This feature originally appeared in the June issue of MReport.
Carissa Robb serves as President and COO of Constant. She most recently served as SVP and Head of U.S. Loan Servicing for TD Bank, overseeing operational units responsible for servicing a $150 billion dollar portfolio of consumer, residential, and commercial accounts. She joined TD Bank in 2009 to develop the loss mitigation program for distressed real estate and built the governance and control framework for TD Bank’s Loan Servicing and Collections division.
Robb recently spoke with MReport on whether the housing and mortgage industries are more prepared to handle a crisis than they were a decade ago, and what AI can do to assist the hardship process.
What are some of the differences between the Great Recession and the challenges faced today? While there are some similarities, there are also several differences, the biggest one being volume. In 2008, there was an equity challenge for homes. They weren’t holding their value. This, coupled with the fact that adjustable rates were causing homes to be unaffordable, meant many people were underwater.
Also, unemployment wasn’t nearly as severe as it is right now. Today, 33 million people have sought unemployment benefits. We saw a lot of short sales and loan modifications in 2008-2009. Yet many people still lost their homes. Then you add on the sheer volume of unemployment from this crisis, and the ability for people to purchase these homes becomes a question mark due to damaged credit and insufficient income. People, especially small business owners, are going to need more time to recover.
How quickly they can return to income levels that support their existing monthly payments or a modification of that payment is something that is unclear. That’s where a lot of my anxiety looking at this crisis and the duration of this crisis comes from. Will they have enough time to get back on their feet and will there be buyers for borrowers willing to sell their homes via a short sale. Mortgage forbearances have already doubled past the defaulted peak rate back in 2008. But the other interesting fact beyond the sheer volume that’s different from the Great Recession is that investors moved fast this time.
This was a real challenge in 2008. Relief was slow to come because the playbooks were still being written. But this time you saw a rapid approach to offer everybody relief in the form of short term forbearances, collection moratoriums and extensions. In 2008-2009, a lot of people were dragged through a process that was frustrating and caused credit deterioration. I think they avoided that this time by acting quickly—although the pendulum may have swung too far in the opposite direction by not having some barriers around who qualifies for forbearance and who really needed the relief. That is another major difference from 2008-2009 as we look ahead with this pandemic.
Are the housing and mortgage industries more prepared to handle a crisis than they were a decade earlier?
Truthfully, I wish we were a bit further along. The playbooks feel very similar, and that’s disappointing. If you were one of the folks in the trenches during 2008-2009, you’re looking at this with the same sense of urgency to act now and prepare: don’t wait, this is not short term. You’ve got to act now to sustain the servicing demands that are coming from customers. We’ve seen a few players respond with technology, offering self-service options for extensions or forbearance agreements in what seems like more of an automated approach.
Unfortunately, I still see a lot of the fine print that says, “We’ve received your request. We’ll get back to you in 7-10 days” which implies that the backend is still manual. And that’s the disappointing part. The potential for automation is there, but we’re still spinning the dialer and adding FTEs through staffing campaigns. The burden on call centers is immense, which implies that we may still be behind in dealing with the volume from a sustainable, automated, intentional way. This is one macro event, but it’s not going to be the last macro event.
We’ve seen a lot of challenges certainly from the mortgage crisis, but there are also the hurricanes. We’re so focused on helping people out of COVID-19, but we’re heading into hurricane season. 2017 was one of the most challenging years for hurricane relief and billions of dollars in damage, and a lot of customers needed help there as well.
The solutions require more precision. It requires a lot of time and investment to create a playbook that is not one size fits all, but very responsive to the specific hardship, and the severity of that hardship, and really understanding what it means to help a borrower fix the problem. It takes a lot of time and a lot of investment. There has been a lot of financial investment around automating credit origination because the focus is typically on revenue generation.
The back office always suffers. We’ve shortchanged our customers as a result of that, and here we are again without technology advancements to help through a crisis.
What is Constant doing to help automate the hardship process?
This is a passion of mine. There are real people in the middle of this. As a servicer, whether you’re prepared or not, whether you’re offering complex relief options or not, you still have to remember that you’re dealing with people’s lives. A person’s home is the most important asset to them.
At Constant, we’ve leveraged all of our experience servicing customers seeking hardship relief. You don’t have to engage in large staffing campaigns. With the competition for FTEs, you may not even find them or be able to train them in time. Technology and AI can help with this problem and in a very sustained, controlled way.
That’s what we’ve done at Constant. This is not about brand recognition for us. It’s about doing better. There are millions of people that need help. We’ve really tried to apply a lot of logic, a lot of integration work, and leverage our partnerships to say, “We can help you because we can understand what you’re asking for. We understand your financial position” and be very prescriptive about how to unwind that.
This could be the time to sell your home. It could be a short sale, it could be a modification. Modifications don’t just have to be on mortgages. We can look at the affordability and the ability and the willingness of a consumer to repay their debt and structure very precise, sustainable solutions by engaging with that borrower and applying some automation and technology to that process.
What did you learn during your time at TD Bank that you are utilizing at Constant?
Constant is a culmination of my experiences, both at TD and at other points of my career. I was a consumer advocate for a while, and a bankruptcy paralegal prior to joining TD. One thing I learned is that you have to be cognizant of the consumer and the different flavors and aspects of a hardship.
The second is that infrastructure is not going to be the same. You can’t design a solution that integrates with the Cadillac API-based servicing platform that can handle any level of automation. We had to develop something that can integrate and accept that servicing platforms are going to be at various levels of maturity. It doesn’t mean you can’t improve your self-service game with a partner that’s good in that field.
The last thing is preparedness. It’s not a lack of goodwill or good faith intentions to help customers out of the problem. It’s not that anybody is sitting there saying, “We don’t want to help,” or, “we don’t want to take the time to help.” It’s just that when you come out of something like the last recession, where we were dealing with massive volumes of foreclosure mediation, short sales, right through 2016, it takes a lot out of you in the back office.
It takes a lot out of your collection staff, your servicing staff, your loss mitigation underwriting staff, your foreclosure team. When you get a little bit of reprieve from that, because delinquency trends start to improve, the foot comes off the gas pedal. But this is actually the time when you should keep your head down and prepare. Do post mortems and prepare for the next event. Had we done that a bit better after the last recession, I think we’d have been more prepared for COVID-19. Not the events, not the crisis itself, but the financial resolution, the technology solution for how we navigate the mass influx of customers that need help.