Description
This paper analyzes the multinational enterprise’s decision to voluntarily disclose information
regarding its investments, a choice we term investment transparency. When disclosing
investment information, managers must weigh the costs and benefits of reducing asymmetries
between the firm and three stakeholder audiences: capital markets, civil society and
governments. We use a unique transaction-level dataset of reserve acquisitions by oilindustry
multinationals compiled by IHS Herold to examine managerial decisions to reveal
or withhold value-relevant information about firm investment.
Multinational investment and voluntary disclosure:
Project-level evidence from the petroleum industry
Anthony P. Cannizzaro
?
, Robert J. Weiner
1
George Washington University, Department of International Business, 2201 G Street NW, Suite 401, Washington, DC 20052, United States
a r t i c l e i n f o
Article history:
Available online 19 February 2015
a b s t r a c t
This paper analyzes the multinational enterprise’s decision to voluntarily disclose informa-
tion regarding its investments, a choice we term investment transparency. When disclosing
investment information, managers must weigh the costs and bene?ts of reducing asymme-
tries between the ?rm and three stakeholder audiences: capital markets, civil society and
governments. We use a unique transaction-level dataset of reserve acquisitions by oil-
industry multinationals compiled by IHS Herold to examine managerial decisions to reveal
or withhold value-relevant information about ?rm investment. Contrary to the agency-the-
oretic motivations traditionally ascribed to voluntary disclosure, our results suggest insti-
tutional and informational factors drive investment transparency. We ?nd that ?rms
disclose less in cross-border transactions, more when societal expectations of transparency
are high, and less when faced with political risk. These results should be of interest to
scholars of accounting and international business, as well as managers and policy makers
involved in the ongoing debate on transparency in the extractive industries.
Ó 2015 Elsevier Ltd. All rights reserved.
Introduction
How transparent are multinational enterprises (MNEs)
regarding their investments? In this study, we use the glo-
bal market for petroleum reserves as a laboratory to exam-
ine investment transparency – value-relevant information
MNEs choose to disclose voluntarily about investment pro-
jects. For a given investment, ?rms may disclose no infor-
mation, partial information, or full information about the
value of the investment.
We use a unique transaction-level dataset compiled by
IHS Herold, which allows us to identify which party dis-
closes each transaction and how much information is
revealed about the investment. We ?nd that ?rms disclose
less about cross-border than domestic investment. This
result is robust to controls for the ?rm’s capital needs,
national institutions, ownership of the ?rm and character-
istics of the investment. Further, we ?nd that ?rms invest-
ing in countries with strong transparency norms (proxied
by government ?scal openness, freedom of the press, and
quality of the accounting system), and strong political con-
straints are more likely to disclose partial information.
Firms from countries marked by less political risk and cor-
ruption are more likely to disclose full information.
We draw on theories of voluntary disclosure and the
institutional and political economy literatures to suggest
that MNEs use voluntary disclosure strategically to man-
age information asymmetries between the ?rm and three
primary stakeholder groups: capital markets, civil society,
and governments. Our results do not support traditional
agency-theoretic motivations such as increasing disclosure
to secure external ?nancial resources, and increased dis-
closure in multinational operations. This runs counter to
the view in the literature that MNEs disclose more in
response to capital market demands for information about
their operations abroad (Cahan, Rahman, & Perera, 2005).http://dx.doi.org/10.1016/j.aos.2015.01.002
0361-3682/Ó 2015 Elsevier Ltd. All rights reserved.
?
Corresponding author. Tel.: +1 202 994 6880; fax: +1 202 994 7422.
E-mail addresses: [email protected] (A.P. Cannizzaro), rweiner@gwu.
edu (R.J. Weiner).
1
Tel.: +1 202 994 5981; fax: +1 202 994 7422.
Accounting, Organizations and Society 42 (2015) 32–47
Contents lists available at ScienceDirect
Accounting, Organizations and Society
j our nal homepage: www. el sevi er. com/ l ocat e/ aos
Rather, institutional factors such as societal expecta-
tions of transparency and political risk play an important
role in multinational disclosure. Our results are consistent
with the view that voluntary disclosure is driven by social
norms (Cho, Guidry, Hageman, & Patten, 2012), as MNEs
seek to be perceived as legitimate by key stakeholders
(Parsons, 1960). Further, our ?ndings are consistent with
MNEs’ strategically reducing transparency to protect
investments from government predation.
Multinational enterprises typically operate numerous
investment projects in a variety of countries and institu-
tional environments. Despite empirical evidence that
exposure to international markets increases the complex-
ity of assessing and transmitting value-relevant informa-
tion (Callen, Hope, & Segal, 2005; Hope, Kang, Thomas, &
Vasvari, 2008; Portes & Rey, 2005; Thomas, 1999), the lit-
erature has only recently begun to explore how countries
matter for corporate transparency (Healy & Serafeim,
2013; Shi, Magnan, & Kim, 2012).
We extend this literature beyond ?nancial-statement
reporting by examining value-relevant disclosures speci?c
to ?rminvestments. The importance of investments to ?rm
strategy is highlighted in the scholarly literatures in
?nance (Maksimovic & Phillips, 2001; Maksimovic,
Phillips, & Prabhala, 2011) and strategic management
(Capron, Mitchell, & Swaminathan, 2001), yet is underrep-
resented in the literature on voluntary disclosure.
More broadly, society has long been concerned over
accountability and corruption in foreign investment.
Widely-held suspicions of MNEs are based in part on per-
ceptions that their secrecy masks illicit behavior. Stiglitz
(2008) argues that MNEs are more likely than domestic
?rms to exploit asymmetries in bargaining power and
information, and use cross-border transactions to avoid
accountability. Transfer pricing, the ability to manipulate
internal prices to shift pro?ts between subsidiaries in dif-
ferent tax jurisdictions, is an advantage of multinationality
(Eden, 2012). Meek and Thomas (2004) note opacity of for-
eign operations as an ongoing issue in international disclo-
sure research.
Our study is relevant to debates over transparency ini-
tiatives by governments, intergovernmental organizations
(IGOs), and civil society (through non-governmental orga-
nizations – NGOs). This debate is particularly contentious
in the extractive industries, of which petroleum is the larg-
est. Survey evidence suggests that petroleum and mining
are among the industries most prone to bribery
(Transparency International, 2011). Darby (2009) notes
that failure by MNEs in the extractive industries to disclose
information is often interpreted as these ?rms having
something to hide. A majority of the 20 worst-performing
countries on the Transparency International (2012) corrup-
tion index are natural-resource rich. Both IGOs and NGOs
2
have been expanded or created to push increased transpar-
ency, discourage corruption, and address other ills of
resource-rich societies (Durnev & Guriev, 2007; Jensen &
Johnston, 2011).
Recent years have seen policy initiatives and disclosure
rule changes designed to enhance transparency. In the Uni-
ted States, Section 1504 of the Dodd-Frank Wall Street
Reform and Consumer Protection Act (‘‘Dodd-Frank,’’
passed in 2010), amends the Securities Exchange Act of
1934 to require disclosure of government payments by
extractive industry ?rms listed on US exchanges. In 2012,
the Securities and Exchange Commission adopted rule
13(q)-1, which requires US-listed ?rms in extractive indus-
tries to include project-level disclosures of payments to
governments at home and abroad in their annual reports,
starting in late 2013.
3
However, following industry chal-
lenges the rule was vacated.
4
The European Union, Hong
Kong, and Canada have enacted or have committed to enact
similar regulations on listed extractive industry MNEs.
5
Project-by-project disclosure requirements, intended to
create accountability and reduce corruption, have proven
extremely controversial (Hunt, 2011). Strong opposition
from listed petroleum MNEs, such as the effort that chal-
lenged the 2013 Dodd-Frank rule, are grounded in claims
that project-level reporting will be detrimental to ?rms
by disclosing private information to governments and
competitors, many of which are state-owned.
6
Numerous
state-owned ?rms are unlisted, and hence not bound by
these rules. In contrast, NGOs assert that project-level dis-
closure will not have signi?cant consequences for competi-
tiveness (Rosenblum & Maples, 2009).
The debate over mandatory project-level disclosure
raises the question of the extent to which such data are
now reported voluntarily. Thus, we examine investment-
level disclosure patterns and the institutional factors driv-
ing them. If investment project disclosures are costly to
?rms, these costs should in?uence managerial decisions
to reveal information. Relating disclosure decisions to
MNEs’ investment locations, our approach integrates exist-
ing insights from the accounting literature on corporate
transparency with work from institutional theory and
political economy to model the decision to disclose. Our
work complements recent research on corporate-level
disclosure of performance and payments to governments
in publicly-traded petroleum MNEs (Healy & Serafeim,
2013).
2
The primary IGO example is the Extractive Industries Transparency
Initiative (EITI), an international collaboration between governments,
businesses, and civil society groups that promotes disclosure of aggregate
?rm payments. NGO examples include Oxfam International (an NGO that
promotes poverty alleviation worldwide), and Publish What you Pay, a
global network of NGOs (including Transparency International and Global
Witness) devoted to promoting transparency in the oil industry.
3
The language of the law is broadly interpreted as requiring issuers to
disclose granular, disaggregated information on a project-by-project basis.
Hunt (2011) provides a comprehensive review of the legislation.
4
The rule was vacated in July 2013 by the US District Court for the
District of Columbia (see Memorandum Opinion ?led July 2, 2013 for Civil
Action Number 12-1668 (JDB)).
5
For example, new EU Transparency and Accounting Directives require
country-by-country and project-by-project disclosure of all government
payments over €100,000 (European Commission MEMO/13/541 dated June
12, 2013).
6
SEC Release No. 34-67717; File No. S7-42-10. Final Rule Making on
Disclosure of Payments by Resource Extraction Issuers, 17 CFR Parts 240 & 249,
summarizes inter alia industry views.
A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47 33
Background and empirical setting
A rich literature examines corporate disclosure deci-
sions. This literature (surveyed in Beyer, Cohen, Lys, &
Walther, 2010) focuses primarily on corporate-level
reporting, and is silent about the motivations for invest-
ment-level disclosure. Exceptions include Kanodia and
Lee (1998) who study the effect of disclosure on invest-
ment, and Chen, Tan, Cheng, and Gong (2013) who exam-
ine domestic investment projects by listed ?rms in China,
where reporting of new projects is mandatory but disclo-
sure of details is not.
Consolidated corporate-level ?nancial statements
re?ect outcomes of multiple managerial decisions, making
empirical identi?cation at the investment-level problem-
atic. In contrast, we focus on individual investments
through asset acquisitions, both domestic and abroad.
The assets we examine are petroleum reserves (the quan-
tity of oil and gas in the ground that is extractable econom-
ically at current prices and costs). The importance of
reserves combined with the ubiquitous state role in the
industry often links them to corruption, and non-transpar-
ent or non-competitive sales conditions. As the Economist
notes, ‘‘Deals for oil?elds can be as opaque as the stuff that
is pumped from them’’ (Economist, 2013).
Reserves provide several advantages for research on
disclosure. First, they are a critical asset to ?rms in the
upstream segment of the industry (exploration and pro-
duction); as the ?rm’s inventory of future production
(and hence a predictor of future earnings), they are central
to valuation and borrowing capacity (Arnott, 2004; Chung,
Ghicas, & Pastena, 1993; Muñoz, 2009; Osmundsen, 2010).
Research has demonstrated that reserves are value-rele-
vant (Misund, Osmundsen, & Asche, 2005; Taylor,
Richardson, Tower, & Hancock, 2012), and that reserve
restatements (analogous to asset restatements in manufac-
turing ?rms) are associated with abnormal stock returns
(Berry & Wright, 2001).
7
Second, reserves are suf?ciently homogenous to be
comparable across geographic boundaries, ?rm bound-
aries, and time. As inventory to be produced in the future,
reserves are similar across ?rms, and market comparisons
are more easily assessed than for assets such as plant and
equipment or organizational divisions. Reserves are largely
homogeneous except for cost differences (for which we
control in our analysis); intangibles such as brand name
or goodwill do not play a role. This homogeneity facilitates
outsiders’ investment valuation if reserve size is disclosed.
Third, research that models determinants of voluntary
disclosure identi?es heterogeneity of private information
as a potentially confounding factor (Chen et al., 2013;
Ellis, Fee, & Thomas, 2012). Some ?rms may be less trans-
parent than others for the simple reason that they have lit-
tle private information or that their private information is
of little value. This is dif?cult to control for in empirical
research. By examining a large number of similar projects,
we avoid this problem.
Finally, acquisitions of reserves, like acquisitions of
?rms, are major management decisions, undertaken infre-
quently and followed by analysts and the trade press. It
thus ?ts well with the agency motivation for disclosure
(discussed below) prevalent in the literature. Purchasing
assets is consistent with both value creation (if the ?rm
can get more out of these assets than it paid for them),
and value destruction though hubris (here overpaying for
assets), empire-building (purchasing assets to increase
?rm size), or both (Hayward & Hambrick, 1997; Hope &
Thomas, 2008).
Unlike most real assets, petroleum reserves are actively
traded in a decentralized global market with many players
(although each ?rm trades infrequently, as shown below).
8
The large number of transactions facilitates empirical
inquiry. The advantage of our database of reserve-transac-
tion announcements is that each transaction effectively cor-
responds to a managerial decision on whether or not to
reveal information for a speci?c investment project. We
focus on disclosure of the two central attributes of each
investment – the size of the reserve acquired and the price
paid. For each transaction, these details are either provided
or redacted.
Theoretical development and hypotheses
We de?ne investment transparency as discretionary
disclosure of value-relevant information at the invest-
ment-level (Verrecchia, 2001). Managers are neither
required to nor prohibited from disclosing information on
reserve investments.
9
Failure to reveal either the price paid
in a transaction or the quantity of reserves purchased makes
a price-per-barrel market comparison impossible, obscuring
valuation. Such valuations may reveal information regarding
the MNE’s exploration success and extraction costs, affecting
negotiations with host governments and the potential for
corrupt activities.
Payoffs to disclosure re?ect information asymmetries
between management and three distinct stakeholder audi-
ences: capital markets, civil society, and governments.
First, capital markets value transparency as a governance
mechanism. Disclosure can help align manager and share-
holder interests, thereby lowering the MNE’s cost of capi-
tal. Second, disclosure of investment details may help
grant the ?rm legitimacy in the eyes of society. Lastly, dis-
closure may magnify political risk and corruption faced by
the ?rm.
7
Because reserves are central to the industry, their reporting has its own
terms and accounting rules, with standards set out by inter alia, the US
Securities and Exchange Commission, Canadian Securities Administrators,
State Commission for Reserves of the Russian Federation, Norwegian
Petroleum Directorate, the Committee for Mineral Reserves International
Reporting Standards, and the Society of Petroleum Engineers. Auditing by
independent specialists with expertise in engineering and geology is
common practice.
8
Reserves are treated as assets in this paper, as well as in the industry
and in ?nancial reporting. Ownership refers to control and cash ?ow rights
from reserves, which are technically property of either private landowners
or governments that receive royalties from the ?rms operating them.
9
US-listed ?rms must ?le 8-K reports for material transactions, but have
discretion over what is material and what to report. Rosenblum and Maples
(2009) ?nd that while many host countries prohibit ?rms from disclosing
details of petroleum contracts, none prevents disclosure of asset payment
or size.
34 A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47
Capital markets
Twostrands of the accountingliterature suggest con?ict-
ing hypotheses about the relationship between capital mar-
kets and investment transparency. The global-diversi?cation
strand of the literature suggests complexity of international
operations raises agency costs, increasing managers’ incen-
tive to disclose. Cahan et al. (2005) argue that international
operations increase ?rm complexity. This complexity gives
international managers information advantages over inves-
tors by virtue of their local knowledge, increasing the scope
for non-value-maximizing behavior. More disclosure
reduces this scope, thus enhancing ?rm value. Hope and
Thomas (2008) ?nd evidence of such agency con?icts in
MNE operations abroad by demonstrating that ?rms that
fail to disclose geographic earnings experience greater for-
eign sales growth, but reductions in foreign pro?ts, consis-
tent with managerial empire building. However, these
studies use foreign sales or number of foreign subsidiaries
to proxy international complexity, and are unable to
directly compare foreign and domestic investments.
The investor-sophistication strand yields the opposite
prediction. Managers respond strategically to the sophisti-
cation of the ?rm’s investors, because uninformed investors
alter the bene?ts of disclosing information. Dye (1998) the-
orizes that the primary bene?t to the ?rm of withholding
information arises from an uninformed investor’s inability
to recognize whether nondisclosure is explained by having
little to disclose, or by hiding information. Dye’s formal
model demonstrates that as the probability investors are
uninformed increases, ?rm incentives to disclose decrease.
In the international context, ?nance research ?nds that
equity investors are typically less informed about foreign
assets. Home bias is the widely documented tendency for
investors to overweight domestic companies in their port-
folios (French & Poterba, 1991). Shareholders ?nd it more
dif?cult to assess the value and risks of investments abroad
than at home because information ?ows are not frictionless
across borders (Ahearne, Griever, &Warnock, 2004; Coval &
Moskowitz, 1999; Kang & Stulz, 1997). This home bias
extends beyond securities to ?rmoperations abroad. Exam-
ining foreign earnings, Thomas (1999) ?nds that investors
place a lower value on foreign information because it is
more dif?cult to assess, resulting in a bias towards domes-
tic investment. The effect on corporate transparency is that
cross-border frictions reduce the bene?t to managers of
disclosing foreign acquisitions relative to domestic. Thus,
?rms should disclose less in cross-border transactions.
The two strands of the literature thus yield opposite
predictions regarding disclosure and multinational invest-
ment. Ultimately, which effect dominates is an empirical
question. Accordingly, we test the following hypotheses:
H1A (Global Diversi?cation Hypothesis). Firms will volun-
tarily disclose more information about cross-border invest-
ments than domestic investments.
H1A
0
(Investor Sophistication Hypothesis). Firms will volun-
tarily disclose less information about cross-border invest-
ments than domestic investments.
Agency theory also suggests that investment transpar-
ency may be more important for ?rms in need of additional
capital. New investment projects in general give managers
an information advantage. For oil industry MNEs, the
unobservable nature of underground reserves creates a
great deal of uncertainty around the production potential
of a new project. Firms necessarily invest time, technology
and human capital in making proprietary estimates of this
potential. The resulting reduction in managerial uncer-
tainty creates the potential for con?ict between insiders
and investors. For example, Bertrand and Mullainathan
(2000) ?nd that oil industry executive compensation is
unrelated to performance when monitoring is poor.
These information asymmetries and agency con?icts
increase the ?rm’s cost of capital as investors consider
these costs and adjust their expectations accordingly
(Diamond & Verrecchia, 1991). Investment disclosures
may provide MNEs in need of external ?nancing a solution
to the agency problem by reducing information asymme-
tries, lowering monitoring costs and signaling to investors
that managerial interests are aligned with their own (Leuz
& Verrecchia, 2000). Thus, we expect ?rms in need of addi-
tional external ?nancing should be more likely to reveal
information regarding reserve transactions.
H1B. Firms with greater reliance on external ?nancing will
voluntarily disclose more information about investments.
Civil society and corporate legitimacy
Researchers have long acknowledged that society and
societal values have a role to play in understanding ?rm
governance issues (Licht, 2001; Licht, Goldschmidt, &
Schwartz, 2005; Stulz & Williamson, 2003; Wolf &
Weinschrott, 1973). Studies focusing on ?rm accounting
highlight the importance of societal values for annual
report disclosures (Hope, 2003) and cross-country differ-
ences in accounting conservatism (Salter, Kang, Gotti, &
Doupnik, 2013).
Institutional theory suggests disclosure may be prefera-
ble if conforming to societal norms of transparency grants
the ?rm legitimacy; a perception that the ?rm’s actions are
desirable in the eyes of key stakeholders (Parsons, 1960;
Suchman, 1995). MNE managers respond strategically to
societal pressures, as legitimacy is often required to oper-
ate successfully abroad (Kostova & Zaheer, 1999; Oliver,
1991). Increasing voluntary disclosures when normative
expectations of transparency are high increases the likeli-
hood that ?rms are perceived as legitimate. Cho et al.
(2012) examine voluntary environmental disclosures and
?nd that ?rms’ reputations improve with greater disclo-
sure, even when actual environmental performance is
poor. Additional evidence in the accounting literature sug-
gests that legitimacy can be obtained via auditing (Free,
Salterio, & Shearer, 2009), and by adopting more rigorous
accounting standards (Guerreiro, Rodrigues, & Craig, 2012).
Transparency may increase MNE legitimacy even when
voluntary disclosures provide only limited information, as
the act of disclosure itself may help legitimize the ?rm
(March & Olsen, 1984). For example, Khanna, Palepu, and
A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47 35
Srinivasan (2004) show that foreign ?rms investing in the
United States tend to adopt US disclosure practices. In
addition, conforming to societal expectations, even
if only symbolically, may alleviate regulatory scrutiny
(Gabbioneta, Greenwood, Mazzola, & Minoja, 2013).
Thus, if legitimacy-seeking is the motivation, partial disclo-
sure may be preferable to either no disclosure or full
disclosure.
As legitimacy is required in all institutional environ-
ments in which the ?rm operates (Kostova & Zaheer,
1999), normative pressures may originate from either the
MNE’s country of origin (home country), or the country
in which an investment resides (the host country). Thus,
we argue that home and host countries marked by societal
expectations of transparency should exert more pressure
on MNEs to reveal information about investments.
H2A. Firms acquiring assets in countries with higher
normative expectations of transparency will voluntarily
disclose more information about investments.
H2B. Firms coming from countries with higher normative
expectations of transparency will voluntarily disclose more
information about investments.
Political risk
Greif (2005, p. 728) notes that institutions that ‘‘reveal
wealth’’ are only optimal when there are constraints that
curb the coercive power of the state. Unless constrained,
government of?cials have the discretion to expropriate
value for the state (Stulz, 2005) or misuse their of?ce to
extract bribes (Rose-Ackerman, 2003). Thus, MNEs operat-
ing in environments with weak institutional constraints on
government and its agents are incentivized to be less
transparent; the same information asymmetries capital
markets and civil society seek to curb may shield the
MNE from these political risks.
After initial costs of a foreign investment are sunk, the
MNE faces the risk that the host government may alter
the terms of a contract ex post, or violate agreements
entirely through expropriation of ?rm assets. This results
in the challenge ?rst identi?ed by Vernon (1971) as the
obsolescing bargain, a time-inconsistency problem stem-
ming from the government’s discretion to change the
terms of any agreement (Henisz & Williamson, 1999).
Firms typically have little recourse for mitigating this form
of risk once capital is invested in the host government’s
jurisdiction.
Political risks are especially salient in the petroleum
industry where rents are a high proportion of product
value. In addition, reserves are considered national patri-
mony, typically under state control, and are often located
in countries marked by weak institutions including poor
property-rights protection and unstable or autocratic gov-
ernments (‘‘resource curse’’). Extractive industries, particu-
larly petroleum, are at greatest risk for expropriation
(Guriev, Kolotilin, & Sonin, 2011; Hajzler, 2012).
Government of?cials may also attempt to extract rents
for themselves, creating costs of disclosure for both ?rms
that behave ethically and those that are complicit in cor-
rupt practices. The former may be disadvantaged in com-
peting for contracts if state of?cials favor opaque ?rms to
discourage corrupt practices from being revealed (Shleifer
& Vishny, 1993). For the latter, transparency exacerbates
the risk of detection (Healy & Serafeim, 2013).
Such risk arises from both home- as well as host-coun-
try governments (Cuervo-Cazurra, 2006; Fisman & Miguel,
2007). For example, the large role of opaque offshore ?nan-
cial centers in international capital ?ows suggests MNE
managers recognize political risks from home govern-
ments (Hines, 2010). Empirical evidence demonstrates that
home-country political risk reduces petroleum reserve val-
ues (Click, Jeong, & Weiner, 2013). Information asymmetry
between the ?rm and host- and home-country govern-
ments increases uncertainty about asset value, reducing
the attractiveness of predation, and hence expropriation
risk. Thus, we expect that ?rms coming from or operating
investments in countries with weak political institutions
will disclose less.
H3A. Firms acquiring assets in countries with greater
political risk will voluntarily disclose less information
about investments.
H3B. Firms coming from countries with greater political
risk will voluntarily disclose less information about invest-
ments.
Agency costs and political risks may not operate inde-
pendently. MNEs face a twin agency problem of managerial
and state discretion – the simultaneous threat of diversion
by ?rm insiders and expropriation by predatory govern-
ments (Stulz, 2005). When the latter poses a signi?cant
risk to the ?rm, the transparency decisions the ?rm would
otherwise adopt to compensate for the former may no
longer be value-enhancing.
Durnev and Fauver (2008) formalize Stulz’s twin agency
problem, showing that the agency costs of insider diver-
sion and the political costs of government expropriation
are complements. Extending the twin agency problem to
investment transparency, we theorize that MNEs subject
to higher agency costs are also more susceptible to political
costs. Ceteris paribus, these ?rms are less likely to disclose
than would be predicted by agency costs or political costs
alone.
H3C. The negative effect of political risk on investment
transparency will be stronger for ?rms with less reliance
on external ?nance.
Methodology
Sample
As noted above, our sample of oil and natural gas
reserve transactions comes from a database of public
36 A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47
announcements compiled by IHS Herold, an advisory ?rm
that conducts independent research and collects ?nancial
and operational information on the global petroleum
industry.
10
Widely-used investment databases (e.g., SDC
Platinum and ORBIS) do not indicate which counterparty
disclosed (or did not disclose) information, presenting an
identi?cation challenge in project-level analysis. In contrast,
the Herold database reports the information source for each
investment, enabling us to relate disclosure to the ?rm mak-
ing the disclosure decision.
Reserve transactions are voluntarily disclosed through
releases in the trade and business press. For each deal,
we obtain the announcement date, counterparty names
and home countries, reserve location, reserve type, trans-
action value (if reported), and estimated reserve size (if
reported). This data is then matched with detailed ?rm
information from Thompson-Reuters’ SDC Platinum data-
base. Information regarding price and quantity transacted
are necessary for the market to fully evaluate a reserve
transaction, yet ?rms frequently make only one piece of
information or neither available.
In total, our sample covering the years 2000 through
2011 is comprised of 1210 buyer-announced domestic
and cross-border reserve transactions by 702 ?rms across
64 countries.
11
We include only transactions that have been
reported by acquirers, as sellers’ investments are no longer
affected by host-country institutions once they are
divested. We exclude transactions with multiple buyers,
those where buyer assets are less than $1 million, and failed
transactions.
Measuring investment transparency
The literature uses several proxies for transparency. The
most common is earnings management, which allows
managers to obscure information from shareholders. Earn-
ings re?ect ?rm-wide outcomes, however, rather than spe-
ci?c investments. Closer to our interest in project-level
transparency is the inclusion of earnings forecasts in cor-
porate-level reports, as these forecasts may be tied to spe-
ci?c ?rm investments (Shi et al., 2012). However, these
measures are limited to publicly-traded ?rms, (a subsam-
ple of the population), and may re?ect cross-country vari-
ation in reporting rules.
To examine investment transparency, we code each
transaction by whether the announcement discloses min-
imal information (only the existence of a transaction is
revealed), partial information (exclusively price paid or
reserve size is revealed), or full information (both
price and size are revealed). Full information facilitates
comparison with market data, while partial information
permits limited insight into the ?rm’s investment
decision.
Independent variables
Need for external ?nancing is an established proxy for
the potential for agency con?icts between management
and capital markets (Jensen, 1986; Rajan & Zingales,
1998) as ?rms in need of ?nancing are incentivized to bind
themselves to better governance and transparency to
ensure that outside capital remains available. Following
Rajan and Zingales (1998), we calculate dependence on
external ?nance as the quantity capital expenditures less
cash ?ow, divided by capital expenditures. These data are
obtained through the SDC Platinum database. We follow
the literature in Winsorising at the 5% level to account
for extreme observations and potentially spurious outliers
(Doidge, Karolyi, & Stulz, 2004; Durnev & Kim, 2005).
We look to press freedom, government budget trans-
parency and accounting system quality to proxy for socie-
tal pressures to disclose. The press functions as a conduit
for societal isomorphic pressures. Openness in the press
serves both as a model of societal expectations of transpar-
ency, and a transmission mechanism to communicate
stakeholder expectations to managers (Bednar, 2012). Joe,
Louis, and Robinson (2009) show that media coverage of
board ineffectiveness forces ?rms to improve corporate
governance practices. Similarly, Dyck and Zingales (2004)
demonstrate that public opinion pressures (proxied by
newspaper circulation) signi?cantly reduce private bene-
?ts of control for ?rm insiders. We use the Press Freedom
Index compiled by Reporters Without Borders (Press Free-
dom); the index covers 179 countries from the years
2002 to 2012. This index is a compilation of 44 complimen-
tary indicators that together constitute a measure of the
state of media freedom within a country. For consistency
across measures (higher values indicate stronger institu-
tions), we reverse-code the Press Freedom Index.
Budget transparency re?ects information available to
civil society regarding ?scal processes, which have a repu-
tation for opacity in many oil producing countries. We
measure government openness with the International Bud-
get Partnership’s Open Budget Index (OBI), which assesses
94 countries based on government transparency and
accountability in the budgeting process.
To measure the importance of accounting quality across
countries, we use an index of audit and enforcement stan-
dards developed in Brown, Preiato, and Tarca (2013). Data
for this measure are drawn from a multi-wave survey con-
ducted by the International Federation of Accountants. The
index covers 51 countries for the years 2002, 2005, and
2008.
To measure political risk, we employ the widely-used
International Country Risk Guide (ICRG) political risk rating
compiled by the PRS Group. The ICRG rating describes the
risk posed by governments across an array of 12 political
risk components (government stability, socioeconomic
conditions, investment pro?le, internal con?ict, external
con?ict, prevalence of corruption, the degree to which
the military intercedes in politics, religious tensions, ethnic
tensions, law and order, democratic accountability, and
bureaucracy quality).
We also consider an alternative measure of political risk
that focuses on the constraints placed on governments. The
10
Herold databases have been used in research in accounting (Misund
et al., 2005; Teall, 1992), economics (Thompson, 2001), and ?nance
(Heaney & Grundy, 2011; Miller & Upton, 1985). Our database has been
used in strategy research (Click & Weiner, 2010; Jandhyala & Weiner, 2014).
11
The small number of deals per ?rm re?ects the major, infrequent
nature of reserve acquisitions. Our results below do not depend on a few
?rms; the top 10 by deal frequency account for less than 25% of the sample.
A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47 37
Political Constraints 3 index (POLCON) is designed to cap-
ture the political structures within a country that support
a government’s ability to make credible regulatory com-
mitments. Governments facing weaker constraints are
freer to pursue predatory actions, and thus pose a greater
risk to ?rms that disclose investment information
(Henisz, 2006). Higher values of the ICRG and POLCON indi-
ces indicate less political risk and greater constraints,
respectively.
Transparency International’s Corruption Perceptions
Index (CPI) is the corruption measure in widest use.
Intended to capture both the administrative and political
aspects of corruption, the index measures perceptions of
multiple forms of graft. These include bribery of public of?-
cials, kickbacks in public procurement, and embezzlement
of public funds (Transparency International, 2012). While
measures based on expert perceptions have a number of
shortcomings (Thomas, 2010), this measure is appropriate
in the examination of managerial decision-making. Shared
perceptions of corruption are likely to in?uence the cogni-
tive processes employed by decision makers. Greater index
values indicate lower perceptions of corruption.
We also use Global Integrity’s anti-corruption index
(Global Integrity) as a measure of cross-national differences
in the costs of corruption faced by ?rms. This index mea-
sures the accessibility and effectiveness of anti-corruption
mechanisms across countries rather than corruption
directly. The index score is compiled from 300 integrity
indicators gathered from experts and professionals in each
country. Greater scores indicate more constraints on
corruption.
Controls
We control for factors that may shape the incentives of
multinationals to reveal information. First is the price of
oil. High oil prices are associated with greater oil-company
pro?tability and better upstream investment opportuni-
Table 1
Variable de?nitions.
Variables Description Data source
Investment Transparency Categorical: minimal disclosure, partial disclosure, full
disclosure
Herold Database
Assets Log value of buyer’s assets SDC Platinum
Need for External Finance (CapEx – Cash Flow)/CapEx (Rajan & Zingales, 1998) SDC Platinum
Buyer SOE dummy Equals 1 if buyer is a state-owned enterprise SDC Platinum
Company Websites
Petroleum Intelligence Weekly, 2013
Buyer Listed dummy Equals 1 if buyer is listed on a stock exchange Herold Database
SDC Platinum
Company Websites
Petroleum Intelligence Weekly, 2013
Oil Price Log value of 12 month strip price of oil Herold Database
Conventional Reserve
Dummy
Equals 1 if conventional reserve type
b
Herold Database
Cross-border dummy Equals 1 if cross-border transaction Herold Database
Press Freedom Index (PFI) Original scale: 0–115; standardized to l = 0, r = 1 Reporters without borders
Lower values indicated greater press freedom (reverse coded for
analysis)
Data available at en.rsf.com
a
Open Budget Index (OBI) Original scale: 0–100; standardized to l = 0, r = 1 International budget partnership
Higher values indicated more government openness Data available at internationalbudget.org
a
Accounting Quality Index Original scale: 0–56; standardized to l = 0, r = 1 Brown, Preiato, and Tarca (2013)
Higher values indicate greater accounting quality
ICRG Rating Original scale: 0–100; standardized to l = 0, r = 1 The PRS Group
Higher values indicate less political risk
Political Constraints (POLCON
3)
Original scale: 0–1; standardized to l = 0, r = 1 Prof. Witold Henisz, Wharton School
Higher values indicate less political risk Data available at www-
management.wharton.upenn.edu/henisz/
Corruption Perceptions Index
(CPI)
Original scale: 0–10; standardized to l = 0, r = 1 Transparency International
Higher values indicate less corruption Data available at cpi.transparency.org
Global Integrity Index Original scale: 0–100; Standardized to l = 0, r = 1 Global Integrity, 2013
Higher values indicate less corruption Data available at Globalintegrity.org
a
a
Data also available through the World Bank Actionable Governance Indicators (AGI) data portal: agidata.org.
b
Reserves extractable with basic technology (includes royalty interests). High-tech reserve types include bitumen, coalbed, deepwater, enhanced,
frontier, heavy oil, LNG, shale, nonconventional, shallow water, tight gas, and diversi?ed.
38 A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47
ties, suggesting more disclosure (Hossain, Ahmed, &
Godfrey, 2005). At the same time, high oil prices generate
greater rent, which promotes corruption (Jeong & Weiner,
2012) and may reduce incentives for disclosure. Thus, the
predicted sign of the effect is thus unclear. Information
on petroleum prices is from the New York Mercantile
Exchange (NYMEX).
We control for physical characteristics of the reserve.
Deep water, heavy oil and shale reserves, for example,
require both greater capital and technological know-how
to extract than conventional oil reserves. We include a
dummy that indicates whether a reserve is conventional
or unconventional (‘‘high-tech’’), as there is potentially less
private information to reveal for conventional assets.
Reserve characteristics are available through the Herold
database.
Next, we control for characteristics of the ?rm. In
regressions where we test need for external ?nance, we
account for ?rm size by taking the log value of ?rm assets
as reported by SDC Platinum. We also control for the listing
status of the ?rm, as private ?rms tend to be closely held,
and thus have less need of communicating information to a
broad group of investors. Information on listing is obtained
from the Herold database and SDC Platinum. Where histor-
ical listings are missing or inaccurate in these sources,
information is hand-collected from company websites,
stock exchanges, and industry publications.
Last, we include a dummy for state-ownership of the
buyer. While existing theory on voluntary disclosure has
little to say on the differences in transparency between pri-
vate corporations and state-owned ?rms, managers in
state-owned ?rms likely face a different set of incentives.
State-owned enterprises face ‘‘soft’’ budget constraints
(Kornai, 1986), as governments step into ?ll budget short-
falls. While a state-owned enterprise may require addi-
tional external ?nancing to fund investment projects, it is
not subject to the discipline of the capital market
(Megginson, 2005), and thus does not face the same incen-
tive to reveal information as publicly traded ?rms.
A summary of variable de?nitions and data sources is
included in Table 1.
Results
Descriptive statistics
Table 2 reports descriptive statistics for our sample. Of
the 1210 transactions in the database, both price paid
and reserves purchased are revealed (‘‘full transparency’’
below) in 657 transactions (54% of the sample); only one
piece of information is revealed (‘‘partial transparency’’)
in 365 transactions (30%); and neither pricing nor quantity
of reserves is disclosed (‘‘minimal transparency’’) in 188
transactions (16%).
In addition, 372 transactions are cross-border (31% of
the sample). The cross-border and domestic sub-samples
are strikingly different; e.g., only 35% of cross-border trans-
actions are fully transparent vs. 63% of domestic transac-
tions. Table 3 presents differences in disclosure between
cross-border and domestic investments for ?rms that
undertake both. These ?rms are far less likely to fully dis-
close details about their foreign investments than domes-
Table 2
Descriptive statistics.
Variable All transactions Cross-border transactions
Mean SD Min Max Obs Mean SD Min Max Obs
Investment Transparency
Full disclosure 0.54 0.50 0.00 1.00 1210 0.35 0.48 0.00 1.00 372
Partial disclosure 0.30 0.46 0.00 1.00 1210 0.41 0.49 0.00 1.00 372
Minimal disclosure 0.16 0.36 0.00 1.00 1210 0.23 0.42 0.00 1.00 372
Assets (Log) 6.60 2.63 0.34 12.59 742 7.54 3.12 0.69 12.59 179
Need for External Finance À0.16 1.16 À3.09 1.75 688 À0.36 1.34 À3.09 1.75 162
Buyer SOE dummy 0.08 0.27 0.00 1.00 1210 0.16 0.37 0.00 1.00 372
Buyer Listed dummy 0.82 0.39 0.00 1.00 1210 0.78 0.41 0.00 1.00 372
Oil Price (12 month strip, log) 4.01 0.50 2.96 4.99 1210 4.01 0.55 2.96 4.99 372
Conventional Reserve dummy
*
0.58 0.49 0.00 1.00 1210 0.49 0.50 0.00 1.00 372
Cross-border dummy 0.31 0.46 0.00 1.00 1210
Press Freedom Host Rating 0.88 0.15 0.00 1.00 967 0.79 0.20 0.10 1.00 282
Press Freedom Home Rating 0.91 0.13 0.00 1.00 943 0.88 0.18 0.12 1.00 258
OBI Host Rating 71.03 18.25 0.00 87.73 792 57.96 22.82 0.00 87.73 273
OBI Home Rating 75.01 12.49 30.63 88.43 711 67.19 15.19 30.63 88.43 192
Accounting Quality Host Rating 46.99 11.66 6.00 56.00 958 40.02 15.92 6.00 56.00 239
Accounting Quality Home Rating 47.46 10.29 10.00 56.00 1001 42.76 12.37 10.00 56.00 282
ICRG Host Rating 80.46 9.57 36.50 97.00 1207 74.72 13.33 36.50 97.00 369
ICRG Home Rating 82.93 6.16 44.00 97.00 1175 82.79 7.09 54.00 97.00 337
POLCON Host Rating 0.40 0.12 0.00 0.69 1210 0.35 0.18 0.00 0.69 372
POLCON Home Rating 0.42 0.09 0.00 0.69 1174 0.42 0.13 0.00 0.69 336
CPI Host Rating 7.12 2.02 1.78 9.47 1210 5.90 2.63 1.78 9.47 372
CPI Home Rating 7.62 1.45 2.18 9.47 1175 7.49 1.66 2.32 9.47 337
Global Integrity Host Rating 80.40 8.19 41.96 86.73 1100 74.64 10.96 41.96 86.73 296
Global Integrity Home Rating 81.61 6.63 49.79 87.94 1055 78.70 8.65 49.79 87.94 251
*
Royalty interests also coded as conventional; Higher cost reserve types include bitumen, coalbed, deepwater, enhanced, frontier, heavy oil, LNG, shale,
nonconventional, shallow water, tight gas, and diversi?ed.
A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47 39
tic, and are more likely to disclose partial information.
These differences are statistically signi?cant at the 0.1%
level (v
2
= 21.5).
Panels AandBof Table 4present the correlationmatrices
for the full sample and cross-border sub-sample, respec-
tively. Panel A shows that Full Disclosure is positively corre-
lated with Need for External Finance, Oil Price, Conventional
Reserves, and all institutional variables, and is negatively
correlated with cross-border transactions. Not surprisingly,
institutional variables are correlated (the median correla-
tion among host-country institutional variables in cross-
border transactions is 0.71). Due to the potential for multi-
collinearity among institutional measures, we allow each
of these measures to enter in separate regressions.
Multinomial logit results
As our dependent variable representing choice of invest-
ment transparency is categorical with three outcomes (full,
Table 3
Disclosure by ?rms with both crossborder and domestic transactions.
Number of
transactions
Full
disclosure
Partial
disclosure
Minimal
disclosure
Total
Crossborder 47 51 16 114
41% 45% 14% 100%
Domestic 86 25 10 121
71% 21% 8% 100%
Total 133 76 26 235
Differences are signi?cant (Pearson v
2
(2) = 21.5, Pr = 0.000).
Table 4
Correlations.
Panel A: Full Sample 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
1 Full Disclosure 1.00
2 Partial Disclosure -0.72 1.00
3 Minimal Disclosure -0.47 -0.28 1.00
4 Press Freedom Host Rating 0.09 -0.03 -0.08 1.00
5 Press Freedom Home Rating 0.12 -0.11 -0.02 0.48 1.00
6 OBI Host Rating 0.10 -0.04 -0.08 0.73 0.28 1.00
7 OBI Home Rating 0.13 -0.12 -0.02 0.52 0.67 0.65 1.00
8 Accounting Quality Host Rating 0.14 -0.08 -0.10 0.65 0.40 0.70 0.49 1.00
9 Accounting Quality Home Rating 0.22 -0.15 -0.13 0.37 0.48 0.33 0.53 0.69 1.00
10 ICRG Host Rating 0.09 -0.03 -0.09 0.79 0.38 0.77 0.46 0.62 0.24 1.00
11 ICRG Home Rating 0.10 -0.07 -0.05 0.38 0.75 0.17 0.55 0.25 0.38 0.43 1.00
12 POLCON Host Rating 0.05 0.02 -0.09 0.71 0.35 0.50 0.43 0.51 0.32 0.55 0.36 1.00
13 POLCON Home Rating 0.03 -0.01 -0.03 0.34 0.73 0.12 0.54 0.27 0.24 0.26 0.58 0.46 1.00
14 CPI Host Rating 0.10 -0.02 -0.11 0.86 0.47 0.79 0.53 0.70 0.35 0.89 0.45 0.69 0.38 1.00
15 CPI Home Rating 0.12 -0.07 -0.07 0.43 0.81 0.20 0.60 0.34 0.54 0.38 0.86 0.43 0.68 0.51 1.00
16 Global Integrity Host Rating 0.08 -0.04 -0.07 0.76 0.31 0.92 0.62 0.68 0.37 0.65 0.17 0.46 0.13 0.67 0.21 1.00
17 Global Integrity Home Rating 0.10 -0.12 0.02 0.39 0.67 0.57 0.92 0.41 0.51 0.27 0.41 0.22 0.31 0.31 0.44 0.63 1.00
18 Assets (Log) 0.06 0.01 -0.11 -0.19 -0.23 -0.13 -0.24 -0.16 -0.22 -0.23 -0.32 -0.17 -0.14 -0.25 -0.32 -0.11 -0.10 1.00
19 Need for External Finance 0.01 -0.02 0.00 0.13 0.11 0.14 0.19 0.12 0.16 0.07 0.11 0.09 0.08 0.12 0.16 0.08 0.09 -0.46 1.00
20 Buyer SOE dummy -0.13 0.05 0.12 -0.31 -0.52 -0.31 -0.53 -0.33 -0.48 -0.26 -0.42 -0.23 -0.28 -0.31 -0.51 -0.35 -0.52 0.35 -0.23 1.00
21 Buyer Listed dummy 0.28 -0.24 -0.09 0.03 0.16 0.00 0.14 0.00 0.11 0.03 0.15 0.09 0.17 0.05 0.18 -0.01 0.09 0.01 0.04 -0.15 1.00
22 Oil Price (12 month strip, log) 0.15 -0.10 -0.08 -0.07 -0.06 -0.01 -0.08 0.44 0.55 -0.17 -0.25 -0.04 -0.06 -0.08 -0.06 0.02 -0.03 0.13 0.11 0.01 0.02 1.00
23 Conventional Reserve dummy 0.18 -0.10 -0.12 -0.11 -0.06 -0.07 -0.04 -0.12 -0.06 -0.07 -0.04 -0.08 -0.09 -0.11 -0.07 -0.10 -0.07 -0.15 0.06 -0.02 0.02 0.05 1.00
24 Cross-border dummy -0.25 0.16 0.14 -0.37 -0.11 -0.52 -0.38 -0.35 -0.29 -0.40 -0.01 -0.23 0.04 -0.41 -0.06 -0.43 -0.25 0.20 -0.10 0.22 -0.06 0.00 -0.13 1.00
Panel B: Cross-border Sub-sample 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
1 Full Disclosure 1.00
2 Partial Disclosure -0.62 1.00
3 Minimal Disclosure -0.41 -0.46 1.00
4 Press Freedom Host Rating -0.13 0.17 -0.05 1.00
5 Press Freedom Home Rating 0.13 -0.14 0.00 0.05 1.00
6 OBI Host Rating -0.10 0.16 -0.07 0.66 0.04 1.00
7 OBI Home Rating -0.03 -0.04 0.08 0.05 0.62 0.02 1.00
8 Accounting Quality Host Rating -0.07 0.13 -0.08 0.65 0.18 0.66 0.10 1.00
9 Accounting Quality Home Rating 0.24 -0.12 -0.15 0.05 0.37 -0.01 0.26 0.23 1.00
10 ICRG Host Rating -0.09 0.10 -0.01 0.71 0.06 0.74 0.18 0.70 -0.08 1.00
11 ICRG Home Rating 0.16 -0.17 0.00 -0.04 0.68 0.09 0.49 0.04 0.35 0.06 1.00
12 POLCON Host Rating -0.08 0.11 -0.03 0.66 0.09 0.39 0.17 0.46 0.11 0.43 0.11 1.00
13 POLCON Home Rating 0.02 -0.05 0.03 0.11 0.73 0.09 0.43 0.13 0.05 0.07 0.44 0.18 1.00
14 CPI Host Rating -0.10 0.12 -0.03 0.81 0.13 0.74 0.20 0.77 -0.01 0.89 0.14 0.57 0.14 1.00
15 CPI Home Rating 0.17 -0.16 -0.01 -0.03 0.73 0.08 0.46 0.05 0.53 0.02 0.85 0.11 0.49 0.09 1.00
16 Global Integrity Host Rating -0.08 0.20 -0.14 0.81 0.01 0.86 0.03 0.78 0.05 0.73 0.05 0.50 0.05 0.77 0.04 1.00
17 Global Integrity Home Rating 0.13 -0.13 0.00 0.02 0.77 0.05 0.87 0.15 0.41 0.13 0.62 0.09 0.40 0.15 0.58 0.04 1.00
18 Assets (Log) -0.08 0.12 -0.05 0.07 -0.14 0.04 -0.12 -0.09 -0.36 -0.04 -0.34 0.00 -0.01 -0.04 -0.34 -0.07 -0.22 1.00
19 Need for External Finance 0.03 -0.05 0.02 0.10 0.12 0.15 0.19 0.03 0.19 0.02 0.20 0.05 0.08 0.07 0.26 0.08 0.28 -0.52 1.00
20 Buyer SOE dummy -0.10 -0.02 0.14 -0.03 -0.46 -0.08 -0.38 -0.14 -0.44 -0.04 -0.41 -0.06 -0.19 -0.07 -0.53 -0.17 -0.50 0.36 -0.31 1.00
21 Buyer Listed dummy 0.30 -0.30 0.01 -0.09 0.30 -0.05 0.26 -0.16 0.22 -0.10 0.16 0.01 0.19 -0.11 0.24 -0.12 0.31 0.08 0.07 -0.10 1.00
22 Oil Price (12 month strip, log) 0.10 0.01 -0.12 -0.04 -0.09 -0.05 -0.33 0.30 0.50 -0.15 -0.25 -0.04 -0.06 -0.12 -0.05 -0.04 -0.21 0.07 0.00 0.07 0.07 1.00
23 Conventional Reserve dummy 0.15 -0.06 -0.10 -0.27 -0.04 -0.19 -0.09 -0.27 -0.03 -0.23 -0.03 -0.17 -0.10 -0.30 -0.05 -0.17 -0.03 -0.16 0.04 0.00 0.02 0.10 1.00
Notes: Correlations for country-level institutional measures are shaded.
40 A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47
partial, and minimal disclosure), we specify a multinomial
logit model (Greene, 2000). We report the results for each
model inthree columns. The ?rst two columns compare Par-
tial Disclosure andFull Disclosure to a baseline of Minimal Dis-
closure; the third column compares Full Disclosure to a
baseline of Partial Disclosure. For each independent variable
we present the coef?cient estimate (beta) with the odds
ratio (calculated as e
beta
) in italics. The odds ratio represents
the change inthe odds of moving fromthe baseline level to a
higher disclosure level for a one unit increase in the inde-
pendent variable; ratios greater than one indicate increased
odds, ratios less than one, decreased odds.
Our ?rm-level data are for the most recent reporting
date prior to the transaction announcement. Institutional
variables are for the year prior to announcement to re?ect
information available to managers. Following the treat-
ment of institutional variables in prior work (Guerreiro
et al., 2012), we scale all country-level indices to have
mean zero and standard deviation of one to enable compa-
rability across coef?cients (Tabachnick & Fidell, 2001).
Standard errors are robust to heteroskedasticity, and clus-
tered by home country.
12
Table 5 presents the multinomial logit results of our
regression of investment transparency on ?rm-level deter-
minants. Our ?rst hypothesis tests whether multinational
investment is associated with greater disclosure (H1A) or
less disclosure (H1A
0
) than domestic investment, as pre-
dicted by the global-diversi?cation and investor-sophistica-
tion theories, respectively. The investor sophistication view
(H1A
0
) is supported: cross-border transactions are less
transparent. The bivariate difference noted in Table 3 holds
in a multivariate setting. The coef?cient for the Cross-bor-
der dummy for Full Disclosure in Model 1 is negative and
signi?cant; this coef?cient remains signi?cant in later
speci?cations that control for host- and home-country
characteristics. The odds ratios of 0.178 (respectively,
0.330) on Full vs. Minimal (respectively, Full vs. Partial)
can be interpreted as an 82.8% (respectively, 67.0%)
decrease in the odds of full disclosure vs. minimal disclo-
sure (respectively, vs. partial disclosure) in cross-border
transactions.
In addition, when we separately estimate our cross-bor-
der and domestic sub-samples, we see that the Buyer Listed
dummy is negative and signi?cant in cross-border transac-
tions (Model 2), but insigni?cant in domestic transactions
(Model 3). These results indicate that publicly-listed ?rms
are less transparent than unlisted ?rms in cross-border,
Table 5
Multinomial logit regressions of investment transparency on ?rm-level determinants.
Variables (1) Full Sample (2) Cross-border Sub-sample (3) Domestic Sub-sample
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Need for External Finance À0.033 À0.106 À0.073 À0.026 À0.137 À0.110 0.048 À0.012 À0.060
(0.125) (0.103) (0.101) (0.183) (0.114) (0.154) (0.191) (0.163) (0.123)
Odds Ratio 0.968 0.899 0.930 0.974 0.872 0.896 1.049 0.988 0.942
Assets (Log) 0.119 0.168 0.049 0.144 À0.014 À0.158 0.205 0.390
**
0.185
***
(0.149) (0.185) (0.103) (0.152) (0.167) (0.101) (0.198) (0.175) (0.070)
Odds Ratio 1.126 1.183 1.050 1.155 0.986 0.854 1.228 1.477 1.203
Cross-border dummy À0.616 À1.724
***
À1.108
***
(0.440) (0.473) (0.247)
Odds Ratio 0.540 0.178 0.330
Buyer SOE dummy À1.207
**
À2.653
**
À1.446
**
À1.589
*
À1.423 0.166 À1.568
*
À4.495
***
À2.927
***
(0.564) (1.089) (0.685) (0.849) (0.957) (0.615) (0.918) (1.202) (0.649)
Odds Ratio 0.299 0.070 0.236 0.204 0.241 1.181 0.208 0.011 0.054
Buyer Listed dummy À0.881 À0.628 0.254 À15.869
***
À14.524
***
1.345 0.491 0.337 À0.154
(0.749) (0.888) (0.496) (0.896) (1.269) (0.869) (0.606) (0.602) (0.529)
Odds Ratio 0.414 0.534 1.289 0.000 0.000 3.838 1.634 1.401 0.857
Oil Price (12 month strip, log) 0.459
***
1.096
***
0.636
***
0.768
**
0.753
*
À0.016 0.197 1.005
***
0.807
***
(0.158) (0.270) (0.184) (0.350) (0.427) (0.260) (0.193) (0.198) (0.196)
Odds Ratio 1.582 2.992 1.889 2.155 2.123 0.984 1.218 2.732 2.241
Conventional Reserve dummy 0.576
**
1.298
***
0.722
***
0.783
**
1.054
**
0.271 0.492
*
1.454
***
0.962
***
(0.245) (0.304) (0.222) (0.397) (0.490) (0.568) (0.263) (0.363) (0.187)
Odds Ratio 1.779 3.662 2.059 2.188 2.869 1.311 1.636 4.280 2.617
Observations 688 162 526
Firms 344 116 255
Pseudo R-squared 0.113 0.065 0.112
Notes: For the ?rst two columns in each regression, partial disclosure and full disclosure are compared to a baseline of minimal disclosure. For the third
column, full disclosure is compared to a baseline of partial disclosure. Institutional variables are standardized and oriented such that higher values indicate
stronger institutions. Odds ratios are presented in italics. Robust standard errors clustered by home country are in parentheses. Constant values are not
reported.
*
Signi?cant at the 10% level.
**
Signi?cant at the 5% level.
***
Signi?cant at the 1% level.
12
Because of the small number of deals per ?rm, we are not able to
cluster errors at the ?rm level.
A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47 41
but not in domestic, transactions. Listed-?rm shareholders
rely more on corporate reporting than shareholders of
unlisted ?rms (which tend to be closely-held). If cross-bor-
der information frictions reduce the bene?ts of disclosure,
it follows that disclosures will be less useful for listed
?rms’ foreign transactions.
The agency-theoretic hypothesis that ?rms in need of
external ?nance will make more transparent investments
(H1B) is not supported; the coef?cients on Need for External
Finance are not statistically signi?cant in any speci?cation.
Together with H1A, these results suggest that agency-the-
oretic considerations are not a dominant driver of invest-
ment-level disclosures.
Table 6 introduces institutional factors. The models are
designed to test the hypothesis of greater disclosure when
host- and home- countries’ societal institutions are more
transparent. We examine the cross-border sample, as
home and host measures are highly correlated in the full
sample where the preponderance of domestic transactions
means that home- and host-countries are the same for
about 70% of the observations.
13
Examining host countries ?rst, we ?nd that partial dis-
closure is favored over both full disclosure and minimal
disclosure in all speci?cations, although statistical signi?-
cance varies. The odds ratios for Partial vs. Minimal in Mod-
els 1 and 2 indicate that one standard deviation increases
in Press Freedom and OBI result in 29.2% and 70.4%
increases in the odds of partial disclosure, respectively.
As discussed above, managers may choose to appear more
transparent in the host country even if the act of disclosure
itself provides limited information. These results are con-
sistent with seeking legitimacy (H2A).
We ?nd mixed results for the hypothesis that MNEs
from more transparent societies behave more transpar-
ently (H2B). The coef?cients on Home Country Rating in
Model 3 suggests that better home country accounting
quality results in greater odds of full disclosure (a one stan-
dard deviation increase in accounting quality increases the
odds of full disclose by 59.0% over minimal disclosure, and
89.6% over partial disclosure). This result supports the
hypothesis that ?rms from countries with higher expecta-
tions of transparency disclose more abroad. The results for
Press Freedom and OBI, however, are inconsistent or in the
wrong direction.
In Table 7, we present the results of multinomial logit
regressions of investment transparency on home- and
host-country political risk and corruption. We ?nd little
support for a relationship between host-country political
risk and investment transparency (H3A). Only the coef?-
cients for POLCON in Model 2 are signi?cant, suggesting
that ?rms investing in countries with better political con-
straints tend to prefer partial disclosure over full disclosure
Table 6
Multinomial logit regressions of investment transparency on societal institutions for cross-border transactions.
(1) Press Freedom (2) Open Budget Index (3) Accounting Quality
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Host Country Rating 0.256
**
À0.050 À0.306
***
0.533
***
0.173 À0.360
***
0.135 À0.036 À0.171
(0.124) (0.090) (0.093) (0.186) (0.215) (0.126) (0.181) (0.113) (0.145)
Odds Ratio 1.292 0.951 0.736 1.704 1.189 0.698 1.145 0.965 0.843
Home Country Rating À0.219
**
À0.028 0.191
*
À0.412
*
À0.344
**
0.068 À0.176 0.464
**
0.640
***
(0.093) (0.085) (0.104) (0.243) (0.150) (0.224) (0.142) (0.188) (0.204)
Odds Ratio 0.803 0.972 1.210 0.662 0.709 1.070 0.839 1.590 1.896
Buyer SOE dummy À0.978
**
À0.872
**
0.106 À1.217
*
À0.856 0.361 À0.710
**
À1.005
**
À0.294
(0.432) (0.401) (0.474) (0.678) (0.539) (0.501) (0.355) (0.418) (0.441)
Odds Ratio 0.376 0.418 1.112 0.296 0.425 1.435 0.492 0.366 0.745
Buyer Listed dummy À0.226 1.329
**
1.555
**
À0.363 0.898 1.261 À0.736 0.810 1.546
**
(0.585) (0.601) (0.611) (0.877) (1.145) (0.840) (0.552) (0.677) (0.747)
Odds Ratio 0.798 3.777 4.735 0.696 2.455 3.529 0.479 2.248 4.693
Oil Price (12 month strip, log) 1.202
**
0.595 À0.608 0.288 0.318 0.031 1.051
***
0.120 À0.931
*
(0.581) (0.420) (0.372) (0.343) (0.380) (0.335) (0.376) (0.316) (0.480)
Odds Ratio 3.327 1.813 0.544 1.334 1.374 1.031 2.861 1.127 0.394
Conventional Reserve dummy 0.249 0.013 À0.236 0.490 0.953
*
0.462 0.703
*
0.546 À0.158
(0.265) (0.293) (0.354) (0.551) (0.524) (0.606) (0.390) (0.607) (0.612)
Odds Ratio 1.283 1.013 0.790 1.632 2.593 1.587 2.020 1.726 0.854
Observations 258 127 210
Firms 184 89 150
Pseudo R-squared 0.060 0.092 0.098
Notes: For the ?rst two columns in each regression, partial disclosure and full disclosure are compared to a baseline of minimal disclosure. For the third
column, full disclosure is compared to a baseline of partial disclosure. Institutional variables are standardized and oriented such that higher values indicate
stronger institutions. Odds ratios are presented in italics. Robust standard errors clustered by home country are in parentheses. Constant values are not
reported.
*
Signi?cant at the 10% level.
**
Signi?cant at the 5% level.
***
Signi?cant at the 1% level.
13
The median correlation between home and host institutional measures
is 0.51 in the full sample and 0.06 in the cross-border sample; however, our
results are robust across samples.
42 A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47
or minimal disclosure. The odds ratio on Host Country Rat-
ing in Model 2 suggests that a one standard deviation
increase in political constraints increases the odds of Par-
tial vs. Minimal by approximately 16% and the odds of Par-
tial vs. Full by approximately 15%. In contrast, we ?nd no
signi?cance for Host Country Rating on ICRG, or either of
our corruption measures.
Our results on Home Country Rating show a generally
strong relation between freedom from political risk and
corruption, and investment transparency. Our strongest
result comes from the ICRG political risk rating and the
CPI rating. The coef?cients on Home Country Rating in
Model 1 suggest ?rms in better-governed countries tend
to favor full disclosure. A one standard deviation increase
in the quality of political institutions increases the odds
of full disclosure over partial by more than 55%. Further-
more, when we introduce ?nancing needs in Table 8,
Models 1 and 2, we see that ?rms consistently prefer
higher levels of disclosure as home political institutions
improve. However, the Home Country Rating coef?cients
for our alternative measure of political risk, POLCON,
are insigni?cant across all speci?cations in Tables 7 and
8. Overall, we interpret these ?ndings as largely consis-
tent with the hypothesis that ?rms are more open with
their investments when home country institutions are
better.
Table 8 also tests the ‘‘Twin Agency Problem’’ that polit-
ical risks and agency problems are complements. We ?nd
little agreement on the signs of these coef?cients, and none
are statistically signi?cant save for the Partial vs. Minimal
coef?cients in Models 1 and 3. These vary in signi?cance
but are not directionally consistent with our hypothesis.
Thus, H3C is not supported.
Discussion and conclusion
Our results support the view that investment transpar-
ency represents a strategic choice. On one hand, civil soci-
ety demands transparency in exchange for legitimacy. On
the other, reducing information asymmetries between
the ?rm and interested third parties such as predatory
governments and rent-seeking public of?cials exposes
the MNE to political risk and the costs of corruption.
Foreign transactions are signi?cantly less transparent
across all speci?cations and samples. These ?ndings are
robust to the MNE’s institutional environment – less infor-
mation is disclosed about asset purchases abroad, even
after controlling for home and host country institutions.
This ?nding supports the investor-sophistication strand
rather than the global-diversi?cation strand of the account-
ing literature: cross-border information frictions reduce
the ?rm’s incentive to disclose.
This ?nding suggests that capital markets, the primary
audience examined in traditional studies of voluntary
disclosure, are less in?uential in the multinational
context. Instead, we ?nd that institutional rather than
agency-theoretic considerations drive MNE investment-
level disclosures; need for external ?nancing is unrelated
to the ?rm’s choice of investment transparency. Our theory
Table 7
Multinomial logit regressions of investment transparency on political risk and corruption in cross-border transactions.
(1) ICRG (2) POLCON (3) CPI (4) Global Integrity
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Host Country Rating 0.090 À0.004 À0.094 0.149
**
À0.017 À0.165
***
0.151 0.033 À0.118 0.148 0.047 À0.100
(0.096) (0.087) (0.095) (0.062) (0.063) (0.063) (0.126) (0.077) (0.119) (0.220) (0.192) (0.130)
Odds Ratio 1.094 0.996 0.910 1.161 0.983 0.848 1.163 1.034 0.889 1.160 1.048 0.905
Home Country Rating À0.233
**
0.209 0.441
***
À0.108 À0.082 0.026 À0.321
***
0.070 0.390
***
À0.101 0.075 0.176
(0.109) (0.182) (0.154) (0.101) (0.131) (0.079) (0.120) (0.144) (0.134) (0.113) (0.138) (0.174)
Odds Ratio 0.792 1.232 1.554 0.898 0.921 1.026 0.725 1.073 1.477 0.904 1.078 1.192
Buyer SOE dummy À0.925
***
À0.736
**
0.189 À0.599 À0.922
***
À0.323 À1.091
***
À0.786
**
0.305 À1.036
**
À0.904
**
0.132
(0.353) (0.325) (0.447) (0.378) (0.320) (0.407) (0.417) (0.313) (0.461) (0.481) (0.452) (0.627)
Odds Ratio 0.397 0.479 1.208 0.549 0.398 0.724 0.336 0.456 1.357 0.355 0.405 1.141
Buyer Listed dummy À0.401 1.391
**
1.792
***
À0.267 1.601
***
1.868
***
À0.245 1.516
***
1.761
***
À0.670 0.645 1.315
*
(0.473) (0.573) (0.607) (0.509) (0.572) (0.566) (0.486) (0.571) (0.584) (0.547) (0.735) (0.685)
Odds Ratio 0.670 4.019 6.001 0.766 4.958 6.475 0.783 4.554 5.818 0.512 1.906 3.725
Oil Price (12 month strip, log) 0.559
**
0.827
***
0.269 0.578
**
0.694
***
0.116 0.616
**
0.695
***
0.079 0.810
*
0.811
**
0.001
(0.268) (0.232) (0.249) (0.245) (0.222) (0.227) (0.250) (0.215) (0.206) (0.471) (0.369) (0.221)
Odds Ratio 1.749 2.286 1.309 1.782 2.002 1.123 1.852 2.004 1.082 2.248 2.250 1.001
Conventional Reserve dummy 0.361
*
0.803
**
0.443 0.368 0.774
**
0.406 0.379 0.817
**
0.437 0.047 0.477 0.431
(0.206) (0.382) (0.435) (0.234) (0.330) (0.385) (0.232) (0.382) (0.436) (0.337) (0.456) (0.502)
Odds Ratio 1.435 2.232 1.557 1.445 2.168 1.501 1.461 2.264 1.548 1.048 1.611 1.539
Observations 334 336 337 202
Firms 235 236 237 148
Pseudo R-squared 0.074 0.063 0.072 0.053
Notes: For the ?rst two columns in each regression, partial disclosure and full disclosure are compared to a baseline of minimal disclosure. For the third
column, full disclosure is compared to a baseline of partial disclosure. Institutional variables are standardized and oriented such that higher values indicate
stronger institutions. Odds ratios are presented in italics. Robust standard errors clustered by home country are in parentheses. Constant values are not
reported.
*
Signi?cant at the 10% level.
**
Signi?cant at the 5% level.
***
Signi?cant at the 1% level.
A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47 43
does not explicitly explain this ?nding, but we can offer
two insights. First, the ?nancial reporting context in which
prior work has established a link between voluntary dis-
closure and the ?rm’s cost of capital considers the ?rm as
a whole. Yet, the external ?nancing needs of the entire ?rm
may not be representative of the ?nancing needs of an
individual investment project. Second, this result may be
a secondary effect of home bias resulting from investors’
discounting foreign information, as it is consistent with
foreign investment disclosures being less useful for raising
capital.
Countries marked by high normative expectations of
transparency tend to encourage greater transparency in
investment disclosures. Yet, home- and host-country
norms have notably different effects. Greater host-country
norms are strongly correlated with partial disclosure,
while ?rms from home countries with higher expectations
of transparency are less likely to release only partial
information.
We interpret these results as consistent with legiti-
macy-seeking behavior. Partial transparency is adequate
in the host country. Limited disclosure is less likely to sat-
isfy civil society in the home country, however, for two
reasons. The ?rst involves familiarity; home country stake-
holders are better acquainted with the ?rm, and thus may
need more speci?c information to access whether corpo-
rate actions are consistent with the values of society. The
second is one of embeddedness. While oil industry MNEs
operate assets in several countries around the world, it is
the institutional logic of the home-country that is most
Table 8
Multinomial logit regressions of investment transparency on political risk and ?rm-level interactions in cross-border transactions.
(1) ICRG (2) ICRG (3) POLCON (4) POLCON
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Host Country Rating 0.173 0.076 À0.097 0.254 0.113 À0.141 0.035 À0.036 À0.072 0.067 À0.009 À0.076
(0.161) (0.175) (0.197) (0.199) (0.162) (0.209) (0.102) (0.155) (0.119) (0.125) (0.151) (0.124)
Odds Ratio 1.189 1.079 0.908 1.289 1.120 0.868 1.036 0.965 0.931 1.069 0.991 0.927
Home Country Rating À0.055 0.546
**
0.601
***
0.133 0.879
***
0.746
***
0.026 À0.130 À0.156 0.044 À0.096 À0.140
(0.236) (0.241) (0.148) (0.282) (0.277) (0.160) (0.203) (0.210) (0.176) (0.213) (0.231) (0.149)
Odds Ratio 0.946 1.726 1.824 1.142 2.408 2.109 1.026 0.878 0.856 1.045 0.908 0.869
Need for External
Finance
À0.146 À0.128 0.018 À0.035 À0.179 À0.144 À0.200 À0.316 À0.116 À0.025 À0.169 À0.144
(0.211) (0.231) (0.211) (0.177) (0.181) (0.185) (0.232) (0.216) (0.152) (0.194) (0.115) (0.150)
Odds Ratio 0.864 0.880 1.018 0.966 0.836 0.866 0.819 0.729 0.890 0.975 0.845 0.866
Need for Ext.
Fin. Â Host Rating
À0.247
*
À0.046 0.201 À0.228
**
À0.162 0.066
(0.136) (0.151) (0.123) (0.104) (0.103) (0.085)
Odds Ratio 0.781 0.955 1.223 0.796 0.850 1.068
Need for Ext.
Fin. Â Home Rating
0.268 0.411 0.144 À0.018 0.015 0.033
(0.285) (0.258) (0.104) (0.150) (0.110) (0.130)
Odds Ratio 1.307 1.508 1.155 0.982 1.015 1.034
Assets (Log) 0.155 0.053 À0.102 0.144 0.067 À0.077 0.122 À0.045 À0.167
*
0.139 À0.022 À0.161
(0.156) (0.163) (0.097) (0.140) (0.151) (0.101) (0.154) (0.172) (0.095) (0.161) (0.175) (0.098)
Odds Ratio 1.168 1.054 0.903 1.155 1.069 0.926 1.130 0.956 0.846 1.149 0.978 0.851
Buyer SOE dummy À1.440
*
À1.150 0.289 À1.472 À1.155 0.317 À1.559
**
À1.603
*
À0.044 À1.503
*
À1.539
*
À0.035
(0.825) (0.952) (0.834) (0.905) (1.016) (0.857) (0.789) (0.928) (0.667) (0.807) (0.927) (0.636)
Odds Ratio 0.237 0.317 1.335 0.229 0.315 1.373 0.210 0.201 0.957 0.222 0.215 0.966
Buyer Listed dummy À15.749
***
À15.702
***
0.048 À14.235
***
À14.202
***
0.033 À15.524
***
À13.352
***
2.171 À15.469
***
À13.421
***
2.048
(1.102) (1.036) (0.980) (1.061) (1.070) (1.180) (1.156) (1.358) (1.478) (1.097) (1.489) (1.511)
Odds Ratio 0.000 0.000 1.049 0.000 0.000 1.034 0.000 0.000 8.767 0.000 0.000 7.752
Oil Price (12 month
strip, log)
0.809
**
1.038
***
0.228 0.844
**
1.133
***
0.290 0.880
***
0.923
**
0.043 0.743
**
0.793
*
0.050
(0.379) (0.390) (0.223) (0.384) (0.390) (0.213) (0.312) (0.425) (0.285) (0.344) (0.463) (0.279)
Odds Ratio 2.246 2.824 1.256 2.326 3.105 1.336 2.411 2.517 1.044 2.102 2.210 1.051
Conventional Reserve
dummy
0.953
***
1.158
**
0.205 1.060
***
1.283
**
0.223 0.933
**
1.046
*
0.114 0.820
**
0.955
*
0.135
(0.315) (0.584) (0.712) (0.373) (0.626) (0.750) (0.378) (0.562) (0.581) (0.375) (0.529) (0.619)
Odds Ratio 2.593 3.184 1.228 2.886 3.607 1.250 2.542 2.846 1.121 2.270 2.599 1.145
Observations 158 158 160 160
Firms 113 113 114 114
Pseudo R-squared 0.110 0.113 0.080 0.069
Notes: For the ?rst two columns in each regression, partial transparency and full transparency are compared to a baseline of minimal transparency. For the
third column, full transparency is compared to a baseline of partial transparency. Institutional variables are standardized and oriented such that higher
values indicate stronger institutions. Odds ratios are presented in italics. Robust standard errors clustered by home country are in parentheses. Constant
values are not reported.
*
Signi?cant at the 10% level.
**
Signi?cant at the 5% level.
***
Signi?cant at the 1% level.
44 A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47
likely to inform managerial preferences for disclosure.
Indeed, of all of our measures of transparency norms, our
strongest results suggest that ?rms from home countries
with higher accounting standards are the most likely to
fully disclose.
This work extends the accounting literature on volun-
tary disclosure (Healy & Palepu, 2001; Verrecchia, 2001)
empirically and theoretically. We present a novel way of
measuring voluntary disclosure that, unlike prior studies
that look at aggregate outcome-based measures, leverages
missing data on reserve transactions to test theories about
disclosure decisions at the investment-level. We demon-
strate that transparency is not merely a ?rm characteristic,
but a non-market strategy ?rms employ to manage various
stakeholder audiences. This concept complements recent
work by Ahern and Sosyura (2014), who examine how
?rms strategically control information dissemination in
the media. We also contribute to a nascent but growing lit-
erature exploring whether countries matter for transpar-
ency and voluntary disclosure (Healy & Serafeim, 2013;
Shi et al., 2012).
This study is not without limitations. While our data
allow us to circumvent a number of empirical challenges,
as in all transaction-level research we are unable to
observe acquisitions that are never announced or captured
by our data provider.
14
This is unlikely to be a serious prob-
lem, given the high pro?le and extensive scrutiny of our
empirical setting. Still, our study assumes such transactions,
if they exist, introduce no systematic bias.
We are also unable to determine systematically
whether or not the prices disclosed are over- or under-pay-
ments, as reserve and pricing data are available only when
disclosed. The question of whether managers reduce trans-
parency to hide overpayment would be an interesting
extension of this work. Additionally, an alternative
explanation for why some multinationals might or might
not disclose is the presence of a ?rm-wide policy on
disclosure. While this is a possibility, we have found no
examples of such a policy. Further, it is inconsistent with
our sample. The data in Table 3 indicate signi?cant varia-
tion within ?rms across transactions, suggesting this is
not the case.
The implications of this study should be of interest to
managers and policy makers. We contribute to the ongoing
debate in the extractive industries by scrutinizing the stra-
tegic dimensions of disclosure. While ceteris paribus,
greater transparency should be good for ?rm stakeholders
and society, removing all information asymmetries may
leave the ?rm vulnerable to external threats. Our theory
and evidence suggest the proper management of transpar-
ency and information asymmetries should be considered
an important component of MNE strategy.
We believe this study offers a number of promising ave-
nues for future research. First, we limit our study to the
global petroleum industry, which is particularly sensitive
to institutional risks. Further research is needed to general-
ize these ?ndings. Second, institutional determinants such
as normative expectations and political risk may interact,
creating potential second-order effects. Lastly, our research
raises the question of whether traditional theories of trans-
parency and voluntary disclosure are applicable to emerg-
ing-market multinationals, many of which are owned or
managed by the state. While we ?nd state-owned ?rms
to be less transparent on average, the political risks to SOEs
likely differ from privately held companies as they are
already the property of a sovereign government. We leave
these questions to future research.
Acknowledgements
We are grateful to IHS Herold for access to data; to
Meghana Ayyagari, Reid Click, Liesl Riddle, Jennifer Spen-
cer, Youli Zou, the Editor (Robert Bloom?eld), two anony-
mous reviewers; and participants at the GW
International Business Workshop and the Academy of
International Business annual meeting for comments;
and to the GW CIBER and Institute for Integrating Statistics
in Decision Sciences for ?nancial support. All omissions
and mistakes are the authors’.
References
Ahearne, A., Griever, W., & Warnock, F. (2004). Information costs and
home bias: An analysis of US holdings of foreign equities. Journal of
International Economics, 62(2), 313–336.
Ahern, K. R., & Sosyura, D. (2014). Who writes the news? Corporate press
releases during merger negotiations. The Journal of Finance, 69(1),
241–291.
Arnott, R. (2004). Oil and gas reserves: Communication with the ?nancial
sector. Oxford Institute for Energy Studies & Chatham House. .
Bednar, M. K. (2012). Watchdog or lapdog? A behavioral view of the
media as a corporate governance mechanism. Academy of
Management Journal, 55, 131–150.
Berry, K. T., & Wright, C. J. (2001). The value relevance of oil and gas
disclosures: An assessment of the market’s perception of ?rms’ effort
and ability to discover reserves. Journal of Business Finance &
Accounting, 28(5–6), 741–769.
Bertrand, M., & Mullainathan, S. (2000). Do CEOs set their own pay? The
ones without principals do (No. w7604). National Bureau of Economic
Research.
Beyer, A., Cohen, D. A., Lys, T. Z., & Walther, B. R. (2010). The ?nancial
reporting environment: Review of the recent literature. Journal of
Accounting and Economics, 50(2), 296–343.
Brown, P. R., Preiato, J. P., & Tarca, A. (2013). Measuring country differences
in enforcement of accounting standards: An audit and enforcement
proxy. Available at SSRN 2286029.
Cahan, S. F., Rahman, A., & Perera, H. (2005). Global diversi?cation and
corporate disclosure. Journal of International Accounting Research, 4,
73–94.
Callen, J. L., Hope, O. K., & Segal, D. (2005). Domestic and foreign earnings,
stock return variability, and the impact of investor sophistication.
Journal of Accounting Research, 43(3), 377–412.
Capron, L., Mitchell, W., & Swaminathan, A. (2001). Asset divestures
following horizontal acquisitions: A dynamic view. Strategic
Management Journal, 22, 817–844.
Chen, J. J., Tan, Y., Cheng, X., & Gong, S. (2013). Do higher value ?rms
voluntarily disclose more information? Evidence from China. In Asian
Finance Association 2013 Conference. .
Cho, C. H., Guidry, R. P., Hageman, A. M., & Patten, D. M. (2012). Do actions
speak louder than words? An empirical investigation of corporate
environmental reputation. Accounting, Organizations and Society,
37(1), 14–25.
Chung, K.-H., Ghicas, D., & Pastena, V. (1993). Lenders’ use of accounting
information in the oil and gas industry. Accounting Review, 68(4),
885–895.
Click, R., Jeong, Y., & Weiner, R. J. (2013). Does foreign investment carry
baggage from home? Working paper.
14
We thank an anonymous reviewer for bringing this to our attention.
A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47 45
Click, R., & Weiner, R. J. (2010). Resource nationalism meets the market:
Political risk and the value of petroleum reserves. Journal of
International Business Studies, 41, 783–803.
Coval, J. D., & Moskowitz, T. J. (1999). Home bias at home: Local equity
preference in domestic portfolios. The Journal of Finance, 54(6),
2045–2073.
Cuervo-Cazurra, A. (2006). Who cares about corruption? Journal of
International Business Studies, 37(6), 807–822.
Darby, S. (2009). The case for company-by-company reporting of data in the
extractive industries transparency initiative. New York, NY: The
Revenue Watch Institute.
Diamond, D. W., & Verrecchia, R. E. (1991). Disclosure, liquidity, and the
cost of capital. Journal of Finance, 46, 1325–1359.
Doidge, C., Karolyi, G. A., & Stulz, R. M. (2004). Why are foreign ?rms listed
in the US worth more? Journal of Financial Economics, 71(2), 205–238.
Durnev, A., & Fauver, L. (2008). Stealing from thieves: Firm governance and
performance when states are predatory (CEI working paper series 2008–
12). Center for Economic Institutions, Institute of Economic Research,
Hitotsubashi University. Retrieved 24.05.12.
Durnev, A., & Guriev, S. (2007). The resource curse: A corporate transparency
channel (CEFIR working paper w0108). Center for Economic and
Financial Research.
Retrieved 25.05.12.
Durnev, A., & Kim, E. (2005). To steal or not to steal: Firm attributes, legal
environment, and valuation. The Journal of Finance, 60(3), 1461–1493.
Dyck, A., & Zingales, L. (2004). Private bene?ts of control: An international
comparison. The Journal of Finance, 59(2), 537–600.
Dye, R. A. (1998). Investor sophistication and voluntary disclosures.
Review of Accounting Studies, 3(3), 261–287.
Eden, L. (2012). Transfer price manipulation. In P. Reuter (Ed.), Draining
development: Controlling ?ows of illicit funds from developing countries
(pp. 205–234). Washington, DC: The World Bank.
Ellis, J. A., Fee, C. E., & Thomas, S. E. (2012). Proprietary costs and the
disclosure of information about customers. Journal of Accounting
Research, 50(3), 685–727.
Fisman, R., & Miguel, E. (2007). Corruption, norms, and legal enforcement:
Evidence from diplomatic parking tickets. Journal of Political Economy,
115(6), 1020–1048.
Free, C., Salterio, S. E., & Shearer, T. (2009). The construction of
auditability: MBA rankings and assurance in practice. Accounting,
Organizations and Society, 34(1), 119–140.
French, K., & Poterba, J. (1991). Investor diversi?cation and international
equity markets. American Economic Review, 81, 222–226.
Gabbioneta, C., Greenwood, R., Mazzola, P., & Minoja, M. (2013). The
in?uence of the institutional context on corporate illegality.
Accounting, Organizations and Society, 38(6–7), 484–504.
Greene, W. H. (2000). Econometric analysis. Upper Saddle River, NJ:
Prentice Hall.
Greif, A. (2005). Commitment, coercion, and markets: The nature and
dynamics of institutions supporting exchange. In Handbook of new
institutional economics (pp. 727–786). US: Springer.
Guerreiro, M. S., Rodrigues, L. L., & Craig, R. (2012). Voluntary adoption of
international ?nancial reporting standards by large unlisted
companies in Portugal-Institutional logics and strategic responses.
Accounting, Organizations and Society, 37(7), 482–499.
Guriev, S., Kolotilin, A., & Sonin, K. (2011). Determinants of nationalization
in the oil sector: A theory and evidence from panel data. Journal of
Law, Economics, and Organization, 27(2), 301–323.
Hajzler, C. (2012). Expropriation of foreign direct investments: Sectoral
patterns from 1993 to 2006. Review of World Economics, 148(1),
119–149.
Hayward, M. L., & Hambrick, D. C. (1997). Explaining the premiums paid
for large acquisitions: Evidence of CEO hubris. Administrative Science
Quarterly, 103–127.
Healy, P. M., & Palepu, K. G. (2001). Information asymmetry, corporate
disclosure, and the capital markets: A review of the empirical
disclosure literature. Journal of Accounting and Economics, 31(1),
405–440.
Healy, P., & Serafeim, G. (2013). What impedes oil and gas companies’
transparency (HBS working paper 12-038). Harvard Business School.
.
Heaney, R., & Grundy, B. (2011). Valuation of crude oil and gas reserves.
Available at SSRN 1966829.
Henisz, W. J. (2006). Polcon_2005 codebook. Retrieved 17.10.12.
Henisz, W., & Williamson, O. (1999). Comparative economic
organization – Within and between countries. Business and Politics,
1(3), 261–277.
Hines, J. R. (2010). Treasure islands. The Journal of Economic Perspectives,
24(4), 103–125.
Hope, O. K. (2003). Disclosure practices, enforcement of accounting
standards, and analysts’ forecast accuracy: An international study.
Journal of Accounting Research, 41(2), 235–272.
Hope, O. K., Kang, T., Thomas, W. B., & Vasvari, F. (2008). The effects of
SFAS 131 geographic segment disclosures by US multinational
companies on the valuation of foreign earnings. Journal of
International Business Studies, 40(3), 421–443.
Hope, O. K., & Thomas, W. B. (2008). Managerial empire building and ?rm
disclosure. Journal of Accounting Research, 46(3), 591–626.
Hossain, M., Ahmed, K., & Godfrey, J. M. (2005). Investment opportunity
set and voluntary disclosure of prospective information: A
simultaneous equations approach. Journal of Business Finance &
Accounting, 32, 871–907.
Hunt, S. (2011). Re?ning black gold: The Dodd-Frank act and corruption in
the oil industry. UCLA Journal of International Law and Foreign Affairs,
16, 41–163.
Jandhyala, S., & Weiner, R. J. (2014). Institutions sans frontières:
International agreements and foreign investment. Journal of
International Business Studies, 45(6), 649–669.
Jensen, M. (1986). Costs of free cash ?ow, corporate ?nance, and
takeovers. The American Economic Review, 76(2), 323–329.
Jensen, N., & Johnston, N. (2011). Political risk, reputation, and the
resource curse. Comparative Political Studies, 44(6), 662–668.
Jeong, Y., & Weiner, R. J. (2012). Who bribes? Evidence from the United
Nations’ oil-for-food program. Strategic Management Journal, 33(12),
1363–1383.
Joe, J. R., Louis, H., & Robinson, D. (2009). Managers’ and investors’
responses to media exposure of board ineffectiveness. Journal of
Financial and Quantitative Analysis, 44(03), 579–605.
Kang, J. K., & Stulz, R. M. (1997). Why is there a home bias? An analysis of
foreign portfolio equity ownership in Japan. Journal of Financial
Economics, 46, 3–28.
Kanodia, C., & Lee, D. (1998). Investment and disclosure: The disciplinary
role of periodic performance reports. Journal of Accounting Research,
33–55.
Khanna, T., Palepu, K. G., & Srinivasan, S. (2004). Disclosure practices of
foreign companies interacting with US markets. Journal of Accounting
Research, 42(2), 475–508.
Kornai, J. (1986). The soft budget constraint. Kyklos, 39(1), 3–30.
Kostova, T., & Zaheer, S. (1999). Organizational legitimacy under
conditions of complexity: The case of the multinational enterprise.
Academy of Management Review, 24(1), 64–81.
Leuz, C., & Verrecchia, R. (2000). The economic consequences of increased
disclosure. Journal of Accounting Research, 38, 91–124.
Licht, A. (2001). The mother of all path dependencies: Toward a cross-
cultural theory of corporate governance systems. Delaware Journal of
Corporate Law, 26, 147–205.
Licht, A., Goldschmidt, C., & Schwartz, S. (2005). Culture, law, and
corporate governance. International Review of Law and Economics,
25(2), 229–255.
Maksimovic, V., & Phillips, G. (2001). The market for corporate assets:
Who engages in mergers and asset sales and are there ef?ciency
gains? The Journal of Finance, 56(6), 2019–2065.
Maksimovic, V., Phillips, G., & Prabhala, N. R. (2011). Post-merger
restructuring and the boundaries of the ?rm. Journal of Financial
Economics, 102(2), 317–343.
March, J. G., & Olsen, J. P. (1984). The new institutionalism:
Organizational factors in political life. American Political Science
Review, 78, 734–749.
Meek, G. K., & Thomas, W. B. (2004). A review of markets-based
international accounting research. Journal of International Accounting
Research, 3(1), 21–41.
Megginson, W. (2005). The ?nancial economics of privatization. New York,
NY: Oxford University Press.
Miller, M. H., & Upton, C. W. (1985). A test of the hotelling valuation
principle. The Journal of Political Economy, 93(1), 1–25.
Misund, B., Osmundsen, P., & Asche, F. (2005). The value-relevance of
accounting ?gures in the international oil and gas industry – Cash ?ows
or accruals? Working paper 43/05. University of Stavanger. .
Muñoz, J. S. (2009). Financing of oil and gas transactions. Texas Journal of
Oil, Gas, and Energy Law, 4(2), 225–267.
Oil companies in emerging markets: Safe sex in Nigeria. Economist, June
15, 2013. .
46 A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47
Oliver, C. (1991). Strategic responses to institutional processes. Academy
of Management Review, 16, 145–179.
Osmundsen, P. (2010). Chasing reserves: Incentives and ownership. In
Energy, natural resources and environmental economics (pp. 19–38).
Berlin Heidelberg: Springer.
Parsons, T. (1960). Structure and process in modern societies. Glencoe, IL:
Free Press.
Portes, R., & Rey, H. (2005). The determinants of cross-border equity
?ows. Journal of International Economics, 65(2), 269–296.
Rajan, R., & Zingales, L. (1998). Financial development and growth.
American Economic Review, 88, 559–586.
Rose-Ackerman, S. (2003). Corruption. The Encyclopedia of Public Choice,
67–76.
Rosenblum, P., & Maples, S. (2009). Contracts con?dential: Ending secret
deals in the extractive industries. New York: Revenue Watch.
Salter, S. B., Kang, T., Gotti, A. P. G., & Doupnik, T. S. (2013). The role of
social values, accounting values and institutions in determining
accounting conservatism. Management International Review, 53(4),
607–632.
Shi, Y., Magnan, M., & Kim, J. (2012). Do countries matter for voluntary
disclosure? Evidence from cross-listed ?rms in the US. Journal of
International Business Studies, 43, 143–165.
Shleifer, A., & Vishny, R. W. (1993). Corruption. Quarterly Journal of
Economics, 108, 599–617.
Stiglitz, J. E. (2008). 2007 Grotius lecture regulating multinational
corporations: Towards principles of cross-border legal frameworks
in a globalized world balancing rights with responsibilities. American
University International Law Review, 23(3), 451–558.
Stulz, R. (2005). Limits of globalization. Journal of Finance, 60, 1595–1638.
Stulz, R., & Williamson, R. (2003). Culture, openness, and ?nance. Journal
of Financial Economics, 70(3), 313–349.
Suchman, M. (1995). Managing legitimacy: Strategic and institutional
approaches. Academy of Management Review, 20(3), 571–610.
Tabachnick, B. G., & Fidell, L. S. (2001). Computer-Assisted Research Design
and Analysis (Vol. 748). Boston: Allyn and Bacon.
Taylor, G., Richardson, G., Tower, G., & Hancock, P. (2012). The
determinants of reserves disclosure in the extractive industries:
evidence from Australian ?rms. Accounting & Finance, 52(s1),
373–402.
Teall, H. D. (1992). Information content of Canadian oil and gas
companies’ historic cost earnings and reserves disclosures.
Contemporary Accounting Research, 8(2), 561–579.
Thomas, W. B. (1999). A test of the market’s mispricing of domestic and
foreign earnings. Journal of Accounting and Economics, 28(3), 243–267.
Thomas, M. A. (2010). What do the worldwide governance indicators
measure? European Journal of Development Research, 22, 31–54.
Thompson, A. C. (2001). The Hotelling Principle, backwardation of futures
prices and the values of developed petroleum reserves—the
production constraint hypothesis. Resource and Energy Economics,
23(2), 133–156.
Transparency International (2011). Bribe payers index 2011. .
Transparency International (2012). Corruption perceptions index 2012.
.
Vernon, R. (1971). Sovereignty at Bay: The Multinational Spread of U.S.
Enterprises. New York, NY: Basic Books.
Verrecchia, R. (2001). Essays on disclosure. Journal of Accounting and
Economics, 32, 97–180.
Wolf, C., & Weinschrott, D. (1973). International transactions and
regionalism: Distinguishing ‘‘insiders’’ from ‘‘outsiders’’. The
American Economic Review, 63(2), 52–60.
A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47 47
doc_133651385.pdf
This paper analyzes the multinational enterprise’s decision to voluntarily disclose information
regarding its investments, a choice we term investment transparency. When disclosing
investment information, managers must weigh the costs and benefits of reducing asymmetries
between the firm and three stakeholder audiences: capital markets, civil society and
governments. We use a unique transaction-level dataset of reserve acquisitions by oilindustry
multinationals compiled by IHS Herold to examine managerial decisions to reveal
or withhold value-relevant information about firm investment.
Multinational investment and voluntary disclosure:
Project-level evidence from the petroleum industry
Anthony P. Cannizzaro
?
, Robert J. Weiner
1
George Washington University, Department of International Business, 2201 G Street NW, Suite 401, Washington, DC 20052, United States
a r t i c l e i n f o
Article history:
Available online 19 February 2015
a b s t r a c t
This paper analyzes the multinational enterprise’s decision to voluntarily disclose informa-
tion regarding its investments, a choice we term investment transparency. When disclosing
investment information, managers must weigh the costs and bene?ts of reducing asymme-
tries between the ?rm and three stakeholder audiences: capital markets, civil society and
governments. We use a unique transaction-level dataset of reserve acquisitions by oil-
industry multinationals compiled by IHS Herold to examine managerial decisions to reveal
or withhold value-relevant information about ?rm investment. Contrary to the agency-the-
oretic motivations traditionally ascribed to voluntary disclosure, our results suggest insti-
tutional and informational factors drive investment transparency. We ?nd that ?rms
disclose less in cross-border transactions, more when societal expectations of transparency
are high, and less when faced with political risk. These results should be of interest to
scholars of accounting and international business, as well as managers and policy makers
involved in the ongoing debate on transparency in the extractive industries.
Ó 2015 Elsevier Ltd. All rights reserved.
Introduction
How transparent are multinational enterprises (MNEs)
regarding their investments? In this study, we use the glo-
bal market for petroleum reserves as a laboratory to exam-
ine investment transparency – value-relevant information
MNEs choose to disclose voluntarily about investment pro-
jects. For a given investment, ?rms may disclose no infor-
mation, partial information, or full information about the
value of the investment.
We use a unique transaction-level dataset compiled by
IHS Herold, which allows us to identify which party dis-
closes each transaction and how much information is
revealed about the investment. We ?nd that ?rms disclose
less about cross-border than domestic investment. This
result is robust to controls for the ?rm’s capital needs,
national institutions, ownership of the ?rm and character-
istics of the investment. Further, we ?nd that ?rms invest-
ing in countries with strong transparency norms (proxied
by government ?scal openness, freedom of the press, and
quality of the accounting system), and strong political con-
straints are more likely to disclose partial information.
Firms from countries marked by less political risk and cor-
ruption are more likely to disclose full information.
We draw on theories of voluntary disclosure and the
institutional and political economy literatures to suggest
that MNEs use voluntary disclosure strategically to man-
age information asymmetries between the ?rm and three
primary stakeholder groups: capital markets, civil society,
and governments. Our results do not support traditional
agency-theoretic motivations such as increasing disclosure
to secure external ?nancial resources, and increased dis-
closure in multinational operations. This runs counter to
the view in the literature that MNEs disclose more in
response to capital market demands for information about
their operations abroad (Cahan, Rahman, & Perera, 2005).http://dx.doi.org/10.1016/j.aos.2015.01.002
0361-3682/Ó 2015 Elsevier Ltd. All rights reserved.
?
Corresponding author. Tel.: +1 202 994 6880; fax: +1 202 994 7422.
E-mail addresses: [email protected] (A.P. Cannizzaro), rweiner@gwu.
edu (R.J. Weiner).
1
Tel.: +1 202 994 5981; fax: +1 202 994 7422.
Accounting, Organizations and Society 42 (2015) 32–47
Contents lists available at ScienceDirect
Accounting, Organizations and Society
j our nal homepage: www. el sevi er. com/ l ocat e/ aos
Rather, institutional factors such as societal expecta-
tions of transparency and political risk play an important
role in multinational disclosure. Our results are consistent
with the view that voluntary disclosure is driven by social
norms (Cho, Guidry, Hageman, & Patten, 2012), as MNEs
seek to be perceived as legitimate by key stakeholders
(Parsons, 1960). Further, our ?ndings are consistent with
MNEs’ strategically reducing transparency to protect
investments from government predation.
Multinational enterprises typically operate numerous
investment projects in a variety of countries and institu-
tional environments. Despite empirical evidence that
exposure to international markets increases the complex-
ity of assessing and transmitting value-relevant informa-
tion (Callen, Hope, & Segal, 2005; Hope, Kang, Thomas, &
Vasvari, 2008; Portes & Rey, 2005; Thomas, 1999), the lit-
erature has only recently begun to explore how countries
matter for corporate transparency (Healy & Serafeim,
2013; Shi, Magnan, & Kim, 2012).
We extend this literature beyond ?nancial-statement
reporting by examining value-relevant disclosures speci?c
to ?rminvestments. The importance of investments to ?rm
strategy is highlighted in the scholarly literatures in
?nance (Maksimovic & Phillips, 2001; Maksimovic,
Phillips, & Prabhala, 2011) and strategic management
(Capron, Mitchell, & Swaminathan, 2001), yet is underrep-
resented in the literature on voluntary disclosure.
More broadly, society has long been concerned over
accountability and corruption in foreign investment.
Widely-held suspicions of MNEs are based in part on per-
ceptions that their secrecy masks illicit behavior. Stiglitz
(2008) argues that MNEs are more likely than domestic
?rms to exploit asymmetries in bargaining power and
information, and use cross-border transactions to avoid
accountability. Transfer pricing, the ability to manipulate
internal prices to shift pro?ts between subsidiaries in dif-
ferent tax jurisdictions, is an advantage of multinationality
(Eden, 2012). Meek and Thomas (2004) note opacity of for-
eign operations as an ongoing issue in international disclo-
sure research.
Our study is relevant to debates over transparency ini-
tiatives by governments, intergovernmental organizations
(IGOs), and civil society (through non-governmental orga-
nizations – NGOs). This debate is particularly contentious
in the extractive industries, of which petroleum is the larg-
est. Survey evidence suggests that petroleum and mining
are among the industries most prone to bribery
(Transparency International, 2011). Darby (2009) notes
that failure by MNEs in the extractive industries to disclose
information is often interpreted as these ?rms having
something to hide. A majority of the 20 worst-performing
countries on the Transparency International (2012) corrup-
tion index are natural-resource rich. Both IGOs and NGOs
2
have been expanded or created to push increased transpar-
ency, discourage corruption, and address other ills of
resource-rich societies (Durnev & Guriev, 2007; Jensen &
Johnston, 2011).
Recent years have seen policy initiatives and disclosure
rule changes designed to enhance transparency. In the Uni-
ted States, Section 1504 of the Dodd-Frank Wall Street
Reform and Consumer Protection Act (‘‘Dodd-Frank,’’
passed in 2010), amends the Securities Exchange Act of
1934 to require disclosure of government payments by
extractive industry ?rms listed on US exchanges. In 2012,
the Securities and Exchange Commission adopted rule
13(q)-1, which requires US-listed ?rms in extractive indus-
tries to include project-level disclosures of payments to
governments at home and abroad in their annual reports,
starting in late 2013.
3
However, following industry chal-
lenges the rule was vacated.
4
The European Union, Hong
Kong, and Canada have enacted or have committed to enact
similar regulations on listed extractive industry MNEs.
5
Project-by-project disclosure requirements, intended to
create accountability and reduce corruption, have proven
extremely controversial (Hunt, 2011). Strong opposition
from listed petroleum MNEs, such as the effort that chal-
lenged the 2013 Dodd-Frank rule, are grounded in claims
that project-level reporting will be detrimental to ?rms
by disclosing private information to governments and
competitors, many of which are state-owned.
6
Numerous
state-owned ?rms are unlisted, and hence not bound by
these rules. In contrast, NGOs assert that project-level dis-
closure will not have signi?cant consequences for competi-
tiveness (Rosenblum & Maples, 2009).
The debate over mandatory project-level disclosure
raises the question of the extent to which such data are
now reported voluntarily. Thus, we examine investment-
level disclosure patterns and the institutional factors driv-
ing them. If investment project disclosures are costly to
?rms, these costs should in?uence managerial decisions
to reveal information. Relating disclosure decisions to
MNEs’ investment locations, our approach integrates exist-
ing insights from the accounting literature on corporate
transparency with work from institutional theory and
political economy to model the decision to disclose. Our
work complements recent research on corporate-level
disclosure of performance and payments to governments
in publicly-traded petroleum MNEs (Healy & Serafeim,
2013).
2
The primary IGO example is the Extractive Industries Transparency
Initiative (EITI), an international collaboration between governments,
businesses, and civil society groups that promotes disclosure of aggregate
?rm payments. NGO examples include Oxfam International (an NGO that
promotes poverty alleviation worldwide), and Publish What you Pay, a
global network of NGOs (including Transparency International and Global
Witness) devoted to promoting transparency in the oil industry.
3
The language of the law is broadly interpreted as requiring issuers to
disclose granular, disaggregated information on a project-by-project basis.
Hunt (2011) provides a comprehensive review of the legislation.
4
The rule was vacated in July 2013 by the US District Court for the
District of Columbia (see Memorandum Opinion ?led July 2, 2013 for Civil
Action Number 12-1668 (JDB)).
5
For example, new EU Transparency and Accounting Directives require
country-by-country and project-by-project disclosure of all government
payments over €100,000 (European Commission MEMO/13/541 dated June
12, 2013).
6
SEC Release No. 34-67717; File No. S7-42-10. Final Rule Making on
Disclosure of Payments by Resource Extraction Issuers, 17 CFR Parts 240 & 249,
summarizes inter alia industry views.
A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47 33
Background and empirical setting
A rich literature examines corporate disclosure deci-
sions. This literature (surveyed in Beyer, Cohen, Lys, &
Walther, 2010) focuses primarily on corporate-level
reporting, and is silent about the motivations for invest-
ment-level disclosure. Exceptions include Kanodia and
Lee (1998) who study the effect of disclosure on invest-
ment, and Chen, Tan, Cheng, and Gong (2013) who exam-
ine domestic investment projects by listed ?rms in China,
where reporting of new projects is mandatory but disclo-
sure of details is not.
Consolidated corporate-level ?nancial statements
re?ect outcomes of multiple managerial decisions, making
empirical identi?cation at the investment-level problem-
atic. In contrast, we focus on individual investments
through asset acquisitions, both domestic and abroad.
The assets we examine are petroleum reserves (the quan-
tity of oil and gas in the ground that is extractable econom-
ically at current prices and costs). The importance of
reserves combined with the ubiquitous state role in the
industry often links them to corruption, and non-transpar-
ent or non-competitive sales conditions. As the Economist
notes, ‘‘Deals for oil?elds can be as opaque as the stuff that
is pumped from them’’ (Economist, 2013).
Reserves provide several advantages for research on
disclosure. First, they are a critical asset to ?rms in the
upstream segment of the industry (exploration and pro-
duction); as the ?rm’s inventory of future production
(and hence a predictor of future earnings), they are central
to valuation and borrowing capacity (Arnott, 2004; Chung,
Ghicas, & Pastena, 1993; Muñoz, 2009; Osmundsen, 2010).
Research has demonstrated that reserves are value-rele-
vant (Misund, Osmundsen, & Asche, 2005; Taylor,
Richardson, Tower, & Hancock, 2012), and that reserve
restatements (analogous to asset restatements in manufac-
turing ?rms) are associated with abnormal stock returns
(Berry & Wright, 2001).
7
Second, reserves are suf?ciently homogenous to be
comparable across geographic boundaries, ?rm bound-
aries, and time. As inventory to be produced in the future,
reserves are similar across ?rms, and market comparisons
are more easily assessed than for assets such as plant and
equipment or organizational divisions. Reserves are largely
homogeneous except for cost differences (for which we
control in our analysis); intangibles such as brand name
or goodwill do not play a role. This homogeneity facilitates
outsiders’ investment valuation if reserve size is disclosed.
Third, research that models determinants of voluntary
disclosure identi?es heterogeneity of private information
as a potentially confounding factor (Chen et al., 2013;
Ellis, Fee, & Thomas, 2012). Some ?rms may be less trans-
parent than others for the simple reason that they have lit-
tle private information or that their private information is
of little value. This is dif?cult to control for in empirical
research. By examining a large number of similar projects,
we avoid this problem.
Finally, acquisitions of reserves, like acquisitions of
?rms, are major management decisions, undertaken infre-
quently and followed by analysts and the trade press. It
thus ?ts well with the agency motivation for disclosure
(discussed below) prevalent in the literature. Purchasing
assets is consistent with both value creation (if the ?rm
can get more out of these assets than it paid for them),
and value destruction though hubris (here overpaying for
assets), empire-building (purchasing assets to increase
?rm size), or both (Hayward & Hambrick, 1997; Hope &
Thomas, 2008).
Unlike most real assets, petroleum reserves are actively
traded in a decentralized global market with many players
(although each ?rm trades infrequently, as shown below).
8
The large number of transactions facilitates empirical
inquiry. The advantage of our database of reserve-transac-
tion announcements is that each transaction effectively cor-
responds to a managerial decision on whether or not to
reveal information for a speci?c investment project. We
focus on disclosure of the two central attributes of each
investment – the size of the reserve acquired and the price
paid. For each transaction, these details are either provided
or redacted.
Theoretical development and hypotheses
We de?ne investment transparency as discretionary
disclosure of value-relevant information at the invest-
ment-level (Verrecchia, 2001). Managers are neither
required to nor prohibited from disclosing information on
reserve investments.
9
Failure to reveal either the price paid
in a transaction or the quantity of reserves purchased makes
a price-per-barrel market comparison impossible, obscuring
valuation. Such valuations may reveal information regarding
the MNE’s exploration success and extraction costs, affecting
negotiations with host governments and the potential for
corrupt activities.
Payoffs to disclosure re?ect information asymmetries
between management and three distinct stakeholder audi-
ences: capital markets, civil society, and governments.
First, capital markets value transparency as a governance
mechanism. Disclosure can help align manager and share-
holder interests, thereby lowering the MNE’s cost of capi-
tal. Second, disclosure of investment details may help
grant the ?rm legitimacy in the eyes of society. Lastly, dis-
closure may magnify political risk and corruption faced by
the ?rm.
7
Because reserves are central to the industry, their reporting has its own
terms and accounting rules, with standards set out by inter alia, the US
Securities and Exchange Commission, Canadian Securities Administrators,
State Commission for Reserves of the Russian Federation, Norwegian
Petroleum Directorate, the Committee for Mineral Reserves International
Reporting Standards, and the Society of Petroleum Engineers. Auditing by
independent specialists with expertise in engineering and geology is
common practice.
8
Reserves are treated as assets in this paper, as well as in the industry
and in ?nancial reporting. Ownership refers to control and cash ?ow rights
from reserves, which are technically property of either private landowners
or governments that receive royalties from the ?rms operating them.
9
US-listed ?rms must ?le 8-K reports for material transactions, but have
discretion over what is material and what to report. Rosenblum and Maples
(2009) ?nd that while many host countries prohibit ?rms from disclosing
details of petroleum contracts, none prevents disclosure of asset payment
or size.
34 A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47
Capital markets
Twostrands of the accountingliterature suggest con?ict-
ing hypotheses about the relationship between capital mar-
kets and investment transparency. The global-diversi?cation
strand of the literature suggests complexity of international
operations raises agency costs, increasing managers’ incen-
tive to disclose. Cahan et al. (2005) argue that international
operations increase ?rm complexity. This complexity gives
international managers information advantages over inves-
tors by virtue of their local knowledge, increasing the scope
for non-value-maximizing behavior. More disclosure
reduces this scope, thus enhancing ?rm value. Hope and
Thomas (2008) ?nd evidence of such agency con?icts in
MNE operations abroad by demonstrating that ?rms that
fail to disclose geographic earnings experience greater for-
eign sales growth, but reductions in foreign pro?ts, consis-
tent with managerial empire building. However, these
studies use foreign sales or number of foreign subsidiaries
to proxy international complexity, and are unable to
directly compare foreign and domestic investments.
The investor-sophistication strand yields the opposite
prediction. Managers respond strategically to the sophisti-
cation of the ?rm’s investors, because uninformed investors
alter the bene?ts of disclosing information. Dye (1998) the-
orizes that the primary bene?t to the ?rm of withholding
information arises from an uninformed investor’s inability
to recognize whether nondisclosure is explained by having
little to disclose, or by hiding information. Dye’s formal
model demonstrates that as the probability investors are
uninformed increases, ?rm incentives to disclose decrease.
In the international context, ?nance research ?nds that
equity investors are typically less informed about foreign
assets. Home bias is the widely documented tendency for
investors to overweight domestic companies in their port-
folios (French & Poterba, 1991). Shareholders ?nd it more
dif?cult to assess the value and risks of investments abroad
than at home because information ?ows are not frictionless
across borders (Ahearne, Griever, &Warnock, 2004; Coval &
Moskowitz, 1999; Kang & Stulz, 1997). This home bias
extends beyond securities to ?rmoperations abroad. Exam-
ining foreign earnings, Thomas (1999) ?nds that investors
place a lower value on foreign information because it is
more dif?cult to assess, resulting in a bias towards domes-
tic investment. The effect on corporate transparency is that
cross-border frictions reduce the bene?t to managers of
disclosing foreign acquisitions relative to domestic. Thus,
?rms should disclose less in cross-border transactions.
The two strands of the literature thus yield opposite
predictions regarding disclosure and multinational invest-
ment. Ultimately, which effect dominates is an empirical
question. Accordingly, we test the following hypotheses:
H1A (Global Diversi?cation Hypothesis). Firms will volun-
tarily disclose more information about cross-border invest-
ments than domestic investments.
H1A
0
(Investor Sophistication Hypothesis). Firms will volun-
tarily disclose less information about cross-border invest-
ments than domestic investments.
Agency theory also suggests that investment transpar-
ency may be more important for ?rms in need of additional
capital. New investment projects in general give managers
an information advantage. For oil industry MNEs, the
unobservable nature of underground reserves creates a
great deal of uncertainty around the production potential
of a new project. Firms necessarily invest time, technology
and human capital in making proprietary estimates of this
potential. The resulting reduction in managerial uncer-
tainty creates the potential for con?ict between insiders
and investors. For example, Bertrand and Mullainathan
(2000) ?nd that oil industry executive compensation is
unrelated to performance when monitoring is poor.
These information asymmetries and agency con?icts
increase the ?rm’s cost of capital as investors consider
these costs and adjust their expectations accordingly
(Diamond & Verrecchia, 1991). Investment disclosures
may provide MNEs in need of external ?nancing a solution
to the agency problem by reducing information asymme-
tries, lowering monitoring costs and signaling to investors
that managerial interests are aligned with their own (Leuz
& Verrecchia, 2000). Thus, we expect ?rms in need of addi-
tional external ?nancing should be more likely to reveal
information regarding reserve transactions.
H1B. Firms with greater reliance on external ?nancing will
voluntarily disclose more information about investments.
Civil society and corporate legitimacy
Researchers have long acknowledged that society and
societal values have a role to play in understanding ?rm
governance issues (Licht, 2001; Licht, Goldschmidt, &
Schwartz, 2005; Stulz & Williamson, 2003; Wolf &
Weinschrott, 1973). Studies focusing on ?rm accounting
highlight the importance of societal values for annual
report disclosures (Hope, 2003) and cross-country differ-
ences in accounting conservatism (Salter, Kang, Gotti, &
Doupnik, 2013).
Institutional theory suggests disclosure may be prefera-
ble if conforming to societal norms of transparency grants
the ?rm legitimacy; a perception that the ?rm’s actions are
desirable in the eyes of key stakeholders (Parsons, 1960;
Suchman, 1995). MNE managers respond strategically to
societal pressures, as legitimacy is often required to oper-
ate successfully abroad (Kostova & Zaheer, 1999; Oliver,
1991). Increasing voluntary disclosures when normative
expectations of transparency are high increases the likeli-
hood that ?rms are perceived as legitimate. Cho et al.
(2012) examine voluntary environmental disclosures and
?nd that ?rms’ reputations improve with greater disclo-
sure, even when actual environmental performance is
poor. Additional evidence in the accounting literature sug-
gests that legitimacy can be obtained via auditing (Free,
Salterio, & Shearer, 2009), and by adopting more rigorous
accounting standards (Guerreiro, Rodrigues, & Craig, 2012).
Transparency may increase MNE legitimacy even when
voluntary disclosures provide only limited information, as
the act of disclosure itself may help legitimize the ?rm
(March & Olsen, 1984). For example, Khanna, Palepu, and
A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47 35
Srinivasan (2004) show that foreign ?rms investing in the
United States tend to adopt US disclosure practices. In
addition, conforming to societal expectations, even
if only symbolically, may alleviate regulatory scrutiny
(Gabbioneta, Greenwood, Mazzola, & Minoja, 2013).
Thus, if legitimacy-seeking is the motivation, partial disclo-
sure may be preferable to either no disclosure or full
disclosure.
As legitimacy is required in all institutional environ-
ments in which the ?rm operates (Kostova & Zaheer,
1999), normative pressures may originate from either the
MNE’s country of origin (home country), or the country
in which an investment resides (the host country). Thus,
we argue that home and host countries marked by societal
expectations of transparency should exert more pressure
on MNEs to reveal information about investments.
H2A. Firms acquiring assets in countries with higher
normative expectations of transparency will voluntarily
disclose more information about investments.
H2B. Firms coming from countries with higher normative
expectations of transparency will voluntarily disclose more
information about investments.
Political risk
Greif (2005, p. 728) notes that institutions that ‘‘reveal
wealth’’ are only optimal when there are constraints that
curb the coercive power of the state. Unless constrained,
government of?cials have the discretion to expropriate
value for the state (Stulz, 2005) or misuse their of?ce to
extract bribes (Rose-Ackerman, 2003). Thus, MNEs operat-
ing in environments with weak institutional constraints on
government and its agents are incentivized to be less
transparent; the same information asymmetries capital
markets and civil society seek to curb may shield the
MNE from these political risks.
After initial costs of a foreign investment are sunk, the
MNE faces the risk that the host government may alter
the terms of a contract ex post, or violate agreements
entirely through expropriation of ?rm assets. This results
in the challenge ?rst identi?ed by Vernon (1971) as the
obsolescing bargain, a time-inconsistency problem stem-
ming from the government’s discretion to change the
terms of any agreement (Henisz & Williamson, 1999).
Firms typically have little recourse for mitigating this form
of risk once capital is invested in the host government’s
jurisdiction.
Political risks are especially salient in the petroleum
industry where rents are a high proportion of product
value. In addition, reserves are considered national patri-
mony, typically under state control, and are often located
in countries marked by weak institutions including poor
property-rights protection and unstable or autocratic gov-
ernments (‘‘resource curse’’). Extractive industries, particu-
larly petroleum, are at greatest risk for expropriation
(Guriev, Kolotilin, & Sonin, 2011; Hajzler, 2012).
Government of?cials may also attempt to extract rents
for themselves, creating costs of disclosure for both ?rms
that behave ethically and those that are complicit in cor-
rupt practices. The former may be disadvantaged in com-
peting for contracts if state of?cials favor opaque ?rms to
discourage corrupt practices from being revealed (Shleifer
& Vishny, 1993). For the latter, transparency exacerbates
the risk of detection (Healy & Serafeim, 2013).
Such risk arises from both home- as well as host-coun-
try governments (Cuervo-Cazurra, 2006; Fisman & Miguel,
2007). For example, the large role of opaque offshore ?nan-
cial centers in international capital ?ows suggests MNE
managers recognize political risks from home govern-
ments (Hines, 2010). Empirical evidence demonstrates that
home-country political risk reduces petroleum reserve val-
ues (Click, Jeong, & Weiner, 2013). Information asymmetry
between the ?rm and host- and home-country govern-
ments increases uncertainty about asset value, reducing
the attractiveness of predation, and hence expropriation
risk. Thus, we expect that ?rms coming from or operating
investments in countries with weak political institutions
will disclose less.
H3A. Firms acquiring assets in countries with greater
political risk will voluntarily disclose less information
about investments.
H3B. Firms coming from countries with greater political
risk will voluntarily disclose less information about invest-
ments.
Agency costs and political risks may not operate inde-
pendently. MNEs face a twin agency problem of managerial
and state discretion – the simultaneous threat of diversion
by ?rm insiders and expropriation by predatory govern-
ments (Stulz, 2005). When the latter poses a signi?cant
risk to the ?rm, the transparency decisions the ?rm would
otherwise adopt to compensate for the former may no
longer be value-enhancing.
Durnev and Fauver (2008) formalize Stulz’s twin agency
problem, showing that the agency costs of insider diver-
sion and the political costs of government expropriation
are complements. Extending the twin agency problem to
investment transparency, we theorize that MNEs subject
to higher agency costs are also more susceptible to political
costs. Ceteris paribus, these ?rms are less likely to disclose
than would be predicted by agency costs or political costs
alone.
H3C. The negative effect of political risk on investment
transparency will be stronger for ?rms with less reliance
on external ?nance.
Methodology
Sample
As noted above, our sample of oil and natural gas
reserve transactions comes from a database of public
36 A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47
announcements compiled by IHS Herold, an advisory ?rm
that conducts independent research and collects ?nancial
and operational information on the global petroleum
industry.
10
Widely-used investment databases (e.g., SDC
Platinum and ORBIS) do not indicate which counterparty
disclosed (or did not disclose) information, presenting an
identi?cation challenge in project-level analysis. In contrast,
the Herold database reports the information source for each
investment, enabling us to relate disclosure to the ?rm mak-
ing the disclosure decision.
Reserve transactions are voluntarily disclosed through
releases in the trade and business press. For each deal,
we obtain the announcement date, counterparty names
and home countries, reserve location, reserve type, trans-
action value (if reported), and estimated reserve size (if
reported). This data is then matched with detailed ?rm
information from Thompson-Reuters’ SDC Platinum data-
base. Information regarding price and quantity transacted
are necessary for the market to fully evaluate a reserve
transaction, yet ?rms frequently make only one piece of
information or neither available.
In total, our sample covering the years 2000 through
2011 is comprised of 1210 buyer-announced domestic
and cross-border reserve transactions by 702 ?rms across
64 countries.
11
We include only transactions that have been
reported by acquirers, as sellers’ investments are no longer
affected by host-country institutions once they are
divested. We exclude transactions with multiple buyers,
those where buyer assets are less than $1 million, and failed
transactions.
Measuring investment transparency
The literature uses several proxies for transparency. The
most common is earnings management, which allows
managers to obscure information from shareholders. Earn-
ings re?ect ?rm-wide outcomes, however, rather than spe-
ci?c investments. Closer to our interest in project-level
transparency is the inclusion of earnings forecasts in cor-
porate-level reports, as these forecasts may be tied to spe-
ci?c ?rm investments (Shi et al., 2012). However, these
measures are limited to publicly-traded ?rms, (a subsam-
ple of the population), and may re?ect cross-country vari-
ation in reporting rules.
To examine investment transparency, we code each
transaction by whether the announcement discloses min-
imal information (only the existence of a transaction is
revealed), partial information (exclusively price paid or
reserve size is revealed), or full information (both
price and size are revealed). Full information facilitates
comparison with market data, while partial information
permits limited insight into the ?rm’s investment
decision.
Independent variables
Need for external ?nancing is an established proxy for
the potential for agency con?icts between management
and capital markets (Jensen, 1986; Rajan & Zingales,
1998) as ?rms in need of ?nancing are incentivized to bind
themselves to better governance and transparency to
ensure that outside capital remains available. Following
Rajan and Zingales (1998), we calculate dependence on
external ?nance as the quantity capital expenditures less
cash ?ow, divided by capital expenditures. These data are
obtained through the SDC Platinum database. We follow
the literature in Winsorising at the 5% level to account
for extreme observations and potentially spurious outliers
(Doidge, Karolyi, & Stulz, 2004; Durnev & Kim, 2005).
We look to press freedom, government budget trans-
parency and accounting system quality to proxy for socie-
tal pressures to disclose. The press functions as a conduit
for societal isomorphic pressures. Openness in the press
serves both as a model of societal expectations of transpar-
ency, and a transmission mechanism to communicate
stakeholder expectations to managers (Bednar, 2012). Joe,
Louis, and Robinson (2009) show that media coverage of
board ineffectiveness forces ?rms to improve corporate
governance practices. Similarly, Dyck and Zingales (2004)
demonstrate that public opinion pressures (proxied by
newspaper circulation) signi?cantly reduce private bene-
?ts of control for ?rm insiders. We use the Press Freedom
Index compiled by Reporters Without Borders (Press Free-
dom); the index covers 179 countries from the years
2002 to 2012. This index is a compilation of 44 complimen-
tary indicators that together constitute a measure of the
state of media freedom within a country. For consistency
across measures (higher values indicate stronger institu-
tions), we reverse-code the Press Freedom Index.
Budget transparency re?ects information available to
civil society regarding ?scal processes, which have a repu-
tation for opacity in many oil producing countries. We
measure government openness with the International Bud-
get Partnership’s Open Budget Index (OBI), which assesses
94 countries based on government transparency and
accountability in the budgeting process.
To measure the importance of accounting quality across
countries, we use an index of audit and enforcement stan-
dards developed in Brown, Preiato, and Tarca (2013). Data
for this measure are drawn from a multi-wave survey con-
ducted by the International Federation of Accountants. The
index covers 51 countries for the years 2002, 2005, and
2008.
To measure political risk, we employ the widely-used
International Country Risk Guide (ICRG) political risk rating
compiled by the PRS Group. The ICRG rating describes the
risk posed by governments across an array of 12 political
risk components (government stability, socioeconomic
conditions, investment pro?le, internal con?ict, external
con?ict, prevalence of corruption, the degree to which
the military intercedes in politics, religious tensions, ethnic
tensions, law and order, democratic accountability, and
bureaucracy quality).
We also consider an alternative measure of political risk
that focuses on the constraints placed on governments. The
10
Herold databases have been used in research in accounting (Misund
et al., 2005; Teall, 1992), economics (Thompson, 2001), and ?nance
(Heaney & Grundy, 2011; Miller & Upton, 1985). Our database has been
used in strategy research (Click & Weiner, 2010; Jandhyala & Weiner, 2014).
11
The small number of deals per ?rm re?ects the major, infrequent
nature of reserve acquisitions. Our results below do not depend on a few
?rms; the top 10 by deal frequency account for less than 25% of the sample.
A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47 37
Political Constraints 3 index (POLCON) is designed to cap-
ture the political structures within a country that support
a government’s ability to make credible regulatory com-
mitments. Governments facing weaker constraints are
freer to pursue predatory actions, and thus pose a greater
risk to ?rms that disclose investment information
(Henisz, 2006). Higher values of the ICRG and POLCON indi-
ces indicate less political risk and greater constraints,
respectively.
Transparency International’s Corruption Perceptions
Index (CPI) is the corruption measure in widest use.
Intended to capture both the administrative and political
aspects of corruption, the index measures perceptions of
multiple forms of graft. These include bribery of public of?-
cials, kickbacks in public procurement, and embezzlement
of public funds (Transparency International, 2012). While
measures based on expert perceptions have a number of
shortcomings (Thomas, 2010), this measure is appropriate
in the examination of managerial decision-making. Shared
perceptions of corruption are likely to in?uence the cogni-
tive processes employed by decision makers. Greater index
values indicate lower perceptions of corruption.
We also use Global Integrity’s anti-corruption index
(Global Integrity) as a measure of cross-national differences
in the costs of corruption faced by ?rms. This index mea-
sures the accessibility and effectiveness of anti-corruption
mechanisms across countries rather than corruption
directly. The index score is compiled from 300 integrity
indicators gathered from experts and professionals in each
country. Greater scores indicate more constraints on
corruption.
Controls
We control for factors that may shape the incentives of
multinationals to reveal information. First is the price of
oil. High oil prices are associated with greater oil-company
pro?tability and better upstream investment opportuni-
Table 1
Variable de?nitions.
Variables Description Data source
Investment Transparency Categorical: minimal disclosure, partial disclosure, full
disclosure
Herold Database
Assets Log value of buyer’s assets SDC Platinum
Need for External Finance (CapEx – Cash Flow)/CapEx (Rajan & Zingales, 1998) SDC Platinum
Buyer SOE dummy Equals 1 if buyer is a state-owned enterprise SDC Platinum
Company Websites
Petroleum Intelligence Weekly, 2013
Buyer Listed dummy Equals 1 if buyer is listed on a stock exchange Herold Database
SDC Platinum
Company Websites
Petroleum Intelligence Weekly, 2013
Oil Price Log value of 12 month strip price of oil Herold Database
Conventional Reserve
Dummy
Equals 1 if conventional reserve type
b
Herold Database
Cross-border dummy Equals 1 if cross-border transaction Herold Database
Press Freedom Index (PFI) Original scale: 0–115; standardized to l = 0, r = 1 Reporters without borders
Lower values indicated greater press freedom (reverse coded for
analysis)
Data available at en.rsf.com
a
Open Budget Index (OBI) Original scale: 0–100; standardized to l = 0, r = 1 International budget partnership
Higher values indicated more government openness Data available at internationalbudget.org
a
Accounting Quality Index Original scale: 0–56; standardized to l = 0, r = 1 Brown, Preiato, and Tarca (2013)
Higher values indicate greater accounting quality
ICRG Rating Original scale: 0–100; standardized to l = 0, r = 1 The PRS Group
Higher values indicate less political risk
Political Constraints (POLCON
3)
Original scale: 0–1; standardized to l = 0, r = 1 Prof. Witold Henisz, Wharton School
Higher values indicate less political risk Data available at www-
management.wharton.upenn.edu/henisz/
Corruption Perceptions Index
(CPI)
Original scale: 0–10; standardized to l = 0, r = 1 Transparency International
Higher values indicate less corruption Data available at cpi.transparency.org
Global Integrity Index Original scale: 0–100; Standardized to l = 0, r = 1 Global Integrity, 2013
Higher values indicate less corruption Data available at Globalintegrity.org
a
a
Data also available through the World Bank Actionable Governance Indicators (AGI) data portal: agidata.org.
b
Reserves extractable with basic technology (includes royalty interests). High-tech reserve types include bitumen, coalbed, deepwater, enhanced,
frontier, heavy oil, LNG, shale, nonconventional, shallow water, tight gas, and diversi?ed.
38 A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47
ties, suggesting more disclosure (Hossain, Ahmed, &
Godfrey, 2005). At the same time, high oil prices generate
greater rent, which promotes corruption (Jeong & Weiner,
2012) and may reduce incentives for disclosure. Thus, the
predicted sign of the effect is thus unclear. Information
on petroleum prices is from the New York Mercantile
Exchange (NYMEX).
We control for physical characteristics of the reserve.
Deep water, heavy oil and shale reserves, for example,
require both greater capital and technological know-how
to extract than conventional oil reserves. We include a
dummy that indicates whether a reserve is conventional
or unconventional (‘‘high-tech’’), as there is potentially less
private information to reveal for conventional assets.
Reserve characteristics are available through the Herold
database.
Next, we control for characteristics of the ?rm. In
regressions where we test need for external ?nance, we
account for ?rm size by taking the log value of ?rm assets
as reported by SDC Platinum. We also control for the listing
status of the ?rm, as private ?rms tend to be closely held,
and thus have less need of communicating information to a
broad group of investors. Information on listing is obtained
from the Herold database and SDC Platinum. Where histor-
ical listings are missing or inaccurate in these sources,
information is hand-collected from company websites,
stock exchanges, and industry publications.
Last, we include a dummy for state-ownership of the
buyer. While existing theory on voluntary disclosure has
little to say on the differences in transparency between pri-
vate corporations and state-owned ?rms, managers in
state-owned ?rms likely face a different set of incentives.
State-owned enterprises face ‘‘soft’’ budget constraints
(Kornai, 1986), as governments step into ?ll budget short-
falls. While a state-owned enterprise may require addi-
tional external ?nancing to fund investment projects, it is
not subject to the discipline of the capital market
(Megginson, 2005), and thus does not face the same incen-
tive to reveal information as publicly traded ?rms.
A summary of variable de?nitions and data sources is
included in Table 1.
Results
Descriptive statistics
Table 2 reports descriptive statistics for our sample. Of
the 1210 transactions in the database, both price paid
and reserves purchased are revealed (‘‘full transparency’’
below) in 657 transactions (54% of the sample); only one
piece of information is revealed (‘‘partial transparency’’)
in 365 transactions (30%); and neither pricing nor quantity
of reserves is disclosed (‘‘minimal transparency’’) in 188
transactions (16%).
In addition, 372 transactions are cross-border (31% of
the sample). The cross-border and domestic sub-samples
are strikingly different; e.g., only 35% of cross-border trans-
actions are fully transparent vs. 63% of domestic transac-
tions. Table 3 presents differences in disclosure between
cross-border and domestic investments for ?rms that
undertake both. These ?rms are far less likely to fully dis-
close details about their foreign investments than domes-
Table 2
Descriptive statistics.
Variable All transactions Cross-border transactions
Mean SD Min Max Obs Mean SD Min Max Obs
Investment Transparency
Full disclosure 0.54 0.50 0.00 1.00 1210 0.35 0.48 0.00 1.00 372
Partial disclosure 0.30 0.46 0.00 1.00 1210 0.41 0.49 0.00 1.00 372
Minimal disclosure 0.16 0.36 0.00 1.00 1210 0.23 0.42 0.00 1.00 372
Assets (Log) 6.60 2.63 0.34 12.59 742 7.54 3.12 0.69 12.59 179
Need for External Finance À0.16 1.16 À3.09 1.75 688 À0.36 1.34 À3.09 1.75 162
Buyer SOE dummy 0.08 0.27 0.00 1.00 1210 0.16 0.37 0.00 1.00 372
Buyer Listed dummy 0.82 0.39 0.00 1.00 1210 0.78 0.41 0.00 1.00 372
Oil Price (12 month strip, log) 4.01 0.50 2.96 4.99 1210 4.01 0.55 2.96 4.99 372
Conventional Reserve dummy
*
0.58 0.49 0.00 1.00 1210 0.49 0.50 0.00 1.00 372
Cross-border dummy 0.31 0.46 0.00 1.00 1210
Press Freedom Host Rating 0.88 0.15 0.00 1.00 967 0.79 0.20 0.10 1.00 282
Press Freedom Home Rating 0.91 0.13 0.00 1.00 943 0.88 0.18 0.12 1.00 258
OBI Host Rating 71.03 18.25 0.00 87.73 792 57.96 22.82 0.00 87.73 273
OBI Home Rating 75.01 12.49 30.63 88.43 711 67.19 15.19 30.63 88.43 192
Accounting Quality Host Rating 46.99 11.66 6.00 56.00 958 40.02 15.92 6.00 56.00 239
Accounting Quality Home Rating 47.46 10.29 10.00 56.00 1001 42.76 12.37 10.00 56.00 282
ICRG Host Rating 80.46 9.57 36.50 97.00 1207 74.72 13.33 36.50 97.00 369
ICRG Home Rating 82.93 6.16 44.00 97.00 1175 82.79 7.09 54.00 97.00 337
POLCON Host Rating 0.40 0.12 0.00 0.69 1210 0.35 0.18 0.00 0.69 372
POLCON Home Rating 0.42 0.09 0.00 0.69 1174 0.42 0.13 0.00 0.69 336
CPI Host Rating 7.12 2.02 1.78 9.47 1210 5.90 2.63 1.78 9.47 372
CPI Home Rating 7.62 1.45 2.18 9.47 1175 7.49 1.66 2.32 9.47 337
Global Integrity Host Rating 80.40 8.19 41.96 86.73 1100 74.64 10.96 41.96 86.73 296
Global Integrity Home Rating 81.61 6.63 49.79 87.94 1055 78.70 8.65 49.79 87.94 251
*
Royalty interests also coded as conventional; Higher cost reserve types include bitumen, coalbed, deepwater, enhanced, frontier, heavy oil, LNG, shale,
nonconventional, shallow water, tight gas, and diversi?ed.
A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47 39
tic, and are more likely to disclose partial information.
These differences are statistically signi?cant at the 0.1%
level (v
2
= 21.5).
Panels AandBof Table 4present the correlationmatrices
for the full sample and cross-border sub-sample, respec-
tively. Panel A shows that Full Disclosure is positively corre-
lated with Need for External Finance, Oil Price, Conventional
Reserves, and all institutional variables, and is negatively
correlated with cross-border transactions. Not surprisingly,
institutional variables are correlated (the median correla-
tion among host-country institutional variables in cross-
border transactions is 0.71). Due to the potential for multi-
collinearity among institutional measures, we allow each
of these measures to enter in separate regressions.
Multinomial logit results
As our dependent variable representing choice of invest-
ment transparency is categorical with three outcomes (full,
Table 3
Disclosure by ?rms with both crossborder and domestic transactions.
Number of
transactions
Full
disclosure
Partial
disclosure
Minimal
disclosure
Total
Crossborder 47 51 16 114
41% 45% 14% 100%
Domestic 86 25 10 121
71% 21% 8% 100%
Total 133 76 26 235
Differences are signi?cant (Pearson v
2
(2) = 21.5, Pr = 0.000).
Table 4
Correlations.
Panel A: Full Sample 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
1 Full Disclosure 1.00
2 Partial Disclosure -0.72 1.00
3 Minimal Disclosure -0.47 -0.28 1.00
4 Press Freedom Host Rating 0.09 -0.03 -0.08 1.00
5 Press Freedom Home Rating 0.12 -0.11 -0.02 0.48 1.00
6 OBI Host Rating 0.10 -0.04 -0.08 0.73 0.28 1.00
7 OBI Home Rating 0.13 -0.12 -0.02 0.52 0.67 0.65 1.00
8 Accounting Quality Host Rating 0.14 -0.08 -0.10 0.65 0.40 0.70 0.49 1.00
9 Accounting Quality Home Rating 0.22 -0.15 -0.13 0.37 0.48 0.33 0.53 0.69 1.00
10 ICRG Host Rating 0.09 -0.03 -0.09 0.79 0.38 0.77 0.46 0.62 0.24 1.00
11 ICRG Home Rating 0.10 -0.07 -0.05 0.38 0.75 0.17 0.55 0.25 0.38 0.43 1.00
12 POLCON Host Rating 0.05 0.02 -0.09 0.71 0.35 0.50 0.43 0.51 0.32 0.55 0.36 1.00
13 POLCON Home Rating 0.03 -0.01 -0.03 0.34 0.73 0.12 0.54 0.27 0.24 0.26 0.58 0.46 1.00
14 CPI Host Rating 0.10 -0.02 -0.11 0.86 0.47 0.79 0.53 0.70 0.35 0.89 0.45 0.69 0.38 1.00
15 CPI Home Rating 0.12 -0.07 -0.07 0.43 0.81 0.20 0.60 0.34 0.54 0.38 0.86 0.43 0.68 0.51 1.00
16 Global Integrity Host Rating 0.08 -0.04 -0.07 0.76 0.31 0.92 0.62 0.68 0.37 0.65 0.17 0.46 0.13 0.67 0.21 1.00
17 Global Integrity Home Rating 0.10 -0.12 0.02 0.39 0.67 0.57 0.92 0.41 0.51 0.27 0.41 0.22 0.31 0.31 0.44 0.63 1.00
18 Assets (Log) 0.06 0.01 -0.11 -0.19 -0.23 -0.13 -0.24 -0.16 -0.22 -0.23 -0.32 -0.17 -0.14 -0.25 -0.32 -0.11 -0.10 1.00
19 Need for External Finance 0.01 -0.02 0.00 0.13 0.11 0.14 0.19 0.12 0.16 0.07 0.11 0.09 0.08 0.12 0.16 0.08 0.09 -0.46 1.00
20 Buyer SOE dummy -0.13 0.05 0.12 -0.31 -0.52 -0.31 -0.53 -0.33 -0.48 -0.26 -0.42 -0.23 -0.28 -0.31 -0.51 -0.35 -0.52 0.35 -0.23 1.00
21 Buyer Listed dummy 0.28 -0.24 -0.09 0.03 0.16 0.00 0.14 0.00 0.11 0.03 0.15 0.09 0.17 0.05 0.18 -0.01 0.09 0.01 0.04 -0.15 1.00
22 Oil Price (12 month strip, log) 0.15 -0.10 -0.08 -0.07 -0.06 -0.01 -0.08 0.44 0.55 -0.17 -0.25 -0.04 -0.06 -0.08 -0.06 0.02 -0.03 0.13 0.11 0.01 0.02 1.00
23 Conventional Reserve dummy 0.18 -0.10 -0.12 -0.11 -0.06 -0.07 -0.04 -0.12 -0.06 -0.07 -0.04 -0.08 -0.09 -0.11 -0.07 -0.10 -0.07 -0.15 0.06 -0.02 0.02 0.05 1.00
24 Cross-border dummy -0.25 0.16 0.14 -0.37 -0.11 -0.52 -0.38 -0.35 -0.29 -0.40 -0.01 -0.23 0.04 -0.41 -0.06 -0.43 -0.25 0.20 -0.10 0.22 -0.06 0.00 -0.13 1.00
Panel B: Cross-border Sub-sample 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
1 Full Disclosure 1.00
2 Partial Disclosure -0.62 1.00
3 Minimal Disclosure -0.41 -0.46 1.00
4 Press Freedom Host Rating -0.13 0.17 -0.05 1.00
5 Press Freedom Home Rating 0.13 -0.14 0.00 0.05 1.00
6 OBI Host Rating -0.10 0.16 -0.07 0.66 0.04 1.00
7 OBI Home Rating -0.03 -0.04 0.08 0.05 0.62 0.02 1.00
8 Accounting Quality Host Rating -0.07 0.13 -0.08 0.65 0.18 0.66 0.10 1.00
9 Accounting Quality Home Rating 0.24 -0.12 -0.15 0.05 0.37 -0.01 0.26 0.23 1.00
10 ICRG Host Rating -0.09 0.10 -0.01 0.71 0.06 0.74 0.18 0.70 -0.08 1.00
11 ICRG Home Rating 0.16 -0.17 0.00 -0.04 0.68 0.09 0.49 0.04 0.35 0.06 1.00
12 POLCON Host Rating -0.08 0.11 -0.03 0.66 0.09 0.39 0.17 0.46 0.11 0.43 0.11 1.00
13 POLCON Home Rating 0.02 -0.05 0.03 0.11 0.73 0.09 0.43 0.13 0.05 0.07 0.44 0.18 1.00
14 CPI Host Rating -0.10 0.12 -0.03 0.81 0.13 0.74 0.20 0.77 -0.01 0.89 0.14 0.57 0.14 1.00
15 CPI Home Rating 0.17 -0.16 -0.01 -0.03 0.73 0.08 0.46 0.05 0.53 0.02 0.85 0.11 0.49 0.09 1.00
16 Global Integrity Host Rating -0.08 0.20 -0.14 0.81 0.01 0.86 0.03 0.78 0.05 0.73 0.05 0.50 0.05 0.77 0.04 1.00
17 Global Integrity Home Rating 0.13 -0.13 0.00 0.02 0.77 0.05 0.87 0.15 0.41 0.13 0.62 0.09 0.40 0.15 0.58 0.04 1.00
18 Assets (Log) -0.08 0.12 -0.05 0.07 -0.14 0.04 -0.12 -0.09 -0.36 -0.04 -0.34 0.00 -0.01 -0.04 -0.34 -0.07 -0.22 1.00
19 Need for External Finance 0.03 -0.05 0.02 0.10 0.12 0.15 0.19 0.03 0.19 0.02 0.20 0.05 0.08 0.07 0.26 0.08 0.28 -0.52 1.00
20 Buyer SOE dummy -0.10 -0.02 0.14 -0.03 -0.46 -0.08 -0.38 -0.14 -0.44 -0.04 -0.41 -0.06 -0.19 -0.07 -0.53 -0.17 -0.50 0.36 -0.31 1.00
21 Buyer Listed dummy 0.30 -0.30 0.01 -0.09 0.30 -0.05 0.26 -0.16 0.22 -0.10 0.16 0.01 0.19 -0.11 0.24 -0.12 0.31 0.08 0.07 -0.10 1.00
22 Oil Price (12 month strip, log) 0.10 0.01 -0.12 -0.04 -0.09 -0.05 -0.33 0.30 0.50 -0.15 -0.25 -0.04 -0.06 -0.12 -0.05 -0.04 -0.21 0.07 0.00 0.07 0.07 1.00
23 Conventional Reserve dummy 0.15 -0.06 -0.10 -0.27 -0.04 -0.19 -0.09 -0.27 -0.03 -0.23 -0.03 -0.17 -0.10 -0.30 -0.05 -0.17 -0.03 -0.16 0.04 0.00 0.02 0.10 1.00
Notes: Correlations for country-level institutional measures are shaded.
40 A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47
partial, and minimal disclosure), we specify a multinomial
logit model (Greene, 2000). We report the results for each
model inthree columns. The ?rst two columns compare Par-
tial Disclosure andFull Disclosure to a baseline of Minimal Dis-
closure; the third column compares Full Disclosure to a
baseline of Partial Disclosure. For each independent variable
we present the coef?cient estimate (beta) with the odds
ratio (calculated as e
beta
) in italics. The odds ratio represents
the change inthe odds of moving fromthe baseline level to a
higher disclosure level for a one unit increase in the inde-
pendent variable; ratios greater than one indicate increased
odds, ratios less than one, decreased odds.
Our ?rm-level data are for the most recent reporting
date prior to the transaction announcement. Institutional
variables are for the year prior to announcement to re?ect
information available to managers. Following the treat-
ment of institutional variables in prior work (Guerreiro
et al., 2012), we scale all country-level indices to have
mean zero and standard deviation of one to enable compa-
rability across coef?cients (Tabachnick & Fidell, 2001).
Standard errors are robust to heteroskedasticity, and clus-
tered by home country.
12
Table 5 presents the multinomial logit results of our
regression of investment transparency on ?rm-level deter-
minants. Our ?rst hypothesis tests whether multinational
investment is associated with greater disclosure (H1A) or
less disclosure (H1A
0
) than domestic investment, as pre-
dicted by the global-diversi?cation and investor-sophistica-
tion theories, respectively. The investor sophistication view
(H1A
0
) is supported: cross-border transactions are less
transparent. The bivariate difference noted in Table 3 holds
in a multivariate setting. The coef?cient for the Cross-bor-
der dummy for Full Disclosure in Model 1 is negative and
signi?cant; this coef?cient remains signi?cant in later
speci?cations that control for host- and home-country
characteristics. The odds ratios of 0.178 (respectively,
0.330) on Full vs. Minimal (respectively, Full vs. Partial)
can be interpreted as an 82.8% (respectively, 67.0%)
decrease in the odds of full disclosure vs. minimal disclo-
sure (respectively, vs. partial disclosure) in cross-border
transactions.
In addition, when we separately estimate our cross-bor-
der and domestic sub-samples, we see that the Buyer Listed
dummy is negative and signi?cant in cross-border transac-
tions (Model 2), but insigni?cant in domestic transactions
(Model 3). These results indicate that publicly-listed ?rms
are less transparent than unlisted ?rms in cross-border,
Table 5
Multinomial logit regressions of investment transparency on ?rm-level determinants.
Variables (1) Full Sample (2) Cross-border Sub-sample (3) Domestic Sub-sample
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Need for External Finance À0.033 À0.106 À0.073 À0.026 À0.137 À0.110 0.048 À0.012 À0.060
(0.125) (0.103) (0.101) (0.183) (0.114) (0.154) (0.191) (0.163) (0.123)
Odds Ratio 0.968 0.899 0.930 0.974 0.872 0.896 1.049 0.988 0.942
Assets (Log) 0.119 0.168 0.049 0.144 À0.014 À0.158 0.205 0.390
**
0.185
***
(0.149) (0.185) (0.103) (0.152) (0.167) (0.101) (0.198) (0.175) (0.070)
Odds Ratio 1.126 1.183 1.050 1.155 0.986 0.854 1.228 1.477 1.203
Cross-border dummy À0.616 À1.724
***
À1.108
***
(0.440) (0.473) (0.247)
Odds Ratio 0.540 0.178 0.330
Buyer SOE dummy À1.207
**
À2.653
**
À1.446
**
À1.589
*
À1.423 0.166 À1.568
*
À4.495
***
À2.927
***
(0.564) (1.089) (0.685) (0.849) (0.957) (0.615) (0.918) (1.202) (0.649)
Odds Ratio 0.299 0.070 0.236 0.204 0.241 1.181 0.208 0.011 0.054
Buyer Listed dummy À0.881 À0.628 0.254 À15.869
***
À14.524
***
1.345 0.491 0.337 À0.154
(0.749) (0.888) (0.496) (0.896) (1.269) (0.869) (0.606) (0.602) (0.529)
Odds Ratio 0.414 0.534 1.289 0.000 0.000 3.838 1.634 1.401 0.857
Oil Price (12 month strip, log) 0.459
***
1.096
***
0.636
***
0.768
**
0.753
*
À0.016 0.197 1.005
***
0.807
***
(0.158) (0.270) (0.184) (0.350) (0.427) (0.260) (0.193) (0.198) (0.196)
Odds Ratio 1.582 2.992 1.889 2.155 2.123 0.984 1.218 2.732 2.241
Conventional Reserve dummy 0.576
**
1.298
***
0.722
***
0.783
**
1.054
**
0.271 0.492
*
1.454
***
0.962
***
(0.245) (0.304) (0.222) (0.397) (0.490) (0.568) (0.263) (0.363) (0.187)
Odds Ratio 1.779 3.662 2.059 2.188 2.869 1.311 1.636 4.280 2.617
Observations 688 162 526
Firms 344 116 255
Pseudo R-squared 0.113 0.065 0.112
Notes: For the ?rst two columns in each regression, partial disclosure and full disclosure are compared to a baseline of minimal disclosure. For the third
column, full disclosure is compared to a baseline of partial disclosure. Institutional variables are standardized and oriented such that higher values indicate
stronger institutions. Odds ratios are presented in italics. Robust standard errors clustered by home country are in parentheses. Constant values are not
reported.
*
Signi?cant at the 10% level.
**
Signi?cant at the 5% level.
***
Signi?cant at the 1% level.
12
Because of the small number of deals per ?rm, we are not able to
cluster errors at the ?rm level.
A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47 41
but not in domestic, transactions. Listed-?rm shareholders
rely more on corporate reporting than shareholders of
unlisted ?rms (which tend to be closely-held). If cross-bor-
der information frictions reduce the bene?ts of disclosure,
it follows that disclosures will be less useful for listed
?rms’ foreign transactions.
The agency-theoretic hypothesis that ?rms in need of
external ?nance will make more transparent investments
(H1B) is not supported; the coef?cients on Need for External
Finance are not statistically signi?cant in any speci?cation.
Together with H1A, these results suggest that agency-the-
oretic considerations are not a dominant driver of invest-
ment-level disclosures.
Table 6 introduces institutional factors. The models are
designed to test the hypothesis of greater disclosure when
host- and home- countries’ societal institutions are more
transparent. We examine the cross-border sample, as
home and host measures are highly correlated in the full
sample where the preponderance of domestic transactions
means that home- and host-countries are the same for
about 70% of the observations.
13
Examining host countries ?rst, we ?nd that partial dis-
closure is favored over both full disclosure and minimal
disclosure in all speci?cations, although statistical signi?-
cance varies. The odds ratios for Partial vs. Minimal in Mod-
els 1 and 2 indicate that one standard deviation increases
in Press Freedom and OBI result in 29.2% and 70.4%
increases in the odds of partial disclosure, respectively.
As discussed above, managers may choose to appear more
transparent in the host country even if the act of disclosure
itself provides limited information. These results are con-
sistent with seeking legitimacy (H2A).
We ?nd mixed results for the hypothesis that MNEs
from more transparent societies behave more transpar-
ently (H2B). The coef?cients on Home Country Rating in
Model 3 suggests that better home country accounting
quality results in greater odds of full disclosure (a one stan-
dard deviation increase in accounting quality increases the
odds of full disclose by 59.0% over minimal disclosure, and
89.6% over partial disclosure). This result supports the
hypothesis that ?rms from countries with higher expecta-
tions of transparency disclose more abroad. The results for
Press Freedom and OBI, however, are inconsistent or in the
wrong direction.
In Table 7, we present the results of multinomial logit
regressions of investment transparency on home- and
host-country political risk and corruption. We ?nd little
support for a relationship between host-country political
risk and investment transparency (H3A). Only the coef?-
cients for POLCON in Model 2 are signi?cant, suggesting
that ?rms investing in countries with better political con-
straints tend to prefer partial disclosure over full disclosure
Table 6
Multinomial logit regressions of investment transparency on societal institutions for cross-border transactions.
(1) Press Freedom (2) Open Budget Index (3) Accounting Quality
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Host Country Rating 0.256
**
À0.050 À0.306
***
0.533
***
0.173 À0.360
***
0.135 À0.036 À0.171
(0.124) (0.090) (0.093) (0.186) (0.215) (0.126) (0.181) (0.113) (0.145)
Odds Ratio 1.292 0.951 0.736 1.704 1.189 0.698 1.145 0.965 0.843
Home Country Rating À0.219
**
À0.028 0.191
*
À0.412
*
À0.344
**
0.068 À0.176 0.464
**
0.640
***
(0.093) (0.085) (0.104) (0.243) (0.150) (0.224) (0.142) (0.188) (0.204)
Odds Ratio 0.803 0.972 1.210 0.662 0.709 1.070 0.839 1.590 1.896
Buyer SOE dummy À0.978
**
À0.872
**
0.106 À1.217
*
À0.856 0.361 À0.710
**
À1.005
**
À0.294
(0.432) (0.401) (0.474) (0.678) (0.539) (0.501) (0.355) (0.418) (0.441)
Odds Ratio 0.376 0.418 1.112 0.296 0.425 1.435 0.492 0.366 0.745
Buyer Listed dummy À0.226 1.329
**
1.555
**
À0.363 0.898 1.261 À0.736 0.810 1.546
**
(0.585) (0.601) (0.611) (0.877) (1.145) (0.840) (0.552) (0.677) (0.747)
Odds Ratio 0.798 3.777 4.735 0.696 2.455 3.529 0.479 2.248 4.693
Oil Price (12 month strip, log) 1.202
**
0.595 À0.608 0.288 0.318 0.031 1.051
***
0.120 À0.931
*
(0.581) (0.420) (0.372) (0.343) (0.380) (0.335) (0.376) (0.316) (0.480)
Odds Ratio 3.327 1.813 0.544 1.334 1.374 1.031 2.861 1.127 0.394
Conventional Reserve dummy 0.249 0.013 À0.236 0.490 0.953
*
0.462 0.703
*
0.546 À0.158
(0.265) (0.293) (0.354) (0.551) (0.524) (0.606) (0.390) (0.607) (0.612)
Odds Ratio 1.283 1.013 0.790 1.632 2.593 1.587 2.020 1.726 0.854
Observations 258 127 210
Firms 184 89 150
Pseudo R-squared 0.060 0.092 0.098
Notes: For the ?rst two columns in each regression, partial disclosure and full disclosure are compared to a baseline of minimal disclosure. For the third
column, full disclosure is compared to a baseline of partial disclosure. Institutional variables are standardized and oriented such that higher values indicate
stronger institutions. Odds ratios are presented in italics. Robust standard errors clustered by home country are in parentheses. Constant values are not
reported.
*
Signi?cant at the 10% level.
**
Signi?cant at the 5% level.
***
Signi?cant at the 1% level.
13
The median correlation between home and host institutional measures
is 0.51 in the full sample and 0.06 in the cross-border sample; however, our
results are robust across samples.
42 A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47
or minimal disclosure. The odds ratio on Host Country Rat-
ing in Model 2 suggests that a one standard deviation
increase in political constraints increases the odds of Par-
tial vs. Minimal by approximately 16% and the odds of Par-
tial vs. Full by approximately 15%. In contrast, we ?nd no
signi?cance for Host Country Rating on ICRG, or either of
our corruption measures.
Our results on Home Country Rating show a generally
strong relation between freedom from political risk and
corruption, and investment transparency. Our strongest
result comes from the ICRG political risk rating and the
CPI rating. The coef?cients on Home Country Rating in
Model 1 suggest ?rms in better-governed countries tend
to favor full disclosure. A one standard deviation increase
in the quality of political institutions increases the odds
of full disclosure over partial by more than 55%. Further-
more, when we introduce ?nancing needs in Table 8,
Models 1 and 2, we see that ?rms consistently prefer
higher levels of disclosure as home political institutions
improve. However, the Home Country Rating coef?cients
for our alternative measure of political risk, POLCON,
are insigni?cant across all speci?cations in Tables 7 and
8. Overall, we interpret these ?ndings as largely consis-
tent with the hypothesis that ?rms are more open with
their investments when home country institutions are
better.
Table 8 also tests the ‘‘Twin Agency Problem’’ that polit-
ical risks and agency problems are complements. We ?nd
little agreement on the signs of these coef?cients, and none
are statistically signi?cant save for the Partial vs. Minimal
coef?cients in Models 1 and 3. These vary in signi?cance
but are not directionally consistent with our hypothesis.
Thus, H3C is not supported.
Discussion and conclusion
Our results support the view that investment transpar-
ency represents a strategic choice. On one hand, civil soci-
ety demands transparency in exchange for legitimacy. On
the other, reducing information asymmetries between
the ?rm and interested third parties such as predatory
governments and rent-seeking public of?cials exposes
the MNE to political risk and the costs of corruption.
Foreign transactions are signi?cantly less transparent
across all speci?cations and samples. These ?ndings are
robust to the MNE’s institutional environment – less infor-
mation is disclosed about asset purchases abroad, even
after controlling for home and host country institutions.
This ?nding supports the investor-sophistication strand
rather than the global-diversi?cation strand of the account-
ing literature: cross-border information frictions reduce
the ?rm’s incentive to disclose.
This ?nding suggests that capital markets, the primary
audience examined in traditional studies of voluntary
disclosure, are less in?uential in the multinational
context. Instead, we ?nd that institutional rather than
agency-theoretic considerations drive MNE investment-
level disclosures; need for external ?nancing is unrelated
to the ?rm’s choice of investment transparency. Our theory
Table 7
Multinomial logit regressions of investment transparency on political risk and corruption in cross-border transactions.
(1) ICRG (2) POLCON (3) CPI (4) Global Integrity
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Host Country Rating 0.090 À0.004 À0.094 0.149
**
À0.017 À0.165
***
0.151 0.033 À0.118 0.148 0.047 À0.100
(0.096) (0.087) (0.095) (0.062) (0.063) (0.063) (0.126) (0.077) (0.119) (0.220) (0.192) (0.130)
Odds Ratio 1.094 0.996 0.910 1.161 0.983 0.848 1.163 1.034 0.889 1.160 1.048 0.905
Home Country Rating À0.233
**
0.209 0.441
***
À0.108 À0.082 0.026 À0.321
***
0.070 0.390
***
À0.101 0.075 0.176
(0.109) (0.182) (0.154) (0.101) (0.131) (0.079) (0.120) (0.144) (0.134) (0.113) (0.138) (0.174)
Odds Ratio 0.792 1.232 1.554 0.898 0.921 1.026 0.725 1.073 1.477 0.904 1.078 1.192
Buyer SOE dummy À0.925
***
À0.736
**
0.189 À0.599 À0.922
***
À0.323 À1.091
***
À0.786
**
0.305 À1.036
**
À0.904
**
0.132
(0.353) (0.325) (0.447) (0.378) (0.320) (0.407) (0.417) (0.313) (0.461) (0.481) (0.452) (0.627)
Odds Ratio 0.397 0.479 1.208 0.549 0.398 0.724 0.336 0.456 1.357 0.355 0.405 1.141
Buyer Listed dummy À0.401 1.391
**
1.792
***
À0.267 1.601
***
1.868
***
À0.245 1.516
***
1.761
***
À0.670 0.645 1.315
*
(0.473) (0.573) (0.607) (0.509) (0.572) (0.566) (0.486) (0.571) (0.584) (0.547) (0.735) (0.685)
Odds Ratio 0.670 4.019 6.001 0.766 4.958 6.475 0.783 4.554 5.818 0.512 1.906 3.725
Oil Price (12 month strip, log) 0.559
**
0.827
***
0.269 0.578
**
0.694
***
0.116 0.616
**
0.695
***
0.079 0.810
*
0.811
**
0.001
(0.268) (0.232) (0.249) (0.245) (0.222) (0.227) (0.250) (0.215) (0.206) (0.471) (0.369) (0.221)
Odds Ratio 1.749 2.286 1.309 1.782 2.002 1.123 1.852 2.004 1.082 2.248 2.250 1.001
Conventional Reserve dummy 0.361
*
0.803
**
0.443 0.368 0.774
**
0.406 0.379 0.817
**
0.437 0.047 0.477 0.431
(0.206) (0.382) (0.435) (0.234) (0.330) (0.385) (0.232) (0.382) (0.436) (0.337) (0.456) (0.502)
Odds Ratio 1.435 2.232 1.557 1.445 2.168 1.501 1.461 2.264 1.548 1.048 1.611 1.539
Observations 334 336 337 202
Firms 235 236 237 148
Pseudo R-squared 0.074 0.063 0.072 0.053
Notes: For the ?rst two columns in each regression, partial disclosure and full disclosure are compared to a baseline of minimal disclosure. For the third
column, full disclosure is compared to a baseline of partial disclosure. Institutional variables are standardized and oriented such that higher values indicate
stronger institutions. Odds ratios are presented in italics. Robust standard errors clustered by home country are in parentheses. Constant values are not
reported.
*
Signi?cant at the 10% level.
**
Signi?cant at the 5% level.
***
Signi?cant at the 1% level.
A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47 43
does not explicitly explain this ?nding, but we can offer
two insights. First, the ?nancial reporting context in which
prior work has established a link between voluntary dis-
closure and the ?rm’s cost of capital considers the ?rm as
a whole. Yet, the external ?nancing needs of the entire ?rm
may not be representative of the ?nancing needs of an
individual investment project. Second, this result may be
a secondary effect of home bias resulting from investors’
discounting foreign information, as it is consistent with
foreign investment disclosures being less useful for raising
capital.
Countries marked by high normative expectations of
transparency tend to encourage greater transparency in
investment disclosures. Yet, home- and host-country
norms have notably different effects. Greater host-country
norms are strongly correlated with partial disclosure,
while ?rms from home countries with higher expectations
of transparency are less likely to release only partial
information.
We interpret these results as consistent with legiti-
macy-seeking behavior. Partial transparency is adequate
in the host country. Limited disclosure is less likely to sat-
isfy civil society in the home country, however, for two
reasons. The ?rst involves familiarity; home country stake-
holders are better acquainted with the ?rm, and thus may
need more speci?c information to access whether corpo-
rate actions are consistent with the values of society. The
second is one of embeddedness. While oil industry MNEs
operate assets in several countries around the world, it is
the institutional logic of the home-country that is most
Table 8
Multinomial logit regressions of investment transparency on political risk and ?rm-level interactions in cross-border transactions.
(1) ICRG (2) ICRG (3) POLCON (4) POLCON
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Partial vs.
Minimal
Full vs.
Minimal
Full vs.
Partial
Host Country Rating 0.173 0.076 À0.097 0.254 0.113 À0.141 0.035 À0.036 À0.072 0.067 À0.009 À0.076
(0.161) (0.175) (0.197) (0.199) (0.162) (0.209) (0.102) (0.155) (0.119) (0.125) (0.151) (0.124)
Odds Ratio 1.189 1.079 0.908 1.289 1.120 0.868 1.036 0.965 0.931 1.069 0.991 0.927
Home Country Rating À0.055 0.546
**
0.601
***
0.133 0.879
***
0.746
***
0.026 À0.130 À0.156 0.044 À0.096 À0.140
(0.236) (0.241) (0.148) (0.282) (0.277) (0.160) (0.203) (0.210) (0.176) (0.213) (0.231) (0.149)
Odds Ratio 0.946 1.726 1.824 1.142 2.408 2.109 1.026 0.878 0.856 1.045 0.908 0.869
Need for External
Finance
À0.146 À0.128 0.018 À0.035 À0.179 À0.144 À0.200 À0.316 À0.116 À0.025 À0.169 À0.144
(0.211) (0.231) (0.211) (0.177) (0.181) (0.185) (0.232) (0.216) (0.152) (0.194) (0.115) (0.150)
Odds Ratio 0.864 0.880 1.018 0.966 0.836 0.866 0.819 0.729 0.890 0.975 0.845 0.866
Need for Ext.
Fin. Â Host Rating
À0.247
*
À0.046 0.201 À0.228
**
À0.162 0.066
(0.136) (0.151) (0.123) (0.104) (0.103) (0.085)
Odds Ratio 0.781 0.955 1.223 0.796 0.850 1.068
Need for Ext.
Fin. Â Home Rating
0.268 0.411 0.144 À0.018 0.015 0.033
(0.285) (0.258) (0.104) (0.150) (0.110) (0.130)
Odds Ratio 1.307 1.508 1.155 0.982 1.015 1.034
Assets (Log) 0.155 0.053 À0.102 0.144 0.067 À0.077 0.122 À0.045 À0.167
*
0.139 À0.022 À0.161
(0.156) (0.163) (0.097) (0.140) (0.151) (0.101) (0.154) (0.172) (0.095) (0.161) (0.175) (0.098)
Odds Ratio 1.168 1.054 0.903 1.155 1.069 0.926 1.130 0.956 0.846 1.149 0.978 0.851
Buyer SOE dummy À1.440
*
À1.150 0.289 À1.472 À1.155 0.317 À1.559
**
À1.603
*
À0.044 À1.503
*
À1.539
*
À0.035
(0.825) (0.952) (0.834) (0.905) (1.016) (0.857) (0.789) (0.928) (0.667) (0.807) (0.927) (0.636)
Odds Ratio 0.237 0.317 1.335 0.229 0.315 1.373 0.210 0.201 0.957 0.222 0.215 0.966
Buyer Listed dummy À15.749
***
À15.702
***
0.048 À14.235
***
À14.202
***
0.033 À15.524
***
À13.352
***
2.171 À15.469
***
À13.421
***
2.048
(1.102) (1.036) (0.980) (1.061) (1.070) (1.180) (1.156) (1.358) (1.478) (1.097) (1.489) (1.511)
Odds Ratio 0.000 0.000 1.049 0.000 0.000 1.034 0.000 0.000 8.767 0.000 0.000 7.752
Oil Price (12 month
strip, log)
0.809
**
1.038
***
0.228 0.844
**
1.133
***
0.290 0.880
***
0.923
**
0.043 0.743
**
0.793
*
0.050
(0.379) (0.390) (0.223) (0.384) (0.390) (0.213) (0.312) (0.425) (0.285) (0.344) (0.463) (0.279)
Odds Ratio 2.246 2.824 1.256 2.326 3.105 1.336 2.411 2.517 1.044 2.102 2.210 1.051
Conventional Reserve
dummy
0.953
***
1.158
**
0.205 1.060
***
1.283
**
0.223 0.933
**
1.046
*
0.114 0.820
**
0.955
*
0.135
(0.315) (0.584) (0.712) (0.373) (0.626) (0.750) (0.378) (0.562) (0.581) (0.375) (0.529) (0.619)
Odds Ratio 2.593 3.184 1.228 2.886 3.607 1.250 2.542 2.846 1.121 2.270 2.599 1.145
Observations 158 158 160 160
Firms 113 113 114 114
Pseudo R-squared 0.110 0.113 0.080 0.069
Notes: For the ?rst two columns in each regression, partial transparency and full transparency are compared to a baseline of minimal transparency. For the
third column, full transparency is compared to a baseline of partial transparency. Institutional variables are standardized and oriented such that higher
values indicate stronger institutions. Odds ratios are presented in italics. Robust standard errors clustered by home country are in parentheses. Constant
values are not reported.
*
Signi?cant at the 10% level.
**
Signi?cant at the 5% level.
***
Signi?cant at the 1% level.
44 A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47
likely to inform managerial preferences for disclosure.
Indeed, of all of our measures of transparency norms, our
strongest results suggest that ?rms from home countries
with higher accounting standards are the most likely to
fully disclose.
This work extends the accounting literature on volun-
tary disclosure (Healy & Palepu, 2001; Verrecchia, 2001)
empirically and theoretically. We present a novel way of
measuring voluntary disclosure that, unlike prior studies
that look at aggregate outcome-based measures, leverages
missing data on reserve transactions to test theories about
disclosure decisions at the investment-level. We demon-
strate that transparency is not merely a ?rm characteristic,
but a non-market strategy ?rms employ to manage various
stakeholder audiences. This concept complements recent
work by Ahern and Sosyura (2014), who examine how
?rms strategically control information dissemination in
the media. We also contribute to a nascent but growing lit-
erature exploring whether countries matter for transpar-
ency and voluntary disclosure (Healy & Serafeim, 2013;
Shi et al., 2012).
This study is not without limitations. While our data
allow us to circumvent a number of empirical challenges,
as in all transaction-level research we are unable to
observe acquisitions that are never announced or captured
by our data provider.
14
This is unlikely to be a serious prob-
lem, given the high pro?le and extensive scrutiny of our
empirical setting. Still, our study assumes such transactions,
if they exist, introduce no systematic bias.
We are also unable to determine systematically
whether or not the prices disclosed are over- or under-pay-
ments, as reserve and pricing data are available only when
disclosed. The question of whether managers reduce trans-
parency to hide overpayment would be an interesting
extension of this work. Additionally, an alternative
explanation for why some multinationals might or might
not disclose is the presence of a ?rm-wide policy on
disclosure. While this is a possibility, we have found no
examples of such a policy. Further, it is inconsistent with
our sample. The data in Table 3 indicate signi?cant varia-
tion within ?rms across transactions, suggesting this is
not the case.
The implications of this study should be of interest to
managers and policy makers. We contribute to the ongoing
debate in the extractive industries by scrutinizing the stra-
tegic dimensions of disclosure. While ceteris paribus,
greater transparency should be good for ?rm stakeholders
and society, removing all information asymmetries may
leave the ?rm vulnerable to external threats. Our theory
and evidence suggest the proper management of transpar-
ency and information asymmetries should be considered
an important component of MNE strategy.
We believe this study offers a number of promising ave-
nues for future research. First, we limit our study to the
global petroleum industry, which is particularly sensitive
to institutional risks. Further research is needed to general-
ize these ?ndings. Second, institutional determinants such
as normative expectations and political risk may interact,
creating potential second-order effects. Lastly, our research
raises the question of whether traditional theories of trans-
parency and voluntary disclosure are applicable to emerg-
ing-market multinationals, many of which are owned or
managed by the state. While we ?nd state-owned ?rms
to be less transparent on average, the political risks to SOEs
likely differ from privately held companies as they are
already the property of a sovereign government. We leave
these questions to future research.
Acknowledgements
We are grateful to IHS Herold for access to data; to
Meghana Ayyagari, Reid Click, Liesl Riddle, Jennifer Spen-
cer, Youli Zou, the Editor (Robert Bloom?eld), two anony-
mous reviewers; and participants at the GW
International Business Workshop and the Academy of
International Business annual meeting for comments;
and to the GW CIBER and Institute for Integrating Statistics
in Decision Sciences for ?nancial support. All omissions
and mistakes are the authors’.
References
Ahearne, A., Griever, W., & Warnock, F. (2004). Information costs and
home bias: An analysis of US holdings of foreign equities. Journal of
International Economics, 62(2), 313–336.
Ahern, K. R., & Sosyura, D. (2014). Who writes the news? Corporate press
releases during merger negotiations. The Journal of Finance, 69(1),
241–291.
Arnott, R. (2004). Oil and gas reserves: Communication with the ?nancial
sector. Oxford Institute for Energy Studies & Chatham House. .
Bednar, M. K. (2012). Watchdog or lapdog? A behavioral view of the
media as a corporate governance mechanism. Academy of
Management Journal, 55, 131–150.
Berry, K. T., & Wright, C. J. (2001). The value relevance of oil and gas
disclosures: An assessment of the market’s perception of ?rms’ effort
and ability to discover reserves. Journal of Business Finance &
Accounting, 28(5–6), 741–769.
Bertrand, M., & Mullainathan, S. (2000). Do CEOs set their own pay? The
ones without principals do (No. w7604). National Bureau of Economic
Research.
Beyer, A., Cohen, D. A., Lys, T. Z., & Walther, B. R. (2010). The ?nancial
reporting environment: Review of the recent literature. Journal of
Accounting and Economics, 50(2), 296–343.
Brown, P. R., Preiato, J. P., & Tarca, A. (2013). Measuring country differences
in enforcement of accounting standards: An audit and enforcement
proxy. Available at SSRN 2286029.
Cahan, S. F., Rahman, A., & Perera, H. (2005). Global diversi?cation and
corporate disclosure. Journal of International Accounting Research, 4,
73–94.
Callen, J. L., Hope, O. K., & Segal, D. (2005). Domestic and foreign earnings,
stock return variability, and the impact of investor sophistication.
Journal of Accounting Research, 43(3), 377–412.
Capron, L., Mitchell, W., & Swaminathan, A. (2001). Asset divestures
following horizontal acquisitions: A dynamic view. Strategic
Management Journal, 22, 817–844.
Chen, J. J., Tan, Y., Cheng, X., & Gong, S. (2013). Do higher value ?rms
voluntarily disclose more information? Evidence from China. In Asian
Finance Association 2013 Conference. .
Cho, C. H., Guidry, R. P., Hageman, A. M., & Patten, D. M. (2012). Do actions
speak louder than words? An empirical investigation of corporate
environmental reputation. Accounting, Organizations and Society,
37(1), 14–25.
Chung, K.-H., Ghicas, D., & Pastena, V. (1993). Lenders’ use of accounting
information in the oil and gas industry. Accounting Review, 68(4),
885–895.
Click, R., Jeong, Y., & Weiner, R. J. (2013). Does foreign investment carry
baggage from home? Working paper.
14
We thank an anonymous reviewer for bringing this to our attention.
A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47 45
Click, R., & Weiner, R. J. (2010). Resource nationalism meets the market:
Political risk and the value of petroleum reserves. Journal of
International Business Studies, 41, 783–803.
Coval, J. D., & Moskowitz, T. J. (1999). Home bias at home: Local equity
preference in domestic portfolios. The Journal of Finance, 54(6),
2045–2073.
Cuervo-Cazurra, A. (2006). Who cares about corruption? Journal of
International Business Studies, 37(6), 807–822.
Darby, S. (2009). The case for company-by-company reporting of data in the
extractive industries transparency initiative. New York, NY: The
Revenue Watch Institute.
Diamond, D. W., & Verrecchia, R. E. (1991). Disclosure, liquidity, and the
cost of capital. Journal of Finance, 46, 1325–1359.
Doidge, C., Karolyi, G. A., & Stulz, R. M. (2004). Why are foreign ?rms listed
in the US worth more? Journal of Financial Economics, 71(2), 205–238.
Durnev, A., & Fauver, L. (2008). Stealing from thieves: Firm governance and
performance when states are predatory (CEI working paper series 2008–
12). Center for Economic Institutions, Institute of Economic Research,
Hitotsubashi University. Retrieved 24.05.12.
Durnev, A., & Guriev, S. (2007). The resource curse: A corporate transparency
channel (CEFIR working paper w0108). Center for Economic and
Financial Research.
Retrieved 25.05.12.
Durnev, A., & Kim, E. (2005). To steal or not to steal: Firm attributes, legal
environment, and valuation. The Journal of Finance, 60(3), 1461–1493.
Dyck, A., & Zingales, L. (2004). Private bene?ts of control: An international
comparison. The Journal of Finance, 59(2), 537–600.
Dye, R. A. (1998). Investor sophistication and voluntary disclosures.
Review of Accounting Studies, 3(3), 261–287.
Eden, L. (2012). Transfer price manipulation. In P. Reuter (Ed.), Draining
development: Controlling ?ows of illicit funds from developing countries
(pp. 205–234). Washington, DC: The World Bank.
Ellis, J. A., Fee, C. E., & Thomas, S. E. (2012). Proprietary costs and the
disclosure of information about customers. Journal of Accounting
Research, 50(3), 685–727.
Fisman, R., & Miguel, E. (2007). Corruption, norms, and legal enforcement:
Evidence from diplomatic parking tickets. Journal of Political Economy,
115(6), 1020–1048.
Free, C., Salterio, S. E., & Shearer, T. (2009). The construction of
auditability: MBA rankings and assurance in practice. Accounting,
Organizations and Society, 34(1), 119–140.
French, K., & Poterba, J. (1991). Investor diversi?cation and international
equity markets. American Economic Review, 81, 222–226.
Gabbioneta, C., Greenwood, R., Mazzola, P., & Minoja, M. (2013). The
in?uence of the institutional context on corporate illegality.
Accounting, Organizations and Society, 38(6–7), 484–504.
Greene, W. H. (2000). Econometric analysis. Upper Saddle River, NJ:
Prentice Hall.
Greif, A. (2005). Commitment, coercion, and markets: The nature and
dynamics of institutions supporting exchange. In Handbook of new
institutional economics (pp. 727–786). US: Springer.
Guerreiro, M. S., Rodrigues, L. L., & Craig, R. (2012). Voluntary adoption of
international ?nancial reporting standards by large unlisted
companies in Portugal-Institutional logics and strategic responses.
Accounting, Organizations and Society, 37(7), 482–499.
Guriev, S., Kolotilin, A., & Sonin, K. (2011). Determinants of nationalization
in the oil sector: A theory and evidence from panel data. Journal of
Law, Economics, and Organization, 27(2), 301–323.
Hajzler, C. (2012). Expropriation of foreign direct investments: Sectoral
patterns from 1993 to 2006. Review of World Economics, 148(1),
119–149.
Hayward, M. L., & Hambrick, D. C. (1997). Explaining the premiums paid
for large acquisitions: Evidence of CEO hubris. Administrative Science
Quarterly, 103–127.
Healy, P. M., & Palepu, K. G. (2001). Information asymmetry, corporate
disclosure, and the capital markets: A review of the empirical
disclosure literature. Journal of Accounting and Economics, 31(1),
405–440.
Healy, P., & Serafeim, G. (2013). What impedes oil and gas companies’
transparency (HBS working paper 12-038). Harvard Business School.
.
Heaney, R., & Grundy, B. (2011). Valuation of crude oil and gas reserves.
Available at SSRN 1966829.
Henisz, W. J. (2006). Polcon_2005 codebook. Retrieved 17.10.12.
Henisz, W., & Williamson, O. (1999). Comparative economic
organization – Within and between countries. Business and Politics,
1(3), 261–277.
Hines, J. R. (2010). Treasure islands. The Journal of Economic Perspectives,
24(4), 103–125.
Hope, O. K. (2003). Disclosure practices, enforcement of accounting
standards, and analysts’ forecast accuracy: An international study.
Journal of Accounting Research, 41(2), 235–272.
Hope, O. K., Kang, T., Thomas, W. B., & Vasvari, F. (2008). The effects of
SFAS 131 geographic segment disclosures by US multinational
companies on the valuation of foreign earnings. Journal of
International Business Studies, 40(3), 421–443.
Hope, O. K., & Thomas, W. B. (2008). Managerial empire building and ?rm
disclosure. Journal of Accounting Research, 46(3), 591–626.
Hossain, M., Ahmed, K., & Godfrey, J. M. (2005). Investment opportunity
set and voluntary disclosure of prospective information: A
simultaneous equations approach. Journal of Business Finance &
Accounting, 32, 871–907.
Hunt, S. (2011). Re?ning black gold: The Dodd-Frank act and corruption in
the oil industry. UCLA Journal of International Law and Foreign Affairs,
16, 41–163.
Jandhyala, S., & Weiner, R. J. (2014). Institutions sans frontières:
International agreements and foreign investment. Journal of
International Business Studies, 45(6), 649–669.
Jensen, M. (1986). Costs of free cash ?ow, corporate ?nance, and
takeovers. The American Economic Review, 76(2), 323–329.
Jensen, N., & Johnston, N. (2011). Political risk, reputation, and the
resource curse. Comparative Political Studies, 44(6), 662–668.
Jeong, Y., & Weiner, R. J. (2012). Who bribes? Evidence from the United
Nations’ oil-for-food program. Strategic Management Journal, 33(12),
1363–1383.
Joe, J. R., Louis, H., & Robinson, D. (2009). Managers’ and investors’
responses to media exposure of board ineffectiveness. Journal of
Financial and Quantitative Analysis, 44(03), 579–605.
Kang, J. K., & Stulz, R. M. (1997). Why is there a home bias? An analysis of
foreign portfolio equity ownership in Japan. Journal of Financial
Economics, 46, 3–28.
Kanodia, C., & Lee, D. (1998). Investment and disclosure: The disciplinary
role of periodic performance reports. Journal of Accounting Research,
33–55.
Khanna, T., Palepu, K. G., & Srinivasan, S. (2004). Disclosure practices of
foreign companies interacting with US markets. Journal of Accounting
Research, 42(2), 475–508.
Kornai, J. (1986). The soft budget constraint. Kyklos, 39(1), 3–30.
Kostova, T., & Zaheer, S. (1999). Organizational legitimacy under
conditions of complexity: The case of the multinational enterprise.
Academy of Management Review, 24(1), 64–81.
Leuz, C., & Verrecchia, R. (2000). The economic consequences of increased
disclosure. Journal of Accounting Research, 38, 91–124.
Licht, A. (2001). The mother of all path dependencies: Toward a cross-
cultural theory of corporate governance systems. Delaware Journal of
Corporate Law, 26, 147–205.
Licht, A., Goldschmidt, C., & Schwartz, S. (2005). Culture, law, and
corporate governance. International Review of Law and Economics,
25(2), 229–255.
Maksimovic, V., & Phillips, G. (2001). The market for corporate assets:
Who engages in mergers and asset sales and are there ef?ciency
gains? The Journal of Finance, 56(6), 2019–2065.
Maksimovic, V., Phillips, G., & Prabhala, N. R. (2011). Post-merger
restructuring and the boundaries of the ?rm. Journal of Financial
Economics, 102(2), 317–343.
March, J. G., & Olsen, J. P. (1984). The new institutionalism:
Organizational factors in political life. American Political Science
Review, 78, 734–749.
Meek, G. K., & Thomas, W. B. (2004). A review of markets-based
international accounting research. Journal of International Accounting
Research, 3(1), 21–41.
Megginson, W. (2005). The ?nancial economics of privatization. New York,
NY: Oxford University Press.
Miller, M. H., & Upton, C. W. (1985). A test of the hotelling valuation
principle. The Journal of Political Economy, 93(1), 1–25.
Misund, B., Osmundsen, P., & Asche, F. (2005). The value-relevance of
accounting ?gures in the international oil and gas industry – Cash ?ows
or accruals? Working paper 43/05. University of Stavanger. .
Muñoz, J. S. (2009). Financing of oil and gas transactions. Texas Journal of
Oil, Gas, and Energy Law, 4(2), 225–267.
Oil companies in emerging markets: Safe sex in Nigeria. Economist, June
15, 2013. .
46 A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47
Oliver, C. (1991). Strategic responses to institutional processes. Academy
of Management Review, 16, 145–179.
Osmundsen, P. (2010). Chasing reserves: Incentives and ownership. In
Energy, natural resources and environmental economics (pp. 19–38).
Berlin Heidelberg: Springer.
Parsons, T. (1960). Structure and process in modern societies. Glencoe, IL:
Free Press.
Portes, R., & Rey, H. (2005). The determinants of cross-border equity
?ows. Journal of International Economics, 65(2), 269–296.
Rajan, R., & Zingales, L. (1998). Financial development and growth.
American Economic Review, 88, 559–586.
Rose-Ackerman, S. (2003). Corruption. The Encyclopedia of Public Choice,
67–76.
Rosenblum, P., & Maples, S. (2009). Contracts con?dential: Ending secret
deals in the extractive industries. New York: Revenue Watch.
Salter, S. B., Kang, T., Gotti, A. P. G., & Doupnik, T. S. (2013). The role of
social values, accounting values and institutions in determining
accounting conservatism. Management International Review, 53(4),
607–632.
Shi, Y., Magnan, M., & Kim, J. (2012). Do countries matter for voluntary
disclosure? Evidence from cross-listed ?rms in the US. Journal of
International Business Studies, 43, 143–165.
Shleifer, A., & Vishny, R. W. (1993). Corruption. Quarterly Journal of
Economics, 108, 599–617.
Stiglitz, J. E. (2008). 2007 Grotius lecture regulating multinational
corporations: Towards principles of cross-border legal frameworks
in a globalized world balancing rights with responsibilities. American
University International Law Review, 23(3), 451–558.
Stulz, R. (2005). Limits of globalization. Journal of Finance, 60, 1595–1638.
Stulz, R., & Williamson, R. (2003). Culture, openness, and ?nance. Journal
of Financial Economics, 70(3), 313–349.
Suchman, M. (1995). Managing legitimacy: Strategic and institutional
approaches. Academy of Management Review, 20(3), 571–610.
Tabachnick, B. G., & Fidell, L. S. (2001). Computer-Assisted Research Design
and Analysis (Vol. 748). Boston: Allyn and Bacon.
Taylor, G., Richardson, G., Tower, G., & Hancock, P. (2012). The
determinants of reserves disclosure in the extractive industries:
evidence from Australian ?rms. Accounting & Finance, 52(s1),
373–402.
Teall, H. D. (1992). Information content of Canadian oil and gas
companies’ historic cost earnings and reserves disclosures.
Contemporary Accounting Research, 8(2), 561–579.
Thomas, W. B. (1999). A test of the market’s mispricing of domestic and
foreign earnings. Journal of Accounting and Economics, 28(3), 243–267.
Thomas, M. A. (2010). What do the worldwide governance indicators
measure? European Journal of Development Research, 22, 31–54.
Thompson, A. C. (2001). The Hotelling Principle, backwardation of futures
prices and the values of developed petroleum reserves—the
production constraint hypothesis. Resource and Energy Economics,
23(2), 133–156.
Transparency International (2011). Bribe payers index 2011. .
Transparency International (2012). Corruption perceptions index 2012.
.
Vernon, R. (1971). Sovereignty at Bay: The Multinational Spread of U.S.
Enterprises. New York, NY: Basic Books.
Verrecchia, R. (2001). Essays on disclosure. Journal of Accounting and
Economics, 32, 97–180.
Wolf, C., & Weinschrott, D. (1973). International transactions and
regionalism: Distinguishing ‘‘insiders’’ from ‘‘outsiders’’. The
American Economic Review, 63(2), 52–60.
A.P. Cannizzaro, R.J. Weiner / Accounting, Organizations and Society 42 (2015) 32–47 47
doc_133651385.pdf