Foreign Exchange Risk, Hedging, and Tax‐Motivated Outbound Income Shifting

• DOI: http://doi.org/10.1111/1475-679X.12326
• Date: 01 September 2020
• Published date: 01 September 2020
• Author: ZERO DENG

DOI: 10.1111/1475-679X.12326

Journal of Accounting Research

Vol. 58 No. 4 September 2020

Printed in U.S.A.

Foreign Exchange Risk, Hedging,

and Tax-Motivated Outbound

Income Shifting

ZERO DENG∗

Received 24 April 2019; accepted 27 May 2020

ABSTRACT

Although outbound income shifting to low-tax jurisdictions provides tax sav-

ings, it is often accompanied by nontax costs. In this study, I examine whether

foreign exchange (FX) risk constrains tax-motivated outbound income shift-

ing by U.S. multinational corporations. My findings indicate that exposure

to greater currency volatility is associated with less outbound income shift-

ing, and this effect is stronger for firms with foreign affiliates using foreign

functional currencies. I also investigate whether hedging facilitates outbound

income shifting. Consistent with hedging lowering costs associated with ex-

change rate volatility, I find that U.S. firms that use more currency derivatives

∗Oregon State University.

Accepted by Regina Wittenberg Moerman. I thank my dissertation committee: Dan Givoly

(Chair), Henock Louis, John Moran, and Karl Muller. I especially thank Rick Laux for gen-

erously giving time and advice throughout this research. I have significantly benefited from

insightful comments from Badryah Alhusaini, Lisa De Simone, Apoorv Gogar,Ed Ketz, Charles

Lee, Lillian Mills, Jared Moore, Jed Neilson, Renzi Shen, Syrena Shirley, Logan Steele, Laura

Wellman, Ryan Wilson, Hal White, an anonymous associate editor and referee, and workshop

participants at California State University at Sacramento, Oregon State University, Penn State

University, and the 2018 AAA Doctoral Consortium. I also extend thanks to Scott Dyreng for

making available his Exhibit 21 data. All errors are my own.

953

© University of Chicago on behalf of the Accounting Research Center, 2020

954 z. deng

tend to shift more income to low-tax foreign jurisdictions. Overall, these find-

ings suggest that FX risk is an important cost of outbound income shifting.

JEL codes: F31, G32, H26, M40

Keywords: corporate taxation; income shifting; foreign exchange risk;

derivatives

1. Introduction

Tax-motivated income shifting by U.S. multinational corporations (MNCs)

has attracted significant attention in recent years. The popular press pro-

vides anecdotal evidence suggesting that U.S. MNCs engage in extensive

outbound income shifting to avoid U.S. taxes (Bowers and Drucker [2017],

Cohen [2017]). However,empirical estimates generally show a modest mag-

nitude of outbound income shifting, raising the question of why more in-

come is not shifted offshore given the tremendous potential to save taxes

(Dharmapala [2014]). One explanation is that when firms engage in out-

bound income shifting, they incur significant costs that constrain their abil-

ity to take full advantage of these opportunities.1With a few exceptions in

the extant literature, we have a very limited understanding about the trade-

offs U.S. MNCs face when they make outbound income shifting decisions

(Blouin [2012], Dharmapala [2014]).2

A largely overlooked cost of outbound income shifting in the literature

is foreign exchange (FX) risk.3When a corporation shifts earnings to low-

tax foreign countries, the earnings will most likely be denominated in for-

eign currencies and exposed to currency risk.4During the last two decades,

U.S. MNCs have become increasingly concerned about FX exposure due

to the significant increase in exchange rate volatility (Apte [2010]), which

rendered their earnings and cash flows more volatile through FX transla-

tion effects. For example, the surge in the U.S. dollar (USD) against for-

eign currencies in 2015 depressed earnings and cash flows at many major

1The modest income shifting magnitude may also be driven by the exclusion of firms with

loss affiliates (Hopland et al. [2017, 2019], De Simone, Klassen, and Seidman [2017]) and the

use of accounting databases that exclude data on profits in tax haven jurisdictions (Clausing

[2019]).

2These exceptions include Klassen and Laplante [2012b], Dyreng and Markle [2016],

Chen, Hepfer, Quinn, and Wilson [2018], and Gallemore, Huang, and Wentland[2018]. See

subsection 2.1 for more details.

3Foreign exchange risk, also known as currency risk or exchange rate risk, is the financial

risk that earnings and investments denominated in foreign currencies will change in value

due to unexpected exchange rate variations.

4Using affiliate-level data provided by the U.S. Bureau of Economic Analysis, Robinson

and Stocken [2013] report that in 2004, 81% of U.S. MNCs’ foreign affiliates used a foreign

functional currency (i.e., the currency in which a foreign entity’s earnings are denominated).

In this study, I use firm-level data from Compustat and find that 89% of U.S. MNCs had at

least one affiliate using a foreign functional currency during 1993–2017. See subsection 3.2

for descriptive statistics.

fx risk, hedging, and outbound income shifting 955

U.S. MNCs (Francis, Mitchell, and Ziobro [2015], McLaughlin and Valdma-

nis [2015]). In light of these threats to financial performance, corporate

treasurers face substantial challenges in managing FX risk through hedg-

ing because the strategy requires accurate exposure information, reliable

transaction forecasts, and ongoing risk monitoring (Deloitte [2016, 2017]).

Therefore, exposure to significant currency risk is likely to dampen the at-

tractiveness of outbound income shifting due to the greater uncertainty

about financial performance and expensive hedging costs.

This study examines whether FX risk constrains U.S. MNCs’ tax-

motivated outbound income shifting activities. To examine this question, I

measure FX risk using the exchange rate stability ratings provided by the In-

ternational Country Risk Guide (ICRG) and estimate income shifting using

the approach developed by Collins, Kemsley, and Lang [1998] with the mul-

tiperiod regression variables suggested by Klassen and Laplante [2012b].

Due to the adverse FX effects on financial performance, I predict that ex-

posure to greater FX risk is associated with less outbound income shifting

to low-tax jurisdictions. Nonetheless, U.S. firms may avoid FX risk by con-

ducting transactions in USD or shifting income to countries with minimal

FX volatility. Therefore, whether FX risk affects outbound income shifting

is an empirical question.

Using a sample of U.S. MNCs with tax incentives to shift income out of

the United States from 1993 to 2017, I examine how the outbound income

shifting intensity varies cross-sectionally with the subsidiary-weighted mea-

sure of FX risk (FXR i sk ). I find that exposure to greater currency volatility

is associated with less outbound income shifting by U.S. firms, consistent

with the expectation that FX translation effects increase financial report-

ing costs. The economic magnitude suggests that a 10-percentage point de-

crease in average foreign tax rates translates into an additional $19.6 mil-

lion of income shifted out of the United States by a firm exposed to less

risky currencies relative to one exposed to more risky currencies (defined

as FXR is k in the 25th percentile vs. 75th percentile).

Tobetter understand the relation between FX risk and outbound income

shifting, I examine whether the constraining effect of FX risk is stronger

for MNCs with foreign affiliates using foreign currencies as their functional

currencies. An affiliate’s functional currency “is the currency of the envi-

ronment in which [the] entity primarily generates and expends cash” (ASC

830-10-45-2). Therefore, the affiliate’s functional currency (USD vs. non-

USD) determines whether the shifted income will be exposed to FX volatil-

ity. Following Robinson and Stocken [2013], I use nonzero changes in the

cumulative translation adjustment (CTA) account to capture firms with af-

filiates using foreign functional currencies. Consistent with my expectation,

I find a larger effect of FX risk for these firms relative to those with affili-

ates using the USD as the functional currency. However, 89% of the sam-

ple firms use foreign currencies and, therefore, I acknowledge that these

results may be driven by the limited variation in the functional currency

measure.