It’s the time of year for resolutions to be made, but as we all know, these often don’t stick. The best intentions, backed by poor planning and ability to execute, often fail. This is so often the case for mortgage loan quality initiatives because of the effort required and seeming lack of ROI. This doesn’t have to be the case if you follow these simple quality workout steps:
1. Don’t skip the pre-workout prep
2. Use the right equipment
3. Know where to apply a more intense workout for better results
4. Make workouts a life-long commitment
Don’t skip the pre-workout prep
You wouldn’t start a workout without stretching or properly fueling the body. Then why do we begin a loan file audit review without prepping the loan file data and documentation? If you’re taking short cuts and relying on only a subset of the loan file documentation, or worse yet, using only LOS data, your loan quality processes are likely to be met with cramps and fatigue.
Many lenders rely on OCR (optical character recognition) technology that can recognize a document based on key phrases and anchor terms, but is not a foolproof approach. For example, OCR software cannot easily or effectively identify boundaries between documents, nor accurately index and classify multiple versions of the same document. The appeal of OCR as a "set it up once and forget it" technology can doom your quality management strategy.
A better way to create a foundation of granular data and loan file transparency is a multi-tier ability to verify and re-verify data — delivering near 100% accuracy rates. Technology and workflows must address both structured and unstructured documents in a loan file that is often 450 pages or more, containing dozens of documents and forms. While structured documents can be identified through common terms, nonstandard and unstructured documents, like gift letters, have inconsistent wording and no common structure. They require different workflow processing and technology that can identify them, flag them with lower confidence score rates and forward them to skilled inspectors for evaluation.
Once all documents are indexed and classified, data can then be validated and compared side by side against the same data element captured from multiple documents in the loan file. This allows for quick and efficient inspection of key data elements and the ability to determine problems in the file.
Use the right equipment
The right equipment can enhance any workout and, more importantly, improve results. Just like the right running shoe can minimize strain and provide support so the runner can go the distance, the right tools for a quality workout can result in accurate defect detection and condition clearing prior to close and before a loan is sold.
Investors have done a great job filling the gaps of credit, compliance and collateral risk by buttoning up their audit check lists. This requires more thorough quality assurance and due diligence practices for both loan originators and correspondent loan aggregators. With validated and verified data, automation of audit rules can eliminate routine manual workflows and focus audit staff on defects and rebuttal procedures.
Beyond audit check-lists, technology-enabled workflow allows all loans to be reviewed through consistently repeatable processes, provides access to relevant data and source documents in the context of an audit step and results in increased loan file review production. A virtual loan file can be electronically audited and reviewed simultaneously across different audit processes, improving efficiency. And, quality analytics and business intelligence integrated into the workflow enables the identification of the actor or process that is contributing to manufacturing defects.
Know where to apply a more intense workout for better results
Being fit from head to toe is the goal, but let’s face it, some parts of the body need a little more work. This is the case for evaluating the quality of mortgage loans, in that some loans are more complex and potentially higher risk than others. A prudent strategy is to review these types of loans in more detail to understand and manage the potential risk they represent.
Some lenders may not want to leave anything to chance and may audit 100% of their loans in pre-close reviews. For others, loan selection for pre-closing quality assurance audits is often accomplished through adverse sampling. With either of these methods however, insight regarding default and repurchase risk associated with loans is often missing.
Predictive analytics and modeling techniques are key in today’s risk management practices and can provide default and repurchase scoring that can help lenders get a better view of risk in the production pipeline. Loans scored for pre-close review enable a more segmented approach to allocate quality management resources across loans with varying degrees of risk. To address GSE requirements, scoring can also be used to identify loans for targeted or discretionary reviews. And, by running all production through scoring, lenders can also use this information to evaluate month to month risk trends in production.
Make workouts a life-long commitment
There is no “done” in a workout commitment, otherwise it has an adverse impact on long-term health and stability. And, similarly, there is no “done” in a life-of-loan quality workout, otherwise expect an unhealthy or less-than-optimal future. Things change over time related to the borrower, the property, the market and a myriad of other factors that can raise and lower risk in your mortgage loan portfolio.
Life-of-loan quality management is an ongoing effort that demands the most current data related to the mortgage loan file. Lenders who leverage enterprise technology and services to maintain accurate, granular data can reap the benefits of a decision-making data set that can be leveraged across functions, streamlining costs.
Bank and credit unions can balance the risk in their mortgage loan portfolios and leverage timely loan-level data and analytics to help calculate loan loss reserves. The predictive dimension of this will become even more critical as depositories brace for Current Expected Credit Losses (CECL). FASB intends to release the final standard Accounting for Financial Instruments – Credit Impairment in the Spring of 2016.
When CECL becomes implemented, institutions will need tools to better measure the risk of default. Technology is enabling the availability of these types of tools for financial institutions of all sizes, empowering them to address CECL in a supportable and meaningful way.
At the end of the day, loan portfolio managers and servicers need to fill in the gaps of what is NOT known about loan performance and analyze borrower credit and collateral health and expected losses. Leveraging an enterprise platform model, statistical data and business intelligence can empower multiple functions with tools for more effective segmentation and comprehensive portfolio evaluation.
Follow these four quality workout steps for success in 2016! Don’t skip the pre-workout prep; get your loan files in shape with verified, validated data. Use the right equipment; take advantage of new advances in technology that enable audit rules automation and structured workflow for consistent process execution. Know where to apply a more intense workout for better results; leverage advanced tools that can give your quality management workout more focus and better results. And finally, make workouts a life-long commitment; life-of-loan quality management is the path to optimize performance for the highest return.
