United States General Accounting Office
GAO Testimony
Before the Committee on the Budget, U.S. Senate
For Release on Delivery
Expected at
9:30 a.m.
SOCIAL SECURITY
Tuesday,
February 23, 1999
AND SURPLUSES
GAO’s Perspective on
the President’s
Proposals
Statement of David M. Walker
Comptroller General of the United States
GAO/T-AIMD/HEHS-99-95
Mr. Chairman and Members of the Committee:
It is a pleasure to be here today to discuss the President’s recent proposal
for addressing Social Security and use of the budget surplus. These
proposals address some of the most important issues facing the nation,
both now and over the longer term. As you know, both GAO as an
institution and I as an individual have a long-standing interest in these
issues.
The President’s proposal is complex, which makes it all the more important
for us to focus our attention on what it does—and what it does not do—for
our long-term future. In summary, the President’s proposal:
• Reduces debt held by the public from current levels, thereby also
reducing net interest costs, raising national saving, and contributing to
future economic growth.
• Fundamentally changes Social Security financing in two ways:
• It promises general funds in the future by, in effect, trading publicly
held debt for debt held by the Social Security Trust Fund (SSTF).
• It invests some of the trust fund in equities with the goal of capturing
higher returns over the long term.
• Does not have any effect on the projected cash flow imbalance in the
Social Security program’s taxes and benefits, which begins in 2013.
• Does not represent a Social Security reform plan and does not come
close to “saving Social Security.”
Context: Long-Term It is important to look at the President’s proposal in the context of the fiscal
situation in which we find ourselves. After nearly 30 years of unified
Outlook Is Important budget deficits, we look ahead to projections for “surpluses as far as the
eye can see.” At the same time, we know that we face a demographic
tsunami in the future that poses significant challenges for the Social
Security system, Medicare, and our economy as a whole. In this context,
we should recognize that the President uses a longer-term framework for
resource allocation than has been customary in federal budgeting.
Although all projections are uncertain—and they get more uncertain the
farther out they go—we have long held that a long-term perspective is
important in formulating fiscal policy for the nation. Each generation is in
part the custodian for the economy it hands the next and the nation’s long-
term economic future depends in large part on today’s budget decisions.
This perspective is particularly important because our model and that of
Page 1 GAO/T-AIMD/HEHS-99-95
the Congressional Budget Office (CBO) continue to show that absent a
change in policy, the changing demographics to which I referred above will
lead to renewed deficits. This longer-term problem provides the critical
backdrop for making decisions about today’s surpluses.
Surpluses are the result of a good economy and difficult policy decisions.
They also provide a unique opportunity to put our nation on a more
sustainable path for the long term, both for fiscal policy and the Social
Security program itself. Current decisions can help in several important
respects: (1) current fiscal policy decisions can help expand the future
capacity of our economy by increasing national savings and investment,
(2) engaging in substantive reforms of retirement and health programs can
reduce future claims, (3) by acting now, we have the opportunity of
phasing in changes to Social Security and health programs over a sufficient
period of time to enable our citizens to adjust, and (4) failure to achieve
needed reforms in the Social Security and Medicare programs will drive
future spending to unsustainable levels and eventually “squeeze out” most
or all discretionary spending. If we let the achievement of a budget surplus
lull us into complacency about the budget, then in the middle of the
21st century, we could face daunting demographic challenges without
having built the economic capacity or program/policy reforms to handle
them.
The Proposal Before turning to the context for and analysis of the President’s proposal,
let me briefly describe it. The President proposes to use approximately
two-thirds of the total projected unified budget surpluses over the next
15 years to reduce debt held by the public and to address Social Security’s
financing problem. His approach to this, however, is extremely complex
and confusing. The President proposes to “transfer” an amount equal to a
portion of the projected surplus to the Social Security and Medicare trust
funds.1 This transfer is projected to extend the solvency of Social Security
from 2032 to 2049. His proposal to permit the trust fund to invest in
equities is expected to further extend trust fund solvency to 2055. He calls
on the Congress to work with him on program changes to get to 2075.
1
In this testimony, I will address only the Social Security portion of this transfer. The issues are similar
but not identical for the Medicare trust fund transfer.
Page 2 GAO/T-AIMD/HEHS-99-95
To understand and evaluate this proposal, it is important to understand the
nature of the federal budget, how trust funds fit into that budget, and the
challenges of “saving” within the federal budget.
Can We Save for the Future The federal budget is a vehicle for making choices about the allocation of
in the Federal Budget? scarce resources. It is different from state budgets in ways important to
this discussion. Most states use “fund budgeting” in which pension funds
that are separate and distinct legal entities, build up surpluses that are
routinely invested in assets outside the government (e.g., readily
marketable securities held in separate funds). In contrast, the federal
government’s unified budget shows all governmental transactions and all
funds are available for current activities, including current-year activities of
all trust funds. We cannot park our surplus in a cookie jar. The only way to
save in the federal budget is to run a surplus or purchase a financial asset.
When there is a cash surplus it is used to reduce debt held by the public.
Therefore, to the extent that there is an actual cash surplus, debt held by
the public falls.
This presents a problem for any attempt to “advance fund” all or part of
future Social Security benefits. Advance funding within the current
program would mean increasing the flows to the SSTF. Although it is
officially “off budget,” the fact remains that the SSTF is a governmental
fund. In the federal budget, trust funds are not like private trust funds.
They are simply budget accounts used to record receipts and expenditures
earmarked for specific purposes. A private trust fund can set aside money
for the future by increasing its assets. However, under current law, when
the SSTF’s receipts exceed costs, they are invested in Treasury securities
and used to meet current cash needs of the government. These securities
are an asset to the trust fund, but they are a claim on the Treasury. Any
increase in assets to the SSTF is an equal increase in claims on the
Treasury. One government fund is lending to another. The transactions net
out on the government’s books. Given this investment policy, any increase
in the trust fund balances would only become an increase in saving if this
increment were to add to the unified budget surplus (or decrease the
unified budget deficit) and thereby reduce the debt held by the public.
This is also the only way in which an increase in the SSTF balance could be
a form of advance funding.
How do these transactions affect the government’s debt? Gross federal
debt is the sum of debt held by the public and debt held by governmental
accounts—largely trust funds. This means that increases in the trust fund
Page 3 GAO/T-AIMD/HEHS-99-95
surplus will increase gross debt unless debt held by the public declines by
at least the same amount. Any reform of Social Security that increases the
annual SSTF surplus would increase debt held by government accounts
since, under current law, any excess of revenues over benefit payments is
loaned to Treasury for current needs. As a result, total government debt
would go up unless these surpluses were used to reduce debt held by the
public by an equivalent amount.
For most people, the different types of “debt” in the federal budget may be
confusing—especially since what is “good news” for a trust fund may be
“bad news” for total debt and vice versa. This is so because total debt (or
gross debt) is the sum of two very different types of debt—debt owed to the
public and debt owed by one part of the government (general fund) to
another part of the government (trust funds). Therefore, if a trust fund
surplus grows faster than debt held by the public falls, total debt grows—
even if the trust fund surplus is created as an attempt to “save” or to “pre-
fund” some of the future benefit payments. These contradictory
movements emphasize the need to differentiate between different types of
debt and what they mean. Both debt held by the public and debt held by
trust funds are important--but for different reasons. Analytically, therefore,
what is most important is not whether total debt increases but rather the
reasons behind the increase--does it represent an attempt to “advance
fund” through substantive reform or merely the promise of future
resources?
Debt held by the public and debt held by trust funds represent very
different concepts. Debt held by the public approximates the federal
government’s competition with other sectors in the credit markets. This
affects interest rates and private capital accumulation. Further, interest on
debt held by the public is a current burden on taxpayers. Reducing this
burden frees up capacity to meet future needs.
In contrast, debt held by trust funds performs an accounting function and
currently represents the cumulative annual surpluses of these funds (i.e.,
excess of receipts over disbursements plus accrued interest). Importantly,
debt held by the SSTF does not represent the actuarial present value of
expected future benefits for either current or future participants. Nor does
this debt have any of the economic effects of borrowing from the public. It
is not a current transaction of the government with the public; it does not
compete with the private sector for available funds in the credit market. It
reduces the need to borrow from the public and so may hold down interest
rates. Unlike debt held by the public, debt held by trust funds does not
Page 4 GAO/T-AIMD/HEHS-99-95
represent an immediate burden on current taxpayers. Rather, it is a claim
on future resources. The surplus is held in Treasury securities that give the
SSTF a claim on the Treasury equal to the value of those securities. When
the securities have to be redeemed, the Treasury must come up with the
cash. At that time, taxpayers will see some combination of a lower surplus,
lower spending, higher taxes, and/or greater borrowing from the public.
If borrowing from the public is increased to cover this cash need, there
could be upward pressure on interest rates. In addition, because debt held
by the trust fund is not equal to future benefit payments--it is not a measure
of the unfunded liability of the current system--it cannot be seen as a
measure of this future burden. Nevertheless, it provides an important
signal of the existence of this burden. Whether the debt constitutes a new
economic burden for the future or merely recognizes an existing one
depends on whether these currently promised benefits would be paid even
in the absence of the securities.
How Does the President’s This information is important to understand the President’s proposal
Proposal Work? because, in large part, he proposes a set of transactions that, in effect, trade
debt held by the public for debt held by the SSTF.2 By running a cash
surplus over the next 15 years, debt held by the public falls. To “save” this
surplus, the President proposes to “transfer” it to the trust fund in the form
of increased Treasury securities. Under his proposal, debt held by the
public falls, but debt held by the trust funds increases. Because he shows
the transfer as a subtraction from the surplus—a new budgetary concept—
he shows no surplus. As a result, he attempts to save some of the projected
surplus by hiding it.
The mechanics of the proposed transfer of surpluses to the SSTF are
complex and difficult to follow. Few details have been made available, and
there is conflicting information on exactly how it would work. Figures 1
and 2 are flowcharts representing our best understanding of the Social
Security portion of this transfer. Since it is impossible to understand the
changes proposed by the President without understanding the present
system, figure 1 shows the flows under the current system. Under current
law, annual cash flow surpluses (largely attributable to excess payroll taxes
over benefits payments and program expenses) are invested in Treasury
2
Paying down publicly held debt and issuing new special securities to the SSTF are two different
transactions. Nevertheless, the effect is as if the securities are exchanged.
Page 5 GAO/T-AIMD/HEHS-99-95
securities.3 This excess “cash” is commingled with other revenues and
used to finance other governmental activities. In this way, SSTF surpluses
have helped and continue to help finance the rest of the government. This
year, the SSTF surplus is expected to exceed the general fund deficit so
there is also a surplus in the unified budget. Over the entire 15-year period,
more than half of the projected unified surplus is composed of Social
Security surpluses. Absent any change in policy, these unified surpluses
will be used to reduce the debt held by the public.
Figure 1: Current Social Security Flows
Unified Budget
FICA Taxes Other Taxes
FICA Surplus
Social Security
General Fund
Trust Fund
Special Treasuries
Discretionary Mandatory
Benefits Spending Spending
*Unified surplus would
pay down the debt held
by the public
*Unified surplus = FICA surplus + general fund surplus
Source: GAO Analysis.
Under the President’s proposal, this would continue. However, as shown in
figure 2, at the point where total tax receipts are allocated to pay for
3
This presentation is somewhat simplified. In reality, FICA taxes are collected with income and
corporate taxes by the Treasury and then allocated by the Treasury to Social Security, Medicare, or the
general fund. In addition, a portion of income taxes paid on Social Security benefits flow into the SSTF.
The expenditure side of the SSTF transactions is also simplified since administrative expenses also
flow from the trust fund. These elements are unchanged by the President’s proposal and do not change
the flows critical to understanding it.
Page 6 GAO/T-AIMD/HEHS-99-95
government activities, a new financing step would be added to “transfer” a
portion of the unified budget surpluses to the Social Security and Medicare
trust funds. The unified budget would do this by providing a new set of
securities for these trust funds. However, the excess cash would still be
used to reduce the debt held by the public.
Figure 2: Social Security Flows Under President’s Proposal
Unified Budget
FICA Taxes Other Taxes
FICA Surplus
Social Security
Trust Fund General Fund
Special Treasuries
Discretionary Mandatory
Benefits Spending Spending
Transfer for
Trust Fund
Special treasuries and
Purchase of authority to purchase Unified surplus would pay down
equities equities the debt held by the public
Source: GAO Analysis.
In essence, this exchanges debt held by the public for debt held by the trust
funds. While there are many benefits to reducing publicly held debt, it is
important to recognize that under the current law baseline—i.e., with no
changes in tax or spending policy—this would happen without crediting
additional securities to the trust funds.
The administration has defended this approach as a way of assuring both a
reduction in debt held by the public and giving Social Security first claim
on what they call the “debt-reduction dividend” to pay future benefits.
However, issuing these additional securities to the SSTF is a discretionary
act with major legal and economic consequences for the future. Some
Page 7 GAO/T-AIMD/HEHS-99-95
could view this as double counting—or double-crediting. Importantly, to
the extent it appears that way to the public, it could undermine confidence
in a system that is already difficult to explain. However, the debate over
double counting focuses on the form of the proposal rather than its
substance. Although form is important when it interferes with our ability
to understand the substance—and I think this proposal falls into that
trap—the important debate must be on the substance of the proposal.
This proposal represents a fundamental shift in the way the Social Security
program is financed. It moves it away from payroll financing toward a
formal commitment of future general fund resources for the program. This
is unprecedented. Later in my statement, I will discuss the implications of
this proposal for overall fiscal policy and for the Social Security program.
Government Financing The President’s proposals would have the effect of reducing debt held by
the public from the current level of 44 percent of Gross Domestic Product
and Debt (GDP) to 7 percent over the 15-year period. The President notes that this
would be the lowest level since 1917. Nearly two-thirds of the projected
unified budget surplus would be used to reduce debt held by the public.
Because the surplus is also to be used for other governmental activities, the
amount of debt reduction achieved would be less than the baseline (i.e., a
situation in which none of the surplus was used), but nonetheless the
outcome would confer significant benefits to the budget and the economy.
Our previous work on the long-term effects of federal fiscal policy has
shown the substantial benefits of debt reduction.4 One is lowering the
burden of interest payments in the budget. Today, net interest represents
the third-largest “program” in the budget, after Social Security and Defense.
Interest payments, of course, are a function of both the amount of debt on
which interest is charged and the interest rate. Thus, at any given interest
rate, reducing publicly held debt reduces net interest payments within the
budget. For example, CBO estimates that the difference between spending
the surplus and saving the surplus is $123 billion in annual interest
payments by 2009--or almost $500 billion cumulatively between now and
then. Compared to spending the entire surplus, the President’s proposal
would also substantially reduce projected interest payments. Lower
interest payments lead to larger surpluses; these in turn lead to lower debt
4
Budget Issues: Analysis of Long-Term Fiscal Outlook (GAO/AIMD/OCE-98-19, October 22, 1997).
Page 8 GAO/T-AIMD/HEHS-99-95
which leads to lower interest payments and so on: the miracle of
compound interest produces a “virtuous circle.” The result would be to
provide increased budgetary flexibility for future decisionmakers who will
be faced with enormous and growing spending pressures from the aging
population.
For the economy, lowering debt levels increases national saving and frees
up resources for private investment. This in turn leads to increased
productivity and stronger economic growth over the long term. Over the
last several years, we and CBO have both simulated the long-term
economic results from various fiscal policy paths. These projections
consistently show that reducing debt held by the public increases national
income over the next 50 years, thereby making it easier for the nation to
meet future needs and commitments. Our latest simulations done for this
committee, as shown in figure 3, illustrate that any path that saves all or a
significant share of the surplus in the near term would produce
demonstrable gains in per capita GDP over the long run.5 This higher GDP
in turn would increase the nation’s economic capacity to handle all its
commitments in the future.
5
The “on-budget balance” path assumes that any surplus in the non-Social Security part of the budget is
“spent” on either a tax cut or spending increases or some combination but assumes the current law path
for the Social Security trust fund. Thus, the surplus in the Social Security trust fund remains untouched
until it disappears in 2013 after which the unified budget runs a deficit equal to the SSTF deficit. The
“Save the Surplus” path assumes no changes in current policies and that budget surpluses through 2024
are used to reduce debt held by the public. The “No Surplus” path assumes that permanent increases in
discretionary spending and tax cuts deplete the surpluses but keep the budget in balance through 2009.
Thereafter, deficits reemerge as spending pressures grow.
Page 9 GAO/T-AIMD/HEHS-99-95
Figure 3: GDP Per Capita Under Alternate Fiscal Policy Simulations
Per capita 1998 dollars
65,000
60,000
55,000
50,000
45,000
40,000
35,000
1998 level
30,000
25,000
1998 2002 2006 2010 2014 2018 2022 2026 2030 2034 2038 2042 2046 2050
On-budget balance Save the surplus No surplus
Source: GAO Analysis.
Under the President’s proposal, debt held by trust funds goes up more
rapidly than debt held by the public falls, largely due to these additional
securities transferred to the trust funds. Gross debt, therefore, increases.
It is gross debt—with minor exceptions—that is the measure that is subject
to the debt limit. The current limit is $5.95 trillion. Under the President’s
plan, the limit would need to be raised sometime during 2001. Under either
the CBO or the Office of Management and Budget baseline (i.e., save the
entire surplus), the limit would not need to be raised during at least the
next 10 years. Since other proposals to use the surplus would also bring
forward the time when the debt limit would have to be raised, the impact of
the President’s proposal on debt is in part a “compared to what?” question.
In figure 4, we show the debt subject to limit under the baseline, the
Page 10 GAO/T-AIMD/HEHS-99-95
President’s proposal, and a hypothetical path we have labeled “on-budget
balance.”6
Figure 4: Debt Subject to Limit Under Baseline and President’s Proposal
Dollars in billions
$9,000
$8,500
$8,000
$7,500
$7,000
$6,500
$6,000 Debt Limit $5,950
$5,500
$5,000
$4,500
$4,000
President's Budget with Social Security Proposal On-Budget Balance CBO Baseline
Source: OMB, CBO, Senate Budget Committee, and GAO Analysis.
Figures 5 and 6 below compare the composition of debt under the same
three paths: the baseline (save the entire surplus), the President’s proposal
(including both the Social Security proposal and the other spending), and
“on-budget balance.” Figure 5 shows debt held by the public under all
three scenarios, and figure 6 shows debt held by governmental accounts.
As figure 5 shows, debt held by the public falls under all three scenarios.
Since the baseline assumes the entire surplus is devoted to reducing debt
held by the public, it shows the greatest drop. Under the “on-budget
6
The baseline is the CBO baseline. It assumes that none of the surplus is used for tax cuts or spending
increases. “On-budget balance” assumes that any surplus in the non-Social Security part of the budget
is “spent” on either a tax cut or spending increases or some combination but that the surplus in the
Social Security trust fund remains untouched. There is no “on-budget” surplus until 2001.
Page 11 GAO/T-AIMD/HEHS-99-95
balance” path there are no tax cuts or spending increases until there is an
on-budget balance in 2001 while under the President’s proposal spending
increases and tax cuts are front-loaded. As a result, the President’s
proposal is projected to reduce debt held by the public less than the “on-
budget balance” path during these 10 years.
Figure 6 shows the impact of the President’s proposal to transfer securities
to the SSTF. The projections for debt held by government accounts are the
same for the baseline and the “on-budget balance” paths since neither
changes current law. Under the President’s proposal, however, debt held
by the SSTF increases as securities are transferred to it. This leads to the
increase shown in figure 6.
Figure 5: Composition of Federal Debt: Debt Held by the Public
Dollars in billions
$4,000
$3,500
$3,000
$2,500
$2,000
$1,500
$1,000
$500
$0
President's Budget with Social Security Proposal On-Budget Balance CBO Baseline
Source: OMB, CBO, Senate Budget Committee, and GAO Analysis.
Page 12 GAO/T-AIMD/HEHS-99-95
Figure 6: Composition of Debt: Debt Held by Government Accounts
Dollars in billions
$6,000
$5,000
$4,000
$3,000
$2,000
$1,000
$0
President's Budget with Social Security Proposal On-Budget Balance CBO Baseline
Note: Debt held by government accounts is the same under CBO baseline and the on-budget balance
path.
Source: OMB, CBO, Senate Budget, and GAO Analysis.
While reducing debt held by the public appears to be a centerpiece of the
proposal—and has significant benefits—as I noted above, the transfer of
unified surpluses to Social Security is a separate issue. The transfer is not
technically necessary: whenever revenue exceeds outlays and the cash
needs of the Treasury—whenever there is an actual surplus—debt held by
the public falls. The President’s proposal appears to be premised on the
belief that the only the way to sustain surpluses is to tie them to Social
Security. He has merged two separate questions: (1) how much of the
surplus should be devoted to reducing debt held by the public and (2) how
should the nation finance the Social Security program in the future.
Let me turn now to the question of Social Security financing.
Page 13 GAO/T-AIMD/HEHS-99-95
Social Security The President proposes two changes in the financing of Social Security: a
pledge of general funds in the future and a modest amount of investment in
Financing equities. Both of these represent major shifts in approach to financing the
program.
General Fund Financing By, in effect, trading debt held by the public for debt held by the trust funds,
the President is committing future general revenues to the Social Security
program. This is true because the newly transferred securities would be in
addition to any buildup of payroll tax surpluses. Securities held by the
SSTF have always represented annual cash flows in excess of benefits and
expenses, plus interest.7 Under the President’s proposal, this would no
longer continue to be true. The value of the securities held by the SSTF
would be greater than the amount by which annual revenues plus interest
exceed annual benefits and expenditures.
This means that for the first time there is an explicit general fund subsidy.
This is a major change in the underlying theoretical design of this program.
Whether you believe it is a major change in reality depends on what you
assume about the likely future use of general revenues under the current
circumstances. For example, current projections are that in 2032 the fund
will lack sufficient resources to pay the full promised benefits. If you
believe that this shortfall would—when the time came—be made up with
general fund moneys, then the shift embedded in the President’s proposal
merely makes that explicit. If, however, you believe that there would be
changes in the benefit or tax structure of the fund instead, then the
President’s proposal represents a very big change. In either case, the
question of bringing significant general revenues into the financing of
Social Security is a question that deserves full and open debate. The debate
should not be overshadowed by the accounting complexity and budgetary
confusion of the proposal.
One disconcerting aspect of the President’s proposal is that it appears that
the transfers to the trust fund would be made regardless of whether the
expected budget surpluses are actually realized. The amounts to be
7
Cash flow into the SSTF is composed of payroll taxes and a portion of the income taxes paid on Social
Security benefits. Income taxes make up a relatively small component of the surplus. Interest paid to
Social Security is analogous to interest paid on publicly held debt. Both come from the general fund.
Interest on publicly held debt is paid in cash while interest to the trust fund is credited in the form of
additional Treasury securities.
Page 14 GAO/T-AIMD/HEHS-99-95
transferred to Social Security apparently would be written into law as
either a fixed dollar amount or as a percentage of taxable payroll rather
than as a percentage of the actual unified surplus in any given year. These
transfers would have a claim on the general fund even if the actual surplus
fell below the amount specified for transfer to Social Security—and that
does present a risk.8 However, it is important to emphasize that any
proposal to allocate surpluses is vulnerable to the risk that those projected
surpluses may not materialize. Proposals making permanent changes to
use the surplus over a long period of time are especially vulnerable to this
risk.
The history of budget forecasts should remind us not to be complacent
about the certainty of these large projected surpluses. In its most recent
outlook book, CBO compared the actual deficits or surpluses for 1988-1998
with the first projection it produced 5 years before the start of each fiscal
year. Excluding the estimated impact of legislation, CBO says its errors
averaged about 13 percent of actual outlays. Such a shift in 2004 would
mean a surplus $250 billion higher or lower; in 2009, the swing would be
about $300 billion. Accordingly, we should consider carefully any
permanent commitments that are dependent on the realization of a long-
term forecast.
Investment in Equities Under current law, the SSTF is required to invest only in securities that are
issued or backed by the Treasury. The President proposes changing
current law to allow the SSTF to invest a portion of its assets in equities.
His proposal calls for the fund to gradually invest 15 percent of its total
assets in the equity market. According to the administration’s estimates,
the SSTF’s equity holdings would represent only a small portion—about 4
percent—of the total equity market. To insulate investment decisions from
political considerations, the administration proposes investing passively in
a broad-based stock index and creating an independent board to oversee
the portfolio.
8
It is worth noting that something like this happens now. Treasury does not track how much of the
revenues it collects are for Social Security and how much for income taxes. It credits the SSTF with
funds equal to the appropriate tax rate applied to the taxable wage base—whether or not those FICA
taxes were actually paid.
Page 15 GAO/T-AIMD/HEHS-99-95
Last year, we reported on the implications of allowing the SSTF to invest in
equities.9 In that report, we concluded that stock investing offers the
prospect of higher returns in exchange for greater risk. We found that, by
itself, stock investing was unlikely to solve Social Security’s long-term
financing imbalance but that it could reduce the size of other reforms
needed to restore the program’s solvency. We also concluded that investing
in a broad-based index would help reduce, but not eliminate, the possibility
of political influence over stock selections. However, the issue of how to
handle stock voting rights could prove more difficult to resolve. If the
government voted its shares, it would raise concerns about potential
federal involvement in corporate affairs. If the government chose not to
vote, it would affect corporate decision-making by enhancing the voting
power of other shareholders or investment managers. The model
applicable to passive private sector investment managers under the
Employee Retirement Income Security Act may be relevant to the
resolution of this issue.
Stock investing would have approximately the same impact on national
saving as using the same amount of money to reduce debt held by the
public. Both approaches would add about the same amount of funds to
private capital markets, meaning that national saving would essentially be
unchanged. From a budget accounting standpoint, they are not the same.
Under current scoring rules the purchase of equities would be counted as
an outlay, even though it is a financial transaction, because it is a transfer of
funds from a governmental entity to a nongovernmental entity. The
proposal apparently would change that. The administration proposes to
show the entire transfer to the SSTF as a reduction in the surplus and the
equity purchases would be part of that. The purchase of equities has
another financial impact: since part of the surplus would be used to
purchase equities, debt held by the public would be reduced less in the near
term than if that amount went to reduce publicly held debt. However, in
the future, claims on the Treasury would be lower because the program
would rely in part on stock sales to pay benefits.
9
Social Security Financing: Implications of Government Stock Investing for the Trust Fund, the Federal
Budget, and the Economy (GAO/AIMD/HEHS-98-74, April 22, 1998).
Page 16 GAO/T-AIMD/HEHS-99-95
Have Other Countries Although the dilemma we are facing of whether and how to save for the
future is a very difficult one, it is not unique. A look at other democracies
Tackled These shows that surpluses are difficult to sustain. However, several nations have
Problems? succeeded in sustaining surpluses. In those nations, political leaders were
able to articulate a compelling rationale to justify the need to set aside
current resources for future needs.
For example, those countries that have come to the conclusion that the
debt burden matters make it an explicit part of their fiscal decision-making
process. Australia, New Zealand, and the United Kingdom all attempt to
define prudent debt levels as a national goal to strive for. These debt goals
can prove important in times of surplus. New Zealand, for example, used
its debt goals as justification for maintaining spending restraint and
attempting to run sustained surpluses. They promised that once they met
their initial debt target they would give a tax cut. Importantly, when they
hit that specified debt target, they delivered on their promise of tax cuts.
Other countries have saved for the future by separating their pension or
Social Security-related assets from the rest of the government’s budget.
For example, the Canadian Pension Plan is completely separate from both
federal and provincial budgets. When the fund earns surplus cash, it is
invested in provincial debt securities and, starting this year, in the stock
market. Sweden also maintains a pension fund outside the government’s
budget and invests assets in stocks and bonds.
Norway may be the most dramatic example of setting aside current
surpluses to address long-term fiscal and economic concerns. Norway
faces the two-edged problem of a rapidly aging population and declining oil
revenues—a significant source of current government revenue. To address
these long-term concerns, Norway started setting aside year-end budget
surpluses in 1996 to be invested in foreign stocks and bonds. Their express
intention is to draw down these assets to pay for the retirement costs for
their baby boomers.
It should be noted that other nations that have attempted to directly
address their debt and pension problems have usually done so during or
shortly after a fiscal or economic crisis. Fortunately, we do not have that
problem. Instead, we have a unique opportunity to use our current good
fortune to meet the challenges of the future.
Page 17 GAO/T-AIMD/HEHS-99-95
Social Security Reform Finally, it is important to note that the President's proposal does not alter
the projected payroll tax and benefit imbalances in the Social Security
Is Still Needed program. In addition, it does not come close to “saving Social Security.”
Benefit costs and revenues currently associated with the program will not
be affected by even 1 cent. Figure 7, which shows Social Security's payroll
tax receipts and benefit payments, illustrates this point. Without the
President's proposal, payroll tax receipts will fall short of benefit payments
in 201310; with the President's proposal, payroll tax receipts also fall short
of benefit payments in 2013—the graph doesn't change at all. Under the
President’s proposal, expected stock market returns would be used to fill
part of this gap, but from 2013 on the trust funds will need cash from
redeemed Treasury securities, whether or not the President's proposal is
adopted.
10
Cash inflows actually consist of payroll taxes plus the income taxes paid on Social Security benefits.
Cash outflows are almost entirely made up of benefit payments, but they also include the fund’s
administrative expenses.
Page 18 GAO/T-AIMD/HEHS-99-95
Figure 7: SSTF Projected Cash Income and Outflow through 2019
Dollars in trillions
$1.6
$1.4
$1.2
$1.0
$0.8
$0.6
$0.4
$0.2
$0.0
08
16
02
04
07
11
19
99
13
98
00
01
03
05
12
14
15
18
06
09
10
17
20
20
20
20
20
20
20
19
20
19
20
20
20
20
20
20
20
20
20
20
20
20
Cash Income Cash Outgo
Source: Social Security Trustees 1998 Report, Intermediate Assumptions.
Under the President’s proposal, the changes to the Social Security program
will be more perceived than real: although the trust funds will appear to
have more resources as a result of the proposal, in reality, nothing about
the program has changed. The proposal does not represent Social Security
program reform, but rather a different means to finance the current
program. Although the President has called for bipartisan cooperation to
make programmatic changes, one of the risks of his proposal is that the
additional years of financing it provides could very well diminish the
urgency to achieve meaningful changes in the program. This would not be
in the overall best interests of the nation.
To achieve long-term solvency and sustainability, the Social Security
program itself must be reformed. The demographic trends that are driving
the program's financial problems affect the program well into the future.
The impending retirement of the baby boom generation is the best known
Page 19 GAO/T-AIMD/HEHS-99-95
of these trends, but is not the only challenge the system faces. If this were
so, perhaps a one-time financing strategy could be sufficient. But people
are retiring earlier, birth rates have fallen, and life expectancies are
increasing—all these factors suggest that Social Security's financial
problems will outlive the baby boom generation and continue far into the
future. These problems cannot be addressed without changes to the Social
Security program itself.
Changes to the Social Security system should be made sooner rather than
later. The longer meaningful action is delayed, the more severe such
actions will have to be in the future. Changes made today would be
relatively minor compared to what could be necessary years from now,
with less time for the fiscal effects of those changes to build. Moreover,
acting now would allow any benefit changes to be phased in gradually so
that participants would have time to adjust their saving or retirement goals
accordingly. It would be tragic indeed if this proposal, through its
budgetary accounting complexity, masked the urgency of the Social
Security solvency problem and served to delay much-needed action.
There is another reason to take action on Social Security now. Social
Security is not the only entitlement program needing urgent attention. In
fact, the issues surrounding the Medicare program are much more urgent
and complex. Furthermore, the many variables associated with health care
consumption and Medicare costs and the personal emotions associated
with health decisions make reform in this program particularly difficult.
To move into the future without changes in Social Security or health
programs is to envision a very different role for the federal government.
Assuming no financing or benefit changes, our long-term model (and that
of CBO) shows a world in 2050 in which Social Security and health care
absorb an increasing share of the federal budget. (See figure 8.) Budgetary
flexibility declines drastically and there is increasingly less room for
programs for national defense, the young, infrastructure, and law
enforcement—i.e., essentially no discretionary programs at all. Eventually,
again assuming no program or financing changes, Social Security, health,
and interest take nearly all the revenue the federal government takes in by
2050. This is true even if we assume that the entire surplus is saved and
these continued surpluses reduce interest from current levels. As shown in
figure 9, the picture below is even more dramatic if we assume the entire
Page 20 GAO/T-AIMD/HEHS-99-95
surplus is used.11 In that scenario, lower GDP and higher interest
payments lead to a world in which revenues cover only Social Security,
health, and interest in 2030. And in 2050 revenues don’t even cover Social
Security and health!
Figure 8: Composition of Spending as a Share of GDP Under “Save the Surplus”
Simulation
80%
70%
60%
50%
40%
30% Revenue
20%
10%
0%
1998 2030 2050
-10%
Social Security Medicare & Medicaid Net Interest All other spending
Source: GAO Analysis.
11
Our "No Surplus" simulation is not a forecast but rather an illustration of the implications of enacting
permanent tax cuts and/or spending increases that eliminate projected surpluses and the fiscal
pressures posed by the aging of the baby boom generation. This simulation shows ever-increasing
deficits that result in declining investment, a diminishing capital stock, and a collapsing economy. In
reality, these economic consequences would inevitably force policy changes to avert such a
catastrophic outcome.
Page 21 GAO/T-AIMD/HEHS-99-95
Figure 9: Composition of Spending as a Share of GDP Under “No Surplus”
Simulation
80%
70%
60%
50%
40%
30%
Revenue
20%
10%
0%
1998 2030 2050
Social Security Medicare & Medicaid Net Interest All other spending
Source: GAO Analysis.
Although views about the role of government differ, it seems unlikely that
many would advocate a government devoted solely to sending checks and
health care reimbursements to the elderly.
Let us address Social Security for the long term today so that the nation can
turn its attention to these other more pressing and difficult issues early in
the new millenium. Look again at figure 8: Social Security is not the fastest
growing portion of those bars—health grows faster.
Much remains to be done in reforming entitlement programs, and engaging
in meaningful Social Security reform would represent an important and
significant first step. The Congress and the administration, working
together, can find a comprehensive and sustainable solution to this
important challenge.
I recognize, though, that restoring Social Security solvency is not easy.
However, it is easy lifting compared to what faces us in connection with the
Page 22 GAO/T-AIMD/HEHS-99-95
Medicare program. Ultimately, any reforms to Social Security will address
not only the relatively narrow question of how to restore solvency and
assure sustainability but will also go to the larger question of what role
Social Security and the federal government should play in providing
retirement income. There are many proposals being made to address these
questions; choosing among them will involve difficult and complex choices,
choices that will be critically important to nearly every American’s
retirement income.
In my view, progress is likely to be greatest if we see these choices not as
“either/or” decisions but rather as an array of possibilities along a
continuum. Combining elements of different approaches may offer the
best chance to produce a package that addresses the problem
comprehensively for the long term in a way that is meaningful and
acceptable to the American people. For example, such a continuum may
identify individual accounts that could serve as a voluntary or mandatory
supplement to a financially sound and sustainable base defined benefit
structure. In addition, master trust principles can be used to provide for
collective investment of base defined benefit and individual account funds
in ways that would serve to prevent political manipulation of investments.
In order to help structure these choices, I would suggest five criteria for
evaluating possible Social Security proposals.
Sustainable solvency: A proposal should eliminate the gap between trust
fund resources and expenditures over 75 years, and have the ability to
sustain a stable system beyond that time period.
Equity: A proposal should create no "big winners" or "big losers." Those
who are most reliant on Social Security for retirement and disability
income should continue to receive adequate support; those who contribute
the most would also benefit from participation in the system, and
intergenerational equity would improve.
Adequacy: Consistent with Social Security’s social insurance feature, a
proposal should provide for a certain and secure defined benefit promise
that is geared to providing higher replacement rates for lower-income
workers and reasonable minimum benefits to minimize poverty among the
elderly.
Feasibility: A proposal should be structured so that it could be
implemented within a reasonable time period, it could be readily
Page 23 GAO/T-AIMD/HEHS-99-95
administered, and the administrative costs associated with it would be
reasonable.
Transparency: A proposal should be readily understandable to the general
public and, as a result, generate broad support.
Applying such criteria will require a detailed understanding of the possible
outcomes and issues associated with the various elements of proposals.
We are working to provide the data, information, and analysis needed to
help policymakers evaluate the relative merits of various proposals and
move toward agreement on a comprehensive Social Security reform
proposal.
Conclusions Budget surpluses provide a valuable opportunity to capture significant
long-term gains to both improve the nation’s capacity to address the
looming fiscal challenges arising from demographic change and aid in the
transition to a more sustainable Social Security program. The President’s
proposal may prompt a discussion and decision on both how much of our
current resources we want to save for the future and how we can best do
so. The President’s proposal is both wide ranging and complex, and it
behooves us to clarify the consequences for both our national economy and
the Social Security program.
A substantial share of the surpluses would be used to reduce publicly held
debt, providing demonstrable gains for our economic capacity to afford our
future commitments. In this way, the proposal would help us, in effect,
prefund these commitments by using today’s wealth earned by current
workers to enhance the resources for the next generations.
Saving a good portion of today’s surpluses can help future generations of
workers better afford the billowing costs of these commitments, but this is
only one side of the equation. We must also reform the programs
themselves to make these commitments more affordable. Even if we save
the entire surplus over the next 50 years Social Security and health
programs will double as a share of the economy and consume nearly all
federal revenues--essentially crowding out all other spending programs.
Thus, it is vital that any proposal to expand economic growth be
accompanied by real entitlement reform.
The transfer of surplus resources to the trust fund, which the
administration argues is necessary to lock in surpluses for the future,
Page 24 GAO/T-AIMD/HEHS-99-95
would nonetheless constitute a major shift in financing for the Social
Security program, but it would not constitute real Social Security reform
because it does not modify the program’s underlying commitments for the
future. Moreover, the proposed transfer may very well make it more
difficult for the public to understand and support the savings goals
articulated. Several other nations have shown how debt reduction itself
can be made to be publicly compelling, but only you can decide whether
such an approach will work here.
I am very concerned that enhancing the financial condition of the trust fund
alone without any comprehensive and substantive program reforms may, in
fact, undermine the case for fundamental program changes. In addition,
explicitly pledging federal general revenues to Social Security will limit the
options for dealing with other national issues.
The time has come for meaningful Social Security reform. Delay will only
serve to make the necessary changes more painful down the road. We must
be straight with the American people; achieving the goal of “saving Social
Security” will require real options to increase program revenues and/or
decrease program expenses. There is no “free lunch.” After all, we have
much larger and more complex challenges to tackle like the Medicare
program.
As you consider various proposals, you should consider the following
questions.
• How much of the unified budget surplus should go to debt reduction
versus other priorities?
• If we are to use some portion of the surplus to reduce publicly held debt,
is the President’s proposed approach the way to do this?
• Should Social Security be financed in part by general revenues?
• Should the SSTF invest in the stock market?
• How can we best assure the solvency, sustainability, equity, and integrity
of the Social Security program for current and future generations of
Americans?
• How can we best increase real savings for the future?
• How can we best assure the public’s understanding of and support for
any comprehensive Social Security reform proposal?
We at GAO stand ready to help you address both Social Security reform and
other critical national challenges. Working together, we can make a
positive and lasting difference for our country and the American people.
(935302) Leter Page 25 GAO/T-AIMD/HEHS-99-95
Ordering Information
The first copy of each GAO report and testimony is free.
Additional copies are $2 each. Orders should be sent to the
following address, accompanied by a check or money order made
out to the Superintendent of Documents, when necessary, VISA and
MasterCard credit cards are accepted, also.
Orders for 100 or more copies to be mailed to a single address are
discounted 25 percent.
Orders by mail:
U.S. General Accounting Office
P.O. Box 37050
Washington, DC 20013
or visit:
Room 1100
700 4th St. NW (corner of 4th and G Sts. NW)
U.S. General Accounting Office
Washington, DC
Orders may also be placed by calling (202) 512-6000
or by using fax number (202) 512-6061, or TDD (202) 512-2537.
Each day, GAO issues a list of newly available reports and
testimony. To receive facsimile copies of the daily list or any list
from the past 30 days, please call (202) 512-6000 using a touchtone
phone. A recorded menu will provide information on how to obtain
these lists.
For information on how to access GAO reports on the INTERNET,
send an e-mail message with “info” in the body to:
info@www.gao.gov
or visit GAO’s World Wide Web Home Page at:
http://www.gao.gov
United States Bulk Rate
General Accounting Office Postage & Fees Paid
Washington, D.C. 20548-0001 GAO
Permit No. GI00
Official Business
Penalty for Private Use $300
Address Correction Requested
Social Security and Surpluses: GAO's Perspective on the President's Proposals
Published by the Government Accountability Office on 1999-02-23.
Below is a raw (and likely hideous) rendition of the original report. (PDF)