Determinants of the variability in corporate effective tax rates and tax reform: Evidence from...

Available online at www.sciencedirect.com

Journal of Accounting and Public Policy 26 (2007) 689–704

www.elsevier.com/locate/jaccpubpol

Determinants of the variability in corporateeffective tax rates and tax reform:

Evidence from Australia

Grant Richardson a,*, Roman Lanis b

a Department of Accountancy, Faculty of Business, City University of Hong Kong,

83 Tat Chee Avenue, Kowloon Tong, Hong Kong, People’s Republic of Chinab School of Accounting, Faculty of Business, University of Technology – Sydney,

Cnr of Quay Street and Ultimo Road, Haymarket, Sydney, NSW 2000, Australia

Abstract

This study examines the determinants of the variability in corporate effective taxrates in Australia spanning the Ralph Review of Business Taxation reform. Our resultsindicate that corporate effective tax rates are associated with several major firm-specificcharacteristics, including firm size, capital structure (leverage) and asset mix (capitalintensity, inventory intensity and R&D intensity). While the Ralph Review tax reformhad a significant impact on many of these associations, corporate effective tax rates con-tinue to be associated with firm size, capital structure and asset mix after the tax reform.� 2007 Elsevier Inc. All rights reserved.

Keywords: Corporate effective tax rates; Ralph Review tax reform; Size; Capital structure;Asset mix

0278-4254/$ - see front matter � 2007 Elsevier Inc. All rights reserved.doi:10.1016/j.jaccpubpol.2007.10.003

* Corresponding author.Tel.: +86 2788 7923; fax: +86 2788 7944.E-mail address: [email protected] (G. Richardson).

mailto:[email protected]

690 G. Richardson, R. Lanis / Journal of Accounting and Public Policy 26 (2007) 689–704

1. Introduction

Corporate effective tax rates (ETRs)1 are often used by policy-makers andinterest groups as a tool to make inferences about corporate tax systemsbecause they provide a convenient summary statistic of the cumulative effectof various tax incentive and corporate tax rate changes (Kern and Morris,1992, p. 81; Gupta and Newberry, 1997, p. 1). Evidence in the US has shownthat ETRs vary across firms and over time, which suggested that the corporatetax system was inequitable and so was used as a major justification for taxreform (Shevlin and Porter, 1992, p. 60). However, there is a lack of researchon ETRs and tax reform, especially in countries outside the US.

A potential reason for the lack of research in this area is that corporate taxreform is infrequent; hence the opportunity for undertaking such research islimited. The Ralph Review of Business Taxation represented a major eventin corporate tax reform in Australia (Cooper et al., 2002, p. 20; Gilderset al., 2004, p. 16). The Ralph Review was given a mandate by the AustralianGovernment to broadly assess the adequacy of the country’s business incometax policy (Ralph, 1999, p. 10). It submitted its proposals to the AustralianGovernment on July 30, 1999. Several of the Ralph Review’s key proposalscould affect the characteristics normally associated with ETRs. Accelerateddepreciation was recommended for removal. Moreover, a phased-in reductionof the corporate tax rate was also suggested. The Australian Governmentaccepted these key proposals, and they were codified in the Income Tax Assess-ment Act (1997), with application from the 1999–2000 tax year.

The Ralph Review tax reform provides a unique opportunity to treat thisimportant tax policy event as a natural experiment for examining the determi-nants of the variability in corporate ETRs in Australia spanning the taxreform. In so doing, our study adds to the sparse literature about ETRs andtax reform. Moreover, our results provide some further insights into ETRs thatshould be useful to policy-makers. We investigate the impact of firm size, cap-ital structure (leverage) and asset mix (capital intensity, inventory intensity andR&D intensity) on ETRs covering the Ralph Review. Formal tests of theimpact of the tax reform on these associations and whether ETRs are associ-ated with these characteristics after the tax reform are also considered.

1 There are many different types of corporate ETRs in the literature (see, e.g. Plesko, 2003, p.206). Distinctions are made between average ETRs and marginal ETRs. Average ETRs are definedas tax liability divided by income, while marginal ETRs are defined as the change in tax for a givenchange in income. The suitability of each type depends on a study’s research question. AverageETRs are appropriate to examine the distribution of tax burdens across firms or industries, whilemarginal ETRs are suitable to analyze the incentives of new investments (Gupta and Newberry,1997, p. 1). This study uses the term ETRs to denote average effective tax rates, and two differentmeasures are used to improve the robustness of our results: income tax expense divided by bookincome and income tax expense divided by operating cash flows.

G. Richardson, R. Lanis / Journal of Accounting and Public Policy 26 (2007) 689–704 691

We present evidence that shows a significant negative association betweenETRs and firm size. We also find that ETRs have a significant negative asso-ciation with capital structure in terms of leverage. A significant negative (posi-tive) association is also observed between ETRs and asset mix for capitalintensity and R&D intensity (inventory intensity). While the Ralph Reviewhad a major impact on many of these associations, ETRs are still associatedwith firm size, capital structure and asset mix following the tax reform.

The remainder of the paper is organized as follows. Section 2 reviews the majordeterminants of ETRs and develops hypotheses. Section 3 describes the researchdesign. Section 4 reports and analyzes the results. Finally, Section 5 concludes.

2. Determinants of ETRs and hypotheses

2.1. ETRs and firm size

There are two competing views about the association between ETRs andfirm size: the political cost theory and the political power theory. Specifically,under the political cost theory, the higher visibility of larger and more prosper-ous firms causes them to become victims of greater regulatory actions by gov-ernment and wealth transfers (Watts and Zimmerman, 1986, p. 235). As taxesare one part of the total political costs borne by firms, this theory claims thatlarger firms have higher ETRs (Zimmerman, 1983, p. 119). The alternativeview under the political power theory is that larger firms have lower ETRsbecause they have substantial resources available to them to manipulate thepolitical process in their favor, engage in tax-planning and organize their activ-ities to achieve optimal tax savings (Siegfried, 1972, pp. 32–36).

Studies of US firms on the association between ETRs and firm size have pro-duced conflicting results. While Zimmerman (1983) finds a positive associationbetween ETRs and firm size, Porcano (1986) observes a negative associationbetween these variables. Finally, based on empirical evidence, Gupta and New-berry (1997, p. 28) assert that the inconsistent results suggest that firm-size effectscould be sample-specific and not likely to exist over time in firms with longerhistories.

Australian research on ETRs and firm size is almost non-existent. Tran(1997, p. 529), however, observes a negative association between ETRs andfirm size. In another study undertaken to identify the causes of this negativeassociation, Tran (1998, p. 282) finds that larger firms benefited more fromtax-planning (tax incentives) than smaller firms. We therefore expect a negativeassociation between ETRs and firm size in our study. Finally, we also test thevalidity of Gupta and Newberry’s (1997, p. 28) assertion that firm-size effectsmight be sample-specific and unlikely to exist over time in firms with longer his-tories, using Australian data.


2.2. ETRs and firms’ financing and investment decisions

Firms’ financing decisions could impact on ETRs because tax statutes nor-mally allow differential tax treatment to the capital structure decisions of firms(Gupta and Newberry, 1997, p. 7). Consider, for example, the situation inwhich a firm relies more heavily on debt financing rather than equity financingto support its business operations. Given that interest expenditure is taxdeductible while dividends are not, firms with higher leverage are expected tohave lower ETRs. Research by Stickney and McGee (1982) and Gupta andNewberry (1997) finds a negative association between ETRs and leverage.

Firms’ investment decisions might also impact on ETRs. As tax statutesusually permit taxpayers to write-off the cost of depreciable assets over periodsshorter than their economic lives; firms that are more capital-intensive areexpected to have lower ETRs (Stickney and McGee, 1982, p. 142). To theextent that inventory intensity is a substitute for capital intensity, inventory-intensive firms should possess higher ETRs (Zimmerman, 1983, p. 130). Guptaand Newberry (1997) provide evidence that firms with a larger proportion offixed assets have lower ETRs due to tax incentives, while firms with a greaterproportion of inventory have higher ETRs. Finally, R&D expenditure fur-nishes an investment tax shield for R&D-intensive firms. This suggests a neg-ative association with ETRs (Gupta and Newberry, 1997, p. 15).

Following from the above discussion, we hypothesize that:

H1: ETRs are negatively associated with firm size.H2: ETRs are negatively associated with firm leverage.H3: ETRs are negatively associated with firm capital intensity.H4: ETRs are positively associated with firm inventory intensity.H5: ETRs are negatively associated with firm R&D intensity.

2.3. ETRs and tax reform

We also investigate whether the Ralph Review tax reform had an effect onETRs, and if it impacted on the associations between ETRs and variablesreflecting the outcomes of firms’ financing and investment decisions. The keyRalph Review tax-reform proposals are summarized in Table 1, along withtheir estimated corporate tax revenue impacts over the period 1999–2005.

Table 1 shows that several of the tax reforms were designed to increase cor-porate tax revenue, so they should be considered base-broadening tax reforms.The replacement of accelerated depreciation with an effective life depreciationregime was the most important base-broadening tax reform. High-level taxdesign reforms, such as the removal of the 13-month prepayment rule that pre-viously allowed firms to claim tax deductions in advance, are the next most

Table 1Estimated effects of the Ralph Review tax reform on corporate tax revenues over the period 1999–2005 (AUD$Billions)

AUD$B AUD$B

Tax reforms increasing corporate tax revenue

Removal of accelerated depreciation 10.87High-level tax design reforms .97Changes to the taxation of investments .89Integrity measures .53Capital gains tax reforms .23 13.49

Tax reforms decreasing corporate tax revenue

Reduction in corporate tax rates (14.45)Changes to the taxation of income from entities (2.02)Small business measures (2.01) (18.48)Net corporate tax revenue impact (4.99)

Source: Ralph (1999, p. 698).


important base-broadening tax reform. Finally, changes to the taxation ofinvestments, such as the removal of taxation on the profit on sale of deprecia-ble assets, integrity measures that restricted the use of unrealized losses forfirms and capital gains tax reforms which removed asset indexation were alsointended to be base-broadening.

Table 1 also illustrates that several of the tax reforms were designed todecrease corporate tax revenue. The phased-in reduction of the corporate taxrate for the 2000–2001 tax year (from 36% to 34%), and for the 2001–2002tax year and thereafter (from 34% to 30%) was the most important tax reformmeasure designed to decrease corporate tax revenue. The remaining tax reformmeasures were estimated to have a negative impact on tax revenue. Theseinclude changes to the taxation of income from business entities, such asrefunding dividend imputation credits, and small business measures, such asallowing cash accounting.

On the basis of the tax revenue estimates summarized in Table 1, the overallimpact of the Ralph Review tax reform is expected to cause a net reduction in cor-porate tax revenue for the Australian Government of AUD$4.99 billion. Hence,the tax reform is expected to have a significant negative association with ETRs.

It is also possible that the Ralph Review tax reform impacted on the asso-ciations between ETRs and variables reflecting the outcomes of firms’ financingand investment decisions. In terms of financing, where firms rely more heavilyon debt financing in their capital structure, this is expected to increase ETRsafter the tax reform, as the reduction in the corporate tax rate decreases thetax savings on interest. For investment, where firms are capital-intensive, thisis also expected to increase ETRs after the tax reform, due to the removal ofaccelerated depreciation and a reduction in the corporate tax rate. To the


extent that inventory intensity is a substitute for capital intensity, inventory-intensive firms should have lower ETRs after the tax reform. Finally, the RalphReview proposed no change in tax policy for the R&D tax concession, so thetax reform is not expected to have an impact on the association between ETRsand R&D expenditure.

Following from the above discussion, we hypothesize that:

H6: ETRs are negatively associated with the Ralph Review tax reform.H7: The association between ETRs and firm leverage is positively impacted

upon by the Ralph Review tax reform.H8: The association between ETRs and firm capital intensity is positively

impacted upon by the Ralph Review tax reform.H9: The association between ETRs and firm inventory intensity is negatively

impacted upon by the Ralph Review tax reform.H10: The association between ETRs and firm R&D expenditure is not

impacted upon by the Ralph Review tax reform.

3. Research design

3.1. Sample and data

Our sample consists of a single panel of publicly-listed Australian firms col-lected from the Aspect Financial Database over the period 1997–2003. How-ever, the year 2000 was excluded because this is a transitional tax year interms of the Ralph Review proposals, and prior research (e.g. Dhaliwal andWang, 1992; Scholes et al., 1992; Guenther, 1994) shows that firms normallyrespond to tax legislation changes one year after tax legislation becomes oper-ative. The final sample consists of 92 firms (552 firm years) after excluding firmsthat fall into the following categories:

(a) Financial firms, since government regulation faced by these firms is likelyto affect their ETRs differently from other firms.

(b) Foreign firms, as these firms’ financing and investment decisions may beimpacted upon by resident country tax laws that differ from Australiantax laws.

(c) Firms with missing data and/or no activity firms.(d) Firms with negative income or tax refunds, since their ETRs are distorted

(e.g. Zimmerman, 1983; Omer et al., 1993).(e) Firms that have NOL carry-forwards because their ETRs are difficult to

interpret (e.g. Wang, 1991) and are not included in traditional ETRresearch (e.g. Wilkie and Limberg, 1993).

(f) Firms with ETRs exceeding one, since this can cause model estimationproblems (e.g. Stickney and McGee, 1982; Gupta and Newberry, 1997).

Table 2Sample reconciliation

1997–1999 and 2001–2003a

All firms in the Aspect Financial Database excluding financialinstitutions and foreign firms

1529

Less:

Firms with missing data and/or no activity (1347)Firms with negative income or tax refunds (56)Firms with NOL carry-forwards (13)Firms with ETRs exceeding one (21)Final sample (number of firms) 92Final sample (firm years) 552

a The year 2000 was excluded from the sample.


A summary of the sample reconciliation is presented in Table 2.

3.2. Dependent variable

The dependent variable is represented by ETRs. In conventional research,ETRs are measured based on information collected from financial statementsas tax liability divided by income. However, the appropriate definitions of boththe numerator and the denominator of this equation are open to debate (e.g.Shevlin and Porter, 1992; Wilkie and Limberg, 1993; Plesko, 2003).

The issue of which taxes to include in the numerator of the equation is rel-evant because any significant omission can bias the overall tax burdens offirms. Some researchers (e.g. Porcano, 1986; Gupta and Newberry, 1997) usethe income tax expense of firms and make no adjustment for deferred taxexpense. Others (e.g. Stickney and McGee, 1982; Omer et al., 1993) arguefor an adjustment to income tax expense by subtracting the deferred taxexpense portion. Income tax expense is used as the numerator of our equationwithout any deferred tax adjustment, since Australian firms are not required toreport deferred tax expense in their financial statements under AccountingStandard AASB 1020: Accounting for Income Tax (Tax-Effect Accounting).

The issue of how income should be measured in the denominator of theequation arises due to the difference between accounting (book) income andtaxable income. The choices include taxable income, book income and cashflow from operations. Taxable income should not be used if the purpose of astudy is to capture the impact of tax incentives on ETRs. If both the numerator(income tax expense) and the denominator (income) are after tax incentives,then any systematic variation in ETRs because of tax incentives will not bedetected (Gupta and Newberry, 1997, p. 12). We use book income as the pri-mary income measure in the denominator. Cash flow from operations is used


as an alternative measure, as this controls for systematic differences in account-ing method choices that are related to firm size (Zimmerman, 1983, p. 123).

In short, two different measures of ETRs are employed as the dependentvariable to improve the robustness of our results. The first (ETR1) is definedas income tax expense divided by book income. The second (ETR2) is definedas income tax expense divided by operating cash flows.

3.3. Independent variables

3.3.1. Firm-specific variables

Firm-specific variables are denoted by proxies for firm size, capital structure(financing) and asset mix (investing). Firm size (SIZE) is measured as the naturallogarithm of total assets (at book value). Financial leverage (LEV) is included toproxy for firms’ capital structure, and is measured as the long-term debt dividedby total assets (both at book values). Three independent variables are included inthe study to proxy for firms’ asset mix: capital intensity (CINT), inventory inten-sity (INVINT) and R&D intensity (RDINT). Specifically, CINT is measured asthe net property, plant and equipment divided by total assets (both at book val-ues). INVINT is measured as inventory divided by total assets (both at book val-ues). Finally, RDINT is measured as R&D expenditure divided by net sales.

3.3.2. Control variablesFirms’ operations could also impact on ETRs. Wilkie (1988) finds that

ETRs are a function of the ratio of tax incentives to book income, where taxincentives (e.g. depreciation) are items that cause book income to differ fromtaxable income. To the extent that tax incentives are not proportionatelyrelated to book income, ETRs can change simply due to changes in bookincome. Hence, we use return on assets (ROA) to control for changes in firms’operating results. ROA is measured as pre-tax income divided by total assets.We expect ROA to have a positive sign because an increase in return on assetsleads to an increase in ETRs (Gupta and Newberry, 1997, p. 15).

Industry-sector (INSEC) dummy variables defined at the two-digit GlobalIndustry Classification Standard (GICS) code level are also included as controlvariables in our study, given that it is possible for ETRs to fluctuate across dif-ferent industry sectors (see, e.g. Omer et al., 1993; McIntyre and Nguyen, 2000;Derashid and Zhang, 2003). We incorporate nine INSEC dummy variables inour study: energy, materials, industries, consumer discretionary, consumer sta-ples, health care, information technology, telecommunications and utilities.2

No sign predictions are made for the INSEC dummies.

2 With utilities being the omitted sector in our regression model.


3.3.3. Tax reform variables

To formally test the Ralph Review’s impact on ETRs, a period dummy var-iable (TREF) is included in our study (coded 1 if the observation is for thepost-tax-reform period, 0 otherwise), along with interaction terms comprisingthe TREF dummy variable multiplied by each of the firm-specific variables andcontrol variables. The TREF coefficient provides a test of the mean shift inETRs after tax reform. The interaction term coefficients permit testing forslope shifts in each of the firm-specific and control variables after tax reformand therefore determine whether these associations changed after the taxreform. However, the firm-specific and control variable coefficients provide evi-dence on their associations with ETRs in the pre-tax-reform period, while thesum of these coefficients and the coefficients of their corresponding interactionterms with TREF provide evidence on whether these firm-specific and controlvariables were associated with ETRs after the tax reform.

3.4. Regression model

Our empirical analysis involves estimating the following regression model:

ETRit ¼ a0 þ b1SIZEit þ b2LEVit þ b3CINTit þ b4INVINTit

þ b5RDINTit þ b6ROAit þ b7�14INSECit þ b15TREFit

þ b16TREF � SIZEit þ b17TREF � LEVit þ b18TREF

� CINTit þ b19TREF � INVINTit þ b20TREF �RDINTit

þ b21TREF �ROAit þ b22�29TREF � INSECit þ eit ð1Þ

where the dependent variable, ETRit, is the corporate effective tax rate proxyfor firm i in year t. The independent variables (with subscripts omitted) includeproxies for firm size (SIZE), capital structure (LEV), asset mix (CINT, IN-VINT and RDINT), firm operations (ROA), industry sectors (INSEC), theRalph Review tax reform (TREF) and interaction terms (TREF * SIZE, TRE-F * LEV, TREF * CINT, TREF * INVINT, TREF * RDINT, TREF * ROAand TREF * INSEC). Finally, all variables are computed based on financialstatement data gathered from the Aspect Financial Database.

4. Results and analyses

4.1. Descriptive statistics

Table 3 reports descriptive statistics for the dependent variable (ETR1 andETR2) in panel A, and selected independent variables in panel B over the per-iod 1997–2003.

Table 3Descriptive statistics of effective tax rates and selected independent variables over the period 1997–2003 (n = 552)

ETR1 ETR2

Panel A: Dependent variable

Mean .356 .229Median .314 .210Standard deviation .254 .135

SIZE LEV CINT INVINT RDINT ROA

Panel B: Selected independent variables

Mean 19.445 .189 .393 .153 .002 .109Median 19.205 .171 .359 .106 0 .092Standard deviation 1.954 .232 .217 .156 .012 .163

Variable definitions: ETR1 is income tax expense divided by book income, ETR2 is income taxexpense dividend by operating cash flows, SIZE is the natural logarithm of total assets, LEV islong-term debt divided by total assets, CINT is net property, plant and equipment divided by totalassets, INVINT is inventory divided by total assets, RDINT is R&D expenditure divided by netsales, and ROA is pre-tax income divided by total assets.


For the dependent variable, ETR1 has a mean of .356 and a median of .314,and ETR2 has a mean of .229 and a median of .210. Given that both ETR mea-sures have the same numerator, and that operating cash flows are normallygreater than book income, the mean for ETR1 is greater than ETR2, asexpected. For the independent variables, SIZE has a mean of 19.445 and amedian of 19.205, LEV has a mean of .189 and a median of .171, CINT hasa mean of .393 and a median of .359, INVINT has a mean of .153 and a med-ian of.106, RDINT has a mean of .002 and a median of 0, and ROA has amean of .109 and a median of .092. A reasonable level of consistency isobserved between the mean and median for all variables.

4.2. Regression results

Table 4 (panel A) reports pooled cross-sectional OLS regression results ofETRs on several firm characteristics over the period 1997–2003 (coefficient esti-mates with t-statistics in parentheses). To obtain robust standard errors in ourregressions, we use the Huber/White/Sandwich estimator of standard errors(see, e.g. Froot, 1989; Wooldridge, 2002). In addition, t-statistics for tests ofthe coefficient sums are presented in Table 4 (panel B).

Both regression models in Table 4 (panel A) are significant (p < .01), withadj. R2s of 13% for ETR1 and 19% for ETR2. Concerning the significanceof the regression coefficients for the firm-specific variables, Table 4 (panel A)shows that SIZE has a significant negative association with ETR1 (p < .01)and ETR2 (p < .10). Our results are consistent with H1 and prior Australian

Table 4Pooled cross-sectional OLS regression results of effective tax rates on various firm characteristicsover the period 1997–2003 (n = 552)

Variable Predicted sign ETR1 ETR2

Panel A: coefficient estimates (with t-statistics in parentheses)a

Intercept .875 .396(4.734)*** (3.808)***

SIZE � �.024 �.007(�2.766)*** (�1.400)*

LEV � �.113 �.290(�1.326)* (�4.028)***

CINT � �.150 �.077(�2.542)*** (�1.638)**

INVINT + .497 .209(2.630)*** (2.944)***

RDINT � �.731 �.345(�2.125)** (�1.264)*

ROA + .020 .298(.164) (3.239)***

TREF � �.447 �.232(�1.943)** (�1.933)**

TREF*SIZE ? .013 .001(1.181) (0.167)

TREF*LEV + .497 .361(3.448)*** (4.349)***

TREF*CINT + .013 .039(.163) (0.722)

TREF*INVINT � �.355 �.168(�1.671)** (�2.074)**

TREF*RDINT ? .302 .048(.717) (.148)

TREF*ROA � �.971 �.289(�3.587)*** (�1.901)**

Adj. R2 0.13 0.19F-value 8.27 8.87(Two-tailed p-value) (.01) (.01)

Variable Hypothesis ETR1 ETR2

Panel B: t-statistics for hypotheses tests of significance of independent variables in the post-tax-reform

period based on the coefficient estimates reported in Panel Ab

SIZE b1 + b16 = 0 �1.817* �2.247**

LEV b2 + b17 = 0 �3.316*** �1.734**

CINT b3 + b18 = 0 �2.574*** �1.414*

INVINT b4 + b19 = 0 1.449* 1.061RDINT b5 + b20 = 0 �4.235*** �1.692**

ROA b6 + b21 = 0 3.929*** 4.829***

Variable definitions: ETR1 is income tax expense divided by book income, ETR2 is income taxexpense dividend by operating cash flows, SIZE is the natural logarithm of total assets, LEV islong-term debt divided by total assets, CINT is net property, plant and equipment divided by total

(continued on next page)


Table 4 (continued)assets, INVINT is inventory divided by total assets, RDINT is R&D expenditure divided by netsales, ROA is pre-tax income divided by total assets, TREF is a dummy variable coded 1 if theobservation is for the post-tax-reform period, 0 otherwise, TREF * SIZE is an interaction term,TREF * LEV is an interaction term, TREF * CINT is an interaction term, TREF * INVINT is aninteraction term, TREF * RDINT is an interaction term, and TREF * ROA is an interaction term.Our regression model also includes dummy control variables for industry-sector effects.

ETR1it or ETR2it ¼ a0 þ b1SIZEit þ b2LEVit þ b3CINTit þ b4INVINTit þ b5RDINTit

þ b6ROAit þ b7�14INSECit þ b15TREFit þ b16TREF � SIZEit

þ b17TREF � LEVit þ b18TREF � CINTit þ b19TREF � INVINTit

þ b20TREF �RDINTit þ b21TREF �ROAit þ b22�29TREF � INSECit

þ eit

Note, the coefficients for the industry-sector dummy control variables and corresponding inter-action terms are untabulated.*, **, *** Significant at .10, .05 and .01 levels. p-Values are one-tailed for directional hypothesesand two-tailed otherwise.

a Regression estimates are based on the Huber/White/Sandwich estimator of standard errors (see,e.g. Froot, 1989; Wooldridge, 2002).

b The t-statistics were calculated as follows: (bi + bj)/[Var(bi) + Var(bj) + 2 cov(bi, bj)]1/2.


research by Tran (1997); using a different database, time period and researchdesign. It appears that larger firms do possess superior economic and politicalpower relative to smaller firms and are able to reduce their tax burdens. Over-all, the Australian findings indicate that firm-size effects are not sample-spe-cific, but have a robust (negative) association with ETRs over time in firmswith longer histories. Hence, these findings contradict Gupta and Newberry’s(1997, p. 28) assertion. Research on ETRs in several less-developed Asia–Paci-fic countries: Korea, Malaysia, Taiwan and Thailand (see, e.g. Kim and Limp-aphayom, 1998; Derashid and Zhang, 2003) also shows a consistent negativeassociation between ETRs and firm-size in firms with longer histories. Whilein the US, it might be possible that firm-size effects exhibit the tendencies doc-umented by Gupta and Newberry (1997, p. 28), these findings are not general-izable to other countries, where the evidence shows that firm size (politicalpower theory) is an important predictor of ETRs. In addition, LEV is includedin our study as a proxy for firms’ capital structure. The results indicate that ithas a significant negative association with ETR1 (p < .10) and ETR2 (p < .01),and is consistent with H2. Because interest expenditure is tax-deductible, firmswith higher leverage have lower ETRs. For the asset-mix variables of CINT,INVINT and RDINT, Table 4 (panel A) demonstrates a number of significantassociations. CINT has a significant negative association with ETR1 (p < .01)and ETR2 (p < .05). These results are consistent with H3 such that firms thatare more capital-intensive have lower ETRs. INVINT has a positive associa-tion with ETR1 and ETR2 (p < .01). These results are consistent with H4,showing that inventory-intensive firms have higher ETRs. RDINT has a signif-


icant negative association with ETR1 (p < .05) and ETR2 (p < .10). Theseresults are consistent with H5, suggesting that firms that are more R&D-inten-sive have lower ETRs. Finally, ROA is included in our study to control for firmoperations. The results show that it has a significant positive association withETR2 (p < .01), as expected.

For the tax-reform variables in Table 4 (panel A), the coefficient of TREF isnegative and significant for ETR1 and ETR2 (p < .05). These results indicatethat ETRs decreased in the post-tax-reform period, and are consistent withH6. As expected, the corporate tax rate reductions dominated the base-broad-ening rule changes. The coefficients of the interaction terms between TREF andsome of the independent variables are also significant. This suggests that theassociation between these variables and ETRs changed after the tax reform.TREF * LEV is positive and significant for ETR1 and ETR2 (p < .01). Theseresults show that firms with higher leverage experienced an increase in ETRsafter the tax reform, because a reduction in the corporate tax rate decreasedthe tax savings associated with interest expenditure; this is consistent withH7. TREF * INVINT is negative and significant for ETR1 and ETR2(p < .05). These results show that inventory-intensive firms encountered adecrease in their ETRs after the tax reform, which is consistent with H9.Because inventory intensity is a substitute for capital intensity, inventory-inten-sive firms have lower ETRs after the tax reform. TREF*RDINT is insignifi-cant, so tax reform did not impact on ETRs through different levels of R&Dintensity, and these results are consistent with H10. Finally, TREF * ROA isnegative and significant for ETR1 (p < .01) and ETR2 (p < .05). These resultsare consistent with the expected impact of the corporate tax rate reductions.

Table 4 (panel B) also reports the t-statistics for hypotheses tests of signifi-cance of independent variables in the post-tax-reform period, based on the coef-ficient estimates listed in panel A of Table 4. The results indicate that SIZE, LEV,CINT, INVINT and RDINT are significantly associated with ETRs (ETR1 orETR2 at p < .10 or better) after the tax reform. These results show that whilethe Ralph Review impacted on many of these associations, ETRs continue tobe associated with firm size, capital structure and asset mix after the tax reform.

4.3. Robustness checks

As a robustness check of our results, we re-estimated our regression modelsby dividing our sample into the pre-(1997–1999) and post-(2001–2003) RalphReview tax-reform periods3 to determine whether the associations identified

3 Similar to Gupta and Newberry (1997), each period included only those firms with the requisitedata in all years of that period (i.e., the same firms were not required to be in both periods). Thisyielded a sample of 128 firms (384 firm years) for 1997–1999 and 150 firms (450 firm years) for2001–2003.


in Table 4 between ETRs and SIZE, LEV, CINT, INVINT and RDINTremain intact. The results (not reported) show that the coefficient estimatesfor SIZE, LEV, CINT, INVINT and RDINT had identical signs and similarlevels of statistical significance in the pre- and post-tax-reform periods to thosereported in Table 4.

To further assess the robustness of the full sample results in Table 4, we per-formed some other robustness checks. First, variance inflation factors (VIFs)were computed for each independent variable in our regression models. Over-all, the VIFs indicate that multicollinearity is not problematic in any of themodels. Second, to deal with potential outlier problems, we re-estimated ourregression models after excluding several outliers, based on the method sug-gested by Neter et al. (1996). Our results in terms of sign and statistical signif-icance are comparable to those reported in Table 4. Finally, we re-estimatedour regression models to include NOL carry-forward firms. The results showthat the coefficient estimates had similar signs, but weaker levels of statisticalsignificance, as expected (see, e.g. Wang, 1991).

5. Conclusions

This paper examines the determinants of the variability in corporate ETRsin Australia spanning the Ralph Review tax reform. We find a significant neg-ative association between ETRs and firm size. We also observe that ETRs havea significant negative association with capital structure for leverage. A signifi-cant negative (positive) association is also found between ETRs and asset mixfor capital intensity and R&D intensity (inventory intensity). While the RalphReview impacted on many of these associations, ETRs continue to be associ-ated with firm size, capital structure and asset mix after the tax reform. Ourstudy adds to the scarce literature on ETRs and tax reform. Moreover, theresults of this study provide some additional insights into ETRs that shouldbe useful to policy-makers.

Our study has several limitations. First, the sample is drawn from publicly-listed Australian firms. Because of data unavailability, it was not possible toinclude unlisted firms in our sample. Second, we constructed our ETR mea-sures using financial statement data since tax return data are private andunavailable. The literature (see, e.g. Plesko, 2003) questions the accuracy offinancial-statement-based ETR measures, so our results should be interpretedwith some caution. Finally, our ETR model may be incomplete. For example,the extent of firms’ foreign operations and ownership structure might impacton ETRs. We excluded these variables due to data and cost constraints. Futureresearch could consider these issues.


Acknowledgement

The authors would like to thank Terry Shevlin, Joseph Weintrop, and sem-inar participants in the Department of Accountancy at the City University ofHong Kong for helpful comments. Moreover, we are particularly grateful toMartin Loeb and the two anonymous reviewers for their insightful suggestions.Finally, the authors acknowledge financial support through the Frank BurkeScholarship 2004, provided by CPA Australia.

References

Cooper, G.S., Krever, R.E., Vann, R.J., 2002. Income Taxation: Commentary and Materials, 4thed. Australian Tax Practice, Sydney, NSW.

Derashid, C., Zhang, H., 2003. Effective tax rates and the industry policy hypothesis: evidence fromMalaysia. Journal of International Accounting, Auditing and Taxation 12, 45–62.

Dhaliwal, D., Wang, S., 1992. The effect of book income adjustment in the 1986 alternativeminimum tax on corporate reporting. Journal of Accounting and Economics 15 (1), 7–26.

Froot, K.A., 1989. Consistent covariance matrix estimation with cross-sectional dependence andheteroskedasticity in financial data. Journal of Financial and Quantitative Analysis 24 (3), 333–355.

Gilders, F., Taylor, J., Richardson, G., Walpole, M., 2004. Understanding Taxation Law: AnInteractive Approach, second ed. LexisNexis Butterworths, Sydney, NSW.

Guenther, D., 1994. Earnings management in response to corporate tax rate changes: evidencefrom the 1986 Tax Reform Act. The Accounting Review 69 (1), 230–243.

Gupta, S., Newberry, K., 1997. Determinants of the variability on corporate effective tax rates:evidence from longitudinal data. Journal of Accounting and Public Policy 16 (1), 1–34.

Kern, B.B., Morris, M.H., 1992. Taxes and firm size: the effect of tax legislation during the 1980s.Journal of the American Tax Association 14 (1), 80–96.

Kim, A., Limpaphayom, P., 1998. Taxes and firm size in Pacific-Basin emerging economies. Journalof International Accounting, Auditing and Taxation 7 (1), 47–68.

McIntyre, R.S., Nguyen, T.D.C., 2000. Corporate Income Taxes in the 1990s. Institute onTaxation and Economic Policy, Washington, DC.

Neter, J., Wasserman, W., Kutner, M., 1996. Applied Linear Regression Models, third ed. Irwin,Chicago, IL.

Omer, T., Molloy, K., Ziebart, D., 1993. An investigation of the firm size–effective tax rate relationin the 1980s. Journal of Accounting, Auditing and Finance 8 (2), 167–182.

Plesko, G.A., 2003. An evaluation of alternative measures of corporate tax rates. Journal ofAccounting and Economics 35 (2), 201–226.

Porcano, T., 1986. Corporate tax rates: progressive, proportional, or regressive. The Journal of theAmerican Tax Association 7 (2), 17–31.

Ralph, J.T., 1999. Review of Business Taxation, Full Report. AGPS, Canberra.Scholes, M., Wilson, P., Wolfson, M., 1992. Firms’ responses to anticipated reductions in tax rates:

the Tax Reform Act of 1986. Journal of Accounting Research 30 (Suppl.), 161–185.Shevlin, T., Porter, S., 1992. The corporate tax comeback in 1987: some further evidence. The

Journal of the American Tax Association 14 (1), 58–79.


Siegfried, J., 1972. The relationship between economic structure and the effect of political influence:empirical evidence from the federal corporation income tax program. Ph.D. dissertation,University of Wisconsin.

Stickney, C., McGee, V., 1982. Effective corporate tax rates: the effect of size, capital intensity,leverage, and other factors. Journal of Accounting and Public Policy 1 (2), 125–152.

Tran, A.V., 1997. The gap between accounting profit and taxable income. Australian Tax Forum13 (4), 507–534.

Tran, A.V., 1998. Causes of the book-tax income gap. Australian Tax Forum 14 (3), 253–286.Wang, S., 1991. The relation between firm size and effective tax rates: a test of firms’ political

success. The Accounting Review 66 (1), 58–169.Watts, R., Zimmerman, J., 1986. Towards a Positive Theory of Accounting. Prentice-Hall, New

Jersey.Wilkie, P., 1988. Corporate average effective tax rates and inferences about relative tax preferences.

The Journal of the American Tax Association 10 (1), 75–88.Wilkie, P., Limberg, S., 1993. Measuring explicit tax (dis)advantage for corporate taxpayers: an

alternative to average effective tax rate. The Journal of the American Tax Association 15 (1),46–71.

Wooldridge, J.M., 2002. Econometric Analysis of Cross Section and Panel Data. MIT Press,Cambridge, MA.

Zimmerman, J., 1983. Taxes and firm size. Journal of Accounting and Economics 5 (2), 119–149.

Date post:	25-Oct-2016
Category:	Documents
Upload:	grant-richardson
View:	212 times
Download:	0 times

Determinants of the variability in corporate effective tax rates and tax reform: Evidence from...

Documents