The effect of the alternative minimum tax on corporate tax burdens

The effect of the alternative minimum tax on corporate tax burdens

Quarterly Review of Economicsand Face, Vol. 33, No. 4, Summe r 1993, pages 123439 CopyrightQ 1993 by Board of Trustees of the University of Illinois A...

1MB Sizes 0 Downloads 1 Views

Quarterly Review of Economicsand Face, Vol. 33, No. 4, Summe r 1993, pages 123439 CopyrightQ 1993 by Board of Trustees of the University of Illinois All rights of reproductionin any form reserved. ISSN 99355797 ‘lke

The Effect of the Alternative Minimum Tax on Corporate Tax Burdens THOMAS C. OMER and DAVID A. ZIEBART University of Illinois at Urbana-Champaign

The Tax Refm Act of 1986 raised the corporate alternative minimum tax @MT) and explicitly linked corporate taxation tofinancial accounting -Book Income”. Congress added the book income adjustment (although temporarily) to eliminate highly publicized instances in which corporations with substantial book income paid no tax. The intent of this study is to assess the incidence of the AMT and the industry groups expected to pay additional taxes urulerthe new law. Our results indicate that whit%corporate taxespaid increased aper implementation of the law, the incidence of the new law did not necessarily fall on those fii expected to pay the new tax. Firms with increased tax burdens did not necessarily incur lower tax burdens prior to the implementation of the new AMT, nor were thq the most p-ofitable. Consequently, thepotential negative impact of this new law may have overshadowed its expected revenue generation and corredive benefits.

The 1986 Tax Reform Act contained a new corporate alternative minimum tax (AMT) that was motivated by the perception that some U.S. corporations did not pay their “fair” share of the corporate tax burden. These corporations reported substantial earnings to their stockholders via their financial statements but paid little, or no, income taxes to the federal government, Citizens for Tax Justice (1985). The AMT contained several provisions intended to curb future discrepancies between reported profits and taxes paid. In particular, a new provision that specifically incorporated the financial statement income numbers was implemented. This provision was referred to as the book income adjustment (BIA). Unfortunately, in devising this means of taxing high profit/low tax corporations, Congress likely increased the compliance costs of all corporations because the BIA required that a substantial number of corporations with differences between accounting income and taxable income estimate their AMT liabilities even though the firm would not ultimately owe the additional tax under theAh4T rules. The BIA was also a concern to the Financial Accounting Standards Board (FASB) and American Institute of Certified Public Accountants (AICPA). They felt it was likely to provide incentives for many corporations to alter reported 123

124

QUARTERLY REVIEW OF ECONOMICS AND FINANCE

accounting financial

numbers. The FASB and AICPA suggested that alteration statement

information

would likely have a negative

of reported

impact

on the

reliability of corporate financial statements.’ This study looks at the incidence of the AMT across firms and industries determine

the extent to which Congress’ objective

of firms reporting incidence include

of increasing

to

the tax burden

high profits but historically paying little tax was achieved. If the

of the tax is much broader than expected, potentially

less reliable

financial

the potential negative effects and increased

compliance

costs. We suggest that these negative effects may have outweighed

statements

the benefits

associated with changes in the tax burden of target firms. Using firm characteristics

and predictions

derived from previous research, incidence

of industry susceptibility

to the AMT

we provide evidence on the extent to which the

of the AMT affected high profit/low tax firms. Our results support the

notion that the AMT had a much broader impact than originally intended but the change in the relation between reported profitability The remainder corporations

of this paper is organized

determined

and tax payments was small.

as follows. Section two reviews how

their AMT liabilities.

Section

likely to affect payment of the AMT, our operational and estimates of the corporate

three

discusses

measures

factors

of these factors,

tax burden. Section four contains our analysis and

results while the last section provides a summary and our conclusions.

1990

THE Am--PRE

The AMT was a separate system for calculating under

regular

rules were required

to calculate

The AMT calculation adjustments, alternative preferences

of firms that,

tax rules, had little or no tax liability for the current

period. Any firm having preferences

operating

the tax liabilities

and adjustments

taxable income

and adjustments.

taxable

income

before

any net

added AMTp-eferences, added or subtracted AMT

added the BIA, and subtracted minimum

reporting

with the AMT

an AMT liability.

started with regular

loss (NOL) deduction,

associated

the AMT NOL deduction

to arrive at

(AMTI) . Table 1 provides a list of the AMT

AMT preferences

are items permanently

excluded

from taxation under regular income tax rules or items that provided substantial tax deferral. AMT adjustments are items for which the timing of the deduction was different under AMT and regular tax rules. Even though the AMT preferences and adjustments were extensive, Congress added the BIA and required that it be calculated only after all adjustments and preferences had been considered. However, the BIA was considered

a temporary

part of the new AMT and was

only in effect for the tax years 1987, 1988, and 1989. The Treasury was charged with developing

a new adjustment

factor that did not depend on accounting

book

THE EFFECT OF THE ALTERNATIVE

ALTEBNATIVE

Table 1.

MINIMUM TAX PREFERENCES

Preferences

MINIMUM

TAX

125

AND ADJUSTMENTS

Adjustments

The excess percentage depletion over adjusted basis Depreciation adjustment of depletable property after 1986.

for assets placed in service

The excess intangible drilling costs over 65 percent of taxable income.

Amortization and capitalization ofmine exploration and development costs.

The excess of the reserve for bad debts over actual experience.

Circulation expenditure

Taxexempt interest on primte activity bonds issued after g/08/86

Long term contract adjustment entered into after 03/01/86

The excess allowable charitable contribution property over property’s adjusted basis.

amortization.

of Adjustment for depreciation of pollution facilities placed in service after 12/31/86

Theexcessofaccelerateddepreciationoverstraightline.

Adjustment for use of the installment report gains Adjustment

control

method

to

of net operating losses deduction.

Book income adjustment

income. The Treasury produced

for contracts

(BIA)

the adjusted current earnings

(ACE) adjustment

that was effective for tax years after December 31,1989. The BIAwas 50 percent of the difference between the firm’s adjusted net book income

and AMTI before

firm’s adjusted applicable

calculating

net book income

the BIA or subtracting

was the net income

financial statement.2After

adding the BIAand

AMTI was reduced by the standard exemption by 20 percent.‘This this amount,

the AMT foreign tax credit (FIX)

were subtracted

amount

gave the tentative minimum

AMT NOLs. The

or loss reported

on its

subtracting AMT NOLs, ($40,000)

tax (TMT)

and multiplied

before credits. From

and investment

tax credit (ITC)

giving TMT. Finally, the firm’s regular tax liability, net of foreign

tax credits and possessions

tax credits, was subtracted

from TMT giving the AMT

liability. If the AMT liability was positive the firm owed the AMT liability plus its regular tax liability. If the AMT liability was zero or negative the firm owed only its regular tax liability. The AMT provided partial relief in situations where accounting require different

recognition

and tax rules

of income and expense. In such instances, the AMT

tax credit could be used against the regular tax liability in future periods to reduce the double tax. On the other hand, this credit applied only to timing difference reversals and did not alleviate problems

associated

addition,

the credit could only be used in years that the firm’s regular tax liability

exceeded

the firm’s regular tax liability plus the AMT liability. The calculation

subject to AMT rules.

whenever

municipal

a firm had preferences

bond

differences

tax and accounting

the AMT liability occurred

rules, for example

with permanent

between

interest.

In of

and adjustments

126

QUARTERLY REVIEW OF ECONOMICS AND FINANCE

WHYFIRMSWOULDLIKELYPAYTHEAMT Factors Influencing

AMT Payment

Our analysis extends previous simulation

research

that evaluated the impact

of the new AMT. We focus on industry predictions about payment of the AMT and individual firm characteristics expected to influence the incidence of the AMT. A review of these studies follows. G. A. Harter (1986) analyzed the AMT impact on corporate

tax liabilities

in

response to assertions that the AMT would increase the corporate tax burden by $139 billion but would also encourage growth. Any detrimental effects of the AMT would fall on capital intensive, slow growth, low profit firms deemed less important to economic growth. It should be noted that this argument seems contrary to Congress’s expressed desire to increase taxes for high profit/low tax firms. Harter concluded

that the incidence

of the minimum tax could be traced to high growth,

low profit firms (many for an extended

period of time).

Harter

(1986,

p. 300)

states “The proposed alternative minimum tax has the character of a tax on growth.” Robert B. Lucke, Mark R. Eisenach, and Larry L. Dildine (LED) (1986) investigated the effect of the Senate’s version of the AMT using simulated data for proto-typical firms in the retailing, durable manufacturing, nondurable manufacturing, and air transportation industries. These industries were chosen because of their asset composition (such as more or less depreciable assets or more diverse asset types). LED’s results suggested that firms in the airline and nondurable manufacturing industries were more likely to pay the AMT than firms in the retail and durable manufacturing industries. LED also hypothesized a relation between the likelihood and firm-specific

characteristics.

They concluded

debt, higher growth, and lower profits had an increased AMT. LED presented

of paying the AMT

that newer firms with higher probability of paying the

the following reasons for linking the incidence

of the AMT

to these firm characteristics. Firms with higher debt were more likely to pay the AMT because larger interest payments reduce taxable income and as a result the BIAwould tend to be greater in proportion to taxable income. Growing firms were more likely to pay the AMT because of the large depreciation preference generated. Less profitable firms were more likely to pay the AMT because taxes on taxable income would not increase as fast as taxes under the AMT even though regular tax rates were higher. Profitable firms were less likely to pay taxes under the AMT because regular taxes would increase faster as taxable income increases. Newer firms were more likely to be subject to the AMT because they would have a relatively high proportion of tax depreciation to taxable income. LED concluded that the combined effect of these firm characteristics would lead to an increased

probability of paying the AMT. They also concluded

that the

THE EFFECT OF THE ALTERNATIVE MINlMUM TAX set of firms likely to pay additional Congress’ original intent. While LED’s simulation only tax preference component

taxes under the AMT was much broader

results are generally

item is accelerated

applicable

depreciation

of the BIA), Lowell Dworin (1987b)

indicates

the BIA and depreciation

adjustments

tax base. Dworin (1987a)

used the Statistics of Income

information

to assess the expected

to companies

(including

127

than whose

the depreciation

that, cross-sectionally,

were the major determinants (SOI)

of the AMT

1981 corporate

impact of the AMT. He suggested

tax

(1987a,

p.

253-254) that previous simulation studies had not adequately addressed the impact of the AMT for two reasons. First, the studies assumed non-stochastic economic changes. Thus, changes in the environment were excluded from the analysis. Second, the studies had not allowed for firms to react to the AMT. In particular, changes in financing and investment strategies could have a significant effect on the impact of the AMT. Dworin (1987b)

determined

that firms with characteristics

of an “airline” or a

“nondurable manufacturer” would attempt to reduce their average growth rate and their debt to equity ratio. The extent of their response was determined by their exposure to the AMT. He also pointed out that under stochastic economic conditions, the period of time that firms would be subject to the AMT would be much longer than expected. He suggested that the range of years firms would be subject to Ah4T and the size of the additional tax bill would vary significantly. Some firms would pay AMT taxes for as many as 28 out of 50 years. Overall, the studies discussed above suggest that the incidence of the AMT may not be highly associated with the firm’s profitability. Other factors such as industry, growth, and age would also affect the incidence of the tax. Unfortunately, few studies have used individual firm data to assess the distribution of taxes prior to the new AhIT nor the distribution after its implementation. This study investigates the distribution of corporate taxes prior to implementation of the AMT in 1986 and immediately following its implementation in 1987. We rely on predictions from earlier studies to determine the association between the likelihood of AMT payment (what firms and industries were most and least likely to pay the AMT) and profit as well as the association between preAh4T and post&MT tax burdens and other factors cited as influencing

AMT payment.

Tax Burden Because much of the publicity regarding problems with the corporate tax laws has focused on firms’ effective tax rates (ETBs), our analysis focuses on the relation between firms’ ETBs and those characteristics that have been predicted as increasing the likelihood of paying additional taxes under the AMT. ETB estimation is an often debated issue in determining a firm’s tax burden. Estimating ETBs requires

a measure of taxes paid and a measure of corporate

income.

Don

128

QUARTERLY REVIEW OF ECONOMICS AND FINANCE

Fullerton (1984) notes that in its most simplistic form, “last year’s tax as a percentage of last year’s income may be a good summary of the burden or redirected

income flow.” However, this computation

paid and income. The estimation nature of corporate restricted

requires a consensus on taxes

of individual firm ETRs is exacerbated by the proprietary tax return information. Consequently, most researchers are

to estimating

ETRs from corporate

financial

statement

information

or

using aggregated Statistics of Income (SOI) data. Previous research incorporating ETR measures use a wide range of definitions

for taxes paid and corporate

taxable

income (Don Fullerton, (1982a, 198213, 1984); Thomas Porcano, (1986); Terry Shevlin, (1987); Clyde Stickney and V.E. McGee, (1982); R.L.Weiss, (1979); Jerold Zimmerman, (1983)). In most cases, selection of various measures were based on ease of computation, consistency with reported IRS summary statistics, or other criteria established by the researcher. For this study, two ETR measures were chosen. The first measure is Zimmerman’s (1983) ETR definition. This measure was chosen because of its reported consistency with IRS SO1 summary statistics. Zimmerman used COMPUSTAT financial statement information to calculate overall ETRs for sample firms from 1941 to 1981. His measure of a firm’s ETR is: ETRI

= Income Tax Expense - Change in Deferred Operating

Tax Liability

(1)

Cashflows

In defending his choice, Zimmerman [17, p. 1371 states, “the magnitudes, time trends and cross-sectional differences in tax rates are comparable using IRS and COMPUSTAT data.” The second ETR measure is: ETR2 =

Income Tax Expense - Deferred Pre-Tax

Book Income

-

Tax Expense

Deferred Corporate

This measure was used in Stickney and McGee

(2)

Tax Expense

Marginal Tax Rate

(1982)

to examine

the relations

between firm size, leverage, capital intensity, and other factors and ETRs. Thomas Omer, Karen Molloy, and David Ziebart (1991) suggest that measures like Stickney and McGee’s are much less susceptible to systematic differences in the reporting of deferred taxes across firms.

ANALYSIS Since LED’s study provides both industry and individual firm ing payment of the AMT, the results of their study are used analysis. Using COMPUSTAT data, we selected sample firms to calculate both ETR measures and the firm characteristics

predictions regardas the basis for this with data available suggested by LED.

All non-regulated

industry firms on the COMPUSTAT

income were considered.

The initial search resulted in a sample of 984 firms with

627 in LED’s four general two-digit SIC Manufacturing product

tape with positive operating

industry groupings.

Industry groups were based on

classifications for retail, airline, and manufacturing was further separated into durable and nondurable

lines. All firms not in the LED industries

grouP* We estimated

the four firm characteristics

were combined

that LED expected

industries. groups by into a single

to be determi-

nants of AMT payment as follows. (1) Debt is proxied by two measures;

the first

measure is total debt minus cash and short term investments divided by total assets (DEBTI). This measures provides an indication of the overall change in the firm’s debt level.4 The second measure is long term debt divided by long term debt plus equity (DEBT2) is also proxied

and indicates only the change in long term debt. (2) Profitability by two measures, the first measure is operating income after

depreciation divided by total current assets plus net plant assets (PROPITR) and represents the return to total producing assets. The second measure is reported accounting net income (PROPITG) and was chosen because many of the complaints about the lack of tax payments are associated with firms that reported large accounting profits. (3) Growth is estimated by the change in sales over five years (GROWTH).5 (4) Age is estimated by the ratio of accumulated depreciation to cost of total plant assets (AGE). On average, the longer a firm is in existence the larger the ratio of accumulated depreciation to total plant assets. Descriptive statistics for ETRl, ETR2, PROPITR, PROPITG, DEBTI, DEBT2, GROWTH, and AGE are presented in Table 2 for the total sample for 1986 and 1987. Tables 3 and 4 provide descriptive statistics for firms in the LED industries for 1986 and 1987. The results in Table 2 indicate

that, on average, our sample firms continued

to grow over 1986 and 1987 and increased

their long term debt. Both effective tax

rate measures indicate an increase in the tax burden of firms in our sample. On average the profit rate stayed the same (13 percent) while reported accounting profits increased.6 Results in Tables 3 and 4 indicate except

increased

for the retail industry which experienced

tax burdens

across LED groups

a slight decline.

The airline

transportation industry experienced the greatest increase in tax burden with an average 1986 ETRl and ETR2 of four percent, and 13 percent respectively; and 1987 estimates of 12 percent and 33 percent for ETRl and ETR2. This represents an increase greater than 250 percent from 1986 to 1987. On average, all industry groups experienced growth over the period. Profit rates remained the same for the airline and nondurable manufacturing groups while the retail industry experienced a slight decline and durable manufacturers experienced a slight increase. Gross accounting profits increased for all industry groups with the airline industry reporting the largest increase.

130

QUARTERLY

REVIEW OF ECONOMICS

AND FINANCE

Table 2. SUMMARY STATISTICS FOR CORPORATE RATIO, GROWTH AND AGE FOR TOTAL SAMPLE Variable

Std. Dev.

Mean

Median

ETRS, PROFIT, DEBT Minimum

Maximum

Panel A: Total Sam/& I986 N-984 DEBT1 # DEBT2 # AGE# [email protected] PROFITR # [email protected] El-RI **# JiX’R2 **#

0.27 0.46 0.38 94.92 0.13 152.30 0.06 0.29

0.19

0.24

0.00

0.54

0.38

0.00

0.15

0.38

0.00

652.84

17.37

-7643.80

0.10

0.24 585.00

-1.00 -2200.00

14.96

0.11 0.28

1.60 12.00 1.00 12952.60 1.00 8805.00 1.oo 1.00

0.05 0.38

-0.90

-1.00

Panel I% TolalSampk 1987 N-984 DEBT1 # DEBT2 # AGE # [email protected] PROFITR + [email protected] El’Rl **# El-R2 **#

0.27 0.50 0.39 125.71 0.13 194.89 0.07 0.31

0.19 1.10 0.16 503.41 0.22 637.12 0.12 0.27

0.24

0.00

0.38 0.39 24.49 0.10 19.20 0.06 0.38

0.00

1.10 32.30 0.90 6626.70 1.00 9552.00 1.00 1.oo

0.00

-4151.40 -1.00 447.40 -1.00 -1.00

LEGEND DEBT1

= Total Debt

DEBT2

= Long Term Debt /Long Term Debt plus Stockholders

AGE GROWTH PROFlTR PROFITG ETRl El’R2

= Accumulated

minus

CAsh and Short Term Investments/Total

Depreciation/Total

Assets.

Equity.

Plant Assets

= Change in Net Sales over previous 5 years. = Operating Income after Depreciation/Total = Financial Statement Accounting

Current Assets plus Net Plant Assets.

Income.

= Income Tax Expense minus Deferred Tax Liability/Operating

Cashflows.

= Income Tax Expense minus Deferred Tax Expense/PROFlTG(Deferred Expense/Corporate Marginal Tax Rate).

Passage of the AMT was grounded

in Congress’

perception

Tax

that the relation

between profits and tax payments, at least for many large firms, was weak prior to 1986. Given this assumption, we expect a weak positive relation between profit and ETR measures prior to implementation of the AMT. The results provided in Table 5 Panel A indicate that for 1986, across all firms in the sample, the strength of the relation between profits and ETRs depends on which measures are considered. The correlation of .47 between gross profits (PROFITG) and ETRl is positive and significant. The correlation of .32 between PROFITG and ETR2 is also positive and significant. The correlations between profit rate (PROFITR)

and ETRI

and ETR2 are positive and significant

at .58 and

THE EFFECT OF

THE ALTERNATIVE MINIMUM

TAX

Table 3. SUMMAR Y STATISTICS FOR CORPORATE EXR!3, PROFlT, DEBT RATIO, GROWTH, AND AGE BY LED INDUSTRY GROUPS 1986 Variable

Std. Dev.

Mean

Median

Minimum

MaximUll

0.59

A Relail N- 75 DERTI #

0.21

0.13

0.19

0.00

DEBT2 #

0.39

0.22

0.37

0.00

1.59

AGE #

0.30

0.13

0.30

0.05

0.77

188.77

392.26

44.67

-123.03

1892.80

0.11

0.08

0.10

-0.30

0.36

117.10

227.73

21.28

-30.75

1028.00

[email protected] PROFITR # [email protected] ETRl**#

0.05

0.07

0.04

-0.17

0.24

ETR2 **#I

0.41

0.21

0.45

-0.72

1.00

Ii

1hrd.k manujncluring N-295

DEBT1 #

0.26

0.18

0.23

0.00

1.09

DEBT2 #

0.41

0.45

0.32

0.00

5.35

0.43

0.14

0.44

94.85

380.57

7.40

0.07

0.10

0.09

-0.55

0.40

90.75

371.76

7.07

-513.21

3689.00

AGE# [email protected] PROFlTR # [email protected]

0.02 -386.08

0.77 4382.00

ETRI **#

0.04

0.11

0.04

-0.52

0.47

ETR2 l*#

0.29

0.31

0.39

-1.00

1.00

DEBT1 #

0.24

0.16

0.23

0.00

DEBT2 #

0.47

0.80

0.35

0.00

C. Non-lhrrtble manujrrtluringN-248

AGE # [email protected] PROFlTR #

0.40

0.12

0.41

34.60

1123.20

19.88

0.04 -7643.79

1.08 12.02 0.75 12952.20

0.13

0.09

0.13

-0.24

0.68

310.05

9609.02

46.03

-2200.00

8805.00

l!TRl **#

0.08

0.13

0.08

-0.87

0.72

lZR2

0.34

0.27

0.42

-1.00

1.00

PROFITGQ **#

Il.

Airline lmnsporlalion

N- 9

DEBT1 #

0.23

0.12

0.26

0.06

0.36

DEBT2 #

0.59

0.23

0.41

0.27

0.99

AGE #

0.33

0.12

0.31

0.18

232.63

345.65

144.94

GROW’[email protected] PROFITR #

-151.66

0.53 1042.94

0.05

0.09

0.08

-0.16

0.15

60.55

224.96

65.45

-465.30

311.89

GTRl **#

0.04

0.07

0.02

-0.02

0.24

ETR2 **#

0.13

0.19

0.04

0.00

0.47

[email protected]

131

132

QUARTERLY REVIEW OF ECONOMICS AND FINANCE

Table 4. SUMMAR Y STATISTICS FOR CORPORATE EI-RS, PROFIT, DEBT RATIO, GROWTH, AND AGE BY LED INDUSTRY GROUPS 1987 Variable

Mean

Std. Dev.

Median

Minimunl

MaXitllUln

A. Relail N-75 DEBT1 #

0.24

0.15

0.21

0.03

DEBT2 #

0.42

0.25

0.38

0.00

1.80

AGE #

0.31

0.13

0.30

0.07

0.80

233.49

461.76

56.59

-453.31

0.10

0.08

0.10

-0.26

0.30

129.83

253.85

23.22

-14.87

1171.00

[email protected] PROFlTR # [email protected]

0.70

2516.60

El’Rl **#

0.05

0.07

0.05

-0.17

0.30

El’R2 **#

0.40

0.18

0.42

a.09

1.00

DEBT1 #

0.26

0.18

0.23

0.00

1.10

DEBT2 #

0.39

0.33

0.32

0.00

2.80

B. IIurable manu/ncluring N-295

AGE # [email protected] PROFITR # [email protected]

0.44

0.14

0.45

148.61

550.86

13.42

0.08

0.11

0.09

XI.82

0.30

128.01

392.55

9.19

-275.62

3207.00

0.00 -208.18

0.80 6626.70

ETRl **#

0.05

0.10

0.05

-0.54

0.40

ETR2 **#

0.33

0.27

0.39

-1.00

1.00

Non-Durdlc manufncturingN-248 DEBT1 #

0.25

0.20

0.22

0.00

1.10

DEBT2 #

0.57

2.10

0.36

0.00

32.30

AGE # [email protected] PROFlTR # [email protected]

0.41

0.10

0.41

65.80

546.50

32.45

0.10 -4151.40

0.80 2635.50

0.13

0.10

0.13

-0.20

0.60

399.5 1

1051.30

71.55

-104.20

9552.00

ETRl **#

0.09

0.10

0.08

-0.20

0.50

ETR2 **#

0.38

0.20

0.41

-1.00

1.oo

DEBT1 #

0.23

0.11

0.22

0.11

0.40

DEBT2 #

0.58

0.30

0.52

0.33

1.30

AGE#

0.32

0.10

0.33

0.14

0.50

350.65

440.52

250.77

-24.68

1417.10

0.05

0.05

0.06

-0.05

0.10

160.12

232.62

168.73

-264.35

447.20

ETRl l*#

0.12

0.11

0.11

0.00

0.30

El.R2 **#

0.33

0.23

0.34

0.00

0.70

D. Airline lransportahn N-9

[email protected] PROFITR # [email protected]

N&-x Set Table 2.

THE EFFECT OF THE ALTXRNATIVE MINIMUM TAX

T&125. SPEARMAN CORRELATIONS: Variable

DEBT1 DEBT2

AGE

133

TOTAL SAMPLE

GROWTH PROFlTR PROFlTG

ETRl

ETR2

PanelA I986 N - 984 DEBT1

1.00

DEBT2

0.72

-0.20

-0.12

-0.14

0.47

-0.21

-0.14

l.ocl

-0.11

0.01

0.02

-0.07

-0.11

-0.07

-0.29

-0.19

-0.14

-0.05

-0.05

0.37

0.55

0.22

0.17

1.00

0.55

0.58

0.44

1.00

0.47

0.32

1.00

0.48

AGE

1.00

CROWl’H

1.00

PROFlTR PROFITG El-RI

1.00

E-l-R2 PanelB 1987NDEBT1 DEBT2 AGE GROWTH PROFTlX PROFlTG ETRl ETRP

1.00

984

0.71

-0.22

-0.12

-0.10

-0.29

-0.21

-0.17

1.00

-0.29

-0.01

0.01

-0.13

-0.12

-0.10

-0.24

-0.13

-0.10

-0.08

-0.03

0.38

0.59

0.24

0.15

1.00

0.53

0.56

0.33

1.00

0.53

0.29

1.00

0.35

1.00

1.00

1.00

.44, respectively. Contrary to Congress’ belief that the relation was weak, it appears that firms that were profitable tended to have higher ETRs prior to implementation of the AMT. Table 5 Panel B provides the correlation between our profit and ETR measures after the implementation of the AMT in 1987. The results are mixed, with the correlation between PROFITG and ETRl increasing to .53, while at the same time the correlation between PROF’ITR and ETR2 falls to .33. Table 5 Panels A and B also provide correlations between the other firm characteristics suggested by LED. In 1986, AGE, DEBTl, and DEBT2 are negatively correlated with both ETR measures. GROWTH is positively correlated with both ETR measures. For 1987 the correlations between ETR measures and these firm characteristics remain essentially the same. In both years the correlations between ETRs and these factors are weak. The results in Table 5 suggest that prior to implementation of the AMT firms that were newer, more profitable, had less debt, and were growing faster paid higher taxes. After the AMT was implemented there appears to be no substantial change in the set of firms paying higher taxes. Thus, in this sample the AMT appears to have had little effect on the distribution of corporate taxes. This appears to conflict with the Citizens for Tax Justice (1988) appraisal that the new law had a significant effect on tax payments of high profit/low tax firms.

134

QUARTERLY REVIEW OF ECONOMICS AND FINANCE

Table 6.

SPEARMAN CORRELATIONS:

Variable

DEBT1

DEBT2

AGE

FIRMS IN LED INDUSTRIES

GROWTH PROFlTR PROFlTG

ETRl

ETR2

PanelA 1986 N - 627 DEBT1

1.00

DEBT2

0.78

-0.17

-0.11

-0.21

-0.31

-0.22

1.oo

-0.30

-0.01

-0.07

-0.13

-0.13

-0.04

-0.36

-0.23

-0.15

-0.10

-0.15

0.40

0.52

0.21

0.13

1.00

0.56

0.63

0.43

1.00

0.47

0.24

1.00

0.39

AGE

1.oo

GROWTH

1.00

PROFlTR PROFlTG JCl’Rl El’Rf

-0.10

1.00 PanelB 1987N-627

DEBT1

1.00

DEBT2

0.75

-0.20

-0.09

-0.19

-0.31

-0.25

-0.14

1.00

-0.30

-0.30

-0.08

-0.11

-0.13

-0.06

-0.30

-0.18

-0.11

-0.12

-0.14

0.45

0.60

0.26

0.11

1.00

0.58

0.66

0.28

1.00

0.57

0.20

1.00

0.27

AGE

1.oo

GROWTH

1.00

PROFlTR PROFlTG ETRl

1.00

ETR2

Variable

GROUP PROFITR PROFITG

ETRl

ETR2

Panel C 1986 N - 62 7 GROUP

1.00

PROFlTR

0.215

0.169

0.158

1.oo

0.558

0.633

0.425

1.00

0.469

0.241

1.00

0.395

PROFITG FTRl

-0.022

1.00

GTR2 PanelDl987N-627 GROUP PROFlTR PROFlTG GTRl ETR2

1.oo

0.175

0.240

0.202

0.027

1.00

0.581

0.655

0.277

1.00

0.572

0.199

1.00

0.269 1.00

Table 6, Panels A and B, provide the correlations teristics

and ETR

measures

excluding

between

the firm charac-

firms that are not in the LED industry

groups. Note that the results are very similar to those presented in Table 4. However, Panels C and D of Table 6 provide correlations between ETRs and firm characteristics and a variable representing LED’s predictions about which industries were most and least likely to pay the AMT. For each firm the GROUP

variable

was assigned the rank value of its industry group based on LEDs ranking of most and least likely to pay the AMT. The GROUP

variable contains values that range

from one to four where one is the industry group least likely to pay the AMT and four is the industry group most likely to pay the AMT. In Panel C (prior to implementation of the AMT) the correlation

between

GROUP and ETR2 is negative (although weak). This suggests that firms least likely to pay the AMT were paying higher taxes prior to the implementation of the AMT. The correlation between ETRl and GROUP is positive. The correlations between both measures of profit and both ETR measures are positive. Panel D indicates that after the AMT was implemented the correlations between ETR measures and GROUP became stronger for ETRl and changed to positive for ETR2. Thus, both measures of tax burden indicate an increase in the taxes paid by firms most likely to pay the AMT. Results at the industry level indicate that the Tax Reform Act of 1986 increased the ETRs of industry groups that LED predicted would be most likely to pay the AMT. However, the change in the relation between profits and ETRs is dependent on the measure of profits and the measure of ETR. Consequently, we include an additional set of analyses that looks at the relation between individual firm characteristics of age growth and profit, debt, and changes in ETRs from 1986 to 1987 to determine firm’s ETR.

the extent to which these factors contributed

to changes in the

Based on the results of LED’s analysis, firms that are most likely to pay the AMT are newer firms with high debt, high growth rates, and low profit rates. The effect of these factors on changes in ETRs from 1986 to 1987 are tested by regressing the change in ETRs from 1986 to 1987 on the change in debt position, profitability, growth, and age of the firm. Using changes for the AGE and GROWTH variables requires a different interpretation of the coefficients associated with these variables. Because AGE is measured by accumulated depreciation divided by total plant assets, the change in this variable from 1986 to 1987 represents the depreciation expense for 1987. LED suggest that firms with higher levels of depreciation

are more likely to pay

the AMT because the rules for depreciation expense are more restrictive for AMT purposes than for regular tax purposes. These restrictions will increase the AMT tax base and will tend to increase the amount of tax paid under AMT rules faster than taxes paid under regular tax rules. The change in the GROWTH variable can be interpreted as the rate at which the firm is growing. LED suggest that firms

136

QUARTERLY REVIEW OF ECONOMICS AND FINANCE

Table 7. REGRESSION RESULTS BEIWEEN CHANGES IN ETRAND CHANGES IN LED FIRM CHARACTERISTICS 1986 TO 1987 Panel A [email protected] Variables, Change in ETRI AND ETRL-Total Sample

Variable

Dependent

Variable

AETRl

Dependent

Variable

AETRl

AETR2

AETR2

Variable

INTERCEPT STD ERROR T-STATISTIC

0.0116 0.0050 2.289*

0.0229 0.0098 2.329*

INTERCEPT STD ERROR T-STATISTIC

0.0081 0.0049 1.655

0.0187 0.0097 1.917

ADEBTl STD ERROR T-STATISTIC

-0.1403 0.0431 -3.255**

-0.2221 0.0836 -2.656**

DDEBT2 STD ERROR T-STATISTIC

0.0541 0.0069 7.822**

0.0533 0.0137 3.893**

AAGE STD ERROR T-STATISTIC

-0.0946 0.0757 -1.250

0.0509 0.1469 0.346

AAGE STD ERROR T-STATISTIC

-0.0462 0.0732 -0.631

0.1180 0.1451 0.813

AGROWI-H STD ERROR T-STATISTIC

0.000003 0.000009 0.323

0.000008 0.00001 0.441

AGROWH STD ERROR T-STATISTIC

0.000003 0.000009 0.369

0.000008 0.00001 0.472

APROFITR+ STD ERROR T-STATISTIC

0.1058 0.0349 3.029**

0.2251 0.0677 3.322**

APROFITR+ STD ERROR T-STATISTIC

0.1288 0.0338 3.81l**

0.2566 0.0669 3.833**

ADJ R2

0.019

0.019

ADJ R*

0.067

0.027

Panel B Dependenl Variabb ETRI AND ETRZ-LED Industry GVU/IS

Dependent

Variable

Dependent

Variable

Variable

ETRl

ETR2

Variable

ETRl

ETRP

INTERCEPT STD ERROR T-STATISTIC

0.0107 0.0049 2.170*

0.0322 0.0126 2.549*

INTERCEPT STD ERROR T-STATISTIC

0.0074 0.0044 1.658

0.0288 0.0124 2.306*

ADEBTl STD ERROR T-STATISTIC

-0.1081 0.0460 -2.349*

-0.1525 0.1180 -1.293

DDEBT2 STD ERROR T-STATISTIC

0.0601 0.0052 1.1465

0.0532 0.0146 3.648**

AAGE STD ERROR T-STATISTIC

a.0252 0.0768 -0.329

-0.2142 0.1968 -1.088

AAGE STD ERROR TSTATISTIC

0.0104 0.0695 0.150

4.1705 0.1934 -0.881

AGROWI-H STD ERROR T-STATISTIC

0.000001 0.000007 0.177

0.000007 0.00002 0.375

AGROWTH STD ERROR T-STATISTIC

0.000001 0.000007 0.240

0.000007 0.00002 0.400

APROFITR+ STD ERROR T-STATISTIC

0.2880 0.0621 4.638**

0.7789 0.1591 4.894**

APROFITR+ STD ERROR T-STATISTIC

0.3415 0.0554 6.158

0.8422 0.1543 5.459**

A?] R*

0.048

0.040

AQj R*

0.207

0.058

Nda seeTable2. + Similar rush

are found using I’ROFITC in place OlPROFUR. Achangc from IWG to 1937 Significantatlas than .05. bvwail test. l * Signiliant al lar than LW5.twuail test

invariable

l

THE EFFECT OF THE ALTJZNATIVE

MINIMUM TAX

137

that are growing will be more likely to pay the AMT. Thus, we would also expect that firms growing at faster rates would also be more susceptible to payment of the AMT. Estimates of the regression coefficients and their t-statistics are provided in Table 7 Panels A and B. The results in Panel A suggest that, for the total sample, the change in ETRs is associated with the change in PROFIT and the change in DEBT. However, the association

between

change

in DEBT

and change

in ETRs

depends

on the

measure of DEBT considered. When total debt to total assets is used (DEBTl) the coefficient is negative and significant. This suggests that increases in total debt generally increased the debt based tax shields and mitigated any positive changes in ETRs across firms. This result is consistent across ETR measures. On the other hand, when only the change in long term debt to equity (DEBT2)

is considered

the coefficient for DEBT is positive and significant. This suggests that as long term debt increased ETRs increased. This result is consistent with LEDs predictions about the effect of debt on payment of the AMT. However, referring to Table 2, note that total debt remained constant from 1986 to 1987, suggesting that short term debt based tax shields decreased and offset any increased likelihood of paying the AMT that was associated with increases in long term debt. The coefficient for the PROFIT measure is positive and significant. This suggests that as firms increased profits from 1986 to 1987 they paid more taxes. Whether the increased taxes paid is associated with the AMT is unknown because higher profits would have also caused higher

regular

taxes. The coefficients

for

GROWTH and AGE are not significant. This suggests that the amount of asset based tax deductions and the rate at which the firm was growing had little affect on the change in the firms’ ETR The results suggest that, firms with higher profits that were increasing their long term debt levels appear to pay higher taxes in 1987. Table 7 Panel B provides the regression results when firms not in LED’s industry groups are excluded from the sample. The results are consistent with Panel A Profitable firms increasing their long term debt experienced higher taxes in 1987. The analysis presented

in Table 7 is limited because the ability of the model to

explain the variation in ETR changes is poor. The low coefficient of determination indicates that LEDs factors are not the only determinants of the firm’s effective tax rate. Consequently, additional work is necessary to determine what factors, in addition, to those described by LED, would likely cause a firm to pay the AMT.

SUMMARY This study is an empirical analysis of the incidence of the changes in the tax burden of corporations after implementation of the Tax Reform Act of 1986. Because the Act contained a controversial addition to the Alternative Minimum

138

QUARTERLY REVIEW OF ECONOMICS AND FINANCE

Tax, the Book Income Adjustment, the FASB and AICPA were concerned with its affect on the reporting of accounting information and the possible inequities associated with the new law. The results suggest that, on an industry basis, the industry groups most likely to pay the AMT experienced increased tax burdens as measured by ETRs. However, it also appears that firms not likely to pay the AMT also experienced a higher tax burden. At the individual firm level, only PROFIT change in a firm’s ETR and only the coefficient

and DEBT appear to affect the for long term DEBT is consistent

with LED’s prediction that firms with higher debt would pay higher taxes. GROWTH and AGE appear to have no affect on the changes in tax burdens from 1986 to 1987 but are weakly associated with the level of firms’ tax burdens. Overall, the results suggest some consistency between expected payment of the AMT and low ETRs prior to implementation of the AMT. However, because the model estimated in Table 7 explains little of the variation in ETRchanges, and firms not likely to pay the AhIT also experienced higher tax burdens, additional analyses to determine the consequences of the Tax Reform Act of 1986 should be conducted. Questions that are as yet unanswered include: (1) what factors in addition to those suggested by LED affect payment of the AMT, (2) were firms able to escape the effects of the BIA through changes in their reported accounting income; and finally, (3) were the additional compliance costs imposed by the new law in excess of revenues generated. Because the relation between likelihood of AMT payment and low ETRs appears to be weak, we suggest that Congress should question the corrective benefits of the MT (shifting the corporate tax burden) in relation to the negative effects associated with compliance costs, financial statement reporting, and inequities associated with application of the AMT. Alternative tax reform measures, for example, credit limits, increased rates, limitations on cost recovery methods, or a larger tax base, are likely candidates for consideration.

NOTES 1. For the 1990 tax year the BIA was replaced with the adjusted current earnings ACE provision. 2. A corporation’s applicable financial statement means any income statement from a priority list of statements. In that list SEC statements are given top priority. Net book income must take into account all items of revenue, expense, and gain and loss attributable to the taxpayers’ normal method of accounting. It includes any extraordinary items, income or loss from discontinued changes.

operations,

and cumulativeadjustments

resultingfrom

accounting

3. Theexemption amountwasreduced by25percentoftheexcessofAhfl”Iover$150,000. Thus, firms with ANTI greater than $310,000 had an exemption amount of $0. 4.

This definition

is the same as that described by LED.

method

THE EFFECT OF THE ALTERNATIVE 5.

The growth measure

for 1986 is the firm’s change

MINIMUM

in sales revenue

TAX

139

from 1981 to 1986.

The 1987 measure is the change in sales revenue from 1982 to 1987. 6. The values for ETRI, ETR2 and PROFlTR were truncated at -1 .OOand 1 .OOto remove the effects of outliers. mean.

In all cases the oudiers

were more than 20 standard

deviations

from the

REFERENCES 111

PI PI [41 [51 WI 171 181 WI 1101 [Ill 1121 USI P41

D51

Ml [171

Citizens for Tax Justice. 1985. Corporate Taxpayers & Corporate Free L.oaders, Robert McIntyre Ed. Citizns fat- Tax Justice. 1988. The Gnporate Tax Gmeba& Gnporale Income Taxes After Tax Refm Robert McIntyre Ed. Dworin, Lowell. 1987a. “Impact of the Corporate Alternative Minimum Tax.” Pp. 253-278 in Compendium of Tax Reseanh 1987, Office of Tax Analysis, Department of Treasury, Washington D.C. Dworin, Lowell. 1987b. “Impact of the Corporate Alternative Minimum Tax.” National Tnr Journal (September): 505613. Fullerton, Don. 1984. Which Effective Tax Rate?” National TaxJournal (March): 23-42. -. 1982a. Why Do Varying Effective Corporate Tax Rates Matter.” Tax Notes, 1980 Effective Tax Rate Supplement (March). -. 1982b. “Why Do Varying Effective Corporate Tax Rates Matter.” Tax Notes, 1981 Supplement (August). Harter, GA 1986. “How Tax Reform Would Affect Companies With Different Growth and Profitability Characteristics.” Tax Notes (April 21): 297-301. Lucke, Robert B., Mark R. Eisenach, and Larry L. Dildine. 1986. “The Senate Alternative Minimum Tax: Does It Snare only the Tax ‘Abuser’?” Tax Notes (August 18): 681-690. Omer, Thomas C., Karen H. Molloy, and David Ziebart. 1991. “Measurement of Effective CotpotateTaxRatesUsingFinancialStatementInformation.” TheJoumaloftheAmerkan Taxation Association (Spring): 57-72. Pechman, Joseph. 1977. Federal Tar Policy. The Brookings Institute,Washington D.C. Porcano, Thomas M. 1986. “Corporate Tax Rates: Progressive, Proportional, or Regressive.” Journal of the American Taxation Association (Spring): 17-31. Shevlin, Terry. 1987. “Taxes and Off-BalanceSheet Financing: Research and Develop ment Limited Partnerships.“AcwvntingRevieu (July): 480-509. Stickney, Clyde P. and VE. McGee. 1982. “Effective Corporate Tax Rates: The Effect of Size, Capital Intensity, Leverage, and Other Factors.” Joumal of Accounting and Public Portly (March): 125-152. U.S. Congress, Joint Committeeon Taxation. 1987. GeneralExplanation ofthe TaxRefm Adof 1986, [H.R. 3838,99th Congress; Public Law99-5141, CommerceClearingHouse, Chicago, (May). Weiss, R.L. 1979. “Effective Corporation Income Tax Rates.” National Tax Journal (March) : 380-389. Zimmerman, Jerold L. 1983. “Taxes and Firm Size.” Journal of Accounting and Economics (August): 119-150.