Subscribe
Share
Share
Embed
Market Pulse is a monthly podcast by Equifax, in partnership with Moody’s Analytics. Equifax hosts bring you interviews with industry experts on the latest economic and credit insights that can help drive better business decisions. Whether you’re in financial, mortgage, auto or another service industry, we help make sense of the latest economic conditions that impact you. This podcast series supplements our Market Pulse webinars, which occur on the first Thursday of each month.
Welcome to a special edition
of Market Pulse from Equifax,
where we break down what's shaping the
mortgage industry today and what happens
next.
Welcome to a Lab edition of the
Equifax Market Pulse podcast
recorded at NBA Annual25 in
Las Vegas. I'm Jennifer Henry.
I'm the managing director and Chief
Strategy Officer for government credit
capital markets and housing
strategy at Equifax.
I'm joined today by Andrew Davidson,
President of Andrew Davidson and Company
one of the nation's foremost experts on
mortgage analytics, credit
risks and capital markets,
and also the founder of Financial
Lifecycle Education Companies.
It's great to have you here, Andy.
Oh, Jennifer, it's great
to be here with you today.
All right, Andy, to set the
stage for our listeners,
could you please start by explaining in
simple terms what a credit score is and
what it aims to measure?
Oh, sure. You know anyway,
thanks for mentioning our not-for-profit
financial life cycle education.
What we've done is combine finance and
mathematics for educating high school
students about financial topics.
And one of the areas we've been
working on recently is credit scores.
And incidentally just Friday we,
I was at a math conference basically
talking about a new lesson that we're
creating about how credit scores work.
And what I've learned from that side and
as well as the Andrew Davidson Company
mortgage side,
is that many people don't really
understand what a credit score is.
It's just a number. Mm-Hmm
. To them.
And what we've been trying to do is
delve into what exactly is a credit score
you know, for our mortgage clients
as well as for these students.
And the way I like to think about it
is a credit score is something that's
generated using the data in a data
file and notably only the data
that's in the data file.
And then the credit modeling
companies vantage score and FICO
have done historical analysis of how
people with similar credit files have
performed in the past,
whether or not they've become delinquent
on one or more of their accounts.
Interesting.
And then what happens is they take
that experience and they apply that
model to the data that's in the
credit file now for each individual.
And so once you understand that the credit
score is really a model applied to a
credit file,
you can start to understand a little bit
better the strengths and weaknesses of
that credit score.
That's really interesting. Can you
elaborate a little bit more on that?
So like how do credit scores differ based
upon those different groups of people?
Yeah, so you know,
a lot of people think you have one
credit score and the credit score is
everything to know
about you. But you know,
we now have two different companies
creating credit scores as well.
We have three different bureaus with data.
So the credit score could differ either
based on the data by those credit
bureaus or by which scoring
company that you've used.
And an analogy I like to use is
imagine you're at a company or
two different companies that we're going
to evaluate people based on athletic
ability.
That's interesting.
And one of 'em said we're going
to look primarily at weight.
We'll look at some other things, but
weight is the thing we concentrate on.
And another one says,
height is really the thing that
we determines athletic ability.
And let's say the sport that they chose
to use their scores on or develop their
scores from was throwing a javelin. Okay,
maybe not something everyone
does every day, .
But so they develop a score.
One has more of a weight aspect to it,
the other one has more
of a height aspect to it.
And they both do a really good job of
forecasting how far people can throw a
javelin.
But you can imagine that the score that
is based on weight might have a tall
person and a short person
in the same category, right.
Getting the same javeon distance and
then the one based on height could have a
lighter person and a heavier person
still in that same height category.
And you know, they are going to have
like these different aspects to them.
So they both do a good job of
forecasting javelin throwing,
but let's say you're going to use
it for football or basketball,
it could be that one of those scores is
going to do a lot better than the other.
Right? Because a score isn't just how
well you throw javeon the javelin,
it's also which bucket you were put
in, right? Where you put in a bucket,
other people your same height or you put
in a bucket with other people your same
weight.
And I think that's what people really
haven't understood about credit scores is
that it's all about how the people are
grouped together and how they've come up
with a score for that group, not
just for you individually. That's.
Interesting. So
how do you think that that is
going to impact the industry
now that the FHFA has announced that
they're going to move forward with
acceptance of this
additional credit score,
having vantage now be part of the process?
Yeah, so I think there's sort
of two big issues. One is that,
you know,
people just haven't really started to
focus on this because the process of
adding vantage score has been a lot of
start and stop and there's been other
confusion brought into
that and it's a big effort.
And so people just haven't spent the
time to put in that effort for something
that's not certain yet. But I
think even more fundamentally,
everyone has just viewed credit score as
just a credit score and it's as though
there was only one and as though it was
always correct and that was the credit
score for that person.
And I think this is really going to have
people start to focus more intensely on
what a credit score means,
why you could have different credit
scores from different scoring companies or
why the scores different from scores
calculated on the data from each of
the bureau. And this fundamental
idea that it's data and a model
you know, data, original data
and a model based on that.
And then that model applied to current
data is something people don't really
think about that carefully.
And this is really going to force people
to rethink what a credit score is and
how they use it. And I think
it's going to be a long process.
Yeah. Well that was going to
be my next question is, okay,
so it's announced that effective
immediately the GSEs are going to be
able to accept these new, well at least
Vantage for in addition to the FIO five.
So what is the industry
readiness? Like what do you, you,
you said you think it's
going to take a long time.
Like what do you think readiness is
and what do you think that the industry
needs to be able to get
ready to adopt these scores?
Yeah,
I think it's important to think about
like what do you think of as the industry
and our focus is really primarily
on the investor community and
how they incorporate the data that they
receive from Fannie or Freddy or from
another originator through a trustee
and through a complicated process
into their investment decision.
So on the front end, you know,
it's mostly an operational question
as to whether or not a loan will get
accepted by the GSE and how to transmit
that data for the ultimate investor.
They also have to make decisions
about the risk associated with
that data. And people I say have just
become very comfortable with the idea.
They just have one credit score.
So there's just so many processes
that have to happen first,
the data has to get to the people who
are making these decisions and that means
it's gotta flow through many
different processes. Second,
they need to have models
that they can use on that.
And even as a model provider,
we don't know yet what kind of models
they will need. And it often takes,
you know, months to years for
people to adopt a new model.
And then even beyond that,
people have a certain intuition
from something they've
worked with for years and
years. And developing a new intuition,
a level of comfort with
something new also takes time.
So while it may be mostly in like an
administrative aspect on the origination
side, on the investor side,
it's really a complex process of
information flow modeling and comfort.
And I think all of those
just take a lot of time.
And do you have any perspective on like
what type of information that you think
that the investor community
is lacking at this time?
Yeah, I think so. One is they haven't
really seen the scores live, right?
So there was a data set provided
on Vantage score over a certain
time period, which I
think was very valuable,
but that's not the same as it flowing
through your actual decision process.
The second thing is, as I
sort of mentioned earlier,
I don't think people have really focused
on what a credit score is and why you
could have a vantage score and a
classic FICO score that differ by 50
points, right? And 50 points is a lot.
It could be double the
amount of credit risk.
It could be a five or 10%
change in prepayment speeds.
So understanding that it doesn't
mean one is right and one is wrong,
it's just that there was a different
grouping of people may affect them.
This is further compounded
by lender choice.
So investors are always fearful that
the person selling to them knows more
than they do and is going
to take advantage of them
in what they call adverse
selection. Right? And when
you have adverse selection,
you're going to either say, I'm not
going to participate in that market,
or I'm going to treat things like
they're the worst possible that I could
imagine. and it's people,
it's going to take a while for people
to understand how this lender choice is
working and how much worse
is the performance then if
you were given all of one
score all of the other, or ideally
both, you know, of two, you know,
both scores and these are
just things that are unknown
and investors, you know, they take risk
but they don't like the unknown. Right.
Investors like to know which risks they're
taking and unknown risk is not one of
their favorite things. Yep.
So is there anything that you think that
we could be doing as industry leaders
to bring those groups together
to start sharing information?
Yeah,
I think we really need to open up the
black clocks of what a credit score
is. And it's not like anyone
who's been keeping it a secret.
Mm-Hmm . But it hasn't
really been the focus of education.
Yeah. 'cause people haven't really
needed to know there's one score,
you don't really need to know anything
about it, it just, you get it Right. You.
Didn't need to know. Yeah.
It and it didn't really matter what
that score was 'cause you knew how it
worked. Mm-Hmm . Right.
You know, like even for consumers,
people don't really describe
what a credit score is,
a ranking of log odds ratios.
Mm-Hmm . Right.
Instead they just say, oh, you
know, above this is excellent,
this range is good, this
range is poor. You know,
but investors need to think
about numerically, right?
And then they also need to
have intuition and you know,
I think we need to just start
really educating people.
Like one of the people questions
I ask people is you know,
credit scores are some sort of
measure of risk, maybe a ranking.
How much change in credit score does
it take for the risk to double? Yep.
And I would say almost no one has ever
answered that question correctly when
I've asked them the answer case people
know is around 40 points. Right.
and so if you're looking at credit scores
that might differ by 20 to 40 points,
that could be 50% or you know,
double the amount of risk
if you don't recalibrate.
That's pretty impactful. Yes.
So I'd just shift gears a little bit so
we know that the credit scores are based
upon the underlying data
that's being contributed. Yeah.
So that's the credit data from
Equifax, Experian, and TransUnion.
There's been a lot of pressure in the
market to potentially move from a tri
merge to a buy merge or a
single file credit report.
And you've had the for thought over the
past couple of years to kind of start
looking at that data and to really
understand what the potential
impacts could potentially be or
could be if we were to move to
one of those models.
And I just wanted to get your insights
on what the impacts could be to the
investors, consumers, lenders, and
all of the folks in the process.
Yeah. Just as we were talking about
the difference between, you know,
two different models, the weight
model and the height model.
Mm-Hmm .
The other piece that we talked about in
the definition of a credit score is that
it's based on the data in the credit file.
And originally the credit score companies,
I understand it was sort of had
a regional focus each of them.
And so it was important to combine
them together in order to get sort of a
fuller picture. So the regional
differences are largely gone now.
Mm-Hmm .
But each of the companies has focused
on different types of additional data.
Mm-Hmm .
So we still face the fact that
if you pick data from firm A, B,
or C,
the data in the file that you're using
to calculate the credit score is going to
be different. Right. And so once again,
you can get different credit
scores based on which bureau
and you know,
obviously what we want is to bucket people
or put them in the group that's most
representative of their risk.
We don't want a credit
score that's too high.
'cause You may end up lending to someone
who can't afford that loan and you
don't want a credit score that's too low.
Because it may cost the person
more than it would cost otherwise.
And so bringing in variety of data
sources is actually very useful for that.
Now there may be some situations where
we don't need that much data that the
person has a large thick file at all
three bureaus and you're just not going to
see that much difference.
But as you start to get to borrowers
thinner files or new people you're trying
to bring in where people have
had a credit problem in the past,
the fuller picture may
be more valuable there.
And obviously there's costs
associated with that. And you know,
at some point the market's going to have
to figure out what trade off it wants
between sort of more information
and you know, lower cost. Right.
If you're going to have more
information, it's going to cost you more.
But once again, I think once again,
people haven't really focused on what
the difference is between the credit
bureaus and how they've
operated over time. Yeah.
Now obviously there's people at banks
who are doing credit card lending can go
all of this stuff, you know,
intensely well mm-hmm .
Build their own models.
But most of the investor base certainly
in the securitized market just isn't
that focused on all these details.
So on that note,
the difference in the data from the
credit bureaus is becoming even more
profound because of the director
initiatives to provide more
innovation into the different credit
bureaus.
And I know that we here at Equifax are
really focused on providing alternative
data to the consumers.
And I know our counterparts at Experian
and TransUnion are doing the same.
And I know you've had the opportunity
to look at some of the alternative data,
especially the data coming from
Equifax around telco utility.
And I just wanted to get your
perspective on, you know,
how do you think that that
is impacting borrowers,
especially as we're coming
into these new proposed models?
Yeah. So right now I think
there's an opportunity, right?
So we've been doing one thing the
same way for, you know, many decades.
And if we have to start changing,
I think it's a good time to also
rethink what we're doing and you
know, there's so much more data.
Now the utility data is really great
example and we have a lot of people who
operate in a very different
economy than they used to.
So not everyone has the same job
at the same firm for 20 or 30
years earning a paycheck every other week.
And so as we have people in more
variety of economic circumstances and
we have more data, it would be
useful to bring that in. And,
but we don't think the solution is to
take all of that data and then put it into
one score. Right? So obviously
that would be a useful thing to do,
but that's not really
going to be sufficient.
So we would hope is that in addition
to improved scores with more data and
better data,
we also start to let the investors know
and the credit decision makers know
a little bit more about the underlying
components of how that score was
developed. And the utility data is
like a great opportunity rental data.
There's another one,
there's many places where people
who might have had trouble with some
traditional credit might be doing
really well on something like their
utility bills or their rent,
which may be a much better indication
of how they'll do on a mortgage. Right.
And how they did on either some
medical bill or some small department
store credit card where they had
a dispute, right. So that we,
there's just a fuller picture
that we can get of a borrower.
The other thing that we've really
learned over time is that like the credit
score itself is only about whether
it's not someone becomes delinquent.
Mm-Hmm .
It's not really about whether or not
they can sustain their financial position
over time.
And we've actually found there's some
people who have low credit scores because
they miss payments from time to time
because their income isn't steady. Yeah.
Or because they had an unexpected expense
who actually always become current
again. Right. So that's still going
to produce a low credit score,
but they may still be
a very viable borrower.
For home. Right. And still
have the ability to pay and.
House and actually moving
into home actually maybe
make them more stable. Right.
So that we really need to start thinking
of not just a single measure of credit
risk,
but think of a variety of measures and
how they interact to describe that whole
person, right? Not just their weight,
not just their height, but their weight,
their height, how fast
they can run. You know,
do they have experience
throwing a javelin?
But we can get a lot more information
into this decision. That's.
Great. Well,
do you have anything else that you
would like our listeners to know about
credit, credit scoring?
Anything else that you would like to
leave them with as a parting note?
Yeah, I think as a parting note, I would
just say, Hey, this is an opportunity.
It's a little scary,
it's threatening some of our
business models are being upset.
You know, even on our side,
we don't know exactly what models
we're going to provide to our clients.
People are starting to ask questions,
but I think we should try and take this
as a learning opportunity to create
something better. Obviously we've been
doing something one way, it's fine,
but it's certainly not the best.
So let's try and improve what
we're doing in the future.
Thank you so much for your insights.
We really appreciate your perspectives
on what's happening in the industry.
For our listeners who would like to reach
out to you, where can they find you?
You know, they can find our
website ad-co.com and there's
contact us there on the financial side.
It's fi cycle F-I-C-Y-C-L e.org
if they're interested in the
financial education. Anyway,
thank you so much for having me.
It's always pleasure to talk
to you and work with Equifax.
Likewise, Andrew, you're always a great
partner and we really appreciate it.
The information and opinions provided
in this podcast are intended as general
guidance only and are subject
to change without notice.
The views presented during the podcast
are those of the presenter as of the
date.
This podcast was recorded and do not
necessarily reflect official positions of
Equifax.
Investor analysts should direct
inquiries using the contact us box on the
investor relations section@equifax.com.