JACOB J. LEW
OFFICE OF MANAGEMENT AND BUDGET
U.S. HOUSE OF REPRESENTATIVES
COMMITTEE ON THE BUDGET
May 20, 1999
Mr. Chairman, Representative Spratt, and Members of the Committee:
Thank you for the opportunity to appear before your committee today to discuss
H.R. 853, the "Comprehensive Budget Process Reform Act of 1999," as introduced by
Representatives Nussle and Cardin on February 25, 1999.
I would like to begin by emphasizing that the Budget Enforcement Act (BEA) has worked. The
BEA has imposed an essential discipline on discretionary spending by means of enforceable
discretionary spending caps. And the statutory pay-as-you-go (PAYGO) requirements have
ensured that new mandatory spending and new tax cuts are paid for with offsetting spending
reductions and revenue increases.
In short, the BEA's spending caps and PAYGO requirements have, over the last decade, helped
reduce and eliminate the deficit and produce a surplus for the first time in 29 years. These tools
for fiscal discipline, together with the 1993 and 1997 Budget Reconciliation Acts, have been key
to our success.
Moreover, it should be noted that the PAYGO rules have been instrumental in the President's
commitments to "save Social Security first" and to strengthen Medicare. By requiring that new
spending and tax cuts be fully paid for, PAYGO has effectively prevented the spending of
projected surpluses before the solvency of Social Security and Medicare have been secured. In
addition, PAYGO has started us down the road toward substantial debt reduction.
We have a process that has worked. Before we make any changes to the current budget rules, we
need to ask why the changes are needed, and to consider very carefully all of their consequences.
H.R. 853 would make several far-reaching changes to the current budget process. Transforming
the Concurrent Budget Resolution into a Joint Budget Resolution presented to the President for
signature, is a concept which this Administration has in the past supported.
However, as I will explain, the Administration is strongly opposed to the bill's serious weakening
of the PAYGO rules and its establishment of an automatic continuing resolution. In addition, we
have concerns about changes to the emergency procedures, the appropriations "lockbox" and
other provisions in the bill.
I. Repealing PAYGO in an Era of Surpluses.
H.R. 853 would effectively repeal PAYGO in an era of surpluses. It would amend the BEA to
permit on-budget surpluses to be spent on tax cuts or mandatory spending increases, without pay-as-you-go offsets. To understand fully the implications of this change, a brief review of the
PAYGO rules will be useful.
Background.--The Budget Enforcement Act of 1990 set up separate enforcement mechanisms
for: (1) discretionary spending; and (2) revenues and direct spending. These mechanisms --
annual caps on discretionary spending and a pay-as-you-go requirement for revenues and direct
spending -- replaced the largely ineffective Gramm-Rudman-Hollings regime of declining annual
The PAYGO process originally required that changes in direct spending and revenues, combined,
not increase the deficit in any year through FY 1995. The Budget Reconciliation Act of 1993
(OBRA-93) extended this requirement through FY 1998, and the 1997 Balanced Budget Act
(BBA) further extended PAYGO for all legislation enacted through 2002.
PAYGO applies not to each new law individually, but to the cumulative effect of all new laws
enacted since a designated starting point. The original starting point was all legislation enacted
subsequent to the 1990 BEA. The current starting point for PAYGO calculations is legislation
enacted since the BBA in 1997. OMB is required to maintain a "PAYGO scorecard" of both
deficit and savings effects from all direct spending and revenue legislation enacted since the BBA
and through 2002. Deficit effects of such legislation are calculated for the budget year and each
of the ensuing four years (so that PAYGO will be enforced through 2006, for legislation which is
enacted in 2002.)
OMB enforces the PAYGO requirements through "sequestration." If at the end of a
congressional session, the scorecard shows a combined net deficit increase (or surplus reduction)
for the fiscal years just-beginning and just-ended, OMB is required to implement across-the-board
cuts in all non-exempt direct spending programs in amounts sufficient to eliminate the deficit
increase (or restore the surplus). These across-the-board cuts are called sequestration. About 80
percent of outlays associated with direct spending programs are statutorily exempt from automatic
sequestration cuts. Exempt programs include Social Security, Federal retirement and disability
programs, net interest, certain low-income programs, veterans' compensation and pensions, and
regular State unemployment insurance benefits.
Under the automatic sequestration of non-exempt programs, the sequester calculations are made
so that two programs with automatic spending increases (COLAs) -- the special milk program and
vocational rehabilitation -- are cut first, followed by two special-rule programs (Stafford loans,
formerly called guaranteed student loans, and foster care and adoption assistance), and then
Medicare and the remaining non-exempt direct spending programs. Automatic cuts in Medicare
under PAYGO are limited to four percent, but there is no limit to the cuts which can be imposed
on non-exempt direct spending programs.
The PAYGO requirements apply only to new legislation, not to changes in spending levels under
existing law. For example, the estimated increase in mandatory spending resulting from a new
law that broadened a beneficiary population would have to be offset, or it would trigger a
sequester. However, if a beneficiary population as defined under existing law simply grew, the
increased spending would not have to be offset. This is the key to PAYGO's success: it holds
people responsible for legislative changes they can control - not for economic changes beyond
their direct control.
Under current law, PAYGO applies whether the Federal government is running a deficit or a
surplus. Therefore, tax cuts or direct spending increases that would cause a reduction in
on-budget surpluses must be fully offset, just as legislation causing or increasing on-budget
deficits must be offset.
Title VII of H.R. 853 would fundamentally change current law by permitting tax cuts or new
direct spending legislation to be enacted without offsets -- up to the amount of projected
on-budget surpluses. For example, the bill would permit a large tax cut or more spending to be
enacted without any offsets, as long as the amount of the tax cuts does not cause or increase an
on-budget deficit. This effectively repeals the pay-as-you-go requirement in an era of surpluses.
The Administration strongly opposes this repeal of the PAYGO rules. The Administration has
proposed a framework for allocating projected budget surpluses over the next 15 years, but
strongly believes that after Social Security and Medicare have been strengthened the pay-as-you-go
disciplines should continue - as they did following OBRA-93 and the BBA of '97. H.R. 853,
however, would set into permanent law the principle that any amount of projected on-budget
surpluses could be spent on new tax cuts or new direct spending programs without offsets.
To understand the dangers of this approach, consider this year's Congressional Budget
Resolution, H.Con.Res. 68 (Budget Resolution). The Budget Resolution calls for a tax cut of
$778 billion over the next ten fiscal years -- which amounts to nearly all of the currently projected
on-budget surpluses for that period. The PAYGO repeal called for in H.R. 853 would permit
enactment of these permanent, and very expensive, tax cuts without any offsetting revenues or
spending cuts. The tax cut would create large negative balances on the PAYGO scorecard for
each of the subsequent years on the assumption that these negative balances will be offset by the
actual surpluses when the time comes.
Now consider what would happen if the economy grows a bit more slowly than is currently
projected, and the on-budget surpluses over the next 10 years, in the absence of legislation, turn
out to be half of what is currently projected. The tax cuts would already be in permanent law, but
the surpluses which were supposed to finance the tax cuts would not have materialized. We could
face a net deficit big enough to trigger a 100 percent PAYGO sequestration. That means that
Medicare spending would be automatically cut by 4 percent; spending for all of the non-exempt
mandatory spending programs--child support enforcement, social services block grants, veterans
education and readjustment benefits, CCC, crop insurance, and others would be eliminated; and
there might still be some deficit remaining. Or, to avoid such a sequestration, Congress and the
Administration would be forced to slash selected mandatory and discretionary spending
programs. One of the principal reasons we need to maintain the fiscal discipline of the PAYGO
rules during a time of surplus, as well as during deficit periods, is the relative uncertainty of
The PAYGO rules have been and continue to be a pillar of fiscal discipline. They have saved the
surpluses for Social Security and Medicare, and have reduced our public debt. We urge the
Committee to maintain this discipline.
II. Automatic Continuing Resolution.
Title VI of H.R. 853 would establish an automatic continuing resolution (auto-CR) which would
continue funding at the previous year's levels, in the absence of regular appropriations. Similar
proposals have been under discussion in the past, particularly since the government shutdowns of
1995. The government shutdowns during the 104th Congress were unnecessary and very costly,
and -- as the President has said -- should never happen again.
However, an auto-CR is an irrational and unworkable response. Congress should not undermine
the ability to respond to a changing world by substituting an automatic funding mechanism for the
hard work and judgment that results from bicameral action and presidential approval.
In addition, under this bill, auto-CRs would last for the whole year, unless replaced by regular
appropriations. Full year CRs could therefore trigger a sequester if they result in spending levels
greater than the caps of that year.
An auto-CR could disrupt the funding of government programs in other ways. For example, an
auto-CR could be a powerful incentive for filibusters in the Senate. A minority of 41 in the Senate
could impose a freeze on selected programs - defense or non-defense -- simply by filibustering
the relevant appropriations bills. Alternatively, a minority of 41 could prevent program
reductions, where the savings are needed to fund higher priorities. In fact, such a minority could
perpetuate programs with no review or reform whatsoever.
In short, it is the Congress' constitutional responsibility to make decisions about appropriate
funding levels for the government's activities. Putting appropriations on auto-pilot would be a
I would remind the Committee that in 1997, the President vetoed an emergency flood
supplemental because it attempted to enact an auto-CR that would have undermined the
III. Emergency Spending.
Title II of H.R. 853 would repeal the BEA "emergency" procedures. Those procedures currently
provide for the upward adjustment of the discretionary spending caps to accommodate emergency
spending. For this purpose, spending is deemed an "emergency" when it is jointly designated as
such by the President and the Congress. (Though seldom used, the BEA also permits designation
of direct spending and revenue provisions as emergencies; in such cases, the costs of such
legislation are not placed on the PAYGO scorecard.)
H.R. 853 would replace the current-law emergency procedures with a requirement that both the
President's Budget and the Congressional Budget Resolution include a "reserve" for emergencies
that is not less than the average for emergency spending in the preceding five years.
Use of the reserve fund is made contingent upon the Budget Committee Chairmen certifying that
the spending meets a new statutory definition of "emergency." "Emergency" is defined in the bill
as a "situation that requires new BA and outlays...for the prevention or mitigation of, or response
to, loss of life or property, or a threat to national security; and is unanticipated"; the bill defines
"unanticipated" as "sudden,...urgent,...unforeseen,...and temporary."
In addition, under H.R. 853, any legislation which proposes emergency spending that would
exceed the emergency reserve would be automatically referred to the Budget Committees for not
more than 3 days. The Budget Committees would determine whether to report an amendment
exempting the emergency spending from the discretionary caps or PAYGO requirement, as
H.R. 853, as introduced, does not address the issue of whether the discretionary spending caps
would be adjusted upward to levels sufficient to accommodate inclusion of an "emergency
reserve." If there is an insufficient upward adjustment, the fencing off of funds for this emergency
reserve would make already extremely tight spending caps that much tighter. The Administration
would strongly oppose a significant tightening of the discretionary caps.
But even if there is an intention to fully adjust the caps for such a reserve, the Administration
would still have concerns about the advisability of this proposal in its current form. Consider
Table 1, which shows emergency spending in each year since enactment of the BEA. As the
table shows, emergency spending is by its very nature inherently unpredictable. If an emergency
reserve is created, based on a five-year average, it could end up being too little to cover
emergencies in some years; while in other years, it would end up being too much, which would
divert scarce resources from other needs.
Moreover, the President, the Congress, and the Nation need to be able to respond quickly to
emergencies - whether it is for military and humanitarian needs in Kosovo, aid to victims of
tornadoes, farmers struggling with low prices, or assistance desperately needed by hurricane or
earthquake victims. The current process, which permits emergency spending only when it is
jointly designated, in law, by the President and the Congress, can already take months. H.R. 853,
by contrast, would further encumber the process by requiring the Budget Committees to
determine whether particular emergencies meet a rigid statutory definition. This additional
encumbrance is unnecessary and could have very negative consequences when emergency relief
is urgently needed.
IV. Putting the Squeeze on Appropriations: the Lockbox and Baselines.
Title VI of the bill would establish procedures to give Members offering Floor amendments
cutting appropriations the option to allocate the savings either as offsets for other spending, or as
savings to go into a "lockbox." The lockbox savings would automatically reduce the
Appropriations Committees' 302 allocations; and by operation of language to be included in the
appropriations bills, would also reduce the statutory caps.
The Administration has concerns about this proposal because it has the potential to further reduce
already tight discretionary spending caps. In an era of very tight discretionary spending limits,
savings from lower priority appropriations should continue to be available for higher priorities.
In addition, the lockbox mechanism itself, is unworkable. For example, a Senator could offer an
amendment to reduce funding in a particular appropriation bill and direct all of the savings to the
lockbox. Even if the House cuts nothing from that bill, H.R. 853 requires that one-half of the
Senate's cut would remain in the lockbox -- automatically reducing the 302(a) allocations in the
Senate and in the House. So you could end up with a circumstance where a Senate amendment
has lowered the House Appropriations Committee's 302(a) allocation -- contrary to the levels the
House had adopted in the Budget Resolution.
In addition to appropriations being squeezed by a lockbox mechanism, H.R. 853 purports to
mandate that the Congressional Budget Office (CBO) and OMB use the prior fiscal year's level
without adjustment for inflation as their baselines for projections of discretionary spending in
future years. The results would be that when estimating future surpluses or deficits, the Congress
would be assuming a hard freeze on discretionary programs, rather than estimating the inflation-adjusted costs of continuing current services. The result would be a substantial under-estimate of
what it would cost to continue current government operations and services - resulting in more
pressure on discretionary appropriations.
V. Accrual budget for Federal insurance programs.
Title VI of the bill mandates budgeting for Federal insurance programs on the basis of the net
present value of the risk assumed in a given year, instead of the traditional cash basis of payouts
minus premium collections. This approach is generally analogous to budgeting for credit
programs under the Federal Credit Reform Act. The requirements would apply to deposit
insurance, pension guarantees, flood and crop insurance, the Overseas Private Investment
Corporation's insurance program, and other insurance programs. The bill provides for several
years of experimentation, publication of advisory estimates, and transparency for the models and
data used. In addition, it would require reports by OMB, CBO, and GAO on the feasibility of
risk-assumed budgeting for insurance programs. It would require the President actually to base
the budgets for insurance programs on risk-assumed estimates beginning with the FY 2006
We agree that risk-assumed estimates -- if they are reliable and well understood -- would have
considerable merit for scoring insurance programs in the budget. However, the use of this
methodology, outside of the comparatively ordered world of contractual arrangements between
lenders and borrowers, is sufficiently difficult that OMB would oppose a statutory deadline for its
implementation. Estimates for some programs could change substantially from year to year with
shifts in interest rates and other long-range assumptions. Producing the estimates would require
highly sophisticated estimating models that neither we nor the private sector have now or are
likely to have any time soon. Whether such models could be developed in time to meet the
requirements of the bill is highly uncertain. While we understand the bill sponsors' desire to set a
firm target date for implementing this change, we donot believe it is realistic at this time.
VI. Ten-year Limits on Program Authorizations and Entitlements.
Title IV of H.R. 853 requires committees to submit schedules for reauthorizing, within 10 years,
all programs in their jurisdiction, including entitlements. It also prohibits the consideration of
new direct spending programs unless their duration is limited to 10 or fewer years. And it
guarantees Members the right to offer amendments subjecting proposed entitlements to the
The apparent objective of this title is to limit the enactment of new entitlement benefits. The
Administration believes the right approach is not to put arbitrary roadblocks in the way of new
direct spending, but to maintain the current law PAYGO rules so that new direct spending is paid
for, and must compete against alternative uses of available funds. It is highly ironic that H.R. 853
on the one hand seeks to rein in the creation of new entitlement authority, at the same time that it
repeals the pay-as-you-go requirements when surpluses exist.
I want to thank the Committee for this opportunity to present the Administration's views on H.R.
853 and would be happy to answer any questions you may have.