Latest / Market Pulse / Driving Efficiency and Resilience in the Mortgage Industry

Transcript

Driving Efficiency and Resilience in the Mortgage Industry · Market Pulse

0:02
Welcome to a special edition
0:05
where we break down what's shaping the
0:09
next.
0:14
Welcome to a Lab edition of the
0:19
recorded at NBA Annual25 in
0:24
I'm the managing director and Chief
0:28
capital markets and housing
0:32
I'm joined today by Andrew Davidson,
0:34
President of Andrew Davidson and Company
0:40
mortgage analytics, credit
0:43
and also the founder of Financial
0:48
It's great to have you here, Andy.
0:50
Oh, Jennifer, it's great
0:52
All right, Andy, to set the
0:55
could you please start by explaining in
0:59
what it aims to measure?
1:00
Oh, sure. You know anyway,
1:02
thanks for mentioning our not-for-profit
1:07
What we've done is combine finance and
1:12
students about financial topics.
1:14
And one of the areas we've been
1:18
And incidentally just Friday we,
1:21
I was at a math conference basically
1:25
creating about how credit scores work.
1:28
And what I've learned from that side and
1:32
mortgage side,
1:33
is that many people don't really
1:36
It's just a number. Mm-Hmm
1:39
And what we've been trying to do is
1:43
you know, for our mortgage clients
1:47
And the way I like to think about it
1:51
generated using the data in a data
1:56
that's in the data file.
1:58
And then the credit modeling
2:03
have done historical analysis of how
2:08
performed in the past,
2:09
whether or not they've become delinquent
2:13
Interesting.
2:14
And then what happens is they take
2:19
model to the data that's in the
2:23
And so once you understand that the credit
2:28
credit file,
2:29
you can start to understand a little bit
2:32
that credit score.
2:34
That's really interesting. Can you
2:37
So like how do credit scores differ based
2:41
Yeah, so you know,
2:43
a lot of people think you have one
2:46
everything to know
2:50
we now have two different companies
2:53
We have three different bureaus with data.
2:55
So the credit score could differ either
3:00
bureaus or by which scoring
3:04
And an analogy I like to use is
3:08
two different companies that we're going
3:13
ability.
3:14
That's interesting.
3:15
And one of 'em said we're going
3:20
We'll look at some other things, but
3:23
And another one says,
3:24
height is really the thing that
3:28
And let's say the sport that they chose
3:32
scores from was throwing a javelin. Okay,
3:35
maybe not something everyone
3:37
But so they develop a score.
3:40
One has more of a weight aspect to it,
3:43
the other one has more
3:45
And they both do a really good job of
3:50
javelin.
3:51
But you can imagine that the score that
3:56
person and a short person
3:59
Getting the same javeon distance and
4:03
lighter person and a heavier person
4:08
And you know, they are going to have
4:12
So they both do a good job of
4:16
but let's say you're going to use
4:20
it could be that one of those scores is
4:24
Right? Because a score isn't just how
4:28
it's also which bucket you were put
4:31
other people your same height or you put
4:35
weight.
4:36
And I think that's what people really
4:40
that it's all about how the people are
4:43
with a score for that group, not
4:47
Interesting. So
4:50
how do you think that that is
4:55
now that the FHFA has announced that
4:59
acceptance of this
5:01
having vantage now be part of the process?
5:04
Yeah, so I think there's sort
5:10
you know,
5:10
people just haven't really started to
5:14
adding vantage score has been a lot of
5:17
confusion brought into
5:21
And so people just haven't spent the
5:24
that's not certain yet. But I
5:29
everyone has just viewed credit score as
5:33
there was only one and as though it was
5:37
score for that person.
5:39
And I think this is really going to have
5:42
what a credit score means,
5:44
why you could have different credit
5:48
why the scores different from scores
5:53
the bureau. And this fundamental
5:59
you know, data, original data
6:02
And then that model applied to current
6:05
think about that carefully.
6:07
And this is really going to force people
6:11
how they use it. And I think
6:15
Yeah. Well that was going to
6:18
so it's announced that effective
6:23
able to accept these new, well at least
6:28
So what is the industry
6:31
you said you think it's
6:34
Like what do you think readiness is
6:39
needs to be able to get
6:42
Yeah,
6:42
I think it's important to think about
6:46
and our focus is really primarily
6:52
how they incorporate the data that they
6:56
another originator through a trustee
7:02
into their investment decision.
7:06
it's mostly an operational question
7:09
accepted by the GSE and how to transmit
7:14
They also have to make decisions
7:19
that data. And people I say have just
7:23
They just have one credit score.
7:25
So there's just so many processes
7:28
the data has to get to the people who
7:33
it's gotta flow through many
7:37
they need to have models
7:41
And even as a model provider,
7:43
we don't know yet what kind of models
7:47
you know, months to years for
7:51
And then even beyond that,
7:52
people have a certain intuition
7:56
years. And developing a new intuition,
7:59
a level of comfort with
8:03
So while it may be mostly in like an
8:08
side, on the investor side,
8:10
it's really a complex process of
8:15
And I think all of those
8:18
And do you have any perspective on like
8:21
that the investor community
8:24
Yeah, I think so. One is they haven't
8:29
So there was a data set provided
8:34
time period, which I
8:37
but that's not the same as it flowing
8:42
The second thing is, as I
8:45
I don't think people have really focused
8:50
could have a vantage score and a
8:55
points, right? And 50 points is a lot.
8:58
It could be double the
9:01
It could be a five or 10%
9:05
So understanding that it doesn't
9:10
it's just that there was a different
9:14
This is further compounded
9:19
So investors are always fearful that
9:24
than they do and is going
9:28
selection. Right? And when
9:31
you're going to either say, I'm not
9:34
or I'm going to treat things like
9:37
imagine. and it's people,
9:40
it's going to take a while for people
9:43
working and how much worse
9:48
score all of the other, or ideally
9:52
both scores and these are
9:58
and investors, you know, they take risk
10:02
Investors like to know which risks they're
10:06
their favorite things. Yep.
10:09
So is there anything that you think that
10:13
to bring those groups together
10:18
Yeah,
10:18
I think we really need to open up the
10:23
is. And it's not like anyone
10:26
Mm-Hmm . But it hasn't
10:30
Yeah. 'cause people haven't really
10:33
you don't really need to know anything
10:35
Didn't need to know. Yeah.
10:37
It and it didn't really matter what
10:40
worked. Mm-Hmm . Right.
10:43
people don't really describe
10:46
a ranking of log odds ratios.
10:49
Instead they just say, oh, you
10:52
this range is good, this
10:55
but investors need to think
10:58
And then they also need to
11:02
I think we need to just start
11:05
Like one of the people questions
11:10
credit scores are some sort of
11:14
How much change in credit score does
11:18
And I would say almost no one has ever
11:23
I've asked them the answer case people
11:28
and so if you're looking at credit scores
11:32
that could be 50% or you know,
11:34
double the amount of risk
11:38
That's pretty impactful. Yes.
11:41
So I'd just shift gears a little bit so
11:45
upon the underlying data
11:49
So that's the credit data from
11:53
There's been a lot of pressure in the
11:58
merge to a buy merge or a
12:02
And you've had the for thought over the
12:07
looking at that data and to really
12:12
impacts could potentially be or
12:17
one of those models.
12:18
And I just wanted to get your insights
12:23
investors, consumers, lenders, and
12:27
Yeah. Just as we were talking about
12:30
two different models, the weight
12:34
Mm-Hmm .
12:35
The other piece that we talked about in
12:38
it's based on the data in the credit file.
12:41
And originally the credit score companies,
12:44
I understand it was sort of had
12:47
And so it was important to combine
12:51
fuller picture. So the regional
12:55
Mm-Hmm .
12:56
But each of the companies has focused
13:00
Mm-Hmm .
13:01
So we still face the fact that
13:06
or C,
13:07
the data in the file that you're using
13:11
be different. Right. And so once again,
13:13
you can get different credit
13:18
and you know,
13:19
obviously what we want is to bucket people
13:23
representative of their risk.
13:26
We don't want a credit
13:29
'cause You may end up lending to someone
13:32
don't want a credit score that's too low.
13:35
Because it may cost the person
13:39
And so bringing in variety of data
13:45
Now there may be some situations where
13:49
person has a large thick file at all
13:53
see that much difference.
13:55
But as you start to get to borrowers
13:58
to bring in where people have
14:02
the fuller picture may
14:06
And obviously there's costs
14:09
at some point the market's going to have
14:12
between sort of more information
14:16
If you're going to have more
14:19
But once again, I think once again,
14:22
people haven't really focused on what
14:26
bureaus and how they've
14:30
Now obviously there's people at banks
14:33
all of this stuff, you know,
14:36
Build their own models.
14:37
But most of the investor base certainly
14:42
that focused on all these details.
14:44
So on that note,
14:45
the difference in the data from the
14:49
profound because of the director
14:52
initiatives to provide more
14:57
bureaus.
14:58
And I know that we here at Equifax are
15:03
data to the consumers.
15:04
And I know our counterparts at Experian
15:09
And I know you've had the opportunity
15:12
especially the data coming from
15:16
And I just wanted to get your
15:20
how do you think that that
15:22
especially as we're coming
15:28
Yeah. So right now I think
15:31
So we've been doing one thing the
15:35
And if we have to start changing,
15:38
I think it's a good time to also
15:43
know, there's so much more data.
15:45
Now the utility data is really great
15:49
operate in a very different
15:53
So not everyone has the same job
15:57
years earning a paycheck every other week.
16:00
And so as we have people in more
16:05
we have more data, it would be
16:10
but we don't think the solution is to
16:13
one score. Right? So obviously
16:17
but that's not really
16:19
So we would hope is that in addition
16:24
better data,
16:25
we also start to let the investors know
16:29
a little bit more about the underlying
16:34
developed. And the utility data is
16:39
There's another one,
16:39
there's many places where people
16:43
traditional credit might be doing
16:48
utility bills or their rent,
16:50
which may be a much better indication
16:53
And how they did on either some
16:58
store credit card where they had
17:03
there's just a fuller picture
17:06
The other thing that we've really
17:10
score itself is only about whether
17:14
Mm-Hmm .
17:14
It's not really about whether or not
17:19
over time.
17:20
And we've actually found there's some
17:24
they miss payments from time to time
17:28
Or because they had an unexpected expense
17:31
again. Right. So that's still going
17:35
but they may still be
17:38
For home. Right. And still
17:40
House and actually moving
17:44
So that we really need to start thinking
17:49
risk,
17:50
but think of a variety of measures and
17:53
person, right? Not just their weight,
17:58
their height, how fast
18:00
do they have experience
18:02
But we can get a lot more information
18:05
Great. Well,
18:08
do you have anything else that you
18:11
credit, credit scoring?
18:14
Anything else that you would like to
18:18
Yeah, I think as a parting note, I would
18:22
It's a little scary,
18:23
it's threatening some of our
18:28
You know, even on our side,
18:29
we don't know exactly what models
18:33
People are starting to ask questions,
18:35
but I think we should try and take this
18:38
something better. Obviously we've been
18:42
but it's certainly not the best.
18:43
So let's try and improve what
18:46
Thank you so much for your insights.
18:48
We really appreciate your perspectives
18:52
For our listeners who would like to reach
18:55
You know, they can find our
19:00
contact us there on the financial side.
19:03
It's fi cycle F-I-C-Y-C-L e.org
19:08
financial education. Anyway,
19:11
It's always pleasure to talk
19:14
Likewise, Andrew, you're always a great
19:19
The information and opinions provided
19:23
guidance only and are subject
19:26
The views presented during the podcast
19:30
date.
19:30
This podcast was recorded and do not
19:34
Equifax.
19:37
Investor analysts should direct
19:41
investor relations section@equifax.com.