Entrenchment in status positions and the adoption of new norm-deviant organizational practices: Evidence from the Japanese banking industry,1983

Abstract

We examine the effect of status entrenchment on the adoption of new norm-deviant organizational practices. Identifying organizational age and status mobility as factors affecting entrenchment, we extend the middle-status conformity theory by explicating how entrenchment moderates the relationship between status and adoption. Using original data from the Japanese loan syndication market, we show that young and new-in-status banks have a lower propensity to follow status-based adoption behavior than actors entrenched in the same status positions. We discuss implication of these results for the understanding of new practice adoption and organizational status effects.

Keywords

Japanese banks legitimacy middle-status conformity norm-deviant behavior organizational status practice adoption

Scholars have long noted that organizational status plays an important role in the adoption of novel practices. Early research (Becker, 1970; Marsh & Coleman, 1956) suggested a U-shaped relationship between status and adoption, formally restated as the middle-status conformity theory by Phillips and Zuckerman (2001). This theory suggests organizations’ legitimacy concerns vary systematically with their positions in the status hierarchy and, consequently, organizational status variously enables or impedes the adoption of practices that deviate from established norms or challenge the status quo (Durand & Kremp, 2016; Phillips & Zuckerman, 2001).

An implicit but central assumption in extant theory has been that organizations occupying comparable status positions face equivalent conformity pressures and legitimacy considerations. This assumption rests on a largely static conceptualization of status—defined as “unearned ascriptions of social rank” (Washington & Zajac, 2005, p. 282)—in which organizations receive benefits and act according to their positions in the status hierarchy. Recent research has challenged this conception by highlighting status fluidity and its implications for organizational action. Askin and Bothner (2016) demonstrate how organizations often refuse to accept status demotion, choosing instead to maintain behaviors associated with their previous higher status. Krishnan and Kozhikode (2015) find that the specter of status loss may drive organizations to take drastic and illegal actions. Jensen (2003) shows that firms shed stigmatized partners for fear of losing status and its benefits. From this perspective, status is better conceived of as “prized territory in a turf war” (Askin & Bothner, 2016, p. 26), to be gained and lost, than as an ascribed organizational trait.¹

The recognition that status is fluid holds important implications for our understanding of new practice adoption. Status fluidity implies that organizations’ entrenchment in status positions varies, making application of status-based expectations and norms less straightforward. Consequently, organizations’ predispositions to follow status-based adoption behavior may be affected. Specifically, if actors of equivalent status differ in how strongly they identify with their position, it may impact their conformity pressures and adoption behavior. To date, scholars have offered little insight into this possibility. In particular, we lack an understanding of whether organizations that acquired status positions recently—and, whose place in the hierarchy may therefore be less entrenched—have the same propensity to adopt norm-deviant practices as more entrenched occupants of equivalent status positions. Examining this question is imperative because it offers new insight into how status impacts the adoption and the spread of new practices.

In this paper, we extend the middle-status conformity theory by considering the effect of status entrenchment on the adoption of new practices. We argue that weakly entrenched high- and low-status organizations are less likely to adopt new norm-deviant practices, as compared to more entrenched organizations of equal status. Consequently, the U-shaped relationship between status and adoption postulated by the middle-status conformity theory is less pronounced for weakly entrenched organizations. We test our theory using data from Japan’s banking industry between 1983 and 2005, a particularly suitable setting for our analysis.

New Practice Adoption and Status Entrenchment

The adoption of new technical and organizational practices offers firms opportunities to garner competitive advantages vis-a-vis competitors. Early adopters can distinguish themselves in the market (Abrahamson, 1991), enhance customer sentiment (Kamins & Alpert, 2004), improve standings with key stakeholders (George, Dahlander, Graffin, & Sim, 2016), and gain recognition as ‘innovators’ or ‘market leaders’ (e.g., Rindova, Pollock & Hayward, 2006). Yet, new practices also frequently constitute a break with established social norms (Abrahamson, 1991), inciting vigorous contestation. Even when practices offer unequivocal instrumental benefits, organizations may refrain from adoption due to sanctions and stigmatization by other field members (Jonsson, 2009).

Given these potential consequences, organizations’ decisions to adopt new and controversial practices are influenced by legitimacy considerations; these, in turn, depend on their status positions. Phillips and Zuckerman (2001) explicate a U-shaped relationship between status and practice adoption, based on a two-stage model of categorization and selection. In the first stage, audiences—e.g., customers and clients—categorize organizations into three groups: preferred “full-fledged players” with unquestionable legitimacy (high-status actors), peripheral organizations with marginal legitimacy (middle-status actors), and nominally illegitimate “mere candidates” (low-status actors). In the second stage, audiences select suitable organizations from these groups, based on their unique product offerings.

The basis for Phillips and Zuckerman’s middle-status conformity lies in the trade-off between prioritizing a higher categorization (which requires actors to demonstrate conformity with legitimate practices) and selection (which calls for differentiation vis-a-vis other offerings through non-conformity). Organizations categorized as high status are seen as legitimate “beyond doubt” (Phillips & Zuckerman, 2001, p. 385), allowing them to engage in maximum levels of non-conformity to entice audience selection. Meanwhile, the illegitimacy of low-status actors means they both have little to lose and feel less pressure for conformity. Consequently, they too are incentivized to adopt non-conformist practices and behaviors in an effort to gain selection. In contrast, the emphasis of middle-status actors lies in maintaining their current status and hopefully moving up the hierarchy. As a result, they are more likely to focus on categorization, specifically by adopting conventional practices that “attest to their competence” and social worth in line with existing conventions and practices (Durand & Kremp, 2016, p. 66).

In specifying the middle-status conformity theory, Phillips and Zuckerman (2001) explicitly emphasize actors’ “social-psychological orientations” towards their own status position. Specifically, they suggest that the effects of status build on high- and low-status actors’ assumptions that their positions are relatively fixed. This may not always be the case, however, as high-status actors “may not feel secure enough to defy convention if the achievement of high-status is recent” (p. 389), while low-status actors may not identify as the “relatively permanent outsiders” (p. 388) who can engage in norm-deviant behavior.

Subsequent research has recognized this potential misalignment between actors’ social-psychological orientations and their structural status positions. High-status organizations have been shown to suffer “status anxiety” due to association with stigmatized partners (Jensen, 2006; Jonsson, Greve, & Fujiwara-Greve, 2009). Such status anxiety can impact actors’ propensity to engage in non-conformist behavior or adopt new practices that threaten their legitimacy (Krishnan & Kozhikode, 2015). Other works have identified how low-status organizations refuse to accept their marginal positions, choosing instead to explicitly engage in behaviors associated with higher status positions, in the hopes of moving up the hierarchy (Delmestri & Greenwood, 2016).

Together, these works show how structural status positions do not by themselves determine actors’ behaviors; in order to take effect, status must also be entrenched in the minds of both actors and audiences (cf. Phillips, Turco, & Zuckerman, 2013). To date, however, few works have specified how actors come to be entrenched in a particular status position, nor have they examined the effects of varying levels of entrenchment on the adoption of norm-deviant practices. In particular, we know little about the behaviors of actors who do not feel secure in their high-status positions or who do not accept the permanency of their low- or middle-status positions.

Status entrenchment and the adoption of norm-deviant practices

In this paper we address these questions by focusing on how status entrenchment affects the adoption of norm-deviant practices. We define entrenchment as the extent to which status positions are taken-for-granted by both the organizations occupying these positions, and external audiences (Prato, Kypraios, Ertug, & Lee, 2018).² We build our notion of entrenchment on an emerging research stream that explicitly recognizes heterogeneity and mobility in status positions. Bothner, Smith, and White (2010), for example, distinguish between robust and fragile status positions, based on the concentration of an actor’s network ties. From this perspective, fragile status positions are those characterized by few endorsements, and where such endorsements accumulate from precariously positioned actors.

While Bothner et al.’s (2010) fragility conceptualization centers on status’ structural characteristics, entrenchment emphasizes the cognitive aspect of status (Bitektine, 2011). As Jensen, Kim, and Kim (2011) note, status represents a “cognitive category” (p. 91), one which audiences and producers use as a sensemaking device to delineate the “minimal attitudes and behaviors and culturally defined expectations” necessary for participation in a market (p. 92). In order for this sensemaking device to function, however, status positions must be taken for granted. Entrenchment is thus the degree to which status has taken cognitive root.

Status positions become entrenched and taken-for-granted over time through processes of repeated exposure and observation (Bitektine, 2011, p. 160). By continuously evaluating producers, audiences form assumptions and beliefs; as these are reinforced, audiences apply status-specific criteria to organizations without referencing their structural status positions. Organizations themselves similarly take their position for granted and begin to act accordingly. As Podolny (1993) notes, such cognitively entrenched taken-for-granted assumptions are distinct from the actual quality of the organization’s offerings and products.

We suggest that organizational tenure and status stability constitute two crucial factors of status entrenchment. Our motivation for selecting these two factors is that tenure determines an organization’s cumulative repeated exposure to audience evaluators, while status stability reinforces pre-existing evaluations, thus increasing taken-for-grantedness. In the following we explicate these mechanisms, as well as their impact on the adoption of norm-deviant practices.

Entrenchment through tenure

Entrenchment is affected by an organization’s tenure in the field, such that young organizations’ status positions are weakly entrenched. While age has long been recognized as a key determinant of performance and survival (see discussion in Hannan & Carroll, 2004), prior research has mostly focused on how age impacts internal organizational factors, such as routines and practices (Le Mens, Hannan, & Polos, 2015), or organizational experience and learning (Thornhill & Amit, 2003). Age, however, also impacts how organizations are perceived by external audiences, including customers, suppliers, analysts, and regulators (e.g., Choi & Shepherd, 2005). Young organizations lack the histories that audiences rely on for status classification; as such histories emerge, however, they play a bigger role than current indicators of performance or affiliations in shaping audiences’ taken-for-granted assumptions and beliefs about the organizations (Kodeih & Greenwood 2014). Stuart, Hoang, and Hybels (1999) find that audiences place less weight on links to prominent investors as a sign of status when evaluating older biotech firms, as compared to younger biotech firms. Similarly, Singh, Tucker, and House (1986) show that the effect of endorsement-granting ties on survival is especially strong for young organizations but declines for older entities. Through repeated audience exposure, organizational age serves to entrench a cognitive status classification, separate from that of the structural status position.

Entrenchment through status stability

Cognitive entrenchment also depends on the stability of the observed positions, i.e., the extent to which the organizations’ status position remains unchanged. High levels of status mobility, i.e., significant shifts in an organization’s status position, generate uncertainty and ambiguity about the permanence of the organization’s position (cf. Zhao & Zhou, 2011). This uncertainty relates not to an organization’s current structural status positions as such, but rather to its longevity and sustainability. Under conditions of significant mobility organizational status is less likely to become taken for granted and cognitively entrenched, both among observing audiences and within an organization.

In sum, cognitive status entrenchment is a function of both age and status mobility; the younger and more status-mobile an organization is, the less cognitively entrenched its status. In the following we develop hypotheses about how these entrenchment effects moderate the relationship between status and the adoption of norm-deviant practices.

Hypotheses Development

Status and the adoption of loan syndication in the Japanese banking industry

Situating our analysis in the Japanese banking industry, we focus on how varying levels of entrenchment impacted the probability of adopting loan syndication. Loan syndication is a financial practice wherein borrowers’ loans are split into tranches and distributed to multiple lenders, who can trade the tranches on the secondary market. Because loan tranches can be distributed to numerous co-lenders and subsequently traded among licensed firms, financial institutions enter the practice independently by participating in loan distribution and trading.

Loan syndication first emerged in the Japanese market in the early 1980s. During that era, Japan’s corporate lending market was dominated by the so-called main-bank system, a format based on bilateral relationships wherein individual banks supplied the full loan amount to clients, charged low interest, and maintained clients’ loans as a form of relational commitment (Scher, 1998). Syndicated lending represented a major break from the institutionalized practice of bilateral lending and significantly challenged the established norms in the Japanese banking market:

Loan syndication was something completely new. . .clients were confused and some of them upset. They worried what this would mean, what would happen to their relationships [with their main banks]. . .they said they wouldn’t agree to it. (Edman, 2016, p. 70)

As this quote implies, loan syndication threatened the trust built up between banks and their customers through carefully cultivated embedded relationships (Scher, 1998). The adoption of syndication therefore carried a clear risk of de-legitimization, yet it also offered significant benefits, primarily in the form of lower risk exposure through diversification, as well as higher returns due to collection of administrative fees.

Ties that underpin Japanese banks’ status positions are based on close customer relationships that are difficult to sever, yet they are not necessarily sacrosanct (Scher, 1998). Because of regulations on competition and price-setting, Japanese banks have traditionally competed on the number and strength of their client relationships. The strength of a bank–client relationship is based on the amount of outstanding loans; even today, Japanese firms publish a list of their top 10 banks based on outstanding borrowings. Banks occupying the top spot in these rankings often receive first priority when clients seek auxiliary services, including M&A transactions, equity market operations, securitization, and so on.

Banks’ client relationships also function as a source of competitive advantage by enticing new customers. In the relationship-based Japanese economy (Uchida & Udell, 2010), banks often serve as network conduits, providing access to new suppliers, partners, and customers. Customers thus enter into banking relationships not simply to gain funds, but to widen their own networks, using the focal bank as an intermediary (Sheard, 1989). In this way, loans are seen as “treasures” that serve to cement and ensure the banks’ positions with individual customers. Consequently, both borrowers and banks are acutely conscious of status within the field in general, and their own status positions in particular.

As noted above, the practice of loan syndication itself challenged many of the norms in the Japanese banking system and was hence potentially de-legitimizing, yet not deviant enough to warrant automatic de-legitimation. Together, these features of the Japanese banking industry suggest applicability of the middle-status conformity theory, because it satisfies the theory’s scope conditions: (1) a status hierarchy that is relatively stable, but not so rigid as to preclude status mobility; (2) a market where organizations are conscious of status and its effects, such that high-status actors derive substantial benefits, while low-status actors have an incentive to gain status; and (3) a practice being adopted that carries risk of de-legitimization, but not to the extent of completely discrediting high-status actors (Phillips & Zuckerman, 2001). Accordingly, we hypothesize a U-shaped relationship between organizational status and propensity to adopt loan syndication:

Hypothesis 1: In the Japanese banking industry, there is a U-shaped relationship between bank status and the adoption of loan syndication, such that high- and low-status banks are more likely to adopt loan syndication than middle-status banks.

Age-based entrenchment and the adoption of new norm-deviant practices

As noted above, the U-shaped relationship between status and the adoption of deviant practices is underpinned by a two-stage mechanism, wherein audiences first categorize organizations into different status groups, subsequently selecting a suitable organization for transactions (Phillips & Zuckerman, 2001). We suggest that entrenchment moderates the U-shaped relationship by impacting the categorization aspect of this mechanism.

First, organizational age is frequently employed by audiences to categorize firms and their offerings (Singh et al., 1986). Most young firms are grouped in middle- or low-status positions due to their lack of network ties (Bitektine, 2011), although some achieve high-status positions because of relationships with significant partners and benefactors (Stuart et al., 1999). Given the firms’ lack of history, these initial categorizations are based on structural status positions, as opposed to cognitively taken-for-granted status. Moreover, these structural status positions are not fixed; young organizations’ relational ties and networks are new and untested (Santos & Eisenhardt, 2009) and thus more likely to break (Podolny, Stuart, & Hannan, 1996). Consequently, young firms’ initial structural status positions have yet to become taken-for-granted social facts among audiences.

The lack of status entrenchment creates both opportunities and risks for young organizations. For those categorized into low-status positions, a lack of entrenchment offers the possibility to ascend the status hierarchy; consequently, young organizations often put significant effort into signaling their worth to external audiences (Deutsch & Ross, 2003). As Phillips and Zuckerman (2001) note, audiences categorize actors based on their conformity with prevailing norms, hence young actors in low-status positions frequently emphasize conformity with industry norms and conventions to increase their chances of achieving a higher status categorization (Durand & Kremp, 2016; Zimmerman & Zeitz, 2002).

At the other end of the status hierarchy, young high-status actors may feel uncertain about the durability of their new positions with audiences (Prato et al., 2018). Previous research has found that managers promoted to senior positions at a young age are more aware and anxious about their position, as compared to older managers in the same position (Pearce & Xu, 2012). In much the same way, young high-status organizations may find it difficult to take their positions for granted (cf. Blader & Chen, 2012). Consequently, in an effort to be “more cautious so as not to expose and undermine their status-inconsistent standing” (Prato et al., 2018, p. 5), these organizations are likely to eschew risky non-conformist or deviant behaviors in favor of products and practices that signal their commitment to the status quo, thereby minimizing the risk of status loss (Phillips et al., 2013).

Young middle-status organizations face both the opportunity to move into high-status positions and the risk of losing their current status position. Similarly to low and high-status entities, we would thus expect middle-status entities to also avoid deviant practices in favor of those that confirm with existing norms and behaviors.

In sum, we suggest that younger organizations across the status hierarchy are likely to emphasize conformity and avoid deviant practices, but for different reasons. For low-status young organizations, conformity offers a path up the status hierarchy, towards a classification as a peripheral or even “fully fledged” player. Conversely, for high-status young organizations, conformity offers a way of reassuring audiences that their status positions are well deserved (Prato et al., 2018). Young middle-status organizations are guided by both of these considerations. The full implication of young age is thus to reduce non-conformist actions across all status positions. Formally we suggest:

Hypothesis 2: Organizational age moderates the U-shaped relationship between status and adoption of loan syndication, so that (1) young high- and low-status Japanese banks are less likely to adopt loan syndication than old high- and low-status banks, respectively, and (2) young middle-status Japanese banks are as likely to adopt loan syndication as old banks of equal status.

Status mobility and the adoption of norm-deviant practices

High- and low-status organizations adopt norm-deviant practices in large part because their categorizations carry audiences’ expectations and assumptions about their behavior. Such expectations and assumptions, however, depend crucially on audiences’ ability to classify organizations into distinct status categories. Audiences often struggle to classify organizations that span multiple categories. Zuckerman (1999) showed how stock analysts were unclear about how to categorize firms operating in multiple industries and sectors, while Hsu (2006) demonstrated how actors with roles in multiple genres received lower evaluations as compared to those typecast into one genre. Importantly, audiences’ ambiguity in these cases arose not from an inability to observe a categorical location, but rather from actors’ simultaneous position in multiple locations (Hsu, Hannan, & Koçak, 2009).

We suggest that the same kind of ambiguity arises when audiences observe organizations in multiple status categories over time. Although audiences recognize an organization’s status position at any one point in time, mobility across status positions may make audiences uncertain as to whether the organization truly belongs in its current category. Audiences that are ambiguous about an organization’s classification may be less likely to apply behavioral expectations afforded to more entrenched actors (Hsu, 2006), thereby influencing the organizations’ actions.

At the top of the status hierarchy, audiences may be reluctant to consider newly ascended organizations as being beyond reproach. Such classification ambiguity constitutes a form of identity threat for the organization (Elsbach & Kramer, 1996). Specifically, organizations may feel that their true status position is not being fairly recognized (cf. Sauder, 2006). Elsbach and Kramer (1996) suggest that organizations respond to identity threats by re-emphasizing their belonging to the category in question; in our case, this implies adhering to practices that signal conformity with the prevailing status quo. Such adherence is not pursued to avoid penalties, but rather to guarantee the rewards that come with high status (Durand & Kremp, 2016). In other words, newly ascended high-status organizations are likely to eschew norm-deviant practices in favor of standard practices in order to safeguard their positions.

Conversely, at the bottom of the status hierarchy, classification ambiguity implies that audiences are less likely to dismiss newly descended organizations as ‘permanent outsiders,’ i.e., unambiguously low-status firms that cannot be expected to abide by established social norms. In these instances, ambiguity offers immunity and exemption from status-related expectations. Specifically, it allows organizations to abstain from behaviors that could jeopardize their legitimacy and potential rebound into the status hierarchy. Avoiding legitimacy-threatening behaviors constitutes part of organizations’ defense mechanisms (Elsbach & Kramer, 1996) as they seek to regain their status position. Askin and Bothner (2016) thus show how firms experiencing status loss react by increasing prices, in an effort to signal their continued quality, i.e., belonging to the higher status category.

For organizations moving to the middle of the status hierarchy, categorization ambiguity implies that audiences may be less willing to treat them as occupying intermediate status positions, and thus may not expect maximum conformity. Consequently, these organizations may be less likely to follow behavioral patterns applicable to middle-status organizations that did not experience a status change.

Taken together, these mechanisms imply that the U-shaped relationship between status and adoption of norm-deviant practices is less pronounced for new-in-status organizations. Compared to organizations entrenched in the status hierarchy, recently arrived high- and low-status organizations should be less likely to engage in behaviors that may harm their legitimacy, while middle-status organizations may not show the same degree of conformity as their entrenched counterparts. Formally, we suggest:

Hypothesis 3: Status mobility moderates the U-shaped relationship between status and adoption of loan syndication, such that (1) Japanese banks that acquired high- or low-status status positions recently are less likely to adopt loan syndication and (2) Japanese banks that acquired middle-status positions recently are more likely to adopt loan syndication, compared to banks of equal status that did not experience status mobility.

Method

Data

We used the Financial Quest Nikkei NEEDS database—one of the most comprehensive datasets on Japanese companies (Delios & Henisz, 2003)—to gather data on performance indicators, number of employees, and foreign affiliates of Japanese banking institutions. We excluded foreign lenders because they may be less embedded in the market and hence guided by different institutional demands (Edman, 2016). We used the NEEDS database to construct a two-mode network of bank–client relationships. Using data on listed companies’ lenders, including the amount supplied by each bank, we identified each bank’s lending relationships between 1983 and 2005, as well as clients’ bank rankings. Using this information, we calculated our independent variable, status (see below).

We collected data on Japanese lending syndicates from the Thomson Financial One database which provides the names of loan syndicates’ participants and borrowers (client firms), as well as transaction dates, size, domicile, and purpose. We used the full extent of the database on Japanese syndications, from January 1, 1983. We searched for information on possible older syndications using Nikkei Telecom 21—a database of Japan’s major financial newspapers—through 1970 but found none, thereby concluding that there is minimal risk of left censoring. We gathered data on Japanese banks’ categorical membership (i.e., licensed banks, city banks, and regional banks) from the Japan Banker’s Association, the country’s main banking industry association.

Independent variables

Status

Because “status is fundamentally a relational concept” (Washington & Zajac, 2005, p. 286), it takes form through affiliations and relationships with other field actors (Sauder, Lynn, & Podolny, 2012); multiple affiliations with high-level actors enhance and strengthen status (Bothner, Podolny, & Smith, 2011; Podolny, 1993). Accordingly, we follow prior literature on organizational status (e.g. Bothner, Kang, & Lee, 2006; Jensen, 2003; Shipilov & Li, 2008) to operationalize status by Bonacich centrality within the network of relationships. In constructing our status variable we utilized information on banks’ relative standing to one another. We created the status variable as follows.

Lending relationships represent a two-mode network with banks and borrowers as nodes. First, we used a one-mode network projection (Wasserman & Faust, 1994) to establish connections between banks (i.e., two banks have a tie if they lend to the same borrower in the same year). We constructed this projection using a five-year moving window to account for fluctuation in the number of ties and tie decay. In the highly status- and affiliation-conscious Japanese lending market, borrowers rank their lenders in order of importance and prefer to transact with the highest ranking bank first. Thus, loan amounts indicate banks’ standings among specific clients, with the highest ranking bank providing the largest loan. This information, like the order of underwriters on “tombstone” ads in United States investment banking (Podolny, 1993), indicates the relative standing of banks to one another, thereby providing the basis for our status calculation.

Next, we assigned directed ties (indicating deference) from lower to higher ranking banks with the same borrower, weighting these by loan amount to incorporate information on the relative standing of different banks. We aggregated this information across all banks, forming a weighted network of directed ties. We calculated the Bonacich centrality measure on this directed weighted network of bank-to-bank ties. Banks with higher Bonacich centrality have more prominent positions in the market’s social structure. Because Bonacich centrality is a recursive measure (taking into account the centrality of a focal bank’s partners, as well as its own centrality) and asymmetric (indicating differentiation in a bank’s social standing), this operationalization corresponds to the theoretically desirable way of measuring organizational status as “accumulated deference” (Washington & Zajac, 2005).

New-in-status

To operationalize how recently a bank gained its status position, we needed to identify banks experiencing status mobility. We also needed to distinguish between substantive status mobility (significant status gain or loss) and random oscillations or insignificant status changes. Following Askin and Bothner’s (2016) logic of ascertaining status mobility, we identified banks that experienced a status gain or loss exceeding two standard deviations from their mean over five prior years. Formally, we designated banks as new-in-status if they satisfied the following condition:

New-in-status = 1 if $(R_{i t} \geq ({\bar{R}}_{i (t - 5)} + 2 \times σ_{i (t - 5)}) | R_{i t} \leq ({\bar{R}}_{i (t - 5)} - 2 \times σ_{i (t - 5)}))$ and 0 otherwise, where $R_{i t}$ is bank i’s status in current year (t), ${\bar{R}}_{i (t - 5)}$ is bank i‘s mean status in five prior years, and $σ_{i (t - 5)}$ is the standard deviation of bank i’s status over the preceding five years. By this operationalization, 16% of all bank-years were designated as new-in-status. This variable is time varying.

Organizational age

We calculated age as the number of years since founding.

Control variables

An alternative explanation of status effect on syndication is that organizational resources drive adoption. Well-funded banks may lend to more borrowers, resulting in higher prominence in the lending network and greater capability to start syndication. It should be noted, however, that higher status does not automatically follow from the availability of funds. In the highly stratified Japanese financial world, great importance is attached to the identity of business partners. Borrowers are wary of obtaining loans from low-status financial institutions, hence actors are unlikely to become significant lenders based on resources alone. Nonetheless, to control for this alternative explanation we included variables measuring banks’ resources. Our data include information on banks’ total assets, cash reserves, liabilities, and revenues. Exploratory analyses revealed high correlations between some variables, potentially resulting in a multicollinearity problem. To avoid this, we include total assets and liabilities but omit cash reserves and revenues. The remaining variables are not correlated to the extent of creating a multicollinearity problem (Greene, 2003) and are included in the analysis (see Table 1).

Table 1.

Correlations and Descriptive Statistics for Variables in Analyses.

		mean	st.dev.	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16
1	Total assets/10⁶	5.10	15.09
2	Liabilities /10⁵	1.84	2.16	0.12
3	Profitability (ROA)	0.00	0.03	−0.02	−0.01
4	Bad loans /10⁶	0.08	0.25	0.18	0.06	−0.23
5	Client base /10⁶	0.96	0.75	0.38	0.12	−0.03	0.21
6	Number of offices	92.90	43.56	0.22	0.05	0.03	0.17	0.75
7	Foreign branches	0.46	1.65	0.88	0.26	−0.02	0.07	0.26	0.09
8	Foreign capital	0.00	0.01	0.01	0.00	0.01	−0.01	0.01	0.01	0.08
9	Regional bank	0.86	0.35	−0.67	−0.32	0.07	−0.16	−0.02	0.23	−0.66	0.01
10	Merger bank	0.05	0.01	0.04	0.00	0.02	−0.02	0.01	0.01	0.03	0.00	0.02
11	New-in-status	0.16	0.37	0.01	−0.03	0.01	0.05	0.06	0.04	−0.01	0.00	−0.01	0.01
12	Age	36.62	24.93	−0.05	−0.07	−0.01	0.07	0.16	0.23	−0.08	0.00	0.15	0.15	0.10
13	Status	0.52	0.26	0.11	0.05	−0.02	0.19	0.18	0.26	0.07	0.01	−0.12	−0.13	−0.06	0.06
14	Status squared	0.71	0.17	0.11	0.06	−0.02	0.20	0.19	0.25	0.07	−0.01	−0.14	−0.14	−0.05	0.05	0.99
15	Status-mobile from 1st tercile	0.04	0.10	−0.04	−0.03	−0.02	0.02	−0.06	−0.03	−0.04	0.00	0.03	0.03	0.23	0.01	−0.04	−0.05
16	Status-mobile from 2nd tercile	0.07	0.11	−0.02	−0.02	0.01	0.05	0.02	0.05	−0.02	0.01	0.03	0.03	0.23	0.04	0.06	0.07	−0.02
17	Status-mobile from 3rd tercile	0.05	0.12	0.02	−0.02	0.01	0.04	0.11	0.14	0.02	0.00	0.04	0.04	0.36	0.14	0.06	0.06	0.00	−0.03

Underperforming banks may use syndication as a means of getting more deals, thereby increasing profitability. To control for the possibility that poor financial performance drove the adoption of syndication we included variables capturing profitability (operationalized by return on assets, ROA, a widely used measure in the financial literature) and bad loans (calculated as the sum of loan amounts not repaid to banks by borrowers due to bankruptcy and nonperforming loans).

Client base (number of borrowers)

We included a count of a bank’s clients in order to control for both the influence of a broader market niche, and the possibility that syndication is driven by a need to consolidate the bank’s resources with other lenders. The number of borrowers also controls for the effect of a bank’s network on adoption (cf. Palmer & Barber, 2001). In particular, centrally embedded major banks have a far larger client base than weakly embedded fringe actors (Lincoln & Gerlach, 2004).

Number of offices

We include a count of bank offices to account for size and concomitant organizational inertia due to bureaucratization and routine ossification, which may make larger banks less flexible in adopting new practices.

Foreign branches

While loan syndication was new to Japan in the mid-1980s, the practice had been in existence in the global financial markets since the late 1970s. The adoption of a new practice may be influenced by a firm’s previous exposure to that practice; we control for this with a count of the bank’s foreign branches.

Regional banks

We included a dummy variable for regional banks to control for different opportunities and motivations presented by geographic location.

Merger bank

The Japanese banking industry went through a period of consolidation in the 1990s; as a result, some banks in our data were created through mergers. To control for the possibility of status spillover from “ancestor banks” (and, perhaps, different assumptions about their legitimacy among audiences) we included a dummy variable indicating whether a bank was formed through a merger.

Foreign capital

The presence of foreign shareholder capital has been shown to be positively related to Japanese firms’ propensity to challenge institutionalized behaviors (Ahmadjian & Robinson, 2001) and thus may affect banks’ propensity to adopt norm-deviant practices. We included a dummy variable indicating whether a bank has foreign owners.

Model

Since we were interested in banks’ propensity to start loan syndication, event history models that estimate the hazard of starting syndication as a function of independent variables were appropriate for our analyses. One methodological issue to address concerned the type of model to use and, more specifically, the shape of the underlying distribution of events with respect to time. While non-parametric Cox models provide flexibility in that they make no assumptions about the underlying hazard rate, they can be inefficient (Hosmer, Lemeshow, & May, 2008). We tested accelerated time models based on Weibull and Gompertz distributions, and compared their fit to one another, as well as the Cox and exponential models, which assume a constant hazard rate (Tuma & Hannan, 1984). Based on model fit statistics (LR tests and AIC comparisons) we established that Gompertz models provided the best fit to the data. The model takes the form:

h (t) = e^{(α + γ t)} e^{(β X)}

where h(t) is the hazard rate, $β$ is the vector of coefficients, α is a constant, and γ is the time function parameter (expected to be > 0). Since we are interested in decisions to start loan syndication, banks are in our risk set as long as they are active in the market (provide loans) and exit the risk set after their first instance of syndication.³

Results

Table 2 presents our models. Model 1 includes control variables only. Model 2 adds the moderating variables, age and new-in-status. The coefficient on the new-in-status variable is negative and highly statistically significant (p < 0.05), indicating that banks with recently acquired status positions were less likely to start syndication, compared to banks with long-lasting status positions. The age variable’s coefficient is positive and highly statistically significant (p < 0.05), indicating that younger banks were less likely to start syndication than banks with longer market tenure.

Table 2.

Gompertz Models: Adoption of Loan Syndication by Japanese Banks, 1983–2005.

	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Const.	−8.373***	−10.204***	−10.042***	−10.269***	−8.562***	−12.314***
	(1.135)	(1.842)	(1.534)	(1.575)	(1.537)	(2.782)
Total assets	3.365	3.365	7.484**	7.221**	7.345**	7.327**
	(2.047)	(3.005)	(3.100)	(3.414)	(3.463)	(3.444)
Liabilities	−0.064	−0.444	−0.799**	−0.868**	−0.844**	−0.845**
	(0.231)	(0.316)	(0.378)	(0.384)	(0.389)	(0.379)
Profitability (ROA)	9.359	12.565	17.214	17.111	15.178	13.643
	(10.759)	(18.503)	(18.077)	(19.295)	(18.183)	(16.655)
Bad loans	0.469	0.328	−0.340	−0.072	−0.330	−0.216
	(0.441)	(0.485)	(1.379)	(1.419)	(1.367)	(1.640)
Client base	−0.793**	−0.997**	−0.967**	−1.207**	−0.991**	−1.000**
	(0.294)	(0.454)	(0.460)	(0.470)	(0.460)	(0.407)
Nr offices	0.030***	0.025***	0.026***	0.027***	0.030***	0.030***
	(0.004)	(0.006)	(0.006)	(0.006)	(0.006)	(0.006)
Nr foreign branches	−0.505***	−0.434**	−0.545**	−0.533**	−0.545**	−0.546**
	(0.155)	(0.222)	(0.224)	(0.244)	(0.204)	(0.224)
Regional bank	−0.600	−0.446	0.758	1.310	0.622	1.672
	(0.964)	(1.259)	(1.626)	(1.461)	(1.466)	(1.835)
Merger bank	0.323	0.335	0.391	0.442	0.259	0.362
	(1.322)	(1.425)	(1.324)	(1.473)	(1.168)	(1.365)
Foreign capital	0.001	0.001	0.001	0.001	0.001	0.001
	(0.921)	(0.921)	(0.921)	(0.921)	(0.921)	(0.921)
New-in-status		−0.113**	−0.123**	−0.154**	−0.143	−0.173
		(0.056)	(0.061)	(0.077)	(0.104)	(0.177)
Age		0.030***	0.011***	0.014**	0.011***	0.016**
		(0.010)	(0.004)	(0.007)	(0.005)	(0.008)
Status			−7.280	−8.380*	−8.532	−9.387
			(7.621)	(5.771)	(8.966)	(9.237)
Status sq			6.634**	9.078***	8.989*	9.736*
			(3.312)	(1.202)	(5.989)	(6.487)
Status × Age				−0.082*		−0.085
				(0.051)		(0.064)
Status sq × Age				0.078**		0.087**
				(0.039)		(0.043)
Status × New-in-status					4.954***	4.888**
					(1.057)	(2.432)
Status sq × New-in-status					−4.821**	−4.911**
					(2.399)	(2.443)
N (bank years)	2,081	2,081	2,081	2,081	2,081	2,081
Mean VIF	1.42	1.61	1.85	2.08	2.31	2.64
Log likelihood	289.15***	297.34***	305.42***	307.67***	309.34***	314.31***

Note: standard errors in parentheses. *p < .10; **p < .05; ***p < .001.

Model 3 adds the main and squared terms of our independent variable, status. In this model, the coefficient for the main effect of status is negative, while the coefficient for the squared term of status is positive and statistically significant (p = 0.042). This combination indicates a U-shaped relationship between status and the propensity to adopt loan syndication, lending support to hypothesis 1.

In the next set of models (4 and 5), we examine the interaction effects of the moderating variables on the relationship between status and syndication adoption. In model 4, the interaction between age and status is negative, and that between age and the squared term of status is positive and significant (p < 0.05). Figure 1 depicts this relationship. In the case of older banks, high and low status results in greater propensity to syndicate than middle status. In younger banks, the line representing the relationship between status and propensity to syndicate is flatter, indicating that status has a relatively smaller impact on the adoption of syndication when status positions are not cemented. Compared to older banks, younger low- and high-status banks are less likely to adopt syndication, whereas there is no difference between older and younger middle-status banks. These results lend support to hypothesis 2.

Figure 1.

The moderating effect of age on organizational status in adoption of loan syndication by Japanese banks, 1983–2005.

In model 5 the interaction between new-in-status and status variables is positive and significant (p = 0.001), while that between new-in-status and the squared term of status is negative and significant (p = 0.048), providing support hypothesis 2. Figure 2 illustrates this relationship.⁴ In the case of banks entrenched in their status positions (i.e., status positions not acquired recently), high and low status results in a greater propensity to syndicate than middle status—similar to the effect of being new in a status position. In the case of banks that are new in their status positions, the U-shaped relationship holds, but its shape is flatter, indicating that weakly entrenched status positions have a relatively smaller impact on the adoption of syndication. These results lend strong support to hypothesis 3.

Figure 2.

The moderating effect of newness in a status position on organizational status in adoption of loan syndication by Japanese banks, 1983–2005.

Model 6 combines the interactions of age and new-in-status variables with the main and squared effects of status. The direction and statistical significance of the interaction effects remain in model 6, indicating that the two status entrenchment effects obtain independently and together. This speaks to the robustness of results from models 4 and 5.

As Figures 1 and 2 illustrate, at medium levels of status the difference between entrenched and un-entrenched banks’ propensity to adopt syndication is smallest (insignificant for the age variable), indicating that the flatter U-shape is due to the diminished propensity of high- and low-status banks to adopt syndication.

Robustness checks

To ensure the robustness of our result and validity of our interpretation we perform additional analyses following state-of-the-art methodological recommendations (Haans, Pieters, & He, 2015; Lind & Mehlum, 2010), as well as prior empirical studies (e.g., Matusik & Fitza, 2012; Phillips & Zuckerman, 2001). First, we identify the inflection point in the curvilinear effect of status to ensure it falls within the range of available data: we take the first (partial) derivative of the model equation with respect to status (X): $\frac{\partial Y}{\partial X} = - 7.280 + 2 \times 6.634 X$ . Equating this to 0 and solving for X yields the inflection point at x = 0.55, well within the range of the independent variable’s observed values (0–1), providing confidence in the results’ robustness (Haans et al., 2015). We next split the sample according to the inflection point. We denote the subsample with x ⩽ 0.55 “low-status” and that with x > 0.55 “high-status” and perform analyses of the (status × new-in-status) and (status × age) interactions on each subsample. The results, shown in Table 3, should be consistent with the curvilinear specification used on the full sample.

Table 3.

Gompertz Models: Adoption of Loan Syndication by Japanese Banks of Low and High Status, 1983-2005.

	Low status					High status
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
Const.	−11.206***	−10.508***	−10.508***	−10.886***	−16.139***	−15.997***	−16.086***	−15.702***
	(3.095)	(2.903)	(2.903)	(3.024)	(5.028)	(4.983)	(5.011)	(4.892)
Total assets	8.347	9.245*	9.774*	9.402*	11.926**	11.469**	10.149**	11.587**
	(5.911)	(6.012)	(4.997)	(6.015)	(5.731)	(5.158)	(5.312)	(6.003)
Liabilities	−0.997	−1.126*	−1.133	−1.130*	−1.255*	−1.298**	−1.253*	−1.302**
	(0.737)	(0.647)	(0.647)	(0.647)	(0.652)	(0.640)	(0.656)	(0.643)
Profitability (ROA)	−10.286	−11.193	−11.138	−11.219	14.653	16.363	13.490	14.669
	(8.257)	(8.986)	(8.942)	(9.007)	(11.763)	(13.137)	(10.830)	(11.777)
Bad loans	5.707**	5.753*	5.467*	5.475*	−1.256	−1.240	−1.840	−1.104
	(2.480)	(2.548)	(2.471)	(2.548)	(1.772)	(1.635)	(1.851)	(1.389)
Client base	−2.219***	−2.275**	−2.128***	−2.175**	0.189	0.144	0.155	0.038
	(0.640)	(0.656)	(0.654)	(0.676)	(0.670)	(0.682)	(0.648)	(0.655)
Nr offices	0.040	0.047***	0.041***	0.042***	0.016*	0.015*	0.015*	0.015*
	(0.012)	(0.012)	(0.012)	(0.012)	(0.009)	(0.009)	(0.009)	(0.009)
Nr foreign branches	−0.245	−0.180	−0.180	−0.169	−0.476	−0.512*	−0.470	−0.502
	(0.427)	(0.430)	(0.430)	(0.437)	(0.300)	(0.301)	(0.299)	(0.297)
Regional bank	0.787	0.932	0.939	0.777	−1.772	−1.418	−1.791	−1.202
	(0.633)	(0.750)	(0.756)	(0.625)	(1.742)	(1.867)	(1.859)	(1.841)
Merger bank	0.002	0.002	0.002	0.002	0.731	0.698	0.734	0.868
	(0.888)	(0.732)	(0.637)	(0.222)	(1.864)	(1.236)	(1.464)	(1.672)
Foreign capital	0.001	0.001	0.001	0.001	0.001	0.001	0.001	0.001
	(0.921)	(0.921)	(0.921)	(0.921)	(0.921)	(0.921)	(0.921)	(0.920)
New-in-status	−0.717**	−0.849*	−0.808*	−0.751	2.241**	2.549**	3.077*	3.306
	(0.351)	(0.511)	(0.486)	(0.515)	(1.138)	(1.256)	(1.854)	(2.962)
Age	0.025**	0.034***	0.025**	0.032*	−0.034**	−-0.044*	−0.043**	−0.045
	(0.013)	(0.011)	(0.013)	(0.019)	(0.017)	(0.026)	(0.022)	(0.469)
Status	−2.175**	−3.314**	−2.658*	−3.441	4.719***	5.118*	5.911*	5.647
	(1.104)	(1.682)	(1.601)	(9.217)	(1.204)	(3.065)	(3.552)	(4.948)
Status × Age		−0.031**		−0.033**		0.081**		0.076**
		(0.016)		(0.017)		(0.041)		(0.039)
Status × New-in-status			2.382**	2.193**			−4.614**	−3.971**
			(1.209)	(1.113)			(2.310)	(2.016)
N (bank years)	936	936	936	936	1,145	1,145	1,145	1,145
Mean VIF	1.78	2.24	2.36	2.54	1.94	2.21	2.15	2.47
Log Likelihood	139.25***	148.31***	151.22***	156.29***	210.21***	212.55***	212.44***	217.38***

Note: standard errors in parentheses. *p < .10; **p < .05; ***p < .001.

Models 1 and 5 include control variables, moderating variables, and status. The status variable coefficient is highly significant in both models (p = 0.034 and p = 0.003). It is negative in the “low-status” subsample and positive in the “high-status” subsample, confirming the robustness of the full sample’s curvilinear specification. Models 2 and 3 evaluate the interaction of status with age and new-in-status variables, respectively, in the “low-status” subsample, while models 6 and 7 do so in the “high-status” subsample. Again, consistent with the full sample specification, each interaction is significant. Graphs (a) to (d) in Figure 3 illustrate these interactions and their consistency with the full sample model: un-entrenched banks (young banks and new-in-status) are less likely to adopt syndication than highly entrenched low- and high-status banks.

Figure 3.

The moderating effect of age and newness in a status position on organizational status in adoption of loan syndication by Japanese banks, 1983–2005. Low- and high-status subsamples.

Another robustness check concerned banks’ legal standing, to which status is often linked (e.g., Krishnan & Kozhikode, 2015). Organizations engaging in illicit behavior may be more prone to also adopt norm-deviant behavior. We collected data on banks’ illegal behavior from Japan’s financial regulatory agency—the Financial Services Agency—and included this variable in our models. Few banks were fined by the agency during the period of our analysis and we found no significant effect of this variable.

Status mobility origin and adoption behavior

While our analyses lend support to hypotheses 1 to 3, we consider the following additional question: Do banks’ adoption behaviors depend on where in the status hierarchy new-in-status banks arrive from? Thus far we have treated all new-in-status banks equally, disregarding their previous status positions. However, the propensity of actors to engage in norm-deviant practices may differ depending on their origin status position. To address this question, we split the full sample into three terciles, corresponding to low, middle, and high status.⁵ We identified the tercile of origin for banks that experienced status mobility and were new in their status positions. Accordingly, we created three dummy variables: status-mobile from 1st tercile, status-mobile from 2nd tercile, and status-mobile from 3rd tercile, based on the bank’s location in the previous five years.⁶ Comparing the interaction effects of these variables with the status variable for appropriate terciles allows us to assess differences in the effects of status mobility on banks’ propensity to engage in syndication. This analysis is reported in Table 4.

Table 4 contains three panels corresponding to sub-samples based on the 1st, 2nd, and 3rd status terciles. The results show that in the 3rd status tercile, new-in-status banks arriving from the 1st or the 2nd tercile are less likely to syndicate, compared to banks entrenched in the 3rd tercile positions. The same pattern holds in the 1st status tercile. In the 2nd status tercile, the coefficient on the status variable and the interactions with origin terciles are not significant. This finding is consistent with the curvilinear relationship between status and adoption: in the middle of the status hierarchy the relationship between status and adoption is expected to be neither positive nor negative.

Table 4.

Gompertz Models: Adoption of Loan Syndication by Japanese Banks, 1983-2005. Tercile Analysis.

	1st status tercile				2nd status tercile				3rd status tercile
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8	Model 9	Model 10	Model 11	Model 12
Const.	−8.396	−8.767***	−8.497***	−8.322***	−11.292***	−11.068***	−11.682***	−11.380***	−15.902***	−15.945***	−15.861***	−15.861***
	(5.780)	(2.159)	(2.853)	(2.188)	(4.573)	(4.568)	(4.653)	(4.647)	(9.389)	(9.720)	(4.822)	(9.812)
Total assets	20.381	12.883	22.123	11.319	27.453**	27.746**	27.702**	28.114**	8.164	9.221	9.363	9.993*
	(26.570)	(28.236)	(25.767)	(29.742)	(13.936)	(14.084)	(14.062)	(14.271)	(6.280)	(5.841)	(6.659)	(5.624)
Liabilities	−2.568	−1.545	−2.457	−1.115	−3.061**	−3.084**	−3.122**	−3.149**	−0.837	−0.916	−0.924	−1.029
	(2.937)	(3.222)	(2.961)	(3.314)	(1.132)	(1.150)	(1.138)	(1.158)	(0.571)	(0.600)	(0.609)	(0.645)
Profitability (ROA)	12.702	11.085	12.392*	12.448	11.056	11.297	13.937	11.427	2.245	2.751	3.235	3.936
	(12.616)	(10.399)	(7.985)	(12.983)	(11.230)	(11.784)	(12.091)	(11.290)	(2.718)	(2.028)	(3.796)	(3.298)
Bad loans	8.223	10.813*	8.023	10.109*	2.860	2.098	2.872	2.389	−1.290	−1.349	−1.205	−1.302
	(5.075)	(6.295)	(5.394)	(6.555)	(3.544)	(3.805)	(3.151)	(3.484)	(1.829)	(2.993)	(1.893)	(2.049)
Client base	0.505	0.772	0.569	0.883	−1.046*	−1.033*	−1.045*	−1.416*	−1.303	−1.141	−1.501	−1.621
	(1.199)	(1.094)	(1.223)	(1.118)	(0.605)	(0.602)	(0.609)	(0.911)	(0.836)	(0.836)	(0.837)	(0.837)
Nr offices	0.004	0.005	−0.007	−0.011	0.048**	0.048**	0.044**	0.044**	0.038**	0.032**	0.028**	0.046*
	(0.045)	(0.046)	(0.051)	(0.053)	(0.017)	(0.017)	(0.017)	(0.016)	(0.012)	(0.012)	(0.011)	(0.012)
Nr foreign branches	1.244	1.010	1.103	0.940	−0.288	−0.282	−0.253	−0.267	−1.101***	−1.107***	−1.210**	−1.223**
	(1.033)	(1.152)	(0.994)	(1.110)	(0.436)	(0.457)	(0.451)	(0.477)	(0.322)	(0.338)	(0.340)	(0.317)
Regional bank	1.749	1.726	1.543	1.964	0.030	0.026	0.040	0.059*	4.662*	5.080*	4.810*	5.283**
	(1.764)	(1.562)	(1.471)	(1.977)	(0.021)	(0.018)	(0.040)	(0.035)	(2.481)	(2.651)	(2.509)	(2.688)
Merger bank	0.276	0.326	0.252	0.322	0.003	0.004	0.004	0.005	1.211	1.320	1.324	1.443
	(0.321)	(0.343)	(0.289)	(0.300)	(0.006)	(0.006)	(0.006)	(0.006)	(1.235)	(1.132)	(1.314)	(1.379)
Foreign capital	0.001	0.001	0.001	0.001	0.001	0.001	0.001	0.001	0.001	0.001	0.001	0.001
	(0.031)	(0.031)	(0.031)	(0.031)	(0.031)	(0.031)	(0.031)	(0.031)	(0.031)	(0.031)	(0.031)	(0.031)
Age	−0.002	−0.002	−0.002	−0.001	0.017	0.017	0.017	0.017	0.007	0.007	0.008	0.009
	(0.021)	(0.022)	(0.020)	(0.022)	(0.013)	(0.013)	(0.013)	(0.013)	(0.013)	(0.013)	(0.014)	(0.014)
status-mobile from 1st tercile (†)					0.643**	0.795**	0.742*	0.765*	1.026**	1.286**	1.534*	1.413
					(0.315)	(0.397)	(0.388)	(0.321)	(0.492)	(0.587)	(1.166)	(1.650)
status-mobile from 2nd tercile	−0.892**	−1.014	−1.082*	−1.218					9.343**	1.034*	0.964	1.113
	(0.356)	(0.987)	(0.588)	(0.879)					(4.672)	(0.616)	(0.866)	(1.113)
status-mobile from 3rd tercile (†††)	−0.673*	−0.861**	−0.853**	−0.887	0.973**	1.056	1.124**	1.332*
	(0.456)	(0.463)	(0.424)	(0.870)	(0.501)	(1.120)	(0.600)	(0.766)
Status	−4.022**	−4.554	−4.713	−4.736	0.084	0.094	0.078	0.087	5.208**	5.122	5.265*	5.717
	(1.904)	(4.766)	(4.030)	(4.401)	(1.121)	(1.340)	(1.224)	(1.213)	(2.433)	(5.257)	(4.203)	(4.404)
Status × Mobile from 1st tercile (Ƭ)							0.397	0.431		−2.342***		−2.619**
							(0.621)	(0.514)		(0.588)		(1.212)
Status × Mobile from 2nd tercile (ƬƬ)		2.186***		2.019**							−3.281***	−3.010***
		(0.633)		(0.993)							(0.845)	(0.836)
Status × Mobile from 3rd tercile (ƬƬƬ)			1.253**	1.601**		0.753		0.643
			(0.634)	(0.743)		(0.764)		(0.754)
Wald test				(ƬƬ=ƬƬƬ) 2.03				(†=†††) 3.44*				(Ƭ=ƬƬ) 1.53
N (bank years)	692	692	692	692	692	692	692	692	692	692	692	692
Log Likelihood	112.81***	115.33***	116.82***	121.68***	100.12***	104.22***	101.14***	103.23***	120.41***	126.13***	128.53***	131.16***

Note: standard errors in parentheses. *p < .10; **p < .05; ***p < .001.

To assess whether there is material difference in the propensity to start syndicating between banks arriving from different terciles we performed the Wald test for difference in interaction coefficients in the 1st and 3rd status tercile subsamples. The Wald statistic for both the 1st and 3rd terciles are not significant at p = 0.05. Thus, we fail to conclude that the propensity to adopt syndication varies depending on the origin tercile of status-mobile banks. This implies that at the high end of the status distribution, new-in-status banks are less likely to adopt syndication than banks entrenched in the 3rd tercile, but that there is no difference between banks arriving from the middle or the bottom of the status distribution. Analogous results hold in the 1st tercile of the status distribution.

In the 2nd tercile where neither status nor the interaction between status and origin tercile are significant, we performed a test of difference in the main effect of origin tercile. The test statistic (3.44 on 1 d.f.) is marginally statistically significant (p = 0.087). We interpret this result as indicating a difference in the level of adoption between new-in-status and entrenched banks in the 2nd status tercile (higher for new-in-status banks, consistent with the curvilinear specification in Table 2), but no substantial difference in the adoption propensity of banks arriving from the 1st or the 3rd status tercile. This lends support to the proposition that the effect of status mobility on adoption of norm-deviant behavior is driven by actors’ level of entrenchment in their current positions, as opposed to their origin status positions.

Overall, the status tercile analyses indicate that regardless of origin, new-in-status banks are less likely to follow the adoption behavior prevalent at the newly acquired status level: banks that recently moved into low- or high-status positions are less likely to adopt syndication, while those moving into the middle of the status hierarchy are more likely to adopt syndication, compared to banks entrenched in the respective status positions. The finding that the same phenomenon (being in a recently acquired status position) based on two different mechanisms (age and status mobility) leads to the same type of behavior speaks to the validity of our explanation focusing on socio-psychological aspects of being new in a status position. It also helps us address the question of whether it is being new in status or status change per se (status gain or loss) that drives adoption patterns. Since our empirical findings indicate that both age- and status mobility-based mechanisms result in the same pattern of adoption behavior, we have greater confidence that the observed effects are due to being new in status, rather than status change.

Discussion and Conclusions

In this study we examine how status entrenchment impacts the effect of status on the adoption of norm-deviant practices. Our core argument is that weakly entrenched organizations are less likely to exhibit adoption behaviors commonly associated with their status position. Proposing that entrenchment depends on status mobility and age, we elaborate on their underlying mechanisms.

Examining the adoption of loan syndication in the Japanese banking industry yields support to our theorizing. The relationship between status and adoption is U-shaped and moderated by status mobility and age. A flatter U-shape for new-in-status and younger banks indicates that these organizations are less likely to follow the adoption pattern prevalent at their status levels. Status mobility and young age among low- and high-status banks reduce these banks’ propensity to adopt syndication, while status mobility increases adoption among middle-status banks.

Our results indicate that status mobility tempers organizations’ propensity to engage in status-based behavior, regardless of their original status. This finding is consistent with the notion that status-based behavior with respect to norm-deviance is determined by organizations’ legitimacy concerns, which are moderated by status entrenchment. While we find little evidence that this dynamic is affected by the organizations’ previous status, the results merit further examination in future work.

This study joins a growing literature on the implications of status fluidity for organizational action (Adut, 2005; Askin & Bothner, 2016). While prior research has examined organizational responses to status change itself, our contribution lies in elaborating how being new in status affects the extent to which organizations engage in status-specific behaviors. By highlighting how entrenchment affects actors’ propensity to adopt new non-conformist practices, our findings emphasize the importance of considering not only actors’ status positions, but also how stable and taken-for-granted these are, among both the organizations themselves and audiences. Future research should provide additional examination of this phenomenon by analyzing other possible factors affecting actors’ identification with their status positions, e.g., how different (and perhaps divergent) status markers, or speed of status mobility, affect entrenchment and adoption behavior.

By considering actors’ entrenchment in status positions, we contribute to overcoming the implicit but prevailing assumption that actors in the same status positions face equivalent legitimacy considerations and behavioral predispositions. Extending earlier work (e.g., Jensen, 2003, Krishnan & Kozhikode, 2015, Prato et al., 2018), our conceptualization of “status entrenchment” relaxes this assumption, thereby developing the idea that actors of equal status may differ in their motivations and, as a result, have different propensities for non-conformity (Phillips & Zuckerman, 2001).

Generalizability and boundary conditions

We believe the relationships between status entrenchment and the adoption of norm-deviant practices will hold in contexts beyond Japanese banking and loan syndication. The mechanisms that we theorize and test are not specific to Japan; extant work has highlighted the effects of age, status mobility, and status anxiety in a host of different national and industry contexts (Askin & Bothner, 2016; Delmestri & Greenwood, 2016; Krishnan & Kohzikode, 2015). As discussed in the methods section, the Japanese loan syndication setting was selected because it corresponds to the scope conditions that are crucial in testing status effects (Phillips & Zuckerman, 2001; Podolny et al., 1996); in this way it serves as a particularly revelatory setting for examining our hypotheses. Furthermore, Japanese firms—and banks in particular—have been shown to display significant levels of mimetic isomorphism and homogeneity in their strategies (Delios & Henisz, 2003; Greve, 2000). This trait minimizes the risk that banks’ adoption behaviors are due to exposure to different customers with heterogeneous demands and risk profiles.

Our findings are subject to several boundary conditions. First, we expect our theorizing to hold more for organizational as opposed to technical practices. Because technical innovations often provide instrumental benefits that depend less on social acceptance, status and legitimacy may play smaller roles in adoption (e.g., Kimberly & Evanisko, 1981). Adoption of technical innovations may also require a greater ability to overcome organizational inertia and engage in exploration, where younger organizations have an advantage (e.g., Hannan & Carroll, 2004). Thus our findings regarding the age effect may especially be less applicable to adoption of technical innovations. Second, our hypotheses assume markets where quality and performance are difficult to evaluate, or alternatively, where non-market factors associated with status constitute significant measures of worth. This assumption comes from the recognition that in markets where audiences receive easily verifiable information about performance, status will be less important in evaluating actors. In such contexts, audiences may be quicker to withdraw endorsements, status may be less “sticky,” and entrenchment processes may be less pronounced.

Our results are more likely to be replicable in settings characterized by relatively stable organizational populations and audiences. The formation of audiences’ perceptions—on which our mechanisms depend—will be hampered when audiences shift rapidly. Similarly, when organizational populations are characterized by large inflows and outflows, the effects of behavioral incentives linked to status positions may be dampened. Finally, as noted above, previous work has indicated a high level of imitation and homogeneity among Japanese banks. While this trait helps to minimize the risk of confounding effects, it also reduces generalizability to markets characterized by greater heterogeneity and significant market segmentation. In such settings, organizations’ adoption of norm-deviant practices may be driven more by focus on different customer segments, as opposed to status. Therefore future work should focus on replication in such markets.

Footnotes

Acknowledgements

We are grateful to Matthew Bothner, Francois Collet, Gokhan Ertug, Yonghoon Lee, and Matteo Prato for valuable comments on earlier drafts of the article and to seminar participants at ESMT, Hitotsubashi University, Stanford University, Waseda University, and the COORS Tokyo group for useful feedback.

Funding

Jesper Edman acknowledges funding from the Japan Society for the Promotion of Science, grant number 25780232.

ORCID iDs

Jesper Edman

Alex Makarevich

Notes

Author biographies

Jesper Edman is an Associate Professor at Waseda University in Tokyo, Japan. His research focuses on how organizational identities shape the diffusion and impact of global practices on local institutions and market structures.

Alex Makarevich is an Associate Professor of Management at California State University, East Bay. He studies how embeddedness in inter-organizational relationships shapes the performance of firms and their alliances in a variety of empirical contexts.

References

Abrahamson

(1991). Managerial fads and fashions: The diffusion and rejection of innovations. Academy of Management Review, 16, 586–612.

Adut

(2005). A theory of scandal: Victorians, homosexuality, and the fall of Oscar Wilde. American Journal of Sociology, 111, 213–248.

Ahmadjian

C. L.

Robinson

(2001). Safety in numbers: Downsizing and the deinstitutionalization of permanent employment in Japan. Administrative Science Quarterly, 46, 622–654.

Askin

Bothner

M. S.

(2016). Status-aspirational pricing: The “Chivas Regal” strategy in US higher education, 2006–2012. Administrative Science Quarterly, 61, 217–253.

Becker

M. H.

(1970). Sociometric location and innovativeness: Reformulation and extension of the diffusion model. American Sociological Review, 35, 267–282.

Bitektine

(2011). Toward a theory of social judgments of organizations: The case of legitimacy, reputation, and status. Academy of Management Review, 36, 151–179.

Blader

S. L.

Chen

Y. R.

(2012). Differentiating the effects of status and power: A justice perspective. Journal of Personality and Social Psychology, 102, 994.

Bothner

M. S.

Kang

J. H.

Lee

(2006). Status volatility and organizational growth in the US venture capital industry. Academy of Management Proceedings, 1, 1–6.

Bothner

M. S.

Podolny

J. M.

Smith

E. B.

(2011). Organizing contests for status: The Matthew effect vs. the Mark effect. Management Science, 57, 439–457.

10.

Bothner

M. S.

Smith

E. B.

White

H. C.

(2010). A model of robust positions in social networks. American Journal of Sociology, 116(3), 943–992.

11.

Cattani

Ferriani

Negro

Perretti

(2008). The structure of consensus: Network ties, legitimation, and exit rates of US feature film producer organizations. Administrative Science Quarterly, 53, 145–182.

12.

Choi

Y. R.

Shepherd

D. A.

(2005). Stakeholder perceptions of age and other dimensions of newness. Journal of Management, 31, 573–596.

13.

Delios

Henisz

W. J.

(2003). Political hazards, experience, and sequential entry strategies: The international expansion of Japanese firms, 1980–1998. Strategic Management Journal, 24, 1153–1164.

14.

Delmestri

Greenwood

(2016). How Cinderella became a queen: Theorizing radical status change. Administrative Science Quarterly, 6, 1–44.

15.

Deutsch

Ross

T. W.

(2003). You are known by the directors you keep: Reputable directors as a signaling mechanism for young firms. Management Science, 49, 1003–1017.

16.

Durand

Kremp

P. A.

(2016). Classical deviation: Organizational and individual status as antecedents of conformity. Academy of Management Journal, 59, 65–89.

17.

Edman

(2016). Cultivating foreignness: How organizations maintain and leverage minority identities. Journal of Management Studies, 53, 55–88.

18.

Elsbach

K. D.

Kramer

R. M.

(1996). Members’ responses to organizational identity threats: Encountering and countering the Business Week rankings. Administrative Science Quarterly, 41, 442–476.

19.

George

Dahlander

Graffin

S. D.

Sim

(2016). Reputation and status: Expanding the role of social evaluations in management research. Academy of Management Journal, 59, 1–13.

20.

Greene

W. H.

(2003). Econometric analysis. Delhi: Pearson Education.

21.

Greve

H. R.

(2000). Marketing niche entry decisions: Competition, learning, and strategy in Tokyo banking, 1894–1936. Academy of Management Journal, 43, 816–836.

22.

Haans

R. F.

Pieters

Z. L.

(2015). Thinking about U: Theorizing and testing U- and inverted U-shaped relationships in strategy research. Strategic Management Journal, 37, 1177–1195.

23.

Hannan

M. T.

Carroll

(2004). Demography of corporations and industries. Princeton, NJ: Princeton University Press.

24.

Hsu

(2006). Jacks of all trades and masters of none: Audiences’ reactions to spanning genres in feature film production. Administrative Science Quarterly, 51, 420–450.

25.

Hsu

Hannan

M. T.

Koçak

. (2009). Multiple category memberships in markets: An integrative theory and two empirical tests. American Sociological Review, 74, 150–169.

26.

Hosmer

D. W.

Lemeshow

May

. (2008) Applied survival analysis. Hoboken, NJ: John Wiley & Sons.

27.

Jensen

(2003). The role of network resources in market entry: Commercial banks’ entry into investment banking, 1991–1997. Administrative Science Quarterly, 48, 466–497.

28.

Jensen

(2006). Should we stay or should we go? Accountability, status anxiety, and client defections. Administrative Science Quarterly, 51, 97–128.

29.

Jensen

Kim

B. K.

Kim

(2011). The importance of status in markets: A market identity perspective. In Pearce

J. L.

(Ed.), Status in management and organizations (pp. 87–117). Cambridge: Cambridge University Press.

30.

Jonsson

(2009). Refraining from imitation: Professional resistance and limited diffusion in a financial market. Organization Science, 20, 172–186.

31.

Jonsson

Greve

Fujiwara-Greve

(2009). Undeserved loss: The spread of legitimacy loss to innocent organizations in response to reported corporate deviance. Administrative Science Quarterly, 54, 195–229.

32.

Kamins

M. A.

Alpert

F. H.

(2004). Corporate claims as innovator or market leader: Impact on overall attitude and quality perceptions and transfer to company brands. Corporate Reputation Review, 7(2), 147–160.

33.

Kimberly

J. R.

Evanisko

M. J.

(1981). Organizational innovation: The influence of individual, organizational, and contextual factors on hospital adoption of technological and administrative innovations. Academy of Management Journal, 24, 689–713.

34.

Kodeih

Greenwood

(2014). Responding to institutional complexity: The role of identity. Organization Studies, 35, 7–39.

35.

Krishnan

Kozhikode

R. K.

(2015). Status and corporate illegality: Illegal loan recovery practices of commercial banks in India. Academy of Management Journal, 58, 1287–1312.

36.

Le Mens

Hannan

Polos

. (2015). Organizational obsolescence, drifting tastes, and age dependence in organizational life chances. Organization Science, 26, 550–570.

37.

Lincoln

J. R.

Gerlach

M. L.

(2004). Japan’s network economy: Structure, persistence, and change. Cambridge: Cambridge University Press.

38.

Lind

J. T.

Mehlum

(2010). With or without U? The appropriate test for a U-shaped relationship. Oxford Bulletin of Economics and Statistics, 72(1), 109–118.

39.

Marsh

C. P.

Coleman

A. L.

(1956). Group influences and agricultural innovations: Some tentative findings and hypotheses. American Journal of Sociology, 61, 588–594.

40.

Matusik

S. F.

Fitza

M. A.

(2012). Diversification in the venture capital industry: Leveraging knowledge under uncertainty. Strategic Management Journal, 33, 407–426.

41.

Merton

R. K.

Merton

R. C.

(1968). Social theory and social structure. New York: Simon & Schuster.

42.

Palmer

Barber

B. M.

(2001). Challengers, elites, and owning families: A social class theory of corporate acquisitions in the 1960s. Administrative Science Quarterly, 46, 87–120.

43.

Pearce

J. L.

Q. J.

(2012). Rating performance or contesting status: Evidence against the homophily explanation for supervisor demographic skew in performance ratings. Organization Science, 23, 373–385.

44.

Phillips

D. J.

Turco

C. J.

Zuckerman

E. W.

(2013). Betrayal as market barrier: Identity-based limits to diversification among high-status corporate law firms. American Journal of Sociology, 118, 1023–1054.

45.

Phillips

D. J.

Zuckerman

E. W.

(2001). Middle-status conformity: Theoretical restatement and empirical demonstration in two markets. American Journal of Sociology, 107, 379–429.

46.

Podolny

J. M.

(1993). A status-based model of market competition. American Journal of Sociology, 98, 829–872.

47.

Podolny

J. M.

Stuart

T. E.

Hannan

M. T.

(1996). Networks, knowledge, and niches: Competition in the worldwide semiconductor industry, 1984–1991. American Journal of Sociology, 102, 659–689.

48.

Prato

Kypraios

Ertug

Lee

Y. G.

(2018). Middle-status conformity revisited: The interplay between achieved and ascribed status. Academy of Management Journal, doi.org/10.5465/amj.2017.0316.

49.

Rindova

V. P.

Pollock

T. G.

Hayward

M. L.

(2006). Celebrity firms: The social construction of market popularity. Academy of Management Review, 31, 50–71.

50.

Santos

F. M.

Eisenhardt

K. M.

(2009). Constructing markets and shaping boundaries: Entrepreneurial power in nascent fields. Academy of Management Journal, 52, 643–671.

51.

Sauder

(2006). Third parties and status position: How the characteristics of status systems matter. Theory and Society, 35, 299–321.

52.

Sauder

Lynn

Podolny

J. M.

(2012). Status: Insights from organizational sociology. Annual Review of Sociology, 38, 267–283.

53.

Scher

M. J.

(1998). Japanese interfirm networks and their main banks. Basingstoke, UK: Palgrave Macmillan.

54.

Sheard

(1989). The main bank system and corporate monitoring and control in Japan. Journal of Economic Behavior and Organization, 11, 399–422.

55.

Shipilov

A. V.

S. X.

(2008). Can you have your cake and eat it too? Structural holes’ influence on status accumulation and market performance in collaborative networks. Administrative Science Quarterly, 53, 73–108.

56.

Singh

J. V.

Tucker

D. J.

House

R. J.

(1986). Organizational legitimacy and the liability of newness. Administrative Science Quarterly, 31, 171–193.

57.

Stuart

T. E.

Hoang

Hybels

R. C.

(1999). Interorganizational endorsements and the performance of entrepreneurial ventures. Administrative Science Quarterly, 44, 315–349.

58.

Thornhill

Amit

(2003). Learning about failure: Bankruptcy, firm age, and the resource-based view. Organization Science, 14, 497–509.

59.

Tuma

N. B.

Hannan

M. T.

(1984). Social dynamics: Models and methods. Orlando, FL: Academic Press.

60.

Uchida

Udell

(2010). Banking in Japan. The Oxford handbook of banking, 2nd edition. Oxford: Oxford University Press.

61.

Washington

Zajac

E. J.

(2005). Status evolution and competition: Theory and evidence. Academy of Management Journal, 48, 282–296.

62.

Wasserman

Faust

(1994). Social network analysis: Methods and applications (Vol. 8). Cambridge: Cambridge University Press.

63.

Zhao

Zhou

(2011). Status inconsistency and product valuation in the California wine market. Organization Science, 22, 1435–1448.

64.

Zimmerman

M. A.

Zeitz

G. J.

(2002). Beyond survival: Achieving new venture growth by building legitimacy. Academy of Management Review, 27, 414–431.

65.

Zuckerman

E. W.

(1999). The categorical imperative: Securities analysts and the illegitimacy discount. American Journal of Sociology, 104, 1398–1438.

Entrenchment in status positions and the adoption of new norm-deviant organizational practices: Evidence from the Japanese banking industry,1983–2005

Abstract

Keywords

New Practice Adoption and Status Entrenchment

Status entrenchment and the adoption of norm-deviant practices

Entrenchment through tenure

Entrenchment through status stability

Hypotheses Development

Status and the adoption of loan syndication in the Japanese banking industry

Age-based entrenchment and the adoption of new norm-deviant practices

Status mobility and the adoption of norm-deviant practices

Method

Data

Independent variables

Status

New-in-status

Organizational age

Control variables

Client base (number of borrowers)

Number of offices

Foreign branches

Regional banks

Merger bank

Foreign capital

Model

Results

Robustness checks

Status mobility origin and adoption behavior

Discussion and Conclusions

Generalizability and boundary conditions

Footnotes

Acknowledgements

Funding

ORCID iDs

Notes

Author biographies

References