The Budget and Economic Outlook: 2018 to 2028

APPENDIX

The Eects of the 2017Tax Act on CBO’s

Economic and Budget Projections

Overview

In December2017, Public Law 115-97, referred to here

as the 2017tax act, was enacted. e act made import-

ant changes to the tax system that apply to both busi-

nesses and individuals. Consequently, the Congressional

Budget Oce had to estimate its eects when preparing

its new baseline projections, which incorporate the

assumption that current laws aecting taxes and spend-

ing generally do not change. In those projections for the

2018–2028period, the act’s changes boost economic

output and increase budget decits, on net.

What Are the Act’s Major Provisions?

e 2017tax act changes corporate and individual tax

rates and includes various provisions that aect how

businesses and individuals calculate their taxable income.

Among other things, the act lowers the top corporate

income tax rate to 21percent. It changes the way that

the foreign income of U.S. corporations is taxed, and it

reduces some incentives for corporations to shift prots

outside the United States. For the next eight years, the

act lowers individual income tax rates and broadens the

total amount of income subject to that tax. Also for the

next eight years, it increases the tax exemptions for prop-

erty transferred at death and for certain gifts. Starting

next year, it eliminates the penalty for not having health

insurance—a penalty imposed under a provision of the

Aordable Care Act generally called the individual man-

date. And it changes the measure of ination that is used

to adjust certain tax parameters.

What Are the Act’s Projected Economic Eects?

In CBO’s assessment, the 2017tax act changes busi-

nesses’ and individuals’ incentives in various ways. On

net, those changes are expected to encourage saving,

investment, and work.

CBO projects that the act’s eects on the U.S. economy

over the 2018–2028period will include higher levels

of investment, employment, and gross domestic prod-

uct (GDP). For example, in CBO’s projections, the act

boosts average annual real GDP by 0.7percent over the

2018–2028period. Analysis of the act’s economic eects

is complicated by its mix of permanent and temporary

provisions; of particular note is that it lowers the corpo-

rate income tax rate permanently but individual income

tax rates only through 2025. As a result, the projected

economic eects vary over the 11-year period; the largest

eects on the economy occur during the period’s middle

years.

CBO’s projections of the act’s economic eects are based

partly on projections of the act’s eects on potential

GDP—the economy’s maximum sustainable level of

production. In the agency’s projections, the act increases

the level of potential GDP by boosting investment and

labor. By lowering the corporate income tax rate, the act

gives businesses incentives to increase investment, and

by lowering individual income tax rates through 2025,

it gives people incentives to increase their participation

in the labor force and their hours worked, expanding the

potential labor supply and employment. Other provi-

sions of the tax act, including a limit on deductions for

state and local taxes and for mortgage interest, will push

down residential investment, but the overall eect on

investment is positive. One result of the act will dampen

those positive eects on potential output: It will increase

federal decits and therefore increase federal borrowing

and interest rates, crowding out some private investment.

In CBO’s projections, the act initially boosts real GDP

(that is, GDP adjusted to remove the eects of ina-

tion) in relation to real potential GDP, inuencing other

economic variables, such as ination and interest rates.

GDP is pushed up in relation to potential GDP because

the act increases overall demand for goods and services

(by raising households’ and businesses’ after-tax income).

e heightened economic activity subsequently generates

more demand for labor and consequently higher wages.

In response, the labor force participation rate (which is

the percentage of people in the civilian noninstitutional-

ized population who are at least 16years old and either

Appendix B

106 The BudgeT and economic ouTlook: 2018 To 2028 APRIL 2018

working or seeking work) rises, as do the number of

hours worked, and the unemployment rate goes down.

e largest positive eects occur during the 2018–

2023period. After income tax rates rise as scheduled at

the close of 2025, the growth of overall demand is damp-

ened in relation to the growth of potential output.

Among the eects of the initially stronger output growth

are slightly higher ination and an increase in the

exchange value of the dollar. Furthermore, CBO expects

the Federal Reserve to respond to the stronger labor

market and increases in inationary pressure by pushing

short-term interest rates higher over the next few years.

Long-term interest rates are also expected to rise.

Just as the tax act is projected to boost real GDP, it

is expected to increase income for labor and business

over the 2018–2028period. e act will also aect

the relationship between GDP and gross national

product (GNP). GNP diers from GDP by includ-

ing the income that U.S. residents earn from abroad

and excluding the income that nonresidents earn from

domestic sources; it is therefore a better measure of

the income available to U.S. residents. Because the act

reduces the amount of net foreign income earned by

U.S. residents in CBO’s projections, it increases GNP

less than it increases GDP.

What Are the Act’s Projected Budgetary Eects?

To construct its baseline budget projections, CBO incor-

porated the eects of the tax act, taking into account

economic feedback—that is, the ways in which the act

is likely to aect the economy and in turn aect the

budget. Doing so raised the 11-year projection of the

cumulative primary decit (that is, the decit excluding

the costs of servicing the debt) by $1.3trillion and raised

projected debt-service costs by roughly $600billion. e

act therefore increases the total projected decit over the

2018–2028period by about $1.9trillion.

Before taking economic feedback into account, CBO

estimated that the tax act would increase the primary

decit by $1.8trillion and debt-service costs by roughly

$450billion. e feedback is estimated to lower the

cumulative primary decit by about $550billion, mostly

because the act is projected to increase taxable income

and thus push tax revenues up. And that feedback raises

projected debt-service costs, because even though the

reduction in primary decits means that less borrowing

is necessary, the act is expected to result in higher interest

rates on debt, which are projected to more than oset the

eects on debt-service costs of the smaller debt. On net,

economic feedback from the act raises debt-service costs

in CBO’s projections by about $100billion.

What Uncertainty Surrounds CBO’s Estimates?

CBO’s estimates of the economic and budgetary eects

of the 2017tax act are subject to a good deal of uncer-

tainty. e agency is uncertain about various issues—for

example, the way the act will be implemented by the

Treasury; how households and businesses will rearrange

their nances in the face of the act; and how households,

businesses, and foreign investors will respond to changes

in incentives to work, save, and invest in the United

States. at uncertainty implies that the actual outcomes

may dier substantially from the projected ones.

The Major Provisions of the Act

e 2017tax act makes important changes to the tax sys-

tem that apply to both businesses and individuals. ey

include changes to corporate and individual tax rates

and a variety of provisions that aect how businesses and

individuals calculate their taxable income. e changes

have important eects on incentives to save, invest, and

work.

Together, CBO estimates, the act’s provisions reduce,

on net, the user cost of capital, which is the gross

pretax return on investment that provides the required

return to investors after covering taxes and depreci-

ation. at required return can be thought of as the

return that investors would have received if they had

used their funds to make another, equally risky invest-

ment. erefore, all things being equal, as the user

cost of capital falls, investment rises, and vice versa. In

addition, the smaller user cost of capital implies lower

eective marginal tax rates on capital income.

By CBO’s

1. e eective marginal tax rate on capital income is the share of

the return on an additional investment made in a particular year

that will be paid in taxes over the life of that investment. Unlike

statutory tax rates, eective marginal tax rates account for the

tax treatment of depreciation and various other features of the

tax code. For descriptions of CBO’s method of estimating the

eective marginal tax rate on capital income, see Congressional

Budget Oce, Taxing Capital Income: Eective Marginal Tax

Rates Under 2014 Law and Selected Policy Options (December

2014), Appendix A, www.cbo.gov/publication/49817, and

Computing Eective Tax Rates on Capital Income (December

2006), www.cbo.gov/publication/18259. For a description of

the relationship between the eective marginal tax rate and the

user cost of capital, see page 30of the December 2014 report,

in which the user cost of capital is found by summing the real

before-tax rate of return required to cover certain costs (ρ) and

the rate of depreciation (δ).

107appendiX B: The eFFecTS oF The 2017TaX acT on cBo’S economic and BudgeT proJecTionS The BudgeT and economic ouTlook: 2018 To 2028

estimate, the act reduces the eective marginal tax rate

on capital income, averaged over all types of investment,

by between 1.4percentage points and 3.4percentage

points from 2018 to 2028 (see Table B-1). at in turn

stimulates personal saving.

In addition, CBO estimates that the act reduces, on

net, the eective marginal tax rate on labor income by

2.2percentage points in 2018 and by slightly smaller

amounts through 2025, thereby encouraging work.

Beginning in 2026, the act is projected to boost the rate,

2. e eective marginal tax rate on labor income is the share

of an additional dollar of such income that is unavailable to a

worker because it is paid in federal individual income taxes and

payroll taxes or oset by reductions in benets from government

programs. at rate, like the eective marginal tax rate on capital

income, diers from statutory tax rates by taking into account

dierent features of the tax code (for example, the gradual

reduction in the value of the earned income tax credit as income

rises). For more information on how changes in after-tax wages

distort incentives to work, see Robert McClelland and Shannon

Mok, A Review of Recent Research on Labor Supply Elasticities,

Working Paper 2012-12 (Congressional Budget Oce, October

2012), www.cbo.gov/publication/43675.

as temporary measures that lower it expire and provisions

that push it up continue.

Changing the Corporate Income Tax Rate

Before the act was passed, businesses subject to the

corporate income tax faced a graduated rate structure.

e statutory tax rates were 15percent, 25percent,

34percent, and 35percent, depending on the business’s

income. e act replaces that structure with a single rate

of 21percent, beginning in 2018. at change lowers,

on average, the tax rate paid by businesses subject to the

corporate income tax. e change also contributes to

the reduction of the eective marginal tax rate on capital

income.

e corporate income tax distorts domestic economic

incentives, aecting the decisions made by corporations

and investors.

In addition, variation among the cor-

porate tax systems of dierent countries distorts deci-

sions about where to locate international investment.

3. For more information on how the corporate income tax distorts

economic incentives, see Congressional Budget Oce, Corporate

Income Tax Rates: International Comparisons (November 2005),

pp. 1–9, www.cbo.gov/publication/17501.

Table B-1 .

Projections of Eective Marginal Federal Tax Rates

Percent

2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028

Labor Income

Rate Under Prior Law 29.4 29.5 29.7 29.8 30.0 30.1 30.2 30.4 30.5 30.6 30.7

Rate Under the 2017 Tax Act 27.2 27.4 27.6 27.7 27.9 28.1 28.2 28.5 30.6 30.7 30.8

Dierence (Percentage points) -2.2 -2.2 -2.2 -2.1 -2.1 -2.0 -1.9 -1.9 * 0.1 0.1

Capital Income

Rate Under Prior Law 16.5 16.8 17.9 17.9 17.9 17.9 17.9 17.9 17.9 18.0 18.0

Rate Under the 2017 Tax Act 14.7 14.7 14.6 14.5 15.4 15.7 16.1 16.5 16.0 16.5 16.5

Dierence (Percentage points) -1.8 -2.1 -3.3 -3.4 -2.5 -2.2 -1.9 -1.4 -1.9 -1.5 -1.5

Source: Congressional Budget Oce.

The eective marginal tax rate on labor income is the share of an additional dollar of such income that is unavailable to a worker because it is paid in

federal individual income taxes and payroll taxes or oset by reductions in beneﬁts from government programs, averaged among workers with weights

proportional to their labor income.

The eective marginal tax rate on capital income is the share of the return on an additional investment made in a particular year that will be paid

in taxes over the life of that investment. The before- and after-tax rates of return used to calculate that eective tax rate are weighted averages of

the rates for every combination of asset type, industry, form of organization, and source of ﬁnancing; the weights used are the asset values of each

combination. All of those rates of return incorporate estimated values for interest rates on corporate debt, rates of inﬂation, and returns paid by

Ccorporations on equity that are consistent with recent trends and with CBO’s economic forecast. Speciﬁcally, CBO has incorporated a nominal interest

rate on debt for corporate securities of 5.8 percent; a rate of inﬂation, measured by the price index for urban consumers, of 2.4 percent; and a real

return on equity of 5.2 percent.

* = between zero and 0.05 percentage points.

108 The BudgeT and economic ouTlook: 2018 To 2028 APRIL 2018

Reducing the corporate income tax rate in the United

States reduces those distortions in several important

ways. First, it reduces the pretax return required to

induce businesses to invest. at reduces the user cost of

capital and should therefore increase investment. Second,

it makes debt nancing less advantageous in relation to

equity nancing—because businesses may deduct the

interest on debt from their taxable income, and the value

of that deduction becomes smaller when tax rates are

lower. ird, the reduction in corporate income taxes

increases U.S. and foreign investors’ incentives to invest

and to locate activities in the United States rather than

abroad.

Fourth, it reduces the incentive to shift income

from the United States to lower-tax countries.

Changing International Taxes

e act changes how the United States taxes the foreign

income of U.S. corporations. It also imposes a onetime

tax on previously untaxed foreign prots. And it adds

measures to discourage prot shifting, a practice in

which multinational corporations lower their tax liabili-

ties by shifting reported taxable income from aliates in

countries with higher corporate tax rates to aliates in

countries with lower ones.

Changing the Taxation of Foreign Income. ere are

two broad ways in which a country may tax the foreign

income earned by a domestic corporation. Under a pure

worldwide system, any foreign income is taxed immedi-

ately by the corporation’s home country. Under a pure

territorial system, the corporation’s home country does

not tax foreign income at all.

Under prior law, the United States had a system that

more closely resembled worldwide taxation. However,

only some types of foreign income—generally those

that the government regarded as being passive (such as

interest income) or highly mobile—were taxed as the

income was earned. Taxes on many types of foreign

4. For more information about those incentives, see Congressional

Budget Oce, Taxing Capital Income: Eective Marginal Tax

Rates Under 2014 Law and Selected Policy Options (December

2014), www.cbo.gov/publication/49817. For more information

about location decisions, see Congressional Budget Oce,

International Comparisons of Corporate Income Tax Rates (March

2017), www.cbo.gov/publication/52419.

5. For a more detailed description of both approaches, see

Congressional Budget Oce, Options for Taxing U.S.

Multinational Corporations (January 2013), www.cbo.gov/

publication/43764.

income earned by a U.S. corporation’s foreign subsidiary

were deferred until the income was repatriated—that is,

distributed to the U.S. parent company. Earnings were

considered repatriated if, for example, they were paid out

to shareholders as dividends, used to buy back shares,

or used to fund an investment in physical capital in the

United States.

e 2017tax act replaces that approach with a system

that may more closely resemble territorial taxation.

Dividends that a U.S. parent company receives from

its foreign subsidiaries will now be exempt from U.S.

taxation. However, foreign income that the government

regards as passive or highly mobile will still be taxed as

the income is earned.

Because the repatriation of foreign earnings triggered

tax liability under prior law, some corporations behaved

as though they were constrained in how they could

use foreign earnings. e new dividend exemption will

eliminate that constraint. As a result, corporations will

probably repatriate a larger share of their foreign earnings

by reducing the amount that they reinvest in foreign

economies.

However, the dividend exemption is anticipated to

encourage some further prot shifting, because corpora-

tions that shift prots from the United States to lower-

tax countries can now repatriate them without paying

taxes. at increase in prot shifting will reduce the

amount of income subject to U.S. taxes.

Onetime Tax on Previously Untaxed Foreign

Prots.e tax act also addresses the treatment of undis-

tributed foreign earnings that accumulated before the

taxation of foreign income was changed (see Box B-1).

It imposes a tax on those undistributed foreign earn-

ings, with separate rates for cash assets (15.5percent)

and noncash assets (8percent). Corporations must pay

the tax regardless of whether they actually repatriate

the earnings to the United States—a requirement often

called “deemed repatriation”—and have the option of

spreading the payment of the tax over eight years. e

tax should have only a limited eect on the decisions

that corporations make, because it applies only to their

existing stock of foreign earnings.

Measures to Reduce Prot Shifting. e act con-

tains several provisions to reduce corporations’ incen-

tive to shift prots out of the United States. Two

109appendiX B: The eFFecTS oF The 2017TaX acT on cBo’S economic and BudgeT proJecTionS The BudgeT and economic ouTlook: 2018 To 2028

provisions—which impose a tax on global intangible

low-tax income (GILTI) and create a deduction for

foreign- derived intangible income (FDII)—reduce cor-

porations’ incentive to locate high-return assets (which

are often intangible assets, such as intellectual property,

or IP) in low-tax countries. e provisions reduce that

incentive by applying special treatment to prots that

exceed a specied return on tangible assets (such as

equipment and structures).

6. e GILTI provision imposes a tax on foreign income that

exceeds a 10percent return on foreign tangible assets if a

In addition to reducing prot shifting through the

location of intangible assets, the GILTI and FDII

provisions aect corporations’ decisions about where to

locate tangible assets. By locating more tangible assets

abroad, a corporation is able to reduce the amount of

foreign income that is categorized as GILTI. Similarly,

by locating fewer tangible assets in the United States, a

multinational corporation’s average foreign tax rate is below a

certain threshold. e FDII deduction applies to U.S. prots that

exceed a 10percent return on U.S. tangible assets. e deduction

is proportional to the share of U.S. income that is derived from

foreign sales.

Box B-1 .

Repatriation of Undistributed Foreign Earnings

Before the 2017 tax act was enacted, a multinational corpo-

ration (MNC) could defer paying taxes on foreign earnings

until they were distributed to the MNC’s parent company in

the United States. Earnings were considered distributed if, for

example, they were paid out to shareholders as dividends,

used to buy back shares, or used to fund an investment

in physical capital in the United States. To avoid the tax

cost, MNCs left large amounts of earnings in their foreign

subsidiaries—a total of $2.6 trillion as of 2015, according to the

sta of the Joint Committee on Taxation.

The 2017 tax act mandates “deemed repatriation” for those

accumulated foreign earnings, which means that MNCs will pay

U.S. taxes on the earnings even if they are not distributed to

the United States. The act thus eliminates the tax disincentive

to distribute those earnings. As a result, MNCs are expected

to end up deploying the earnings in their domestic operations

more often.

The Congressional Budget Oce projects that deemed repa-

triation will have some eects on MNCs’ ﬁnancial decisions.

Before the change in law, some MNCs, to avoid paying the

tax cost of using foreign earnings to fund investments and

payments to shareholders, used borrowed funds for those

purposes, in CBO’s judgment. Because MNCs can no longer

avoid the tax cost, CBO projects that some will reduce their

borrowing. Also, some of the previously undistributed earnings

can be paid to shareholders through share repurchases and

larger dividends.

1. Thomas A. Barthold, Joint Committee on Taxation, letter to the Honorable

Kevin Brady, Chairman, House Ways and Means Committee, and the

Honorable Richard Neal (August 31, 2016), https://go.usa.gov/xQrVY.

On the whole, however, CBO projects that the economic

eects of deemed repatriation will be small. The MNCs that

refrained from distributing their foreign earnings tended to

be established corporations in the high-tech sector that faced

low costs in funding domestic activities and probably did not

forgo worthy investments as a result of keeping their earnings

undistributed. Furthermore, even though the term “repatria-

tion” suggests that the undistributed funds will return to the

United States from abroad, they are often already invested

in dollar-denominated ﬁxed-income securities issued by U.S.

borrowers. The funds are outside the United States only in

the sense of being owned by a foreign subsidiary of a U.S.

corporation. In fact, MNCs have held a substantial fraction

of their undistributed funds as long-term Treasury securities,

CBO estimates. Finally, over the past decade, MNCs have paid

large amounts of cash to their shareholders through share

repurchases even as they have kept earnings undistributed,

so it is unlikely that the foreign earnings represent pent-up

dividends or investments waiting to happen.

In CBO’s projections, the eects of deemed repatriation on

MNCs’ ﬁnancial decisions lead to a small decrease in the cor-

porate spread, which is the dierence between corporate and

U.S. government interest rates. Corporations are expected to

reduce their holdings of U.S. government debt and reduce their

borrowing. As they reduce holdings of federal debt, interest

rates for it will rise; meanwhile, as they borrow less, interest

rates for corporate debt will fall. The resulting decrease in

the corporate spread should support additional corporate

investment but put some upward pressure on the interest rates

of Treasury notes.

110 The BudgeT and economic ouTlook: 2018 To 2028 APRIL 2018

corporation can increase the amount of U.S. income that

can be deducted as FDII. Together, the provisions may

increase corporations’ incentive to locate tangible assets

abroad. (Like prot shifting, such decisions change the

locations of reported prots—but they are not classied

as prot shifting, because they involve actual economic

activity rather than simply reporting.)

Another provision, the base erosion and antiabuse tax

(BEAT), limits the ability of both U.S. and foreign

multinational corporations to use related-party transac-

tions to shift prots from the United States to lower-tax

countries. (Related-party transactions are transactions

between the aliates of a multinational corporation.)

BEAT imposes a minimum tax on relatively large multi-

national corporations, which must pay the larger of two

amounts: their regular tax liability, and a tax at a spec-

ied rate (generally 10percent) on a broader measure

of U.S. taxable income that is adjusted for related-party

transactions.

Changing the Taxation of Domestic Business Activity

e 2017tax act makes numerous changes to tax

provisions that aect both corporate and noncorporate

businesses. ose changes limit or eliminate some tax

preferences and thus increase the tax base (that is, the

total amount of income subject to tax); provide incen-

tives for certain types of investments by allowing busi-

nesses to deduct the costs more rapidly; and create a new

deduction for certain owners of pass-through businesses

(which are businesses whose prots are taxed not directly

through the corporate income tax but when their owners

pay income tax on their share of prots).

On net, those

changes reduce the eective marginal tax rate on capital

income paid by corporate and noncorporate businesses.

Base Expansion. e act expands the business income

tax base in a number of ways. One is a new limit on net

interest deductions; another modies the treatment of

losses.

Interest Limit. Under prior law, a business could generally

deduct its interest expense when calculating its taxable

income. For businesses whose gross receipts are greater

than $25 million, the act limits the deduction of interest

7. For more information, see “Key Methods at CBO Used

to Estimate the Macroeconomic Eects of the 2017 Tax Act”

(supplemental material for e Budget and Economic Outlook:

2018 to 2028, April 2018), https://go.usa.gov/xQcZD.

expense to an amount equal to a business’s interest

income plus 30 percent of its adjusted taxable income.

e measure of adjusted taxable income used for that

determination excludes interest income and expense. It

also excludes deductions for depreciation and similar

costs through 2021but then includes them. Business

interest that is not deducted because it exceeds the limit

may be carried forward—that is, potentially claimed

in a future year. Special rules apply to pass-through

businesses.

Limiting the deductibility of interest creates an incen-

tive to reduce existing debt and reduces the incentive to

issue new debt, particularly for companies that already

have substantial amounts of debt. Limiting interest

deductions may also increase multinational corporations’

incentive to borrow through aliates that are not in

the United States instead of through aliates that are.

at would increase prots reported by aliates that

are in the United States. In addition, the change in the

denition of adjusted taxable income in 2022lowers

businesses’ capacity to deduct interest, encouraging larger

investment and depreciation deductions before 2022.

Limits on the Use of Net Operating Losses. Under prior

law, a net operating loss could be deducted from tax-

able income up to 2years in the past and up to 20years

in the future. For losses occurring after 2017, the act

restricts the deduction to future income (for most

industries), and it restricts the deduction to 80percent of

taxable income. In addition, the 20-year limit is repealed.

For the owners of pass-through businesses, trade or

business losses can be used to oset current-year income

from other sources. e act limits that current-year

deduction to $500,000annually for joint returns and

$250,000 for single returns. Any excess loss can be

deducted as a net operating loss in the future.

Overall, those provisions treat losses less generously

than prior law did. Restricting the deduction of losses to

future income will mean that companies will no longer

be able to use losses in a way that creates a current-year

refund. at change may especially hurt corporations

without many liquid assets. In addition, the changes

reduce corporations’ incentive to claim various deduc-

tions that can result in losses.

Deductions for Capital Investments. When a business

invests in a tangible asset, it generally deducts the cost

111appendiX B: The eFFecTS oF The 2017TaX acT on cBo’S economic and BudgeT proJecTionS The BudgeT and economic ouTlook: 2018 To 2028

of the investment over time until it has deducted the

full purchase price of the asset. For each type of asset,

tax law and regulations prescribe a depreciation schedule

that determines the amount to be deducted each year.

Under certain circumstances, however, the cost of the

asset can be fully “expensed”—that is, fully deducted in

the year it is placed in service. e 2017tax act expands

those circumstances for many types of tangible assets but

restricts them for certain intangible ones—specically,

research and development (R&D) and software devel-

opment. It increases the base amount of tangible equip-

ment that can be expensed under section 179 of the tax

code to $1million, and it increases the base amount

of investment at which that expensing begins to phase

out to $2.5million. e act also temporarily increases

the percentage of the investment in new tangible equip-

ment that businesses can expense from 50percent of the

acquisition cost to 100percent; between 2023 and 2027,

that “bonus depreciation” will be phased down to zero

in 20-percentage-point increments.

In contrast, invest-

ment in R&D and software development must now be

deducted in equal proportions over ve years if the costs

are incurred in 2022or later; in the past, that investment

could be expensed.

e speed with which businesses can deduct their capital

spending aects the pretax rate of return needed to

induce a new investment; it thus aects the user cost of

capital as well. Expensing reduces the user cost of capital

by allowing businesses to deduct the cost of investment

from their taxable income more quickly. e expansion

of expensing for tangible assets should result in more

investment in the qualifying types of assets. However,

those types were already treated more favorably than

nonqualifying types of tangible assets (mostly buildings),

and the expansion of expensing will widen that disparity.

e result will be some distortion in favor of the quali-

fying types. Requiring R&D and software development

costs to be deducted over ve years rather than imme-

diately will increase the cost of capital and thus reduce

those types of investment.

Deduction for Certain Owners of Pass-rough

Businesses. e prots of pass-through businesses are

allocated to their owners, added to their taxable income,

and often taxed through the individual income tax.

8. e bonus depreciation percentage was 50percent in 2017;

under prior law, it was scheduled to be 40percent in 2018,

30percent in 2019, and zero thereafter.

e rate at which those prots are taxed consequently

depends on which tax bracket the owner is in. rough

2025, individual income tax rates are generally lower

under the 2017tax act than they would have been under

prior law, but not by nearly as much as the corporate

income tax rate. However, the act also provides a tem-

porary new deduction to many owners of pass-through

businesses through 2025. e deduction is equal to

20percent of qualied business income, which includes

the reasonable compensation of owners for services

rendered to the business. Eligibility for the deduction

depends on both the owner’s income and the nature of

the business. e deduction phases out with income for

owners of personal-service businesses (such as law rms,

medical practices, and consulting rms). For other own-

ers, the deduction may be limited by the wages that the

business pays and the property that it owns.

Because it has the same eect as a reduction in the tax

rate, the deduction for pass-through businesses lowers

the cost of capital for qualifying companies and reduces

the disparity between the tax treatments of debt- and

equity-nanced investment. It also reduces the disparity

between the treatments of capital income earned by cor-

porations and of capital income earned by pass-through

businesses. However, it may result in dierent tax rates

for dierent sources of labor income. at dierence

could occur because the deduction gives owners of pass-

through businesses an incentive to underreport their

reasonable compensation—a tactic that has been used

successfully to avoid self-employment taxes in the past

and that is not available to wage earners. In addition,

the deduction’s dierent treatment of dierent industries

could further aect economic decisions.

Changing Individual Income Taxes

e 2017tax act changes individual income taxes,

lowering statutory tax rates but also broadening the tax

base through various provisions. On net, the act reduces

marginal tax rates: Provisions that reduce statutory rates

and expand the standard deduction push marginal rates

down, an eect only partly oset by provisions that limit

itemized deductions and eliminate personal exemptions.

Most of the provisions involving individual income taxes

expire at the end of 2025. e temporary nature of those

provisions will aect the behavior of some taxpayers;

they will try to earn more during the years when rates

are lower or to delay deductible expenses—whose value

rises as rates increase—until after 2025. Many other

112 The BudgeT and economic ouTlook: 2018 To 2028 APRIL 2018

taxpayers will not change their behavior as a result of the

provisions’ temporary nature. at might occur because

they cannot change the timing of their taxable income,

because they expect policymakers to permanently extend

the provisions, or because they are unaware of the expira-

tion dates.

Temporary Reduction in Individual Income Tax Rates.

Under prior law, taxable ordinary income earned by most

individuals was subject to the following seven statutory

rates: 10percent, 15percent, 25percent, 28percent,

33percent, 35percent, and 39.6percent.

Dierent

rates applied to dierent brackets of people’s taxable

ordinary income. e 2017tax act retains the seven-rate

structure but reduces most of the rates; the new rates are

10percent, 12percent, 22percent, 24percent, 32per-

cent, 35percent, and 37percent. e act also expands

the width of the brackets, increasing the number of

taxpayers subject to lower rates.

e lower tax rates are projected to increase the supply

of labor.

Because they will increase after-tax returns on

investment, they are also anticipated to boost investment

by pass-through businesses, which are taxed through the

individual income tax.

Temporary Reduction in the Amount of Income

Subject to the Alternative Minimum Tax. Some tax-

payers are subject to the alternative minimum tax

(AMT), which was intended to impose taxes on higher-

income people who use tax preferences to greatly reduce

or even eliminate their liability under the regular income

tax. e AMT allows fewer exemptions, deductions, and

tax credits than the regular income tax does, and tax-

payers are required to pay the AMT if it is higher than

their regular tax liability. e 2017tax act temporarily

increases the income levels at which the AMT takes

eect. As a result, less income is subject to the AMT.

9. Taxable ordinaryincome is all income subject to the individual

income tax (other than most long-term capital gains and

dividends) minus adjustments, exemptions, anddeductions.

10. For more information, see “Key Methods at CBO Used

to Estimate the Macroeconomic Eects of the 2017 Tax Act”

(supplemental material for e Budget and Economic Outlook:

2018 to 2028, April 2018), https://go.usa.gov/xQcZD.

11. For discussion of that kind of taxation, see Congressional Budget

Oce, Taxing Businesses rough the Individual Income Tax

(December 2012), www.cbo.gov/publication/43750.

e changes to the AMT have eects similar to those

resulting from the reductions in statutory rates. However,

the eect on the labor supply is likely to be smaller,

because higher-income people, most of whom are already

working full time, are less likely to increase their supply

of labor than the population as a whole is.

Temporary Changes to the Standard Deduction and

Itemized Deductions. When preparing their income tax

returns, taxpayers may either take the standard deduc-

tion, which is a at dollar amount, or itemize—that is,

deduct certain expenses, such as state and local taxes,

mortgage interest, charitable contributions, and some

medical expenses. Taxpayers benet from itemizing

when the value of their deductions exceeds the standard

deduction. Under prior law, however, the total amount

of most itemized deductions was generally reduced by

3percent of the amount by which a taxpayer’s adjusted

gross income exceeded a specied threshold.

Other

restrictions applied to specic itemized deductions.

e 2017tax act nearly doubles the size of the standard

deduction and repeals the overall limit on itemized

deductions, but it also eliminates many small itemized

deductions and reduces the amounts that can be claimed

for two widely used deductions. First, deductions for

state and local taxes—the sum of property taxes and

either income or sales taxes—may not exceed $10,000.

Second, taxpayers may deduct the interest on no more

than $750,000 of home mortgage debt, a reduction from

$1.1million under prior law.

e combination of the higher standard deduction and

the restrictions on the two widely used deductions has a

number of eects:

•

e number of taxpayers itemizing deductions

is expected to fall from 49million in 2017 to

18million in 2018.

•

After-tax income changes for many taxpayers. e

increase in the standard deduction causes after-

tax income to rise for most taxpayers who did not

previously itemize deductions. After-tax income also

rises for some higher-income taxpayers, because the

eect of restricting the two widely used deductions

is oset by the repeal of the limit on total itemized

12. Adjusted gross income includes income from all sources not

specically excluded by the tax code, minus certain deductions.

113appendiX B: The eFFecTS oF The 2017TaX acT on cBo’S economic and BudgeT proJecTionS The BudgeT and economic ouTlook: 2018 To 2028

deductions. However, after-tax income falls for some

homeowners and residents of states and localities with

high taxes.

•

e restrictions aect the mix of investment.

By applying caps to state and local property tax

deductions and by limiting the amount of deductible

mortgage interest, the act reduces the incentive to

buy a house, or to invest in housing in other ways,

in relation to the incentive to make other kinds of

investment.

Temporary Repeal of Personal Exemptions and

Expansion of the Child Tax Credit. Under prior law,

taxpayers could generally claim a personal exemption for

themselves and each dependent. at exemption reduced

their tax burden. In addition, taxpayers with income

below specied thresholds were eligible for a tax credit

of up to $1,000 for each qualifying child under the age

of 17.

at credit was partially refundable (mean-

ing that eligible people received money back from the

government if the value of the credit was greater than the

amount of taxes that they owed).

e act repeals the personal exemption but doubles the

size of the maximum child tax credit for most eligi-

ble taxpayers; in addition, eligibility for the credit is

extended to include more higher-income taxpayers. e

maximum refundable portion is increased to $1,400.

Taxpayers can also claim a new $500nonrefundable tax

credit for each dependent who is not a qualifying child.

e eects of those provisions vary among groups of tax-

payers. After-tax income is projected to decline for most

taxpayers, including those without dependents who will

no longer benet from the personal exemption and many

other taxpayers for whom the expanded credits do not

compensate for the loss of the personal exemption. For

many lower-income taxpayers with children, however,

after-tax income will increase. at eect occurs because

many people with low income do not pay income taxes

and will therefore not be aected by the elimination of

the personal exemption but will still benet from the

expanded refundable credit if they have children.

13. For more information about the child tax credit, see

Congressional Budget Oce, Refundable Tax Credits (January

2013), www.cbo.gov/publication/43767.

Changing the Estate and Gi Taxes

e value of property transferred at death and of certain

gifts made during a person’s lifetime is subject to the

federal estate and gift taxes.

However, such transfers up

to a certain cumulative dollar amount are exempt from

taxation. e 2017tax act doubles the amount between

2018 and 2025.

at increase gives people a greater incentive to hold

assets and transfer them at death. In addition, the expira-

tion of the increase at the end of 2025 is likely to induce

people to make gifts before 2026.

Eliminating the Penalty for Not Having Health

Insurance

e Aordable Care Act includes a provision, generally

called the individual mandate, that requires most people

to have health insurance meeting specied standards

and that imposes a penalty on those who do not comply

(unless they have an exemption). Under prior law, the

size of the penalty was the greater of two quantities: a

xed amount specied in law, or a specied fraction of

a household’s income. e tax act reduces the size of the

penalty to zero, starting in 2019.

Because the size of the penalty increased with house-

hold income, it acted as a tax on income. In addition,

it encouraged some people to buy subsidized insur-

ance through the marketplaces established under the

Aordable Care Act; the result was that they faced higher

marginal tax rates, because those subsidies shrink as

income rises. Both of those eects discouraged work, so

the elimination of the penalty is projected to increase the

labor supply slightly.

In addition, eliminating the penalty is expected to make

insurance premiums in the nongroup market, where

insurance is purchased individually, higher than they

would otherwise have been. Insurers are required to

provide coverage to any applicant, and they cannot vary

premiums to reect enrollees’ health status or to limit

14. For more information about those taxes, see Congressional

Budget Oce, Federal Estate and Gift Taxes (December 2009),

www.cbo.gov/publication/41851.

15. For further discussion of those eects, see Edward Harris and

Shannon Mok, How CBO Estimates the Eects of the Aordable

Care Act on the Labor Market, Working Paper 2015-09

(Congressional Budget Oce, December 2015), www.cbo.gov/

publication/51065.

114 The BudgeT and economic ouTlook: 2018 To 2028 APRIL 2018

coverage of preexisting medical conditions. ose fea-

tures are most attractive to applicants with relatively high

expected costs for health care, so eliminating the penalty

will tend to reduce insurance coverage less among older

and less healthy people than among younger and health-

ier people, boosting premiums.

Requiring an Alternative Ination Measure to Adjust

Tax Provisions

Many parameters of the tax system are adjusted for ina-

tion, including the individual income tax brackets. ose

adjustments prevent a general increase in prices from

increasing taxes. Under prior law, most of those adjust-

ments were based on changes in the consumer price

index for urban consumers (CPI-U), which is a measure

of ination calculated by the Bureau of Labor Statistics

(BLS). Beginning in 2018, the measure used for adjust-

ing most parameters of the tax system will be changed to

the chained CPI-U. Whereas the CPI-U measures ina-

tion in the price of a xed “basket” of goods, the chained

CPI-U allows for adjustments in spending patterns by

consumers; also, unlike the CPI-U, it is little aected

by statistical bias related to the sample sizes BLS uses in

computing each index. For both reasons, the chained

CPI-U grows more slowly than the CPI-U does.

CBO’s projections, the former grows more slowly than

the latter by 0.25percentage points per year, on average.

e change in the measure of ination will increase

revenues because it will accelerate a phenomenon called

real bracket creep, in which income is pushed into higher

and higher tax brackets because it is rising faster than

ination. Real bracket creep results in individuals’ facing

higher marginal tax rates, so it reduces the incentive

to work. Unlike many of the tax act’s changes to the

individual income tax, this change is permanent, and the

resulting increase in revenues will grow over time.

In 2026, the temporary provisions of the act that push

down marginal tax rates will have expired. Because the

change in the measure of ination pushes up marginal

16. For more discussion, see Congressional Budget Oce, Repealing

the Individual Health Insurance Mandate: An Updated Estimate

(November 2017), www.cbo.gov/publication/53300.

17. For more information, see the testimony of Jerey Kling,

Associate Director for Economic Analysis, Congressional Budget

Oce, before the Subcommittee on Social Security of the House

Committee on Ways and Means, Using the Chained CPI to Index

Social Security, Other Federal Programs, and the Tax Code for

Ination (April 18, 2013), www.cbo.gov/publication/44083.

rates, the eective marginal rate on labor income will be

higher, beginning in that year, than it would have been

under prior law, CBO estimates.

How the Act Aects the Economic Outlook

In CBO’s projections, the eect of the 2017tax act is to

boost the average amount of real GDP by 0.7percent

over the 2018–2028period (see Table B-2). Real GDP

is boosted by 0.3percent in 2018 and by 0.6percent

in 2019, and the eect peaks at 1.0percent in 2022. In

later years, the eect is smaller, and by 2028it has fallen

to an increase of 0.5percent. at pattern arises because

the act’s eects on real GDP growth are positive initially

and then negative.

Like real GDP, real potential GDP is higher in every

year of the 11-year period because of the tax act. But

through 2022, the increase in real GDP is greater than

the increase in real potential GDP (see Figure B-1). e

result is that the positive output gap—the amount by

which real GDP exceeds real potential GDP—is larger

than it would have been otherwise. (Even without the

act, real GDP would have been greater than real poten-

tial GDP in CBO’s baseline projections.)

at larger output gap through 2022puts some upward

pressure on prices. Ination (as measured by the price

index for personal consumption expenditures) is pro-

jected to be slightly higher than it would have been

otherwise over the rst several years of the period and

then to be unchanged.

In CBO’s projections, the larger output gap and greater

inationary pressure prompt the Federal Reserve to

respond by pushing interest rates higher over the next

few years than they would have been without the tax act.

e rate for 3-month Treasury bills is higher by 0.5per-

centage points by 2022, and the rate for 10-year Treasury

notes is 0.2percentage points to 0.3percentage points

higher during the 2018–2022period. ose higher inter-

est rates and the end of the act’s cuts in personal income

taxes in 2025slow the growth of real GDP, reducing the

pressure on prices and interest rates. However, as a result

of greater federal borrowing and certain provisions of the

tax act that aect portfolio decisions, interest rates on

10-year notes are still slightly higher by 2028 than they

would have been otherwise.

e projected gains in output generate increases in

income for the employees and owners of the businesses

115appendiX B: The eFFecTS oF The 2017TaX acT on cBo’S economic and BudgeT proJecTionS The BudgeT and economic ouTlook: 2018 To 2028

Table B-2 .

Economic Eects of the 2017 Tax Act

Average

2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028

2018–

2022

2023–

2028

2018–

2028

Output (Percent)

Real GDP 0.3 0.6 0.8 0.9 1.0 0.9 0.9 0.9 0.6 0.6 0.5 0.7 0.7 0.7

Real potential GDP 0.2 0.4 0.6 0.8 0.9 0.9 0.9 0.9 0.7 0.6 0.5 0.6 0.8 0.7

Nominal GDP 0.4 0.8 0.9 1.1 1.2 1.2 1.2 1.1 0.9 0.8 0.8 0.9 1.0 0.9

Real GNP 0.1 0.4 0.5 0.5 0.6 0.6 0.6 0.6 0.3 0.2 0.1 0.4 0.4 0.4

Contribution of Components to Real GDP

(Percentage points)

Private consumption 0.4 0.6 0.6 0.6 0.8 0.9 0.8 0.7 0.5 0.4 0.3 0.6 0.6 0.6

Private nonresidential ﬁxed investment 0.2 0.4 0.5 0.5 0.4 0.3 0.3 0.3 0.2 0.1 0.1 0.4 0.2 0.3

Private residential investment * -0.1 -0.1 -0.1 -0.1 -0.1 -0.1 -0.1 * * * -0.1 -0.1 -0.1

Government consumption and investment ** ** 0.1 0.1 0.1 0.1 0.1 0.1 0.1 ** ** ** 0.1 0.1

Net exports -0.3 -0.3 -0.2 -0.2 -0.2 -0.2 -0.1 -0.1 ** ** 0.1 -0.2 * -0.1

Exports -0.2 -0.1 -0.1 -0.1 * * * ** ** ** ** -0.1 ** *

Imports

-0.1 -0.2 -0.2 -0.1 -0.1 -0.2 -0.1 -0.1 * ** ** -0.2 -0.1 -0.1

Potential Labor and Productivity (Percent)

Potential labor force 0.1 0.3 0.4 0.4 0.5 0.5 0.5 0.4 0.3 0.2 0.2 0.3 0.4 0.3

Potential average labor hours 0.1 0.2 0.3 0.3 0.3 0.3 0.3 0.3 0.2 0.1 * 0.3 0.2 0.2

Potential total labor hours 0.2 0.5 0.7 0.8 0.8 0.8 0.8 0.7 0.5 0.3 0.1 0.6 0.5 0.6

Potential labor productivity * -0.1 * ** 0.1 0.1 0.1 0.2 0.3 0.3 0.3 * 0.2 0.1

Employment and Unemployment

Total nonfarm employment (Percent)† 0.2 0.5 0.6 0.7 0.7 0.7 0.7 0.7 0.6 0.6 0.6 0.5 0.6 0.6

Unemployment rate (Percentage points) * -0.1 -0.1 -0.1 -0.1 * * ** ** ** ** -0.1 ** *

PCE Price Level (Percent)

** ** 0.1 0.1 0.1 0.1 0.2 0.2 0.2 0.2 0.2 0.1 0.2 0.1

Interest Rates (Percentage points)

Federal funds rate 0.1 0.2 0.4 0.4 0.5 0.3 ** * * * ** 0.3 0.1 0.2

Three-month Treasury bills 0.2 0.2 0.4 0.4 0.5 0.3 0.1 * * * ** 0.3 0.1 0.2

Ten-year Treasury notes 0.3 0.2 0.2 0.2 0.2 0.1 ** ** ** ** ** 0.2 ** 0.1

International Measures

Net international lending as a percentage

of GDP (Percentage points) -0.4 -0.4 -0.4 -0.4 -0.5 -0.5 -0.4 -0.3 -0.2 -0.2 -0.2 -0.4 -0.3 -0.4

Net international income as a percentage

of GDP (Percentage points) -0.1 -0.2 -0.3 -0.3 -0.4 -0.4 -0.3 -0.3 -0.3 -0.3 -0.4 -0.3 -0.3 -0.3

Export-weighted exchange rate (Percent)

1.8 1.7 1.9 1.9 1.8 1.7 1.6 1.6 1.6 1.5 1.5 1.8 1.6 1.7

Memorandum:

Real GDP Growth (Percentage points) 0.3 0.3 0.2 0.1 0.1 * * -0.1 -0.2 -0.1 -0.1 0.2 -0.1 **

PCE Price Inﬂation (Percentage points) ** ** ** ** ** ** ** ** ** * * ** ** **

Source: Congressional Budget Oce.

Real values are nominal values that have been adjusted to remove the eects of inﬂation.

GDP = gross domestic product; GNP = gross national product; PCE = personal consumption expenditures; * = between -0.05 percent or percentage points

and zero; ** = between zero and 0.05 percent or percentage points.

a. A negative value indicates an increase in imports.

b. A higher value indicates an increase in the exchange value of the dollar.

[†Values corrected on April 17, 2018]

116 The BudgeT and economic ouTlook: 2018 To 2028 APRIL 2018

that produce the output. So employees’ total compen-

sation rises in CBO’s projections, as do their wages and

salaries. (Total compensation includes not only wages

and salaries but also bonuses, stock options, benets, and

the employer’s share of payroll taxes for social insurance

programs.) Corporate prots and business income also

increase.

Other organizations have also estimated the economic

eects of the 2017tax act (see Box B-2). e forecasts

vary, but most show increases in the level of real GDP

over the rst few years and a more moderate increase

by2027.

Eects on Potential Output

Various provisions of the 2017tax act directly aect the

productive potential of the U.S. economy. ey do so

by promoting increases in investment and the potential

labor supply. e act is also projected to raise measured

total factor productivity, which is the average real output

per unit of combined labor and capital services. On net,

the act is projected to raise the level of potential output

throughout the 2018–2028period. e eect on poten-

tial output peaks at 0.9percent in the middle years of

the period and declines to 0.5percent in 2028. In CBO’s

projections, the act’s contribution to real GDP at the end

of the period results from an increase in the amount of

potential output.

Private Investment. Increases in investment boost

potential output by increasing the stock of capital

goods—structures, equipment, intangible assets, and

inventories—that are used to produce output. e

act aects private investment through three channels:

Figure B-1 .

Economic Eects of the 2017 Tax Act at a Glance

-1.0

-0.5

0.5

1.0

1.5

0.2

0.4

0.6

0.8

1.0

-0.2

0.2

0.4

0.6

0.1

0.2

Percent Percentage of Potential GDP

Percentage PointsPercent

The act boosts real GDP in relation to real potential GDP in the

near term.

That change creates additional excess demand in the

economy, raising the output gap, . . .

. . . putting some upward pressure on consumer prices, . . . . . . and pushing up interest rates.

1 2

3 4

All Other Eects

Eects of the

2017 Tax Act

3-Month

Treasury

Bills

Real GDP

Real Potential

GDP

10-Year

Treasury Notes

2017 2019 2021 2023 2025 2027

Source: Congressional Budget Oce.

Real values are nominal values that have been adjusted to remove the eects of inﬂation. Potential GDP is CBO’s estimate of the maximum sustainable

output of the economy. Excess demand exists when the demand for goods and services exceeds the amount that the economy can sustainably supply.

The output gap is the dierence between GDP and CBO’s estimate of potential GDP and is expressed as a percentage of potential GDP. Consumer

prices are measured by the price index for personal consumption expenditures.

GDP = gross domestic product.

117appendiX B: The eFFecTS oF The 2017TaX acT on cBo’S economic and BudgeT proJecTionS The BudgeT and economic ouTlook: 2018 To 2028

Box B-2 .

Comparison With Other Organizations’ Estimates

Various organizations other than the Congressional Budget

Oce have estimated the economic eects of the 2017 tax act.

In general, the organizations expect the act to increase the

level of real gross domestic product (GDP) throughout the peri-

ods that they examine. Many of the forecasts follow a pattern

similar to the one followed by CBO’s projections: increasing

positive eects on real GDP over the ﬁrst several years, then a

moderation, and then a more muted eect by 2027.

In the organizations’ projections for the 2018–2022 period,

the act’s expected average eect on real GDP ranges from

0.3percent to 1.3 percent; CBO’s projection is 0.7 percent. For

the 2023–2027 period, the average eect ranges from 0.3 per-

cent to 2.9 percent; CBO’s projection is 0.8 percent. In 2027,

the projected eect ranges from −0.1 percent to 2.9 percent;

CBO’s projection is 0.6 percent.

CBO limited its comparison to forecasts that broadly exam-

ined the ﬁnal version of the tax act. Other forecasts examined

earlier versions of the act or only parts of it, so CBO did not

include them in the comparison.

Assorted Estimates of the Eects of the 2017 Tax Act on the Level of Real GDP

Percent

First Five Years

Tenth

Year Average

2018 2019 2020 2021 2022 2027

2018–

2022

2023–

2027

2018–

2027

Moody's Analytics 0.4 0.6 0.2 0.1 0.0 0.4 0.3 0.3 0.3

Macroeconomic Advisers 0.1 0.3 0.5 0.6 0.6 0.2 0.4 0.5 0.5

Tax Policy Center

0.8 0.7 0.5 0.5 0.5 * 0.6 0.3 0.5

International Monetary Fund 0.3 0.9 1.2 1.2 1.0 -0.1 0.9 0.3 0.6

Joint Committee on Taxation – – – – – 0.1 to 0.2 0.9 0.6 0.7

Congressional Budget Oce 0.3 0.6 0.8 0.9 1.0 0.6 0.7 0.8 0.7

Goldman Sachs 0.3 0.6 0.7 0.7 0.7 0.7 0.6 0.7 0.7

Tax Foundation 0.4 0.9 1.3 1.8 2.2 2.9 1.3 2.9 2.1

Penn Wharton Budget Model – – – – – 0.6 to 1.1 – – –

Barclays 0.5 – – – – – – – –

Sources: Congressional Budget Oce and the organizations listed above.

Real values are nominal values that have been adjusted to remove the eects of inﬂation.

GDP = gross domestic product; –=not available; * = between -0.05 percent and zero.

a. Values are for ﬁscal years.

changes in incentives, crowding out (which occurs when

larger federal decits reduce the resources available for

private investment), and changes in economic activity.

18. CBO estimated the act’s eects on investment in 32 types of

equipment, 23 types of nonresidential structures, 3 types of IP

products, 3 types of residential capital, and inventories. For more

information, see “Key Methods at CBO Used to Estimate

the Macroeconomic Eects of the 2017 Tax Act” (supplemental

material for e Budget and Economic Outlook: 2018 to 2028,

April 2018), https://go.usa.gov/xQcZD.

Some of the changes to investment are nanced by

domestic investors and some are nanced by foreign

investors, resulting in changes to international invest-

ment ows.

In CBO’s projections, total business xed investment—

which consists of investment in nonresidential struc-

tures, equipment, and IP products—is higher in every

year from 2018 through 2028 than it would otherwise

have been. It is boosted by changes in incentives and

118 The BudgeT and economic ouTlook: 2018 To 2028 APRIL 2018

stronger economic activity but dampened by crowding

out from increased federal borrowing (see Figure B-2).

By contrast, residential investment is lower in every year

from 2018 through 2028 than it would otherwise have

been. Incentives to undertake residential investment are

reduced through 2025by limits on the deductibility of

property taxes and mortgage interest, as well as by fewer

households’ itemizing deductions. Residential investment

is reduced throughout the entire period by crowding out.

Changes in Incentives. e tax act aects investment

in the United States by changing incentives to invest,

including the user cost of capital and thus the minimum

return that an investment must achieve to be protable.

e act reduces the user cost of capital in various ways.

Some provisions do so by reducing statutory tax rates.

Extending bonus depreciation also reduces the user cost

of capital. However, the act increases the user cost of

capital for owner-occupied housing from 2018 through

2025 and for research and development beginning in

2022.

e act species several signicant changes in 2026that

aect the user cost of capital for pass-through businesses

and for homeowners. As a result, their response to the

19. e incentives and crowding out that aect business xed

investment also aect investment in inventories.

tax act depends partly on their expectations of future tax

policy. In CBO’s projections, 20percent of investment is

made by businesses and households that expect provi-

sions scheduled to end in 2026actually to do so, and

80percent of investment activity is consistent with the

provisions’ being extended.

(e act also includes some

less signicant changes in scal policy over the 11-year

period, and CBO incorporated the projection that all

businesses and households behave as if they expect those

changes to occur.)

e tax act aects the user cost of capital in dierent

ways for the three kinds of xed business investment and

for residential investment (see Figure B-3).

•

Investment in equipment is projected to benet the

most from changes in the user cost of capital because

of lower statutory tax rates and the extension of

100percent bonus depreciation through 2022. e

allowed amount of bonus depreciation declines over

the following several years, and by 2027, the increase

20. ose projections of expectations are based on historical

responses to extensions of major tax provisions. For more

information, see “Key Methods at CBO Used to Estimate

the Macroeconomic Eects of the 2017 Tax Act” (supplemental

material for e Budget and Economic Outlook: 2018 to 2028,

April 2018), https://go.usa.gov/xQcZD.

Figure B-2 .

Eects of the 2017 Tax Act on Business Fixed Investment

Billions of Dollars

Page 1 of 1

Crowding Out

-60

-40

-20

100

2017 2019 2021 2023 2025 2027

Changes in

Incentives

Changes in

Economic Activity

Business ﬁxed investment is greater

through 2028 because the positive

eects of changes in incentives and

economic activity oset the negative

eects of crowding out.

Source: Congressional Budget Oce.

Business ﬁxed investment is businesses’ purchases of equipment, nonresidential structures, and intellectual property products. The changes in

incentives consist of changes in the user cost of capital, which is the gross pretax return on investment that provides the required return to investors

after covering taxes and depreciation, and changes in the beneﬁts of locating business establishments in the United States. Changes in economic

activity consist of changes in demand for goods and services and changes in the supply of labor. Crowding out occurs when larger federal deﬁcits

reduce the resources available for private investment.

119appendiX B: The eFFecTS oF The 2017TaX acT on cBo’S economic and BudgeT proJecTionS The BudgeT and economic ouTlook: 2018 To 2028

in investment that is due to changes in the user cost

of capital stems almost entirely from the reduction in

the corporate tax rate.

•

Investment in nonresidential structures also benets

from lower statutory tax rates. In addition, certain

types of structures with relatively short tax lives, such

as oil derricks, benet from bonus depreciation. But

by 2027, as with the previous category, the increase

in investment that is due to changes in the user cost

of capital stems almost entirely from the reduction in

the corporate tax rate.

•

Investment in IP products is boosted by changes

in the user cost of capital through 2021. However,

in contrast to its treatment of equipment, the tax

act makes depreciation less generous for R&D

and for software development beginning in 2022.

Consequently, starting in that year, investment in IP

products is lower than it would otherwise have been.

•

e bulk of residential investment is in owner-

occupied housing. e tax act increases the user cost

of capital for homeowners from 2018 to 2025by

limiting the deductibility of property taxes and

mortgage interest and by reducing the number of

households that itemize. at increase outweighs a

reduction in the user cost of capital for the people or

pass-through businesses that own most rental housing

and that will benet from lower individual tax rates

during that period. Beginning in 2026, the act has

little impact on the user cost of residential capital.

e tax act also increases incentives to invest in the

United States by encouraging rms to locate their estab-

lishments here. e primary means of encouragement is

the reduction in the statutory corporate tax rate in the

Figure B-3 .

Eects of the 2017 Tax Act on Investment Through Changes in Incentives

Billions of Dollars

-40

-20

-40

-20

2017 2019 2021 2023 2025 2027

Equipment

Nonresidential Structures

Intellectual Property

Products

Residential

Structures

The act’s changes in the tax treatment

of depreciation eventually reduce the

eects of incentives on investment in

equipment and intellectual property

products.

Limits on the tax deductibility of

payments for property taxes and

mortgage interest, along with a drop

in the number of households that take

itemized deductions, reduce spending on

residential structures through 2025.

Source: Congressional Budget Oce.

The changes in incentives consist of changes in the user cost of capital, which is the gross pretax return on investment that provides the required return

to investors after covering taxes and depreciation, and changes in the beneﬁts of locating business establishments in the United States.

120 The BudgeT and economic ouTlook: 2018 To 2028 APRIL 2018

United States. However, that eect is partly oset by

other changes. For example, the GILTI and FDII provi-

sions may increase the incentive to locate tangible assets

outside the United States.

Furthermore, although the increased incentives to locate

establishments in the United States will boost total

investment, that eect is muted by the amount of labor

available, in CBO’s estimation. In other words, barring

a change in the amount of labor supplied in the United

States, business location decisions are projected to have

only a limited eect on investment. at is because the

additional labor used by an establishment locating in the

United States is no longer available to other establish-

ments. So the increased investment by the new establish-

ment is partly oset by reduced investment by existing

establishments.

Crowding Out. CBO estimates that greater federal

borrowing ultimately reduces private investment. When

the government borrows, it borrows from people and

businesses whose savings would otherwise be nancing

private investment. Although an increase in government

borrowing strengthens the incentive to save, the resulting

rise in saving is not as large as the increase in government

borrowing; national saving, or the amount of domes-

tic resources available for private investment, therefore

falls. However, private investment falls less than national

saving does in response to government decits, because

the higher interest rates that are likely to result from

increased federal borrowing tend to attract more foreign

capital to the United States. In CBO’s assessment, the

crowding out of private investment occurs gradually, as

interest rates and the funds available for private invest-

ment adjust in response to increased federal decits.

e reduction in private investment resulting from

crowding out occurs primarily because of higher interest

rates, so the eects on dierent categories of investment

depend on how sensitive they are to interest rates. In

general, interest rates constitute a larger share of the user

cost of capital for types of capital that depreciate slowly,

so changes in interest rates have a larger eect on invest-

ment in those types of capital. For example, a 1percent

rise in mortgage rates would have a larger impact on

residential investment than a 1percent rise in corpo-

rate bond rates would have on businesses’ purchases of

computers. Consequently, investment in residential and

nonresidential structures bears a disproportionate share

of the impact of larger decits. e act’s crowding-out

eects vary not only by type of investment but also as

time passes; the strongest eects occur in 2022, when the

act’s eects on the decit are largest (see Figure B-4).

Changes in Economic Activity. When demand for their

output increases, businesses invest in capital to meet

that additional demand; the expanded investment then

increases the potential output of the economy, because

a larger capital stock increases the businesses’ ability to

produce output. e impact on investment is greatest

during the period in which demand is accelerating. Once

businesses have invested enough to meet the additional

Figure B-4 .

Eects of the 2017 Tax Act on Investment Through Crowding Out

Billions of Dollars

Page 1 of 1

-30

-25

-20

-15

-10

-5

2017 2019 2021 2023 2025 2027

Equipment

Nonresidential Structures

Intellectual Property Products

Residential

Structures

The reduction of investment resulting

from crowding out is greatest in 2022,

when the eects of the act on the

federal deﬁcit are the largest.

Source: Congressional Budget Oce.

Crowding out occurs when larger federal deﬁcits reduce the resources available for private investment.

121appendiX B: The eFFecTS oF The 2017TaX acT on cBo’S economic and BudgeT proJecTionS The BudgeT and economic ouTlook: 2018 To 2028

demand, the only further stimulus to investment is the

need to gradually replace the additional capital.

In CBO’s projections, the tax act increases demand

primarily by increasing households’ demand for goods

and services over the next few years, widening the output

gap. Consequently, rms engage in investment to meet

that demand beyond what they would do in response to

changing tax incentives. e act’s eect on investment

through that channel is positive during the period when

the output gap is growing more rapidly than it would

have in the absence of the act and negative when it is

growing less rapidly.

e act is also projected to expand investment through

another change in economic activity: increasing the labor

supply. Businesses must purchase additional capital for

the new workers to use. However, because rms adjust

their stocks of capital more slowly than they adjust the

number of their employees, the response of investment

to changes in the labor supply is gradual.

How the Increase in Investment Is Financed. e projected

increase in U.S. investment would be nanced by private

domestic and foreign saving. In CBO’s projections, the

private domestic saving rate initially rises in response to

the higher after-tax rates of return on U.S. investment

resulting from the tax act. In addition, because the act

boosts U.S. economic output, national income rises,

and total private domestic saving grows. (However,

some portion of the increased private domestic saving

is used to nance increased federal borrowing, reducing

the amount of saving available for private investment.)

Earnings subject to deemed repatriation are expected to

be used primarily to reduce corporate debt and thus to

contribute only slightly to nancing the increase in pri-

vate investment (see Box B-1 on page 109). Meanwhile,

increases in the rate of return on investment in the

United States in relation to the rate in other countries

will attract additional inows of foreign saving. CBO

estimates, therefore, that a substantial portion of the

increase in private investment will be nanced through

those inows.

Potential Labor Supply. In CBO’s projections, the

2017tax act also boosts potential output by increasing

the potential supply of labor through increases in the

potential labor force participation rate and in hours

worked per worker. e potential labor force participa-

tion rate is higher by an annual average of 0.2 percentage

points during the 2018–2028 period; the peak eect is

0.3 percentage points in 2023 and 2024.

Total potential hours worked, the result of increases

in both the potential labor force participation rate

and average weekly hours, rise by an annual average of

nearly 0.6percent. e peak increase in potential hours

worked—more than 0.8percent—occurs in 2023; by

2028, the eect has dwindled to about 0.1percent. CBO

estimates that more than half of the projected eects on

the overall potential labor supply result from increases in

the potential labor force participation rate. e remain-

der result from increases in average weekly hours.

ose eects occur because the tax act changes incentives

to work, particularly by lowering statutory individual

income tax rates and by making other changes that lower

marginal tax rates through 2025.

In the following

years, however, most of the relevant provisions that lower

tax rates expire, and marginal rates will be higher than

under prior law, primarily because of the new measure

of ination that the act species for adjusting various

parameters of the tax system. As a result, the act reduces

incentives to work in those years. An exception is the

act’s elimination of the penalty for not having health

insurance. at elimination is permanent, so its eect on

the potential labor supply—slightly increasing it, in part

because the size of the penalty increased as household

income increased, causing it to act as a tax on income—

is projected to be permanent.

CBO expects that it will take time for people to respond

to provisions in the act. e agency’s estimates therefore

account for the time that it takes for people to under-

stand the act’s eects and to make adjustments in how

much they work. For example, the estimates reect the

speed with which people are expected to increase their

supply of labor in response to lower tax rates in the early

years of the 11-year period and to decrease that supply

after provisions expire later on.

21. Even if that estimate of the relative shares were dierent, the

estimated change in total potential hours worked would not

change, and therefore the estimate of potential output would not

either.

22. For more information, see “Key Methods at CBO Used

to Estimate the Macroeconomic Eects of the 2017 Tax Act”

(supplemental material for e Budget and Economic Outlook:

2018 to 2028, April 2018), https://go.usa.gov/xQcZD.

122 The BudgeT and economic ouTlook: 2018 To 2028 APRIL 2018

Also, as with expectations about capital costs, CBO

incorporated the projection that 20percent of people

anticipate the scheduled expiration of many of the bill’s

provisions in 2025. ose people respond by supplying

more labor in the years when tax rates are scheduled

to be temporarily low. ey also begin reducing their

supply of labor even before the rates are scheduled to

increase, because such adjustment is costly. People who

are projected to be surprised by the act’s change in tax

rates have more muted responses to the lower rates before

2025 and also a more muted response to the increase

afterward. Taken together, over the 11-year period,

CBO’s projections of the average labor response to the

tax act are not much aected by the agency’s projections

of people’s dierent expectations.

Potential Productivity. Over the rst few years of the

2018–2028period, CBO projects, the 2017tax act

will not have much net eect on potential labor pro-

ductivity, which is dened as real potential output per

potential hour of labor (see Table B-2 on page 115). If

the contribution of capital to output rises more than the

contribution of potential hours of labor, potential labor

productivity rises. At rst, the act is projected to boost

hours and capital by similar amounts, so the eect on

potential labor productivity is small. But in later years,

the contribution of capital to output has increased more

than the contribution of potential hours, and by 2027,

potential labor productivity is increased by 0.3percent.

Because the increase in the level of potential labor pro-

ductivity is roughly unchanged between 2027 and 2028,

it has little eect on potential output growth by the end

of the 11-year period.

e act is also projected to raise potential output slightly

by discouraging prot-shifting strategies that histori-

cally have suppressed measured total factor productiv-

ity. e act is expected to encourage rms to claim as

domestic production the services of IP that were pre-

viously claimed as production abroad (see Box B-3 on

page 124). In CBO’s estimation, even though the rms

made that claim, those services have been and continue

to be generated by IP assets that are included in estimates

of the domestic capital stock. As a result, the shift in the

reported location of services associated with that IP will

result in an increase in measured domestic output even

though there is no corresponding increase in measured

domestic inputs of labor or capital. Another way of look-

ing at the shift is that more reported production is being

generated by the same measured amount of labor and

capital. at is the denition of an increase in total factor

productivity. CBO has therefore adjusted its projections

of potential total factor productivity by only a slight

amount each year to account for the anticipated increase

in output that is not matched by an increase in inputs.

Eects on Actual Output

In CBO’s projections, the 2017tax act boosts the

demand for goods and services, accelerating the growth

of actual output in relation to the growth of potential

output over the rst half of the 2018–2028period. As

a result, the output gap is 0.1percentage point larger

between 2018 and 2022 than it would have been other-

wise, on average. Heightened overall demand is projected

to increase consumer spending, increase employment

further above CBO’s estimate of its potential level,

reduce net exports (that is, exports minus imports), and

slightly increase ination. However, because most pro-

visions of the act that relate to individual income taxes

expire and thus subtract from overall demand after 2025,

the output gap is 0.1percentage point smaller in 2026

and slightly smaller in 2027 than it would have been

otherwise.

Consumer Spending. e eect of the act on real GDP

over the next few years derives largely from its impact on

consumer spending. e act reduces individual income

tax revenues, increasing households’ disposable income

and thereby their spending. e changes to individual

income taxes include temporary changes to tax rates,

the standard deduction, the personal exemption, the

child tax credit, itemized deductions, and the alternative

minimum tax.

Higher- and lower-income households adjust their

spending dierently, on average, in response to such

increases in disposable income. CBO accounted for

those dierences by assessing the distribution of tax cuts

among income groups.

In CBO’s assessment, lower-

income households spend a larger share of the additional

income in such cases than higher-income households do.

CBO’s estimate of the overall eect on consumer spend-

ing also incorporates the agency’s assessment of the act’s

23. For more information, see “Key Methods at CBO Used

to Estimate the Macroeconomic Eects of the 2017 Tax Act”

(supplemental material for e Budget and Economic Outlook:

2018 to 2028, April 2018), https://go.usa.gov/xQcZD.

123appendiX B: The eFFecTS oF The 2017TaX acT on cBo’S economic and BudgeT proJecTionS The BudgeT and economic ouTlook: 2018 To 2028

impact on equity and housing wealth. In CBO’s projec-

tions, lower corporate taxes contribute to the boost in

consumer spending by increasing the after-tax earnings

of businesses, thereby raising the equity wealth of busi-

nesses’ shareholders. Countering that eect are the act’s

changes related to the standard deduction for individuals

and to the treatment of state and local taxes and mort-

gage interest deductions, which are expected to make

house prices lower than they would be otherwise. CBO

does not expect the provisions that govern repatriation of

businesses’ foreign earnings to aect consumer spending

signicantly (see Box B-1 on page 109).

Furthermore, CBO’s estimate of the act’s impact on

consumer spending accounts for the elimination of the

penalty for not having health insurance. at change

means that people will be less likely to obtain coverage,

decreasing subsidies and aecting consumer spending.

Analysis of the act’s eect on consumer spending is com-

plicated by the fact that most of the changes to individ-

ual income taxes are scheduled to end after 2025. What

people expect about expirations matters; a change in dis-

posable income that they consider transitory is likely to

aect their spending less than one that they expect to last

longer. In CBO’s projections, about 80percent of con-

sumer spending is undertaken by people who believe that

the individual income tax cuts will be extended beyond

2025, and the remainder is undertaken by people who

believe that they will end as scheduled. (ose specica-

tions are analogous to what CBO used for expectations

of scal policies aecting decisions to work and invest.)

But CBO’s estimate of the overall change in consumer

spending in the next few years would not change very

much if the agency used dierent specications, because

the expectations in this case relate to relatively distant

events.

In later years, the end of most provisions related to

individual income taxes slows the growth of consumer

spending. In CBO’s projections, those changes subtract

from disposable income and overall demand in 2026

and2027.

Net Exports. In the near term, the act is projected to

boost real imports, reduce real exports, and therefore

lower real net exports. In CBO’s projections, imports

rise in the near term because the act raises the domestic

demand for goods and services. For example, the capital

investment stimulated by the act will raise demand for

imported capital goods (such as computers and machine

tools) and for imported materials (such as steel and

aluminum). Furthermore, when the domestic economy

is operating above its potential, as it is in CBO’s projec-

tions, additional increases to production are costly and

dicult, making the propensity to import goods and ser-

vices particularly strong. And higher domestic demand

can push exports down as rms concentrate on satisfying

that demand.

In addition, CBO expects the act to moderately increase

the exchange value of the dollar in 2018 (see Table B-2

on page 115).

Increased demand for U.S. assets,

which results mainly from the increase in the rate of

return on those assets, strengthens the dollar in CBO’s

projections. at stronger dollar causes export prices

to rise and import prices to decline. Consequently, real

exports decrease, real imports increase, and real net

exports fall.

CBO expects the act’s initial eects on real net exports to

begin to dissipate after 2019. One reason is that the act’s

eect on the exchange value of the dollar is projected

to gradually decline after 2020. In addition, the expi-

ration of the cuts in individual income taxes dampens

consumer spending and thus imports. By 2026, CBO

expects the act’s eect on real net exports to disappear.

e Labor Market. Over the next few years, the wider

output gap, and the resulting increase in demand for

labor and upward pressure on wages, are projected to

raise employment and hours worked further above

CBO’s estimate of their potential levels. e agency

expects the tax act to initially lower the unemploy-

ment rate by a small amount, slightly widening the gap

between that rate and the natural rate of unemploy-

ment over the 2018–2022 period. (e natural rate of

unemployment is the rate of unemployment that results

from all sources except uctuations in overall demand.)

e unemployment rate is projected to be, on average,

0.1percentage point lower—and the labor force partic-

ipation rate and total hours worked to be, respectively,

0.2percentage points and 0.7percent higher—than they

would have been otherwise between 2018 and 2022.

24. CBO’s measure of the exchange value of the dollar is an export-

weighted average of the exchange rate indexes between the dollar

and the currencies of leading U.S. trading partners. An increase

in that measure indicates that the dollar is appreciating.

124 The BudgeT and economic ouTlook: 2018 To 2028 APRIL 2018

Box B-3 .

The Eects of Proﬁt Shifting on Economic Statistics

The proﬁt-shifting strategies used by multinational corporations

(MNCs) aect many economic indicators. All of the strategies

distort data about U.S. taxable income by inﬂating reported

foreign income while reducing reported domestic income. But

the strategies alter other statistics in dierent ways.

Although the 2017tax act includes a number of provisions that

discourage proﬁt shifting, it may encourage some proﬁt shifting

by exempting foreign dividends from U.S. taxation. On net, the

Congressional Budget Oce projects, the changes in tax law

will reduce proﬁt shifting by roughly $65billion per year, on

average, over the next 11years. Most of that projected reduc-

tion can be attributed to less use of the debt allocation and

intellectual property (IP) transfer strategies discussed below.

Locating MNCs’ Debt in High-Tax Countries. By allocating

a greater share of debt, and the associated deduction for

interest payments, to high-tax countries, an MNC can reduce

the amount of taxable income reported in those high-tax

countries.

In CBO’s projections, the reduction in proﬁt shifting

through decisions about debt location accounts for about half

of the $65billion total reduction in proﬁt shifting resulting from

the tax act.

When a U.S. aliate of an MNC borrows from a foreign bank

on behalf of the entire MNC (thus allocating debt to the United

States), that loan shows up in U.S. international investment

position accounts as an increase in foreign-owned U.S. assets.

The result is a reduction in the United States’ net international

investment position.

Locating debt in the United States can alter net international

lending—which is national saving minus domestic investment—

if that debt is borrowed from foreign investors. Net interna-

tional lending is also equal to the sum of net international

income (which is the dierence between the income earned by

1. MNCs use many strategies to shift proﬁts to low-tax countries. For purposes

of simpliﬁcation, CBO has categorized all of them into the three types

described here.

2. The same incentive exists for a variety of other costs that beneﬁt an MNC,

such as costs for headquarters. CBO focuses on debt both because it is the

mechanism that this strategy usually employs and because the choice of

where to locate debt has economic eects that are similar to those resulting

from the use of the other mechanisms.

U.S. residents from foreign sources and the income earned by

foreign individuals from U.S. sources) and net exports (which

are exports minus imports). The reason that locating debt in

the United States aects net international lending is that the

reduction in the U.S. net international investment position

leads to a reduction in net international income. Because

there is no corresponding change in net exports, net interna-

tional lending declines, along with gross national product. But

because reported production is unaected, gross domestic

product (GDP) is unchanged.

The act’s reduction in the U.S. corporate tax rate, combined

with the new rules governing the deduction of interest, will

reduce some use of this strategy. Before the act was enacted,

a relatively high statutory tax rate made the United States an

attractive location for debt. But now, because the United States

is unlikely to continue to be the highest-taxed jurisdiction for

many MNCs, some will move their debt to aliates in countries

with a higher corporate tax rate.

Transferring Intellectual Property. When an MNC moves its IP

from an aliate in a high-tax country to an aliate in a low-tax

country, that MNC can report less of its taxable income in the

high-tax country and more in the low-tax country. CBO projects

that the tax act’s reductions in proﬁt shifting through the trans-

fer of IP will account for roughly one-third of the total projected

reduction in proﬁt shifting over the next 11years.

Proﬁt shifting through the international transfer of IP distorts

real U.S. product statistics (that is, statistics adjusted to remove

the eects of inﬂation) and real GDP. Royalties and other

revenues derived from IP are counted in the national income

and product accounts—ocial U.S. accounts that track the

amount and composition of GDP, the prices of its components,

and the way in which the costs of production are distributed as

income—as real production of IP services. When IP assets are

transferred from the United States to another country, the real

services derived from those assets are attributed not to the

United States but to the other country, so real net exports and

real GDP are reduced. However, unlike locating debt in high-

tax countries, transferring IP has no eect on net international

lending, because any reductions to net exports associated

with IP transfers are matched by an additional dollar of net

international income.

Continued

125appendiX B: The eFFecTS oF The 2017TaX acT on cBo’S economic and BudgeT proJecTionS The BudgeT and economic ouTlook: 2018 To 2028

CBO estimates that the reduction in the U.S. corporate tax

rate, combined with the new rules governing the treatment of

income from high-return investments (much of which is derived

from IP), will reduce corporations’ incentives to shift proﬁts by

transferring IP outside the United States. However, that eect

is expected to be modest. IP is especially easy to relocate, so

MNCs are typically able to locate it in whichever aliates face

the lowest tax rate on the income that it generates. Because

tax havens outside the United States will continue to have

relatively low tax rates, CBO projects that most IP currently

located there will remain there. For newly created or future

IP, the changes resulting from the tax act and the ﬁxed costs

of transferring IP to foreign aliates will probably deter some

small amount of proﬁt shifting.

Setting Transfer Prices. MNCs can reduce their U.S. taxes by

strategically setting transfer prices—the prices that aliates of

the same MNC charge each other across national boundaries.

To minimize proﬁts earned in high-tax countries, MNCs can

systematically overstate the prices that aliates in high-tax

countries pay for imports from foreign aliates and understate

the prices that aliates in high-tax countries charge for exports

3. Technically, transferring IP to aliates in low-tax countries can also be

categorized as strategically setting transfer prices. However, proﬁt shifting

through IP transfers and proﬁt shifting through setting the transfer prices of

tangible assets distort statistics in dierent ways.

to foreign aliates.

CBO projects that reduced proﬁt shifting

through that strategy will account for only a small portion of

the projected $65billion annual reduction in proﬁt shifting.

That strategy tends to distort reported economic statistics

about trade prices: In CBO’s view, the ocial U.S. export price

indexes are lower than they would have been otherwise, and

import price indexes are higher. Those inaccuracies distort

overall U.S. price indexes that use trade prices as an input,

such as the GDP deﬂator.

By distorting economic statistics about trade prices, the stra-

tegic setting of transfer prices also aects the national income

and product accounts. The strategy leads nominal exports to

be understated and nominal imports to be overstated, thereby

reducing ocial measures of net exports and nominal GDP.

Strategically setting transfer prices alters the composition of

net international lending. But like transfers of IP, the strategy

has no eect on the total amount of net international lend-

ing, because each dollar that the strategy removes from net

exports is oset by a dollar of foreign proﬁt added to net inter-

national income. And because transfer prices do not aect total

national income, gross national product (the sum of domestic

income and net international income) is likewise unchanged.

4. MNCs are required to set transfer prices similar to the prices that would

be paid for goods and services in market-based transactions. However, for

some traded goods and services, it is dicult to ﬁnd comparable market

prices. For those transactions, MNCs have more leeway to strategically set

transfer prices to minimize tax liability.

Box B-3. Continued

The Eects of Proﬁt Shifting on Economic Statistics

And nonfarm employment is projected to be, on aver-

age, about 0.6percent higher over the 11-year period,

representing about 0.9 million jobs (see Table B-2 on

page 115).*

Ination. CBO expects the 2017tax act to have a

positive but small eect on consumer price ination

over the next few years. at expectation results from

CBO’s estimates that the act will only slightly widen the

gap between the actual and natural rates of unemploy-

ment and that the link between general price ination

and labor market conditions has been weak in recent

years. In addition, the act is expected to slow growth in

the prices of imported goods, slightly dampening the

inationary pressure from labor markets, particularly in

the near term. Finally, expectations of ination, which

have been low and relatively stable since the late 1990s,

are expected to remain close to the Federal Reserve’s

long-run goal in the coming years, as consumers and

businesses expect the central bank to successfully adjust

monetary policy to prevent ination from deviating

excessively from its target.

As a result, core PCE ination—that is, ination for

personal consumption expenditures, excluding prices for

25. For more information, see “Key Methods at CBO Used

to Estimate the Macroeconomic Eects of the 2017 Tax Act”

(supplemental material for e Budget and Economic Outlook:

2018 to 2028, April 2018), https://go.usa.gov/xQcZD.

[*Values corrected on April 17, 2018]

126 The BudgeT and economic ouTlook: 2018 To 2028 APRIL 2018

food and energy—is expected to be very slightly higher

each year between 2018 and 2025. e total PCE price

index is expected to rise slightly more quickly than that,

as is the consumer price index; both are projected to be

higher by 0.1percent through 2023, on average, than

they would have been in the absence of the act and to be

higher by 0.2percent in 2028.

Eects on Interest Rates

In response to the projected widening of the output gap

and the greater inationary pressure, CBO expects the

Federal Reserve to raise short-term interest rates more

rapidly over the next few years than it would have if the

2017tax act had not been enacted. As a result, the fed-

eral funds rate (the interest rate that nancial institutions

charge each other for overnight loans of their monetary

reserves) is projected to be 0.5percentage points higher

in 2022 than it otherwise would have been. e faster

increase in interest rates is expected, in turn, to restrain

the boost in output by dampening consumption and

investment spending, thereby limiting the increase in

demand for labor and keeping ination close to the

central bank’s long-term goal. CBO’s projections include

a slight and temporary reduction in short-term interest

rates by the Federal Reserve in response to the end of

most of the act’s individual income tax provisions after

2025, but there is no net eect on short-term rates by

the end of the 11-year period.

e eects on long-term interest rates follow a similar

pattern. However, because long-term rates are partly

determined by the average of expected short-term rates,

the eect on long-term rates is larger initially but more

muted overall.

CBO’s projections of interest rates over the 11-year

period are also based on the agency’s projections of a

number of factors that aect the interest rates of U.S.

Treasury securities over the longer run. On net, those

factors are projected to result in rates of longer-term

Treasury notes that are somewhat higher as a result of the

tax act, even as rates of shorter-term Treasury securities

are roughly unaected. In CBO’s projections, factors

that increase the interest rates of Treasury securities over

the period include the increase in federal borrowing and

the increase in the after-tax rate of return on capital.

Additional factors that increase the rates of longer-term

Treasury securities include the reduction in companies’

holdings of such securities following deemed repatria-

tion of foreign holdings and an increase in the premium

incorporated in the rates of such securities. e tax act

increases that premium in CBO’s projections because

with greater upward pressure on ination, longer-term

Treasury securities become less valuable as a hedge

against unexpectedly low ination. e main factor that

decreases the interest rates of Treasury securities over the

period is the increase in net foreign investment.

Eects on Income

e economic eects of the tax act include not just

greater GDP but also higher overall income. Domestic

income that derives from the production of goods and

services—for labor, employees’ compensation and their

wages and salaries; for businesses, corporate prots and

proprietors’ income—is projected to rise with GDP.

Flows of net international income also change, reecting

the tax act’s eects. And businesses see changes in income

in addition to those associated with production, which

will aect taxable business income.

Employees’ Compensation and Wages and Salaries.

Employees’ total compensation in the economy behaves

in a pattern similar to that projected for total GDP. Over

the 2018–2028period, the act is projected to increase

such compensation by an annual average of 0.9percent;

the peak eect is 1.0 percent in 2023. Average total

wages and salaries follow a similar pattern—gaining

0.9percent, on average, and peaking at an increase of

about 1.1percent in 2023.

Corporate Prots and Proprietors’ Income. In CBO’s

projections, domestic corporate prots increase over the

11-year period, becoming 7.1percent larger in 2028

than they would have been without the 2017tax act.

e increase occurs partly because of greater total GDP

and partly because of lower net interest payments by

corporations. at second eect happens for two reasons.

First, corporations are expected to reduce their debt and

interest payments in response to the act’s less favorable

treatment of interest costs. Second, corporations are esti-

mated to have held debt in the United States to nance

domestic investment while they had substantial holdings

of foreign prots. As those prots are repatriated, the

corporations are expected to reduce their debt and inter-

est payments (see Box B-1 on page 109).

26. For more information, see “Key Methods at CBO Used

to Estimate the Macroeconomic Eects of the 2017 Tax Act”

(supplemental material for e Budget and Economic Outlook:

2018 to 2028, April 2018), https://go.usa.gov/xQcZD.

127appendiX B: The eFFecTS oF The 2017TaX acT on cBo’S economic and BudgeT proJecTionS The BudgeT and economic ouTlook: 2018 To 2028

In addition, the change in the deductibility of net oper-

ating losses alters taxable corporate income. e act lim-

its the deductibility of those losses, so corporate income

rises. But they may be deducted from future income, so

the act largely alters when taxable corporate income will

be reported rather than permanently increasing it.

In CBO’s projections, nonfarm proprietors’ income rises

by 1.2percent over the 2018–2022period before falling

back to a 0.3percent gain by 2028, roughly following

the pattern projected for overall economic activity. Over

the 2018–2028period, the increase averages 0.9percent.

Prot Shifting and Foreign Income. e act includes

changes to the treatment of international income that

will aect how multinational corporations shift their

prots among aliates in order to lower their tax liabil-

ities. ree of the most widely used prot-shifting strat-

egies are locating debt in aliates in countries with high

corporate income tax rates, transferring intellectual prop-

erty, and strategically setting transfer prices (the prices

that aliates charge each other across national boundar-

ies; see Box B-3 on page 124). Such prot shifting dis-

torts the national income and product accounts—ocial

U.S. accounts that track the amount and composition

of GDP, the prices of its components, and the way in

which the costs of production are distributed as income.

Prot shifting also lowers taxable corporate income in

the United States—by roughly $300billion each year,

recent estimates from the economic literature suggest.

CBO attributes most of that amount to decisions about

the location of debt and transfers of IP.

27. at estimate was informed by CBO’s calculations and by

Fatih Guvenen and others, Oshore Prot Shifting and Domestic

Productivity Measurement, Working Paper 23324 (National

Bureau of Economic Research, April 2017), www.nber.org/

papers/w23324; Kimberly A. Clausing, “e Eect of Prot

Shifting on the Corporate Tax Base in the United States and

Beyond,” National Tax Journal, vol. 69, no. 4 (December 2016),

pp. 905–934, http://dx.doi.org/10.17310/ntj.2016.4.09;

Kimberly A. Clausing, e Eect of Prot Shifting on the Corporate

Tax Base in the United States and Beyond (available at SSRN,

November 2015, updated June 2016), pp. 905–934, http://

dx.doi.org/10.2139/ssrn.2685442; and Gabriel Zucman,

“Taxing Across Borders: Tracking Personal Wealth and Corporate

Prots,” Journal of Economic Perspectives, vol. 28, no. 4 (Fall

2014), pp. 121–148, http://dx.doi.org/10.1257/jep.28.4.121.

For a discussion of prot shifting and taxable income, see

Congressional Budget Oce, An Analysis of Corporate Inversions

(September 2017), www.cbo.gov/publication/53093.

In CBO’s projections, the provisions of the tax act reduce

prot shifting and the resulting statistical distortions,

on net. at change in the reported location of prots is

expected to result in an increase in taxable income even

though there is no direct increase in measured income

from domestic inputs of labor or capital. All told, the

reduction in prot shifting raises income reported in

the United States by roughly $65billion each year,

on average, in CBO’s projections over the 11-year

period. Changes in the location of debt and transfers

of IP account for most of that reduction in total prot

shifting.

Eects on Gross National Product. e 2017tax act is

expected to aect GDP and GNP dierently. It raises

the projected level of real GDP by an annual average

of 0.7percent over the 11-year period, an increase of

about $710per person (in 2018dollars). Real GNP, by

contrast, increases by 0.4percent, on average, or about

$470per person.

e act is expected to increase GNP

less than it increases GDP because it shrinks U.S. net

international income (see Table B-2 on page 115).

ere are two reasons for that decline in net income

ows to the United States. First, the increase in foreign

investment in the United States that is associated with

greater private investment and increased government

borrowing generates a fall in net international lending,

which is national saving minus domestic investment.

In CBO’s projections, the act decreases net international

lending over the next 11years by an average of 0.4per-

cent of GDP (see Figure B-5). e additional income

generated by the foreign investment in the United States

accrues to foreign investors.

e second reason is that the act alters the rates of return

earned on international assets. As the after-tax prot-

ability of U.S. investments rises because of the taxact,

foreign investors earn a higher return on their U.S.

assets. In addition, the reported rate of return that U.S.

investments earn abroad will decline after 2023 as the act

28. e peak eects for the per-person amounts occur in 2024,

at $900for real GDP per person and $640for real GNP per

person; by 2028 the amounts are $550for real GDP per person

and $250for real GNP per person.

29. In the national income and product accounts, net international

lending is called “net lending to the rest of the world.” Over most

of the past 40years, it has been negative, indicating that the

United States is a net borrower. CBO projects that net lending

will remain negative from 2018 through 2028.

128 The BudgeT and economic ouTlook: 2018 To 2028 APRIL 2018

discourages U.S. companies from shifting their taxable

income from the United States to aliates in foreign

countries. By altering the relative rates of return on inter-

national assets through those changes, the act reduces net

international income and shrinks the dierence between

GDP and GNP.

How the Act Aects the Budget Outlook

e 2017tax act had signicant eects on CBO’s bud-

getary projections for the 2018–2028period. e agency

took two steps to incorporate those eects into the

projections. First, CBO estimated the act’s direct eects,

which are the eects on the budget that do not take

into account any changes to the aggregate economy. For

example, this step incorporated the ways in which the

act’s reduction in tax rates will diminish federal revenues

through its eects on taxpayers’ behavior. Second, CBO

considered macroeconomic feedback—that is, the ways

in which the act will aect the budget by changing the

overall economy (such as by increasing wages, prots,

and interest rates). Incorporating both kinds of eects

boosts the projected primary decit by a cumulative

$1.272trillion over the course of the 11-year period.

After debt service too is incorporated, the projected de-

cit is higher by $1.854trillion (see Table B-3).

Before incorporating macroeconomic feedback, CBO

estimates that the tax act would increase the primary

decit by a cumulative $1.843trillion over the 11-year

period—increasing it through 2026 and decreasing

it thereafter.

ose decit increases would increase

debt-service costs in every year and by growing amounts

that total $471billion over the period.

ose increases would be partially oset by macroeco-

nomic feedback. In CBO’s projections, macroeconomic

feedback reduces the primary decit by a cumulative

$571billion over the 2018–2028period. at reduction

mainly results from the act’s boost to taxable income,

which increases revenues. e eects on the primary

decit, like those on taxable income, are largest in the

early years, peaking in 2019 and then getting smaller.

Macroeconomic feedback also raises debt-service costs

through two partly osetting eects: e reduction in

the primary decit lowers federal borrowing and thus

debt-service costs, but the act also leads to higher interest

rates and thus increases the cost of federal borrowing.

30. ose direct eects on the primary decit primarily reect the

cost estimate produced by the sta of the Joint Committee on

Taxation. See Joint Committee on Taxation, Estimated Budget

Eects of the Conference Agreement for H.R. 1, the “Tax Cuts And

Jobs Act,” JCX-67-17 (December 18, 2017), https://go.usa.gov/

xQczr (PDF, 37 KB). However, in contrast to the cost estimate,

the estimates reported in this appendix extend through 2028 and

include debt-service costs. e direct eects shown in Table B-3

also reect a number of technical revisions. e sources of those

revisions include information about the implementation of the

tax act learned in recent months.

Figure B-5 .

Eects of the 2017 Tax Act on Net Foreign Transactions

Percentage of Gross Domestic Product

Growth in the federal deﬁcit and

in investment increase borrowing

from foreigners, which reduces net

international income.

Page 1 of 1

Net International

Lending

Net International

Income

-0

0.2

2017 2019 2021 2023 2025 2027

Source: Congressional Budget Oce.

Net international income is the dierence between the income earned by U.S. residents from foreign sources and the income earned by foreign

individuals from U.S. sources. Net international lending is a measure that summarizes a country’s transactions with the rest of the world; it consists of

net exports, net international income, and net transfers.

129appendiX B: The eFFecTS oF The 2017TaX acT on cBo’S economic and BudgeT proJecTionS The BudgeT and economic ouTlook: 2018 To 2028

On net, macroeconomic feedback from the act raises

projected debt-service costs by $110billion over the next

11years.

Uncertainty Surrounding CBO’s Estimates

CBO’s estimates of the economic and budgetary eects

of the 2017tax act are subject to signicant uncertainty.

e agency is particularly uncertain about how the act

will be implemented; what policies state governments

and foreign countries might change in response to the

act; what expectations people have about future scal

policy; how businesses will rearrange their nances in the

face of the act; how households, businesses, and foreign

investors will respond to changes in incentives to work,

save, and invest in the United States; and how changes in

economic activity will aect labor and capital income.

Implementation

How the Treasury ultimately implements the tax act will

partly determine how businesses and households respond

to the various provisions. For example, CBO’s projec-

tions of the new deduction for owners of pass-through

businesses incorporate the expectation that the Treasury

will be able to enforce the limits that the act places on

the types of income that are eligible for the deduction.

States’ and Foreign Countries’ Responses

If state governments and foreign countries change their

own scal policies in unanticipated ways in response to

the tax act, those changes will have implications for the

act’s economic and budgetary eects. For example, many

state governments could choose not to incorporate some

of the act’s provisions—such as those involving personal

deductions and bonus depreciation—in their own tax

systems. at step would signicantly aect how house-

holds and rms chose to adapt to the changes. Foreign

governments might reduce their corporate tax rates or

adjust their tax rules in unanticipated ways in response

to the changes in U.S. tax law. In particular, if foreign

governments signicantly lowered their tax rates on cor-

porate income, that would dampen net inows of foreign

capital. In addition, foreign governments are expected

to challenge several of the new tax rules with the World

Trade Organization. If those challenges are broadly suc-

cessful, the United States could be subject to retaliatory

taris unless the tax provisions were changed.

Table B-3 .

Contributions of the 2017 Tax Act to CBO’s Baseline Budget Projections

Billions of Dollars

Total

2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028

2018–

2022

2018–

2028

Eects Without Macroeconomic Feedback

Eects on the Primary Deﬁcit

194 281 307 304 263 218 183 164 36 -60 -46 1,349 1,843

Eects on Debt-Service Costs 3 8 17 29 39 48 55 63 68 70 71 97 471

Eects on the Deﬁcit

197 289 325 333 302 266 238 227 104 10 25 1,445 2,314

Eects of Macroeconomic Feedback

Eects on the Primary Deﬁcit

-33 -67 -65 -58 -55 -49 -47 -49 -48 -50 -51 -278 -571

Eects on Debt-Service Costs 0 5 12 18 23 27 23 13 3 -4 -11 59 110

Eects on the Deﬁcit

-33 -61 -53 -41 -31 -22 -24 -36 -44 -54 -62 -219 -461

Total Contributions to Baseline Projections

Eects on the Primary Deﬁcit

160 214 243 246 208 169 136 115 -12 -110 -97 1,071 1,272

Eects on Debt-Service Costs 3 14 29 47 63 74 78 76 71 66 60 156 582

Eects on the Deﬁcit

164 228 272 292 271 243 214 191 59 -43 -37 1,226 1,854

Source: Congressional Budget Oce.

a. Macroeconomic feedback refers to the ways in which the act would aect the budget by changing the economy.

b. The primary deﬁcit is the deﬁcit excluding debt-service costs.

c. Positive numbers indicate an increase in the deﬁcit; negative numbers indicate a decrease in the deﬁcit.

130 The BudgeT and economic ouTlook: 2018 To 2028 APRIL 2018

People’s Expectations

In CBO’s projections, 20percent of households and

businesses expect scal policy to change over the 2018–

2028period as the tax act species; others are surprised

by those changes. Such expectations can have important

eects on how households and businesses respond to the

act. For example, if more people expect the reduction in

individual income tax rates to be temporary, as the act

species, more may shift their supply of labor from later

years into the years before rates are scheduled to go up.

If that happened, the timing of CBO’s projections would

change, but the average eect over the 11-year period

would not be strongly aected.

Prot Shiing by Multinational Corporations

e eect of the tax act’s international provisions on

prot shifting by multinational corporations is particu-

larly uncertain. One source of uncertainty is the pro-

visions’ complexity, which makes it dicult to predict

how and when corporations might respond to them.

CBO is also uncertain about how foreign governments

might change their tax rules in response to the act. For

instance, those governments might lower their own

corporate income tax rates to better compete for interna-

tional investment; that change would dampen the act’s

expected eect on prot shifting. And CBO is uncertain

about whether the provisions will be deemed compliant

with international rules.

Decisions to Work, Save, and Invest

Many economic eects of the new legislation stem from

its eects on individuals’ decisions to work and save

and on businesses’ decisions to invest. CBO’s estimates

of those eects reect the agency’s assessment of how

changes in individual and corporate tax rates aect the

supply of labor and the user cost of capital, as well as

its assessment of how changes in individuals’ disposable

income and wealth aect consumer spending. CBO

tries to produce assessments that lie in the middle of the

distribution of possible outcomes. But if fewer people

than CBO expects respond to lower marginal tax rates by

participating in the labor force, for example, the boost

in potential GDP will likewise be smaller than CBO pro-

jects. Another example involves the expected response of

international investors to the reduction in U.S. corporate

tax rates. If they increase investment more than CBO

expects, capital stock will increase more and the eects

on actual and potential output will be larger.

Some eects may dier from CBO’s assessments because

those eects may depend on economic conditions in a

way that the agency has not incorporated. For example,

CBO has not accounted for the extent to which the act’s

limits on the deductibility of net operating losses could

discourage investment more during periods of economic

weakness than in periods of economic strength. (e

eect of those limits is uncertain for other reasons as

well. For example, they could dampen the positive incen-

tives to invest that result from other provisions in the tax

act, a possibility that CBO has not accounted for in its

projections.)

Changes in Economic Activity

CBO projects that the tax act will increase labor income

and capital income, boosting demand for goods and

services over the next several years. But demand may

respond more or less to those changes in income than

CBO estimates. Moreover, the changes in economic

activity resulting from the act may have smaller or larger

eects on businesses than CBO estimates. For example,

if businesses increase investment more than expected in

response to increases in economic activity, labor produc-

tivity and wages will rise faster than they do in CBO’s

projections.