Determinants of the use of various control mechanisms in US–Chinese joint ventures

Description
This studyof US–Chinese joint ventures examined the effects of relative partner knowledge and specific asset
investments on the usage of various types of control mechanisms. These controls included expatriate staffing, socialisation
practices, delegated decision-making responsibilities, parent companycommunications and manager performance
incentives. Based on field visits and surveydata, we found that partner knowledge and specific asset investments
influenced a broad set of controls

Determinants of the use of various control mechanisms
in US–Chinese joint ventures
Peter Chalos
a,
*
, Neale G. OÕConnor
b
a
Department of Accounting (M/C 006), College of Business Administration, University of Illinois at Chicago, 601 South Morgan Street,
Chicago, IL 60607 7123, USA
b
Department of Accountancy, City University of Hong Kong, Hong Kong, China
Abstract
This study of US–Chinese joint ventures examined the e?ects of relative partner knowledge and speci?c asset
investments on the usage of various types of control mechanisms. These controls included expatriate sta?ng, sociali-
sation practices, delegated decision-making responsibilities, parent company communications and manager perfor-
mance incentives. Based on ?eld visits and survey data, we found that partner knowledge and speci?c asset investments
in?uenced a broad set of controls. Whilst the US joint venture partners considered controls to be particularly useful for
the selective transmission and protection of their knowledge, the Chinese partners viewed these same controls as a
means to selectively share and protect their speci?c asset investments in the ventures.
Ó 2004 Elsevier Ltd. All rights reserved.
Introduction
Foreign direct investment in China has grown
dramatically in the past decade, reaching US$52.7
billion in 2002 (PeopleÕs Daily, 2003). Given Chi-
naÕs recent entry into the World Trade Organisa-
tion, it is anticipated that future foreign direct
investment will increase 15–20% annually (China
Statistical Yearbook, 2002). Although joint ven-
tures (hereafter JVs) between overseas companies
and domestic ?rms remain one of the predominant
modes of entry into China, recent empirical evi-
dence from ?rms such as Rockwell, Caterpillar,
Kodak, and General Motors suggests that local
Ôenvironmental unpredictability, lack of bureau-
cratic transparency and partner unreliabilityÕ make
the management of JVs di?cult (Cooper & John-
son, 2000, p. 7). The lack of e?ective rule of law
heightens these concerns for management (Su,
1999). Surveys suggest that between one third to
one half of JVs are unpro?table (Cooper &
Johnson, 2000).
Despite the poor ?nancial results of so many
Chinese JVs, no consensus has emerged as to the
reasons for such lacklustre performance. One re-
cently proposed explanation among others sug-
gests that the management controls that are
implemented often do not capture the unique
contributions of the alliance partners (Child &
Yan, 1999; Yan & Luo, 2001). For example, each
JV partner may contribute unique knowledge such
as technical know-how and/or specialised assets
such as machinery that are di?cult to substitute or
independently develop without the other partner.
*
Corresponding author.
E-mail addresses: [email protected] (P. Chalos), acno@ci-
tyu.edu.hk (N.G. OÕConnor).
0361-3682/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.aos.2003.10.005
Accounting, Organizations and Society 29 (2004) 591–608
www.elsevier.com/locate/aos
Knowledge is considered to be an intangible asset
while specialised assets are considered to be spe-
ci?c or non-transferable outside of the venture. In
either case, the market value of these assets outside
of the alliance is di?cult to objectively determine.
This limits the e?ectiveness of equity ownership as
a control determinant. The lack of an e?ective rule
of law to protect such contributions via contrac-
tual stipulations compounds the issue. According
to Geringer and Frayne (1990, p. 104), ÔParents are
often unable to rely solely on their (equity) own-
ership position and related formal controls to
ensure that their objectives are adequately con-
sidered, instead requiring recourse to other modes
of in?uenceÕ. It appears to be essential that both
foreign and local partners have a clear and implicit
understanding of e?ective management controls
beyond the initial JV agreement contract. Dis-
cussants of these challenges have concluded that
while critical to success JV controls remain largely
unexplored and poorly understood (Chalos,
OÕConnor, & Xu, 1999; Child & Yan, 1999; Ink-
pen & Beamish, 1997).
This study examines unexplored determinants
of the use of JV controls in US–Chinese JVs sug-
gested by the above literature and our own ?eld
results. We examine the proposition that the use of
speci?c JV control mechanisms are motivated not
so much by partner equity as by the selective
sharing and protection of intangible partner
knowledge and assets that are speci?c to the JV.
These asset contributions determine the competi-
tive advantage of the venture (Beamish & Delios,
2001; Dyer & Singh, 1998; Yan & Luo, 2001).
Following calls by JV researchers (Hitt, Dacin,
Levitas, Arregle, & Borza, 2000), this study
examines the e?ect on JV controls of partner
knowledge sharing and protection (Inkpen, 1995;
Kogut, 1988) and speci?c non-transferable asset
investments in the JV (Beamish & Banks, 1987;
Hennart, 1991; Williamson, 1985). We examine
relationships from the perspective of the US and
Chinese partner. Few studies have considered the
perceptions of both local and foreign partners with
regard to management control issues. Such an
investigation is critical to developing an under-
standing of how each partner perceives and uses
controls in the management of the venture.
The remainder of the paper is organised as
follows. The actual controls that are used in JVs
are ?rst described. Partner factors that may a?ect
these controls are proposed. Speci?cally, the rela-
tionships of partner equity ownership, relative
knowledge and speci?c non-transferable asset
investments to management controls are hypoth-
esised. The research methods explain the basis for
the measurement and validation of the survey
instrument. The survey results report the statistical
relationships between equity ownership, partner
knowledge and asset investments speci?c to the
venture for each control. The paper concludes with
a discussion of the implications of the dissimilar
control perceptions that exist between the US and
Chinese JV respondents.
Theory development
A joint venture agreement legally creates a JV
through a contract and identi?es the major rights
and obligations of the participants. The JV
agreement typically identi?es the provisions for
management and the performance of JV obliga-
tions, including the transfer of knowledge. How-
ever, given contractual limitations, joint venture
partners typically rely on various control mecha-
nisms. These controls serve to in?uence behaviour
in areas that could not be included in the JV
agreement and to ensure the ful?lment of parts of
the agreement that are di?cult to legally enforce.
The aim of such controls is to align partner dif-
ferences and to promote a better understanding of
the intangible strengths and relationships that each
partner brings to the alliance.
Previous studies of alliances have examined
control mechanisms within the framework of cul-
tural controls, behavioural actions and outcome
results controls (Dekker, 2004; Groot & Merchant,
2000; Mjoen & Tallman, 1997). Broadly speaking,
as discussed in the literature: (i) cultural controls
encourage the alignment of partner values and
interests; (ii) behavioural controls direct and
monitor managers; and (iii) output controls
motivate managers to meet strategic goals and
objectives (Eisenhardt, 1985; Merchant, 1998;
Ouchi, 1980). Following the international JV
592 P. Chalos, N.G. O’Connor / Accounting, Organizations and Society 29 (2004) 591–608
control typology of Groot and Merchant (2000, p.
582) we examined : (i) two types of cultural con-
trols, expatriate sta?ng across four areas of
management and socialisation practices; (ii) two
types of behavioural controls, delegated decision
responsibility and parent company communica-
tions; and (iii) one type of outcome control,
manager performance incentives. The utilisation of
each JV control mechanism is brie?y reviewed
below.
1
Expatriate sta?ng has been characterised as a
common JV cultural control (Baliga & Jaeger,
1984; Chang & Taylor, 1999; Groot & Merchant,
2000). Expatriates are frequently employed in key
JV positions, including production, marketing, ?-
nance and administration. Underlying the use of
expatriate sta?ng as a cultural control mechanism
is evidence that this leads to better transmission of
common values and goals between foreign and
local partners (Isobe, Makino, & Montgomery,
2000; Lyles & Salk, 1996).
Socialisation practices have been found to
facilitate the transmission of organisational culture
in Chinese JVs (Firth, 1996). Socialisation is a
cultural control designed to promote congruent
expectations and mutual commitments through
which JV managers learn to share common atti-
tudes and knowledge of the organisation (Nonaka
& Takeuchi, 1995). Socialisation includes such
practices as training, mentoring, technical skill
transmission and the development of a common
organisational culture (Bauer, Morrison, & Call-
ister, 1998; Worm & Frankenstein, 2000).
Foreign (Chinese deputy) general manager deci-
sion-making responsibility refers to the degree of
control of strategic and operational decisions that
is placed in the hands of the foreign or Chinese
(deputy) general manager. The locating of such
authority is a fundamental behavioural control
challenge for management (Contractor & Lor-
ange, 1988; Groot & Merchant, 2000). Decision
authority over various aspects of JV operations
has been found to reside with managers who have
adequate experience and expertise. For example,
foreign partners have been found to be less likely
to delegate decision responsibilities to local part-
ners when the local partners are inexperienced
(Inkpen & Beamish, 1997).
Parent company communication refers to the
frequency of operational and ?nancial communi-
cation between the venture and the foreign and
Chinese parent company. Frequent communica-
tion enhances the transparency of partner actions
that builds trust and leads to greater cooperation
(Lane & Lubatkin, 1998). Viewed in this way,
parent company communication acts as a
behavioural control. Inkpen and Crossan (1995)
report that frequent meetings between venture
managers and head o?ce personnel, plant visits
and information sharing led to greater collabora-
tion.
Manager performance incentives involve the use
of budget performance linked rewards to foster
outcomes that are in the interests of both JV
partners (Khanna, Gulati, & Nohria, 1998). Per-
formance linked rewards promote goal congruence
and act as positive motivational control in JVs
(Geringer & Hebert, 1989). In the case of ?edging
JVs, the partnersÕ joint performance expectations
have been found to a?ect the performance incen-
tive system (Groot & Merchant, 2000).
A striking feature of cultural, behavioural and
outcome control mechanisms is that their purpose,
as de?ned and explored in the JV literature, has
little to do with equity ownership. While each
partnerÕs equity share is designed to represent
respective asset contributions, the intangible nat-
ure of partner assets at the formation of the alli-
ance suggests the limitations of equity as a control
determinant. Indeed, we argue that these non-
equity contributions are likely to be critical to one
aspect of control system design, the choice and the
use of various control mechanisms.
1
In this study, the identi?cation of speci?c controls came
from: the US–Chinese JV literature (for a review, see Child &
Faulkner, 1998); interviews with Chinese–US JV auditors and
managers of multinationals located in Hong Kong; and site
visits to four JVs in mainland China. The ?ve controls were
found to be most common out of a larger range of controls that
were observed in our site visits. For example, we found the
appointment of expatriates frequently existed in JV agreements.
Incentive issues, encouraging Chinese responsibility and
accountability for decision making, were also crucial. The lack
of adequate business orientation among Chinese managers
further suggested the importance of socialisation and commu-
nication integrating control mechanisms.
P. Chalos, N.G. O’Connor / Accounting, Organizations and Society 29 (2004) 591–608 593
Results of JV studies of equity ownership on
controls have been inconsistent. Some authors
have found evidence that the degree of ownership
increases control (Youse?, 1975), while others
have reported inconsistent results (Dang, 1977).
Groot and Merchant (2000, p. 606) speculated that
Ôunequal ownership may have signi?cant e?ects on
decision-making (controls)Õ. Mjoen and Tallman
(1997) found no linkage between equity and stra-
tegic controls or between equity and operational
controls (e.g. the allocation of decision-making
responsibility and the use of manager performance
incentives). This ?nding is consistent with earlier
work that reported the use of speci?c control
mechanisms to be of more relevance than overall
equity control to JV partners (Geringer, 1986;
Schaan, 1983).
Equity ownership appears to be a means of
controlling the overall direction of an alliance
rather than to be a determinant of speci?c man-
agement controls. For the US (Chinese) partner,
gaining higher equity ownership guarantees
greater control over the composition of the board
of directors. As the board appoints the JVÕs up-
per-level management, including expatriates (local
managers), the dominant partner has the poten-
tial to exercise signi?cant control over the long-
term strategic direction of the venture. However,
this does not axiomatically generalise to other
types of control mechanisms that require the
cooperation of employees involved in the daily
operations of the JV. For example, the in?uence
of dominant equity ownership on cultural,
behavioural and outcome controls at the opera-
tional level may be stymied by minority partner
resistance and lack of cooperation. Recent case
studies of JVs have found that partner equity is
not the only independent variable explaining
variation in the uses of behavioural control
mechanisms (Child & Faulkner, 1998). This
viewpoint reinforces ChildÕs (1994) contention
that majority equity ownership position, by itself,
cannot be used to enforce controls over a JV
without jeopardising its success. Hence, we pro-
pose the following hypothesis:
H1: US equity ownership is positively associated
with the proportion of expatriate sta?ng.
Selective dissemination and protection of for-
eign partner technical and business knowledge has
been speculated by JV researchers to in?uence
cultural, behavioural and outcome controls (Groot
& Merchant, 2000; Inkpen & Beamish, 1997).
Groot and Merchant (2000, p. 603) argue that
ÔpartnersÕ relative knowledge . . . seems to partly
explain the partnersÕ di?ering control foci and
their control intensityÕ. The acquisition of US
partner knowledge, commonly de?ned as the rel-
ative level of technical, production, marketing and
operational knowledge (and as de?ned in this
study) contributed by the US partner relative to
the Chinese partner, is a primary reason for JV
formation (Yan & Luo, 2001). Indeed, the will-
ingness of foreign ?rms to share knowledge has
been found to be a primary criterion in the selec-
tion of the JV foreign partner by local ?rms to a
much greater degree than for foreign ?rms in the
selection of their local partners (Hitt et al., 2000).
In US–Chinese JVs, it is common for the US
partner to provide technical information and
managerial knowledge. While important, Chinese
competencies in terms of local business connec-
tions, environmental and institutional knowledge
tend not to generate sustained competitive
advantage. Rather it is the superior technology
and management contributed by the US partner
that redress the operational de?ciencies of the
Chinese partner needed to sustain a competitive
position in the Chinese market (Naughton, 1995).
JV partners prosper in part by developing
knowledge sharing routines. However, the extent
to which the foreign partner shares knowledge
with the local partner varies according to both the
foreign partnerÕs control abilities and intentions.
The level of US partner knowledge a?ects controls
in two ways. First, the US partner has a strong
incentive to share non-proprietary technical
knowledge and business knowledge with the Chi-
nese partner. The Chinese partner shares this
objective. Sharing non-proprietary knowledge al-
lows both partners to generate superior economic
returns than would otherwise be the case. Even
when the local partner has unlimited access to the
foreign partnerÕs skills, the knowledge required to
eliminate partner dependency is usually more dif-
?cult for the local partner than the foreign partner
594 P. Chalos, N.G. O’Connor / Accounting, Organizations and Society 29 (2004) 591–608
to acquire. Thus, high levels of knowledge trans-
mission by the foreign partner have been found to
reduce inter-partner con?ict by promoting learn-
ing for the local partner, but only con?rmation of
managerial, not technical knowledge, has been
found (Steensma & Lyles, 2000).
Second, in JVs where the US partner contrib-
utes more knowledge of management, production
and patented know-how, greater potential exists
for the Chinese partner to expropriate this
knowledge and to act opportunistically in a com-
petitive fashion. To avoid this, the foreign partner
may take precautions by using controls to delib-
erately impede the transfer of certain types of
knowledge (Tiemessen, Lane, Crossan, & Inkpen,
1997).
2
Despite reforms in China, intellectual
property rights remain a serious problem for US
?rms. There are numerous cases in which local JV
partners or local contractors have pirated aspects
of product design. Even when foreign partners
seek damages through the court system, foreign
partners are frustrated by lack of power to enforce
payment (Weldon & Vanhonacker, 1999).
Accordingly, foreign partners are motivated to
protect proprietary resources against opportunism
by establishing controls that protect these re-
sources (Dyer & Singh, 1998). The success of
knowledge transfer thus not only depends on the
communication ability of the foreign partner as
well as the receptivity of the less experienced local
partner, but also depends on the e?cacy of the
cultural, behavioural and outcome controls
implemented in the JV.
3
We suggest the following knowledge-control
relationships. The ?rst is that both partners should
in?uence cultural, behavioural and outcome con-
trols that assist in the dissemination and receptiv-
ity of US partner knowledge, as it is in their
mutual economic interest. This knowledge in-
cludes all non-proprietary business and technical
knowledge that would bene?t the alliance. Second,
since proprietary knowledge is more likely to be
held by the foreign partner, who also has greater
experience, the US partner is likely to in?uence the
control system to a greater degree than the Chinese
partner. In terms of the individual controls, as
US partner knowledge increases, there is a greater
need to employ expatriates in upper-level man-
agement across functional areas to improve both
the protection and transmission of knowledge.
Similarly, there is a greater scope for the use of
socialisation practices to promote shared organi-
sational values between the JV partners and local
partner receptivity to the knowledge of the foreign
partner. When US partner knowledge is greater,
there is a need for decision responsibilities to be
initially assumed by the foreign general manager
for overall operational responsibility. This facili-
tates the gradual knowledge transition to the local
partner and encourages eventual acceptance of
greater managerial responsibilities once the local
partner has gained experience. The gradual values
and knowledge transition roles provided by soci-
alisation practices and the locating of decision
authority also promote trust which helps mitigate
opportunistic behaviour of the local partner in the
longer term.
With higher US partner knowledge contribu-
tion, we also expect that ?nancial and operational
budget communications with the parent ?rms will
be more frequent. The greater frequency of parent
company communications should encourage goal
setting and the transmission of learning feedback
with respect to routine business operations that
serves the learning objectives of both partners. In
addition, frequent communications enhance part-
ner transparency and cooperation. Finally, there is
potentially greater scope for the use of manager
performance incentives to promote a business
culture wherein managers and their subordinates
are made aware of relationships between knowl-
edge, e?ort and performance outcomes.
Summarising, both partners have an economic
incentive to share knowledge through the available
2
A reviewerÕs distinction of controls that both protect
proprietary knowledge as well as share non-proprietary knowl-
edge is appreciated.
3
To date however, evidence on the use of various control
mechanisms, similar to those examined in the present study, for
the purposes of knowledge transmission is limited to the
multinational subsidiary context (Gupta & Govindarajan,
2000). While the ownership structure admittedly di?ers between
JVs and multinational subsidiaries (i.e. knowledge is less likely
to be expropriated), both institutional structures face similar
challenges in transmitting foreign knowledge to local managers
via a common management control system.
P. Chalos, N.G. O’Connor / Accounting, Organizations and Society 29 (2004) 591–608 595
mechanisms of the control system but the US
partner has an additional concern, the protection
of proprietary knowledge. We expect that both
partners would perceive controls to be a positive
function of the greater relative knowledge contri-
bution of the foreign partner and the US partner
more so than the Chinese partner. Based on the
above, we propose the following hypotheses:
H2: US partner knowledge is positively associated
with the use of management controls for US
and Chinese partners. These controls include:
(a) the proportion of expatriate sta?ng; (b)
socialisation practices; (c) delegated decision
responsibilities to the foreign general man-
ager; (d) parent company communications;
and (e) manager performance incentives.
H3: The relationship between US partner knowl-
edge and the use of management controls is
stronger for the US JV partners than for the
Chinese partners. These controls include: (a)
the proportion of expatriate sta?ng; (b) soci-
alisation practices; (c) delegated decision
responsibilities to the foreign general man-
ager; (d) parent company communications;
and (e) manager performance incentives.
Our ?eld study results and those of others
indicate that asset speci?c investments are impor-
tant to US–Chinese ventures (for a review, see
Child & Faulkner, 1998; Yan & Luo, 2001) and
may be expected to have implications for the use of
controls. Speci?c asset investments in this study
included: employee development, supplier rela-
tionships, regulatory agency relationships, avail-
ability of alternative partners and local
investments in land, buildings and special purpose
machinery. These asset investments correspond
closely to WilliamsonÕs (1985) classi?cation of
asset speci?c investments in human assets, alterna-
tive partners and physical assets. Specialised assets
however have little value outside of the relation-
ship, and thus represent increased risk to the asset
holder. Hence, the partner with greater speci?c
assets in the venture is naturally more likely to
exercise control over these assets.
Local partners are interested in promoting
their unique asset contributions by establishing
controls that protect and promote their invest-
ments (Dyer & Singh, 1998). The foreign partner
also has speci?c investments in employee training
and governmental dispensations that are not
transferable outside of the alliance. While both
partners have speci?c investments, the Chinese
partner typically has fewer partner alternatives
than does the US partner, suggesting that the
local partner is more locked into the JV rela-
tionship. The local partner also has stronger
relational ties to suppliers, government agencies,
state employees and other agency relationships
than does the foreign partner due to a history of
transactions, reputation, cultural a?nity and
geographical proximity (Beamish & Lu, 2001;
Park & Luo, 2001). Finally, since the local part-
ner typically provides the JV property, plant and
equipment, the local partner has potentially more
physical assets than the foreign partner that are
speci?c to the venture.
Since the local partnerÕs physical, partner and
relational assets are more speci?c to the alliance
than the foreign partnerÕs assets, as these speci?c
asset investments increase, the local partner will be
more inclined to institute cultural, behavioural and
outcome controls to selectively share and protect
these assets. Child, Yan, and Ku (1997) have
suggested that Chinese partners focus on speci?c
controls as a means of economising on costs and
protecting their investments. The Chinese partner
is typically protected by regulatory agencies that
provide costless monitoring of foreign investors.
The resulting collusion provides the Chinese
partner with controls not available to the foreign
partner. For example, local government bargain-
ing power may be used to strengthen monitoring
and control of JV operations (Luo, Shenkar, &
Nyaw, 2001). This is more cost e?ective for the
alliance, bene?ting both partners.
We suggest the following speci?c asset-control
relationships. As speci?c asset investments in-
crease, the desire for greater local management
sta?ng and decision responsibilities should in-
crease. The use of socialisation practices also
serves to promote local partner speci?c assets in
human resources. For example, Chinese partners
often negotiate to include terms for the training of
local managers. Many of the large multinational
596 P. Chalos, N.G. O’Connor / Accounting, Organizations and Society 29 (2004) 591–608
?rms (e.g. Motorola, GE and Lucent) run accel-
erated management programmes that include
training, action learning, project management,
expatriate coaching and secondment (Dayao,
1997). The greater the human asset training
investment, the greater the need for controls.
Equally, as asset speci?c investments increase,
there is a greater requirement for communications
with the parent company and local state authori-
ties that are charged with monitoring the JV
activities and that ultimately provide a key source
of bargaining power for the local partner. Finally,
since risk increases as asset speci?c investments
increase, outcome controls serve to mitigate this
risk. The larger the potential loss from asset
expropriation, for example such as a partner
developing his own supplier contacts or leaving the
venture, the more the need to implement perfor-
mance incentives that align partner interests
(Parkhe, 1993).
The speci?c assets contributed by the foreign
partner are less prone to the costs of opportunism.
For example, investments in site-speci?c machin-
ery may be lost with less damage to a foreign
parent companyÕs strategic mission. The foreign
partner also has more partner alternatives. Local
partners on the other hand may perceive
behavioural and cultural controls to be more
e?ective in protecting physical property, plant and
equipment assets, safeguarding regulatory tax and
tari? dispensations, and guaranteeing contractual
promises subject to local laws. Since access to local
distribution channels, suppliers, workforce and
regulatory agencies are largely controlled by the
local partner, the foreign partner is often in a
subordinate position. It is therefore in the foreign
partnerÕs interest to have the local partner institute
the cultural, behavioural and outcome controls
necessary to safeguard asset investments speci?c to
the JV. Based upon the above, we propose the
following hypotheses:
H4: Asset speci?c investments are positively asso-
ciated with the use of management controls
for Chinese and US partners. These controls
include: (a) the proportion of non-expatriate
sta?ng; (b) socialisation practices; (c) dele-
gated decision responsibilities to the Chinese
deputy manager; (d) parent company commu-
nications; and (e) manager performance
incentives.
H5: The relationship between asset speci?c
investments and the use of management
controls is stronger for the Chinese JV
partners than for the US partners. These
controls include: (a) the proportion of non-
expatriate sta?ng; (b) socialisation practices;
(c) delegated decision responsibilities to the
Chinese deputy manager; (d) parent com-
pany communications; and (e) manager per-
formance incentives.
Methods
Survey instrument design: site visits
To ensure a valid survey design, we conducted
extensive site visits to US–Chinese JVs in Hong
Kong, Beijing and Shanghai before administering
our survey. The purpose of these visits was to
identify in speci?c terms the controls that were
used. We initially developed our instrument in
English over a six-month period, during which we
held numerous discussions with CPA practitioners
and US and Hong Kong general managers who
were doing business in China. Based on their
comments, we made survey revisions. To ensure
internal reliability, a bilingual Chinese research
professor translated the ?nal version of the
questionnaire into Mandarin. Another Chinese
professor, who was formerly a Ministry of Trade
and Economic Co-operation (MOFTEC) em-
ployee in Shanghai, back-translated, changed and
made corrections to the questionnaire.
4
Addi-
tional discussions with a partner at Grant
Thornton and with the chief accountant of the
Shanghai Stock Exchange provided further clari-
?cation of both the content and the translation of
the Chinese version (see Table 1). We mailed a
pilot sample of 20 questionnaires, and followed
this with telephone calls to clarify any ambigui-
ties.
4
A copy of the English and Chinese version of the
questionnaire is available from the authors.
P. Chalos, N.G. O’Connor / Accounting, Organizations and Society 29 (2004) 591–608 597
Table 1
Survey items of variable constructs
Variable # Item Response scale
Panel A: control antecedents
US equity ownership 1. % US equity held by the respondentÕs parent
company
Percentage US equity held
US partner knowledge (contri-
bution relative to the Chinese
partner)
Please indicate the extent to which each of the
following factors were provided by the respec-
tive partners:
Items 1–3
1. In comparison with your partner, who has a
better understanding of the production func-
tion (manufacturing technology, raw materials
price and availability) of the JV?
1 ¼My partner has a much better
understanding
2. In comparison with your partner, who has a
better understanding of the sales and market-
ing function (competitorsÕ actions, market
demand, product attributes) of the JV?
7 ¼I have much better understanding
(reverse scored for Chinese)
3. In comparison with your partner, who has a
better understanding of what can be achieved
in the JV?
4. Technology––knowledge, patents and produc-
tion processes
Items 4 and 5
5. Management expertise––production, adminis-
tration etc.
1(7) ¼Provided largely by foreign (Chi-
nese) parent (reverse scored for both
groups)
Asset speci?c investments (of
each individual partner)
1. Before the JV began, to what extent were
alternatives available to enter into other JVs if
negotiations failed?
1 ¼To a very little extent
2. To what extent is your investment operation
site speci?c (e.g. imported machinery, special
purpose of land)?
7 ¼To a very large extent (reverse scored
for item 1)
3. To what extent is your investment human asset
speci?c (heavy investment in training and
learning)?
Panel B: controls
Expatriate sta?ng 1. The number of foreign and number of Chinese
?nance and administration managers
Percentage of total foreign (expatriates)
managers out of the total number of
managers in: ?nance and administration;
manufacturing; sales; other
2. The number of foreign and number of Chinese
manufacturing managers
3. The number of foreign and number of Chinese
sales managers
4. The number of foreign and number of Chinese
managers in other functions
Socialisation practices What is the extent to which each of the
following training practices is true of your JV?
1. Chinese managers are sent to the foreign parent
company for in-house training and develop-
ment
1 ¼To a very little extent
2. Accountability and responsibility is encour-
aged through training programs conducted in-
house or locally
7 ¼To a very large extent
3. Information sharing of problem areas is
encouraged in operations (e.g. production
quality)
598 P. Chalos, N.G. O’Connor / Accounting, Organizations and Society 29 (2004) 591–608
Control antecedents
US equity ownership
We measured equity ownership as the percent-
age of US equity ownership held. The Chinese
partner held the remaining percentage. This was
consistent with the primary measure of ownership
structure in the JV literature (Geringer & Hebert,
1989; Parkhe, 1993).
US partner knowledge
Knowledge contribution refers to the relative
di?erences in knowledge between the JV partners
with respect to knowledge of manufacturing and
raw material availability and price; sales and
marketing; factors of JV success; technology,
patents and proprietary processes; and manage-
ment expertise in production and administration.
The ?ve-item scale comprised three items that we
Table 1 (continued)
Variable # Item Response scale
4. Frequent meetings are used to encourage inter-
action between managers and their subordinates
5. Foreign managers perform a coaching role in
developing awareness of business skills, quality
and costs in the JV
6. Virtually all employees understand the factors
that ensure the JVÕs future success
Foreign (Chinese deputy)
general manager decision-
making responsibility
What is the extent to which authority is
delegated to the foreign (Chinese) general
manager from the Board of Directors to make
the following decisions for the joint venture?
1. Development of new products and projects 1 ¼Extremely low
2. The hiring and ?ring of personnel 7 ¼Extremely high
3. Sourcing of inputs (e.g. materials, parts, etc.)
4. Operating procedures and schedules
5. Pricing of products/outputs
6. Distribution of products/outputs
Parent company
communications
1. What is the extent to which you communicate
with the foreign (Chinese) parent while build-
ing the annual operating budget?
1 ¼Little extent
7 ¼Great extent
2. How frequently does informal communication
take place between you and the foreign (Chi-
nese) head o?ce?
Item 2
1 ¼Quarterly, 2 ¼Monthly, 3 ¼Weekly,
4 ¼Daily
3. How often do you report ?nancial data to the
foreign (Chinese) parent?
Items 3 and 4
4. How often do you report non-?nancial data to
the foreign (Chinese) parent (e.g. quality con-
trol reports, labour productivity)
1 ¼Annually, 2 ¼Quarterly,
3 ¼Monthly, 4 ¼Weekly
Manager performance
incentives
1. The extent to which compensation (salary)
contracts clearly specify how compensation is
related to manager performance relative to the
divisionÕs budget
1 ¼Extremely low
2. The extent to which managers whose divisionÕs
performance relative to their budget are among
the top 25% are given larger ?nancial rewards
than those given to managers among the
bottom 25%
7 ¼Extremely high
3. The extent to which manager ?nancial rewards
increase as their actual performance >budgeted
performance
P. Chalos, N.G. O’Connor / Accounting, Organizations and Society 29 (2004) 591–608 599
adapted from the work of Shields and Young
(1993) designed to gauge the relative amount of
information held by each partner in the areas of
production, sales and marketing and JV perfor-
mance capability. Two additional items examined
relative partner expertise in the areas of tech-
nology (knowledge, patents and production pro-
cesses) and management expertise (production
and administration). We used the relative
knowledge contribution by the US JV partner to
maintain the consistency of the hypothesised
relationships across the two samples.
Asset speci?c investments
We developed three items from the work of
Williamson (1985) and measures that were used in
the Yan and Gray (1994) study. These items in-
cluded the amount of alternative JV investments
that were available to the partners before they en-
tered into the JV, the extent to which the invest-
ment was operation or site speci?c (e.g. imported
machinery or special purpose land), and the extent
to which the investment was human resource spe-
ci?c (whether there was a heavy investment in
training and learning). For the last two items,
respondents were asked to indicate the level of their
own asset speci?c investments in the venture.
Controls
Expatriate sta?ng
We measured the proportion of expatriate
(non-expatriate) sta?ng by the percentage of for-
eign (local) to total managerial representatives in
four managerial positions that were found to be
most common in US–Chinese JVs (?nance and
administration, sales and marketing, production
and other).
5
We also measured expatriate sta?ng
by the number of managers, controlling for size
and found no di?erence between the measures. We
treated each managerial function equally for two
reasons. First, apart from the employment of an
expatriate as a general manager position, there is
no empirical study that has found one function to
be more important than another. Second, it is
likely that the JVs may place di?erent emphasis
across the functions depending on situational
factors, such as the speci?c expertise of the foreign
partner (e.g. in production or sales), strategy
(market expansion or export) or industry (Yan &
Luo, 2001). Our measures did not permit the
testing of these factors.
Socialisation practices
Socialisation was measured using six questions
from the 15-item instrument that was developed by
Pascale (1985) and adapted by Chow, Shields, and
Wu (1999). The items that were removed were
those that did not apply to the Chinese environ-
ment. The relevant areas include, training pro-
grammes, foreign head o?ce visits, coaching,
information sharing, meetings and interaction and
understanding organisational goals for success.
Foreign (Chinese deputy) general manager deci-
sion-making responsibility
We asked the US (Chinese) respondent how
much decision-making authority was given to the
US (Chinese deputy) general manager by the
board of directors in six areas of operation. We
adapted the measure from the work of Killing
(1983), who asked managers to assess joint deci-
sion making in product pricing, product design,
production scheduling, the production process,
quality standards, the replacement of a functional
manager, budget sales targets, budget cost targets
and budget capital expenditure. The questionnaire
asked the extent to which the board of directors
delegated authority to the general manager in
terms of product development, the hiring and ?r-
ing of personnel, the sourcing of inputs, operating
procedures, pricing and distribution.
Parent company communications
In measuring communication, we began with
the frequency of communication construct devel-
5
We did not test to see how many locals (e.g. mainland
Chinese) the foreign partner hired. Typically, and especially at
the time of our data collection, most local representatives that
were hired as expatriates were those from other Asian countries
such as Hong Kong and Taiwan. These representatives were
considered to be expatriates in our analysis. Although there
may be degrees of di?erence, the mainland Chinese still perceive
Hong Kong and Taiwanese nationals to be expatriates (i.e. they
do not a?ord them the same trust as they do their own).
600 P. Chalos, N.G. O’Connor / Accounting, Organizations and Society 29 (2004) 591–608
oped by Parkhe (1993) and modi?ed it according
to the feedback that we received during the site
visits. This feedback provided the basis for the
design of questions about the frequency of com-
munication between the JV and the parent ?rms in
four areas: budget development, informal com-
munication, operational reporting and ?nancial
reporting.
Manager performance incentives
Functional manager performance incentives
were measured using three questions that were
adapted from the work of Shields and Young
(1993). These questions asked about the extent to
which compensation was related to managerial
performance; the extent to which managers in the
top 25% of performers were given larger rewards
than those in the bottom 25%; and the extent to
which ?nancial rewards increased as actual per-
formance exceeded budgeted performance. The
three items capture the concept of the extent to
which incentives are linked to budget performance
and have been used as part of a measure of per-
formance rewards in the multinational subsidiary
context (Chow et al., 1999).
Respondent sample
The primary source of ?rm selection for the
survey of US JV managers was the American
Chamber of Commerce Directory of companies
that do business in China (China Statistical Press,
1996). We targeted 320 companies and made
telephone calls to all of the companiesÕ o?ces in
Hong Kong to con?rm whether they had a man-
ufacturing JV in China. The application of these
criteria generated a data set of 242 ?rms with JVs
that were mainly located in Beijing and Shanghai.
The questionnaires were distributed through the
Hong Kong representative o?ces of the US part-
ners. A monetary incentive of HK $50 (approxi-
mately US$8) accompanied each survey, along
with a sponsoring letter from the chief ?nancial
controller in the Hong Kong representative o?ce.
Extensive follow-up was undertaken, in which all
of the JV representative o?ces in Hong Kong were
contacted by telephone during a six-month period
in order to check for the completion of the survey.
A total of 242 questionnaires were mailed and 118
were returned, a response rate of 49%. One of the
questionnaires was incomplete and thus discarded,
which left a sample of 117 responses for analysis.
As it was not possible to pair match the
respondents, we drew the Chinese sample from a
list of US–Chinese JVs at the Shanghai Univer-
sity of Finance and Economics (SUFE).
6
The
list included more than 500 manufacturing JVs
that were targeted with the collaboration of two
management professors at SUFE. To obtain
reliable survey data from the Chinese managers,
we followed several procedures, as suggested by
site visit interviewees and the China management
literature (Adler, Campbell, & Laurent, 1989).
7
We removed all references to the foreign source
of the project. The cover letters and return
envelopes indicated that this was a China project
undertaken by a management professor at
SUFE. A sponsoring letter of approval was ob-
tained from MOFTEC and attached to each
survey, and a monetary incentive of 30 Yuan
(approximately US$4) accompanied each ques-
tionnaire. Of the 500 questionnaires that we
mailed, 160 were returned for a response rate of
32%. Fifteen of the questionnaires were incom-
plete and thus discarded, reducing the sample to
145 responses.
6
The names of Chinese deputy general managers were not
available for the US–Chinese JVs that were listed in the
American Chamber of Commerce Directory. Hence, we would
have needed the foreign manager to pass on the survey to the
Chinese manager for completion, thus violating the anonymity
required to obtain a reliable response from the Chinese
manager, likely biasing responses.
7
Apart from language and the relative newness of manage-
ment research in China, factors such as uncertainty about the
sharing of information with strangers (which indicates that
there are remnants of paranoia from the Cultural Revolution)
and uncertainty over Communist Party support for foreign
academic activities in China have fostered a lack of trust in
foreign academic research. In addition, a discussion with a
China management researcher convinced us that having some
level of government approval would facilitate the completion of
questionnaires by local managers (Yadong Luo, personal
conversation, 1998).
P. Chalos, N.G. O’Connor / Accounting, Organizations and Society 29 (2004) 591–608 601
Results
Sample validity and reliability checks
The survey instrument included several cate-
gorical and descriptive questions, in addition to
nine parts with responses anchored on seven-item
Likert scales. To ensure external validity, we de-
signed the instrument according to previous survey
and case study results from the JV and manage-
ment control literature, as well as observations
from our JV site visits. The average date of the
formation of the JVs was 1992 for both samples,
which meant that the mean age of the ventures was
approximately ?ve years when we conducted the
survey. The equity split was, on average, 60% US
and 40% Chinese for both samples. Firm size
ranged from 40 to several thousand employees.
Firms were spread across a spectrum of industries
in both samples.
The JVs in the US sample averaged 328
employees and those in the Chinese sample
averaged 498 employees. The absolute value of
the investment in each ?rm ranged from US$6
million to US$28 million.
8
Both groups had, on
average, several years of experience with other
JV partners, and both expected a 4–6 year pay-
back period on their investments. As prior JV
relationships with other partners in developing
nations have been found to reduce the foreign
partnerÕs dependence on the local partner for
information (Chalos et al., 1999), partner expe-
rience was examined as a covariate. None of the
hypothesised results di?ered when the covariate
was included. Analyses were conducted within
each sample, controlling for di?erences between
mean responses on the same variable in the two
respondent groups.
We examined survey item reliability in several
ways. One reliability check involved CronbachÕs
(1951) alpha coe?cient. To test construct reli-
ability, we calculated alpha coe?cients for each
set of surveyed items that comprised the mea-
sured variables. As can be seen in Table 2, all of
the alpha coe?cients were above 0.65, a rea-
sonable internal reliability score (Carmines &
Zeller, 1979). A second reliability check involved
con?rmatory factor analysis. Con?rmatory rather
than exploratory factor analysis was used be-
cause each of the variables was based on prior
constructs and observational data from our site
visits. We compiled the results of the eigenvalue
loadings for all of the model variables from a
rotated varimax factor structure of the survey
data.
The constructs all had eigenvalues greater
than one, corroborating the alpha ?ndings. To-
gether the factors explained 65.3% of the vari-
ance for the US managers and 65.2% of the
variance for the Chinese managers. All of the
loadings were greater than 0.40 and quantita-
tively similar across the constructs for each
respondent group, indicating high respondent
convergence (Harman, 1967). Table 2 shows the
descriptive statistics and correlation matrix of all
of the variables for each sample group. Corre-
lations between the independent variables were
uniformly low for both groups. Variance in?a-
tion and eigenvalue scores of independent vari-
able multicollinearity were low and are reported
in Table 2 (Kennedy, 1992).
Tests of hypotheses
Table 2 includes alpha coe?cients, means and
bivariate correlations for all of the variables for
the two respondent samples. To explore the in?u-
ence of equity ownership, partner knowledge and
speci?c asset investments on management con-
trols, we ran OLS regressions of the three inde-
pendent variables on each control for each
respondent sample (10 regressions in all: 5 controls
à 2 respondent samples). This enabled us to test
our hypotheses by examining the signi?cance of
the slope of each of the three independent vari-
ables against each of the ?ve controls for the two
8
In 1996, the average contractual commitment of all foreign
enterprises in China was US$3.02 million (China Statistical
Yearbook, 2001). While the value of investment in our sample
was larger than the average, it was signi?cantly less than the
investments that were made by the largest US ?rms in China.
Luo and OÕConnor (1998) reported that the top ten US ?rms in
China invested US$3.6 billion across 70 JVs, with an average
investment of US$50 million.
602 P. Chalos, N.G. O’Connor / Accounting, Organizations and Society 29 (2004) 591–608
respondent samples. This indicated the association
between each control and equity ownership, part-
ner knowledge and speci?c asset investments
within each group. It also enabled testing the sig-
ni?cance of the di?erences between the slopes of
similar relationships across the two samples. We
examined the following relationships for each
sample.
Y
1ÁÁÁ5
¼ b
0
þ b
1
v
1
þ b
2
v
2
þ b
3
v
3
þ e
Ã
ð1Þ
where Y
1ÁÁÁ5
represented management controls
and v
1ÁÁÁ3
represented US equity ownership, US
partner knowledge and asset speci?c invest-
ments.
9
All of the regressions were signi?cant on all
controls for the two respondent groups (see Table
3).
10
Hypothesis one (H1) posited that expatriate
sta?ng, a cultural control, would be positively
associated with US equity ownership. This was
supported. Both US and Chinese respondents
perceived expatriate sta?ng to be positively asso-
ciated with equity ownership (US b ¼ 0:413;
p < 0:01; Chinese b ¼ 0:175; p < 0:05).
11
Consis-
tent with the ?rst hypothesis, equity ownership
was not associated with any other control mech-
anisms from the perspective of both partners.
Neither partner perceived equity ownership to be
su?cient justi?cation for the adoption of a
T
a
b
l
e
2
D
e
s
c
r
i
p
t
i
v
e
s
t
a
t
i
s
t
i
c
s
a
n
d
c
o
r
r
e
l
a
t
i
o
n
m
a
t
r
i
x
o
f
v
a
r
i
a
b
l
e
s
U
S
(
C
h
i
n
e
s
e
)
r
e
s
p
o
n
d
e
n
t
s
F
o
r
e
i
g
n
g
e
n
e
r
a
l
m
a
n
a
g
e
r
s
(
n
¼
1
1
7
;
R
a
b
o
v
e
d
i
a
g
o
n
a
l
)
,
C
h
i
n
e
s
e
d
e
p
u
t
y
m
a
n
a
g
e
r
s
(
n
¼
1
4
5
;
R
b
e
l
o
w
d
i
a
g
o
n
a
l
)
M
e
a
n
s
R
a
n
g
e
A
l
p
h
a
E
q
u
i
t
y
K
n
o
w
.
A
s
s
e
t
E
x
p
a
t
.
S
o
c
i
a
l
.
D
e
c
Õ
n
C
o
m
m
Õ
n
I
n
c
e
n
t
i
v
e
s
U
S
e
q
u
i
t
y
o
w
n
e
r
s
h
i
p
(
%
)
6
2
.
6
2
(
6
1
.
0
3
)
0
–
1
0
0
–
–
0
.
3
0
5
Ã
Ã
Ã
0
.
0
9
3
0
.
4
6
8
Ã
Ã
Ã
)
0
.
0
4
4
0
.
1
0
2
0
.
1
3
9
0
.
0
7
5
U
S
p
a
r
t
n
e
r
k
n
o
w
l
e
d
g
e
2
8
.
5
8
(
1
9
.
7
0
)
Ã
Ã
5
–
3
5
0
.
7
0
(
0
.
7
4
)
0
.
4
7
0
Ã
Ã
Ã
–
)
0
.
1
5
7
Ã
0
.
3
2
4
Ã
Ã
0
.
3
3
0
Ã
Ã
Ã
0
.
3
3
8
Ã
Ã
Ã
0
.
2
5
6
Ã
Ã
0
.
2
2
4
Ã
A
s
s
e
t
s
p
e
c
i
?
c
i
n
v
e
s
t
m
e
n
t
s
1
1
.
7
7
(
1
1
.
7
9
)
3
–
2
1
0
.
6
5
(
0
.
7
7
)
)
0
.
0
5
2
0
.
2
0
0
Ã
Ã
–
0
.
1
3
7
0
.
1
9
3
Ã
Ã
0
.
0
7
4
0
.
1
0
4
0
.
2
7
8
Ã
Ã
E
x
p
a
t
r
i
a
t
e
s
t
a
?
n
g
1
3
2
.
2
4
(
1
0
9
.
7
8
)
0
–
4
0
0
0
.
6
7
(
0
.
8
0
)
0
.
3
9
4
Ã
Ã
Ã
0
.
5
1
4
Ã
Ã
Ã
0
.
2
3
7
Ã
–
0
.
3
1
9
Ã
Ã
0
.
3
0
2
Ã
0
.
0
9
9
0
.
2
5
5
Ã
S
o
c
i
a
l
i
s
a
t
i
o
n
p
r
a
c
t
i
c
e
s
3
0
.
4
3
(
2
4
.
1
4
)
Ã
Ã
6
–
4
2
0
.
7
9
(
0
.
8
6
)
0
.
0
7
6
)
0
.
0
1
6
0
.
5
5
2
Ã
Ã
Ã
0
.
1
5
9
–
0
.
2
2
5
Ã
0
.
2
6
7
Ã
Ã
0
.
5
3
8
Ã
Ã
Ã
D
e
c
i
s
i
o
n
-
m
a
k
i
n
g
r
e
s
p
o
n
s
i
b
i
l
i
t
y
3
3
.
3
8
(
2
0
.
7
4
)
Ã
Ã
6
–
4
2
0
.
8
2
(
0
.
9
4
)
)
0
.
3
6
1
Ã
Ã
Ã
)
0
.
6
8
9
Ã
Ã
Ã
0
.
3
7
7
Ã
Ã
Ã
)
0
.
4
1
5
Ã
Ã
Ã
0
.
2
3
7
Ã
Ã
–
)
0
.
0
3
6
0
.
2
5
4
Ã
Ã
P
a
r
e
n
t
c
o
m
p
a
n
y
c
o
m
m
u
n
i
c
a
t
i
o
n
s
1
3
.
8
8
(
1
0
.
8
7
)
Ã
4
–
1
9
0
.
6
6
(
0
.
7
6
)
0
.
0
4
9
0
.
0
0
0
0
.
4
8
9
Ã
Ã
Ã
0
.
1
1
8
0
.
6
0
7
Ã
Ã
Ã
0
.
3
3
0
Ã
Ã
Ã
–
0
.
4
3
1
Ã
Ã
Ã
M
a
n
a
g
e
r
p
e
r
f
o
r
m
a
n
c
e
i
n
c
e
n
t
i
v
e
s
1
1
.
9
7
(
1
2
.
6
7
)
3
–
2
1
0
.
8
2
(
0
.
8
8
)
0
.
0
2
4
)
0
.
1
0
5
0
.
4
1
5
Ã
Ã
Ã
0
.
2
4
9
Ã
0
.
3
8
6
Ã
Ã
Ã
0
.
3
5
1
Ã
Ã
Ã
0
.
1
7
9
Ã
–
S
i
g
n
i
?
c
a
n
c
e
o
f
t
-
s
t
a
t
i
s
t
i
c
f
o
r
i
n
d
e
p
e
n
d
e
n
t
s
a
m
p
l
e
m
e
a
n
s
t
-
t
e
s
t
a
n
d
P
e
a
r
s
o
n
b
i
v
a
r
i
a
t
e
c
o
r
r
e
l
a
t
i
o
n
s
:
Ã
p
<
0
:
0
5
,
Ã
Ã
p
<
0
:
0
1
,
Ã
Ã
Ã
p
<
0
:
0
0
1
(
t
w
o
-
t
a
i
l
e
d
t
e
s
t
)
.
T
h
e
o
r
e
t
i
c
a
l
r
a
n
g
e
s
h
o
w
n
.
T
h
e
a
c
t
u
a
l
r
a
n
g
e
w
a
s
t
h
e
s
a
m
e
a
s
t
h
e
t
h
e
o
r
e
t
i
c
a
l
r
a
n
g
e
f
o
r
a
l
l
v
a
r
i
a
b
l
e
s
e
x
c
e
p
t
U
S
e
q
u
i
t
y
o
w
n
e
r
s
h
i
p
(
5
–
9
5
)
,
a
s
s
e
t
s
p
e
c
i
?
c
i
n
v
e
s
t
m
e
n
t
s
(
3
–
1
9
)
a
n
d
m
a
n
a
g
e
r
p
e
r
f
o
r
m
a
n
c
e
i
n
c
e
n
t
i
v
e
s
(
3
–
2
0
)
.
9
Post hoc comparisons of interaction e?ects were found to
be either small or insigni?cant, and did not change the
interpretation of the main-e?ects (reported in Table 3).
10
In the regression equations, the adjusted R
2
was signi?-
cantly higher for the local respondents than it was for the US
respondents. Chow tests con?rmed the signi?cance of these
di?erences (see Table 3, Panel B).
11
We also measured the proportion of the board of directors
representing the foreign partner. Child and Yan (1999) model
this construct as mediating the relationship between equity
ownership and the employment of expatriates. This relationship
was evident. US equity was signi?cantly correlated with foreign
board percentage for both the foreign (0.846; p < 0:01) and the
Chinese (0.325; p < 0:01) samples. The foreign board percent-
age in turn was signi?cantly correlated with expatriates for both
the foreign (0.333; p < 0:01) and the Chinese (0.603; p < 0:01)
samples. This is also consistent with the theory for H1.
P. Chalos, N.G. O’Connor / Accounting, Organizations and Society 29 (2004) 591–608 603
broader set of behavioural, outcome and cultural
controls.
Hypothesis two (H2) predicted that perceptions
of higher US partner knowledge relative to the
Chinese partner would be positively related to
cultural, behavioural and outcome controls for
both partners. For the US respondents, four out of
the ?ve controls were signi?cant in the expected
direction (socialisation practices b ¼ 0:364;
p < 0:01; foreign decision-making responsibility
b ¼ 0:329; p < 0:01; parent company communi-
cations b ¼ 0:220; p < 0:10; and manager perfor-
mance incentives b ¼ 0:178; p < 0:10). These
?ndings supported hypotheses H2b, H2c, H2d
and H2e for US respondents. For the Chinese
respondents, two of the ?ve controls were signi?-
cant and positive (expatriate sta?ng b ¼ 0:221;
p < 0:01 and foreign decision-making responsibil-
ity, as shown by the negative beta for Chinese
decision-making responsibility b ¼ À0:619;
p < 0:01). These ?ndings support H2a and H2c for
Chinese respondents. T-tests of the bs on each
control between the two samples indicated that the
US partner perceived socialisation practices
(t ¼ À3:030; p < 0:01), foreign decision-making
responsibilities (t ¼ 1:634; p < 0:10) and manager
performance incentives (t ¼ À2:129; p < 0:05) to
be signi?cantly greater than did the Chinese part-
ner, supporting H3b, H3c and H3e.
12
Hypothesis four (H4) proposed that asset
speci?c investments would be positively associ-
ated with the use of controls for the Chinese
partner. Four of the ?ve controls were signi?cant
and in the expected direction for the Chinese
respondents (socialisation practices b ¼ 0:561;
p < 0:01; Chinese decision-making responsibility
b ¼ 0:277; p < 0:01; parent company communi-
Table 3
Standardized OLS regression coe?cients of equity ownership, partner knowledge and asset speci?c investments on individual controls:
US (Chinese) respondents
Expatriate sta?ng Socialisation
practices
Decision-making
responsibility
Parent company
communications
Performance
incentives
Panel A: standardized regression coe?cients
US equity ownership 0.413
ÃÃÃ
(0.175
ÃÃ
) )0.194
ÃÃ
(0.092) 0.019 ()0.055) 0.089 (0.050) )0.017 (0.083)
US partner knowledge 0.080 (0.221
ÃÃÃ
) 0.364
ÃÃÃ
(0.029) 0.329
ÃÃÃ
()0.619
ÃÃÃ
) 0.220
Ã
(0.055) 0.178
Ã
()0.080)
Asset speci?c investments )0.108 (0.114) 0.127 (0.561
ÃÃÃ
) 0.013 (0.277
ÃÃÃ
) 0.052 (0.500
ÃÃÃ
) 0.247
ÃÃÃ
(0.407
ÃÃÃ
)
Adjusted R
2
0.175
ÃÃÃ
(0.101
ÃÃÃ
) 0.134
ÃÃÃ
(0.302
ÃÃÃ
) 0.089
ÃÃÃ
(0.541
ÃÃÃ
) 0.049
ÃÃ
(0.231
ÃÃÃ
) 0.089
ÃÃÃ
(0.162
ÃÃÃ
)
Panel B: t-statistics of di?erences in standardized regression coe?cients between US and Chinese respondents
US partner knowledge )0.420 )3.030
ÃÃÃ
1.634
Ã
1.252 )2.129
ÃÃ
Asset speci?c investments )1.860
Ã
4.232
ÃÃÃ
)3.835
ÃÃÃ
4.831
ÃÃÃ
0.797
Chow F statistic (d.f. 8, 255)
: b/w US and Chinese
regressions
1.478 15.686
ÃÃÃ
12.021
ÃÃÃ
14.812
ÃÃÃ
1.839
Ã
Signi?cance levels:
Ã
p < 0:10,
ÃÃ
p < 0:05,
ÃÃÃ
p < 0:01 (two-tailed test).
Tests of multicollinearity: a high variance in?ation factor (VIF) of the regression coe?cient upon the error term is one indication of
multicollinearity. ÔFor standardized data, VIF
I
> 10 indicates harmful collinearityÕ (Kennedy, 1992, p. 183). The VIF factors ranged
from 1.002–1.919. The condition indices (CIs) ranged from 2.094–13.582 between antecedents, below the moderate dependency range
suggested by Belsley (1991, p. 56).
12
We ran various sensitivity tests on the signi?cance of the
regressions, by including size (# of employees) and JV duration
and found the signi?cance of the coe?cients to be unchanged.
We tested other formulas for the measure of expatriate sta?ng
to provide more assurance of our ?ndings. For example, we
tested the same model using a measure of the absolute numbers
of expatriates with size included. The results were unchanged.
When the percentage of expatriate sta?ng in each functional
category was entered in the regression model, the US partner
knowledge coe?cient was signi?cant for production and sales
(Chinese respondents only), but was insigni?cant for the
percentage of expatriate sta?ng in the ?nance and administra-
tion category. The regression model for the percentage of
expatriates in the ‘‘other’’ category was not signi?cant. While
these results admittedly are ad hoc, it is plausible that the
consistent results found for only manufacturing and sales
managers could re?ect the fact that these functions are areas of
greater foreign partner expertise.
604 P. Chalos, N.G. O’Connor / Accounting, Organizations and Society 29 (2004) 591–608
cations b ¼ 0:500; p < 0:01; and manager per-
formance incentives b ¼ 0:407; p < 0:01). This
supports H4b, H4c, H4d and H4e. Hypothesis
four (H4) predicted that perceptions of asset
speci?c investments would be positively related to
the use of management controls for the US
partner as well. Only one of the ?ve controls was
found to be signi?cant (manager performance
incentives b ¼ 0:247; p < 0:01), supporting H4e.
T-tests of b di?erences on controls between Chi-
nese and US respondents were signi?cant in the
expected direction for Chinese partners on three
of the ?ve controls. As speci?c asset investments
increased, Chinese respondents perceived signi?-
cantly more socialisation practices (t ¼ 4:232;
p < 0:01), more decision-making responsibility
for the Chinese deputy general manager (t ¼
À3:835; p < 0:01), and greater parent company
communications (t ¼ 4:831; p < 0:01) than did
the US respondents. This supports H5b, H5c and
H5d.
Discussion
Recent observations from ?eld studies of US–
Chinese JVs have led to propositions regarding
the determinants of JV control mechanisms. We
framed the propositions as testable hypotheses of
the e?ects of partner equity ownership, knowledge
dependence and asset speci?c investments on JV
controls. Controls suggested by the joint venture
literature have included expatriate sta?ng, social-
isation practices, delegated decision responsibili-
ties, parent company communications and
incentives. We examined the hypothesised control
determinants from the perspective of both JV
partners. Based on ?eld visits and survey data, we
found that partner equity ownership in?uenced
expatriate sta?ng but none of the other control
mechanisms. Relative to partner knowledge
dependence and speci?c asset investments, bar-
gaining dominance conferred by majority equity
ownership appeared to o?er little explanation of
JV control mechanisms. This ?nding concurs with
Mjoen and Tallman (1997), who found the role of
equity as a determinant of control to be Ôhighly
questionableÕ and who suggested instead that
controls over JV activities seemed to be more
descriptive of actual JV control mechanisms.
Consistent with evidence of motives in JV
partner selection (Hitt et al., 2000), we found
knowledge dependency and asset speci?c transac-
tion costs to be determinants of controls to varying
degrees for each partner. Beamish Delios (2001),
and Dyer Singh (1998), argue that intangible
partner knowledge and speci?c asset contributions
determine the control mechanisms of the venture.
Our results partly supported these propositions
and others developed from recent case studies
(Child & Faulkner, 1998; Yan & Luo, 2001). The
descriptive results indicated that US partner knowl-
edge was perceived to be signi?cantly greater than
that of the Chinese partner from both partnersÕ
perspectives. Statistically, US partner knowledge
relative to the Chinese partner in?uenced a broad
set of controls, suggesting that partner knowledge
has some bearing upon the design of the controls
used in the venture. The US partner perceived
socialisation practices, foreign manager decision
responsibilities, parent company communications
and manager performance incentives to be e?ective
control mechanisms that were positively related to
the di?usion of US knowledge. As US knowledge
increased, the Chinese partner also preferred
greater foreign manager decision responsibilities as
well as greater expatriate sta?ng. This suggests
that the Chinese partner was willing to give up
decision control in cases where a knowledge de?-
ciency existed.
In contrast to the US partner, however the
Chinese partner did not perceive any positive
relationships between US knowledge and sociali-
sation practices, parent company communications
or manager performance incentives. These strong
perceptual di?erences indicate fundamental dis-
agreements between the partners with respect to
the e?cacy of several controls in transmitting
partner knowledge and improving learning in the
alliance. It appears critical to determine: what
knowledge each partner can and is willing to
contribute to the JV; what strategic motives
underlie this knowledge contribution; and whether
this contribution makes each partner better o? in
the long run. Conceivably, these control di?er-
ences were deliberate on the part of the US
P. Chalos, N.G. O’Connor / Accounting, Organizations and Society 29 (2004) 591–608 605
partner, who was perhaps reluctant to share cer-
tain knowledge that might eventually be used by
the Chinese partner as a JV rival. Alternatively,
the controls may simply not have been put to
optimal use from the perspective of the local
partner in disseminating knowledge.
Speci?c asset investments had signi?cant e?ects
on the promulgation of management controls
from the viewpoint of the Chinese partner, but to a
lesser extent from the viewpoint of the US partner.
Investments in speci?c assets by the Chinese
partner were positively and strongly associated
with manager performance incentives, socialisa-
tion, parent company communications, and more
Chinese manager decision-making responsibilities.
The use of these controls was perceived to increase
as Chinese partner speci?c asset investments in-
creased, thus supporting the evidence for transac-
tion cost minimisation through JV controls found
by Mjoen and Tallman (1997). US partners in
contrast perceived only manager performance
incentives to be related to increases in their speci?c
asset investments, suggesting that the foreign
partner had little con?dence in control mecha-
nisms to protect their speci?c asset investments.
Knowledge dependence and speci?c asset
investments mirror the di?erent contributions,
goals and motives of each partner. Since the
Chinese partner is knowledge dependent on the
foreign partner, control mechanisms must be ?nely
balanced between cooperative learning for the
good of the venture and competitive learning that
threatens the ventureÕs existence. Similarly, since
the US partner is dependent upon the speci?c as-
sets contributed by the Chinese partner, control
mechanisms must be ?nely balanced between
maintaining local relationships for the good of the
venture and fostering relationships that encourage
partner defection. The di?erent partner percep-
tions of controls that were found are an indication
of perhaps unavoidable control misalignment be-
tween the partners. Partner di?erences in their
perceptions of control mechanisms are determined
largely by their respective motives. This represents
an inherent source of instability within the JV
(Child & Faulkner, 1998; Yan & Luo, 2001). The
design of the JV control system serves to manage
but not eliminate both collaborative and compet-
itive forces that are simultaneously at work within
the alliance.
The limitations of this study must be considered
when interpreting the signi?cance of its conclu-
sions. Response anonymity did not permit pair-
wise matching of US–Chinese responses by ?rm.
Pairwise matching of speci?c JV partners would
have minimised inter-?rm variability. Statistical
checks indicated that the US and Chinese
respondent samples did not signi?cantly di?er in
size, industry range, investment, or alliance dura-
tion. Response bias between the two samples re-
mains a possibility. Similarities or di?erences
found in US–Chinese JVs are not necessarily
generalisable to other JV environments. Finally,
although the survey metrics were based upon
extensive site visits and the current literature,
important variables may have been omitted. Fu-
ture research should corroborate these ?ndings
with additional ?eld studies in other JV environ-
ments, incorporating other antecedents and con-
trols. Exploring the relation of controls to JV
instability and performance would also be a
worthwhile extension of the present study.
Acknowledgements
We wish to thank participants of the Interna-
tional Accounting Conference, Cambridge Uni-
versity, 2000, the AAANZ Conference (Cairns,
Australia, 1999) and seminars at the City Univer-
sity of Hong Kong (2001), the Hong Kong Uni-
versity of Science and Technology (2003) and the
University of Illinois at Chicago (2002). We are
especially grateful for the comments that we re-
ceived from Chee Chow, Yadong Luo, Michael
Firth, Maris Martinsons, Yanni Yan and two
anonymous reviewers. The work described in this
paper was substantially supported by a grant from
the Research Grants Council of the Hong Kong
Special Administrative Region, China. [Project
No. City U 9040381].
References
Adler, N. J., Campbell, N., & Laurent, A. (1989). In search of
appropriate methodology: From outside the PeopleÕs
606 P. Chalos, N.G. O’Connor / Accounting, Organizations and Society 29 (2004) 591–608
Republic of China looking in. Journal of International
Business Studies, 20(1), 61–74.
Baliga, B. R., & Jaeger, A. M. (1984). Multinational corpora-
tion: Control systems and delegation issues. Journal of
International Business Studies (Fall), 25–40.
Bauer, T. N., Morrison, E. W., & Callister, R. R. (1998).
Organizational socialization: A review and directions for
future research. Research in Personnel and Human Resources
Management, 16, 149–214.
Beamish, P. W., & Banks, J. C. (1987). Equity joint ventures
and the theory of the multinational enterprise. Journal of
International Business Studies (Summer), 1–16.
Beamish, P., & Delios, A. (2001). Survival and pro?tability:
The roles of experience and intangible assets in foreign
subsidiary performance. Academy of Management Journal,
44, 1028–1039.
Beamish, P. W., & Lu, J. (2001). Internationalization and
Performance of SMEs. Strategic Management Journal,
22(June/July), 565–586.
Belsley, D. A. (1991). Conditioning diagnostics: Collinearity and
weak data in regression. New York: John Wiley.
Carmines, E. G., & Zeller, R. A. (1979). Reliability & validity
assessment. Sage University Paper Series on Quantitative
Applications in the Social Sciences, 7-017. Beverly Hills,
CA: Sage.
Chalos, P., OÕConnor, N. G., & Xu, Z. (1999). A comparative
study of budgetary controls in Chinese state and foreign
owned enterprises. Advances in Management Accounting, 7,
169–185.
Chang, E., & Taylor, M. S. (1999). Control in multinational
corporations: The case of Korean manufacturing subsidi-
aries. Journal of Management, 25(4), 541–565.
Child, J. (1994). Management in China during the age of reform.
Cambridge: Cambridge University Press.
Child, J., & Faulkner, D. (1998). Strategies of cooperation:
Managing alliances, networks and joint ventures. London:
Oxford University Press.
Child, J., & Yan, Y. (1999). Investment and control in
international joint ventures: The case of China. Journal of
World Business, 24(1), 3–15.
Child, J., Yan, Y., & Ku, Y. (1997). Ownership and control in
sino-foreign joint ventures. In P. W. Beamish & P. J. Killing
(Eds.), Cooperative strategies: Asian perspectives (pp. 181–
225). San Francisco: New Lexington Press.
China Statistical Press (1996). In American Chamber of
Commerce (Ed.), American companies in Hong Kong.
Torrance, CA: Caravel, Inc.
Chow, C. W., Shields, M. D., & Wu, A. (1999). The
Importance of National Culture in the design of and
preference for management controls for multi-national oper-
ations. Accounting, Organizations and Society, 24, 441–
461.
Contractor, F. J., & Lorange, P. (1988). The strategy
and economics basis for cooperative ventures. In F. J.
Contractor & P. Lorange (Eds.), Cooperative strategies in
international business (pp. 3–28). Lexington: Lexington
Books.
Cooper, H., & Johnson, I. (2000). CongressÕs vote primes ?rms
in U.S. to boost China deals. Asian Wall Street Journal, 26–
28 May 2000, pp. 1, 7.
Cronbach, L. J. (1951). Coe?cient alpha and internal structure
of tests. Psychometrika (September), 297–334.
Dang, T. (1977). Ownership, control and performance of the
multinational corporation: A study of U.S. wholly-owned
subsidiaries and joint ventures in the Philippines and Taiwan.
Unpublished doctoral dissertation, University of California,
Los Angeles.
Dayao, D. (Ed.) (1997). How do you train 10,000 Chinese?
World Executive’s Digest (January), p. 16.
Dekker, H. C. (2004). Control of inter-organizational relation-
ships: evidence on appropriation concerns and coordination
requirements. Accounting, Organizations and Society, 29(1),
27–49.
Dyer, J. H., & Singh, H. (1998). The relational view: Cooper-
ative strategy and sources of interorganizational competitive
advantage. Academy of Management Review, 23(4), 660–
679.
Eisenhardt, K. M. (1985). Control: Organizational and
economic approaches. Management Science, 31(2), 134–
149.
Firth, M. (1996). The di?usion of managerial accounting
procedures in the PeopleÕs Republic of China and the
in?uence of foreign partnered joint ventures. Accounting,
Organizations and Society, 21(7/8), 629–654.
Geringer, J. M. (1986). Criteria for selecting partners for joint
ventures in industrialized market economies. Ph.D. disserta-
tion, University of Washington, Seattle.
Geringer, J. M., & Frayne, C. A. (1990). Human resource
management and international joint venture control: A
parent company perspective. Management International
Review, 30(Special issue), 103–120.
Geringer, J. M., & Hebert, L. (1989). Control and performance
of international joint ventures. Journal of International
Business Studies, 20(2), 235–254.
Gupta, A. K., & Govindarajan, V. (2000). Knowledge ?ows
within multinational corporations. Strategic Management
Journal, 21, 473–496.
Groot, L. C. M., & Merchant, K. A. (2000). Control of
international joint ventures. Accounting, Organizations and
Society, 25, 579–607.
Harman, H. H. (1967). Modern factor analysis. Chicago:
University of Chicago Press.
Hennart, J. (1991). The transaction cost theory of joint
ventures: an empirical study of Japanese subsidiaries in
the United States. Management Science, 37, 483–
497.
Hitt, M. A., Dacin, T. M., Levitas, E., Arregle, J. L., & Borza,
A. (2000). Partner selection in emerging and developed
market contexts: Resource-based and organizational learn-
ing perspectives. Academy of Management Journal, 43(3),
449–467.
Inkpen, A. (1995). The management of international joint
ventures: An organizational learning perspective. London:
Routledge.
P. Chalos, N.G. O’Connor / Accounting, Organizations and Society 29 (2004) 591–608 607
Inkpen, A., & Beamish, P. W. (1997). Knowledge, bargaining
power and the instability of international joint ventures. The
Academy of Management Review, 22(1), 177–202.
Inkpen, A., & Crossan, M. M. (1995). Believing is seeing: Joint
ventures and organization learning. Journal of Management
Studies, 32, 595–618.
Isobe, T., Makino, S., & Montgomery, D. B. (2000). Resource
commitment, entry timing, and market performance of
foreign direct investments in emerging economies: The case
of Japanese international joint ventures in China. Academy
of Management Journal, 43(3), 468–484.
Kennedy, P. (1992). A guide to econometrics (3rd ed.).
Cambridge, MA: MIT Press.
Khanna, T., Gulati, R., & Nohria, N. (1998). The dynamics of
learning alliances: Competition, cooperation and relative
scope. Strategic Management Journal (March).
Killing, J. P. (1983). Strategies for joint venture success. New
York: Praeger.
Kogut, B. (1988). Joint ventures: Theoretical and empirical
perspectives. Strategic Management Journal, 9(4), 319–332.
Lane, P., & Lubatkin, M. (1998). Relative absorptive capacity
and inter-organizational learning. Strategic Management
Journal, 19(5), 461–478.
Luo, Y., & OÕConnor, N. G. (1998). Structural changes of
foreign direct investment in China: An evolutionary per-
spective. Journal of Applied Management Studies, 7(1), 74–
89.
Luo, Y., Shenkar, O., & Nyaw, M.-K. (2001). A dual parent
perspective on control and performance in international
joint ventures: Lessons from a developing economy. Journal
of International Business Studies, 32(1), 41–58.
Lyles, M. A., & Salk, J. E. (1996). Knowledge acquisition from
foreign parents in international joint ventures: An empirical
examination in the Hungarian context. Journal of Interna-
tional Business Studies, 27, 877–903.
Merchant, K. A. (1998). Modern management control systems:
Text & cases. London: Prentice-Hall International.
Mjoen, H., & Tallman, S. (1997). Control and performance in
international joint ventures. Organization Science, 8(3), 257–
274.
Naughton, B. (1995). Growing out of the plan: Chinese economic
reform, 1978–1993. New York: Cambridge University Press.
Nonaka, I., & Takeuchi, H. (1995). The knowledge-creating
company. New York: Oxford University Press.
Ouchi, W. G. (1980). Markets, bureaucracies, and clans.
Administrative Science Quarterly, 25, 129–141.
Park, S. H., & Luo, Y. (2001). Guanxi and organisational
dynamics: Organisational networking in Chinese ?rms.
Strategic Management Journal, 22(5), 455–477.
Parkhe, A. (1993). Strategic alliance structuring: A game
theoretic and transaction cost examination of inter-?rm
cooperation. Academy of Management Journal, 36(4), 794–
829.
Pascale, R. (1985). The paradox of ‘‘corporate culture’’:
Reconciling ourselves to socialization. California Manage-
ment Review, 27, 26–41.
PeopleÕs Daily. 2003. Overseas investment to China reaches
US$52.7 billion. Retrieved 16th June 2003, from http://
english.peopledaily.com.cn/home.shtml.
Schaan, J. L. (1983). Parent control and joint venture success:
The case of Mexico. Unpublished doctoral dissertation.
University of Western Ontario.
Steensma, H. K., & Lyles, M. A. (2000). Explaining IJV
survival in a transitional economy through social exchanged
and knowledge-based theories. Strategic Management Jour-
nal, 21, 831–851.
Shields, M., & Young, S. M. (1993). Antecedents and conse-
quences of participative budgeting: Evidence on the e?ects
of asymmetrical information. Journal of Management
Accounting Research (Fall), 265–280.
State Statistical Bureau (2002). China Statistical Yearbook
2002. Beijing: China Statistical Press.
Su, Z. (1999). Successful managers in international joint-
ventures in China. The International Scope Review, 1(1),
1–22.
Tiemessen, I., Lane, H. W., Crossan, M. M., & Inkpen, A. C.
(1997). Knowledge management in international joint ven-
tures. In P. W. Beamish & J. P. Killing (Eds.), Cooperative
strategies: North American perspective (pp. 370–399). San
Francisco: New Lexington Press.
Weldon, E., & Vanhonacker, W. (1999). Operating a foreign-
invested enterprise in China: Challenges for managers and
management researcher. Journal of World Business, 34(1),
94, Spring 99, 14p.
Williamson, O. E. (1985). The economic institutions of capital-
ism. New York: Free Press.
Worm, V., & Frankenstein, J. (2000). The dilemma of mana-
gerial cooperation in Sino-Western business operations.
Thunderbird International Business Review, 42(3), 261–283.
Yan, A., & Gray, B. (1994). Bargaining power, management
control, and performance in United States–China joint
ventures: A comparative case study. Academy of Manage-
ment Journal, 37, 1478–1517.
Yan, A., & Luo, Y. (2001). International joint ventures: Theory
and practice. New York: M.E. Sharpe, Inc.
Youse?, S. (1975). Contextual factors in?uencing control
strategy of multinational corporations. Academy of Man-
agement Journal (March), 136–143.
608 P. Chalos, N.G. O’Connor / Accounting, Organizations and Society 29 (2004) 591–608

doc_559149477.pdf