[Senate Hearing 113-190]
[From the U.S. Government Publishing Office]
S. Hrg. 113-190
HOUSING FINANCE REFORM: DEVELOPING A PLAN FOR A SMOOTH TRANSITION
=======================================================================
HEARING
before the
COMMITTEE ON
BANKING,HOUSING,AND URBAN AFFAIRS
UNITED STATES SENATE
ONE HUNDRED THIRTEENTH CONGRESS
FIRST SESSION
ON
EXAMINING VARIOUS OPTIONS FOR A TRANSITION TO A NEW HOUSING FINANCE
SYSTEM THAT ACHIEVES THE ACCESSIBILITY, AFFORDABILITY, AND STABILITY
GOALS OF HOUSING FINANCE REFORM WHILE MINIMIZING DISRUPTIONS TO BOTH
THE PRIMARY AND SECONDARY MORTGAGE MARKETS
__________
NOVEMBER 22, 2013
__________
Printed for the use of the Committee on Banking, Housing, and Urban
Affairs
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Available at: http: //www.fdsys.gov /
________
U.S. GOVERNMENT PRINTING OFFICE
86-866 PDF WASHINGTON : 2014
-----------------------------------------------------------------------
For sale by the Superintendent of Documents, U.S. Government Printing
Office Internet: bookstore.gpo.gov Phone: toll free (866) 512-1800
DC area (202) 512-1800 Fax: (202) 512-2104 Mail: Stop IDCC,
Washington, DC 20402-0001
COMMITTEE ON BANKING, HOUSING, AND URBAN AFFAIRS
TIM JOHNSON, South Dakota, Chairman
JACK REED, Rhode Island MIKE CRAPO, Idaho
CHARLES E. SCHUMER, New York RICHARD C. SHELBY, Alabama
ROBERT MENENDEZ, New Jersey BOB CORKER, Tennessee
SHERROD BROWN, Ohio DAVID VITTER, Louisiana
JON TESTER, Montana MIKE JOHANNS, Nebraska
MARK R. WARNER, Virginia PATRICK J. TOOMEY, Pennsylvania
JEFF MERKLEY, Oregon MARK KIRK, Illinois
KAY HAGAN, North Carolina JERRY MORAN, Kansas
JOE MANCHIN III, West Virginia TOM COBURN, Oklahoma
ELIZABETH WARREN, Massachusetts DEAN HELLER, Nevada
HEIDI HEITKAMP, North Dakota
Charles Yi, Staff Director
Gregg Richard, Republican Staff Director
Laura Swanson, Deputy Staff Director
Glen Sears, Deputy Policy Director
Erin Barry Fuher, Professional Staff Member
Michelle Maiwurm, Legislative Assistant
Greg Dean, Republican Chief Counsel
Chad Davis, Republican Professional Staff Member
Mike Lee, Republican Professional Staff Member
Michael Bright, Republican Senior Financial Advisor
Dawn Ratliff, Chief Clerk
Taylor Reed, Hearing Clerk
Kelly Wismer, Hearing Clerk
Shelvin Simmons, IT Director
Jim Crowell, Editor
(ii)
C O N T E N T S
----------
FRIDAY, NOVEMBER 22, 2013
Page
Opening statement of Chairman Johnson............................ 1
Opening statements, comments, or prepared statements of:
Senator Crapo................................................ 2
WITNESSES
James Millstein, Chairman and Chief Executive Officer, Millstein
and Company.................................................... 4
Prepared statement........................................... 24
Responses to written questions of:
Chairman Johnson......................................... 70
John Bovenzi, Partner, Oliver Wyman.............................. 5
Prepared statement........................................... 36
Responses to written questions of:
Chairman Johnson......................................... 73
Mark Zandi, Chief Economist and Cofounder, Moody's Economy.com... 7
Prepared statement........................................... 41
Responses to written questions of:
Chairman Johnson......................................... 74
David Min, Assistant Professor of Law, University of California,
Irvine School of Law........................................... 8
Prepared statement........................................... 59
Responses to written questions of:
Chairman Johnson......................................... 77
(iii)
HOUSING FINANCE REFORM: DEVELOPING A PLAN FOR A SMOOTH TRANSITION
----------
FRIDAY, NOVEMBER 22, 2013
U.S. Senate,
Committee on Banking, Housing, and Urban Affairs,
Washington, DC.
The Committee met at 10:02 a.m., in room SD-538, Dirksen
Senate Office Building, Hon. Tim Johnson, Chairman of the
Committee, presiding.
OPENING STATEMENT OF CHAIRMAN TIM JOHNSON
Chairman Johnson. The Committee will come to order. Today
we continue our series of hearings to better inform the
Committee's efforts to reform our housing finance system. While
we have spent time examining what the different components of a
new system should look like, today we focus on how to develop a
credible plan that will seamlessly transition us from the
current system to the new one.
This is a critical issue, especially as we consider the
significance of the housing market to the overall economy. If
the transition is not properly managed, or does not have enough
flexibility built in, then we are asking for trouble and we
could end up with potential market disruptions, which would
impede economic growth. That is the last thing we want.
There are a number of transition issues worth discussing,
starting with how best to wind down Fannie Mae and Freddie Mac
while we build a stronger system. A key goal is ensuring
taxpayers are fully repaid as we also consider what well-
functioning systems and assets could be utilized in a new
system. We should seek to find options that make the best use
of existing resources to avoid making the transition needlessly
inefficient, costly, and complex.
We also need to examine the timing and sequence of the
transition plan. There should be transparency with respect to
the details of the transition to provide market participants
the certainty they need to make long-term plans and decide what
roles they may want to play.
While there should be clear goals to make sure we
eventually reach the desired outcome, we should also consider
allowing for some overlap so that the new system is well-tested
and fully functioning before we turn off the lights on the old
system. There is simply no need to roll the dice when we are
talking about nearly 20 percent of our economy.
A few other questions to contemplate: What is the best way
to verify that risk-sharing structures work and sufficient
private capital is able to provide mortgage credit through good
times and bad? What kinds of emergency powers are needed to
deal with unforeseen events during the transition and minimize
market disruptions? If a common securitization platform will
help deliver a single security, should the platform be fully
operational before shutting down the old system? And what steps
are needed to make sure legacy MBS issued by Fannie and Freddie
do not become orphaned?
Given our witnesses' expertise in housing finance, as well
as some of our witnesses' experiences managing other
transitions, I look forward to hearing their views on these
important questions and other issues. With that, I turned to
Ranking Member Crapo for his opening statement.
STATEMENT OF SENATOR MIKE CRAPO
Senator Crapo. Thank you, Mr. Chairman. The issue of how we
transition into a future system is one of the most important
topics we will cover in this series of housing finance reform
hearings. Numerous challenges will face our regulators and
market participants as we move toward a new housing finance
system.
These challenges are real and must be addressed. But they
must also be weighed in the context of the consequences of the
status quo. The status quo means Government control of nearly
all of our Nation's mortgage-backed securities market, stifling
financial innovation. The status quo means little, if any,
reduction in the $5 trillion of outstanding mortgage debt to
which the taxpayer is currently exposed.
The status quo means continued legal and market
uncertainty, creating costs that we may be realizing for
decades to come. These realities prove that inaction carries
its own dangers in our housing market, as does action that
merely gives the appearance of change without recognizing the
mistakes of the past.
With that in mind, I look forward to hearing from our
witnesses regarding issues we need to consider, and
recommendations they have to most effectively transition to a
reformed system. During yesterday's hearing, we heard
recommendations on how to best equip the future regulator to
meet the transitional issues it might face in preparing to
assume its responsibilities.
I expect that today's hearing will give us the opportunity
to pair the lessons learned yesterday with today's relevant
information, particularly as it pertains to how to handle the
assets of Fannie Mae and Freddie Mac. As we proceed with this
discussion, there are many questions that must be considered,
such as, while Fannie and Freddie are unique institutions, what
do past experiences suggest are processes to maximize the
return on our investment for the taxpayer, while also
minimizing disruption in our markets?
How do we allow adequate time to achieve and measure the
appropriate benchmarks of the transition while still ensuring
accountability in the process? What is the best way of dealing
with the legacy obligations of Fannie and Freddie to ensure
adequate protection for both taxpayers and investors?
While these questions do not represent an exhaustive list,
they do provide an insight into the complexity that we must
address. S.1217 seeks to address the questions we will face
through a 5-year phase-in for the future regulator, FMIC. This
phase-in would coincide with the beginning of a wind down of
Fannie and Freddie supervised by the Federal Housing Finance
Administration, or FHFA.
Once the 5-year phase-in is complete and FMIC becomes
operational, Fannie and Freddie's charters would be revoked and
they would cease writing new business. However, the wind down
of the companies would continue, and during that time, the
obligations of Fannie and Freddie would be explicitly
guaranteed by the Federal Government, offset by a continued
revenue stream from the companies.
Additionally, so long as FMIC did not believe any actions
interfered with its duties, FHFA would be free to sell or
transfer the assets and business lines of Fannie and Freddie to
aid in their resolution. One suggested use for some
infrastructure of these companies is to aid in the
establishment of a small bank mutual company prescribed in
S.1217. Another destination for this infrastructure is likely
within the common securitization platform envisioned in the
bill.
There has been much discussion of these concepts and other
proposals for how Fannie and Freddie's assets might best be
utilized in the future housing finance system. This discussion
is productive and we must consider all perspectives so that we
reach an optimal solution for consumers and taxpayers.
However, the consequences of inaction far outweigh the
value of comprehensive reform. The status quo is not an option.
We now have momentum of both houses of Congress and the White
House working toward achieving real reforms. We must build on
this momentum.
I welcome the views of all who seek to aid in this effort
and look forward to the testimony we will hear today. Thank
you, Mr. Chairman.
Chairman Johnson. Thank you, Senator Crapo. Are there any
other Members who would like to give brief opening statements?
I would like to remind my colleagues that the record will be
open for the next 7 days for additional statements and other
materials.
I would like to introduce our witnesses that are here with
us today. First, Mr. James Millstein is the Chairman and CEO of
Millstein and Company. Previously Mr. Millstein served as the
Chief Restructuring Officer at the Treasury Department.
Next will be Mr. John Bovenzi is a Partner of Oliver Wyman.
Mr. Bovenzi served over 20 years in high-level positions at the
FDIC and played a key role with the Resolution Trust
Corporation. Next is Dr. Mark Zandi, Chief Economist at Moody's
Analytics where he directs economic research encompassing
macroeconomics, financial markets, and public policy.
And Professor David Min, Assistant Professor of Law at the
University of California, Irvine School of Law. Previously
Professor Min served as a staff attorney at the SEC and as
counsel to Senator Schumer, and he helped the lead for American
Progress Working Group to develop a housing finance reform
proposal.
We welcome all of you here today and thank you for your
time. Mr. Millstein, you may begin your testimony.
STATEMENT OF JAMES MILLSTEIN, CHAIRMAN AND CHIEF EXECUTIVE
OFFICER, MILLSTEIN AND COMPANY
Mr. Millstein. Chairman Johnson, Ranking Member Crapo,
Senators Warner and Corker, thanks for inviting me here to
testify on the transition.
I have spent the entirety of my 30-year professional career
as a lawyer, investment banker, and as a public servant in
doing corporate restructuring work. During the recent financial
crisis, as the Chairman mentioned, I was the Chief
Restructuring Officer at the Treasury Department where my
primary responsibility was managing the restructuring and the
exit from our substantial investment in AIG.
I also am an adjunct professor of law at Georgetown where I
teach a course on the Federal regulation of financial
institutions. I am here today because embedded in the task of
reforming our Nation's housing finance system is a
restructuring of the two largest players in that system today,
Freddie Mac and Fannie Mae.
Because they are central to mortgage credit formation in
the United States, winding them down, as some Members of
Congress and the Administration have advocated, will have
significant and, I believe, adverse consequences for mortgage
credit availability and for the nascent housing and economic
recovery.
Rather than winding them down, I urge you to consider a
restructuring alternative that addresses the fundamental causes
of the companies' insolvency, eliminates the private gain/
public loss nature of their current Government charters,
generates a significant profit to Treasury for supporting their
solvency during the recent crisis, and most importantly,
ensures a smooth transition to a new housing finance system
that better protects taxpayers against future losses while
providing for the continuing availability of mortgage credit to
the creditworthy.
There appears to be a growing consensus in the policy
community around the basic architecture of that new housing
finance system. A Federal guarantee on qualified mortgage
product is required to ensure the widespread availability of a
30-year fixed-rate product, and to sustain the deep and liquid
mortgage securities funding markets that complement balancing
lending for the banking industry.
The Government guarantee should be explicit, structured as
reinsurance, priced at arm's length by an independent agency
and available to reimburse investor losses only after a thick
layer of private first-loss capital provided by well-
capitalized mortgage insurers or subordinated capital provided
by investors has been exhausted.
The Government reinsurer also needs to be a strong
regulator with authority over all issuers, guarantors and
servicers with whom it interacts in the new system. And I
commend Senators Corker and Warner for having put together a
coalition around a bill that has all of these elements in it.
However, the transition contemplated by the bill out of the
conservatorships to this new system is fraught with danger and
needs serious rethinking to mitigate three significant risks
that any credible transition plan must address.
First, the fragile economic recovery cannot afford the risk
of a significant disruption in mortgage credit availability.
Second, the Government must end its ongoing backstop of Fannie
Mae and Freddie Mac solvency and conservatorship in a way that
minimizes the likelihood that Treasury will need to cover
future losses on the $5.5 trillion of liabilities the Treasury
now backstops.
While the substantial fees and net interest margin which
the companies are currently earning and paying over to Treasury
may look like an asset to be seized by taxpayers as the quid
pro quo for the bailout, it could easily turn out to be a
substantial liability if there were another significant housing
downturn.
Third, there must be a credible path toward the development
of the substantial layer of private first loss capital on which
the functioning of the new system will depend. If you build the
new Government reinsurer but the required layer of first-loss
capital does not come in size or at a pace of your contemplated
wind down of Fannie and Freddie, the whole system will shut
down before it has a chance to start.
In my view, one fundamental choice will determine whether
you successfully mitigate these risks in the transition. Do you
recapitalize and privatize the guarantee businesses within
Fannie Mae and Freddie Mac using their assets and operations to
create a set of well-capitalized private market players who can
ensure that sufficient first-loss capital is in place to allow
the new system to function as contemplated?
Or do you make the bet that if you build the new
reinsurance system, players yet to be named with capital yet to
be raised will generate the sizable amount of first-loss
capital required to make the new system function on the 5-year
timetable that your bill would wind Fannie and Freddie down.
As currently drafted, 1217 chooses the latter course, and
for that reason, I believe it puts the entire reform project at
risk of failure. I appreciate the desire to avoid recreating
the dangerous duopoly that contributed to the financial crisis
and, unfortunately, I am running out of time, but I have laid
out in my written testimony five or six steps that are all
interrelated to avoid that outcome in connection with the
privatization of the mortgage guarantee businesses, and I am
happy to answer questions in the body of the hearing. Thank you
very much.
Chairman Johnson. Thank you. Mr. Bovenzi, you may proceed.
STATEMENT OF JOHN BOVENZI, PARTNER, OLIVER WYMAN
Mr. Bovenzi. Good morning, Chairman Johnson, Ranking Member
Crapo, and Members of the Committee. My name is John Bovenzi, a
partner at Oliver Wyman, and I want to thank you for affording
me the opportunity to be here to speak about housing finance
reform.
My perspective draws on 28 years of experience at the FDIC
and the Resolution Trust Corporation, and I would like to start
with the RTC's experience in creating a new Federal agency.
First, the RTC showed that it is possible to start a new
Federal agency and be successful in a relatively short period
of time.
However, stakeholders must have patience. The complexities
of the issues that must be addressed virtually ensure that
there will be bumps, missteps, and delays along the way. New
agencies need governance structures, information systems,
staff, policies, procedures. The FDIC was able to provide the
RTC with a lot of support, but it was not nearly enough.
There was a great deal of frustration with the RTC early
on, but some perspective is necessary. The RTC managed its way
through those problems, saved taxpayers money, and finished its
work early.
Second, the leadership is critically important. The person
in charge needs to possess the skills to work effectively with
a large number of stakeholders and the managerial skills to
address the many operational issues faced by a new agency.
Sometimes there is a tendency to focus on high-level policy
issues and not give sufficient attention to the critical
operational details.
Third, the employees of the two Government-sponsored
enterprises in FHFA are the people who have the experience and
the expertise to effectively transfer critical functions to a
new agency. Their importance should not be undervalued or lost
if Congress decides to move in this direction, because they are
the ones who will ultimately determine success or failure.
Regarding the lessons that may be learned from the FDIC's
experience, I would like to comment on three areas related to
governance, financial strength, and supervision. First,
independence in the system of checks and balances are two
important features of the FDIC's governance structure. The
FDIC's five-person Board of Directors and a strong Office of
the Inspector General have served the agency well as part of an
overall system of checks and balances, and those features have
been included in the proposed structure for FMIC.
Second, much has been learned about what is required to
maintain a strong deposit insurance fund. The FDIC's fund
became insolvent during the two most recent financial crises.
The agency had to substantially raise bank insurance premiums
during these crisis periods when banks could least afford to
pay them, and this, among other things, exacerbated the credit
crunches that existed.
As a result, Congress granted the FDIC much greater
premium-setting flexibility and the agency now has the
authority to set the size of the deposit insurance fund high
enough to withstand similar crisis periods. And FMIC would need
sufficient flexibility and authority to manage the size of its
fund so it, too, can protect taxpayers during economic
downturns.
Third, the FDIC has a wide range of supervisory and
enforcement powers. The proposed bill would grant some, but not
all of these authorities to the FMIC. It is worth considering
whether FMIC should be granted broader supervisory and
enforcement authority.
Finally, I would like to make a few comments about the
sales processes. The FDIC and the RTC experimented with a large
variety of sales processes and learned much through trial and
error. Two key principles emerged. First, virtually all sales
were subject to inclusive, open, and transparent competitive
bidding processes.
The RTC and the FDIC did not engage in negotiated sales
with individual buyers, recognizing that open competition would
maximize value and reduce the possibilities for fraud or abuse.
Second, the RTC and the FDIC partnered with the private
sector in the disposition of many of their assets. Through
loss-sharing transactions with healthy banks and equity
partnerships with private sector investors, the agencies found
they could benefit from the added value the right management
could bring to those assets.
Thank you and I look forward to your questions.
Chairman Johnson. Thank you. Dr. Zandi, you may proceed.
STATEMENT OF MARK ZANDI, CHIEF ECONOMIST AND COFOUNDER, MOODY'S
ECONOMY.COM
Mr. Zandi. Thank you, Mr. Chairman, Ranking Member Crapo,
Senators Corker and Warner, thank you for the opportunity to be
here today. I am an employee of the Moody's Corporation, but my
views are my own and not those of the Moody's Corporation. You
should also know that I am on the Board of MGIC, the Nation's
largest private mortgage insurance company. I am also on the
Board of TRF, one of the Nation's largest CDFIs, so it is
important for you to recognize the context of today's
discussions.
Most fundamentally, a successful transition from the
current housing finance system to the future system means that
activities in the mortgage market cannot be disrupted. Mortgage
credit must flow smoothly. This is, obviously, very key to the
housing market, and by extension, the broader economic
recovery. It is particularly important over the next several
years as the economy continues to try to recover.
And because of the size of the U.S. mortgage market, it is
key to the global financial system, and so it is very important
that this all works out well; otherwise, we will disrupt a very
important part of that global financial system. So this is a
very important task.
Now, of course, any discussion of the transition process
presupposes an end state for the future housing finance system,
and I will say S.1217, the Corker-Warner legislation, has, in
my view, a very good vision of where the system should go, a
hybrid system with an explicit Government backstop to the
system.
I think under this vision that is in the legislation, the
30-year fixed-rate, prepayable mortgage will remain a mainstay
of the housing market and I think that is very important. It
also ensures affordable mortgage loans to most middle income
Americans through most economic times, and that is also very
key.
I would like to suggest a few adjustments to S.1217 to help
facilitate the transition process because it is so key, and I
should say, there are a lot of moving parts here and a lot of
things to consider. I am just going to consider a few things
for you to contemplate.
The first is, I think there needs to be some flexibility
with regard to the 5-year deadline. I think a hard and fast
deadline in the context of the complexity of this process and
the uncertainty involved creates some potential for disruption.
Moreover, you need to consider the economic environment and
the environment in the financial system as this process
unfolds. It may very well be the case that the day you want to
switch on the lights to the new system the economic and
financial market environment would not be conducive to that. So
I think there needs to be some flexibility around this process,
and I think the transition time line should be based on
benchmarks, hitting certain targets, and I will mention two
things that are very key.
One is a well-functioning operational common securitization
platform. This is vital to any future housing finance system.
If that is not up and running and working well, then the system
will not operate well. So that is very important.
And I also would argue there needs to be a common security.
This is becoming an increasing problem in the current--the way
the system is currently designed, and this would also address
some issues with regard to legacy MBS issues. So I think this
is very, very important to consider in terms of transition.
The other adjustment I would make is that I think the move
from the current capitalization of the system to the
capitalization of the system envisaged in the future needs to
be phased in over time and there needs to be some flexibility
around that as well. So just based on my calculation, the
current system is effectively capitalized to about a 2.5-
percent loss rate. Under 1217, we envisaging a 10-percent
capitalization.
Moving from 2.5 to 10, even under the best assumptions, and
we have done a fair amount of calculations here, would add 40
basis points to mortgage rates, which in the grand scheme of
things may not be a whole lot, but it will also be in the
context of rising mortgage rates because of what the Federal
Reserve Board is doing. So I think--and that is the minimum and
assuming everything goes well in the transition process.
So I do think that there needs to be some flexibility with
regard to how those capital levels are phased in. Also in that
context, we are only talking about one part of the housing
finance system. There are other parts that need to be
calibrated to. You know, what is going on with regard to bank
capital requirements, what is going on in the private level
securities market, what are decisions being made by the FHA. So
phasing in capital requirements are very important.
There are many other small adjustments I would make, but
those are the two key things that I would focus on. I think
they are very doable and at the end of the day, I think we will
end up with a system that is meaningfully better than the one
that we--certainly much better than the one we have now.
Thank you again for the opportunity to speak to you today.
Chairman Johnson. Thank you. Mr. Min, please proceed.
STATEMENT OF DAVID MIN, ASSISTANT PROFESSOR OF LAW, UNIVERSITY
OF CALIFORNIA, IRVINE SCHOOL OF LAW
Mr. Min. Chairman Johnson, Ranking Member Crapo, Senators
Corker and Warner, thank you for the opportunity to testify
today on this incredibly important topic. For the purposes of
my testimony, I am going to assume that the system end state
that we transition into will be some variation of S.1217, the
bill proposed by Senators Corker and Warner.
I think that the Corker-Warner bill envisions a so-called
hybrid system in which the Federal Government provides explicit
and price reinsurance on mortgage-backed securities created by
approved bond issuers in a model based loosely on the Deposit
Insurance model, the FDIC. I believe that Corker-Warner
provides a good framework for long-term housing finance reform,
as I discuss in my written testimony, but the observations I
make today hopefully hold true regardless of whether we go
Corker-Warner or in some other direction.
As a threshold matter, I just want to note again, echoing
some of the other panelists, how massive and complex this
proposed transition will be. Fannie and Freddie currently
account for more than $5 trillion in mortgages, and since the
crisis, these enterprises have been responsible for more than
60 percent of new mortgage originations, about $1.7 trillion
each year, an amount that is slightly more than 10 percent of
GDP. These are the most systemically important companies we
have ever attempted to resolve.
Now, the Federal Government has some experience in winding
down large and systemically important institutions. AIG and GM
are recent examples that come to mind. But here we are talking
about winding down two massive and economically vital
institutions, while simultaneously transitioning their core
functions into a newly created set of institutions. I believe
there is no precedent for this. And if we screw this transition
up, we are talking about catastrophic damage to a very fragile
housing sector and the broader economy.
So as a threshold matter, I believe the guiding principle
for policy makers thinking about transition must be the same
that governs doctors, do no harm. Avoiding the disruption in
mortgage liquidity must be a paramount concern in structuring
the transition from the current system to the future one.
We must also take steps to ensure that sufficient finances
be available for affordable rental housing. In the aftermath of
the financial crisis and housing crisis, policy makers have
generally sought to deemphasize home ownership and shrink the
Federal Government's footprint in housing. If we are successful
in these objectives but unsuccessful in redirecting finance to
rental housing, millions of working class households will pay
the price.
With that in mind, I want to talk about the Corker-Warner
transition plan which lays out a few detailed steps. The
Corker-Warner bill contemplates a transition period of no more
than 5 years following its enactment, during which time Fannie
and Freddie would be phased out and the infrastructure for the
new system, including the FMIC at the heart of the framework,
would be established.
Upon enactment, Corker-Warner would eliminate the
affordable housing goals currently in place and begin to
gradually reduce the loan limits over time, along with their
GSE portfolios. Once the FMIC is certified as operational, an
event that must occur again within 5 years of enactment, Fannie
and Freddie are to be dechartered on their outstanding legacy
obligations explicitly guaranteed with full fit and credit of
the United States.
At a high level, I think this plan provides a thoughtful
template for transition, but leaves certain issues unresolved,
several of which I identify in my written testimony. Like Dr.
Zandi, I have a few recommendations I think that could help
improve this transition.
First, I think we should delegate more responsibility to
regulators and remove some of the predesignated timetables in
place. In a number of ways, I think this transition plan in
place seeks to micromanage the transition process. Specific
timeframes are put in place for winding down the GSEs, lowering
loan limits, and getting the FMIC up and running.
What we have seen, of course, is that, you know, the best
laid plans do not always work the way we planned. Capital
requirements are highly detailed in this plan, but I think that
all of these things are highly complex and technical issues
that are best resolved by regulators looking closely at
accumulated data, rather than by legislation based on
assumptions that may or may not turn out to be true.
Second, I think transition should be phased in over time,
echoing Dr. Zandi, rather than through the on/off model
currently contemplated. As recent events may highlight,
unanticipated problems can and do arise, particularly with any
transition as complex as moving trillions of dollars in
mortgage origination finance from one platform to another.
We may end up flipping the switch on the new system, only
to discover that the lights do not turn on. I believe a
preferable approach is to adopt a phased-in approach or
piecemeal approach to transition, turning over small, but
increasingly larger parts of the mortgage markets to the new
infrastructure.
For example, rather than preparing the CSP, Common
Securitization Platform, to handle all the mortgage financed by
GSE securitization, we could start with a dedicated subset,
like 15-year fixed-rate mortgages or high-cost conforming
mortgages. Such an approach, I believe, would allow regulators
to test the new system in a meaningful way and develop that
data that can help them perfect the new architecture. I think
it would also help build investor liquidity in the MBS which is
critically important.
Third, convert legacy securities into the new MBS. Make
this an option. You obviously cannot force investors to do
this. Fourth, preapprove the new MBS for use in markets like
TBA, for use as collateral in markets like the Fed discount
window lending, repo, and derivatives markets.
And finally, I think that the regulators need to be given
more tools to prepare for another countercyclical downturn like
we have today. Obviously, the role that Fannie and Freddie are
playing today is critically important. I think regulators need
more power to be able to deal with situations like these. I am
out of time, so I apologize.
Chairman Johnson. Thank you all for your testimony. As we
begin questions, I will ask the clerk to put 5 minutes on the
clock for each Member. This is a question for Mr. Millstein,
Dr. Zandi, and Professor Min. Mr. Millstein, what are the most
important elements of wind down of Fannie Mae and Freddie Mac?
And should the charters be repealed prior to the new system
being fully tested and operational? Should it be a target-based
or time line-based approach?
Mr. Millstein. I think it should be a target-based, not a
time line-based approach, and as I said in my opening remarks,
I think the wind down--I think you have two problems, and the
biggest problem here is, you have got a new system you are
building that requires a lot of first-loss capital. You cannot
issue a new guarantee unless you have your first-loss capital
in place. So there will be no guarantees in the absence of that
capital being there.
Today, six out of ten mortgages in America are being
provided by these two entities you are proposing to wind down.
Getting the timing of that wind down right with the wind up of
your new system, to me, is a very, very--you have made this
more complex than it needs to be.
So what I have urged and what the plan I proposed suggests
is, take the guts of the mortgage guarantee businesses of the
two entities, capitalize them so they can play the role of
first-loss provider in your system, and when they are properly
capitalized--because today they have no capital other than the
backstop from Treasury.
When they are properly capitalized, turn your system on,
because now you have got your two--at least two, maybe four if
you do it with the multifamily businesses as separated from the
single family--you have four separate entities who can play the
role of first-loss provider.
In that context, tear up their charters. They are no longer
Government-sponsored enterprises. They are just private
mortgage insurers competing with MGIC, Dr. Zandi's company, and
every other private mortgage insurer for the business of
acquiring and pooling mortgage loans into conforming MBS, and
buying reinsurance from the Government.
Chairman Johnson. Dr. Zandi, do you agree?
Mr. Zandi. I agree it should be a target-based time line or
transition, not a time line-based transition. I mentioned a
couple of benchmarks and targets that I would focus on, the
common securitization platform being very important. That is
key to allowing for bringing down the barriers of entry to
allow more private capital to come in. So that needs to be
operational, working well, and functioning properly before we
can transition to the new system.
I mentioned the common security. I think that is also going
to be quite important. That helps address also bringing down
barriers to entry. It allows smaller lenders to participate in
the system and that is going to be very key to this all working
out well. And there are others, other key benchmarks. So I
think it needs to be target-based.
I do think it would be, to answer your question explicitly
about the charter, I think the approach I would take is to put
the two institutions into receivership, take their assets, put
them into a limited life regulated entity. This is a structure
that was designed as part of the HERA legislation for this
purpose. It allows the institutions to operate normally.
Liquidity would continue to flow to the mortgage market. The
market would not be disrupted, but the assets could then be
deployed in a way to allow for a greater competition in the
mortgage guarantee market.
The concern I have is that if we just take Fannie Mae and
Freddie Mac, restructure them and throw them out into the
world, that the world will be dominated by the entities that
follow Fannie Mae and Freddie Mac. Nothing will change. It will
be a duopoly or potentially even a monopoly and I do not think
that is appropriate.
So I think we should take those assets and we should allow
1,000 flowers to bloom, allow other guarantors to come in, more
private capital to come in, and they can license those assets
in their own functions and it would create greater competition
and ultimately a more resilient, better, stronger market at a
lower cost to taxpayers and to mortgage borrowers.
Chairman Johnson. Mr. Min.
Mr. Min. To answer your questions, I think it should be
target-based. I am a little reluctant to say their charters
should be repealed until we know there is enough liquidity in
the rest of the system. We are in the aftermath, of course, of
a 100-year flood. We are in a credit downturn.
It is not clear to me where the sources of private mortgage
funding are going to come from right now. As I detail in my
written testimony, bank deposits do not appear to be a likely
source. Private label securitization, which was responsible for
so much mortgage financing in the mid-2000s, has scaled back
completely.
So I do not know where this is going to come from. I do not
know how long it will take. I think we need to achieve those
liquidity benchmarks. Just to give one anecdote on sort of the
target-based versus time line, you know, the securitization
platform, the common CSP, which Dr. Zandi described, is at the
heart of this new plan.
It is necessary either for Corker-Warner or for the PATH
Act; and yet, if you are following what is going on, it has
been a year and a half after the FHFA first announced its plans
to develop the CSP. We have had literally no movement. The last
press release they issue announced that they had leased some
office space. It does not appear that they have had a lot
development toward that goal.
If that is the sort of timeframe we are looking at, we have
to be cautious about proceeding ahead based on assumptions that
things will happen that may take more time.
Chairman Johnson. Mr. Bovenzi, based on your RTC
experience, to what extent should flexibility be built into the
transition process to allow a regulator to address potential
problems and to avoid market disruptions?
Mr. Bovenzi. Well, like the other speakers, I believe you
need to build in a great deal of flexibility. Congress can set
the basic parameters, but you cannot anticipate everything, and
even if Congress could, facts change, markets do not stay the
same and there will be new issues to address that could not
really have been foreseen.
So I think the RTC experience shows that need for
flexibility. They tried out some different things that did not
work before finding out what did work for them. And it was not
a smooth path for awhile, but it got to the right place. That
flexibility was critical in getting there.
Chairman Johnson. Senator Crapo.
Senator Crapo. Thank you, Mr. Chairman. Dr. Zandi and Mr.
Millstein, the two of you have differing perspectives on how to
best capitalize and establish a reformed market. But it seems
to me, as I read your testimony, that there is an agreement
between you that any current revenues of Fannie and Freddie
should only be used to, either, one, repay the taxpayers or
make the taxpayers whole, or two, offset the cost of
backstopping and capitalizing a new system.
If I am correct about that, we have already seen this money
used to offset a temporary tax cut and there are now other
proposals out there advocating its use for other initiatives. I
would like both of you to please elaborate on your views. Do I
understand your testimony correct? Do you believe we should
limit the utilization of these assets to either making the
taxpayers whole and utilizing the necessary assets to
capitalize a new system?
Mr. Zandi. You have characterized my view accurately. I do
not believe that it is appropriate to use these resources for
other purposes, and I think they should be used to ultimately
repay taxpayers. That is an open question, exactly what that
means and how much that will cost.
And moreover, I do think it is undesirable to use Fannie
Mae and Freddie Mac to generate revenues for other Government
functions. The best example of that, of course, was the 10
basis point increase in the G-fee to help pay for the payroll
tax holiday. I view that as a mistake, and hopefully we do not
go down that same path again.
So I do not think that is the appropriate use of these
institutions and the profits they are now generating. It should
be used to repay taxpayers and used to help facilitate whatever
future housing finance system we evolve to.
Senator Crapo. Thank you.
Mr. Millstein. I agree with that. I think the alternative
to the system we currently have where all of their profits are
being swept into the Federal Treasury is, in effect, you are
taxing the mortgage market to pay for other things. And here
you are embarked upon a great legislative reform effort. These
entities are producing $25 billion to $30 billion a year of
cash-flow that can be used to recapitalize a grossly
undercapitalized housing finance system.
It seems to me that as part of your reform effort, you
should seize those revenues and use them for this public
purpose of fixing the housing market.
Senator Crapo. Thank you. And, Dr. Zandi, I noted in my
opening statement that we cannot allow our legitimate concerns
about unknowns to cause us to forget the equally legitimate
concerns about the status quo. In your testimony, you laid out
several of those concerns. Could you further expand on why you
believe that inaction carries its own dangers?
Mr. Zandi. Yeah. I think the current system is a real
problem. You know, clearly, what we just discussed is an issue.
As Fannie and Freddie continue to generate profits, the
inclination will be to use that for other Government purposes,
and again, I do not think that is appropriate. So that is
clearly a risk.
And also, over time, with those profits getting
incorporated into the budget and the budget process, that will
make it more difficult to actually address the reform of Fannie
Mae and Freddie Mac. So the longer it takes us to do this, the
more difficult it will become.
Moreover, you know, the fact that Fannie Mae and Freddie
Mac are in this limbo status is creating its own set of issues
and problems. You know, increasingly over time, this will
become more apparent. I will just give you an example of what I
mean.
Right now, mortgage credit is very tight for potential
first-time home buyers. The average credit score for someone
selling a loan or purchasing a loan to Fannie Mae is 760. That
means you have to be in the top third of the distribution of
credit scores. That is not consistent with a first-time home
buyer.
One of the reasons for that is because of rep and warranty
policies between Fannie, Freddie, and the lenders, and this is
not getting resolved, certainly not quickly enough, and one of
the reasons I would proffer is because of this limbo status. So
that is just an example of what I mean. There are many others
that will become more evident going forward.
And then finally, there is no reason for it. Right? I mean,
there is plenty of private capital out there. They are all very
interested in participating in this market. It is clearly
obvious in lots of different places and there is no reason why
taxpayers should be taking credit risk on mortgage loans.
Senator Crapo. Thank you. And, Mr. Bovenzi, you noted many
valuable lessons learned from a positive nature about the RTC
that could provide insight into our decision making as we
approach Fannie Mae and Freddie Mac. Are there any cautionary
lessons that we should also glean from the history of the RTC?
Were there actions corrected at the time or merely thought you
had gained through your benefit of hindsight?
Mr. Bovenzi. There are probably many cautionary lessons
given that it was not, you know, a smooth path. I mean, there
was a time when the RTC was accused of ruining the housing
markets in the United States, but looking back with some
perspective, they are now viewed as having been a success in
terms of straightening it out.
So I think, you know, there is a danger to everybody having
a very short-term perspective of what needs to be done versus a
longer-term perspective. As we are hearing, this is not a
simple matter either way. You make changes and there will be
some disruption; you do not make changes and you have issues as
well.
So I think what I have tried to lay out were things to
think about if Congress goes down this route that are
cautionary to think about, like how do we treat the employees
in the GSEs who have the expertise to help make a smoother
transition and issues like that from the RTC's and FDIC's
experience.
Senator Crapo. Thank you.
Chairman Johnson. Senator Warner.
Senator Warner. Thank you, Mr. Chairman. First of all, I
just want to thank you and Ranking Member Crapo for the
attention you have given this issue. I believe this is our
ninth hearing this fall, and while there is a lot of other
activities going on in the body that are showing, perhaps,
disruption, I think your leadership of this Committee has shown
a real focus and attention on this issue, and even during these
challenging weeks, continuing to advance the process forward. I
think it is very helpful, one.
Two, I think this is about our 20th housing finance hearing
we have had since 2011, and again, I think you and your team
have built a very solid record here. A couple of comments
before I get to my question. I think Dr. Zandi has pointed out
some of the challenges around 1217 and the legislation we have
worked on.
We do think there needs to be high, solid capital
standards. I do think that at some point, though, the
comparison to the status quo, which is right now no private
capital risk, this is purely a public sector socialized risk,
it is really not even an apples-to-apples comparison. Any
system, even Mr. Millstein's system, would have higher costs
involved because there would be additional private capital.
I also wanted to comment to Dr. Min. I think one of the
comments you made about the notion that any future market
rental housing, other kind of first-time home buyer access,
perhaps not completely eloquently put together, but are market
access fund, recognizing there is that appropriate role, I
think, is something that we have taken a pretty good stab at.
I concur with a lot of the folks on the panel, that we have
got to get this right in this transition period. The only
hesitancy--one of the hesitancies I have, if you move simply
from a target versus time line, those targets always can slide.
We have seen that constantly.
I think Dr. Zandi's comments about, you know, how quickly
FHFA is moving toward the common securitization platform is
trying to get this judgment right, and I would welcome from all
on the panel, and really appreciate your testimonies, you know,
what are those speed bumps, metrics, hurdles, whatever we want
to call it, along the way so that we do not have this simple
moment where we flip on the switch.
There has got to be these points, and I think again, Dr.
Zandi--well, Professor Min, you have got a number of those, but
clearly, getting the securitization platform right and the
common security right is very important.
I guess what I would hope as well is, Dr. Min, to make sure
that we do not try to recreate the wheel, you know, a lot of
the activity going on at FHFA right now, I think, are efforts
to make sure that we transfer over the assets.
I want to agree with Mr. Millstein. I think it is really
important that we keep the people and intellectual capital
assets from Fannie and Freddie, for a variety of reasons,
including the fact that a lot of those folks live in Virginia.
So I guess what I would ask is, how do we make sure--let us
take, for example, the question around the starting up an
issuer.
Do you, Professor Min or Dr. Zandi, want to make a couple
comments on what should be some of those gating things we
should look at?
Mr. Min. I will leave you some time. I think obviously you
need capital. As I point out in my written testimony, I think
that there is a lot of capital. It obviously depends on the
cost of capital, reverses return analysis that you are well
familiar with. But I think on top of that, there needs to be
some sense that there is going to be investor liquidity in this
market, and so for that reason alone, the CSP and the single
security are important.
I think as Dr. Zandi and I both point out in our
testimonies, right now there is significant pricing differences
between Fannies and Freddies. If you simply do not have a
single security, those differences in pricing liquidity will be
greatly exacerbated as you add a third, fourth, or more
issuers.
That is why that particular product is so important. I
think at the same time, you want to make sure that a lot of the
agency investors, the agency liquidity that is there transfers
over, because I think that itself will be a very important
calculation for investors.
Is there a market key? Will it be there when we startup the
operation for that switch? And that is why I have recommended
some of the steps I have about getting liquidity in place.
Mr. Zandi. Senator, so when you say issuer, I take that to
be issuer guarantor, and that is a matter of some discussion
and debate as well, but I am just going to assume that. You
know, I do think that it will be very important in the
transition process that the Government is helping to stand up
additional guarantors. I mean, we have got Fannie and Freddie
and the infrastructure is there and we can make this work so
that they go out into the marketplace and liquidity is
sustained and everything is working properly.
But to really make this a well-functioning system for the
future, we need to be able to have more than two guarantors out
there. We need many. It depends on the structure of the market
and what the scale economies are, but my guess is that the
market could reasonably support four or five, six guarantors,
and that would be competitive market.
But that is not going to happen, I do not think, on its
own. It certainly is not going to happen if we just throw
Fannie Mae and Freddie Mac, whoever we turn them into, out into
the marketplace. That is just going to shut down the market to
competition.
Even if you do the common securitization platform, even if
you have a common security that will help bring down various
entry, even if you stipulate that these new entities are SIFI
institutions and they have higher capital requirements and
raise their cost of capital, I still fear that given these
legacy relationships they have, it will be very difficult to
generate a competitive market.
So it is going to be important, in the transition process,
that we have a clearly articulated means for standing up,
helping stand up new guarantors into the marketplace to make
this a competitive market. And we can talk about how to do
that, but that is key to this all working out in the end.
Senator Warner. Thank you, Mr. Chairman. I just want a
final quick comment. It is just that I also want to echo and
appreciate Senator Crapo's comments about the challenges of the
status quo and the very real threat. Even though I think the
majority of us all feel very strongly, we ought to keep the G-
fees in this industry, there remains that threat. So I
appreciate your comments as well.
Chairman Johnson. Senator Corker.
Senator Corker. Thank you, Mr. Chairman. I appreciate you
having all of these hearings, and Ranking Member Crapo, and
working with him to make that happen, and all of you for coming
to them still trying to adjust to having all my rights stripped
away yesterday by people I have worked with for a long, long
time to solve issues in a bipartisan way without any
discussion.
But anyway, Mr. Min, I want to tell you, I do appreciate
some of the concerns you raised about mortgage-backed
securities and legacy mortgage-backed securities. The notion
that you mentioned about basically turning them into FMIC-
backed mortgage-backed securities is exactly what 1217
envisions. So I appreciate the contribution you made to this
hearing in stating that, and it is exactly what we would like
to do.
Dr. Zandi, I know we have worked together for a long, long
time to try to figure out a way to move past the system that we
have, and I know that you know we have had people on the left,
on the right, in the center. I think you attended a dinner
about 15 months that had multiple kinds of folks there to try
to figure out how we begin to look at this system.
I think there is no question that you agree, as someone who
spent a lot of time in this world, that we need to move past
the system we have now where private sector investors do
really, really well when the market is good, and then the
taxpayers do really bad when things go wrong. You agree with
that, is that correct?
Mr. Zandi. Yes, sir, I do.
Senator Corker. And I know that we have looked a lot at
capital and, you know, how much capital is necessary. Today, in
a $5 trillion market, which 90 percent of new loans today are
coming through some Government entity, but under the 1217 bill,
there would need to be about $500 billion in capital, is that
correct, under the 1217 arrangements?
Mr. Zandi. Yeah, 5 percent of the $5 trillion.
Senator Corker. No, 10 percent of that.
Mr. Zandi. Excuse me, 10 percent of the $5 trillion.
Senator Corker. I know that was a Freudian slip.
Mr. Zandi. Yeah, that was Freudian. I am sorry. I am just
trying to slip down----
Senator Corker. I am not my normal energetic self. I
apologize.
Mr. Zandi. As you could tell, neither am I.
Senator Corker. And I know that you and I looked at a chart
that actually looked at nonagency capital returns over the last
few days about $636 billion coming over the next 7 years back
into the system. So the notion of trying to have about $500
billion, quote, quote, quote, is that is what we build back up
to, does not seem in any way to be farfetched, does it?
Mr. Zandi. I think we can do it. I think there is, as we
can tell from what is going on in the marketplace in recent
months, a great deal of interest and participating in taking on
mortgage credit risk. We can see it from the proposals that
have come forward in the last couple weeks.
Senator Corker. Yes.
Mr. Zandi. We can see it in my world, in the PMI world that
I am a board member of MGIC. So there is clearly a great deal
of interest in private capital coming into this marketplace,
yes.
Senator Corker. I noticed, over the last couple of weeks,
we had an offer from a gentleman named Berkowitz from down near
Florida to buy actually the credit risk portion of the remnants
of Fannie and Freddie, and while the offer likely will be
rejected because of the amount, in talking with people that are
related to that type of offer, in many ways, does that not give
us hope that there are people out there that are willing to
invest in the credit risk portion only and it is a matter of
just getting the economics right?
But it does show that there is interest out there for
people to participate as buying guarantors in a new system. Is
that correct?
Mr. Zandi. Yeah, that is exactly right. I mean, if you look
at the proposal, there are $17 billion of real capital in the
proposal, at least that is the number he put forward, and by my
calculation, $17 billion gets you pretty close to what you need
to get to that 10-percent capital in year one of the new
system.
So immediately I saw that, I go, Oh, OK. So there is
private capital out there willing to participate in this
marketplace and it gets you pretty close to where you want to
be in year one. So I took a great deal of solace in that. And I
agree, I would not go with the proposed structure that he has,
but certainly, you know----
Senator Corker. But my sense is, and I have read some
quotes since those discussions matured a little bit and
developed, it seems to me that the offer that he made was very
exactly in line with 1217. In other words, you would have to
have 1217 pass to create this modular competitive system for
that type of offer to even work.
So it seems to me that what we are already seeing out there
is hedge funds and private equity and other folks who are
willing to capitalize this system in such a way as to make it
competitive and modular. Is that correct?
Mr. Zandi. Exactly right. I mean, when I was talking about
those new guarantors coming into the marketplace, you could
envisage what he is proposing as, in fact, one of those new
guarantors. So it fits exactly in with the vision you have that
is in 1217.
Senator Corker. Mr. Millstein, I know we have had a number
of meetings in our office. I know that you were in the public
sector and now are out in the private sector over the last
couple of years with your firm. I know that when you are in the
public sector, you learn about just sort of the way Government
works and you can figure out ways of, you know there is a
degree of incompetence and people are afraid of change and
there are ways of sort of making money off of that.
I know that--I appreciate your testimony today, but I do
want the Committee to know that unless Mr. Millstein is willing
to, under oath, say differently, he does benefit personally in
the event, if we have the fear and inability to move beyond the
system that we have, in other words, if we were to do what he
espoused, he and his family would benefit greatly personally. I
just think that----
Mr. Millstein. Senator, let me interrupt you just so we do
not waste any more time on that. I no longer----
Senator Corker. Well, actually----
Mr. Millstein. I no longer----
Senator Corker. ----I am just making a statement----
Mr. Millstein. I know, but I want to make sure you are----
Senator Corker. ----less of a statement of fact and you are
willing to----
Mr. Millstein. So I am going to tell you right now, that I
do not own those securities any longer. I do own a home and I
am a small-time real estate developer, so I have a deep
interest in how you do this, to make sure you do it right, but
I am no longer a stockholder.
Senator Corker. So I guess--well, very good. That is good
to hear. Right.
Chairman Johnson. Professor Min, in addition to single
family, could you elaborate and what key issues should we
consider in structuring an orderly transition for multifamily
housing reform?
Mr. Min. Thank you for that question. I did not have a
chance to get into in my oral testimony. I think that--so as it
is structured now, I think that the Corker-Warner plan
envisages the market access fund, which is actually made up of
several discrete funds receiving or being responsible for the
bulk of multifamily.
In addition to that, I think you all received at least one
or two proposals, thinking about ways to create multifamily
specific guarantor issuers under the new system envisioned by
Corker-Warner. So I think those would be the two main engines
for multifamily, and I think that in theory, as I have
researched this issue, that should actually work.
But the sort of issues I raised in my testimony, my written
testimony, are about getting from Point A to Point B. I think
these are the segments, particularly in the affordable rental
space, that are most likely to experience a vacuum as we reduce
the footprint of the GSEs and try to transition to that Point B
where these institutions will be responsible for filling that
space.
So what I propose in my testimony is, is two steps.
Basically, let us try to get a running start for multifamily
housing for institutions that are dedicated for this affordable
rental space, including starting prefunding the market access
fund now, siphoning off a little bit of the G-fee, which is
fairly ample at this point, and start to prefund that so that
it really is in place and operational when that transition
happens.
The second step is to go along with the plan proposed by
former Assistant Secretary Dr. Bostick of trying to create
multifamily subsidiaries of Fannie and Freddie and allowing
them a head start so that they can be operational again as we
get into this transition period.
I think the concern is that this is a space that purely
private capital has not always gone into and it has been--it is
an increasingly important part of the system, particularly as I
mentioned, as we think about pulling back from single family
and pulling back the Federal Government's footprint. So those
are the steps I would outline.
Chairman Johnson. Senator Crapo, do you have any follow-up?
Senator Crapo. Yes, just one last question, Mr. Chairman.
And I guess I will direct this to Dr. Zandi, but any of the
others who would like to jump in on it, feel welcome.
On this question related to the Chairman's earlier question
about whether we needed to have a target focus like achieving
benchmarks before we actually pull the trigger and move forward
with the finalization, or whether we needed time lines as
Senate Bill 1217 currently has, and I know Senator Warner got
into this.
I understand the comments that all of you have already made
about the fact that this is complicated and a rigid time line
might not be sufficiently flexible to allow us to get it right.
That being said, I kind of want to focus on the other side of
that, and that is, if you just continue to try to look at
benchmarks, I look at the--I am an opponent of quantitative
easing, and I look at it the way that we are trying to get out
of quantitative easing right now with benchmark targets that we
are not meeting.
And we just seem to always stay where we are. I know that
may not be a perfect example, but my point is, how do we
ultimately have some accountability here in the sense that we
get there if we start focusing on a benchmark process in order
to implement the legislation? Dr. Zandi.
Mr. Zandi. Yeah, an excellent point and it is a tension
that Senator Warner highlighted as well. I think the approach
would be to have very explicit benchmarks and then to make sure
that there is clear oversight with regard to how those
benchmarks are achieved or not achieved and why.
So if you come up to a certain--it is not that we should
not have a time line. We should have a time line, but with
benchmarks in the time line, and as we approach them, then if
we are not achieving them, then we need--there should be
oversight and there should be accountability with regard to why
and using that as a mechanism for ensuring we get back on the
time line.
But there may be reasons why you might want to delay,
because there are things that are completely out of our
control. I will give you just a blue sky example, but it is
very possible. You get up to the point where you want to turn
on the light switch and 2 weeks before, there was a financial
event in Europe and the financial markets are in turmoil.
Now, do you really want to turn on the light switch at that
point? The answer is probably not. Nobody would really want to
do that. So you need to have some flexibility with regard to
how you do this.
Senator Crapo. All right. Thank you.
Mr. Millstein. Can I answer that, though?
Senator Crapo. Yes.
Mr. Millstein. Look, the system that Senators Warner and
Corker have designed is one that requires a good deal of first-
loss capital. And we here can speculate on how much there is
and when it will come and how much appetite there is. But at
the end of the day, your legislation is depending upon the
investment decisions of tens of thousands of investment
managers, and you cannot command them to show up on cue.
They are going to have to see the new system design and see
it operate a little bit before they are going to start coming
in droves. Today, with all due respect to my colleague, Mr.
Zandi, today there is $8 billion of capital in the private
mortgage insurance business.
Yes, Mr. Berkowitz has shown up with an offer that says he
will put $17 billion in, but there is a long way between that
offer and a closing. A lot of other things have to happen. But
the most important lesson, I think, in terms of this benchmark
versus time line is, you are trying to design a system that
will induce people to put capital into new mortgage insurers
and into new first-loss securities that do not exist today, in
a system with a guarantee that does not exist today, for a
market of new securities that does not exist today.
It is going to take time and you are depending on tens of
thousands of individual investment managers to play with you,
to shake your hand and say, Yes, I will help you build this
market. They may come on a 5-year timetable, but if they do
not, when you flip your switch, your system is going to shut
mortgage availability down, nothing any of you want to do.
So that is the risk, to me, of having a very hard time line
built into this, and that is the risk, to me, of taking the
assets under your control today, that you could recapitalize
today, to make sure that this system functions. The risk of
just saying, ``No,'' in order to preserve the possibility that
MGIC will be able to raise $125 billion of capital in order to
play my first-loss role, I am going to trash the assets that
are currently doing this for me in this market. That, to me, is
crazy, crazy.
Mr. Bovenzi. If I could make one comment? There may be some
precedents for finding something in between. I think when the
FDIC sets up a bridge bank for a failed institution, there is a
set timeframe for how long the Government should own that bank
before it has been sold or restructured.
But because of the public policy issue of not having
nationalized banks, if there are reasons why that life of the
bridge bank should extend, I think the FDIC Board of Directors
could, you know, do that and have flexibility to have it come
forward for, you know, a board to say, OK, there are certain
reasons why we need to extend this beyond.
So maybe there is some areas where there may be precedents
where you can try to set some form of timeframe with
benchmarks, but flexibility to extend if necessary.
Chairman Johnson. Dr. Zandi, do you care to respond?
Mr. Zandi. Thank you. I agree that there is a lot of moving
parts here and there is a lot of uncertainty. A lot of the
uncertainty revolve around issues of private capital and what
cost will the private capital come into the system. And there
is a lot of codependencies here.
I mean, how much capital comes in depends on how much
clarity you provide and oversight you provide. So there is a
lot of reasons to be nervous about the process. But let me say
that this is very doable. It is not like we cannot do this, and
we can, and the economics of this are such that there will be
investors and there will be private capital. It will come in.
And this can work out very, very gracefully. We just have to be
very careful about how we do this and articulate it.
Chairman Johnson. Does Senator Warner have any follow-up?
Senator Warner. I just really appreciate the conversation
back and forth here. I saw one of the groups who are interested
in this issue this week and they were talking about how the
housing market has recovered a bit, but there are still
challenges.
And I pressed this particular group, there was like, you
know, when are you ever going to hit that magic time? I mean,
if we could predict when that total solidness is there, that
may just be an unobtainable goal. I would reflect that we are 5
years after the crisis now. Obviously the housing market's
recovery has not been to that whole 5 years.
I would point out that in the wild and wooly final days of
Dodd-Frank, there were efforts to, on the floor of the Senate,
to unwind the GSEs with no transition on a 2-year basis and
completely zero them out that got a lot of votes. And, you
know, I think we have, in a broadly bipartisan way,
directionally set a path for reform that the Chairman and the
Ranking Member are improving upon.
And if there are moments along the way during this 5-year
period where we will have--I am not sure what the right term
is--a speed bump or indicators, you know, such as the common
securitization platform that is independent of--I think the
appropriate questions Mr. Millstein has raised about private
capital showing up.
You have got FHFA trying risk-sharing models now. We have
had testimony at these hearings about trying to create kind of
blended securities during the interim that would again give us,
I think, better indices. And, you know, my point is, you have
got to start this process, and, you know, waiting for the
perfect moment leaves also a huge amount of uncertainty over
this market that is terribly important to our economy.
And as we have perhaps repeatedly shown, as well
intentioned as we may be on this panel about not using G-fees
or other fees for other purposes, locking in a reform basis may
be one of the best ways to ensure that we do not do that.
I would also make a comment, as one of the things that I
think was an area that needed a great deal of improvement in
1217 was multifamily. I think there are a lot of Members on
this panel from both sides who are working on improving that.
And, frankly, in that area more than ever, the do no harm
mantra is important because they did not create the challenges.
That part of the business did not create the challenges.
So I guess it is not as much a question other than the fact
that I think I would at least take your admonitions to heart,
but not to the point of saying--as an excuse to say we have got
to spend another months, years, whatever, studying this issue.
I think it is time to proceed.
And again, I am going to close by simply thanking the Chair
and the Ranking Member because you have brought an attention
and focus to this that I think has really advanced the debate
greatly.
Mr. Zandi. Can I make one quick comment just to reinforce
something the Senator said about timing? You could make a very
good case that timing is incredibly good right now. Right? I
mean, the credit environment is excellent. The number of
mortgage loans that are 30-day delinquent is as low as it has
been in the data that I have back into the 1980s.
Sixty-day delinquency, that is 2 months late, is as low as
it has been, and even 90-day delinquencies are approaching a
record low. And private capital is very interested in
participating now. And you also have very low yields and, you
know, here is an opportunity to get the yield.
So I would echo that this is a--you know, there are
challenges in the environment that we exist and we have a 7-
percent unemployment rate and we have got to make sure that
this all works out well. But I would argue that there is no
better time for doing this than literally right now.
Senator Warner. Right.
Mr. Zandi. And then I would say, simply to add, that if
Senator Crapo, thinking along the continuum, but QE is going to
end at some point. Interest rates are turning back to kind of
more normal levels would make it even harder to make this
transition. If not now, when?
Chairman Johnson. Mr. Millstein, do you have any closing
comments?
Mr. Millstein. Well, I want to join with Senator Warner in
saying that I agree, the most important thing we could do for
private capital beginning to come back into this market is to
pass legislation telling the markets where the Government is
going to move with regard to Fannie and Freddie and ending of
the conservatorships. They are at the center of the housing
market today.
They are operating with no capital and not seeking a
capital return, making the pricing of their guarantees, I
think, a discount to what private parties would do. And
clarifying the future state of Fannie and Freddie to the market
is critically important for private capital formation to
return.
My caution, despite what Senator Corker may have
characterized my remarks to be, my caution is not about
legislating. My caution is about having a mechanical line down
of the two entities that are providing almost all credit
formation in the conforming market today.
You have run significant risks, but telling the markets
through your legislation where you are going and what your
objectives are, competitive, first-loss, mortgage guarantee
industry providing substantial capital in front of a new
Government regulator who is strong, but guaranteeing qualified
product, that is an important signal.
It will allow private capital to start developing plans
around your new structure. But the one caution I am giving you
is, while they are doing that, the pace at which they come you
cannot command. You cannot predict and you cannot write a
timetable to be certain of, and therefore, you need to take
what you have got that is working and transition it to having a
role in the new system so that you have continuous mortgage
credit formation along the way.
Chairman Johnson. Thank you all, our witnesses, for being
here with us for an unusual hearing. Given the number of
Members urging the Committee to move quickly, we anticipated
greater attendance. I also want to thank----
Senator Warner. Mr. Chairman, I am proud to be here.
Chairman Johnson. I also want to thank Senator Crapo for
his thoughtful questions, good partnership and commitment to
housing finance reform. This hearing is adjourned.
[Whereupon, at 11:20 a.m., the hearing was adjourned.]
[Prepared statements and responses to written questions
supplied for the record follow:]
PREPARED STATEMENT OF JAMES MILLSTEIN
Chairman and Chief Executive Officer, Millstein and Company
November 22, 2013
Introduction
Chairman Johnson, Ranking Member Crapo, Members of the Committee,
thank you for the opportunity to testify on the development of an
effective transition plan for the U.S. housing finance system.
I have spent the entirety of my 30-year professional career--as a
lawyer, banker, and public servant--in the corporate restructuring
business. I have restructured companies as diverse as American
Airlines, WorldCom, and Charter Communications in the United States,
Cadillac Fairview in Canada, United Pan European Communications,
EuroDisney, and Marconi in Europe, and Daewoo Corporation in Korea.
During the recent financial crisis, I served as the Chief Restructuring
Officer of the U.S. Department of the Treasury. In that role, my
primary responsibilities were managing, restructuring and designing the
exit from the Department's substantial investments in AIG and Ally
Financial.
I am here today because embedded in the task of reforming our
Nation's housing finance system is a restructuring of the two largest
players in that system: Freddie Mac and Fannie Mae. These companies now
operate in conservatorship under the control and direction of the
Federal Housing Finance Agency. Because they are central to mortgage
credit formation in the United States today, ``winding them down'' as
some members of Congress and the Administration suggest is certain to
have significant and adverse consequences for mortgage credit
availability and for the nascent housing and economic recovery. Rather
than wind down, I urge you to consider a restructuring alternative that
addresses the fundamental causes of the companies' insolvency,
eliminates the private gain/public loss nature of their current
Government sponsorship, generates a significant profit to Treasury for
supporting their solvency, and, most importantly, ensures a smooth
transition to a new housing finance system that better protects
taxpayers against future losses while providing for the continuing
availability of credit to the creditworthy.
There appears to be a growing consensus in the policy community
around the basic architecture of that new housing finance system. A
Federal guarantee on qualified mortgage products is required to ensure
the widespread availability of a 30-year fixed-rate product, and to
sustain the deep and liquid mortgage securities funding markets that
have developed over the past 30 years to complement balance sheet
lending from the U.S. banking system. The guarantee should be explicit
and structured as reinsurance, available to reimburse investor losses
only after a layer of private ``first-loss'' insurance provided by
well-capitalized mortgage insurers or subordinated capital provided
through structured product markets has been exhausted. The reinsurance
should be priced at arm's length by an independent agency required to
use its reinsurance fees to build a reserve fund to protect taxpayers
against future loss should that reinsurance ever be called. Finally, in
contrast to the system prevailing before 2008, the Government reinsurer
also needs to be a strong regulator with authority over all issuers,
guarantors and servicers with whom it interacts in the new system. In
this regard, I commend Senators Corker and Warner and the coalition of
other members of this panel behind S.1217 for putting out a bill with
all of these elements in it.
However, the transition to this new system contemplated by S.1217
is fraught with difficulty and needs serious rethinking to mitigate
three significant risks that any credible transition plan must address.
First, our fragile economic recovery cannot afford the risk of a
significant disruption in mortgage credit. Borrowing rates will need to
rise in the new system to reflect the cost of the first-loss capital
and new reserves required to protect taxpayers on their guarantee. At
the same time we need to protect against a significant contraction in
the availability of housing credit that would push us back into
recession. Second, the Government must end its ongoing backstop of
Fannie Mae and Freddie Mac in conservatorship in a way that minimizes
the likelihood that Treasury will need to cover future losses on their
$5.5 trillion of liabilities. While the substantial guarantee fees and
net interest margin which the companies are currently earning and
paying over to Treasury may look like an asset to be seized by
taxpayers as the quid pro quo for their bailout, it could easily turn
out to be a substantial liability if there were another significant
housing downturn. Managing that liability in a responsible way to avoid
future taxpayer losses is a critical challenge of the transition.
Third, there must be a credible path toward the development of the
substantial layer of private ``first loss'' capital on which the
functioning of the new system will depend. If you build the new
Government reinsurer but the required layer of first-loss capital
doesn't come in the size or at the pace of your contemplated wind down
of Fannie and Freddie, the whole system will shut down before it has a
chance to start. The idea that ``if you build it, they will come,'' may
work in the movies, but you are playing with the Nation's housing
finance system. Hope is not a credible strategy.
You have to make a fundamental choice in meeting these challenges
in the transition: Restructure Fannie and Freddie and use their assets
and operations to create a well-capitalized set of private market
players who can ensure that the new system functions as contemplated,
or wind them down on the bet that if you build the new reinsurance
system, new private players with the sizeable capital required to make
the new system function will come. My concern with both S.1217 and the
Protecting American Taxpayers and Homeowners Act introduced in the
House of Representatives is that each is based on the bet that to-be-
named new players with capital yet to be raised will show up right on
queue as the two institutions at the center of the current system are
mechanically wound down. As I hope to demonstrate in the following
testimony, we don't have to gamble with the future of the housing
market. There is a better alternative.
Evolution of the Government's Role in the Conforming Mortgage Market
A responsible transition begins with a clear understanding of the
status quo and how it arose.
In the late 19th and early 20th centuries, there were various
attempts to use bond markets to fund housing and commercial real
estate. But the lack of adequate securities law and insurance company
regulation, the absence of uniform contractual investor protections,
poor underwriting, and outright fraud led to repeated funding market
collapses, subjecting the U.S. economy to painful downturns. The 1870s
and 1880s featured a permutation of covered bonds, where mortgage
originators issued debt to the public collateralized by pools of the
mortgages they had originated. While this market functioned for a time
and funneled investor dollars into housing finance, it eventually
collapsed because the originators violated their purported underwriting
standards and packed the pools with nonconforming collateral. In the
1900s, New York title guarantee companies originated mortgages, insured
them, and sold participation certificates backed by them (an early form
of mortgage-backed securities). These title insurance companies
eventually failed, and the mortgage securities markets they supported
collapsed, because of poor underwriting, thin capitalization, and weak
State insurance regulation. The 1920s featured the issuance of single-
property real estate bonds, each governed by a separate set of
indenture provisions, the proceeds of which were used to finance large
construction projects. Poor underwriting and weak investor protections
led to its eventual collapse. The same decade (the roaring 20s) also
saw bank and thrift failures at an average rate of 600 per year (in a
banking system with approximately 10,000 banks and thrifts), a crisis
by today's standards and significantly disruptive to home lending and
local economic activity. The pace of bank and thrift failures peaked 5
years after the Crash of 1929 when, in 1933, roughly 4,000 banks and
thrifts failed, resulting in widespread foreclosure and a severe
contraction of housing finance credit.
In response to the housing and banking crisis of the 1930s, the
Federal Government restructured the banking system and significantly
expanded its role in housing finance. Among other things, the Banking
Act of 1933 created and Banking Act of 1935 expanded the authority of
the Federal Deposit Insurance Corporation (FDIC), an independent agency
of the Government chartered to provide Federal deposit insurance to
banks to prevent the bank runs that forced the Roosevelt Administration
to impose a national 2-month long bank holiday in early 1933. The acts
also provided the FDIC with regulatory authority over its member banks,
initially funding its reserve fund with loans from the Treasury and
Federal Reserve. Those loans were repaid with interest after member
bank insurance fees began to accumulate. When faced with widespread
bank failures during the recent financial crisis, the FDIC's Deposit
Insurance Fund also fell into deficit. However, instead of drawing on
its line of credit with Treasury to replenish its coffers, the FDIC
pulled forward insurance assessments and imposed additional fees on its
member banks. The Dodd-Frank Act of 2010 requires the FDIC's Deposit
Insurance Fund to reach 1.35 percent of insured deposits by 2020.
In the 1930s, Congress also addressed mortgage finance directly. In
1932 it created and capitalized the Reconstruction Finance Corporation
(RFC), which made loans to, among others, banks and mortgage
associations. \1\ In 1932, Congress established the Federal Home Loan
Bank System in order to create additional funding for home loans
originated by savings and loan institutions. Federal Home Loan Banks
(FHLBanks) make loans to member institutions secured against eligible
collateral--typically mortgages--and issue debt to the public to fund
such lending activity. The cost of that funding is generally lower than
an individual member can obtain because the debt is the joint and
several obligation of all FHLBanks, which operate under Government-
sponsored charters. The FHLBanks are capitalized by their members,
whose borrowing limits are proportionate to their respective capital
contributions. The FHLBanks are regulated by the Federal Housing
Finance Agency and have a minimum capital requirement of 4 percent of
assets. In exchange for their Federal charters and exemption from State
taxation, FHLBanks pay an assessment of 10 percent of annual earnings
for affordable housing programs.
---------------------------------------------------------------------------
\1\ The RFC funded various relief projects during the Depression
and authorized loans and investments to support the Government's
efforts during World War II. It also established multiple companies to
carry out its mission. The RFC was ultimately disbanded in 1957.
---------------------------------------------------------------------------
These efforts helped stabilize banks and other mortgage providers
in the 1930s. But many would-be homebuyers in the 1930s remained shut
out of the mortgage market, and home construction remained muted. The
average mortgage required a large downpayment, had a maturity of 3 to 5
years, and featured large balloon payments at maturity. Although most
loans were renewable at maturity the interest rate would reset,
subjecting borrowers to the risk of significant interest rate movement
over the short life of the mortgage loan, with no ability to hedge that
risk.
The National Housing Act of 1934 established the Federal Housing
Administration (FHA) to address this problem and facilitate credit for
home construction and repairs to a broader swath of borrowers. By
offering insurance in exchange for a fee and assuming a first layer of
risk, the FHA made possible the issuance of fixed-rate, long-term
mortgage with regular monthly payments. \2\ The act authorized the
creation of a reserve fund to support claims made on the Government's
insurance: the Mutual Mortgage Insurance Fund (MMI Fund). The act also
provided initial capital for the fund, and it has since been funded
through premiums on insured mortgage loans. Today the MMI Fund is
required to maintain a reserve of 2 percent of insured loans based on
projected losses over a 30-year horizon. During most of its history, a
portion of premiums collected in excess of that reserve minimum have
been transferred to Treasury. Between 2001 and 2007, for example, the
program transferred approximately $14 billion to Treasury. Earlier this
year, for the first time in FHA's history, it borrowed $1.7 billion
from Treasury to bring the MMI Fund reserve up to its congressionally
mandated minimum level. The FHA single-family mortgage insurance
program generally targets first-time and lower-income homebuyers,
although during the recent crisis, when other private mortgage insurers
failed or became undercapitalized, the FHA significantly expanded its
footprint to ensure credit availability.
---------------------------------------------------------------------------
\2\ As opposed to short-term balloon payment mortgages that were
more traditionally available.
---------------------------------------------------------------------------
The same act that created the FHA also authorized the agency to
create ``national mortgage associations'' to purchase and sell FHA-
insured mortgages. The objective was to create additional liquidity for
housing credit beyond the then FDIC-guaranteed deposit-based funding
available in the banking system or though the discounting of mortgages
at the FHLBanks. Similar to the FHLBanks, the national mortgage
associations would tap capital markets to fund FHA-insured mortgage
originations. Unlike the FHLBanks, they would not be cooperatives.
Instead, it was contemplated that they would have a broad base of
private equity investors. However, 4 years after passage of the
National Housing Act in 1934, no national mortgage association charters
were ever taken out by the private sector. As a result, at the urging
of the Roosevelt Administration, the Government-owned RFC began buying
FHA-insured loans and in 1938 formed a subsidiary that became the only
chartered national mortgage association: the Federal National Mortgage
Association (Fannie Mae).
Fannie provided a secondary mortgage market into which originators
could sell loans, which freed capital and provided funding so that
those originators could recycle the funds and extend additional
mortgage credit. In 1954, after 16 years as a purely Government entity,
the Federal National Mortgage Association Charter Act converted Fannie
into a mixed-ownership corporation, where the Government held preferred
stock and private investors held its common stock. In 1968, in order to
remove its growing balance sheet liabilities from the Federal budget,
Congress split the company in two, leaving behind the Government
National Mortgage Association (Ginnie Mae), a Government entity that
began guaranteeing passthrough securities backed by mortgages insured
by FHA, the Department of Veterans Affairs, and Farmers Home
Administration and fully privatized the ownership of Fannie Mae.
However, the Government charter remained with the privatized Fannie,
and with that charter came the obligations to serve the public policy
ends of increasing home ownership and supporting low- and moderate-
income housing. This original policy error in the 1968 privatization
created a private shareholder-owned company with a public mission,
allowing Republican and Democrat Administrations alike to pressure
Fannie to increase credit availability to serve political ends. It also
allowed Fannie to use its public mission as political ammunition to
fend off challenges from banks to its market power and to its
relatively lax regulatory oversight.
In 1970 Congress created the Federal Home Loan Mortgage Corporation
(Freddie Mac) to compete with Fannie, expand the secondary market for
mortgages, and help thrifts manage interest rate risk. Freddie had the
same charter and implied Government guarantee as Fannie, but it was
initially capitalized and owned by the FHLBanks. In 1971 Freddie issued
its first mortgage backed security (MBS). Securitizing mortgages
purchased from thrifts defined the company's business model over the
next few decades, while Fannie continued primarily to purchase and hold
mortgage loans in its portfolio funded with balance sheet borrowing
from the capital markets. Both Government-sponsored enterprises (GSEs)
were permitted to purchase non-FHA-insured loans.
Inflation, interest rate volatility, economic downturns, and the
beginning of a 30-year wave of bank deregulation in the late 1970s and
early 1980s wreaked havoc on savings and loan associations, also known
as ``thrifts'', as well as on Fannie. Thrifts funded most of the long-
term mortgages they held with short-term obligations, largely deposits.
Until the 1980s regulators imposed ceilings on the rates of interest on
savings and time deposits that thrifts could pay. The Depository
Institutions Deregulation and Monetary Control Act of 1980 phased out
those limits. \3\ Meanwhile the then relatively new money market funds
began to grow rapidly with little regulatory impediment, competing with
thrifts for deposits. As a result, the rates paid on deposits
increased, and the thrifts faced a mismatch between their funding costs
(deposits) and the earnings on their primary assets. Fannie, as a
balance sheet lender, was exposed to similar risks, although it funded
itself in the capital markets, largely through long-term debt, rather
than with deposits. To try to help thrifts mitigate losses from the
mismatch between their deposit funding costs and the interest rates on
their long-term mortgage assets, the Garn-St. Germain Depository
Institutions Act of 1982 permitted them to expand into corporate
lending, an area for which they had little underwriting experience. In
relatively short order, thrifts experienced significant losses during
the recession that occurred during the mid-1980s, and the sector
virtually collapsed by the late 1980s, simultaneously putting the
FHLBanks under pressure. The same act also authorized banks to provide
adjustable-rate mortgages.
---------------------------------------------------------------------------
\3\ This was the first in a series of legislative deregulatory
initiatives for the financial sector over a 30-year period, which
culminated in the repeal of the separation of commercial and investment
banking that had been provided by the Banking Act of 1933.
---------------------------------------------------------------------------
Regulatory forbearance and a rapid change in its funding profile
away from long-term debt toward short-term debt permitted Fannie to
weather that particular storm, and by the early 1990s it began to shift
away from a long-term buy and hold strategy to the Freddie
securitization model. Meanwhile, in 1989 Congress reorganized and
privatized Freddie, and it opened membership in the FHLBanks to
commercial banks. The latter action more than offset the losses
suffered by the FHLBanks during the savings and loan crisis of the late
1980s. As a result of the expansion of FHLBank membership to include
commercial banks, FHLBank assets increased by a factor of six to
roughly $1 trillion.
During the 1980s, as the savings and loan crisis intensified and
many thrifts failed and withdrew from the mortgage funding markets, the
GSEs, FHA, VA, Ginnie, and the FHLBanks--filled the void and increased
their respective share of credit exposure to the residential mortgage
market. The figure to the right [below] illustrates that between 1982
and the mid-1990s, as the dominant savings and loan share of the market
shrank, the GSEs and Government agencies went from less than 10 percent
of the market to roughly 50 percent of it.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
After the collapse of the savings and loan associations in the
1980s, and the concomitant rise of the GSEs and Government agencies in
the market, the only other significant change in mortgage funding was
the creation and growth of the private-label mortgage backed security
(PLS) market. Like MBS issued by the GSEs, PLS facilitated the pass
through of funds from security investors to mortgage originators.
However, instead of leaving the credit risk with the Government or one
of its Government-sponsored proxies, the PLS market passed credit risk
onto private investors. Regulators and rating agencies facilitated the
growth of this market by lowering capital charges imposed on banks to
hold PLS, especially certain highly rated allegedly riskless tranches
of them, and by permitting their use as collateral in short-term
funding markets. While the single-family residential mortgage market
roughly doubled in size between 2000 and 2007 to over $10 trillion, PLS
outstanding in that market more than quadrupled, increasing from
approximately $400 billion to $2.3 trillion. A significant portion of
that increase represented PLS composed of subprime and Alt-A mortgages,
PLS which proved, we now know, to be rife with poor underwriting,
misrepresentations, and outright fraud.
Encouraged by legislated charter amendments in 1992 imposing new
affordability goals, growing competition from issuance in the PLS
markets, and private shareholder return expectations, Fannie and
Freddie used their Government-subsidized balance sheets to purchase
riskier assets, including PLS backed by subprime and Alt-A mortgages.
By contrast, the conforming loans bundled in MBS which the enterprises
guaranteed were quite conservatively underwritten and exhibited default
rates and loss severities modest by comparison to the default rates and
loss severities exhibited by the PLS which they bought. But weak
regulation of the enterprises failed to deter them from what would
prove to be a path toward self-destruction. And regulatory arbitrage
encouraged banks to purchase large volumes of PLS as well, effectively
setting them on the same path as Fannie and Freddie.
House prices began to collapse in 2006. PLS investors fled the
market and MBS issuance other than what the GSEs supported came to a
complete halt by the middle of 2008. In addition, banks cut back
sharply on balance sheet lending. The entire sector of private mortgage
insurers, who had conceptually provided credit enhancement for banks
and the enterprises, became insolvent. The thin layers of capital at
Fannie and Freddie proved woefully inadequate to absorb losses from the
risky assets they had acquired in their portfolios.
And so it was, in the summer of 2008, that the Federal Government
found the economy spiraling deep into recession with the continuing
collapse in home prices and mortgage credit formation adding weight to
that fall. No one was left to take credit risk in mortgages except for
Government agencies and the GSEs. And by early September, it seemed
that the GSEs might be on the brink of failing.
Treasury intervened. It sought and obtained legislation that
created a new regulator, the Federal Housing Finance Authority (FHFA),
over the entities with authority to place Fannie and Freddie into a
Government supervised conservatorship or receivership. Treasury had
hoped that the mere existence of that resolution authority would assure
markets that the Government would ultimately make good on what had been
an implicit guarantee of the enterprises' liabilities, and that Fannie
and Freddie would be able to fund themselves in private markets without
the need for formal Federal Government support.
However, as financial conditions continued to deteriorate, in
September 2008 the FHFA placed Fannie and Freddie into Federal
conservatorships. To ensure that they would remain solvent and continue
to provide funding to the mortgage markets, Treasury entered into
Preferred Stock Purchase Agreements through which it committed to
inject fresh capital into the companies. Treasury also took a warrant
on 79.9 percent of each company's common stock. This investment
structure allowed Treasury to achieve its goals without taking the
liabilities of the enterprises onto the Federal balance sheet. \4\ This
Treasury backstop convinced agency MBS and debt investors that they
were not at risk of default, and they continued to buy GSE MBS and
funded debt throughout the crisis.
---------------------------------------------------------------------------
\4\ Accounting rules could require consolidation if the Government
owns 80 percent or more of a private company.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Through the mechanism of the Preferred Stock Purchase Agreements,
Treasury transformed what had previously been an implicit guarantee of
two private companies into an explicit guarantee of their balance sheet
solvency. Rather than extending the guarantee directly to their
liabilities, however, Treasury guaranteed each entities' ability to pay
their obligations when due by committing to an open ended purchase of
sufficient preferred stock to cover the entities' losses.
Between the fall of 2008 and the end of 2011, to cover both
accounting reserves and allowances and real cash credit losses at the
enterprises, as well as a mandatory 10-percent dividend on the amount
of its outstanding senior preferred stock, Treasury provided $187
billion to Fannie and Freddie. However, as the housing market
stabilized in 2011, the enterprises' losses abated, and they started to
become extraordinarily profitable. Prices of securities in their
portfolios were rallying, guarantee fees had increased, and large
reversals in reserves and allowances seemed likely. In August of 2012,
Treasury and FHFA amended the terms of the PSPAs, changing the fixed
10-percent dividend on the outstanding preferred to a variable dividend
equal to 100 percent of the enterprises' earnings in any quarter. As a
result, by the end of this year--six quarters after the amendment--
Fannie and Freddie will have returned to taxpayers almost the entirety
of the $187 billion that Treasury invested in them. \5\
---------------------------------------------------------------------------
\5\ Treasury invested $187 trillion in the two companies. Through
the end of the third quarter of 2013, they had returned $146 billion
through dividends on Treasury's preferred securities. In their recent
earnings releases, the companies announced that they would send another
$39 billion in dividends by the end of December 2013. That will bring
total dividends to $185 billion.
---------------------------------------------------------------------------
Nevertheless, more than 5 years after Fannie and Freddie were
placed into conservatorship, the PLS market is moribund, balance sheet
lending by banks remains weak, and those private mortgage insurers that
have survived only as a result of regulatory forbearance and continuing
fee streams related to MBS guaranteed by the GSEs. As a result, Fannie
and Freddie, together with FHA, VA, and Ginnie, underwrite the credit
risk on and facilitate funding for 9 out of every 10 new mortgages
written in the United States today. After 5 years of conservatorship,
the GSEs remain liable for $4.4 trillion securities backed by single-
family mortgages--nearly half of the outstanding mortgage debt on
single-family residences in the United States--as well as $1.1 trillion
of on-balance-sheet funded debt. But, as a result of the August 2012
amendment, and the sweep of all of their earnings to the Treasury
Department, neither company has any material capital standing between
the taxpayers and potential future losses on those $5.5 trillion of
liabilities.
Lessons From Evolution for the Transition
One of the most important lessons from the evolution of the
mortgage markets in the United States is that the willingness of
private investors to take on mortgage credit risk is quite volatile,
subject to huge swings in appetite between irrational exuberance and
fear. After the turn of the century, PLS investors piled into a loosely
supervised, poorly underwritten market with securities whose underlying
collateral support was both opaque and riddled with fraud and
misrepresentation. In this respect, the PLS market of the 2000s was not
at all unlike the private mortgage markets that briefly flowered in the
1870s and 1880s and then again in the 1910s and 1920s. They were robust
for a time and then floundered on their own excesses.
Today, institutional investors remain wary of doing business with
any of the previous participants in the underwriting chain. And it is
safe to assume that it will be a long time before investors trust any
of the mortgage brokers and bank aggregators that managed to survive
the crisis, or the ratings agencies that were complicit in their flawed
offerings. The few PLS deals that have been done in the last few years
are being done almost entirely by new nonbank finance companies against
mortgages of high credit quality with huge layers of ``first loss''
protection ahead of the investors.
The thrift industry has all but disappeared as a source of mortgage
funding and overall bank portfolio lending is down 20 percent from
2008. Although banks have worked through the majority of their troubled
residential mortgages, capital requirements have increased, making it
more expensive to hold mortgage loans on balance sheet. And while
deposits at banks have expanded slowly since the crisis, they would
need to allocate their entire deposit base to fund the existing
mortgage market. That means that banks would have to stop all
commercial and consumer lending, grinding our economy to a halt, and
the fate of the sector would be tied to a single asset class.
Private mortgage insurers have begun to add to their aggregate
credit risk and capital over the past 2 years. But they still have only
half the $17 billion in capital they had at their peak--well short of
the capital that would be required to adequately support the likely
$750 billion to $1 trillion in annual demand for conforming mortgage
loans over the next 5 years.
Another important lesson from history is that change is slow. It
took 25 years and a major change in the regulatory landscape governing
the thrift industry for the GSEs and Government agencies to obtain a
majority of the credit risk in the U.S. mortgage market. Assuming that
an objective is to prevent the mortgage credit market from shrinking
dramatically, it will likely take an equally long time to shift a
material amount of that credit risk back into private hands. That is
because Fannie and Freddie are the central pipes of the current
mortgage funding system. Everyone participating in the origination,
aggregation, securitization, and servicing of mortgage credit risk in
America connects to them. They are the only standard setters with any
market credibility. They own nearly half the credit risk in the
mortgage market. Banks depend on them to move risk off their balance
sheets while capturing profits from the spread between the primary and
secondary market. Private mortgage insurers depend on credit
enhancement requirements that Fannie and Freddie impose on conforming
mortgages. Rate investors rely on their guarantees to funnel investment
funds into the housing market. And, without their participation, the
so-called To-Be-Announced market would shrink considerably (only Ginnie
securities would remain), and rate locks for consumers on conforming
mortgages would disappear. They have done more mortgage modifications
and workouts than any other lender in the system. In short, while it
could be done, winding Fannie and Freddie down will entail a massive
restructuring of the entire infrastructure of the mortgage funding
system. It is not going to be as simple as flipping a light switch or
clearing a field and building a baseball diamond.
A final takeaway is that the now express Government guarantee of
Fannie and Freddie's solvency has left the Government with a rather
complicated liability management problem. The credit investor most at
risk from a botched transition, in which mortgage credit availability
contracts, house prices decline and delinquencies and defaults spike
again, is the Treasury Department. Treasury now effectively stands
behind those $5.5 trillion in liabilities, with virtually no capital in
front of it. \6\ What now looks like a huge asset could quickly turn
into a massive liability for taxpayers if a ``wind down'' throws the
mortgage funding markets into disarray.
---------------------------------------------------------------------------
\6\ Under the terms of the PSPAs, each company currently has a
capital reserve of $3 billion, which is required to decline to $0 in
2018.
---------------------------------------------------------------------------
A Smooth Transition: Transforming the Government Guarantee and Managing
Existing Liabilities
Given that history and the current state of the housing market, I
see one viable path to the end state envisioned in S.1217 that does not
put the economy or taxpayers at risk: restructure, recapitalize, and
privatize the single-family and multifamily guarantee businesses of
Fannie and Freddie. Doing so would solve the thorniest problem in the
transition: ensuring the creation of a durable layer of private capital
ahead of the FMIC backstop.
Note that I am not proposing that you reinstate the precrisis GSE
model. On the contrary, I suggest that you tear it down. Eliminate the
``national mortgage association'' charters dating back to 1934 that
enabled Fannie and Freddie to operate Government-sponsored hedge funds,
using Government-subsidized cost of funds to purchase risky mortgage
assets on a highly levered basis. Wind those portfolios down to levels
necessary to operate a cash window to facilitate the purchase of
mortgage loans from small banks and credit unions and to manage the
workout of troubled mortgages that they guarantee. Fully privatize
their single-family and multifamily businesses separately, and subject
them to strong safety and soundness and capital adequacy regulation.
Create conditions for competition among private sector participants,
while establishing a credible resolution regime for all participants,
including those who may otherwise be deemed too big to fail. Substitute
their existing access and affordability requirements with FMIC
regulatory authority and dedicated Federal vehicles subject to
congressional oversight and funded through fees on all securities
issued in the conforming market. In short, I propose fixing the
problems with these businesses and using what remains to bridge to your
new system.
Below I provide a framework for evaluating any transition plan,
followed by a more detailed explanation of what I propose. Finally, I
evaluate the transition plan in S.1217.
Framework for Transition
Any responsible transition plan for reforming the Government's role
in housing finance should satisfy the following 10 criteria:
1. Avoid a sharp contraction in mortgage credit that would depress
house prices and risk another recession;
2. Avoid saddling Treasury with losses from the remaining
liabilities at Fannie and Freddie to which it is currently
exposed;
3. Establish an adequate and durable layer of private capital ahead
of the FMIC and taxpayers;
4. Preserve liquidity between the old and new systems, including
through the TBA market;
5. Maintain secondary market access for small and community banks
without forcing them to go through large U.S. banks;
6. Prevent concentration of mortgage risk in large U.S. banks;
7. Facilitate competition among private loss providers on FMIC-
reinsured MBS;
8. Avoid consolidating GSE obligations onto the Federal balance
sheet;
9. Contribute to deficit reduction; and
10. Respect the rule of law.
A Realistic Way Forward
Again, I urge you to use the two companies currently under your
control and central to the current mortgage system to transition
smoothly to a new, safer system.
Use them to build capital cushions at each issuer adequate to
protect the Government on the $4.4 trillion in outstanding MBS that the
Government has backstopped through the conservatorships. To build that
capital, FHFA should immediately direct the enterprises to increase
their guarantee fees to market levels, reflecting a return on capital
that other private insurers have to factor into their fees. And
Treasury should suspend its profit sweep and allow the companies to
retain their earnings. Building capital to restore the companies'
solvency is one of the express mandates of the statute authorizing the
creation of the conservatorships, a mandate that has been entirely
ignored during the past 5 years.
Meanwhile, begin building reserves at the new Mortgage Insurance
Fund to protect the Federal Mortgage Insurance Corporation when the new
system is turned on. Ten basis points of guarantee fees on the existing
GSE MBS books should be reallocated to start building the FMIC reserve
fund. In exchange, once the FMIC is up and running, it could reinsure
the outstanding $4.4 trillion of legacy Fannie and Freddie MBS. This
would relieve Treasury of any further obligation to backstop future
losses on those securities through the PSPAs. It would also create
continuity between the securities issued under the old regime and
securities issued in the new regime, ensuring that securities in the
new regime are issued into a deep and liquid market based on the same
reinsurance. Without this bridge, legacy GSE MBS will be orphaned from
the new market, and FMIC securities will suffer in pricing during the
transition.
The personnel, assets, and operations associated with the single-
family and multifamily businesses in each enterprise should be
separated. Each separated business should be licensed to purchase FMIC
reinsurance. And when they attain adequate capitalization levels, those
businesses should be rechartered under State insurance and corporate
charter statutes. Equity in each of the separately capitalized
businesses should be sold to the highest bidder or in an initial public
offering. This separation would promote focused management and reduce
consolidated market power that could squeeze out competition. The
privatized single-family businesses should be permitted to maintain
portfolios of sufficient size to operate a cash window for small banks,
credit unions, and other originators, to manage troubled mortgages
which they have guaranteed, and to conduct basic Treasury operations.
None of these private companies would have special privileges--no
implicit or explicit guarantee of their liabilities that Fannie and
Freddie enjoy today as a result of their ``national mortgage
association'' charters and the express backstop from Treasury of their
solvency in conservatorship. They will not have the ability to issue
Government guaranteed debt to fund expansive on-balance-sheet mortgage
portfolio investments, which Fannie and Freddie had the ability to do
under the Government-sponsorship model. And each of the separated
businesses and new entrants should be subject to a resolution regime
that could facilitate their failure without wrecking mortgage credit
formation or the functioning of the mortgage market generally. In the
case of future failure, shareholders can be wiped out and any mortgage
guarantee infrastructure transferred to new ownership. That regime
should be modeled on the FDIC's orderly liquidation authority for banks
and financial institutions deemed systemically important by the FSOC.
Finally, conduct the ritual slaughter that so many are demanding.
The highly levered investment portfolios each firm ran before the
crisis--in effect, their Government-sponsored hedge funds--should
continue to be wound down under Federal supervision. The public
charters that allowed private shareholders to benefit on the backs of
taxpayers should be terminated. Affordability and access requirements
that have been suspended by the FHFA in conservatorship will end with
the termination of the charters. Fannie and Freddie should ultimately
be placed into receivership and liquidated.
I appreciate the trepidation of recreating two dominant players in
the mortgage market by releasing parts of Fannie and Freddie back into
the wild, however they may have been restructured. Therefore, an
important element of both the transition and the end state is to ensure
that there is a competitive marketplace for conforming mortgage credit
risk.
There are at least six concrete, mutually reinforcing steps that
the FHFA and FMIC can take during the transition to promote a more
competitive market structure while ensuring continuity through the
recapitalization and privatization of the mortgage guarantee
businesses:
1. Common MBS Security: The FMIC could work with private market
participants to establish a common To-Be-Announced market for
securities eligible for FMIC insurance and facilitate options
for multilender pools of eligible single-family mortgages. A
single FMIC security could remove the largest barrier to entry
for new issuers to compete with Fannie and Freddie. The
difference in liquidity between Fannie and Freddie securities
has given Fannie a more competitive position in the MBS market
over Freddie. To combat Fannie's advantage over Freddie and
both companies' advantages over any new issuer, a common TBA
market through which a single FMIC security could be issued
would defeat the competitive advantages the privatized mortgage
guarantee businesses would otherwise have over any new issuer
or guarantor. Establishing a single security and extending it
to legacy GSE MBS could be accomplished in 2 to 3 years.
2. Capital Surcharges: Legislation could require the FMIC to impose
heightened capital and other heightened prudential requirements
on any issuer or bond guarantor that establishes a dominant
market position. This would be similar to the heightened
requirements being debated for large banks and other financial
institutions designated by the Financial Stability Oversight
Council. New entrants with lower capital requirements should
then be able to offer the same first loss protection to
taxpayers on better terms while achieving comparable return on
equity for their investors. S.1217 imposes a hard limit on
market share, which risks discontinuities and inefficiencies
that will translate into unnecessarily higher mortgage pricing.
Adverse capital charges rely on economic incentives to achieve
the same desired outcome, while providing flexibility to the
regulator to preserve efficiencies in the market. If the
consensus is to impose a hard limit on market share, I urge you
to base that limit on some empirical analysis of the benefits
and costs of the market structure which that hard limit will
dictate.
3. Infrastructure Licensing: At least during the transition, the
FMIC could coordinate with the FHFA to allow new MBS guarantors
to use the issuance infrastructure of the enterprises and/or
the common securitization platform in exchange for a fee to
help establish and grow their market share. This could
eliminate or at least mitigate another substantial barrier to
entry, and ensure that there are existing, viable competitors
when the restructured single-family and multifamily businesses
of Fannie and Freddie are privatized.
4. Common Securitization Platform: Meanwhile, efforts to establish a
common securitization platform (CSP) that could serve as a
market utility should continue. However, it is important to
recognize that the platform will take several years to reach
its first milestone--bond administration functions--and as
currently envisioned even in its final phase it will support
only a quarter of the aggregation and issuance chain necessary
to generate FMIC-reinsured MBS. It is more realistic during the
transition to expect that the FMIC will rely on a combination
of issuers (similar to Ginnie Mae) and whatever is available at
the CSP, which will continue to evolve. The FMIC need not wait
on the CSP to reach a particular level of maturity to begin
offering reinsurance. The FMIC should supervise the CSP, but
its operations could be managed by private market participants.
5. Pricing: Raising guarantee fees at the enterprises to take into
account a proper return on capital charge will also have the
collateral benefit of creating a pricing umbrella under which
new private investors and insurers can compete.
6. Equal Access: In all events, the FMIC should offer reinsurance to
all new entrant first-loss providers who meet its capital and
other eligibility requirements.
With respect to minimum capital requirements, it would be a mistake
to set capital requirements at arbitrarily high levels. Doing so will
increase borrowing costs with little to no incremental improvement in
the safety of the system. It will trap capital that could otherwise be
used productively in our economy. And it may impair returns on MBS
guarantee businesses to such a degree that you will deter the
investment necessary to capitalize new entrants. It would also be a
mistake to set capital standards for first-loss providers or for a
subordinated layer of first-loss capital in reform legislation that are
substantially higher or lower than those required of other providers of
mortgage credit, such as banks. To do so would create a type of
regulatory arbitrage between the capital required for mortgages held in
whole loan form on balance sheet versus mortgages held in MBS form with
FMIC reinsurance, an arbitrage that was partially responsible for the
flight of mortgage credit out of the banking system and into the
previously undercapitalized GSE system historically. Forcing mortgage
credit back into the banking system by having substantially higher
capital requirements for FMIC first-loss providers is not only
inconsistent with heightened capital standards coming out of the crisis
for banks going forward. It would also exacerbate the Too Big To Fail
problem with our largest banks if in fact better returns on balance
sheet mortgage lending leads to accelerated asset growth at the largest
banks.
Based on recent experience with the largest mortgage credit shock
in any of our lifetimes, and consistent with what banking regulators
are targeting, a total capital requirement between 4 and 5 percent
would be adequate to protect the FMIC against loss. Such a level would
be consistent with the 10-percent capital standard that S.1217 calls
for, assuming a relatively conservative risk-weighting for conforming
mortgages. I warn against expanding the set of eligible capital beyond
the definition accepted by banking regulators. Regarding the
capitalization level for the MIF, we believe that a reserve balance of
1.5 percent is appropriate and achievable within an acceptable
timeframe without imposing undue costs on the conforming market. This
is greater than the 1.35-percent minimum level that the FDIC is
targeting for the Deposit Insurance Fund, and behind 4 to 5 percent of
first-loss capital, it would have been more than adequate to protect
taxpayers against loss during the recent crisis.
Allowing the single-family and multifamily businesses at the heart
of today's conforming market to be recapitalized and privatized rather
than liquidated would provide continuity in the transition. Meanwhile,
the six tools identified above could be used to ensure that a more
diversified set of first-loss providers could enter the market and
provide a check on the perpetuation of the dangerous duopoly that now
exist under the conservatorships. In addition, when the new system is
switched on and the FMIC begins to offer reinsurance on new conforming
MBS, there would be at least two issuers/guarantors with sufficient
capital and capacity to provide small banks and credit unions an
effective conduit to the secondary market without having to sell their
customer relationships to the countries' largest banks. And those two
issuer/guarantors would be able to provide a counterweight to the large
banks in the mortgage market generally. In turn, mortgage risk would
not become concentrated in the banking sector. Legacy MBS obligations
would follow the privatized companies, not be absorbed onto the Federal
Government's balance sheet. Privatization proceeds would flow to the
Government and generate over $100 billion in deficit reduction. \7\
---------------------------------------------------------------------------
\7\ Since Fannie and Freddie were put into Federal
conservatorships, the Congressional Budget Office (CBO) has treated
them as consolidated entities of the Government that confer a subsidy
(budget cost) to the market because CBO estimates that the price the
companies charge to guarantee mortgages against default is lower than a
private entity would charge. Severing the single-family and multifamily
businesses' special relationship with the Government and divesting the
Government's financial interests in those businesses subsequently in
the market would eliminate the guarantee-fee subsidy and generate cash
proceeds from sales that CBO could score as a substantial negative
subsidy (budget benefit) over the budget window.
---------------------------------------------------------------------------
AIG Precedent
During my recent tenure as the Chief Restructuring Officer at
Treasury, I had primary responsibility for the oversight of the
Government's capital commitments to AIG, which rivaled in size the
amount of capital invested to date in Fannie and Freddie. After a
series of restructurings of that $182 billion commitment, we designed
and implemented a recapitalization plan for AIG that involved (i)
selling off almost half of its insurance businesses to generate
sufficient proceeds to repay its debt to the Federal Reserve and (ii)
exchanging Treasury's $50 billion of preferred stock into 92 percent of
the common equity of the company. Treasury then sold the common stock
into the public markets in a series of secondary offerings in 2011 and
2012, which fully eliminated Government ownership of AIG. In the end,
taxpayers made almost $23 billion on an investment that the OMB
initially projected would result in $50 billion of losses for the
Government.
As with Fannie and Freddie, at the height of the financial crisis
regulators determined that the potential failure of AIG could threaten
the stability of the financial system. Failures in management and
regulation were blamed for allowing the company to reach that point.
However, once taxpayer capital was committed, Treasury and the Federal
Reserve Bank of New York went to work figuring out how to fix the
company and recover the taxpayer support, all while protecting broader
financial stability. Approximately $2 trillion of notional derivatives
at AIG Financial Products were wound down and substantially derisked
before the recapitalization was consummated in early 2011. Operating
businesses were sold as going concerns in value-maximizing transactions
in order to reduce the company's complexity, shrink its balance sheet
and repay its Government support. Financial leverage was reduced to
responsible levels. Management refocused on AIG's core property and
casualty and life insurance businesses, which remain today important
cornerstones of the global insurance landscape and integral parts of
the daily risk management realities for countless policyholders.
The key lesson from this process is that tried and true methods of
corporate reorganization within our existing rule of law can be used to
move forward with reform. Privatizing recapitalized and newly State-
chartered mortgage-guarantee businesses would enable Treasury to
recover its substantial investment in the companies and begin moving
toward a safer housing finance system driven by market incentives, with
private capital first in line for losses. Taxpayers deserve both
outcomes. Once the companies have enough capital to cover their ``first
loss'' insurance exposure, Treasury should outline a detailed
privatization plan in consultation with various stakeholders. The firms
could then be released from Government control and Treasury's ownership
stakes in the restructured entities sold to private investors over
time.
Maximizing the asset value of what the Government controls today
and selling it off over time to ensure the new system has at least two
well-capitalized first-loss providers is a far safer bet in transition
than hoping that new capital sources and avenues for funding mortgages
will arrive in appropriate size to support the market according to an
arbitrary wind down time line. Moreover, it would provide a clear
roadmap to the desired end state that all private market participants
can plan for and invest around.
The Transition Proposed by S.1217
The transition suggested in S.1217 has elements of the above
proposal but ultimately rests on a leap of faith, faith that if we
build a new system, new private capital will come in sufficient size
and speed to allow the new system to meet the future demand for
mortgage funding currently being channeled through Fannie and Freddie.
If it does not, the new system will shut down before it ever has a
chance to get off the ground. As a result, the implementation of the
new system contemplated in S.1217 carries significant execution risk,
depending on capital yet to be raised by players yet to be named. It
also puts taxpayers at risk, not only from potential shocks resulting
from the contemplated break-up of the two largest players in today's
mortgage market, but also from the $4.4 trillion of MBS liabilities
which the bill puts on the Federal Government's balance sheet. Further,
by imposing capital requirements on first-loss providers in the new
system higher than those are imposed on regulated depositary
institutions, S.1217 ignores the lessons of the recent crisis which
suggest that capital will flow to that part of the financial system
where permitted leverage is the greatest and leveraged returns are the
highest.
S.1217 requires that Fannie and Freddie cease doing any new MBS
business and wind their portfolios down to $0 by the FMIC certification
date; that is within 5 years after the bill's enactment. This arbitrary
deadline risks significant dislocation in mortgage credit formation if
private capital is not raised in sufficient amounts to substitute for
Fannie and Freddie on that timetable.
The bill would consolidate $4.4 trillion of GSE MBS liabilities
onto the balance sheet of the Federal Government. This poses two
problems. It will create discontinuities in the trading markets for
those legacy MBS and the new FMIC-reinsured MBS, orphaning the existing
securities and creating significant liquidity constraints on the new
FMIC MBS which will negatively affect mortgage pricing during the
transition. Separately, putting the full faith and credit behind those
contingent liabilities could balloon the Federal debt. This would
complicate an already difficult debate over the sustainability of U.S.
debt and austerity measures.
S.1217 provides that Fannie and Freddie be repurposed in three
ways. First, parts of their businesses are to be sold to a mutual with
small bank members. I am sympathetic to the desire to provide small
banks with access to the secondary mortgage market away from large
banks. Fannie and Freddie do that today. Why create execution risk of
trying to recreate Freddie out of spare parts from both companies? The
bill is also unclear how the mutual would be capitalized, governed, or
established to compete on pricing with large banks. Second, the bill
takes the multifamily businesses ``at no cost'' and puts them inside
the FMIC. Although it may be a placeholder for a better plan, in its
current form, it is neither legal nor workable. Third, the bill would
allow certain businesses to be sold as going concerns. But here too, it
would strip them of assets and purport to sell them without any
capital. A financial institution without capital is like a widget
factory with no assembly line: you can sell the building but you are
not going to get going-concern value for the business that it houses.
As a drafting matter, if this is the path you want to pursue, S.1217
needs to be redrafted to expressly override HERA's provisions that
obligate the conservator or receiver to maximize the sale proceeds of
the assets it seeks to liquidate.
More generally, S.1217 requires 10-percent capital at first-loss
providers. Assuming the conforming market is $5 trillion in 7 to 10
years after a significant portion of the legacy books have rolled over,
and assuming that the bill means for that capital to be defined as a
percentage of total assets, a 10-percent capital requirement means that
$500 billion of first-loss capital needs to be raised to backstop the
conforming MBS market. That is a gargantuan sum relative to the
existing equity capital in the banking and insurance sectors. On top of
this, the bill requires another 2.5 percent, or $125 billion, in the
MIF to be raised within 15 years. Again, the bill provides no direction
of how either such capital level will be reached. But each is critical
to the functioning of the new system. In short, the bill is based on a
leap of faith that such a substantial amount of capital will be raised
by players yet to be named.
Conclusion
By highlighting the important role that the mortgage guarantee
businesses of Fannie and Freddie play in today's mortgage funding
markets, and by taking on the politically charged idea that the safest
and surest path to raising the significant amount of new capital on
which the new system depends and ensuring continuity in mortgage credit
formation during the transition is to recapitalize and privatize those
businesses, I hope that my testimony today will stimulate a frank
conversation about how to handle the transition. I do not envy you the
task ahead: To say that housing finance reform is the most complicated
policy, economic and corporate finance challenge I've seen in my 30
years in the restructuring business would be a gross understatement.
As I noted previously, you have a fundamental choice in meeting the
challenges in transition: Fix what you have and use it to move to a
better-capitalized system of mortgage funding, or destroy what is
working today and make a leap of faith that new capital will be raised
by players to be named. For the sake of the Federal budget, the
stability of the mortgage funding markets and the value of the single-
most important asset for most Americans, I hope that I have persuaded
you that the choice is clear.
______
PREPARED STATEMENT OF JOHN BOVENZI
Partner, Oliver Wyman
November 22, 2013
Introduction
Good morning Chairman Johnson, Ranking Member Crapo, and Members of
the Committee. My name is John Bovenzi, and I am a Partner at Oliver
Wyman, a business unit of Marsh and McLennan Companies (MMC). I would
like to thank you for affording me an opportunity to share my
perspective on housing finance reform.
Much of my perspective on housing finance reform draws on my 28
years of experience at the Federal Deposit Insurance Corporation (FDIC)
and the Resolution Trust Corporation (RTC). I served as Deputy to the
Chairman of the FDIC from 1989 through 1992, the period of time when
the FDIC was responsible for establishing and managing the RTC. From
1992 to 1999, I was Director of the FDIC's Division of Resolutions and
Receiverships and played a key role in merging the RTC into the FDIC.
From 1999 to 2009, I served as Deputy to the Chairman and Chief
Operating Officer at the FDIC.
I believe there is much of value in the FDIC's and the RTC's
experience that can be helpful to the Committee as it determines the
best path forward for housing finance reform.
Overview of the FDIC and the RTC
First, let me briefly provide an overview of the two agencies'
missions and responsibilities.
As you know, the FDIC is an independent agency created by Congress
in the aftermath of the Great Depression. Its mission is to maintain
stability and public confidence in our Nation's financial system, and
it has three primary roles through which it carries out this mission:
(1) by insuring deposits, (2) by examining and supervising financial
institutions for safety and soundness and consumer protection, and (3)
by managing receiverships of failed institutions.
The RTC was a temporary Federal agency established under the
Financial Institutions Reform, Recovery, and Enforcement Act (FIRREA)
in August of 1989. Its mission was to manage the assets and liabilities
of Savings & Loan Institutions (S&Ls) that had been placed into
conservatorship. The agency's goal was to dispose of these assets as
quickly as practicable at maximum value in order to reduce taxpayer
expense.
The RTC resolved 747 S&Ls with assets totaling nearly $400B during
its 6-year existence. While heavily criticized at the time, today, the
RTC is widely viewed as a success story. The total cost to taxpayers
from the failed S&Ls wound down by the RTC was about $80B, a far lower
number than originally projected. After its work was done, the RTC
demonstrated that a Government agency can put itself out of business
effectively and efficiently once its mission has been accomplished.
How the FDIC's and the RTC's Experience Can Be Applied to a New Federal
Mortgage Insurance Corporation
Title 1 of the ``Housing Finance Reform and Taxpayer Protection
Act'' would create a new Federal Mortgage Insurance Corporation (FMIC)
to provide insurance protection for mortgage-backed securities. In the
proposed legislation, the new FMIC is modeled to a large extent after
the FDIC, so observations on the FDIC's structure and experience may be
useful. Also, as a start-up Federal agency, the RTC experience in
establishing itself also should be of value.
Lessons From the RTC's Experience
I'll start with the RTC's experience in creating a new Federal
agency, since I believe that is where we can learn the most. There are
three important points I would like to make at the outset.
First, the creation and subsequent experience of the RTC show that
a new Federal agency can start up and be successful in a relatively
short period of time. However, the RTC experience also shows that the
complexity of the political and operational issues that must be
addressed requires that stakeholders show some degree of patience.
There will be bumps, missteps, and delays along the way.
Second, the leadership of such an organization is critically
important. The Director will need to possess both the skills to work
effectively with a large number of stakeholder groups, as well as the
managerial skills to effectively address the many operational issues
that will be faced by a new agency.
Third, ultimately the employees of the two Government-sponsored
enterprises and the Federal Housing Finance Agency (FHFA) will
determine whether the start-up is a success or a failure. They are the
people who have the ability to effectively transfer critical functions
to a new agency. Their experience and expertise should not be
undervalued or lost if Congress decides to move in the direction of the
proposed legislation.
Let me elaborate on these three points.
Regarding the first point, that time and patience will be
necessary, when the RTC was created, it needed a new governance
structure, new information systems, new staff, and new policies and
procedures. The FDIC was able to provide a great deal of support, but
not nearly enough. Initially, seven hundred employees were transferred
from the FDIC to the RTC. Still, the RTC needed to hire many more
employees and contractors. Both the internal hiring and contractor
procurement processes had to be fair and transparent, with all of the
necessary controls, including background checks. This took time, when
people outside the agency were more focused on seeing immediate
results.
To get off to a faster start, the RTC initially adopted many of the
FDIC's policies and procedures, but these proved to be insufficient.
The RTC's mission and duties were not the same as the FDIC's so most of
those policies and procedures needed to be revised or created from
scratch, generally with sufficient time for public comment and
revisions based on those comments. Information systems were an even
greater challenge. The FDIC's systems were not suited for the RTC's
needs. New systems had to be created, only to be populated with poor
data from insolvent S&Ls. Finally, as a political compromise, the RTC's
governance structure started out with two Boards of Directors, one for
policy and one for operations. The blurred line between these two sets
of responsibilities led to finger pointing and a lack of
accountability.
As a result, the RTC's start-up went slower than what most
observers had hoped for and there was a great deal of frustration with
the RTC. But some perspective is necessary. The RTC successfully
managed its way through those problems and today is widely viewed as a
success story. The agency saved taxpayers money and finished its work
early.
Regarding the second point, that the new agency's leadership will
need to possess both leadership and management skills, I'll simply say
that while this may be obvious to most people, there is an occasional
tendency for the leadership of a Government agency to focus almost
exclusively on high level policy issues and not give sufficient
attention to operational details. A director need not personally focus
on all of the operational details involved in the start-up of a
Government agency, in fact that would be counterproductive. However,
that person must have a clear appreciation of the significance of
internal operations and ensure there are clear delegations of authority
and accountability.
As to the third point, that a new agency's success or failure will
be determined by the employees of the GSEs and the FHFA, there are a
few issues that Congress should consider based on the RTC's experience.
Title 3 of the proposed legislation abolishes the Federal Housing
Finance Agency and transfers its staff, infrastructure, technology and
other resources to the FMIC, but the bill is silent as to the fate of
the employees of Fannie Mae and Freddie Mac. If those two enterprises
are to be shut down some of their operations will have to be
transferred to the FMIC or elsewhere, which means some jobs would
become available in other organizations. But the uncertainty
surrounding how many jobs will be available, on what terms, and who
will get them will create significant complications in ensuring a
smooth transition. The experience and the expertise of Fannie Mae's and
Freddie Mac's employees will be needed to have an effective transition,
so some thought needs to be given as to how that skill and talent can
be preserved.
As a limited life agency, the RTC's employees knew that by doing
their jobs correctly they might be putting themselves out of a job. The
same applied to the FDIC's employees who were responsible for handling
the spike in bank closings. Eventually the economy would recover and
their workload would vanish.
Certain steps were taken to mitigate the harmful effects on the
RTC's and the FDIC's employees. By law, RTC employees also were FDIC
employees, thus they had the same rights as FDIC's employees. This
meant that when the two agencies were merged together the career civil
servants in each agency had equal rights to the remaining jobs. Other
FDIC and RTC employees who had been hired on a temporary basis were in
a more tenuous position. In most cases, they were not likely to have
their contracts renewed once the workload diminished to the point where
staff reductions were necessary.
To the credit of the employees at both agencies, they continued to
do their jobs effectively even though they did not know if, or for how
long, those jobs might last. Indeed after the two agencies were merged
together, the FDIC had to undergo a large and painful downsizing given
the substantial reduction in its workload in going from the crisis to
the postcrisis period.
Throughout that difficult period, constant and clear communication
was critical. Employees needed as much information as possible so they
could better plan for their future. The same would be true here. It
will not be possible to provide jobs for everyone and some attrition
will certainly occur, but should Congress go down this path, the
employees at the GSEs should know that they will be treated fairly and
respectfully.
Lessons From the FDIC's Experience
Regarding lessons that may be learned from the FDIC's experience, I
would like to make some comments on three broad areas: (1) corporate
governance, (2) financial strength, and (3) the supervision of
financial-sector participants.
Corporate Governance
Independence and a system of checks and balances are two important
features of the FDIC's governance structure that have served the agency
well over time. These features also are part of the proposed structure
for the Federal Mortgage Insurance Corporation.
Once certain parameters have been established around an agency's
power and authority, an independent structure allows an agency to carry
out its duties in a responsive manner. Because the market is constantly
changing, an agency needs the ability to continually assess new
information and adapt to those changes.
The FDIC has a five person Board of Directors, each appointed by
the President and confirmed by the Senate. No more than three Board
members may be from the same political party. In my view, this
structure has served the FDIC well over time by providing appropriate
checks and balances on important policy decisions.
The FDIC also has a strong Office of the Inspector General that
provides independent reviews of the agency's operations to prevent
waste, fraud and abuse. The Inspector General is appointed by the
President and reports directly to Congress as well as to the FDIC
Chairman. This too has served the FDIC well as part of an overall
system of checks and balances. The creation of an Office of the
Inspector General is an important safeguard that has been included in
the proposed bill.
Based on my experience it appears that the proposed bill covers the
most important aspects of creating a strong governance structure.
Financial Strength
Much has been learned over the past 30 years about what is required
to maintain a deposit insurance fund strong enough to not have to rely
on taxpayer support during a financial crisis. During the 1980s and
early 1990s nearly 3,000 insured depository institutions became
insolvent and were closed. As a result, the Bank Insurance Fund became
insolvent. While the FDIC did not have to rely on taxpayer support, it
did have to substantially raise bank deposit insurance premiums during
the crisis period, when banks could least afford to pay more. This had
several adverse effects, not the least of which was exacerbating the
credit crunch that existed at that time.
Because of that experience, Congress relaxed some of the controls
on the FDIC's ability to manage the size of its deposit insurance fund,
but it was not enough. During the 2008 financial crisis the FDIC's
deposit insurance fund again had insufficient funds. The agency did not
have to rely on taxpayer support, but once more it did have to charge
banks substantially higher premiums when banks could least afford to
pay them. As a result Congress further relaxed the controls on the
FDIC's ability to assess high enough premiums over the course of the
business cycle. Now the agency has the authority it needs to build the
deposit insurance fund to high enough levels that it can withstand a
crisis period. This new authority allows the FDIC to charge higher
premiums during the healthy part of the economic cycle, so it will not
be forced to further dampen credit availability during an economic
downturn.
The proposed bill includes targets for the size of the FMIC's
Mortgage Insurance Fund. After the first 5 years of its existence the
FMIC is expected to have charged participating institutions fees
sufficient to create a fund that is 1.25 percent of all outstanding
covered securities. After 10 years the ratio is targeted to be 2.5
percent of all outstanding covered securities. It is difficult to know
what size fund is needed to protect taxpayers against losses. For many
years the FDIC's statutory target ratio of the deposit insurance fund
to insured deposits was 1.25 percent. Prior to the 1980's the rationale
was that while arbitrary, history had shown that the 1.25 percent
target ratio worked. In the aftermath of two financial crises it was
clear the 1.25 percent ratio did not work. Since then the FDIC analyzed
what fund size would have been necessary to keep the deposit insurance
fund from becoming insolvent. That review led the agency to raise its
target ratio to 2 percent.
The important point is that the FMIC will need sufficient
flexibility and authority to manage the size of the Mortgage Insurance
Fund based on continuous analysis so it can protect taxpayers against
losses during economic downturns.
Supervision of Financial-Sector Participants
The FDIC sets standards for bank behavior. The agency has the
authority to set entry standards for groups that seek to obtain bank
charters and the authority to remove deposit insurance protection for
banks that aren't meeting those standards. In between, the FDIC has a
wide range of formal and informal enforcement actions it can employ to
force banks to meet its supervisory standards without removing a bank's
deposit insurance coverage. These authorities include issuing formal
cease-and-desist orders, civil money penalties, and agreeing to
informal memoranda of understanding. The FDIC also has examination
authority to ensure banks are in compliance with FDIC supervisory
standards and to determine whether enforcement actions are necessary.
Also, to help ensure that supervisory actions are taken in a timely
manner, the FDIC is subject to Prompt Corrective Action requirements,
which mandate that certain supervisory actions be taken as bank capital
levels drop below prescribed levels. In their entirety these powers are
an important part of safeguarding the financial system and protecting
the deposit insurance fund.
The proposed bill would grant some, but not all, of these
authorities to the FMIC. The new agency would have authority to
determine entry standards for mortgage servicers, issuers, and
guarantors. Those standards track the FDIC standards in many respects,
since they include a review of the financial history of the applicant,
its capital adequacy, the character of management, and the risk posed
to the insurance fund. The bill also empowers the FMIC to issue civil
money penalties and revoke its approval if a participating institution
does not continue to meet its standards. However, the bill does not
grant the FMIC examination authority, nor does it allow for a full
range of enforcement actions.
Based on my experience, it would be worth considering whether the
FMIC should be granted broader supervisory and enforcement authorities
beyond controlling entry and exit into and out of the program and the
ability to issue civil money penalties. The FDIC has rarely used its
power to revoke deposit insurance coverage, finding it to be a
cumbersome process compared to its other enforcement alternatives such
informal memoranda of understanding and formal cease-and-desist orders.
The FMIC likely would have the same experience. Other more practical
enforcement tools may be more effective in helping the FMIC accomplish
its objectives. Also, consideration should be given to giving the FMIC
examination authority. While off-site monitoring can be used to help
monitor bank behavior, over time the FDIC has found there is no
substitute for the direct interaction with bank management that occurs
during the examination process.
How the RTC's Experience Can Be Applied to the Proposed Wind Down of
Fannie and Freddie
Title 5 of the ``Housing Finance Reform and Taxpayer Protection
Act'' requires that Fannie Mae and Freddie Mac be wound down and phased
out of business over a 5-year period. The RTC had a similar requirement
in its original charter. By statute the agency, which was created in
1989, had to be wound down and merged into the FDIC by year-end 1996.
It accomplished that objective a year earlier than originally planned.
Given those similarities there may be some valuable lessons based on
the RTC's experience should Congress determine that it wants to wind
down Fannie and Freddie's operations. We already covered issues related
to the treatment of the GSE's employees so I won't repeat those
concerns here, rather I'll talk briefly about governance issues and
sales processes.
Corporate Governance
Many of the same governance principles that apply to the creation
of a new agency also apply to the wind down of an existing agency or
agencies. Strong oversight is critical because taxpayer dollars and
important public policy objectives are at stake.
The RTC was governed by a Board of Directors with additional
oversight provided by Congress, an Office of the Inspector General, and
the General Accounting Office (GAO), among others. Such checks and
balances, while introducing some degree of inefficiency, are well worth
the costs in order to ensure there are strong oversight, effective
internal controls, and fair processes.
Sales Processes
According to the proposed legislation, Fannie Mae and Freddie Mac
should have no more than $552.5B in real estate related assets
(mortgage loans and mortgage-backed securities) by year-end 2013. The
bill requires that these assets be reduced by at least 15 percent a
year over a 5-year period. Any remaining assets are to be put into
receivership after that point. This is not significantly different than
what the RTC was charged with accomplishing. Most of the $400B in
assets from the insolvent S&Ls that the RTC was responsible for also
were real estate related assets.
The RTC experimented with a large variety of sales processes for
the different types of assets it managed. It learned much through trial
and error, but a few key principles emerged to help guide the agency.
First, virtually all sales were subject to an inclusive, open and
transparent competitive bidding process. The RTC did not engage in
negotiated sales with individual buyers for pools of assets, despite
the desire for such by many potential buyers. The agency recognized
that open competition would maximize value and that it also reduced the
possibilities for fraud or abuse. Given that a number of the insolvent
S&Ls that were costing taxpayer money had committed fraud and abuse, it
was that much more important that the Government cleanup be beyond
reproach.
Second, the RTC partnered with the private sector in the
disposition of many its assets. For pools of assets that required
particular expertise, the RTC found it best to sell a portion of the
pool to private-sector investors with the required expertise and retain
partial ownership of the assets. Such partnerships allowed the RTC, and
hence taxpayers, to benefit from the added value the right management
could bring to those assets as well as from any appreciation in assets
value over time due to an improving economy.
Such public/private-sector partnerships in managing and disposing
of assets aren't without their challenges. Often both sides have a
healthy degree of mistrust for one another. The private sector often
views the Government as an unreliable and slow business partner, while
the Government often sees the private sector as overly focused on its
financial returns and under appreciative of the types of processes and
controls that must be put into place whenever taxpayer money is at
stake.
These differences can be overcome by clarifying up front what the
expectations are for each business partner. The Government needs to
understand that financial incentives for the private sector maximize
value for taxpayers. Private sector asset managers need to understand
that they have to comply with certain processes and oversight that they
may view as inefficient and time consuming, but that are necessary to
show the public that the overall process is being managed in a way that
treats people fairly and shows them that their money is not being
wasted.
During the most recent financial crisis the FDIC effectively used
public/private equity partnerships (and the closely related loss-
sharing agreements it entered into with the acquirers of insolvent
banks) to manage many of the assets it was responsible for as receiver
for failed banks. It found that these partnerships greatly enhanced
asset values and returns to failed-bank creditors, including the FDIC's
deposit insurance fund.
Such agreements between the public and private sector, while
valuable in certain situations, are not necessarily the preferable
sales technique in all situations. Some assets can be sold outright and
still maximize value, in part by eliminating ongoing commitments and
administrative burdens on the part of the Government. Each asset
category and situation should be evaluated on its own merits to
determine the best strategy.
As the Committee and the Congress deliberate further on this
important issue, I and my colleagues at Oliver Wyman are ready to
collaborate with you to offer our experience and expertise on this key
public policy matter.
______
PREPARED STATEMENT OF MARK ZANDI
Chief Economist and Cofounder, Moody's Economy.com
November 22, 2013
Much of the debate over the future of the Nation's housing finance
system has focused on the system's end state--whether housing finance
should be privatized, retain some form of Government backstop, or even
remain effectively nationalized as it is today. No matter which goal is
chosen, however, reform will not succeed without an effective
transition. A clearly articulated plan for getting from here to there
is vital; otherwise policy makers will be appropriately reluctant to
move down the reform path.
For the purposes of this testimony, it is assumed that the future
housing finance system will be a hybrid system, much like that proposed
in recent legislation introduced by Senators Corker and Warner, S.1217.
That is, private capital will be responsible for losses related to
mortgage defaults, but in times of financial crisis, when private
capital is insufficient to absorb those losses, the Government will
step in. Mortgage borrowers who benefit from the Government backstop
will pay a fee to compensate the Government for potential losses.
While there are advantages and disadvantages to any housing finance
system, a hybrid system is the most likely to be implemented. Such a
system will preserve the long-term fixed-rate mortgage as a mainstay of
U.S. housing, and it will ensure that affordable mortgage loans are
available to most middle-income Americans through good and bad times.
Taxpayers will backstop the system, but it will be designed so that
lenders and borrowers bear the ultimate cost.
A hybrid system will require substantial new private capital.
Currently, little private capital is involved in making mortgage loans;
the Federal Government acts as the Nation's principal mortgage
originator via Fannie Mae, Freddie Mac, and the Federal Housing
Administration. How much private capital will be needed depends on many
factors, but assuming the new system's requirements are consistent with
those applied to the Nation's largest banks, as much as $175 billion in
today's dollars might have to be raised.
For context, this amounts to more than the equity raised in the 10
largest initial public offerings in U.S. history combined, including
those for the insurer AIG and the credit-card giant Visa. Such a large
amount will not be easy to raise quickly. Any viable transition plan
must therefore clearly determine where the private capital will come
from and at what cost.
The transition plan must also spell out the fate of Fannie Mae and
Freddie Mac. While few wish to return to the old system, which was
dominated by these thinly capitalized, too-big-to-fail behemoths, the
consensus stops there. Some insist that Fannie and Freddie be
completely dismantled, while others propose using their current profits
to recapitalize and ultimately reprivatize them.
Dismantling the two institutions would risk disrupting the flow of
mortgage credit, which, for all their faults, Fannie and Freddie have
continued to provide efficiently through the Great Recession and
subsequent recovery. On the other hand, recapitalizing and privatizing
the institutions could leave them in control of U.S. housing finance.
It is unclear who could compete with them; without such competition,
the future system will eventually resemble the old one. A dominant
duopoly will allow the entities to overcharge for their services and
will become the taxpayers' problem if they blunder again.
A host of smaller but still critical technical and legal issues
must also be resolved in the transition to a new system. Moving Fannie
and Freddie from their current conservatorship status to receivership
to new ownership will be complicated. Their mortgage securities must be
managed by whoever succeeds them at least as efficiently as they are
doing. Shifting oversight authority from the Federal Housing Finance
Agency, Fannie's and Freddie's current regulator, to the overseer of
the future system will also involve many steps. And ensuring that small
lenders have access to the Government backstop for the mortgages they
originate will not be easy.
Some initiatives necessary to reshape the housing finance system
are already under way and should be nurtured. The FHFA is developing a
common securitization platform, which will be important no matter the
system's final form. The platform should support greater transparency,
which in turn will promote better credit risk management and lower
future mortgage defaults, more liquidity, better access for small
lenders and increased competition.
The FHFA is also requiring Fannie and Freddie to share more risk
with private investors, including private mortgage insurers and
investors. This should provide information and experience necessary for
the risk-sharing envisaged under most housing finance reform proposals.
The transition to the future housing finance system will require
legislation and take years to implement, but cannot begin unless there
is a clearly laid-out road to reform. With such a road map, it is
plausible that housing finance reform could become law soon. It is
exciting to think that the new housing finance system could conceivably
be in place at the start of the next decade.
A more detailed description of the road to housing reform is
provided in the paper as an appendix to this testimony, ``The Road to
Reform'', Mark Zandi and Cristian deRitis, Moody's Analytics white
paper, September 2013.
Transition Objectives
The transition from the current, largely nationalized housing
finance system to the future hybrid system must protect the economic
recovery. Government support to the housing finance system cannot be
withdrawn too quickly without undermining the housing recovery, which
is vital to the broader economic recovery. Mortgage credit conditions
are still very tight: Lenders remember the massive losses suffered
during the housing crash and are uncertain about a number of regulatory
issues. Prematurely withdrawing Government support would exacerbate
this problem.
Taxpayers should be made financially whole during the transition.
The Government's support to Fannie and Freddie should be repaid, along
with the cost of backstopping the rest of the financial system when
Fannie and Freddie failed, and the costs associated with setting up a
new financial system. Taxpayers should also receive a return on their
financial support commensurate with the risks they have taken.
Private capital standing in front of the Government's guarantee
must be adequate to absorb mortgage losses resulting from all but the
most severe financial crises and economic downturns. This is necessary
to protect the Government against losses and avoid future bailouts. A
substantial amount of private capital, from varied sources, will be
needed by the future housing finance system.
The transition to the new housing finance system must reduce the
system's reliance on large and complex financial institutions such as
Fannie Mae and Freddie Mac. The housing finance system's design must
ensure that institutions in the system can fail without catastrophic
economic consequences.
Access to affordable owner-occupied and rental housing must be
maintained through the transition. This has become even more important
in the wake of the Great Depression and the significant destruction of
homeowners' equity in the Great Recession, ongoing financial pressure
on low-income households, and changing demographics.
Legacy Fannie and Freddie Securities
Investors in legacy Fannie and Freddie MBS and debt securities must
be protected. The Federal Government now guarantees existing MBS and
bond obligations of Fannie and Freddie through agreements between the
Treasury Department and the two firms. This must continue through the
transition period. Not doing so would undermine investors' faith in the
U.S., raising borrowing costs and exacerbating the Nation's fiscal
problems. This is a legacy of the old system, and while the new system
should avoid re-creating this obligation, we cannot retroactively
change expectations without damaging the Nation's credibility in global
credit markets.
S.1217 addresses this issue by providing an explicit guarantee on
the ``payments of all amounts which may be required to be paid under
any obligation'' of the Government-sponsored enterprises. Legacy GSE
MBS would thus be backed by the Government's full faith and credit,
much like a GNMA MBS. This support applies to mortgage-backed
securities that have been issued by Fannie and Freddie in the years
leading up to the ``certification date,'' when the GSEs stop issuing
MBS.
Legacy GSE MBS must be made fungible with Government-backed MBS in
the new housing finance system. In S.1217 this would be MBS backed by a
Government regulator--call it the Federal Mortgage Insurance Corp. One
possibility is to establish a resecuritization process whereby
investors in legacy MBS are able to, but not required to, convert them
into FMIC MBS. These new MBS would be deliverable into the new to-be-
announced market, and would simply require a new CUSIP number and a
matching-up of payment delays.
Investors should be able to exchange legacy GSE MBS for the new
FMIC MBS indefinitely and without cost. When the existing stock of
legacy securities outstanding becomes small enough so that the costs of
maintaining the exchange program exceed its benefits, some type of
``clean up'' call may be appropriate.
Common Security
Smoothing the transition to these new securities would be the
development of a common Government-guaranteed security prior to the
full implementation of the new housing finance system. This would
improve liquidity in the TBA market and result in lower mortgage rates.
A common security would also lower entry barriers to the guarantor
market, as no guarantor would have an advantage because of the
liquidity of the securities they back.
This is a problem in the current housing finance system, as Freddie
Mac securities are much less liquid than Fannie Mae securities. Fannie
and Freddie split the MBS market 60-40, but on a typical day the
trading volume of Fannie MBS is 10 times greater than that of Freddie
MBS. To compensate, Freddie is forced to charge a lower guarantee fee
than Fannie. In the second quarter of 2013, Fannie's average G-fee was
57 basis points, compared with Freddie's 51 basis points.
There are some modest differences between the securities--Freddie
pays investors more quickly than Fannie and its securities prepay a bit
more quickly--but the key difference is their liquidity. This liquidity
difference makes the mortgage market less efficient and less
competitive, and leads to higher costs for mortgage borrowers and
taxpayers.
A potential near-term fix to this problem would be to make Fannie
and Freddie securities fungible, creating a common TBA security. That
would require a change to the good-delivery guidelines for TBA, to
allow the delivery of either Fannie or Freddie securities into the same
contract. The securities themselves would not change; their separate
TBA markets would simply be merged. Both securities would still be
separately identifiable and tradable, only the TBA trades would be
merged. Not only would this interim step improve liquidity, it would
demonstrate investor interest in a truly common security that would be
an important feature of the future hybrid housing finance system.
The Future of Fannie and Freddie
A critical question in the transition to a future housing finance
system is what to do with Fannie Mae and Freddie Mac. For all that is
wrong with the current system, Fannie and Freddie are doing an
effective job buying conforming mortgages, bundling them into MBS with
a Government guarantee, and selling them to global investors. The
mortgage market is not working as well as it should, but it is working.
Whatever is done with Fannie and Freddie must not disrupt this flow of
mortgage credit, for the sake of the housing and economic recoveries.
Arguably the most straightforward approach, with the least amount
of near-term risk, would be to recapitalize and reprivatize Fannie and
Freddie. Both are currently profitable, as a result of improving
mortgage credit conditions and their higher guarantee fees. The two
agencies' profits are flowing to the U.S. Treasury, rapidly repaying
the $188 billion Fannie and Freddie received from taxpayers in order to
stay in business. The GSEs are on track to repay the Treasury's
investment by the end of this year.
After that, their profits could be used to build the capital
necessary for them to become private guarantors in the future finance
system. Once appropriately capitalized, they would be reprivatized,
with the Government selling them to private investors to maximize the
return to taxpayers.
There is a considerable downside to this approach, however: The
future housing finance system could again be dominated by Fannie and
Freddie or their successors. The system could encourage competition,
for example, by establishing a new common securitization platform run
as a Government utility that produces a single Government-backed
security. The reincarnated Fannie and Freddie would also likely be
classified as systemically important financial institutions, or SIFIs,
and thus face stiffer capital and liquidity requirements. This would
raise their cost of capital vis-a-vis newer entrants, further
supporting competition.
But the two giant firms would still have considerable advantages of
size and scale, important legacy relationships, and entrenched software
and systems. Most likely this approach would create a hybrid system
dominated by a duopoly, firms with significant power over the mortgage
and housing markets that would be much too big to fail. The arrangement
would be uncomfortably similar to the dysfunctional system that
prevailed prior to the Great Recession.
An alternative approach would be to simply put Fannie and Freddie
into receivership and liquidate their assets. Guarantors in the hybrid
system would be largely new entities, begun by those purchasing
Fannie's and Freddie's assets. There is significant risk in this
approach, as there would be no assurance that the new guarantors would
be able to continue the institutions' activities, at least not in a
timely way. The chance of a disruption in the flow of mortgage credit
would be uncomfortably high.
A better approach would be for the Government to put Fannie and
Freddie into receivership, and to strip them of their key assets. They
would then be rechartered as new private guarantors, able to license
back these assets from the Government receiver. Their operations would
not be disrupted, ensuring that the mortgage market functioned smoothly
through the transition. But to level the competitive playing field, any
other new guarantors could also license the same key assets from the
receiver. This would facilitate easy entry into the guarantor market
and thus encourage competition.
The current Senior Preferred Stock Purchase Agreement between the
U.S. Treasury and Fannie and Freddie would need to be restructured to
permit the redemption of the Treasury's senior preferred shares and the
cancelation of its warrant in the firms. The restructured SPSPA would
determine the appropriate compensation taxpayers require from Fannie
and Freddie for their financial support.
Fannie and Freddie would be put into receivership, and their
operating assets and liabilities moved into limited life regulated
entities, or LLREs, allowing them to maintain their operations
independent of the resolution process. This is similar to the procedure
envisaged in Dodd-Frank for failing SIFIs. The assets of the LLREs
would then be sold or licensed back to Fannie's and Freddie's successor
firms, which would be chartered as independent guarantors, and to the
new competitor guarantors.
Fannie's and Freddie's $4.5 trillion legacy guaranty book would not
be included in the assets transferred from the Government receiver to
the LLREs. More private capital would be needed to support the legacy
books than could be raised in a reasonable period, ensuring that the
new housing finance system would never get going. The receiver would
engage the new guarantors to manage the loans in the legacy books,
providing a steady source of revenue.
Sources of Private Capital
A substantial amount of private capital will thus be necessary to
support the future housing finance system. Over time, some will come
through the guarantors' retained earnings. This will not help in the
early years, but under conservative assumptions, retained earnings
could eventually provide as much as one-third of the guarantors'
capital requirements.
The equity market is another potential source for early capital.
Some financial institutions have held big initial public offerings in
the recent past: AIG, Visa, and Bank of America each raised close to
$20 billion in equity. The guarantors in the future housing finance
system should see returns on equity similar to those of the money-
center banks and life insurers, or about 10 percent. This would be
consistent with a valuation of 100 percent of tangible book value and a
price-earnings multiple of 10. The guarantors' return on equity would
be less than the 15-percent ROE that private mortgage insurers have
historically received, although this appears to have declined to near
13 percent in the current low interest-rate environment. It is
encouraging that many private mortgage insurers have been able to raise
significant equity capital in recent months.
But it is hard to see the equity market producing all the remaining
capital needed by the guarantors. Equity investors will be rightly
nervous about the new system, and will question the guarantors'
earnings prospects in a highly regulated and mature market. The
guarantors' earnings may also be relatively volatile, fluctuating with
the housing and business cycles, and their market share will shift
against the nonguaranteed part of the mortgage finance system. And of
course there is the reputational risk associated with playing a pivotal
role in the provision of mortgage credit.
Equity investors in the new guarantors would likely include those
currently taking equity stakes in private mortgage insurers.
Shareholders in the Nation's largest PMI companies include mutual funds
such as Fidelity and the Vanguard Group, pension funds such as TIAA-
CREF, asset management firms such as Goldman Sachs Asset Management and
State Street Global Advisors, hedge funds such as Paulson & Co. and
Citadel, and diversified financial institutions such as BlackRock. A
wide range of global reinsurers are also providing capital relief to
the PMI companies and would likely be interested in taking stakes in
the new guarantors.
Yet, even if the guarantors can raise the amount of equity
envisaged from public markets, a capital shortfall will remain. This
would be temporarily filled by the Nation's large mortgage originators
through a seller-financing arrangement. In the hybrid system assumed
here, originators would not be permitted to own guarantors, but there
would be an exception while the system is being established. In that
period, originators would be required to temporarily take equity in the
guarantors in partial payment for the Government-guaranteed mortgages
they sell. The equity received by the originators as payment would be
valued at 100 percent of tangible book value.
The success of requiring large originators to temporarily hold
equity in the guarantors hinges on several factors. Most importantly,
the originators, which include the Nation's largest banks, would need
to have excess capital. Capital ratios in the banking system are at a
record high and rising: According to the Federal Deposit Insurance
Corp., the Tier 1 capital ratio for all banks is above 9 percent and
climbing. Banks are also making record profits, and although their
recent profitability is temporarily supported by improving credit
quality and the resulting release of loan loss reserves, they should
have plenty of excess capital given their long-term earnings power and
more limited growth opportunities postregulatory reform.
While bank originators may object to this arrangement, they also
have a strong incentive to ensure that guarantors in the new hybrid
system are well-capitalized. Originators will prefer a well-functioning
housing finance system, with a Government backstop and a TBA market, to
alternatives that require them to hold many more mortgages on their
balance sheets. However, since the banks' investments in the guarantors
would have pedestrian returns, and since a 100-percent risk-weighting
would be capital-intensive, bank originators would be expected to sell
their stakes in the guarantors as soon as their capital is no longer
needed. There would also be a reasonable divesture period, in case they
are unexpectedly slow to sell their shares.
Critical to this arrangement's success is that even with their
equity stakes, the large bank originators should have no control over
the guarantors. Otherwise, small lenders would be appropriately nervous
about their ability to compete. Large originators would receive
nonvoting or B-shares as payment from the guarantors. This is similar
to the arrangement Visa set up with its bank members when it designed
its IPO. Once the B-shares were sold to nonoriginator investors, they
would become voting A-shares.
Common Securitization Platform
A well-functioning common securitization platform is an important
requirement for a successful transition to a new housing finance
system. All non-Ginnie Mae, Government-guaranteed securities should use
a common securitization platform. Although not required, nonguaranteed
securities could use the same platform.
The common securitization platform would produce a more liquid
market, facilitate loan modifications in future downturns, and give
issuers operating flexibility at a low cost. It would also allow for a
robust TBA market. Such a platform is also important for lowering
barriers of entry into the future mortgage guarantor market, allowing
for more competition and reducing too-big-to fail risks.
The securitization facility would leverage current efforts by the
FHFA to develop a single platform for Fannie Mae and Freddie Mac
securities. For a fee, the securitization facility would provide a
range of services, including mortgage loan note tracking, master
servicing, data collection and validation to improve transparency and
integrity, and bond administration.
Mortgage loans included in securities that use the common
securitization facility would be covered by a uniform pooling and
servicing agreement and uniform servicing standards that encourage
prudent underwriting and align investor and borrower interests. This
would encourage the adoption of similar standards for other mortgages.
The common securitization platform would permit multiple
originators to sell mortgages into single securities with access to the
Government guarantee. In return, the originators would receive pro rata
shares of the security. Pooling requirements would be largely the same
as for typical single-originator securities, and they would be good for
delivery into the TBA market. Originators could thus easily convert
securities to cash before the securities were created, an especially
important feature for smaller originators.
Transition Contingencies
It is important to recognize the possibility that the transition
process may not go as smoothly as planned. The transition involves
complex changes to the legal and operational framework at the center of
housing finance. It also involves the development of new guarantors and
securities and new oversight responsibilities over a wide range of
institutions and activities. Given all these moving parts, it is
plausible to think that things will not come together as quickly as
hoped.
As such, any legislation to reform the housing finance reform
system should allow for some flexibility in the timing of the
transition process. In S.1217 the transition process must be completed
within 5 years. There should be some flexibility in this deadline, as
the FMIC needs the ability to speed or slow the process if it
jeopardizes the housing market and capital markets more broadly.
Suppose, for example, that there is a major financial crisis in year
five of the transition. The FMIC should thus have the authority to
reduce or even eliminate the private capital first-loss requirement in
cases of significant financial market disruption. It is thus also
critical that the GSEs be able to continue their business operations
until the new system is fully operational.
Conclusions
Since the Government took over Fannie Mae and Freddie Mac during
the financial collapse 5 years ago, effectively nationalizing the
Nation's housing finance system, nothing meaningful has changed. The
Government still makes nearly nine of every 10 U.S. mortgage loans.
This is bad for both taxpayers and homebuyers.
Taxpayers are on the hook for potential losses on the hundreds of
billions of dollars in mortgages that Fannie and Freddie insure each
year. This is not necessary: Private investors are willing to take on
much of this risk and, with some safeguards, are capable of doing it.
The longer Fannie and Freddie stay in Government hands, the more
lawmakers will be tempted to use them for purposes unrelated to
housing. This has already happened. Last year's payroll tax holiday was
partially paid for by raising the premiums Fannie and Freddie charge
homebuyers for providing insurance. Mortgage borrowers will be paying
extra as a result over the next decade.
The housing market's revival has allowed Fannie and Freddie to
again turn large profits, amounting to tens of billions of dollars each
year. Policy makers may begin to rely on these profits to fund
Government spending, making it especially hard to let Fannie and
Freddie go.
Policy makers may also eventually be tempted to make Fannie and
Freddie lend to people who really cannot afford mortgages. This is
partly how the two institutions got into financial trouble during the
housing bubble--they took on more risk than they should have to meet
their housing-affordability goals. Helping disadvantaged households
become homeowners is laudable, but experience shows that politically
driven help can be abused.
The bigger problem now is the limbo status of Fannie and Freddie,
which fosters indecision at the two institutions and by their
regulator, the FHFA. Lenders who do business with Fannie and Freddie
are unsure of the rules, and are thus extra cautious, keeping credit
overly tight for potential homebuyers. This is evident in the average
credit scores of borrowers through Fannie and Freddie, which today are
in the top third of all of credit scores.
Lawmakers recognize the current situation's dangers and have
introduced legislation to reform the Nation's housing finance system.
Yet these legislative efforts lack a clear plan for getting from the
current housing finance system to the future one. The transition cannot
be bungled: The Nation's economic recovery depends on housing, which in
turn depends on the flow of mortgage credit. The $10 trillion U.S.
mortgage market is also critically important to the entire global
financial system.
Yet while the transition will be complicated and rife with risk, it
is eminently doable.
The Federal Government has unwound much of its extraordinary
intervention in the economy prompted by the Great Recession. Fiscal
stimulus has been replaced by fiscal austerity. The Trouble Asset
Relief Program bailout fund will soon be history. The Federal Reserve
is planning to begin normalizing monetary policy. That leaves Fannie
and Freddie and the Nation's housing finance system as the largest
piece of unfinished business. It is time to finish it.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
______
PREPARED STATEMENT OF DAVID MIN
Assistant Professor of Law, University of California, Irvine School of
Law
November 22, 2013
Chairman Johnson, Ranking Member Crapo, and Members of the
Committee, my name is David Min and I am an Assistant Professor at the
University of California, Irvine School of Law, where I teach and
research in the area of banking and capital markets regulation. Before
coming into academia, I spent over a decade working in financial
markets law and policy, both in private practice and in the Federal
Government. Most recently, I served as the Associate Director for
Financial Markets Policy for the Center for American Progress, where I
was responsible for managing the activities of the Mortgage Finance
Working Group organized by CAP. I am here, however, in my individual
capacity, and not as a representative of either CAP or the Mortgage
Finance Working Group.
For the purposes of my testimony, I will assume that the system of
housing finance that we transition into will be some variation of
S.1217, the bill proposed by Senators Bob Corker (R-TN) and Mark Warner
(D-VA). \1\ The Corker-Warner bill envisions a so-called ``hybrid''
system, in which the Federal Government provides explicit and priced
reinsurance on mortgage-backed securities insured by approved bond
guarantors, in a model based loosely on the deposit insurance model of
the Federal Deposit Insurance Corporation. As I have noted elsewhere,
the Federal Government has provided, in one way or another, a
catastrophic backstop on most residential mortgage funding since the
New Deal's banking and housing reforms, and this role has been
inextricably linked to a number of key policy objectives, including
financial stability, broad and constant liquidity, and the wide
availability of the 30-year fixed-rate mortgage that has become the
hallmark of U.S. housing finance. \2\
---------------------------------------------------------------------------
\1\ Housing Finance Reform and Taxpayer Protection Act of 2013,
S.1217, 113th Cong. (2013) (hereinafter ``Corker-Warner'' or the
``Corker-Warner bill'').
\2\ See, e.g., David Min, ``How Government Guarantees in Housing
Finance Promote Stability'', 50 HARV. J. LEG. 437 (2013); David Min, et
al., ``The Future of U.S. Housing Finance: Five Points of View'', 17 J.
STRUCTURED FIN. 36 (2011). Before Agency securitization evolved to
dominate U.S. mortgage markets, federally insured depository
institutions were the primary source of residential mortgage funding.
In both types of financing, the Federal Government holds the
catastrophic tail risk. Since the 1940s, this Government backing has
existed for the vast majority of U.S. home loans, typically more than
70 percent. Id.
---------------------------------------------------------------------------
The transition being contemplated would be the largest such
undertaking in history, and one that, to the best of my knowledge, has
no close precedents. Fannie and Freddie currently hold slightly more
than $5 trillion in mortgage-related assets. \3\ Since the sudden and
steep decline in private mortgage finance that occurred in 2008, the
two enterprises have been responsible for more than 60 percent of the
new mortgage originations, about $1.7 trillion each year, an amount
equivalent to slightly more than 10 percent of our Nation's annual
gross domestic product. The Federal Government has some experience in
resolving failed institutions--recently, the Government's interactions
with AIG and General Motors come to mind, and before that, we had the
experience of the Resolution Trust Company in resolving hundreds of
failed thrifts. But I can think of no instance in which we have tried
to simultaneously resolve large failed institutions and transition
their core economic functions into a newly created set of institutions,
certainly not on the scale imagined by Corker-Warner.
---------------------------------------------------------------------------
\3\ According to their Form 10-K filings, at the end of 2012,
Fannie held $3.064 trillion in mortgage related assets, and Freddie
held $2.005 trillion.
---------------------------------------------------------------------------
The experience of the past decade has shown us that the current
housing finance model, which relies predominantly on the mortgage-
backed securities issued by the Government-sponsored enterprises (GSEs)
\4\ Fannie Mae and Freddie Mac, well serves a number of critically
important policy goals, including offering a broadly available and
affordable 30-year fixed-rate mortgage product, meeting the credit
needs of underserved populations and product types such as rural areas
and multifamily housing, and providing countercyclical liquidity when
other sources of housing finance have dried up. Unfortunately, this
experience also illustrates a number of serious flaws with the GSE
model. Because these enterprises were publicly backed, with private
shareholders, they continuously sought to maximize profits by
increasing their risk-taking, creating a ``heads I win, tails you
lose'' dynamic between their shareholders and taxpayers. Moreover,
protections against taxpayer exposure were clearly insufficient in this
framework. In my view, the Corker-Warner bill is a good starting point
for thinking about how to keep the parts of the current system that
worked, while eliminating the parts that proved problematic.
---------------------------------------------------------------------------
\4\ Following the conservatorship of Fannie and Freddie, many have
ceased calling these firms ``Government-sponsored,'' since they are
effectively seen as part of the Federal Government. Technically,
however the conservatorship was structured in a way that kept the
enterprises private and lacking an explicit Government guarantee, so
the term ``Government-sponsored enterprise'' or ``GSE'' is still
accurate.
---------------------------------------------------------------------------
This same balance needs to be reflected in how we think about
transitioning to the future housing finance system outlined in Corker-
Warner. Thus, while safety and soundness and taxpayer protection are
obviously important policy goal, I believe the most important priority
in structuring the transition should be to ensure that there continues
to be sufficient liquidity across all market segments. Given the
complexity and size of the GSEs' current operations, this is a
tremendously complex and multilayered task. Moreover, the stakes could
not be higher, as major hiccups would have devastating effects on an
already-stagnant economy and financial system.
This is particularly true for financing affordable multifamily
housing, which is a primary source of rental housing in this country.
In the aftermath of the housing crisis, policy makers have generally
sought to reduce the emphasis on home ownership and shrink the Federal
Government's role in housing finance, which is reflected in Corker-
Warner. But achieving these objectives will naturally mean that
affordable rental housing will be much more important, both from a
social and economic perspective. Thus, one of the most important
elements of the transition will be ensuring that we maintain sufficient
liquidity for rental housing, such as multifamily housing.
The guiding principle for legislators and regulators who are
structuring our housing finance transition must first and foremost be,
``Do no harm.'' Avoiding the disruption of mortgage liquidity, either
systemwide or in individual market segments, should be a paramount
concern during this period. A failure to adhere to this principle would
be catastrophic for the housing markets and the broader economy.
Assessing the Corker-Warner Transition Plan
The Corker-Warner bill contemplates a transition period of no more
than 5 years following its enactment, during which time Fannie and
Freddie would be phased out and the infrastructure for the new system,
including the Federal Mortgage Insurance Corporation (FMIC) at the
heart of this framework, is established. \5\ Upon enactment, Corker-
Warner would eliminate the affordable housing goals currently in place
for Fannie and Freddie, \6\ and begin to gradually reduce the high cost
area loan limits, which currently stand at 150 percent of the
conforming loan limit (now set at $417,000) to 115 percent of the
conforming loan limit within 5 years. \7\ The mortgage assets held in
Fannie and Freddie's investment portfolios would be reduced by 15
percent each year until the FMIC is certified as being operational; at
the end of that year, their remaining assets would be used to wind down
the enterprises and help cover the costs of any remaining legacy
guarantees. \8\
---------------------------------------------------------------------------
\5\ Corker-Warner 501-506.
\6\ Corker-Warner 506.
\7\ Corker-Warner 504.
\8\ Corker-Warner 505.
---------------------------------------------------------------------------
Upon FMIC certification, an event which must occur within 5 years
of enactment, the charters for Fannie and Freddie are repealed and
these firms barred from conducting any new business. \9\ At this time,
outstanding ``legacy'' debt obligations (bonds and mortgage-backed
securities) issued by the GSEs would be explicitly guaranteed with the
full faith and credit of the United States. \10\
---------------------------------------------------------------------------
\9\ Corker-Warner 501.
\10\ Corker-Warner 501.
---------------------------------------------------------------------------
At a high level, Corker-Warner provides a thoughtful template for
long-term mortgage finance reform. But transitioning to the new system
that Corker-Warner creates will be a long and difficult process. The
Corker-Warner bill provides some broad guidance and mandates on the
question of transition, but many issues remain unresolved and need to
be addressed before we move on. I discuss some of these below.
Developing a Common Securitization Architecture
Central to the Corker-Warner framework is the development of a new
infrastructure for issuing securities with a common Government
guarantee. Currently, Fannie and Freddie each have their own
securitization architectures, but creating a common securitization
platform (CSP) is a prerequisite to opening the new system up to a
multitude of issuers. \11\ Creating the CSP is also important for
creating a single security, which many see as a precondition for a
successful transition towards the new system, because of the
differences in liquidity and pricing that are likely to develop in a
system that has more than two issuers.
---------------------------------------------------------------------------
\11\ See, Edward J. DeMarco, Acting Director, Federal Housing
Finance Agency, ``The Conservatorships of Fannie Mae and Freddie Mac'',
remarks before the National Association of Federal Credit Unions
Congressional Caucus, Washington, DC (Sept. 13, 2012).
---------------------------------------------------------------------------
Even in the current environment, with two virtually identical
issuers enjoying the same Government guarantee, investors clearly
prefer Fannie obligations over Freddie obligations, and as a result,
Fannies trade in deeper and more liquid markets and enjoy better
pricing. \12\ As of June 2012, the spread between 30-year 4.5-percent
Fannie MBS and Freddie MBS was about 48 cents. \13\ These spreads are
certain to widen with the entry of additional issuers, unless a common
security is created. Thus, moving towards a single security seems to be
an important part of any transition towards the new system. A single
security should also improve liquidity in the important ``To Be
Announced'' (TBA) market, the forward market that is responsible for
more than 90 percent of the trading volume in agency MBS (and which
allows borrowers to ``lock in'' their rates). \14\
---------------------------------------------------------------------------
\12\ See, American Securitization Forum, ``Discussion of a
Proposed Single Agency Security'' 1-2, ASF White Paper Series, July 2,
2012.
\13\ Id. at 3.
\14\ See, generally, James Vickery and Joshua Wright, ``TBA
Trading and Liquidity in the Agency MBS Market'', Federal Reserve Bank
of NY Economic Policy Review, May 2013.
---------------------------------------------------------------------------
In theory, establishing a CSP and single security should not be
overly difficult. After all, Ginnie Mae securities have a large number
of issuers and a shared Government guarantee, and they effectively
trade as a single security. Several white papers have been written
describing best practices in creating a single security, and they
generally share the same recommendations. \15\ We need a common
platform, such as the CSP, and standardization of terms and contracts,
including loan delivery and pooling requirements, remittance
requirements, underwriting guidelines, servicing standards, and
disclosure policies.
---------------------------------------------------------------------------
\15\ See, e.g., id.; Mortgage Bankers Association, ``Ensuring
Liquidity Through a Common, Fungible GSE Security'', in Key Steps on
the Road to GSE Reform, Sept. 11, 2013; American Securitization Forum,
``Discussion of a Proposed Single Agency Security'', supra n. 12;
Richard Johns, Executive Director, ``Structured Finance Industry Group,
Essential Elements of Housing Finance Reform'', Testimony Before the
U.S. Senate Committee on Banking, Housing, and Urban Affairs (Sept. 12,
2013).
---------------------------------------------------------------------------
But in reality, moving towards a common platform and single
security may be quite difficult. The technical challenges alone are
likely to be very challenging. Fannie and Freddie each created and
perfected their securitization infrastructures over many years.
Integrating these systems together into an open securitization platform
that can be utilized by any approved issuer will be a painstaking task.
But as recent history teaches us, developing a complex technology
infrastructure can be much more difficult than originally anticipated.
It took Wells Fargo 3 years to integrate its data systems with those of
Wachovia, following its acquisition of the Charlotte-based bank holding
company. Bank of America did not finish integrating its data systems
with those of Merrill Lynch until September of this year, nearly 5
years after Merrill was acquired. \16\
---------------------------------------------------------------------------
\16\ See, ``Bank of America Merrill Lynch Completes Transaction
Data Repository'', Press Release, Sept. 16, 2013, available at http://
newsroom.bankofamerica.com/press-releases/commercial-and-middle-market-
banking/bank-america-merrill-lynch-completes-transaction.
---------------------------------------------------------------------------
Indeed, it is worth noting that the Federal Housing Finance
Agency's progress towards creating the CSP is proceeding exceedingly
slowly. At this point, more than 18 months after the common
securitization platform was first publicly announced by the FHFA, \17\
the only public announced progress towards creating this CSP has been
the filing of a certificate of formation for a limited liability
company and the signing of a lease for office space. \18\ The slow pace
of CSP development does not bode well for the relatively aggressive
time line envisioned by Corker-Warner, which calls for the FMIC to be
certified as operational within 5 years.
---------------------------------------------------------------------------
\17\ See, Federal Housing Finance Agency, ``A Progress Report on
the Common Securitization Infrastructure'' (Apr. 30, 2013), available
at http://www.fhfa.gov/webfiles/25144/
WhitePaperProgressReport43013.pdf. The idea for creating a common
securitization platform was first announced in February 2012 by FHFA,
which followed up with a detailed proposal that was released to the
public in October 2012.
\18\ See, Federal Housing Finance Agency, ``FHFA Announces
Significant Steps in Organization of Joint Venture To Establish Common
Securitization Platform'', News Release (Oct. 7, 2013).
---------------------------------------------------------------------------
Achieving Liquidity for the New MBS
Another important transition issue that must be considered is how
to best scale up liquidity for the new securities guaranteed by the
FMIC in the Corker-Warner framework. Corker-Warner essentially
envisions an on/off progression, in which the GSEs are shut down at the
same time that the FMIC opens for business. Once the FMIC is certified
as operational, the GSEs lose their charters and are barred from
conducting any new business. The concern with this approach, of course,
is that when this handoff occurs, there is a lack of demand for the new
securities, due to any number of factors (such as a reluctance by
investors to be early adopters). If that were to occur, the sudden drop
in liquidity could be quite problematic for the housing markets.
Responsibly Reducing High Cost Area Loan Limits
Corker-Warner calls for scaling back the size of the Federal
Government's footprint, with so-called ``high cost'' loan limits being,
at least in the transition phase, the primary focus of this reduction.
But it is not clear how much private nonguaranteed liquidity is
currently available to fill the vacuum that will be created. Some have
suggested that depository institutions should play a greater role in
financing home loans, reprising the role they once played in
originating and holding mortgages to term (as opposed to simply
originating mortgages with the intent to sell these to secondary market
actors, which has increasingly displaced originate-to-hold lending).
But the fact is that bank deposits are simply not a large enough source
of funding at this time to replace much of the activity that Fannie and
Freddie currently do. As Figure 1 illustrates, the total amount of U.S.
bank deposits is barely sufficient to meet U.S. housing finance needs.
Moreover, as my fellow witness Jim Millstein has noted, we have been
experiencing a net decline in real estate loans held by commercial
banks, suggesting that traditional bank balance sheets are an unlikely
source of increased housing finance in the near future. \19\
---------------------------------------------------------------------------
\19\ Jim Millstein, Chairman and Chief Executive Officer,
Millstein and Co., ``A Blueprint for Finance Reform in America'',
Remarks before the Woodrow Wilson International Center for Scholars,
May 22, 2012.
---------------------------------------------------------------------------
Similarly, it is improbable that private-label securitization,
which accounted for so much volume during the housing boom years of
2002-07, will be able to replace much of the vacuum left by a reduced
Government footprint in the near future. Since 2008, private-label
securitization of mortgages has essentially been nonexistent. Figure 2
lists the underwriting characteristics for all of the private-label
mortgage securitization deals that have taken place since the 2008
financial crisis. The credit characteristics are extremely high, and
the volume is still very low. As Georgetown Law professor Adam Levitin
has described, the PLS market is currently a ``market for lemons'' and
is likely to stay that way for some time, until investors regain
confidence in the integrity of the highly informationally asymmetric
PLS process. \20\ At the same time, the implementation of the
``Qualified Mortgage'' (QM) standard, which provides safe harbor for
mortgage originators, and the ``Qualified Residential Mortgage'' (QRM)
standard, which provides an exemption from the risk retention
requirements of Dodd-Frank, are likely to have some impact on the
availability of private, nonguaranteed mortgage finance, but it is too
early to tell what this impact might be.
---------------------------------------------------------------------------
\20\ Adam J. Levitin, Professor of Law, Georgetown University Law
Center, ``Housing Finance Reform: Fundamentals of a Functioning
Private-Label Mortgage Backed Securities Market'', Testimony before the
U.S. Senate Committee on Banking, Housing, and Urban Affairs (Oct. 1,
2013).
---------------------------------------------------------------------------
In the near term, aggressively lowering loan limits may lead to a
gap in the availability of mortgages in high-cost areas, which could
adversely affect the housing markets in those regions.
Ensuring the Continued Flow of Mortgage Finance for Underserved Market
Segments
Historically, the GSEs have played an important role in providing
mortgage credit to underserved market segments, such as rural housing
and housing for lower-income households. They have played a
particularly important role in providing financing for affordable
multifamily rental housing, with roughly two-thirds of the multifamily
units they finance being affordable to households earning less than 80
percent of area median income. \21\ GSE financing for affordable
multifamily rental housing has come from both their guarantee
activities and purchases for their investment portfolios, and has been
motivated at least in part by their affordable housing goals. The GSEs'
footprint in multifamily housing finance has generally been much more
variable than their single family market share, shrinking down to about
25 percent of the market when market conditions are good and increasing
to fill the vacuum when market conditions deteriorate (70 percent at
the height of the crisis). \22\
---------------------------------------------------------------------------
\21\ See, Ethan Handelman, Vice President for Policy and Advocacy,
National Housing Conference, ``Housing Finance Reform: Essential
Elements To Provide Affordable Options for Housing'', Testimony Before
the U.S. Senate Committee on Banking, Housing, and Urban Affairs (Nov.
7, 2013). As Mr. Handelman outlines, one key reason why the GSEs play
such a dominant role in affordable multifamily is that, unlike purely
private sources of mortgage capital (such as insurance funds or pension
funds), they offer long-term products, which are critical for many of
the specific needs of developing and maintaining affordable multifamily
deals.
\22\ See, Shekar Narasimhan, Managing Partner, Beekman Advisors,
Inc., ``Housing Finance Reform: Essential Elements of the Multifamily
Housing Finance System'', Testimony before the U.S. Senate Committee on
Banking, Housing, and Urban Affairs (Oct. 9, 2013).
---------------------------------------------------------------------------
Corker-Warner calls for an aggressive reduction in the GSE
portfolios over a period of 5 years, and for the elimination of the
affordable housing goals. Replacing these mechanisms would be a
separate set of entities that have been collectively described as the
``Market Access Fund,'' which would be funded by a small levy, between
5 to 10 basis points of outstanding mortgage guarantees. \23\ The
Market Access Fund would, as my fellow witness Mark Zandi has
described, attempt to provide both direct subsidies and explicit credit
enhancement to foster greater access to securitization in underserved
market segments, particularly in affordable multifamily housing. \24\
The existing multifamily guarantee business would be transferred to the
FMIC, as Corker-Warner currently stands, \25\ although as Dr. Zandi has
noted, this is likely a placeholder, as it is difficult to imagine a
regulator running a business. \26\
---------------------------------------------------------------------------
\23\ This includes the National Housing Trust Fund and Capital
Magnet Fund that were established but not implemented under the Housing
and Economic Recovery Act of 2008, as well as other funds meant to
provide targeted credit support meant to facilitate securitization of
niche products. See, Corker-Warner, 401-403; Ellen Seidman, Phillip
Swagel, Sarah Wartell, and Mark Zandi, ``A Pragmatic Plan for Housing
Finance Reform'', June 19, 2013, available at http://www.urban.org/
uploadedpdf/412845-Housing-Finance-Reform.pdf.
\24\ Mark Zandi, Chief Economist and Cofounder, Moody's Analytics,
``Essential Elements of Housing Finance Reform'', Testimony before the
U.S. Senate Committee on Banking, Housing, and Urban Affairs (Sept. 12,
2013).
\25\ Corker-Warner 601.
\26\ Mark Zandi and Cristian deRitis, ``Evaluating Corker-
Warner'', Moody's Analytics Report (Sept. 2013).
---------------------------------------------------------------------------
One concern with the proposed transition is that it may wind down
key aspects of the current system that have provided financing for
these underserved segments--the investment portfolios and the
affordable housing goals--without having fully established the Market
Access Fund. Given economic and demographic trends, we have already
seen a sharp increase in the demand for affordable rental housing. As
policy makers seek to deemphasize home ownership and reduce the Federal
Government's footprint in single-family housing finance, we should
expect to see this demand increase. Thus, it is imperative that we
avoid leaving vacuums in the availability of mortgage credit, which
would be devastating for renters, rural homeowners, lower income
families, and many others who are vulnerable and struggling to make
ends meet during a period of economic stagnation.
Attracting Sufficient and Appropriately Priced Capital Into the New
System
Another important concern is bringing in sufficient capital to fund
the new private MBS issuers that are central to the hybrid system
envisioned by Corker-Warner. While some have raised concerns about the
availability of private capital willing to serve as equity in this new
system, \27\ I am optimistic that there is a large pool of capital to
draw upon, and I think that last week's proposal from Fairholme Capital
Management gives us some evidence of that. The question, of course, is
on what terms this capital is available.
---------------------------------------------------------------------------
\27\ Mark Zandi and Cristian deRitis, ``The Road to Reform'' 5-7,
Moody's Analytics Report (Sept. 2013).
---------------------------------------------------------------------------
The core economics of this business are strong, as Figure 3
indicates. Since their conservatorship, the GSEs have been steadily
improving their performance, as the impaired loans guaranteed during
the 2003 to 2007 period are written off and the new books of business
from 2008 and onward grow to become a larger proportion of their
balance sheets. That being said, there are a number of variables that
will affect how much capital is available and on what terms. These
include capital requirements, \28\ expected market size and share
(which is affected by, among other things, loan limits and barriers to
entry), and the pricing of the Government guarantee.
---------------------------------------------------------------------------
\28\ While it is generally thought that equity investors prefer
higher leverage, so as to maximize their potential returns, there is
some evidence that lower leverage (such as created by higher capital
requirements) leads to higher returns over time. See, generally,
Malcolm Baker and Jeffrey Wergler, ``Do Strict Capital Requirements
Raise the Cost of Capital? Bank Regulation and the Low Risk Anomaly'',
NBER Working Paper No. 19018 (2013).
---------------------------------------------------------------------------
Corker-Warner requires that private capital representing no less
than 10 percent of the guaranteed MBS be placed in a first loss
position, \29\ although as former Treasury official Phillip Swagel has
noted, Corker-Warner contemplates that this private capital may be
tranched, which would lower its costs. \30\ Some have argued that this
capital level is too high, and will thus lead to both a dearth of
private capital and a sharp increase in mortgage rates. For example,
Laurie Goodman and Jun Zhu conduct an empirical analysis and conclude
that 4-5 percent capital would have covered all of the GSEs' losses
coming out of the 2007-08 mortgage crisis. \31\ Taking into account the
proposed Mortgage Insurance Fund, which is required to reach a reserve
level of 2.5 percent of outstanding principal balance within 10 years
(1.25 percent within 5 years), Corker-Warner effectively contemplates a
12.5-percent buffer against taxpayer loss, \32\ on top of any
improvements to mortgage loss rates that may accrue as a result of QM
and QRM. This compares to the current system, in which Fannie and
Freddie had minimum capital requirements of 2.5 percent of assets plus
0.45 percent of adjusted off-balance sheet obligations (including
guaranteed mortgage-backed securities). \33\
---------------------------------------------------------------------------
\29\ Corker-Warner 202.
\30\ Phillip Swagel, ``Housing Finance Reform: Essential Elements
of a Government Guarantee for Mortgage-Backed Securities'', Testimony
before the U.S. Senate Committee on Banking, Housing, and Urban Affairs
(Oct. 31, 2013).
\31\ Laurie S. Goodman and Jun Zhu, ``The GSE Reform Debate: How
Much Capital Is Enough?'', Urban Institute, Oct. 24, 2013.
\32\ There are two other layers of protection against taxpayer
loss that Corker-Warner increases. Private mortgage insurance currently
covers a little less than 15 percent of mortgages securitized by the
GSEs and takes a first loss position between 6 percent and 35 percent.
See, Fannie Mae 2013 Third Quarter Credit Supplement; Freddie Mac
October 2013 Investor Presentation. Under Corker-Warner, PMI is left in
place, covering loans with an LTV over 80, but its coverage amounts are
increased. Corker-Warner 2. Homeowner equity is also likely to be
increased, by virtue of the 5-percent downpayment requirement created
in Corker-Warner. Corker-Warner 2. Fannie Mae's current average loan-
to-value ratio on all single-family loans they have guaranteed is 67.1
percent. See, Fannie Mae 2013 Third Quarter Credit Supplement. Freddie
Mac's average loan-to-value ratio on its single-family conventional
guaranty book of business is 73 percent. See, Freddie Mac October 2013
Investor Presentation.
\33\ See, Office of Federal Housing Enterprise Oversight,
``Mortgage Market Note: Fannie Mae and Freddie Mac Capital'', July 17,
2008. Fannie and Freddie were also subject to risk-based capital
requirements, based on a 10-year stress test model. Id. The two
enterprises were actually subject to surplus capital requirements due
to alleged accounting improprieties and other issues, from 2004 until
the time of their conservatorship. See, Federal Housing Finance Agency,
``Capital Prior to Conservatorship'', available at http://www.fhfa.gov/
Default.aspx?Page=146. Following their conservatorship, FHFA has
suspended capital classifications and announcements. See, Federal
Housing Finance Agency, ``Capital Under Conservatorship'', available at
http://www.fhfa.gov/Default.aspx?Page=78.
---------------------------------------------------------------------------
Maintaining Sources of Countercyclical Liquidity
A longer-term transition goal should be to preserve sources of
countercyclical liquidity. The unfortunate fact about private bank
capital is that it is highly procyclical, chasing profits during credit
booms and becoming overly risk averse during credit contractions. As a
result, the Government is typically the only game in town when it comes
to countercyclical liquidity. We need only look to our current mortgage
markets to see this phenomenon on display. Since the credit contraction
began in 2007, Agency securitization has been responsible for virtually
all housing finance, accounting for over 90 percent of residential
mortgage originations. Without this countercyclical liquidity, it is
certain that the housing bust would have been far worse, with extremely
negative effects on the broader economy. Before this most recent
housing crisis, the last great housing crisis we had occurred in the
1930s, when we did not have in place any sources of countercyclical
liquidity. The result was a 50 percent national delinquency rate and a
10-percent foreclosure rate. \34\
---------------------------------------------------------------------------
\34\ See, David Min, ``How Government Guarantees in Housing
Finance Promote Stability'', supra n. 2.
---------------------------------------------------------------------------
Of course, in winding down Fannie and Freddie, Corker-Warner
eliminates the two largest sources of countercyclical mortgage
liquidity. Corker-Warner recognizes the need for such a function,
however, and thus, in the presence of ``unusual and exigent''
circumstances, allows for the issuance of securities that do not have
10-percent private capital in a first loss position. In the event that
we have another mortgage crisis, this exigency clause may not be
sufficient to meet the liquidity needs of the market. Based on
observations of the current experience, in which Fannie and Freddie
have been responsible for roughly two-thirds of all mortgage
originations, it may be the case that greater emergency powers are
appropriate.
Recommendations for Transition
Given the issues with the transition contemplated by Corker-Warner,
what should we do next? I lay out some recommendations below.
Delegate More Responsibility to Regulators and Remove Arbitrary
Timetables
In a number of different ways, Corker-Warner looks to micromanage
the transition process. The GSEs are given specific time lines for
lowering their loan limits and winding down their portfolios, and the
FMIC is provided with very specific capital requirements as well as a
specific schedule for implementing the CSP and ending the activities of
the GSEs. But as the above analysis demonstrates, these are highly
technical issues that would benefit greatly from dedicated expertise
and data. Is 10 percent the right level of capital? Will winding down
the GSE portfolios by 15 percent a year have an adverse effect on the
housing markets (particularly underserved and vulnerable areas)? Will
private sources of mortgage finance come into the market if the GSEs
lower their loan limits each year? What if we shut down the GSEs, the
CSP opens for business and liquidity is lacking? All of these are
questions that are best answered by regulators making decisions based
on data analysis, rather than by legislation making choices based on
assumptions that may or may not turn out to be correct.
Regulators should be given greater discretion and encouraged to
respond to developments on the ground, with broad principles guiding
their actions rather than detailed and specific rules. Timetables
should not be dictated ex ante, but rather should be developed in
response to data-driven analysis. It may be useful to compare the roles
of the FHFA and FMIC with those of Federal banking regulators, who
enjoy very broad discretion and expansive powers to promulgate and
enforce regulations based on their regulatory goals. Given the
complexity of the transition we are anticipating, giving regulators
more flexibility in their actions and timetable would seem a prudent
and more effective course of action.
Thus, I believe that Corker-Warner should not attempt to create
specific capital requirements, or create specific timetables, but
should instead substitute high level regulatory targets and mandates,
while leaving the specifics to the regulators. Greater flexibility and
delegation to regulators are preferable for managing a transition of
this scale and scope.
Phase in the Transition in Parts, Not All at Once
The current transition plan contemplated by Corker-Warner
effectively calls for flipping on a switch (certification of the FMIC),
at which point the GSEs will turn off and the FMIC and CSP will turn
on. But as recent events may highlight, unanticipated problems may
arise, particularly with any transition as complex as moving a trillion
dollars in mortgage origination financing over from one platform to
another. Flipping the switch may lead us to discover that the lights
are not working, or only working in parts of the building.
Rather, I believe a preferable approach would be to adopt a
piecemeal approach to transition, turning over small (but increasingly
greater) parts of the mortgage markets to the new infrastructure. For
example, rather than preparing the CSP architecture to handle the
mortgages financed by GSE securitization all at once, we could first
start with a dedicated subset of mortgages, such as 15-year fixed-rate
mortgages, or ``high cost'' conforming mortgages. Such an approach
would have a myriad of benefits. First, it would allow regulators to
test the new system in a meaningful way, and develop data that can help
them perfect the new infrastructure. Second, it would help build
investor liquidity in the new MBS being produced. Instead of requiring
investors to all become early adopters, a piecemeal approach to
transition would build volume over time in specific product segments.
Third, to the extent that there were problems with liquidity in the
transition, such an approach would leave in place the GSEs to pick up
any slack that might be needed.
Under this approach, transition could proceed based on meeting
specified liquidity benchmarks, and not on a preordained timeframe.
Such an approach might actually proceed more quickly than the
transition called for by Corker-Warner, since this would allow for
earlier partial certifications of FMIC, rather than the all-or-nothing
approach currently specified in the legislation. This would also wind
down the GSEs in an orderly fashion by removing increasingly larger
parts of their business and transferring them to the new system.
Convert Legacy Securities Into New FMIC-Backed MBS
One way to build liquidity in the new system is to offer all
holders of legacy securities the option of converting their securities
into the new, explicitly guaranteed MBS created under the new housing
finance regime. \35\ Assuming transition was phased in as described in
the previous section, each class of securities could be converted at
the time that an equivalent product was offered by the FMIC. This
approach would build immediate volume into the new architecture, which
would improve liquidity and lower prices.
---------------------------------------------------------------------------
\35\ Compelling investors to convert their securities could raise
contractual and other issues, and thus is likely to raise more problems
than it solves. However, if there were cost-effective ways to encourage
these investors to convert, those might be worth considering as well.
---------------------------------------------------------------------------
Preapprove the New MBS for Use in TBA Market and as Collateral
Another relatively simple step that could improve liquidity for the
new security is to ensure, ahead of time, that it will be accepted for
delivery into the TBA market. As I discussed previously, the TBA market
is an enormous futures market that is responsible for over 90 percent
of the trading in Agency MBS, and thus is a critical source of
liquidity. On their face, these new securities should have no problem
fitting into the TBA market, as they are Government-backed (an
important de facto requirement for TBA trading) and seem to possess all
of the other predicate characteristics. \36\ As part of the transition
process, regulators should open up discussions with the Securities
Industry and Financial Markets Association (SIFMA), which sets
standards for the TBA market, and take any steps necessary to ensure
that the new MBS are accepted for TBA trading.
---------------------------------------------------------------------------
\36\ See, generally, Vickery and Wright, ``TBA Trading and
Liquidity in the Agency MBS Market'', supra n. 14.
---------------------------------------------------------------------------
Similarly, regulators should seek to preapprove the new MBS as
collateral in the various markets and transactions in which Agency MBS
is accepted as collateral, such as the Fed's discount window lending,
the OTC derivatives market (standards set by the International Swaps
and Derivatives Association), and repo markets (standards set by
SIFMA). Given that the new MBS carry an explicit Government guarantee
(typically the most important requirement), this should not be a
difficult task, but simply setting expectations ahead of time may have
a large beneficial impact on liquidity.
Give a Running Start to Institutions Focused on Underserved Markets
As I described earlier, the transition process may have particular
issues in maintaining liquidity in underserved market segments. As the
GSEs are unwound, it may be the case that the new infrastructure is not
yet set up well enough to fill the void. To help alleviate this
problem, it makes sense to give a head start to the new institutions
tasked with serving these markets.
Thus, it may make sense to start funding the Market Access Fund
immediately, taking these funds out of the G-fee that is currently
being levied by the GSEs. Since 2008, Fannie and Freddie have financed
roughly $2.5 trillion in new mortgage originations, and they are
charging 50 basis points on these. Taking even a small amount out of
this could go a long way in getting the MAF up and running, so that it
is able to take on a greater share of the underserved market once
transition is underway.
Similarly, it would be useful to immediately fund and activate MBS
guarantors with a specific focus on affordable housing finance, with an
eye towards immediately becoming part of the new Corker-Warner
architecture. Some of you may be familiar with the plan put forth by
Raphael Bostic, Shekar Narasimhan, and Mark Willis, which proposes the
immediate spin-off of the multifamily securitization assets and
business of Fannie and Freddie into a new joint subsidiary. \37\ This
new multifamily entity would, for a fee, piggyback off of the
guarantees of Fannie and Freddie, until such time as the GSEs were
eliminated and the FMIC was operational. At that point, the new
multifamily entity would convert into an issuer in the new Corker-
Warner system.
---------------------------------------------------------------------------
\37\ Raphael Bostic, Shekar Narasimhan, and Mark Willis,
``Multifamily Finance Reform'', June 24, 2013, available at http://
www.beekmanadvisors.com/presentations/
Multifamily%20Finance%20Reform_Moving%20to%20a%20Solution-2013-06-24-
DRAFT%20FOR%20DISCUSSION.pdf.
---------------------------------------------------------------------------
Whether or not the Bostic/Narasimhan/Willis plan is adopted, it
provides an interesting template for thinking about how to serve
affordable housing finance needs. As this plan illustrates, it is
critical to get things up and running immediately, to give these
entities a running start and thus help to ensure that liquidity in
these underserved markets will not be lacking when transition occurs.
Providing Expanded Emergency Powers to FMIC To Deal With Housing Crises
Finally, we should think about the importance of tools that can
allow our housing finance system to respond to emergency situations. In
addition to the current provisions articulated in Corker-Warner, which
allow for FMIC to guarantee MBS that don't meet the 10-percent private
capital requirement, other powers should be provided, which allow FMIC
to effectively provide countercyclical liquidity in the event of
another crisis, like the one we are currently emerging from. At a bare
minimum, this should include the ability to raise loan limits and lower
its insurance fees. Policy makers may want to consider the feasibility
of emergency powers that would allow for expanded eligibility for FMIC
guarantees, or the ability to (temporarily) directly invest in mortgage
assets.
Conclusion
The topic of transitioning into the new housing finance system of
the future is a critically important but highly complex one. The
Corker-Warner transition plan provides us with a good starting point to
start thinking about some of these difficult issues, and I appreciate
the opportunity to discuss this topic with you today. Thank you again
for holding this hearing, and for the opportunity to testify. I look
forward to your questions.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
RESPONSES TO WRITTEN QUESTIONS OF
CHAIRMAN JOHNSON FROM JAMES MILLSTEIN
Q.1. Should a new system be designed to issue a single security
regardless of how many guarantors or issuers there may be? If
so, how should the issuance of a single security be structured
during transition? How should the issuance of new guaranteed
MBS be introduced while ensuring that legacy GSE securities do
not become orphaned?
A.1. Yes, the new system should provide for a single conforming
To-Be-Announced (TBA) eligible pass-through single-family
mortgage-backed security (MBS) reinsured by the Federal
Government through the Federal Mortgage Insurance Corporation
(FMIC), regardless of the number of guarantors or issuers. Such
a ``single security'' is necessary to avoid liquidity
advantages any one issuer or guarantor could obtain, which
would likely, over time, pose an insurmountable barrier to
entry to competitors for the majority of investor demand for
conforming MBS.
The first step toward a single security during transition
could be to issue a common GSE MBS. The Federal Housing Finance
Agency (FHFA) could work with the Securities Industry and
Financial Markets Association to amend the good-delivery
guidelines to provide that such a security would be TBA
eligible. The Securities and Exchange Commission may need to
grant an additional exemption from Regulation AB to ensure that
is the case. The enterprises would also need to standardize
requirements and processes to be able to issue a common GSE
security. We believe that all of this could be accomplished
within 2 years if the enterprises are given a clear directive.
Note that depending on the rate at which the FMIC is stood
up, the enterprises could, during conservatorship, instead seek
to issue a single FMIC MBS. Elements of a Common Securitization
Platform (CSP) could be used for such issuance, but it would
not be necessary to wait for such a platform to be developed.
In the interim, Fannie Mae and Freddie Mac could alter their
requirements and processes to do it, similar to issuers
participating in the Ginnie Mae II program, which provides for
a common security.
To avoid orphaning the market for legacy GSE securities,
creating liquidity premia for new issuance, and turning away
current investors, the FMIC could reinsure legacy GSE
securities in exchange for a fee to appropriately compensate it
for the risk of doing so. Alternatives, such as an exchange of
old securities for new securities, carry significantly higher
execution risk and would be difficult if not impossible to
coordinate given the widespread investor base.
Although the single security would account for most of the
new conforming market, the FMIC should also work with
participants in the new conforming market to facilitate
issuance of alternative securities to satisfy different
investment demands. For example, the Government-sponsored
enterprises (GSEs) accounted for over $200 billion of issuance
of collateralized mortgage obligations (CMOs) in 2012. That
represented approximately 20 percent of total agency issuance
and a significant portion of funding for U.S. residential
mortgages. The FMIC should attempt to retain this volume of
demand by allowing multiple structured forms of conforming MBS
to issue. And the FMIC could prevent the liquidity barriers
among issuers of alternative forms of conforming MBS by
employing multiple issuer pools, similar to the Ginnie Mae II
securities backed by fixed-rate and adjustable-rate mortgages.
Q.2. Please discuss whether or not licensing of Fannie Mae and
Freddie Mac's assets can be an important tool in managing their
wind down in a manner that maximizes value and minimizes
disruption to the housing finance market. Are there other
authorities that could be helpful to a regulator in managing
their wind down?
A.2. Developing an issuance infrastructure to compete with
Fannie Mae and Freddie Mac is another barrier to entry for
private competitors. Building such an infrastructure from
scratch will require substantial time and expense for any new
entrant. While a Common Securitization Platform might
eventually provide new entrants with a utility for
securitization and bond administration functions, and while the
FMIC might also promulgate national standards for origination
and servicing, the build-out of a network of mortgage
originators and the systems for sourcing and screening their
loans to be underwritten or qualified to be included in a
guaranteed pool of loans for eventual securitization in a FMIC
reinsured mortgage-backed security are critical functions in
the MBS issuance chain that are not, to my knowledge, currently
contemplated to be included in the CSP.
To allow bond guarantors to compete on equal footing with
the GSEs in the interim, Fannie Mae and Freddie Mac could be
directed to license or offer necessary elements of their
issuance infrastructure for a fee that would reflect run-rate
operating expenses for a private competitor. In a sense, these
elements of GSE infrastructure would serve as a market utility.
And they could do so before the CSP is launched. As more
issuance functions are performed by the CSP and private
competitors develop their own networks and systems, they would
rely less on the legacy GSEs.
During the transition to the new system while the old GSE
model is wound down, there are indeed additional authorities
that would help a regulator to maximize value and minimize
disruption to the housing finance market. Most importantly, the
mortgage guarantee businesses of Fannie Mae and Freddie Mac
need to be recapitalized and ultimately privatized to ensure
that there is adequate private capital to support desired
levels of conforming issuance through the transition and in the
end state, and to protect taxpayers against loss on the legacy
book of GSE MBS before that book is reinsured by the FMIC (as
described above in Answer 1). The FHFA, in conjunction with the
FMIC and Treasury, need the authority and directive to
accomplish this, which would have the added benefit of
maximizing the value of the legacy GSE estates for their
largest owner: taxpayers. A regulator needs discretion to
increase fees charged by the GSEs to guarantee MBS depending on
the risk of providing that guarantee, market conditions, the
relative costs of private competitors, and the need to
recapitalize the mortgage guarantee businesses of Fannie Mae
and Freddie Mac. A regulator also needs discretion to change
conforming loan limits depending on market conditions. Further,
a regulator should have the authority to supervise and
establish requirements for participants in the conforming MBS
chain, including capital requirements for entities assuming
credit risk or providing credit enhancement on individual
mortgages or pools, and fees and processes for servicing those
mortgages. Finally, a regulator should have the authority to
direct the legacy GSEs to experiment with risk-sharing
structures, infrastructure leasing and other avenues to
facilitate private competition in conforming mortgage credit.
Q.3. Should a common securitization platform be fully
operational before the old system is shut down? Are there other
goals that should be achieved before the old system is shut
down to help make the transition as effective and smooth as
possible?
A.3. It is not necessary to wait to launch the new system until
a CSP is fully operational. As I explain in my responses to the
first and second questions, while a CSP is being established,
the legacy GSEs could facilitate the issuance of a single
security and could be directed to license issuance
infrastructure to private competitors. Both steps could
accelerate the entry of new guarantors into the market and
thereby allow the FMIC to offer secondary-loss insurance on
conforming MBS relatively quickly.
Moreover, it is unclear what it would mean for the CSP to
be fully operational. To my knowledge, plans for the CSP are
still being developed, and the most likely first phase would
provide for a portion of the securitization and bond
administration functions currently performed by the GSEs to be
performed by the CSP. That first phase will take several years.
And beyond that initial phase, I am skeptical that a CSP will
ever perform all of the additional elements in the issuance
chain, as I explain in my response to the second question. To
my knowledge, no one is contemplating that the CSP would
purchase loans for cash from originators and warehouse them on
its balance sheet until they can be securitized.
There are important goals that should be achieved before
the legacy GSE model is shut down. The mortgage guarantee
businesses of Fannie Mae and Freddie Mac should be
recapitalized and ultimately privatized to ensure that there is
adequate private capital to support desired levels of
conforming issuance through the transition and in the end
state, and to provide an adequate capital cushion in front of
the FMIC's reinsurance so as to protect taxpayers against loss
on that reinsurance. A single conforming TBA-eligible pass-
through single-family mortgage-backed security should be
established. The FMIC should be established and its Mortgage
Insurance Fund should be capitalized from fees paid by the GSEs
for the reinsurance of their legacy MBS. The new regulator
should also have the resources necessary to supervise
participants in the conforming MBS chain in order to prevent
market disruption and to protect taxpayers against potential
losses on the FMIC's insurance. Standards for originating and
servicing conforming mortgages should be established.
However, the FMIC should not establish standards for
securitized conforming mortgages in a vacuum. It should work
closely with borrower, lender, servicer, and investor groups to
develop standards that are not only theoretically sound but
also workable and sensible in practice. To that end, the newly
formed CSP is attempting to establish an industry advisory
group that would be similar to the Treasury Borrowing Advisory
Committee. Such a group would increase the likelihood that the
CSP can be useful across the mortgage market. The FMIC should
establish similar advisory committees to facilitate meaningful
vehicles for private-sector input for the new system before the
legacy GSE model is shut down.
------
RESPONSES TO WRITTEN QUESTIONS OF
CHAIRMAN JOHNSON FROM JOHN BOVENZI
Q.1. In an earlier Housing Finance Reform hearing the Committee
held on regulatory issues, witnesses called for the new housing
finance regulator to have more supervisory and enforcement
authorities than simply approval and denial of participation.
Do you agree that is something we should consider adding? If
so, why is it important to equip a regulator with explicit
supervisory and enforcement authorities?
A.1. Based on my experience at the FDIC, I believe it is worth
considering whether the FMIC should be granted broader
supervisory and enforcement authorities beyond controlling
entry and exit into and out of the program and the ability to
issue civil money penalties. Those three authorities are
important, but in most instances appropriate behavior can be
maintained by less extreme enforcement measures than by forcing
an institution's exit from the program. The FDIC has found its
authority to issue informal memoranda of understanding and
formal cease-and-desist orders to be quite effective in that
regard. Moreover, the FDIC has rarely used its power to revoke
deposit insurance coverage, finding it to be a cumbersome
process compared to these other enforcement alternatives.
The FDIC also has examination authority to ensure banks are
in compliance with FDIC supervisory standards and to determine
whether enforcement actions are necessary. Consideration also
should be given to giving the FMIC examination authority. While
off-site monitoring can be used to help monitor an
institution's behavior, over time the FDIC has found there is
no substitute for the direct interaction with bank management
that occurs during the examination process.
Finally, to help ensure that supervisory actions are taken
in a timely manner, the FDIC is subject to Prompt Corrective
Action requirements, which mandate that certain supervisory
actions be taken as bank capital levels drop below prescribed
levels. In their entirety these powers are an important part of
safeguarding the financial system and protecting the FDIC's
deposit insurance fund.
Q.2. What should we consider when it comes to employee
retention and related issues in the transition?
A.2. Based on my experience at the FDIC and RTC I believe it is
critical to provide open and clear communication to employees
as to their job status. As a limited life agency, the RTC's
employees knew that by doing their jobs correctly they might be
putting themselves out of a job. The same applied to the FDIC's
employees who were responsible for handling the spike in bank
closings. Eventually the economy would recover and their
workload would vanish. But they were the ones who had the
experience and the expertise to determine whether their
agencies would succeed or fail. The same is true here.
Ultimately the employees of the two Government-sponsored
enterprises and the Federal Housing Finance Agency (FHFA) will
determine whether a new start-up agency succeeds or fails.
Uncertainty surrounding how many jobs will be available, on
what terms, and who will get them will create significant
complications in ensuring a smooth transition. There may not be
answers to all of these questions, but employees can understand
that as long as they are kept informed in a timely manner and
believe they will be treated fairly, even if there aren't
enough jobs for everyone at the end of the transition.
Q.3. Please discuss whether or not licensing of Fannie Mae and
Freddie Mac's assets can be an important tool in managing their
wind down in a manner that maximizes value and minimizes
disruption to the housing finance market. Are there other
authorities that could be helpful to a regulator in managing
their wind down?
A.3. Licensing certain of Fannie Mae's and Freddie Mac's key
assets could be helpful in certain situations in order to
preserve value and minimize disruption. This would allow the
Government to obtain additional revenue by selling licenses to
a number of private-sector firms. For other assets where
licensing isn't appropriate, it would be important to ensure
that they are sold through open and competitive processes
rather than through negotiated sales. It may also be valuable
to form partnerships with private-sector participants for the
management and sale of certain assets. Such partnerships have
served the FDIC well in maximizing value from its failed-bank
receiverships.
Q.4. Should a common securitization platform be fully
operational before the old system is shut down? Are there other
goals that should be achieved before the old system is shut
down to help make the transition as effective and smooth as
possible?
A.4. As a general matter, to ensure a smooth transition, it is
important not to shut down one system before its replacement is
operational. This may mean some extra costs in the short run,
but stability to the system is too important to risk otherwise.
Thus, some patience and flexibility in approach is required as
transitions as complicated as a moving to a new housing finance
system generally don't move as fast as one hopes or initially
anticipates.
------
RESPONSES TO WRITTEN QUESTIONS OF
CHAIRMAN JOHNSON FROM MARK ZANDI
Q.1. Acting Director DeMarco indicated that FHFA is planning
further increases to guarantee fees. Do you believe further
increases are justified in light of current market conditions?
A.1. No, I don't believe further GSE G-fee increases are
appropriate at this time. The housing recovery has been hurt by
the close to 100 basis point increase in fixed mortgage rates
over the past year. The housing recovery is vital to the
broader economic recovery, job growth and lower unemployment.
With the Federal Reserve tapering its bond-buying program,
long-term interest rates are likely to rise further in coming
months. In this context, increasing G-fees and thus mortgage
rates would not be productive at this time.
While it is desirable for the GSEs to reduce their
footprint in the mortgage market and allow more private lending
to occur, private lending remains constrained. The private
residential mortgage backed securities market is moribund due
to a range of factors, including a lack of regulatory clarity
over such things as the QRM rule, and will be unlikely to
provide sufficient credit, at least not for much of 2014.
The current G-fees charged by the GSEs is consistent with
an approximately 2.5-percent capitalization (this estimate is
based on a number of assumptions). While this is an
insufficient level of capitalization for the housing finance
system in the long-run, it is sufficient for the very immediate
future.
Q.2. Should a new system be designed to issue a single security
regardless of how many guarantors or issuers there may be? If
so, how should the issuance of a single security be structured
during transition? How should the issuance of new guaranteed
MBS be introduced while ensuring that legacy GSE securities do
not become orphaned?
A.2. Yes, the future housing finance system should be designed
to issue a common Government-guaranteed security. This would
improve liquidity in the TBA market and result in lower
mortgage rates. A common security would also lower entry
barriers into the guarantor market, as no guarantor would have
an advantage because of the liquidity of the securities they
back.
This is a problem in the current housing finance system, as
Freddie Mac securities are much less liquid than Fannie Mae
securities. Fannie and Freddie split the MBS market 60-40, but
on a typical day the trading volume of Fannie MBS is 10 times
greater than that of Freddie MBS. To compensate, Freddie is
forced to charge a lower G-fee than Fannie. There are some
modest differences in the securities--Freddie pays investors
more quickly than Fannie and its securities prepay a bit more
quickly--but the key difference is their liquidity. This
liquidity difference makes the mortgage market less efficient
and less competitive, and leads to higher costs for mortgage
borrowers and taxpayers.
A potential fix to this problem in the transition would be
to make Fannie and Freddie securities fungible, creating a
common TBA security. That would require a change to the good-
delivery guidelines for TBA, to allow the delivery of either
Fannie or Freddie securities into the same contract. The
securities themselves would not change; their separate TBA
markets would simply be merged. Both securities would still be
separately identifiable and tradable, only the TBA trades would
be merged. Not only would this interim step improve liquidity,
it would demonstrate investor interest in a truly common
security that would be an important feature of the future
hybrid housing finance system.
The future housing finance system should not orphan the
currently close to $5 trillion in legacy MBS that are already
guaranteed by the GSEs and effectively backed by taxpayers. Any
reform process should make clear that the United States will
stand behind these legacy obligations. In effect, legacy GSE
MBS would become like Ginnie Mae MBS, with a full faith and
credit wrap guarantee on the entire security.
Legacy MBS should also be made fungible with the new
Government-backed MBS. This would require a restructuring
process whereby holders of these legacy MBS can convert them
into the new MBS, deliverable into the new TBA market. While
there would be some technical issues, including requiring a new
CUSIP number and a matching-up of payments delays, they could
be easily addressed.
Q.3. Should a common securitization platform be fully
operational before the old system is shut down? Are there other
goals that should be achieved before the old system is shut
down to help make the transition as effective and smooth as
possible?
A.3. Yes, a common securitization platform should be fully
operational before the old system is shutdown. The CSP would
produce a more liquid market, facilitate loan modifications in
future downturns, and give issuers operating flexibility at a
low cost. It would also allow for a robust TBA market.
The CSP would leverage current efforts by the FHFA to
develop a single platform for Fannie Mae and Freddie Mac
securities. For a fee, the securitization facility would
provide a range of services, including mortgage loan note
tracking, master servicing, data collection and validation to
improve transparency and integrity, and bond administration.
Mortgage loans included in securities that use the CSP
(including all mortgages that benefit from the Government
guarantee plus some nonguaranteed loans) could be covered by a
uniform pooling and servicing agreement and uniform servicing
standards that encourage prudent underwriting and align
investor and borrower interests. This would encourage the
adoption of similar standards for other mortgages.
The CSP would permit multiple originators to sell mortgages
into single securities with access to the Government guarantee.
In return, the originators would receive pro rata shares of the
security. Pooling requirements would be largely the same as for
typical single-originator securities, and they would be good
for delivery into the TBA market. Originators could thus easily
convert securities to cash before the securities were created,
an especially important feature for smaller originators.
Other goals that should be achieved before the old system
is shut down include the issuance of a common Government-
guaranteed security (see response to Question 2), and well-
developed risk-sharing by the GSEs. Enticing private capital
back into the housing finance system is part of the transition
to any future housing finance system. It is thus encouraging
that the FHFA has mandated Fannie and Freddie to begin that
process through risk-sharing with capital market investors and
private mortgage insurers. The goals so far are modest and
should be steadily expanded. What is learned from these efforts
will be instrumental to ensuring there is enough private
capital to support the future housing finance system.
------
RESPONSES TO WRITTEN QUESTIONS OF
CHAIRMAN JOHNSON FROM DAVID MIN
Q.1. In the written testimony you provided you state,
``Regulators should be given greater discretion and encouraged
to respond to developments on the ground . . . .'' How can a
regulator set capital requirements in a manner that protects
taxpayers while keeping in mind broader market dynamics,
including access to credit?
A.1. As I state in my written testimony, determining the
specific level of capital necessary to protect taxpayers from
losses is a largely empirical question that ought to be studied
in great detail. A number of factors are still unknown,
including, importantly, how much investment capital will come
into the new system contemplated by Corker-Warner and at what
price. With that being said, I would offer the following
principles in thinking about how to balance the different
policy objectives of reducing taxpayer risk and keeping access
to credit broadly and consistently available.
First, capital requirements for the new MBS issuers should
generally be harmonized with those for other financial
institutions, in order to prevent capital arbitrage.
Second, consistent with many of the proposals put forth in
Basel III and by U.S. banking regulators, it is appropriate to
think about countercyclical capital measures, such as
contingent capital or countercyclical capital buffers.
Regulators may also consider following the lead of the Federal
Reserve and conducting stress tests to ensure that banks are
sufficiently capitalized during credit expansions. On the flip
side, during credit downturns, it is necessary to ensure
sufficient liquidity, either through relaxing countercyclical
capital requirements or other activities.
Finally, in thinking about the ``right'' level of capital
that the new MBS issuers should hold, it may be worth
considering the highly disparate performances of different
mortgage products. While bank capital requirements currently do
not differentiate between different types of home loans, or
different types of MBS, the starkly different delinquency rates
between say 30-year fixed-rate mortgages originated for Agency
securitization and 5-year adjustable-rate mortgages originated
for private-label securitization suggest that such a
differentiation may be quite appropriate. Such an approach
would be somewhat consistent with the treatment of home loans
under Basel III's new standardized approach, and with the
Qualified Residential Mortgage exemption to risk retention
under Dodd-Frank.
Q.2. Should a common securitization platform be fully
operational before the old system is shut down? Are there other
goals that should be achieved before the old system is shut
down to help make the transition as effective and smooth as
possible?
A.2. As I stated in my written testimony, I believe the
overriding policy goal of any transition should be to do no
harm. Thus, it is imperative that the new common securitization
platform (CSP) be fully operational before we shut down the old
system. But any such endeavor will inevitably have unexpected
developments. Therefore, testing out the new system is clearly
necessary before the old system is shut down, to avoid
jeopardizing the mortgage markets and the broader economy. This
reasoning is why I strongly advocate a piecemeal approach to
transition. The new system should start with small dedicated
pieces of the business currently held by Fannie and Freddie,
such as 15-year fixed-rate mortgages or high cost conforming
mortgages, before moving on to absorb larger chunks of the
conforming market. Such an approach would allow regulators to
work out the kinks in the new system before shutting down the
old one.