Advancing the MSR businesses into the future
Do you remember the car from Back to the Future movie? Aesthetically, the DMC DeLorean was the perfect centerpiece for one of the most successful franchises in Hollywood history. For true car fanatics, however, this was a bit of a miss.
Despite its stunning lines and overall “freshness”, the DeLorean was massively overweight for a sports car. It also housed a modest 130-horsepower V6 engine capable of doing 0-60 in rather pedestrianized 9.5 seconds. In fact, his top speed was only 110 mph, barely faster than the 88 mph Marty McFly needed to get back to 1955.
The mortgage industry is a lot like the DeLorean these days. Much of the technological sophistication of recent years has taken place upstream of transactions and in electronic closing technologies, both of which are very visible in the borrower’s experience.
If you lift the hood, however, much of our industry is still fueled by technology that was built over a decade ago, which creates downstream impacts in the aftermarket, especially when it comes to trading. mortgage management rights (MSR). Fortunately, however, that is starting to change.
Currently, initiators selling service rights have extraordinary difficulty in meeting the multiplicity of buyer delivery requirements. As a result, buyers often have to enter information and process documents manually when transferring loans into their management systems or delivering loan files to sub-contractors. This creates additional costs and delays, which can ultimately make it difficult to comply with the required timely communications to borrowers regarding who is managing their loan.
The source of friction is the lack of a standard process for exchanging MSRs. Documents and data are received from sellers in so many different ways that buyers cannot cope with the incongruities. For example, document images are classified using different taxonomies, stacking orders are not consistent, and image quality varies, so no two loans are identical in terms of structure. or readability. Buyers receive loan packages from different sellers who look completely different from each other. All of this means buyers have to manually go through each loan and re-stack documents in any order they want and determine what is missing from individual loan records.
There are also nuances that lead to data discrepancies in the loan files which result in additional costs. These can go unnoticed when the work is performed by staff whose expertise is insufficient to identify these problems. For example, for a loan with co-borrowers, the fact that a credit report exists is not enough. Is this credit report joint or unique? Do you have a loan application for each borrower? For any loan, is the content composed of a hyphen consistent throughout the loan file?
Correcting these inconsistencies between documents and data files using manual procedures is rife with human error. Rules-based technology can capture these nuances much more effectively. Instead of reviewing all loan records for the few loans that have deviations, staff can focus on clearing terms for the loans that have problems.
The lack of standardization in the MSR trade also limits business opportunities. Banks, lenders, and investors who buy MSR want to transact quickly and get back into the market and do more, but every time they buy loans they are given a big task of due diligence and paperwork. , which means they can only do so many trades. Banks are particularly crippled because they have particular compliance burdens that make it more expensive and time consuming to check loans.
There are MSR trading platforms that make it easy to trade the deal, of course. But loan delivery has not improved significantly in years. Sellers struggle to meet buyers’ demands, which means buyers still have to process documents and resolve discrepancies through a combination of manual labor and piecemeal automation that still isn’t robust enough.
Recently, however, a leap forward has taken place in the area of MSR transfers that could eliminate friction for all parties, including the consumer.
Take the plunge
Over the past couple of years, some parts have come into place that will help streamline MSR trades. The Industry Loan Application Dataset (ILAD), a loan application “super-set” based on MISMO v3.4, can make it easier for buyers and sellers to exchange loan information digitally. Until recently, however, the biggest piece of the puzzle – the development of technology that could standardize and standardize both documents and MSR datasets regardless of the original date of the loan – was missing. Now that piece has arrived.
By applying automated and repeatable processes, new technologies streamline MSR transactions by standardizing the naming and stacking of documents, independent of individual loan vendor procedures or protocols. Technologies today can extract data from hundreds of data fields on dozens of the most common loan documents, allowing buyers and sellers to identify and resolve data inconsistencies faster.
Key to these technologies is Automated Document Recognition (ADR) technology based on machine learning and Automated Data Extraction (ADE) tools. Together, they can perform multiple automated tasks and apply automated rules for any type of document verification, document-to-document comparison, and document-to-data comparison. The configurations can then be used to define a buyer-specific naming convention and stacking order for integration into a service system. It’s significantly faster than sending large files to a foreign partner and waiting days for the files to come back.
Essentially, these technologies consolidate the many manual steps of MSR transfers into highly automated steps, which are both configurable and universally applicable to all loan cases. They are also much more powerful than standard optical character recognition (OCR) tools. For example, on loans involving two borrowers, most current OCR systems often fail to detect whether there are credit reports for both borrowers or whether all pages of the credit report are on file. New technologies, however, can.
These technologies can be implemented with extraordinary speed, often in as little as two days. They also allow sellers to drop and drag loan files to a secure online portal used by both buyers and sellers. If there is something wrong with one or more loans, such as a missing document, both parties can see the errors and correct them through a communication platform.
Even if a buyer outsources their due diligence to a third party vendor, that third party can log into the portal and perform the required work in a secure web environment. Additionally, using APIs, data and documents can be easily integrated into a buyer’s service system.
This “leap” in MSR trading innovation has already been achieved with great success by Freddie Mac thanks to his Freddie Automated Servicing Transfer (FASTSM). The technology behind this tool has dramatically streamlined MSR transactions, saving time and reducing costs by standardizing and simplifying the exchange of documents and data during service transfers. FAST uses the same kind of document processing automation and machine learning data extraction technologies mentioned earlier, which can create a verified and validated data stream from digital images and scanned documents.
The real price: lower costs
Speeding up MSR transactions by applying automation to normalize and standardize the process creates enormous potential for cost savings, so much so that it could catch buyers off guard. If buyers can automate 80% of the tasks related to processing documents and identifying and resolving data discrepancies, they only need to apply their full-time employees to the 20% that are exceptions with conditions that must be cleared.
This means that they can drastically reduce personnel costs or simply deploy their teams to other tasks. For many buyers, this could also be an opportunity to reduce the army of business process outsourcing (BPO) staff needed to work on important trades and the associated FTE expense. By automating MSR transfer tasks, buyers can end up with adequate internal staff to perform the due diligence work themselves.
There are also other advantages. Since purchased MSRs require fewer manual keys, there is less chance of errors. By reducing the time it takes to complete transfers and absorb loans into their management system, buyers can move on to the next MSR acquisition much faster. This is good news, as we have seen an increase in transactions over the past 60-90 days.
These benefits seem to be the reason why the number of buyers who automate MSR transfer work has skyrocketed over the past year. In fact, our own platform, LoanLogics IDEA, offers 95% coverage of top co-show buyers for MSRs. Last year alone, we also stepped up our support for the bulk market, helping over 50% of these large buyers.
When these technologies are combined with data standards, they can revolutionize secondary and service markets. By standardizing MSR negotiation tasks, the bottlenecks created by the large disparities in the delivery of loan files simply disappear.
Currently, the only obstacle that remains is the will to change. In fact, for all of its flaws, there are still plenty of people who prefer the DeLorean, even though today’s hybrid cars would leave it in the dust. But for most drivers, and for most mortgage participants, what really matters is what’s under the hood.