Counting the Benefits of Loan Engineering
July 15, 2022
When we talk at Candor about the potential benefits to lenders of employing a machine-based approach to loan underwriting, people often assume that the agenda is to eliminate people. In fact, the people engaged in loan underwriting are often among the most valuable and capable individuals in the enterprise, one reason why they are often redeployed into risk management and distressed servicing roles when volumes fall and defaults rise.
Over years of research and development, we’ve discovered that the big ROI win that comes from engineering the loan process is to let people do those tasks that lead to more volume and more revenue, and with lower risk, while letting machines do what they do best, namely count and collate. According to the MBA’s Quarterly Performance Survey for 2021, personnel costs for underwriting account for just 6.4% of average loan production expenses (ALPE), while support and other costs involved in loan underwriting were 21.3% of ALPE.
What these numbers suggest is that making the most valuable people in the enterprise more effective and able to handle more completed loan files per unit of time could result in far greater returns than reducing headcount. Moreover, the ability to scale internal volumes easily, without needing to hire or release additional staff to cope with business changes that result from shifting trends in the economy or interest rates, represents an additional area of potential savings.
Of the more than $10,000 in direct loan production expenses on an average loan in 2022, this according to the MBA Performance Survey, underwriting represents almost a quarter of the expense. Managing this part of the expense ledger, something that historically has not been easy or even practical for many lenders, presents some new opportunities for banks and independent mortgage banks to capture savings while at the same time driving productivity. How does Candor drive productivity? With several key process steps:
Imagine being able to scale your lending operations from the level necessary to handle a 10 million loan year like 2020, but then throttle back to the capacity adequate for a 5 million loan year like that predicted for 2022 without the need for excessive headcount reductions. Having a scalable, flexible system for taking a loan application to loan surety, and do this in a matter of hours rather than days, helps protect valuable team members and drive ROI.
Allowing machines to process the loan through to surety enables to underwriters to review the final product and send the loan to funding and close. Making loan engineering science a core part of your operations means that we can achieve significant cycle time reductions, produce error free and consistent credit risk analysis, and thereby enable decisions that are backed by a comprehensive defect policy.
Not only does the Candor approach to loan underwriting increase productivity and decrease expenses, both absolutely and in terms of the cost volatility per loan, but the output from the system adds an additional level of savings and risk management for lenders. By encapsulating all of the relevant information used in the underwriting process into an encrypted document, Candor provides the lender, the end investor and the guarantor, with a complete inventory of the loan file.
Having a standardized and immutable record of how the loan was underwritten has numerous applications in terms of risk and compliance. First, when the mortgage note is sold into the secondary market, the complete file accompanies the asset. When loan servicing asset is sold, the complete file goes with the servicing asset. And in the event that a guarantor initiates a loan repurchase query, both parties to the sale of the note possess the complete record file that is associated with the loan.
Not only does Candor provide surety with respect to the creation and sale of the mortgage loan, but the way in which the data behind the loan underwriting is created and preserved creates another level of surety in terms of defect, lending bias and tail-risk in terms of potential repurchase claims. More than simply automating much of the loan underwriting process, Candor is creating a new language to capture and convey the key data in a given loan file.
We’ve learned since going live in 2020 is that once we taught a machine how to ensure the integrity of the data and the decision, the mortgage origination process became incredibly fast, shrinking loan origination cycles. Until today, mortgage firms basically have no upside in terms of operating leverage. The industry has spent billions on robotic systems that automate what is still a largely manual process. Loan underwriting is too complex, too variable for such an approach. That’s why we decided to apply the basic tools of behavioral science to underwriting mortgage loans.
Want to learn more? Contact us