A10 Efferin&Powell2007 Ethnography

Accounting, Organizations and Society 32 (2007) 223–262
www.elsevier.com/locate/aos
Management control, culture and ethnicity

in a Chinese Indonesian company
Sujoko Efferin a, Trevor Hopper b,c,*
a
Universitas Surabaya, Surabaya, Indonesia
b
Manchester Business School, University of Manchester, Manchester M13 9PL, UK
c
Stockholm School of Economics, Sweden and Victoria University, Wellington, New Zealand
Abstract
This study explores socio-cultural aspects of management control in a Chinese Indonesian manufacturing company.
Ethnographic data collection methods were combined with grounded theory data analysis to explore how cultures, eth-
nic differences, history, politics, and commercial considerations shaped management controls. A combination of emic
and etic methods were used to generate grounded comparisons with nomethetic research on culture and control in a
cultural contingency tradition.
Chinese Indonesians own most Indonesian private domestic capital despite being an ethnic minority (3–4% of pop-
ulation) and having suffered extensive discrimination. The case links the Chinese businessmen’s values to socialisation
during childhood and then examines how their interaction with the Javanese culture of pribumi employees, ethnic ten-
sions between employers and employees, and organisational and economic factors affected management control. Con-
sistent with previous cultural contingency research the Chinese owners’ preferences resided with controlling behaviour
through personnel and behavioural controls, low budget participation, centralisation, subjective rather than objective
controls, and tentatively, few rewards tied to results and the use of group rewards. Whether Chinese managers exhibited
longer term orientations concerning planning and rewards could not be ascertained. However, ethnic tensions and com-
mercial considerations mitigated the owners’ ability to control according to cultural preferences. Based upon these find-
ings reflections on past research and suggestions for further developments are made with respect to methods,
methodology, and incorporating a broader range of theories and issues, especially ethnicity, politics, and history.
Ó 2006 Elsevier Ltd. All rights reserved.
*
Corresponding author. Address: Division of Accounting Introduction
and Finance, Manchester Business School, University of
Manchester, Crawford House, Booth Street East, Manchester
There has been considerable interest in whether
M13 9PL, UK. Tel.: +44 161 275 4014; fax: +44 161 275 4023.
E-mail addresses: s_efferin@ubaya.ac.id (S. Efferin), Tre- national cultures produce different control systems
vor.Hopper@mbs.ac.uk (T. Hopper). (see Bhimani, 1999; Harrison & McKinnon, 1999).
0361-3682/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.aos.2006.03.009
224 S. Efferin, T. Hopper / Accounting, Organizations and Society 32 (2007) 223–262
However, the research is beset with theoretical and source of action and meaning. This is so in Indone-
methodological controversy (Baskerville, 2003; sia, which has many ethnic groups. However, in
Baskerville-Morley, 2005; Hofstede, 2003) and the political arena the primary ethnic distinction
inconsistent and problematical results (Harrison has lain between pribumi and Chinese. Politicians’
& McKinnon, 1999). Bhimani (1999) argues the accusations that Indonesian Chinese contaminate
two main conceptual approaches used to date, or resist Indonesian national culture led to politi-
structural contingency and culture based idea- cal suppression of Chinese culture and influence.
tional theories, need supplementing with theories Chinese Indonesian business practices may draw
of societal effects, new institutional sociology, from Confucianism but they must also co-opt sup-
and ‘new’ accounting history, despite their differ- port in hostile local environments marked by
ent assumptions about what homogenises control resentment of Chinese businessmen, mutual ethnic
systems, the nature of controls, appropriate suspicions, and a history of state discrimination
research methodologies, and their focus of analy- (Ong, 1999; Redding, 1993; Redding & Whitley,
sis. This paper illustrates Bhimani’s claim that the- 1990; Yeung, 1999). As Bhimani (1999), Wickra-
ories can be complementary and progress lies in masinghe and Hopper (2005), Wickramasinghe,
theoretical triangulation rather than continuous Hopper, and Rathnasiri (2004) argue, addressing
sniping over the barricades of the accounting multi-culturalism required addressing history, pol-
methodology ‘science wars’.1 itics, and ethnic conflict.
There has been particular interest in how Chi- Cultural contingency research based on surveys,
nese culture impinges on management control.2 often using cultural constructs from Hofstede
Our original primary research interest was similar – (1980), has neglected issues of ethnicity and
it lay in exploring whether the cultural beliefs of multi-culturalism. For example, in multi-cultural
Chinese owners of an Indonesian manufacturing societies, who is Chinese can be problematic, being
company were consistent with Confucianism – the an issue of subjective definition rather than ances-
foundation of Chinese culture – and how, if at all, try. Cultural contingency research gives little
these impinged on the firm’s management control explanation of what Chinese values are, why and
system (MCS). However, the Chinese owner-man- how they emerged, why they matter, and how they
agers operated in a multi-cultural milieu for their influence controls. Hence it is unsurprising that
employees were predominately pribumi (mainly it has proven inconclusive (Baskerville, 2003;
Javanese – see glossary for definition of Chinese Chenhall, 2003; Harrison & McKinnon, 1999).
and Indonesian terms). Thus, if Chinese culture is Moreover, culture is neither totalising nor deter-
important for MCSs then Javanese culture and ministic: managers have choices and can act con-
inter-cultural interactions should be too. trary to cultural beliefs. They must consider
Multi-culturalism is often associated with eth- economic considerations, competition, and factors
nic differentiation. Ethnicity is a source of group such as organisational size and technology when
identity: it not only attributes characteristics exercising control. As the researchers realised in
(whether founded or imaginary) to members’ focal the course of fieldwork, cultural beliefs may be
group but also to other ethnic groups. Ethnicity marginal, consequently the central research ques-
defines the self in relation to others and can be a tion broadened to:
How does the socio-cultural environment of
Chinese Indonesian businessmen influence
1
An adaption of a point made by David Cooper in a plenary the design and operation of their company’s
at the Global Management Accounting Research Conference,
Michigan State University, 2004.
MCS. Is it a consequence of Confucian val-
2
Studies of Chinese culture and MCSs have often been on ues, Javanese values, ethnic differences, state
non-mainland Chinese operating in multi-cultural environ- threats, or best business practice?
ments. For ease of argument we equate the minority Indonesian
Chinese culture with that of mainland Chinese, i.e. a national The authors believed ethnographic fieldwork could
culture. inform this by building theory bottom up, avoid-
S. Efferin, T. Hopper / Accounting, Organizations and Society 32 (2007) 223–262 225
ing unduly predetermined categories and causal The paper initially outlines the model of MCS
chains, being longitudinal, and embracing a and features of Chinese culture that provided etic
broader set of issues. categories of analysis. It outlines propositions of
However, the researchers had a dilemma. anticipated relationships between culture and
Understanding the social construction and mean- MCSs derived from previous research for further
ing of an MCS requires emic analysis (which examination. The paper then discusses factors
describes indigenous values of a particular society not captured by the etic analysis of field data sub-
using semiotic ethnography based on ‘thick sequently analysed in an emic fashion. These were
description’ from fieldwork) whereas most the Javanese culture of the firm’s workers and the
accounting research on Chinese culture and con- local context – especially the history of ethnic dif-
trol is etic (it applies broader predetermined theo- ferentiation and discrimination against Chinese
retical models across several societies). An etic Indonesians. Then the research methods are
view is how outsiders see and interpret a phenom- explained and justified in greater detail. The empir-
enon, whereas an emic view is how insiders/partic- ics initially describe the company’s key actors,
ipants interpret the phenomenon as part of their history, structure, membership, technology, and
world (Marshall, 1998). markets (also significant for understanding the
The researchers’ inclinations to emic research MCS). It then analyses the Chinese owners’ values,
were tempered by a desire to engage with prior etic their sources, and how these factors – especially
research, build knowledge cumulatively, and use cultures, perceived best business practice, and eth-
theories in a complementary, pluralistic manner nic tensions; shaped result, action, and cultural
(Bhimani, 1999). Hence grounded data was ana- controls. The conclusions summarise the findings
lysed using etic categories from prior research and their implications for future research.
whilst using emic analysis to create new categories
and concepts when etic categories did not suffice.
Whether to concentrate on building rich descrip- Management control systems and cultural
tions of social groups or make cross-cultural com- contingencies
parisons lies at the heart of methodological
debates within ethnography between semiotic Features of a MCS
and behavioural approaches. The emic and etic
distinction was developed by the linguist Kenneth The researchers’ definition of an MCS was
Pike and adapted by Goodenough to develop broad – namely a system within social, cultural,
typologies for cross-cultural comparison derived political, and economic environments used by
from field data (Sanday, 1979) to reconcile semi- management to align employee behaviour with
otic and behavioural approaches. The researchers organisational objectives and to manage internal
do not claim combined emic and etic methods interdependencies (such as management–worker
are superior, or that emic analysis precludes com- and inter-departmental relations), and external
parative analysis (though it is beyond many emic relationships (with the state, society, customers,
researchers’ remit, and its creation of overlapping and suppliers) (Euske & Riccaboni, 1999).
but different categories hinders comparison), Merchant’s model of MCSs (1998) was used for
rather they claim that these methods were effective etic classification as it embraces a wide range of
for their research aims. As Sanday (1979, p. 34) formal and social controls, is operationally well
remarks, ‘‘The main differences are whether the defined, has informed or is consistent with MCS
primary focus is on the whole, the meaning, or models in previous research, and is not confined
the behaviour and the degree to which the analytic to large organisations (Davila, 2005). Manage-
goal is diagnosis or explanation. Which mode one ment control is notoriously difficult to define. Mer-
adopts is a matter of taste not dogma. . . . What chant’s model was used not because it is the ‘best’
counts in the long run is not how the facts are con- but because it provided effective etic categories to
structed but whether they make sense.’’ analyse emic findings, which enabled results to be
compared with previous research and be linked to (Redding, 1993), and social relations based on
internal and external factors that emerged as sig- jen that embraces chung and shu (Suryadinata,
nificant.3 Merchant distinguishes three types of 1978). Jen is like the Christian doctrine of ‘‘loving
control: result, action, and personnel. They are others as oneself’’ but extends to justice and equal-
not mutually exclusive; rather MCSs differ accord- ity – similar to secular concepts of law and order.
ing to emphases on each. In Confucianism humanity depends upon commu-
Result control defines outputs expected from nity status derived from human reciprocity. Chung
employees. Their achievement is often reinforced involves sincerity and honesty whereas shu empha-
by rewards. It is most useful if what constitutes sises altruism. Both emphasise ‘doing to others as
effective performance is known, employees can you would wish to be done to yourself’.
influence results, outputs are measurable, and Confucian values of vertical and horizontal
employees perceive managers’ authority as legiti- social order influence Chinese business culture
mate (Merchant, 1998). Action control monitors (Suryadinata, 1978). Vertical order covers hierar-
means (behaviour) rather than ends (results) by chical social relationships, such as husband and
prohibiting undesirable acts (behavioural con- wife, parents and children, masters and servants,
straints), deriving desired employee behaviour rulers and subjects. Each social position has
from plans (pre-action review), and monitoring ascribed responsibilities and duties known as li,
behaviour by direct observation or formal controls the most important being filial piety (hsiao). Par-
(action accountability). Personnel/cultural controls ents are expected to educate their children and
(hereafter called cultural control) focus upon direct them towards a correct life, whereas children
recruitment, training, job design, and promoting should respect parents, care for them, and protect
shared norms and values to induce employee self- their dignity. Chinese society is marked by strong
control (Merchant, 1998, p. 121). family collectivism (Redding, 1993): it assumes
people who neglect family responsibilities will
Dimensions of Chinese culture not honour obligations to outsiders. The need to
fulfil individual responsibility (li) underlies all
The research also needed to determine etic cate- other values. Hsiao extends to societal responsibil-
gories of Chinese culture. The researchers’ defini- ities, notably respecting, obeying, and demonstrat-
tion of culture is, ‘‘a shared way of life that ing loyalty to superiors, being wise rulers, and
includes values, beliefs, and norms transmitted caring for subordinates. Failure to do this will
within a particular society from generation to gen- embarrass not just the perpetrators but also their
eration’’ (Scupin, 1998, p. 36). It provides ‘‘. . . the parents.
bases for choice, by a social group, of particular Self-identity and social order in Chinese culture
ends and of particular means by which these ends is embedded in horizontal social relationships
are to be accomplished’’ (Lachman, Nedd, & Hin- (Redding, 1993). Individual ‘face’ and reputation
ings, 1994, p. 41), i.e. it defines what is good or bad, stems from social group membership. Chinese
holy or unholy, beautiful or ugly. Individuals judge society consists of concentric circles. The core is
organisations and respond according to stable core the family, surrounded by a lineage group or
cultural values and less enduring peripheral values extended family. Family resources should be pro-
whose breach brings less severe sanctions. tected and enhanced to foster self-sufficiency.
Confucianism, the foundation of Chinese cul- Relations between each circle are cemented by
tural values (Suryadinata, 1978), espouses moral mutual trust (Fei, 1939; quoted in Redding,
laws, tao, that emphasise social order, harmony 1993, p. 58). Friends and acquaintances are trusted
according to established mutual dependencies in
3
guanxi – ethnic based networks that, inter alia,
The authors do not deny the value of more emic inclined
research on national cultures and control such as Ahrens
protect family resources. Everyone has a sense of
(1997), Ansari and Bell (1991), and Euske and Riccaboni whom to trust and co-operate with, for what pur-
(1999). Indeed they commend it. poses. Outside these circles polite but guarded rela-
tions prevail. Socialisation by parents and schools and employees, especially if both are Chinese,
(Redding, 1993) and Chinese Indonesian’s harsh making it relatively easy to establish discipline
historical experiences have reinforced and shaped and stable hierarchies. Patrimony may foster
their values over generations (Yeung, 1999), pro- organisational adaptability, goal congruence, com-
moting guanxi networks as a defensive form of sol- pliant and diligent subordinates, and stable
idarity, co-operation, and self-reliance within organisational membership. On the other hand it
families. can produce nepotism, organisational cliques,
Personal trust derives from meeting personal non-family members with insufficient authority,
obligations, reputation and ‘face’. Neutral, imper- information secrecy, subjective performance evalu-
sonal relationships are difficult as they make deter- ation, restricted promotion opportunities, risk
mining trustworthiness difficult. Good social avoiding employees, and constraints on organisa-
relations and mutual trust are valued – ‘who you tional growth.
know’ is as important as ‘what you know’ (Redd- If delineations of effective MCS practices derive
ing, 1993). Hierarchical decisions such as firing or from the West, and Chinese culture differs from
hiring, rewards, discipline, and evaluating employ- Western culture, then MCS practices may differ
ees tend to be based on friendship, trust, emotional in Chinese organisations (O’Connor, 1995; Tsui,
ties and loyalty rather than results and efficiency. 2001). Merchant, Chow, and Wu (1995) conjecture
Cultural controls are legitimate in Chinese organ- how this may occur using five dimensions of
isations, are easily established, and tend to super- national cultures delineated by Hofstede (1980)
sede bureaucratic controls (Redding, 1993). and Hofstede and Bond (1984): namely collectiv-
Major values derived from li and hsiao are qin ism, masculinity, power distance, uncertainty
jian, ke ji, qin fen, and jing shen (Cleary, 1992; avoidance, and Confucian dynamism.4 Our
Xu, 2000). Qin jian (frugality) emphasises saving, research adapted Merchant et al.’s (1995) review
conserving resources, and displaying wealth mod- to generate seven propositions that guide our
estly. Ke ji (asceticism) stresses controlling desires empirical analysis and provide linkages to previous
for wealth, power and pleasures and fulfilling research. They are that Chinese managers will use:
social responsibilities. Qin fen advocates diligence
and seeking knowledge and wisdom. Jing shen 1. Personnel and action controls rather than result
embraces prudence and foresight. According to controls (Harrison, 1993; Merchant, 1998;
Confucius (Cleary, 1992, p. 25), ‘‘people who do Ouchi, 1979, 1980).
not think far enough ahead inevitable have worries 2. Little participation due to patrimony (Lau &
near at hand’’. Tan, 1998). Controls will be centralised due to
Confucian values have produced a patrimonial lower individualism (Birnbaum & Wong, 1985;
Chinese business style (Redding, 1993, p. 155). Harrison, McKinnon, Panchapakesan, &
Power stems from ownership: owners and employ- Leung, 1994; Lau, Low, & Eggleton, 1995;
ees view companies as family property and O’Connor, 1995).
management as akin to regulating a family. 3. Subjective rather than objective controls due to
Owner-managers instil values of personal trust, an emphasis on trust and personal relations
money consciousness and prudence on the family’s (Merchant et al., 1995; Redding, 1993).
behalf. Critical positions tend to be assigned to 4. Few rewards tied to results (outputs) to avoid
family or trusted members of lineage groups. loss of face and risk associated with uncertainty
Leadership is autocratic but paternalistic. Employ- (Merchant et al., 1995).
ers are responsible for employee welfare, allocating
jobs, stewardship of resources, helping the ineffi-
4
cient, providing security for the old and showing Hofstede and related accounting work, and the relationship
of the five dimensions of national cultures, especially Chinese
understanding. In return employees should show culture, are not discussed for reasons of space. Fuller exposi-
unquestioning obedience and diligence. The values tions are in Lau and Tan (1998), Merchant et al. (1995), and
of li and hsiao are powerful norms for employers Tsui (2001).
5. Group-based rewards due to values of collectiv- of human behaviour’’ refers to an individual’s

ism and shared risk (Merchant et al., 1995). external actions, motions, postures, and speech,
6. Few long-term incentives as collectivist values whereas batin ‘‘the inner realm’’ refers to their
decrease short term gaining (Merchant et al., emotional life: – fuzzy, shifting private feelings in
1995). their phenomenological immediacy (Geertz, 1960,
7. Long-run time horizons when planning using p. 232). Correct combinations of alus-kasar and
subjective information garnered from guanxi lair-batin enable even an uncultivated peasant to
or social networks (Harrison et al., 1994). attain the level of a hyper-civilised divine king
(Geertz, 1960, p. 233). The closer one is to alus
There has been empirical support for some and the more batin experiences one has, the more
propositions, e.g. participation and centralisation ideal one is. This produces three clusters of cul-
but other studies have produced surprises. For tural values: social order involving andap-asor
example, Merchant et al.’s (1995) study of (Tai- and bapakism, social harmony (rukun), and mysti-
wanese) Chinese and USA managers found con- cism including rituals such as slametan and
textual factors such as management education kenduri.
and experience, beliefs about stock markets, busi- Social order is manifest in Javanese etiquette.
ness growth, type of industry, and labour force This has rigid, formal rules of interaction within
mobility affected management controls more than social hierarchies, including linguistic forms and
cultural factors. A laboratory study by Awathsi, andap-asor, which means to humble oneself
Chow, and Wu (1998) found USA subjects used politely and demonstrate correct behaviour.
group controls in situations of interdependency Everyone should know their position and that of
and made more self-sacrifices than Chinese sub- others: behaving and choosing words correctly
jects, contrary to expectations. The failure to con- according to who is being addressed is a mark of
sistently corroborate links between Chinese culture respect. Words have a status hierarchy ranging
and controls have several possible reasons, includ- across alus to kasar, and Javanese pattern speech
ing assuming that national cultures are unitary on this axis according to the addressee’s status
whereas societies are often multi-cultural and beset and the conversation’s context. Failure to show
by ethnic tensions, as in this study, where Chinese andap-asor can cause shame (sungkan) for the reci-
Indonesian businessmen employed pribumi, mostly pient, especially if they cannot behave or reply as
Javanese, workers. Thus there needs to be an elab- well as the first person (Geertz, 1960, 1961).
oration of Javanese culture and its possible conse- Although modern education has brought more
quences for control. egalitarian attitudes demonstrating andap-asor is
still valued.
Javanese culture Bapakism is a Javanese form of paternalism and
patronage (Geertz, 1961; Rademakers, 1998).
There are no studies of Javanese culture and Bapak literally means father but can also mean a
MCSs to the best of our knowledge hence we charismatic figure that cares for community mem-
turned to anthropological work for insight. The bers. Bapaks demand respect, obedience and loy-
pillars of Javanese culture are alus-kasar and lair- alty from subordinates. They can claim
batin (Geertz, 1960). Spiritual excellence stems protection, gifts and help when needed but they
from alus and kasar. Alus means pure, refined, must listen to, empathise with and proffer advice.
polite, exquisite, ethereal, subtle, civilised, and Giving compliments is important: their proper and
smooth. God and his mystical experience are alus. constant delivery motivates subordinates and com-
Human behaviour and actions are alus if they fol- mands respect.
low appropriate manners and etiquette. Kasar Rukun is the maintenance of social harmony
means the opposite: impolite, rough, and uncivi- (Geertz, 1961, p. 149). A society without overt
lised. Everyone from peasant to king is ordered expression of divisive opinions and feelings is
between these two poles. Lair, ‘‘the outer realm rukun. This is manifest through collective (musya-
warah) and unanimous (mufakat) decisions, and outside the workplace (Bhimani, 1999) or demon-
co-operation (gotong royong). Reaching harmoni- strate links between beliefs and action – how and
ous compromise without intense feelings or expres- whether culture is enacted within MCSs is
sion of resentment is ideal. This can promote neglected (Harrison & McKinnon, 1999): action
evasion, covert disobedience and mutual avoid- may involve agency or responses to other commer-
ance in social relations: when conflicts burst open cial or social pressures. Sixth, there is undue reli-
they can be traumatic and severe. ance on problematical concepts and survey
Javanese have strong mystical beliefs. Many research instruments derived from Hofstede, often
believe ancestors, places, and spirits can communi- in tandem with problematical ones from contin-
cate. Maintaining harmonious relations with spir- gency theory and research on accounting perfor-
its is important for a tranquil life. This requires mance measurement (Baskerville, 2003; Chenhall,
following rituals such as slametan/kenduri – a com- 2003; Harrison & McKinnon, 1999; McSweeney,
munal feast on important occasions symbolising 2002) – building concepts and factors bottom up
the mystic and social unity of participants (Geertz, through field studies has been neglected.
1960). Though it incorporates some Islamic Bhimani (1999) argues that theories of Societal
elements, most Javanese regard slametan as dis- Effects, New Institutional Sociology, and the
tinctively Javanese (traditional spiritualism) and ‘New’ (especially Foucauldian) accounting can
pre-Islamic or even Hindu in inspiration. The address these deficiencies despite their different
goals are psychological: the absence of aggressive assumptions. For example, whereas contingency
feelings, no disturbance, and building unity theory matches predetermined dimensions of cul-
through social compromise (Beatty, 1999). ture and controls, seeks universal results involving
Prima facie, Javanese and Chinese cultures are functional relationships, assumes purposeful con-
similar in that both emphasise paternalism, hierar- trols can be constructed, and uses nomothetic
chy, reputation, social harmony, and social order methods (systematic studies, often statistical, to
though Javanese culture has a distinctive mysti- establish general relationships) the other theories
cism and set of behavioural expectations. If so, are more inclined to incorporate context, the
there may be little cultural dissonance if MCSs emergence of systems, and ideographic methods.
reciprocate these values. In contrast, the ‘new’ history stresses uniqueness,
complexity, and serendipity; how external institu-
Problems of cultural contingency tions bear upon the emergence of controls;
rationalities as products of historical power-
Cultural contingency studies have come under knowledge regimes; and longitudinal ideographic
growing attack. First, analyses may be too general- methods; new institutionalism focuses on subjec-
ised – they presume what is Chinese is relatively tivity and contextualised conceptions of social real-
homogenous within and across nations. They pay ity; and societal effects has rigorous grounding of
little heed to cultural diversity in societies and contextual factors expressed as theoretical proposi-
interactions between cultures (Baskerville-Morley, tions suitable for deeper comparative analysis. We
2005). Second, such studies tend to be static – they leave it to others to demonstrate the potential of
ignore how and why cultures change (Bhimani, these alternative theories. Our research does not
1999). Third, they ignore the impact of historical explicitly incorporate the formal substance of any
and external organisational factors involving, inter of them. However, we share Bhimani’s advocacy
alia, political and economic institutions and strug- of more contextual, emergent, and ideographic
gles upon culture and control (Bhimani, 1999). research, and we believe our research methods that
Fourth, they assume who is Chinese is objectively combine the emic and the etic, and incorporate the
ascertainable when it may be subjectively defined. politics of ethnicity, multiculturalism, history,
Fifth, they fail to establish how cultural values business pragmatism, socialisation, and action
are imbued by socialisation in the family, educa- over time to study culture and control, reflect the
tion systems, and social experiences within and spirit of what Bhimani intended.
Ethnicity, Indonesian politics, and Chinese tions such as traders, police, military, and bureau-
businesses crats for particular ethnic groups exacerbated
ethnic differentiation. Ethnicity rather than class
From the outset the researchers knew about the mobilised competition for resources, a common
history of ethnic strife upon Chinese businesses view being, ‘‘We’ve got less because they’ve got
and Indonesian politics. However, the original more’’ (Fenton, 1999, p. 47). This continues today
focus was upon Chinese culture and controls, in Indonesia – ethnicity rather than class is corre-
hence the emphasis upon their etic categorisation. lated with economic segregation, social ranking,
The significance of ethnicity and politics was a and power. Despite only representing 6–8 million
minor ‘surprise’ revealed by fieldwork and subse- of 200 million Indonesians, the Chinese share of
quent data analysis. For ease of exposition ethnic- private domestic capital far exceeds that of any
ity and Indonesian politics are discussed here but other ethnic group (Leong & Lim, 1992; Robison,
the reader should be aware that the processes of 1986)6 yet they have suffered political, social, and
discovery were, as is oft so in case study research, economic deprivation (Heryanto, 1997, 1998).
more iterative and ‘messy’ than the paper’s chro- The earliest permanent Chinese settlements in
nology may suggest. Indonesia date from the late 13th century (Fryer
Indonesia, despite the similarity of Javanese & Jackson, 1977; Rickleffs, 1993). Until the late
and Chinese culture has national integration prob- 18th century, most migrants were males who inter-
lems due to religious and ethnic divisions. Approx- married with locals, adopted local lifestyles, and
imately 85% of Indonesians are Moslems. The are now regarded as indigenous pribumi. However,
remainder include Christians, Hindus, Buddhists, immigration waves after World War I brought
Khonghuchus,5 and followers of Aliran. Moslems numerous Chinese women to Indonesia. Chinese
disagree whether the state should be secular or fol- children with Chinese parents raised in a Chinese
low Islamic law. The largest ethnic groups are culture rapidly increased. They became differenti-
Javanese (45%) and Sundanese (15%). The other ated as non-pribumi. However, definitions of ethnic
400 ethnic groups complain about ‘Javanisation’ Chinese in Indonesia do not rest simply on biolog-
of politics and culture but the major ethnic dichot- ical criteria. For example, many considered as Chi-
omy is between the Chinese (3% of population) nese are by ancestry less than one-quarter so,
and others (the so-called pribumi) (Brown, 1994). whilst others with a greater Chinese lineage are
Ethnicity classifies people according to alleged considered by themselves and others to be indige-
physical and social identities (Fenton, 1999). Eth- nous pribumi (Skinner, 1963). Who is Chinese is
nic group boundaries may be symbolic (language, a social construction stemming from how individ-
ancestry, religion, kinship, or culture) and may uals function within society and with whom they
have a material and class base. Ethnicity may identify.
involve racial and religious stereotyping and be Chinese Indonesians can be classified into totok
empirically unfounded (Eriksen, 1993) but it influ- and jiaosen. Totoks still practise and educate their
ences social action. Culture is often sub-conscious children in Confucian values, speak Mandarin or
and taken-for-granted whereas ethnicity more con- other Chinese dialects, and celebrate Chinese tra-
sciously defines the self and others. For example, ditional events. Most embrace Buddhism or
groups use ethnic claims to differentiate themselves Khong Hu Chu religions. In contrast, most jiaosen
and stigmatise other groups to mobilise political are Christians (some embrace Islam), follow few
programmes. Chinese traditions in daily life, speak little or no
The state can shape ethnicity (Brown, 1994).
Colonial rulers imported labour to Indonesia and
6
adopted divide and rule policies. Reserving occupa- The share of Chinese capital in the Indonesian economy is
controversial (Amir, 1997; Hadiz, 1997; Kwik, 1997; Ning,
1997) and it is difficult to get reliable data. Estimates vary from
less than 50% to 70%. Nevertheless, it is widely accepted that
5
A derivation of Taoism, Buddhism and Confucianism. Chinese capital is the largest source of private domestic capital.
Chinese dialect, and adopt local customs. Their foreign investments, and establishing political sta-
culture is a mix of Western, local, and Chinese/ bility through authoritarian rule (Robison, 1986).
Confucian influences. Younger Chinese are The new political leaders saw Chinese Indonesians
increasingly becoming jiaosen but many Chinese as engines of economic development and less
parents still teach Confucian values to their off- threatening to their power than pribumi (Robison,
spring, and many jiaosen separate religion and cul- 1986). Nevertheless the Chinese suffered discrimi-
ture, confining religion to personal faith, whereas nation, resentment, scapegoating, harassment,
culture defines proper social conduct. and extortion. They joined the West, Communism,
During Dutch colonial rule (pre-independence) and Fundamentalist Islam as the fourth major
the Chinese were used as economic bridges to other ‘Other’ in New Order political discourse (Heryanto,
ethnic groups and many prospered. Following 1998). The suppression of Chinese culture was jus-
independence, the Soekarno regime (1945–1966) tified rhetorically as being alien to ‘Indonesian
incorporated anti-foreign and wealth redistribu- national culture’. Public Chinese festivals were
tion sentiments (Chalmers, 1997) in economic banned, Chinese schools closed, entry quotas
programmes to encourage indigenous ownership applied to public universities and schools, and Chi-
and ethnicity. The Benteng (‘Fortress’) Programme nese symbols removed (for example customs decla-
(1950–1957) and the Economic Urgency Pro- ration forms categorised printed materials in
gramme (1951–mid-1950s) promoted pribumi Chinese characters alongside pornography, arms,
economic power relative to Dutch and Chinese and narcotics). The ‘othering’ of ethnic Chinese
capital. The Benteng programme restricted import- was justified by references to their origins outside
ing to registered pribumi merchants. The Economic Indonesia;7 Chinese colonial history; and allega-
Urgency Programme tried to create a strong tions that they socially segregated themselves, over-
pribumi business class by funding modernisation stated their religious and cultural traditions,
of small-scale pribumi manufacturers and restricted achieved economic dominance through economic
some markets to them. It failed because brokers crimes, and their primary loyalties lay with main-
with political connections won licences and sold land China and communism (Heryanto, 1998). In
them to more skilled capitalists, mostly ethnic short, Chinese Indonesians were portrayed as lesser
Chinese. In 1959 a ban on aliens retailing outside Indonesians who contaminated authentic Indone-
designated urban centres undermined the Chinese sian identity.
role as middlemen in national marketing. Many There were frequent, mass attacks on Chinese
rural Chinese were forcibly transferred into large property and life (Heryanto, 1997; Ning, 1997).
towns. Refusal to obey evacuation orders enraged Indonesian newspapers reinforced Chinese stig-
army personnel who used harsh physical measures matisation (Coppel, 1983, p. 158). Anti-Chinese
to enforce compliance (Skinner, 1963). This regula- attacks were often reported as natural and sponta-
tion only applied to Chinese without Indonesian neous – a populist search for justice provoked by
citizenship but it made all Chinese insecure for Chinese economic domination (Heryanto, 1998).
many families were a mix of Indonesian and The Chinese turned to the state for protection.
Chinese citizens. Following regulatory pressures This led Chinese businessmen to becoming ‘pariah
and attacks on their properties many Chinese left entrepreneurs’ having ‘wealth without power’
Indonesia. (Wang Gungwu, quoted in Blusse, 1990) in an
Soekarno’s nationalistic economic policy pro- environment of conflicts, contradictions, and co-
hibited international capital inflows and its con- operation. They relied upon powerful politico-
frontation with ethnic Chinese capital resulted in bureaucrats for protection who exploited Chinese
inadequate investment (Hill, 1996; Hill & Mackie,
1994). The New Order State (1966–1998) assumed
power when the economy collapsed. Its focus and 7
Arabs, Indians, and Indo-Europeans also had external
legitimacy rested on economic development, origins but did not suffer discrimination being too few to
through state, private (mainly ethnic Chinese) and constitute a threat.
business skills for personal economic and political lytical categories from field data and not generalise
ends (Mackie, 1992; Robison, 1986). findings beyond the site. However, many ethnogra-
Soeharto’s fall changed policies. The new phers wish to generalise which requires systematic,
regime brought a more democratic state, civil not cumulative work, better suited to etic approaches.
military rule, and ethnic tolerance and pluralism. Consequently, anthropological research commonly
This improved the position of Chinese Indonesians uses a mixture of emic and etic approaches (Mar-
but conflicts between employers (mainly Chinese) shall, 1998). This research pursued this path in
and workers (mainly pribumi) persisted partly order to engage with prior MCS and Chinese cul-
because wages deteriorated following high Indone- ture research using etic categorisations. The
sian inflation8 after the 1997–1998 Asian financial research model, derived from the review of culture,
crises. According to Ministry of Labour statistics ethnicity, and MCSs above, is summarised in Fig. 1.
(Binawas, 2000), there were 145 strikes in Indone- It provided etic categories for analysing the emic
sia during 1999 involving 48,239 workers and data collected. The MCS propositions were origi-
915,105 working hours9 but the actual number nally restricted to Chinese preferences for action
was greater since many strikes went unreported. and cultural controls. The remainder were added
The institutionalisation of ethnicity over 350 years subsequently to engage with prior cultural contin-
permeates economic activities and MCS practices gency research.
in Indonesia for it fosters ethnic stereotypes and As Bhimani (1999) notes, conventional contin-
suspicions. For example, many pribumi regard gency theory is convergent: it implicitly assumes
Chinese businessmen as exploitative and unscru- that factors such as size, technology, competition
pulous whereas many Chinese believe pribumi are will make MCSs across the world similar. Its
unreliable, lazy, and untrustworthy. incorporation of national cultures is significant
for it introduces ideational notions (Bhimani,
1999). The presumption that people behave simi-
Research methods larly according to their membership in a wider
society assumes cultural beliefs influence human
Methodology action and perceptions, thus it introduces subjec-
tive issues more commonly studied by qualitative
Most research on MCSs and Chinese culture is research methods. However, most work on
etic: its categories draw from concepts of bureau- national culture and MCSs, including that on
cracy and legal-rational authority in Western soci- overseas Chinese, tends to follow nomothetic
eties. Thus ethnographically inclined researchers research methods that test hypothesised associa-
face a dilemma. They can respect cultural relativism tions of predetermined attributes of culture, often
and use emic methods exclusively to generate ana- using research instruments from Hofstede, with
dimensions of MCSs measured by instruments
8
Inflation was 11.79%, 77.54%, 2.01%, 9.35%, 12.55% in
from previous contingency work. As argued previ-
1997, 1998, 1999, 2000, and 2001, respectively (Badan Pusat ously, this approach is problematic.
Statistik Indonesia/Indonesian Office of Statistics). The ontological assumption here was that MCS
9
Major examples include strikes of: 4000 workers from a practices are a product of individual meanings,
garment factory in Bogor-West Java that led to the destruction shared negotiations and are socially constructed
of the company’s office (Kompas, 15 December 2000); 500
workers from a shoe factory in Bekasi-West Java who occupied
(Hopper & Powell, 1985) – hence ethnographic
an inter-city highway for three hours (Kompas, 4 February methods to collect data. The epistemological
2000); 2000 workers from four companies following the assumption was that understanding derives from
dismissal of a worker in Surabaya (Surabaya Post, 22 February close observation of everyday interactions and
1999), and thousands of workers following the dismissal of actors’ explanations – hence grounded theory to
seven workers accused of being provocateurs in an electrical
appliances manufacturing company in Surabaya, which led to
analyse data to test prior categories and relations
the destruction of the company’s marketing office (Media and form new ones (Strauss & Corbin, 1998). We
Indonesia, 26 October, 2000). do not deny that social structures capable of gen-
Confucian Values Ethnicity of Chinese Javanese Values

Indonesian
• Vertical Order • Javanese etiquette
o Ascribed duties (li) • Economically powerful (alus-kasar)
o Filial piety hsiao) • Politically vulnerable • Language (lair-batin)
• Horizontal Order • Socially alienated • Correct behaviour
o Family centrism • Stigmatisation, (andap-asor)
o Collectivism harassment, extortion, • Social harmony
o Guanxi discrimination from other (rukun)
o Trust ethnic groups and state • Rituals & mysticism
o Reputation & face (slametan)
• Personal Qualities • Father figure
• Frugality (qin jian) (bapakism)
• Diligence (qin fen)
• Prudence (jing shen) Company external
• Asceticism (ke ji) & internal
interdependencies
Management Control System
• Personnel and action controls not result controls Other factors

• Little participation and centralisation
• Subjective not objective controls
• Few rewards tied to results
• Group-based rewards when rewards used
• Few long-term incentives
• Emphasis on subjective information from social
networks
• Long-run time horizons
Fig. 1. Management control in a Chinese Indonesian business.
eralisation exist but they are created and sustained notions and frames of reference of the everyday
in social interaction (Scapens & Macintosh, 1990, world’ (quoted in Alvesson & Skoldberg, 2000, p.
1996). They are subjectively created and sustained 33). Thus categories and relationships of cultural
and are capable of exception, change, and contes- contingency research may be useful for compara-
tation. They are not ‘laws’ as nomethetic cultural tive analysis but they need corroboration from
contingency research might imply. We question field data, relating to action, not be seen as abso-
Bhimani’s division of possible research methodol- lute or necessarily persistent, and be open to new
ogies to study culture between nomothetic and ide- elements. Treated in such a manner cultural con-
ational on philosophical and practical grounds for tingency work can form part of ‘integrative eth-
whilst it may be useful for expositional purposes it nography’ (Baszanger & Dodier, 2004) consistent
is invalid for, as Bhimani points out, once contin- with anthropologists’ and sociologists’ arguments
gency research embraces ideational notions associ- that understanding phenomena requires an itera-
ated with culture it enters an ideational terrain. tion of etic and emic views (Scupin, 1998; Smith
However, we also reject versions of grounded the- & Young, 1998; Wilk, 1996).
ory that restricts ethnography to rich descriptions
of observations on a particular site (see Alvesson Data collection and analysis
& Skoldberg, 2000; Silverman, 2004). As Bour-
dieu, Chamboredon, and Passeron (1991) note, Gaining access was not easy: overseas Chinese
‘Social science must create its own social concepts businessmen are usually guarded about their com-
formed in an entirely different purview than the panies and the research questions are sensitive in
Indonesia. To overcome this, trust and confidence (Spradley, 1980); and prevent participation dilut-
between the researcher and the respondents needed ing critical analysis (Hammersley & Atkinson,
establishment and personal networks pursued. 1995).
Access in Friends Company10 stemmed from the The owners and key employees fully supported
principal researcher’s previous involvement as an the research but the researcher’s ethnicity and
informal advisor to the company and friendship entry via the owners was a barrier with pribumi
with the owners since 1988. This made him an blue-collar employees, who were reluctant to be
‘insider’ with empathy and access to the owners interviewed for fear it might affect their jobs.
and managers, facilitated social interaction in nat- Hence, data about them came mainly from casual,
ural settings, and enabled him to participate in friendly conversations and interviews with external
work activities. It gave a deep understanding of Javanese figures and, to offset possible researcher
the firm’s activities, personnel, and physical infra- bias, Javanese researchers were trained to follow
structure, and revealed factors that pass unnoticed a rigorous research protocol involving the second
by other research methods. non-Chinese researcher to collect data from prib-
Ethical considerations were important. To umi employees. However, events from a pribumi
ensure participants’ willingness to participate and perspective are not presented. The principal
to protect them; confidentiality and anonymity researcher’s position within Friends, his involve-
were guaranteed; a tape recorder was used only ment in its affairs, and his ethnicity and values
with permission; and the researcher disclosed to probably skewed his behaviour and interpretations
respondents his true identity, research purposes of events towards the Chinese owners’ perspective
and how the findings would be disseminated. (Hammersley & Atkinson, 1995). The study is not
The field researcher’s Chinese Indonesian eth- objective in a positivistic sense. Its objectivity lies
nicity had strengths and weaknesses. It sensitised in its openness, willingness to listen and observe,
and gave him access to emic views of Chinese and representing respondents’ views as accurately
Indonesian businessmen on sensitive matters. as possible by diligent fieldwork and analysis
Good access is difficult without prior social rela- (Strauss & Corbin, 1998).
tions and reciprocity. Feedback on MCS systems The fieldwork lasted 1 year. Data came from
was an entry condition. The owners were con- interviews, documents, and participant observa-
scious of their lack of modern managerial knowl- tions (Mason, 1996; Spradley, 1980). Repeated
edge and wished to use external advice where semi-structured interviews, i.e. ‘‘conversations with
appropriate. During the research a consultant a purpose’’ (Burgess, 1984, in Mason, 1996) were
was employed to improve organisation structures held with the four owners and key employees,
and systems of accounting but the owners would including the consultant. Further interviews were
ask the researcher, who had no specific assigned held with five Chinese businessmen from other
duties, for comments on this and other matters. companies, four pribumi businessmen who deal
The owners however decided whether to act with Chinese businesses, and a pribumi Islamic
accordingly. Spradley (1980) identifies four modes intellectual. Initial interviews sensitised the
of participant observation ranging from non-par- researcher to issues for exploration in subsequent
ticipative passive to active involvement. The for- interviews. Interviews with the Chinese owners
mer helps maintain research objectivity but limits explored ethnicity, control, Chinese social identity,
insights from social engagement. The research social vulnerability, solidarity, family welfare, what
was inclined to ‘moderate’ participant observation organisations mean to them, personal values, and
rather than ‘action’ research to balance participa- their future. The management consultant and the
tion and observation, be an insider and an outsider Chinese businessmen gave insights into manage-
ment in Chinese companies. Trained pribumi
research assistants conducted some interviews with
10
For reasons of confidentiality the names of the company pribumi businessmen and the Islamic intellectual to
and personnel have been changed. encourage frankness about pribumi perceptions.
Questions during interviews and observations develop by breaking it down into its subcategories
helped open up lines of enquiry and direct theoret- (when, where, why, how, and so on), and finally
ical sampling. All interviews, observations, and integrate the major categories into a model.
documentary analysis were in Indonesian and Details of the coding and their explanation and
interviews were tape-recorded and transcribed. justification are given in the Appendix.
Documents studied included charts of accounting Theoretical sampling tested the internal consis-
systems and procedures, associated reports, manu- tency and completeness of the model. Constantly
als and documents, budget forms, the cash book, comparing concepts against field data revealed
and financial statements. The researcher had full variations among concepts, made categories den-
access to these and they provided an initial under- ser, verified and confirmed links between the cate-
standing of work processes but analysis of inter- gories/concepts being developed, and developed
views and observations transpired to be more new lines of enquiry and theory evolution. Sam-
important for formulating findings. Field notes pling did not necessarily entail gathering new data:
written during or immediately after each session returning to old transcriptions/memos was some-
about the situation, discussions, and native times sufficient. Sampling continued until catego-
phrases and terms were cross-checked with inter- ries were saturated, i.e. no new or relevant data
view data to promote balance and improve data emerged, variation in categories’ dimensions and
reliability and validity. properties were well demonstrated, and relation-
Data analysis consisted of transcription and ships between categories were established and val-
microanalysis (Strauss & Corbin, 1998, Chapter idated. Reaching absolute theoretical saturation
5). Microanalysis (see Appendix) helped generate proved difficult as new data continually emerged
concepts and relationships at the outset, identified but the major grounded concepts and interrela-
and linked native terms and beliefs, guided further tionships proved robust.
interviews and observations, constructed explana-
tory statements for comparison with data catego-
ries and relations in previous research, Friends company
systematically coded them into categories with a
central idea, and finally integrated them into a Brief history
model. Preliminary results formulated questions
for subsequent interviews and the focus of subse- Friends Company is situated in an industrial
quent observations. Interrelations between data park in East Java. It is owned by four sharehold-
categories and explanatory statements of relations ers: Mr. O, Mr. W, Mr. H, and Mr. A. It produces
(what, when, how and why) between them were plastic sheets in rolls or forms such as plastic bags/
constructed. Data categories were derived from sachets, inner layers of paper bags and appliances
the theoretical issues of the study (result control, for industry and households. Customer require-
action control, personnel/cultural controls, Chi- ments vary according to materials used, sheet
nese and Javanese cultural values, ethnicity, tech- dimensions, colour, bundle, model (with or with-
nical aspects of organisational activities, history out handle) and weight. Selling prices are based
of the company and its environment). Those pro- on weight and specifications.
cedures were often conducted simultaneously with Friends was established in June 2000. Its previ-
no rigid, clear-cut boundary between them, and ous owners for 30 years, Mr. Ed and his father,
categories and linkages were constantly refined had grown the company from 20 to 250 workers
and modified to ensure the final categorisation by 1994–1997. Mr. Ed is a 56-year-old totok Chi-
adequately reflected data. nese Indonesian with no formal education beyond
Concepts and their relationships were then sys- Junior High School. He used traditional Chinese
tematically coded. Similar concepts were grouped family business practices, spurning modern man-
within a category with one central idea. Once iden- agement techniques. There was a simple organisa-
tified a category became easier to remember and to tional structure, a strong paternal culture, and
centralised decision making – the owner subjec- ments, provides financial information to the
tively assessed performance and determined sala- owners, and protects company assets against
ries. In 1998–2000 the firm experienced cash flow fraud. Its responsibilities are not quantifiable so
problems: management could not cope with performance is judged qualitatively.
expansion and a customer defaulted on a large bill. Production processes are simple and well under-
Employees decreased to around 150 and in 1999 stood. In contingency theory parlance there is low
Mr. Ed offered his company for sale. An interme- task uncertainty. All production is job order: cus-
diary introduced Mr. Ed to Mr. O, who convened tomers’ requirements determine product character-
a meeting with three of his closest friends, Mr. W, istics – variations need different inputs and
Mr. A and Mr. H. In May 2000, the four friends completion times. Extruders process plastic ores
purchased the firm. into plastic rolls despatched to buyers or processed
They introduced modern management, aggres- further. Scrap and defects, which can also be pur-
sive marketing, recruited new employees, and chased externally, are recycled into lower quality
immediately hired a management consultant, Mr. plastic ores and re-used, which enables selling prices
L, an old friend of Mr. A, to establish a new to be modified. Mixing new and cheaper recycled
accounting system and organisation structure. plastic ores reduces costs more than minimising
They invested in machine repairs, maintenance, waste but excessive defects and scrap indicate waste
some new machines, factory enlargement, and of new ores or over-use of recycled materials. Pro-
redesigned layouts. Maximum production capacity duction volume per shift influences unit costs as
was 46.7 tons/month but in June 2001 usable workers’ wages are time not output based.
capacity was only 22 tons. Monthly sales were The market for Friend’s products is large for
circa Rp 190,000,000 (£15,417). Production is plastic sheets have diverse uses in households and
labour-intensive and there are three 8 h shifts. Pro- industry. About 90% of Friend’s sales are to indus-
duction costs were direct materials – 50%, direct try: the remainder is to independent wholesalers.
labour – 40%, and factory overheads – 10%. Five large companies (including Friends) in East
Friend’s total assets in the June 2001 report are Java compete for large orders but many small
valued at Rp 5.5 billion (£458,300). There were firms compete for small orders. Friend’s market
190 factory workers and 25 staff. is characterised by a concentrated and longstand-
Fig. 2 details Friend’s organisational structure. ing customer base (mainly 10 big companies); rev-
It has three main responsibility centres: Market- enue determined by order size not frequency; and
ing, Manufacturing, and Finance and Accounting few competitors. Bargaining power with customers
(F&A) Divisions.11 Manufacturing is a cost centre. is weak: losing one significantly reduces revenue.
Mr. H is the director responsible for production,
delivery, scheduling, maintaining machines and Social backgrounds of organisational members
factory facilities, inventories, and recruiting manu- The four owners (messrs. O, W, H and A) and
facturing employees. Marketing is a revenue and two senior employees (Mr. U and Cik K) play
expense centre. Mr. A is the director. It determines key roles. Mr. O is a 35-year-old bachelor from a
selling prices, sales terms and conditions, market- rich jiaosen Catholic Chinese Indonesian business
ing strategies, sales training, sales incentives, and family with an MBA in finance from a USA uni-
liaises with Manufacturing over delivery times. versity. He is not formally involved in Friend’s
F&A is a discretionary expense centre. Mr. W, management but attends meetings involving strat-
the President Director, is its director. It supports egy: the other shareholders hold him in high
the line with mainly non-financial data, adminis- esteem as he is an experienced manager. Mr. W
ters petty cash, processes cash receipts and pay- (28 years) and Mr. H (32 years) are brothers from
a rich, totok Catholic Chinese business family.
Their father has several businesses and both were
11
Division is used by Friends instead of department and is used raised with strong Chinese values. Mr. A is a 28-
accordingly here. year-old bachelor from a prosperous jiaosen Cath-
President Director
Mr W
Purchasing
Secretary
Officer
Manufacturing Director
F&A Director
Marketing Director
Mr H Assistant for
Mr W
Manufacturing Mr A
Director
Accounting
Supervisor Head of the Dept Head of the Dept
of Personnel and of Production Sales Supervisor
General Affairs Head of the Dept
Mr U of Warehousing Cik L
Petty Accounting Security Maintenance Production Ware housing Customer Sales

Staffs QC staff administration PPC staff Salesman Services Admin.
Cashier Staffs Officers coordinator staffs
Heads of
Technicians Production
Shifts
Foremen
Factory
Workers
Fig. 2. Friends company–organisation structure.
olic Chinese family who did not finish his under- We want to be very prudent in choosing
graduate degree. The four owners have been close partners. Trust must be developed first
friends from high school or university whence before all else. Only after getting close, can
many business relationships among Chinese Indo- we decide to do business together. In fact,
nesians stem. Knowing a person’s character is a we met with many other people during that
prerequisite for developing trust and business rela- period of 10 years. Yet, we feel that they do
tions, as Mr. H and Mr. W commented: not match our criteria.’’ (Mr. W)
‘‘All partnerships begin from friendship . . . The other two key players, Cik K and Mr. U,
We must understand the personalities of joined Friends in the Mr. Ed era. Cik K (the senior
our intended partners. If we are not assured, marketing supervisor) is about 40 years, unmarried,
probably the relationship will be limited to and from a middle-class but not wealthy totok Chi-
mere friendship.’’ (Mr. H) nese family. She is a senior high school graduate and
‘‘We have known each other for more than has worked in Friends for 17 years. Her nickname,
10 years. . . . Our characters match, we are Cik K, denotes employees’ recognition of her
all cengli, and we can give in to each other. seniority, authority, and social rank. She is well
known to important Chinese competitors, buyers, our workers’ rights. Subsequently no significant dis-
and suppliers, and respected in their guanxi, some- putes or strikes have occurred.
times being almost a family member. She remains Pribumi workers considered Mr. U as their
an important business contact. Cik K played a vital bapak. Although Mr. U is not Javanese, employees
role in saving Mr Ed’s business when it was near felt stronger emotional and social ties with him
bankruptcy. She spoke to suppliers on behalf of than the owners because they believed he was prib-
Mr. Ed and guaranteed payment. The suppliers umi and thus would guarantee fair treatment from
trusted her and continued to supply materials, the Chinese owners. The owners needed Mr. U to
which enabled Mr. Ed to repay his debts. secure pribumi workers’ loyalty and effort, as Mr.
This reveals four important points. First, the O recognised:
importance Chinese, external and internal to
‘‘We need Mr. U because he is our mediator
Friends, attach to integrity, trustworthiness, loy-
in dealing with the workers. He looks intelli-
alty and expertise: without this Cik K’s guarantee
gent and cengli. Yet this means that he has
would not have been valued, nor would the new
the capability to provoke them for his own
owners have taken pains to retain her. Second,
interests. . . . In the future, we have to be able
the importance Cik K plays in the guanxi of this
to control him so he will not turn against
business, which will not be forgotten. Third, the
us.’’
high value placed on trust by Chinese business-
men: when Cik K gave her guarantee she gave Thus the owners saw Mr. U as an important but
no money, written assurance, or valuables as a potentially dangerous employee. Given volatile
pledge. Her word was enough. Fourth, how in a Indonesian politics and the history of agent provoc-
Chinese company personal and business relation- ateurs in Chinese companies he could exploit his
ships are inseparable, especially if the employer influence over workers to mobilise ethnic clashes.
maintains kinship-like relations. Cik K regarded The owners co-operated with Mr. U whilst simulta-
Mr. Ed not only as her formal superior and neously trying to reduce their dependency on him.
employer but also as a senior family member The organisation structure (see Fig. 2) reveals
whom she must respect and obey, consistent with four management tiers: top management consist-
Confucian values of hsiao. For Cik K, working ing of the President Director and divisional Direc-
in Friends not only meant pursuing a career but tors; the middle-level consisting of five supervisors;
also respecting her long relationship with Mr. Ed. the lower-level consisting of staff in the F&A,
Mr. U, the head of the production department Manufacturing, and Marketing Divisions; and
(circa 38 years) is a nominal Moslem from an eco- the bottom-level comprising security officers and
nomically comfortable Bugis–Banjar12 family. He factory workers. For convenience, bottom-level
has an MBA from a Jakarta university and has employees are called blue-collar and the remainder
worked 6 years for Friends. Previously he was a white-collar.
manager in a bank, which acquainted him with Chi- Differences in ethnic identity, gender, education,
nese businessmen. Mr. Ed was his customer for and religion permeate divisional membership. Only
4 years and a formal business relationship changed three Manufacturing employees are Chinese – all
into friendship. The Indonesian government liqui- senior managers with a degree. There are 46 male
dated the bank and, following a chance encounter, and 155 female employees but all 11 senior manag-
Mr. Ed recruited Mr. U to handle production and ers are men. Only four pribumi employees – all
worker affairs. The company was beset by strikes senior managers – are educated beyond school.
as workers went unpaid due to financial difficulties. The women have particularly low education –
Mr. U calmed the situation and got Mr. Ed to hon- junior high or below. All employees are Moslems
apart from seven Christians, none of whom are
12
Bugis and Banjar are predominantly Moslem pribumi ethnic
shop floor workers. Most Manufacturing employ-
groups. The Bugis were originally from Southern Sulawesi and ees are pribumi (mainly Javanese), female, not edu-
the Banjar from Kalimantan. cated beyond junior high school, and Moslems.
In the Marketing Division there are five Chinese related to Chinese and Javanese cultural values
and two pribumi. All have tertiary education (see Fig. 1), ethnic tensions, modern management,
except Cik K. There are four Christians, a Bud- and business pragmatism.
dhist, an unknown, and a Moslem. In the F&A
Division there are three Chinese and three pribumi Result control
– all Christians with tertiary education and, except
for the Director, women. White-collar employees Result controls existed in three areas: budget
have similar religions, minority identities, school- setting, sales targets, and production defect and
ing, educational attainments, and attitudes, which scrap rates.
made communication easier and reduced ethnic
tensions. Setting budgets
In summary, there are few Chinese and Chris- Meetings of the four owners, occasionally Mr.
tian employees (5% and 7.5%, respectively) but L (the management consultant), and Mr. W and
top management is their preserve: none are Mr. H’s father, established budgets and targets
employed in bottom level jobs. Males and tertiary by consensus. The importance of budgets for con-
educated personnel are a minority (25.7% and trolling operations varied. The sales budget was
8.9%, respectively) but they dominate top and crucial as it determined overall performance.
middle management positions and most supervi- Other budgets were used to predict working capi-
sory ones. In contrast, blue-collar employees are tal requirements: the emphasis lay on financing
pribumi, mainly women, almost all Moslems, with and planning rather than management control.
modest education. There is a high correspondence The father advised on Indonesian politics, eco-
between ethnic identity, religion, education, hier- nomic developments, workers, technical matters,
archical level, and formal authority. In Friends, and their likely effect upon plans. The owners took
Chinese means being Christian, educated, senior heed because of his experience and their respect for
management, and well rewarded. filial piety (li) within family-centric businesses. The
Most pribumi workers had negative stereotypes owners would use conventional, modern, MCS
of Chinese employers as wealthy but unscrupulous practices but how was influenced by culture, ethnic
and cunning opportunists prepared to do anything problems, and business pragmatism.
for profit. They claimed ethnic exploitation was Politics and economics are intertwined in Indo-
common in Indonesia and working hard only ben- nesia (Coppel, 1983; Heryanto, 1997, 1998; Robi-
efited the Cino. On the other hand, the owners son, 1986). Government stability affected riots
worried that pribumi employees would extort them against Chinese businesses, exchange rates, and
if not properly controlled. Ethnic antagonism was inflation – all critical to Friend’s sales. For exam-
exacerbated by the Indonesian economic crisis. ple, in 2000 exchange rates fluctuated from Rp
Friends paid wages above government rates and 7000–11,000: $US1 according to presidential suc-
honoured workers’ legal rights but wages were cession conflicts, legal actions against members
deteriorating in real terms, partly because market of the previous regime, and religious and ethnic
prices for plastic sheets had risen less than infla- riots. A depreciating Rupiah increases many prices
tion. This increased workers’ resentment, created with demand consequences. During the New
motivational problems, and increased Friend’s Order era an authoritarian state guarded business
exposure to strikes. Workers suspected the owners stability and business flowed from good relation-
used economic excuses for ethnic exploitation. ships with it, but Soeharto’s fall changed this dra-
matically. The current political situation, whilst
less discriminatory, remains volatile. Reformists’
The company’s MCS demands for New Order members to be tried for
political and economic crimes can provoke mass
Data on control was analysed using etic catego- riots. Given the history of anti-Chinese attacks
ries of result, action, and cultural controls and there was anxiety that dissidents would hire
provocateurs to provoke strikes or attack Chinese considerations such as: types of orders to pursue,
properties to fuel national chaos. likely competition, which ores to use, capacity,
The owners’ network or guanxi, often based on likely performance, and production costs. Output
discussions in pubs, restaurants, or golf clubs, pro- targets were derived from four budgets: sales, pro-
vided crucial information on Indonesian politics duction, direct material purchases, and marketing
and the economy. Friends’ owners checked, dis- and general expenses. Key performance indicators
cussed, and reconfirmed predictions with friends were sales growth, product costs, productivity, and
prior to budget meetings. The hostile environment quality. Budgets incorporated three scenarios –
invoked solidarity amongst Chinese Indonesians, optimistic, medium, and pessimistic. The sales
fostered the guanxi as a medium of information budget was broken down by product and customer
exchange, and reinforced the owners’ predilection type over time and by machines used (new or old).
to act according to personal trust. The production budget was more problematic:
Both totok and jiaosen Chinese used the guanxi standard costing was impractical due to the multi-
for self-protection, albeit differently. A Chinese ple product dimensions, though labour and over-
senior management consultant noted: heads were relatively fixed. Hence the emphasis
on scrap and defect rates as a surrogate for mate-
‘‘There are two types of Chinese business:
rial usage control.
traditional Chinese and modern Chinese.
All directors discussed each budget and deci-
The former emphasises family networks
sions required unanimity. There was no division
whereas the latter emphasises social net-
according to divisional roles, no hint of budgetary
works from church or other associations.’’
slack, or any manipulation of divisional perfor-
Nevertheless, every business and family group had mance results because budgets were about ‘family’
its own guanxi: when linked to friends’ guanxis money and chung (sincerity and honesty) was
they were extensive. Becoming a member is not taken for granted. However, such data did not
easy: it requires at least one member’s personal pass beyond the inner group. Behaviour was con-
recommendation and a record of trustworthiness sistent with collective values, qin jian (frugality)
and reliability. Once accepted, a member enjoys and jing shen (prudence and thinking ahead). Vari-
favourable treatment and support but loss of trust ances were investigated in a spirit enunciated as,
makes transacting with other Chinese difficult. ‘‘We are all new in this business, so let’s learn
Reputation must be protected and promises together and support each other’’. Goal congru-
be honoured. A Chinese businessman illustrated ence, personal trust, informality, and family collec-
this: tivism were important. Friendship prior to being
business partners was essential and simplified bud-
‘‘Once, we delivered goods to our buyer. He
geting. Budgets were only used for delegated con-
checked everything and received them but
trol with respect to sale and scrap rates.
when he wanted to use them the goods
started discolouring. We cannot say, ‘You
Sales targets and commissions
have checked and signed the delivery order’,
although legally we can do so. Finally, we
The sales target, derived from the sales budget,
negotiated how we would share the loss. This
was important. Mr. A gave each salesman a
was the etiquette adopted to create trust so
monthly sales target derived from the medium
we get repeat orders from them.’’
level budget. Every Saturday he discussed weekly
The guanxi helped find credible suppliers and buy- results and problems with Marketing employees.
ers, finance, trustworthy employees, and business A monthly formal performance review was nor-
partners, provided information and help, and re- mally symbolic because problems were picked up
duced exposure to the state. previously. More than 50% of the salesmen’s
Once political trends were determined, their income came from commission based on individ-
budget ramifications were translated into technical ual sales to motivate order maximisation. Mr. A
gave lower monthly commission based on aggre- tor (Mr. H), the production head (Mr. U) and
gate sales to other Marketing employees to pro- workers responsible for the defects. Mr. H and
mote co-operation, sales support, and client Mr. U were penalised to show divisional members
service but fixed salaries formed most of their were treated equally but workers carried the bulk
income. Only salesmen were punished for not of the penalty.
reaching targets. If monthly targets were unmet, Penalties needed sensitive handling to prevent
salesmen could recover by meeting their 3-month blue-collar employees perceiving them as wage
targets or, failing this, their 6-month targets. If this reductions and ethnic discrimination. Mr. U, being
was not achieved the formal rule was dismissal, the workers’ bapak, became a guarantor of man-
though this was the owners’ prerogative. Ethnic agement’s good intentions, as he elaborated:
antagonisms between employees and owners were
‘‘A financial penalty is effective since it
low in Marketing and there was minimal conflict
directly affects their daily income. The only
over commissions.
thing we have to be careful about is the
amount and the way we explain it. They will
Production defect and scrap rates
not resist if they make mistakes. We need to
teach them how to be disciplined and
Manufacturing costs were critical – they consti-
accountable.’’
tuted the bulk of costs. The owners scrutinised
production volumes, the direct material purchases Mr. U explained to workers how improved prod-
budget, material price and mix variances, and pro- uct quality would improve Friend’s future, jobs
duction defect and scrap rates but only production and wages, and he promised penalties would not
defect and scrap rates were passed down as opera- be extortionate. The owners’ reliance upon Mr.
tional targets. Delegated output targets and associ- U13 contravened their preference for appointing
ated rewards were technically difficult due to the close, trusted colleagues, often family, known to
lack of a standard product. be loyal, sincere and honest. Their dependence
Scrap was pieces of plastic sheets not delivered on him meant they had to sacrifice values of patri-
to customers. The standard rate (0.5% per batch) monial, autocratic management, and sharply
was derived directly from the budget. Scrap rates demarcated employer and employee responsibili-
were checked at the end of each batch. Excess ties. Similarly, applying financial penalties to Mr.
scrap indicated worker carelessness in setting-up H because of his functional responsibility violated
machines or faulty machines. No rewards or sanc- the owners’ values of collectivism, shared responsi-
tions were tied to scrap rates: they merely helped bility, and vertical order (li and hsiao). Chinese
prevent workers stealing scrap. values had to be modified in the face of ethnic
Product quality, essential for customer satisfac- tensions.
tion and repeat orders, was measured by produc- Thus the owners used result controls for critical
tion defects. Mr. H transmitted defect rate variables especially in sales and production. Aca-
targets set by the owners independently of the bud- demics are highly respected in Indonesia and the
get to the production head (Mr. U) who relayed owners, who believed their knowledge of modern
them to foremen and workers. Constant deviation managerial techniques was limited, often sought
from normal rates indicated problems with work- to discuss such issues with the researcher. They fre-
ers, machines, or material quality (the purchasing quently mentioned the need for professional man-
officer’s responsibility). Quality control staff esti- agement using modern techniques such as budgets,
mated defects by end of batch sampling and, if performance-based rewards, and reliable account-
above 2%, reworked batches could be demanded. ing information.
Weekly and monthly defect rate reports went to
the President Director (Mr. W) and Mr. H. 13
After the research Mr. U left Friends to work again for Mr.
Defects above standard brought a penalty of Rp Ed. Since then a pribumi personnel officer has struggled to
2000 per kilo levied on the Manufacturing Direc- maintain control of production workers.
Result control in production was difficult as a to secure us.’’ Standardised procedures and docu-
standard output could not be ascertained and ment flows across major transaction cycles (selling,
material purchases were subject to currency fluctu- purchasing, cash receipts, and cash disbursement)
ations and some customers supplying their own separated functions (custody, recording, checking,
raw materials. Nevertheless, the owners did not and approval) across the three owner/managers.
extend budget participation beyond their inner cir- The accounting system programmed behaviour
cle and their guanxi due to values of patrimony through rules and procedures but financial control
and familial control. Subjective political consider- was not delegated. Rather than leaving financial
ations initially predominated budgeting though stewardship to ‘unknown’ professionals, the own-
the conclusions were translated into detailed bud- ers monopolised critical financial functions such
gets. However, ethnic tensions affected employee’s as approving bills and purchases to personally
perceived legitimacy of controls – hence the own- protect family assets and cement ties of personal
ers had to work through the bapak, Mr. U, and trust and protect their reputation with customers
subject Mr. H to penalties for production defects, and their guanxi. However, outside inner circles
in violation of their beliefs that owners’ responsi- the accounting system was designed to protect
bilities were collective. The owners were not averse resources in a climate of distrust. Mr. L, the con-
to implementing delegated result controls but they sultant who designed Friend’s accounting system
had to accommodate a complex mix of interacting commented: ‘‘The most important thing for them
factors. Pragmatic and commercial reasons could is to avoid being cheated by their employees. They
coincide with cultural values but enacting the lat- are so worried about this.’’ The accounting system
ter could be inconsistent, for example delegated evolved in a context of ethnic prejudice and suspi-
result control may violate patrimonial values but cion, as Mr. H noted: ‘‘Dealing with (blue-collar
be congruent with frugality (qin jian) and seeking employees) is tricky. I don’t want to be over-sensi-
knowledge and wisdom (qin fen). Control practices tive but this system enables us to secure our inter-
needed adapting to commercial considerations, ests and avoid unnecessary confrontation.’’ Mr.
technologies, customers, the legacy of employee Ed had relied on spasmodic personal surveillance
relations left by Mr. Ed, local politics, ethnic ten- for internal control, which resulted in a major
sions, and desires to incorporate best international fraud. The current owners wished to avoid this
business practice as well as cultural beliefs. by adopting accounting procedures that sup-
pressed ethnic distrust by appearing neutral, objec-
Action control tive, and administered discipline and tidiness
regardless of rank or ethnicity. Such a bureau-
In Friends action control was enacted through cratic control system was culturally legitimate for
the accounting system (behavioural constraint owners and employees alike. The owners realised
and pre-action review), authority limits for deter- they could not rely on traditional, Chinese family
mining selling prices (pre-action review), and business practices like trust and informal transact-
action accountability. ing exclusively. Their financial caution combined
Chinese values of frugality and money conscious-
Accounting system ness (qin jian), and prudence/thinking ahead (jing
Accounting systems prohibited certain shen) with business pragmatism.
employee behaviour. Formal procedures for pro-
duction orders, production controls, billing, issues Authority limits in determining selling prices
of cash receipts, payroll, purchasing and receiving, Each salesman had limited discretion to negoti-
and bank and petty cash disbursements were ate selling prices. If a customer’s wanted a price
designed to prevent fraud. As Mr. W observed: less than the salesman’s discretionary limit he
‘‘We cannot just rely upon employees but if we needed approval from Cik K – the marketing
don’t believe in them and delegate our authority supervisor. If the price was below her discretionary
we cannot work as well. Hence, we need a system limit (which was higher) the transaction went to
Mr. A who had no limit. If he was unsure, which to his colleagues. . . . Even if a duty is not for-
occurred frequently with large orders, he secured mally mine, I don’t mind helping them out-
a consensus decision from the owners. Thus, side working hours.’’
bureaucratic rules provided limited delegation
According to traditional Chinese management
within Chinese cultural preferences for formal
values authority stems from personal ownership
managerial hierarchy and collective decision-mak-
– not written, binding agreements. Constraining
ing by owners. Formal limits of authority on
behaviour by codified rules contradicts both Chi-
employees reconciled distrust of outsiders with
nese and pribumi employees’ belief that delegating
pragmatic recognition of needs for constrained
authority and responsibility is the employer’s priv-
delegation. Individual authority and responsibil-
ilege, so employees never sought written job
ity, and performance evaluation of owners acting
descriptions. Given the ethnic antagonism of
as divisional managers was irrelevant due to the
blue-collar employees the owners were wary of
owners’ collectivist values and mutual personal
doing so. Mr. U commented: ‘‘They don’t need
trust.
too many formalities such as a written job descrip-
tion . . . [employees] may use it as a weapon to
Action accountability
neglect their responsibilities. . . . Instead of pro-
Elsewhere definitions of acceptable action lay in
moting order you may get more trouble and con-
unwritten, shared understandings of Friend’s
flict.’’ The symmetry between Javanese values of
members. Employees were expected to obey own-
bapakism emphasising loyalty and obedience of
ers’ orders, be loyal, i.e. show filial piety (hsiao);
servants and Chinese values of vertical order
be diligent (qin fen), helpful, and above all, trust
meant owners and pribumi employees had com-
the wisdom and prudence of the owners (jing shen),
mon cultural expectations regarding performance,
demonstrate honesty by being truthful and sincere
written job descriptions, and rules.
(chung), and elevate company interests above per-
The owners’ direct monitoring of employee
sonal goals and accomplish assigned duties (li).
behaviour was largely confined to white-collar
These expectations had cultural legitimacy with
employees – observing blue-collar employees was
owners and employees alike.
impractical due to their number and ethnic ten-
For the owners this derived from Chinese values
sions. Mr. U and worker informants were a crucial
of respecting vertical order (li and hsiao). Mr. W
source of information on Manufacturing employ-
commented: ‘‘If there is a highly trustworthy,
ees’ behaviour: they were the eyes and ears of own-
loyal, committed and capable person, he will be
ers in the factory. They reported which employees
our right-hand and be given more authority. Such
were obedient, loyal, honest, and skilled, and they
an employee is ideal.’’ Chinese employees shared
relayed ethnic tensions promptly so precautions
expectations of filial piety (hsiao) due to their Con-
could be taken quickly. Workers trusted Mr. U
fucian parental education. For example, Cik K
as their bapak and relayed their aspirations
commented:
and complaints to him. Mr. H also cultivated per-
‘‘One has to serve one’s master as best as one sonal relationships with informants – experienced
can. In the past, my mother worked for a blue-collar employees with proven loyalty to the
company and she also did her best. When previous owner, who often gleaned sensitive infor-
Mr. Ed had problems with the workers in mation. Often Mr. H relied more on their observa-
the past he was taken to the Bakorstanas. I tions than supervisors’ reports. The owners’
was there with him from 11 a.m. till 2 a.m. reliance on Mr. U and worker informants contra-
the next day. He was the only one interro- vened their preference for relying on trusted people
gated by the officer but I was there to help but, given mutual distrust between pribumi work-
him as best as I could. . . . Employees’ skill ers and the owners, and the practical constraints
is important. However, I always observe on personally exerting action control, they were
which one of them is ready to lend a hand pragmatic in sacrificing core values.
During Mr. Ed’s ownership there was no stan- end of Ramadan) and non-Moslem employees
dard incentive scheme. His arbitrary determination before the Christmas/New Year holidays. THR
of wages produced discrepancies between new and levels for factory workers are regulated by the gov-
old employees’ salaries that compounded employee ernment: for other employees it equals a month’s
perceptions of ethnic discrimination and little con- salary and allowances.
nection between effort and rewards. Both blue- Deviance brought punishment though white-
and white-collar employees suspected that Mr. Ed and blue-collar employees were treated differently.
favoured Chinese employees. Mr. U confirmed this White-collar employees were initially rebuked ver-
when explaining why so many strikes occurred: bally by their superior who, if ignored, would tell
the owners. Cik K, as patron of white-collar
‘‘During the era of Mr. Ed the ethnic dis-
employees, determined whether Marketing and
crimination was very apparent, especially in
F&A employees had responded properly and
terms of salary. . . . There was a Chinese
whether the owners should issue a final warning.
employee in the production department. He
Normally, as she pointed out, she would rectify
received Rp 1,000,000 for his first salary
problems personally:
while my first salary was only Rp 400,000.
Many employees asked me: ‘Look at him, ‘‘At first, the salesmen . . . stayed in the office
do you accept this?’ The ethnic gap is much and made calls to prospective customers.
less now and this is more effective for con- They would go out only if they were sure that
trolling employees.’’ the customers wanted to buy something.
Then, I was angry with them: ‘There is no
The new owners systematised salaries according
way customers want to buy something unless
to position, length of service, dependants and
you see and approach them seriously face to
achievement to promote perceptions of fairness.
face so you can see their reaction clearly’.
Workers’ salaries now consist of the main salary
Since then they have become more active in
based on government UMR policy, a night shift
approaching customers.’’
premium, length of service, and a skill allowance,14
paid on daily attendance. Wages of other blue-col- The owners did not normally admonish blue-collar
lar employees (heads of shift, technicians, and employees. The first warning came from their
foremen) comprise the main salary (standardised superior (normally a foreman). If unheeded the
below junior management salaries) and allowances matter was referred to Mr. U who administered
for length of service, dependants, and skills. another warning. Workers who argued with super-
White-collar employees received their main sal- visors were scolded. Mr. U reported unresponsive
ary plus allowances for length of service, depen- recalcitrant workers to Mr. H who rarely adminis-
dants, and achievement. Standard wages between tered verbal punishments. He would check why the
Divisions varied and, except for achievement dispute occurred and dismiss employees deemed
allowances, across management levels. Achieve- unsatisfactory. The owners’ reliance upon Mr. U
ment allowances were based solely upon the own- ran counter to their values but only he could
ers’ subjective judgment. Salesmen also received a verbally punish blue-collar employees without
transport allowance and, like other Marketing provoking ethnic tensions: the owners had to co-
staff, sales commission. In addition, employees operate with Mr. U to exert discipline in the
received an annual incentive called Tunjangan Hari factory.
Raya (THR) on their main religious days. Moslem Verbal warnings and rebukes have similar con-
employees and factory workers receive this before notations for Chinese and pribumi employees due
the Eid-Al-fitre/Idul Fitri celebration (usually the to their paternalistic cultures. Early warnings are
given indirectly and politely. If this fails a rebuke
is given. At this point the employee’s faults are
14
The skill allowance is for the few workers able to operate perceived as almost intolerable and if no improve-
extruders – an essential operation. ment materialises the only punishment left is dis-
missal. For the Chinese, employee failure to porting documents prior to short-listing for
respond to hierarchical authority (hsiao) and per- interviews. However, this was not invariably
form ascribed duties (li) is a serious insult. Java- adhered to, and blue- and white-collar employees
nese etiquette (andap-asor) emphasises using were recruited differently.
appropriate linguistic forms, bapakism entails obe- Selection criteria for blue-collar employees
dience and loyalty of subordinates to superiors, included relevant experience and skills but also
and rukun stresses social harmony. If a superior personal character and religion. Mr. U sought sub-
admonishes using strong linguistic forms then the missive and obedient employees unlikely to
subordinate knows the severity of the situation. inflame ethnic conflicts. Although Mr. U is a Mos-
Verbal sanctions rather than formal controls had lem he was suspicious of graduates from Islamic
considerable significance amongst all employees. schools and Javanese with strong Islamic back-
Thus the owners replaced some personalised, grounds15 believing they misunderstood Islamic
rudimentary, centralised, and informal controls principles (such as fighting to spread the faith)
of Mr. Ed’s ‘traditional’ Chinese management and might provoke unrest. He was proud of his
era with ‘modern’ systems recommended by the reputation for creating harmonious industrial rela-
consultant, Mr. L. Later the owners and consul- tions and eliminating strikes, and he avoided jeop-
tants systematised and formalised wages and ardising this. When recruiting local applicants Mr.
rewards. Bureaucratic innovations to programme U went to a local warung at lunchtime where a
behaviour did not conflict with Chinese or Java- contact helped him ascertain whether applicants
nese cultural attributes and their objectivity and were suitable. He explained:
perceived fairness helped legitimate managerial
‘‘I co-operate with a local criminal . . . He
authority by defusing ethnic tensions. Neverthe-
interviews the local applicants. If it is OK,
less, general expectations of behaviour resided in
he recommends them to me and I do the next
Chinese values (reciprocated in Javanese values)
interview. . . . I don’t really know how the
emphasising loyalty, obedience, honesty, and
interview goes but if there is a problem . . .
accomplishing assigned duties, rather than
the local criminal will rebuke him. He does
achievement. Financial systems reinforced this
not receive money from us and is not even
being procedural and not used for result control.
formally employed by us but he is very proud
Evaluations of white-collar workers’ behaviour
of it. He thinks that he is a very important
and ensuing bonuses derived from the owners’
representative of this company for the local
subjective assessments but this was impossible in
people and nobody can work here without
production due to ethnic tensions, hence the
his approval. . . . We know many of the local
owners’ reliance on Mr. U, in violation of their
people are criminals so we let local people do
cultural preferences. Like result controls, adopting
the selection.’’
action controls stemmed from a complex mix of
factors. In contrast, recruitment of white-collar workers
was the preserve of the owners. Criteria included
Cultural controls competency and skills, personality, and socio-cul-
tural background (ethnicity, religion, and gender).
Cultural controls were used extensively. They The owners preferred Chinese for vital positions
embraced selective recruitment, creating divisions believing that their parents instilled respect for ver-
amongst blue-collar employees, building personal tical order, maintaining trust, and they shared the
relationships, and respecting religious customs. owners’ social vulnerability. Accounting positions
were particularly important because the owners
Selective recruitment
Formally, personnel were recruited through 15
This reflects differentiation between Javanese abangan and
standard administrative procedures. Job appli- santri (orthodox-leaning Moslems). Mr. U believed an abangan
cants submitted application letters, CVs, and sup- was a more suitable worker.
valued confidentiality and preventing fraud. Pur- an industrial area with a history of strikes and
chasing was vital because materials were the bulk riots: during the May 1998 anti-Chinese riots the
of Friend’s costs: a dishonest employee might local community saved many factories from mass
negotiate secret commissions with suppliers. Mr. attacks, as a Chinese businessman confirmed:
W commented:
‘‘There was a riot in this area due to the elec-
‘‘I prefer to recruit tenglang as accountants tion of a local village chief. Some factories
since we are also tenglang. . . . Trustworthi- were attacked . . . but I know one of the pro-
ness is an absolute prerequisite for working vocateurs. My workers also know the local
in accounting. All positions from supervisor people very well. During the riot they stood
level and above should be occupied by trust- in front of my factory and it was
worthy employees. It would be better if they untouched.’’
were all Chinese. . . . Why? Because . . . we
During the New Order era Chinese businessmen
have the same ancestors and we are the
protected their businesses through state agencies.
minority here.’’
Previously the military protected company assets
Recruitment of Chinese employees operated and broke strikes by, for example, arresting work-
through the owners’ guanxi. When a vital position ers’ leaders.
fell vacant the owners asked friends and relatives
‘‘During Mr. Ed’s time workers were kept in
to suggest candidates. If this failed they advertised
the dark and were not allowed to know
in newspapers and those selected were referred to
about government regulations on workers’
friends for comment. Recommendation from the
rights. If they dared protest they would be
guanxi was crucial, for example a Chinese Assis-
. . . fired or moved to another department.
tant to the Manufacturing Director was personally
Mr. Ed often used military forces to suppress
introduced and recommended by the owners’
them but this did not solve the problem.
relatives.
Strikes were repeated for months. In the
Supervisor positions upwards were important
past, the military had absolute power and
because of their performance ramifications. The
everybody was scared of them.’’ (Mr. U)
owners knew Chinese supervisors would encounter
problems controlling pribumi workers so they Following political reforms and the new demo-
appointed pribumi supervisors of pribumi subordi- cratic government such behaviour by military
nates. Personal recommendations from the own- and state apparatuses ceased. Mr. H commented:
ers’ guanxi were valued for pribumi applicants to
‘‘Military symbols are no longer effective in
supervisory positions. If this was unavailable then
controlling the workers. Instead of being
religion and gender were important. The owners
obedient they will be very angry if we invite
assumed non-Moslem pribumi were less antagonis-
soldiers here. The soldiers are also aware of
tic towards Chinese than Moslem pribumi, partly
this.’’
because in the New Order era Chinese children
attended school with Christians. Generally female Friends now had to control employees itself –
pribumi were considered more cooperative than hence its emphasis on cultural controls.
male pribumi. Chinese values of li and hsiao and personal
trust are similar to Javanese values of bapakism,
Dividing workers and building personal relations rukun, and andap-asor. Common collective values
Friend’s recruitment practices were preoccupied normally increase employee cohesiveness and
with recruiting trustworthy employees who would identification with owners but in Friends this
not exploit ethnic tensions. Good relations with was offset by ethnic tensions, especially with
the local Javanese community provided protection blue-collar employees. Consequently, Mr. U pre-
against riots and crimes. Chinese fears of stigmat- vented group solidarity, close employee ties, and
isation are not without foundation. Friends is in informal leaders emerging by constantly changing
shift membership. Cohesive groups may be pro- Mr. H used a friendly, informal and sympa-
ductive but ones with hostile attitudes could create thetic approach with blue-collar employees, often
effective resistance and threaten workers’ loyalty paying compliments, which is important in Java-
to him. nese culture: their proper use can stabilise and
Yet simultaneously he and the owners culti- solidify superior–subordinate relationships. How-
vated personal relationships with employees to ever, workers’ ethnic suspicions and their complex
gain their loyalty, reduce ethnic suspicions, and expectations regarding appropriate speech stymied
exhibit welfare and care. The owners cultivated Mr H’s attempts to divert their loyalty from Mr.
personal relations with white- and blue-collar U. Gaining affinity to Javanese employees requires
employees differently. Mr. H, the Manufacturing knowledge of their culture, for example, appropri-
Director, tried to build personal relationships with ateness of words in replies influences attitudes.
blue-collar employees. He recounted how: Also it can spawn complaints and demands, as
Mr. U noted:
‘‘I use a personal approach towards the
workers. I like to know their problems and ‘‘During the Mr. Ed’s time . . . there was no
make friendly jokes with them so that our mediator. As a result they pressed him and
relationship is more than merely formal because he always did what they asked they
employer and employee. I learned this from asked for even more. They thought that
my father. . . . Mr. U said that Mr. I is the Mr. Ed could be easily squeezed. Strikes
most juaji person here and even a provoca- took place very easily for trivial reasons.’’
teur. He is a technician and has worked here
Hence the owners continued to rely on Mr. U as a
for more than 27 years. Mr. U said he was
mediator.
dangerous and persuaded me to sack him.
How to relate to pribumi workers and whether
So I went downstairs and had a conversation
to use mediators such as Mr. U preoccupied the
with him. I found nothing wrong with him.
Chinese Indonesian businessmen not in Friends:
He told me that his kidney was painful so I
told him to go to the pharmacy and buy ‘‘I inherited many problems from my father.
medicine, no need to go to the doctor, and . . . If the workers needed something they
we will refund the money.’’ talked directly to him and if my father did
not grant it they would strike. My father’s
Such behaviour, consistent with Chinese values
position was weak since he had no
of patrimonial leadership and reciprocity, enabled
mediator.’’
Mr. H to assess employees’ behaviour, personali-
‘‘In my opinion we cannot treat them in an
ties and skills, and determine who had influence.
egalitarian way. The distance between supe-
This tallied with Javanese values of bapakism for
riors and subordinates must be preserved.
a bapak must listen, and give sympathy, empathy,
. . . (Otherwise) they will dare to argue with
and advice on personal problems to gain personal
us. I once approached them personally in
loyalty, as Mr. U recognised.
the past and they complained and demanded
‘‘Workers are very demanding. . . . We must many things. However, they dare not do so
follow their ways of thinking, listen to their with our production supervisor.’’
personal problems, give advice, and use ‘‘The absence of a mediator does not pro-
simple and understandable language to com- duce problems in my company . . . since we
municate with them. We become their coun- have emphasised from the beginning that
sellor for their personal problems such as the all are equal here and we will not discrimi-
illness of a wife and the circumcision of a son nate between superior and subordinate. We
but we must never be emotionally involved. told them: ‘Anyone who does wrong will be
. . . A personal approach is much more effec- punished for their own benefit’. . . . But we
tive than formal ways.’’ must keep our distance from them. If we
crack jokes with them too often they will no nasi tumpeng when the rice cone top was given to
longer respect us, although occasionally Mr. W as President Director. The ceremony
making jokes or associating with them is enabled the owners to convey their appreciation
important.’’ of Javanese culture and religion, and assert com-
mon group membership. It brought owners and
Nevertheless, despite their differences on how to
employees closer, promoted common interests,
relate to pribumi employees, all the Chinese busi-
and de-emphasised ethnic differences. The owners
nessmen believed ethnic prejudices fostered by
also granted workers a short break with extra
state stigmatisation of ethnic Chinese meant many
snacks and hot tea daily when fasting during Ram-
pribumi employees viewed personal approaches by
adan ceased. Despite the owners being Christians
Chinese employers suspiciously. On the other hand
this demonstrated respect for employees’ religious
owners must reciprocate expectations of care
duties, which resonated with Javanese bapakism
rather than merely exploiting employees.
values.
Developing social relations with white-collar
Cultural controls proved very important in
employees fell mainly upon the owners, especially
Friends, as the pains taken over selecting and
Mr. W, who maintained personal, informal com-
recruiting employees, ensuring they had desired val-
munication with Chinese and Javanese employees
ues, fostering social relations, and exhibiting
alike. Cik K was an important mediator between
respect for Javanese culture and religion indicate.
employers and white-collar employees. Her com-
The owners preferred Chinese employees, especially
ments are pertinent:
in key positions, believing they shared their values
and social vulnerability. However, Javanese values
‘‘I know all of the (Marketing and F&A)
were similar to Chinese ones – the problem here
employees personally. This is very important
lay in perceived potential ethnic tensions. Thus,
for making them open and honest towards
whilst the owners exercised cultural controls over
us. . . . The company has to take care of its
white-collar employees, blue-collar employees were
employees so they will do their best for the
selected by Mr. U who employed his own criteria
company as a way of expressing their grati-
and contacts for selecting and rendering employees
tude. The owners must realise that although
docile, submissive, and thus controllable.
they have a lot of money, they cannot buy
Cultural controls filled gaps left by result and
everything with it. We can only buy loyalty
action controls. Result control needs quantifiable
with care.’’
performance standards, which was not always pos-
Religious, ethnic and educational bonds between sible. Action control helped deal with qualitative
white-collar employees and the owners made social aspects of operations but required continuous sur-
relationships easier. In the factory, however, reli- veillance and considerable resources. But action
gious customs were sensitive. Any action con- and result controls only dealt with symptoms
strued as disrespect was deemed an insult but and not the causes of latent ethnic antagonisms.
respecting customs brought positive reactions. Cultural controls address these more fundamen-
The owners knew this and responded accordingly, tally by trying to inculcate alternative beliefs
for example regarding slametan and Ramadan. enacted in employees’ daily activities. Cultural
Machines in the factory often broke down controls are more subtle than result and action
despite frequent repairs. Some workers attributed controls which are overt visible tools to secure
this to a penunggu caused by the owners not hold- the owners’ interests. Nevertheless, cultural con-
ing the ritual ceremony of slametan since acquiring trols did not supplant result and action controls
the company. Following workers’ requests a – in practice they were intertwined, complemen-
slametan was held in October, 2000, attended by tary, and only separable for analytic purposes.
a kyai, worker representatives, the three owner- Result and action controls dealt directly with the
managers, and Mr. U. It began with Islamic prayer process and outputs of activities, giving concrete
led by the kyai and concluded with the sharing of directions on how to achieve organisational objec-
tives, whereas personnel/cultural controls helped what you can get in return if you want to pay
suppress ethnicity and promoted employees’ com- something. This is all to prevent us from
mitment to company aims. being cheated in the future. My father also
gave examples of cengli and non-cengli
behaviours.’’ (Mr. H)
Values, socialisation and ethnicity
The owners strongly held Chinese values of hsiao
The empirical analysis examined whether emic and li. Mr. H commented:
data on the Chinese owners’ values incorporated
‘‘For us Chinese, there is no mature person.
Confucian cultural values of respecting vertical
All people regardless of age are immature
and horizontal relations, and personal qualities16
in the eyes of their parents. I have children,
(see Fig. 1); how values were inculcated during
but they will never be mature enough in my
socialisation; and how these values bore upon con-
eyes and neither am I in the eyes of my
trols in the business. However, it is not sufficient to
father. For example, my grandmother is
just match behaviour, controls, and comments to
92 years old, and even now, she will admon-
predetermined categories of culture, it must be
ish my father if he has not had dinner by 12
demonstrated that beliefs derived from socialisa-
o’clock. . . My father has told me to learn
tion in a specific community.
from him since he has tasted much more salt
The owners’ values of Confucian order were
than me’’.
imbued from childhood by parents, schooling,
and daily life in Indonesia. All respect social strat- Vertical order was evident in Friend’s MCS.
ification and appropriate behaviour when interact- The importance of li was marked by expectations
ing with others. The owners’ parents instilled the of loyal, obedient employees; the owners’ preroga-
import of trust, reputation and personal relation- tive to: assign duties, determine achievement
ships, for example by using classical Chinese sto- allowances for white-collar employees, select
ries as analogies: white-collar and supervisory employees, spurn
job descriptions or widespread delegated targets,
‘‘I have been fond of Chinese stories since I
and discipline by verbal admonishments. The own-
was a kid. The most impressive one is Sam
ers’ respect for hsiao was manifest in the budget
Kok. I got two lessons from this. First, you
involvement of Mr. H and Mr. W’s father,
need to know your personal strengths and
attempts to develop personal relations with
those of your opponents before making deci-
employees, and beliefs that ownership legitimated
sions. Secondly, if you want to control some-
the owners’ authority. They valued loyal, obedient
body you need to know his character. For
subordinates who did not argue openly. Mr. W
example, Lu Po was a strong and brave gen-
stated: ‘‘An ideal employee is one who has disci-
eral. However, he was easy to bribe and eas-
pline, and is loyal and highly committed to the
ily betrayed his masters. This made him easy
company. I really dislike employees not like that.’’
to tame.’’ (Mr. W)
Running the business was dominated by prefer-
‘‘We were often taught by our father about
ences for autocratic, patrimonial, harmonious
(the importance of) knowing people’s per-
family-like relationships coupled to obligations
sonal characters. For example, from Chinese
for employee welfare.
kung fu stories, we were told what ruthless
Respect for horizontal order was manifest in
and good characters are, and so on. Consider
MCS practices. Family centrism was evident in
the confinement of planning and budgeting to
16 the inner ‘family’ circle, the appointment of own-
The values jen (justice/equality), chung (sincerity and
honesty) and shu (altruism) being fundamental values that ers to the Presidency and divisional headships,
permeate others were not used for coding but they are referred and their personal involvement in key account-
to in the text. ing and purchasing tasks. Within the inner circle
collectivism prevailed, underpinned by beliefs that be cengli, or not well known. The owners were
all members were jen, chung, and shu. For exam- open and cooperative with insiders but maintained
ple, the owners collectively determined budgets, distant but polite relationships with others. Mr. W
sales prices, and company and divisional goals. commented: ‘‘We have to maintain our etiquette
Beliefs about the importance of trust were on how to be polite with anybody else according
inbred by the owners’ parents. According to Mr. to our Chinese customs.’’ Who to trust was
W, ‘‘Every time my father visits us, he teaches learned from parents, and friends and relatives in
us. One can either occupy a respected position in guanxi. Parents passed on distrust of pribumi but
society or be the trash of society. . . (To be some pribumi and Chinese groups were considered
respected) we must be able to make people trust more trustworthy than others.
us.’’ Business transactions often relied on trust:
‘‘The Bugis and other ethnic groups from
‘‘I don’t like people who cannot be trusted. outer islands are more cengli than the Java-
When I was in the iron trade, one transaction nese. . . . People from Madura, although
could be as much as one billion Rupiahs17 many are involved in crimes, are tamsunya.
and we didn’t use any written documents. A Maduran will defend his boss with his life
Everything was based upon telephone con- even if he has to fight with another Maduran.
versations’’ (Mr. W) People from Jember, both Chinese and prib-
umi, are generally tricky and deceitful – just
However, consistent with jing shen trust relations
like Chinese from Medan.’’ (Mr. O)
were guarded:
Ancestral origins in China (e.g. Hokkian, Can-
‘‘Never fully trust other people. If you have
ton, Tio Ciu) were unimportant – Indonesian ori-
Rp 1000, do not give it all to someone even
gins were believed to denote varying degrees of
your own wife, brother, etc. because it may
trustworthiness.
be a matter of life and death for us. They
Ethnic stereotyping and distrust was reinforced
may be good persons but they can be care-
by the education system which separates most Chi-
less. Hence, all crucial decisions must be in
nese and pribumi in Indonesian cities18 and experi-
our own hands.’’ (Mr. W)
ences of social and state discrimination. For
Maintaining reputation and ‘face’ were vital for example, in 1986 all pupils (including Mr. W and
maintaining trustworthiness for parents had Mr. A) in predominantly Chinese Christian sec-
taught the owners that this determined social sta- ondary schools in Surabaya were suddenly sent
tus. Hence their direct involvement in purchasing home to avoid a possible anti-Chinese riot. As
and financial matters which, inter alia, enabled recently as May 1998 anti-Chinese riots in Jakarta,
them to liaise with guanxi and protect their reputa- Medan and Solo resulted in about 50,000 Chinese
tion. The guanxi was vital for support, and infor- fleeing Indonesia and not wanting to return.
mation about planning, budgets, and prospective Socialisation of values of frugality (qin jian) and
supervisors and white-collar staff. Knowing per- asceticism (ke ji) came from parents recounting
sonal characteristics was essential to distinguish family history. Most immigrant Chinese, including
insiders from outsiders. Insiders were well known, the owners’ ancestors, migrated to escape famine,
could be trusted, confided in, and helpful. Thus poverty and war in China. Parents frequently
Cik K, being trusted, operated at the fringes of recounted stories about early generations’ poverty
the inner circle, giving advice and exercising dele- and problems to encourage their offspring to exer-
gated authority to allocate tasks and administer cise values of qin fen (diligence), qin jian (frugality),
discipline in administrative functions. Outsiders jing shen (prudence and foresight) and ke ji (ascet-
were deemed to possess unacceptable attitudes,
18
Chinese are the majority in private Christian schools and
17
About £83,300. pribumi in state/other private schools.
ics). Buying the business was not an easy decision for your working capital and 20% for your
for the owners. They were inexperienced, knew lit- family. If something happens to your busi-
tle about the industry, had to borrow extensively ness it will deplete your working capital. If
and, apart from Mr. O, had secure jobs in parents’ the capital is used up – mbo-hwat – the last
businesses. However, it was also an opportunity to resort is the saving portion. This philosophy
take heed of parental exhortations to expand fam- is applied in this company. . . . Once we can
ily welfare, as Mr. W reflected: repay the entire loan we will not distribute
all the profit to shareholders. 50% will be
‘‘The Chinese are more established in busi-
retained for new investment. When the sav-
ness because when their ancestors first came
ings are big enough we will use them for
to Indonesia they were so poor they did not
our pensions. Why? If something happens
want their children to suffer like them. They
in the future we can still survive. Who could
worked very hard, lived very frugally and
predict that the May 1998 tragedy would
saved money little by little. Finally, their
happen? This principle has been agreed by
descendants had enough money to run their
all the shareholders. . . This company must
own business. Their children then continued
survive whatever happens. However, our
to develop their parents’ business or estab-
welfare must be protected as well.’’
lished new businesses.
The owners were constantly alert to social and
On another occasion he stated:
political developments in Indonesia for, as Mr. H
‘‘(To be respected) we have to save money. If stated, ‘‘If we can save money, then we can go to
we put one drop of water in a pool every day the USA, Australia or anywhere. We have to be
in the future the pool will be full of water. prepared for the worst.’’
No matter how little money we receive we However, the owners were not immune from
have to save it. If we already have a lot of wider social values and they internalised consumer-
money, money will seek us out. Otherwise ism. Contrary to frugality (qin jian) and asceticism
we have to look for it.’’ (ke ji) they spent money liberally when relaxing, pos-
sibly due to their youth. Frugality (qin jian), asceti-
Chinese values regarding personal qualities
cism (ke ji), prudence (jing shen), diligence (qin fen)
were imbued by parents but reinforced by contin-
and family-centrism remained core business values
ual experiences of problematical ethnicity in daily
but did not necessarily extend to their private life.
life. The owners recounted many painful and trau-
They displayed their wealth amongst close friends
matic ethnic events at the hands of some state offi-
but exhibited modesty in public, for example, stay-
cials and locals who viewed Chinese as targets for
ing in cheap hotels, consistent with jing shen and
extortion and harassment. Hence most state offi-
ke ji. Both values were evident in the owners’ presen-
cers and locals were viewed as potentially danger-
tation of self, especially in uncertain situations. For
ous. Relationships were handled with caution and
example, at work and when dealing with perceived
care, and values of prudence and foresight fused
outsiders, the owners sported a modest appearance
into business practices. A successful business built
and plain clothing. However, when meeting pro-
on jing shen (prudence and foresight), ke ji (ascet-
spective buyers or special guests they dressed well.
ics) and qin jian (frugality) was seen not as an
They used inexpensive cars for work despite having
option but a must for Chinese survival in Indone-
luxury cars at home. When visiting state offices they
sia. Having resources may avert calamities such
wore tidy, plain clothes. They hid their wealth from
as being expelled, as Mr. H recounted:
outsiders to avoid being taken undue advantage of,
‘‘If you have money, you need to save it for and to promote freer interaction which made
your future. My father taught me that if character assessment easier. Hence, modesty and
you earn money you should divide it into prudence in daily life stemmed from avoiding extor-
three portions: 50% for your savings, 30% tion as well as cultural values.
Factors influencing management control ees were low. But they had to use formal,
standardised, rule-based, objective wage systems
We now wish to examine the propositions for blue-collar employees. Despite Mr. H’s
about Chinese culture and MCSs summarised in attempts to befriend workers, the owners could
Fig. 1. There was tentative evidence that Chinese not monitor their behaviour directly and they
Indonesian owner–managers emphasised person- relied, somewhat reluctantly, on Mr. U, pribumi
nel (cultural) and action controls rather than result supervisors, and informants.
controls as claimed by Harrison (1993), Merchant The owners placed great store on cultural con-
(1998), and Ouchi (1979, 1980) but the research trols. Recruiting officially was formal and objective
uncovered a complex mix of action, cultural, and but in practice informal procedures predominated.
result controls that defies simple classification or The owners personally recruited white-collar
dichotomies. employees according to: ethnicity, religion, and
The Chinese owners selectively combined objec- gender; recommendations from guanxi; and per-
tive and subjective action controls. Bureaucratic ceptions of trustworthiness. The owners preferred
methods, being objective, formal and rule-bound Chinese white-collar employees believing they held
contradict Chinese managers’ inclinations for Confucian values, were easier to deal with, and
informal controls and patrimonial management. were more controllable. For pribumi supervisors
However, Friend’s owners used bureaucratic con- the owners used the guanxi and ranked candidates
trols pragmatically and selectively. They would according to perceived trustworthiness, religion,
violate cultural values – possibly due to their mod- ethnicity and gender. In contrast, when recruiting
ern, secular, education – if business circumstances blue-collar workers the owners sacrificed values
warranted it. For example, length of service wage of personal trust for more urgent, practical prob-
allowances reduced staff turnover, and objective lems related to ethnicity. They relied on Mr. U,
rewards promoted perceptions of equity. More- who assessed candidates by his perceptions of
over, as Harrison (1993) and Lau et al. (1995) sug- desirable characteristics. Mr. U, despite his bapak
gest, some aspects of bureaucracy are not status, protected his reputation for tranquil
antithetical to some Chinese (or Javanese) values, industrial relations by dividing employees and
for example, length of service allowances rewarded stamping on hints of unrest. The owners’ depen-
loyalty. Being objective they could defuse ethnic dence upon Mr. U worried them. Consequently,
tensions. they tried to develop personal relationships with
Nevertheless, formal action controls were used blue-collar employees by respecting Javanese pat-
sparingly. The owners were unwilling to abandon terns of interaction, observing slametan rituals,
subjectively defining appropriate employee behav- and providing sustenance during Ramadan, con-
iour. For example, they used accounting to pre- sistent with Chinese values of patrimony and
scribe behaviour but not for performance reciprocity.
evaluation; written job descriptions were neither Thus the owners’ use of cultural controls was a
sought nor valued; and, where possible, employees complex blend of fulfilling Javanese and Chinese
were evaluated according to Chinese values, e.g. values of patrimonial leadership and care, main-
loyalty and obedience rather than efficiency or taining good relations with the local community,
results. However, this also tallied with pribumi respecting workers’ religious and traditional
employees’ understanding of legitimate authority beliefs, and developing personal relations with
relations. employees despite mutual ethnic distrust and
The owners influenced white-collar workers’ different expectations for speech and interaction.
behaviour by subjectively determined ‘achieve- Yet the owners’ ethnic fears spawned subversive
ment allowances’, rebukes and dismissals. The forms of cultural control using informers; selective
message conveyed was, ‘Everything you do is taken recruitment based on ethnicity, gender and reli-
into account.’ This was possible because ethnic ten- gion; and reluctant dependence upon a pribumi
sions between employers and white-collar employ- mediator who, despite his position of bapak, weak-
ened shop floor resistance by creating worker divi- employers responsible for employee welfare, allo-
sions and his own brand of selective recruitment. cating jobs, and stewardship of resources.
Although the owners favoured behavioural MCS literature tends to dichotomise autocratic
(action) and cultural controls, they experimented and participative budgeting. In Friends budgeting
with result controls for key economic variables. contained elements of both. It was autocratic in
The budget was not used for delegated control or that it operated within concentric circles delineat-
performance evaluation, consistent with the find- ing insiders and outsiders but it was participative
ings of Harrison et al. (1994). Budgets were used when it involved insiders. The owners believed it
for financial planning and control within the inner was imprudent (jing shen) to trust employees with
group and the owners retained personal control financial information – even Cik K and Mr U were
over key financial positions. Budget targets were excluded. The owners kept Friend’s finances secret
not assigned to employees or owners acting as for fear of precipitating salary demands or extor-
divisional heads apart from sales targets and pro- tion, though employees did not expect budget par-
duction defect rates. The owners would pragmati- ticipation. Chinese employees’ respect for li, hsiao,
cally consider Western, modern approaches and and authority deriving from ownership was
selectively use result controls tied to rewards in broadly consistent with pribumi employees’ beliefs
critical areas. about bapakism and social order.
In summary, the owners preferred cultural and It is argued that Chinese business values of
action controls to align employee behaviour with ownership being the only legitimate source of
Chinese values but this preference was constantly authority encourage subjective rather than objec-
adjusted in the light of ethnic tensions and com- tive information and controls (Merchant et al.,
mercial effectiveness. They preferred informal con- 1995; Redding, 1993). Javanese values of social
trols but would use formal result controls for order and bapakism have similar repercussions.
pragmatic business reasons. Moreover, objective Hence, Indonesian employees (whether Chinese
bureaucratic procedures sometimes allayed ethnic or not) may grant Chinese owners the prerogative
suspicions and were not inconsistent with some to make hard decisions such as hiring and firing,
Chinese and Javanese values. evaluating employees’ performance, determining
There was support for claims that Chinese rewards, allocating resources, and exercising
organisations are centralised (Birnbaum & Wong, discipline and control. When beliefs in familial dis-
1985; Harrison et al., 1994). The owners filled all cipline based upon ownership permeate organisa-
divisional headships and executed key activities tional culture, then employees will tend to do as
in areas such as purchasing. Budget involvement told, not answer back, and do not need controlling
was restricted to the owners and their inner sanc- by a formal bureaucratic system. Personal trust
tum. Here rules defining individual responsibilities and relationships between employers and employ-
and power were superfluous, for participation and ees displace more neutral means of coordination.
trust was total, discussions open and frank, Evidence on this was mixed during this research.
responsibility collective, and respect came from During Mr. Ed’s ownership, which interviewees
expertise. Values of personal trust, consensus and frequently referred to as ‘traditional Chinese busi-
collective accountability derived from long per- ness style’, controls were entirely subjective, lead-
sonal relationships and shared interests prevailed, ing to major commercial and industrial relations
consistent with the claims of Birnbaum and Wong problems. The new owners introduced systematic,
(1985), Harrison et al. (1994), and Merchant et al. objective, and quantified planning and rewards for
(1995). This derived from Chinese values of dis- commercial and ethnic reasons. However, they
trust of outsiders, authority stemming from own- rejected formal job descriptions and budget dele-
ership, and protecting family wealth through gation, and retained subjective controls where pos-
diligence (qin fen) and prudence (jing shen). Out- sible due to their cultural preferences, which could
side the ‘family’ budgeting was autocratic and reciprocate with those of employees, for example,
paternalistic – consistent with li – which makes no employee sought or expected formal job
descriptions. Hence, information and controls vac- planning. Without a suitable control group it can-
illated between objective and subjective depending not be ascertained whether planning in Friends
on business pragmatism, cultural preferences, and was longer-run than in comparable Western com-
ethnic tensions, which made ascertaining the own- panies, as suggested by Harrison et al. (1994). The
ers’ preference for a type of information at a par- owners were not averse to long-run planning – it
ticular time difficult. For example, important just was not practical in their situation. If there
contextual information for budgets, often gar- is a Chinese tendency to long time horizons it is
nered socially within guanxi, was primarily subjec- likely to be context related.
tive but its ramifications were translated into
objective, quantified budgets. Whether the initial
reliance on subjective external information owed Conclusions
more to the Indonesian business context, ethnic
distrust, or Chinese business values is difficult to This research links overseas Chinese business-
ascertain. Also, objective systems are not necessar- men’s values to organisational controls, Confucian
ily antithetical to Chinese business values, for values, socialisation by parents, ancestral history,
example, salary increments based on length of ser- schooling, and formative experiences in the local
vice rewarded loyalty. community, especially ethnic stigmatisation. Java-
Merchant et al. (1995) argue that Chinese val- nese and Chinese Indonesian values are similar
ues such as loss of face militate against tying and, for jiaosen Chinese Indonesians, increasingly
rewards to performance. This was largely so in intertwined. Chinese values of jen, personal trust,
Friends. Except for Marketing, where ethnic dis- reputation, face and family reliance, and Javanese
trust was low, rewards were not tied to targets. values of etiquette and rukun both presume reci-
Payment-by-result systems were rejected within procal social ties underpin relationships. Both sub-
Manufacturing due to possible ethnic complica- jugate individual rights to collective interests to
tions and measurement problems. Nevertheless, maintain social order and harmony, and empha-
performance related rewards were not rejected sise social hierarchy. Javanese etiquette (linguistic
absolutely, e.g. sanctions were imposed for excess form and andap-asor) regulates behaviour accord-
defects. Merchant et al. (1995) also argue that Chi- ing to social hierarchies. Bapakism entails mutual
nese collectivist values incline owners towards obligations between leaders and subordinates,
group not individual rewards. Again, the evidence and obedience, consistent with Chinese li and hsiao
from the few result controls tied to rewards used values. Exceptions were Javanese slamatan (rituals
was mixed. The owners established individual tar- and mysticism) and religious duties. However,
gets and commissions for salesmen (who worked although the owners struggled with Javanese
independently) and personal achievement allow- speech and interaction patterns this was not a sig-
ances for other Marketing staff but elsewhere they nificant problem as they reciprocated other expec-
used group-based rewards to encourage collectiv- tations. MCS problems in multi-ethnic situations
ism, for example defect rate penalties in Manufac- are often attributed to cultural differences. How-
turing and group commission for other Marketing ever, the values of the Chinese owners and pribumi
staff. Business contingencies – not merely values employees were complementary: it was the history
drove the formulation of rewards. of state discrimination and wealth differentials that
There were no long-run incentives, contrary to fuelled ethnic tensions.
claims that Chinese values encourage them (Mer- The Chinese owners’ orientation and prefer-
chant et al., 1995). Environmental scanning was ences resided with controlling behaviour subjec-
considerable but confined to the forthcoming bud- tively (action controls) and aligning employee
get period and information was mainly subjective and employer values through cultural controls,
and socially derived as Harrison et al. (1994) though they were not averse to using result con-
argue. However, the volatile Indonesian context trols for critical economic factors, or formal
militated against longer-run more quantitative bureaucratic action controls when measurement
was feasible. This study confirms previous research Second, this research illustrates the limitations
claiming that Chinese owned businesses have low of static accounting studies that ignore how and
budget participation (Lau & Tan, 1998), centrali- why cultures interact dynamically. For example,
sation (Birnbaum & Wong, 1985; Harrison et al., in Friends the culture of Chinese members varied
1994; Lau et al., 1995), tendencies to subjective between totok and jiaosen, which reflected their
rather than objective controls (Merchant et al., assimilation of other Indonesian cultures. Chinese
1995; Redding, 1993) and, tentatively, few rewards Indonesian culture may become more jiaosen if
tied to results and greater use of group rewards political repression and ethnic conflict subsides,
(Merchant et al., 1995). Whether Chinese manag- and pluralism and integration increase in the new
ers exhibited longer term orientations concerning political climate. Also, although the owners per-
planning and rewards (Harrison et al., 1994; Mer- ceived themselves as Chinese Indonesians, they
chant et al., 1995) could not be ascertained. saw their business practices as rupturing the ‘tradi-
However, the issues transcend beyond testing tional’ Chinese methods of Mr. Ed, partly because
models of culture and accounting to the issues on their age and education had exposed them to other
page 7 expressing reservations about the conceptu- cultures, including Western life-styles, aspirations,
alising of culture in many accounting studies. The and methods of conducting business. Culture
first question alleged that most studies are too gen- stems not just from antecedents but also from con-
eral and neglect multi-culturalism. Support for this tact and conflicts with other cultures, and associ-
allegation came from this study. Indonesian cul- ated political dynamics.
tures generally and within Friends were diverse, Third, the problems of accounting studies that
and Chinese culture was not unitary. For example, neglect historical and external organisational fac-
who is Chinese was not homogenous within Indo- tors involving, inter alia, the impact of political
nesia or Friends, depending partly on the extent and economic institutions and struggles upon cul-
individuals and groups of Chinese had assimilated ture and control, are illustrated in this study. For
other indigenous cultures. In Friends cultural example, Indonesian Chinese culture relates to
diversity was marked, especially between Chinese Chinese immigration patterns and how some Chi-
and pribumi, and respondents claimed each culture nese were used politically and economically. For
varied regionally. National culture is an abstract example, the economic role ascribed to Chinese
concept that can dehumanise, deny free will, and businessmen by colonial rulers partly laid founda-
rationalise dominant groups’ discrimination tions for ethnic policies post-independence,
against minorities. For example, the threats of eth- whereby Chinese businessmen became ‘pariah
nic conflict perceived by Friend’s Chinese owners entrepreneurs’-abused and discriminated by state
stemmed from rhetoric of politicians claiming organs and politicians who used them for eco-
(however dubious empirically) that Chinese Indo- nomic ends whilst, paradoxically, affording them
nesians contaminated a notion of Indonesian cul- a degree of privilege and protection. This was asso-
ture and national identity propounded by the ciated with ‘Cold War’ politics and allegations that
state, which led to Chinese Indonesians being con- indigenous Chinese identified with mainland Com-
stituted as marginal ‘others’. As Bhimani (1999) munism (whether spurious or not). Such regional
notes, delineations of national culture can owe and global political and economic relations perco-
more to power/knowledge relations exercised by lated down to Friends. For example, the owners’
powerful institutions such as the state than beliefs beliefs were coloured by experiences at school
grounded in a community. Accounting studies and in the community, e.g. riots against indige-
have not pursued such ends but uncritical employ- nous Chinese and their property, and dealings with
ment of national culture carries this implicit dan- state officials. Ethnic tensions in Friends were a
ger. Thus when national culture is used its latent residual from previous eras. The potential
empirical veracity needs establishing with respect threats contributed to Friend’s owners fostering
to antecedents, inculcation, and incorporation into personal relations with employees, taking pains
action. to respect pribumi beliefs, cultivating employee
informers, garnering intelligence from their qin jian and ke ji in their private life, though when
guanxi, adopting bureaucratic systems, and presenting self in public they took pains to act to
employing a ‘pribumi’ mediator. However, this the contrary. Also they would sacrifice values
was dynamic. For example, Friend’s owners’ had occasionally for commercial reasons, for example:
to be creative in avoiding civil and industrial rela- formal targets tied to rewards for salesmen and
tions unrest as in the new democratic state they production personnel, and an objective wages sys-
could no longer turn to the military. tem for pribumi employees, violated beliefs about
This brings into play the fourth point of dispu- subjective performance evaluation being the own-
tation with contemporary research on culture and ers’ prerogative; and Mr. U incurring penalties
control, namely that identifying who is Chinese for production waste transgressed beliefs of collec-
outside mainland China is more problematic than tive responsibility. The Chinese owners mobilised
often assumed. This study found who is Chinese cultural preferences when their benefits appeared
was socially defined by the actors themselves and to exceed costs. Thus they adapted some modern,
those around them – not biologically determined. ‘Western’ controls that ran counter to their cul-
‘Who and what is Chinese?’ are ethnic constructs tural beliefs, though some aspects of bureaucracy
that overlap with claims to cultural distinctiveness. resonated with shared cultural values of employers
Definitions of cultural and ethnic identity are sub- and employees. Indeed, they actively sought out
jective, dynamic, complex, and unstable. As over- information and advice on Western business meth-
seas Chinese businessmen operate in multi-ethnic ods. Cultural determinism must be avoided and
societies, cultural and ethnic identities become heed paid to individual agency, choice, and con-
intertwined with indigenous cultures and context text. For example, control exercised within Friends
leading to gradations and different meanings of could owe more to company size, its technology,
being Chinese. Culture is a complex web of shift- and family ownership than culture. It is difficult
ing patterns linking people in different locales to ascertain definitively that culture, socialisation
and social formations: it is unwise to compare and action formed a causal chain despite linkages
and contrast ‘Chinese culture’ across locations being established. Also, the behaviour and values
without careful studying context and dynamics, of Friend’s owners may resemble those of ethnic
or to assume cultures are homogenous. minority immigrant businessmen elsewhere. That
According to the fifth criticism, accounting is, the values and behaviour of Chinese Indonesian
studies fail to connect values back to socialisation businessmen may owe more to being an ethnic
or forward into action. The notion of culture pre- minority operating under difficult conditions
sumes certain values are inculcated in specific rather than a distinctive cultural heritage (Light
social formations. However, accounting studies & Rosenstein, 1995). This is difficult to refute cat-
do not demonstrate how cultural values are egorically for it was difficult to discriminate
imbued by socialisation in the family and educa- between the effects of ethnic tensions and Confu-
tion systems, and social experiences within and cian values – they were intertwined. Demonstrat-
outside the workplace (Bhimani, 1999). In con- ing links to parental socialisation helped, prima
trast, this study established a prima facie link facie, to ascertain links to heritage but this was
between the owners’ beliefs, socialisation in Chi- also linked to ethnic experiences overseas. Further
nese families, and Chinese ancestry. However, this comparative research, more sensitive to local con-
must be treated cautiously for cultures may not be text, is required.
enacted within MCSs (Harrison & McKinnon, The last criticism accused accounting studies of
1999), values vary between core and peripheral, undue reliance on problematical concepts and sur-
and they can be drawn on selectively according vey research instruments derived from Hofstede,
to the situation. Thus overseas Chinese business- often in tandem with problematical ones from
men’s beliefs and actions may be contingent and contingency theory and research on accounting
only intermittently bear upon control locally. For performance measurement (Baskerville, 2003;
example, the owners would transgress values of Chenhall, 2003; Harrison & McKinnon, 1999;
McSweeney, 2002), and neglecting building con- ology, and the ‘New’ (especially Foucauldian) His-
cepts and factors bottom up through field studies. tory as Bhimani (1999) commends but we are
On close reading leading contributions on Chinese unconvinced that this is the optimal strategy as
culture and MCS (e.g. Awathsi et al., 1998; Mer- his proposals, like his targets of criticism, are
chant et al., 1995; O’Connor, 1995; Tsui, 2001) accounting-centric for they draw mainly from
do link Confucian values to control in their litera- extant accounting research. Our plea is much sim-
ture review but they then use Hofstede’s five pler – we simply wish accounting studies involving
dimensions and measures of national cultures to culture to engage more directly with the rich tapes-
identify Chinese cultural attributes. But the con- try of theories and methods in anthropology and
nection of Hofstede’s measures to overseas Chi- ethnography, of which Bhimani’s commendations
nese culture is surmise and not empirically are a part, albeit a vital one.
grounded and accounting studies do likewise.
The struggle to link Chinese culture to MCS fac-
tors may be because Hofstede’s tools are inade- Acknowledgements
quate surrogates of Chinese culture, which is
formed by a rich mosaic of factors. ‘Hofstede’ Our thanks to the insightful comments of Wai
studies, given their simple and generalised cultural Fong Chua, David Cooper, Colwyn Jones, Kari
dimensions, and their static and non-contextual Lukka, Cyrille Sardais; colleagues at the Global
nature, may be ineffective for extending work in Management Accounting Research Conference,
this area. Michigan State University, 2004; the International
Lastly, like any case study, our research has Management Accounting III Conference, Univer-
methodological limitations. Nevertheless, we hope siti Kebangsaan, Malaysia, 2004; University of
it demonstrates how combining ethnographic New South Wales, 2005, and HEC, Paris, 2005;
methods of data collection with a version of and the two anonymous referees; and help on liter-
grounded theory for data analysis can combine ature sources from Graeme Harrison and Fadi
etic and emic insights on culture and MCSs. The Alkaraan. The usual disclaimers apply.
aim is to complement and develop more conven-
tional survey work – not to discount it. However,
survey research has employed dubious etic catego- Appendix. Using grounded theory to analyse data
ries and questionable research instruments arms-
length from subjects with little regard to context Grounded theory helped interpret, verify, orga-
and action. But, as Bhimani (1999) notes, it has nise and conceptualise data, and produce theoret-
entered ideational territory: cultural contingency ical propositions. Grounded theory was developed
research cannot divorce itself from methodological by two sociologists, Barney Glaser and Anselm
and research method issues that ensue. Ethno- Strauss in the 1960s but their methodological
graphic grounded studies can identify issues and approaches subsequently diverged (Parker & Rof-
concepts bottom-up to improve surveys but they fey, 1997). Glaser advocates specifying research
are more than pilot stages. They offer rich detail questions from actors’ perceptions after site entry
that surveys or laboratory studies cannot capture whereas Strauss and Corbin commend determin-
that can challenge, enrich, and provide insight into ing the topic of enquiry beforehand. Glaser prefers
conflicting or inconclusive results elsewhere. All is analytical methods that allow research problems
not lost for etic inclined cultural contingency to emerge naturally by constantly comparing inci-
researchers but they must incorporate emic issues. dents whereas Strauss and Corbin advocate more
We hope more researchers in this vein triangulate structured analyses arguing that researchers need
theory and methods and avoid privileging the specific, systematic guidance for interpretation.
quantitative over the qualitative or the etic over Finally, Glaser leaves testing and verification of
the emic. Such work could beneficially incorporate emerging concepts to researchers using quantita-
theories of Societal Effects, New Institutional Soci- tive methods, while Strauss and Corbin advocate
provisional qualitative testing of concepts and the- During interviews with the owners about the
ory during the research. This research, whilst con- MCS, they were asked about their cultural values,
ceding merits of Glaser’s approach for emic social backgrounds, the history of their friend-
research, opted for Strauss and Corbin’s methods ship, their perception of pribumi and Chinese,
in the belief that it is impractical (perhaps impossi- their business philosophies, etc. to derive their
ble) to enter the field without theoretical socio-cultural thoughts, i.e. their emic viewpoint.
constructs. Some employees were also interviewed then about
According to Strauss and Corbin (1998), their opinions of the company and ethnic beliefs
grounded theory is generated by continuous inter- to garner their emic perspective. This was the
play between data collection and analysis based most difficult part because the researcher had to
on asking questions and making comparisons. be careful in presenting himself and show empa-
The latter compares properties and dimensions of thy. The researcher acted differently during inter-
concepts emerging from the data with specific inci- views with the company’s owners and employees.
dents, objects or actions derived from prior litera- With the owners, the speech and body language
ture, or the researcher’s experiences, to enable the tried to project being more of a management
meaning of concepts to be examined and devel- advisor to maintain openness. In contrast, with
oped. The initial interviews explored the owners’ employees, the emphasis was on presenting the
involvement in the company and its history. Then, self as an academic researcher to gain trust. Aca-
they were asked how they performed management demicians have high status with Indonesian peo-
control. However, rather than use this term (which ple, especially since the 1998 political reforms.
did not exist in their lexicon) phrases were used such However, all subjects knew the researcher was
as, monitoring employees, observing employee both a management advisor and academic
behaviour, assessing employees’ performance, researcher.
motivating employees, rewards and punishment, Concurrently, some interviews were performed
allocating tasks. Information on the owners’ and with some Chinese businessmen and pribumi intel-
key employees’ roles in these processes, who helped lectuals outside the company about their personal
them, and why, helped incorporate an emic view. experiences of Chinese-pribumi interactions for
Then, answers and personal observations were comparison and validation, especially of the essen-
combined and categorised according to Merchant’s tially etic models derived from literature on Confu-
division of action, result and personnel/cultural cianism and Chinese Indonesian ethnicity. This
controls to derive the initial ‘‘model’’ of the MCS provided a bridge to make sense of the emic views
(etic view) and rudimentary causal relations. This of participants, for example, to understand why
provided the initial categories to classify the next the owners considered ethnicity of applicants was
interview questions. The initial interviews and cate- important when recruiting, and why some workers
gories resemble Strauss and Corbin’s open coding viewed the owners with suspicion. Sometimes ten-
processes. tative conclusions were fed back to the owners and
Next, some employees were interviewed and employees, e.g. ‘‘is it true that you do not trust
observations conducted to validate this model. them because of these reasons?’’ Thus the process
Where there were discrepancies or inconsistencies was as follows: emic view ! making sense of this
further interviews helped revise the model and through etic claims ! gaining confirmation from
incorporate emic data. Some memos were then the actors (reconciling etic and emic views). How-
made to trace relationships amongst categories ever, in some cases, this proved impossible. For
(Strauss and Corbin’s axial coding). For example, example one could not ask workers ‘‘do you sus-
what was the relationship between financial penal- pect him because he is a Chinese?’’ They would
ties and the roles of Mr. U and Mr. H? The results not dare answer for fear of giving offence. But
helped determine the next interview questions. some confirmation could be gleaned from key
These processes were performed iteratively and employees close to workers such as Mr. U and
continuously. Cik K.
The interviews were also the basis for categoris- apply them flexibly and creatively to their own
ing the influence of culture and ethnicity upon the materials.’’ (ibid, 1998, p. 14). However, we wish
owners’ beliefs and actions using open coding. to make it clear that we did not conduct a full
Interview results were continuously integrated to grounded theory study for we did not produce
the ‘‘model’’ of MCS via axial coding from ques- comprehensive theory. What we did is apply
tions about the owners’ values, and their percep- grounded theory methods to analyse data
tion of how ethnic differences influenced their (grounded analysis).
exercise of management control. There was no
clear boundary between open and axial coding.
Sometimes answers about relationships amongst References
categories (axial coding) would come accidentally
when seeking descriptions of a specific category Ahrens, T. (1997). Talking accounting: An ethnography of
(open coding) and vice versa. After refinements management knowledge in British and German Brewers.
using selective coding (here findings not estab- Accounting, Organizations and Society, 22, 617–637.
Alvesson, M., & Skoldberg, K. (2000). Reflexive methodology.
lished as related to the research questions and London: Sage.
resultant model are stripped out) a general picture Amir, M. S. (1997). The non-pribumi and social justice. In I.
of the phenomenon under study was derived. Dur- Chalmers & V. R. Hadi (Eds.), The politics of economic
ing the validation and refinement of this during development in Indonesia (pp. 211–213). London: Routledge.
further interviews and observations some new Ansari, S. L., & Bell, J. (1991). Symbolism, collectivism and
rationality in organizational control. Accounting, Auditing
findings about new categories (open coding) and and Accountability Journal, 4, 4–27.
new relationships among categories (axial coding) Awathsi, V. N., Chow, Chee W., & Wu, A. (1998). Performance
emerged but not frequently. measures and resource expenditure choices in a teamwork
Documentary analysis was only used in the environment: The effects of national culture. Management
beginning of the study. Incorporation of the emic Accounting Research, 9, 119–138.
Baskerville, R. F. (2003). Hofstede never studied culture.
was crucial during interviews for many native con- Accounting, Organizations and Society, 28, 1–14.
cepts, terms, and thoughts appeared to lie outside Baskerville-Morley, R. F. (2005). A research note: The unfin-
the conventional boundaries of the MCS litera- ished business of culture. Accounting, Organizations and
ture. The etic view was crucial for initially framing Society, 30, 389–392.
the research and conveying the views of actors and Baszanger, I., & Dodier, N. (2004). Ethnography: Relating the
part to the whole. In D. Silverman (Ed.), Qualitative
making sense of what happened to non-Indonesian research: Theory, method and practice (2nd ed.). London:
readers. However, some data related to culture Sage.
and ethnicity was removed during the refinement Beatty, A. (1999). Varieties of Javanese religion: An anthropo-
of the ‘‘model’’/general picture during selective logical account. Cambridge: Cambridge University Press.
coding as it had no direct relationships with the Bhimani, A. (1999). Mapping methodological frontiers in cross-
national management control research. Accounting, Orga-
MCS. The theoretical reviews that formed etic nizations and Society, 24, 413–440.
models of MCS, Confucianism, Javanese culture, Binawas, D. (2000). Data direktorat persyaratan tenaga kerja.
and Chinese Indonesian ethnicity proved to be a Republik Indonesia: Departemen Tenaga Kerja.
two-edged sword. They helped the interviews to Birnbaum, P. H., & Wong, G. Y. Y. (1985). Organizational
be done more systematically but, to some extent, structure of multinational banks in Hong Kong from a
culture-free perspective. Administrative Science Quarterly,
they also may have restricted the ‘‘model’’/general 30, 262–277.
picture that emerged despite the emic analysis. Blusse, L. (1990). The role of Indonesian Chinese in shaping
Do the methods depart from Strauss and Cor- modern Indonesian life: A conference in retrospect. In
bin? We certainly made adjustments to make the Proceedings in the symposium the role of the Indonesian
study feasible. This may be within Strauss and Chinese in shaping modern Indonesian life, July 13–15, 1990.
Cornell University and Southeast Asian Studies Summer
Corbin’s intent for they acknowledge that their Institute.
book is to make students: ‘‘. . . understand the pur- Bourdieu, P., Chamboredon, J.-C., & Passeron, J.-C. (1991).
pose of procedures rather than to memorize the The craft of sociology: Epistemological preliminaries. Berlin:
techniques themselves . . . they should be able to de Gruyter.
Brown, D. (1994). The state and ethnic politics in Southeast Asia. Heryanto, A. (1998). Ethnic identities and erasure: Chinese
London: Routledge. Indonesians in public culture. In Southeast Asian identities:
Burgess, R. G. (1984). In the field: An introduction to field Culture and the politics of representation in Indonesia,
research. London: George Allen and Unwin. Malaysia, Singapore, and Thailand (pp. 95–114). Singapore:
Chalmers, I. (1997). Introduction. In I. Chalmers & V. R Institute of Southeast Asian Studies.
Hadiz (Eds.), The politics of economic development in Hill, H. (1996). The Indonesian economy since 1966: Southeast
Indonesia: Contending perspectives (pp. 1–35). London: Asia’s emerging giant. Cambridge: Cambridge University
Routledge. Press.
Chenhall, R. H. (2003). Management control systems design Hill, H., & Mackie, J. (1994). Introduction. In H. Hill (Ed.),
within its organizational context: Findings from contin- Indonesia’s new order: The dynamics of socio-economic
gency-based research and directions for the future. Account- transformation (pp. xxii–xxxv). St. Leonards: Allen &
ing, Organizations and Society, 28, 127–168. Unwin.
Cleary, T. (1992). The essential Confucius. San Francisco: Hofstede, G. H. (1980). Culture’s consequences: International
Harper Collins. differences in work-related values. Beverly Hills, CA: Sage.
Coppel, C. A. (1983). Indonesian Chinese in crisis. Singapore: Hofstede, G. H. (2003). What is culture? A reply to
Oxford University Press. Baskerville. Accounting, Organizations and Society, 28,
Davila, T. (2005). An exploratory study on the emergence of 811–813.
management control systems: Formalizing human resources Hofstede, G. H., & Bond, M. H. (1984). Hofstede’s culture
in small growing firms. Accounting, Organizations and dimensions: An independent validation using Rokeach’s
Society, 30, 223–248. value survey. Journal of Cross-Cultural Psychology,
Eriksen, T. H. (1993). Ethnicity and nationalism: Anthropolog- 417–433.
ical perspectives. London: Pluto Press. Hopper, T., & Powell, A. (1985). Making sense of research into
Euske, K. J., & Riccaboni, A. (1999). Stability to profitability: the organizational and social aspects of management
Managing interdependencies to meet a new environment. accounting: A review of its underlying assumptions. Journal
Accounting, Organizations and Society, 24, 463–481. of Management Studies, 22, 429–465.
Fei, X. (1939). Peasant life in Communist China. London: Kwik, K. G. (1997). The myth of Chinese economic dominance.
Routledge. In I. Chalmers & V. R. Hadiz (Eds.), The politics of
Fenton, S. (1999). Ethnicity: Racism, class, and culture. Lon- economic development in Indonesia: Contending perspectives
don: Macmillan. (pp. 208–211). London: Routledge.
Fryer, D. W., & Jackson, J. C. (1977). Indonesia. London: Lachman, R., Nedd, A., & Hinings, B. (1994). Analyzing cross-
Ernest Benn Ltd. national management and organizations: A theoretical
Geertz, C. (1960). The religion of Java. Chicago: University of framework. Management Science, 25, 40–55.
Chicago Press. Lau, C. M., Low, L. C., & Eggleton, I. R. C. (1995). The impact
Geertz, H. (1961). The Javanese family: A study of kinship and of reliance on accounting performance measures on job-
socialization. New York: Free Press. related tension and managerial performance: Additional
Hadiz, V. R. (1997). Pribumi-Chinese relations: An introduc- evidence. Accounting, Organizations and Society, 20,
tion. In I. Chalmers & V. R. Hadiz (Eds.), The politics of 359–381.
economic development in Indonesia: Contending perspectives Lau, C. M., & Tan, J. J. (1998). The impact of budget emphasis,
(pp. 204–208). London: Routledge. participation and task difficulty on managerial performance:
Hammersley, M., & Atkinson, P. (1995). Ethnography: Princi- A cross-cultural study of the financial services sector.
ples in practice (2nd ed.). London: Routledge. Management Accounting Research, 9, 163–183.
Harrison, G. L. (1993). Reliance on accounting performance Leong, T. C., & Lim, T. S. (1992). Indonesia: Business and
measures in superior evaluative style: The influence of investment opportunities. Singapore: Cassia Communication.
national culture and personality. Accounting, Organizations Light, I. H., & Rosenstein, C. (1995). Race, ethnicity and
and Society, 18, 319–339. entrepeneurship. New York: Aldine.
Harrison, G. L., & McKinnon, J. L. (1999). Cross-cultural Mackie, J. (1992). Changing patterns of Chinese big business in
research in management control systems design: A review of Southeast Asia. In R. McVey (Ed.), Southeast Asian
the current state. Accounting, Organizations and Society, 24, capitalists. Ithaca, NY: Southeast Asia Program, Cornell
483–506. University.
Harrison, G. L., McKinnon, J. L., Panchapakesan, S., & Marshall, G. (1998). Oxford Dictionary of Sociology. Oxford
Leung, S. (1994). The influence of culture on organizational University Press.
design and planning and control in Australia and the United Mason, J. (1996). Qualitative researching. London: Sage.
States compared with Singapore and Hong Kong. Journal of McSweeney, B. (2002). Hofstede’s model of national cultural
International Financial Management and Accounting, 5, differences and consequences: A triumph of faith – a failure
242–262. of analysis. Human Relations, 55, 89–118.
Heryanto, A. (1997). Silence in Indonesian literary discourse: Merchant, K. A. (1998). Modern management control systems.
The case of the Indonesian Chinese. Sojourn, 12, 26–45. Upper Saddle River, NJ: Prentice-Hall.
Merchant, K. A., Chow, C. W., & Wu, A. (1995). Measure- Spradley, J. P. (1980). Participant observation. Orlando: Har-
ment, evaluation and reward of profit center managers: A court Brace.
cross-cultural field study. Accounting, Organizations and Strauss, A., & Corbin, J. (1998). Basics of qualitative research:
Society, 20, 619–639. techniques and procedures for developing grounded theory.
Ning, H. (1997). The struggles of a pribumi entrepreneur. In I. Newbury Park, CA: Sage.
Chalmers & V. R. Hadiz (Eds.), The politics of economic Suryadinata, L. (1978). Confucianism in Indonesia: Past and
development in Indonesia: Contending perspective present. In L. Suryadinata (Ed.), The Chinese minority in
(pp. 218–222). London: Routledge. Indonesia: Seven papers. Singapore: Chopmen Enterprises.
O’Connor, N. J. (1995). The influence of organizational culture Tsui, J. S. L. (2001). The impact of culture on the relationship
on the usefulness of budget participation by Singaporean– between budgetary participation, management accounting
Chinese managers. Accounting, Organizations and Society, systems and managerial performance. The International
20, 383–403. Journal of Accounting, 3, 125–146.
Ong, A. (1999). Flexible citizenship: The cultural logics of Wickramasinghe, D., & Hopper, T. M. (2005). A cultural
transnationality. London: Duke University Press. political economy of management accounting controls: A
Ouchi, W. G. (1979). A conceptual framework for the design of case study of a textile mill in a traditional Sinhalese village.
organizational control mechanisms. Management Science, Critical Perspectives on Accounting, 16, 473–503.
25, 833–848. Wickramasinghe, D., Hopper, T. M., & Rathnasiri, C. (2004).
Ouchi, W. G. (1980). Markets, bureaucracies, and clans. Japanese cost management meets Sri Lankan politics: State
Administrative Science Quarterly, 25, 129–141. regulation and management controls in a privatised utility.
Parker, L. D., & Roffey, B. H. (1997). Back to the drawing The Accounting, Auditing, and Accountability Journal, 17,
board: Revisiting grounded theory and the everyday 85–120.
accountant’s and manager’s reality. Accounting, Auditing Wilk, R. R. (1996). Economies and cultures: An introduction to
and Accountability Journal, 10(2), 212–247. economic anthropology. Oxford: Westview.
Rademakers, M. F. L. (1998). Market organization in Indone- Xu, Z. G. (2000). The analects of Confucius (Lun Yu Tong Yi).
sia: Javanese and Chinese family business in the Jamu Beijing: People’s Literatures Publisher (Ren Ming Wen Xue
Industry. Organization Studies, 19, 1005–1027. Chu Ban She).
Redding, S. G. (1993). The spirit of Chinese capitalism. New Yeung, H. W. C. (1999). Internationalisation of ethnic Chinese
York: Walter de Gruyter. business firms from Southeast Asia. International Journal of
Redding, S. G., & Whitley, R. D. (1990). Beyond bureaucracy: Urban and Regional Research, 23, 103–127.
towards a comparative analysis of forms of economic
resource coordination and control. In S. R. Clegg & S. G.
Redding (Eds.), Capitalism in contrasting cultures
(pp. 79–104). New York: Walter de Gruyter.
Glossary of Indonesian and Chinese terms
Rickleffs, M. C. (1993). A history of modern Indonesia since c.
1300 (2nd ed.). London: Macmillan. Abangan: Javanese people who are nominal Moslem and
Robison, R. (1986). Indonesia: The rise of capital. North mainly peasants.
Sydney, Australia: Allen & Unwin. Alus: Javanese cultural concept meaning pure, refined, pol-
Sanday, P. R. (1979). The ethnographic paradigm(s). In J. Van ished, polite, exquisite, ethereal, subtle, civilised and
Maanen (Ed.), Qualitative methodology. London: Sage. smooth.
Scapens, R. W., & Macintosh, N. B. (1990). Structuration Andap-asor: Javanese cultural concept meaning to humble
theory in management accounting. Accounting, Organiza- oneself politely towards anyone who of either approxi-
tions and Society, 15, 447–455. mately equal or higher rank.
Scapens, R. W., & Macintosh, N. B. (1996). Structure and Bakorstanas: Coordinating Agency for National Stability – a
agency in management accounting research: A response to powerful semi-military state agency during the New Order
Boland’s interpretive act. Accounting, Organizations and era dealing with subversion and other affairs that may cause
Society, 21, 675–690. public disorder and instability. However, the definition of
Scupin, R. (1998). Cultural anthropology: A global perspective subversion was never clear. Consequently, the agency often
(3rd ed.). Upper Saddle River, NJ: Prentice Hall. protected the interests of powerful state elites and to abuse
Silverman, D. (2004). Introduction. In D. Silverman (Ed.), and extort. The new government of Indonesia has dissolved
Qualitative research: Theory, method and practice (2nd ed.). it.
London: Sage. Bapakism: Javanese form of paternalism/patronage.
Skinner, G. W. (1963). The Chinese minority. In R. T. McVey Batin: Javanese cultural concept meaning the inner realm of
(Ed.), Indonesia. New Haven: Southeast Asia Studies, Yale human experience.
University. Cengli: Colloquial adjective term originally from Hokkian (a
Smith, S., & Young, P. D. (1998). Cultural anthropology: Chinese dialect) used by Chinese Indonesians to describe
Understanding a world in transition. Boston, MA: Allyn and something/someone that is plausible, fair, reasonable or
Bacon. just.
Chinese Indonesian: Ethnic Chinese in Indonesia.Chinese Pribumi: Political term used widely in popular vernacular,
Indonesian rather than Indonesian Chinese is used to stress coined during Dutch colonialism meaning indigenous
the general political affiliation of the ethnic group that Indonesian ethnic groups. It was used extensively during the
identifies itself as Indonesian people with Chinese origin New Order era to exclude and discriminate against ethnic
rather than part of the pan-Chinese global society. Chinese to imply foreign origin. Its common usage over-
Cik: Colloquial Hokkian for elder sister used informally by looks that groups deemed indigenous are, at least partially,
Chinese Indonesians and other ethnic groups for elder, se- descended from various ethnic groups, including early Chi-
nior Chinese women to denote respect and a close rela- nese immigrants.
tionship in an informal situation. Rukun: Javanese cultural concept meaning a state of agree-
Cino: Colloquial Javanese term with negative and racial con- ment of unanimity in a group concerning its means and
notation commonly used to express hatred or derogatory purposes.
feelings towards Chinese people. Sam Kok: Classical story describing hundreds of peoples’ per-
Confucianism: A Chinese philosophy concerned with social, sonal lives and characters in detail during the end of the
moral and governmental problems emphasising social order Han dynasty in China and the rise of three kingdoms.
and harmony. Santri: Javanese people who are pious, orthodox-leaning
Guanxi: Ethnic based trust network among ethnic Chinese. Moslem.
Hokkian: A segment of the Chinese Indonesian community Slametan: Communal feast that symbolises mystic and social
originally from Fukkien-China that has particular dialect. unity of its participants given at celebrations, ameliorations,
Hsiao: Filial piety. A Confucian concept of vertical relation- and sanctifications such as birth, marriage, sorcery, death,
ship between parents and children. house moving, harvest, opening a factory, illness, supplica-
Jiaosen: Chinese Indonesians who adapt better to local cultures tion of guardian spirit, and circumcision. The ritual’s pur-
and customs. Their culture is a mixture of Western, local pose in Javanese society is to create and maintain social
and Chinese/Confucian cultures and customs rather than harmony (rukun).
pure traditional Chinese. Tamsunya: Colloquial Hokkian for attitude of always fulfilling
Javanese: Largest ethnic group in Indonesia mainly occupying promises whatever the consequences.
the island of Java (Jawa). Tao: Confucian concept meaning way or moral law.
Jen: Confucian concept meaning to love man. It consists of Tenglang: Used by Chinese Indonesians to distinguish them-
chung (sincerity and honesty) and shu (altruism). selves from other ethnic groups in Indonesia. It connotes
Juaji: Colloquial Hokkian term for someone who is stubborn, our people or friend.
disobedient, and selfish. Totok (old meaning): Chinese Indonesians who continue to
Kasar: Javanese cultural concept meaning impolite, rough, speak their ancestral language, practice traditional Chinese
uncivilised. customs, and are oriented sentimentally and politically to
Kyai: A traditional Javanese Moslem leader. China/Taiwan.
Lair: Javanese cultural concept meaning the outer realm of Totok (new meaning): Chinese Indonesians who practise tra-
human behaviour. ditional Chinese/Confucian values, speak Mandarin/other
Li: Confucian concept of individual’s ascriptive roles in a Chinese dialects, educate their children according to Chinese
society. values, and celebrate Chinese traditional events.
Mbo-Hwat: Colloquial Hokkian term that means that nothing Tunjangan Hari Raya (THR): Allowance given to employees
can be done to prevent bad things occurring. to celebrate one of their main religious days.
Nasi Tumpeng: Traditional Javanese meal for special occasions Upah Minimum Regiona (UMR): The Indonesian government
made from a large cone of cooked rice and side dishes such minimum wage reviewed annually. It was Rp 328,000 per
as chicken, vegetables, and eggs. The top of the pyramid is month (around £28) in November 2001 for the Surabaya
given to the person regarded as the leader. area.
Penunggu: Javanese term meaning a spirit/ghost guarding a Warung: Traditional Javanese tent cafe selling coffee, tea, tra-
place. Many people in Indonesia believe that every place has ditional cakes, and a limited variety of meals. Poorer people
at least one. use it for social gatherings.

A10 Efferin&Powell2007 Ethnography

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

A10 Efferin&Powell2007 Ethnography

Uploaded by

Copyright:

Available Formats

Accounting, Organizations and Society 32 (2007) 223–262

Management control, culture and ethnicity

5. Group-based rewards due to values of collectiv- of human behaviour’’ refers to an individual’s

Confucian Values Ethnicity of Chinese Javanese Values

Management Control System

• Personnel and action controls not result controls Other factors

Fig. 1. Management control in a Chinese Indonesian business.

Petty Accounting Security Maintenance Production Ware housing Customer Sales

Fig. 2. Friends company–organisation structure.

You might also like

A10 Efferin&amp;Powell2007 Ethnography

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

A10 Efferin&amp;Powell2007 Ethnography

Uploaded by

Copyright:

Available Formats

Accounting, Organizations and Society 32 (2007) 223–262

Management control, culture and ethnicity

5. Group-based rewards due to values of collectiv- of human behaviour’’ refers to an individual’s

Confucian Values Ethnicity of Chinese Javanese Values

Management Control System

• Personnel and action controls not result controls Other factors

Fig. 1. Management control in a Chinese Indonesian business.

Petty Accounting Security Maintenance Production Ware housing Customer Sales

Fig. 2. Friends company–organisation structure.

You might also like

A10 Efferin&Powell2007 Ethnography

A10 Efferin&Powell2007 Ethnography