ACCESS TO CREDIT FORUM Tuesday, April 3, 2018 Marriott Marquis | Washington, DC Lisa Rice, President & CEO National Fair Housing Alliance @NatFairHouse
ACCESS TO CREDIT FORUM
Tuesday, April 3, 2018
Marriott Marquis | Washington, DC
Lisa Rice, President & CEO
National Fair Housing Alliance
@NatFairHouse
Evidence for Expanding
Access to Credit
NFHA National Access to
Credit Forum
Washington, DC
April 3, 2018
Excessively Tight Credit is Keeping Many Credit Worthy Individuals out of Homeownership
3
Cumulative missing loans, 2009-2015: 6.3 million and growing
Why: zero default risk tolerance 2001-2003: reasonable lending standards
All product risk has been eliminated
Tight borrower underwriting risk profile requirements (credit score, income, etc.)
Today’s loans outperform by historical
standards. There is room to expand.
3
Default rates for loans with FICO < 700 and 80-90 LTV
0%
5%
10%
15%
20%
25%
30%
35%
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
1999-2003
2004
2005
2006
2007
2008
2009-2010
2011-3Q16
Year
Source: Fannie Mae Single Family Loan Level Credit Data.
Barriers to Homeownership PersistSavings/Down Payment: consumers often lack awareness of low down payment options
• 53 percent of renters say they rent because they can’t afford a down payment- more than any other reason
• Only 19 percent of borrowers are aware of low down payment options• High rents make it difficult to save, low interest rates make it hard for basic savings to grow
Credit: access is tight by historical standards
• The median FICO score has drifted up 20 points over the past decade• Both the FHA and VA serve more borrowers at the lower end of the spectrum
Income: stagnant wages and variability in income sources hasn’t caught up to current underwriting
• Rents growing faster than incomes
Affordability: home price appreciation and rising interest rates will cause affordability to decline
• If interest rates rise to 5.5 percent, the share of median income devoted to a mortgage payment will surpass the 2001-03 average
Language: Limited English proficiency• If we control for other factors that influence homeownership (e.g., income, age, and race),
neighborhoods with the highest concentrations of LEP residents have homeownership rates 5 percentage points lower than rates in neighborhoods with the median concentration of LEP residents
Sources: Laurie Goodman et al., (2017) Barriers to Accessing Homeownership: Down Payment, Credit, and Affordability. https://www.urban.org/sites/default/files/publication/94801/barriers_to_accessing_homeownership.pdfLaurie Goodman, Ed Golding, Sarah Strochak (2018) Is Limited English Proficiency a Barrier to Homeownership?https://www.urban.org/urban-wire/new-evidence-shows-limited-english-proficiency-barrier-homeownership 4
Blacks experiencing slowest recovery, losing all
homeownership gains made over last 50 years.No major US city has come even
close to closing the gap between
black and white homeownership.
-8%
-6%
-4%
-2%
0%
2%
4%
6%
8%
Black White Hispanic Others
1970-2000 2000-2016
Percentage-point change in homeownership rate
All gains in black homeownership
since the Fair Housing Act have
been erased since 2000
Source: US Census Bureau, American Community Survey, and Urban Institute.
Source: American Community Survey and Urban Institute.Note: Data as of 2016.
5
Government channels have provided expanded
access to credit with FHA and VA lending
1% 2% 4%7%
35%
50%
17% 18% 18% 15%
24%
8%12% 11% 11% 11%
29% 26%
<640 640–660 660–680 680–700 700–760 ≥760
GSE FHA VA
Credit score
Agency Credit Distributions
6
55%
40%
0%5%2%
12%
85%
1%4%14%
5%
77%
≤80% 80–95% 95–96.5% >96.5%
GSE FHA VA
LTV at origination
Agency Loan-to-Value Ratio Distributions
Source: eMBS and Urban Institute.Note: Based on purchase originations.
Nonbanks playing a key role in opening up access to
credit FICO, LTV and DTI across channels
700
710
720
730
740
750
760
770
780
No
v-1
3
Mar
-14
Jul-
14
No
v-1
4
Mar
-15
Jul-
15
No
v-1
5
Mar
-16
Jul-
16
No
v-1
6
Mar
-17
Jul-
17
No
v-1
7
All Median FICO Bank Median FICO
Nonbank Median FICO
GSE FICO: Bank vs. Nonbank
66
68
70
72
74
76
78
80
82
84
86
88
90
No
v-1
3
Mar
-14
Jul-
14
No
v-1
4
Mar
-15
Jul-
15
No
v-1
5
Mar
-16
Jul-
16
No
v-1
6
Mar
-17
Jul-
17
No
v-1
7
All Median LTV
Bank Median LTV
Nonbank Median LTV
GSE LTV: Bank vs. Nonbank
30
32
34
36
38
40
42
No
v-1
3
Mar
-14
Jul-
14
No
v-1
4
Mar
-15
Jul-
15
No
v-1
5
Mar
-16
Jul-
16
No
v-1
6
Mar
-17
Jul-
17
No
v-1
7
All Median DTI
Bank Median DTI
Nonbank Median DTI
GSE DTI: Bank vs. Nonbank
Source: eMBS and Urban Institute.Note: Based on purchase originations.
LTV
≤ 80 80-95 > 95
FICO
≤ 700 6.2 10.7 11.7
700-750 2.0 4.5 5.4
> 750 0.8 2.4 3.5
≤ 30 30-45 > 45
10.5 10.7 11.1
DTI Breakdown: FICO ≤ 700 and LTV 80-95
LTV
≤ 80 80-95 > 95
FICO
≤ 700 26.0 33.7 38.5
700-750 13.4 20.4 21.5
> 750 5.3 11.6 13.8
≤ 30 30-45 > 45
25.6 32.8 39.5
DTI Breakdown: FICO ≤ 700 and LTV 80-95
2007
LTV
≤ 80 80-95 > 95
FICO
≤ 700 4.6 4.8 3.2
700-750 1.5 2.0 2.3
> 750 0.4 0.8 1.0
≤ 30 30-45 > 45
1.2 2.4 1.1
DTI Breakdown: FICO > 700 and ≤ 750,
LTV 80-95
2011
2002
How predictive is DTI? Delinquency rates by year of origination
and credit characteristics
Source: Fannie Mae and Freddie Mac Single Family Loan-Level Credit Data8
11
The Evidence for Expanding Access to Credit
Cheryl Young,Senior Economist
Trulia
22
CONSUMER BRANDS
BUSINESS BRANDS
3
FinancialServices
Health Care
HealthyFood
Active Life
Research Focus: Where You Live Matters
4
Mapping and Measuring Proximityto Financial Services
Metros:● Atlanta● Detroit● Houston● Oakland
Traditional Financial Services:● Banks● Credit Unions● Mortgage Lenders
Alternative Financial Services:● Check-Cashing / Pay-Day
Loans ● Debt Relief Services● Installment Loans● Title Loans
55
Analysis
6
Majority-minority census tracts across Atlanta, Houston, Oakland and Detroit have roughly 33% fewer traditional banking establishments than majority-white tracts.
Traditional Financial Services, Findings Across All Metros
Suggest we add labels on each bar chart
Traditional Financial Service Establishments Per 10,000 People
7
Houston has the largest disparity in traditional financial service establishments between majority-minority and majority-white census tracts.
Traditional Financial Services, Findings by Metro
* Difference is not statistically significant.
Traditional Financial Service Establishments Per 10,000 People, by Metro
8
Traditional Financial Services, Findings by Census Tract Type
The extent of racial disadvantage in terms of proximity to traditional financial services varies by metro; in Houston, majority-black tracts are the most disadvantaged. In Detroit, it’s majority-Hispanic.
Traditional Financial Services Establishments Per 10,000 People, by Census Tract Type
9
Alternative Financial Services, Findings Across All Metros
There are twice as many alternative banking service establishments in majority-minority census tracts than in majority-white census tracts.
Alternative Financial Services Establishments Per 10,000 People
10
Alternative Financial Services, Findings by Metro
Suggest we add labels on each bar chart
The largest disparity is in Oakland, where there are 2.3 times as many alternative financial service providers in majority-minority tracts than white tracts.
Alternative Financial Services EstablishmentsPer 10,000 People, by Metro
11
Alternative Financial Services, Findings by Census Tract Type
Across all four metros, majority-white census tracts have significantly fewer alternative financial service establishments than majority-black or Hispanic tracts.
Alternative Financial Services EstablishmentsPer 10,000 People, by Census Tract Type
12
Key Takeaway: Where you live matters.
1313
Thank you.
Modernizing FHA to Improve
Consumer Access
Laurie Goodman
Co-Director, Housing Finance Policy
Center
Urban Institute
National Access to Credit
Forum
Washington, DC
April 3, 2018
Originators charge more for weaker borrowers; even
though FHA does not do risk-based pricing
3.0
3.2
3.4
3.6
3.8
4.0
4.2
4.4
4.6
4.8
5.0< 640 640 - <660 660 - <680 680 - <700
700 - <760 ≥ 760 All
Interest Rate
Source: eMBS and Urban Institute.
Cost of Servicing
3
$59 $77 $90 $96 $114 $156 $156 $181 $163
$482
$704
$911
$1,246
$2,009
$2,358
$1,965
$2,386
$2,113
2008 2009 2010 2011 2012 2013 2014 2015 2016
Performing Nonperforming
• Servicing nonperforming loans is much more expensive than servicing performing
loans.
• FHA servicing is still more expensive; servicing FHA nonperforming loans is 3 times as
expensive as servicing GSE nonperforming loans.
Source: Mortgage Bankers Association Servicing Operations and Forum and Urban Institute.
Housing Counseling Facilitates LMI Lending
Overcoming the Cost of OriginationUsing Housing Counselors
Reducing Cost of LMI Originations
• Cost of Origination has risen by $3,000 since 2008
• Effectively Using Housing Counseling Saves Money in 4 ways
• UnidosUS has models that links together our network allowing us to
increased counseling capacity and market reach
• Layering modes of counseling to reduce cost and increase effectiveness of
communication.
Spreading the Aspiration ofOwning Home
Communities of color and LMI families generally have lost wealth since the crisis – makes families cautious.
There is distrust of messengers around homeownership.
Many prime lenders have reduced their focused in LMI markets as the segments revenue potential has significantly decreased.
Counseling can help in raising aspirations for homeownership and its potential for wealth promotion in our communities.
Access to Homeownership Supported By MI
We Help People Buy a House, and Keep it Their Home
Forty Percent With Incomes Less Than $75,000
More Than 50% Are First Time Homebuyers
Accessible Through Prudent Low Down Payment Lending with MI
Support Access By Alignment to GSE Programs, Education, and Training
1
0
50,000
100,000
150,000
200,000
250,000
4Q16 1Q17 2Q17 3Q17
MI helped nearly860,000 HOMEOWNERS purchase or refinance
a mortgage in the past year
Source: USMI Member Companies
Improve Access Through Fee Re-Evaluation
0
20
40
60
80
2007 2009 2011 2013 2015 2017
Annualized Fee Example680 Credit Score / 90 LTV
Annual Fee Upfront Fee Adverse Market
Credit Score
LT
V
2We Help People Buy a House, and Keep it Their Home
Fees Impact Access And Limit Choice
Unchanged View of Remote Exposure Since Crisis
Recommend Re-evaluation To Increase Access to Efficient Execution
GSE/MI Eligibility Matrix
Cells Impacted by Fees
Borrowers MutualInsurance Fund
Richard Cooperstein
Access to Credit ForumNational Fair Housing AllianceWashington DC
April 3, 2018
© 2018. Andrew Davidson & Co., Inc. All Rights Reserved
What’s the Goal: Really?
LIQUIDITY
CREDITACCESS
WEALTH
© 2018. Andrew Davidson & Co., Inc. All Rights Reserved
So What’s Better?
75 new owners & 5 lose their homes OR 10,000 new owners &
2000 lose their homes
000’s new owners AND < 10% lose their homes
3% Down OR 2.5% Reserves
SO…
© 2018. Andrew Davidson & Co., Inc. All Rights Reserved
The Groundwork
1. Educatea) Get the under-banked into the system
b) Get merchants to report their payments
2. Data Accessa) Build better scores for under-banked and
under-served populations and markets• FICO XD, Vantage 4.0, Finicity, etc.
b) A data utility with historic performance including new and old scores
© 2018. Andrew Davidson & Co., Inc. All Rights Reserved
Borrowers Mutual Insurance Fund
WHAT: Funds for short-term income interruptions and major maintenance items
WHO: Under-served populations & markets
WHY: Keep those with fragile liquidity in homes
HOW: 2% borrower reserves instead of 3% down + 0.5% from the GSEs: from 10 bps Guar. Fee on all loans
WHERE: Administered by: servicers, Board of non-profits and the GSEs?
© 2018. Andrew Davidson & Co., Inc. All Rights Reserved
Outcomes
RISK: Might be lower than low down-payments
FUNDING: 80% by the borrowers themselves
EFFICIENCY: Mutual structure diversifies risk and extends coverage (4%? per borrower)
FOCUS: The key risk facing new buyers: volatility
ALIGNMENT: Targeted borrowers, GSEs, Budget
© 2018 CoreLogic, Inc. All Rights Reserved. Proprietary & Confidential. One CoreLogic
Evidence for
Expanding
Access to
Credit
April 03, 2018
1
NFHA
© 2018 CoreLogic, Inc. All Rights Reserved. Proprietary & Confidential. One CoreLogic 2
Source: CoreLogic
Home Purchase Lending to Low Credit Score
Borrowers Collapsed and Has Not Recovered
0%
5%
10%
15%
20%
25%
30%
35%Ja
n-9
6
De
c-9
6
No
v-9
7
Oct-
98
Se
p-9
9
Au
g-0
0
Ju
l-0
1
Ju
n-0
2
Ma
y-0
3
Ap
r-0
4
Ma
r-0
5
Fe
b-0
6
Ja
n-0
7
De
c-0
7
Nov-0
8
Oct-
09
Se
p-1
0
Au
g-1
1
Ju
l-1
2
Jun-1
3
Ma
y-1
4
Ap
r-1
5
Ma
r-1
6
Fe
b-1
7
Jan-1
8
Sh
are
of P
urc
ha
se
Ori
gin
atio
ns to
< 6
40
Cre
dit S
co
re B
orr
ow
ers
© 2018 CoreLogic, Inc. All Rights Reserved. Proprietary & Confidential. One CoreLogic 3
> 6%
4% - 6%
2% - 4%
< 2%
Low Credit Purchase Lending Lowest in High Cost MarketsShare of Purchase Loans with Credit Scores < 640
© 2018 CoreLogic, Inc. All Rights Reserved. Proprietary & Confidential. One CoreLogic 4
Source: CoreLogic
0.0%
0.2%
0.4%
0.6%
0.8%
1.0%
1.2%
1.4%
500 550 600 650 700 750 800 850
Purchase Credit Score
2007
Originations
Applications
0.0%
0.2%
0.4%
0.6%
0.8%
1.0%
1.2%
1.4%
500 550 600 650 700 750 800 850
Purchase Credit Score
2017
Originations
Applications
Collapse in Credit Scores Due More to Demand Not Supply
© 2018 CoreLogic, Inc. All Rights Reserved. Proprietary & Confidential. One CoreLogic 5
Source: CoreLogic
Market is Producing a Deficit of 130,000 Low Income Loans Per YearLow-to-Moderate Area Share of Purchase Loans
14%
15%
16%
17%
18%
19%
20%
21%
22%
23%
24%
3 Percentage Points
Below Normal
© 2018 CoreLogic, Inc. All Rights Reserved. Proprietary & Confidential. One CoreLogic 6
Source: CoreLogic
Market is Producing a Deficit of 250,000 Minority Loans Per YearMinority Area Share of Purchase Loans Remains Low
14%
16%
18%
20%
22%
24%
26%
28%
30%
7 Percentage Points
Below Normal
© 2018 CoreLogic, Inc. All Rights Reserved. Proprietary & Confidential. One CoreLogic 7
Affordability Gap is Rapidly RisingSupply Gap Between Entry Home Prices and Luxury Home Prices Rapidly Widening
Source: CoreLogic, Home Price Index
100
140
180
220
260Ja
n-0
0
Oct-
00
Ju
l-0
1
Ap
r-0
2
Ja
n-0
3
Oct-
03
Ju
l-0
4
Ap
r-0
5
Ja
n-0
6
Oct-
06
Ju
l-0
7
Ap
r-0
8
Ja
n-0
9
Oct-
09
Ju
l-1
0
Ap
r-1
1
Ja
n-1
2
Oct-
12
Ju
l-1
3
Ap
r-1
4
Ja
n-1
5
Oct-
15
Ju
l-1
6
Ap
r-1
7
Ja
n-1
8
Entry Home
Price Index
Luxury Home
Price Index
© 2018 CoreLogic, Inc. All Rights Reserved. Proprietary & Confidential. One CoreLogic 8
Where to find more information
Look for regular updates to our housing forecast,
commentary and data at
https://www.corelogic.com/insights-index.aspx
@CoreLogicEcon
@TheSamKhater
The views, opinions, forecasts and estimates herein are those of the CoreLogic Office of the Chief Economist, are subject
to change without notice and do not necessarily reflect the position of CoreLogic or its management. The Office of the
Chief Economist makes every effort to provide accurate and reliable information, however, it does not guarantee
accuracy, completeness, timeliness or suitability for any particular purpose.
CORELOGIC and the CoreLogic logo are trademarks of CoreLogic, Inc. and/or its subsidiaries.