Mortgage fraud is likely to increase – here’s how you can prepare for it.
After a record year that saw nearly $ 3.8 trillion in loan origination, industry observers are predicting a smaller, albeit still substantial, market in 2021 with a significantly larger purchase share. The latest Mortgage Bankers Association estimate predicts that volume will decline by around 16% this year to $ 3.2 trillion as the refinancing market cools and purchase volume climbs to a healthy level of 1 , $ 67 trillion.
The pursuit low interest rate the environment, work-from-home trends and mature millennials are all driving shopping demand. This demand, however, is hampered by weak housing inventories and has pushed home prices to new highs: 9% more year-on-year, with average loan amounts hitting a record 402. $ 200 in December.
If the past is predictive, the upcoming shift to a buying market, combined with rising house prices and the incentive for rapid appreciation, can lead to a increase in fraud for housing programs. The question is: will our industry be ready?
Cradled in complacency
In recent years, the mortgage market has shifted strongly towards refinancing. These types of transactions, which involve fewer participants and often occur between a lender and an existing customer, tend to be less susceptible to fraud.
Purchase transactions, on the other hand, have more players –
buyers, sellers, real estate agents, appraisers, loan officers and mortgage brokers, which increases the risk of fraud.
Many new entrants to the mortgage industry – both new fintech lenders and young employees of traditional businesses – have not experienced the mortgage crisis firsthand. All they have seen is the refi-boom of recent years, in which profit fraud and housing fraud have been extremely low. As a result, they may be less susceptible to common fraud schemes and may believe, “We’re doing everything right. Fraud is not a problem, and we have a fraud alert tool in place anyway. This, of course, is exactly when the risk of fraud increases.
Not a rerun from the early 2000s
Some industry trends are worth watching: for example, the return limited non-QM loans of documents; increased demand for investor loans and fixed and reversible loans; and the resurgence of third party arrangements through wholesale and correspondent loans.
The changing economic landscape, the digital transformation of mortgage origination and real estate settlement, and the pandemic are also creating new challenges in fraud detection and income and employment verification. For example, the growing odd-job economy means that nearly 40 million Americans are now their own employers, not W-2 borrowers. Likewise, advances in digital lending are reducing in-person interactions during build-ups and closings. The trend towards working from home, made necessary by the pandemic, is very likely a permanent option in some companies.
New ways to predict and detect fraud
Manufactured income and undisclosed debt are the most common forms of housing fraud. However, other schemes may involve forged documents, including verification of inflated or fabricated income and employment, synthetic identities, and false returns of investment income through reverse occupancy systems.
Proven fraud alert solutions are effective in detecting defects that could indicate fraud or a compliance issue as the loan application progresses through the workflow process. Likewise, new industry databases are also helping lenders and their technology partners identify employment fraud.
Today’s fraud alert systems make extensive use of available data (natural intelligence) and work well in standard mortgage origination workflows. However, large originators may need a more focused workflow solution that helps them become more efficient by reducing the volume of loans to review and clear, so they can focus their resources on them. riskiest loans. This is where the use of natural and artificial intelligence can change the landscape of fraud detection.
This new generation of fraud detection has just arrived on the market. Specifically, solutions that use predictive modeling and pattern recognition scoring to identify the level of fraud and the risk of prepayment default in a specific application or in a loan portfolio. Larger lenders should consider moving from their current alert-based system and take advantage of today’s predictive modeling and machine learning techniques that can simultaneously examine models and factor-aware submodels. such as synthetic identity, manufactured income, fictitious employment, prepayment default, undisclosed debt and loan. risk of participants. First American, for example, was able to reduce the average review rate from 60-70% to just 10% by using a targeted risk score.
Through the combined use of natural and artificial intelligence, lenders can benefit from this new operational efficiency, while mitigating EPDs, buyouts and long and expensive care and maintenance of distressed assets. These new analyzes will also have practical application in the acquisition of investor loans and the service of transfers.
Fraud detection is a never-ending journey to find the proverbial needle in the haystack. With the evolution of predictive analytics, machine learning and artificial intelligence, lenders should explore opportunities to anticipate the next wave of potential fraud risks to proactively protect themselves.