Comparison of Senate bill with Kolbe-Stenholm

[National Dialogue Moderator <moderator>]

     To the panel and participants:

I have been asked to compare and contrast the Senate bipartisan
bill and the Kolbe-Stenholm legislation.

The most important difference between the two bills is procedural
rather than substantive.  I joined with Senator Breaux as well as
Congressmen Kolbe and Stenholm in introducing legislation during
the last Congress that was modeled on the recommendations of the
National Commission on Retirement Policy.  This year I was actively
involved in negotiating with the sponsors of other reform proposals,
including Senators Kerrey, Moynihan, and Santorum, in the interest
of devising an agreement that could draw bipartisan support.
Because of the larger number of "authors" involved in the Senate
negotiations, it was inevitable that additional compromises must
be reached with those of different viewpoints.  I believe it is
fair to say that, while both the Kolbe-Stenholm and Gregg-Breaux
bills have been altered to reflect our changed economic and political
circumstances in the new year, the Kolbe-Stenholm legislation is
slightly closer to the form of the Commission recommendations than
is the Senate bipartisan agreement.

The Senate negotiating procedure, of course, brings with it advantages
and disadvantages.  The disadvantage was that on some particulars
I was obliged to give ground when I might have preferred a policy
choice closer to the Kolbe-Stenholm position.  But there is also
an advantage to bringing in more people with different ideas,
because this is an important part of producing legislation that
can draw significant Senate support.

I will only touch upon a few of the substantive differences here.
The bills may be viewed as more similar than they are different.
Each would create a personal account with surplus payroll taxes
equal to 2% of taxable payroll.  Each would offer progressive
matches for low-income individuals.  Each would eliminate the lion's
share of the pressure to increase income taxes that would exist
both under current law and under the major reform alternatives that
do not address the form and size of government outlays.  Each makes
a number of changes to improve the work incentives of the traditional
system.

One difference between the two bills is the way in which they add
progressivity to the basic defined benefit.  You may be aware that
one of the criticisms of personal accounts is that they do not have
a progressive or redistributive character, and thus might impair
the "safety net against poverty" that results from the current
system being somewhat redistributive.  Each bill would put in place
enhancements to the progressivity of the basic defined benefit to
adjust for the lack of progressivity in the personal accounts
(though both versions this year also offer progressive matches in
the personal accounts as well.)  The Kolbe-Stenholm bill does this
through a new minimum benefit guarantee that assures that any
individual who has worked a full lifetime would be guaranteed a
benefit at least at the poverty level, which the current system
does not do.  The Senate bill would instead make gradual changes
to the basic OASDI "bend point factor" formula to make it more
progressive.  Each would result in the defined benefit on the
lower-income end being strengthened.

The rationale behind the Senate method here is that some Senate
negotiators did not want to sever the connection between earnings
and benefits for any income group.  A minimum benefit provision
would guarantee a certain level of benefits regardless of one's
earnings history.  The Senate changes to the bend point factors
attempt to achieve a similar gain but without severing the connection
with earnings and contributions.  There are good arguments for each
approach.  Many in the Senate believe that maintaining the connection
between earnings and bnefits is an important work incentive.  But
I also know that the House sponsors are proud of the fact that they
can demonstrate that no individual who works a full liftetime under
their bill would retire in poverty, no matter how low their earnings.

One question asked dealt with how Social Security proposals would
care for those who take time off from work.  Clearly, the "minimum
benefit guarantee" in the House bill (which operates at some level
at any number of years worked at 20 or above) gives added protections
to those whose lifetime earnings levels are low.  It must be
remembered that under current law, benefits will go down for an
individual who has suffered work interruptions, and that it is the
progressivity of the system that acts as a shield against work
interruptions translating into old-age poverty.  Since both proposals
increase the progressivity of the basic benefit, they would enhance
the protections available for those who take time off from work.

Another difference between the bills is the way in which the personal
accounts advance-fund future benefits.  The Kolbe-Stenholm legislation
stays closer to the form of last year's Commission bill, in that
progressive and gradual changes are made to the basic Social Security
benefit formula in accordance with the sponsors' calculation of
the appropriate changes after a portion of benefits have been moved
into a funded system.  The Senate bipartisan bill makes the
calculation more exact, but less predictable.  The adjustment to
the remaining OASDI benefit would be made in exact proportion to
the interest-compounded value of the payroll tax refund into a
personal account in a given year.  That phrase is a mouthful, so
please allow me to translate.  Suppose that you received a $400
refund into a personal account this year.  The calculation would
be made as to what level of benefit this $400 would have provided
to you if it compounded at the Trust Fund interest rate, and this
amount is deducted from your ultimate benefit, to reflect the fact
that this portion of your benefit has been moved into the present
on a funded basis, and off of the future taxpayer's shoulders.
Doing it this way enables beneficiaries to make a very simple
choice.  If they do not wish to assume any new risk, they can just
simply choose to invest their money in a bond that earns the same
rate as the Trust Fund (3%.)  If they believe they can beat that
rate by investing in one of the equities funds provided, they may
choose to do so. But by doing the calculation this way, beneficiaries
are assured that the swap will not change their total benefits if
they simply buy a T-bond.

One important aspect to bear in mind about each of these bills is
that they would significantly reduce the amount of general revenues
required to fund Social Security benefits relative to current law
in the years from 2014 to beyond.  Last year, we each produced
bills that would achieve this by working wholly within the 12.4%
OASDI tax rate.  The personal accounts would receive a 2% contribution,
and the payroll tax would be dropped to 10.4%.  But earlier this
year the President as well as several other legislators produced
approaches to Social Security that did not attempt to keep costs
stable over the next 30 years, leaving in place its current-law
growth to 18% by the year 2030.  It became clear that much of the
rest of the body politic was unwilling to make the choices necessary
to live within the constraints embodied in last year's
Gregg-Breaux-Kolbe-Stenholm legislation.  So we attempted to retool
our legislation in an attempt to find an agreement with those who
had been reluctant to put forward similarly comprehensive proposals.

Both the Senate bipartisan plan and the Kolbe-Stenholm plan would
assume that methdological corrections are made to the Consumer
Price Index, but they differ in the magnitude of those assumptions.
Kolbe-Stenholm assumes that they would have an impact equal to
0.33%, whereas the Senate bill assumed 0.5%, but with a provision
guaranteeing that current retirees would continue to have COLAs
indexed with the old formulation of CPI.  This is by no means a
perfect solution, but it met the desire of Senate negotiators to
say that their legislation would not affect current seniors, even
to the extent that CPI corrections that most economists agree are
necessary are implemented.

As a consequence, the assumptions regarding revenues to the system
are somewhat different under each of the plans.  They are slightly
higher under the Senate bill, though much lower than current law
in such years as the 2020s and the 2030s.  There are two things
that chiefly determine the amount of benefits that reform plans
can promise.  One is the tax levels provided to the system, and
the other is the degree of advance funding (and thus the proportion
of total investment that can receive an improved rate of return.)
Kolbe-Stenholm and the Senate bill attempt to achieve a similar
amount of advance funding, but the total revenues would be slightly
higher on the Senate side.  Consequently, advertised benefits
(calculations are still ongoing) are likely to be slightly higher
in the Senate bill, while on the other hand the House bill is likely
to achieve slightly greater success in reducing currently-projected
tax burdens.

Which option is preferable is likely to be determined by future
negotiations.  By having slightly more revenue in the Senate bill,
we were able to be a bit "softer" on some provisions.  For example,
our bill would gradually raise the normal retirement age to 67 (it
would go there under current law) and then it would stop there.
The House bill would reach the same point, but thereafter both the
normal retirement age and the early eligibility age would continue
to go up by one month every two years, to reflect projected changes
in life expecctancy.  If we wish to avoid these changes, then costs
will be associated with that choice.  The American body politic
will need to decide upon the right balance between tax levels and
the "right" amount of time for individuals to spend in retirement.
As indicated, the current formulation would mean that future
generations would be hit by tax increases of approximately 50%
relative to the high levels already paid today.  The Kolbe-Stenholm
and Senate bills show the different costs associated with each
choice.

I will stop there.  I have done my best to characterize the
differences accurately, and if my House colleagues disagree with
any of my descriptions, my errors are inadvertent.  I do strongly
support the effort being made in the House, and trust that this
sheds some light on the slightly different direction taken by the
Senate bipartisan team.

I will say to readers that there is one other important similarity
between the House and Senate efforts, and that is that we are each
battling against the presumption that nothing on Social Security
can happen this year.  Though it is always difficult politically
to tackle Social Security, it will never get much easier.  In fact,
it only becomes more difficult, and more expensive, with each year
of delay.  It is important to get past the divisive rhetoric about
who is more concerned about America's seniors, and charges that
the replacement of an unfunded benefit promise with a partially
funded personally-owned benefit is a "benefit cut."  Such rhetoric
only serves to make public servants more wary of reaching an
agreement, and makes it much more difficult to achieve success.
Although I do not expect anyone to agree with every element of our
proposal, I do hope that you will express any support that you may
feel for action this year.  In order for success to be achieved,
it is imperative that we be joined by the President and leadership
from both sides in both houses of Congress, in evaluating serious
and comprehensive proposals, rather than continuing the current
stalemate in which "talking point" plans are detailed that in
reality would achieve nothing.  We are doing our best to enable
something constructive to happen.

     Senator Judd Gregg