie is a managing director at McLean, Va.-headquartered Actualize Consulting. Prior to Actualize, Seu served as a vice president for Freddie Mac, where he headed the corporate enterprise data program, and McLuckie worked for nearly 15 years at Fannie Mae, most recently as a vice president. Seu can be contacted at firstname.lastname@example.org, and McLuckie at email@example.com.