Taxes and Financial Constraints: Evidence from Linguistic Cues

• Published date: 01 September 2015
• DOI: http://doi.org/10.1111/1475-679X.12081
• Author: KELVIN K. F. LAW,LILLIAN F. MILLS
• Date: 01 September 2015

DOI: 10.1111/1475-679X.12081

Journal of Accounting Research

Vol. 53 No. 4 September 2015

Printed in U.S.A.

Taxes and Financial Constraints:

Evidence from Linguistic Cues

KELVIN K. F. LAW∗AND LILLIAN F. MILLS†

Received 28 June 2014; accepted 10 March 2015

ABSTRACT

Using a new measure of financial constraints based on firms’ qualitative dis-

closures, we find that financially constrained firms—firms that use more neg-

ative words in their annual reports—pursue more aggressive tax planning

strategies as evidenced by: (1) higher current and future unrecognized tax

benefits, (2) lower short- and long-run current and future effective tax rates,

(3) increase in tax haven usage for their material operations, and (4) higher

proposed audit adjustments from the Internal Revenue Service. We exploit

∗CentER Fellow, Tilburg University; †The University of Texas at Austin.

Accepted by Philip Berger. We appreciate Scott Dyreng, Feng Li, and William McDonald

for sharing their data. We appreciate helpful comments from an anonymous referee, Shannon

Chen, James Chyz, Michael Drake, Jonathan Feinstein (NTA discussant), Shane Heitzman,

Bradford Hepfer, James Hines (Michigan discussant), Petro Lisowsky, Mark Shuai Ma (AAA

discussant), William Mayew, William McDonald, Edmund Outslay, Andrew Schmidt, Casey

Schwab, Terry Shevlin, Erin Towery, workshop participants at the M-TAXI conference (Uni-

versity of Michigan’s Office of TaxPolicy Research), Tilburg University, University of Southern

California, University of Maryland, Brigham Young University, 2014 American Accounting As-

sociating Annual Meeting at Atlanta, and 2014 National Tax Association Annual Conference,

and from the Georgia Tax Reading Group, Iowa Tax Reading Group, Tennessee Tax Read-

ing Group, and Brigham Young University PhD Prep Track students. The Internal Revenue

Service (IRS) provided confidential tax information to one of the authors pursuant to pro-

visions of the Internal Revenue Code that allow disclosure of information to a contractor to

the extent necessary to perform a research contract for the IRS. None of the confidential

tax information received from the IRS will be disclosed in this treatise. Statistical aggregates

will be used so that a specific taxpayer cannot be identified from information supplied by

the IRS. All opinions are those of the authors and do not reflect the views of the IRS. An

Online Appendix to this paper can be downloaded at http://research.chicagobooth.edu/

arc/journal-of-accounting-research/online-supplements.

777

Copyright C, University of Chicago on behalf of the Accounting Research Center,2015

778 K.K.F.LAW AND L.F.MILLS

the unexpected closures of local banks as exogenous liquidity shocks to show

that firms’ external financial constraints affect their tax avoidance strategies.

Overall, the linguistic cues in firms’ qualitative disclosures provide incremen-

tal information beyond traditional accounting variables or commonly used

effective tax rates to reveal and predict tax aggressiveness, both contempora-

neously and in the future.

JEL codes: G30; H25; H26; M41

Keywords: Tax avoidance; unrecognized tax benefits; negative words;

disclosure

1. Introduction

Traditional corporate finance theories posit that firms faced with finan-

cial constraints—broadly defined as frictions that prevent firms from fund-

ing all desired investments (Lamont, Polk, and Sa´

a-Requejo [2001])—have

higher costs of external financing. Financially constrained firms preserve

internal finance to generate funds for future investment opportunities.

One potential source of additional financing that has received little atten-

tion until recently is aggressive corporate tax planning. All else equal, firms

should adopt an equilibrium tax reporting posture that weighs tax savings

against the expected direct and indirect costs of tax aggressiveness.1How-

ever, when the frictions faced in raising external funds increase, financially

constrained firms could pursue more aggressive tax planning on the mar-

gin as a substitute for a more expensive source of external financing from

lenders or capital markets. In this paper, we investigate this argument em-

pirically by examining whether the variation across firms in their financial

constraints explains the variation in their tax aggressiveness. We predict

and test whether more financially constrained firms pursue more aggres-

sive tax planning strategies to provide additional internal funds, compared

to less financially constrained firms.

Our study complements but differs from recent working papers exam-

ining the relationship of financial constraints and aggressive tax planning

activities (e.g., Chen and Lai [2012], Dyreng and Markle [2013], Edwards,

Schwab, and Shevlin [2014]) in two key aspects. First, we use the qualitative

measure proposed by Bodnaruk, Loughran, and McDonald [2013] to mea-

sure financial constraints. They find that firms are more likely to experi-

ence subsequent liquidity events after they use more negative words in their

annual reports, consistent with the disclosure of negative information cap-

turing hard-to-quantify unfavorable aspects of firms’ business environments

(Tetlock, Saar-Tsechansky, and Macskassy [2008]). The linguistic approach

1Prior research shows that aggressive corporate tax planning is often hard to detect and,

even when aggressive tax planning is detected, the associated direct penalties are often small

(Weisbach [2002], Slemrod [2004], Desai and Dharmapala [2006]). Nevertheless, Graham

et al. [2014] find that surveyed managers respond that they want to avoid a reputation for

being tax aggressive, implying that managers perceive an indirect cost of that reputation.

TAXES AND FINANCIAL CONSTRAINTS 779

mitigates the concern of misidentifying accounting variables to capture fi-

nancial constraints.2Second, we rely on a firm-specific measure to identify

financial constraints. Our research design informs investors, analysts, and

researchers of the effect of firm-level constraints on current and future tax

reporting, incremental to the economy-wide measures (e.g., IPOs, GDP)

studied by Edwards, Schwab, and Shevlin [2014].3

Investors, tax authorities, shareholders, and regulators traditionally rely

on firms’ quantitative information to evaluate the extent to which tax avoid-

ance (any reduction in explicit tax paid) is aggressive (unlikely to be sus-

tained if challenged). Financial measures such as effective tax rates (ETRs)

are also the primary source of information for the media’s high-profile in-

vestigations that spotlight firms’ aggressive tax avoidance strategies (e.g.,

the low tax rates of Starbucks, Google, and GE). However, if negative

words in annual reports provide incremental information about aggressive

tax planning beyond current accounting variables, this linguistically based

measure could help researchers and regulators identify tax-aggressive firms

that under-report tax reserves or do not face mandatory disclosure of tax

reserves.4Further, our evidence below linking the use of negative words to

future higher unrecognized tax benefits (UTBs) and lower ETRs suggests

that financial constraints predict future tax aggressiveness.

To measure a firm’s financial constraints, we compute Use of Negative

Words as the fraction of the negative words in a firm’s annual 10-K filing

based on the summary word count files provided by William McDonald.5

Bodnaruk, Loughran, and McDonald [2013] propose a novel measure to

capture firms’ financial constraints based on the fraction of negative words

in the 10-K filings. They show that firms’ use of negative words predicts

subsequent liquidity events including dividend cuts or omissions, debt

downgrades, and lower asset growth, even after controlling for various

2Although the theory that external funds are more expensive than internal costs of funds

for financially constrained firms is clear, the empirical literature provides mixed guidance

on which accounting variables or index measures of constraints to use. Widely used mea-

sures are often minimally or negatively correlated. See Whited and Wu [2006], Hadlock and

Pierce [2010], Bodnaruk, Loughran, and McDonald [2013], Buehlmaier and Whited [2014],

Farre-Mensa and Ljungqvist [2014], and Hoberg and Maksimovic [2015]. Kaplan and Zingales

[1997] and Hadlock and Pierce [2010] conduct textual analyses into the annual reports of

49 manufacturing firms and 356 firms, respectively, using hand-collected measures. However,

such measures are often criticized as being subjective, non-replicable, and non-generalizable

to a large sample.

3Our year fixed effects control for time-series variation in the level of macroeconomic

conditions.

4Examples include firms using international accounting standards (e.g., IFRS), firms using

U.S. Generally Accepted Accounting Principles (GAAP) before 2007, or firms that decrease

reported reserves to minimize tax authorities’ scrutiny, particularly after new tax return re-

quirements in 2010 (Towery [2014]).

5McDonald’s website lists 2,329 negative words (http://www3.nd.edu/mcdonald). Wedo

not directly parse firms’ 10-K filings to get negative word counts, nor have we discovered a way

to extract only the tax footnote. In supplemental tests, we count the negative words in the tax

footnotes of 60 firms with extremely large and small tax reserves.