|
||
Chapter
3
Looking
Inside
Loan
Prospector
Alejandro
Hernandez,
a native
of Colombia,
entered
the
homebuying
process
with
several
strikes
against
him,
recalled
Donna
Van
Osten,
a senior
underwriting
officer
at PHH
Mortgage
Services.
At first
glance,
the
file
on Mr.
Hernandez,
who
was
moving
to Florida
from
Indiana,
“was
not
your
typical
‘slam
dunk’
loan
because
the
borrower
had
a low-down-payment
amount
and
was
taking
on a
significant
increase
in his
housing
payment,”
she
said.
When
Mr.
Hernandez’s
application
was
run
through
automated
underwriting,
a
different
picture
emerged.
The
comprehensive
risk
assessment
identified
enough
positive
factors
in
the
30-year-old’s
profile,
such
as
a
good
credit
history
and
accumulated
savings,
to
rate
Mr.
Hernandez
as
an
excellent
bet
to
handle
housing
expenses
tripling
from
$500
in
rent
to
$1,500
in
mortgage
payments.
Mr.
Hernandez
was
approved.
“Using
traditional
underwriting
methods,
we
would
have
been
concerned
about
granting
this
mortgage,”
observed
Van
Osten.
“Because
this
loan
scored
as
an
accept,
he
was
approved
up
front
and
benefited
from
the
subsequent
reduced
documentation
requirements.”
For
most
people,
purchasing
a
home
is
the
largest
financial
decision
of
their
lives.
Obtaining
a
mortgage
to
finance
such
an
important
purchase
involves
countless
pieces
of
information
ranging
from
income
verifications
to
bank
statements,
from
tax
returns
to
credit
records.
For
applicants
like
Mr.
Hernandez,
lenders
may
face
a
complicated
set
of
circumstances
that
does
not
point
to
an
easy
loan
decision.
Loan
Prospector
excels
in
analyzing
a
multitude
of
factors
simultaneously.
It
balances
the
layers
of
risks
and
compensating
factors
that
make
mortgage
underwriting
so
complex.
Building
Blocks
of
Underwriting
All
mortgage
underwriting,
whether
traditional
or
automated,
is
based
on
a
wide
variety
of
factors,
broadly
categorized
as
the
“three
Cs”:
collateral,
credit
reputation
and
capacity,
as
illustrated
in
Exhibit
3.
mortgage
research
consistently
has
shown
that
a
borrower
with
a
significant
financial
stake
in
the
property
is
less
likely
to
default.
The
relationship
between
borrower
equity
and
loan
default
is
shown
in
Exhibit
4,
which
examines
the
foreclosure
experience
of
borrowers
with
differing
down
payments.
For
loans
purchased
by
Freddie
Mac
between
1985
and
1989,
for
example,
borrowers
who
put
down
5
to
9
percent
were
five
times
more
likely
to
enter
foreclosure
than
those
who
made
down
payments
of
20
percent
or
more.
(For
further
explanation
of
exhibit
details,
see
About
the
Data
in
This
Report)
Credit
Reputation.
Mortgage
lenders
also
rely
on
credit
information
compiled
by
national
credit
repositories,
commonly
known
as
credit
bureaus,
to
ascertain
a
borrower’s
track
record
of
handling
credit.
Repositories
can
provide
lenders
with
detailed
credit
files;
they
also
can
provide
a
credit-bureau
score,
which
summarizes
the
information
into
one
number
reflecting
an
individual’s
expected
credit
performance.
The
need
to
consider
so
many
varied
pieces
of
information
increases
the
difficulty
of
making
an
accurate
assessment
of
an
individual’s
credit
profile.
Credit-bureau
scores,
long
used
in
consumer
lending,
address
this
problem.
Based
on
the
statistical
relationship
between
the
information
contained
in
individual
credit
files
and
actual
repayment
experience,
credit-bureau
scores
accurately
summarize
an
individual’s
likelihood
of
repayment.
FICO
scores
are
one
example
of
a
credit-bureau
score.2
FICO
scores
range
in
value
from
about
400,
denoting
the
highest
risk,
to
about
900,
indicating
the
lowest
risk.
Another
example
of
a
credit-bureau
score
is
the
MDS
bankruptcy
score,
for
which
a
lower
score
indicates
lower
risk.3
Freddie
Mac
research
has
shown
that
borrowers
with
strong
credit
profiles
are
significantly
less
likely
to
default
on
their
mortgages.
Based
on
our
1994
purchases,
for
example,
Exhibit
5
shows
that
borrowers
possessing
weak
credit
profiles,
defined
as
FICO
scores
under
620,
were
18
times
more
likely
to
enter
foreclosure
than
borrowers
with
FICO
scores
above
660.
Analyzing
other
data,
researchers
from
the
Federal
Reserve
Board
reached
a
similar
conclusion.
(See
Credit
Scores:
The
View
from
the
Federal
Reserve)
In
a
comprehensive
evaluation
of
the
relationship
between
credit-bureau
scores
and
mortgage
performance,
they
concluded:
Capacity.
A
borrower’s
financial
wherewithal
to
repay
a
mortgage
constitutes
the
last
of
the
trio
of
credit
determinants.
Typically,
capacity
is
evaluated
using
two
ratios
that
express
the
percentage
of
an
applicant’s
income
needed
to
cover
monthly
debt
obligations,
including
the
mortgage
payment.
5
Borrower
savings,
referred
to
as
cash
reserves,
also
are
used
to
assess
capacity.
The
linkage
between
total-debt-to-income
ratios
and
foreclosure
rates
is
demonstrated
in
Exhibit
6.
For
example,
based
on
Freddie
Mac’s
1994
purchases,
borrowers
with
total-debt
levels
greater
than
36
percent
of
their
incomes
were
twice
as
likely
to
enter
foreclosure
as
those
with
ratios
below
30
percent.
While
capacity
is
an
important
underwriting
component,
debt-to-income
ratios
generally
are
less
powerful
predictors
of
loan
performance
than
other
factors.
This
sample
points
to
both
down
payments
and
credit-bureau
scores
as
better
indicators
of
mortgage
risk.
Layering
of
Risk
Once
the
components
of
a
mortgage
application
have
been
analyzed,
a
lender
must
determine
whether
the
risks
associated
with
collateral,
credit
reputation
and
capacity
combine
to
make
an
investment-quality
mortgage.
Default
probabilities
will
grow
when
multiple
risk
factors
are
present,
which
is
known
as
layering
of
risk.
Risks
can
be
layered
across
the
three
Cs.
For
example,
Freddie
Mac
has
found
that
borrowers
with
both
smaller
down
payments
(collateral)
and
riskier
credit
profiles
experience
dramatically
higher
defaults
than
borrowers
with
only
one
of
these
two
risk
factors
present.
Layering
also
can
appear
within
one
of
the
three
Cs.
In
terms
of
capacity,
for
example,
a
borrower
may
possess
both
a
high
debt-to-income
ratio
and
minimal
reserves.
Along
with
analyzing
layers
of
risk,
the
need
to
identify
strengths
that
offset
those
risks
further
complicates
the
lending
decision.
Yet
this
is
the
job
the
human
underwriter
is
expected
to
perform.
The
Loan
Prospector
Difference
Like
traditional
underwriting,
automated
underwriting
evaluates
mortgage
applications
on
the
basis
of
the
three
Cs—collateral,
credit
reputation
and
capacity.
However,
Loan
Prospector
represents
a
quantum
leap
forward
in
the
ability
to
identify
sound
mortgage
loans.
Its
unique
strength
lies
in
the
ability
to
analyze
a
multitude
of
factors
simultaneously.
Consider,
for
instance,
the
need
to
balance
10
or
more
elements
in
a
loan
application.
The
possible
combinations
of
these
factors
alone
can
number
in
the
thousands.
Human
underwriters
cannot
be
expected
to
assess
them
accurately
and
consistently
from
application
to
application.
In
contrast,
statistically
based
automated
underwriting
systems
such
as
Loan
Prospector
are
designed
to
handle
risk
combinations
numbering
in
the
millions.
Accurate
and
Predictive
The
proof
of
any
underwriting
system
lies
in
its
ability
to
assess
risk.
By
using
our
technology
to
evaluate
previously
purchased
loans,
we
compared
actual
mortgage
performance
with
the
risk
classification
that
would
have
been
provided
by
Loan
Prospector.
The
upshot:
Loan
Prospector
excels
at
identifying
which
loans
will
perform
and
which
will
not.
A
look
at
the
foreclosure
experience
of
Freddie
Mac’s
1994
mortgage
purchases
illustrates
this
point.
Mortgages
classified
as
caution
by
Loan
Prospector,
as
shown
in
Exhibit
7,
entered
foreclosure
at
about
32
times
the
rate
of
those
in
the
accept
category.
Automated
underwriting
combines
the
best
in
statistical
analysis
with
state-of-the-art
technology
that
collects
and
sorts
a
wealth
of
critical
information.
Its
accurate
assessment
of
mortgage
applicants
will
open
the
doors
to
homeownership
for
more
American
families.   
Footnotes:
|
||
|
Good
underwriting
is
the
foundation
of
a
healthy
mortgage
finance
system.
An
accurate
assessment
of
risk
is
key
to
helping
families
buy
homes
they
can
afford
and
keep.
Mistakes
that
lead
to
foreclosure
can
be
devastating
not
only
to
borrowers,
but
also
to
the
neighborhoods
in
which
they
live.
Credit
files
contain
extensive
information
about
open
and
closed
credit
accounts,
called
tradelines.
For
each
tradeline,
for
example,
the
credit
file
tracks
how
much
of
the
available
credit
limit
has
been
used,
the
consumer’s
history
of
repaying
the
account
and
whether
payment
is
up-to-date
or
delinquent.
Credit
files
also
document
the
number
and
nature
of
recent
credit
inquiries,
which
are
requests
by
potential
credit
grantors
to
review
a
credit
file.
In
addition,
credit
files
contain
information
from
public
records,
such
as
declarations
of
bankruptcy
and
unpaid
judgments.
Built
on
the
past
performance
of
similar
loans,
Loan
Prospector
strengthens
the
underwriting
process
by
accurately
assessing
the
layering
of
risks
and
compensating
factors.
In
balancing
the
strengths
in
a
loan
application
against
risk
factors
that
traditionally
lead
to
loan
denial,
Loan
Prospector
can
identify
many
new
families
who
represent
acceptable
mortgage
credit
risks.
1.
