Economic Policy Uncertainty and Firm Tax Avoidance

1

Economic Policy Uncertainty and Firm Tax Avoidance

Huu Nhan Duong Email: [email protected]; Phone: +61 3 99032032 Department of Banking and Finance, Monash University, Clayton, VIC 3800, Australia Ferdinand Gul Email: [email protected] Deakin Business School, Deakin University, Burwood, VIC 3125, Australia Justin Hung Nguyen Email: [email protected] School of Accounting and Commercial Law, Victoria University of Wellington, Wellington 6140, New Zealand My Nguyen

Email: [email protected] ; Phone: +61 3 99255683 School of Economics, Finance and Marketing, RMIT University, Melbourne, VIC 3001 Australia

This version: 10 Aug 2017

2

Economic Policy Uncertainty and Firm Tax Avoidance

Abstract

We investigate whether and how economic policy uncertainty is related to firm tax avoidance.

We predict that an increase in policy uncertainty results in greater financial constraints, which

in turn, lead firms to increase tax avoidance activities. We find a strong positive association

between economic policy uncertainty and firm tax avoidance. This relation is robust to

alternative measures of tax avoidance and several tests to address endogeneity concerns. Firms

use several strategies to avoid tax including tax deferrals and shelters. Further analysis shows

that the effect of policy uncertainty on tax avoidance is less pronounced for firms with higher

level of cash holdings. Overall, our findings highlight the importance of uncertainty around

government policy in determining firm tax avoidance activities.

Key words: Policy uncertainty, tax avoidance, financial constraints

JEL Classification: G18, G31, G32, G34

3

1. Introduction

In this paper, we empirically investigate whether and how economic policy uncertainty is

systematically related to corporate tax avoidance. Our investigation is motivated by the central

role of politics on corporate taxation and that a country’s corporate tax regimes are products of

a political process. In the United States, many businesses are now facing with substantial

uncertainty from a host of policy changes, especially on corporate tax policy such as the

expiration of the Bush tax-cuts in 2010, the alternative minimum tax policy (AMT), the

expiration of a two-percentage-point federal payroll tax cut and over 55 other tax breaks that

are set to expire in 2013. Whether corporate tax rates are scheduled to go up is far from certain

for firms, which make it difficult for firms to forecast their future earnings and financing

sources. These policy uncertainties may also cause firms’ existing tax planning strategies

become ineffective and encourage them to increase their tax avoidance activities while their

current tax planning are still effective under the current regime. Despite this potential

interaction between policy uncertainty and corporate taxation, the effect of policy uncertainty

on corporate tax avoidance remains unexplored from the academic literature.

We hypothesize that economic policy uncertainty may impact on corporate tax

avoidance through a financial constraint channel. The literature on economic policy

uncertainty and financial constraints states that policy uncertainty will lead to capital supply

frictions and increase firm financing costs (Pástor and Veronesi, 2012, 2013; Gungoraydinoglu

et al., 2017). When traditional debt and equity financing sources become costlier and more

difficult to obtain, firms may acquire alternative source of funds through their tax planning by

reducing their current reported taxable income or increase tax credits, thereby decreasing cash

taxes paid (Edwards et al., 2015). Similarly, Law and Mills (2015) state that when the frictions

in external finance increase, financially constrained firms will pursue more aggressive tax

planning on the margin as a substitute for a more expensive source of external financing from

4

lenders or capital markets. Edwards et al. (2015) suggest several reasons why constrained firms

will employ tax saving as a source of fund. First, tax savings, unlike many other cost-cutting

techniques (i.e. reducing advertising, research and development, capital expenditure and

staffing), are less likely to adversely affect the firm’s operations. Second, firms are also highly

likely to have additional opportunities to generate current cash tax savings via deferral-based

tax planning strategies. Third, constrained firms are also receptive to employ tax avoidance

activities as a source of generate more cash (Hanlon et al., 2017). Collectively, the financial

constraints channel suggests that firms heighten their tax avoidance activities when policy

uncertainty increases.

We examine the relation between policy uncertainty and corporate tax avoidance from

1988 to 2014 using the Baker et al. (2016)’s (hence BBD thereafter) policy uncertainty index.

This index is a weighted average measure of the frequency of articles containing key terms

related to policy uncertainty in 10 leading U.S. newspapers, uncertainty about future changes

in the federal tax code measured by the dollar impact of tax provisions set to expire in the near

future, and the forecast disagreement concerning government spending and consumer price

index which are used as proxy for uncertainty about future fiscal and monetary policy. While

election years are also used in the literature as a measure of policy uncertainty, we employ

Baker et al. (2016)’s index instead as this index also captures policy uncertainty unrelated to

elections or outside of election years (Gulen and Ion, 2016). It also accounts for the effect of

elections as well as the extent to which the election outcomes are uncertain (Gulen and Ion,

2016).

We find a positive association between policy uncertainty and firm tax avoidance. Our

estimation suggests that when policy uncertainty doubles, firms on average lower their cash

tax effective rates by 1.97%, implying a reduction of annual tax expense and annual tax

payment of $1.81 million and $4.06 million on average, respectively. Further, we find that the

5

positive impact of policy uncertainty on corporate tax avoidance persists for the following two

years and is economically stronger when 100% jump in policy uncertainty results in an 2.97%

surge in tax avoidance (which equivalent to a reduction of income tax payment of $6.13 million

on average per firm-year). This positive impact, however, disappears after three years.

The BBD index may capture the effects of general economic uncertainty that potentially

confounds our findings of a positive relationship between policy uncertainty and tax avoidance.

We address this concern by controlling for several proxies for economic uncertainty suggested

by Gulen and Ion (2016). These include the election year (Julio and Yook, 2012), the

Livingstone survey of uncertainty about future economic growth, cross-sectional standard

deviation of firm-level profit growth, the VXO index of implied volatility, cross-sectional

dispersion in stock returns and the Jurado et al. (2015)’s index. Our findings are qualitatively

unchanged when we augment our baseline regression model with these six general economic

uncertainty proxies.

Another potential issue with using the BBD index is that it may capture the effects of

other non-policy related factors such as currency uncertainty or labour market variation which

raises an error-in measurement concern that could potential bias the association between policy

uncertainty and tax avoidance (Nguyen and Phan, 2017). To address the error-in-measurement,

we regress the measure of the news-based component of the US BBD index on the Canada

index to obtain the regression residuals. Given that many of the shocks that affect economic

uncertainty in the United States will also affect general economic uncertainty in Canada, the

residuals should be free from potential confounding effects of macroeconomic forces common

to both countries. The results further corroborate our findings that firms engage in more tax

avoidance activities when policy uncertainty increases.

To further alleviate endogeneity concerns, we follow Gulen and Ion (2016) and use an

instrumental variable specification in which a measure of political polarisation in the United

6

States is used as an instrument for policy uncertainty. The results of these tests confirm our

main findings that policy uncertainty is positively associated with corporate tax avoidance.

In addition, we use alternative measures to capture aggressive and deliberate corporate

tax avoidance practices. We find that policy uncertainty is positively associated with the

probability that a firm undertakes a tax-sheltering transaction in given year, the likelihood that

a firm is tax dodger in a given year in that it reports a positive pre-tax income but pays no tax.

Moreover, the baseline results are also robust when we use long-run tax rates as alternative

proxy for corporate tax avoidance.

We next examine a specific mechanism by which policy uncertainty increases firm tax

avoidance, specifically through its impact on firm financial constraints. We perform two

separate analyses. First, we provide evidence that the aggregate bank credit conditions, as

proxied by the spread of commercial and industrial loan rates (on loans greater than USD 1

million) over the federal funds rates, tighten when policy uncertainty increases. Second, prior

studies show that firms have a precautionary motive to hold more cash when financial

constraints increase (Opler et al., 1999; Bates et al., 2009). Thus, if policy uncertainty affects

corporate tax avoidance through its effect in increasing financial constraints, the effects of

policy uncertainty on firm tax avoidance should be less severe for firms with higher cash

holdings. We find support for this empirical prediction. We further show that the moderating

effect of cash holdings on the relation between policy uncertainty and tax avoidance is stronger

for financially constrained firms at which we characterize as being smaller in size, higher

dividend payout ratio, younger age and not having debt and paper rated, than for unconstrained

firms. Taken in their entirety, our these results support the “financial constraints” hypothesis

that firm tax avoidance increases when financial constraints heighten in the period of high

policy uncertainty.

7

Our study contributes to the literature in several ways. First, we expand the literature

on the determinants of corporate tax behaviour. Much of the prior research focuses on cross-

sectional variation in tax avoidance and identifies firm-level factors associated with firm tax

avoidance, such as financial leverage (Lisowsky, 2010), ownership (Chen et al., 2010; Cheng

et al., 2012; Badertscher et al., 2013), executives (Dyreng et al., 2010) and corporate

governance (Armstrong et al., 2015). We contribute to this literature by uncovering the

association between economic policy uncertainty and tax avoidance.

Our study also contributes to the literature on political costs and taxes, which is dated

back to Zimmerman (1983) who envisions taxes as one component of political costs. Most of

the researchers focus on reputational costs and scrutiny effects on valuation of tax benefits.

They suggest that public pressure from disclosure of firms' tax haven subsidiaries causes firms

to curtail their tax avoidance (Dyreng et al., 2016). Likewise, Hanlon and Slemrod (2009),

Graham et al. (2014), Lisowsky et al. (2013) and Gallemore et al. (2014) examine whether

reputational costs result in aggressive tax behaviour, where reputation costs are defined to

include political costs. While political uncertainty is broadly captured by our economic

uncertainty measures, our study fundamentally differs from the prior works. Specifically, rather

than focusing on political scrutiny, which mitigates tax avoidance, we study economic policy

uncertainty, which we hypothesize exacerbates tax avoidance.

The remaining of this article proceeds as follows. Section 2 provides details on data and

variable description. Section 3 discusses the main findings and implications while Section 4

concludes the paper.

8

2. Data and sample selection

2.1 Sample and data sources

Our data comes from several sources. The financial information for all publicly traded firms

with headquarters located in the USA between 1988 and 2014 from Standard and Poor’s

Compustat database. The monthly economic policy uncertainty index of Baker et al. (2016) is

sourced from http://www.policyuncertainty.com. Following the prior literature (Chen et al.,

2010; Dyreng et al., 2016; Cen et al., 2017), we remove firm-year observations with negative

pre-tax income or book value with non-positive sales or with total assets of less than $1 million.

Firms from the financial services and utilities industries are also excluded. We also drop firm-

year observations with unavailable information from Compustat to calculate the key tax

avoidance variables and other control variables. These screening criteria yield a final full

sample of 65,822 firm-year observations.

2.2 Measures of firm tax avoidance

Following Hasan et al. (2017), we use the firm’s cash effective tax rate (CETR) which equals

to cash taxes paid divided by pre-tax book income before special items. This measure reflects

both temporary (i.e. tax deferral differences) and permanent differences and is also unaffected

by tax accruals. For an ease of our interpretation, CETR is multiplied by -1 and denoted as

TA_CETR as measures of firm tax avoidance. By definition, higher TA_CETR implies greater

tax avoidance. The definition and detailed calculation of this variable are provided in Appendix

A.

2.3 Measures of economic policy uncertainty

The economic policy uncertainty index (PU) is developed by Baker et al. (2016) which is a

weighted average of the three components. The first component quantifies the volume of news-

9

based policy uncertainty every month starting from January 1985. This is done by searching

on the 10 leading newspapers: USA Today, Miami Herald, Chicago Tribune, Washington Post,

Los Angeles Times, Boston Globe, San Francisco Chronicle, Dallas Morning News, New York

Times and Wall Street Journal that contain the following key words: “uncertainty” or

“uncertain”; “economic” or “economy”; and one of the following policy terms: “congress”,

“deficit”, “Federal Reserve”, “legislation”, “regulation” or “White House”. To control for

changing in the volume of articles across newspapers and time, total numbers of word counts

are scaled by the total numbers of articles in the same newspaper and month, which yields a

monthly policy uncertainty series for each newspaper. These monthly newspaper-level

uncertainty series are then standardized by unit standard deviation from 1985 to 2010 and then

averaged across the ten papers per month. Finally, the series are then normalized to a mean of

100 from 1985 to 2009.

The second component of the PU index measures the level of uncertainty related to

future changes in the tax code. It is a transitory measure constructed by the number of

temporary federal tax code provisions set to expire in the contemporaneous calendar year and

future ten years and reported by the Joint Committee on Taxation. The third and final

component is the CPI disagreement and expenditure dispersion. It is measured by the

forecasters’ disagreement (the interquartile range of forecast) over future outcomes about

inflation rates and federal government purchases, respectively.

The overall measure of policy uncertainty is calculated by normalising each of the three

components above and then weighted average of the resulting series, using a weight of one-

half for the news-based component, one-sixth of the tax component and one-third for the

forecaster disagreement component. Baker at al. (2016) use several robustness tests. These

include comparing the index with the Chicago Board Options Exchange Market Volatility

Index (VIX); controlling for the potential for political slant to skew newspaper coverage

10

of policy uncertainty; and using uncertainty indicators based on the Beige Book releases before

each regularly scheduled meeting of the Federal Open Market Committee (FOMC). All the

tests confirm that policy uncertainty index remains a consistent measure of economic policy

uncertainty,

2.4 Control variables

We identify several control variables following the literature including firm size (SIZE),

financial leverage (LEVERAGE), cash holdings (CASH), profitability (ROA), loss carry-

forwards (NOL), equity income (EQUITY INCOME), capital investment (PPE) asset

intangibility (INTANGIBLE) and foreign income (FOREIGN_INCOME). We include a size

proxy which is the natural logarithm of the firm’s market capitalisation (SIZE). Prior studies

provide conflicting evidence of the association between tax avoidance and firm size. Consistent

with the “political cost” hypothesis, larger firms have greater incentive to engage in tax

avoidance activities (Zimmerman, 1983). Large firms are often more sophisticated and better

equipped to structure complex tax-reduction transactions (Hanlon et al., 2005). However, some

other studies (Jacob, 1996; Gupta and Newberry, 1997) do not find a statistically significant

relationship.

Financial leverage (LEVERAGE) is included to capture the effect of the tax shield on

debt, which higher corporate tax shields can reduce marginal tax rates the incentives for

incremental tax planning (Graham, 1996a; 1996b, 2000). Newberry and Dhaliwal (2001) also

argue that multinationals can place debt in high-tax locations to reduce their effective tax rates.

They can also structure off-balance sheet financing to maximize interest deductions without

decreasing book income (Mills and Newberry, 2004) or can structure debt to use foreign tax

credits (Newberry, 1998). Collectively, these studies suggest that increased levels of debt are

positively associated with firm tax avoidance.

11

We also control for cash holdings (CASH) to capture the firm tax planning incentives.

The association between cash holdings on firm tax avoidance is not determined the priori. On

the one hand, firms with more cash have less incentive to defer taxes (Cen et al., 2017). On

the other hand, tax aggressive firms may also hold more cash as a precautionary motive for

future settlement with the Internal Revenue Service (IRS) (Hanlon et al., 2017).

ROA is a firm’s operating income scaled by lagged total assets. It is used to control for

the effect of firm profitability on taxes and we expect a positive association between

profitability on both ETR and CETR following Edwards et al. (2015). Firm loss carry-forwards

(NOL) is also included as loss carry-forward may also cause a firm’s tax rate to differ from the

statutory rate (Auerbach and Poterba, 1987).

EQUITY_INCOME is also included because prior research suggests that economies of

scale and firm complexity resulting in greater equity income are positive associated with tax

avoidance (Rego, 2003; Chen et al., 2010). We also control for the existence of foreign

jurisdictions (FOREIGN_INCOME) and asset intangibility (INTANGIBLE) because these are

likely to affect both firms’ likelihood of using debt and firms’ possibility of engaging in tax-

avoiding behaviour. Specifically, firms taking advantage of foreign tax rate differentials should

avoid more tax on average and so we expect FOREIGN_INCOME to be positively associated

with tax avoidance.

2.5 Descriptive statistics

Table 1 reports descriptive statistics of our tax avoidance measures (Panel A), economic policy

uncertainty index (Panel B) and control variables (Panel C) used in our baseline regression in

Equation 1 below. The mean values of (inverse) cash effective tax rates (TA_CETR) and

effective tax rates (TA_ETR) are -26.68% and -30.31% respectively. The mean of value of

DTAX is -0.001.

12

3. Research methodology, results and discussions

3.1 Policy uncertainty and corporate tax avoidance

To investigate the relationship between policy uncertainty and corporate tax avoidance, we use

the following augmentations of panel regressions common to the tax avoidance literature:

��_��,�� = �� + ��,� + �� ,�,� + �� + ��,�� (1)

Here, i indexes firms, t indexes calendar quarters and l ∈ {1,2,3) stands for the year lead

between the dependent and independent variables. Our standard errors are clustered at the firm

level to correct for potential cross-sectional correlation in the error term ��,�� (Petersen, 2009).

TA_CETR is an inverse measure of firm cash effective tax rates or greater tax avoidance for a

firm i from year t to t+l. For each firm i, the policy uncertainty variable (PU) is measured as

the natural logarithm of the arithmetic average of the PU index in the twelve months of the

firm’s fiscal year t. Since the majority of the explanatory power of the overall policy uncertainty

comes from its news-based component, for brevity, we follow Gulen and Ion (2016) to present

the regression results from the news-based index. This also serves to eliminate any possible

confusion as to which of the components of the PU index is driving our results. Nevertheless,

our results are qualitatively the same if we use the overall index instead.

Similar to Gulen and Ion (2016), we do not include the time fixed effects in our

specification since doing so will automatically absorb all explanatory power of the policy

uncertainty variables. Even though the baseline model includes firm attribute control variables,

it might still omit some unknown firm characteristics that affect uncertainty and corporate tax

avoidance. To ease this concern, we use firm fixed-effect regressions and industry fixed effects

to control for the influences of unknown time-invariant firm level and industry level factors,

respectively. To reduce the impact of extreme outliers, all variables have been winsorised at

the 1% and 99% level.

13

<Insert Table 2 here>

Table 2 displays the results of the association between tax avoidance and policy uncertainty.

The first model (Column (1)) contains no controls aside from policy uncertainty index while

in the second to fifth models (in Columns (2) to (5) include the full set of controls and industry

fixed effects. Also, cash tax effective rates in current year (t) is replaced by one-year (t+1),

two-year (t+2) and three-year (t+3) leads as dependent variables in Columns (2) to (5),

respectively. Results in Columns (1) to (5) suggest that an increase in policy uncertainty leads

to lower cash tax effective rates or higher firm tax avoidance firm cash holdings both in the

current year and the following years. In particular, the coefficient of PU of 0.0197 (Column

(2)) indicates that when policy uncertainty doubles, firms on average lower their cash tax

effective rates by 1.97% implying a reduction of annual tax expense and annual tax payment

of $1.81 million and $4.06 million on average, respectively. Further, the coefficient of PU

remains significant in Columns (3) and (4), suggests that the positive impact of policy

uncertainty on tax avoidance persists after two years and the effect is economically stronger

when 100% jump in policy uncertainty results in an 2.97% surge in the tax avoidance in two

years later. This positive impact, however, disappears after three years as suggested by the

insignificant explanatory power of PU on ��_��in Column (5). The adjusted R2

increases considerably from 0.013 in the first model to 0.061 suggesting the addition of a

powerful set of controls.

3.2 Control for confounding effect of economic uncertainty

The BBD index may be highly likely correlated with other sources of general economic

uncertainty such as recessions, wars, financial crisis that potentially confound our findings of

a positive relationship between policy uncertainty and firm tax avoidance. To control for this

possible contamination, we follow Gulen and Ion (2016) to include several proxies for

14

economic uncertainty and separately run time-series regressions of PU on a list of macro

uncertainty variables. First, we include the GDP forecast from Livingston survey published by

the Philadelphia Federal Reserve, we calculate the coefficient of variation of GDP forecast as

a proxy for expected economic growth uncertainty (GDPDIS).1 Second, we compute the annual

cross-sectional standard deviation of firm profit growth as a proxy for future profitability

variation, where firm profit growth is defined as the ratio of the change in net income to average

sales (SDPROFIT). Third and fourth, to control for the equity market uncertainty, we calculate

the monthly standard deviation of stock returns (SDRETURN) and the Chicago Board Options

Exchange’s VXO index of implied volatility (VXO). Fifth, we use another comprehensive

measure of aggregate uncertainty (JLN), developed by Jurado et al. (2015) which is based on

the co-movement in the unpredictable component of a big number of economic indicators.

Finally, we follow Julio and Yook (2012) to construct an election year dummy (ELECYEAR)

which is equal one on the years of presidential elections. Through our sample period 1985-

2014, there were seven U.S. presidential elections happening every four years in 1988, 1992,

1996, 2000, 2004, 2008, and 2012. We take natural logarithms of all of these economic

uncertainty measures (except for the election year dummy) and gradually add each of them and

then all of them to Equation (1). The results of this regression analysis are presented in Table

3 below.


Except minor discrepancy in the result when GDP forecast is added to Equation (1) in Column

2, the regression results provided in Table 3 show that the positive association between PU and

firm tax avoidance remains highly statistically significant in the presence of these five other

macro-economic uncertainty variables. After all the economic uncertainty controls are

introduced as in Column (7), the coefficient of PU_LOG is 0.0102 suggesting that a 100%

1 Biannual GDP forecasts from the Livingstone survey of the Philadelphia Federal Reserve Bank.

15

increase in policy uncertainty, the firm cash tax effective rates is reduced or tax avoidance

increases by 1.02%. Additionally, in relation to the results of the control variables for firm tax

avoidance, both Tables 2 and 3 indicate the significant explanatory powers of these proxies on

firm tax avoidance and the signs of their coefficient estimates are generally consistent with

predictions. The statistically significant evidence also reveals that the explanatory power of

policy uncertainty on is not fully absorbed by any of these six proxies, that highlights the

robustness of our baseline results. This also supports the argument of Gulen and Ion (2016)

that BBD index comprises macroeconomic uncertainty information that is not captured by any

of the other well-known measures adopted in the existing literature.

Another potential issue with using the BBD index as a proxy for policy uncertainty is

that it may capture the effects of other non-policy related factors, such as currency uncertainty,

which may cause an error-in-measurement concerns that could potentially bias our estimation.

To address this error-in-measurement issue, we follow Gulen and Ion (2016) to extract the

economic uncertainty components from the original PU measure. We do so by using the two-

step regression approach. First, we regress the measure of the news-based component of the

U.S. BBD policy uncertainty index on the Canadian news-based uncertainty measure together

with other six macro-economic variables described above. We then obtain the regression

residuals (RPU) which are the difference between the actual and the predicted U.S news-based

uncertainty measure. Canadian news-based uncertainty index is chosen due to the close

relationship between the US and Canadian economies and, thus, any aggregate shock to Canada

would affect the U.S. as well. Hence, if the BBD index partially captures policy-unrelated

economic uncertainty, the inclusion of the Canadian index helps to remove the economic

uncertainty in U.S. that is derived from economic and policy uncertainty in Canada. This

technique presents an econometric advantage compared to the previous one which just includes

the six macroeconomic variables. This is because this approach helps mitigate the concern of

16

multi-collinearity resulted from the inclusion of too many correlated variables such as PU and

macroeconomic variables into one regression.

In particular, we follow Gulen and Ion (2016) to propose an augmented monthly time-

series model:

USPUt = α0 + β1CANPUt + βkMACRO_VARIABLESk,t + ϵt (2)

Here, USPUt and CANPUt are the logarithm transformation of news-based policy uncertainty

measures developed by BBD for the United States (U.S.) and Canada. The term

MACRO_VARIABLESt represents a vector of six direct measures of macroeconomic

uncertainty for U.S. as defined above. The residuals obtained from running Equation (2) should

represent a “cleaner” policy uncertainty index as it is exempt from the direct and indirect

sources of general economic uncertainty. We then aggregate those monthly residuals in

Equation (3) to yearly level using arithmetic average, and denote the new and cleaner measure

of policy uncertainty for US as RPU. We then repeat the baseline analysis in Equation (1) with

PU being replaced by RPU to be the main variable of interest. Specifically, we run the

following model:

��_��,�� = �� + ��,� + �� ,�,� + �� + ��,�� (3)

The regression result using Equation (3) is presented in Column (8) of Table 3. This

result confirms our main findings that policy uncertainty is positively associated with firm cash

holdings. The relation remains statistically and economically significant when an economic-

free policy uncertainty measure is adopted. In particular, the result indicates that a doubling in

the residual policy uncertainty leads to a surge by 2.49% in the corporate cash-to-assets ratio.

The larger positive coefficient on policy uncertainty suggests that the cleaner measure, i.e.

exempt from aggregate economic shocks, even possesses stronger explanatory power over cash

holdings. This evidence strengthens our argument that policy-related uncertainty indeed

positively drives corporate cash holdings.

17

3.3 Addressing endogeneity concern: Instrumental variable analysis

Another concern with our regression analysis described above is that despite the inclusion of

both firm and industry fixed effects, our policy uncertainty and corporate tax avoidance may

be jointly correlated with the unobservable factors which raise an endogeneity concern in our

models. We address this by conducting an instrument variable analysis. Following McCarty et

al. (1997) and Gulen and Ion (2016), we use the partisan polarization measure (POLAR) as an

instrument for policy uncertainty. This measure is based on the DW-NOMINATE scores

developed by McCarty et al. (1997) to track legislators’ ideological positions over time. In

particular, the measure is calculated as the difference in the first dimension of the DW-

NOMINATE scores between the Republican (code: 200) and Democratic (code: 100)2 parties.

We measure the polarizations for the members in both the Senate and House of Representatives

as alternative instruments. It is argued that partisan polarisation makes it more difficult to build

legislation, resulting in policy gridlock and greater variation in policy (McCarty, 2004). Thus,

partisan polarisation is used as an instrument for policy uncertainty because it is directly related

to policy uncertainty and is unlikely to have a direct impact on firm tax avoidance. In particular,

we execute a two-stage regression strategy as follows:

PUt = α0 + β1POLARt + βkMACRO_VARIABLEk,t + ϵt (4)

TA_CETRi,t+l = α0 + β1FPUi,t + βjCONTROLj,i,t + �iFirmi + ϵi,t (5)

Similar to Equation (2), Equation (4) is a monthly time-series regression where a measure of

political polarization (POLAR) is further added to the model. The fitted values of PU estimated

from Equation (4) are aggregated to yearly level to be the key variable of interest, FPUi,t, in

Equation (5). The specification of Equation (5) is the same with Equation (1), except that the

2 Data are obtained from http://voteview.org/dwnomin_comparison.htm for the period 1998-2014, that is the maximum availability period.

18

original news-based PU is replaced by the fitted PU. Firm-level controls, firm fixed and cluster

effects are included in Equation (5) as in Equation (1).


The regression results using Equation (6) are documented in Table 4. In Column (1) to (4), we

add one more year lead in each model to examine the impact of policy uncertainty on firm tax

avoidance over time. The significantly positive coefficients of the fitted PU from Columns (1)

to (4) confirm our baseline result of the positive association between policy uncertainty and

firm tax avoidance. This impact, however, reverses in year 4, suggesting that firms decrease

their tax avoidance activities when policy uncertainty is resolved in the future. The coefficient

estimates of PU also reveal that firms increase their tax avoidance initially when PU increases

and reduce through time when the uncertainty becomes less severe. Economically, after

controlling for potential endogeneity issue between policy uncertainty and tax avoidance, the

impact of policy uncertainty on corporate tax avoidance becomes much stronger. In particular,

the coefficients of the fitted PU in Column (2) reports that a doubling in the level of policy

uncertainty leads to a reduction by 5.97% in the cash tax effective rates (i.e. equivalent to an

average reduction of tax payment of $12.33 million per firm) in the following year.

As a robustness check, in Columns (6) and (8) we report results when CANPU is

included from Equation (5). In Columns (7) and (8) we replace the Senate DW-NOMINATE

with House DW-NOMINATE scores as the instrumental variable. The results on the coefficients

of the fitted PU consistently further corroborate our findings of a negative relationship between

policy uncertainty and tax avoidance.

3.4 Alternative measures of tax avoidance

Several alternative measures of tax avoidance are also used to ensure the robustness of our

results. These include TA_ETR, DTAX, DEFERAL, TA_CETR5, CASH_RATIO, Low_CETR,

19

CETR_dodger, CURRENT_ETR, SHELTER_LEVEL and SHELTER_DUMMY. The firm

effective tax rate (ETR) which is defined under the Generally Accepted Accounting Principles

(GAAP). ETR is defined as total tax expenses divided by pre-tax book income before special

items. This measure captures firm practices that reduce tax expenses for financial reporting

purposes. One drawback of this measure is that it only reflects tax avoidance strategies that

generate permanent differences and does not capture the effects of temporary book-tax

differences (i.e. deferral strategies). It is also subject to GAAP tax accruals such as the valuation

allowance and unrecognised tax benefits.

DTAX is used to capture the tax practices that drive a permanent difference between

book income and tax income (Frank et al., 2009). This measure is more likely to reflect a

deliberate attempt to avoid tax. Following Frank et al. (2009), we estimate DTAX as the residual

of a regression of permanent book-tax differences on various non-tax planning determinants.

Conceptually, DTAX captures tax avoidance activities that are more aggressive and directly

affect net income through a reduction in total tax expense.

DEFERAL is included following Edwards et al. (2016) which equals to -1 times the

ratio of deferred tax expenses to pre-tax income adjusted for special items as an alternative

measure of firm tax avoidance.

Dyreng et al. (2008) argue that the numerator in the annual TA_ETR formula may

include taxes paid on earnings in a different fiscal year period which may cause a mismatch

problem. To alleviate this problem, we use long run tax rates measured over a five-year period

as an alternative measure of firm tax avoidance following Dyreng et al. (2008). A firm that is

successful in avoiding paying tax over a long period of time (i.e. 5 years) is considered as an

aggressive tax avoider. As a result, TA_CETR5 is -1 times the sum of total tax paid over five

years (t to t+4) scaled by pre-tax income net of total special items over the same accumulation

period.

20

Hassan et al. (2017) also argue that more tax aggressive firms are those that pay

significantly lower taxes when compared to their peers belonging to the same industry. As a

result, based on measures of tax avoidance using CETR, they use dummy variables to capture

firms paying significantly lower taxes than their counterparts in a given year. Specifically,

Low_CETR is constructed analogously to capture firms paying lower CETR when compared to

their industrial counterparts in a given year.

Low_CETR captures firms that pay fewer taxes than their counterparts but they do not

necessarily reflect the most extreme case of tax avoidance practices, especially those that are

labelled under the Center of Tax Justice as “tax dodgers”. “Tax dodgers” companies are those

that are profitable but pay no corporate tax rates. As a result, following Hassan et al. (2017) we

also include ETR_Dodger which equals to one if a firm has a positive pre-tax profit and a zero

ETR in a given year and zero otherwise. CETR_Dodger equals to one if a firm has positive

pre-tax profit and a zero CETR in a given year and zero otherwise.

We also employ a tax-shelter prediction score (SHELTER_LEVEL) as computed in

Wilson (2009) and SHELTER_DUMMY following Hassan et al. (2017). Tax shelters refer to

those complex transactions used by corporations to obtain significant tax benefits probably

never intended by the tax code (Hanlon and Slemrod, 2009). To capture conforming tax

avoidance which occurs when a firm lowers its taxes by reducing both taxable income and pre-

tax accounting income, we follow Cen et al. (2017) to use CASH RATIO which uses operating

cash flows as the denominator. Specifically, the CASH RATIO measure is defined as cash taxes

paid divided by pre-tax operating cash flows adjusted for extraordinary items and discontinued

operations. To further test the robustness of our results, we use a cash tax differential (CTD)

measure developed by Henry and Sansing (2014) which is calculated as the difference between

cash taxes paid and the product of statutory tax rate and pre-tax income, scaled by lagged total

21

assets. Finally, we also estimate effective tax rates using current tax expenses

(CURRENT_ETR) to capture current taxes owed to the tax authorities.

Table 5 present the results associated with these variables. For dummy variables, such

as LOW_CETR, CETR_DODGER and SHELTER_DUMMY, we use logistic regressions.

Across all our models, we find that the coefficients on policy uncertainty is positive and

significant suggesting that in the period of higher policy uncertainty, firms are likely to pay

significant higher tax rates, dodge as well as shelter more their taxes. It also indicates a positive

association between long run corporate tax avoidance and policy uncertainty in the US.


3.5 Financial constraints

The literature suggests a negative association between policy uncertainty and bank

credit growth at both firm and aggregate levels (Bordo et al., 2016). If policy uncertainty

exacerbates financial constraints, we expect that firms will increase their tax avoidance for their

precautionary incentives. To test the financial constraints mechanism, we first examine if

aggregate bank credit is tightened due to heightened policy uncertainty, and as the results, firms

will increase tax avoidance.

3.5.1 Policy uncertainty and credit market conditions

To investigate the effect of policy uncertainty on general credit market conditions, the

following regression model is used:

CISPREADt = α0 + β1PUt + βkMACRO_VARIABLESk,t + δtQuartert + ϵt (6)

Equation (6) is quarterly time-series regression of a proxy for credit market conditions with

CISPREAD is run on news-based measure of policy uncertainty, PU, together with six

22

macroeconomic variables described above. Following Harford (2005), Officer (2007), and

Harford et al. (2014), we measure credit market conditions by CISPREAD which is the spread

of commercial and industrial loan rates (on loans greater than USD 1 million) over the federal

funds rate.3 The authors argue that larger CISPREAD indicates that credit conditions are more

tightening. We also include four quarter dummies to account for the possible seasonality as

well as time trend effects on credit supply. The results for this test are displayed in Table 6.


The result shows that commercial and industrial loans become costlier when policy

uncertainty is more heighten, manifested by the positive coefficient for PU. This makes it

harder for firms to access these main sources of external finance. In sum, the results provide

evidence that policy uncertainty exacerbates the credit market conditions at aggregate level that

is consistent with findings of Bordo et al. (2016).

3.5.2 Policy uncertainty, tax avoidance and cash holdings

Our findings so far suggest that firm tax avoidance increases when financial constraints

heighten in the period of high uncertainty. In this section, we examine whether cash holdings

serve as a moderating channel to alleviate the positive impact of policy uncertainty on firm tax

avoidance. Specifically we argue if firms have a precautionary motive to hold more cash when

financial constraints increase (Opler et al., 1999; Bates et al., 2009), the effects of policy

uncertainty on firm tax avoidance should be less severe for cash-rich firms. Hence, the

moderating effect of cash holdings on the relation between policy uncertainty and tax

avoidance is expected to be stronger for more financially constrained firms.

3 Following Harford et al. (2014), the spread of commercial and industrial loan rates (on loans greater than USD 1 million) over the federal funds rate are collected from the Federal Reserve Senior Loan Office (SLO) survey published in January, 2017.

23

This analysis will help explain why many U.S (especially multinational) firms hold

more cash in the period of higher uncertainty instead of engaging more in tax avoidance. This

is because the latter activities are usually challenged by the foreign and state jurisdictions. For

example, the recent Wall Street Journal highlights that France has challenged Google for its

tax avoidance activities and demanded €1.7 billion in back taxes and penalties. Likewise, Apple

has been challenged by tax authorities in Australia and Amazon.com has been challenged by

the France and various U.S states (Pfanner, 2012). We argue that if those firms have not saved

enough cash, then paying the tax, including both the penalties and interests, could force them

to forgo capital spending or raise external funds (which are very costly, especially in the period

of high economic policy uncertainty). As a result, when faced with greater uncertainty, firms

may have precautionary motives to have sufficient cash on hand to avoid paying more tax

penalties resulting from their tax avoidance activities.

To test this hypothesis, we estimate the following model:

TA_CETRi,t = α0 + β1CASHi,t + β2PUi,t*CASHi,t + βjCONTROL VARIABLESj,i,t + �iFirmi +

δtYeart + ϵi,t (7)

In this Equation (7), all the variables are the same as in Equation 1 and the variables of interest

is the interaction term, PU*CASH, that capture the impact of cash holdings on the association

between policy uncertainty and tax avoidance. If cash holdings weaken the positive impact of

policy uncertainty on capital investment, the coefficient of the interaction term should be

negative.

We further divide the sample into financial constrained firms (FC) and unconstrained

firms (UC) following Almeida et al. (2004) and Denis and Sibilkov (2010) to test if the

moderating role of cash holding is more pronounced for more financially constrained firms.

Since there is no agreement in the literature regarding the classification of constrained versus

unconstrained firms, we rely on the following five well documented categorization schemes,

24

including firm size, dividend payout ratio, age, debt and paper ratings. According to Almeida

et al. (2004) and Hadlock and Pierce (2010), financially constrained firms are those that are

small, young, low both short-term and long-term credit quality and hence are more vulnerable

to capital market frictions. The four classification schemes are classified as below:

Scheme #1: We rank firms based on their asset size per year and assign to the financially

constrained (unconstrained) group those firms in the bottom (top) three deciles of the

annual size distribution.

Scheme #2: Similarly, firms are ranked based on their payout ratio for every year over

the 1988-2014 and allocate those firms in the bottom (top) three decides to the

financially constrained (unconstrained) categories. The payout ratio is computed by

taking the common dividend paid divided by operating income. Note that firms who do

not pay dividend for a particular year are assigned zero value for their payout ratio.

Scheme #3: We classify financially unconstrained firms are those that have their debt

rated by Standard & Poor’s (S&P Long-term Senior Debt rating) and their debt not in

default (rating of “D”). Firms that do not have their debt rated but report positive long-

term debt are defined as financially constrained.

Scheme #4: Firms are classified as financially unconstrained if they have their short-

term rated by S&P’s and their debt is not in default. Firms are defined as financially

constrained if they have positive short-term debt but are not rated by S&P’s.

Scheme #5: We calculate firm age by taking the difference between the year of interest

and IPO year. For every year in the sample period, we again rank firms by their ages

and assign those firms in bottom (top) three deciles into financially constrained

(unconstrained) groups.

We then rerun Equation (7) separately on the two groups for each classification scheme

and their results are reported in Table 7 below.

25


The absence of all macro-level independent variables described in Equation (2) above allows

us to include both year and industry fixed effects in the regression models. In relation to the

full sample result, Column (1) shows that the coefficient on the interaction term, PU*CASH, is

negative and statistically significant as expected suggesting the mitigating role of cash holdings

on the impact of policy uncertainty on tax avoidance. Columns (2) through (11) of Table 7

present regression results on subgroups of constrained (FC) and unconstrained (UC) firms

using five aforementioned classification schemes. We find that the coefficients of the

interaction term, PU*CASH, are negative and statistically significant for the FC subsample

while obtaining statistically insignificant coefficient for the interaction term PU*CASH for

financially unconstrained firms. In other words, the results indicate that the increase in cash

reserves is likely to discourage financially constrained firms to engage in tax avoidance

activities induced by higher policy uncertainty. The results strongly support our hypothesis that

cash holdings serve as a mechanism to mitigate the positive association between policy

uncertainty and tax avoidance, and the moderating impact is more pronounced for more

financially constrained firms.

4. Conclusions

In this paper we empirically investigate the impact of economic policy uncertainty on firm tax

avoidance. We find a strong and economically meaningful positive association between

economic policy uncertainty and firm tax avoidance. This relation is robust to alternative

measures of tax avoidance and several tests to address endogeneity concerns that arise from

the possibility that the measure of policy uncertainty may inadvertently capture economic

uncertainty. In addition, firms use several strategies to avoid tax including tax deferrals and

shelters. Further analysis shows that the effect of policy uncertainty on tax avoidance is less

26

pronounced for firms with higher level of cash holdings. Overall, our findings shed more lights

on the importance of uncertainty around government policy in determining firm tax avoidance

activities, thereby contributing to the emerging literature on the economic effect of policy

uncertainty.

27

References

Almeida, H., Campello, M., & Weisbach, M. S. (2004). The cash flow sensitivity of cash. Journal of Finance, 59(4), 1777-1804.

Armstrong, C. S., Blouin, J. L., Jagolinzer, A. D., & Larcker, D. F. (2015). Corporate governance, incentives, and tax avoidance. Journal of Accounting and Economics, 60(1), 1-17.

Auerbach, A. J., & Poterba, J. M. (1987). Tax-loss carryforwards and corporate tax incentives. In M. Feldstein (Ed.), The effects of taxation on capital accumulation (pp. 305-343). Chicago: The University of Chicago Press.

Badertscher, B. A., Katz, S. P., & Rego, S. O. (2013). The separation of ownership and control and corporate tax avoidance. Journal of Accounting and Economics, 56(2–3), 228-250.

Baker, S. R., Bloom, N., & Davis, S. J. (2016). Measuring economic policy uncertainty. Quarterly Journal of Economics, 131(4), 1593-1636.

Bates, T. W., Kahle, K. M., & Stulz, R. M. (2009). Why do U.S. firms hold so much more cash than they used to? Journal of Finance, 64(5), 1985-2021.

Cen, L., Maydew, E. L., Zhang, L., & Zuo, L. (2017). Customer–supplier relationships and corporate tax avoidance. Journal of Financial Economics, 123(2), 377-394.

Chen, S., Chen, X., Cheng, Q., & Shevlin, T. (2010). Are family firms more tax aggressive than non-family firms? Journal of Financial Economics, 95(1), 41-61.

Cheng, C. S. A., Huang, H. H., Li, Y., & Stanfield, J. (2012). The effect of hedge fund activism on corporate tax avoidance. Accounting Review, 87(5), 1493-1526.

Denis, D. J., & Sibilkov, V. (2010). Financial constraints, investment, and the value of cash holdings. Review of Financial Studies, 23(1), 247-269.

Dyreng, S., Hanlon, M., & Maydew, E. L. (2010). The effects of executives on corporate tax avoidance. Accounting Review, 85(4), 1163-1189.

Dyreng, S. D., Hoopes, J. L., & Wilde, J. H. (2016). Public pressure and corporate tax behavior. Journal of Accounting Research, 54(1), 147-186.

Edwards, A. S., Schwab, C. M., & Shevlin, T. (2015). Financial Constraints and Cash Tax Savings. The Accounting Review, 0(0), null.

Gallemore, J., Maydew, E. L., & Thornock, J. R. (2014). The reputational costs of tax avoidance. Contemporary Accounting Research, 31(4), 1103-1133.

Graham, J. R. (1996a). Debt and the marginal tax rate. Journal of Financial Economics, 41(1), 41-73.

Graham, J. R. (1996b). Proxies for the corporate marginal tax rate. Journal of Financial Economics, 42(2), 187-221.

Graham, J. R. (2000). How big are the tax benefits of debt? Journal of Finance, 55(5), 1901-1941.

Graham, J. R., Hanlon, M., Shevlin, T., & Shroff, N. (2014). Incentives for tax planning and avoidance: Evidence from the Field. Accounting Review, 89(3), 991-1023.

Gulen, H., & Ion, M. (2016). Policy uncertainty and corporate investment. Review of Financial Studies, 29(3), 523-564.

Gungoraydinoglu, A., Çolak, G., & Öztekin, Ö. (2017). Political environment, financial intermediation costs, and financing patterns. Journal of Corporate Finance, 44, 167-192.

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

28

Hadlock, C. J., & Pierce, J. R. (2010). New evidence on measuring financial constraints: Moving beyond the KZ index. Review of Financial Studies, 23(5), 1909-1940.

Hanlon, M., Hoopes, J. L., & Shroff, N. (2014). Effect of tax authority monitoring and enforcement on financial reporting quality. Journal of the American Taxation Association, 36(2), 137-170.

Hanlon, M., Maydew, E. L., & Saavedra, D. (2017). The taxman cometh: Does tax uncertainty affect corporate cash holdings? Review of Accounting Studies, 22(3), 1198-1228.

Hanlon, M., Mills, L. F., & Slemrod, J. B. (2005). An empirical examination of corporate tax noncompliance. In A. Auerbach, J. R. Hines, & J. Slemrod (Eds.), Taxing Corporate Income in the 21st Century (pp. 171–210): Cambridge, MA: Cambridge University Press.

Hanlon, M., & Slemrod, J. (2009). What does tax aggressiveness signal? Evidence from stock price reactions to news about tax shelter involvement. Journal of Public Economics(93), 126-141.

Hasan, I., Hoi, C.-K., Wu, Q., & Zhang, H. A. O. (2017). Does social capital matter in corporate decisions? Evidence from corporate tax avoidance. Journal of Accounting Research, 55(3), 629-668.

Jacob, J. (1996). Taxes and transfer pricing: Income shifting and the volume of intrafirm transfers. Journal of Accounting Research, 34(2), 301-312.

Julio, B., & Yook, Y. (2012). Political uncertainty and corporate investment cycles. Journal of Finance, 67(1), 45-83.

Jurado, K., Ludvigson, S. C., & Ng, S. (2015). Measuring uncertainty. American Economic Review, 105(3), 1177-1216.

Law, K. K. F., & Mills, L. F. (2015). Taxes and financial constraints: Evidence from linguistic cues. Journal of Accounting Research, 53(4), 777-819.

Lisowsky, P. (2010). Seeking shelter: Empirically modeling tax shelters using financial statement information. Accounting Review, 85(5), 1693-1720.

Lisowsky, P., Robinson, L., & Schmidt, A. (2013). Do publicly disclosed tax reserves tell us about privately disclosed tax shelter activity? Journal of Accounting Research, 51(3), 583-629.

McCarty, N. M., Poole, K. T., & Rosenthal, H. (1997). Income redistribution and the realignment of Americanpolitics.Washington, DC: SEI Press., 1-72.

Mills, L. F., & Newberry, K. J. (2004). Do foreign multinationals' tax incentives influence their U.S. income reporting and debt policy? National Tax Journal, 57(1), 89-107.

Newberry, K. J. (1998). Foreign tax credit limitations and capital structure decisions. Journal of Accounting Research, 36(1), 157-166.

Newberry, K. J., & Dhaliwal, D. S. (2001). Cross-Jurisdictional Income Shifting by U.S. Multinationals: Evidence from International Bond Offerings. Journal of Accounting Research, 39(3), 643-662.

Nguyen, N. H., & Phan, H. V. (2017). Policy uncertainty and mergers and acquisitions. Journal of Financial and Quantitative Analysis Forthcoming.

Opler, T., Pinkowitz, L., Stulz, R., & Williamson, R. (1999). The determinants and implications of corporate cash holdings. Journal of Financial Economics, 52(1), 3-46.

Pástor, Ľ., & Veronesi, P. (2012). Uncertainty about government policy and stock prices. The Journal of Finance, 67(4), 1219-1264.

Pástor, Ľ., & Veronesi, P. (2013). Political uncertainty and risk premia. Journal of Financial Economics, 110(3), 520-545.

Petersen, M. A. (2009). Estimating standard errors in finance panel data sets: Comparing approaches. Review of Financial Studies, 22(1), 435-480.

29

Pfanner, E. (2012). European countries seek more taxes from U.S. multinationals. The Wall Street Journal(November 19).

Rego, S. O. (2003). Tax-Avoidance Activities of U.S. Multinational Corporations*. Contemporary Accounting Research, 20(4), 805-833.

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

30

Tables

Table 1: Summary Statistics - Period 1988-2014

Variable Obs Mean Std. Dev. Min Max

Tax Avoidance

TA_CETR 65,822 -0.2668 0.2219 -1.0000 0.0000

TA_ETR 65,822 -0.3031 0.1764 -1.0000 0.0000

DTAX 34,577 -0.0011 0.0636 -0.1823 0.2405

DEFERRAL 63,438 -0.0108 0.3440 -1.9599 1.4488

TA_CETR5 65,822 -0.2836 0.3478 -2.4906 0.4021

LOW_CETR 65,822 0.1700 0.3756 0.0000 1.0000

CETR_DODGER 65,822 0.0670 0.2500 0.0000 1.0000

SHELTER_DUMMY 34,921 0.3161 0.4650 0.0000 1.0000

SHELTER_LEVEL 34,921 3.5603 1.9664 0.4430 13.1800

CASH_RATIO 57,369 -0.2037 0.1695 -0.7985 0.0000

CTD 65,822 0.1142 0.5664 -0.8095 4.7362

CURRENT_ETR 63,336 -0.2904 0.2936 -1.9639 0.6485

Firm Controls

SIZE 65,822 5.7089 2.1032 -4.5805 14.4142

MTB 65,822 2.9500 2.9009 0.4010 19.0110

LEVERAGE 65,822 0.2107 0.2872 0.0000 1.9459

CASH 65,822 0.1974 0.2874 0.0001 1.8315

NOL 65,822 0.3037 0.4599 0.0000 1.0000

ROA 65,822 0.1258 0.1507 -0.1363 1.0173

EQUITY_INCOME 65,822 0.0008 0.0042 -0.0071 0.0302

PPE 65,822 0.3389 0.3459 0.0030 2.2742

INTANGIBLE 65,822 0.1558 0.2572 0.0000 1.5791

FOREIGN_INCOME 65,822 0.4909 8.2173 -77.9279 1027.0520

31

Table 2: Policy Uncertainty and Corporate Tax Avoidance

(1) (2) (3) (4) (5)

VARIABLES TA_CETR (t)

TA_CETR (t)

TA_CETR (t+1)

TA_CETR (t+2)

TA_CETR (t+3)

PU 0.0215*** 0.0197*** 0.0362*** 0.0296*** 0.0035 [6.34] [5.12] [8.75] [6.85] [0.74]

SIZE

-0.0004 0.0031*** 0.0037*** 0.0038***

[-0.63] [4.12] [4.76] [4.59]

MTB

0.0074*** 0.0083*** 0.0050*** 0.0042***

[16.74] [17.06] [9.85] [7.70]

LEVERAGE

-0.0047 -0.0016 -0.0003 0.0069

[-0.87] [-0.25] [-0.04] [0.94]

CASH

0.0478*** 0.0631*** 0.0569*** 0.0414***

[10.77] [11.85] [10.25] [7.00]

NOL

0.0610*** 0.0511*** 0.0393*** 0.0322***

[21.95] [17.09] [12.44] [9.65]

ROA

-0.0377*** -0.2569*** -0.2079*** -0.1782***

[-4.00] [-24.17] [-19.36] [-15.65]

EQUITY_INCOME

0.8410*** 0.0930 -0.2355 -0.1463

[2.88] [0.27] [-0.65] [-0.39]

PPE

0.0487*** 0.0751*** 0.0617*** 0.0547***

[10.17] [13.96] [10.90] [8.91]

INTANGIBLE

-0.0035 0.0113* 0.0104 0.0141*

[-0.62] [1.78] [1.54] [1.91]

FOREIGN_INCOME

-0.0003*** -0.0003*** -0.0003*** -0.0005***

[-3.89] [-3.48] [-3.07] [-3.41]

Observations 80,142 65,822 51,724 44,823 39,885

Adjusted R-squared 0.013 0.044 0.061 0.042 0.033

SIC3 FE Yes Yes Yes Yes Yes

Year FE No No No No No

Firm Cluster Yes Yes Yes Yes Yes

Robust t-statistics in brackets

*** p<0.01, ** p<0.05, * p<0.1

32

Table 3: Control for Macro-economic Uncertainty (1) (2) (3) (4) (5) (6) (7) (8)

Original PU Cleaner PU

VARIABLES TA_CETR (t) TA_CETR (t)

PU 0.0197*** -0.0050 0.0195*** 0.0205*** 0.0159*** 0.0179*** 0.0102** 0.0319***

[5.10] [-1.18] [5.02] [5.27] [4.16] [4.28] [2.02] [5.09]

ELECYEAR -0.0002 0.0153***

[-0.14] [7.30] GDPDIS 0.0260*** 0.0178***

[9.88] [5.40] SDPROFIT -0.0009 -0.0006

[-0.54] [-0.33] VXO -0.0052* 0.0970***

[-1.92] [6.24] SDRETURN -0.0820*** -0.0931***

[-13.06] [-11.27] JLN -0.0043* -0.0895***

[-1.76] [-6.53] SIZE -0.0004 -0.0015** -0.0004 -0.0004 -0.0018** -0.0005 -0.0022*** -0.0007 [-0.63] [-2.17] [-0.61] [-0.65] [-2.56] [-0.78] [-3.10] [-0.93]

MTB 0.0074*** 0.0075*** 0.0074*** 0.0073*** 0.0073*** 0.0077*** 0.0077*** 0.0075*** [16.73] [16.90] [16.71] [16.70] [16.51] [16.52] [16.46] [16.02]

LEVERAGE -0.0047 -0.0008 -0.0047 -0.0045 0.0002 -0.0080 -0.0006 -0.0034 [-0.86] [-0.16] [-0.87] [-0.82] [0.03] [-1.41] [-0.11] [-0.59]

CASH 0.0478*** 0.0454*** 0.0478*** 0.0477*** 0.0446*** 0.0475*** 0.0428*** 0.0470*** [10.77] [10.24] [10.77] [10.74] [10.04] [10.42] [9.39] [10.22]

NOL 0.0610*** 0.0577*** 0.0611*** 0.0610*** 0.0578*** 0.0646*** 0.0603*** 0.0604*** [21.95] [20.55] [21.91] [21.92] [20.70] [22.13] [20.36] [20.40]

ROA -0.0377*** -0.0335*** -0.0377*** -0.0376*** -0.0308*** -0.0377*** -0.0291*** -0.0372*** [-4.00] [-3.56] [-4.00] [-3.99] [-3.27] [-3.88] [-2.99] [-3.72]

EQUITY_INCOME 0.8410*** 0.8115*** 0.8418*** 0.8377*** 0.8392*** 0.8133*** 0.8009*** 0.7936** [2.88] [2.80] [2.88] [2.87] [2.89] [2.71] [2.68] [2.53]

PPE 0.0487*** 0.0495*** 0.0487*** 0.0487*** 0.0487*** 0.0493*** 0.0495*** 0.0455*** [10.17] [10.37] [10.16] [10.15] [10.21] [9.90] [9.99] [9.02]

INTANGIBLE -0.0035 -0.0109* -0.0034 -0.0035 -0.0120** -0.0032 -0.0155*** -0.0057 [-0.62] [-1.95] [-0.61] [-0.63] [-2.13] [-0.54] [-2.65] [-0.97]

FOREIGN_INCOME -0.0003*** -0.0003*** -0.0003*** -0.0003*** -0.0003*** -0.0003*** -0.0002*** -0.0002**

[-3.89] [-3.85] [-3.88] [-3.90] [-3.73] [-3.07] [-2.95] [-2.02] Observations 65,822 65,822 65,822 65,822 65,822 60,295 60,295 56,007

Adjusted R-squared 0.044 0.046 0.044 0.044 0.047 0.044 0.049 0.042

SIC3 FE Yes Yes Yes Yes Yes Yes Yes Yes

Year FE No No No No No No No No

Firm Cluster Yes Yes Yes Yes Yes Yes Yes Yes

Robust t-statistics in brackets *** p<0.01, ** p<0.05, * p<0.1

33

Table 4: 2SLS - Political Polarization IV

(1) (2) (3) (4) (5) (6) (7) (8) Include Canada PU in the 1st stage regression? No Yes No Yes

VARIABLES TA_CETR

(t)

TA_CETR

(t+1)

TA_CETR

(t+2)

TA_CETR

(t+3)

TA_CETR

(t+4) TA_CETR

(t)

TA_CETR

(t)

TA_CETR

(t)

FPU (Senate) 0.0267*** 0.0597*** 0.0671*** 0.0383*** -0.0175**

[4.90] [10.60] [11.76] [6.42] [-2.56]

FPU (Senate) 0.0339***

[6.38]

FPU (House) 0.0275***

[5.01]

FPU (House) 0.0318*** [6.01]

SIZE -0.0004 0.0014 0.0022** 0.0019* 0.0023** -0.0005 -0.0005 -0.0005 [-0.50] [1.49] [2.14] [1.86] [2.13] [-0.61] [-0.52] [-0.58]

MTB 0.0060*** 0.0075*** 0.0040*** 0.0037*** 0.0026*** 0.0061*** 0.0061*** 0.0061*** [11.50] [12.87] [6.50] [5.90] [4.04] [11.60] [11.51] [11.57]

LEVERAGE 0.0036 0.0074 0.0077 0.0156* 0.0293*** 0.0041 0.0037 0.0039 [0.50] [0.88] [0.90] [1.67] [2.95] [0.59] [0.52] [0.56]

CASH 0.0460*** 0.0585*** 0.0489*** 0.0318*** 0.0355*** 0.0457*** 0.0460*** 0.0458*** [7.79] [7.92] [6.45] [3.99] [4.35] [7.73] [7.78] [7.75]

NOL 0.0463*** 0.0357*** 0.0216*** 0.0188*** 0.0209*** 0.0459*** 0.0462*** 0.0460*** [13.45] [9.74] [5.69] [4.84] [5.17] [13.35] [13.43] [13.38]

ROA -0.0179 -0.2551*** -0.2060*** -0.1756*** -0.1493*** -0.0172 -0.0178 -0.0175 [-1.41] [-17.16] [-14.10] [-11.56] [-9.24] [-1.36] [-1.40] [-1.38]

EQUITY_INCOME 0.4451 -0.0739 -0.2759 -0.3378 -0.7284 0.4424 0.4445 0.4435 [1.22] [-0.18] [-0.64] [-0.73] [-1.44] [1.21] [1.22] [1.21]

PPE 0.0491*** 0.0739*** 0.0668*** 0.0572*** 0.0452*** 0.0492*** 0.0491*** 0.0492*** [7.97] [10.34] [9.13] [7.37] [5.44] [7.98] [7.97] [7.98]

INTANGIBLE -0.0153** -0.0018 -0.0001 0.0022 0.0095 -0.0156** -0.0154** -0.0155** [-2.22] [-0.22] [-0.01] [0.24] [1.02] [-2.27] [-2.23] [-2.25]

FOREIGN_INCOME -0.0002** -0.0000 -0.0002* -0.0005*** -0.0005*** -0.0002** -0.0002** -0.0002** [-2.11] [-0.10] [-1.88] [-3.06] [-4.18] [-2.13] [-2.12] [-2.13] Observations 36,267 29,251 25,975 23,872 20,943 36,267 36,267 36,267


SIC3 FE Yes Yes Yes Yes Yes Yes Yes Yes

Year FE No No No No No No No No

Firm Cluster Yes Yes Yes Yes Yes Yes Yes Yes


*** p<0.01, ** p<0.05, * p<0.1

34

Table 5: Alternative Measures of Tax Avoidance (1) (2) (3) (4) (5) (6) (7)

(8) (9) (10)

OLS

Logit

VARIABLES TA_ETR (t)

DTAX (t) DEFERRAL (t)

TA_CETR5 (t)

SHELTER_LEVEL (t)

CASH_ RATIO (t)

CURRENT_ETR (t)

LOW_CETR(t) CETR_DODGER

(t) SHELTER_DUMMY

(t)

PU 0.0224*** 0.0034** 0.0175*** 0.0258*** 0.2980*** 0.0389*** 0.0409***

0.1828*** 0.4414*** 0.5728*** [6.81] [2.30] [2.92] [4.54] [6.58] [13.68] [7.76]

[3.32] [5.74] [7.03]

SIZE -0.0009* 0.0031*** 0.0019** 0.0035*** 0.4615*** -0.0030*** -0.0018**

-0.2579*** -0.2710*** 1.0004*** [-1.68] [11.30] [2.39] [3.76] [46.65] [-5.45] [-2.06]

[-24.80] [-20.19] [34.59]

MTB 0.0056*** 0.0004 -0.0030*** 0.0073*** -0.0239*** 0.0055*** 0.0031***

0.0913*** 0.0621*** -0.0755*** [14.55] [1.52] [-5.07] [12.14] [-3.37] [16.10] [5.36]

[17.18] [8.73] [-6.49]

LEVERAGE -0.0120*** -0.0060** 0.0068 -0.0040 -0.5406*** -0.0190*** 0.0129*

0.2780*** 0.2988*** -0.8221*** [-2.65] [-2.41] [0.90] [-0.54] [-6.88] [-4.10] [1.84]

[3.92] [3.00] [-5.20]

CASH 0.0915*** 0.0004 -0.0787*** 0.0225*** 0.6685*** 0.0925*** 0.0111*

1.0398*** 0.2087** 0.1476 [22.81] [0.14] [-11.99] [3.62] [8.85] [22.54] [1.76]

[17.53] [2.49] [1.42]

NOL 0.0284*** 0.0004 0.0036 0.0633*** 1.5040*** 0.0537*** 0.0552***

0.7777*** 0.4227*** 2.2244*** [12.69] [0.36] [0.97] [16.23] [35.04] [25.71] [15.14]

[22.06] [9.50] [36.46]

ROA -0.2834*** 0.2101*** 0.2187*** 0.0731*** 1.5740*** -0.3845*** -0.1513***

-2.8176*** -2.6674*** 2.6749*** [-31.09] [28.83] [13.38] [5.04] [10.43] [-42.18] [-10.55]

[-18.59] [-10.25] [11.45]

EQUITY_INCOME 1.0310*** -0.1610 0.1938 0.8702** 10.2973*** -0.2291 1.1509***

10.7198** 14.8570*** 19.5944*** [3.85] [-1.43] [0.52] [2.33] [3.97] [-0.96] [3.07]

[2.54] [2.79] [3.45]

PPE 0.0353*** -0.0016 0.0046 0.0444*** 0.0014 0.1242*** 0.0755***

0.5631*** 0.6301*** -0.0828 [8.96] [-0.72] [0.72] [6.93] [0.02] [30.41] [12.69]

[8.93] [7.91] [-0.50]

INTANGIBLE 0.0131*** 0.0004 -0.0176** -0.0150* 0.2734*** 0.0266*** -0.0071

-0.3601*** -0.7875*** -0.1344 [2.79] [0.15] [-2.24] [-1.94] [2.94] [5.85] [-0.96]

[-4.37] [-6.39] [-0.89]

FOREIGN_INCOME -0.0000 0.0001 -0.0001 -0.0004*** 0.0043** -0.0001 -0.0003**

-0.0233 -0.0328** 0.1200*** [-0.05] [1.24] [-1.32] [-5.30] [2.13] [-0.86] [-2.54]

[-1.53] [-2.29] [2.92]

Observations 65,822 34,577 63,438 65,822 34,921 57,369 63,336

65,822 65,822 34,921

Adjusted R-squared 0.061 0.187 0.008 0.021 0.463 0.150 0.025

SIC3 FE Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes

Year FE No No No No No No No

No No No

Firm Cluster Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes

Pseudo R-squared

0.10 0.085 0.45


*** p<0.01, ** p<0.05, * p<0.1

35

Table 6: Policy Uncertainty and Credit Market Conditions

VARIABLES CISPREAD (t)

(1) (2) (3) (4) (5) (6) (7) (8)

PU 0.4086*** 0.4139*** 0.2249*** 0.4144*** 0.4550*** 0.4008*** 0.3948*** 0.1653*** [6.22] [5.94] [3.68] [5.96] [7.12] [7.12] [5.11] [2.94]

ELECYEAR -0.0178 0.0609 [-0.40] [1.38]

GDPDIS 0.2882*** 0.4626*** [7.66] [10.25]

SDPROFIT 0.0093 -0.1598*** [0.37] [-7.90]

VXO -0.1904*** 0.4302* [-3.76] [1.81]

SDRETURN -0.9374*** -0.0013 [-6.30] [-0.01]

JLN -0.1026** -0.5406** [-2.16] [-2.50]

Constant -1.2993*** -1.3191*** -1.8063*** -1.3537*** -0.9496*** -1.3067*** -0.9664*** -1.5896*** [-4.28] [-4.13] [-6.71] [-4.07] [-3.06] [-5.00] [-2.82] [-6.12]

Quarter FE Yes Yes Yes Yes Yes Yes Yes Yes

Observations 122 118 118 118 118 114 102 102


Robust t-statistics in brackets *** p<0.01, ** p<0.05, * p<0.1

36

Table 7: The Moderating of Cash Holdings and Financial Constraints Channel

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) TA_CETR (t) FC Criteria

Full Size Payout Ratio Firm Age Debt Rating Paper Rating

Sample FC UC FC UC FC UC FC UC FC UC PU*CASH -0.0368*** -0.0546** 0.0133 -0.0396** -0.0200 -0.0652** -0.0232 -0.0305** -0.0467 -0.0406*** 0.1567** [-2.81] [-1.99] [0.63] [-2.01] [-0.42] [-2.36] [-0.41] [-2.16] [-1.07] [-3.02] [2.19] SIZE -0.0036*** -0.0134*** 0.0006 -0.0073*** 0.0048*** 0.0047* -0.0084*** -0.0062*** -0.0020 -0.0035*** 0.0059* [-5.00] [-5.18] [0.32] [-6.53] [3.74] [1.91] [-3.57] [-6.53] [-1.00] [-4.10] [1.92] MTB 0.0074*** 0.0117*** 0.0042*** 0.0140*** 0.0068*** 0.0085*** 0.0151*** 0.0083*** 0.0043*** 0.0074*** 0.0057*** [16.69] [11.20] [6.98] [14.88] [5.07] [5.03] [7.26] [17.52] [3.32] [16.23] [3.00] LEVERAGE 0.0062 -0.0205* 0.0089 0.0020 0.1073*** 0.0242** 0.0323 -0.0044 0.0372*** 0.0051 0.0557* [1.14] [-1.69] [1.09] [0.25] [6.69] [2.37] [1.49] [-0.74] [2.59] [0.91] [1.95] CASH 0.2074*** 0.2578** 0.0038 0.2055** 0.0937 0.3233** 0.1983 0.1824*** 0.2257 0.2240*** -0.7271** [3.48] [2.04] [0.04] [2.28] [0.43] [2.57] [0.76] [2.84] [1.12] [3.66] [-2.13] NOL 0.0532*** 0.0805*** 0.0211*** 0.0597*** 0.0246*** 0.0476*** 0.0592*** 0.0605*** 0.0281*** 0.0562*** 0.0209*** [18.69] [14.96] [5.00] [15.81] [5.14] [6.88] [7.26] [18.16] [5.67] [18.66] [2.95] ROA -0.0233** 0.1428*** -0.1122*** 0.0249* -0.0806*** -0.0494** -0.0802** -0.0159 -0.0237 -0.0192** -0.1121** [-2.47] [7.67] [-7.09] [1.76] [-2.93] [-2.34] [-2.06] [-1.60] [-0.87] [-2.01] [-2.00] EQUITY_INCOME 0.7475** 0.9562* 0.2540 1.3836*** 0.5151 0.6655 1.0504 0.7456** 0.5919 0.7583** 0.4938 [2.56] [1.66] [0.68] [3.16] [1.13] [0.88] [1.23] [2.07] [1.28] [2.42] [0.71] PPE 0.0511*** 0.0462*** 0.0658*** 0.0473*** 0.0717*** 0.0379*** 0.1014*** 0.0489*** 0.0654*** 0.0503*** 0.0646*** [10.73] [4.83] [8.58] [7.11] [5.98] [3.92] [5.69] [9.60] [6.02] [10.30] [3.03] INTANGIBLE -0.0243*** -0.0111 -0.0327*** -0.0330*** -0.0377** -0.0533*** -0.0148 -0.0174*** -0.0441*** -0.0245*** -0.0349 [-4.30] [-0.91] [-3.84] [-3.91] [-2.39] [-4.75] [-0.63] [-2.81] [-3.40] [-4.23] [-1.45] FOREIGN_INCOME -0.0003*** -0.0001 -0.0003*** -0.0004 -0.0003*** -0.0007 0.0015*** 0.0002 -0.0003*** -0.0005 -0.0002*** [-3.46] [-0.02] [-4.10] [-0.52] [-5.13] [-1.54] [4.47] [0.42] [-4.51] [-1.12] [-3.73] Observations 65,822 19,788 19,692 36,081 17,563 9,492 6,771 50,210 15,612 60,486 5,336 Adjusted R-squared 0.060 0.057 0.092 0.054 0.101 0.078 0.076 0.059 0.078 0.059 0.100 SIC3 FE Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Year FE Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Firm Cluster Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Robust t-statistics in brackets

*** p<0.01, ** p<0.05, * p<0.1

37

Appendix Appendix A1 Variable definitions

This table defines all variables used in Equation 1. The databases used to source the items necessary to calculate each variable are also provided.

Panel A: Tax avoidance measures

Variables Measures Definition TA_CETR Cash effective tax rate Cash tax paid (txpd) divided by pre-tax book income (pi) less special items (spi). When the denominator

is zero or negative, CETR is set as missing. CETR is truncated to the range [0,1]. TA_CETR is defined as -1 times CETR.

TA_ETR Effective tax rate Total tax expense (txt) divided by pre-tax income, which is the difference between pre-tax book income (pi) and special items (spi). If the denominator is zero or negative, ETR is set as missing. ETR is truncated to the range [0,1]. TA_ETR is defined as -1 times ETR.

DTAX Discretionary permanent book-tax difference

DTAX is the residuals (�) of the following regression estimated by two-digit SIC code and fiscal year where all variables (including the intercept ( ��)) are scaled by beginning-of-year total assets (at) following Frank et al. (2009):

��,� = �� + �� + �� + �� + �� + ��∆��

+ �� + �� Where: �� = pre-tax book income (pi) for firm i in year t; ��=current deferral tax expenses (txfed) for firm i in year t; �� =current foreign tax expense (txfo) for firm i in year t; ��= deferred tax expense (txdi) for firm i in year t; ��= statutory tax rate in year t (35%); ��= goodwill and other intangibles (intan) for firm i in year t; ��= income (loss) reported under the equity method (esub) for firm i in year t; ��= income (loss) attributable to minority interest (mii) for firm i in year t; ��= current state income tax expense (txs) for firm i in year t; ∆��=change in the net operating loss carryforwards (tlcf) for firm i in year t; ��= one-year lagged PERMDIFF for firm i in year t; and �� = discretionary permanent difference (��) for firm i in year t. Following Frank et al. (2009) and Hassan et al. (2017), missing values of these variables are handled as follows: If minority interest (mii), current foreign tax expense (txs), income from unconsolidated entities (esub) or current state tax expense (txs) is missing on Compustat, we set MI, CFOR, UNCON or CSTE to zero. If current deferral tax expense (TXFED) is missing on Compustat, we set the value of CFTE to: total tax expense (txt) less current foreign tax expense (txfo) less current state tax expense (txs) less deferred tax expense (txdi). If information for goodwill and other intangibles (INTANG) is missing on Compustat, we set the value for INTANG to zero. If INTANG=C, then we set the value of INTANG to that for goodwill (GDWL).

DEFERRAL -1 times the ratio of deferred tax expense to pre-tax income adjusted for special items (txdfed+txdfo)/(pi-spi); if missing (txdfed+txdfo) then txdi/(pi-spi)

38

TA_ETR5 Long term effective tax rates Five year effective tax rate: txt/(pi-spi). Both txt and pi-spi are cumulated over five years before calculation.

TA_CETR5 Long term cash effective tax rates

Five year cash ETR: txpd/(pi-spi). Both txpd and pi-spi are cumulated over five years before calculation.

LOW_ETR Bottom quintile of the ETR distribution for all firms

A dummy variable which equals to 1 if a firm’s ETR belongs to the bottom quintile of the ETR distribution for all firms with the same two-digit SIC code in a given year and zero otherwise.

LOW_CETR Bottom quintile of the CETR distribution for all firms

A dummy variable which equals to 1 if a firm’s CETR belongs to the bottom quintile of the CETR distribution for all firms with the same two-digit SIC code in a given year and zero otherwise.

ETR_DODGER Tax dodgers A dummy variable which equals to 1 if a firm has a positive pre-tax profit and a zero ETR in a given year and zero otherwise.

CETR_DODGER Tax dodgers A dummy variable which equals to 1 if a firm has a positive pre-tax profit and a zero CETR in a given year and zero otherwise.

SHELTER_DUMMY SHELTER_LEVEL CASH_RATIO Firm cash ratio Cash tax paid divided by pre-tax operating cash flows adjusted for extraordinary items and discontinued

operations. This is txpd/(oancf+txpd-xidoc). CTD Cash tax differential Cash tax differential of Henry and Sansing (2014) which is estimated as the difference between cash

taxes paid and the product of statutory tax rate and pre-tax income, scaled by lagged total assets. (txpd-0.35*(pi-spi)).

CURRENT_ETR Current effective tax rate (txt-txdi)/(pi-spi) Panel B: Economic Policy Uncertainty PU News-based economic policy

uncertainty Count the numbers of key words on the 10 leading newspapers and then scale by the total numbers of articles in the same newspaper and month, which yields a monthly policy uncertainty series for each newspaper. These monthly newspaper-level uncertainty series are then standardized by unit standard deviation from 1985 to 2010 and then averaged across the ten papers per month. Finally, the series are then normalized to a mean of 100 from 1985 to 2009.

Panel C: Control Variables SIZE Firm Size Natural logarithm of the market value of equity (prcc_f * csho) for a firm at the beginning of the year. MTB Market to book ratio Market value of equity (prcc_f * csho), scaled by book value of equity. LEVERAGE Leverage Long term debt (dltt) scaled by lagged assets (at) CASH Cash holding Firm cash holding defined as cash and marketable securities (che) divided by lagged assets (at) NOL Net loss carry forward A dummy variable that equals to one if loss carry forward (tlcf) for a firm is positive and zero otherwise ROA Return on assets It is measured as operating income (pi-xi) scaled by lagged assets (at) EQUITY_INCOME Equity income Equity income in earnings (esub) for a firm in a given year, scaled by lagged assets (at) PPE Property, plant and equipment Property, plant and equipment (ppent) for a firm in a given year, scaled by lagged assets (at) INTANGIBLE Intangible assets Intangible assets (intan) for a firm in a given year, scaled by lagged assets (at) FOREIGN_INCOME Foreign income Foreign income (pifo) for a firm in a given year, scaled by lagged assets (at). Missing values in pifo are

set to zero.

39

Appendix A2: Control for Firm Fixed Effects (1) (2) VARIABLES TA_CETR (t) TA_CETR (t) PU 0.0179*** 0.0218***

[4.65] [4.26] SIZE 0.0035** 0.0018

[2.50] [1.10] MTB 0.0051*** 0.0053***

[10.14] [9.89] LEVERAGE -0.0089 -0.0075

[-1.44] [-1.15] CASH 0.0113** 0.0074

[2.06] [1.30] NOL 0.0499*** 0.0523***

[15.31] [15.26] ROA 0.1140*** 0.1225***

[10.53] [10.83] EQUITY_INCOME 1.8123*** 1.6672***

[6.00] [5.31] PPE -0.0292*** -0.0281***

[-4.82] [-4.45] INTANGIBLE 0.0025 0.0009

[0.38] [0.13] FOREIGN_INCOME -0.0001 -0.0001

[-0.57] [-0.66] ELECYEAR

0.0104*** [5.03]

GDPDIS

0.0016 [0.46]

SDPROFIT

-0.0018 [-1.08]

VXO

0.0674*** [4.37]

SDRETURN

-0.0639*** [-7.32]

JLN

-0.0642*** [-4.74]

Observations 64,575 59,062 Adjusted R-squared 0.217 0.218 Firm FE Yes Yes Year FE No No Firm Cluster Yes Yes Robust t-statistics in brackets

*** p<0.01, ** p<0.05, * p<0.1

40

Appendix A3: First Difference Model

(1) (2) VARIABLES ΔTA_ETR (t) ΔPU 0.0303*** 0.0433***

[5.37] [6.21] ΔSIZE 0.0280*** 0.0296***

[9.11] [9.02] ΔMTB 0.0012* 0.0013*

[1.80] [1.76] ΔLEVERAGE -0.0107 -0.0115

[-1.60] [-1.64] ΔCASH -0.0185*** -0.0184***

[-2.80] [-2.70] ΔNOL 0.0113*** 0.0121***

[2.59] [2.65] ΔROA 0.2434*** 0.2417***

[16.95] [16.35] ΔEQUITY_INCOME 2.1745*** 2.0430***

[4.65] [4.17] ΔPPE -0.0493*** -0.0487***

[-6.81] [-6.47] ΔINTANGIBLE -0.0041 -0.0048

[-0.54] [-0.60] ΔFOREIGN_INCOME -0.0001 -0.0002

[-0.73] [-1.62] ΔELECYEAR

0.0057** [2.47]

ΔGDPDIS

-0.0110*** [-2.58]

ΔSDPROFIT

0.0017 [0.95]

ΔVXO

0.0234 [1.41]

ΔSDRETURN

-0.0040 [-0.31]

ΔJLN

-0.0152 [-1.08]

Observations 51,433 46,586 Adjusted R-squared 0.022 0.023 SIC3 FE Yes Yes Year FE No No Firm Cluster Yes Yes Robust t-statistics in brackets

*** p<0.01, ** p<0.05, * p<0.1

Economic Policy Uncertainty and Firm Tax Avoidance

Documents