Does location matter for disclosure? Evidence from geographic segments

• Author: Arnt Verriest,Edith Leung
• Published date: 01 May 2019
• Date: 01 May 2019
• DOI: http://doi.org/10.1111/jbfa.12375

DOI: 10.1111/jbfa.12375

Does location matter for disclosure? Evidence

from geographic segments

Edith Leung1Arnt Verriest2

1Department of Accounting, Auditing, and

Control, ErasmusSchool of Economics, Erasmus

University Rotterdam, Rotterdam, the

Netherlands

2EDHEC Business School, Roubaix, France

Correspondence

ArntVerriest, EDHEC Business School, 24

AvenueGustave Delory,59057, Roubaix Cedex 1,

France.

Email:arnt.verriest@edhec.edu

Abstract

The segment disclosures of multinational companies provide strate-

gic information. We use the location characteristics of geographic

segments to identify the reasons for withholding or disclosing seg-

ments. We examine segment data from around the adoption of IFRS

8, a reporting standard that requires firms to reveal more disag-

gregated information. Consistent with a proprietary cost motive for

nondisclosure, we find that segments in regions thatare deemed bet-

ter for business tend to be hidden, while higher entry barriers for

a segment are positively related to disclosure. These effects appear

to be stronger for firms for which proprietary cost motives are more

important. Among the previously unrevealed segments, proprietary

costs explain the nondisclosure of segment earnings and other rele-

vant financial information for investors.

KEYWORDS

IFRS 8, proprietary disclosure costs, segment reporting

JEL CLASSIFICATION

M41

1INTRODUCTION

Afirm's expansion into foreign markets complicates the processing, gathering, and analysis of information for man-

agers (Egelhoff,1991), investors (Blanco, Garcia Lara, & Tribo, 2015; Callen, Hope, & Segal, 2006; Hope, Kang, Thomas,

& Vasvari, 2009), and financial analysts (Duru & Reeb, 2002; Herrmann, Hope, & Thomas, 2008). This is primarilydue

to national differences in, for example, labor regulations, tax regimes, legal institutions, business practices, and gen-

eral cultural differences (Dunning, 1993; Reeb, Kwok, & Baek, 1998). The significant increase in internationalization

and foreign operations among many companies overthe last few decades has made access to information about firms’

international operations even more important. Access to high-quality information about a firm's foreign operations

is vital for investors’ and analysts’ ability to analyze its current performance accurately and forecast its future per-

formance. Geographic information is particularly important for analyzing firms with substantial foreign operations,

including many European firms.

Despite the importance of information on corporate foreign operations for evaluating firm performance, many

firms are reluctant to provideextensive information on their geographic segments. Research on information disclosure

typically assumes the existence of disclosure costs to explain why firms do not fully disclose all the information in

J Bus Fin Acc. 2019;46:541–568. wileyonlinelibrary.com/journal/jbfa c

2019 John Wiley & Sons Ltd 541

542 LEUNG ANDVERRIEST

their possession, even when there is a clear demand for it from external parties. The cost most commonly cited as an

example is proprietary cost (Verrecchia, 2001)—the cost of disclosing strategicinformation to potential competitors

that can harm the disclosing firm (Darrough & Stoughton, 1990).

This study investigates whether the economic characteristics of a segment's location or region make firms reluc-

tant to disclose geographic segment information. First, we exploita change in segment disclosure regulations that has

resulted in more geographic segment disclosure: since January 2009, all firms reporting under IFRS have needed to

prepare and disclose their segment reports in compliance with IFRS 8, which replaced the old standard, IAS 14. The

new standard forces firms to disclose information on all “material” countries. Therefore, IFRS 8 provides more disag-

gregated, and higher quality, geographic segments.1Despite some initial concern as to whether firms would use the

discretion in defining “materiality” for geographic segments, prior literaturefinds that IFRS 8 has resulted in more dis-

closure (Bugeja, Czernkowski, & Moran, 2015; Cereola, Nichols, & Street, 2017; Crawford,Extance, Helliar, & Power,

2012; Leung & Verriest, 2015; Nichols, Street, & Cereola, 2012). The fact that, under IFRS 8, firms are required to

restate their segments for the year prior to adoption to allow investors to compare financial statements over time is

crucial for our methodology.This feature enables us to identify which geographic segments are newly disclosed under

IFRS 8 for the pre-adoption year, since we also have the original segment disclosures under IAS 14. Since we exam-

ine data for the same year under two different disclosure standards, we effectively use each firm as its own control.

Changes in segment disclosures are therefore likelyto be due to disclosure incentives ratherthan c hanges in the firm's

economic condition. We interpret newly disclosed segments as segments that management wanted to hide and seg-

ments that are revealedunder both standards as those that management wanted to disclose. Subsequently, we investi-

gate how these “new” geographic segments differ in nature from the “old”segments. This framework provides us with

an excellent opportunity to investigatewhether motives such as proprietary costs affect disclosure without having to

rely on the traditional, flawed proxies for these costs, such as industry concentration(Ali, Klasa, & Yeung, 2009; Ded-

man & Lennox,2009). Second, we apply our methodology to a sample of European firms. European multinationals have

much greater foreign sales and operations than their US-based peers. Information on their geographic segments is

therefore particularly important for investors. Our sample contains 1,277 stock-listed firms.

Following theoretical work on disclosure policies (e.g., Leuz& Verrecchia, 2000; Verrecchia, 2001), we expect firms

to hide those geographic segments that reveal positiveor “good” information that competitors can use to their advan-

tage. If IFRS 8 forces firms to provide more geographic disclosure, we expect firms to reveal the segments they had

previously decided to conceal. Specifically,we predict that firms are more likely to reveal “good” geographic segments

after the introduction of IFRS 8. Totest this notion empirically, we consider various characteristics of ageographic seg-

ment based on its location, including its attractiveness for doing business, its future economic prospects, and its credit

risk. We predict that firms making disclosure decisions will hide segments that are attractive for business, have posi-

tive economic prospects, and have low credit risk. Next,we consider the extent to which a country has entry barriers

for new market participants. Wepredict that lower entry barriers decrease the likelihood of disclosure. Prior research

suggests that revealing information about segments in areas where it is easier for other firms to enter can be more

costly,as this may reduce a firm's competitive advantage more quickly (Leuz, 2003; Raith, 2003).

Our empirical findings are consistent with these predictions. First, we find that newly revealedgeographic segments

tend to be more economically attractive countries or locations. Specifically, we find that activities in more attractive

business locations, more financially stable countries, and countries with lower credit risk are more likely to be hidden.

Second, we find that newly revealed segments havelower entry barriers. In other words, multinational firms are more

likely to hide locations to which other firms may easily gain access. Both sets of results are in line with the notion that

firms have strong incentivesto hide information that may harm their competitive stance.

We extend our analyses in three ways. First, we investigate whether the effects are stronger in sample subsets

where we would aprioriexpect the proprietary costs of disclosure to be more relevant. In line with prior findings

(Ellis, Fee, & Thomas, 2012), we indeed find stronger indications of proprietary cost motives in more R&D-intensive

industries.

1By“disaggregation,” we are referring to whether segments are reported at the country, multi-country, or continent level.