ȱ
ȱǱȱȱȱȱ
ȱ ǯȱ ȱ
ȱȱȱȱ¢ȱ
ȱǯȱ ȱ
ȱȱȱȱ
ȱȱ
ȱȱȱ¢ȱ
¢ȱŘŝǰȱŘŖŖşȱ
ȱȱȱ
ŝȬśŝŖŖȱ
ǯǯȱ
ŚŖŗŖŚȱ
ȱȱȱ
Pr
epared for Members and Committees of Congress

ȱǱȱȱȱȱ
ȱ
¢ȱ
Recent policies have sought to contain damages spilling over from housing and financial markets
to the broader economy, including monetary policy, which is the responsibility of the Federal
Reserve, and fiscal policy, including a tax cut in February 2008 of $150 billion and two
extensions of unemployment compensation in June and November of 2008.
Over the past few months, the government has also intervened in specific financial markets,
including financial assistance to troubled firms, including legislation granting authority to the
Treasury Department to purchase $700 billion in assets. The broad intervention into the financial
markets has been passed to avoid the spread of financial instability into the broader market but
there are disadvantages, including leaving the government holding large amounts of mortgage
debt.
With the worsening performance of the economy beginning in September 2008, Congress passed
and President Obama signed a much larger stimulus packages comprised of spending and tax
cuts. The American Recovery and Reinvestment Act of 2009 (ARRA, P.L. 111-5), a $787 billion
package with $286 billion in tax cuts and the remainder in spending, was signed into law on
February 17, 2009. It includes spending for infrastructure, unemployment benefits, and food
stamps, revenue sharing with the States, middle class tax cuts, and business tax cuts.
The need for additional fiscal stimulus depends on the state of the economy. The National Bureau
of Economic Research (NBER), in December 2008, declared the economy in recession since
December 2007. Growth rates, after two strong quarters, were negative in the fourth quarter of
2007, positive in the first and second quarters of 2008, and a negative 0.5% in the third quarter.
Preliminary estimates put growth at a negative 6.2% for the fourth quarter of 2008, the worst
since 1982. According to one data series, employment fell in every month of 2008. The
unemployment rate, which rose slightly in the last half of 2007, declined in January and February
of 2008, but began rising in March and in January 2009 stood at 7.6%. Some forecasters believe
that the ongoing financial turmoil will result in a recession that is deeper and longer than average.
Fiscal policy temporarily stimulates the economy through an increase in the budget deficit which
leads to an increase in total spending in the economy, either through direct spending by the
government or spending by the recipients of tax cuts or government transfers. There is a
consensus that certain proposals, ones that result in more spending, can be implemented quickly,
and leave no long-term effect on the budget deficit, would increase the benefits and reduce the
costs of fiscal stimulus relative to other proposals. Economists generally agree that spending
proposals are somewhat more stimulative than tax cuts since part of a tax cut may be saved by the
recipients. The most important determinant of the effect on the economy is the stimulus’ size. The
2008 stimulus package increased the deficit by about 1% of GDP. The ARRA would increase the
budget deficit by about 1.3% in 2009 and an additional 2.2% (or 3.5% overall) in 2010. CBO
projects that the ARRA would raise GDP by a range of 1.4% to 3.8% in 2009 compared to what it
otherwise would have been.
ȱȱȱ

ȱǱȱȱȱȱ
ȱ
ȱ
The Current State of the Economy .................................................................................................. 1
The 2009 Stimulus Package ............................................................................................................ 4
Preliminary Discussions...................................................................................................... 4
House Proposal ................................................................................................................... 5
Senate Proposal................................................................................................................... 6
The American Recovery and Reinvestment Act of 2009 .................................................... 7
Discussion........................................................................................................................... 8
Issues Surrounding Fiscal Stimulus............................................................................................... 10
The Magnitude of a Stimulus .................................................................................................. 10
Bang for the Buck ....................................................................................................................11
Timeliness ............................................................................................................................... 14
Long-term Effects ................................................................................................................... 15
Should Stimulus be Targeted? ................................................................................................. 16
Is Additional Fiscal Stimulus Needed? ................................................................................... 16
Policies Previously Adopted ............................................................................................. 17
Interventions for Financial Firms and Markets ............................................................................. 18

ȱ
Table 1. Zandi’s Estimates of the Multiplier Effect for Various Policy Proposals ....................... 13
Table 2. Timing of Past Recessions and Stimulus Legislation ...................................................... 14

ȱ
Author Contact Information .......................................................................................................... 20

ȱȱȱ

ȱǱȱȱȱȱ
ȱ
he National Bureau of Economic Research (NBER) has declared the U.S. economy to be
in recession since December of 2007. With the worsening performance of the economy,
T congressional leaders and President Obama proposed much larger stimulus packages. The
American Recovery and Reinvestment Act of 2009 (ARRA), an $820 billion package with $275
billion in tax cuts (offset by a $7 billion gain from the treatment of built in losses) and the
remainder in spending, was passed by the House on January 28 (H.R. 1). It contained
infrastructure spending, revenue sharing with the States, middle class tax cuts, business tax cuts,
unemployment benefits, and food stamps. Similar legislation was passed in the Senate on
February 10 (a substitute for H.R. 1) and would cost $838 billion, with $292 billion in tax cuts.
The version of the bill signed into law on February 17, 2009 (P.L. 111-5) was a $787 billion
package with $286 billion in tax cuts and the remainder in spending.
Numerous actions have already been taken to contain damages spilling over from housing and
financial markets to the broader economy. These policies include traditional monetary and fiscal
policy, as well as federal interventions into the financial sector. In February 2008, in response to
weaker economic growth, an economic stimulus package of approximately $150 billion was
adopted.1 A provision that was considered (but not enacted) in the February stimulus bill was a
26-week extension of unemployment benefits; this extension was eventually enacted.2
A number of financial interventions have also been undertaken, before and after financial market
conditions worsened significantly in September 2008. The Federal Reserve has reduced short-
term interest rates to zero and introduced a number of facilities, providing direct assistance to the
financial system that would eventually surpass $1 trillion. In October 2008, legislation granting
the Treasury Department authority to purchase up to $700 billion in assets through the Troubled
Assets Relief Program (TARP) was adopted.3
This report first discusses the current state of the economy, including measures that have already
been taken by the monetary authorities. The next section reviews the economic stimulus package.
The following section assesses the need for, magnitude of, design of, and potential consequences
of fiscal stimulus. The final section of the report discusses recent and proposed financial
interventions.
ȱȱȱȱȱ¢Śȱ
The need for fiscal stimulus depends, by definition, on the state of the economy. According to the
National Bureau of Economic Research (NBER), the official arbiter of the business cycle, the

1 A second stimulus plan (H.R. 7110) passed the House on September 26, but was not passed by the Senate. It included
$36.9 billion on infrastructure ($12.8 billion highway and bridge, $7.5 billion water and sewer, $5 billion Corps of
Engineers); $6.5 billion in extended unemployment compensation; $14.5 billion in Medicaid, and $2.7 billion in food
stamp and nutrition programs.
2 For a discussion of the tax, housing, and unemployment legislation adopted in the 110th Congress see CRS Reports
RS22850, Tax provisions of the Economic Stimulus Package, by Jane G. Gravelle; RS22172, The Conforming Loan
Limit, by Eric Weiss and Mark Jickling, and RS22915, Temporary Extension of Unemployment Benefits: Emergency
Unemployment Compensation (EUC08), by Julie Whittaker.
3 See CRS Report RL34427, Financial Turmoil: Federal Reserve Policy Responses, by Marc Labonte, for a discussion
of Federal Reserve Policy and CRS Report RL34730, The Emergency Economic Stabilization Act and Current
Financial Turmoil: Issues and Analysis, by Baird Webel and Edward V. Murphy.
4 This section was prepared by Marc Labonte of the Government and Finance Division.
ȱȱȱ
ŗȱ

ȱǱȱȱȱȱ
ȱ
economy has been in recession since December 2007. It defines a recession as a “significant
decline in economic activity spread across the economy, lasting more than a few months” based
on a number of economic indicators, with an emphasis on trends in employment and income.5 But
because a recession is defined as a lasting decline, the NBER typically does not declare a
recession until it is well under way. The current recession was declared in late November of 2008.
After two strong quarters, economic growth fell by 0.2% in the fourth quarter of 2007. It then
increased by 0.9% in the first quarter of 2008 and 2.8% in the second quarter of 2008. Real GDP
decreased by 0.5% in the third quarter, however. Preliminary estimates of the fourth quarter
indicate a decline of 6.2%, the worst since 1982. The latest consensus forecast predicts that GDP
will continue to contract until the second half of 2009 and the rate of decline will accelerate, with
output falling by 1.9% for 2009 and unemployment reaching a high of 8.8%.6 If correct, this
recession would be the longest in the period since the Great Depression.
According to one data series, employment fell in every month of 2008. The deepening of the
downturn following September can also be seen in the movement of the unemployment rate. The
unemployment rate, which was 4.8% in February 2008, rose to 6.1% in August and September,
and has steadily risen since, reaching 7.6% in January 2009.
After a long and unprecedented housing boom, house prices have fallen 11% since their peak in
April 2007,7 and residential investment has fallen by almost half. This marked possibly the first
year of falling house prices since the Great Depression, according to one organization which
compiles the data.8 The decline in residential investment has acted as a drag on overall GDP
growth, while the other components of GDP grew at more healthy rates until the third quarter of
2008. Many economists argued that the housing boom was not fully caused by improvements in
economic fundamentals (such as rising incomes and lower mortgage rates), and instead
represented a housing bubble—a situation where prices were being pushed up by “irrational
exuberance.”9
Most economists believed that a housing downturn alone would not be enough to singlehandedly
cause a recession.10 But in August 2007, the housing downturn spilled over to widespread
financial turmoil.11 Triggered by a dramatic decline in the price of subprime mortgage-backed
securities and collateralized debt obligations, large losses and a decline in liquidity spread
throughout the financial system. Although the real production of goods and services showed
unexpected resilience until the fourth quarter of 2008, the ability of private borrowers to access
credit markets remained restricted throughout the year. Evidence of a credit crunch was seen in
the persistence of wide spreads between the interest rates that private borrowers paid for credit
and the yields on Treasury securities of comparable maturity.

5 National Bureau of Economic Research, The NBER’s Recession Dating Procedure, January 7, 2008.
6 Blue Chip, Economic Indicators, vol. 34, no. 2, February 10, 2009.
7 Based on the Federal Housing Finance Administration’s Purchase-Only House Price Index, a national measure of
single-family houses with conforming mortgages based on resale data.
8 Michael Grynbaum, “Home Prices Sank in 2007, and Buyers Hid,” New York Times, January 25, 2008. Prices are
compiled by the National Association of Realtors.
9 For more information, see CRS Report RL34244, Would a Housing Crash Cause a Recession?, by Marc Labonte.
10 See, for example, Frederic Mishkin, “Housing and the Monetary Transmission Mechanism,” working paper
presented at the Federal Reserve Bank of Kansas City symposium, August 2007.
11 See CRS Report RL34182, Financial Crisis? The Liquidity Crunch of August 2007, by Darryl E. Getter et al.
ȱȱȱ
Řȱ

ȱǱȱȱȱȱ
ȱ
The Federal Reserve (Fed) was forced to create unusually large amounts of liquidity to keep
short-term interest rates from rising in August 2007, and has since reduced interest rates
significantly. The Fed has gradually reduced the federal funds target rate from 5.25% to a range
of 0 to 0.25%, as of December 16, 2008. In addition, the Fed has lent directly to financial
institutions through an array of new facilities, and the amounts of loans outstanding has at times
exceeded a trillion dollars.12 A reduction in lending by financial institutions in response to
uncertainty or financial losses is another channel through which the economy entered a recession.
To date, financial markets remain volatile, new losses have been announced at major financial
institutions, and responses outside traditional monetary policy have been undertaken. Last March,
the financial firm Bear Stearns encountered liquidity problems, was purchased, after a plummet in
stock value, by JPMorgan Chase with financial assistance from the Fed. Then in July, the
government sponsored enterprises (GSEs) Fannie Mae and Freddie Mac experienced rapidly
falling equity prices in response to concerns about the value of their mortgage backed securities
assets. In July, Congress authorized Treasury to extend the GSEs an unlimited credit line (which
has not been utilized to date) in the Housing and Economic Recovery Act of 2008 (P.L. 110-289)
because of concern that the failure of a GSE would cause a systemic financial crisis. The federal
government took control of Fannie Mae and Freddie Mac in early September.
According to news reports in the fall of 2008, government officials decided not to intervene on
behalf of Lehman Brothers and Merrill Lynch;13 on September 14, Bank of America took over
Merrill Lynch without federal intervention, and on September 15, Lehman Brothers filed for
bankruptcy. The Treasury and Federal Reserve were trying to engineer a private bailout of the
nation’s largest insurance company, AIG, but on September 16 seized control with an $85 billion
emergency loan, which would later be increased.14
On September 18, Administration and Federal Reserve officials with the bipartisan support of the
Congressional leadership, announced a massive intervention in the financial markets.15 The
proposal asked for authority to purchase up to $700 billion in assets over the next two years. The
Treasury had also provided insurance for money market funds, where withdrawals have been
significant. These proposals suggested that government economists see problems with the
transmission of traditional monetary stimulus into the financial sector and ultimately into the
broader economy, where a significant contraction of credit could significantly reduce aggregate
demand. Although the legislation passed with some delay, the stock market fell significantly. The
original proposal had discussed buying mortgage related assets, particularly mortgage-backed
securities, but the Treasury indicated it will spend the initial $250 billion on preferred stock in
financial institutions. The Federal Reserve has also announced purchases of commercial paper,
$200 billion of asset backed securities, and $600 billion of mortgage-related securities; the
government has also announced a plan to guarantee certain assets of Citigroup and Bank of
America.

12 See CRS Report RL34427, Financial Turmoil: Federal Reserve Policy Responses, by Marc Labonte.
13 David Cho and Neil Irwin, “No Bailout: Feds Made New Policy Clear in One Intense Weekend,” Washington Post,
September 16, 2008, pp. A1, A6-A7.
14 Glenn Kessler and David S. Hilzenrath, “AIG at Risk; $700 Billion in Shareholder Value Vanishes,” Washington
Post, September 16, 2008; U.S. Seizes Control of AIG With $85 Billion Emergency Loan, Washington Post,
September 17, 2008, pp. A1, A8.
15 See CRS Report RS22957, Proposal to Allow Treasury to Buy Mortgage-Related Assets to Address Financial
Instability, by Edward V. Murphy and Baird Webel.
ȱȱȱ
řȱ

ȱǱȱȱȱȱ
ȱ
At the same time the economy and financial sector had been grappling with the housing
downturn, energy prices had risen significantly, from $48 per barrel in January 2007 to $115
dollars on April 30, 2008 and $144 as of July 2, 2008. After that, oil prices began a downward
trend, and had fallen below $70 by October and $60 by the end of November. The price reached
$43 per barrel on December 10; and remains at less than $50 per barrel. Most recessions since
World War II, including the most recent, have been preceded by an increase in energy prices.16
Energy prices had gone up almost continuously in the current expansion, however, without
causing a recession, which may point to the relative decline in importance of energy consumption
to production. Although a housing downturn, financial turmoil, or an energy shock might not be
enough to cause a recession in isolation, the combination was sufficient.
ȱŘŖŖşȱȱŗŝȱ
¢ȱȱ
On December 15, House Speaker Pelosi suggested a $600 billion package with $400 billion of
spending and $200 billion in tax cuts as a starting point for discussion. It was reported that the
package would include infrastructure spending, aid to the states, unemployment compensation,
and food stamps. Earlier, on December 11, Finance Committee Chairman Baucus suggested that
half of an expected $700 billion plan might be in tax cuts; he mentioned child tax credits, state
and local property tax deductions, the R&D tax credit, the marriage penalty, tax exempt bonds
and energy incentives. House Republican Leader Boehner proposed a tax package that included
increases in the child tax credit, suspending the capital gains tax on newly acquired assets,
increasing expensing, extending bonus depreciation and raising the share of costs expensed from
50% to 75%, extending net operating loss carrybacks to three years, lowering the corporate tax
rate from 35% to 25%, and expanding energy subsidies.
Reports on December 29 suggested that then President Elect Obama would propose a package of
$670 billion to $770 billion, but that additions in Congress might raise the total to $850 billion.
The package was reported to include $100 billion in aid to the States to fund Medicaid, possibly
with additional grants, and at least $350 billion for public works, alternative energy, health care
and school modernization, and expanding unemployment insurance and food stamp benefits. The
package would also include middle class tax cuts, possibly including changes to the child credit,
state and local property taxes, marriage penalties, the R&D tax credit and tax exempt bonds.
Following a meeting between President Elect Obama and Congressional leaders on January 5,
news reports indicated that the share of the package directed at tax cuts would increase to about
40%, perhaps $300 billion. President Elect Obama suggested a credit for working families of up
to $1,000 for couples and $500 for singles. Business provisions that were discussed included
extensions of the bonus depreciation and small business expensing enacted in February 2008 that
expired at the end of 2008 as well as an extended net operating loss carryback provision that was
discussed but not enacted in 2008. Also discussed was an expansion of the first-time homebuyers

16 For more information, see CRS Report RL31608, The Effects of Oil Shocks on the Economy: A Review of the
Empirical Evidence, by Marc Labonte.
17 This section was prepared by Jane Gravelle and Thomas Hungerford, Government and Finance Division.
ȱȱȱ
Śȱ

ȱǱȱȱȱȱ
ȱ
credit adopted in the 2008 housing legislation and expanding renewable energy incentives. A
payroll tax holiday was also discussed.
News reports on January 9 indicated some resistance of Congressional lawmakers to two
provisions in President Elect Obama’s plan: a $3,000 tax credit for employers who hire new
workers and the working families credit which provides for a credit of 6.2% of earnings up to a
ceiling of $500 for individuals and $1,000 for married couples. Some were concerned that the
employer tax credit would not benefit distressed firms and will be difficult to administer. There
were also concerns about the effects of a tax benefit of small magnitude having an effect if
reflected in withholding, although many economists suggest that a larger fraction of income
received in small increments is spent.
ȱȱ
The House proposal of the American Recovery and Reinvestment Act (H.R. 1) was comprised of
both spending and tax cuts. The spending proposals, which total $518.7 billion, include the
following:
• $54 billion for energy efficiency ($32 billion to improve the energy grid and encourage
renewable energy, $16 billion to repair and retrofit public housing; $6 billion to
weatherize modest income homes).
• $16 billion for science and technology ($10 billion for research, $6 billion to expand
broadband access in rural and underserved areas).
• $90 billion for infrastructure ($30 billion for highways, $31 billion for public
infrastructure that leads to energy cost savings, $19 billion for clean water, flood control
and environmental infrastructure, $10 billion for transit and rail).
• $141.6 billion for education ($41 billion to local school districts dedicated to specific
purposes, $79 billion to prevent cutbacks in state and local services including $39 billion
to local school districts and public colleges and universities distributed through existing
formulas, $15 billion to states for meeting performance measures, $25 billion to states for
other needs, $15.6 billion to increase the Pell grant by $500, $6 billion for higher
education modernization).
• $24.1 billion for health ($20 billion in health information technology and $4.1 billion for
preventive care).
• $102 billion for transfer payments ($43 billion for unemployment benefits and job transit
benefits, $39 billion to cover health care for unemployed workers, $20 billion for food
stamps)
• $91 billion to the States ($87 billion in general revenues by temporarily increasing the
Medicaid matching rate and $4 billion for law enforcement)
The proposal also contained $275 billion in tax cuts, which was reduced by a small revenue gain
from limits on built-in losses. The tax elements included:
ȱȱȱ
śȱ

ȱǱȱȱȱȱ
ȱ
• Temporary income tax cuts for individuals included the Making Work Pay tax credit—a
6.2% credit for earnings with a maximum of $500 for singles and $1,000 for couples,
phased out for taxpayers with incomes over $75,000 ($150,000 for joint returns)—with a
$145.3 billion 10-year cost, $4.7 billion for a temporary increase in the earned income
credit, $18.3 billion to make the child credit fully refundable, $13.7 billion to expand
higher education tax credits and make them 40% refundable (the refundability feature
accounts for $3.5 billion).
• Tax provisions for business, which would have lost revenue in FY2009-FY2010 and gain
revenue thereafter, included $37.8 billion for extending bonus depreciation, $59.1 billion
for a temporary five year loss carryback for 2008 and 2009 (except for recipients of
TARP funds; and electing firms would reduce losses by 10%), and $1.1 billion for
extending small business expensing.
• A series of provisions related to tax exempt bonds aimed at aiding State and local
governments, which would have cost $1.3 billion for FY2009-2010, but $37.3 billion
from FY2009-FY2019. Almost half the revenue loss would have arisen from allowing
taxable bond options which to make bonds attractive to tax exempt investors. Other major
provisions measured by dollar cost were qualified school construction bonds, recovery
zone bonds, and provisions to allow financial institutes more freedom to buy tax exempt
bonds.
• A permanent provision would have repealed the 3% withholding for government
contractors, which would not have lost revenue until 2011 and would have cost $10.9
billion for FY2009-2019.
• Energy provisions, some permanent and some temporary, would have totaled $5.4 billion
in FY2009-FY2011 and $20.0 billion in FY2009-2019. There was also a provision
substituting grants for credits for certain energy projects which would have shifted
benefits to the present.
• The proposal also included a provision to eliminate the requirement for paying back
credit for first time homebuyers unless they sell their homes within three years ($2.5
billion for 2009-2019). There was also a substitution of grants for the low income
housing credit, which would have shifted benefits to the current year ($3 billion).
• A minor provision ($208 million for FY2009-2019) would have provided incentives for
hiring unemployed veterans and disconnected youth.
• Repeal (prospectively) a Treasury ruling made in 2008 that allowed financial institutions
to carry over losses in an acquisition (gains $7 billion for FY2009-FY2010).
The bill passed the House on January 28, with an additional of $3.7 billion, primarily for transit
funds, bringing the total cost to $820 billion.
ȱȱ
The Senate passed a bill (a substitute for H.R. 1) with $838 billion in spending and tax cuts. The
higher cost was primarily due to the addition of the Alternative Minimum Tax (AMT) “patch”
(see below).
ȱȱȱ
Ŝȱ

ȱǱȱȱȱȱ
ȱ
The tax provisions were similar to the House tax bill in many respects. The bill, however, did not
fully refund the child credit and made 30% of tuition credits refundable, but allowed an exclusion
of up to $2,400 of unemployment benefits. The Senate bill also included a $15,000
homeownership tax credit at a 10-year cost of $35 billion and an above-the-deduction for certain
costs of a new automobile purchase ($11 billion 10-year cost). It would not have required net
operating losses to be reduced, as in the House bill. It added provisions for businesses including
an election to accelerate alternative minimum tax and research and experimentation credits in lieu
of bonus depreciation, a deferral of tax on income from cancellation of indebtedness, an increase
in the exclusion for small business stock. It also altered the size and mix of tax exempt bond
provisions, with the total cost of $22.6 billion, and changed some energy provisions. The bill also
included a $300 per adult payment to individuals eligible for Social Security, Railroad
Retirement, Veterans benefits, and Supplemental Security Income, at a cost of $17 billion, and
provided a one year increase in the Alternative Minimum Tax exemption (the AMT “patch”), at a
cost of $70 billion. Overall, the tax cuts were $368.4 billion. The measure also included $87
billion in Medicaid funding for the states, $20 billion to provide health insurance for unemployed
workers, and $16 billion to provide for health information technology.
The details of the spending provisions amounted to $290 billion in discretionary spending and
$260 billion in direct spending.18 One category of provisions would have provided $116 billion
for infrastructure and science including $5.9 billion for the Department of Homeland Security and
border stations, $7 billion for broadband technology, provisions in infrastructure and science for a
variety of federal programs (e.g. $4.6 billion for the corps of engineers, $9.3 billion for defense
and veterans), $27 billion for highways, $8.4 billion for mass transit, $10.9 billion for grants and
other transportation, $8.6 billion for public housing, $15 billion for environmental programs, and
$4.3 billion for science. The bill would have provided $84 billion for education and training, with
the majority, $79 billion, in grants to states and localities, and also included $13 billion in Title 1,
and $3.9 billion in Pell grants. Energy programs accounted for $43 billion; $23 billion would
have been provided for nutrition, early childhood and similar programs, and $14 billion for
health. The Senate bill also contained a limit on executive compensation at firms receiving
assistance from the Troubled Asset Relief Program (TARP).
ȱȱ¢ȱȱȱȱȱŘŖŖşȱ
The American Recovery and Reinvestment Act of 2009 (P.L. 111-5) was signed by President
Obama on February 17, 2009. The version of the Act signed into law has several provisions
similar to the House and Senate proposals. The total 10-year cost, at $787 billion, is lower than
both versions initially passed by the House and Senate, however. The spending parts of the Act
account for 63.7% of the total cost ($501.6 billion) and the tax provisions account for 36.3%
($285.6 billion). The Act includes the $70 billion AMT “patch” and an executive compensation
limitation for TARP recipients.
Many of the tax and spending provisions of the Act were scaled down from the House and Senate
proposals. The Making Work Pay tax credit was scaled back to $116.2 billion between FY2009
and FY2011 and provides a tax credit of up to $400 for a single taxpayer and $800 of joint
taxpayers. The temporary increase in the earned income credit is projected to cost $4.7 billion
over 10 years and the child tax credit is projected to cost $14.8 billion. The Act also includes a
$8,000 first-time homebuyer credit with a 10-year cost of $6.6 billion, an above-the-line

18 The direct spending also includes $83.7 billion in the refundable portion of tax credits.
ȱȱȱ
ŝȱ

ȱǱȱȱȱȱ
ȱ
deduction for sales tax on a new automotive purchase ($1.7 billion), a $2,400 exclusion of
unemployment compensation benefits ($4.7 billion), and a $250 payment to recipients of Social
Security, SSI, Railroad Retirement benefits, and certain veterans benefits ($14.2 billion).19
The business tax provisions are projected to lose $75.9 billion in revenues for FY2009 and
FY2010, but gain revenue in the future; the total 10-year revenue loss is projected to be $6.2
billion. Other business tax provisions include an extension of bonus depreciation ($5.1 billion
revenue loss), a five-year carryback of net operating losses for small businesses ($0.9 billion),
delayed recognition of certain cancellation of debt income ($1.6 billion), an increase in the small
business capital gains exclusion from 50% to 75% ($0.8 billion), and incentives to hire
unemployed veterans and disconnected youth ($0.2 billion).
The energy tax provisions amount to $20 billion over 10 years and $3.4 billion for FY2009-
FY2011. The major energy provision is a long-term extension and modification of renewable
energy production tax credits ($13.1 billion over 10 years). The Act alters the size and mix of tax
exempt bond provisions with a total 10-year cost of $25 billion.
The discretionary appropriations provisions of the Act provide $311 billion in appropriations
between FY2009 and FY2019. Investments in infrastructure and science account for $120 billion
and education and training programs are to receive $106 billion. Health programs are set to
receive $14.2 billion, the Supplemental Nutrition Assistance Program (formerly food stamps) will
receive $20 billion, Head Start $2.1 billion, and $2 billion for the child care development block
grant. Almost $40 billion will be used for investments in energy infrastructure and programs.
Increases for direct spending programs include $57.3 billion for assistance to unemployed
workers and struggling families, $25.1 billion for health insurance assistance, $20.8 billion for
health information technology, and $90.0 billion for state fiscal relief.
ȱ
Fiscal stimulus is only effective when the policy options increase aggregate demand. Many
economists view fiscal policy as less effective than monetary policy in an open economy. As
mentioned earlier in this report, however, several monetary policy options have already been
employed for several months.
Fiscal stimulus can involve tax cuts, spending, or a combination of both. Tax cuts may be less
effective than spending because some of the tax cut may be saved, which diminishes the
effectiveness of the stimulus. Some argue that tax cuts that are temporary, that appear in a lump
sum rather than in withholding, or that are aimed at higher income individuals are more likely to
be saved. Some evidence suggests that two-thirds of the 2001 tax rebate was spent within two
quarters.
The challenge to spending programs is that there may be a lag time for planning and
administration before the money is spent. For that reason, infrastructure spending is often
discussed in the context of “ready-to-go” projects where all of the planning is in place and the
only missing factor is funding. The U.S. Conference of Mayors has identified $73 billion of these
projects and urged some funds to be given directly to localities; the American Association of State

19 There is also a $250 tax credit for federal and state pensioners not eligible for Social Security with a $218 million 10-
year cost.
ȱȱȱ
Şȱ

ȱǱȱȱȱȱ
ȱ
highway and Transportation Officials has identified $64 billion of these projects; the National
Association of Counties has identified $9.9 billion. Some analysts suggest that aid to state and
local governments may be more quickly spent because these governments are likely to cut back
on spending in downturns due to balanced budget requirements, and the aid may forestall these
cuts.20 The Congressional Budget Office (CBO) score for the spending (discretionary and direct)
portion of the Act estimates that about 21% will be spent in FY2009, and 38% in FY2010.21
Overall, about 74% of the spending and tax provisions are estimated to reach the public by the
end of FY2010. However, if the AMT “patch” is omitted then about 70% is estimated to reach the
public by the end of FY2010.
The receipt of tax cuts can also be delayed. For example, according to Joint Committee on
Taxation estimates of the Making Work Pay credit revenue losses, 17% of the total would be
received in FY2009.22 The benefit is provided in the form of withholding; since the measure was
not in place on January 1, some benefit would be delayed until tax returns are filed. Close to 50%
would be received in FY2009 if a rebate mechanism were used (based on estimates of a similar
provision considered in 2008 at about the same time of the year, 93% of the rebate was projected
to be received in the current fiscal year). There is some limited evidence that periodic payments
are more likely to be spent than lump sum payments, but that evidence is subject to uncertainty
and is not of a magnitude that the withholding approach would result in a larger short run
stimulus than a rebate.23 In the second year, 57% would be received.

20 See CRS Report R40107, The Role of Public Works Infrastructure Spending in Economic Stimulus, by Claudia
Copeland, et al., and CRS Report 92-939, Countercyclical Job Creation Programs, by Linda Levine for a discussion of
some of these issues.
21 CBO, Letter to the Honorable Nancy Pelosi, Estimated Budget Impact of the Conference Agreement for H.R. 1,
February 13, 2009.
22[http://www.house.gov/jct/x-19-09.pdf].
23 This issue does not address the difference between temporary and permanent tax cuts; economists expect the latter to
have more effect on consumption, but a permanent tax cut would result in budget pressures after recovery. Alan S.
Blinder, “Temporary Income Taxes and Consumer Spending” The Journal of Political Economy, Vol. 89, February
1981, pp. 26-53, found the rebate 38% as effective as a permanent change and a withholding approach 50%, suggesting
that the rebate would be 75% as effective as withholding. James M. Poterba, “Are Consumers Forward Looking?”
American Economic Review, Vol. 78, (May 1988), pp. 413-418 found only 20% spent. Many economists have
reservations about estimates using aggregate data, however, because of the difficulties of determining the
counterfactual. For that reason, many researchers turned to comparisons of households with different amounts of tax
cuts. Two studies of spending out of refunds (lump sum receipts) and spending out of withholding in the first Reagan
tax cut found that 35% to 60% of refunds were spent but 60% to 90% of the withholding was spent (See Nicholas
Souleles, “The Response of Household Consumption to Income Tax Refunds,” American Economic Review, vol. 89
(September 1999), pp. 947-958; and Nicholas Souleles, “Consumer Response to the Reagan Tax Cuts,” Journal of
Public Economics. Vol. 85, pp. 99-120.). This research suggests a significant fraction of a temporary tax cut is spent,
but that the lump sum has an effect that is about two thirds of the effect of withholding. This comparison is, however,
somewhat clouded by the possibility that individuals may use tax refunds as a method of forced savings and not intend
to spend them. In both cases, however, there is evidence of an effect for temporary tax cuts. Research on the 2001
rebate also indicates a significant amount was spent: David S. Johnson, Jonathan A. Parker, and Nicholas S. Souleles,
Household Expenditures and the Income Tax Rebate of 2001,”American Economic Review, Vol. 96, December 2006,
pp. 1589-1610 find over two thirds spent within two quarters. For other research see CRS Report RS21126, Tax Cuts
and Economic Stimulus: How Effective Are the Alternatives?, by Jane G. Gravelle. Not included in that discussion are
survey data asking individuals about their spending, as individuals themselves may not know what they spent. A
preliminary study of the 2008 rebate also found significant spending: Christian Broda and Jonathan Parker, “The
Impact of the 2008 Tax Rebates on Consumer Spending: Preliminary Evidence,” Mimeo, University of Chicago and
Northwestern University, July 29,2008: [http://online.wsj.com/public/resources/documents/WSJ-
2008StimulusStudy.pdf]
ȱȱȱ
şȱ

ȱǱȱȱȱȱ
ȱ
Several studies have estimated the effects of the proposed package on the economy. Romer and
Bernstein estimate an increase of 3.7 million jobs by the fourth quarter of 2010; Zandi estimates
3.3 million in 2010.24 Citing uncertainty surrounding the effects of fiscal stimulus, CBO projects
that the ARRA would boost GDP in 2009 by a range of 1.4% to 3.8% and employment by a range
of 0.8 million to 2.3 million compared to what it otherwise would have been. In 2010, CBO
projects that the ARRA would boost GDP by 1.1% to 3.3% and employment by 1.2 million to 3.6
million in 2010 compared to what it otherwise would have been. Starting in 2014, CBO projects
that the ARRA would cause GDP to be slightly lower than it otherwise would have been due to
“crowding out” effects described in the section titled “Long-term Effects”.
ȱȱȱŘśȱ
ȱȱȱȱȱ
The most important determinant of a stimulus’ macroeconomic effect is its size. The 2008
stimulus package (P.L. 110-185) increased the budget deficit by about 1% of gross domestic
product (GDP). In a healthy year, GDP grows about 3%. In the moderate recessions that the U.S.
experienced in 1990-1991 and 2001, GDP contracted in some quarters by 0.5% to 3%. (The U.S.
economy has not experienced contraction in a full calendar year since 1991.) Thus, a swing from
expansion to recession would result in a change in GDP growth equal to at least 3.5 percentage
points. A stimulus package of 1% of GDP could be expected to increase total spending by about
1%.26 To the extent that spending begets new spending, there could be a multiplier effect that
makes the total increase in spending larger than the increase in the deficit. Offsetting the
multiplier effect, the increase in spending could be neutralized if it results in crowding out of
investment spending, a larger trade deficit, or higher inflation. The extent to which the increase in
spending would be offset by these three factors depends on how quickly the economy is growing
at the time of the stimulus—an increase in the budget deficit would lead to less of an increase in
spending if the economy were growing faster.
Thus, if the recession is mild, additional stimulus may not be necessary for the economy to
revive. If, on the other hand, the economy has entered a deeper, prolonged recession, as some
economists believe to be likely, then fiscal stimulus may not be powerful enough to avoid it.
Since the current recession has already lasted longer than the historical average, it may end before
further fiscal stimulus can be enacted. Economic forecasts are notoriously inaccurate due to the
highly complex and changing nature of the economy, so there is significant uncertainty as to how
deep the downturn will be, and how much fiscal stimulus would be an appropriate response.
The American Recovery and Reinvestment Act of 2009 will increase the budget deficit by about
1.3% in 2009 and an additional 2.2% (or 3.5% overall) in 2010. Some believe that circumstances

24 Christina Romer and Jared Bernstein, “The Job impact of the American Recovery and Reinvestment Plan, Chair,
Nominee Designate Council of Economic Advisors and Office of the Vice President Elect, January 9, 2009,
[http://otrans.3cdn.net/45593e8ecbd339d074_l3m6bt1te.pdf]; Mark Zandi, “The Economic Impact of the American
Recovery and Reinvestment Act,” January 21, 2009, [http://www.economy.com/mark-zandi/documents/
Economic_Stimulus_House_Plan_012109.pdf].
25 This section was prepared by Marc Labonte, Government and Finance Division.
26 See, for example, “Options for Responding to Short-term Economic Weakness,” Testimony of CBO Director Peter
Orszag before the Committee on Finance, January 22, 2008.
ȱȱȱ
ŗŖȱ

ȱǱȱȱȱȱ
ȱ
warrant a larger stimulus – GDP in FY2009 is expected to contract by more than size of the
stimulus in 2009. Others have expressed reservations that the deficit is already too large and, at
least with respect to spending, it would be difficult to spend such large amounts without financing
wasteful projects. Although the Act authorizes $379 billion of spending in 2009, CBO estimates a
outlays of only $120 billion because it does not project that executive agencies can spend the total
amount authorized in this fiscal year.
ȱȱȱȱ
In terms of first-order effects, any stimulus proposal that is deficit financed would increase total
spending in the economy.27 For second-order effects, different proposals could get modestly more
“bang for the buck” than others if they result in more total spending. If the goal of stimulus is to
maximize the boost to total spending while minimizing the increase in the budget deficit (in order
to minimize the deleterious effects of “crowding out”), then maximum bang for the buck would
be desirable. The primary way to achieve the most bang for the buck is by choosing policies that
result in spending, not saving.28 Direct government spending on goods and services would
therefore lead to the most bang for the buck since none of it would be saved. The largest
categories of direct federal spending are national defense, health, infrastructure, public order and
safety, and natural resources.29
Higher government transfer payments, such as extended unemployment compensation benefits or
increased food stamps, or tax cuts could theoretically be spent or saved by their recipients.30
While there is no way to be certain how to target a stimulus package toward recipients who would
spend it, many economists have reasoned that higher income recipients would save more than
lower income recipients since U.S. saving is highly correlated with income. For example, two-
thirds of families in the bottom 20% of the income distribution did not save at all in 2004,
whereas only one-fifth of families in the top 10% of the income distribution did not save.31
Presumably, recipients in economic distress, such as those receiving unemployment benefits,
would be even more likely to spend a transfer or tax cut than a typical family.
The effectiveness of tax cuts also depends on their nature. As discussed above, tax cuts received
by lower income individuals are more likely to be spent. Some economists have also argued that
temporary individual tax cuts, such as the 2001 and 2008 rebates, are more likely to be saved;
however, evidence on the 2001 tax rebate suggests most was eventually spent, and debate
continues on the effect of the 2008 rebate. Most evidence does not suggest that business tax cuts
would provide significant short-term stimulus. Investment incentives are attractive, if they work,
because increasing investment does not trade off short term stimulus benefits for a reduction in

27 There may be a few proposals that would not increase spending. For example, increasing tax incentives to save
would probably not increase spending significantly. These examples are arguably exceptions that prove the rule.
28 Policies that result in more bang for the buck also result in more crowding out of investment spending, which could
reduce the long-term size of the economy (unless the policy change increases public investment or induces private
investment).
29 For the purpose of this discussion, government transfer payments, such as entitlement benefits, are not classified as
government spending.
30 Food stamps cannot be directly saved since they can only be used on qualifying purchases, but a recipient could
theoretically keep their overall consumption constant by increasing their other saving.
31 Brian Bucks et al., “Recent Changes in U.S. Family Finances: Evidence from the 2001 and 2004 Survey of
Consumer Finances,” Federal Reserve Bulletin, vol. 92, February 2006, pp. A1-A38.
ȱȱȱ
ŗŗȱ

ȱǱȱȱȱȱ
ȱ
capital formation, as do provisions stimulating consumption. Nevertheless, most evidence does
not suggest these provisions work very well to induce short-term spending. This lack of
effectiveness may occur because of planning lags or because stimulus is generally provided
during economic slowdowns when excess capacity may already exist. Of business tax provisions,
investment subsidies are more effective than rate cuts, but there is little evidence to support much
stimulus effect. Temporary bonus depreciation is likely to be most effective in stimulating
investment, more effective than a much costlier permanent investment incentive because it
encourages the speed-up of investment. Although there is some dispute, most evidence on bonus
depreciation enacted in 2002 nevertheless suggests that it had little effect in stimulating
investment and that even if the effects were pronounced, the benefit was too small to have an
appreciable effect on the economy. The likelihood of the remaining provisions having much of an
incentive effect is even smaller. Firms may, for example, benefit from the small business
expensing, but it actually discourages investment in the (expanded) phase out range.32 Net
operating losses carrybacks do not increase incentives to spend, but do target cash to troubled
businesses.
Mark Zandi of Moody’s Economy.com has estimated multiplier effects for several different policy
options, as shown in Table 1.33 The multiplier estimates the increase in total spending in the
economy that would result from a dollar spent on a given policy option. Zandi does not explain
how these multipliers were estimated, other than to say that they were calculated using his firm’s
macroeconomic model. Therefore, it is difficult to offer a thorough analysis of the estimates. In
general, many of the assumptions that would be needed to calculate these estimates are widely
disputed (notably, the difference in marginal propensity to consume among different recipients
and the size of multipliers in general), and no macroeconomic model has a highly successful track
record predicting economic activity. Thus, the range of values that other economists would assign
to these estimates is probably large. Qualitatively, most economists would likely agree with the
general thrust of his estimates, however—spending provisions have higher multipliers because
tax cuts are partially saved, and some types of tax cuts are more likely to be saved by their
recipients than others. As discussed above, a noticeable increase in consumption spending has not
yet accompanied the receipt of the rebates from the first stimulus package. (Note, however, that
these effects do not account for the possibility of extensive delay in direct spending taking place.)

32 For more information, see CRS Report RS21136, Government Spending or Tax Reduction: Which Might Add More
Stimulus to the Economy?, by Marc Labonte; CRS Report RS21126, Tax Cuts and Economic Stimulus: How Effective
Are the Alternatives?, by Jane G. Gravelle; CRS Report RL31134, Using Business Tax Cuts to Stimulate the Economy,
by Jane G. Gravelle; and CRS Report RS22790, Tax Cuts for Short-Run Economic Stimulus: Recent Experiences,
coordinated by Jane G. Gravelle. Also see Fiscal Policy for the Crisis, IMF Staff Position Note, December 29, 2008,
SPN/08/01 [http://www.imf.org/external/np/pp/eng/2008/122308.pdf].
33 Mark Zandi, “Washington Throws the Economy a Rope,” Dismal Scientist, Moody’s Economy.com, January 22,
2008.
ȱȱȱ
ŗŘȱ

ȱǱȱȱȱȱ
ȱ
Table 1. Zandi’s Estimates of the Multiplier Ef ect for
Various Policy Proposals
One-year change in real GDP
Policy Proposal
for a given policy change per
dollar
Tax Provisions
Non-refundable rebate
1.02
Refundable rebate
1.26
Payroll tax holiday
1.29
Across the board tax cut
1.03
Accelerated depreciation
0.27
Extend alternative minimum
0.48
tax patch
Make income tax cuts expiring
0.29
in 2010 permanent
Make expiring dividend and
0.37
capital gains tax cuts
permanent
Reduce corporate tax rates
0.30
Spending Provisions
Extend unemployment
1.64
compensation benefits
Temporary increase in food
1.73
stamps
Revenue transfers to state
1.36
governments
Increase infrastructure
1.59
spending
Source: Mark Zandi, Moody’s Economy.com.
ȱ
The CBO rankings of multipliers are similar to Zandi’s.34 For government purchases and transfers
to state and local governments for infrastructure, their multipliers are 1.0 in the low scenario and
2.5 in the high. For transfers to state and local governments not for infrastructure, the multipliers
are 0.7 and 1.9. CBO sets the multipliers for transfers at 0.8 to 2.2, for temporary individual tax
cuts at 0.5 to 1.7, and the tax loss carryback at 0 to 0.4. As with Zandi, these effects do not
incorporate differentials in the rate of spending, however. In particular, they note that
infrastructure spending will likely be delayed, while transfers would occur very quickly. Unlike
Zandi, CBO emphasizes the broad uncertainty inherent in estimating multipliers.

34 Congressional Budget Office, “The State of the Economy and Issues in Developing an Effective Policy Response,”
Testimony of Douglas W. Elmendorf, Director, House Budget Committee, January 27, 2009.
ȱȱȱ
ŗřȱ

ȱǱȱȱȱȱ
ȱ
ȱ
Timeliness is another criterion by which different stimulus proposals have been evaluated. There
are lags before a policy change affects spending. As a result, stimulus could be delivered after the
economy has already entered a recession or a recession has already ended. First, there is a
legislative process lag that applies to all policy proposals—a stimulus package cannot take effect
until bills are passed by the House and Senate, both chambers can reconcile differences between
their bills, and the President signs the bill. Many bills get delayed at some step in this process. As
seen in Table 2, many past stimulus bills have not become law until a recession was already
underway or finished.35
Table 2. Timing of Past Recessions and Stimulus Legislation
Beginning of Recession
End of Recession Stimulus Legislation Enacted
Nov. 1948
Oct. 1949
Oct. 1949
Aug. 1957
Apr. 1958
Apr. 1958, July 1958
Apr. 1960
Feb. 1961
May 1961, Sep. 1962
Dec. 1969
Nov. 1970
Aug. 1971
Nov. 1973
Mar. 1975
Mar. 1975, July 1976, May 1977
July 1981
Nov. 1982
Jan. 1983, Mar. 1983
July 1990
Mar. 1991
Dec. 1991, Apr. 1993
Mar. 2001
Nov. 2001
June 2001
Source: Bruce Bartlett, “Maybe Too Little, Always Too Late,” New York Times, Jan. 23, 2008.
Second, there is an administrative delay between the enactment of legislation and the
implementation of the policy change. For example, although the 2008 stimulus package was
signed into law in February, the first rebate checks were not sent out until the end of April, and
the last rebate checks were not sent out until July. When the emergency unemployment
compensation (EUC08) program began in July 2008 there was about a three week lag between
enactment and the first payments of the new EUC08 benefit. Many economists have argued that
new government spending on infrastructure could not be implemented quickly enough to
stimulate the economy in time since infrastructure projects require significant planning. (Others
have argued that this problem has been exaggerated because existing plans or routine
maintenance could be implemented more quickly.) Others have argued that although federal
spending cannot be implemented quickly enough, fiscal transfers to state and local governments
would be spent quickly because many states currently face budgetary shortfalls, and fiscal

35 The International Monetary Fund recently analyzed the “timeliness, temporariness, and targeting” of U.S. tax cuts
from 1970 to 2008 in International Monetary Fund, World Economic Outlook, Washington, D.C., Oct. 2008, p. 172.
ȱȱȱ
ŗŚȱ

ȱǱȱȱȱȱ
ȱ
transfers would allow them to avoid cutting spending.36 ARRA granted $379 billion of budget
authority in 2009, but CBO projects that only $120 billion will be outlayed in 2009.37
Finally, there is a behavioral lag, since time elapses before the recipient of a transfer or tax cut
increases their spending. For example, the initial reaction to the receipt of rebate checks was a
large spike in the personal saving rate (see above). It is unclear how to target recipients that
would spend most quickly, although presumably liquidity-constrained households (i.e., those with
limited access to credit) would spend more quickly than others. In this regard, the advantage to
direct government spending is that there is no analogous lag. Although monetary policy changes
have no legislative or administrative lags, research suggests they do face longer behavioral lags
than fiscal policy changes because households and business generally respond more slowly to
interest rate changes than tax or transfer changes.
Ȭȱȱ
A main factor in another round of fiscal stimulus may be the size of the current budget deficit.
The 2009 stimulus package is relatively large, and CBO projects the deficit will already exceed
$1 trillion in 2009. Deficits of this magnitude would set a peacetime record relative to GDP.
Although current government borrowing rates are extremely low (because of the financial
turmoil), there is a fear that a deficit of this size could become burdensome to service when
interest rates return to normal. A larger deficit could eventually crowd out private investment, act
as a drag on economic growth, and increase reliance on foreign borrowing (which would result in
a larger trade deficit). By doing so, the deficit places a burden on future generations, and could
further complicate the task of coping with long-term budgetary pressures caused by the aging of
the population.38 In the highly unlikely, worst case scenario, if too much pressure is placed on the
deficit through competing policy priorities, then investors could lose faith in the government’s
ability to service the debt, and borrowing rates could spike.
Many of these issues could be minimized if the elements of the stimulus package are temporary –
an increase in the budget deficit for one year would lead to significantly less crowding out over
time than a permanent increase in the deficit. There is often pressure later to extend policies
beyond their original expiration date, however. Among policy options, increases in public
investment spending would minimize any negative effects on long-run GDP since decreases in
the private capital stock would be offset by additions to the public capital stock. Also, tax
incentives to increase business investment would offset the crowding out effect since the increase
in aggregate spending was occurring via business investment.
The direct effect of the American Recovery and Reinvestment Act on the budget deficit is
relatively small after 2011, although it leads to a permanent increase in interest payments on the
national debt if not offset by future policy changes.

36 Transfers to state and local governments could be less stimulative than direct federal spending because state and local
governments could, in theory, increase their total spending by less than the amount of the transfer. (For example, some
of the money that would have been spent in the absence of the transfer could now be diverted to the state’s budget
reserves.) But if states are facing budgetary shortfalls, many would argue that in practice spending would increase by as
much as the transfer.
37 Congressional Budget Office, Cost Estimate for the Conference Agreement for H.R. 1, Letter to Honorable Nancy
Pelosi, Feb. 13, 2009.
38 See CRS Report RL32747, The Economic Implications of the Long-Term Federal Budget Outlook, by Marc Labonte.
ȱȱȱ
ŗśȱ

ȱǱȱȱȱȱ
ȱ
ȱȱȱǵȱ
It is clear that the slowdown has been concentrated in housing and financial markets to date.
Some economists have argued that as long as problems remain in these depressed sectors, then
generalized stimulus will only postpone the inevitable downturn. For example, as long as
financial intermediation remains impaired, access to credit markets will be limited and it will be
difficult for stimulus to lead to sustained growth. (As noted above, separate legislation to support
housing and financial markets was enacted in 2008.) If so, fiscal stimulus may, at most, provide a
temporary boost as long as those problems are outstanding, but cannot singlehandedly shift the
economy to a sustainable path of expansion. For example, the first stimulus package, enacted in
the first quarter of 2008, did not prevent the economy from deteriorating further in the third
quarter of 2008. Other economists argue that if the current housing bust is being caused by the
unwinding of a bubble, then the government could be unable to reverse unavoidable market
adjustment that is bringing those markets back to equilibrium. But some would argue that the best
way to help a troubled sector is by boosting overall demand.
ȱȱȱȱǵȱ
The economy naturally experiences a boom and bust pattern called the business cycle. A
recession can be characterized as a situation where total spending in the economy (aggregate
demand) is too low to match the economy’s potential output (aggregate supply). As a result, some
of the economy’s labor and capital resources lay idle, causing unemployment and a low capacity
utilization rate, respectively. Recessions generally are short-term in nature—eventually, markets
adjust and bring spending and output back in line, even in the absence of policy intervention.39
Policymakers may prefer to use stimulative policy to attempt to hasten that adjustment process, in
order to avoid the detrimental effects of cyclical unemployment. By definition, a stimulus
proposal can be judged by its effectiveness at boosting total spending in the economy. Total
spending includes personal consumption, business investment in plant and equipment, residential
investment, net exports (exports less imports), and government spending. Effective stimulus
could boost spending in any of these categories.
Fiscal stimulus can take the form of higher government spending (direct spending or transfer
payments) or tax reductions, but generally it can boost spending only through a larger budget
deficit, as is the case with ARRA. A deficit-financed increase in government spending directly
boosts spending by borrowing to finance higher government spending or transfer payments to
households. A deficit-financed tax cut indirectly boosts spending if the recipient uses the tax cut
to increase his spending. If an increase in spending or a tax cut is financed through a decrease in
other spending or increase in other taxes, the economy would not be stimulated since the deficit-
increasing and deficit-decreasing provisions would cancel each other out.
How much additional spending can stimulate economic activity depends on the state of the
economy at that time. When the economy is in a recession, fiscal stimulus could mitigate the
decline in GDP growth by bringing idle labor and capital resources back into use. When the
economy is already robust, a boost in spending could be largely inflationary—since there would

39 For more information, see CRS Report RL34072, Economic Growth and the Business Cycle: Characteristics,
Causes, and Policy Implications, by Marc Labonte.
ȱȱȱ
ŗŜȱ

ȱǱȱȱȱȱ
ȱ
be no idle resources to bring back into production when spending is boosted, the boost would
instead bid up the prices of those resources, eventually causing all prices to rise. The recession
appears to have deepened in the fourth quarter of 2008. By historical standards, the recession
would be expected to end before fiscal stimulus could be delivered, but forecasters are predicting
this recession will be longer than usual. Most of the stimulus provided by ARRA will be delivered
by 2011, and CBO is projecting that there will still be a significant output gap at that point.
Because total spending can be boosted only temporarily, stimulus has no long-term benefits, and
may have long-term costs. Most notably, the increase in the budget deficit “crowds out” private
investment spending because both must be financed out of the same finite pool of national saving,
with the greater demand for saving pushing up interest rates.40 To the extent that private
investment is crowded out by a larger deficit, it would reduce the future size of the economy since
the economy would operate with a smaller capital stock in the long run. In recent years, the U.S.
economy has become highly dependent on foreign capital to finance business investment and
budget deficits.41 Since foreign capital can come to the United States only in the form of a trade
deficit, a higher budget deficit could result in a higher trade deficit, in which case the higher trade
deficit could dissipate the boost in spending as consumers purchase imported goods. Indeed,
conventional economic theory predicts that fiscal policy has no stimulative effect in an economy
with perfectly mobile capital flows.42 Some economists argue that these costs outweigh the
benefits of fiscal stimulus.
ȱ¢ȱȱȱ
Stimulus has also been delivered from other fiscal changes and monetary policy. First, the federal
budget has automatic stabilizers that cause the budget deficit to automatically increase (and
thereby stimulate the economy) during a downturn in the absence of policy changes. When the
economy slows, entitlement spending on programs such as unemployment compensation benefits
automatically increases as program participation rates rise and the growth in tax revenues
automatically declines as the recession causes the growth in taxable income to decline.
Second, any consideration for additional stimulus has to include the effects of stimulus previously
enacted. According to the Congressional Budget Office (CBO), the total deficit in FY2008 was
$455 billion, or 3.2% of gross domestic product, sharply higher than the FY2007 deficit of $162
billion. In January 2008, CBO had projected that under current policy the budget deficit would
increase by $56 billion in 2008 compared to 2007. When the cost of the February 2008 stimulus
package and part of the cost of financial market intervention in the fall of 2008 is added, the
increase in the deficit for one year rose by nearly $300 billion. CBO projects the deficit will
increase further in 2009, to $1.2 trillion or 8.3% of GDP, in the absence of additional stimulus.
These increases in the deficit would also be expected to have a stimulative effect on aggregate
spending.

40 Crowding out is likely to be less of a concern when the economy is in recession since recessions are typically
characterized by falling business investment.
41 If foreign borrowing prevents crowding out, the future size of the economy will not decrease but capital income will
accrue to foreigners instead of Americans.
42 For more information, see CRS Report RS21409, The Budget Deficit and the Trade Deficit: What Is Their
Relationship?, by Marc Labonte and Gail E. Makinen.
ȱȱȱ
ŗŝȱ

ȱǱȱȱȱȱ
ȱ
Third, the Federal Reserve has already delivered a large monetary stimulus. By the end of April
2008, the Fed had reduced overnight interest rates to 2% from 5.25% in September 2007.43 On
December 16, the interest rate was lowered to a targeted range of 0% to 0.25%. Typically, lower
interest rates stimulate the economy by increasing the demand for interest-sensitive spending,
which includes investment spending, residential housing, and consumer durables such as
automobiles. Yet, the potential for stimulus caused by lower interest rates can be limited if tight
credit markets constrain borrowing. In addition, lower interest rates can stimulate the economy by
reducing the value of the dollar, all else equal, which would lead to higher exports and lower
imports.44
One might take the view that the Federal Reserve has chosen a monetary policy that it believes
will best achieve a recovery given the actions already taken. If it has chosen that policy correctly,
an argument can be made that an additional fiscal stimulus is unnecessary since the economy is
already receiving the correct boost in spending through lower interest rates and through the first
stimulus package. In this light, additional fiscal stimulus would be useful only if monetary policy
is unable to adequately boost spending—either because the Fed has chosen an incorrect policy or
because the Fed cannot boost spending enough through lower interest rates and direct assistance
to the financial sector to spark a recovery, and direct intervention in financial markets is not
adequate. 45 (Now that interest rates have fallen to zero, the Fed can no longer reduce rates to
stimulate the economy, but it can increase – and has increased - its direct assistance to the
financial sector.)
ȱȱȱȱȱȱ
A number of direct interventions in the economy occurred in 2008 which could be seen as a type
of stimulus, in part because of credit problems. One indication of restricted credit despite
stimulative Federal Reserve monetary policy was the failure of mortgage rates to fall
significantly. Instead, the spread between Treasuries and Government Sponsored Enterprise
(GSE) bonds remained elevated over the summer. The newly created Federal Housing Finance
Agency (FHFA) cited the persistence of this wide spread as a major factor in its decision to place
the GSEs in conservatorship in September. During the week of September 15-19, financial
markets were further disturbed by the bankruptcy of investment bank Lehman Brothers and
Federal Reserve intervention on behalf of the insurer AIG. These actions eroded market
confidence further, resulting in a sudden spike of the commercial paper rate spread from just
under 90 basis points to 280 basis points, a spike that in times past might have been called a
panic. If financial market confidence is not restored and private market spreads remain elevated,
the broader economy could slow more due to difficulties in financing consumer durables,
business investment, college education, and other big ticket items.

43 For interest rate changes see CRS Report 98-856, Federal Reserve Interest Rate Changes: 2001-2008, by Marc
Labonte and Gail E. Makinen.
44 For more information, see CRS Report RL30354, Monetary Policy and the Federal Reserve: Current Policy and
Conditions, by Gail E. Makinen and Marc Labonte.
45 Fed Chairman Ben Bernanke may have hinted at the latter case when he testified that “fiscal action could be helpful
in principle, as fiscal and monetary stimulus together may provide broader support for the economy than monetary
policy actions alone.” Quoted in Ben Bernanke, “The Economic Outlook,” testimony before the House Committee on
the Budget, January 17, 2008.
ȱȱȱ
ŗŞȱ

ȱǱȱȱȱȱ
ȱ
In September 2008, Administration and Federal Reserve officials with the bipartisan support of
the Congressional leadership, announced a massive intervention in the financial markets,
requesting authority to purchase up to $700 billion in assets over the next two years. The Treasury
had also provided insurance for money market funds, where withdrawals have been significant.
Congressional leaders and other Members raised a number of issues and made some additional
proposals, which included setting up an oversight mechanism, restrictions on executive
compensation of firms from which assets are purchased, acquiring equity stakes in the
participating firms, and allowing judges to reduce mortgage debt in bankruptcies (not included in
the final Act).
Later, in October 2008, legislation (P.L. 110-343) was enacted to allow an initial $250 billion of
financing with an additional $100 billion upon certification of need, with Congress allowed 30
days to object to the final $350 billion. The plan has oversight by an Inspector General, audit by
the Government Accountability Office, setting standards of appropriate compensation, and
providing for equity positions in all participating companies. The final package also added an
expansion of deposit insurance coverage. There remained, however, concerns about how to price
acquired assets in a way that balances protection of taxpayers with providing adequate assistance
to firms. The Treasury had indicated use of a reverse auction mechanism to purchase mortgage
backed securities, where companies will bid to sell their assets. It is not clear how well such an
auction would work with heterogeneous assets.46
The Treasury subsequently announced that it would use the first $250 billion authorized to
purchase preferred stock in financial institutions and has now indicated it will use subsequent
funds for capital injections, consumer credit (such as auto loans, student loans, small business
loans, and credit cards) and mortgage assistance.47 Congressional leaders urged Treasury to
provide $25 billion in aid to U.S. auto manufacturers.48 On November 10, a restructuring of
government assistance to AIG was announced which increased the amount at risk from $143.7
billion to $173.4 billion, extended the loan length and reduced the interest rate. The Federal
Reserve also announced on October 14 that it would begin purchasing commercial paper.49 News
reports indicated the Federal Deposit Insurance Corporation (FDIC) had a plan, supported by
many congressional Democrats, to offer financial incentives to companies that agree to reduce
monthly mortgage payments, but that this plan was opposed by the Bush Administration.50 On
November 23, the government announced a plan to assist Citicorp, and on November 25 the
Federal Reserve revealed plans to purchase $200 billion in asset backed securities through the
Term Asset-Backed Securities Loan Facility (TALF); these securities are based on auto, credit
card, student and small business loans. The Federal Reserve also announced a plan to purchase
$600 billion of mortgage related securities owned or guaranteed by the housing GSEs.
Much of the intervention up to this point had been in the financial markets. However, the Detroit
automakers (GM, Ford, and Chrysler) asked for $34 billion in loans to forestall bankruptcy. After
Congress did not adopt an emergency loan of $14 billion in a special post-election session in

46 See CRS Report RL34707, Auction Basics: Background for Assessing Proposed Treasury Purchases of Mortgage-
Backed Securities, by D. Andrew Austin.
47 Testimony of Interim Assistant Secretary for Financial Stability Neel Kashkari before the House Committee on
Oversight and Government Reform, Subcommittee on Domestic Policy, November 14, 2008.
48 David M. Herszenhiorn, “Chances Dwindle on Bailout Plan for Automakers,” New York Times, November 14, p. A1.
49 Federal Reserve Board Press Release, October 14, 2008.
50 Buinyamin Appelbaum, FDIC Details Plan to Alter Mortgages, Washington Post, November 14, 2008, p. A1.
ȱȱȱ
ŗşȱ

ȱǱȱȱȱȱ
ȱ
December 2008, the Administration announced, on December 19, that it would provide $17.4
billion from TARP: $9.4 billion to GM and $4 million to Chrysler. An additional $4 billion would
be made available for GM if the remaining $350 billion in TARP funds is approved. On
December 30, $6 billion in TARP funds were provided for GMAC, the auto financing company.
On January 16, 2009, Administration officials indicated they would provide $20 billion to
guarantee assets of Bank of America; funds have also been provided to small banks.
Legislation has been introduced (H.R. 384) by the Chairman of the Financial Services Committee
to regulate the spending of the final $350 billion, but many of the provisions could also be
achieved through an agreement between the Congress and the Administration. There is interest in
directing some of the funds to directly aid mortgage holders to avoid foreclosure and small
business loans, as well as considering oversight issues. Congress could have enacted legislation to
disallow the release. However, on January 15, the Senate defeated a proposal to block the
spending of the additional funds.
Among the issues of concern with financial interventions is whether an ad hoc, case-by-case
intervention is likely to be a successful strategy. A case-by-case strategy can create uncertainty
and also moral hazard (causing firms to undertake too much risk if they expect to be rescued).
The creation of TARP represents a shift to a more broad-based approach. The approach of a broad
based intervention could take the form of the purchase of troubled assets (as originally proposed
or through a “bad bank”) or the injection of capital (such as the Treasury’s decision to purchase
preferred stock).51

ȱȱȱ

Jane G. Gravelle
Marc Labonte
Senior Specialist in Economic Policy
Specialist in Macroeconomic Policy
jgravelle@crs.loc.gov, 7-7829
mlabonte@crs.loc.gov, 7-0640
Thomas L. Hungerford

Specialist in Public Finance
thungerford@crs.loc.gov, 7-6422

51 These issues are discussed in more detail in CRS Report RL34730, The Emergency Economic Stabilization Act and
Current Financial Turmoil: Issues and Analysis, by Baird Webel and Edward V. Murphy.
ȱȱȱ
ŘŖȱ