The Past, Present and Future of Fannie Mae, Freddie Mac, and FHA Edward Pinto Codirector AEI International Center on Housing Risk [email protected]HousingRisk.org May 6, 2014 The views expressed are those of the author alone and do not necessarily represent those of the American Enterprise Institute. 1
25
Embed
The past, present, and future of Fannie Mae, Freddie Mac, and FHA
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Mortgage Trouble Occurs When Home Values Grow Faster than Fundamentals
5
The Abandonment of Sound Mortgage Underwriting Standards
• Necessary pre-condition: GSE abandonment of sound underwriting
– Gail Cincotta’s 1991 Senate Banking testimony: "Lenders will respond to the most conservative standards unless [the GSEs] are aggressive and convincing in their efforts to expand historically narrow underwriting."
6
An Explicit Government Guarantee Is Not the Solution
Experience has shown that any federal insurance program with an explicit guarantee will fail to protect taxpayers. • Assumes the government would be better at pricing risk than
the market. – Every government insurance program has failed or is insolvent.
• Encourages too much of the wrong kind of debt for our economy. – Debt that bids up existing housing assets and the land they sit on,
creating a temporary wealth effect and a crowding out of capital investment needed for a productive and growing economy and jobs growth.
– The result will be another artificial housing boom and consequent bust.
• Inevitably leads to politicized credit standards. – Will once again put lower-income families into housing they can’t afford,
with the same disastrous results. 7
Most Likely Outcome: Status Quo
No imperative to pass legislation.
• GSE charters are perpetual.
• Loan limits set at no lower than 2007 market peak.
• Treasury obligated to provide GSEs an additional $200+ billion.
• No time limit on conservatorship.
• Exempt from QM’s 43% total DTI limit until Jan. 2021.
• GSEs give the appearance of being profitable and these “profits” go to Treasury.
• New leadership at FHFA. 8
Given These Facts What Will AEI’s New Housing Risk Center Do to Fill the Void?
The AEI/Pinto-Oliner Risk Indices:
• National Mortgage Risk Index (NMRI) – Classifies loans into low, medium, and high risk based on default
experience of 2007 vintage loans with similar characteristics. – Currently covers nearly all gov’t-guaranteed mortgages for home
purchases (about 75% of all purchase loans) – Limited to purchase loans now; refi loans to be added in 2014.
• 27-metro Composite Collateral Risk Index (CRI) – Metro areas chosen to represent housing market in U.S. as a whole. – Expected release dates: DC-area index, mid-2014; other metro areas
throughout 2014 and 2015.
• Capital Adequacy Index (CAI) – Assesses capital reserves at institutions with large mortgage
exposure. – Under development. Expected release date: 2015.
9
Measuring Stressed Default Rates
• Takeaway: Huge spread of default rates across risk buckets
• All 320 risk buckets for home purchase loans are shown at Periodic Table on HousingRisk.org
10
Risk Bucket FICO CLTV Total DTI Default Rate
Very Low ≥ 770 61-70% ≤ 33% 0.8%
Low 720-769 76-80% 34-38% 4.2%
Medium 690-719 81-85% 39-43% 9.3%
High 660-689 91-95% 44-50% 22.7%
Very High 620-639 > 95% > 50% 45.8%
Note: Default rates represent cumulative defaults through year-end 2012 for Freddie Mac’s 2007
vintage of acquired loans. The loans included in the calculation are all primary owner-occupied, 30-year
fixed-rate, fully amortizing, fully documented, home purchase loans.
* Index changes measured from first month shown. Source: AEI International Center on Housing Risk, www.HousingRisk.org. Separate index not available for VA guaranteed loans. RHS is Rural Housing Service.
Fannie/Freddie and FHA/RHS indices edged up further in March, while composite index was little changed because of a decline in the FHA loan share.
• Mortgage risk indices for most metro areas are below the national average, but are trending up
along with home price overvaluation.
• Metro area/county indices for FHA not yet available.
• However, FHA’s California volume is disproportionately in Riverside-San Bernardino and the Central Valley (defined in next slide), a signal of heightened default risk in those areas.
15
Fannie/Freddie MRI Fitch Home Price
Overvaluation
Metro Area/County Oct-Dec 2012 Jan-Mar 2014 2012 2013:Q3
Note: Fannie Mae MRIs are estimated using the Fannie Mae California state MRI and the Freddie Mac ratio of the metro area/county MRI to the
California state MRI. San Francisco includes Marin, San Francisco, and San Mateo counties; Oakland includes Alameda and Contra Costa counties.
Fitch estimates from Fitch Ratings, “U.S. RMBS Sustainable Home Price and Economic Risk Factor Report – January 2014”.
Driver of Housing Policy: Homeownership Rates
– Homeownership rate differences among minorities and whites are worthy of national discussion, as are potential solutions.
– According to the Census Bureau, the homeownership rates for Non-Hispanic Blacks (Blacks), Hispanics, and Non-Hispanic Whites (Whites) are 43.2 percent, 45.5 percent, and 73.4 percent respectively.
– Credit is a key input into the ultimate sustainability of any loan and the stability of both neighborhood and national housing markets.
• The credit profile for these groups is vastly different.
– A 2007 Federal Reserve report to Congress on credit scoring found that the median scores for Blacks, Hispanics, and Whites were 618, 670, and 737 respectively (medians interpolated from the Fed data).
– The same report found credit scores to be “predictive of credit risk for the population as a whole and for all major demographic groups” and further “credit characteristics included in credit history scoring models do not serve as substitutes, or proxies, for race, ethnicity, or sex.”
– Blacks and Hispanics are expected to account for well over half of the growth in households from 2010 to 2025.
16
While Leverage Has Increased, the Homeownership Rate Has Been Stagnant since 1960
* Interpolated between 1950 and 1960
**Estimated loan term for 1970
***Adjusted for the 8.70% and 5.64% of loans that were seriously delinquent in 2010 and 2013 respectively (assumes 70% of homeowners have mortgages). 17
15
17
19
21
23
25
27
29
31
40%
50%
60%
70%
80%
90%
100%
FHA: average LTV (1940-1967), median LTV (1980-2013 (left axis)
US homeownership rate(left axis)*,***
FHA: average term inyears (right axis)**
Risk Level and Default Rate Correlate for all Vintages
Combined LTV and FICO Highly Correlated to Default Rate for Home Purchase Loans (2007 Vintage)
• t
19
Combined LTV and FICO Highly Correlated to Default Rate for Cash Out Refinance Loans (2007 Vintage)
• t
20
Contentious Issues: Duty to Serve and Disparate Impact
• One view: “narrow” underwriting prevents low-income and minority borrowers from getting loans to the same extent as higher income/non-minority ones.
• Another view: the QM credit box is broad, deep, and pro-cyclical during a boom—all FHA (and Fannie and Freddie) are eligible – High risk lending has a disparate impact on low-income and
minority families and neighborhoods
21
A Thought Experiment
• To raise Black and Hispanic homeownership rates to the White rate of 73.4 percent would require the origination of 10 million 30 year loans totaling $1.7 trillion with an average credit score of 630, a down payment of 3.5 percent and a median debt-to-income ratio of 42%.
• These loans clearly present high risks to homebuyers, neighborhoods, lenders, mortgage insurers, and taxpayers. – 35-30% default rate under stress
– Defaults not evenly distributed—greater concentration in low-income and minority neighborhoods
22
A Second Thought Experiment
• A home loan with a 630 FICO, 4% down payment and 42% DTI. • FHA’s advantages and cross subsidies allow substantial underpricing of
high risk loans compared to the private sector. – 12% rate on a private 30 year fixed rate loan (rate based on Barclay’s
data) with a loan payment constant of 12.34—clearly a high risk loan. • Amortized LTV after year 5 – 94%.
– 4.25% rate on an FHA 30 year loan with 1.75% upfront MIP and 1.35% annual MIP with a loan payment constant of 7.25—clearly still a high risk loan. • This is due to fact that the reduced loan constant was allowed to be fully
capitalized into a 67% higher home price. • Amortized LTV after year 5 – 89%.
• An alternative: – Allow the FHA home buyer to purchase a 32% higher priced home by reducing
the term to 15 years, the rate to 3.25%, and the MIP to 0.65% (loan payment constant = 9.15). • This dramatically reduces risk to the borrower and neighborhood. • Amortized LTV after year 5 – 70%.
• Year 5: three times the wealth building effect: $10,500 ($150,000 home price) vs. $31,000 ($119,000 home price)
23
Insanity: doing the same thing over and over again and expecting different results – Albert Einstein
• 60 years of policy failure—increasing leverage will increase homeownership. – FHA’s failed effort to serve lower-income borrowers in the 1960s and early-1970s
was chronicled in Cities Destroyed for Cash.
– The affordable housing mandates contained in the 1992 GSE reform act did not end well.
– Duty to Serve and Disparate Impact in the context of the pro-cyclical QM credit box would have the same result:
• Increased leverage bids up the price of existing housing, making it less affordable, and leads to volatile home prices for those least able to cope.
• “Mortgage has become just another word for trouble…instead of a straight, broad highway to debt-free ownership” for families with the ability, desire and discipline to become homeowners. (FHA 1935)