Printer Friendly

What originations can learn from workouts.

Bill Kelvie, the former chief information officer at Fannie Mae and part of the force that introduced the first generation of Desktop Underwriter[R] (DU) to the mortgage world back in 1994, made an interesting observation recently. He said, "We're learning a lot about originations from these loan workouts ..."

Whatever Bo knows about football and baseball, Kelvie knows about originations. But there is more to his observation than meets the eye.

If one roiled the DU launch tape forward a bit, one might recall that Fannie Mae (and Kelvie) also ushered in the second and third generations of DU before 2000, suggesting that Fannie Mae recognized the benefits of its DU decisioning experiences and continued to benefit from advancements in both knowledge and technology.

In hindsight, today we might consider three generations of a decisioning solution within a (not particularly rambunctious) six-year mortgage period a glacial pace by today's benchmarks.

Or would we?

Fast-forward to 2010, and take a peek at originations again. How much adjusting and tuning of automated underwriting have occurred in the past six years? While we may be operating with a smaller array of mortgage product offerings, we do seem to be doing less automated decisioning and more manual underwriting/decisioning than perhaps ever before.

Why? A quick tour through just about every servicing operation in America provides some clues.

In the attempt to make a nonperforming loan perform again, there seems to be no limit to the amount of legwork being done. As Kelvie noted, there is lots to learn from what was done (and not done) at the time of originations. In addition to the obvious--those loans based on stated income and assets--there is the growing evidence that a single property value will never suffice in the future, that a probability of default will become a permanent part of the evaluation, and that a loan's net present value (NPV) may be as relevant as its loan-to-value (LTV) ratio or interest rate in determining what is the best offering.

But a clearer picture of what went wrong, what are the key indicative variables and where things can be made right are all emerging from these loan workouts. And, true to Kelvie's observation, there is a lot to be learned from yesterday, today and tomorrow's loan workout scenarios.

In short, should you consider every loan a loan workout, just at a different stage? How would that impact your origination strategies? Perhaps we should talk about 10 things we should learn about originations from loan workouts.

1. Get the borrower into the right option the first time

Some servicers are struggling with this--offering the Obama modification (the Home Affordable Modification Program [HAMP]) first because the numbers work, but the chances of success are not always numerically based. Those servicers that are getting the borrower into the right program the first time are experiencing success rates echelons beyond the industry averages. So we know it's possible.

Translated to originations, this strategy would probably not include the program that pays the highest loan commission, or even the one that makes the company the most immediate profit. However, when all factors impacting the borrower's loan performance are weighed using good decisioning functionality to evaluate all possible loan options, chances for a successful outcome escalate.

2. Throw out preliminary product pricing

As the world has learned from trial modifications, they are [expletive deleted]. A trial mod is just a feel-good exercise that makes parties on both sides feel as if they've accomplished something. They really haven't.

Trial mods are the loan-workout equivalent of preliminary product pricing. Forget preliminary product pricing. Do the full eligibility evaluation right up front. Start with the borrower there. You have actually accomplished something at that point and if we've learned anything from servicing, it will entail considerably less rework and legwork.

3. Drive with data, not paper

Third-party data, information and analytical capabilities have become so rich, broad and deep that the paper versions for most documentation are superfluous. Servicers using more data and information are having richer conversations with borrowers, and in many cases achieving more rapid and higher levels of success.

These same data sources are as relevant in originations as servicing. As important, nearly all of the required regulatory and closing documents can be done electronically and traced through eVaults, ending the ridiculous paper-file shuffle that only aids the ability to lose things.

4. Advance borrower communications

As consumers, borrowers have become accustomed to real-time interactions, whether a live chat with their bank or Land's End, or secure e-mail tracking events such as iTunes[R] purchases. Aside from those delinquent borrowers who won't reach out to communicate, the nonperforming borrowers who are trying to stay in their homes duly appreciate the armada of communications options some servicers provide, and they use them.

Secure e-mail, chat, 24/7 toll-free numbers, text messaging and, especially, electronic documents (which save not only on time and postage, but the research and tracking of lost documents) have all reduced time to completion and boosted the odds of a successful loan mod. The loan officer and/or processor need not be the only channel of communications. Many borrowers are very good at it.

5. Keep programs current with automation

Since its introduction the Obama plan (aka HAMP) has undergone more than a hundred changes and/or enhancements. Keeping people trained and operations current on the program's features is both time-consuming and fraught with potential inconsistencies. The only way to effectively stay up-to-date is through automation.

The same is true of origination product offerings. Substituting rate sheets and fliers for automation is a breeding ground for disappointments for both the borrower and the loan officer. The lesson here is automate wherever possible.

6. Make decisions transparent and auditable

As we've learned from the Making Home Affordable Compliance administration, every piece of data used to run the waterfall, calculate intermediate values and generate the net present value must be saved, whether the borrower qualifies for the Obama plan or not (if you have signed up to be a HAMP servicer). This can be accomplished using hundreds of Microsoft[R] Excel[TM] spreadsheets or through automated decisioning.

Given the amount of manual underwriting (and multiple iterations of manual underwriting) occurring in many origination shops today, a similar requirement for underwriting could be a show-stopper. It doesn't have to be.

7. Use outcomes for analytical purposes (and better offerings)

At present, servicers are incented to track reperforming borrower performance for up to three years. For some servicers, this requirement is also driving some advanced analytical considerations of why some borrowers perform and some do not, and why.

The nuances associated with successful reperformance can then be blended into non-Obama loan workout offerings that could result in even higher levels of reperformance All of this information is relevant to formulating new and more effective origination strategies--and probably should be.

8. Automate levels of discretion and approvals

For every 10 workout agents hired by servicers, at least two supervisors and one manager may be required. Pure mathematics can be used to determine how many possible outcomes 10 agents, two supervisors and one manager could produce on three loan workout options, wreaking havoc when it comes to any form of consistency of repeated outcomes. But if the decisioning is automated along with the approval process (using rules-driven decisioning), consistency will be front-and-center.

Given the amount of manual underwriting and iterations of manual checking going on in the origination process, automating the loan approval process could go a long way toward a forward-thinking, successful origination strategy. While there are recognizable nuances to this approach (such as the Federal Housing Administration eagle), these nuances actually create a greater need to automate the approval process.

9. Be more efficient

Twelve months after the Making Home Affordable program (now programs) announcements began, the efforts to implement them are still a work in progress. Borrowers, servicing staff and service providers read the announcements, attend informational sessions, whiteboard, discuss and deploy, yet are still trying to figure out how to make it work. Little time has been spent on how to implement effectively or efficiently. Some not-insignificantly-sized companies are still debating whether to build or buy a solution. As a result, borrowers in need are waiting months for assistance--and now even more than a year.

Efficiency doesn't mean cutting corners, it means getting the effort done with the right metrics in mind. As you contemplate a new origination program or offering, balance what the optimal offering (including process) looks like, with how you are able to offer it. If the gap is huge, it will not work. The trial-to-permanent mod process has proven this in servicing. It doesn't need to be replicated in originations.

10. Eliminate product silos

Adding to nonperforming borrowers' frustration is that the queue they work through changes as their status changes. Effectively they start over as they move from collections to loss mitigation to loan workout to short sale and so on. While this is hard on the borrower, it creates significant inefficiencies for the staff in each silo (e.g., department) who will also have to start over each time.

Historically lenders have offered different products and/or different product features, depending on the origination channel. As a result, fulfillment centers evolved per channel--an unnecessary burden not only on cost but also on the organization's ability to sustain compliance, efficiencies, consistency and a smooth investor delivery.

Investors are done with surprises. The only way to deliver effectively these days is to consistently offer products. Siloed product offerings are history.

There may be less focus on automated decisioning in the originations arena than ever, in part because there are fewer products but more so because the decisioning focus is almost exclusively on the loan work-out arena. Focus, time, budget, energy--there are only so many resources a mortgage entity can deploy at any one time. But there is a lot those responsible for origination strategies could learn from this loan work-out era, in addition to making sure this phase becomes a bad memory--not the prototype.

Linda C. Simmons is general manager, mortgage, with Overture Technologies Inc., Bethesda, Maryland. She can be reached at
COPYRIGHT 2010 Mortgage Bankers Association of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2010 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:The Deciding Advantage; Federal National Mortgage Association
Comment:What originations can learn from workouts.(The Deciding Advantage)(Federal National Mortgage Association)
Author:Simmons, Linda C.
Publication:Mortgage Banking
Geographic Code:1USA
Date:Jun 1, 2010
Previous Article:The on-demand economy.
Next Article:RESPA readiness: here are some best practices for hitting a moving target known as RESPA.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters