ONE VOICE. ONE VISION. ONE RESOURCE.

GSE Reform: Consumer

Costs in a Reformed
System -
MBAs Voodoo Economics
In evaluating any proposal for GSE reform, three major objectives must be
balanced: protecting taxpayers, attracting capital to Guarantors, and ensuring
consumers and borrowers have access to affordable financing. MBAs proposal
carefully considers each of these priorities, and achieves such a balance. This
paper expands upon the discussion of consumer costs in MBAs proposal.

17870 mba.org/gsereform
 MBA understands that certain stakeholders are primarily concerned with the
impact of housing finance reform on consumer costs. This analysis reviews
the different components of reform that may impact consumer costs, and
concludes that costs under the MBA proposal are likely to be quite similar
to those seen in the mortgage market today. Whether costs are modestly
higher or lower will depend on how the different components noted here are
determined through the political process, but stakeholders concerned with
consumer costs should not forget that a more stable system, as envisioned
in MBAs proposal, has benefits for consumers, as it ensures that borrowers
can get loans even during downturns and periods of market disruption.

The difficulty in arriving at a precise estimate with

respect to consumer cost is that several components
of housing finance reform actively under debate
will impact costs. Until these debates are resolved,
it is impossible to be confident about the total
impact on cost. Most prominently, these six factors
under debate will all impact consumer costs:

A full faith and credit guarantee behind MBS,

Guarantor capital requirements, return

targets, and pricing behavior,
. Cost to buy CSP
MIF premiums, . Cost to buy GSE
systems (AU,
Affordable housing fees,
DU...)
The credit box, and . SG&A for
guarantors...
Capital requirements for MBS investors.
A Full Faith and Credit
Guarantee Behind MBS
While most of the other factors listed above may
contribute to somewhat higher consumer costs
through reform, those increases would likely be
offset by a move to an explicit government guarantee
of eligible MBS. A primary reason for the higher
price and lower rates on Ginnie Mae securities
relative to Fannie Mae and Freddie Mac MBS is the
full faith and credit guarantee on Ginnies. Fannie
Mae and Freddie Mac and their securities are now
explicitly backed by the Treasury through the
Preferred Stock Purchase Agreements (PSPAs),
but even the relatively small distinction in this
environment leads to a marked difference in price.
Move to explicit guarantee on $3-5 trillion would increase cost of
UST debt which would need to be passed on to taxpayers
GSE REFORM: CONSUMER COSTS IN A REFORMED SYSTEM
Mortgage Bankers Association, July 2017. All rights securities
Ginnie
would
need to
acquire
ability to
price
credit
risk. Also
need to
At a consumer level, this benefit from investors
acquire
valuing the explicit guarantee on Ginnies can be
systems
seen in the spread between mortgage rates on
conventional vs. FHA loans, which has ranged from 10 to take
to 40 basis points in recent years as shown in Figure 1 over
(page 4). This spread indicates the magnitude of the
lending &
potential benefit of moving to an explicit guarantee
behind the MBS. The spread has varied over time originate
due to investor perceptions regarding the relative in market
value of the securities, a value impacted by differing crises or
prepayment speeds, default rates, trading liquidity, right to
and other factors. The fact that the conventional
hold in
conforming market is larger than the Ginnie market
may indicate that a conventional MBS with an explicit illiquid
guarantee could trade even better than a Ginnie MBS, markets
leading to a further reduction in consumer costs.
Ginnie doesn't take up-front credit risk but
retains tail risk in lender failure... add R&W
losses resulting from lender failures on
3 as
reserved. in the crisis to assumptions
those are tail risk losses to taxpayer
 FIGURE 1: SPREAD BETWEEN CONVENTIONAL FIGURE 2: SPREAD BETWEEN CONVENTIONAL
CONFORMING AND FHA 30-YEAR MORTGAGE RATE CONFORMING 30-YEAR MORTGAGE RATE
AND 10-YEAR TREASURY RATE

0.4 3.2

2.8
0.3

2.4

0.2

2.0

0.1
1.6

0.0 1.2
2014 2015 2016 2017 2000 2005 2010 2015

Another Everyone agrees with this. I

math free SOURCE: MBAs Weekly Applications Survey argued this from 2000-2006.
SOURCE: Federal Reserve and MBA
assertion. HERA empowers FHFA to
You can While the precise impact on consumer costs from address the nead for
have an true housing finance reform may be difficult to gauge, Capital Requirements, Return meaningful capital at the
explicit Targets, and Pricing Behavior GSEs
we know that attempts to shortcut reform through
guarantee recap and release would lead to much higher costs for MBAs proposal recommends that Guarantor capital
behind consumers. Global investors have been clear that they requirements be much more rigorous than the
strongly have no appetite for returning to a world of implicit requirements that were applied to Fannie Mae and
capitalized guarantees and the resulting instability. Consumers Freddie Mac pre-crisis, which were proven to be
entities, it would be much worse off in such an unstable system. much too low. Our proposal advocates a level of
needed be capital that would be sufficient to survive a severe
at the The ultimate benefit of legislative reform to stress level of losses, and also be devised to be
securities consumers is that it provides an explicit guarantee consistent with required capital for others institutions
level with that the system will be more stable over time, and like banks who are major investors in mortgages.
all of the hence the mortgage market will be available to With such a capital requirement, we would expect
arbitrages, consumers, even during severe downturns a benefit consumer costs to be comparable to what we see
complexity, that would be worth the tradeoff of modestly higher in the market today, but higher than those in the
costs and costs should these be required. Figure 2 shows the woefully undercapitalized pre-crisis GSE system.
distortions spread between 30-year mortgage rates and 10-year Some proposals have argued for a much higher
of that Treasury rates. What jumps off the page is that this level of capitalization, which would directly lead to
large of an spread nearly doubled during the crisis. This is the higher consumer costs, while others have argued
expansion instability that the MBA proposal seeks to solve for for much lower standards, which would lead to a
through a combination of sufficient private capital much less stable system. The lower costs would be
and an appropriately priced government backstop. illusory, as they could not be maintained over time.
3.5-4% capital at GSEs + CRT would cover all, including 100-year flood losses. You can add a limited and
explicit, fully paid for, government guarantee behind that as a cat reinsurance policy. Such an approach puts
an explicit guarantee behind the TBA market without any real tail risk to taxpayers. The MBAs securities
guarantee approach is a subsidy that, given the size
GSE REFORM: of the
CONSUMER COSTSmortgage
IN A REFORMED market,
SYSTEM will increase Treasury debt costs.
4
Mortgage Bankers Association, July 2017. All rights reserved.
Like matter in physics, neither risk or cost can be destroyed, only moved and hidden.
Clearly MBA has not thought through their economic assumptions
or, doesn't understand markets or economics. Multiple guarantors
requires multiple duplicitous cost centers (capital, systems, G&A,
modeling, staffing, sales...). Coupled with less volume and smaller
scale at each guarantor the returns to make the economics work
would have to be higher ---> higher costs to consumers.
MIF Premiums
In addition to the level of capital, the return targets
for the Guarantors are a critical variable that impact Several proposals under consideration would move
costs. Historically, given their unassailable duopoly the system from one that included implicit guarantees
position in the guarantee business, and a regulatory that supported the GSEs to explicit guarantees
framework that did not permit oversight of returns, behind just the MBS, not the Guarantors. Under
the GSEs achieved above-market levels of returns. the MBA proposal, this explicit guarantee would
MBAs proposal will lead to a more competitive be paid for through premiums that would build up
secondary market with the potential for new entrants over time in a Mortgage Insurance Fund (MIF). As
if returns are above market over a sustained period substantial private capital would stand in front of
of time, or if the stable returns under utility-style the MIF, it would only be covering catastrophic risk.
regulation are deemed attractive by certain investors. This risk would be difficult to price precisely, as
truly catastrophic financial market events are quite
Under the utility framework, MBA envisions that rare and difficult to predict. MBAs view is that MIF
the regulator will monitor the market to ensure premiums should be set at a reasonable level, and
adequate but not excessive returns, measured on a the MIF should be allowed to build up over time to a
through-the-cycle basis. This more competitive and level consistent to cover crises similar in magnitude
more transparent approach to pricing, along with a to what we have seen historically. If these premiums
cultural shift from the GSE growth stock model to are set too low, there is an increased risk that the MIF
a Guarantor focused on steady, utility-like returns, would run dry during a crisis, and taxpayer support
should also lead to less pressure on consumer costs. would be needed. (MBAs proposal would adjust
Furthermore, by ensuring a level playing field for premiums on a go forward basis in such an event to
lenders of all sizes and business models, MBAs pay back taxpayers and replenish the MIF, similar to
proposal would support a vibrantly competitive the FDICs Deposit Insurance Fund.) If MIF premiums
and dynamic primary market which will deliver are set too high, the MIF would grow quickly, but
mortgages at competitive rates to consumers. 1- if a ithousing downturn occurred
would needlessly before thecosts.
add to consumer build-up of sufficient
capital at the MIF, taxpayers would own the risk 2- The Fed/OCC/FDIC
The GSEs "growth stock model" resulted from the
& state bank regulators would not allow MIF to increase assessments
1992 GSE reform act and the use of the portfolios
at theINtime
GSE REFORM: CONSUMER COSTS that banks
A REFORMED need to rebuild capital 3- In failure the MIF would
SYSTEM
for income generation. HERA permanently ended 5
Mortgage Bankers Association,
be in July
a2017. All rights
battle withreserved.
the FDIC as receiver leading to battle between MBS
that model.
investors and depositors.
 Affordable Housing Fees The Credit Box
Roughly one-third of existing-home sales today go Higher risk loans have higher expected losses. Subject
to first-time homebuyers, down from a historical to a QM-like standard that would set eligibility for
average closer to 40%. For first-time buyers securitization through the Guarantors, these entities
and others on the margin, higher costs or more would set underwriting criteria and make pricing
stringent standards can mean being shut out of the decisions consistent with prudent risk management.
market altogether. Efforts to extend affordability If the eligibility standard is set very conservatively, for
and access to underserved borrowers are one example, if it were consistent with the QM parameters
of the items that FHFA or its successor would absent the benefits of the patch, mortgage credit
closely monitor in the system we envision. risk would be quite low, and consumer costs may be
lower than in the current market. However, access
MBAs proposal supports charging a fee on MBS to to credit would be unacceptably low in the eyes of
raise funds to support affordable housing initiatives. many stakeholders. On the other hand, if there were
However, debate is ongoing regarding the size of no eligibility standard, and we returned to a market
this fee. Understanding that affordable housing where the GSEs set their own credit standards, with
needs in the country, for both owners and renters, the taxpayers on the hook for any misjudgments, we
are large and growing, setting this fee too high could could be back to a world of negative amortization,
be counterproductive as it would directly lead to interest only, and subprime, which could lead to
higher costs and loss of access for consumers who higher costs for consumers. Getting the balance
may just barely qualify in the conventional market. right, such that the Guarantor credit box represents
It will be important for policymakers to carefully the provision of sustainable credit to qualified
balance the costs and benefits when setting this fee. borrowers without being unduly restrictive, will
be a challenging but worthy goal of reform.
1- The largest banks have massive excess capital and are not
This already exists today AND as we saw with the political
making loans today unless they can sell them to the GSEs.
capture of OFHEO/HUD and during the crisis, politicians would
What the MBA is looking for is a capital arbitrage subsidy for the
(at the urging of the housing industrial complex) push to widen
largest banks in return for finally lending. 2- Bank prudential
the credit box and pass more risk to the taxpayer.
regulators would have incentive to push back onGSE affordable
REFORM: CONSUMER COSTS IN A REFORMED SYSTEM
6
housing goals where they viewed them as risking or requiring
Mortgage Bankers Association, July 2017.GSE
Compare All rights reserved.
underwriting standards to those of the big banks
increases in bank capital and the losses of each. GSEs priced risk quite well in comparison.
Given that, with Ginnie loans, the lender retains the credit risk
(unlike the GSEs who retain the credit risk) and it is clear that
this is a public policy failure and disproportionately transfers risk
to the public and financial benefits to Ginnie lenders. Bad policy
is not its own justification.
Capital Requirements for MBS Investors
FIGURE 3: COMPARISON OF FHA AND GSE PRICING BY
One of the factors leading to higher values on CREDIT SCORE FOR 96.5/97 LTV LOANS (AS OF JULY 2017)
Ginnie Mae securities, and hence lower rates on
FHA and VA loans included in those securities,
is their bank regulatory capital treatment. This is
7.0
connected to, but distinct from, the benefit of the

( B A S E R AT E + L L PA / U P F R O N T M I P + A N N U A L M I P )
full faith and credit backing on Ginnie Mae securities.
For risk-based capital purposes, Ginnie Mae MBS
6.5
receive a 0% weight, while Fannie and Freddie

A L L I N R AT E T O T H E B O R R O W E R
MBS receive a 20% weight, even with the Treasury
backstop behind the companies. Moreover, Ginnie
Mae MBS also are considered high quality liquid 6.0
assets for the liquidity coverage ratio test, while
Fannie and Freddie MBS are haircut with respect
to their ability to meet this liquidity standard. 5.5
This would give the largest banks close to $200
If reform follows MBAs proposal and provides an billion in regulatory capital relief. Coupled with their
explicit guarantee behind MBS issued by Guarantors, 5.0
multiple guarantor model - in which large banks
and bank regulators change risk weights and liquidity would be allowed to join together and own
treatment to match those for Ginnie Mae MBS guarantors, this would lead to greater concentration
today, that will be another factor leading to lower of risks in 4.5
our largest banks and further squeeze
Fannie / Freddie Freddie Home
consumer costs in the new system. In fact, given the smaller banks and independent lenders - turningPossible Advantage
larger size of the conventional market relative to the them into third party originatorsFannie
for HomeReady
the largest banks
FHA

Ginnie market, there may be even greater liquidity 4.0

and their guarantors.
740 720 700 680 660 640 620
in the conventional market. Banks benefit by having
CREDIT SCORE
a more liquid mortgage investment, originators
benefit by having better pricing for their loans/
securities, and consumers benefit from lower rates.

SOURCE: MBA analysis of Fannie Mae,

Other Considerations Freddie Mac, and FHA pricing

Beyond the impact for average pricing, changes in

any of these factors could lead to differential changes
in costs for stronger vs. weaker credit borrowers.
However, when examining the potential impact of
future pricing on different subsets of borrowers, it
is critical to remember that today the GSEs are not
reaching many first-time homebuyers given their
current pricing structure. For many borrowers with
less than perfect credit, FHA offers a competitive
or lower all-in cost as shown in Figure 3. This is true
even for the GSEs affordable housing initiatives.
Conclusion
The ultimate impact on consumer costs from
housing finance reform will be a function of a large
number of factors that are to be decided through
the legislative debate. MBA expects that the lower
costs resulting from the explicit guarantee behind the
MBS will largely if not completely offset the higher
costs that are a function of more rigorous capital
requirements, the MIF premium, and affordable
housing fees. Other factors, including the business
model of the Guarantors, which will impact required
returns, the credit standards for the new system, and
potential changes to bank regulatory treatment of
conventional MBS, will also have an impact. Balancing
taxpayer protection, investor returns, and consumer
costs is critical to realizing a more stable housing
finance system going forward. MBA looks forward to
continuing to contribute to all aspects of this debate.

GSE REFORM: CONSUMER COSTS IN A REFORMED SYSTEM

7
Mortgage Bankers Association, July 2017. All rights reserved.