Impact of QRM
Community
Mortgage
Banking Project
10997 January 27, 2012 2:24 PM
Executive Summary
We strongly support the intention of the Qualified Residential Mortgage
(QRM) exemption from the Risk Retention provisions of the Dodd-Frank
Wall Street Reform and Consumer Protection Act. We believe Congress
effectively threaded the needle by creating an exemption from risk
retention requirements for well-underwritten, fully documented, sound
and sustainable residential mortgages. But it is essential that any risk
retention requirement protect investors from unsustainable lending
practices while not unnecessarily restricting consumers’ access to credit.
Securitization is a valuable liquidity channel for providing borrowers with
affordable mortgage credit. Portfolio lending does not have the capacity
to meet market demands, and the future of the government sponsored
enterprises is uncertain. Without a viable private securitization process,
the nation’s housing finance needs cannot be met. Importantly, aside
from risk retention and QRM requirements, all residential loans, whether
securitized, sold or held in portfolio, will need to meet an “ability to
repay” standard [Qualified Mortgage (QM)] test. The QM standard will
effectively define the broader market for residential loans, with QRM
loans being a narrower subset of the market.
The proposed rule implementing the risk retention and QRM provisions
goes beyond what Congress intended and would drastically limit
affordable mortgage financing options for moderate-income families,
first-time borrowers, borrowers of color and many others.
This presentation uses independent, reliable data to make the
following key points:
• The proposed regulation will hurt consumers by limiting access
to credit for well-qualified borrowers.
• Excluding risky products and requiring sound underwriting, full
documentation and verification — all part of the proposed QRM
standard and a basic requirement of the related QM standard —
addresses the most important factors that drive default risk.
• The proposed down payment, loan-to-value (LTV) and debt-to
income (DTI) requirements are unnecessary and not worth the
societal cost of excluding far too many borrowers from the most
affordable loans.
−− The impact of these standards will be worse for minorities,
first-time borrowers and refinancing homeowners with
limited equity and threatens to disturb the balance between
the rental and homeownership markets.
−− The alternative proposal, a 10 percent down payment
requirement, would be as bad, if not worse than a 20 percent
requirement, as it would reduce liquidity even further.
−− The proposed credit standard will not act as intended. Borrowers
with stronger credit may not satisfy this credit standard, while
those with weaker credit may. Operationally, it will also be difficult
for lenders to handle voluminous credit report data.
• Because government issued/guaranteed loans are exempt from
the rule, prescribing hard-wired down payment, LTV and DTI
standards will in effect shift mortgage risk to the government
because the majority of loans will flow to them rather than
private lenders utilizing mortgage insurance and other credit
enhancements to manage risk and underwrite sustainable loans
for consumers.
• Most new households over the past couple of years have opted
to rent. Rental vacancy rates have been declining. Imposing
additional hurdles to homeownership through the imposition
of an overly narrow QRM could keep more households as
renters, potentially leading to upward pressure on rents.
Impact of QRM
2
We Strongly Support the Elimination of Riskier Loan Features
as Currently Proposed in QM and QRM Regulations
• No negative amortization loans
• No interest-only loans
• No balloon loans
• Loans cannot exceed 30 years in maturity
• No prepayment penalties
• Income, employment, assets and other debts must be verified [ie., no no-doc or low-doc loans]
• Underwriting for ARMs must use fully indexed interest rate
• Total points and fees cannot exceed three percent of loan amount
• Stable payment history
• ARM reset caps of two percentage points per year
• No investor loans — only loans for owner-occupied properties are QRM
• No piggyback seconds
Impact of QRM
3
Primary Concerns
• Risk retention rules do not need to legislate underwriting criteria
−− LTV and DTI do not belong in the rule
−− Product type and documentation requirements in the current proposal address the issue
• LTV mandates will limit access and affordability for
−− First-time homebuyers
−− Minorities
−− Homeowners without 25 percent equity will be unable to refinance
• QRM with LTV and DTI mandates will exclude broad private market participation
• The proposed rule does not reflect the clear intent of Congress to include low down payment loans
that are insured by private mortgage insurance within the QRM exemption.
• Costs of non-QRM loans are being underestimated
Impact of QRM
4
More Than 80 Percent of GSE Business from 1997–2009 Would Not Have Been QRM
Percent of all mortgages that would NOT have met all requirements under the proposed QRM standard, by year of origination
100
80
79.6%
76.7%
80.5%
83.6%
80.6%
83.0%
77.6%
85.6%
88.5%
89.3%
82.6%
75.4%
69.5%
60
40
20
0
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
Source: FHFA.
Impact of QRM
5
Impact of Removing LTV Requirement Relative to Performance of Non-Traditional Loans
QM and QRM Eliminate Riskier Loans
Percent
50
40
30
20
10
0
2005
2006
2007
2008
Origination Year
QRM Delinquency Rate a
QRM — Removing LTV Requirementa
Fannie Mae Serious Delinquency Rate b
Alt-A ARM c
Option ARM c
Subprime ARM c
a. Ever-90 Delinquency Rate
b. Serious Delinquency Rate as of Q2 2011
c. Cumulative Default Rate as of 6/3/11
Note: The FHFA applied a FICO score of 690 as a proxy for the credit criteria in the proposal.
This factor contributed to the relatively strong performance of high LTV loans.
Source: FHFA, Citi, Fannie Mae.
Impact of QRM
6
Conservative Approach: Assumes Every Dollar of Savings Goes Toward Down Payment, 9% Savings Rate (CEX)
Seattle, WA
Median HP: $293K
Income: $58,990 / year
Monthly Savings: $376
Years: 13 / 6.5
20% DP: $58,572
10% DP: $29,286
Chicago, IL
Median HP: $167K
Income: $46,781 / year
Monthly Savings: $298
Years: 9 / 4.5
20% DP: $33,380
10% DP: $16,690
San Francisco, CA
Median HP: $495K
Income: $70,040 / year
Monthly Savings: $447
Years: 18 / 9
20% DP: $98,946
10% DP: $49,473
Los Angeles, CA
Median HP: $304K
Income: $54,828 / year
Monthly Savings: $350
Years: 15 / 7.5
20% DP: $60,884
10% DP: $30,442
Phoenix, AZ
Median HP: $130K
Income: $48,881 / year
Monthly Savings: $312
Years: 7 / 3.5
20% DP: $26,081
10% DP: $13,040
Philadelphia, PA
Median HP: $208K
Income: $36,669 / year
Monthly Savings:$234
Years: 15 / 7.5
20% DP: $41,624
10% DP: $20,812
Houston, TX
Median HP: $154K
Income: $42,797 / year
Monthly Savings: $273
Years: 9 / 4.5
20% DP: $30,737
10% DP: $15,368
Birmingham, AL
Median HP: $140K
Income: $31,704 / year
Monthly Savings:$202
Years: 12 / 6
20% DP: $28,090
10% DP:$14,045
Notes: “DP”: Down Payment. “Median HP”: Median Home Price. “Income”: Median household income,
“Years”: Years to Save for a 20% Down Payment / 10% Down Payment on a Median-priced Home.
Sources: National Association of Realtors, Census Bureau (American Community Survey, American Housing Survey),
Federal Reserve, Bureau of Labor Statistics (Consumer Expenditure Survey)
Impact of QRM
7
Conservative Approach: Assumes Every Dollar of Savings Goes Toward Down Payment, 5% Savings Rate (BEA)
Seattle, WA
Median HP: $293K
Income: $58,990 / year
Monthly Savings: $209
Years: 23 / 12
20% DP: $58,572
10% DP: $29,286
Chicago, IL
Median HP: $167K
Income: $46,781 / year
Monthly Savings: $166
Years: 17 / 8
20% DP: $33,380
10% DP: $16,690
San Francisco, CA
Median HP: $495K
Income: $70,040 / year
Monthly Savings: $248
Years: 33 / 17
20% DP: $98,946
10% DP: $49,473
Los Angeles, CA
Median HP: $304K
Income: $54,828 / year
Monthly Savings: $194
Years: 26 / 13
20% DP: $60,884
10% DP: $30,442
Phoenix, AZ
Median HP: $130K
Income: $48,881 / year
Monthly Savings: $173
Years: 13 / 6
20% DP: $26,081
10% DP: $13,040
Philadelphia, PA
Median HP: $208K
Income: $36,669 / year
Monthly Savings:$130
Years: 27 / 13
20% DP: $41,624
10% DP: $20,812
Houston, TX
Median HP: $154K
Income: $42,797 / year
Monthly Savings: $152
Years: 17 / 8
20% DP: $30,737
10% DP: $15,368
Birmingham, AL
Median HP: $140K
Income: $31,704 / year
Monthly Savings:$112
Years: 21 / 10
20% DP: $28,090
10% DP:$14,045
Notes: “DP”: Down Payment. “Median HP”: Median Home Price. “Income”: Median household income,
“Years”: Years to Save for a 20% Down Payment / 10% Down Payment on a Median-priced Home.
Sources: National Association of Realtors, Census Bureau (American Community Survey, American Housing Survey),
Federal Reserve, Bureau of Labor Statistics (Consumer Expenditure Survey), Bureau of Economic Analysis
Impact of QRM
8
High LTV Lending Is Not a Recent Innovation — Heavily Used by Borrowers of Color
Percent of Homebuyers with LTVs of 90% or Higher
Percent of Homebuyers with LTVs of 95% or Higher
80
60
72%
70
63%
51%
50
44%
60
40
50
44%
40
30
43% 42%
35%
40%
34%
30%
31%
30
20
27% 27%
23%
21%
25%
17%
20
10
10
0
0
All Households
African-American Buyers
1989–97
1997–2005
Hispanic Buyers
All Households
2009
African-American Buyers
1989–97
1997–2005
Hispanic Buyers
2009
Source: MBA Analysis of Census Bureau American Housing Survey Data.
• High LTV loans have been an important component of the home purchase market for decades.
• Borrowers of color have made much greater use of these loans.
• All homebuyers have made more use following the loss of wealth in the recession.
Impact of QRM
9
If Available, Non-QRM Loans Will Cost Significantly More than QRM Loans
Liquidity
• The agency MBS market is extremely liquid due to:
−− The volume of particular securities outstanding.
−− Limited variation across these securities, i.e., the underlying mortgages are homogeneous.
• Liquidity differs considerably by issuer and type of MBS.
−− Fannie MBS trade at a higher price due to better liquidity.
• The non-agency market for securitized products is much less liquid. The result is less liquidity and higher rates for borrowers.
• The private non-QRM market will be left with odd lots of high-risk, heterogeneous loans.
Cost of Capital
• Securitizer / originator will bear costs of equity capital and costs of financing the five-percent risk retention requirement.
• Costs will increase with volume, limiting origination capacity.
Capacity
• Balance sheet constraints will be binding for portfolio lenders.
• Investors must consider counter party strength of securitizers.
Source: SIFMA compilation of Federal Reserve Bank of New York, Municipal Securities Rulemaking Board, FINRA, TRACE and IDC data
Impact of QRM
10
How the Markets Will Treat Non-QRM Loans
•
The proposed QRM definition would be extremely restrictive, limiting LTV on purchase loans to 80 or below and DTI to 28 / 36. In the
proposal, the regulators also included an alternative, 90 LTV 33 / 41 DTI.
• The Alternative QRM approach would bifurcate the home purchase market, negatively impacting the liquidity of non-QRM loans,
thus raising rates to borrowers.
• Those outside the QRM box are going to be less liquid for a variety of reasons.
−− Loans in the QRM box are necessarily going to be more homogeneous. Those in the non-QRM box are going to be more
heterogeneous, with respect to LTV, DTI, product type and other underwriting parameters. For example, investor loans are going
to be confined to the non-QRM space.
−− This means that the QRM pools are more likely to have greater trading liquidity, while non-QRM pools are much more likely to
trade specified, with reduced liquidity.
−− The heterogeneity would reflect both more variation in credit quality and greater variation in convexity / baseline prepay speeds.
−− Note that LTV and DTI are data elements that are amenable to disclosure, while other data elements may not be.
Impact of QRM
11
Impacts on Market Liquidity
DTI
QM Box
33 / 41
Alternative QRM Box
28 / 36
Proposed QRM Box
80
90
LTV
Note: Lenders typically will not lend outside the QM boundary.
Impact of QRM
12
Cost to Borrower Increases Beyond True Economic Cost When Risk Retention Requirements Are Triggered
True Economic Cost
+ Cost of Risk Retention
Cost of Credit
Cost of
risk retention
Cost of
risk retention
Increase in true economic risk is generally
captured by risk-based pricing to some extent
True Economic Cost
Non-QRM Loans
QRM Loans
QRM Proposal
Risk Parameter (LTV, DTI, etc.)
Alternative Proposal
Objective, quantifiable risk parameters can be disclosed
to investors, hence can be risk-based priced
Impact of QRM
13
Liquidity Drives Cost of Risk Retention
Percent of Loans
Low LTV Refi Loans and
Move-up Homebuyers
First-time Homebuyers
Homogeneous Set of Low-risk Loans
Vanilla Loans Go to FHA Liquid
Ginnie Mae Market
Abundant Liquidity
for GSE Securities
Non-QRM Loans Pooled into
Odd Lots, Very Limited Liquidity
Liquid Market for
QRM Securities
Limited Amount of GSE Securitization
Low LTV
QRM Cutoff
High LTV
• Picture illustrates high refi share market of today. Purchase market has much higher LTVs.
Impact of QRM
14
20 Percent Down Does Not Work; 10 Percent Down Is No Better.
Today’s Market: Distribution of LTVs
for Homebuyers in 2009
QRM — 20 Percent Down
Payment: Distribution of LTVs
for Homebuyers in 2009
QRM — 10 Percent Down
Payment: Distribution of LTVs
for Homebuyers in 2009
35
35
35
30
30
30
25
25
25
20
20
20
15
15
15
10
10
10
5
5
5
0
0
0
80%
or less
80%–
90%
90%–
95%
Above
95%
80%
or less
80%–
90%
90%–
95%
Above
95%
80%
or less
80%–
90%
90%–
95%
Above
95%
Source: MBA Analysis of Census Bureau American Housing Survey Data.
• Borrowers with conforming loan
balances across the LTV spectrum
have access to the liquidity of the TBA
market, which provides the lowest rates.
• Borrowers with jumbo balances
pay more for fixed-rate loans.
• In October 2011, 43 percent of purchase
loans went to one of the government
loan programs (FHA, VA or USDA).
• The small share of homebuyers
who can make 20 percent or larger
down payments get access to the
lowest mortgage rates and have
the broadest access to credit.
• All other homebuyers either get
loans through a government program
or pay a higher rate than those who
qualify as QRM.
Impact of QRM
• If regulators opt for the alternative
proposal, it could make the situation
even worse. As shown, the purchase
market would be cut in half.
• Stronger credit borrowers with less
than a 10 percent down payment would
likely opt for a government loan, leaving
minimal liquidity and even higher costs
for those in the non-QRM space.
15
Borrowers of Color Use Government Lending to a Greater Extent
Governmenta Share of Home Purchase Loans by Borrower Characteristic
100
80
60
40
20
0
2006
2007
African-American
2008
Hispanic
2009
2010
Non-Hispanic White
a. FHA, VA, USDA
Source: Federal Reserve Analysis of HMDA data.
• HMDA data show that borrowers of color have already heavily been using government housing programs such as FHA in recent years.
• For example, 81.6 percent of African-American borrowers used a government program to finance the purchase of a home in 2010.
• An overly narrow QRM definition would further increase borrowers of color reliance on these government programs, as they
would be the primary source of low-down payment lending.
Impact of QRM
16
Appendix
Impact of QRM
A1
Difference in Volume and Performance when Removing the Debt-to-Income /
Payment-to-Income Requirements from the QRM Standards
25%
20%
15%
10%
5%
0%
1997
1998
1999
2000
2001
2002
2003
Change in Cumulative Delinquencies
2004
2005
2006
2007
2008
2009
Change in Mortgage Volume
Source: FHFA. “Mortgage Market Note 11-02: Qualified Residential Mortgages.” April 11, 2011.
Debt-to-Income / Payment-to-Income Ratios
FHFA data and analysis show that removing the DTI requirement would result in a very large (up to 24%) increase in loans
that would qualify, but a relatively small increase in cumulative delinquencies.
Impact of QRM
A2
Difference in Volume and Performance when Removing the Product-Type
Requirements from the QRM Standards
25%
20%
15%
10%
5%
0%
1997
1998
1999
2000
2001
2002
2003
Change in Cumulative Delinquencies
2004
2005
2006
2007
2008
2009
Change in Mortgage Volume
Source: FHFA. “Mortgage Market Note 11-02: Qualified Residential Mortgages.” April 11, 2011.
Product Requirements
FHFA found that, “For the 2005–2007 origination years, the requirement for product-type (no non-traditional and
low documentation loans, or loans for houses not occupied by the owner) was the QRM risk factor that most reduced
delinquency rates.”
Impact of QRM
A3
Difference in Volume and Performance when Removing the Loan-to-Value
Requirements from the QRM Standards
25%
20%
15%
10%
5%
0%
1997
1998
1999
2000
2001
2002
2003
Change in Cumulative Delinquencies
2004
2005
2006
2007
2008
2009
Change in Mortgage Volume
Source: FHFA. “Mortgage Market Note 11-02: Qualified Residential Mortgages.” April 11, 2011.
Loan-to-Value
Similarly, FHFA data and analysis show that removing the LTV requirement would result in a very large (up to 17%) increase in loans that would
qualify, but a relatively small increase in cumulative delinquencies, as a result of removing these standards. Note that the QRM proposal would
also require borrowers to pay closing costs out of pocket.
Impact of QRM
A4
FHFA Estimated Impact of Removing LTV Requirement on Ever-90 Delinquency Rates
Percent
12%
10%
8%
6%
4%
2%
0%
1997
1998
1999
2000
2001
2002
2003
QRM Delinquency Rate
2004
2005
2006
2007
2008
2009
Impact of Removing LTV Requirement
Source: FHFA. “Mortgage Market Note 11-02: Qualified Residential Mortgages.” April 11, 2011.
• Removing the LTV requirement does lead to higher relative delinquency rates for most years. However, as FHFA analysis shows, absolute
performance remains strong and ever-90 rates stay below 2 percent because of the high quality imposed by other QRM parameters.
• For the 2005–2007 origination years, even loans within the extremely narrow QRM box had ever-90 delinquency rates near
or above 2 percent. Ten percent unemployment and 30 percent declines in home values can cause even extraordinarily safe loans
to become delinquent.
Impact of QRM
A5
Impact of Increasing Minimum Down Payment on Default Rates
for Loans that Meet Prudent Underwriting Standards
25
24.7%
19.8%
20
16.4%
15
8.5%
10
4.7%
1.8%
1.6%
1.2%
3.7%
1.4%
1.3%
2002
Non-Qualified
2.8%
2.5%
5.6%
3.9%
6.3%
5.8%
4.7%
5.6%
1.8%
1.6%
2.1%
1.2%
1.1%
5
0
4.7%
4.4%
6.8%
6.4%
2003
2004
2005
Qual &> = 5DP
2006
Qual &> = 10DP
2007
2008
Qual &> = 20DP
Source: Vertical Capital Solutions of New York, an independent valuation and advisory firm conducted this analysis
using loan performance data maintained by First American CoreLogic, Inc.
Impact of QRM
A6
Down Payment Constraint
50
40
30
20
10
0
80–85
2007
85–90
90–95
2008
95–100
95+ CLTV
2009
2010
Source: MBA analysis of CoreLogic / LoanPerformance.
• According to CoreLogic data, based on the LTV requirement alone, the proposed QRM rule would have disqualified
48.3%–57.2% of purchase loans made over this four-year period.
• Over these four years, on average, 40 percent of purchase money mortgages had LTVs or CLTVs of 95 percent or higher.
• In 2007 and 2008, many borrowers opted for piggyback loans, leading to higher CLTVs.
• In 2009 and 2010, many more borrowers chose FHA or other government loans, which allowed lower down payments.
For most of this time, 50 percent of purchase loans on owner-occupied residences were insured by FHA.
Impact of QRM
A7
Conservative Approach — Assumes Every Dollar of Savings Goes Toward Down Payment
Down Payment Calculations — Median Household, 9% Savings Rate (CEX)
Birmingham
Philadelphia
Chicago
Seattle
San Francisco
$31,704
$2,642
$2,246
$36,669
$3,056
$2,597
$46,781
$3,898
$3,314
$58,990
$4,916
$4,178
$70,040
$5,837
$4,961
$202
$758
$140,450
$234
$912
$208,120
$298
$376
$900
$1,015
$166,900 $292,860
Required Down Payment (20%) $28,090
Required Down Payment (10%) $14,045
Required Down Payment (5%)
$7,023
$41,624
$20,812
$10,406
$33,380
$16,690
$8,345
12
15
6
3
Median Annual
Household Income
Monthly Income
After-tax Income Monthly Savings
(After-tax incomemonthly expenditures)
Median Gross Rent
Median Home Price
Years to save for
20% down payment
Years to save for
10% down payment
Years to save for
5% down payment
Los Angeles
Phoenix
Houston
$54,828
$4,569
$3,884
$48,881
$4,073
$3,462
$42,797
$3,566
$3,031
$447
$1,303
$494,730
$350
$1,197
$304,420
$312
$912
$130,405
$273
$848
$153,683
$58,572
$29,286
$14,643
$98,946
$49,473
$24,737
$60,884
$30,442
$15,221
$26,081
$13,040
$6,520
$30,737
$15,368
$7,684
9
13
18
15
7
9
7
5
6
9
7
3
5
4
2
3
5
4
2
2
Sources: MBA analysis of Census Bureau, Bureau of Labor Statistics, and National Association of Realtors data.
• This table shows sample calculations regarding how long it would take a typical household in different metros
to save for a 20% down payment on a typical home.
• There are a number of assumptions necessary to complete this calculation. Values are drawn from government survey data.
• Current owners who have lost their equity would need to save for a new down payment, thus this analysis considers all households,
not just renters.
Impact of QRM
A8
Conservative Approach — Assumes Every Dollar of Savings Goes Toward Down Payment
Down Payment Calculations — Median Household, 5% Savings Rate (BEA)
Birmingham
Philadelphia
Chicago
Seattle
San Francisco
$31,704
$2,642
$2,246
$36,669
$3,056
$2,597
$46,781
$3,898
$3,314
$58,990
$4,916
$4,178
$70,040
$5,837
$4,961
$112
$758
$140,450
$130
$912
$208,120
$166
$209
$900
$1,015
$166,900 $292,860
Required Down Payment (20%) $28,090
Required Down Payment (10%) $14,045
Required Down Payment (5%)
$7,023
$41,624
$20,812
$10,406
$33,380
$16,690
$8,345
21
27
10
5
Median Annual
Household Income
Monthly Income
After-tax Income Monthly Savings
(Average BEA personal
savings rate, 2008–2011)
Median Gross Rent
Median Home Price
Years to save for
20% down payment
Years to save for
10% down payment
Years to save for
5% down payment
Los Angeles
Phoenix
Houston
$54,828
$4,569
$3,884
$48,881
$4,073
$3,462
$42,797
$3,566
$3,031
$248
$1,303
$494,730
$194
$1,197
$304,420
$173
$912
$130,405
$152
$848
$153,683
$58,572
$29,286
$14,643
$98,946
$49,473
$24,737
$60,884
$30,442
$15,221
$26,081
$13,040
$6,520
$30,737
$15,368
$7,684
17
23
33
26
13
17
13
8
12
17
13
6
8
7
4
6
8
7
3
4
Sources: MBA analysis of Census Bureau, Bureau of Labor Statistics, and National Association of Realtors data, Bureau of Economic Analysis.
• This table shows sample calculations regarding how long it would take a typical household in different metros
to save for a 20% down payment on a typical home.
• There are a number of assumptions necessary to complete this calculation. Values are drawn from government survey data.
• Current owners who have lost their equity would need to save for a new down payment, thus this analysis considers all households,
not just renters.
Impact of QRM
A9
Conservative Approach — Assumes Every Dollar of Savings Goes Toward Down Payment
Down Payment Calculations — Median Renter Household, 9% Savings Rate (CEX)
Birmingham
Philadelphia
Chicago
Seattle
San Francisco
Los Angeles
Phoenix
Houston
Median Renter Income
$18,071
$20,901
$26,665
$33,624
$39,923
$31,252
$27,862
$24,394
Monthly Income
$1,506
$1,742
$2,222
$2,802
$3,327
$2,604
$2,322
$2,033
After-tax Income $1,280
$1,481
$1,889
$2,382
$2,828
$2,214
$1,974
$1,728
Homeowner Costs
with Mortgage (ACS)
$1,062
$1,172
$1,915
$2,200
$3,079
$2,348
$1,497
$1,453
Monthly Savings
(After-tax incomemonthly expenditures)
$115
$133
$170
$214
$255
$199
$178
$156
Median Gross Rent
$758
$912
$900
$1,015
$1,303
$1,197
$912
$848
Median Home Price
$140,450
$208,120
$166,900 $292,860
$494,730
$304,420
$130,405
$153,683
Required Down Payment (20%) $28,090
$41,624
$33,380
$58,572
$98,946
$60,884
$26,081
$30,737
Required Down Payment (10%) $14,045
$20,812
$16,690
$29,286
$49,473
$30,442
$13,040
$15,368
Required Down Payment (5%)
$7,023
$10,406
$8,345
$14,643
$24,737
$15,221
$6,520
$7,684
Years to save for
20% down payment
20
26
16
23
32
25
12
16
Years to save for
10% down payment
10
13
8
11
16
13
6
8
Years to save for
5% down payment
5
7
4
6
8
6
3
4.6
Sources: MBA analysis of Census Bureau, Bureau of Labor Statistics, and National Association of Realtors data.
• Renters typically have lower incomes than owners.
• This table shows how long it would take a typical renter in different metros to save for a 20% down payment on a typical house.
Impact of QRM
A10
Conservative Approach — Assumes Every Dollar of Savings Goes Toward Down Payment
Down Payment Calculations — Median Renter Household, 5% Savings Rate (BEA)
Birmingham
Philadelphia
Chicago
Seattle
San Francisco
Los Angeles
Phoenix
Houston
Median Renter Income
$18,071
$20,901
$26,665
$33,624
$39,923
$31,252
$27,862
$24,394
Monthly Income
$1,506
$1,742
$2,222
$2,802
$3,327
$2,604
$2,322
$2,033
After-tax Income $1,280
$1,481
$1,889
$2,382
$2,828
$2,214
$1,974
$1,728
Homeowner Costs
with Mortgage (ACS)
$1,062
$1,172
$1,915
$2,200
$3,079
$2,348
$1,497
$1,453
Monthly Savings
(Average BEA personal
savings rate, 2008–2011)
$64
$74
$94
$119
$141
$111
$99
$89
Median Gross Rent
$758
$912
$900
$1,015
$1,303
$1,197
$912
$848
Median Home Price
$140,450
$208,120
$166,900 $292,860
$494,730
$304,420
$130,405
$153,683
Required Down Payment (20%) $28,090
$41,624
$33,380
$58,572
$98,946
$60,884
$26,081
$30,737
Required Down Payment (10%) $14,045
$20,812
$16,690
$29,286
$49,473
$30,442
$13,040
$15,368
Required Down Payment (5%)
$7,023
$10,406
$8,345
$14,643
$24,737
$15,221
$6,520
$7,684
Years to save for
20% down payment
37
47
29
41
58
46
22
30
Years to save for
10% down payment
18
23
15
20
29
23
11
15
Years to save for
5% down payment
9
12
7
10
15
11
6
7
Sources: MBA analysis of Census Bureau, Bureau of Labor Statistics, and National Association of Realtors data. Bureau of Economic Analysis.
• Renters typically have lower incomes than owners.
• This table shows how long it would take a typical renter in different metros to save for a 20% down payment on a typical house.
Impact of QRM
A11
Family Net Worth by Selected Characteristics
Median Net Worth
Family Characteristic
20072009
All Families
125.4
96.0
Less than 20
10.1
7.2
20–39.9
39.132.9
40–59.9
95.472.6
60–79.9
216.7167.5
80–89.9
373.5302.5
90–100
1,205.1894.5
Percentile of income (2007)
Race or ethnicity of respondent (2007)
White non-Hispanic
Non-white or Hispanic
178.8
149.9
32.8
23.3
Housing status (2007)
Owner
244.8192.6
Renter or other
5.5
3.6
Source: Federal Reserve, 2009 Survey of Consumer Finances
Impact of QRM
A12
Trends in Loan-to-Value Ratio by Income and Race / Ethnicity for First-Time Homebuyers, 1989–2005
Income or Race / Ethnicity LTV Category
1989–2005 (%)
1989–1997
(%)
1997–2005
(%)
Change
(%)
Low-income buyers*
80% or less
44.4
45.9
43.1
–2.8
80.1 to 90%
19.3
19.5
19.1
–0.4
90.1 to 95%
12.1
10.2
13.5
3.3
Above 95%
24.3
24.3
24.2
–0.1
Moderate-income buyers
80% or less
41.9
43.4
40.4
–3.0
80.1 to 90%
22.4
24.2
20.8
–3.5
90.1 to 95%
14.4
14.1
14.7
0.6
Above 95%
21.3
18.3
24.1
5.9
High-income buyers
80% or less
45.5
44.5
46.4
1.9
80.1 to 90%
26.0
28.6
23.9
–4.7
90.1 to 95%
13.2
14.0
12.4
–1.6
Above 95%
15.3
12.9
17.3
4.4
• The table to the left shows that African-American,
Hispanic, low-income and first-time homebuyers
have consistently turned to high LTV loans in
greater proportions.
• The table shows data from 1989–2005
calculated from the American Housing Survey
(AHS) conducted by the Census Bureau.
• AHS data for 2009 shows:
80% or less
All African-American
Hispanic
HouseholdsBuyers Buyers
26%
14%
20%
80.1%–90%
17%13%17%
90.1%–95%14%
21%
19%
Above 95%
51%
44%
31%
Source: MBA analysis of AHS data.
White buyers
80% or less
44.4
45.1
43.0
–2.1
80.1 to 90%
23.7
25.0
22.2
–2.8
90.1 to 95%
13.0
12.7
13.4
0.7
Above 95%
19.0
17.2
21.4
4.2
African-American buyers
LTV = loan-to-value
80% or less
37.8
36.7
37.9
1.1
80.1 to 90%
19.9
20.1
20.4
0.3
90.1 to 95%
15.4
16.2
15.0
–1.2
Above 95%
26.8
27.0
26.7
–0.2
Hispanic buyers
80% or less
40.5
42.1
41.0
–1.0
80.1 to 90%
20.4
23.8
18.6
–5.2
90.1 to 95%
14.9
11.6
15.3
3.7
Above 95%
24.2
22.5
25.0
2.5
Impact of QRM
* Low-income homebuyers are defined as those
whose incomes are less than 80 percent of the
area median income.
Source: Tabulations from the 1991–2005 American
Housing Surveys
Table from Belsky and Herbert, “Initial Housing Choices
Made by Low-Income and Minority Homebuyers,”
2008.
A13
Borrowers of Color Use Government Lending Programs to a Greater Extent
Incidence of Selected Types of Home Purchase Loans by Borrower Characteristic
Minority status of borrower
African-American
Hispanic
Non-Hispanic white
2006
Govern-
Port-
ment*GSE Other**folio
13.916.943.2 26
7 18.2 46.5 28.3
9.6
33.2
27.8
29.4
2007 20082009
Govern-
Port-
ment*GSEOther**folio
21.934.215.728.2
Govern-
Port- Govern-
Portment* GSE Other**folio ment*GSEOther**folio
6419.4 5.2 11.4 81.49.22.6 6.8
12.2
37 17.2 33.6
51.5 29.5 6.1
11.5
44
35.4
16.2
28.4
36.2
9.9
13
18.5
73.615.3 4.1 6.9
52.1
28.9
7.1
11.9
* FHA, VA, USDA
** Conventional, non-GSE
Source: Federal Reserve Analysis of HMDA data, Avery et al, 2010.
• HMDA data show that borrowers of color have already been heavily using government housing programs, such as FHA, in recent years.
• For example, 81.4 percent of African-American borrowers in 2009 used a government program to finance the purchase of a home.
• An overly narrow QRM definition would further increase borrowers of color use of these government programs, as they would be
the primary source of low down payment lending.
Impact of QRM
A14
Impact of Regulator’s Proposed Credit Standard
Sample 1: FICO 8 Score (2005 Data)
Sample 2: FICO 8 Score (2008 Data)
25%
25%
20%
20%
Fail QRM
Satisfy QRM
Fail QRM
10%
10%
5%
5%
0%
0%
0
<
45
45
0
<
45
Satisfy QRM
45
15%
0
48 to 4
0 79
to
51 50
0
t 9
54 o 53
0
9
t
57 o 5
0 69
t
60 o 5
0 99
t
63 o 6
0 29
t
66 o 6
0 59
t
69 o 68
0
t 9
72 o 71
9
0
t
75 o 7
0 49
t
78 o 7
0 79
to
81 80
0
to 9
83
9
>
83
9
15%
0
48 to 4
0 79
to
51 50
0
t 9
54 o 53
0
9
t
57 o 5
0 69
t
60 o 5
0 99
t
63 o 6
0 29
t
66 o 6
0 59
t
69 o 68
0
t 9
72 o 71
9
0
t
75 o 7
0 49
t
78 o 7
0 79
to
81 80
0
to 9
83
9
>
83
9
Percent
FICO Score
FICO Score
Source: 2010 Fair Isaac Corporation analysis of credit report data from 2005 and 2008.
Data are population with new mortgage account opened.
• Fair Isaac data show the impact of the proposed credit standard,
i.e., no 30-day delinquencies on any tradeline at closing, no
60-day delinquencies on any tradeline over the prior two years.
• A borrower with a 700 score may have been two payments
behind on a disputed bill 23 months ago, and hence would
fail this portion of QRM.
• The above left chart has data from 2005, the above right
from 2008.
• Conversely, a borrower with a 600 score may have had a
90-day mortgage delinquency 25 months ago, but would
satisfy this portion of QRM.
• This analysis shows that the regulator’s proposal is not acting as
intended. Borrowers with very high FICO scores may not satisfy
this credit standard, while those with very low credit scores do.
• Operationally, it will also be difficult for lenders
to handle voluminous credit report data as opposed
to summary scores.
Impact of QRM
A15
Rent / Own Market Dynamics
Year-over-year Changes in the Number of Owner- and RenterOccupied Households, and Homeownership Rate
2,500
Rental Vacancy Rates: 5 or more units,
for United States (Percent)
70
2,000
15
68
1,500
12
66
1,000
500
64
9
0
62
2011Q1
2010Q1
2009Q1
2008Q1
2007Q1
2006Q1
2005Q1
2004Q1
2003Q1
Change in Renter Households
2002Q1
Change in Owner Households
Homeownership Rate
2001Q1
6
2000Q1
60
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
-1,000
1999Q1
-500
Source: U.S. Census Bureau.
• Household formation slowed considerably over the downturn, but is recovering now.
• Renter household growth is strong, the number of owner households has continued to decline.
• However, rental vacancy rates are dropping. At some point, given the remarkably small number of new units being built,
this will put upward pressure on rents.
• Of course, there is significant variation across markets.
• As rents rise, many of these households may choose to buy.
• The down payment constraint is typically the highest hurdle for first-time homebuyers transitioning from rentals.
• Buying a home is a way to protect against future increases in rent. Of course, owning a home has many additional
expenditures that renters do not face.
Impact of QRM
A16
Loan Characteristics and Performance from Worst Origination Years
Share of Mortgage Loans with Certain Features Cumulative Default Rates
Category
1st Half 2nd Half 1st Half 2nd Half 1st Half2nd Half 1st Half
2005 2005 2006 2006 2007 2007 2008
Category 2005
2006
2007
2.5%
4.4%
4.8%
Loan Type
Prime
61%
61%
58%
57%
69%
78%
79%
Jumbo Prime Fixed
Alt-A
10% 12%14% 15%14% 6% 4%
Alt-A Fixed
Non-prime
24%
Alt-A ARM
Government
24%
23%
24%
12%
7%
1%
5% 4%5% 5%5%9%16%
100% 101%100% 101%100% 100% 100%
Subprime Fixed
Subprime ARM
Amortization Type Option ARM
IO
Option ARM
7.5%
15.5%
15.4%
14.1%
25.7%
25.9%
15.5%
13.8%
22.4%
33.9%
28.6%
13.5%
24.3%
22.1%
9.5%
16% 18%18% 18%18% 12% 5%
4%
5%
9%
8%
4%**
Occupancy Status
Non-Owner Occupied
13%
14%
14%
15%
11%
12%
10%
Note: Categories are not mutually exclusive. For example, there were prime IO loans originated.
* Option ARM volume was not captured in this survey after the first half of 2007
Source: MBA Mortgage Origination Surveys, various years.
Source: Citi Mortgage Credit Weekly, 6 / 3 / 11
• Alt-A loans, low- and no-doc lending, accounted for 10%–15% of loans originated from 2005–2006. These loans defaulted 3×–5× more
frequently than prime fixed loans.
• Non-prime loans made up almost 25% of loans originated in 2005 and 2006, and have also defaulted at very high rates.
• Option ARMs and IOs together accounted for almost 30% of loans originated at the peak of the boom. Option ARMs have defaulted at more
than 5× the rate of prime fixed loans from this period.
Impact of QRM
A17
QRM Only Available to Narrow Slice
Market has already pulled back to safer products
2007 Market
Alt-A Lending
2011 Market
QM Proposal
Alternative QRM Proposal
Option ARMs
QRM Proposal
Subprime 2 / 28s
Where would most borrowers of color and low-income homebuyers fall?
Impact of QRM
A18
Fannie Mae Credit Profile of Key Product Features
Credit Characteristics of Single-Family Conventional Guaranty Book of Business
As of
September 30, 2011
Serious Delinquency Rate
Negative
Amortizing Loans
Interest Only
Loans
Alt-A
Loans
Subprime
Loans
7.79%
15.70%
12.71%
23.91%
Sources: Fannie Mae Book of Business Delinquency Rates as of Q3 2011.
Impact of QRM
A19
For More Information,
Please Contact
Mike Fratantoni
Research and Economics
Mortgage Bankers Association
(202) 557-2935