MARKETING MORTGAGES TO THE tens of millions of Americans who lack
credit scores is a bit like getting dressed in the morning without
turning on the light in your bedroom. You’ll probably put on
matched socks, but you’ll want to double-check to make sure before
you head for the office. [??] There’s no shortage of products and
services to help you find and rate the risk of lending to the 35 million
to 50 million Americans who lack credit scores because they pay cash as
they go, or have simply stopped using credit on a regular basis. [??]
Some of the tools, such as Dayton, Ohio-based LexisNexis[R]‘
ThinDex[R], have been used for some time in the unsecured
consumer-credit markets. Others–including Minneapolis-based Fair Isaac Credit Service Inc.’s FICO[R] Expansion[TM] score and Annapolis,
Maryland-based PRBC[R] (formerly Pay Rent, Build Credit), a credit
bureau that encourages consumers to report, document and view their own
payment histories online–are newer products being pitched to lenders as
a proactive means for making sure borrowers pay the lowest, fairest
mortgage rate based on their payment history. [??] The challenge is
figuring out which ones do the best job of finding and qualifying
credit-worthy but credit-scoreless borrowers.
[ILLUSTRATION OMITTED]
Wait a second
The catch with using any of these tools is the secondary mortgage
market has not yet endorsed them. The Federal Housing
Administration’s (FHA’s) position on Fair Isaac’s
Expansion score is typical of the attitude of the government-sponsored
enterprises (GSEs) toward credit scores targeting the emerging market.
[ILLUSTRATION OMITTED]
“FHA always encourages lenders to attempt to develop a credit
history for those who have no (or few) trade lines with the existing
credit bureaus. If lenders wish to utilize this proprietary, one-source
score to assist them in making the credit decision, they may certainly
do so. At this time, however, we will certainly not set any minimum
score as necessary for obtaining an FHA-insured mortgage (although
credit bureau scores are required to obtain a risk classification from
FHA’s TOTAL Mortgage Scorecard),” says Department of Housing
and Urban Development (HUD) spokesman Lamar Wooley.
Adds Freddie Mac spokesman Brad German, “We’re aware of
some of these products, but they’re not an active part of our
business at this time.”
Fannie Mae last year did two things to help lenders make mortgages
to scoreless borrowers. First, it expanded its guidelines so that if one
borrower has acceptable credit and a co-borrower has no credit, a lender
can accept a portion of the co-borrower’s income when underwriting the loan. Second, it gave housing counseling groups the ability to
capture nontraditional credit data via Fannie Mae’s Home Counselor
Online[TM] software.
Lisa Nelson, Fair Isaac’s vice president of business
operations, who last summer made the rounds pitching the benefits of the
Expansion score to the GSEs and banking regulators, completely
understands the importance of the secondary market’s willingness to
accept the expanded score.
“That’s why we’re putting so much effort into our
conversations with Fannie and Freddie and the large mortgage bureau
resellers to see where they’re going, and pushing hard for
opportunities to partner with people in all aspects of this
industry,” Nelson said.
Other data on the reliability of expanded scoring may come from the
Information Policy Institute, New York, which plans to publish a report
at year-end that will assess the predictive power of various types of
nontraditional data and their impact on consumer access to credit.
To be fair, the secondary market is not uninterested in tools to
tap the underserved markets. Fannie Mae and CitiMortgage Inc., St.
Louis, have both purchased research and credit modeling data from PRBC.
Because most of the products designed to help lenders identify
scoreless consumers and calculate the risk of lending to them involve
databases, the most important questions to ask product providers may be
these: Where does your data come from? How has their reliability been
tested? Who can alter or add information to the database?
If you rely on a nonprofit housing counseling agency to provide
data, are you sure the counselors are scrupulously verifying past
payments? Some counselors are more diligent than others, points out Mark
Catone, senior vice president at The First American Corporation, Santa
Ana, California. “This is what makes lenders and the secondary
market very nervous–whether the data has any integrity,” Catone
says.
If third-party brokers or even your own loan officers collect
nontraditional payment information, its quality must be checked. ABN
AMRO Mortgage Group Inc., Ann Arbor, Michigan, routinely validates and
spot-checks data from nontraditional mortgage credit reports created for
scoreless borrowers.
“If they provided us a cell-phone bill paid in cash for two or
three months running, we might call the cell-phone provider and verify
that information as part of our quality control,” explains Fran
Clemens, ABN AMRO’s senior vice president of emerging markets.
With those questions and concerns in mind, we sat down to talk with
providers of services old and new about what they can do for lenders
interested in helping borrowers without credit scores get the best
possible interest rate for which they qualify.
Finding them
The universe of scoreless borrowers is large and diverse, so your
first step should be to target the segment of this market that best fits
your product mix. You must also distinguish between the underserved and
the uninterested.
Kosta Skoulikaris, a senior marketing consultant for Experian[R]
Costa Mesa, California, says combining the right unregulated marketing
databases with the right multilinear equation can help a lender decide
not only who is likely to respond to an advertising pitch for a
particular mortgage, but also who is likely to be approved. After all,
it’s a waste of money to generate leads that won’t end up as
closed mortgages that meet your credit thresholds.
“We have a solution, VeriScore[SM], that we’ve had
tremendous success with in the financial services vertical using
marketing data to create a custom scoring model to rank and score an
audience that has not been served because they have not had credit or
they had a thin file,” Skoulikaris says. Clients using the product
have seen increases of 25 percent to 35 percent in response rates, he
claims.
The product combines analysis of a company’s previous buyers
with marketing data on 215 million customers, self-reported behavior for
30 million Americans and subscription and catalogue-buying history.
Part of the reason for the successes seen in this market is the
fact that this group of consumers hasn’t been getting three
credit-card offers a week in the mail for the past two decades. But the
more solicitations you send to the same group, the lower your response
rate goes over the long run, he warns.
So your chance to be the first one to mail a mortgage offer to a
scoreless borrower may be decreasing in the future. Experian has seen an
increase in demand from lenders trying to sell financial products to
emerging-market customers. “We’ve seen it within the entire
financial services spectrum specifically, trying to identify the thin
file and no hits,” Skoulikaris says.
Other tools for fine-tuning your marketing include Chicago-based
TransUnion’s expanded score product, which helps score borrowers
who don’t have traditional credit scores because they haven’t
used credit in the past year, or use it only around the holidays.
“We’ve been able to identify up to 35 percent more people
that would be overlooked based upon changed information in their credit
file than using the standard formula of having people meet the credit
score intervals,” explains Chet Wiermanski, vice president of
TransUnion Analytic Decision Services, Chicago.
Experian locates the market
While every market has scoreless borrowers, some offer more
opportunity than others, says Laura DeSoto, senior vice president of
product development and marketing for Experian Credit Information
Solutions, Costa Mesa, California.
“There are certain MSAs [metropolitan statistical areas] that
are very high opportunities–Los Angeles, Philadelphia, Houston, Tampa
[Florida] and Dallas are markets where there is a perception that there
is more opportunity in current renters who would like to be current
mortgage holders,” she says.
Starting with Los Angeles, Experian is putting full-scale effort
into finding data furnishers such as utility, telecommunication and
payday-loan organizations that can supply a mass of local consumer data.
One early project pulled full credit-file information from a
telecommunications firm with about 5 million consumers, 8 percent of
whom were not already in Experian’s credit files.
Telecommunications service is fairly universal, and a significant
portion of the population that has phone service doesn’t have
credit, DeSoto explains.
When the project rolls out nationally, the telecommunications
company will report about 20 million accounts to the company. What
Experian likes about telecommunications is that it mimics traditional
credit with bills paid on time, or 30, 60 or 90 days late. That
contrasts with prepaid phone cards and stored-value cards, for which the
consumer pays prospectively, as well as payday loans, which may have
unusual terms. Interestingly, 99.3 percent of those first 5 million
customers had only positive payment histories, DeSoto says.
“If we can prove the value there [in Los Angeles] to the
lenders and data furnishers and consumers, that would be more potent
than a shotgun approach across the entire country,” DeSoto says.
Trust, but verify
In a market where employees are motivated by loan volume and
throughput, how do you make sure they generate verified, quality credit
data for unscored borrowers?
First American’s Catone knows all about the angst risk
managers experience when they think about the manual underwriting
process often used with scoreless borrowers. “It’s very clear
that the process and that method do not protect the lender or the
investor fully in terms of risk, nor is it fast and easy,” he says.
“When we talk to underwriters, they are extremely concerned
that the proper level of due diligence is not being applied to these
files when they’re being processed, and they’re seeing more
loans of this type.”
To overcome that unease, First American offers its Assisting
Non-Traditional Homebuyers in Emerging Markets (ANTHEM) Suite with
NTreport. The latter is a nontraditional credit report for which the
company first retrieves any available bureau data and then follows up
with supplemental credit information verified by Fair Credit Reporting
Act-certified specialists working in the company’s bilingual call
center.
“Since this is one of the areas First American specializes in,
they are usually able to complete a credit history faster and at a lower
cost than the lender, which, combined with the increased data due
diligence, makes a compelling reason to out-source the data
compilation,” Catone says.
First American also goes one step further to rank, code and score
the information based on the type of credit, length of credit, payment
history and verification methods used to create items in the
nontraditional mortgage credit report (NTMCR).
To create its ANTHEM Score, First American uses the GSE rules for
nontraditional mortgage credit reports overlaid with verifications and
process rules and weights based on the length of the payment history.
“For example, 18 months of rent is an extremely high
weight,” Catone says. “If we don’t get a rent or utility
[payment history], we’ll go to insurance or revolving credit cards.
That can be supplemented by thin-file data [for consumers] with not
enough trade lines.”
First American will have more data on the product going forward,
because Boston-based Massachusetts Housing Finance Agency (Mass Housing)
Massachusetts’ independent affordable-housing agency, is going to
require its lenders to use the service or one that is as good or better
to deliver a score for scoreless borrowers.
“That’s a step up in quality, in their view. It manages
risk better for loans they will package and sell through bonds or
investor tools. It will prove it’s [ANTHEM is] going in the right
direction–or discover things we can improve,” Catone adds.
The ANTHEM report also pulls international credit, which can be
useful in cases where the scoreless borrower is a recent immigrant.
“We just did a report where we had a couple relocate from the
United Kingdom, and we packaged their U.K. credit report together with
credit established here to get four credit lines–and they got a lower
rate,” Catone says.
Pricing varies based on lender volume discounts, but can range
between $40 and $8o depending upon options including verification, data
storage and running a tri-merge credit report. The price also includes a
toll-free consumer help line for consumers who need guidance in
understanding how to establish a nontraditional credit history.
FICO Expansion score
Another scoring tool used in this niche comes from Fair Isaac
Corporation, Minneapolis. The company’s FICO Expansion score hit
the market last fall. Calculated using data collected from
nontraditional sources, reseller-verified information, application and
deal information, it’s typically used when a borrower doesn’t
have a full-blown traditional FICO score, or to supplement a thin file.
“We are targeting the consumer-reporting agencies that have
alternative credit data–companies that would have information that
might pertain to bounced-check activity; rent-to-own or payment plans;
magazine subscriptions; payday-loan information; public-record data,
which isn’t necessarily alternative, since some bureaus have that;
and debit-account behavior. In a second round, going for payroll
check-cashing behavior would be great,” says Fair Isaac’s
Nelson.
Calibrated to the classic FICO score range of 300-850, the
Expansion program can score 80 percent of all thin-file or no-hit
mortgage applicants, with 43 percent of them receiving a score of 660 or
higher, Nelson says.
To create the product, Fair Isaac looked at 200,000 consumers who
were unscorable at a particular point in time. Then it looked further
out in time when the consumers became scorable, and did an analysis to
see if the expansion score truly was predictive.
Lenders are now using the product live in controlled-volume
studies. Some are opening every account they score to see whether the
return is a 500 or an 800 and then using the score to assess performance
and decide where to set a minimum acceptable credit score. Some have set
volume limits to avoid taking on too much exposure until the product is
more established.
The information used by Fair Isaac can be enhanced with information
collected by the lender, but that information doesn’t go into the
database. “For any given transaction, when a lender provides us
information that information is used for that one score only. It’s
not added into the database, and not leveraged for future product
development or future score requests,” Nelson says.
Do-it-yourself scoring
Two other tools target renters who want to use their payment
histories to establish credit. PRBC combines data collection, scoring
and reporting technology with consumer-reported historical data verified
by third parties, including lending officials who are not involved in a
current transaction with the borrower.
“We allow consumers to go back 36 months in time to show
they’ve paid their bills on time using canceled checks, bank
statements, money-order receipts or even billing statements showing
prior months’ payments,” says PRBC Chief Executive Officer
Michael G. Nathans.
That type of data collection raises the issue of using NTMC to
balance poor traditional credit, something the GSEs do not allow.
“It [nontraditional credit] is not to be used as an offset to
derogatory credit,” says Wendy Wood, director of Fannie Mae’s
National Community Lending Center, Washington, D.C.
Nathans disagrees, saying that the Equal Credit Opportunity Act
(ECOA) states that lenders must consider, “on the applicant’s
request, any information the applicant may present that tends to
indicate that the credit history being considered by the creditor does
not accurately reflect the applicant’s creditworthiness.”
PRBC clients can also track payments going forward by bringing in
receipts for verification in hard-copy form or electronically via payees
such as payday lenders or bill-payment centers located in retail stores
that have agreements with PRBC.
PRBC also collects rent-payment information from landlords who are
able to supply data uploads in extensible markup language (XML)
collected from on-site check readers, payment terminals or
property-management software systems.
Lenders pay $2 to $8 for PRBC reports, depending upon volume. In
addition to CitiMortgage, PRBC’s customer list includes Advantage
Credit International, a Pensacola, Florida-based firm supplying credit
information to 7,000 mortgage brokers and other customers, and The
National Credit Reporting Association Inc. (NCRA), Blooming-dale,
Illinois, a trade association for consumer credit-reporting agencies.
The PRBC report includes a bill payment score and merges of PRBC
consumer files with Experian, Equifax Inc. and TransUnion consumer files
into a single machine- or eye-readable MISMO[R]-compliant lender report.
PRBC does not charge consumers a fee for its service, and does not sell
its clients’ information to other companies.
Another product targeting renters, RentReporters, Pasadena,
California, helps lenders verify the accuracy of rent payments. “We
contact the landlord and report to bureaus that accept third-party
reporting. We can give them [lenders] a legitimate reporting on the
landlord account,” explains Crispin Luna, RentReporters’ chief
executive officer and president. Working with title companies,
RentReporters verifies ownership of the rental property to ensure its
actually getting information from the landlord when verifying payment.
“We don’t do relatives and we don’t just put a supplement
on the credit report,” Luna says.
When a consumer signs up with RentReporters, the rent payment is
confirmed each month, which Luna says cuts down on fraud. While your
cousin Joe might lie once and say he rented to you all last year,
he’s less likely to tell the same lie every month for the next
year.
RentReporters also tracks payments for other services, but only
when the information comes directly from the service provider. “We
don’t trust the bill from the consumer. With the advent of
computers and [Adobe[R]] Photoshop[R] anything can be fixed. We only
take verifications from the entity itself or a middle ground between the
consumer and the service provider, such as a bank with online
banking,” Luna says.
Luna promises one other thing: “When you send us a client,
it’s still your client. We don’t remarket or sell leads.”
Consumers pay $89.95 for the first year and a two-year look-back history
to join RentReporters.
Meet Mr. and Mrs. Payday
If you’re considering using data from payday lenders,
it’s important to know who uses payday lenders and how that
population intersects with the scoreless consumer market. According to Teletrack Inc., Norcross, Georgia, which collects and shares credit
information about payday-loan customers and rent-to-own retailers, the
average payday-loan customer earns between $25,000 and $50,000 a year,
and more than 42 percent of payday-loan customers are homeowners.
“What makes our data set unique is the fact that payday
lenders and rent-to-own [stores] do not report their information to the
big-three credit bureaus,” says Todd Cable, Teletrack’s vice
president of marketing. “Someone who falls out of the prime credit
arena has limited places to go to get credit. They tend to go to places
that cater to that clientele–subprime auto finance, buy-here-pay-here
stores, title pawn dealers and subprime mortgage companies. Most likely
there is a credit report, and it’s going to be awful. We’ll
have more recent information [than a traditional credit report], which
also helps businesses when it comes to skip-tracing. We also have
bankruptcy and landlord/tenant data and Treasury [Department] Office of
Foreign Assets Control data.”
Cable says some lenders have incorporated Teletrack data into
predictive subprime scoring models, where it provides a lift when
information isn’t available from a traditional credit report.
To access Teletrack, lenders pay a monthly fee, which Cable says is
“affordable” even to small lenders, for a certain number of
inquiries per month. “If the credit department subscribes to us,
skip-tracing is included, provided they give us their charge-off
data,” Cable adds.
If you pick up the Teletrack data, keep in mind that the company
collects information on charge-offs that are paid and inquiry
information on subprime consumers with no negative credit history, but
does not report positive payment history. That’s why Fair Isaac has
elected not to use the Teletrack information in its scoring model,
Nelson says.
Check the courthouse
Public records can be another source of information about scoreless
borrowers. LexisNexis’ database gathers information from those
records, including bankruptcies, judgments and liens filed in local
courthouses. In courthouses where the records are not yet online, the
company physically sends runners to collect information, says Tom Brown,
LexisNexis’ vice president for financial services solutions.
Lenders can also get data collected from property tax and motor
vehicle records. “The borrower may already have assets that are not
being financed in any way, such as vehicles, boats, RVs [recreation
vehicles] or airplanes, and any lien on those would also show up,”
Brown explains.
Clients using LexisNexis’ ThinDex over the past four years
have included telecommunications, auto lending and mortgage companies.
To make the product function as well as possible, LexisNexis uses a
sample of the customer’s accounts to refine its risk-management
algorithms and customize the database.
Brown adds that because ThinDex is sourced directly by LexisNexis,
the lag time between when information appears at the courthouse and when
it is posted on ThinDex is “a matter of days.”
LexisNexis’ pricing is based on volume.
The old-fashioned way
Until a more substantial verdict on tools for evaluating the
payment history of the unscored masses comes in, you can always
originate mortgages for scoreless borrowers the old-fashioned way: Use
the agency guidelines that explain how to create an NTMCR. Find someone
who’s completely trustworthy in your shop, and have him or her
manually verify the borrower’s past payment history.
The FHA’s guidelines are in chapter 2 of HUD Manual 4155.1
REV-5: Mortgage Credit Analysis for Mortgage Insurance,
One-to-Four-Family Properties.
Freddie Mac’s rules are in its Single-Family Seller/Servicer
Guide, volume 1, chapters 37.4 and 37.11. Fannie Mae’s Selling
Guide, part X (underwriting guidelines), sections 103 and 804, explain
its rules for NTMCRs.
Worth the bother?
With all the risks inherent in lending to scoreless borrowers, is
it really worth it to tackle this growing market? Yes, definitely, say
those in the market already. If the lure of 35 million new customers
doesn’t attract you, perhaps the idea of doing the right thing
will.
“You look at things like utility payments and rent payments
that don’t show up on your standard credit report,” says
Kenneth Wade, chief executive officer of NeighborWorks[R] America,
Washington, D.C., an umbrella organization for hundreds of nonprofit
community development groups.
“When consumers show a track record in those areas, it’s
just as reliable a predictor, but most telecom and utility companies do
not report to credit bureaus. It puts folks who have non-standard credit
at a disadvantage, and they end up in subprime or less-than-advantageous
loan products when they could be in a loan product that was priced
better for them.”
Dona DeZube is a freelance writer based in Clarksville, Maryland.
She can be reached at dezube@comcast.net.
from WordPress http://bit.ly/1nAFiWO
via IFTTT
No comments:
Post a Comment