HOUSING FINANCE POLICY CENTER
HOUSING FINANCE
AT A GLANCE
A MONTHLY CHARTBOOK
February 2015
1
ABOUT THE CHARTBOOK
HOUSING FINANCE POLICY CENTER STAFF
The Housing Finance Policy Center’s (HFPC) mission is to
produce analyses and ideas that promote sound public
policy, efficient markets, and access to economic
opportunity in the area of housing finance. At A Glance, a
monthly chartbook and data source for policymakers,
academics, journalists, and others interested in the
government’s role in mortgage markets, is at the heart of
this mission.
Laurie Goodman
Center Director
We welcome feedback from our readers on how we can
make At A Glance a more useful publication. Please email
any comments or questions to ataglance@urban.org.
To receive regular updates from the Housing Finance
Policy Center, please visit urban.org/center/hfpc to sign up
for our bi-weekly newsletter.
Ellen Seidman
Senior Fellow
Jim Parrott
Senior Fellow
Sheryl Pardo
Associate Director of Communications
Jun Zhu
Senior Financial Methodologist
Wei Li
Senior Research Associate
Bing Bai
Research Associate I
Karan Kaul
Research Associate I
Taz George
Research Associate II
Maia Woluchem
Research Assistant
Alison Rincon
Center Administrator
INTRODUCTION
More than one in three FHA borrowers could save
money by refinancing today
We continued to study the impact of the Federal
Housing Administration’s (FHA) premium cut. After
estimating the effects of the premium cut on borrower
savings (page 33) and on the economics of the MMI
fund in January, we turned our attention to refinance
volumes in February. Our latest blog post summarizes
our estimates of refinance volume, but more
importantly, outlines some of the key drivers,
assumptions and caveats that help put these
estimates in context. Our conclusion is that
approximately 2.4 million, or over 1 in 3, FHA
borrowers can lower their monthly payments upon
refinancing, even after considering the cost of
refinancing. The number of borrowers who actually
end up refinancing will likely be lower and will depend
on individual borrower sensitivity to interest rates,
expectations of future interest rates, and other
factors.
Loss Severity on Residential Mortgages based on
new Freddie Mac data
Our in depth analysis of new Freddie Mac loan-level
records of 17 million loans originated from 1999 to
2013 reveals several interesting findings. One stands
out in particular – that loans with higher LTVs with
added private mortgage insurance have a significantly
lower loss severity than loans with lower LTVs and no
mortgage insurance. In other words, mortgage
insurance is doing precisely what it was designed to
do. We subsequently summarized this in a blog post
titled: Private Mortgage Insurance is better than expected
at protecting taxpayers from losses. This chartbook
contains our quarterly supplement showing the latest
results from the loan-level credit data. We include the
new severity analysis in this chartbook (page 40).
US Department of Treasury's Credit Rating Agency
Exercise
Private label securities (PLS) activity continues at very
low levels (page 10). As part of an effort to revive the
moribund PLS market, the US Treasury Department
recently conducted an exercise with the major credit
rating agencies. This exercise was intended to boost
investor confidence in the ratings process—a critical
piece of the PLS system that failed during the housing
crisis. Treasury’s exercise revealed five important
lessons that we discussed in detail in a recent issue
brief.
Mel Watt and Julian Castro Hill testimonies
Federal Housing Finance Agency (FHFA) Director Mel
Watt and Department of Housing and Urban
Development (HUD) Secretary Julian Castro both
recently testified in front of the House Financial
Services Committee. While Castro faced tough
questions from Republicans concerning the
administration’s decision to lower premiums for FHAinsured mortgages, Watt was grilled for reducing the
down payment requirements for GovernmentSponsored Enterprise (GSE) backed mortgages to 3
percent, and for funding the Housing Trust Funds.
Watt indicated that FHFA could announce a decision
on the guarantee fees charged by Fannie Mae and
Freddie Mac by the end of March. On the issue of
principal reduction, Watt stated that there are
instances in which it could be beneficial to borrowers
without being net present value negative for the GSEs,
but added that a decision was yet to be made.
INSIDE THIS ISSUE
•
Mortgage originations expected to be relatively flat
in 2015 relative to 2014 (page 12)
•
The median FICO score at origination is 50 points
higher than a decade ago (page 14)
•
Home prices are still very affordable on a national
level, by historical standards (page 16)
•
Home prices continue to increase, but the rate of
growth declined slightly in 2014 (page 17)
•
The GSE portfolios are both within $15 billion of their
2015 cap of $399 billion (page 19)
•
Freddie Mac completes a new, $27.6 billion risksharing deal; Fannie has one in the works (page 21)
•
Agency net issuance totaled only $2 billion in January
(page 30)
•
Quarterly supplement: See the details of the GSE’s
composition, defaults, repurchase rates, and
severities (pages 34-40)
CONTENTS
Overview
Market Size Overview
Value of the US Residential Housing Market
Size of the US Residential Mortgage Market
Private Label Securities
Agency Mortgage-Backed Securities
6
6
7
7
Origination Volume and Composition
First Lien Origination Volume & Share
8
Mortgage Origination Product Type
Composition (All Originations & Purchase Originations Only)
9
Securitization Volume and Composition
Agency/Non-Agency Share of Residential MBS Issuance
Non-Agency MBS Issuance
Non-Agency Securitization
10
10
10
Agency Activity: Volumes and Purchase/Refi Composition
Agency Gross Issuance
Percent Refi at Issuance
11
11
State of the Market
Mortgage Origination Projections
Total Originations and Refinance Shares
Housing Starts and Home Sales
12
12
Originator Profitability
Originator Profitability and Unmeasured Costs (OPUC)
13
Credit Availability for Purchase Loans
Borrower FICO Score at Origination Month
Combined LTV at Origination Month
Origination FICO and LTV by MSA
14
14
15
Housing Affordability
National Housing Affordability Over Time
Affordability Adjusted for MSA-Level DTI
16
16
Home Price Indices
National Year-Over-Year HPI Growth
Changes in CoreLogic HPI for Top MSAs
17
17
Negative Equity & Serious Delinquency
Negative Equity Share
Loans in Serious Delinquency
18
18
GSEs under Conservatorship
GSE Portfolio Wind-Down
Fannie Mae Mortgage-Related Investment Portfolio
Freddie Mac Mortgage-Related Investment Portfolio
19
19
CONTENTS
Effective Guarantee Fees & GSE Risk-Sharing Transactions
Effective Guarantee Fees
Fannie Mae Upfront Loan-Level Price Adjustment
GSE Risk-Sharing Transactions
20
20
21
Serious Delinquency Rates
Serious Delinquency Rates – Fannie Mae & Freddie Mac
Serious Delinquency Rates – Single-Family Loans & Multifamily GSE Loans
22
23
Refinance Activity
Total HARP Refinance Volume
HARP Refinances
24
24
GSE Loans: Potential Refinances
Loans Meeting HARP Pay History Requirements
25
Modification Activity
HAMP Activity
New HAMP Modifications
Cumulative HAMP Modifications
26
26
Modification by Type of Action and Bearer of Risk
Changes in Loan Terms for Modifications
Type of Modification Action by Investor and Product Type
27
27
Modifications and Liquidations
Loan Modifications and Liquidations (By Year & Cumulative)
28
Modification Redefault Rates by Bearer of the Risk
Redefault Rate after Modification (12 Months & 24 Months)
29
Agency Issuance
Agency Gross and Net Issuance
Agency Gross Issuance
Agency Net Issuance
30
30
Agency Gross Issuance & Fed Purchases
Monthly Gross Issuance
Fed Absorption of Agency Gross Issuance
31
31
Mortgage Insurance Activity
MI Activity & Market Share
FHA MI Premiums for Typical Purchase Loan
Initial Monthly Payment Comparison: FHA vs. PMI
32
33
33
Special Feature: Loan Level GSE Credit Data
Quarterly Feature: Loan Level Credit Data from the GSEs
Fannie Mae Balance & Default Rate
Freddie Mac Balance & Default Rate
Fannie Mae & Freddie Mac Cumulative Default Rate by Vintage Year
Fannie Mae & Freddie Mac Cumulative Repurchase Rate by Vintage Year
Freddie Mac Loss Severity
34-35
36-37
38
39
40
Related HFPC Work
Publications and Events
41-42
OVERVIEW
MARKET SIZE OVERVIEW
Fed Flow of Funds data from 2014 Q3 indicates a modest increase in the total value of the US residential 1-4 unit
housing market to $21.4 trillion from $21.2 trillion the previous quarter. Household equity drove most of this
increase, growing by $178 billion. With credit standards remaining tight, mortgage debt increased by just $15 billion,
though that was still the largest quarterly increase in the past two years. Agency MBS make up 56.8 percent of the
total mortgage market, private-label securities make up 7.3 percent, and unsecuritized first liens at the GSEs,
commercial banks, savings institutions, and credit unions make up 29.0 percent. Second liens comprise the remaining
6.9 percent.
Value of the US Housing Market
Debt, household mortgages
($ trillions)
Household equity
Total value
25
21.4
20
15
11.5
10
9.9
5
0
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
Q3
Sources: Federal Reserve Flow of Funds and Urban Institute.
Size of the US Residential Mortgage Market
Agency MBS
Unsecuritized first liens
Private Label Securities
Second Liens
($ trillions)
6
5.6
5
4
Debt,
household
mortgages,
$9,833
3
2.9
2
1
0
2000
0.7
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
Sources: Federal Reserve Flow of Funds, Inside Mortgage Finance, Fannie Mae, Freddie Mac, eMBS and Urban Institute.
Note: Unsecuritized first liens includes loans held by commercial banks, GSEs, savings institutions, and credit unions.
2013
2014
Q3
6
OVERVIEW
MARKET SIZE OVERVIEW
As of December 2014, debt in the private-label securitization market totaled $712 billion and was split among
prime (19.7 percent), Alt-A (43.6 percent), and subprime (36.7 percent) loans. In January 2015, outstanding
securities in the agency market totaled $5.64 trillion and were 46.4 percent Fannie Mae, 27.3 percent Freddie Mac,
and 26.3 percent Ginnie Mae.
Private-Label Securities by Product Type
Alt-A
($ trillions)
Subprime
Prime
1
0.8
0.6
0.4
0.31
0.26
0.2
0
1999
0.14
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
December 2014
Sources: CoreLogic and Urban Institute.
Agency Mortgage-Backed Securities
Fannie Mae
Freddie Mac
Ginnie Mae
Total
($ trillions)
6
5.6
5
4
3
2.6
2
1.5
1
0
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
January 2015
Sources: eMBS and Urban Institute.
7
OVERVIEW
ORIGINATION VOLUME
AND COMPOSITION
First Lien Origination Volume and Share
First lien originations for 2014 through Q3 were far below their 2013 pace, totaling $850 billion. The share of bank
portfolio originations rose to nearly 27 percent, while the GSE share has dropped to 51 percent from 61 percent in
2013, reflecting the curtailment of refinancing activity. FHA/VA originations account for another 21 percent, and the
private label origination share remains less than one percent.
Volume, $ trillions
GSE securitization
FHA/VA securitization
PLS securitization
Bank portfolio
$4.0
$3.5
$3.0
$2.5
$2.0
$1.5
$0.23
$1.0
$0.01
$0.18
$0.5
$0.43
$0.0
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013 2014 Q1-3
Share, percent
100%
90%
26.7%
80%
0.7%
70%
21.5%
60%
50%
40%
30%
51.1%
20%
10%
0%
2002
2003
2004
2005
2006
Sources: Inside Mortgage Finance and Urban Institute.
2007
2008
2009
2010
2011
2012
2013
2014
Q1-3
8
OVERVIEW
MORTGAGE
ORIGINATION
MORTGAGE ORIGINATION
PRODUCT
PRODUCT TYPE
TYPE
Adjustable-rate mortgages (ARMs) accounted for as much as 29 percent of all new originations during the peak of the
recent housing bubble in 2005 (top chart). They fell to a historic low of 1 percent in 2009, and slowly grew to 6 percent
of total originations in November 2014, 30 percent higher than the level one year ago. Fifteen year fixed-rate
mortgages (FRMs), predominantly a refinance product, comprise 15.1 percent of new originations. If we exclude
refinances (bottom chart), the share of 30-year FRMs in November 2014 stood at 87 percent, 15-year FRMs at 5.8
percent, and ARMs at 6.1 percent.
All Originations
Fixed-rate 30-year mortgage
Fixed-rate 15-year mortgage
Adjustable-rate mortgage
Other
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
November 2014
Sources: CoreLogic Prime Servicing and Urban Institute.
Purchase Loans Only
Fixed-rate 30-year mortgage
Fixed-rate 15-year mortgage
Adjustable-rate mortgage
Other
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
2000
2001
2002
2003
2004
2005
Sources: CoreLogic Prime Servicing and Urban Institute.
2006
2007
2008
2009
2010
2011
2012
2013
2014
November 2014
9
OVERVIEW
SECURITIZATION VOLUME AND
COMPOSITION
Agency/Non-Agency Share of Residential MBS Issuance
Non-agency single-family MBS
issuance has hovered at or below
3 percent of total issuance since
early 2011, and this share is even
lower if re-REMICs are excluded.
The non-agency market has not
healed from the financial crisis,
with new non-agency issuance
tiny relative to pre-crisis levels.
In 2014, new non-agency deals
totaled $27.5 billion. Over half of
the new issuance in 2014, $18
billion, was scratch and dent
securitizations. In January 2015,
there was just over $1 billion in
new non-agency securitization.
100%
98.8%
Agency share
90%
80%
70%
60%
50%
40%
30%
Non-Agency share
20%
10%
1.23%
Sources: Inside Mortgage Finance and Urban Institute.
Non-Agency MBS Issuance
($ billions)
Non-Agency Securitization
($ billions)
Re-REMICs and other
$1,200
$6
Scratch and dent
Alt A
$1,000
Jan. 2015
2014
2013
2012
2011
2010
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
0%
$5
Subprime
$4
Prime
$600
$3
$2,056
$17,981
$1,597
$674
$10,002
$400
$200
1.057
$800
$2
$1
2014
2013
2012
2011
2010
2009
2008
2007
2006
2005
2004
2003
2002
2001
Sources: Inside Mortgage Finance and Urban Institute.
Jan-12
Mar-12
May-12
Jul-12
Sep-12
Nov-12
Jan-13
Mar-13
May-13
Jul-13
Sep-13
Nov-13
Jan-14
Mar-14
May-14
Jul-14
Sep-14
Nov-14
Jan-15
$0
$-
Sources: Inside Mortgage Finance and Urban Institute.
Note: Monthly figures equal total non-agency MBS issuance
minus Re-REMIC issuance.
10
OVERVIEW
AGENCY ACTIVITY:
VOLUMES AND PURCHASE/
REFI COMPOSITION
Agency issuance totaled $87.2 billion in the first month of 2015, slightly up from $69.8 billion for the same month a
year ago. In January 2015, refinances edged up to 56 and 57 percent of the GSEs’ business, as the average mortgage
rate continued to decline. The Ginnie Mae market has always been more purchase-driven, with refinance volume of
36 percent. Note that the latest uptick does not include the effect of the 50 basis point decrease in the FHA premium.
Agency Gross Issuance
Fannie Mae
Freddie Mac
Ginnie Mae
($ trillions)
$2.5
$2.0
$1.5
$1.0
$0.33
$0.27
$0.5
$0.44
$0.0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Ann.
Sources: eMBS and Urban Institute.
Note: Annualized figure based on data from January 2015.
Percent Refi at Issuance
Freddie Mac
Percent refi
Fannie Mae
Ginnie Mae
Mortgage rate
Mortgage rate
90%
9.0%
80%
8.0%
70%
7.0%
60%
6.0%
50%
5.0%
40%
4.0%
30%
3.0%
20%
2.0%
10%
1.0%
0%
0.0%
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
January 2015
Sources: eMBS and Urban Institute.
Note: Based on at-issuance balance.
11
STATE OF THE MARKET
MORTGAGE ORIGINATION
PROJECTIONS
The forecasts for mortgage originations in 2015 are mixed: The MBA sees a modest increase over 2014 of $81
billion to $1,203 billion, Fannie sees a small increase of $14 billion, and Freddie expects no change. And although
lower projected interest rates have lifted the forecasted refinancing share slightly above last month’s estimates,
both Freddie and Fannie still expect a small decline in in 2015, in Freddie's case to 35 percent, while Fannie predicts
a more modest drop to 41 percent. All three sources expect higher housing starts and home sales in the coming
year.
Total Originations and Refinance Shares
Period
2014 Q1
2014 Q2
2014 Q3
2014 Q4
2015 Q1
2015 Q2
2015 Q3
2015 Q4
FY 2011
FY 2012
FY 2013
FY 2014
FY 2015
FY 2016
Originations ($ billions)
Total, FNMA
Total, FHLMC
Total, MBA
estimate
estimate
estimate
240
250
247
305
320
297
328
350
300
294
280
278
269
320
281
321
350
332
307
300
318
284
230
272
1496
1492
1436
2154
2122
2044
1866
1925
1845
1167
1200
1122
1181
1200
1203
1186
1275
1170
FNMA
estimate
49
39
42
45
50
39
36
38
66
72
60
43
41
35
Refi Share (%)
FHLMC
estimate
44
38
42
38
45
35
30
30
64
70
59
40
35
24
MBA
estimate
50
40
38
46
49
39
34
36
65
71
60
43
39
32
Sources: Mortgage Bankers Association, Fannie Mae, Freddie Mac and Urban Institute.
Note: Shaded boxes indicate forecasted figures. All figures are estimates for total single-family market; column labels indicate source of
estimate. Forecasts include interest rates as well. The yearly averages for 2011, 2012, and 2013 were 4.5%, 3.7%, and 4.0%, respectively. The
three sources project an annual average rate of 4.2% for 2014. For 2015, their projections ranged from 4.0% to 4.4%, and for 2016 they range
from 4.4% to 5.2%.
Housing Starts and Homes Sales
Housing Starts, thousands
Home Sales. thousands
Year
Total, FNMA
estimate
Total,
FHLMC
estimate
Total,
MBA
estimate
Total, FNMA
estimate
Total,
FHLMC
estimate
Total,
MBA
estimate
Existing,
MBA
estimate
New,
MBA
Estimate
FY 2011
FY 2012
FY 2013
FY 2014
FY 2015
FY 2016
609
781
925
997
1155
1319
610
780
920
1000
1200
1400
612
783
930
994
1110
1228
4566
5028
5519
5346
5658
5885
4570
5030
5520
5370
5600
5800
4501
5030
5505
5338
5694
6003
4200
4661
5073
4901
5198
5444
301
369
432
437
496
559
Sources: Mortgage Bankers Association, Fannie Mae, Freddie Mac and Urban Institute.
Note: Shaded boxes indicate forecasted figures. All figures are estimates for total single-family market; column labels indicate source of estimate.
12
STATE OF THE MARKET
ORIGINATOR PROFITABILITY
When originator profitability is high, mortgage rates tend to be less responsive to the general level of interest
rates, as originators are capacity-constrained. When originator profitability is low, mortgage rates are far more
responsive to the general level of interest rates. As interest rates have risen from the lows in 2012, and fewer
borrowers find it economical to refinance, originator profitability is lower. Originator profitability is often
measured as the spread between the rate the borrower pays for the mortgage (the primary rate) and the yield on
the underlying mortgage-backed security in the secondary market (the secondary rate). However, with guarantee
fees up dramatically from 2011 levels, the so-called primary-secondary spread has become a very imperfect
measure to compare profitability across time.
The measure used here, Originator Profitability and Unmeasured Costs (OPUC), is formulated and calculated by
the Federal Reserve Bank of New York. It looks at the price at which the originator actually sells the mortgage into
the secondary market and adds the value of retained servicing (both base and excess servicing, net of g-fees) as
well as points paid by the borrower. This measure was in the narrow range of 2.04 to 2.31 in 2014, and edged up to
2.41 in the first month of 2015.
Originator Profitability and Unmeasured Costs
Dollars per $100 loan
6
5
4
3
2.41
2
1
0
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
January 2015
Sources: Federal Reserve Bank of New York, updated monthly and available at this link:
http://www.ny.frb.org/research/epr/2013/1113fust.html and Urban Institute.
Note: OPUC stands for "originator profits and unmeasured costs" as discussed in Fuster et al. (2013). The OPUC series is a monthly (4-week
moving) average.
13
STATE OF THE MARKET
CREDIT
CREDIT AVAILABILITY
AVAILABILITY FOR
FOR
PURCHASE LOANS
Access to credit has become extremely tight, especially for borrowers with low FICO scores. The mean and median
FICO scores on new originations have both drifted up about 47 and 50 points over the last decade. The 10th
percentile of FICO scores, which represents the lower bound of creditworthiness needed to qualify for a mortgage,
stood at 662 as of November 2014. Prior to the housing crisis, this threshold held steady in the low 600s. LTV levels
at origination remain relatively high, averaging 85, which reflects the large number of FHA purchase originations.
Borrower FICO Score at Origination
FICO Score
90th percentile
Mean
Median
10th percentile
850
800
801
750
749
740
700
662
650
600
550
500
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
Sources: CoreLogic Servicing and Urban Institute.
Note: Purchase-only loans.
2013
2014
November 2014
Combined LTV at Origination
LTV
90th percentile
Mean
Median
10th percentile
110
101
100
90
89
85
80
70
66
60
50
40
30
2001
2002
2003
2004
2005
Sources: CoreLogic Servicing and Urban Institute.
Note: Purchase-only loans.
2006
2007
2008
2009
2010
2011
2012
2013
2014
November 2014
14
San Francisco-Redwood City-South San Francisco CA
San Jose-Sunnyvale-Santa Clara CA
Oakland-Hayward-Berkeley CA
San Diego-Carlsbad CA
Los Angeles-Long Beach-Glendale CA
New York-Jersey City-White Plains NY-NJ
Portland-Vancouver-Hillsboro OR-WA
Seattle-Bellevue-Everett WA
Boston MA
Denver-Aurora-Lakewood CO
Newark NJ-PA
Nassau County-Suffolk County NY
Sacramento--Roseville--Arden-Arcade CA
Baltimore-Columbia-Towson MD
Washington-Arlington-Alexandria DC-VA-MD-WV
Minneapolis-St. Paul-Bloomington MN-WI
Dallas-Plano-Irving TX
Chicago-Naperville-Arlington Heights IL
Philadelphia PA
Fort Worth-Arlington TX
Tampa-St. Petersburg-Clearwater FL
St. Louis MO-IL
Houston-The Woodlands-Sugar Land TX
Pittsburgh PA
Orlando-Kissimmee-Sanford FL
Columbus OH
Riverside-San Bernardino-Ontario CA
Charlotte-Concord-Gastonia NC-SC
Phoenix-Mesa-Scottsdale AZ
Atlanta-Sandy Springs-Roswell GA
San Antonio-New Braunfels TX
Kansas City MO-KS
Cincinnati OH-KY-IN
Las Vegas-Henderson-Paradise NV
Miami-Miami Beach-Kendall FL
Cleveland-Elyria OH
Detroit-Dearborn-Livonia MI
STATE OF THE MARKET
CREDIT AVAILABILITY
AVAILABILITY FOR
FOR
CREDIT
PURCHASE LOANS
Credit has been tight for all borrowers with less-than-stellar credit scores, but there are significant variations across
MSAs. For example, the mean origination FICO for borrowers in San Francisco- Redwood City- South San Francisco,
CA is almost 772, while in Detroit-Dearborn-Livonia, MI it is 721. Across all MSAs, lower average FICO scores tend
to be correlated with high average LTVs, as these MSAs rely heavily on FHA/VA financing.
Origination FICO and LTV by MSA
Origination FICO
Mean origination FICO score
Mean origination LTV
Origination LTV
780
100
770
95
760
90
750
85
740
80
730
75
720
70
710
65
700
60
Sources: CoreLogic Servicing as of November 2014 and Urban Institute.
Note: Purchase-only loans.
15
STATE OF THE MARKET
HOUSING AFFORDABILITY
National Housing Affordability Over Time
$300
$280
$260
$240
$220
$200
$180
$160
$140
$120
$284,619
Credit
Bubble
$238,861
2014
2013
2012
2011
2010
2009
2008
2007
2006
2005
2004
2003
2002
$200,000
2001
Sources: CoreLogic, US Census, Freddie Mac
and Urban Institute.
Note: The maximum affordable price is the
house price that a family can afford putting 20
percent down, with a monthly payment of 28
percent of median family income, at the
Freddie Mac prevailing rate for 30-year fixedrate mortgage, and property tax and insurance
at 1.75 percent of housing value.
Max affordable price
Housing Prices ($ thousands)
2000
Home prices are still very affordable
by historical standards, despite
increases over the last three years.
Even if interest rates rose to 6
percent, affordability would be at the
long-term historical average.
Median sales price
Max affordable price at 6.0% rate
November 2014
Affordability Adjusted for MSA-Level DTI
Ratio
1.4
1.3
1.2
1.1
1
0.9
0.8
Sources: CoreLogic, US Census, Freddie Mac and UI calculations based on NAR methodology.
Note: Index is calculated relative to home prices in 2000-03. A ratio above 1 indicates higher affordability in November 2014 than in
2000-03.
16
Cleveland-Elyria OH
Cincinnati OH-KY-IN
Pittsburgh PA
Columbus OH
Chicago-Naperville-Arlington Heights IL
Las Vegas-Henderson-Paradise NV
Tampa-St. Petersburg-Clearwater FL
St. Louis MO-IL
Detroit-Dearborn-Livonia MI
Newark NJ-PA
Kansas City MO-KS
Sacramento--Roseville--Arden-Arcade CA
Minneapolis-St. Paul-Bloomington MN-WI
Orlando-Kissimmee-Sanford FL
Nassau County-Suffolk County NY
Charlotte-Concord-Gastonia NC-SC
Atlanta-Sandy Springs-Roswell GA
Denver-Aurora-Lakewood CO
Houston-The Woodlands-Sugar Land TX
San Antonio-New Braunfels TX
Fort Worth-Arlington TX
San Diego-Carlsbad CA
Oakland-Hayward-Berkeley CA
Phoenix-Mesa-Scottsdale AZ
Baltimore-Columbia-Towson MD
Dallas-Plano-Irving TX
Riverside-San Bernardino-Ontario CA
Boston MA
New York-Jersey City-White Plains NY-NJ
Seattle-Bellevue-Everett WA
Philadelphia PA
Miami-Miami Beach-Kendall FL
Portland-Vancouver-Hillsboro OR-WA
Washington-Arlington-Alexandria DC-VA-MD-WV
San Jose-Sunnyvale-Santa Clara CA
Los Angeles-Long Beach-Glendale CA
San Francisco-Redwood City-South San Francisco CA
0.7
STATE OF THE MARKET
HOME PRICE INDICES
National Year-Over-Year HPI Growth
The strong year-over-year house price growth through 2013 has slowed somewhat in 2014, as indicated by both
the repeat-sales HPI from CoreLogic and hedonic index from Zillow.
Year-over-year growth rate
20%
15%
10%
CoreLogic HPI
6.6%
5%
0%
5.0%
Zillow HVI
-5%
-10%
-15%
-20%
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
Sources: CoreLogic, Zillow, and Urban Institute.
2014
December2014
Changes in CoreLogic HPI for Top MSAs
Despite rising 28 percent from the trough, national house prices still must grow 15.5 percent to reach pre-crisis
peak levels. At the MSA level, three of the top 15 MSAs have reached their peak HPI– Houston, TX; Dallas, TX; and
Denver, CO. Two MSAs particularly hard hit by the boom and bust– Phoenix, AZ and Riverside, CA– would need
to rise more than 40 percent to return to peak levels.
HPI changes (%)
MSA
United States
New York-Jersey City-White Plains NY-NJ
Los Angeles-Long Beach-Glendale CA
Chicago-Naperville-Arlington Heights IL
Atlanta-Sandy Springs-Roswell GA
Washington-Arlington-Alexandria DC-VA-MD-WV
Houston-The Woodlands-Sugar Land TX
Phoenix-Mesa-Scottsdale AZ
Riverside-San Bernardino-Ontario CA
Dallas-Plano-Irving TX
Minneapolis-St. Paul-Bloomington MN-WI
Seattle-Bellevue-Everett WA
Denver-Aurora-Lakewood CO
Baltimore-Columbia-Towson MD
San Diego-Carlsbad CA
Anaheim-Santa Ana-Irvine CA
2000 to peak
Peak to
trough
Trough to
current
% Rise needed
to achieve
peak
98.8
115.4
181.5
65.1
40.8
159.7
44.4
126.3
194.6
38.3
73.9
94.1
36.4
128.5
148.7
162.3
-32.4
-19.9
-39.0
-36.4
-33.5
-33.3
-12.6
-52.7
-53.3
-13.6
-30.5
-31.8
-14.4
-25.5
-38.2
-36.8
28.0
16.8
42.7
19.8
37.4
26.4
31.8
48.1
47.8
27.1
23.9
34.2
35.6
6.5
36.3
36.8
15.5
6.9
14.9
31.3
9.5
18.6
-13.2
42.8
44.8
-8.9
16.1
9.2
-13.9
26.1
18.6
15.7
Sources: CoreLogic HPIs as of December 2014 and Urban Institute.
Note: This table includes the largest 15 Metropolitan areas by mortgage count.
17
STATE OF THE MARKET
NEGATIVE EQUITY & SERIOUS
DELINQUENCY
Negative Equity Share
With housing prices appreciating through the first three quarters of 2014, residential properties in negative equity
(LTV greater than 100) as a share of all residential properties with a mortgage has dropped to 10.3 percent.
Residential properties in near negative equity (LTV between 95 and 100) comprise another 2.6 percent.
35%
30%
25%
Near or in negative
equity
20%
15%
10%
12.9%
10.3%
Negative equity
5%
3Q14
2Q14
1Q14
4Q13
3Q13
2Q13
1Q13
4Q12
3Q12
2Q12
1Q12
4Q11
3Q11
2Q11
1Q11
4Q10
3Q10
2Q10
1Q10
4Q09
3Q09
0%
Sources: CoreLogic and Urban Institute.
Loans in Serious Delinquency/Foreclosure
Serious delinquencies and foreclosures continue to decline with the housing recovery, but remain quite high
relative to the early 2000s. Loans 90 days delinquent or in foreclosure totaled 4.7 percent in the third quarter
of 2014, down from 5.7 percent for the same quarter a year earlier.
10%
9%
Percent of loans 90
days delinquent or in
foreclosure
Percent of loans in
foreclosure
8%
7%
6%
4.7%
5%
4%
Percent of loans 90
days delinquent
3%
2.4%
2%
2.3%
1%
3Q00
1Q01
3Q01
1Q02
3Q02
1Q03
3Q03
1Q04
3Q04
1Q05
3Q05
1Q06
3Q06
1Q07
3Q07
1Q08
3Q08
1Q09
3Q09
1Q10
3Q10
1Q11
3Q11
1Q12
3Q12
1Q13
3Q13
1Q14
3Q14
0%
Sources: Mortgage Bankers Association and Urban Institute.
18
GSES UNDER CONSERVATORSHIP
GSE PORTFOLIO WIND-DOWN
Fannie and Freddie ended 2014 with portfolios totaling $413.3 billion and $408.4 billion, respectively, well
below the 2014 portfolio cap of $469.625 billion and just above the 2015 cap of $399 billion. Compared to
December 2013, Fannie has contracted by 15.8 percent, and Freddie Mac by 11.4percent. They are shrinking
their less liquid assets (mortgage loans and non-agency MBS) at close to the same pace that they are shrinking
their entire portfolio.
Fannie Mae Mortgage-Related Investment Portfolio Composition
Fannie MBS in portfolio
Non-FNMA agency MBS
($ billions)
Non-agency MBS
Mortgage loans
Current size: $413.3 billion
2015 cap: $399.181 billion
Shrinkage year-over-year: 15.8%
Shrinkage in less-liquid assets yearover-year: 10.1%
900
800
700
600
500
400
300
200
100
0
2006
2007
2008
2009
2010
2011
2012
2013
Sources: Fannie Mae and Urban Institute.
2014
December 2014
Freddie Mac Mortgage-Related Investment Portfolio Composition
FHLMC MBS in portfolio
Non-FHLMC agency MBS
Non-agency MBS
($ billions)
Mortgage loans
Current size: $408.4 billion
2015 cap: $399.181 billion
Shrinkage year-over-year: 11.4%
Shrinkage in less-liquid assets yearover-year: 16.5%
900
800
700
600
500
400
300
200
100
0
2006
2007
2008
Sources: Freddie Mac and Urban Institute.
2009
2010
2011
2012
2013
2014
December 2014
19
GSES UNDER CONSERVATORSHIP
EFFECTIVE GUARANTEE FEES AND
GSE RISK-SHARING TRANSACTIONS
Effective Guarantee Fees
Fannie’s average charged g-fee on new single-family originations was 63.5 bps in Q3 2014, up slightly from 62.6 in
the previous quarter. This is a marked increase over 2012 (39.9 bps) and 2011 (28.8 bps), and has contributed to
the GSEs’ profits. Fannie’s loan-level price adjustments (LLPAs) are shown in the second table. The 25 bp Adverse
Market Delivery Charge has been added to these upfront numbers. The FHFA asked for input about the level of gfees and LLPAs in summer of 2014 and has indicated that a decision about updated g-fees and LLPAs will be
rendered at the end of Q1 or early Q2, 2015.
70
Fannie Mae single-family effective gfee rate
63.5
60
Fannie Mae single-family average
charged g-fee on new acquisitions
50
Freddie Mac management and g-fee
rate
40
41.2
30
32.4
20
10
0
1Q09
2Q09
3Q09
4Q09
1Q10
2Q10
3Q10
4Q10
1Q11
2Q11
3Q11
4Q11
1Q12
2Q12
3Q12
4Q12
1Q13
2Q13
3Q13
4Q13
1Q14
2Q14
3Q14
Sources: Fannie Mae, Freddie Mae and Urban
Institute.
Note: Freddie only reports the effective g-fee
on the entire book of business.
Fannie Mae Upfront Loan-Level Price Adjustments (LLPAs)
LTV
Credit Score
≤60
60.01 – 70 70.01 – 75 75.01 – 80 80.01 – 85 85.01 – 90
90.01 –
95
95.01 – 97
> 740
0.00%
0.25%
0.25%
0.50%
0.50%
0.50%
0.50%
0.50%
720 – 739
0.00%
0.25%
0.50%
0.75%
0.75%
0.75%
0.75%
0.75%
700 – 719
0.00%
0.75%
1.00%
1.25%
1.25%
1.25%
1.25%
1.25%
680 – 699
0.25%
0.75%
1.50%
2.00%
1.75%
1.50%
1.50%
1.25%
660 – 679
0.25%
1.25%
2.25%
2.75%
3.00%
2.50%
2.50%
2.00%
640 – 659
0.75%
1.50%
2.75%
3.25%
3.50%
3.00%
3.00%
2.50%
620 – 639
0.75%
1.75%
3.25%
3.25%
3.50%
3.50%
3.50%
3.25%
< 620
0.75%
1.75%
3.25%
3.25%
3.50%
3.50%
3.50%
3.50%
Product Feature (Cumulative)
Investment Property
1.75%
1.75%
1.75%
3.00%
3.75%
N/A
N/A
N/A
2-unit property
1.00%
1.00%
1.00%
1.00%
1.00%
N/A
N/A
N/A
3-4 unit property
1.00%
1.00%
1.00%
N/A
N/A
N/A
N/A
N/A
Condominiums
0.00%
0.00%
0.00%
0.75%
0.75%
0.75%
0.75%
0.75%
Sources: Fannie Mae and Urban Institute.
Note: Adverse Market Delivery Charge (AMDC) of 0.250% has been added to the LLPA numbers in the matrix by LTV and credit
score. Freddie Mac charges very comparable LLPAs.
20
GSES UNDER CONSERVATORSHIP
GSE RISK-SHARING
TRANSACTIONS
Freddie Mac – Structured Agency Credit Risk (STACR)
Date
Transaction
Reference Pool Size ($ millions)
July 24, 2013
STACR Series 2013 - DN1
$22,584.40
November 12, 2013
STACR Series 2013 - DN2
$35,327.30
February 6, 2014
STACR Series 2014 - DN1
$32,076.80
April 2, 2014
STACR Series 2014 - DN2
$28,146.98
August 6, 2014
STACR Series 2014 - DN3
$19,746.23
August 6, 2014
STACR Series 2014 – HQ1
$9,974.68
September 10, 2014
STACR Series 2014 – HQ2
$33,434.43
October 23, 2014
STACR Series 2014 – DN4
$15,740.71
October 23, 2014
STACR Series 2014 – HQ3
$8,000.61
January 26, 2015
STACR Series 2015 – DN1
$27,600.00
Freddie Mac Total Reference Collateral
$232,675.68
Percent of Freddie Mac’s Total Book of Business
15.1%
Fannie Mae – Connecticut Avenue Securities (CAS)
Date
Transaction
Reference Pool Size ($ millions)
October 24, 2013
CAS 2013 – C01
$26,756.40
January 14, 2014
CAS 2014 – C01
$29,308.70
May 28, 2014
CAS 2014 – C02
$60,818.48
July 25, 2014
CAS 2014 – C03
$78,233.73
November 19, 2014
CAS 2014 – C04
$58,872.70
Fannie Mae Total Reference Collateral
$248,990.01
Percent of Fannie Mae’s Total Book of Business
9.5%
Details of Freddie Mac’s latest capital markets transaction, STACR 2015 – DN1
Amount ($ millions)
Tranche Thickness (%)
CE (%)
Rating
Coupon
(1mL+)
$26,399.54
95.5
4.5
NR
--
M-1, M-1H, Total
$230, $46.4, $276.4
1
3.5
M:A2; D:A
125
M-2, M-2H, Total
$230, $46.4, $276.4
1
2.5
M:Baa1; D:BBB
240
M-3, M-3H, Total
$345, $69.6, $414.6
1.5
1
M: Ba1
415
B, B-H, Total
Reference Pool Size
$75, $201.4, $276.4
$27,643.50
1
0
--
NR
--
1150
--
Class
A-H
Sources: Fannie Mae, Freddie Mac and Urban Institute.
Note: Classes A-H, M-1H, M-2H, and B-H are reference tranches only. These classes are not issued or sold. The risk is retained by Fannie Mae and
Freddie Mac. “CE” = credit enhancement. Under “Rating,” “F” = Fitch, “M” = Moody’s, “D” = DBRS.
21
SERIOUS
DELINQUENCY
GSES UNDER CONSERVATORSHIP
SERIOUS
RATES
RATES ATDELINQUENCY
THE GSEs
Serious delinquency rates of GSE loans continue to decline as the legacy portfolio is resolved and the pristine, post2009 book of business exhibits very low default rates. As of December 2014, 1.89 percent of the Fannie portfolio
and 1.88 of the Freddie portfolio were seriously delinquent, down from 2.38 percent for Fannie and 2.39 percent for
Freddie in December 2013.
Serious Delinquency Rates–Fannie Mae
Single-family: Non-credit enhanced
Single-family: Credit enhanced
Single-family: Total
Percentage of total loans
16%
14%
12%
10%
8%
6%
4%
3.47%
2%
1.89%
1.62%
0%
2006
2007
2008
2009
2010
2011
2012
2013
2014
December 2014
Sources: Fannie Mae and Urban Institute.
Serious Delinquency Rates–Freddie Mac
Single-family: Non-credit enhanced
Single-family: Credit enhanced
Single-family: Total
PMI Credit Enhanced*
Percentage of total loans
10%
9%
8%
7%
6%
5%
4%
3%
2%
1%
0%
2006
3.89%
3.10%
1.88%
1.74%
2007
2008
2009
2010
2011
2012
2013
2014
December 2014
Sources: Freddie Mac and Urban Institute.
Note*: Following a change in Freddie reporting in September 2014, we switched from reporting credit enhanced delinquency rates to PMI
credit enhanced delinquency rates.
22
GSES UNDER CONSERVATORSHIP
SERIOUS DELINQUENCY RATES
Serious delinquencies for FHA and GSE single-family loans continue to decline, but remain high relative to 20052007. FHA delinquencies are declining from a higher relative starting point. GSE multifamily delinquencies have
declined to pre-crisis levels, though they did not reach problematic levels even in the worst years.
Serious Delinquency Rates–Single-Family Loans
FHA
Fannie Mae
Freddie Mac
10%
9%
8%
7%
6.18%
6%
5%
4%
3%
1.96%
2%
1%
3Q14
1Q14
3Q13
1Q13
3Q12
1Q12
3Q11
1Q11
3Q10
1Q10
3Q09
1Q09
3Q08
1Q08
3Q07
1Q07
3Q06
1Q06
3Q05
1Q05
0%
Sources: Fannie Mae, Freddie Mac, MBA Delinquency Survey and Urban Institute.
Note: Serious delinquency is defined as 90 days or more past due or in the foreclosure process.
Serious Delinquency Rates–Multifamily GSE Loans
Fannie Mae
Freddie Mac
Percentage of total loans
1.0%
0.8%
0.6%
0.4%
0.2%
0.0%
2005
0.05%
0.04%
2006
2007
2008
2009
2010
2011
2012
2013
2014
December 2014
Sources: Fannie Mae, Freddie Mac and Urban Institute.
Note: Multifamily serious delinquency rate is the unpaid balance of loans 60 days or more past due, divided by the total unpaid balance.
23
GSES UNDER CONSERVATORSHIP
REFINANCE ACTIVITY
The Home Affordable Refinance Program (HARP) refinances have begun to slow. Two factors are responsible for
this: (1) higher interest rates, leaving fewer eligible loans where refinancing is economically advantageous (in-themoney), and (2) a considerable number of borrowers who have already refinanced. Note that these latest
numbers do not reflect the dramatic drop in interest rates in the Q4 2014 and early 2015. Nonetheless, HARP
refinances total 3.259 million since the Q2 2009 program inception, accounting for 16.1 percent of all GSE
refinances in this period.
Total HARP Refinance Volume
Fannie Mae
(thousands)
Freddie Mac
Total
160
140
120
100
80
60
40
20
0
Jun-12
12
8
5
Sep-12
Dec-12
Mar-13
Jun-13
Sep-13
Dec-13
Mar-14
Jun-14
Sep-14
November 2014
HARP Refinances
November Year-to-date
2014
2014
Inception to
date
2013
2012
2011
Total refinances
134,582
1,377,906
20,250,157
4,081,911
4,750,530
3,229,066
Total HARP refinances
12,492
201,337
3,259,291
892,914
1,074,769
400,024
Share 80–105 LTV
74.5%
72.3%
69.9%
56.4%
56.4%
85.0%
Share 105–125 LTV
16.6%
17.2%
17.2%
22.4%
22.4%
15.0%
Share >125 LTV
8.9%
10.5%
12.9%
21.2%
21%
0%
All other streamlined
refinances
19,947
247,858
3,501,062
735,210
729,235
785,049
Sources: FHFA Refinance Report and Urban Institute.
24
GSES UNDER CONSERVATORSHIP
GSE LOANS:
POTENTIAL REFINANCES
To qualify for HARP, a loan must be originated before the June 2009 cutoff date, have a marked-to-market loan-tovalue (MTM LTV) ratio above 80, and have no more than one delinquent payment in the past year and none in the
past six months. There are 809,330 eligible loans, but 37 percent are out-of-the-money because the closing cost
would exceed the long-term savings, leaving 509,517 loans where a HARP refinance is both permissible and
economically advantageous for the borrower. Loans below the LTV minimum but meeting all other HARP
requirements are eligible for GSE streamlined refinancing. Of the 6,568,948 loans in this category, 5,625,680 are
in-the-money.
More than 70 percent of the GSE book of business that meets the pay history requirements was originated after
the June, 2009 cutoff date. FHFA Director Mel Watt announced in May 2014 that they are not planning to extend
the date.
Total loan count
26,924,865
Loans that do not meet pay history requirement
Loans that meet pay history requirement:
871,297
26,053,568
Pre-June 2009 origination
7,378,278
Post-June 2009 origination
18,675,290
Loans Meeting HARP Pay History Requirements
Pre-June 2009
LTV category
In-the-money
Out-of-the-money
Total
≤80
5,625,680
943,268
6,568,948
>80
509,517
299,813
809,330
Total
6,135,197
1,243,081
7,378,278
LTV category
In-the-money
Out-of-the-money
Total
≤80
4,969,486
10,645,984
15,615,470
>80
1,385,673
1,674,147
3,059,820
Total
6,355,159
12,320,131
18,675,290
Post-June 2009
Sources: CoreLogic Prime Servicing as of December 2014.
Note: Figures are scaled up from source data to account for data coverage of the GSE active loan market (based on MBS data from eMBS).
Shaded box indicates HARP-eligible loans that are in-the-money.
25
MODIFICATION ACTIVITY
HAMP ACTIVITY
In 2014 Q3, new HAMP activity declined in terms of trial and permanent mods relative to a year prior. New trial mods
have averaged just over 30,000 per quarter in 2014, compared to over 45,000 per quarter in 2013. Cumulative
permanent HAMP mods started now total over 1.4 million, and 961,700 of these are active permanent mods.
New HAMP Modifications
New Trial Mods Started
New Permanent Mods Started
New Active Permanent Mods Started
Number of mods
(thousands)
450
400
350
300
250
200
150
100
33
50
30
0
3.2
2009
2010
2011
2012
2013
2014
2014 Q3
Sources: U.S. Treasury Making Home Affordable and Urban Institute.
Cumulative HAMP Modifications
All Trials Mods Started
All Permanent mods started
Active Permanent Mods
Number of mods
(millions)
2.5
2.2
2
1.5
1.4
1
0.96
0.5
0
2009 Q1
2009 Q3
2010 Q1
2010 Q3
2011 Q1
2011 Q3
Sources: U.S. Treasury Making Home Affordable and Urban Institute.
2012 Q1
2012 Q3
2013 Q1
2013 Q3
2014 Q1
2014 Q3
26
MODIFICATION ACTIVITY
MODIFICATION BY TYPE OF
ACTION AND BEARER OF RISK
The share of principal reduction modifications peaked at 20 percent in December 2012 and has now dropped to
6.8 percent. This is to be expected, as increasing home prices have increased equity, reducing the need for
principal reduction and making such modifications less likely to be net-present-value positive. Portfolio loans are
the most likely candidates for principal reduction, followed by private investor loans, because the GSEs and
FHA/VA generally do not allow this type of modification. The FHFA is studying whether a change in this policy is
warranted for GSE modifications.
Changes in Loan Terms for Modifications
Modification Quarter
13Q2
13Q3
13Q4
14Q1
14Q2
14Q3
One quarter
% change
One year
% change
Capitalization
81.6
83.5
87.7
74.3
59
71.1
-20.6
-14.8
Rate reduction
81.0
78.9
76.7
73.3
71.9
66.5
-7.5
-15.7
Rate freeze
5.2
5.5
7
6.5
7.1
7.5
5.9
38.2
Term extension
67.7
69.3
75.9
78
84
82.0
-2.4
18.4
Principal reduction
12.2
13.6
10.5
8.1
5
6.8
36.7
-49.7
Principal deferral
20.5
25.3
30.6
25.1
11.5
15.9
37.8
-37.3
Not reported*
1.5
2.2
0.7
0.7
0.7
0.5
-28.0
-77.4
Sources: OCC Mortgage Metrics Report for the Third Quarter of 2014 and Urban Institute.
Note: This table presents modifications of each type as a share of total modifications. Columns sum to over 100% because loans often receive
modifications with multiple features.
*Processing constraints at some servicers prevented them from reporting specific modified term(s).
Type of Modification Action by Investor and Product Type
Fannie Mae
Freddie Mac
Governmentguaranteed
Private
Investor
Portfolio
Overall
Capitalization
98.8%
96.8%
31.9%
93.9%
93.3%
71.1%
Rate reduction
48.8%
53.8%
77.4%
71.0%
69.4%
66.5%
Rate freeze
9.4%
5.6%
7.3%
5.0%
10.4%
7.5%
Term extension
93.5%
95.2%
97.5%
28.4%
59.0%
82.0%
Principal reduction
0.1%
0.0%
0.4%
21.4%
27.8%
6.8%
Principal deferral
14.3%
14.9%
12.1%
21.2%
24.5%
15.9%
Not reported*
0.1%
0.0%
0.6%
1.2%
0.7%
0.5%
Sources: OCC Mortgage Metrics Report for the Third Quarter of 2014 and Urban Institute.
Note: This table presents modifications of each type as a share of total modifications. Columns sum to over 100% because loans often receive
modifications with multiple features.
*Processing constraints at some servicers prevented them from reporting specific modified term(s).
27
MODIFICATION ACTIVITY
MODIFICATIONS AND
LIQUIDATIONS
Total modifications (HAMP and proprietary) are now roughly equal to total liquidations. Hope Now reports show
7,343,260 borrowers have received a modification since Q3 2007, compared with 7,454,470 liquidations in the
same period. Both liquidation and modification activity are slowing significantly, and through November 2014
ended far below their comparable 2013 total. Only 33,572 modifications were completed in November 2014, a
50 percent reduction from the monthly rate in early 2013.
Loan Modifications and Liquidations
Number of loans (thousands)
1,600
1,400
1,200
1,000
579.5
800
137.7
331.7
600
400
200
2008
2009
2010
2011
2012
2013
Proprietary mods
Liquidations
0
2007 (Q3Q4)
HAMP mods
2014
(Ann.)
Sources: Hope Now Reports and
Urban Institute.
Note: Liquidations includes both
foreclosure sales and short sales.
Annualized figure based on data
from November 2014.
Cumulative Modifications and Liquidations
7.5
Number of loans (millions)
8
5.9
7
6
HAMP mods
5
Proprietary mods
4
Liquidations
1.4
3
2
1
Sources: Hope Now Reports and
Urban Institute.
Note: Liquidations includes both
foreclosure sales and short sales.
0
2007 (Q3Q4)
2008
2009
2010
2011
2012
2013
2014 YTD
28
MODIFICATION ACTIVITY
MODIFICATION REDEFAULT
RATES BY BEARER OF THE RISK
Redefault rates on modified loans have come down dramatically from 2008 to 2014. For the period as a whole,
the steepest drops have been on private label modifications. More recently, there have been sharp declines in the
redefault rates on government-guaranteed modifications, although this product type still has higher redefault
rates than others.
Redefault Rate 12 Months after Modification
80%
70%
Fannie Mae
Freddie Mac
Government-guaranteed
Private
Redefault rate
60%
50%
40%
30%
20%
Portfolio Loans
10%
Overall
0%
2008
2009
2010
2011
2012
2013
Year of modification
Sources: OCC Mortgage Metrics Report for the Third Quarter of 2014 and Urban Institute.
Redefault Rate 24 Months after Modification
80%
70%
Fannie Mae
Freddie Mac
Government-guaranteed
Private
Portfolio loans
Overall
Redefault rate
60%
50%
40%
30%
20%
10%
0%
2008
2009
2010
2011
2012
Year of modification
Sources: OCC Mortgage Metrics Report for the Third Quarter of 2014 and Urban Institute.
29
AGENCY ISSUANCE
AGENCY GROSS AND
NET ISSUANCE
While refinancing activity fell off due to higher interest rates through the course of 2014, newly reduced rates and
lowered FHA premiums make the fate of agency issuance uncertain. Agency gross issuance totaled $87.2 billion in
the first month of 2015, a 25 percent increase year-over-year. Net issuance, which excludes repayments,
prepayments, and refinances on outstanding mortgages, remains low and dominated by Ginnie Mae. This is
unsurprising, given the increased role of FHA and VA since the crisis.
Agency Gross Issuance
Agency Net Issuance
Issuance
Year
GSEs
Ginnie Mae
Total
Issuance
Year
GSEs
Ginnie Mae
Total
2000
$360.6
$102.2
$462.8
2000
$159.8
$29.3
$189.1
2001
$885.1
$171.5
$1,056.6
2001
$367.8
-$9.9
$357.9
2002
$1,238.9
$169.0
$1,407.9
2002
$357.6
-$51.2
$306.4
2003
$1,874.9
$213.1
$2,088.0
2003
$335.0
-$77.6
$257.4
2004
$872.6
$119.2
$991.9
2004
$83.3
-$40.1
$43.2
2005
$894.0
$81.4
$975.3
2005
$174.4
-$42.2
$132.1
2006
$853.0
$76.7
$929.7
2006
$313.6
$0.3
$313.8
2007
$1,066.2
$94.9
$1,161.1
2007
$514.7
$30.9
$545.5
2008
$911.4
$267.6
$1,179.0
2008
$314.3
$196.4
$510.7
2009
$1,280.0
$451.3
$1,731.3
2009
$249.5
$257.4
$506.8
2010
$1,003.5
$390.7
$1,394.3
2010
-$305.5
$198.2
-$107.3
2011
$879.3
$315.3
$1,194.7
2011
-$133.4
$149.4
$16.0
2012
$1,288.8
$405.0
$1,693.8
2012
-$46.5
$118.4
$71.9
2013
$1,176.6
$393.6
$1,570.1
2013
$66.5
$85.8
$152.3
2014
$650.9
$296.3
$947.2
2014
$30.3
$59.8
$90.1
2015 YTD
$59.7
$27.5
$87.2
2015 YTD
-$0.3
$2.3
$2.0
%Change
year-over-year
23.7%
27.4%
24.8%
%Change
year-over-year
-156.1%
-58.4%
-67.6%
2015 Ann.
$715.8
$330.2
$1,046.0
2015 Ann.
-$3.9
$27.7
$23.8
Sources: eMBS and Urban Institute.
Note: Dollar amounts are in billions. Annualized figure based on data
from January 2015.
Sources: eMBS and Urban Institute.
Note: Dollar amounts are in billions. Annualized figure
based on data from January 2015.
30
AGENCY GROSS AND NET
AGENCY GROSS
ISSUANCE &
ISSUANCE
BY MONTH
FED PURCHASES
AGENCY ISSUANCE
Monthly Gross Issuance
Sources: eMBS, Federal Reserve Bank of New York and Urban Institute.
2015
2014
2013
2012
2011
Ginnie Mae
2010
2009
2008
Freddie Mac
2007
2006
2005
2004
2003
Fannie Mae
2002
2001
While government and GSE lending
($ billions)
have dominated the mortgage
market since the crisis, there has
250
been a change in the mix. The Ginnie
Mae share reached a peak of 28
200
percent of total agency issuance in
2010, declined to 25 percent in
2013, and has since then risen to 32
150
percent in January 2015. The recent
increase in the Ginnie Mae share
100
reflects the decline in refinance
activity, as Ginnie Mae is less
impacted by this decline. We would
50
expect the Ginnie Mae share to
decline in the months ahead, as we
see the impact of lower rates and
0
more refinance activity.
January 2015
Fed Absorption of Agency Gross Issuance
In Q1 2014, the Fed began to taper, but gross issuance dropped even more, and Fed absorption reached 74 percent.
Since then, gross issuance increased and the Fed continued to taper, resulting in a steady decline of the absorption
share to 53 and 36 percent in Q2 and Q3, respectively. In October, the Fed announced the end of its purchase
program. However, buying continued at a much reduced level, as the Fed kept reinvesting funds from pay downs on
mortgages and agency debentures into the mortgage market. In January 2015, total Fed purchase edged up to $26
billion, yielding 29.8 percent Fed absorption of gross issuance, up from Q4 2014’s 25 percent.
Gross issuance
($ billions)
Total Fed purchases
250
200
150
100
50
0
2001
2002
2003
2004
2005
2006
2007
2008
Sources: eMBS, Federal Reserve Bank of New York and Urban Institute.
2009
2010
2011
2012
2013
2014
2015
January 2015
31
AGENCY ISSUANCE
MORTGAGE INSURANCE
ACTIVITY
MI Activity
Overall mortgage insurance activity experienced a second consecutive quarter of significant increase in Q3 2014,
reaching $123 billion. Private mortgage insurers led the way, with a $10 billion quarterly increase and a jump in
market share to 44 percent (41 percent for 2014 overall). The FHA share dropped to 30.8 percent for the quarter,
and 34 percent for 2014 overall, despite an uptick in Q3 endorsements to $38 billion from $36 billion.
Total private primary MI
FHA
VA
Total
200
150
$123.0
100
$54.0
50
$38.0
0
1Q11
$31.1
2Q11
3Q11
4Q11
1Q12
2Q12
3Q12
4Q12
1Q13
2Q13
3Q13
4Q13
1Q14
2Q14
3Q14
Sources: Inside Mortgage Finance and Urban Institute.
MI Market Share
Total private primary MI
FHA
VA
24.7%
100%
90%
80%
34.0%
70%
60%
50%
40%
41.3%
30%
20%
10%
0%
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
Q1-3
Sources: Inside Mortgage Finance and Urban Institute.
32
AGENCY ISSUANCE
MORTGAGE INSURANCE
ACTIVITY
FHA premiums rose significantly in the years following the housing crash, with annual premiums rising 170%
from 2008 to 2013 as FHA worked to shore up its finances. In a move announced by President Obama just after
the new year, and effective January 26, annual premiums were cut by 50 bps. We expect this reduction to
significantly mitigate FHA’s problem of adverse selection, in which lower-FICO borrowers disproportionately
gravitate to FHA financing over GSE with PMI. As shown in the bottom table, a borrower putting 3.5% down will
now find FHA more economical regardless of their FICO score.
FHA MI Premiums for Typical Purchase Loan
Case number date
Upfront mortgage insurance premium
(UFMIP) paid
Annual mortgage insurance
premium (MIP)
1/1/2001 - 7/13/2008
7/14/2008 - 4/5/2010*
4/5/2010 - 10/3/2010
10/4/2010 - 4/17/2011
4/18/2011 - 4/8/2012
4/9/2012 - 6/10/2012
6/11/2012 - 3/31/2013a
4/1/2013 – 1/25/2015b
Beginning 1/26/2015c
150
175
225
100
100
175
175
175
175
50
55
55
90
115
125
125
135
85
Sources: Ginnie Mae and Urban Institute.
Note: A typical purchase loan has an LTV over 95 and a loan term longer than 15 years. Mortgage insurance premiums are listed in basis points.
* For a short period in 2008 the FHA used a risk based FICO/LTV matrix for MI.
a
Applies to purchase loans less than or equal to $625,500. Those over that amount have an annual premium of 150 bps.
b
Applies to purchase loans less than or equal to $625,500. Those over that amount have an annual premium of 155 bps.
c
Applies to purchase loans less than or equal to $625,500. Those over that amount have an annual premium of 105 bps.
Initial Monthly Payment Comparison: FHA vs. PMI
Assumptions
Property Value
Loan Amount
LTV
Base Rate
Conforming
FHA
FICO
620 - 639
$250,000
$241,250
96.5
3.90%
3.66%
640 - 659
660 - 679
680 - 699
700 - 719
720 - 739
740 - 759
760 +
FHA MI Premiums
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
FHA UFMIP
0.85
0.85
0.85
0.85
0.85
0.85
0.85
0.85
FHA MIP
PMI
3.75
3.00
2.50
1.75
1.75
1.25
1.00
1.00
GSE AMDC & LLPA*
1.48
1.48
1.48
1.31
1.31
1.10
1.10
1.05
PMI Annual MIP
Monthly Payment
$1,295
$1,295
$1,295
$1,295
$1,295
$1,295
$1,295
$1,295
FHA
$1,542
$1,520
$1,506
$1,451
$1,451
$1,394
$1,387
$1,377
PMI
($247)
($225)
($211)
($156)
($156)
($99)
($92)
($82)
PMI Advantage
Sources: Genworth Mortgage Insurance, Ginnie Mae and Urban Institute.
Note: Mortgage insurance premiums listed in percentage points. Grey shade indicates FHA monthly payment is more favorable. The GSE
monthly payment calculation does not include special programs like Fannie Mae’s MyCommunitMortgage (MCM) and Freddie Mac’s Home
Possible (HP), both offer more favorable rates for low- to moderate-income borrowers.
33
*AMDC=Adverse Market Delivery Charge and LLPA= Loan Level Price Adjustment, both described in detail on page 20.
SPECIAL FEATURE: LOAN LEVEL GSE CREDIT DATA
FANNIE MAE COMPOSITION
Since 2008, the composition of loans purchased by Fannie Mae has shifted towards borrowers with higher FICO
scores. For example, 72.4 percent of loans originated from 2011 to 2013 were for borrowers with FICO scores
above 750, compared to 40.4 percent of borrowers in 2007 and 36.6 percent from 1999-2004.
Balance on 30-year, Fixed-rate, Full-doc, Amortizing Loans
Origination
Year
1999-2004
2005
2006
2007
2008
2009-2010
2011-2013
LTV
Origination
FICO
≤70
70 to 80
80 to 90
>90
Total
≤700
9.4%
15.1%
4.5%
4.5%
33.4%
700 to 750
9.2%
14.2%
3.4%
3.2%
30.0%
>750
15.5%
16.1%
2.7%
2.3%
36.6%
Total
34.0%
45.3%
10.7%
10.0%
100.0%
≤700
12.5%
15.5%
3.4%
2.3%
33.9%
700 to 750
9.8%
13.3%
2.2%
1.4%
26.6%
>750
17.3%
18.6%
2.1%
1.4%
39.5%
Total
39.6%
47.5%
7.7%
5.2%
100.0%
≤700
12.6%
16.2%
3.6%
2.2%
34.6%
700 to 750
8.8%
13.6%
2.2%
1.3%
25.9%
>750
15.7%
20.0%
2.4%
1.4%
39.5%
Total
37.2%
49.8%
8.2%
4.9%
100.0%
≤700
10.7%
15.2%
5.4%
3.2%
34.5%
700 to 750
7.7%
12.5%
3.1%
1.7%
25.0%
>750
15.1%
20.0%
3.3%
2.1%
40.4%
Total
33.5%
47.7%
11.8%
7.0%
100.0%
≤700
7.6%
7.2%
3.0%
2.1%
19.9%
700 to 750
7.8%
11.9%
4.1%
2.7%
26.5%
>750
18.9%
25.5%
5.8%
3.4%
53.6%
Total
34.3%
44.7%
12.9%
8.2%
100.0%
≤700
3.6%
2.9%
0.3%
0.2%
6.9%
700 to 750
8.2%
10.8%
1.7%
0.8%
21.5%
>750
32.3%
33.6%
4.0%
1.7%
71.6%
Total
44.1%
47.3%
5.9%
2.7%
100.0%
≤700
2.5%
3.4%
0.6%
0.8%
7.3%
700 to 750
5.7%
9.3%
2.3%
3.0%
20.2%
>750
26.6%
32.8%
6.5%
6.5%
72.4%
Total
34.8%
45.5%
9.4%
10.3%
100.0%
36.4%
46.1%
9.5%
8.1%
100.0%
Total
Sources: Fannie Mae and Urban Institute.
Note: Fannie Mae loan level credit data includes loans originated from Q1 1999 to Q4 2013. The percentages are weighted by origination
balance.
34
SPECIAL FEATURE: LOAN LEVEL GSE CREDIT DATA
FANNIE MAE DEFAULT RATE
While the composition of Fannie Mae loans originated in 2007 was similar to that of 2004 and earlier vintage years,
2007 loans experienced a much higher default rate due to the sharp drop in home values in the recession.
Originations from 2009 and later have pristine credit characteristics and a more favorable home price environment,
contributing to very low default rates.
Default Rate on 30-year, Fixed-rate, Full-doc, Amortizing Loans
Origination
Year
Origination
FICO
≤700
700 to 750
1999-2004
>750
Total
≤700
700 to 750
2005
>750
Total
≤700
700 to 750
2006
>750
Total
≤700
700 to 750
2007
>750
Total
≤700
700 to 750
2008
>750
Total
≤700
700 to 750
2009-2010
>750
Total
≤700
700 to 750
2011-2013
>750
Total
Total
LTV
Total
≤70
70 to 80
80 to 90
>90
3.6%
4.6%
6.2%
7.1%
4.9%
1.1%
1.9%
2.9%
3.0%
1.9%
0.4%
0.8%
1.5%
1.7%
0.7%
1.5%
2.4%
4.0%
4.6%
2.5%
13.4%
17.3%
20.1%
21.7%
16.4%
6.2%
9.7%
12.8%
13.2%
8.9%
2.2%
4.5%
7.2%
8.2%
3.7%
6.7%
10.1%
14.5%
15.6%
9.4%
17.9%
22.6%
26.4%
27.8%
21.6%
8.6%
13.6%
16.8%
17.3%
12.3%
2.9%
6.0%
9.5%
9.9%
5.1%
9.4%
13.5%
18.8%
19.9%
12.7%
19.3%
23.9%
32.0%
32.8%
24.6%
8.5%
14.1%
20.9%
20.1%
13.6%
2.8%
6.1%
12.1%
11.8%
5.6%
9.4%
13.9%
23.5%
23.4%
14.2%
14.1%
17.5%
24.8%
24.9%
18.1%
5.0%
8.5%
14.1%
13.6%
8.8%
1.2%
2.9%
6.8%
7.3%
3.0%
4.9%
6.8%
13.3%
13.9%
7.6%
3.0%
4.1%
4.4%
5.2%
3.6%
0.7%
1.5%
2.0%
2.4%
1.3%
0.2%
0.5%
0.9%
1.2%
0.4%
0.5%
0.9%
1.4%
1.8%
0.8%
0.5%
0.6%
0.5%
0.5%
0.5%
0.1%
0.2%
0.2%
0.2%
0.2%
0.0%
0.0%
0.1%
0.1%
0.0%
0.1%
0.1%
0.1%
0.2%
0.1%
2.1%
3.4%
5.8%
5.2%
3.3%
Sources: Fannie Mae and Urban Institute.
Note: Fannie Mae loan level credit data includes loans originated from Q1 1999 to Q4 2013, with performance information on these loans
through Q32014. Default is defined as more than six months delinquent or disposed of via short sales, third-party sales, deeds-in-lieu of
foreclosure, or real estate owned (REO acquisitions).
35
SPECIAL FEATURE: LOAN LEVEL GSE CREDIT DATA
FREDDIE MAC COMPOSITION
Since 2008, the composition of loans purchased by Freddie Mac has shifted towards borrowers with higher FICO
scores. For example, 69.5 percent of loans originated from 2011 to Q2 2013 were for borrowers with FICO scores
above 750, compared to 38.9 percent in 2007 and 33.2 percent from 1999-2004.
Balance on 30-year, Fixed-rate, Full-doc, Amortizing Loans
Origination
Year
Origination
FICO
≤700
700 to 750
1999-2004
>750
Total
≤700
700 to 750
2005
>750
Total
≤700
700 to 750
2006
>750
Total
≤700
700 to 750
2007
>750
Total
≤700
700 to 750
2008
>750
Total
≤700
700 to 750
2009-2010
>750
Total
≤700
700 to 750
2011-2Q13
>750
Total
Total
LTV
Total
≤70
70 to 80
80 to 90
>90
7.7%
16.6%
5.5%
5.6%
35.4%
8.9%
15.9%
3.4%
3.2%
31.4%
13.6%
15.5%
2.3%
1.8%
33.2%
30.2%
48.0%
11.2%
10.6%
100.0%
10.6%
16.9%
3.3%
3.0%
33.8%
9.4%
15.5%
2.0%
1.7%
28.5%
15.8%
18.8%
1.7%
1.4%
37.7%
35.7%
51.3%
7.0%
6.0%
100.0%
10.0%
17.3%
3.4%
3.3%
34.0%
8.3%
16.1%
1.9%
1.5%
27.9%
14.4%
20.7%
1.7%
1.3%
38.1%
32.7%
54.1%
7.1%
6.1%
100.0%
9.1%
15.5%
4.6%
4.8%
34.1%
7.5%
14.3%
2.6%
2.6%
27.0%
14.3%
19.5%
2.5%
2.6%
38.9%
30.9%
49.4%
9.7%
10.0%
100.0%
7.3%
8.7%
3.1%
2.2%
21.3%
9.1%
13.1%
3.7%
2.5%
28.3%
21.6%
21.5%
4.7%
2.6%
50.4%
38.0%
43.3%
11.5%
7.2%
100.0%
3.9%
3.2%
0.3%
0.2%
7.7%
9.3%
11.9%
1.7%
0.9%
23.8%
32.4%
31.0%
3.6%
1.4%
68.4%
45.7%
46.1%
5.6%
2.5%
100.0%
3.0%
3.1%
0.6%
0.7%
7.4%
7.4%
10.8%
2.3%
2.7%
23.1%
26.9%
32.0%
5.4%
5.1%
69.5%
37.3%
46.0%
8.2%
8.5%
100.0%
35.0%
47.8%
9.1%
8.0%
100.0%
Sources: Freddie Mac and Urban Institute.
Note: Freddie Mac loan level credit data includes loans originated from Q1 1999 to Q2 2013. The percentages are weighted by
origination balance.
36
SPECIAL FEATURE: LOAN LEVEL GSE CREDIT DATA
FREDDIE MAC DEFAULT RATE
While the composition of Freddie Mac loans originated in 2007 was similar to that of 2004 and earlier vintage years,
2007 loans experienced a much higher default rate due to the sharp drop in home values in the recession.
Originations from 2009 and later have pristine credit characteristics and a more favorable home price environment,
contributing to very low default rates.
Default Rate on 30-year, Fixed-rate, Full-doc, Amortizing Loans
Origination
Year
Origination
FICO
≤700
700 to 750
1999-2004
>750
Total
≤700
700 to 750
2005
>750
Total
≤700
700 to 750
2006
>750
Total
≤700
700 to 750
2007
>750
Total
≤700
700 to 750
2008
>750
Total
≤700
700 to 750
2009-2010
>750
Total
≤700
700 to 750
2011-2Q13
>750
Total
Total
LTV
Total
≤70
70 to 80
80 to 90
>90
2.8%
3.9%
6.1%
6.6%
4.5%
0.9%
1.5%
2.5%
2.7%
1.6%
0.3%
0.7%
1.3%
1.6%
0.7%
1.1%
2.1%
4.0%
4.6%
2.3%
11.2%
15.4%
18.3%
19.8%
14.8%
5.4%
8.8%
11.8%
12.1%
8.1%
1.9%
4.3%
6.7%
7.7%
3.5%
5.5%
9.3%
13.7%
14.9%
8.6%
14.9%
19.8%
22.9%
25.2%
19.2%
7.5%
12.0%
14.4%
14.5%
11.0%
2.5%
5.7%
8.3%
9.1%
4.7%
7.6%
12.1%
17.0%
19.0%
11.4%
15.6%
21.0%
26.4%
29.0%
21.4%
7.3%
12.7%
16.9%
17.2%
12.1%
2.5%
6.0%
9.6%
10.8%
5.2%
7.5%
12.7%
19.5%
21.3%
12.6%
11.8%
15.6%
21.6%
20.8%
15.7%
4.2%
7.7%
12.1%
10.8%
7.4%
1.3%
3.0%
6.3%
5.8%
2.7%
4.0%
7.0%
12.3%
12.0%
6.8%
2.2%
3.1%
3.5%
3.6%
2.7%
0.5%
1.2%
1.5%
1.7%
1.0%
0.1%
0.4%
0.8%
0.8%
0.3%
0.4%
0.8%
1.1%
1.4%
0.7%
0.2%
0.3%
0.2%
0.4%
0.2%
0.1%
0.1%
0.1%
0.1%
0.1%
0.0%
0.0%
0.0%
0.1%
0.0%
0.0%
0.1%
0.1%
0.1%
0.1%
2.1%
3.9%
6.2%
6.9%
3.7%
Sources: Freddie Mae and Urban Institute.
Note: Freddie Mac loan level credit data includes loans originated from Q1 1999 to Q2 2013, with performance information on these loans
through Q4 2013. Default is defined as six months delinquent or disposed of via short sales, third-party sales, deeds-in-lieu of foreclosure, or
real estate owned (REO acquisitions).
37
SPECIAL FEATURE: LOAN LEVEL GSE CREDIT DATA
DEFAULT RATE BY VINTAGE
With cleaner books of business and the housing recovery underway, default rates for the GSEs are much lower than
they were just a few years ago. For Fannie Mae and Freddie Mac’s 1999-2003 vintages, cumulative defaults total
around 2 percent, while cumulative defaults for the 2007 vintage are above 14 percent and 13 percent, respectively.
For both Fannie Mae and Freddie Mac, cumulative defaults from 2009-10 and 2011-2013 are on pace to fall below
pre-2004 levels. For Fannie loans 3.5 years after origination, the cumulative default rates from 2009-10 and 20112013 are about 0.59 and 0.10 percent, respectively, compared to the cumulative default rate from 1999-2003 of
0.64 percent. For Freddie loans 3 years after origination, the cumulative default rates total 0.37 percent from 200910 and 0.07 percent from 2011-Q2 2013, compared to the rate from 1999-2003 of 0.42 percent.
Fannie Mae Cumulative Default Rate by Vintage Year
16%
1999-2003
14%
2004
12%
2005
10%
2006
8%
2007
6%
2008
4%
2009-2010
2%
0%
2011-2013
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Loan age in years
Sources: Fannie Mae and Urban Institute.
Freddie Mac Cumulative Default Rate by Vintage Year
14%
1999-2003
12%
2004
10%
2005
8%
2006
6%
2007
2008
4%
2009-2010
2%
0%
2011-2Q2013
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Loan age in years
Sources: Freddie Mac and Urban Institute.
38
SPECIAL FEATURE: LOAN LEVEL GSE CREDIT DATA
REPURCHASE RATE BY VINTAGE
These figures show the cumulative percentage of fixed-rate, full documentation, amortizing 30-year loans of a given
vintage that Fannie and Freddie have put back to lenders due to reps and warrants violations. Note that the putbacks are generally quite small, with the exception of the 2006-2008 vintages. These numbers exclude loans put back
through global settlements, which are not done at the loan level. Moreover, lenders’ attitudes are formed by the total
share of put-backs on their books. The database used in this analysis, while very characteristic of new production,
excludes many loans that are likely to be put back, including limited documentation loans, non-traditional products
(such as interest-only loans), and loans with pool insurance policies.
Fannie Mae Repurchase Rate by Vintage Year
0.9%
0.8%
0.7%
1999-2003
0.6%
2004
0.5%
2005
0.4%
2006
0.3%
2007
0.2%
2008
0.1%
2009-2010
0.0%
0
1
2
3
4
5
6
7
8
9
10
11
12
13
2011-2013
Loan age in years
Sources: Fannie Mae and Urban Institute.
Freddie Mac Repurchase Rate by Vintage Year
2.0%
1.8%
1.6%
1999-2003
1.4%
2004
1.2%
2005
1.0%
2006
0.8%
2007
0.6%
0.4%
2008
0.2%
2009-2010
0.0%
2011-2Q2013
1
2
3
4
Sources: Freddie Mac and Urban Institute.
5
6
7
8
9
10
11
12
13 14
Loan age in years
39
SPECIAL FEATURE: LOAN LEVEL GSE CREDIT DATA
FREDDIE MAC LOSS SEVERITY
Freddie Mac recently released data
showing the status of loans after
they have experienced a credit event
(default). A credit event is defined as
a delinquency of 180 days or more, a
deed-in-lieu, short sale, foreclosure
sale or REO sale. The top right chart
shows we expect 77.6 percent of
loans experiencing a credit event to
eventually liquidate. The bottom left
table show the severities of the 54
percent of the loans that have
liquidated after experiencing a
credit event. The bottom right table
(the product of the top right and
bottom left tables) shows
the estimated severity on all loans
that have experienced a credit
event.
Estimated liquidation rate on loans that have experienced a
credit event, by origination year and LTV
Year
<=60
60-80
>80
Total
1999-2004
57.0%
72.1%
78.1%
73.9%
2005
65.5%
80.0%
84.7%
80.0%
2006
66.4%
81.2%
84.8%
80.7%
2007
65.6%
79.4%
83.5%
79.7%
2008
62.8%
76.3%
80.6%
76.5%
2009-2010
67.0%
81.1%
88.0%
80.3%
2011-2Q2013
69.2%
81.7%
88.0%
81.6%
Total
63.2%
77.5%
81.1%
77.6%
Estimated severity on loans that have experienced
a credit event, by origination year and LTV
Severity at liquidation,
by origination year and LTV
Year
<=60
60-80
>80
Total
Year
60-80
>80
1999-2004
19.3% 30.4%
15.9%
23.2%
1999-2004
22%
12%
2005
27.7% 39.9%
26.0%
36.0%
2005
32%
22%
2006
33.5% 44.1%
28.0%
40.0%
2006
36%
24%
2007
35.9% 44.2%
28.7%
38.5%
2007
35%
24%
2008
28.6% 40.9%
25.8%
34.6%
2008
31%
21%
2009-2010
20.5% 30.7%
14.1%
26.5%
2009-2010
25%
12%
8.2%
23.8%
8.3%
15.9%
2011-2Q2013
19%
7%
29.2% 39.9%
23.1%
33.9%
Total
31%
19%
2011-2Q2013
Total
Sources: Freddie Mac Single Family Loan-Level Dataset and Urban Institute calculations.
Note: Freddie Mac loan level credit data includes loans originated from Q1 1999 to Q2 2013, with performance information on these
loans through Q4 2013.
40
RELATED HFPC WORK
PUBLICATIONS AND EVENTS
Upcoming Events
Data Talk: Credit Scoring—Getting Beyond the Usual - March 12, 2015, 12:30-2:30pm
Please join us for this data talk focusing on new developments in credit scoring; these developments which
incorporate data not traditionally used in scoring, allow for an expansion in the number of consumers who can be
evaluated, potentially expanding the credit box. Presenters include Sarah Davies, Senior Vice President,
Analytics, Research and Product Management, Vantage Score, and Michael Turner, Founder, President and CEO,
PERC. Ken Brevoort, Chief of Credit Information and Policy, CFPB will be the discussant, Laurie Goodman, Director,
Housing Finance Policy Center, Urban Institute will moderate.
For more information, contact Alison Rincon (Arincon@urban.org)
Publications
Blog Posts
The US Treasury’s Credit Rating Agency Exercise
Authors: Laurie Goodman, Jim Parrott
Date: February 6, 2015
More than one in three FHA borrowers could save money
by refinancing today
Authors: Karan Kaul, Laurie Goodman and Jun Zhu
Date: February 16th, 2015
Loss Severity on Residential Mortgages
Authors: Laurie Goodman, Jun Zhu
Date: February 1, 2015
FHFA’s Federal Home Loan Bank Members Proposal
Overshoots the Mark
Authors: Laurie Goodman, Jim Parrott and Karan Kaul
Date: January 13, 2015
The five major forces driving down mortgage interest rates
Authors: Ellen Seidman and Wei Li
Date: February 12, 2015
Private mortgage insurance is better than expected at
protecting taxpayers from losses
Authors: Laurie Goodman and Jun Zhu
Date: February 3, 2015
A Review and Critique of the 2014 actuarial assessment of
FHA’s Mortgage Insurance Fund
Three predictions about mortgage affordability in 2015
Author: Laurie Goodman
Authors: Karan Kaul
Date: January 6, 2014
Date: February 2, 2015
Servicing: An Underappreciated Constraint to Access to
Credit
Author: Laurie Goodman
Date: December 16, 2014
What a new measure of mortgage denials reveals about
mortgage credit access
Authors: Ellen Seidman and Wei Li
Date: January 30, 2015
A Better Measure of Mortgage Application Denial Rates
Authors: Wei Li and Laurie Goodman
Date: December 2, 2014
Is the homeownership safety net unraveling for seniors?
Authors: Taz George and Ellen Seidman
Date: January 28, 2015
Measuring Mortgage Credit Availability Using Ex-Ante
Probability of Default
Authors: Wei Li and Laurie Goodman
Date: November 18, 2014
Four impacts of the Federal Housing Administration’s
premium cut
Authors: Karan Kaul and Bing Bai
Date: January 21, 2015
Measuring Mortgage Credit Accessibility
Authors: Wei Li, Laurie Goodman, Ellen Seidman, Jim
Parrott, Jun Zhu, and Bing Bai
Date: November 3, 2014
41
RELATED HFPC WORK
PUBLICATIONS AND EVENTS
Publications
Blog Posts
Comment Letter on the CFPB’s HMDA mortgage
data proposal
Authors: Ellen Seidman, Laurie Goodman, Wei Li
Jim Parrott, Kathryn L.S. Pettit, Carlos Martin,
and Peter Tatian
Date: October 22, 2014
Obtaining a mortgage loan: How do we know if it’s too
hard, too easy, or just about right?
Authors: Maia Woluchem and Taz George
Date: January 12, 2015
Assessing the Proposed Housing Goals
Authors: Jim Parrott, Laurie Goodman, Wei Li
Ellen Seidman, and Jun Zhu
Date: October 22, 2014
A New View of the Housing Boom and Bust
Authors: Bing Bai and Taz George
Date: September 17, 2014
Charting the Course to a Single Security
Authors: Laurie Goodman and Lewis Ranieri
Date: September 3, 2014
HARP Significantly Reduced Mortgage
Default Rates
Author: Jun Zhu
Date: September 3, 2014
Putting Mortgage Insurers on Solid Ground
Authors: Mark Zandi, Jim Parrott, and Cristian
deRitis
Date: August 26, 2014
FHA: Time to stop overcharging borrowers for yesterday’s
mistakes
Authors: Laurie Goodman, Bing Bai and Jun Zhu
Date: January 6, 2014
The 2015 slash to VA loan limits will have little impact on
veterans even in hardest-hit DC suburbs
Authors: Laurie Goodman and Jun Zhu
Date: December 29, 2014
Should you rent or buy? What to consider in housing
decisions
Author: Ellen Seidman
Date: December 19, 2014
New measure shows mortgage denial rate is triple
traditional estimates
Authors: Wei Li and Laurie Goodman
Date: December 16, 2014
The biggest obstacle to a mortgage market recover that
nobody is talking about
Authors: Laurie Goodman and Taz George
Date: December 16, 2014
42
Copyright © February 2015. The Urban Institute. All rights reserved. Permission is granted for reproduction of this file, with attribution to the
Urban Institute.
The Urban Institute is a nonprofit, nonpartisan policy research and educational organization that examines the social, economic, and
governance problems facing the nation. The views expressed are those of the authors and should not be attributed to the Urban Institute, its
trustees, or its funders. The Urban Institute’s Housing Finance Policy Center (HFPC) was launched with generous support at the leadership
level from the Citi Foundation and John D. and Catherine T. MacArthur Foundation. Additional support was provided by The Ford Foundation
and The Open Society Foundations.
Ongoing support for HFPC is also provided by the Housing Finance Council, a group of firms and individuals supporting high-quality
independent research that informs evidence-based policy development. Funds raised through the Council provide flexible resources, allowing
HFPC to anticipate and respond to emerging policy issues with timely analysis. This funding supports HFPC’s research, outreach and
engagement, and general operating activities. Funders do not determine research findings or influence scholars’ conclusions. Scholars are
independent and empowered to share their evidence-based views and recommendations shaped by research. The Urban Institute does not
take positions on issues.
43