Actually, that
is the easy question. There seems to be an almost universal
consensus that the agencies should be phased out, and there
are many different ways to do this. Perhaps the cleanest
would be a gradual but preset reduction in the maximum size
of loan that the agencies are authorized to purchase.
The reason
that this phase-out plan, or any other, has not been adopted
is that we need the agencies to keep the housing and
mortgage markets from collapsing. The tough question is not
how we phase out the agencies but how we replace the market
support they now provide.
Right now,
there is no game plan for replacing them. We are just
drifting, one more example of Federal Government
dysfunction. Unless we develop an effective game plan for
replacing them, Fannie and Freddy will be with us
indefinitely.
We lost our
chance to do it the easy way. A phase-out of the agencies
before the financial crisis would have been matched by a
self-generating phase-in of the private sector. The
fixed-rate mortgages that the agencies could no longer
purchase would have been sold into the then-thriving private
secondary market. The overflow of adjustable rate mortgages
would have been placed in the portfolios of the major
depository institutions, who viewed them as prime assets
with manageable interest rate risk and cross-selling
potential.
Both of these
outlets, which would have required nothing of Government,
are now effectively shut. The private secondary market
remains traumatized by the heavy losses sustained on private
mortgage securities following the crisis. The depositories
remain traumatized by the heavy losses they have sustained,
and are still sustaining, from being held legally liable for
misdeeds committed prior to the crisis.
In its zeal to
punish those considered responsible, the Federal Government
has fundamentally altered the risk/reward calculus of those
institutions going forward. They now view home mortgages as
carrying high political risk. They will not be fund sources
in the new housing finance system that must emerge to
replace Fannie and Freddie.
The other
mortgage lenders in our system today are mortgage companies
that originate to sell. They will continue to be needed to
originate conventional loans for sale to Fannie and Freddie,
and FHA loans that are sold to issuers of GNMA securities.
Because they do not have permanent funding capacity, they
cannot fill the void left by a phase-out of Fannie and
Freddie. However, they could convert themselves into
mortgage banks of the type described below, which would give
them permanent funding capacity.
One of the
benefits of a financial crisis is to reveal the types of
institutional arrangements that are not sufficiently robust
to weather the storm, and those that are. Fannie and Freddie
fall in the first group, while the Danish system of mortgage
banks belongs in the second. These banks, privately owned
but regulated by the state, have existed for several hundred
years. They originate mortgages and fund them with mortgage
bonds for which they retain full liability.
While Denmark
was subject to a housing bubble, and a subsequent collapse
in home prices similar to that of the US, there was no
crisis there and no breakdown in the market for mortgage
bonds. During the worst phases of the financial crisis in
the US, it was business as usual in the Danish mortgage
market. Indeed, there has never been a default on a
Danish mortgage bond.
The reason the Danish mortgage bond system weathered the
crisis and the private mortgage-backed security system in
the US collapsed is not difficult to understand. A Danish
mortgage-bond is a full liability of the bank issuing it.
This means that if the pool of mortgages supporting one bond
experiences a high default rate, all the resources of the
issuing bank are available to protect the investors who hold
the bonds issued against that pool. Hence, all the bonds
issued by one bank stand or fail together, and they never
fail.
In contrast, every private mortgage-backed security (MBS)
issued in the US had “credit enhancements” designed to allow
it stand on its own feet. If nine MBSs had more credit
enhancement than they needed and one had less, that one
would fail. The surpluses from the nine good securities
could not be touched, and the issuer of the security with a
deficiency had zero liability.
The Dodd/Frank legislation passed after the crisis took a
swipe at this problem by stipulating that under some
circumstances mortgage security issuers had to retain 5% of
the issue. If that had been done 10 years earlier, it could
have had an enormous impact, but to have any impact today
more powerful medicine is needed.
Specifically, we need a new system of nationally-chartered
mortgage banks, similar to those in Denmark. They could be
regulated by the Federal Housing Finance Agency, which
currently regulates Fannie and Freddie. Depository
institutions would be encouraged to charter mortgage bank
affiliates, and existing mortgage companies would be
encouraged to convert. Such encouragement might include some
temporary Federal assistance in raising the capital that
full-fledged mortgage banks require.
This is not the only possible game plan for replacing Fannie
and Freddie, but I have not seen any that have been as
thoroughly tested as this one.
