The alternative minimum tax and effective marginal tax rates

MITLIBRARIES DUPL

Digitized

by

the

Internet

Archive

in

2011

with

funding

from

Boston

Library

Consortium

Member

Libraries

11 1-15

Massachusetts

Institute

of

Technology

Department

of

Economics

Working

Paper

Series

THE

ALTERNATIVE

MINIMUM

TAX

AND

EFFECTIVE

MARGINAL

TAX

RATES

Daniel

R.

Feenberg

James

M.

Poterba

Working

Paper

03-37

October

2003

Room

E52-251

50

Memorial

Drive

Cambridge,

MA

021

42

This

paper

can

be

downloaded

without

charge from

the Social

Science

Research Network Paper

Collection at

http://ssrn.com/abstract=47060

1

MASSACHUSETTS

INSTITUTE

OF

TECHNOLOGY

The

Alternative

Minimum

Tax

and

Effective

Marginal

Tax

Rates

Daniel R. Feenberg

NBER

James

M.

Poterba

MIT

and

NBER

October2003

ABSTRACT

This paperexaminesthe impact ofthe Alternative

Minimum

Tax on

the weightedaverage

marginal tax rates thatapplytovarious

components

oftaxable income.It also considers the

impact ofseveral

AMT

reformproposalson the

number

of

AMT

taxpayers, the totalrevenue

collected fromthe

AMT,

and the weighted averagemarginal tax rates thatapplytowages,capital

income, anddeductionssuchasstate and localtaxes and charitable gifts.

The

paperuses the

NBER

TAXSIM

model

toproject federalpersonal

income

tax liabilitiesaswell as

AMT

liabilities

between 2003 and2013.

The

AMT

hasonly a

modest

impacton theaveragemarginal tax rates

for

most

sources of

income

because

some

AMT

taxpayersfacehighermarginal taxrates,and

otherslowertaxrates, as aresultofthetax.

The

projections

show

that

modest

increases in the

AMT

exclusion levelhavesubstantial effects on the

number

of

AMT

taxpayers, andthatindexing

the

AMT

parameters

would

reduce the

number

of

AMT

payers in

2010 by more

thansixty

percent. These changes

would

alsoreduce the

AMT's

impacton averagemarginal taxrates.

We

areextremelygratefulto InnaShapiro for assistancewiththe

TAXSIM

program,to

Rosanne

Altshulerforhelpful

comments,

andtothe Smith RichardsonFoundation, theNational

1

The

Alternative

Minimum

Tax

(AMT)

isaprovisionoftheU.S.

income

taxcodethat

currently affects amodest

number

oftaxpayers. Itwill

become

anincreasingly important

component

ofthe federal tax system inthe

coming

decade.

The

minimum

taxbeganin 1970asa

ten percent,andlatera fifteen percent, taxon preferences inexcess of $30,000. Preferences

included accelerated depreciation, oil depletion, andthe capitalgains deduction.

A

deduction of

onehalfofregulartax paid,andlaterofalltax,

was

allowedagainstpreferences. Netoperating

losses andretirement

income

receivedspecialtreatment.

The

minimum

taxexpired in 1981. Beforeits expiration, however,in 1979,a

new

"Alternative

Minimum

Tax" was

establishedwithabase thatincludedall the

components

of

AdjustedGross

Income (AGI)

andthe capital gainsdeductionin additiontopreferences.

The

new

AMT

allowedthe full

amount

of regularincome tax asa credit. Aside from changes in the

treatmentofnet operating lossesand slightchanges intheexclusion amount, thisisthe

AMT

that

iscurrentlyin force,andthat isscheduledtoremain inforce indefinitely.

When

apreferential tax

rate forlong-termcapital gains

was

establishedin 1991, care

was

takentoavoid treating capital

gains asapreference.

A

number

of research studiespublishedsince the late 1990s have identified the

AMT

asa

growing factor inthe

income

tax structure.

Harvey

and Tempalski (1997)

were

among

the firstto

point outthatbecause theexclusionlevel for the

AMT

was

not indexed forinflation, while

most

other keyparametersinthe tax code

were

indexed, the

AMT

would

applyto agrowing

number

of

taxpayers. Rebelein andTempalski(2000),theU.S. Congress Joint

Economic Committee

(2001),

Tempalski (2002), and

Burman,

Gale,andRohaly (2002) provide furtheranalysis ofthegrowing

importanceofthe

AMT.

The

AMT

became

more

importantafterthepassageofthe

Economic

Growth

and

Taxpayer

ReliefReconciliationAct of 2001

(EGTRRA),

which

reduced ordinary

income

tax

two

examples ofstudies that

show

thatbecause

EGTRA

lowered

income

tax liabilities for

many

taxpayers, itwill raisethe

number

of

AMT

payers.

The

Jobs and

Growth

TaxpayerRelief

ReconciliationActof

2003

reinforcesthis effect. For

many

upper middle

income

households

who

willface the

AMT

as a resultofrecenttaxreforms,the

AMT

"takesback" a substantial shareof

the potential tax reliefassociatedwiththesereforms.

Burman,

Gale and Rohaly (2003a,2003b)

suggestthatin2010, 33.8 percentofthe

EGTRRA

tax cuts willbe"recaptured" inthe

form

of

higher

AMT

liabilities. Thispercentageexceeds 65 percent fortaxpayerswith

AGI

between

$100,000 and $500,000.

Sullivan (2002) offersa careful review ofthe issues associated withthe potentialrepealof

the

AMT.

Proponents ofrepealpointtothe potentialcomplianceburdenofthe tax,which can

effectivelyrequiretaxpayersto prepare

two

taxreturnsandto

compare

the results,andtothe

unanticipated impacton taxpayerliabilities.

The

potentialrevenuecostof

AMT

repealisa

substantial impedimenttosystematic reform.

Inthispaper,

we

use the

NBER's

TAXSIM

program

to evaluate the impactofthe

AMT

on

incentivesto

work

andto save.

We

present

some

estimates ofthe growing

number

of

taxpayers

who

will face the

AMT,

but theyare notthe central focusofthe study.

We

emphasize

projectionsforcalendaryear 2010, sinceundercurrentlaw,this istheyear

when

the

AMT

will

affectthe largest

number

oftaxpayers.

A

number

of

income

tax provisionsarecurrently

scheduledtoexpire in2010,so the taxenvironmentin2011 will besubstantially different

from

thatin 2010. Although our projectionsrelyon strongassumptionstoproject the future

number

of

tax returns andthe leveland composition of

income

on thesereturns, theyoffer

some

insight

on

the likely futurecourseofthe

AMT

relativetothe ordinary

income

tax.

The

paperisdividedinto six sections.

The

firstdescribes thebasic structureofthe

AMT

and comparesalternative

minimum

taxable

income

with taxable

income

asdefinedunderthe

anditpresentsour projections ofthe

number

of

AMT

taxpayersinaggregate, andin

income

sub-categories, for futureyears.

Our

results fromthe

TAXSIM

analysis arebroadlyconsistentwith

those fromotherrecent studiesthatproject future

AMT

liabilities. Sectionthree describesthe

probabilitythattaxpayers invarious

income

categories will face the

AMT.

It also studies

how

various reformproposals

would

affectthese probabilities.

The

fourth section investigates the

impact ofthe

AMT

onthe weightedaverage marginaltax rates thatapplyto wages, interest

income,dividend income, andseveral

income

taxdeductions. Itconsidersmarginal tax rates

associated withthe federal

income

tax, aswell asthetotal effective tax ratethatcombines the

federal

income

tax withthe potentially-deductible state

income

tax. Section fiveexploresthe

effectof various reformproposalson the weighted averagemarginal tax rates

on income

and

deductionflows. Thereisa briefconclusion.

1.

The

Structure ofthe Alternative

Minimum

Tax

Although the

AMT

issometimes describedasa

complex

andmysterioustax, for

many

taxpayers

who

face the

AMT,

the

AMT

calculationis straightforward. Fortaxpayers

who

do not

itemize,the "tentative

AMT"

is 26percentof

AG

I inexcessofan exclusionamount. For

taxpayers withalternative

minimum

income

above $175,000,the

AMT

tax rate is 28percent.

The

AMT

is paid asanactual tax liability onlytotheextent that itexceeds thetaxpayer's regulartax

liability. For2010,the year

when

the

AMT

isprojectedtogeneratethe largest revenue flow, the

exclusionequals$45,000 forjoint filersand $33,750forsingle individuals. Fortaxpayerswith

incomesabove $150,000 onjoint returnsand$112,000 on single returns, theexclusionisphased

outat therateof 25 centsforeachdollarof

AGI

abovethe threshold. This implies thatthe

28%

AMT

rate appliestojointfilerswith

AGI

above $206,000,and forsinglefilerswith

income

above

minimum

income

will equal 175,000. For

example

forjointfilers,thisrequires

AGI

-

(45000

-.25*(AGI - _150000))

=

175000.

Severalfeatures ofthe

AMT

arescheduledtochangebetween

2003

and

2010

under

currentlegislation. For example,after2005, thezero bracket

amount

is

no

longeraddedto the

exclusion. For

2003

and2004, theexclusionis $58,000forjointfilers and $40,250 forsingle

filers. It declinesin lateryears.

Foritemizers,the

AMT

calculationis

more

complicated. Allitemized deductions except

those forstateand localtaxes, medical expenses inexcessof

2.5%

of

AGI,

and miscellaneous

deductions are available as deductions against alternative

minimum

income.

The

taxpayer

must

identify eligibledeductionsand subtract

them

from alternative

minimum

income. Itemized

deductions underthe

AMT

are notsubjectto thephase-outthatispresentintheregularpersonal

income

tax. In rarecasesthis

may

invalidatethe usualrule thattaxpayers should itemize if

itemized deductionsexceedthe standard deduction.Foravery few, thedecreased

AMT

from

itemizationwillcompensatefor a largertaxable income.

For

most

taxpayerswith longtermcapital gains, the

AMT

iscalculated

on

non-gain

income

plus

20%

oflong-term gains. Forlow

income

taxpayerswithgains, the

10%

capitalgains

rateisapplied togainsthat

would

notpushtaxable income abovethe

15%

bracketboundary. All

tax credits exceptthe foreigntax creditcan be creditedagainst

AMT

liability,ascanthe

taxpayer's regulartax liability. Sincetaxyear 2000,tax creditsdo notaffectthe total liability

calculation.Priorto that year,the taxformsdistinguishedbetween

AMT

liability,recorded on

form

6251, and credits

whose

value, iftaken,

would

befullyoffsetby the

AMT.

These

were

referredtoas "lostcredits"andthe credit formsincluded lines toavoid takingsuch credits.

While

the

AMT

may

requiretaxpayers to

compute

theirtax liabilityunder

two

regimes,

none

ofthe calculationsrequireanyspecial recordkeeping

beyond

whatis requiredby the regular

true that

some

taxpayers withnetoperating losses,or accelerated depreciation orinothersimilar

esoteric situations are subjecttoadditionalrecord keeping,and

may

besubjecttoadditional

AMT

by

2010, but there are few suchtaxpayers. For

most

taxpayers

who

will face

AMT

liability,the

AMT

calculation isnotparticularly burdensome.

Figure 1 provides informationon the pattern of

AMT

andordinary

income

tax liabilityfor

ataxpayerfilinga joint return with

two

dependentsandclaimingthestandard deduction.

Tax

liabilitiesunderbothtaxschedulesare

shown

atdifferent levels of adjusted gross income.

The

figuregraphs

AMT

andordinary

income

tax liabilityforboth2003 and2010, withthe taxpayer's

real

AGI

held constantin2003 dollars.

The

graphs

show

that

AMT

liabilityexceeds

income

tax

liability fora substantialrangeof incomes, beginningatroughly $50,000, and endingatroughly

$375,000,in2010.

The

figure also

shows

thatthe rangeofincomesover

which

AMT

exceedsthe

income

tax iswiderin

2010

than in2003.

The

fourlinesdemonstrate

how

similar the shapes of

the fourtaxschedules are, sothatasmall shiftleftward in thereal

AMT

schedule,such asthe one

thattakesplace

between

2003 and2010,shifts

much

ofthe

income

distribution ontothe

AMT

schedule.

2. ProjectingFuture

AMT

Liabilities:

The

TAXSIM

Model

The

NBER

TAXSIM

model

isa computer

program

thatcalculates federal incometaxand

payroll tax liabilitiesfora representative sample of U.S. families. Itanalyzes datafromthe

Statistics of

Income

(SOI)Public

Use

File, a stratified

random

sampleof U.S. taxpayersthat

oversampleshigh-incometax returns. Feenbergand Coutts (1993) describethe personal

income

tax sectionsofthe

TAXSIM

model

in

some

detail.

The

TAXSIM

algorithm includesdetailed computer code basedonthe personal

income

tax systemthat, basedoncurrentlegislation, willbe inforceinfuture years.

To

construct

individual

income

tax return file,the 1999

SOI

datafile,to2010. This "aging" processrequires

forecasts ofthe growthrateofthe aggregate

amount

ofvarious

income components

anddeduction

flows.

We

assume

annualgrowthratesof1.2percent forpopulation,2.2 percent for realincomes

anddeductions, and2.5 percent forthe price level forall

income

anddeductionitems. These

growthratesare

drawn

from

CBO

projections.

Long

term gainsare normalizedto5.6 percentof

AGI, which was

theiraverage valueovertheyears 1981-1999.

When

thegrowthrates describedabove are appliedtothe 1999cross-section data

on

tax

returns,the resulting

income

taxrevenue forecasts for2004-2013 aresubstantiallyhigher than

those

made

by

the Congressional

Budget

Office.

A

one-timeten percent reductioninallnominal

magnitudes,appliedfor examplein 1999, brings our aggregaterevenueestimates into

much

closer

agreementwiththe

CBO.

We

thereforereduce allnominal 1999 magnitudes

by

tenpercent

before

we

make

ourfutureprojections.

TAXSIM

calculates federal aswell asstate marginal

income

tax rates.

AMT

payers

cannot deductstate

income

taxes

from

theirtaxableincome, sotheireffective state

income

tax

rate is higher thanthe comparable ratefacing those

who

payordinary federal

income

taxand

claim an itemized deduction for statetaxes.

To

recognizetheimpactofthe effectivestatetax rate

on

the total taxburden

on

wages, interest, dividends, andother

income

components,

we

therefore

calculate federalmarginaltax rates underthe

AMT

aswellasthe

combined

tax ratethatequals

the federalmarginaltaxrateplusthe net-of-federal-deductibilitystate

income

taxrate.

Table 1 presentsinformation

on

our projections regardingthe

AMT

undercurrent law.

The

table

shows

the total

number

oftaxpayers

who we

project will face the

AMT,

thetotal

amount

of

AMT

revenue,andtheratioof revenue

from

the

AMT

torevenue

from

the ordinary

income

tax.

The

table

shows

thatprojected

AMT

revenue peaks in2010,

when

itaccounts for

nearlynine percentoftotal

income

taxrevenue.

We

project that 37milliontaxpayers

-

nearly one

7

have no itemized deductions, no capital gains,no phase-out ofthe

AMT

exclusion, and noother

preferences. Thisgroup of taxpayerswill face relativelysimple

AMT

calculations.

Table 1

shows

thatthe

number

of

AMT

filers andthe revenue collectedfromthe

AMT

declineafter2010,

when

various provisionsof

EGTRRA

expire. Ifthis legislativechange occurs

in2011, the

number

of

AMT

filers will fallby

more

thanonethird,and

AMT

revenues will

decline

by

more

than half.

The

projections inTable 1 aresimilartothoseinseveral other studiesofthe

AMT.

For

example,

Burman,

Gale, and Rohaly (2003b) projectthat there willbe33.1 million

AMT

filers in

2010.

They

note that inthe absenceof

EGTRRA,

they

would

have projected 14.3 million

AMT

filers. Theirprojectionsuggests

AMT

revenue of

$124

billionin2010, only $1 billion lessthan

ourprojection. Kiefer,etal. (2002) projectthatthere willbe35.1 million

.AMT

filersin

2010

and

that the

AMT

willyield $133billioninrevenuethatyear. Their projections includethe effectof

EGTRRA,

but theydonotincorporate any ofthe2003 tax changes.

The

prospective growth of

AMT

liabilitiesandthe

number

of

AMT

taxpayers has

generated a

number

of proposals for legislativereformsthat

would

slowthe growthofthe

AMT.

We

usethe

TAXSIM

model

toevaluatethe impactofsixpotentialreforms

on

the

number

of

AMT

taxpayersandtherevenue generatedby the

AMT.

Table2 presents these findings.

The

first

row

reportsbaselinecalculations for2010, correspondingtothe informationinTable 1.

The

nextsix

rows

considerreform proposals.

The

firstproposal

we

consider

would

allow personal exemptionstobe subtractedfrom

otherincome indefiningalternative

minimum

income.

Such

exemptions arepermittedunderthe

ordinaryincometax,thereby reducing ordinary taxable

income

relativetoalternative

minimum

taxable income.

The

second

row

of Table2

shows

thatthisreform

would

sharplyreduce the

number

of

AMT

taxpayersin2010, from37 millionto 12 million, and it

would

also reduce

8

A

secondpotential reformcalls fortheextensionof

income

taxprovisionsthat expire

between

2005

and

2010

sothatthey

would

remainin forcein2010.

The

two

most

significant

provisionsareone that

would

raisethe

AMT

exclusionin

2010

to $13,000forjointfilers and

$6,500forsingle filers,and one that

would

continuethe preferentialtaxation of dividends.

The

formergreatlyreducesthe

AMT,

andthe latterreduces boththeregular

income

tax andthe

AMT

by

similar amounts.

The

third

row

of Table 2

shows

that extendingall expiring provisions

would

reduce the

number

of

AMT

filers

from

37millionto 14.1 millionin2010,witha decline in total

AMT

revenue from

$126

to $58 billion.

The

third

AMT

reformthat

we

consider involveschanges in the relativetreatment

of

single andmarried taxpayers.

The 2010

jointtaxpayer exclusionof $45,000 underthe

AMT

is

substantiallylessthantwicethe singletaxpayer exclusionof$33,750. In principle, the single

exclusioncould simplybedoubledto $67,500, butthat

would

substantiallyincrease themarriage

bonusforcoupleswithnon-workingspouses.

We

havetherefore

modeled

a lessgenerous option

which

offers the couplethe greaterofthecurrent joint deduction, orthe singlededuction plus the

secondary earner's

wages up

tothe single exclusion. This

means

thatthefull $67,500isoffered

onlyto

two

earner couples, with bothearners

making

more

than $33,750.

Our

analysisofthis

reform

shows

thatitleads to a

more modest

reduction in

AMT

taxpayers thanthe

two

earlier

reformproposals.

The

number

of

AMT

filers in

2010

drops

from

37 millionto25millioninthis

caseand

AMT

revenuedeclines

by

roughly onethird, from

$126

to $87billion.

The

lastthreeproposals

we

consider involvechangestothe

AMT

exclusionlevel.

One

involves indexingtheexclusion, whilethe secondandthird involveraisingtheexclusion

by

$3,000 and$10,000respectively.

The

nominal,un-indexed characterofthe

AMT

exclusionisa

key contributortotheprojected growthof

AMT

revenuesandtaxpayers.

We

project 14.0million

indexation.

AMT

revenuesin thiscase are$48 billion. Indexing theexclusionthereforereduces

AMT

revenueinour projectedpeak revenueyearby oversixtypercent.

The

findings onthe impact ofindexing the

AMT

underscore the importance ofthe

inflationrate indetermining the relativerevenueyield oftheregular

income

taxandthe

AMT.

Figure 1

showed

thatsmallchanges inthe real threshold at

which

taxpayers areaffectedby the

AMT

canhave substantial effectson revenues and onthe

number

of

AMT

filers.

By

comparison,

real growthinthe aggregate

economy

has a

much

more modest

effect on

AMT

revenue, because

realbracketcreepraises the regular

income

tax as well asthe

AMT

base.

The

last

two

reformsthat

we

consider involve specificnominal changes inthe

AMT

exclusion. Raisingthe exclusionby

$3000

in

2010

reducestheprojected

number

of

AMT

taxpayers

by

6.4 million, from 37 millionto31 million. Raisingthe exclusionby $10,000 reduces

the

number

offilers to 17million.

Revenues

drop from

$126

billion in the statusquo,to

$100

billion ifthe exclusion israised

by

$3000, to$65billion ifthe exclusionrisesby $10,000.

The

burden ofthe

AMT

doesnot fall equally acrossthe

income

distribution. Table3

shows

the

AMT

tax burdenas a shareof

AGI

forhouseholds stratified

by

AGI.

The

entries

show

total

AMT

liabilitydividedby total

AGI

forthehouseholds ineachcategory; theyare not

restricted tohouseholdswith

AMT

liability.

The

calculationsapplyto2010, andthey

assume no

changes between

now

andthen in the federal tax law.

The

AMT

burden ishighestforthosewith

AGI

between

$200,000 and$500,000. This groupfaces

AMT

liabilitythataverages

2.7%

of

AGI.

The

burdenis halfasgreat,

1.3%

of

AGI,

fortaxpayers with

AGI

of $50,000 to$75,000. It is

lower for

AMT

taxpayerswith higher

-

greaterthan $500,000,and lower

-

less than $50,000

-AGI.

Figure2 plots information similarto that inTable3.

The

figure

shows

average taxrates,

10

rate fortaxpayerswith

AGI

between

$200,000 and $500,000, andvery littleimpact

on

averagetax

rates forhouseholds

whose 2003

AGI

is

below

approximately$50,000.

3.

The

Probability of Facingthe

AMT:

Current

Tax

Policyand Various

Reforms

The

results inTable3 andFigure2 underscorethe importanceof disaggregating taxpayers

by

income class

when

analyzingthe

AMT.

One

useful

way

to illustrate

how

the

AMT's

impact

varies acrosshouseholds withdifferent characteristics and

income

levelsis toproject the

probability that ataxpayer willbe an

AMT

filer.

Burman,

Gale, andRohaly(2003a, 2003b) and

otherearlierstudies estimatetheAMT-filing probabilities fortaxpayers invarious

income

categories.

Our

analysis explores

how

various reformproposals

would

affect theincome-specific

likelihoodoffacingthe

AMT.

Table4presents resultsbasedon ourprojections ofthe

AMT

in2010.

The

table reports

probabilitiesundertheassumption thatcurrent tax legislationremains in forcethrough

2010

and

underthe variousreformsdescribed inTable 2.

The

results inthe first

column

indicate the

projected

AMT

payment

probabilitiesunderthe status quo.

The

tablepresents results forthree

setsoftaxpayers: the

whole

taxpayer population,the thirtypercentof taxpayers

who

areprojected

toitemize in2010, andthesubsetofalltaxpayers with

two

or

more

dependents.

The

probabilities

offacingthe

AMT

differacrossthesegroups.

The

resultsinthe firstpanelofTable4

show

that thetaxpayers

who

havethe greatest

chance ofenteringthe

AMT

regimein

2010

arethosewith

AGI

between

$75,000 and$500,000.

These

income

thresholdsare specifiedinconstant $2003. At lower incomelevels, the

AMT

exclusion

makes

itunlikelythatataxpayerwill face

AMT

liability.

At

AGI

levelsabove

$500,000, theprogressivityofthe personal

income

tax schedule

makes

itlikely that for

most

taxpayers,personal

income

taxliabilitywillexceed

AMT

liability.

The

resultsunderscorethe

11

facingthe

AMT

is

56.3%

fora $50-75,000

AGI

taxpayer,

77.2%

for ataxpayer with

AGI

between

$75,000 and$100,000, and

more

than

90%

fortaxpayers with

AGI

between $100,000 and

$500,000.

The

results inthesecondpanelof Table4, foritemizers, arebroadlysimilartothosefor

the universeofall taxpayers.

The

distribution ofitemizers across

AGI

categoriesis shifted toward

higherincomes, however,sothat while only

25.7%

ofall taxpayersareprojectedto face the

AMT,

52.5%

ofall itemizers areprojectedto face the

AMT.

The

lastpanelof Table4,

which

presentsresults fortaxpayers with

two

or

more

dependents,

shows

thatthepresence ofdependents

changes theAGI-specificprobability offacingthe

AMT.

In the $25-50,000

AGI

category, for

example,

15.8%

ofall taxpayersareprojected topaythe

AMT,

while

more

than halfofthe

taxpayersin this

AGI

category with

two

or

more

dependents areprojected toface the

AMT.

There isalsoasharp increase intheprobability offacingthe

AMT

atincomes between $50,000

and $75,000.

The

overallprobability offacing the

AMT

isgreaterthan

50%

forthe subsetof

taxpayerswith

two

or

more

dependents.

The

next six

columns

of Table 4presentresults similartothose inthe firstcolumn, but

correspondingtothe various

AMT

reform optionsdescribed above.

The

second

column

considers

theimpact of allowing personal exemptionstobe subtractedfrom alternative

minimum

taxable

income. In thiscase,the overallprobability offacingthe

AMT

drops from

25.7%

to8.4%.

The

effectis strongestfortaxpayers with

AGI

below

$100,000. Forthe$75-100,000

income

category,

forexample,the probabilityoffacingthe

AMT

drops from

77.2%

to 15.8%.

The

effectofthis

reformisalso verypowerfulfortaxpayers

who

havepositiveincome and twoor

more

dependents.

Thisreformreduced theirprobabilityoffacingthe

AMT

in

2010

from

56.1%

to 15.6%.

The

third

column

of Table 4considers theimpactofextendingtheexpiring

EGTRRA

provisionsthrough 2010. Foralltaxpayers, this reformreducestheprobability of payingthe

12

income

AMT

payers. Fortaxpayers with

two

or

more

dependents,the impact ofthisreform is

smallerthanallowing personal exemptionstobe deducted from alternativetaxable income.

The

probabilityoffacing the

AMT

drops from

56.1%

underthe status

quo

to

27.8%

withthe extension

ofexpiringprovisions.

The

next

column

considers the

"AMT

marriagepenalty reform"describedin thelast

section. Thisreform reducesthe percentageoftaxpayers facing the

AMT

from

25.7%

to 17.4%.

The

threepanels ofTable4

show

thatthe impact ofthisreformon theprobabilityof paying

AMT

islessconcentrated atlower

income

levelsthanathigherlevels. Inparticular,the probability that

taxpayersinthe $50-100,000

AGI

range face the

AMT

declines lessforthisreformthan for either

oftheprevious reforms that

we

considered.

The

lastthree

columns

considerchanges inthe

AMT

exclusionlevel.

Comparison

ofthe

findingsinthese

columns

shows

thatindexing generatesthe largestreductionintheprobabilityof

facingthe

AMT

fortaxpayers with

AGI

between

$50,000 and $200,000.

Even

withindexation,

the probabilityof payingthe

AMT

remains

83.7%

fortaxpayerswith

AGI

ofbetween $200,000

and$500,000.

The

probabilityis

92%

forthisgroup

when

the exclusionisraised

by $3000

or

by

$10000.

The

thirdpanelofTable

4

shows

thatvirtuallyallofthetaxpayersinthis

AGI

category

with

two

or

more

dependentsfacethe

AMT.

Indexing the

AMT

exclusion hasa smallereffect

on

thepercentage oftaxpayerswith

two

or

more

dependents

who

paythe

AMT

thanitdoeson the

fraction ofalltaxpayers facing this tax.

Figure 3

shows

the impact ofraising the

AMT

exclusionon

number

of

AMT

taxpayers

and on therevenueyieldofthe

AMT.

The

marginaleffectof an incremental increaseinthe

exclusion declinesas theexclusionrises.

A

$10,000increase intheexclusion in2010, for

example,is predictedtoreduce

AMT

revenue

by

nearly fiftypercent, and ithas a similar

13

the

number

of

AMT

filers

by

28million, from34.8 million to6.7 million, and itreducesthe total

revenue generated

by

the

AMT

from $121.2billionto$41.5 billion.

4.

The

AMT

and Marginal

Tax

Rates

Most

ofthepopular discussionsurroundingthe

AMT

focusesonthe

number

of taxpayers

who may

face thetax, oron theimpactofthe

AMT

on federal taxrevenues.

The

effectofthe

AMT

on incentives forworking, saving,and engaginginvariousactivitiesthatgenerate tax

deductions has received

much

lessattention.

To

explorethe incentive effects ofthe

AMT,

we

compare

the ordinary

income

tax rateandthe

AMT

rate facingtaxpayers

who

areprojected tobe

subject to the

AMT

in 2010. Sinceataxpayer

may

face differentmarginal taxrates ondifferent

income

flows,

we

presenttabulations forvarious

components

oftaxableincome. Foreach

income

component,

we

compute

the fractionofthat

component

that isreceived

by

AMT

taxpayers.

We

then

compute

the percentage

whose

tax raterises, andthepercentage

whose

taxratedeclines, asa

result ofthe

AMT.

We

constructmarginal tax rates

by

calculating theincremental tax that ataxpayer

would

pay ifoneofhis income elementsordeductions

was

onepercentgreaterthanthereportedvalue.

The

tax rateon a specific

income component

foragivenhouseholdisdefinedasthechangeintax

liability dividedby .01 timesthe initial taxreturn entry, i.e.,the changein taxdividedby the

changein

income

or the change inthe deductionamount.

We

calculatethesemarginal tax rates

underthe assumptionthatall currently legislated tax rulesremaininforceunless theyarechanged

by

legislation thathas alreadybeenenacted.

We

therefore

assume

thatphase-outsof

some

ofthe

tax provisions thatwereenacted in2001 and

2003

will take effect asplanned.

4.1 Marginal

Tax

Rate Increasesand Decreases with the

AMT

Table 5presents ourfindings with regardtothe disparity betweenataxpayer'sordinary

14

marginal

income

taxrate. Table 6reports parallel calculations thatconsiderthe impactofthe

AMT

onthe taxpayer's

combined

federaland statetaxrate, netoffederal tax deductibility.

The

first

column

inTable5 considers

wages

and salaries. Nineteenpercentofthe

wages

receivedby

AMT

taxpayersfaces alower marginaltax rateas a resultofthe

AMT,

whilethe remaining 81

percentfaces ahighertax rate.

The most

common

outcome

isa taxrateincreaseof

between

zero

and fivepercentage points, for

example

reflectinga

move

fromthe 33 percent bracketonthe

ordinary

income

tax scheduleto 28percent

on

the

AMT

schedule.

Almost

onetenthofthe

wage

and salary

income

received

by

AMT

payersistaxedata rate 5 to 10 percentagepoints loweras a

resultofthe

AMT,

with aroughly equal

amount

of

wage

income

taxed atarate 5 to 1 percentage

pointshigher.

The

next

two

income

categories that

we

consideraredividends andinterestincome. For

bothofthese income categories, thereisa greaterchance ofa declineinthemarginaltax ratethan

there

was

for

wage

andsalary income.

More

than

32%

ofdividends,and

26%

ofinterestincome,

facesalowermarginaltax rate asaresultofthe

AMT.

A

largeshare ofboth dividendsand

interest

-

35.1 and41.2percent, respectively

-

faces marginaltax rates that arehigher, but

no

more

than fivepercentage pointshigher, as a resultofthe

AMT.

The

remaining

columns

of Table5 present similarcalculations for longtermgains, state

income

tax deductions, andcharitable contributions.

Each

ofthesethree

income

components

or

deductions istreatedina different

way

underthe

AMT.

Forlong-termgains, 12percentofthe

long-term gains reported

by

AMT

taxpayers face the

same

tax ratewiththe

AMT

aswithout it.

Becausestate andlocaltaxes are not deductible incomputing alternative

minimum

income, but

they aredeductible for ordinary

income

tax calculations, virtuallyall taxpayers face

much

higher

marginaltax ratesonthese deductionsunderthe

AMT

than undertheordinary

income

tax. For

such taxpayers,the changeinthemarginaltax rateis large

-

thefullvalueofthe

AMT

rate.

15

charitablecontributionscan be deducted. In partbecausecontributions tendtobe

made

by

taxpayersinupper

income

brackets, for

whom

the

AMT

tends toreducemarginal taxrates, nearly

three quarters of these contributionsaredeductible ata lowermarginal tax rate as aresultofthe

AMT.

Table6

shows

the impact ofthe

AMT

onthe

combined

federal andstatemarginal tax

ratesthat taxpayersface

on income

anddeductions.

The

effective marginalstate

income

tax rate

rises from(l-tfCdcrai)*tSiatetotstatefor an itemizerundertheordinary

income

tax

who

switches to

become

an

AMT

payer.

Comparing

theresults inTables 5 and6

shows

how

substantial thiseffect

canbe. For example, thepercentage of

wage

incomethatexperiencesa to5 percentage point

tax increase as aresult ofthe

AMT

dropsfrom 55 to

44%,

whilethepercentage witha5-10

percentage point increase rises from 10 to

21%.

The

effectofrecognizing state

income

tax rateson the distributionof marginaltax rate

changesforinterest

income

and fordividend income is similarto thatfor

wage

income. For

long-termgains, the shareofgainsreceived

by

AMT

payers with no changein theirmarginal tax rates

declines from

42%

inthe upperpanelof Table 5 to 1

1.7%

inTable 6, andthe sharereceived

by

those with marginaltax rate increases betweenzeroand fivepercentrises from

16%

to

72%.

For

statetax deductions, thereisvery littlechange as aresultofthe inclusionofstatetaxes.

4.2

Weighted Average

Marginal

Tax

Rates

Tables 5 and6 offer

some

insightson

how

the

AMT

affects taxpayerincentives.

The

tables do not consider

how

many

taxpayersface the

AMT,

however, andthey donotyield single

summary

measuresofthe impactofthe

AMT.

To

address these needs,

we

compute

weighted

average marginal tax rates ona rangeofdifferent

income

components

receivedby alltaxpayers.

We

includeboth

AMT

payers andothertaxpayers, andthereby obtainameasure ofthe

AMT's

16

category

by

averaging marginaltax ratesacross taxpayerswith weights equalto theshareof

aggregate

income

or deductions inthatcategory received

by

eachhousehold.

Table7

shows

theweightedaverage marginal

income

tax rates

on

four

income

components

and

on

four deductioncategories.

The

table

shows

weighted average marginal tax

rates everythreeyears

between

2001 and

2013

both with and withoutthe

AMT.

Forthe"without

AMT"

case

we

assume

thatthe

AMT

isrepealed butthatthere are no otherchangesinthe

personal

income

taxschedule. In practice,

AMT

repeal

would

presumablycoincide with changes

inthe

income

tax lawthat

would

restoreatleastpartoftherevenue thatiscurrentlycollected

by

the

AMT.

Because

there are

many

possible

ways

in

which

the

income

taxcouldbe modifiedto

recoup lost

AMT

revenue, however,

we

decided notto

make

anarbitrary choiceandreport the

associatedresults.

Comparing

the

columns

for2001 and

2004

inTable 7

shows

thatthe

AMT

hasa relatively

minor

effect

on

weighted averagemarginal

income

tax rates intheseyears. For2004, for

example,the estimatedweightedaveragemarginal tax rate

on

wages

is

22.6%

with the

AMT,

and

22.3%

withoutit.

The

only

income

ordeduction categoryfor

which

the weighted average

marginaltax ratediverges

by

more

thanonepercentage pointis state

income

taxes,

where

the

disparityin

2004

isnearlysixpercentagepoints.

The

weighted averagemarginaltax rateswithandwithoutthe

AMT

diverge

by

more

in

2010

thaninthe earlieryears ofthedecade. Forwages,the weightedaveragemarginal tax rateis

projectedtobe 1.5percentage points higherin

2010

as a resultofthe

AMT.

Because the

maximum

AMT

rate of28 percentis

much

lowerthanthe

maximum

regular bracketrates,it

may

be surprising that the

AMT

raisesaveragemarginal taxrates.

However,

the lowest

AMT

bracket

rateof26percent ishigher thanthe incometax bracketof15 percentthatanon-itemizing family

offourwill leave only at$78,500 ($2003), andit is slightlyhigher thanthe25percent ordinary

17

showed

a substantial fractionof

wage income

received

by

AMT

taxpayers facing ahighertaxrate

with, ratherthan without,the

AMT.

The

marginal taxratepatterns forinterest

income

and fordividendsare similartothose

forwages. In2004, the

AMT

hasanegligible effect onthe weightedaverage marginal taxratefor

interestincome, and itraisesthe dividendtax rate

by

0.3 percentagepoints.

By

2010,however,

the effects are larger. Thereis a0.9percentage point increase inthe marginal taxrateon interest

income, anda 0.7 percentage point increaseinthe tax rateon dividends.

The

distributionof

dividend

income

is

more skewed

toward high-income householdsthanthe distributionofinterest

income, soa higherfractionofdividend recipients have marginal incometax rates that areabove

theirmarginal taxrateunderthe

AMT.

The

fourth

row

in Table 7

shows

the impactofthe

AMT

on themarginaltaxburden on

capital gains.

The

weightedaveragemarginal tax ratewith the

AMT

is higher thanthatwithout

the

AMT,

with thedifferencereaching its

maximum

at0.7percentagepoints in2010.

The

differenceis0.2 percentagepoints in2004.

The

tablealso

shows

thattheweightedaverage

marginal tax rateundercurrentlawis lowerin

2004

than in2001,reflecting thereduction in

marginal tax rates that

was

enacted in 2003. Forrealized long-termcapital gains,

JGTRRA

reduced theweightedaverage marginal tax rate by aboutfourpercentage points.

The same

tax act

loweredthemarginaltax rateon dividend income,andthe third

row

ofTable7

shows

the impact

ofthischange:a tenpercentage pointdrop inthe weightedaverage marginal taxrate.

The

last fourrowspresentinformation on

how

the

AMT

affects theweightedaverage

marginaltax rates that applytovarious personal incometaxdeductions. Forstateand local

income

taxdeductions,

shown

in

row

five,the effectis dramatic. Thesetax paymentsarenot

deductible from alternative

minimum

taxableincome,but theyaredeductible from taxable

income

undertheordinary

income

tax. In2004, theweightedaveragemarginal tax subsidy on

havebeen 24.1 percent.

The

disparityisthe resultof roughly onethirdofstateand localtax

deductionsbeing claimedby taxpayers

who

are inthe

AMT

regime, and for

whom

these taxes do

not generate adeduction.

The

disparity

between

theaveragemarginal taxratewith andwithout

the

AMT

isprojectedto

grow

overthenext decade. In2010,theweighted averagemarginal tax

rate

on

stateand localtax deductionsisprojectedtobe9.4 percentwiththe current

AMT

in effect,

compared

with26.9 percent ifthe

AMT

did notexist.

The

projectionsalso

show

that expiration

ofthe

EGTRRA

provisionsin2011 bringsthe

two

sets ofweighted average marginaltax rates

into closerproximity,withthe weighted marginaltaxrateunderthe statusquo, withthe

AMT,

rising

from

9.4%

in

2010

to

18.2%

in2013.

The

AMT

hasa smallerimpactonthe weightedaverage marginaltax rates fortheother

deductionsthat

we

consider. In2004, forexample, ithasno effect

on

the weighted average

marginal tax rateon medical deductions,and itaffectsthemortgage interestdeduction

by

0.4

percentage points andcharitablecontributions

by

0.7 percentagepoints. In2010, allofthese

disparities are larger.

The

weighted average marginaltaxrate formedicaldeductions is0.5

percentagepoints lowerwiththe

AMT

thanwithoutit,whileboththe mortgage interestand

charitablecontributions deductionsare atweightedaveragesthat arenearly

two

percentage points

greaterunderthe

AMT

thanwithoutit.

Table 8 presentsweighted average marginaltaxratecalculations similar tothoseinTable

7, butitincludestheeffectofthe

AMT

on net-of-deductibilitystate

income

tax rates aswellas

the effect

on

federalrates.

The

levelofthemarginaltax rates in thistable ishigher than the level

inTable7, andthe differences that resultfromthe

AMT

are also larger. For wages, forexample,

theprojected impactofthe

AMT

in

2010

is an increaseintheweightedaverage marginaltaxrate

of2.2percentagepoints,

compared

with 1.5percentagepoints

when

onlyfederaltax rates are

19

interest,the weightedaveragemarginal taxrate ondeductions risesto2.6 percentage points in

Table 8,

compared

with2.2 percentagepointsinTable 7.

The

weightedaverage federal marginal

income

tax ratesinTable 7 are projections.

To

provide a base forcomparison with boththe level andthevariation in similarmarginal tax rates in

thepast, Table 9 presents data foryearssince 1960on theweighted average marginaltax rates on

wages, several different

components

ofcapitalincome, and several deductions.

The

datainthe

appendixrepresentanupdated version ofthetimeseries formarginal tax ratesin Poterba(2002).

The

table

shows

thatthe changesassociatedwith the

AMT,

evenin2010, are in

many

cases

smallerthanthechangesassociated with significant taxreforms inthepast.

The Tax Reform

Act

of1986, forexample,reducedthe weightedaverage marginal

income

tax rate on

wage

income

by

4.3 percentagepointsover the 1986 to 1988period.

The

marginal

income

tax ratereductions

associated with

ERTA,

the 1981 tax legislation,

were

comparable inmagnitude.

5.

Reform

OptionsandTheirImpacton

Weighted

Average

Marginal

Tax

Rates

The

foregoinganalysis considered

how

various reformproposals

would

affect the

number

of taxpayers facing the

AMT

and the distributionof

AMT

payers across

AGI

categories.

We

have

notconsidered

how

reforms

would

affectweightedaveragemarginal taxrates. Table 10 presents

such informationfor

two

income components,

wages

and interest income, and for state

income

tax

deductions. Thisdeductionisthe onewith the largestchange in the weightedaveragemarginal

taxrateasa resultofthe

AMT.

The

results inTable 10

show

thatall oftheproposed reforms havethe effect of reducing

the marginal taxrateson

wages

and on interest income,although the magnitudeofthese effects

vary acrossproposals.

The

proposalstosubtractpersonalexemptions from alternative

minimum

taxable

income

andto indexthe

AMT

exclusion forinflationarethe

two

thathave the largest

20

overapercentage pointas aresultofthesereforms. Forinterest income, thesetwoproposals

along with extendingthe expiringprovisions in

EGTRRA

and

JGTRA

havethe greatestmarginal

taxrateimpact. Indexingand allowing personalexemptionsunderthe

AMT

have significant

effects on theweightedaverage marginal

income

taxratethatapplies tostate incometax

deductions. Indexing, forexample,

would

raisethismarginal tax rate

by

nearlyfive percentage

points in2010. This isthe largestabsolute effectofany ofthe proposed reforms on any ofthe

marginal tax ratesinTable 10.

Given

thatthe

AMT

has

modest

effectson

most

marginal tax

rates, itisnotsurprising that the effectof

most

ofthereformproposals isalsomodest.

6. ConclusionsandFuture Directions

Thispaperpresents

new

evidence onthegrowth of

AMT

liabilitiesoverthenextdecade,

andthe incentive effects associatedwiththisgrowth.

Our

baseline projectionsconfirmthewidely

documented

patternthatbecausethe

AMT

exclusionlevelis not indexed,there will berapid

growth inthe

number

of

AMT

taxpayers, andinthe

amount

of revenue collected

by

the

AMT,

until 2010. After 2010,

when

a

number

ofprovisions inthe2001 taxreformarescheduledto

phase out,the

number

of

AMT

taxpayerswilldecline, butitwillrise againinsubsequent years

fromthe lower post-2010base.

Althoughthe

AMT

creates substantial changesin

many

aspectsofthe

income

tax system,

we

find that theaverage marginaltax rateson

many

income

components, suchas

wages

and

interestincome, areaffectedonlymodestly by growth ofthe

AMT.

In2010, forexample,

we

project that the

AMT

will raise theweightedaveragemarginaltax rateon

wage

income

by

1.3

percentagepoints relativeto

what

it

would

be ifthe

AMT

were

repealed and

no

othertaxchanges

were enacted. For interest income,the effect ontheweighted average marginal taxrateisjust

below

onepercentagepoint. These changesconceal largerchanges inmarginaltaxrates for

21

ordinaryincometax into the

AMT

brackets of26 and 28percent. Othersdrop from marginaltax

ratesabove 30percentunderthe

income

taxtothe

AMT

rateof 28 percent.

Our

resultsarebased on stylizedassumptions aboutthe rateat

which

aggregate income,

population, andtheprice level will

grow

overthe nextdecade. Allof our analysisusesthe

same

underlyingassumptionsto "age"

income

taxrecords from 1999 through2013. It

would

be useful

togauge the sensitivity ofour findingstotheseassumptions, andtoexplore

how

differencesin

growthrates or

more

importantlyinthe rate ofinflation

between

2003 and

2010

would

affect the

magnitude of

AMT

liabilities.

One

oftheunder-studied issuesassociatedwith the

AMT

concernsthe impactofthistax

onthe tax liabilityofsinglehouseholds andmarried couples.

The

effectofa changein household

status ontax liabilitiesisdifferent

when

thetaxpayers facethe

AMT

than

when

they facethe

regular

income

tax schedule. Gravelle (2001) has notedthis potentiallyimportanteffect.

The

growingsignificanceofthe

AMT

may

thereforecallinto questionthe traditionalanalysisofthe

"marriagepenalty"andrelated featuresofthe

income

tax system. Future research should

considerthisaspectofthe distributionof

AMT

liabilities, along with potentialreformsto address

22

REFERENCES

Burman,

Leonard, WilliamGale, andJeffrey Rohaly.

"The

AMT:Problems

andPotential

Solutions,"National

Tax

Journal 55 (September2002), 555-596.

Burman,

Leonard, WilliamGale, andJeffreyRohaly.

"The Expanding Reach

oftheAlternative

Minimum

Tax." Journal of

Economic

Perspectives 17(Spring 2003a), 173-186.

Burman,

Leonard, WilliamGale, andJeffreyRohaly.

"The

AMT:

Projections and Problems,"

Tax

Notes (July 7, 2003b), 105-117.

Feenberg, DanielR.,and ElisabethCoutts.

"An

Introductiontothe

TAXSIM

Model." Journalof

PolicyAnalysisand

Management

12 (Winter 1993), 189-194.

Gravelle, Jane.

"The

IndividualAlternative

Minimum

Tax: Interactionwith MarriagePenalty

Reliefand Other

Tax

Cuts." Congressional Research Service,June 12,2001.

Harvey, RobertP.andJerry Tempalski.

"The

Individual

AMT:

Why

it Matters,"National

Tax

Journal 50 (1997),453-473.

Kiefer,Donald, RobertCarroll,JanetHoltzblatt, AllenLennan, Janet

McCubbin, David

Richardson, andJerryTempalski.

"The

Economic Growth

and

Tax

Relief Reconciliation

Act of2001:

Overview

andAssessment ofEffects

on

Taxpayers." National

Tax

Journal

55

(March

2002), 89-118.

Poterba, James. "Taxation, Risk-taking, andPortfolio Behavior."In

Alan

Auerbach

and Martin

Feldstein, eds.:

Handbook

of PublicEconomics:

Volume

3 (NorthHolland,

Amsterdam,

2002).

Rebelein, Robert,and JerryTempalski.

"Who

Pays the Individual

AMT?"

Washington: U.S.

Treasury Department,Office of

Tax

AnalysisPaper 87, 2000.

23

Tempalski,Jerry.

"The

Impact ofthe2001

Tax

Billon the Individual

AMT."

Proceedings ofthe

94thAnnual National

Tax

Association Conference onTaxation. National

Tax

Association: 2002. 340-348.

U.S. Congress, Joint

Economic

Committee.

The

Alternative

Minimum

Tax

forIndividuals:

A

24

Table 1: Projected

Number

ofAlternative

Minimum

Tax

Returns

Year

Number

of

AMT

Returns

Percentof

Returns with

AMT

AMT

Revenue

AMT

Revenue/Total

Income Tax

Revenue

2001 2.47 1.9 9.26 1.08

2002

3.56 2.7 11.04 1.33

2003

4.06 3.04 15.55 1.97

2004

3.49 2.59 14.16 1.62

2005

14.62 10.71 35.99 3.69

2006

17.94 12.99 48.22 4.60

2007

20.06 15.78 60.07 5.31

2008

28.35 20.04 81.16 6.68

2009

32.74 22.87 98.92 7.46

2010

37.11 25.61 125.53 8.78 2011 16.80 11.46 45.81 2.70

2012

19.44 13.18 54.6 2.99

2013

22.84 15.21 65.04 3.30

25

Table 2:

Number

of

AMT

Taxpayers

Law

and

Reform

Proposals

and

Ags

;regate

AMT

Liabilities,,2001- 2013, Current

Tax

Filers

Tax

Liability

2001

2004

2007

2010

2013 2001

2004

2007

2010

2013 Status

Quo

2.1 3.3 21.8 37.2 23.8 7.4 11.7 57.5 125.7 68.1

Allow

Personal Exemptions 1.1 1.8 5.7 12.2 5.1 6.2 9.4 24.7 49.7 27.8 Extend Expiring Provisions 2.1 3.3 7.1 14.1 22.9 7.4 11.7 24.8 57.6 101.9

Marriage

Tax

Relief

1.7 2.7 14.2 25.3 15.0 6.8 10.5 40.0 87.4 46.6 Index

AMT

Exclusion 2.1 2.1 9.8 14.0 3.1 7.4 9.3 26.7 47.8 17.9

Add

$3,000 to

AMT

Exclusion 1.5 2.6 16.1 30.9 18.2 6.7 10.4 44.7 101.9 53.8

Add

$10,000to

AMT

Exclusion 1.0 1.7 7.7 17.0 8.6 5.8 8.6 27.0 65.3 33.5

Note: Filers aremeasured in millions,

projections usingthe

NBER TAXSIM

liability in billionsofdollars. Entries are basedon

model, asdescribed inthe text.

Table 3:

AMT

Liabilityas aPercentageofAdjusted Gross Income,

By AGI

Category,

2001-2013

AGI

Class 2001

2004

2007

2010

2013

<25K

0.1 o I o I 0.1 0.1

25-50K

0.1 0.4 0.3

50-75K

0.5 1.3 0.6

75-100K

0.1 0.1 0.8 1.6 0.5

100-200K

0.1 0.1 1.3 2.4 1.2

200-500K

0.4 0.6 1.8 2.7 0.9

>500K

0.3 0.4 0.5 0.6 4

TOTAL

0.1 0.2 0.7 1.3 0.6

26

Table4: Projected Probabilities of

AMT

Payment by AGI, 2010

Status

Quo

Allow

Personal

Exemptions

Extend Expiring Provisions Marriage

Tax

Relief Index

AMT

Exclusion

Add

$3,000to

AMT

Exclusion

Add

$10,000 to

AMT

Exclusion AllTaxpayers

<25K

0.9%

0.6%

0.6%

0.9%

0.6%

0.7%

0.6%

25-50K

15.8 2.2 3.2 13.1 3.2 10.1 3.9

50-75K

56.3 7.2 12.4 33.8 14.7 43.6 16.6

75-100K

77.2 15.8 20.7 41.4 23.9 66.0 28.5

100-200K

94.7 57.5 56.6 67.1 49.7 90.1 66.2

200-500K

91.6 88.9 91.1 89.6 83.7 91.5 91.2

>500K

33.3 30.0 35.5 33.2 31.8 33.3 33.3 All

Incomes

25.7 8.4 9.8 17.4 9.7 21.3 11.7 All Itemizers

<25K

1.2%

0.8%

0.8%

1.2%

0.8%

0.9%

0.8%

25-50K

11.2 2.6 2.6 9.2 2.9 7.7 3.3

50-75K

47.1 7.1 11.0 28.1 13.0 34.9 15.4

75-100K

74.8 16.2 21.2 37.7 25.1 60.9 30.1

100-200K

94.9 58.6 58.7 67.2 52.4 90.1 67.7

200-500K

92.5 90.0 92.1 90.6 84.8 92.5 92.1

>500K

33.9 30.5 36.1 33.8 32.4 33.8 33.8 All

Incomes

52.5 22.2 24.3 34.8 23.9 45.1 29.0

Taxpayerswith

Two

or

More Depe

ndents

<25K

3.4%

2.3%

2.5%

3.4%

2.5%

2.7%

2.5%

25-50K

52.1 3.4 13.5 y).} 11.5 37.1 13.8

50-75K

91.9 17.4 39.9 50.4 48.0 84.5 53.6

75-100K

98.9 30.1 53.6 51.7 63.8 98.2 73.5

100-200K

99.4 56.6 77.1 68.5 82.1 98.9 89.5

200-500K

100.0 98.4 99.9 99.3 95.5 100.0 100.0

>500K

12.3 12.3 12.3 12.3 12.3 12.3 12.3 All

Incomes

56.1 15.6 27.8 35.6 30.6 50.4 34.3

11

Table 5: Effectof

AMT

on Marginal

Tax

Rateson Vai

Tax

Rate Only,

Weighted

by

Amount

of

Income

Items

ious

Income Components,

2010, Federal

Marginal

Tax

Rate

Change

(Percentage Points)

Wages

Dividends Interest

Long

Term

Gains State Taxes Contributions Declineby

>

15 0.0 0.1 0.1 0(1 0.0 1.4 Decline

by

10-15 0.1 0.1 0.0 0.0 0.0 12.2 Decline by 5-10 2.3 6.9 5.0 0.2 0.0 16.9 Decline by 0-5 15 3 25.4 20.9 2.6 0.0 4(> s

No

Change

0.1 0.4 0.3 11.7 15.1 8.5 Increaseby 0-5 43.8 35.1 41.2 70.0 0.0 9.4 Increaseby 5-10 20.6 20.6 19.0 15.2 0.1 6.2 Increaseby 10-15 17.5 9.5 11.2 0.7 4.2 0.0 Increaseby

>

15 1.1 3.0 3.3 0.5 80.6 0.0

28

Table6: Share of

Income

and Deductions Affected

by

Various

Changes

inFederalandEffective

State

Margin

al

Tax

Rates

Due

to

AMT,

2010

Marginal

Tax

Rate

Change

(Percentage Points)

Wages

Dividends Interest

Long

Term

Gains State Taxes Contributions Decline

by

>15

0.0 0.1 0.1 0.0 0.0 1.4 Decline

by

10-15 0.1 0.1 0.0 0.0 0.0 12.2 Decline

by

5-10 2.3 6.8 5.0 0.2 0.0 16.9 Decline

by

0-5 15.3 25.5 20.9 2.6 0.0 46.5

No

Change

0.1 0.4 0.3 11.7 15.1 8.5 Increase

by

0-5 43.8 35.1 41.2 70.0 0.0 9.3 Increase

by

5-10 20.6 20.6 19.0 15.2 0.1 6.2 Increase

by

10-15 17.5 9.5 11.2 0.7 4.3 0.0 Increase

by

>15

1.1 3.0 3.3 0.5 80.5 0.0

29

Table 7:

Weighted

Average

Me

irginal Federal

Tax

Rateson

Income

and Deductions

With

AMT

Without

AMT

2001

2004

2007

2010

2013 2001

2004

2007

2010

2013

Wages

24.1 22.6 23 9 25.1 27.4 23.9 22.3 22.5 22.6 26.6 Interest 32.4 30.0 3 1.4 32.9 35.8 23.5 21.2 21.5 21.9 25.5 Dividends 36.6 26.5 27.7 36.4 39.4 26.1 15.8 15.8 23.5 27.5 Realized

Long-Term

Capital Gains 20.4 16.3 16.6 16.8 15.6 16.9 12.9 12.7 12.2 11.2 State Taxes -17.2 -13.6 -8.0 -6.5 -13.7 -21.1 -19.1 -20.2 -22.3 -26.1 Medical Deductions -12.1 -11.1 -11.1 -11.5 -13.7 -12.1 -11.1 -11.4 -11.9 -14 4

Mortgage

Interest -22.1 -20.8 -22.8 -24.4 -26.5 -21.8 -20.2 -20.6 -21.8 -25.4 Charitable Contributions -21.7 -20.0 -21.9 -23.6 -26.1 -21.5 -19.4 -20.0 -21.6 -25.3

Notes: Without

AMT

scenarioassumes repeal ofthe

AMT,

butno otherchangesto

income

tax

law.

Table 8:

Weighted

Average Marginal

Tax

Rateson

Income

and Deductions, Federal PlusNet

State

Tax

Rates

With

AMT

Without

AMT

2001

2004

2007

2010 2013 2001

2004

2007

2010

2013

Wages

28.6 27.4 29.2 30.8

-

(, 28.4 27.1 27.8 28.3 32.8 Interest 35.9 33.8 35.7 37.7 41.1 27.0 25.0 25.9 26.7 ;o.s Dividends 40.6 30.5 32.0 41.5 45.0 30.0 19.8 20.1 28.6 33.1 Realized

Long-Term

CapitalGains 24.9 21.0 21.5 22.0 21.2 21.3 17.6 17.6 17.4 16.7 State Taxes -19.7 -16.3 -11.0 -10.0 -17.7 -23.6 -21.8 -23.3 -25.8 -30.1 Medical Deductions -14.2 -13.4 -13.7 -15.0 -17.8 -14.3 -13.4 -14.1 -15.0 -17.8

Mortgage

Interest -25.4 -24.3 -26.8 -28.8 -31.3 -25.2 -23.8 -24.6 -26.2 -30.2 Charitable Contributions -24.8 -23.2 -25.5 -27.6 -30.5 -24.6 -22.5 -23.6 -25.6 -29.7

Notes: Without

AMT

scenarioassumesrepeal ofthe

AMT,

butno otherchangesto

income

tax

30

Table9: Weighted AverageMarginal

Tax

andSubsidy Rates, 1960-2002

Year

Wage

Income Interest Income Dividends Realized

LT

Gains Mortgage Interest Pension Income

1960 21.86 n/a 41.64 18.99 n/a n/a

1962 22.11 25.55 42.03 17.53 n/a n/a 1964 20.52 23.18 39.37 17.43 -20.08 n/a 1966 20.11 22.39 37.09 17.73 -19.74 n/a 1967 20.46 22.61 37.86 17.81 n/a n/a 1968 22.89 25.76 40.94 20.27 -22.80 n/a 1969 23.90 26.88 41.34 20.92 n/a n/a 1970 22.89 25.99 38.52 18.08 -22.37 n/a 1971 22.01 25.07 37.16 17.98 n/a n/a 1972 22.46 25.43 36.87 17.86 -22.50 n/a 1973 23.33 26.58 38.00 17.89 -23.43 n/a 1974 24.17 27.53 39.31 18.17 n/a 21.35 1975 24.86 26.62 39.03 17.35 -24.73 21.97 1976 25.71 27.52 41.12 18.24 -25.72 24.70 1977 26.73 27.87 41.23 20.00 -27.62 22.81 1978 28.09 29.06 42.61 19.57 -28.50 23.88 1979 28.22 29.74 43.04 18.34 -28.23 24.03 1980 29.78 31.42 43.01 17.38 -29.12 26.15 1981 30.90 32.15 41.33 18.86 -29.99 27.54 1982 28.86 28.69 35.44 19.18 -27.14 25.89 1983 26.96 26.22 34.04 16.88 -24.83 23.77 1984 26.39 26.56 32.86 17.38 -24.36 24.18 1985 26.59 26.71 32.74 17.35 -24.99 24.10 1986 26.74 25.62 30.95 17.97 -24.82 23.73 1987 23.92 23.16 27.35 24.52 -23.08 n/a 1988 22.47 22.10 24.91 25.51 -22.36 22.80 1989 22.55 22.72 25.04 25.08 -22.44 22.31 1990 22.58 23.55 25.15 24.95 -22.26 22.52 1991 22.55 22.63 25.50 23.89 -21.80 22.22 1992 22.55 22.11 25.28 24.33 -21.63 22.43 1993 23.32 23.56 27.10 25.79 -22.41 22.87 1994 23.61 24.33 27.41 26.11 -22.71 24.02 1995 23.90 25.23 27.85 26.56 -22.77 24.48 1996 24.04 25.09 28.02 26.58 -23.08 25.33 1997 24.46 25.93 28.84 20.34 -23.23 26.08 1998 24.72 26.09 28.75 19.82 -23.26 26.53 1999 25.07 26.47 29.04 19.80 -23.42 27.03 2000 24.99 26.22 28.83 18.53 -23.15 26.82 2001 24.06 25.09 27.90 18.36 -22.32 25.65 2002 24.00 24.99 27.69 18.45 -22.49 25.59

m

I - _{CN NO On oo} rn

—

i <'-. oo CN NO

d

"t r-~'

o

—

. ,

—

i oo i

—

i CN i

—

i cn i t i i

o

<r,

©

ON

o

_d

rn r--' •^

o

cn

C

i i ON 1 ON 1 On t r- 00 ON '* rn On ^t_

o

o

o

SI

^

—

_d

—

' OO -^ cn 1 1 i i r ~. _-1-

_c

_in NO

—

C nO i— i ON 3j CO

O

o

CN ro

*

rn Tf

^

*' ^F 'J /

-i "7 '"7 1 i 1 C ,~ <D ,_^ CN t> CN >n CN in ON

o

o

r-; _{r~ r~ i> C~ r-^ r~} *-, Di s. CN i 1 1 1 i 1 i i _2

m

s. CN _, NO

_q

NO

m

2

<

_O

CN iri in CN

m

iri in in V.

m

m

m

m

m

m

m

o ^ — u

>

o

On ON •* r- r~- nO

_q

d

o

CN r 1 -J

—

! CN

—

• _oi CN

o

/ rn rn

m

ro

m

m

m

—

s c/l *_> u bJD t^ -r NO r- CN f~ CN nO 3— C-

o

o

CN __

d

d

_;

d

_J

d

^H g

"

m

rn

m

m

m

m

O z 1) -5

U

o _<*

~

_<*

_o

_{ON P~; oo in}

₃

5

o

o

CN

~

ON

d

ON On ON ON

o

"

. (N rn CN Ol CN CN d — ^ 3 _o en (D

_

-T On

*

ro •* CN On

o

o

CN CN ,— < CN CN CN CN

—

5

>

c

m

rn

m

CO rn

m

m

Cl> q " 3-

©

ON CN oo

m

—

i CO CU

o

CN f» l>

^

r^ NO r-^ r^ _o CN CN CN CN CN CN CN

^

~u

X

O

—

i NO OO

m

nO r~-

o

CL, —

-

_o

CN in rn rn •<t

m

^J-

^

CN CN CN Ol CN CN CN -J | C 5c r» On

q

>— 1 "* CN NO

—

i 3 CO ca

o

—

CN

m

rn ro' r^i

m

rn

m

to 03 S CN CN CN CN CN CN CN 1) •3" no in NO NO ITj vq in CB

O

o

CN CN CN CN CN CN CN CN 5 > CN CN CN CN CN CN CN CO

<

v TD <L>

<

CO ca

o

o

CN -r «* ^r

*

-i •^ *d-~3 CN CN CN CN CN CN CN 3 -J ca 3

O

5

co CO O £ 3 Pi .2 " S

—

<L> X co op _§ "3

§

'5

5

•- "> x x £

c

&

5 x

5

§

X 3

a

x

O

o

O

X

" f—i

2

f^

C

'35

3

r^ X

O

jH

o

.2

O

w

O

[-H

3

3

o"

5

o

35

H

r/5

<

Oh LU

W W

Ch

& w

<

2

w

<

&o

<

m

Z

Figure1

RegularandTentative

AMT

Effective Rate in2003 and 2010

JointFiling ,

No

Deductions and2 Dependents

0.35

?e \oe i?p ztO

W

3ot itt> Vfft ffr

&

Figure 2

EvolutionofAverageTaxRates 2002-2010

Fixed Distribution ofIncome1995-1999

800 c o J5

h-<

!

11

140 120 100 80 60 40 20

Figure 3

Effect ofRaising Exclusion in201

\AMT

sAMTforAGI>10C )K

^<\

-?Ss_ _{#with/kMT &AG >100K} , v#with

M

*•• AT

/

j?

_—

llliiiij

-——

—

' 8. 10 20 30 40 50 60 70

Increase in ExcercisefromS40,250(joint)

(thousands)

Date

Due

MITLIBRARIES

IP