The Evolution of Structural Embeddedness and Organizational Social Outcomes in a Centrally Governed Health and Human Services Network

Provan, Keith G., Huang, Kun, Milward, H. Brinton, Journal of Public Administration Research and Theory

Much of the literature addressing publicly funded health and human services suggests that services are often fragmented and need to be integrated to keep clients from "falling through the cracks" (Agranoff 1991; Ferlie, Hawkins, and Kewell 2003; President's New Freedom Commission on Memal Health 2003). A fundamental assumption among many policy officials, funders, and service professionals who work in these areas is that an integrated network of service delivery is the most effective approach for providing clients with a seamless continuum of care. This logic is especially powerful when addressing so-called "wicked" problems (O'Toole 1997), such as homelessness, serving the flail elderly (Leishsenring 2004), disaster relief (Kapucu 2005), disabled children (Townsley, Abbott, and Watson 2004), workforce development (Herranz 2008), and serious mental illness (SMI) (Provan and Milward 1995). The problem is compounded by the fact that many of the people served are likely to be "vulnerable" clients (Aday 2001), who have multiple and significant needs and who are likely to have difficulty navigating the maze of a fragmented, un-integrated system on their own.

In the public management literature, the focus on services integration has given way to a broader focus on interorganizational networks (Agranoff and McGuire 2003; Graddy and Chen 2006; Herranz 2008; O'Toole 1997; Provan and Milward 1995). This focus requires a more systematic effort on the part of organizations and their managers to engage in such activities as the colocation of services, joint programs, liaison procedures, common client transfer and referral procedures, and information sharing. Although conclusive research on the impact of networks on client outcomes has been scarce, the general belief is that collaborating service providers will enhance overall effectiveness by utilizing resources more efficiently, recognizing and treating client problems better, and facilitating access to information and expertise (Provan and Milward 2001).

Although client-focused outcome indicators of effectiveness are certainly important, networks are likely to produce other outcomes that may be critical to overall network functioning. In particular, since networks are, essentially, social systems, they are likely to generate social outcomes related to network participants. In many publicly funded networks, especially in health and human services, nonprofit, public, and even for-profit organizations may be embedded in extensive networks of exchange involving referrals, contracts, joint programs, and information sharing. These ties are developed to enhance the level of services provided to clients. But they also have implications for how organizations relate to one another and may, in fact, explain why organizations choose to interact with some providers and not others. Specifically, organizational researchers have found that an organization's embeddedness, or involvement, especially in a voluntary exchange network, is related to a number of social indicators, including the organization's level of trustworthiness, its reputation for high-quality services, and its influence in the network (Galaskiewicz 1985; Laumann, Galaskiewicz, and Marsden 1978; Stuart, Hoang, and Hybels 1999). These aspects of network involvement by individual organizations constitute an important component of the social capital of the community (Putnam 1993) and, when aggregated from the organization to the network level, may contribute significantly to the building of community capacity (Chaskin et al. 2001).

The research reported here is an examination of the relationship between various task-oriented indicators of an organization's network embeddedness and three social outcomes: organizational trustworthiness, reputation, and influence. The research is an extension of previous cross-sectional work by two of the authors, Huang and Provan (2007a), who examined the network when the system was first established. The work presented here is longitudinal, comparing data collected four years after the system's creation. The system was centrally governed by a large national for-profit firm that was initially set up as a network administrative organization and then shifted to a lead organization model (Provan and Kenis 2008).

THEORETICAL BACKGROUND AND HYPOTHESES

The key assumption we make is that an organization's embeddedness with other organizations in a network will be based on its structural position in that network. In operational terms, this refers to the centrality of the organization. Network centrality describes the extent to which an organization is connected to other organizations (either directly or indirectly) within the web of exchanges that comprises a network (Diani 2002; Scott 1991). Central organizations are more embedded in the flow of information and resources in the network than noncentral, or peripheral actors, and have typically been found to have greater power (Cook and Emerson 1978). As noted above, we argue that trustworthiness, reputation, and influence are key social outcomes that are likely to be associated with a position of network embeddedness, operationalized as centrality. Trustworthiness, which is based on trust, is the extent to which an organization can be depended on to honor its commitments with its exchange partners. Reputation refers to perceptions by others that the organization is doing a good job, especially providing high-quality services to its clients. Influence refers to the extent to which an organization's views and actions are taken into consideration by other organizations when these organizations make their own important decisions.

In earlier work (Huang and Provan 2007a), we found modest support for our general arguments. We attributed some of the weakness of the results to the fact that the system was new, and thus, task-based exchanges and social outcomes among service providers had not yet had a chance to fully develop. By examining these relationships using a second wave of data, the strength of these conclusions can be reassessed. Our core research questions are as follows: (a) Is the relationship between network embeddedness and social outcomes proposed previously by Huang and Provan (2007a) supported in an established, as opposed to a newly formed network? (b) Does this relationship strengthen over time as the network evolves from early development toward maturity? and (c) To what extent has agency and overall system performance been affected as the network has evolved? To test the first two general arguments, six hypotheses are proposed. The last research question, on performance, will be examined more generally, to provide a better understanding of whether or not network maturity is related to effectiveness. The research is unusual in that it represents one of the few cases in the public management literature where full relational network data, not ego-centric data, are examined over time as the system evolves, using social network analysis methods (see Isett and Provan 2005, for an exception, and see Rethemeyer 2005, for a critique of how networks are typically studied in the public management literature).

An examination of the evolution in networks using quantitative relational data is exceptionally rare in the literature on organizations, regardless of sector focus. Not surprisingly, there is little theory available to guide development of our specific hypotheses on evolution and network structure. As a result, we provide theoretical support for our static hypotheses while then suggesting more generally how these relationships might be affected by the passage of time, and especially, by the institutionalization of a new system (Scott 2001). Our basic logic is that relationships in networks that are early in their development, following a major institutional and structural change, will be characterized by considerable uncertainty among members. Although relationships in place prior to the change may not disappear, if the change is significant, it is likely to be highly disruptive to the existing network order, resulting in a reassessment of old ties and the formation of many new relationships (c.f. Human and Provan 2000; Powell et al. 2005). As the network system matures over time, knowledge and information about network members and their actions, especially regarding central players, will spread and reputations will become more established. Those organizations that are more embedded in this more stable network will then tend to reap the benefits as a result of their more central position, resulting in higher social outcomes including trustworthiness, reputation, and influence. Conversely, organizations that are able to attain these social outcomes, by whatever means, will become increasingly structurally embedded in the network, as others seek them out for information and resources. We recognize, of course, that this conclusion may not hold for many individual relationships, but we argue that it will hold across the network as a whole.

Trustworthiness

Granovetter (1985:490), in his call for a new economic sociology, argued that the foundations of embedded economic action in social networks rest on "the wide-spread preference for resorting to trusted informants who have dealt with a potential partner and found this partner trustworthy, or even better, for relying on information from one's own past dealings with that person." Consistent with Granovetter's view of trustworthiness, which is based on the direct experience of partners, Gulati (1995) argued that a similar interaction-based trust-building process occurs when strategic alliances are formed by organizations. Later, Gulati and Gargiulo (1999) examined organizational embeddedness and found that organizations rely on information from the network of prior alliances to determine subsequent alliance partners. Negative gossip by third parties about a party's uncooperative behavior significantly reduced the likelihood of direct relations, whereas positive gossip strengthened the likelihood of direct relations. Perrone, Zaheer, and McEvily (2003) also focused on trust in networks arguing that breaches in trust may initiate a reorganization of the social system, where some actors become more central in the social hierarchy and others are pushed toward peripheral status. Based on these findings, organizations that are central, or strongly embedded, in their networks will be able to achieve and maintain their structural position more readily when they are viewed by other network members as having high levels of trustworthiness. Trust and embeddedness are likely to reinforce each other, especially when embeddedness is not based on contractual or mandated ties. Thus, in an established network we expect the following:

[H.sub.1] (static) Provider organizations that are more structurally embedded in a network will have higher levels of trustworthiness.

Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: The Evolution of Structural Embeddedness and Organizational Social Outcomes in a Centrally Governed Health and Human Services Network. Contributors: Provan, Keith G. - Author, Huang, Kun - Author, Milward, H. Brinton - Author. Journal title: Journal of Public Administration Research and Theory. Volume: 19. Issue: 4 Publication date: October 2009. Page number: 873+. © 1999 University of Kansas. COPYRIGHT 2009 Gale Group.

This material is protected by copyright and, with the exception of fair use, may not be further copied, distributed or transmitted in any form or by any means.