Achieving the Benefits of E-Government? An …teep.tamu.edu/Npmrc/Brown2.pdf · Achieving the Benefits of E-Government? An Examination of the ... as homeostasis, cybernetic control,

Achieving the Benefits of E-Government?

An Examination of the Family Support Act of 1988

Mary Maureen Brown [email protected]

UNC Chapel Hill

Shannon Schelin NCSU

Jeffrey Brudney

University of Georgia

NPMRC 2003 Georgetown University

Contemporary organization theorists appear to have reached consensus on the notion that policy success is fundamentally based on the ability to establish and leverage joint partnerships to meet

turbulent environmental conditions. Qualities traditionally considered egalitarian – equality, empowerment, horizontal relationships, and consensus building—are particularly important to

resolving the problems organizations encounter in today’s world (Daft, 2000). The deficiencies of established vertical organizational structures for dealing adequately with government

problems have stimulated voluminous research on the potential benefits of joint partnerships as a feasible alternative to traditional hierarchical/bureaucratic structural forms (Lynn, Heinrich, and Hill, 1999; Fountain 2001; Agranoff and McGuire, 1999; O’Toole, 1988). Lefebvre and Letiche (1999) claim that contemporary theorists have begun to reshape the definition of organization.

Organization is becoming a term to mean “existence in change”. Organization is coming to mean something very different –it is no longer the opposite of indeterminance, but rather an

activity of indeterminance. Calling attention to the necessity of employing joint partnerships, Bardach (1998) argues that many contemporary policy challenges lie across policy domains, outside the jurisdiction of any one agency. He goes on to claim that many serious policy problems go unsolved because of missed opportunities for organizations to collaborate. Milward and Provan (1999) substantiate these findings and demonstrate that public policy arenas are inherently crosscutting. According to their research the requisite knowledge, and corresponding problem solutions, are distributed across a range of agencies and organizations. Without cross agency partnerships, critical policy solutions go unrealized. Yet, cross-agency partnerships by definition lead to greater levels of interconnectedness. Consequently, they can serve to open organizations to environmental threat and uncertainty. The dependency that accompanies interconnectedness can alter the ability of the organization to react effectively to changes in the environment. Over the past decade, increasing attention has been given to the potential utility of complexity theory as a framework for understanding public sector behavior. Complexity theory appears to offer some advantages in understanding how public organizations respond to dynamic events characterized by high levels of uncertainty. This research effort examines the implementation efforts of two major policy initiatives: the Family Support Act of 1988 (FSA-88), and the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA). These two pieces of legislation, in concert, were revolutionary in the manner in which government agencies sought automation (information technology) initiatives to resolve a public problem. Faced with the knowledge that the average state only collected 25 percent of the child support payments that were owed, the two legislative enactments mandated that states 1) establish a state-to-local government wide central registry of child support orders, collections, and disbursements (FSA-88) and 2) integrate local, state, and federal child support collection activities to boost support collection rates (PRWORA). To encourage compliance the Federal government made more than xx available for the automation process. In addition, failure to comply with the enactments would result in substantial financial penalties. In the process of automating the collection activities, many of the states encountered substantial difficulties in meeting the deadline requirements. Independent audits revealed that the failure to

Draft: Not For Distribution or Quotation 2

meet the requisite deadline was largely due to the complexity of the effort. The complexity often triggered personnel shortfalls, requirement changes, and coordination difficulties. This research effort focuses on one aspect of complexity: the pursuant structural adaptations the organizations sought in order to gain stability over the uncertainty associated with the complexity of the effort. Specifically, we examine (1) how the organizations adapted to the complexity, and (2) which adaptation strategies proved important to perceived success. Additionally, we also examine (3) the extent to which the adaptation strategies led to important outcomes measures relating to support collections and case load. The paper begins with an overview of Complex Adaptive Systems when defined by resource exchanges among cross-agency partnerships. Afterwards, the context of the two legislative enactments is provided to promote a clear understanding of the setting in which the study occurred. The methods and findings then follow. In light of the fact that public organizations are increasingly turning to automation to leverage the benefits that cross-cutting programs and policies can provide, this study offers substantial theoretical and applied contributions to our understandings of complexity, uncertainty, and adaptability. Here-to-fore, IT failures were often cast in terms of internal constraints (i.e. leadership, user involvement, planning, etc). This study looks at the field of complexity as a potential means of furthering our understanding of the costs and benefits of employing IT to facilitate the desire for cross-cutting policies and programs. Complex Adaptive Systems Historically, organization theorists have consistently identified that organizations are open systems highly susceptible to environmental influence. Few dispute the notion that organizations are incomplete and require exchanges with other systems to accomplish critical goals and objectives. Open system concepts such as homeostasis, cybernetic control, and dynamic equilibrium are often used to describe organizational phenomena pertaining to stability and adaptability (Morel and Ramanujam, 1999). Organizations are routinely viewed as dynamic systems that actively engage in a number of activities to limit flux and seek equilibrium with the environment (Simon, 1996). Of more recent scholarly concern however, is the knowledge that organizational processes are becoming increasingly complex in their interactions with external agents thereby leaving them highly susceptible to environmental flux. Given the need to employ joint efforts to address mandated requirements, it is becoming more difficult to shield organizational operations from environmental disturbances. Hence, the study of the movement toward greater degrees of complexity, and the consequential vulnerabilities that ensue from environmental perturbations, is the subject of considerable attention. According to Simon (1996) the field of complex adaptive systems provides a useful conceptual framework for studying various organizational phenomena. Not surprisingly then, the study of organizational complexity, and the manners in which organizations adapt to environmental flux, has generated considerable attention and debate over the past two decades. Despite the scholarly activity, there is little consensus over how – or why -- complexity affects organizations. Much of the debate is believed to derive from the methodological problems associated with studying organizational complexity (see xx for a thorough review of the methodological problems associated with complexity).


As the research stands, complex adaptive systems (CAS) is best defined as a conceptual framework capable of promoting an understanding of behavior rather than a single inclusive theory. CAS, as a conceptual framework, is largely concerned with how complex systems relate to dynamic change, adaptation, and evolution. Given the immaturity of the research base, there is no clearly accepted or articulated definition of the concept of CAS, or its central configuration -- complexity. There is little agreement on the definition of the term complexity, and for different research studies it often takes on different meanings. Complexity has been used to reflect such issues as organizational size, structure, and task environment. There does however appear some parsimony on three important themes: 1) complexity signifies a large number of interacting elements, 2) it is associated with emergent nonlinear adaptive properties, and 3) it is vulnerable to high levels of uncertainty and unpredictability. Of central concern is how complex organizations adapt to the uncertain and unpredictable environments they encounter. Simon (1996) defines a complex system as one made up of many parts that have many interactions. And Thompson (1967) describes a complex organization as a set of interdependent parts, which together make up a whole that is interdependent with some larger environment. The phenomenon of “self-organization” was first recognized as an important aspect of change in the physical and biological sciences (Ruelle, 1989; Prigogine and Stengers, 1982; Prigogine and Nikolis 1989; Bak and Chen 1991; Kauffman 1993). These researchers sought to explain unexpected aberrations in the operation of mechanical systems. Specifically, they were interested in how minor fluctuations in performance would cumulate at certain points, eventually leading to large disruptions in the operations of the total system. These minor fluctuations, or points of energy attraction, within the operation of the system were termed “strange attractors” (Ruelle, 1989). Their presence demonstrated the occurrence of an unplanned clustering of energy at specific points in the system that was outside the prescribed plan of operations and occurred without external design. The “strange attractors” shifted the pattern of energy flow within the system, eventually altering the operation of the entire system. CAS researchers often focus on the concept of self-organization, or the tendency for how coherent and purposive wholes emerge from the interactions of simple and sometimes non-purposive components or activities. The conceptual framework includes such ideas as phase changes, fitness landscapes, self-organization, emergence, attractors, symmetry and symmetry breaking, chaos, quanta, the edge of chaos, self-organized criticality, generative relationships, dissipative structures, dissipative energy, bifurcation points, punctuated equilibrium, and increasing returns to scale (see Table 1 for a definition of these fundamental terms). As noted above, organizations are dependent on environmental flows or exchanges involving critical resources such as information, personnel, equipment, or capitol. CAS, when defined by interdependent relationship structures, is often examined in terms of their ability to adapt to changes in the environment. As discussed in detail below, the adaptation can take a variety of forms from static on one extreme to chaotic on the other. A static state reflects the inability of the relationship to adapt requisite policies, procedures, or activities according to changes in the environment. Conversely, the chaotic state represents a hyper-turbulent response to environmental flux. Utilizing these definitions, our definition of complexity is similar to Simon’s and Thompson’s in that we seek to examine how interdependent cross-agency structures evolve


and adapt to environmental flux. Hence the study relies heavily of concepts derived from resource dependency/exchange theory. An understanding of the adaptation patterns of these complex relationships has tremendous managerial implications. Goals and objectives, as well as capitol and opportunity costs are inherently tied to potential activities of adaptation. These secondary adaptive behaviors can cascade in unexpected ways, and thus, can have a tremendous impact on the achievement of critical goals and the final costs associated with any organizational activity. Before exploring the managerial implications of self-organization and complex adaptive systems, further elaboration of the concepts is required. Complex systems are distinguished by a capacity for “self-organization”. They are capable of rearranging and reforming patterns of operation in mutual adaptation to the changing needs and capacities of their components (Comfort, 1994). They are also capable of mutually adapting to the changing demands and opportunities imposed by the environment. The distinguishing characteristic of this process is that it occurs as a result of communication, selection, and adaptation processes within the system itself and between the evolving system and its environment (Comfort, 1994; Kaufmann, 1993). Prigogine identifies three prerequisites for the emergence of self-organization. The first condition is that a system must be open and allow exchanges of energy, matter, and information with the outside world. The second condition is that it be dynamic, far removed from a state of equilibrium. The third condition is that within the system there is feedback that enables special processes to make rapid shifts in behavior. Prigogine points out that it is important that information sharing develops within the system and that the system as a whole produces coherence so that a coherent environment can be generated. Kaufmann (1993) identifies 4 features of a self-organizing system. Self-organization is a continuous process that occurs in social contexts through “communicative acts”. These acts are most often forms of verbal, written, or electronic communication transmitted directly between two or more agents within the system or between the system and its environment. Second, self-organization, coupled with selective tendencies, creates the system’s capacity for adaptation to environmental conditions. Third, self-organization recognizes the influence or control that some units exert over other units in an interdependent system. Finally, self-organizing systems are massively parallel processing systems, where different components perform different functions simultaneously to achieve the goal. Kaufmann’s research draws important, albeit controversial implications. He argues that systems designed for control tend to become paralyzed and can self-destruct in rapidly changing environments. Under situations characterized by environmental flux, lacking the flexibility for spontaneous adaptation, their precise rules for operation lack reliability. Comfort (1994) argues that under extreme conditions, the desire to respond to flux with control may actually stifle any innovative efforts to find more effective means of functioning to altered conditions. Conversely, according to Prigogine and Stengers (1984) and Kauffman (1993) systems without sufficient structure to hold and exchange information can also disintegrate under swiftly changing conditions. According to their research, small changes in operating conditions may lead to large disruptions in performance, or avalanches of disorder.


The notion that the two extreme responses to complexity (control vs flexibility) can be counterproductive is not surprising. According to Kaufmann, creative change is most likely to occur within the narrow region on the edge of chaos. Sufficient control over the structure is required to allow participants to hold and exchange information. But sufficient flexibility to allow mutual adaptation among the participants to substantiate changes in their operating environments is also required. One of the most fundamental implications emerging from the science of complexity is that order naturally emerges in systems, no matter how simple, complex, nonlinear, or chaotic the system is (Lewin, 1999). As Anderson (1999) summarizes, when a system is open to receive energy from the outside, it will tend to create order. When a system becomes closed, it will decay into maximum disorder and chaos. When system boundaries exhibit permeability, natural order evolves through self-organization. Resource dependency theorists often focus on the complexity, or density, of various interdependent relationships as a means for studying organizational phenomena. When considered in light of patterns of interactions grounded in forms of resource exchange, how (and why) the relationship interacts with the environment often takes center stage. Prigogine and Stengers (1984) use chaos to describe how order can arise from complexity through the process of “self-organization”. The resulting self-organizing patterns are often dependent on the schema (mindset) of the involved agents. Additionally, agents do exist outside the boundaries of the CAS and their schemas also serve to determine the rules of interaction concerning how information and resource flows occur. This notion of individual schema as a locus of adaptation has tremendous implications. In human systems, self discrepancy theory (Wiggins 1987) predicts that if an individual’s current state does not match their ideal state, the discrepancy represents one of two possible states: dejection or agitation. In holding with Kaufmann’s and Prigogine’s finding, either of these states may serve as minor fluctuations capable of resulting in major transformation (see below for further elaboration). Seeking to use available resources efficiently and to integrate new resources into existing structures for effective action, organizations respond to environmental flux by changing operating procedures and practices in fundamental ways (Comfort, 1994). Changes occur in part through voluntary selection among alternatives for action and in part through mutual adjustment in performance among participating organizations. Order returns to the system through a creative process of reciprocal exchange, learning, adaptation, and choice among multiple participants operating at multiple levels of responsibility, experience, and knowledge. While adaptive alterations in behavior can result from voluntary/deterministic selection, many times the adaptive behaviors are emergent in nature and nondeterministic. This is commonly seen when small isolated perturbations serve to shift the adaptive mechanisms toward an unpredictable trajectory. For example, a seemingly innocuous unpredicted citizen complaint may resonate in such a way that it takes on a “life of its own” and shifts the organization’s adaptive behaviors in a significant way.


Ongoing work by resource exchange theorists adds substantially to our understanding of some of the strategies managers employ to gain stability over flux. Exchange theorists provide two lines of thought on how complex organizations seek to control environmental perturbations. One strategy is to “buffer” activities from disturbances and the second is to build stability by “bridging” to other units. According to Thompson (1967) the overarching inherent goal of an organization is to reduce to a minimum the extraneous forces that can upset the means it undertakes to pursue various ends. Thompson argues that all organizations are highly motivated to secure enough stability, determinateness, and certainty to be able to function efficiently and effectively in environments that contain unknowns and uncertainties. Coding, stockpiling, leveling, forecasting, and adjusting scale are but some of the explicit strategies organizations employ to control environmental perturbations. Alternatively, bridging is used to modify boundaries to bridge organizations and their exchange partners, competitors, or regulators. According Pfeffer and Salancik (1978), the typical solution to problems of interdependence and uncertainty involves increasing coordination, which translates into increasing mutual control over each other’s activities. All bridging activities tend to share this feature but they differ in the strength and nature of the connections forged. Bridging strategies include bargaining, contracting, co-optation, hierarchical contracts, and joint ventures. Coordination mechanisms that relate to stabilizing the boundaries and buffering the relationship against environmental flux include the institution of rules and programs, schedules, departmentalization, and hierarchy. Yet, as Emerson (1962) points out, as organizations take on more complex and unpredictable tasks it may be impossible to buffer out all traces of uncertainty. Emerson (1962) emphasizes that one actor’s power over another is rooted in the latter’s dependency on resources controlled by the former. The dependence of one actor on another varies as a function of two factors: B’s dependence on A is: (1) directly proportional to the importance B places on the goals mediated by A, and (2) inversely proportional to the availability of these goals to B outside the A-B relationship. Two types of changes can emerge from the buffering and bridging activities: minor changes that affect the routines without affecting the relationship structure, and major changes which affect the structure of the relationship. Holland (1995) notes that complex patterns can arise from the interaction of agents that follow even simple rules. These patterns tend to be emergent in the sense that new properties occur throughout the various levels of the organization. Kauffman (1993) notes that complex systems tend to exhibit “self-organizing” behavior: starting in a random state, they usually evolve toward order instead of disorder. Most approach CAS from the perspective that it is a dynamical system comprised of agents at a lower level of aggregation. Each agent’s behavior is dictated by a schema (a cognitive structure or mindset) that determines what action is taken at any point in time based on the agent’s perception of the environment. According to Levinthal (1997) each individual’s payoff function depends on choices that other agents make, so each agent’s adaptive landscape – mapping its behavior to its realized outcomes – is constantly shifting. Moreover complex adaptive systems evolve over time through the entry, exit, and transformation of agents. Furthermore the linkages between agents may evolve overtime shifting the patterns and strengths of interconnections.


Closely tied to the concept of bounded rationality (March and Simon, 1958), agents are presumed unable to forecast the system level consequences of their individual choices, so they optimize according to their own fitness, not that of the organization. Kauffman’s adaptive landscape metaphor (borrowed from Wright 1931) agents co-evolve on a fitness landscape to a state poised between order and chaos. The landscape on which agents adapt continually shifts, because the payoffs of individual agents depend on the choices that other agents make (Levinthal 1997, McPherson and Ranger-Moore 1991). In keeping with this line of thought Zajaak and Kraatz (1993) contend that organizations purposively adapt to environmental changes through feedback loops that result in restructuring activities. However individuals are free to disrupt institutions and as indicated by Senge (1990 and Hall (1976) behavior patterns can emerge contrary to intention producing unexpected and counterintuitive results. According to Stacey (1995) this is because the choices of agents in human systems are based on perceptions which lead to non-proportional over and under reaction; there are many outcomes possible for any given action; group behavior is more than simply the sum of individual behaviors; and small changes can escalate into major outcomes. Thus choice is not limited to simple heuristics but rather is truly open-ended with unknowable outcomes. More recently, many organization theorists are drawing managerial implications from the strides the sciences have traditionally provided to the field of complex adaptive systems (Anderson, 1999; Stacey, 1999). Seeking to use available resources efficiently and to integrate new resources into existing structures for effective action, organizations respond to environmental flux by changing operating procedures and practices in fundamental ways (Comfort, 1994). Changes occur in part through voluntary selection among alternatives for action and in part through mutual adjustment in performance among participating organizations. Order returns to the system through a creative process of reciprocal exchange, learning, adaptation, and choice among multiple participants operating at multiple levels of responsibility, experience, and knowledge. Self-organization represents a fundamental reallocation of energy and action within a system in order to achieve a larger goal. As Comfort (1994) succinctly states “understanding when, how and where change may occur in dynamic environments is a primary and fundamental challenge for managers.” Managers attempt to dominate, control, arrange, program, and organize the disarray until it is tamed. Rules and order are constructed and maintained. Laws, rules, and regulations are erected to prevent chaos and difference from invading. Indeterminancy and uncertainty are thought of as undesirable. Lewin (1999) claims that trends toward increasing complexity require managing all the organizational levers of dissipative energy. These include allowing promoting emergent processes as self-generated sources of dissipative energy such as facilitating improvisation, product champions, and emergent strategies. Leadership styles that moderate dysfunctional tension and forestall the emergence of chaos will also be critical to adapting to the environmental demands that result from complexity. The research study discussed below considers how organizations adapt to gain stability over the uncertainty associated with complexity. Specifically, the study examines the structural adaptation efforts employed when confronted with legislative enactments mandating radical changes in the structure and functions of child support collection efforts. Before turning to the


research, a discussion of the Family Support Act of 1988 and the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 is required to convey important contextual information about the study. The Automation Requirements of the Family Support Act of 1988 and the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 Background on Child Support Enforcement

The first legislative policy to address child support enforcement was enacted in 1975. The Child Support Enforcement and Paternity Establishment Program (CSE) was instituted by Congress to rectify the problem of delinquent child support (Public Law 93-647). Precipitating this act was Congressional recognition that the composition of the Aid to Families with Dependent Children (AFDC) caseload had changed drastically. In earlier years the majority of children needed financial assistance because their fathers had died; by the 1970s, the majority needed aid because their parents were separated, divorced, or never married. Adding a new part D to title IV of the Social Security Act, this statute, as amended, authorized the use of Federal matching funds for enforcing support obligations by: 1) locating nonresident parents, 2) establishing paternity, 3) establishing child support awards, and 4) collecting child support payments. Since 1981, child support agencies have also been permitted to collect spousal support on behalf of custodial parents, and in 1984 they were required to petition for medical support as part of most child support orders.

Since that time four significant amendments (P.L. 98-378, 100-485, 104-35, and 104-193) have been enacted to the CSE program providing States with additional remedies to collect child support (see table 2). Two of these policies speak directly to automation and have come under the purview of the Office of Child Support Enforcement (OCSE). The desire to automate derived from the difficulties States faced in their attempts to collect delinquent child support payments. Automation was seen as a means for streamlining child support enforcement activities, thereby boosting collection rates.

The first public law that specifically mandated the use of IT resources for support collection was passed in 1988. The Family Support Act of 1988 (FSA88) (P.L. 100-485) established the Federal Office of Child Support Enforcement (OCSE) whose primary mission was to collect delinquent child support. Additionally, FSA88 required States to develop an automated tracking system to manage caseloads and streamline the collection process. To facilitate the automation process, DHHS offered a 90 percent funding match for all automation efforts. In addition, the policy required States to demonstrate that their systems met Federal certification requirements by October 1, 1995 or risk the loss of critical federal funding.

By 1995, 11.6 million custodial mothers of children under the age of 21 were owed child support. Of those, only 7.1 million or 61 percent had a court ordered child support award. About one-third of the 4.5 million mothers without court awards chose not to pursue one for reasons such as they were unable to locate the noncustodial parent or the noncustodial parent was unable to pay. Never-married custodial parents were the group least likely to have a court ordered child support award. Only 44 percent of never-married custodial mothers had support awards


compared with 76 percent of divorced custodial mothers. Moreover, African American custodial mothers and custodial mothers of Hispanic origin were much less likely than their Caucasian counterparts to have child support awards. About 72 percent of Caucasian had child support awards, compared with 45 percent of African American and 47 percent of Hispanics (U.S. Census Bureau, 1999). Both the overwhelming lack of collections, along with the socioeconomic disparities in child support awards and collections, made FSA-88, particularly its automation component, a necessity. Despite the admirable intentions, Montana was the only State that was able to meet the 1995 federally instituted deadline for automation. There were several explanations for the lack of state compliance with the 1995 deadline. One commonly cited problem was the deadline. Many of the states complained that the timeframe did not allow for appropriate due diligence, requirements definition, or stakeholder negotiations to occur. Supposedly, the short timeframe significantly limited the ability of states to develop the plans with sufficient precision to meet the complex automation requirements. Contractor relationship problems were also cited as another hindrance to compliance with the deadline. Over fifty percent of the states noted that the contractors were unable to provide acceptable products within the specified timeframe. Quality problems were a major contributor to poor state-contractor relationships. Furthermore, almost half of the agencies reported staffing shortages that exacerbated implementation issues. To help meet these problems, the OCSE instituted a “transfer” policy. States that transferred in a system that was already operational in another state would receive additional funding. And the deadline was extended, without penalty, until Oct 1, 1997. By the next year, 1996, significant legislative activity was underway to further address the problem of support delinquency. In 1996, Public Law 104-193, the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA), abolished AFDC and related programs and replaced them with a block grant program entitled Temporary Aid to Needy Families (TANF). Under the new law, each State must operate a CSE Program meeting Federal requirements in order to be eligible for TANF funds. In addition to abolishing AFDC, Public Law 104-193 made about 50 changes to the CSE Program, many of them major. These changes include requiring States to increase the identification rate of delinquent parents. Armed with the knowledge of a University of Michigan study of separated parents the Clinton administration sought to extend efforts at the interstate level. In Hill’s study, 12 percent of noncustodial parents lived in different States within one year of divorce. That proportion increased to 25 percent after three years, and to 40 percent after eight years (1988). Estimates based on the Federal income tax refund offset and other sources suggested that approximately 30 percent of all child support cases involved interstate residency of the custodial and noncustodial parents (Weaver & Williams, 1989, p. 510). According to U.S. Census Bureau (1991) data, 20 percent of noncustodial parents lived in a different State than their children, 3 percent lived overseas, and the residence of 11 percent of the noncustodial parents was unknown. Cross-agency inter-jurisdictional integration was seen as imperative to stemming the support delinquency rate.


PRWORA altered substantially the automation requirements for the States. In keeping with the goals of GPRA, the legislation changed the focus to measure performance and program results instead of process. The new automation requirements required State systems to perform functions including: controlling and accounting of Federal, State, and local funds to carry out the child support enforcement program, maintaining data necessary to meet Federal reporting requirements, maintaining data on State performance for calculation of performance indicators, safeguarding of the integrity and security of data in the automated system, developing a state case registry, performing data matches, and providing expedited administrative procedures. Specifically the new automation requirements required the establishment of four overarching integrated systems designed to allow computer-matching capabilities to locate delinquent parents. Figure 1 demonstrates the data flow of the four required systems:

1) The National Directory of New Hires - employers are required to report information about newly hired employees to a state agency (State Directory of New Hires) designated to receive the information. These designated agencies then forward information on to the Federal New Hire Registry. Approximately 30% of child support cases involve noncustodial parents who reside/and or work in a different State than their children. 2) State-Local level Case and Order Registries. 3) The Federal Case and Order Registry. 4) The Federal Parent Locator Service – a computerized national network to provide SSN, addresses and employer and wage information to State and local CSE agencies to establish and enforce child support orders. FPLS pulls data from the IRS, SSA, Department of Defense, Department of Veteran Affairs, and State Employment Security Agencies.

In short, PRWORA required States to extend their automation efforts to include the development of an integrated, automated network. Purportedly, the new network would integrate all information about the location and assets of parents. Thereby requiring States to implement more enforcement techniques, and revising the rules governing the distribution of past due (arrearage) child support payments to former recipients of public assistance. The deadline for PRWORA implementation was set for October 1, 2000. Similar to FSA-88, the Federal government offered matching funds at a rate of 80 percent. By this point the States had one year left to meet the FSA88 certification requirement or lose TANF funding. Furthermore they had the new PRWORA automation requirements that must be met by October of 2000. Unfortunately, the integration requirements meant that many of the States had to retrofit much of the work that had been done to date to meet the FSA-88 requirements to address the looming PRWORA deadline. Despite the two-year FSA-88 extension, by October 1, 1997, only 17 of the 54 State jurisdictions met the FSA certification deadline. Lamentably, these 17 States only accounted for 20 percent of the national caseload.


Among the problems that continued unabated pertaining to staffing, contracting, and planning, many states cited the ACF transfer policy, in which states where offered enhanced funding to use existing automated data systems, as problematic (US DHHS, 1997). The states felt that the transfer policy significantly delayed progress, in lieu of enhancing development and implementation. Purportedly, these delays occurred because of the lack of complete systems available for transfer, as well as the lack of appropriate fit for different states. Finally, 30 percent of the states indicated that the inconsistent interpretations of interstate laws regarding child support created additional complexity and hence delayed their efforts. Ten years after passage, by 1998, 37 states were finally FSA certified. In 1999 five were certified and in 2000 five more became FSA certified. By the 2001 deadline for PRWORA, three more states were FSA-88 certified. But only nine were PRWORA certified. Similar to the Family Support Act of 1988, many states encountered significant delays and failures in their quest for PRWORA certification. In particular, several states were subjected to independent audits, or Independent Verification and Validation Assessments (IV&Vs), in order to ascertain areas of weakness or potential points of failure. For example, North Dakota, was found deficient in many of their project management capabilities. Nevada was cited for the lack of 1) risk management procedures, 2) formal processes for documentation, and 3) appropriate quality assurance methodologies. Other States experienced deficiencies in areas related to accounting, organizing, and planning. Cost estimation methodologies were lacking as was metrics and measures for assessing system capacity needs. By 2003 four states had not yet received FSA-88 certification and 31 States had not been PRWORA certified. In essence, both legislative mandates created, altered, and disbanded many of the existing, autonomous structures of the child support enforcement system and sought to unify efforts under umbrella organizations in order to increase support order establishments, paternity establishments, and collection rates. Due to the need for synchronization (both vertical and horizontal integration) across multiple layers of government, the difficulties apparently compounded. In short, the FSA-88 and PRWORA acted as external chaotic stimuli that could not be buffered. The States were forced to adapt by instituting whole-scale policy, operational, and technological change. Changes that not only called for reorganization, but reengineering across layers of government as well. In light of the resulting complexity, several questions emerge related to how the States met the legislative challenges. These questions draw extensively on complexity theory, in particular organizational aspects. Accordingly, the research questions that form the basis of this inquiry focus on 1) how organizations configure themselves in order to adapt to the externally induced complexity – in essence, what fitness landscape emerges to address the complexity of the policy requirements? 2) xxxx the organizational success factors associated with PRWORA and FSA automation implementation; 3) organizational configurations that contribute to team agitation within IT project design and implementation; and, 4) the actual outcomes realized as a result of automation implementation.


Methods To understand the adaptive strategies of the CSE information technology efforts we employed several data collection strategies. First we conducted a telephone survey of the managers of the IT initiatives for all 54 state level efforts (the sample included American Samoa, Territories of Pacific Islands, Puerto Rico, and Virgin Islands). The survey took place between April and June of 2003 and included three waves of telephone follow-up calls and 2 waves of email follow-ups. Despite the persistence, the survey only generated 29 responses for a total return rate of 54 percent. Ten of the respondents verbally indicated that the political issues surrounding their projects prevented them from responding to the survey. Hence the telephone survey sample is based on 19 respondents. The survey was employed to capture perceptions of flux, adaptation, agitation, and success. We also relied heavily on secondary data published in annual reports to Congress pertaining to variables such as expenditures, schedule delays, certification accomplishments, and case-load. Finally several administrative reports were also analyzed. Many of the states were ordered to undergo independent audits. Also, several General Accounting Reports have been published on the progress of the CSE efforts to date. Many of these data sources can be found on the OCSE web page (www.acf.hhs.gov/programs/cse). The following states are included in the study:

Arkansas Colorado Georgia Indiana Iowa Kentucky Michigan Missouri Montana Nevada North Dakota Pennsylvania South Dakota Texas Utah Virginia Washington West Virginia Wisconsin

As demonstrated, five states fall within the Southern region, nine within the Mid-western region, three in the Western region and one from the South-western region, and one from the Middle-eastern region. The states surveyed account for approximately 42 percent of the total national case-load. The average spending on the information systems to date is roughly $48,000,000. Finally, the average number of system users for the survey respondents is 8,675. Findings One of the primary research questions was to ascertain the type of organizational configurations (loose versus tight) that were employed to implement the FSA and PRWORA policy requirements. Respondents were asked a series of questions to shed light on the configurations of their organization strategies. Specifically they were asked to define the configuration of the organization according to nine characteristics: clarity of boundaries, form of work activities, type of leadership style, direction of team loyalty, clarity of rules, role prescription, formality of structure, and the direction of information flows. The survey questions were intentionally designed as dichotomous variables (where one pole signified a tight configuration and the other a loose configuration) to force the respondent to choose one configuration over another. In total the items provide a fairly comprehensive view of the configuration dynamics of the


organizational arrangement that was established to implement the IT requirements of the FSA and PRWORA. Table 2 presents the respondents’ perceptions of the tight/loose organizational configurations of the IT effort according to the nine “fitness” characteristics. As illustrated in Table 3, the majority of respondents indicated a “tight configuration” on five of the nine fitness characteristics. Boundaries of the IT project were clear, work activities routine and formal, leadership directive, team members elicited loyalty to the organization and rules and procedures were explicit. Eighty-four percent indicated that the boundaries of the IT projects were clear rather than fuzzy in nature. Seventy-eight percent also claimed that the work activities were routine and formal as opposed to ad-hoc and informal. Additionally, 74 percent stated that the leadership style was directive in nature as opposed to nondirective. A majority of respondents (68 percent) also indicated that IT team members were loyal to the organization rather than to the profession. With regard to IT rules and procedures, 63 percent of the respondents noted they were explicit instead of implicit. The four remaining characteristics, information flows, role prescription organizational structure and division of work all demonstrated a loose configuration. In terms of IT information flows, only 47 percent of respondents stated that the information flows were vertical in nature. Fifty-three percent claimed that the information flowed horizontally. In terms of defining staff roles and responsibilities, 63 percent claimed that they were team prescribed. Only 37 percent of the respondents indicated that the roles and responsibilities were prescribed by a central authority figure. When rating the formality of the organization structure, only 32 percent of respondents indicated a formal approach to structure versus 68 percent noting an informal structure. Finally, a minority (28 percent) of respondents perceived the division of work to be highly differentiated, compared to 72 percent who claimed that the work roles overlapped. Figure 2 shows the organizational configuration for each respondent. As illustrated, roughly 50 percent of the sample employed a fairly mixed model of tight and loose configurations. The vast majority of the respondents indicated a fitness landscape that attempted to temper the tight configurations with some adaptability mechanisms. Hence the fitness landscape was one that sought to balance the competing forces and perhaps maximize the benefits that each configuration could provide. Moving on to the second set of research questions, Table 4 reveals an examination of the respondents’ perceptions pertaining to 1) configuration requisites for success, 2) factors that created high levels of team agitation, and 3) the implementation results. Section A of Table 4 reveals the respondents’ perceptions of which of the organizational configurations were important to achieving a successful initiative. The findings are reported in the same order as reported in Table 3. Each respondent was asked to rate (on a seven-point Likert scale) the extent to which each of the configurations was critical to success. Roughly 80 percent of the respondents indicated that negotiable boundaries and routine and formal work activities were highly important to success. Sixty-three percent indicated that directive leadership was highly important and 88 percent claimed that organizational loyalty was highly important. Furthermore, 68 percent viewed team prescribed work roles and explicit rules as critical success factors. The essentialness of horizontal flows was indicated by 82 percent of the population, and 68 percent


indicated that team prescribed work roles was important. Only 47 percent rated hierarchical structures as important and only 42 percent rated differentiated divisions of work as important. Table 3 moves beyond IT success factors and asks respondents about their perceptions of how much agitation the team experienced as it went through the implementation process (section B). Section B of Table 4 focuses on the degree to which the team experienced high levels of agitation. Heavy workloads and conflicting demands were implicated as the greatest contributors to agitation (26 percent and 21 percent, respectively). Sixteen percent of respondents rated too little authority and ambiguous responsibilities as high agitation factors for the team. Only 11 percent of the sample stated that high levels of agitation occurred from the quality sacrifices that resulted from the heavy workload. Finally, only five percent of the respondents claimed that “too little information” and “quality suffers due to compromise” created high levels of team agitation. The final section of Table 4 seeks to assess the outcomes of the IT implementation effort. The overarching finding was that the IT implementation resulted in substantial organizational changes. Seventy-two percent of the respondents indicated that IT implementation created high degrees of change in “interaction patterns with clients.” Another 72 percent noted substantial changes in coordination and communication patterns. Sixty-one percent felt that implementation led to the addition or elimination of organizational units. And finally, 56 percent of respondents noted that IT implementation created high levels of change in management responsibilities. ANOVA The final step of the analysis was to examine the relationship between the organizational configurations (Table 3) and the success requisites, agitation levels, and outcomes demonstrated in Table 4. Analysis of Variance (ANOVA) tests were used to determines whether the means of each grouping of the configuration varied statistically according to the dependent variables of interest: namely success, agitation, and outcomes. Table 5 reveals only those relationships that demonstrated statistical significance. (Given the difficulties of achieving statistically significant parameters in small samples, the threshold for statistical significance was relaxed to .10). Five of the nine configurations demonstrated differences on the dependent variables – organizational structure, work activities, information flows, team member loyalty, and project boundaries. There was no noted difference between the means of the remaining four configurations (leadership, rules and procedures, roles, and division of work). The first ANOVA Model reveals the relationship between the organizational structural configurations (formal versus informal) and success and agitation. The configuration of the organizational structure demonstrated no bearing over implementation outcomes. The lack of any statistically significant relationship between structural configuration and outcomes may be due to the small sample size and the lack of variance rather than the fact that no relationship exists. Given that previous studies have demonstrated relationships between structure and outcomes (see Brown and Brudney, xx), additional research is indicated before any inferences should be made pertaining to structure and outcomes.


When considering structural configuration on success, four relationships demonstrated statistical significance: negotiable boundaries, hierarchical structures, differentiated divisions of work, and team prescribed roles. The formal group rated the importance of hierarchical structures and differentiated divisions of work as twice as important as the informal group. Interestingly, the formal group also rated negotiable boundaries and team prescribed work roles more important than the informal group. At the same time, the formal group tended to witness greater levels of agitation over heavy workloads and too little information (sig =.02 and .07 respectively). The next set of ANOVA models reveals the relationships between routine and ad-hoc activity configurations on success and agitation. Again, the implementation outcomes did not vary between the two groups. Respondents with routine work activities rated organization loyalty, negotiable boundaries, and team-prescribed work roles as more important than their ad hoc counterparts (sig = .00, .01, and .04, respectively). In terms of perceived agitation levels, the routine activity group indicated higher levels of agitation with respect to too little information than did the ad-hoc activity group (sig = .07). The third set of ANOVA models uncovers the relationships between vertical and horizontal information flows on success, agitation, and implementation outcomes. Respondents in the vertical information flow group rated differentiated division of work as more important than their horizontal group counterparts (sig = .04). Conversely, the vertical group indicated that horizontal information flows and leadership direction were less important than did the horizontal information flow group (sig = .00, and .09, respectively). In terms of agitation factors, the vertical information flow group perceived higher levels of agitation pertaining to too little authority and judgment sacrifices compared to the horizontal group (sig = .05 and .09, respectively). Finally, the form of IT information flows does have an impact on implementation outcomes. Respondents in the vertical information flow group rated important communication change outcomes lower than their horizontal group counterparts (sig = .02). Table 4 also considers the relationship between team member loyalty--organization versus profession--and IT project success factors in the fourth set of ANOVA models. Respondents in the organization loyal group rated organizational loyalty, pre-described work activities, team prescribed work roles, and explicit rules as more important success factors than their profession loyal equivalents (sig = .08, .09, .04, and .09, respectively). There was no group variation on agitation factors or implementation outcomes between the loyalty groups. Finally, Table 5 examines the relationship between IT boundaries and success factors and agitation levels. The type of perceived IT boundaries, clear versus fuzzy, demonstrated no bearing over implementation outcomes. Respondents in the clear IT boundaries group rated organizational loyalty and team prescribed work roles as more important to IT project success than did their fuzzy boundaries counterparts (sig = .08 and .06, respectively). On the other hand, the clear boundaries group indicated that too little authority was less agitating to team members than did the fuzzy IT boundaries group (sig = .07). What implications can be harvested from the data analysis described above. Several point of interest surface. First, the federal, state, and local level child support enforcement agencies were


forced to respond to legislative enactments that mandated substantial crosscutting integration. The enactments required major modifications to how operations had traditionally occurred. Hence, policy, operational, and procedural changes had to occur in synchrony with sister states and the federal agencies. The automation effort was designed to enforce the new business rules that would harmonize all the new activities. How the States responded to the complexity of the changes is thus a matter of academic and practitioner interest. The first research question sought to determine how they adapted, the second asked their recommendations on adaptation. The third sought to see how the various configurations affected recommendations, agitation, and perceived outcomes. In short table 3 and figure 2 illustrate that States adapted to the new landscape by employing a mixed model of tight and loose configurations. This finding supports Kauffman’s (1993) theory that complex systems tend to exhibit “self-organizing” behavior: starting in a random state and evolving toward order. The mixture of highly complex and more simplistic organizational characteristics supports the notion that order arises from complexity as a coping mechanism (Prigogine and Stengers, 1984). At the same time, the ANOVAs tended to illustrate that those in “tight” configurations tended to recommend “looser” configurations. And the reverse also holds true. For the most part, the adaptation landscape did not prove influential over outcomes achieved. The small sample size and lack of variance was likely the reason for the failure to achieve statistical significance. This study examined the manner in which State agencies adapted to complex evolving legislative requirements. The findings tend to verify the notions pertaining to “self organization” and that they tend to gravitate toward mixed models of tight and loose coupling. These tight and loose configurations tend to illustrate some effect over implementation process variables, but impacts on outcomes were not noted. By and large the results tend to substantiate the need for continued research on the ramifications of complex adaptive systems.


Table 1 = terms


Table 2: Timeline of Major Policy Initiatives Surrounding

Automation and Child Support Enforcement 1998 Public Law 105-200, the Child Support Performance and Incentive Act of 1998, provides penalties for failure to meet data processing requirements, reforms incentive payments, and provides penalties for violating interjurisdictional adoption requirements. Incentive payments are based on paternity establishment, order establishment, current support collected, cases paying past due support, and cost effectiveness and on a percentage of collections. Incentive payments must be reinvested in the state’s child support program. Public Law 105-187, the Deadbeat Parents Punishment Act of 1998, establishes felony violations for the willful failure to pay legal child support obligations in interstate cases. 1996 Title III of the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) of 1996 (Public Law 104-193) abolished Aid to Families with Dependent Children (AFDC) and established Temporary Assistance for Needy Families (TANF). Each state must operate a Title IV-D child support program to be eligible for TANF funds. States had to comply with numerous changes in child support services. 1995 Public Law 104-35 extends the deadline two years for states to have an automated data processing and information retrieval system. The 90 percent match was not extended. 1988 Public Law 100-485, the Family Support Act of 1988, emphasized the duties of parents to work and support their children, underscoring the importance of child support as the first line of defense against welfare dependence. States were required to: 1) develop mandatory support guidelines; 2) meet paternity standards; 3) respond to requests for services within specified time periods; 5) develop an automated tracking system; 6) provide immediate wage withholding; 8) have parents furnish Social Security number when a birth certificate is issued; and 9) notify AFDC recipients of monthly collections. Provides a 90 percent match to fund State automation needs. Source: http://www.dshs.wa.gov/esa/pdf/2001appendix7.pdf


http://www.dshs.wa.gov/esa/pdf/2001appendix7.pdf

s

State New

Hires

State Child Support Case

SSA, IRS, DOD, VA, State

Employment Agencies

Matches

Federal Parent

Locator System

National Directory of New Hires

Federal Case

Registry

In order to facilitate the matchingcase is required. Given the use oPersonal Responsibility and Welfaccess to new sources for obtaini(professional, driver’s, occupatiopaternity determinations/acknow

The 1996 PRWORA law also reqorders containing records of eachsupport order established or modadministration of many CSE agenregistry, which includes a record penalty charges; amounts collectestandardized information on bothnumber).

In addition to the automated chilestablished an automated directorinformation about employees, inclaw also requires the establishmeregistry of child support orders. registries and are located in the Fallows the states to link up to sev

Figure 1: Child Support System

function of the linked databases, a unique identifier for each f the SSN as the unique child support case identifier, the 1996 are Reform Act gave child support enforcement agencies ng SSNs. These new sources include license applications nal, recreational, and marriage), divorce decrees, support orders, ledgements, and death certificates.

uires states to implement automated registries of child support case in which CSE services are being provided and each ified on or after October 1, 1998. Given the county cies, local registries may be linked to form a central State of the support owed under the order; arrearages, interest or late d and distributed; child's date of birth; any liens imposed; and, parents (i.e. SSN, name, date of birth, and case identification

d support registries, the 1996 law required the States to y of new hires by October 1, 1997. This directory contains luding name, address, SSN, and tax identification number. The nt of a national directory of new hires, as well as a federal These federal directories contain information found in the state ederal Parent Locator System (FPLS). Finally, the 1996 law eral databases and allows the FPLS to be used to establish


parentage; establishing, setting the amount of, modifying, or enforcing child support obligations; or enforcing child custody or visitation orders.


Table 3. Organizational Characteristics of CSE Respondents Boundaries of the IT project were…

Clear 84 percent Fuzzy 16 percent

IT Work Activities were…

Routine & formal

78 Ad-hoc & informal

22

IT Leadership was… Directive 74 Nondirective 26 IT Team Members were…

Loyal to Org 68 Loyal to Profession

32

IT Rules and Procedures were…

Explicit 63 Implicit 37

IT Information Flows were…

Vertical 47 Horizontal 53

IT Roles were Prescribed by…

Authority 37 Team 63

IT Organizational Structure was…

Formal 32 Informal 68

IT Division of work was…

Highly differentiated

28 Overlapping 72


0 2 4 6 8 10

9

7

7

6

6

5

4

4

3

0

tight

con

figur

atio

n

loose configuration

Figure 2: Balance Between Loose and Tight Structural Alternatives


Table 4. Perceptions on Success, Agitation, and Outcomes

Section A. IT Success Requires… Low Med High Mean

(7 point scale)

S.D. (7 point scale)

Negotiable boundaries 5 percent 11 percent 84 percent 5.68 1.24 Routine and Formal Work Activities

11 5 84 5.6 1.61

Directive leadership 26 11 63 3.3 1.75 Organizational Loyalty 6 6 88 5.8 1.24 Explicit Rules 21 11 68 4.8 1.65 Horizontal Info Flows 6 12 82 5.6 1.17 Team Prescribed Work Roles 16 16 68 5.2 1.54 Hierarchical Structures 37 16 47 4.4 1.98 Differentiated Div of Work 26 32 42 4.4 1.73 Section B. Team Agitated by… Low Med High Mean

(5 point scale)


Heavy Workload 42 percent 32 percent 26 percent 3.7 1.0 Conflicting Demands 47 32 21 3.6 1.0 Too Little Authority 68 16 16 3.1 1.4 Ambiguous Resp 84 0 16 2.8 1.3 Quality Suffers from Heavy Workload

53 37 11 3.5 .84

Too Little Information 74 21 5 3.1 1.0 Quality Suffers from Compromise

74 21 5 2.8 1.0

Section C. IT Implementation Resulted in… Low Med High Mean

(5 point scale)


Changes in Interaction Patterns with Clients

17 percent 11 percent 72 percent 5.3 1.8

Changes in Coordination/Comm Procedures

5 22 72 5.3 1.6

Addition / Elimination of Org Units

39 0 61 4.3 2.4

Changes in Management Resp

27 17 56 4.6 2.2


Table 5. Influence of Tight and Loose Configurations IT Organizational Structure

Formal mean Informal F Sig

Success Requires: Negotiable boundaries 6.5 5.3 4.5 .05 Hierarchical Structures 6.1 3.6 10.3 .01 Team Prescribed Work Roles 6.1 4.8 3.9 .07 Differentiated Div of Work 5.6 3.7 6.3 .02 Agitated by: Too Heavy Workload 4.3 3.5 3.6 .07 Too Little Authority 3.6 2.4 7.3 .02 IT Work Activities Routine Ad-Hoc F Sig Success Requires: Organization Loyalty 6.3 4.0 27.8 .00 Negotiable boundaries 6.0 4.3 9.1 .01 Team Prescribed Work Roles 5.5 3.8 5.0 .04 Agitated by: Too Little Information 3.0 2.3 3.8 .07 IT Information Flows Vertical Horizontal F Sig Success Requires: Differentiated Div of Work 5.2 3.6 5.1 .04 Horizontal Info Flows 4.8 6.3 13.9 .00 Leadership Direction 2.5 3.9 3.09 .09 Agitated by: Too Little Authority 3.3 2.3 4.6 .05 Judgment Sacrifices 3.0 2.3 3.3 .09 Project Resulted In… Important Communication Changes

4.4 6.0 6.2 .02

IT Team Members Loyal to Org Loyal to Prof F Sig Success Requires: Organizational Loyalty 6.2 5.0 3.5 .08 Pre-Described Work Activities

6.0 4.6 3.1 .09

Team Prescribed Work Roles 5.6 4.1 4.8 .04 Explicit Rules 5.2 3.8 3.3 .09 IT Boundaries Clear Fuzzy F Sig Success Requires: Organizational Loyalty 6.1 4.6 3.6 .08 Team Prescribed Work Roles 5.5 3.7 4.2 .06 Agitated by: Too Little Authority 2.6 3.6 3.7 .07



References Anderson, P. (1999). Complexity Theory and Organization Science. Organization Science

10:3(216-232). Anderson, P.; Meyer, A.; Eisenhardt, K.; Carley, K.; Pettigrew A. (1999). Introduction to the

Special Issue: Applications of Complexity Theory to Organization Science. Bak, P and Chen K (1991). Self-Organized Criticality. Scientific American 264:1(46-53). Cohen, M. (1999). Commentary on the Organization Science Special Issue on Complexity.

Organization Science 10:3(373-376). Comfort, L. (1994). Self-Organization in Complex Systems. Journal of Public Administration

Research Theory 4:3(393-410). Dooley, K. (1997). A Complex Adaptive Systems Model of Organization Change. Nonlinear

Dynamics, Psychology, and Life Sciences 1:1(69-97) Dooley, K. and Van de Ven, A. (1999) Explaining Complex Organizational Dynamics.

Organization Science 10:3(358-372). Kauffman, S.A. (1993). The Origins of Order. New York: Oxford University Press. Lawrence B (1997) The Black Box of Organizational Demography. Organization Science, 8(1-

22). Lefebvre, E. and Letiche, H. (1999). Managing Complexity from Chaos: Uncertainty,

Knowledge and Skills. Emergence 1:3(7-15). Levinthal D. and Warglien M (1999). Landscape Design: Designing for Local Action in

Complex Worlds. Organization Science 10:3(342-357). Levinthal, D.A. (1997). Adaptation and Rugged Landscapes. Management Science, 43(934-

950). Lewin, A. (1999) Application of Complexity Theory to Organization Science. Organization

Science 10:3(215). Lissack, M (1999). Complexity: The Science, its Vocabulary, and its Relation to Organizations.

Emergence 1:1(110-126). Morel, B. and Ramanujam, R. (1999). Through the Looking Glass of Complexity: The

Dynamics of Organizations as Adaptive and Evolving Systems. Organization Science 10:3(278-293).


Office of Child Support Enforcement. (1994). Annual Report to Congress (for period ending September 30, 1994). Washington, DC: Author.

Office of Child Support Enforcement. (1995a). FYI--Census report: Single parent growth rate stabilizes. Child Support Report, 17(11), p. 5.

Office of Child Support Enforcement. (1995b). President Clinton signs executive order: Federal agencies to facilitate support order establishment and enforcement. Child Support Report, 17(3), pp. 1-2.

Office of Child Support Enforcement. (1996 and various years). Annual Report to Congress (21 annual reports for the years 1976-96). Washington, DC: Author.

Prigogine I, and Stengers, I (1984) Order Out of Chaos. New York: Bantam. Prigogine, I. and Nicolis G. (1977). Self-Organization in Nonequilibrium Systems: From

Dissipative Structures to Order through Fluctuations. New York: Wiley. Ruelle, D. (1989). Chaotic Evolution and Strange Attractors. Cambridge: Cambridge University

Press. Scott W.R. (1992) Organizations: Rational, Natural and Open Systems. Englewood Cliffs, N.J.:

Prentice-Hall. Simon H.A. (1996). The Architecture of Complexity, 3rd ed. Cambridge: MIT Press Stacey, R. (1995). The Science of Complexity: An Alternative Perspective for Strategic Change

Processes. Strategic Management Journal 16:6(477-495). Thompson J. (1967) Organizations in Action, Social Science Bases of Administrative Theory.

New York: McGraw-Hill.

U.S. General Accounting Office. (1987). Child support: Need to improve efforts to identify fathers and obtain support orders (HRD-87-37). Washington, DC: Author.

U.S. General Accounting Office. (1992a). Interstate child support: Wage withholding not fulfilling expectations (HRD-92-65BR). Washington, DC: Author.

U.S. General Accounting Office. (1992b). Child support enforcement--timely action needed to correct system development problems (IMTEC-92-46). Washington, DC: Author.

U.S. General Accounting Office. (1996). Child support enforcement: Early results on comparability of privatized and public offices. (HEHS-97-4). Washington, DC: Author.

U.S. House of Representatives. (1983). Child support enforcement amendments of 1983 (House Report No. 98-527). Washington, DC: U.S. Government Printing Office.



Achieving the Benefits of E-Government? An …teep.tamu.edu/Npmrc/Brown2.pdf · Achieving the Benefits of E-Government? An Examination of the ... as homeostasis, cybernetic control,

Documents