View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

September 15, 1995

eOONOMIC
GOMMeNTORY
Federal Reserve Bank of Cleveland

Should Social Security Be Privatized?
by Jagadeesh Gokhale

M,

Lany developed countries operate
comprehensive public pension programs
to protect the elderly against a wide
range of adverse economic circumstances. In the United States, the Social
Security program has been in operation
for 60 years and has expanded over time
in terms of both payroll tax rates and per
capita benefits. The manner in which
such programs are financed and the
methods used to provide retirement and
other benefits exert a significant influence on national saving, the labor supply, and ultimately, economic growth.
Because a sizable chunk of the U.S. population is expected to begin retiring relatively soon, the rationale, operation, and
magnitude of the Social Security system
are likely to come under increasingly
intense scrutiny by policymakers and the
public alike.1 Several interest groups are
advocating a downsizing of the system,
and the Commission for the Reform of
Entitlements, appointed by President
Clinton, recently recommended that a
portion of Social Security be privatized.
Reforms of public pension systems have
been implemented or are being contemplated in several nations. The reform
effort in Chile, which moved the country
toward privatization in the early 1980s,
has been hailed as an unequivocal success because it transformed a system
headed for bankruptcy into one with significant reserves and the capacity to provide retirement benefits at high rates of
replacement.2 Argentina, Australia,
Bolivia, Columbia, the United Kingdom,
and Peru may soon follow suit. This Economic Commentary takes a brief look at
ISSN 0428-1276

the rationale for public pension programs, discusses the evolution of the
U.S. Social Security system, and examines the issue of privatization.

•

Rationales for Public Funding

There are three main rationales for public intervention in retirement funding.
The primary concern is that a large fraction of the population will not save and
insure adequately during their highearning years and thus will arrive at
retirement in poverty. Both high and low
earners could end up with insufficient
resources because of a lack of foresight
about their future needs, an intense desire
for current consumption despite an
appreciation of future needs, or misinformation about future earning prospects
and disability/mortality probabilities.
Because of the uncertainty surrounding
one's longevity, it makes sense to invest
retirement savings in annuities that provide a steady income until death. Hence,
a second rationale for public intervention
is that the private market for annuities is
thin or that annuities are priced unfairly
because of informational problems.
Anyone offering actuarially fair annuities would attract only relatively healthy
individuals with very low mortality
probabilities, thus making actuarially
fair pricing unprofitable. This kind of
informational problem also affects private disability, spousal, and survivor
insurance. A government, however, can
overcome this problem by mandating
universal coverage through a retirement
insurance program that pays benefits in
the form of annuities — like the Social
Security System in the United States.

The challenge of providing income
security for a growing number of
retirees promises to place Social
Security reform at the center of
future economic debates. Although
the system is projected to be solvent
over the next three-and-a half
decades, its current and prospective
transactions with the general government budget make it an integral part
of the nation's overall fiscal policy.
Hence, judging its solvency independently may prove misleading.

The third major rationale for public pension support is that it is impossible for
the private market to provide mechanisms for risk-sharing between living
and unborn generations. If one generation suffers through a war or a severe
economic depression, there is no way for
private insurers to spread any of the
costs of these events to future generations. But the government can do so by
providing benefits to the unfortunate
generations today and by deferring taxation so that future citizens are required to
pick up part of the tab.

• Evolution of the U.S.
Social Security System
The initial impetus for setting up the
Social Security system in 1935 was provided by the extremely high unemployment rates and the failure of financial
institutions during the Great Depression,
which impoverished a large segment of

the nation's elderly. The system mandated contributions via payroll taxes during one's working years in exchange for
an implicit promise of retirement benefits. At its inception, the payroll tax was
set at 1 percent of earnings up to $3,000
from both employers and employees in
all private-sector commercial and industrial jobs.
The initial setup was designed to accumulate a trust fund that paid benefits
only to contributors, but it was amended
in 1939 to include those elderly who
had not contributed during their working years. This amounted to a switch
from a funded to a pay-as-you-go system and implied that benefits paid to the
initial generations of elderly Americans
would have to be covered by future generations. In addition, the 1939 legislation broke the link between individual
contributions and benefits by introducing dependent benefits.3
The pay-as-you-go system was in effect
until 1983, with current revenues being
immediately transferred to older generations in the form of retirement (as well
as other) benefits. After 1972, benefits
grew rapidly because of their overindexation to inflation, and long-range projections of revenue shortfalls worsened
— primarily as a result of lower post1960s fertility rates that translated into
fewer working-age individuals per
retiree early in the twenty-first century.
Hence, in 1977, a new indexation
method designed to maintain a constant
structure of replacement ratios — the
ratios of real benefits to real past earnings — was introduced.4
Unfortunately, this new approach proved
inadequate to rein in the revenue shortfall, because real benefits were tied to
real average wages rather than to the real
tax base: A smaller future labor force implied a smaller projected tax base despite
rising real average wages. To accommodate higher benefit levels, the 1977 legislation also instituted a gradual increase in
tax rates from 12.1 percent to 15.3 percent by 1990. The law further allowed
for a gradual rise in the maximum taxable income level. Thus, rather than
reduce benefit levels, lawmakers preferred to increase payroll taxes.

FIGURE 1 SOCIAL SECURITY INCOME,
OUTGO, AND ASSETS: 1985-2030a
Millions of dollars
Outgo

3,000
2,500 -

/

2,000

Income

Assets^/

1,500 -

\

1,000
500

\

0
-500

\

-1,000
1985

1990

1995

2000

2005

2010

2015

2020

2025

2030

a. Calendar-year data. Data for 1995 and beyond are projections.
b. End-of-year data.
SOURCE: Board of Trustees of the Social Security Administration, Annual Report, April 1995.

By contrast, the Social Security amendments passed in 1983 did cut benefits,
but the reductions were postponed until
the year 2000. The amendments also
boosted payroll tax rates and imposed a
greater reduction in benefits for early
retirees (those who leave their jobs at
age 62). In addition, a gradual increase
was imposed in the retirement age at
which individuals can qualify for full
benefits, and a quarter of Social Security
payouts became subject to income taxation.5 Recently, the Omnibus Budget
Reconciliation Act of 1993 increased the
taxable fraction to 85 percent.

• Is the Future of
Social Security Secure?
The 1983 amendments extended the
ability of the Social Security system to
meet its obligations for several decades.
Current projections indicate that the system will not run out of money until
about the year 2030 (see figure 1).
Indeed, the 1983 legislation partially
restored the funded nature of Social
Security. The payroll tax hikes, coupled
with the entry of the baby boomers into
the work force, are currently generating
a surplus of $63 billion per year —
funds that are earmarked for paying
benefits to future retirees.
This seemingly secure system may not,
however, translate into a secure future
for those Americans who are due to
retire in the first two decades of the next
century. The fact that the Social Security

Administration's books show adequate
funding for 30 years does not necessarily
mean that the money required to pay
benefits to retiring baby boomers will
actually be available. Individuals who
will turn 65 in the year 2010 may rationally expect to receive their Social Security checks each month. The crucial
question, however, is what the rest of the
federal budget will look like, and how it
will affect the retirement resources of
future retirees. Will most Social Security
benefits be subject to direct taxation?
Would high inflation in the interim erode
the real value of future benefits? Would
the government, if strapped for funds,
cut back on other spending to finance
Social Security payouts? Or might sales
and other consumption taxes be raised so
high that retirees, in attempting to spend
the money, will end up handing a major
share back to the government?
The ability of Congress to fund retirement benefits in the coming years
depends, of course, on whether lawmakers invest current surpluses in incomegenerating assets — income that could
be used to pay future benefits. If the
government spends its revenues on current consumption, providing benefits to
future retirees will be difficult. The law
now requires the Social Security trust
fund to invest its surplus in Treasury
securities. In reality, what this means is
that the funds are available to the government for current spending.6

What fraction of government spending is
devoted to investment? There is no clear
answer to this question. At one extreme,
all government nontransfer spending
could be thought of as investment, since
government operations provide the
economy with a framework of laws and
institutions and ensure national security,
allowing private enterprise to operate
smoothly and to prosper. At the other
extreme, most government nontransfer
spending could be viewed as current
consumption, since it does not result in
the creation of income-generating assets
for the government.
A third perspective is that government
spending on items like education, childhood nutrition, and so on, improves the
earning power of future generations. That
additional income can then serve as a
base for higher payroll taxes and help to
finance the retirement benefits of current
retirees. The efficacy of this channel is
questionable, however, because higher
(anticipated) payroll taxes might dampen
future generations' work incentives.
Hence, when the Treasury securities of
the Social Security trust fund need to be
redeemed to provide benefits to a swelling number of retirees, the absence of
corresponding income-generating assets
and limits on payroll taxes may mean
that future benefits will have to be taxed
or scaled back directly.7
The taxation of Social Security benefits
is already a reality. Today, however,
these benefits are subject only to an
income tax; thus, only retirees with high
asset incomes are affected — that is, the
rich elderly. Future taxation of Social
Security benefits may assume the form
of higher normal retirement ages or
taxes on consumption, both of which
would affect all recipients. Of course,
benefit levels may be reduced directly.
That seems less likely, however, given
that the relatively large number of future
retirees will represent a potent political
pressure group.

• Should We Privatize
Social Security?
Proposals for privatization assume various forms, and their likelihood for success should be judged by whether they
reduce or eliminate the major shortcomings of the current system without introducing new problems. Privatization
should not, of course, abrogate the
accrued benefits of retired generations or
of those about to retire.
Proposals for privatizing Social Security
generally recommend publicly mandated, employer-provided pension plans
for younger workers, say, those age 50
or less (such plans already exist in the
private sector). These workers and their
employers would be required to contribute to 401(k)-type pension plans
instead of toward Social Security payroll
taxes, with the resulting revenue shortfall met through general government
revenues. Such a plan, however, would
also entail a reduction of future implicit
obligations to pay benefits to today's
young workers and, as such, would
break even in present-value terms.
Under such a scheme, employees would
directly control the investment of their
contributions in stocks, bonds, or government securities, thereby reducing the
risk that the funds will be spent by the
government. Separate accounts for
employees and their spouses could be
established to tighten the link between
contributions and benefits, improving
work incentives. Such an approach
would also gradually eliminate the redistribution of resources from younger and
future generations toward older generations with larger consumption propensities. This would increase national saving
and ultimately boost investment and
economic growth.8
Critics point out that under privatization,
funds meant for financing retirement
would be invested in portfolios that
include stocks and hence would be at
greater risk than is currently the case.
However, this criticism loses much of its
force when one considers that investing
for retirement involves time horizons of
25 to 30 years or longer. The choice
between stocks and bonds (government
or corporate) is usually posed as one

between risk and return, with stocks
being the high-risk/high-return investment and bonds being the low-risk/lowreturn investment. Over long periods,
however, it may be that the worst return
earned on stocks will still exceed the best
return on bonds. If so, the risk-versusreturn characterization of the trade-off
would be irrelevant. Indeed, researchers
have demonstrated this to be true during
the 1926-1989 period for investment
horizons of more than 25 years.9
A second criticism is that the intergenerational risk-sharing that is possible under
public funding, as well as the goal of
redistributing monies from the rich to
the poor, may not be feasible under a privatized scheme. However, privatization
does not preclude transfers out of the
government's general budget to generations that suffer unfortunate economic
circumstances or to poor households
from rich ones. Indeed, the systematic
intergenerational resource transfers
entailed in a public pay-as-you-go system and the exposure of funds to government consumption under public funding
are probably far more detrimental to saving and long-term economic growth.

•

Conclusion

Notwithstanding the seemingly secure
future of Social Security for the next
three decades, financing future benefits
for a growing proportion of elderly
Americans will pose a serious economic
challenge. Although there are good reasons for public intervention in retirement
funding, the current financing and spending rules of the U.S. Social Security system contain several features that reduce
work incentives, expose funds to government consumption, and redistribute
resources from low- to high-spending
generations.
The idea of rectifying some of these defects through mandated contributions to
privatized pension schemes, with a
tighter link between benefits and contributions, is gaining ground in many
countries. It is time for U.S. lawmakers
to take a serious look at this option.

•

Footnotes

1. The oldest baby-boom generations are
expected to begin retiring in the year 2010.
2. The replacement rate refers to the ratio
of Social Security retirement benefits to preretirement income.
3. Even today, the existence of dependent
benefits breaks the link between contributions
and benefits, which is a work disincentive for
married women.
4. This method is the Average Monthly
Indexed Earnings (AIME) approach to computing the Primary Insurance Amount (PIA)
for each covered retiree. The computation
uses a wage index to inflate past covered
earnings and applies a regressive formula
to calculate the PIA, which is the monthly
retirement benefit.
5. The normal retirement age is scheduled to
increase from 65 to 67 over a span of 15 years
beginning in the year 2000.
6. Indeed, the stipulation that the trust fund
surplus must be invested in government securities may be causing additional government
spending by making the resources available to
the government at low cost. It is possible, of

Federal Reserve Bank of Cleveland
Research Department
P.O. Box 6387
Cleveland, OH 44101
Address Correction Requested:
Please send corrected mailing label to
the above address.
Material may be reprinted provided that
the source is credited. Please send copies
of reprinted materials to the editor.

course, that investing the surplus in private
capital markets would release an equal
amount of private capital for buying government debt, leaving the government's ability
to finance its spending unaffected. However,
if private lenders are unwilling to hold the
entire additional amount of public debt at the
rates of return being paid to the Social Security trust fund, the cost to the government of
financing the added debt would rise, making
it more difficult to sustain the current level of
public consumption expenditures.
7. Under current projections, trust fund
assets will have to be drawn down beginning
in the year 2020 (see figure 1).
8. The redistribution of resources from
younger to older generations is a consequence of two elements: the pay-as-you-go
nature of Social Security and the growth in
per capita benefits to each successive generation of retirees. Gradually eliminating the
pay-as-you-go setup would prevent growth
in the system from automatically causing an
intergenerational redistribution of resources.
If, under privatization, each generation's benefits were financed by its own prior savings,
benefit levels could rise only if those prior
contributions were increased.

9. See Thomas E. MaCurdy and John B.
Shoven, "Stocks, Bonds, and Pension
Wealth," in David Wise, ed., Topics in the
Economics of Aging, Washington, D.C.:
National Bureau of Economic Research,
1992. Data for years prior to 1926 do indicate periods in which bonds outperformed
stocks. Nonetheless, the overwhelming
weight of the evidence shows that stocks are
a better investment over long horizons. Of
course, the future behavior of capital markets
may not be consistent with past trends.

Jagadeesh Gokhale is an economic advisor
at the Federal Reserve Bank of Cleveland.
The author thanks Seung An for providing
detailed data on Social Security projections.
The views stated herein are those of the
author and not necessarily those of the Federal Reserve Bank of Cleveland or of the
Board of Governors of the Federal Reserve
System.

BULK RATE
U.S. Postage Paid
Cleveland, OH
Permit No. 385