Date of Award




Document Type


Degree Name

Doctor of Philosophy (PhD)


Department of Public Administration and Policy

Content Description

1 online resource (xxiv, 420 pages) : illustrations (some color), color map.

Dissertation/Thesis Chair

R. Karl Rethemeyer

Committee Members

Ashley Fox, Deneen Hatmaker


Affordable Care Act, Exogenous shock, Network analysis, Policy change, Public service delivery, Service delivery network, Health services administration, Breast, System analysis, Medical care, Public administration, Interagency coordination

Subject Categories

Business Administration, Management, and Operations | Public Health | Public Policy


EXECUTIVE SUMMARYFor decades public administration and management scholars have conceptualized organizational networks as solutions to failures of markets and hierarchies (see, for example, Goldsmith and Eggers 2005; Osborne and Gaebler 1992). Relationships among organizations or actors within a network are framed positively as channels through which human, financial, and knowledge resources flow to address complex or “wicked” social problems (see, for example, Rittel and Weber, 1973). However, recent scholarship has sought to pull the curtain back and identify the pitfalls of networked arrangements for public service delivery (see, for example, O’Toole & Meier, 2006; O’Toole & Meier, 2004). Such studies hint that interdependencies among network actors may be a source of instability and vulnerability. Carboni and Milward (2012) theorize that service delivery networks are vulnerable to exogenous shocks, such as natural disasters, new public health threats, terrorist attacks, or policy changes (see, for example, Carboni & Milward, 2012; Mays et al., 2009; Milward, 2014). The theory represents an important milestone in the public administration and management literature because it identifies systemic risk in networks as a vulnerability that could influence the production and delivery of public goods and services. In this study, I propose a novel approach to studying how exogenous shocks influence the production of public goods and services by service delivery networks. I apply the approach to a representative case of 36 service delivery networks that experienced a policy shock, the Patient Protection and Affordable Care Acts (ACA). I use the case to develop an evidence base to answer three questions about how exogenous shocks influence service delivery networks' production and delivery of public goods and services. • How do exogenous shocks influence the production and delivery of public goods and services by service delivery networks? • How do service delivery networks adapt to exogenous shocks? • How do network managers and public managers influence service delivery network adaptation to exogenous shocks? To answer these questions, I follow Rethemeyer and Hatmaker’s (2007) approach to theory extension. First, I develop a synthesis of the public administration and management literature about service delivery networks, the production and delivery of public goods and services, and exogenous shocks. I present Carboni and Milward’s (2012) theory and identify four areas for further study. Next, I develop an approach to evaluating service delivery networks by identifying how exogenous shocks influence network participants and choices over time. I apply the approach to the case of 36 discrete networks operating under the auspices of the New York State Department of Health (NYSDOH) to produce and deliver breast and cervical cancer screening services to uninsured New Yorkers. Although hundreds of thousands of women remained uninsured after the ACA’s implementation in NYS, the number of women screened through the networks declined significantly. To understand how the ACA’s implementation influenced the delivery of cancer screening services to the remaining uninsured, I draw from key informant interviews to define the network context and the arrays of participants and choices across the service delivery networks before and after the policy change. I find that the ACA’s implementation in NYS changed the size and characteristics of the eligible service population. Subsequently, the service delivery networks screened fewer eligible women for breast and cervical cancers because the remaining uninsured were unlikely to seek out primary and preventive care from clinical provider organizations within the networks. Thus, clinical provider organizations were less effective mechanisms to link uninsured women to the cancer screening networks following the policy change. Next, I describe an initiative developed by the NYSDOH to adapt the cancer screening networks to the ACA’s implementation. The initiative sought to engage the remaining uninsured in the local NYS NBCCEDP service delivery networks by mobilizing social service organizations to serve as the primary network referral source. Network managers for each service delivery network were required to use specific network management strategies to achieve this goal. Drawing from the initiative, I develop three expectations about how the local NYS NBCCEDP service delivery networks will adapt to the ACA’s implementation over time. First, the number of social service organizations in the networks will increase over time. Second, interorganizational relationships will increase over time, including resource sharing and referral from network managers to social service organizations. Third, the cancer screening network managers will report more referrals receiving interorganizational relationships with social service organizations over time. To evaluate each expectation, I use four waves of longitudinal ego-centric network data representing January 2015—March 2017. Network managers leading each cancer screening network provided data about the organizations and relationships within each cancer screening network. Network managers validated the data and analysis. I use non-parametric statistical analysis to assess each expectation for the cancer screening networks as a class and descriptive statistics to evaluate variations between the networks. The number of social service organizations increased slightly for the local NYS NBCCEDP service delivery networks as a class over two years. However, key measures of adaptation decreased significantly. Network managers reported fewer resource sharing or referral sending interorganizational relationships over time. Ultimately, at the end of the study period, fewer interorganizational relationships were reported whereby social service organizations sent referrals of eligible, uninsured women to the network managers. However, examining differences between local NYS NBCCEDP service delivery networks highlight important variations that signal that some networks adapt more quickly than others. Evidence from the case suggests that policy changes influence how service delivery networks produce and deliver public goods and services. The case of the local NYS NBCCEDP service delivery networks and the ACA’s implementation in NYS also provides evidence that service delivery networks adapt to policy shocks and that trust, reciprocity, and incentives are important drivers of adaptation in service delivery networks. Additionally, evidence from the case highlights barriers that network managers experience using network management strategies of selective activation and framing to mobilize new network segments. Analysis of network membership and interorganizational relationships over 27 months suggests that adaptation efforts are slow, cumbersome, and may produce unexpected results. I conclude by discussing my findings in the broader public administration and management literature. I develop five propositions about exogenous shocks and service delivery networks based on evidence from the case to support future research about exogenous shocks and the production and delivery of public services by service delivery networks. Proposition 1: Policy changes are intrinsically different from other exogenous shocks because public policies define the goals and resources available to service delivery networks. Proposition 2: Policy changes that influence the participants or choices within a service delivery network are policy shocks. Proposition 3: Underfunding is a systemic weakness that amplifies the effects of policy shocks in service delivery networks. Proposition 4: Service delivery network embeddedness in relevant policy networks increases network robustness or resilience to policy shocks by reducing uncertainty about policy changes. Proposition 5: The emergence of informal interorganizational relationships in service delivery networks following an exogenous shock indicates adaptation. Network managers use of selective activation and framing strategies influences adaptation. Study findings highlight the critical role of public managers and network managers in adapting service delivery networks to policy shocks. It is important for public managers to recognize when they work in a network context and understand how network members contribute to the production and delivery of public goods and services before shocks happen. Public managers that develop relationships by building trust, reciprocity, and good-will with network members to adapt networks to shocks are more successful than those that rely on formal agreements. Finally, public managers have vastly outpaced researchers in developing strategies to adapt networks to shocks (see, for example, Isett et. al. 2011). Public managers should seek to learn from other practitioners about strategies used to respond to exogenous shocks.