Return on Quality

Image by Tessa Hunkin 

Fannie Mae’s new Loan Quality Initiative adds a number of new quality control requirements for originators that must be in place by July 1, 2010.  Perhaps the biggest change is the new requirement for all originators to perform pre-funding quality control reviews, in addition to the existing requirements for post-funding and early payment default (EPD) reviews.  For mortgage quality control professionals, this is another step toward what we at Cogent call “Enterprise QC” — an integrated, end-to-end approach that promotes continuous QC monitoring of all loan origination and servicing processes.

Until now, most lenders have had a disjointed and incomplete approach to quality control across the enterprise. Even among lenders that have been doing some form of pre-funding review, the results are often not available to post-funding reviewers, because there is not a common database for sharing the information.  Although many lenders have begun using automated compliance engines (ACE’s), such as those provided by Mavent and ComplianceEase, the loans that are flagged by the ACE for potential compliance errors are not automatically targeted for post-funding reviews. And QC auditors doing reviews of EPD’s, Repurchases and Claim Denials often do not have access to the data from the pre- and post-funding reviews. On the servicing side, many lenders still do not have a formal quality control process in place, and those that do often do not have access to data from other servicing department audits, let alone audits of originations.

We believe the keys to successful Enterprise QC are: (1) the ability to easily access and manipulate the production and servicing data that are needed to accurately define the populations and select the loans that qualify for each quality control audit, (2) continuous communication between quality control managers and the managers of the processes being audited to ensure that audit checklists always reflect the most current policies and procedures; (3) closed loops for reporting, feedback and response, to ensure that adverse findings are responded to and corrective actions are implemented and documented; and (4) sharing of all quality control data across the enterprise, to maximize the returns from the risk information generated from each QC process.

It’s been no surprise to see a flood of books aiming to unravel the causes of the financial crisis of 2007-2009.  By many accounts, Gregory Zuckerman’s “The Greatest Trade Ever: The Behind-the-Scenes Story of How John Paulson Defied Wall Street and Made Financial History” is the best of the bunch.

Now Michael Lewis, one of our favorite financial authors, has thrown his hat into the ring and released “The Big Short: Inside the Doomsday Machine.”  Lewis is one of the most engaging writerss working today, with a particular flair for apt analogies and clear phrasing (”A credit-default swap was confusing mainly because it wasn’t really a swap at all.  It was an insurance policy…”)

By weaving compelling characters into his narrative, he manages to entertain while informing.  No mean feat.   

For a taste of Lewis’ latest, check out the long excerpt in Vanity Fair.  You’ll see that it reads like a thriller and crystallizes a lot of the thinking that’s surrounded this epic debacle.

(By the way, note the interesting phenomenon on Amazon’s customer reviews of the Michael Lewis book.  Almost every one-star rating bemoans the lack of a Kindle version of the book, not the actual content of the book.)

The recent MBA Mortgage Servicing Conference in San Diego, at which the talk of loan modifications was front and center, reminded us of an article published in the November 2009 issue of Mortgage Banking magazine titled “Servicers as Originators“ [requires subscription]. That article described the most significant recent development in the world of mortgage servicing: the need for mortgage servicers to act like mortgage originators as they re-underwrite loans under various loan modification programs such as HAMP.

While loan modifications are not new, the sheer scale of pending loan modifications has overwhelmed servicers and drastically extended processing times.  At the same time, critical scrutiny of the process and pressure to accelerate the pace of completed modifications, has created fertile conditions for a new loan quality disaster. 

Image by Matthieu

This represents a vastly more complicated “servicing” process than traditional servicers are used to.  Indeed, the modification process is arguably more complex than the origination of a new loan, requiring a re-underwriting of the loan, complete with credit reports, appraisals and verifications (income, asset and employment).  All this in addition to program-specific documentation requirements, which in the case of HAMP, are onerous.

Recognizing this requirement for a different skill set, servicers have been hiring servicing reps with origination experience.  Thus, much of the production staff of now defunct mortgage lenders have new gigs as servicing staff, no doubt helping unemployment numbers in subprime epicenters such as Irvine, CA. 

In addition to massive new hiring, servicers are deploying new or modified software systems to automate what they can; outsourcing various sub-processes where they can; and trying to stay in compliance as program guidelines change.  Which they do, frequently.

As servicers struggle to meet these challenges, it falls on auditing and quality control professionals to ensure that the latest processes and compliance requirements are adhered to.  Unfortunately, this is not an area where software vendors have invested much time or effort.  Except Cogent, as it happens.

Cogent’s ServicingQC system was introduced more than a decade ago and has evolved into the most sophisticated quality control system available for servicing operations.  Moreover, with the release of the CogentQC.NET platform, virtually all of the functionality of Cogent’s ProductionQC system - designed to monitor origination quality - can now be embedded into a ServicingQC system.  Result?  A ready-made platform for servicing quality professionals to monitor their new “servicing” process: loan origination.

Ocwen Financial Corporation, which does residential and commercial loan servicing, special servicing and asset management, is reporting significantly better results in its HAMP modifications than the rest of the industry.  And they attribute it to superior technology. 

In congressional testimony, Ocwen’s President Ronald Faris claimed:

• that Ocwen is converting trial modifications to permanent modifications at a rate that is 10 to 20 times greater than industry average;
• a 3-month re-default rate of less than 5% for Ocwen, versus 18.7% to 33.7% for the industry; and
• a total of 100,000 successful modifications since the beginning of the mortgage crisis.

These impressive numbers could use a little clarification.  So could the statement that Ocwen has spent “$100 million in R&D to build loans servicing technology that incorporates behavioral science for effective customer communication and is also scalable for high volumes.”  But Ocwen is an industry leader and when they report these kinds of results, it’s worth listening.

Image by sgrace

Likewise, their recommendations for improving the HAMP program carry weight.  Among those:

• Lower the borrower debt-to-income (DTI) ratio for modifications to below 31%; in other words, allow for lower monthly payments on modifications  
• Allow for principal reductions on modified loans  
• Make additional funding available for housing counseling groups
• Require underperforming servicers in HAMP to outsource to servicers that perform 

Also worthy of note: Mr. Faris is one of the few industry voices who believes that HAMP is a “well designed response to the mortgage crisis”.  Maybe he’s on to something.

It’s good news that in some quarters, quality control is beginning to matter. But here’s the Economist to remind us that the job of risk manager “is said to have the risk profile of a short option position with unlimited downside and limited upside — something every good risk manager should avoid.”

Small wonder that talent is staying away in droves.  In sales-driven cultures - like mortgage banking - it’s frowned on to discourage transactions, without which money can’t be made.  The bias is to get the deal done.  So risk managers are always swimming against the current. 

However, risk is currently the busiest area for financial recruiters, which means there is a lot of activity.  Chief risk officers are being appointed, risk committees and departments are being formed, and new regulations are announced with frequency. 

But business practices adapt to new structures and find their way around them. (Like fraud, in some ways.)  Once the spotlight is off regulation and compliance, experience suggests that a new innovation will make it seem like “it’s different this time”.  And the Cassandras in risk management will be ignored anew.

So how does a risk manager survive the next bull market?  Hope that new incentives will be put in place to encourage the right behavior.  And show the ROI of risk management - or in the case of mortgage quality control professionals, show the return on quality (ROQ).  In future posts, we’ll try to help formulate that ROQ.

Apparently, in early 2007 - at the peak of the real estate bubble - the Mortgage Bankers Association (MBA) came to the “inescapable conclusion that owning [their]own building was the smartest long-term investment for the association.”  So with $75 million in financing, they purchased a new headquarters building.  Then on February 5, 2010, they announced that they had sold it for a little over $41 million.  Oops.

This is the sort of fiasco that the mortgage industry can dine out on for years.  But it’s just the latest irony that we’ve witnessed recently.  Remember the advocates of unbridled capitalism on Wall St. asking for government assistance so that they could continue to do ”God’s work“?  Then paying out billions in bonuses with taxpayer money?  Irony verging on outrage.

But more ironic to mortgage quality professionals, and likely to have a larger impact, is another major story these days: the recall of millions of cars by Toyota due to quality issues. 

Toyota?  Poor quality?  How could the company that virtually invented the quality revolution in manufacturing over the past 50 years be in this position? 

As one article points out, it was a combination of things: the goal to be Number One in the world driving rapid expansion; increasingly complex products with multiple potential points of risk/failure; uninterrupted success leading to arrogance; a culture that discouraged bad news; and managerial sclerosis. (One might almost be talking about Wall Street).

It might seem like the compeuppance of a world leader in the quality movement would negate the value of a quality culture.  Yet this story may unfold to include other car manufacturers, too.  Everyone is trying to sell as many complex ‘world cars’ as they can, and each car shares components with other cars in manufacturers’ product lines. Will Toyota’s brand be the only one tarnished?  Probably not.  But Toyota will recover, while others may not.  And Toyota at least will re-learn a basic lesson: don’t rest on your laurels. 

Statistics is baffling enough without being footloose with terminology.  So let’s clear up what we mean by a statistical sample.

The term “Statistical Sample” has a very specific meaning in the Cogent system.  It refers to a sample that is randomly selected from the entire population of loans eligible for a particular sample type (aka “audit shell”).  The suggested sample size is calculated every period by the system and is designed to yield a 95% confidence and 2% precision over 12 months.  This is the standard originally established by FNMA, FHLMC, and HUD for lenders who qualify to substitute ’statistical sampling’ for the traditional 10% random sample.

The generic term ’statistical sample’ is not very meaningful, in and of itself.  It simply refers to a sample in which some statistical principle has been employed, without defining which principle.  For example, it could refer merely to a randomly drawn sample, without specifying what population is being drawn from or how much precision will be achieved across what period.

Image by johnbullas
Rick Astley reference 

To illustrate: most Cogent ProductionQC clients have at minimum a “Production” sample type, for which all loans originated in a particular period (typically a month) are eligible.  When a Statistical Sample (in the Cogent definition) is randomly drawn from this population, all loans have the same probability of being selected.  No distinction is made between loan type, loan source or any other loan characteristic.  It is intended  to establish a baseline of overall loan quality across the organization.

In order to achieve a 95% confidence and 2% precision for a particular category of loan, it is necessary to go beyond the “Statistical Sample”.  For example, in the Cogent system, to achieve this standard for all FHA loans originated, define a Targeted Query (Loan Type = FHA) and run the query.  The resulting screen displays all qualifying loans, including qualifying loans  that were randomly drawn previously in the “Statistical Sample” or any other samples in this period.  These count towards the total required.  Use the embedded Cogent Statistical Calculator to calculate the required (”suggested”) sample size for the period.  From the suggested sample size, subtract the number of qualifying loans that have previously been sampled and enter the result in the Sample Size box. The Cogent system will then randomly select the entered number of loans from the qualifying loans.

The Cogent “Stratified Sample” is in effect a pre-defined Targeted Sample.  Most typically, the Cogent system stratifies originations by Source or Channel and automatically tracks and calculates the sample size required for each stratum (Source or Channel), net of qualifying loans randomly drawn in the “Statistical Sample.”  Over 12 months, the Stratified Sample achieves 95% confidence and 2% precision for each stratum.  In Targeted Samples, this automated operation is performed by the user, using Cogent’s embedded tools.

Thus, in order to leverage the Cogent system’s sampling optimization, clients should begin sampling from the broadest category (all loans eligible) to the most narrow category (e.g., individual underwriters).  In this way, all loans selected in previous broad categories are counted towards ever narrower categories, minimizing the number of loans to be sampled and audited.

Image by Rutty 

Berkery Noyes, “the only middle-market investment bank with research and M&A transaction teams dedicated to the mortgage technology, regulatory and compliance market,” has just published its 2009 Recap and 2010 Predictions For the Mortgage Technology Market. 

Providing a view from 30,000 feet, the report lists a number of high-profile M&A transactions from 2009 and offers up a few trends in the mortgage technology space, including:

• accelerating vendor movement towards becoming a complete end-to-end solution provider
• more stringent lending guidelines and regulations increasing the need for compliance, fraud prevention and risk-mitigating technology solutions
• burden of ensuring proper compliance moves to the point of sale rather than merely at closing, and falls on all mortgage industry participants

In the context of these trends, the report states that market participants no longer see compliance and auditing solutions as “nice to have” - they have become “must have” solutions.

We couldn’t agree more.