Vaulting: The act of storing documents safely. In the mortgage environment this is referred to as an eVault, where document storage is electronic and retrievable through a web-based interface.

“What we’ve done is implement a process that has changed people’s behavior. We’re now providing incentives for people who close loans electronically. We’ve been able to demonstrate that it’s a process that is not overwhelming. Once they see that it’s easy to do, they see the greater benefits.” --Steve Trayte, vice president of eMortgages with AmTrust Bank, Cleveland, Ohio.

Data Drive Third-Party Origination Risk Management
Palaparty, Vijay
DALLAS—Risk management in third-party originations centers on effectively and strategically managing data—using technology both to minimize risk as well as to create efficiencies, said panelists here at the Mortgage Bankers Association’s National Technology in Mortgage Banking Conference & Expo.

“Third-party origination strategies can have greater financial, operational and compliance risks,” said Chris Burckhardt, senior vice president and chief information officer at Pulte Mortgage LLC, Englewood, Colo. “As broker and correspondent lending channels continue to be strategic, lenders must find efficient ways to mitigate that risk and build profitable loan portfolios.”

Jay Levine, executive vice president and chief technology officer at Wolters Kluwer Financial Services, Minneapolis, said data and its different applications determine its value. “Securing clean and complete data from third-party originator, managing data in one place, validating data throughout the loan cycle and ensuring third-party origination risk acts in the borrower’s best interest and can improve third-party origination processes using technology,” he said.

“Every risk control effort should have the same strategy,” said Stanley Street, president of Street Resource Group, Atlanta. “Recognition of risk, assessment of potential risks, first execution—taking action on the assessment using technology, oversight, review and follow-up are components of the strategy. Technology implementation goals should also be the same at any level—goals have to be stronger, more robust in depth and accuracy.”

Street said goals should consider risk management roles and strategic corporate goals. “Goals should aim to maintain or increase market share but also protect and promote consumer’s profitability,” he said. “Credentialing, predictive analytics, which is more important among warehouse lenders and originators, and performance monitoring, making sure originators are performing properly with policies, are technology tools to help mitigate third-party origination risk.”

Street added that electronic data standardization and risk-point decision workflow processing enhance the process. “Risk point-decision workflow processing for warehouse lenders is taking points in the loan origination process,” he said. “When lenders make the decision to approve a loan, that is the ultimate risk. So they grab data and identify risk points in the process.”

Teresa Yow, vice president and business IT partner at Synovus Mortgage Corp., Columbus, Ga., said because a majority of borrowers prefer communicating via email, which raises identity theft concerns, lenders must have a secure method of sending sensitive consumer data.

“Integrating risk management in loan origination systems and workflow requirements yields a process that would have to be easy for both consumer and loan officer,” Yow said. “It requires bringing together departments that are players in the plan—they include production, origination, processing, technology, secondary marketing and compliance areas.”

Synovus has a secure document delivery system for origination disclosures, Yow said. Additionally, to avoid fraud, the company implemented a system to watermark documents with disclosure dates. Its email correspondence is auto-generated, validating delivery of documents for compliance.

Yow advocated working toward standardization with investors and integrating risk management into business workflow for security considerations. “Clarifying investor requirements upfront, clarifying documentation required for the loan file, enhancing the web site for customer usability, user training acceptance and borrower acceptance are important considerations,” she said.

“Having a secure web site meeting customer authentication requirements is important,” Yow added. “Ensure compliance requirements are met. Tracking of loan delivery and decreased postage and mailing costs are also benefits of mitigating risk.”
(Back To Top)


AUS Future Relies on Strategy, Data
Palaparty, Vijay
DALLAS—As a business strategy, the automated underwriting system redefines itself in the current lending environment. Going forward, it will need to focus on better serving lenders in decision-making and risk mitigation efforts, panelists said at the Mortgage Bankers Association’s National Technology in Mortgage Banking Conference & Expo.

To maximize AUS potential, panelists agreed that developers should closely tie in its purpose, goals and objectives with the technology driving automation.

“AU is rules-based and affects what you do today in making judgmental decisions,” said Michael Wing, senior vice president of consumer risk management for consumer real estate at Bank of America, Charlotte, N.C. “What is confusing to companies is custom modeling and the issue of owned versus proprietary systems. What is also confusing is all the types of data people have to create a database which underlies the system.”

Wing said automated underwriting systems could be compromised to approve more loans, pinpointing vulnerability in how the systems can be calibrated to satiate market appetite.

“You don’t need technology to say yes or no,” Wing said. “You can create modeling to consistently say yes to more loans. The systems are subject to data quality and when companies stopped verifying pieces of underwriting process, it had nothing to do with AUS. People put rules in there that were executed by the computer. We have to go back to the rules it was based on. People broke those rules and executed loans. The outcomes are what we see today.”

“Automated underwriting simplifies the loan process and decision making process,” said Craig Hughes, corporate secretary and vice president of the mortgage business center at CC Pace Systems, Fairfax, Va.

William Lehman, director at CC Pace, said AUS depends on a company’s business strategy and emphasized that it should not only be an information technology project. “The AU platform is more important than the loan origination system platform because it directs the workflow of the loan,” he said. “In terms of the fraud aspect, the underwriting group represents balance, reduction and efficiency.”

Panelists said Freddie Mac’s Loan Prospector and Fannie Mae’s Desktop Underwriter AU systems seem to be back in favor.

“LP is a tool by which we communicate and make a purchase decision,” said Staci Abel, corporate secretary and vice president at the mortgage business center at Freddie Mac. “Today, it’s a limited, credit risk management tool—a credit view of our consumers.”

AUS systems integrate statistical models in the decision making process. For example, LP uses statistical models in determining a borrower’s creditworthiness and acceptability of the loan application, also pulling credit and loan application data. But statistical models can also be subjective.

“LP is complex and is based on statistical models. Layering the statistical models with rules has served us well,” Abel said. Wing disagreed, saying statistical models are effective, but are only as good the data they are built on. “They are a function of what they are built on,” he said.

AUS interface features were also discussed and how they might help loan originators better determine the best type of loan for the borrower.

“Some originators are inexperienced and when there is more scrutiny, lenders are looking for anything they can to help loan originators to make loan,” Abel said. “They are looking for guidance. However, we don’t see that as our role, but we do give a purchase decision. The lender is still responsible for the loan.”

“Understanding how to model risk on loan products is important,” Wing said. “If you don’t get it right, it can kill you. Folks need to think about that—do they have credit view from their investors? Do they need to have their own system to put that loan through? Are they willing to take on that risk? An AUS system provides loan specific decisions based on human decisions. From a sales perspective, it is to construct a deal that should be approved and the system gives instructions on how to fulfill that deal. It’s a way to get a credit risk perspective and implement risk policy.”
(Back To Top)


Fraud Tools Gain in Sophisitication
Murray, Michael
CHICAGO—As lenders access tools to prevent mortgage fraud against the industry, they will require updated data specific to their region and products, said industry analysts here at the recent Mortgage Bankers Association’s Fraud Issues Conference.

With specific reports to analyze and vendors building tools to sell to the largest number of clients, lenders will need to focus on specific needs, said Mark Nelson, senior vice president of risk management at JP Morgan Chase, New York.

“[Fraudsters] are trying to move as quickly as they can, and it’s important for us to move as quickly as we can,” Nelson said. “Choose tools to test that are most feasible for what you are trying to do.”

Rachel Dollar, CMB, vice president at Smith Dollar PC, Santa Rosa, Calif., said a national appraisal registry/database to help prevent appraiser identity theft is one tool, similar to MBA’s mortgage fraud database objective. The Appraisal Subcommittee, an oversight mechanism for activities of The Appraisal Foundation, has a national registry database that contains information about appraisers who are, or have been, certified or licensed by states, territories or possessions in the United States to perform appraisals in connection with federally-related real estate transactions.

Not all appraisers, however, are in the registry. “There is a need to expand the ASC database and their scope of responsibilities to capture the details into the database for the appraisal registry,” Dollar said.

Feedback from lender colleagues about vendors and flexibility are all important aspects of choosing the proper tool, analysts said. Auditors will want to test the data and physical security of the tool.

Larry Ruder, vice president of fraud at Wells Fargo Home Mortgage, Des Moines, said Wells Fargo requires a source code review to prevent vendors from having a “back door” in which to steal identities. Additionally, Wells Fargo performs IT functions in-house so that vendors do not look into the security “black box.”

Christopher Johnson, in risk mitigation at Washington Mutual, Seattle, said that lenders need to determine the type of fraud in which to focus and test with the most current version, rather than “vaporware,” a product that is announced but fails to materialize.

“Operations and underwriting should determine what they are going to do and how they are going to refer the loans as red flags come up. Analyze the numbers,” Johnson said.

“If nobody else in the processing chain knows how to use it, the tools are ineffective,” Nelson said. “An overall prevention strategy requires a multi-layered approach that includes tools and policy within an organization. When you look at things, try to fit it in to what you are trying to accomplish.”
(Back To Top)


Investors See Cost Savings, Transparency Opportunities in Data
Sorohan, Mike
DALLAS—For industry investors—the government-sponsored enterprises and Wall Street—technology represents an element that not only requires adoption, but leadership as well.

“We’ve talked about the need to really improve data quality and efficiency,” said Ted Adams, director of technology standards at Freddie Mac, McLean, Va., speaking here at the Mortgage Bankers Association’s National Technology in Mortgage Banking Conference & Expo. “It has an impact on our bottom line—every improvement we can eke out is important.”

During 2007, Freddie Mac focused on three key areas: cost, transparency and missed opportunities. “We recognized that we had rising costs and missed opportunities associated with bad data,” Adams said. “We recognized that transparency is critical and we are working on programs that not only ensure that our data are accurate, but transparent.

In taking a renewed approach, Adams said Freddie Mac reached a recognition that data quality is not a technology issue, but a business issue. “We have all the technology support we need, yet we’re still having a data-quality issue—it’s a business issue, not a technology issue,” he said.

As a result, Freddie Mac developed an enterprise data quality program that identifies more than 800 critical data elements and metrics. “One of the things we realized was that to get a better hold of that data, we installed metrics that enabled us to measure the data elements,” Adams said. “As a result, we’ve seen a significant decrease in corrections of data elements, and it’s helped us to manage the entire data set much better.”

Second, Freddie Mac improved its relationship with its seller-delivered data analytics. “We had the same or similar data sets in different areas of the organization,” Adams said. “We expanded our dialogue with our sellers to improve communication.”

Third, in measurement, Freddie Mac kicked off a Six Sigma-like process re-engineering. “We expect that this will reduce and streamline our processes and improve efficiencies,” Adams said.

Additionally, Adams said Freddie Mac has reached out to other industry players, such as MBA, to examine larger-scale initiatives. “Right now, we’re focusing on what we already have and how we can improve those processes,” he said. Adams, a member of the MISMO Governance Committee, said Freddie Mac has had “significant successes” in implementing MISMO standards.

“It’s become a cornerstone of our current environment,” Adams said. “The use of MISMO data set helps us to solve problems and eliminate redundant and contradictory data so that we can use it appropriately and consistently. We’ve reaped benefits of adopting MISMO in Loan Prospector [Freddie Mac’s automated underwriting system]. It’s allowed us to aggregate data and perform functions in-house, which allows us to use data the way we want to use it. We can also change our businesses and terms without it impacting our partners.”

Adams said he expects the GSE to approve more eMortgages this year. “They’re starting to come on pretty strong,” he said. “The numbers are still pretty low, but the growth is pretty dramatic. It looks like we’re going to see a geometric progression. And we’re hearing from our trading partners that the see the cost-savings benefits. One of our customers reports a $75-$100 reduction in closing costs. And it’s being driven by our smaller partners, who are more nimble. And there are a variety of closing solutions out there now that are cost-effective.”

Deborah Holmes, vice president and CIO at Ginnie Mae, Washington, D.C., said her agency has a mandate to expanding eGovernment as part of the President’s Management Agenda, aimed at reducing costs and improving services. Ginnie Mae recently announced an Enterprise Portal, a single-point access point to all of Ginnie Mae’s business applications. The initial phase provides secure, user sign-on and ability to access reporting and feedback services. The first phase expects to implement in November.

“In the future, any business being done by Ginnie Mae will come through the portal,” Holmes said. “We plan for it to be a very simple process.”

Additionally, on April 1 Ginnie Mae will implement unique loan identification numbers for each loan within a pool. “This will improve risk analysis associated with data disclosure,” Holmes said.

Luiz de Toledo, senior vice president and chief administrative officer of technology with Fannie Mae, Washington, D.C., said current market volatility has put pressure on IT initiatives, from both budget and reaction time.

“Lower earnings are causing capital scarcity,” de Toledo said. “At the same time, there is pressure to operate in real time with better analytics. We see three IT priorities: improving credit risk enhancement; providing productivity and infrastructure improve