Cronyism is the soft form of criminal conspiracy, but it’s also not that far removed from the groovier, more celebrated forms of social networking. –An Omniorthogonal blogger (Social Crapital, 2005).

Social networks are closely woven into every aspect of our daily existence. Virtually all economic behavior in modern life is embedded in networks of social relations that condition economic processes (Burt, 1992; Meuer & Krug, 2007; Uzzi, 1997; Sherwood, 2007). The ubiquity of networks, and networking, at the industry, firm, group, and individual levels has attracted significant research attention (see Borgatti & Foster, 2003 for a recent review; Dacin & Delios, 2005 for a special issue on networks in Asia Pacific business in the Asia Pacific Journal of Management; and Parkhe, Wasserman, & Ralston, 2006 for a special issue on network theory development in the Academy of Management Review). Social network theory has become widely used to study issues on organizations, business groups, and strategic alliances (Dacin & Delios, 2005).

While the literature in economics and finance tends to view social networking as dysfunctional, the organizational theory and strategy literatures have focused primarily on the many benefits social networking offers, such as easy access to information, mutual support, cooperation, and trust. We believe that a theory of effective networks requires an understanding of both their upside and downside. In order to address the neglected downside aspect of social networks in the organizational theory and strategy literatures, we discuss in this paper one crucial negative consequence of social networking—cronyism, which means preferential treatment shown to old friends and associates without regard to their qualifications. Cronyism seems natural and inherent in a network as per the definition of network in the Oxford English Dictionary (1989): “an interconnected group of people or organizations having certain connections which may be exploited to gain preferment, information, etc.” Cronyism in social networks may result in their parasitic, mafia-like tendencies that many times may be anti-market and anti-institutional (Brown, 2006; Sherwood, 2007).

A common conception in the West which has held cronyism as a feature endemic to developing countries, particularly in the collectivist cultures of Asia, received support from its apparent manifestation as a cause of the late 1990s Asian financial crisis (Economist, 1998). However, that conception was challenged by the scale of subsequent financial scandals in the West, which often were built on networks of collusion (Davis, 2003).Footnote 1 The lack of academic literature on this important subject contrasts sharply with desires voiced by business and government leaders, in both the East and the West, to limit ethical and legal violations arising from excesses of favoritism in organizational decision making. We seek to redress that gap by discussing how certain network characteristics may affect the occurrence of cronyism.

This paper is organized as follows. Before focusing on cronyism, the next section first reviews some general negative consequences of social networking. It is followed by a section defining cronyism as a form of social exchange. Illustrated in Figure 1, we argue that (1) type of network, clique versus entrepreneurial, and (2) type of competition, inter- versus intra-network, create different patterns of social exchange that lead to different likelihoods of cronyism.

Figure 1
figure 1

A model of networks and cronyism

Full size image

The downside of social networking

Social networking is a very powerful tool, which can be directed toward benevolent as well as malevolent purposes. While the organizational theory and strategy literatures have often focused on the positive consequences of social networking, this section succinctly reviews the negative ones. Sherwood (2007) argued that, in many societies, the institution of social network transacting has served reasonably well as a judicial system substitute and transacting of many kinds is conducted at a relatively low cost (Steier, 2009). This function of social networking is particularly salient in emerging economies (Ren, Au, & Birtch, 2009; Zhang & Ma, 2009). For example, management and strategy scholars in the Asia Pacific region have identified at least two types of basic strategies that influence firm growth in the emerging economies of Asia: (1) a network-centered strategy that utilizes relational capabilities embedded in managers’ external networks—often known as managerial ties (Peng & Luo, 2000) and (2) a market-centered strategy that focuses on firm resources located within the firm’s boundary (Zhang & Li, 2008). Firms in institutionally less developed economies have been suggested to rely more on the network-centered strategy in order to deal with environmental uncertainties and gain access to external resources, such as knowledge for innovation (Gao, Xu, & Wang, 2008). In economies with a better institutional infrastructure, the market-centered strategy relying on the resource-based logic of competitive advantage, which orients the firm toward the efficient enhancement of internal resources, is thought to be more effective. For example, Dieleman and Sachs’ (2006) intensive case study of the Slim Group in Indonesia found that the group’s strategy moved from the extreme of crony capitalism (the relationship-based model) to the existing Western norms for multinational business (the market-based model) as the institutional environment of Indonesia became more developed.

However, social network transacting as a substitute for judicial system may constrain economic development in numerous ways, many of which are little recognized. In an economy dependent on transacting within social networks, intra-network transacting works effectively while inter-network transacting does not, resulting in a fragmented, crony capitalistic economic system (Carney, 2008). There are many other constraints posed by social network transacting, such as proliferation of entry barriers, skewed resource allocation, scarce specialization, diminished choices, exclusion of better qualified actors, restrained competition, poor product quality, and hardening of social stratification (Sherwood, 2007).

Furthermore, it can be costly in terms of money and time to develop and maintain a social network. In the case of guanxi, for example, Chen and Chen (2004) argued that before one can use guanxi, great efforts are needed in initiating and building it. Despite the cost, Wu and Leung’s (2005) study of small and medium sized enterprises in China failed to find any significant effect of managerial ties on competitiveness improvement or overall firm performance. In particular, network ties may not improve firm performance if they are overembedded, because the costs incurred to create and maintain these ties may offset any potential benefits. For instance, Peng, Au, and Wang (2001) studied interlocking directorates in Thailand and concluded that, in an economy largely controlled by ethnic Chinese, having Chinese directors on the board, which may be viewed as networks that are overembedded, did not constitute valuable, unique, and hard-to-imitate resources by and in themselves. Kirchmaier and Stathopoulos (2008) contrasted social network theory, which postulates a positive relationship between social networks and firm performance, and the managerial power approach, which postulates a negative relationship, using a sample of 363 non-financial firms in the UK. The findings of the study lent support to the managerial power approach over social network theory in that the costs of maintaining CEO social networks outweighed the benefits for the firm, with most of the benefits being accrued privately by the CEO, such as higher social status. Further, Kirchmaier and Stathopoulos (2008) found that boards systematically overestimated the value of social networks when hiring a new CEO, and that the expected benefits never materialized as boards failed to adequately monitor highly connected CEOs.

Cronyism as social exchange

Although cronyism is an oft-used term in politics as a downside aspect of social networking, its manifestations are not limited to that sphere. Research suggests that modern organizations are political arenas (Drory & Romm, 1990) and cronyism can pervade private organizations. Khatri and colleagues (Khatri & Tsang, 2003; Khatri, Tsang, & Begley, 2006) have developed a conception of cronyism and attempted to explain variation in rates and types of cronyism across cultures. They argue that cultures may create conditions for cronyism to flourish, but they do not identify the mechanisms that turn cultural propensity into individual acts of cronyism. We posit that networks provide this mechanism. In particular, we hold that type of network, clique or entrepreneurial, and form of network competition, inter- or intra-network, affect the likelihood of cronyist activity.

Khatri et al. (2006: 62) defined cronyism as “a reciprocal exchange transaction where party A shows favor to party B based on shared membership in a social network at the expense of party C’s equal or superior claim to the valued resource.” For cronyism to exist, four conditions must be satisfied: (1) no immediate return of favor, (2) something of value exchanged, (3) shared network membership, and (4) at a third party’s expense. Cronyism is rooted in networks of complex, indirect, and mutually reinforcing social exchanges among actors who undertake implicit, unspecified, and reciprocal transactions with no stipulated time period during which favors must be returned (Heidenheimer, 2002). As argued by Khatri and Tsang (2003) and Khatri et al. (2006), cronyism is distinct from concepts of a similar meaning, such as corruption, nepotism, and guanxi.

According to the network perspective, actors, whether individuals, groups or organizations, are embedded within sets of interconnected relationships that frame their behavior (Brass, Galaskiewicz, Greve, & Tsai, 2004). Contractor, Wasserman, and Faust (2006) define a social network as consisting of a set of actors and one or more relations between the actors. Actors may be any kind of meaningful social unit, including individuals, collective entities, firms, organizations, and divisions within organizations. The relations may be any kind of linkages between actors, including formal role relations, affective expressions (friendship, respect), social interactions, workflows, transfers of material resources (money, goods), flows of nonmaterial resources (information, advice), and business alliances. In its simplest form, a social network is a map of all the relevant ties between the nodes being studied.

People have used the social network metaphor for over a century to connote complex sets of relationships between members of social systems at all levels, from interpersonal to societal. Rather than treating individuals (persons, groups, organizations, states) as discrete units of analysis as is normally done in organizational research, social network analysis focuses on how the structure of ties affects individuals and their relationships. Thus, various levels in social network analysis intermingle, rendering specification of clear levels of analysis somewhat problematic.

A basic principle of network analysis is that “the structure of social relations determines the content of those relations” (Mizruchi, 1994: 330). Scholars have argued that the open-ended, relational features of networks enhance the ability of firms to learn as well as transmit knowledge and skills for innovation (Inkpen & Tsang, 2005; Park, 1996) and to manage in the face of demand uncertainty, rapid product innovation, and hypercompetition (Uzzi, 1997). Networks seem to work well in imperfect or quasi-markets where institutional rules and mechanisms fail to protect the interests of individuals (Burt, 1992; Peng & Luo, 2000). Networks appear to have powerful effects in several important areas of inquiry.

Although network analysis offers a strong tool for examining structures of relations, it is weak in addressing their processes. Social exchange theory provides a possible link. Exchange theory conceives of interpersonal interaction as social exchange and posits that actors form social relations based on the benefits and costs they provide one another (Blau, 1964; Emerson, 1972a, b). The theory has two primary tenets: (1) all behavior is motivated by rewards and punishments, and (2) most interactions consist of the exchange of valued (though not necessarily material) resources (Cook & Whitmeyer, 1992). Its most basic form involves two actors who each possesses a resource the other desires. Desired resources can range from money to admiration (see Cropanzano & Mitchell, 2005 for a discussion of six types of exchange resources) and are specific to the relationship. An exchange of resources is labeled a transaction and multiple transactions between two actors form an exchange relation, which is the smallest unit of analysis in social exchange theory. Such relations can be restricted to a single domain, such as a set of purely business transactions between a purchasing agent and supplier, or they can span many dimensions, such as parent-child relations. Cropanzano and Mitchell (2005) thought that social exchange theory has the potential for providing a unitary framework for much of organizational behavior.

While much early social exchange theorizing (Homans, 1961; Thibaut & Kelley, 1959) focused on the individual level, Emerson (1972a, b) expanded its application to social structure. His primary contribution was to conceptualize interconnected groups of exchange relations as exchange networks in which the frequency or value of one relation affects the frequency or value of others. Emerson’s (1962) discussion of power dynamics, Thibaut and Kelley’s (1959) focus on game theory exchanges, and more recent work on social dilemmas (Yamagishi & Cook, 1993) have acknowledged both the benefits and harms that can arise in networks of social exchange.

Scholars have noted similarities between social exchange theory and social network analysis (Cook & Whitmeyer, 1992). In particular, both conceptualize social structure as sets of actors holding interlinked positions who engage in social relations. Both advance the premise that much behavior in modern life is embedded in networks of social relations that condition response processes (Burt, 1992; Uzzi, 1997). Together they offer the potential for broader scope and more powerful explanation than either alone (Cook & Whitmeyer, 1992). Related to cronyism, network analysis provides the structure in which it develops, while exchange theory identifies the types of relationships that function within specific network structures to facilitate it. In this paper, we employ both network analysis and social exchange theory to analyze cronyism. Before we elaborate on our analysis, the next section discusses a distinction between two types of networks.

Network types

Scholars have identified several characteristics of social network ties. Among the most important are density, strength, directness, intensity, reciprocity, intimacy, and frequency (Mizruchi, 1994). Although in theory networks could be described by a mix of high and low scores on each characteristic, scholars often develop two constellations, one based on high scores on each dimension and the other based on low scores. Granovetter (1973) labeled ties characterized by high scores as strong and those characterized by low scores as weak. Burt (1992) classified groupings of strong ties as clique networks and groupings of weak ties as entrepreneurial networks. The former are typically small and dense, whereas the latter are large, with loosely organized and broadly diffused contacts. In discussing network strategies in the context of institutional transition in Asian emerging economies, Peng and Zhou (2005) also made a distinction between strong-tie-based and weak-tie-based networks. Similarly, in discussing the motivational foundation of social networks, Kadushin (2002) made a distinction between safety networks that are dense and cohesive and effectiveness networks that contain structural holes. Chen and Chen (2009) differentiated social networks into two types—(1) close guanxi, consisting of network ties that are located in the most inner circle of the individual and are characterized by high levels of sentiment and obligation, and (2) distant guanxi, involving network ties that are located at the periphery and carry low levels of sentiment and obligation. In short, while there can be different ways of classifying social networks, the distinction between clique and entrepreneurial networks, which we use in this paper, appears to be a popular classification adopted by researchers.

Clique networks encourage cooperation as tight social relations steer members toward harmonious action. In contrast, entrepreneurial networks often include people with a diversity of styles, interests, and goals. These qualities lead to intra-network competitive behavior because the loose connections and diffused ties encourage members to take care that their own interests are not lost in the shifting tides of the larger network’s activities. Members who compete for influence drop out if their interests are not adequately served. In exchange theory terms, clique networks encourage generalized reciprocal exchanges, where favors done by one actor for a second may be reciprocated eventually by a third actor and are often performed as much to reinforce network solidarity as to contribute to a dyadic relationship. Entrepreneurial networks encourage negotiated exchanges, where the terms are explicitly agreed on beforehand, or restricted reciprocal exchanges, where a direct exchange relation develops between two parties (Ekeh, 1974). Locationally, trust is an attribute of the entire clique network whereas trust in entrepreneurial networks is a localized attribute between the parties that are involved in an exchange transaction (Kadushin, 2002). Finally, clique networks facilitate the development of communal social capital where connections between members promote actions that benefit the entire network. In contrast, entrepreneurial networks encourage the development of individual social capital as members pursue opportunities that accrue benefits to themselves individually (Ibarra, Kilduff, & Tsai, 2005).

Network competition

A social network does not exist in isolation. The nature of its relations with other networks influences the extent to which cronyism is likely to occur. A host of factors influence competition versus cooperation at the network level, including perceptions of interdependence, organizational resource allocation processes, and environmental munificence (Kramer, 1991). Interdependence arising from limited resources has been the most consistently recognized source of inter-group competition (Salancik & Pfeffer, 1974). Such competition provokes cognitive perceptions of the group as a unit and evokes affective reactions from group members, strengthening boundaries between groups. For example, when two ethnic groups in a country compete for the same economic resources, boundaries based on ethnicity become salient and solidarity within each increases (Okamoto, 2003).

Self-categorization theory (Turner, 1987) posits that individuals categorize themselves, among other levels, by virtue of their group memberships. When group-level categories are salient, individuals tend to think in in-group/out-group terms, viewing themselves and other group members as the in-group and all others as out-groups. Kramer (1991) argued that organizational identification is determined foremost by an individual’s primary group in the organization. Not only do people tend to identify the self based on this group, others also categorize them by primary group affiliation. “Inter-group anxiety,” an uneasiness that often accompanies interacting with members of other groups, reduces attempts to go beyond an individual’s main group (Stephan & Stephan, 1985). For example, this social phenomenon plays out powerfully in daily interactions between health care professionals, such as nurses, physicians, technicians, and administrators, and between various specialties. Strong professional cultures in health care have resulted in steep hierarchies and organizational silos, creating barriers to smooth interactions across professional groups and specialties (Khatri, Baveja, Boren, & Mammo, 2006). Ferlie, Fitzgerald, Wood, and Hawkins (2005) studied health care organizations and found that strong social boundaries between health care workers from different professions retarded the spread of innovations and stymied learning and change in these organizations. For example, individual professionals within the so-called multidisciplinary teams often found it difficult to agree to the role redefinitions indicated by evidence-based practice because established professional roles and “jurisdictions” got in the way.

Inter-network competition increases cronyism through its impact on intra-network cohesion. In inter-network competition, networks attempt to satisfy their own concerns at the expense of others (Thomas, 1992). The explicit network-level competition makes a person’s network membership particularly salient as a defining category. The effect of inter-network competition on strengthening intra-network cohesion is one of the oldest findings in the conflict literature (Sherif, Harvey, White, Hood, & Sherif, 1961). Pressure to accommodate fellow network members is especially high when the network is competing with external groups. Members tend to put aside their differences and unite in support of a common cause. Inter-network competition increases cronyism by increasing the pressure on network members to help one another, which is especially easy to justify when the assistance permits in-group members to benefit at the expense of those in competing groups.

A sense of competitiveness is magnified by the tendency of group members to exaggerate their entitlement to disputed resources (Leary & Forsyth, 1987). In order to redress a perceived unfair imbalance, efforts to enhance the standing of fellow group members can seem entirely justified, especially since out-group members are usually viewed as less honest and trustworthy (Brewer, 1979). Group members thus display two biases: in-group favoritism and out-group unfavorability (Labianco, Brass, & Gray, 1998). People allocate higher payoffs to members of their own group than to members of other groups (Messick, 1998). In brief, strong intra-network cohesion coupled with weak inter-network ties create the conditions for unethical behavior (Brass, Butterfield, & Skaggs, 1998), as cronyist attitudes toward in-group members become natural and even expected. Specifically:

Proposition 1a

The greater the extent of inter-network competition, the more likely that cronyism will occur.

At variance with inter-network competition is intra-network competition. As a social network consists of individuals who may have diverse interests, goals, and aspirations, competition among network members is a real possibility (Tjosvold, 1998). Individuals interact with respect to different issues that are of their concern, and gather together on the basis of similar positions on an issue. Murnighan and Brass (1991: 285) define a coalition as “any subset of a group that pools its resources or unites as a single voice to determine a decision for the entire group.” As people may have different opinions on an issue, it is likely that more than one coalition is formed within a network with each holding a significantly different stance on the issue. When these coalitions try to push forward their own agendas, conflicts and intra-network competition will occur.

Examples of intra-network competition often can be found in party politics where certain members of a political party compete for leadership positions within the party. When network members compete for limited resources such as power and status, the competition provokes cognitive perceptions of the self as primary and individuals seek to promote their own interests at the expense of other network members (Schei & Rognes, 2003). Unable to rely on the network itself, they form limited ties with specific network members and even nonmembers to pursue their goals. For example, when a power struggle occurs in a political party, competing members usually try to garner support from other members as well as outside parties such as the media. Nevertheless, intra-network competition is often more transient and dynamic than inter-network competition because, pushed to an extreme, the network will break up, as evidenced by the creation of new political parties as spin-offs from existing ones. In US politics, for example, at the beginning of a presidential election race, there is often keen competition among potential candidates within either or both of the two dominant political parties. Once a candidate has won, intra-party competition is usually suppressed as competition between the candidates nominated by the two parties becomes more salient. Thurman’s (1980) study of office politics in an overseas branch of a large international organization indicates that the composition of a coalition changes as network members’ interests and conflicts shift over time, whereas the social network in which such a coalition is embedded remains intact.

Intra-network competition engenders intra-network fragmentation (De Dreu & Van Vianen, 2001), which emphasizes member differences as individuals favor specific others either to complete an exchange or initiate an alliance (Gimeno, 2004). As a way of garnering support, a network member may engage in cronyist exchanges with specific others who are within or even outside the network at the expense of other network members. In other words, to keep a coalition together, there is a pressure for internally awarded benefits (Murnighan & Brass, 1991). Therefore,

Proposition 1b

The greater the extent of intra-network competition, the more likely that cronyism will occur.

The effect of competition on the likelihood of cronyism is moderated by network type. In the presence of inter-network competition, intra-network cohesion emerges more readily in clique than entrepreneurial networks (Coleman, 1990); in the presence of intra-network competition, intra-network fragmentation emerges more readily in entrepreneurial than clique networks (Uzzi, 1997). These dynamics occur for three main reasons. First, a social network consists of people with different interests and goals. For intra-network cohesion to occur, network members must accommodate fellow members and put aside differences to support a common cause. In other words, they must subordinate self-interest to group interests (Coleman, 1990), a process that is less difficult for clique than entrepreneurial networks, as clique network members tend to define themselves in relation to the network (Oh, Chung, & Labianca, 2004). On the other hand, provoked by intra-network competition, intra-network fragmentation emerges more readily in entrepreneurial networks because members are motivated primarily by their own interests and ties to other members are weaker. When intra-network competition exists, group goals become secondary to their own interests (Madhavan, Gnyawali, & He, 2004).

Second, intra-network cohesion increases when members like, trust, and identify with one another (Bolino, Turnley, & Bloodgood, 2002). They are more likely to like one another when they interact more (Mullen & Copper, 1994). They are more likely to trust one another when they perceive that fellow members are pursuing common goals, not acting only from self-interest (Lewiski, McAllister, & Bies, 1998). They are more likely to identify with one another when they view fellow members as reliable (Hogg & Terry, 2000). Greater interaction, common goals, and reliability tend to characterize clique rather than entrepreneurial networks. In fact, the weak ties fostered by entrepreneurial networks facilitate autonomy, thus increasing group members’ willingness to act in ways that weaken group cohesion and increase fragmentation (Perry-Smith & Shalley, 2003).

Finally, restricted versus generalized exchanges are likely to give rise to intra-network fragmentation versus cohesion, respectively. Ekeh’s (1974) dual exchange theory argues that in restricted exchanges, the mutual reciprocity present in the dyad produces tension and instability as norms of quid pro quo and high accountability generate much effort to maintain equality in exchange rates and settle inequalities quickly. The result is frequent, acutely self-interested conflict over the fairness of exchanges that leads to high turnover in exchange partners. Generalized exchanges, on the other hand, link multiple network members together through their mutual interests (Emerson, 1990). Indirect reciprocity implies shared concerns for network member C, who may be the one to complete an exchange that began when member A provided for member B. Networks characterized by generalized exchange are low in tension and high in stability. Entrepreneurial networks are more likely to be characterized by restricted exchanges, whereas clique networks are more likely to be characterized by generalized exchanges. The direct nature of restricted exchanges (A→B, B→A) better fits the tendency of actors in entrepreneurial networks to firmly defend their self-interests in the face of intra-network competition. In contrast, the indirect nature of generalized exchanges (A→B→C) better fits clique network members’ tendencies toward mutual accommodation.

In sum, network type moderates the relationship between form of competition and likelihood of cronyism. When networks compete, intra-network cohesion is more likely to emerge in clique than entrepreneurial networks. When individuals compete, intra-network fragmentation is more likely to emerge in entrepreneurial than clique networks. Therefore:

Proposition 2a

The positive effect of inter-network competition on the likelihood of cronyism is greater in clique than entrepreneurial networks.

Proposition 2b

The positive effect of intra-network competition on the likelihood of cronyism is greater in entrepreneurial than clique networks.

Discussion

Implications for research

In their paper examining the effects of culture on cronyism, Khatri et al. (2006) advanced propositions on the likelihood of cronyism along the cultural dimensions of individualism-collectivism and verticalness-horizontalness. They also proposed a typology of cronyism that enables us to better understand why a specific type of cronyism is most likely to occur in each cultural configuration. Since cronyism takes place in the context of a social network, the present paper moves to the social network level and develops propositions on factors that have a more proximal impact on cronyism. The next logical step is to link cultural with network factors through positioning the latter as mediating the impact of the former on cronyism. In other words, cultural values flow through network properties to affect individual cronyist behavior.

Although propositions linking cultures with networks are beyond the scope of this paper, we note the parallels of individualist cultures with entrepreneurial networks as well as collectivist cultures with clique networks. For example, individualists define the self as an autonomous entity, deem that personal goals should take priority over group goals, and seek task accomplishment, even at the expense of relationships, qualities generally attributed to entrepreneurial networks; collectivists define themselves through their connectedness to groups, tend to subordinate personal goals to groups goals, and value harmonious relationships, sometimes at the expense of task achievement (Triandis, 1995), qualities generally attributed to clique networks. In one way, these parallels support an argument for the predominance of entrepreneurial networks in individualist cultures and clique networks in collectivist cultures. In another, the closeness of the descriptors across levels constitutes a challenge to anyone seeking to develop propositions linking the two. In particular, there is a great need to sharply delineate the boundaries separating cultures from networks.

Concepts similar to cronyism exist in different cultures, such as guanxi and blat in Chinese and Russian cultures, respectively. Cronyism has the potential of serving as a general concept that encompasses these culture-based concepts, with the benefit that similar phenomena in different cultures can be analyzed within a common theoretical framework. Consider the concept of guanxi, which has been frequently employed by scholars of Chinese management research. Although guanxi is often viewed as an indigenous Chinese construct (Chen & Chen, 2004), cronyism and guanxi share some common attributes. Pye (1992), for example, regards guanxi as a special relationship individuals have with each other with implications of continued exchange of favors. This view is in line with our definition of cronyism. That said, guanxi has its own unique features because of its embeddedness in the Chinese cultural milieu, with its roots anchored in Confucianism (King, 1994). To develop cronyism as a general concept, it is necessary to examine under which circumstances cronyism is able to explain guanxi related phenomena and under which circumstances it is not.

Our discussion indicates the potential value in combining network analysis with social exchange theory. The distinction between clique and entrepreneurial network structures provides a more comprehensive analysis of cronyism when combined with social exchange ideas on different types of exchange relations. For example, efforts in the network literature to characterize types of nodal ties can benefit from social exchange theory’s distinctions among direct, restricted reciprocal, and generalized reciprocal exchanges (Ekeh, 1974). The implications of characterizing networks based on relative proportions of each type of exchange relation should be significant.

Although economics and finance scholars have shown a lot more interest in applying the concept of cronyism to a number of corrupt acts (Brick, Palmon, & Wald, 2003; Johnson & Mitton, 2003; Pagano, 2002), we are beginning to see studies in organizational literature that attempt to systematically analyze the concept and discuss its dynamics (Wan, Yiu, Hoskisson, & Kim, 2008). It may be possible to use existing theories to advance thinking in this area. For example, based on transaction cost theory, Kang (2003) argued that cronyism might improve economic efficiency under certain circumstances. In systems with weak legal, economic, and political institutions, information about market conditions is both scarce and difficult to obtain, and investments and property rights are insecure. The transaction costs of making and keeping agreements and securing property rights can be prohibitively high. Under these conditions, cronyism among a small, stable number of actors can reduce transaction costs because it leads to better information, monitoring and sanctioning, strengthens property rights, and provides alternative means for reciprocity and side-payments. Thus, an application of transaction cost theory to cronyism may shed light on when cronyism is efficient.

Another research direction may analyze when special favors and privileges gain legitimacy. For instance, Johnston (2002) found a strong tendency for lower-middle and middle-status respondents to resent cronyism and for upper-middle and upper-status respondents to tolerate it. Favors and advantages lower- and middle-status people view as illegitimate may be seen by higher status groups as merely the fruits of merit and expertise.

Finally, cronyism and institutional weakness are linked (Krug & Hendrischke, 2001; Sherwood, 2007). Cronyism may flourish when formal systems are unable to cope with demands made on them (Scott, 2002). Vacuous, faulty, contradictory, defective, overly loose, or overly restrictive rules may encourage cronyist social exchanges that enable members to cope with a system’s inherent drawbacks (Caiden, Dwivedi, & Jabbra, 2001). Social networks make sense only in an imperfect system (Burt, 1992). In a well-developed system, individuals need not spend time cultivating connections to get things done. Indeed, the above link may run in the reverse direction, from cronyism to institutions. Cronyist networks may contribute to institutional weakness and pose significant hurdles in reforming a system because clear and effective rules ultimately may hurt them. In sum, strengthening institutions may be an effective means of overcoming cronyism. However, since cronyism is socio-culturally situated, institutional measures without socio-cultural reform may not succeed in curbing it (Park, 2003). A pertinent research question arises: how can institutional measures and socio-cultural reforms be combined to curb cronyism?

Implications for practice

Our propositions suggest that network competition, whether inter- or intra-network, increases the likelihood of cronyism. Those who build corporate governance systems must ask whether their formulations have accounted for this effect. When coupled with suitable merit-based incentives, encouraging competition among employees is likely to improve organizational performance. However, when an organization is described as a hypercompetitive, winner-take-all environment, employees feel intense pressure to produce at any cost and may engage in dubious practices, such as cronyism, in order to meet performance goals. For example, the extent of competition at trading desks in Enron was so keen and unprincipled that traders were afraid to leave their desks for fear that other traders would steal their work (McLean, 2001) and ally with outsiders to exploit it.

Just as cronyism cannot be understood without also understanding its socio-cultural logic (Hooker, 2003), the same is true for measures to curb cronyism. Remedies aimed only at formal structures are unlikely to root it out (Steier, 2009). In fact, emphasis on formal aspects may simply push cronyism deeper into informal social networks that are harder to reach and regulate. For example, Reno (1995) reported that Western governments and the World Bank failed to grasp the dynamics of power in Sierra Leone. Fixed on institutional reform, they did not see the real power wielded through informal political and market networks. Their attempts to impose economic reform and rationalize institutional structures had the opposite effect of deepening opaque socio-economic networks. Recent corporate governance reforms in the United States also concentrated on structural measures, leaving informal networks at the core of cronyism largely intact. If not accompanied by efforts to promote basic US values of fairness and equality of opportunity, these measures seem likely to fail in the long run. Tendencies noted earlier for higher status people to tolerate apparent favoritism pose a further policy dilemma. Their desire for discretion will clash with preferences by those at lower status levels for transparency and objectivity. Moreover, such differential perceptions of legitimacy may vary across cultures.

Conclusions

Although many studies have indicated the benefits of social networks, much less is known about their downside. The impulse to favor relatives, friends, and business associates is so natural that cronyism looms as a major challenge to those concerned with preventing all forms of corrupt acts. In this paper, we explore the dynamics of cronyism as an important potential drawback of networks, and illustrate how social exchange theory has contributed to our analysis. Purportedly a widespread phenomenon around the globe, cronyism has been neglected by organizational researchers. Unless strong countervailing forces prevent it, cronyism may continue to dominate social relations in some countries.

Theoretical advances from studying the downside of networks should contribute to increased knowledge of factors influencing their effectiveness. For managerial practice, a desire to understand the causal dynamics of events such as the Asian financial crisis and US corporate scandals has underscored the need to determine how they came about. In both instances, misuse of network contacts appears to have produced opportunities for corporate mismanagement. An understanding of how these opportunities developed can point toward actions that prevent or inhibit their recurrence. As attention to social networks increases, knowledge of their full range of causes and consequences promises to enrich academics and society.