White House Could Preempt Congress’ Inaction on GSEs

0
110


[ccpw id=”6606″]

Two noted economists seem
to be holding little hope that Congress will take up in any way changing the
status quo of the government-sponsored enterprises (GSEs) before adjourning
next January.  However, they say it is an
issue on the Trump Administration agenda and they expect the Executive Branch will
step into the fray. 
In a paper published
by the Urban Institute titled GSE Reform
is Dead – Long Live GSE Reform
, Jim Parrott and Mark Zandi ask, if Congress
does fail to address GSE reform, what will the executive branch do with the
opportunity?

Parrot is a nonresident fellow at the Urban
Institute.  He owns Falling Creek
Advisors, a financial institution consulting firm and previously served as
White House senior advisor on the National Economic Council. 

Zandi
is chief economist of Moody’s Analytics, where he directs economic research and
frequently comments on monetary and fiscal matters to major news media
outlets.  Among his published books is Financial Shock: A 360º Look at the Subprime
Mortgage Implosion, and How to Avoid the Next Financial Crisis.

The most important player
in any reform of the GSEs will be the Federal Housing Finance Agency (FHFA), the
regulator and conservator of the GSEs Fannie Mae and Freddie Mac.  The two privately held corporations were
placed under federal conservatorship in August 2008 and remain there despite both
returning to profitability in 2008.

The term of the current director, Mel Watt, will expire in
January so the administration will have the opportunity to appoint a new one,
undoubtedly an individual who will share their vision for GSE reform.  The authors say, given the current ideology,
that individual will probably take steps to reduce the GSEs footprint in the
housing finance system.  Also, given the administration’s
public commitment to ending the conservatorship, the Treasury Department could work
with FHFA to release the GSEs.

The authors look at both issues and we will summarize their
analysis in two separate articles.  This
one will look at the options for shrinking the current GSE footprint in the
market.  A second will explore some of
the ways in which the current GSE/government relationship might be altered or
eliminated.

Conservatives
have long been critical of the role the GSEs play in the housing finance
system, placing significant blame on them for the housing crisis and bemoaning
their outsized influence since then. They make the argument that they distort
the market and create excessive taxpayer risk. Thus, it should be expected that
a new director will take steps to reduce this market role.

The
most direct way to accomplish this would be to gradually reduce the conforming
loan limits. 
The current ceiling is
$453,100 in most U.S. counties but can reach as $679,650 in high cost
areas.  Reducing the limit would shrink
the GSE footprint in high income areas, using the rationale that they are the least
in need of government support. Eliminating the outlying higher limits could cut
GSE loan volume as much as 5 percent; reducing the standard limit to $350,000
could cut it by one quarter.  While this
would violate many legislative rules regarding the limits, Parrot and Zandi go
into detail as how this could be done.  

 

 

A
second path would be to tighten minimum credit requirements, for example
reducing the loan-to-value ratio, debt-to-income levels and raising the credit
scores for loans eligible for GSE purchase or guarantee.

 

 

Either
aggressively reducing loan limits or tightening underwriting standards would be
disruptive, but in different ways.  A
loan limit change could leave entire neighborhoods with less access to credit,
possibly all at once. While portfolio lenders might pick up some of the slack, borrowers
who don’t fit into their boxes would face higher mortgage rates.  Tightening credit would push a large number
of borrowers into subsidized government lending channels like FHA which might,
in turn, increase taxpayer risk.

Yet
a third way to reduce the GSE footprint would be to gradually raise the fee
charged for their guarantee.
This is currently at about 60 basis points for all
30-year fixed-rate mortgages (FRMs) The GSEs also charge a loan level price
adjustment, of anywhere from 25 to 375 basis points on higher-risk loans.  FHFA could require the GSEs’ to raise either
or both of these fees and the impacts would differ with their choice.  Boosting the overall guarantee fee would
entice many lower risk borrowers to move to a portfolio lender or FHA.
Increasing the LLPAs would cause the GSE footprint to shrink at the bottom end
of the credit spectrum, pushing higher-risk borrowers toward FHA as well.

Finally, the GSEs could be required to pull back on
the types of loans they purchase. Conservatives argue there is no public policy
reason for the GSEs to do cash-out refinancing, which accounted for 20 percent
of their loan volume last year, or to make investor or second home loans.  Those constituted another 10 percent.

The choices of how to reduce the GSE footprint will
depend on which part a conservative FHFA wants to shrink and the type of market
disruptions it is willing to endure.  The
authors say it is relatively certain it will use a mix of each to shrink the
GSE role in single-family housing and probably will consider ways to reduce
their role in the multi-family sector as well.

Multifamily lending has
increased
over the decades and the GSEs now account for almost half of all outstanding
loans of that type.  The GSEs have seemingly
avoided some of their single-family lending mistakes in this sphere so there
appears to be less pressure to shrink or eliminate that portion of their
businesses.  However, the authors expect
at least some consideration to be given to reducing origination volume caps to
more narrowly target such lending or even to spin the multifamily businesses
out of government control altogether.

Some conservatives have also criticized the cross-subsidy
the GSEs provide.  Freddie and Fannie
charge lower risk borrowers more than their risk levels would dictate so higher
risk borrowers could pay less than theirs would imply. This policy is intended to
put more loans in reach of more borrowers, meet the GSEs’ housing affordability
goals, and allow them to fulfill their mandate to serve rural and manufactured
housing markets.  The cross subsidy is
estimated at $3.8 billion per year.  

Critics argue the cross subsidy is unfair to lower-risk
borrowers
and that it places excessive risk into the system.  They even have even blamed the GSEs affordability
goals in part for the financial crisis. They also argue that those goals increase
demand and push home prices higher, making them unaffordable to those the
system is designed to serve. The authors say these claims have been
consistently proven incorrect, but they remain central to conservative
narratives.

A new director is likely to change this as well.  At the extreme, FHFA could push the GSEs into
full risk-based lending, significantly increasing LLPAs while lowering
guarantee fees for low risk borrowers.

 

 

If the GSEs
were required to remove the cross-subsidy and fully enact risk-based pricing
they would siphon off some lower-risk borrowers from portfolio lenders and lose
some higher risk borrowers to the FHA
or other more heavily subsidized
channels of government support.

In taking
this step, the next FHFA director would also likely roll back the affordability
goals and duty to serve requirements to an extent that the GSEs could easily
meet them without needing a cross-subsidy. While the FHFA is mandated by HERA
to set these requirements, the director has enough discretion to set them in a
way that has little effect. 

 



Original Source