The boundaries of information sharing and integration: A case study of Taiwan e-Government

Government Information Quarterly 29 (2012) S51–S60

Contents lists available at SciVerse ScienceDirect

Government Information Quarterly journal homepage: www.elsevier.com/locate/govinf

The boundaries of information sharing and integration: A case study of Taiwan e-Government Tung-Mou Yang a, Lei Zheng b,⁎, Theresa Pardo c a b c

Department of Library and Information Science, National Taiwan University No. 1, Sec. 4, Roosevelt Rd., Taipei, 10617 Taiwan, R.O.C. School of International Relations and Public Affairs, Fudan University 220 Handan Rd., Shanghai 200433, China Center for Technology in Government, University at Albany, State University of New York 187 Wolf Rd., Suite 301, Albany, NY 12205, USA

a r t i c l e

i n f o

Available online 3 December 2011 Keywords: Information sharing Boundary Electronic government Taiwan

a b s t r a c t The paper explores boundaries of cross-boundary information sharing and integration in the context of Taiwan e-Government by using an integrated framework of boundaries adopted from the literature. The discussion of the various boundaries provides a thorough lens to understand the complexity of crossboundary information sharing and integration. The adopted framework of boundaries is proved to be a useful analytical tool to perceive various vertical and horizontal boundaries in initiatives of crossboundary information sharing and integration in different e-Government contexts. A new process boundary in the vertical dimension is also identified. In addition, the case shows that centralized information systems can help government agencies to cross the boundaries of information sharing and reduce the number of boundaries government agencies may encounter. Lastly, it is perceived that vertical boundaries are not always easier to cross than horizontal boundaries. © 2011 Elsevier Inc. All rights reserved.

1. Introduction Electronic government (e-Government) has been an important strategy for attaining effectiveness and efficiency in government administrations and public services (Dawes, 2008, 2009; Gil-Garcia, Chengalur-Smith, & Duchessi, 2007; Pardo & Tayi, 2007). Researchers define e-Government as the delivery of government services (information, interaction, and transaction) through the use of information and communication technologies to improve daily business, reduce expenses, and increase the quality of services (Bekkers, 2007; GilGarcia & Martinez-Moyano, 2007; Moon, 2002). During e-Government development, cross-boundary information sharing and integration is important when critical information for running government operations are usually scattered around government agencies maintaining respective information systems. The demand for crossboundary information sharing and integration exists not only across different levels of government agencies (the vertical dimension) but also among government agencies with different functionalities (the horizontal dimension) (Gil-Garcia & Martinez-Moyano, 2007; Klievink & Janssen, 2008, 2009; Layne & Lee, 2001; Schooley & Horan, 2007; Siau & Long, 2005). The scope of cross-boundary information sharing and integration also varies in terms of involving national contexts and

⁎ Corresponding author. E-mail addresses: [email protected] (T.-M. Yang), [email protected], [email protected] (L. Zheng), [email protected] (T. Pardo). 0740-624X/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.giq.2011.08.014

levels of government while e-Government develops among local, state, and national governments (Gil-Garcia & Martinez-Moyano, 2007). With cross-boundary information sharing and integration, more effective actions can be applied to solve complex problems, and electronic data can be easier to duplicate and manipulate. Hence, cost is reduced, efficiency is accomplished, and government agencies can act faster to identify problems and react with prompt responsiveness (Gil-Garcia, Soon Ae, & Janssen, 2009; Landsbergen & Wolken, 2001; Luna-Reyes, Gil-Garcia, & Cruz, 2007). Nevertheless, government organizations have different levels, functionalities, goals, values, and cultures, and thus cross-boundary information sharing and integration can become a complex task. In the literature, Gil-Garcia, Pardo, and Burke (2010) provide preliminary definitions of four components of cross-boundary information sharing and integration. The proposed four components are trusted social networks, shared information, integrated data, and interoperable technical infrastructure, which cover both technical and social aspects. In addition, the literature also discusses various barriers and enablers of cross-boundary information sharing and integration from different perspectives including technological, organizational, sociological, ideological, and political contexts in order to comprehend the complexity of cross-boundary information sharing and integration (Dawes, 1996; Dawes, Cresswell, & Pardo, 2009; Fedorowicz, Gogan, & Williams, 2007; Gil-Garcia & Pardo, 2005; Gil-Garcia et al., 2007, 2009; Harrison, Pardo, Gil-Garcia, Thompson, & Juraga, 2007; Pardo & Tayi, 2007; Schooley & Horan, 2007; Yang & Maxwell, 2011; Zhang & Dawes, 2006; Zheng, Jiang, Yang, & Pardo, 2008).

S52

T.-M. Yang et al. / Government Information Quarterly 29 (2012) S51–S60

2. What are the boundaries? However, what are the “boundaries” in cross-boundary information sharing and integration? In existing concepts such as working environment, organizational task, and relations with others, boundaries can be described as discontinuities, representing the incoherence and gaps between two entities (Espinosa, Cummings, Wilson, & Pearce, 2003; Zheng, Yang, Pardo, & Jiang, 2009). Regarding the definition of boundary in the context of cross-boundary information sharing and integration in the public sector, current literature has mainly focused on organizational boundaries. In addition, personal, sectoral, and geographic boundaries are also identified in the literature. Zheng et al. (2009) give the following distinction between “boundaries” and “barriers” in cross-boundary information sharing and integration: “Put simply, a boundary is a line we need or want to cross, and a barrier is what prevents us from crossing. The difficulty of crossing a specific boundary is determined by the existence of certain political, organizational and technological barriers around it. The difficulty of boundary crossing can be much lower when significant barriers do not exist or have been removed. Barriers may therefore be overcome or eliminated with some efforts, but boundaries tend to exist for a long period of time unless significant institutional changes occur to remove them.” The following subsections summarize the discussions about boundaries of information sharing and integration in the literature.

Galivan, 2001), and the information flow across the sectoral boundary has been further increased. Furthermore, geographic boundaries exist in cross-boundary information sharing when government agencies are spread in various geographic locations (Williams et al., 2009). Different cultures and time zones can cause challenges in collaboration work (Espinosa et al., 2003; Orlikowski, 2002). Distance between two geographic locations can also lead to inefficient communication, misunderstanding, and conflict (Armstrong & Cole, 2002; Espinosa et al., 2003). An international boundary exists further when information sharing initiatives involve inter-organizational collaborations across nations (Navarrete, Gil-Garcia, Mellouli, Pardo, & Scholl, 2010). 2.3. Development level and process boundaries Zheng et al. (2009) claim that when gaps exist in the levels of technological, managerial, personnel, and economic developments of participating organizations, a development level boundary can occur. It is observed that information sharing initiatives seem to work better in organizations with similar development levels than those with different development levels. In addition, process boundary also takes place when organizations do not participate in the same business process. It is believed that organizations participating in the same business process are more likely to share information with one another than those that are not (Zheng et al., 2009).

2.1. Organizational boundary 2.4. An integrated framework of the “boundaries” Organizational boundary in information sharing and integration has been studied from both the vertical and horizontal dimensions (Black, Carlile, & Repenning, 2004; Bouty, 2000; Chen, Gangopadhyay, Holden, Karabatis, & McGuire, 2007). Vertical hierarchical structure can hinder information sharing and integration of organizational units residing in different levels (Tsai, 2002). Horizontal departmentization can also impede information sharing among horizontal departments with different functionalities (Argote, Ingram, Levine, & Moreland, 2000; Willem & Buelens, 2007). Due to the differences in expertise, experiences, and regulations in different domains, organizational boundaries exist among different organizations or departments (Carlile, 2004; Espinosa et al., 2003). It is also pointed out in literature that organizational boundaries occur among government agencies due to the differences in their defined missions, utilized resources, organized capacities, assigned responsibilities, and respective accountabilities (Artigas, Elefante, & Marti, 2009; Fedorowicz et al., 2007; Harrison et al., 2007; Kettl, 2006; Luna-Reyes et al., 2007). In the public sector, organizational boundaries are also the result of legislation creating certain departments or agencies and defining their missions (Dawes, 1996; Pardo, Cresswell, Dawes, & Burke, 2004). 2.2. Personal, sectoral, and geographic boundaries Personal boundaries also play an important role in information sharing and integration. Interpersonal relationships and interactions can influence attitudes and intentions to share information. Informal relationships such as personal networks that are not previously arranged and defined by hierarchy and regulation can result in more intense and effective information sharing between departments of an organization (Tsai, 2002; Willem & Buelens, 2007). In addition, a sectoral boundary exists between the public and the private sectors. Organizations in the public sector can own different origins, values, and cultures from those in the private sector (Pardo & Tayi, 2007; Pardo et al., 2004). Researchers suggest that governmental information sharing and integration should not be limited to the public sector only. Both the private and the non-profit sectors need to be included to provide more integrated services to the public (Chan, Lau, & Pan, 2008). There are also more and more governmental information systems outsourced to the private sector (Beyah &

Zheng et al. (2009) proposes a two-dimensional and interactive framework to give a comprehensive illustration of the boundaries in cross-boundary information sharing and integration (see Fig. 1). In the vertical dimension of the framework, there are four boundaries, which are the hierarchical boundary, the personal boundary, the geographic boundary, and the development level boundary. In the horizontal dimension of the framework, there are six boundaries, the departmental boundary, the personal boundary, the geographic boundary, the development level boundary, the process boundary, and the sectoral boundary. When recent research seems to show that boundaries of the vertical dimension are less complex than those of the horizontal dimension (Chan et al., 2008; Klievink & Janssen, 2009; Layne & Lee, 2001), Zheng et al. (2009) claim that boundaries of the vertical and horizontal dimensions are equally important and can exist simultaneously. For instance, crossing a hierarchical boundary in the vertical dimension is not always easier than crossing a departmental boundary in the horizontal dimension as some researchers have suggested (Zheng et al., 2009). While most information-sharing related literature is developed in the context of Western countries, and Zheng et al.'s (2009) framework of boundaries is built based on a case study in Mainland China, it is interesting to further verify the framework of boundaries in other

Fig. 1. A framework of boundaries in cross-boundary information sharing and integration.

T.-M. Yang et al. / Government Information Quarterly 29 (2012) S51–S60

e-Government contexts. In addition, a research gap in the literature also emerges with regard to how these boundaries are bridged to achieve information sharing and integration. The following research questions are proposed for the study: 1. Can the proposed framework of boundaries be applied to other national contexts of e-Government? 2. What are the boundaries of information sharing and integration in other national contexts of e-Government? 3. How are the perceived boundaries bridged to achieve information sharing and integration? 3. Research design and method This study adopts a case study method to extend the framework in literature to a different context. In particular, a case of the e-Networking Project of Government Online Service in Taiwan is studied in the research. This project was a sub-plan of the Challenge 2008—Taiwan National Development Plan proposed by the Cabinet of Taiwan in 2002, and the duration of the project was from 2002 to 2007. The research explores the case through both the vertical and horizontal networks in the context of Taiwan e-Government (see Fig. 2). Qualitative data were collected through semi-structured interviews with key actors from central and local government who participated in the related initiatives. A purposive sampling by using snowball sampling method was adopted to identify relevant interviewees. Purposive sampling is usually employed in the selection of information-rich cases to conduct in-depth study (Wengraf, 2001). In total, twenty-eight participants were interviewed. The average duration of the interviews was about one hour and thirty minutes. In-depth and semi-structured interviews were used to collect data for analysis. Semi-structured interviews have the flexibility to follow up on new information presented in the context of an interview and to explore in-depth new findings. According to Bryman (2004), an interviewer using semi-structured interviews usually has a series of questions that are in the general form of an interview protocol but the sequence of interview questions can vary during interviewing. The interviewer can ask further questions when something important is provided (Bryman, 2004). In addition, government documentation and reports, and related reports and documents from newspapers and non-governmental organizations were also collected and used for the purpose of data analysis. The collected data were transcribed and analyzed by applying grounded theory techniques. Three types of coding, open coding, axial coding, and selective coding, are adopted to analyze and

S53

interpret qualitative data (Strauss and Corbin, 1998). During the data analysis, the qualitative software tool (Atlas.ti) is employed to support coding and analysis activities. 4. Case study Taiwan has started developing its e-Government since 1997. In the past decade, Taiwan e-Government has focused effort on developing initiatives of cross-boundary information sharing and integration. The case of interest in the study, the e-Networking Project of Government Online Service, was started as a follow-up project to continue promoting cross-boundary information sharing and integration when information systems were built across the government agencies of Taiwan. The goal of the project is to provide integrated and efficient services to the public, private firms, public organizations, as well as government agencies through cross-boundary information sharing and integration across several major information systems. The major information systems include the Household Registration Information System (HRIS), the Land Administration Information System (LAIS), the Financial Taxation Information Systems (FTIS), the Commerce Information Systems (CIS), and the Motor Vehicle & Driver Information Systems (MVDIS). The five systems are the backbone information systems and the foundation of Taiwan e-Government. Most government agencies of Taiwan have to acquire information periodically from the five major information systems to run their daily operations. In the e-Networking Project of Government Online Service, in addition to strengthening information sharing and integration across the vertical boundaries of government agencies, the Taiwan government endeavors most of its effort to promote information sharing and integration across the horizontal boundaries. In the project, a Government Service Platform was designed and established to act as an intermediary for government agencies to develop and implement cross-agency service integration. The network of Government Service Platform is a star-shaped network that participating government agencies connect to the platform. The platform is designed to save time and expense so that government agencies can focus on developing innovative services and leave the technical complexity enclosed in the platform. 5. The findings Different boundaries of cross-boundary information sharing and integration identified in the case study are presented and illustrated from both the vertical and horizontal dimensions. In addition, centralized information systems utilized in the both dimensions are observed to help government agencies to cross boundaries

Fig. 2. The institutional context of the research.

S54

T.-M. Yang et al. / Government Information Quarterly 29 (2012) S51–S60

and reduce the number of boundaries government agencies may encounter. 5.1. Vertical dimension In the vertical dimension, flows of cross-boundary information sharing are passed down or up vertically across government agencies. The vertical dimension usually involves the interaction between central government agencies and local government agencies. 5.1.1. Hierarchical boundary A hierarchical boundary is defined as a boundary existing between a higher-level government agency and a lower-level government agency. In the case study, a hierarchical boundary can be recognized between a central government agency and a local government agency, for instance, in the Household Registration Information System (HRIS), three levels of government agencies are involved in crossboundary information sharing and integration. The Department of Household Registration (DHR) is a central government agency under Ministry of the Interior. The Department of Civil Affairs (DCA) and the Household Registration Offices (HRO) are local government agencies under County Government. There is a hierarchical boundary between the DHR and the DCAs. The relationship between government agencies at the two levels is indirect in terms of administrative structure but direct in cross-boundary information sharing. A DCA usually directly reports to a county government. However, the DHR and the DCAs have the same core business (household registration), and the DHR provides policy and operational guidance to the DCAs in the local government level. The HROs are the township government agencies to obtain household information directly from the public. The obtained household information is shared with the DCAs at the county government level. Then the DCAs further share the information to the DHR at the central government level. An interviewee stated: Study participant P13: “…the local Household Registration Offices are very important. They are the front line to the public. Information collected by them is shared to us, the Department of Civil Affairs, and we share the information to the central government agency, the Department of Household Registration. There are total three levels, township level, county government level, and central government level….” On the other hand, a reverse hierarchical boundary also exists when there is an information-sharing requirement between two local government agencies of different counties. For instance, when the DCA in county A requires information sharing from the DCA in county B, the DCA in county A needs to contact the DHR in the central government to acquire information rather than to contact the DCA in county B. Because the Household Registration Information System is built in a centralized hierarchical structure, the information sharing between the two DCAs of different counties needs the DHR to act as an intermediary. Therefore, the information sharing activity between the DHR and the DCAs is bi-directional to cross the hierarchical boundary. Similarly, there are also hierarchical boundaries existing in the other four aforementioned information systems. 5.1.2. Process boundary When the literature identifies a process boundary between organizations with disparate business processes at the horizontal level, the case study in Taiwan finds that a process boundary could also exist between government agencies that are not connected with the same business processes in the vertical dimension. In the case study, if other central government agencies would like to have information sharing from the local government agencies of the five major information systems, they would have to contact the responsible

central government agencies of the five major information systems rather than the local government agencies. However, if a central government agency would like to have information sharing from other local government agencies outside of the five major information systems, a process boundary in the vertical dimension can appear if the central government agency does not have connected business process with the local government agencies. For instance, there is no centralized social affairs information system. To run its new operation, in addition to acquiring information from some major information systems, the Bureau of Labor Insurance needs to have information sharing from every Department of Social Affairs in the local government level. Because of different core businesses and no connected business process, there is a process boundary existing between the Bureau of Labor Insurance in the central government and the Department of Social Affairs of every local government. An interviewee from a central government agency stated: Study participant P8: “…Although we have to acquire information from the local government agencies, we actually have different core business and our business processes are not connected. However, we currently need their information to run our new business….”

5.1.3. Geographic boundary A geographic boundary is a naturally formed boundary that can not be removed unless there is a significant institutional change. When a central government agency has information-sharing activities with a local government agency, a geographic boundary naturally exists between the two government agencies. A geographic boundary could involve properties such as physical distance, geographic terrain, and time zone. For a central government agency, some local government agencies may be located just a few blocks or miles away, but some may be located in remote areas. In the case study, each of the five major information systems involves a responsible central government agency and local government agencies located in different counties, and consequently geographic boundaries exist among them. 5.1.4. Development level boundary Informatization asymmetry can be an example of development level boundary. In the case study, central government agencies and local government agencies located in metropolitan areas have better funding and resources to develop their information infrastructure than do those remote local government agencies. A development level boundary in terms of informatization asymmetry between a central government agency and a remote local government agency is gradually formed and broadened. In addition, it is also evident that some local government agencies located in large cities such as Taipei and Kaoshiung are more advanced in terms of their information infrastructure than central government agencies. Therefore, a reverse informatization asymmetry appears. An interviewee from a central government agency stated: Study participant P4: “…According to our experience, local government agencies in large cities such as Taipei, Tacichung, and Kaoshiung do have much better information environments for us to develop and deploy the information system. They have more information professionals and sufficient funding than do other local agencies in local governments such as Taidung, Yunlin and Miaoli. If we don't give a hand to those local government agencies with weaker information environments, they would never make it…

5.1.5. Personal boundary In addition to the aforementioned four boundaries formed by the organizational structures, the geographic locations, and the

T.-M. Yang et al. / Government Information Quarterly 29 (2012) S51–S60

development divides, there is also an informally formed personal boundary existing in the vertical dimension of cross-boundary information sharing. A personal boundary could appear between a central government agency and a local government agency when they communicate and negotiate an initiative of cross-boundary information sharing. A personal boundary could exist between representative personnel involved in the initiative. A personal boundary is naturally formed thanks to various participating individuals and their different standpoints and perspectives. Therefore, individual characteristics, communication skills, and personal involvement are important influential factors when it comes to cross a personal boundary. Some interviewees stated: Study participant P2: “In the stage of communication and negotiation for information sharing, the leaders and the representatives of government agencies matter. Their personal characteristics and relationship building during interactions matter. It is sometimes more complicated in the aspect of personal interaction than in the aspect of regulation and legislation….” Study participant P10-1: “…The personal interaction, relationship building, and the ability in communication skills are important. When you have to deal with the government agencies in the local government level, it means that you have to interact and negotiate with many people from different areas and with different personal characteristics….” In sum, in the case study, the five identified boundaries in the vertical dimension are hierarchical boundary, process boundary, geographic boundary, development level boundary, and personal boundary.

5.2. Horizontal dimension In the horizontal dimension, flows of cross-boundary information sharing are passed horizontally among parallel government agencies at the same level. The following subsections discuss these horizontal boundaries identified in the case study.

S55

5.2.1. Departmental boundary A departmental boundary is defined as a boundary formed by governmental structure. The boundary can occur between two or more parallel government agencies at the same level. At the central government level, an intra-ministry departmental boundary can exist between two government departments within the same ministry. For instance, the Department of Land Administration needs to have information shared by the Department of Household Registration for identity confirmation of land administration. The two government agencies are both under the Ministry of the Interior (see Fig. 3). An interviewee stated: Study participant P2: “…We also share information with other agencies within the Ministry of the Interior. Ministry of the Interior is the largest governmental organization and contains many different agencies. However, even within the same ministry, we also need to follow the same procedure by using official documents to understand whether the information sharing is necessary. The process cannot be ignored just because we are in the same ministry….” A departmental boundary can be further extended to the government agencies across different ministries. For instance, to achieve efficiency and accuracy in its taxation business, the Financial Data Center requires information shared by the Department of Household Registration. The two central government agencies belong to two different ministries, and clearly an inter-ministry departmental boundary appears between them. Therefore a departmental boundary can exist at both intraministry level and inter-ministry level (see Fig. 3). In this case study, a departmental boundary in the inter-ministry level is believed to be more difficult to cross because of the increasing number of barriers surrounding the boundary. An interviewee stated: Study participant P14: “…I believe the bureaucracy structure is stricter at the central government level. There is a structure formed to separate different government agencies, especially those in different ministries….” At the local government level, there is also a departmental boundary between parallel local government agencies. For instance, if a Local Tax

Fig. 3. The departmental boundary in the horizontal dimension.

S56

T.-M. Yang et al. / Government Information Quarterly 29 (2012) S51–S60

Bureau cannot reach a certain taxpayer via its own information, the agency needs to contact the Department of Civil Affairs in its county to acquire taxpayers' updated contact information. The flow of information sharing is horizontal between the two parallel local government agencies at the same level (see Fig. 3). However, in this case study, a departmental boundary between two local government agencies in two different counties was alleviated to some extent. The five major information systems in the case are designed, deployed, and maintained in a centralized structure. If the Local Tax Bureau of local government B needs to retrieve information from the Department of Civil Affairs of local government A, the local government agency has to go up to the central government level and have the Financial Data Center act as an intermediary to contact the Department of Household Registration to acquire the information rather than directly contact the local agency of local government A (see Fig. 3). The approach is to remove as many departmental boundaries as possible by utilizing the centralized information systems. Otherwise, the number of departmental boundary for a local government agency may increase drastically, and the challenges to cross different departmental boundaries can vary from one another. An interviewee stated: Study participant P3-1: “…If the local government agencies are within the same county, they can directly share information. That is the autonomy of a local government. However, if the information a local government agency needs is national-wide information or is information of other counties, it is necessary for the local government agency to contact and acquire the information through its functional government agency in the central government level….”

5.2.2. Process boundary In the horizontal dimension, a process boundary appears when two government agencies at the same level do not have connected business process. For instance, in its Financial Taxation Information System, the Financial Data Center has its core business connected with other central government agencies such as the agencies of National Tax Administration. Because of this already connected business processes, there is no process boundary between the Financial Data Center and the agencies of National Tax Administration. However, it often happens that a government agency needs to have information shared by other government agencies to run its core business, usually when two government agencies have no connected core business process between them. For instance, the Department of Land Administration focuses on land administration, and the Financial Data Center concentrates on taxation business. There is no connecting core business process between the two central government agencies. However, the Financial Data Center needs information shared by the Department of Land Administration because it believes the land information from the Department of Land Administration is more thorough and accurate to help it to increase taxation revenue. In the situation, not only a departmental boundary but also a process boundary exists between the two central government agencies. The situation also applies to local government agencies in the same county. However, within the scope of the five major information systems, a process boundary does not exist between two local government agencies of different counties. The information sharing activity between the two local government agencies is required to go through their responsible central government agencies by using the centralized information systems. 5.2.3. Geographic boundary A geographic boundary also exists among parallel government agencies. As aforementioned in subsection 5.1.3, a geographic boundary is a naturally formed boundary involving properties such as physical distance, geographic terrain, and time zone. In this case study, the

geographic boundary in the horizontal dimension is relatively narrow because of the small territory of Taiwan. At the central government level, almost all central government agencies are located in the capital area. Therefore, the geographic boundaries among the central government agencies exist but with very narrow gaps. At the local government level, although some counties are located in remote islands, most local government agencies in the same county are located closely. Similarly, because of the five centralized information systems, a geographic boundary of cross-boundary information sharing was alleviated among local government agencies of different counties. Some interviewee pointed out that: Study participant P5-1: “…If every government agency were in the same building, there might be a better chance to develop closer relationship. However, Taiwan is so small that I think there should be few problems regarding where a government agency is located. Especially our central government agencies are all located in Taipei area. The geographic boundary does exist, but it is really small….” Study participant P1: “…Different geographic locations of government agencies don't matter that much. By using the inter-organizational information system, we can easily cross a geographic boundary between two government agencies….”

5.2.4. Development level boundary A development level boundary does not just exist vertically between a central government agency and a local government agency. It also exists horizontally between government agencies at the same level. Parallel government agencies could have different budgets and resources, and their progress in informatization varies. For instance, both the Department of Household Registration and the Department of Land Administration have similar development level in informatization and the information sharing between the two central government agencies are fluent. However, some juridical agencies in central government level have very modest development in terms of informatization, consequently, the information sharing between the juridical agencies and other central government agencies is more limited and less fluent. Similarly, a development level boundary also occurs between two local government agencies at the same level. However, a development level boundary between two local government agencies of different counties does not appear significantly in the case because of the five centralized information systems. From central government agencies, some interviewees stated: Study participant P3: “…there is still paper-based information sharing because the informatization in government agencies is not equivalent. Some central government agencies may have limited budget and their development of informatization is much slower than others….” Study participant P10-1: “…Every government agencies has different ‘readiness’ to manage its information. Some conservative agencies such as judicial agencies operate in an old-fashioned way. There is a huge gap for them to share information to our information system. The ‘readiness’ of informatization between our agency and theirs is totally different….”

5.2.5. Personal boundary A personal boundary in the horizontal dimension forms between representative personnel of government agencies at the same level. The interaction and relationship of personnel and leaders of government agencies matter especially for government agencies at the

T.-M. Yang et al. / Government Information Quarterly 29 (2012) S51–S60

same level. For instance, if the leaders of two parallel government agencies have poor personal relationship, the personal boundary is inevitably broadened, and information sharing between the two parallel government agencies could be more difficult and even be terminated. A personal boundary formed between two parallel government agencies at the same level could be broader than that formed between two hierarchical government agencies having a supervisor–subordinate relationship. Similarly, because of the five centralized information systems, a personal boundary was reduced between two local government agencies of different counties. Some interviewees stated: Study participant P2: “…In many situations, it is the issue of interaction between people. The style and leadership of leaders of government agencies matter. They can influence whether two government agencies cooperate. If the leaders of two government agencies have poor relationship and unfavorable opinions against each other, how can the two government agencies share information to each other?….” Study participant P5: “…if there is a good social network built between two government agencies to foster their relationship. It should be able to benefit their communication and negotiation to achieve efficient information sharing between the two agencies….”

5.2.6. Sectoral boundary In addition to the aforementioned five boundaries in the horizontal dimension between government agencies, there is also a sectoral boundary existing between government agencies and private enterprises or non-profit organizations. Nowadays, the interactions between the public sector and the private sector increase drastically. The crossboundary information sharing between the two sectors can be either one-directional or bi-directional. In the case study, in order to help small and medium companies to obtain bank loans, the Small and Medium Enterprise Administration (SMEA) acts as an intermediary to acquire company registration information from the Department of Commerce and taxation information from the Financial Data Center. Then the SMEA shares the information with banks to evaluate loan applications. The information can help banks to alleviate information

S57

asymmetry, and the chance for small and medium companies to obtain bank loans can thus increase. Furthermore, SMEA also considers acquiring usage information from Taiwan Power Company and Taiwan Water Corporation (pubic owned corporations) to give banks more detailed information to evaluate loan applications from small and medium companies (see Fig. 4). In this example, a sectoral boundary exists between the SMEA in the public sector and banks in the private sector. Similarly, the Department of Land Administration also shares information across sectors with the real estate agents. Some interviewees stated: Study participant P7: “…In our project, in addition to acquiring information from some government agencies, we also need to have information sharing from non-governmental organizations to fulfill the needs of our business….” Study participant P11: “…Some private companies or non-profit organizations also require information sharing from us. However, those organizations are not government agencies. We tend to be more conservative in term of information privacy and confidentiality….”

5.3. Centralized information systems to cross and alleviate boundaries It is observed that centralized information systems in the case study can facilitate government agencies to share information across boundaries. The following subsections discuss how centralized information systems help from both the vertical and horizontal dimensions. 5.3.1. Centralized information systems in the vertical dimension In the case study, the five major information systems were established to connect central government agencies and local government agencies that are connected with same core businesses such as household registration, land administration, and taxation. The centralized information systems are perceived as an effective approach to cross the vertical boundaries between the participating central and local government agencies and to increase the efficiency of information sharing. Because of the connected business processes and the existing supervisor–subordinate relationships between the participating

Fig. 4. The sectoral boundary in the horizontal dimension.

S58

T.-M. Yang et al. / Government Information Quarterly 29 (2012) S51–S60

Fig. 5. The removed boundaries by the centralized information systems.

higher-level and lower-level government agencies, the vertical process boundary and the vertical personal boundary become much narrower than the other three vertical boundaries through the implementation of the five major information systems. In addition, the centralized information systems also remove the horizontal boundaries between local government agencies of different counties as long as their information sharing is within the scope of the centralized information systems. The information sharing between local government agencies of different counties is required to go up to the central government level by having the responsible central government agencies act as intermediaries (see Fig. 5). On the other hand, local government agencies of the same county have autonomy and discretion to directly share information with one another while horizontal boundaries still remain among them. Furthermore, the centralized information systems not only remove the horizontal boundaries between local government agencies of different counties, but also remove the vertical boundaries between central government agencies and local government agencies when they do not have connected core businesses and are within the scope of the centralized information systems. For each centralized information system, there is a single window in the central government level to resolve needs of cross-boundary information sharing. For instance, when central government agency A and local government agency A share the same core business and there is a centralized information system connecting them, other central government agencies can interact directly with central government agency A to acquire the information of local government agency A. There is no need for the other central government agencies to cross the vertical boundaries in order to acquire information from local government agency A (see Fig. 5). 5.3.2. Centralized information systems in the horizontal dimension In the case study, it is observed that the implemented Government Service Platform with star-shaped network acts as an intermediary to

facilitate central government agencies to share information across horizontal boundaries. By participating and connecting to the platform, central government agencies can leave the technological complexity enclosed in the platform, and the development level boundary between two central government agencies to share information can become narrower or even be removed (see Fig. 6). While the platform is a policy implementation by the Cabinet to promote cross-agency integrated service, the platform also acts as a legitimate warrant under legislative support to push and encourage central government agencies to share information across the other horizontal boundaries such as the departmental boundaries, the process boundary, and the personal boundary. The platform is an implementation of a cloud computing concept, Platform as a Service (PaaS) and can also be perceived as a centralized information system deployed in the central government level to facilitate horizontal information sharing across government agencies. 6. Discussion 6.1. The framework of boundaries is generalizable By using Zheng et al.'s (2009) framework of boundaries, boundaries of information sharing and integration in both the vertical and horizontal dimensions are discussed and illustrated in the previous sections. In this case study, five boundaries are identified in the vertical dimension, and they are hierarchical boundary, process boundary, geographic boundary, development level boundary, and personal boundary (see Fig. 7). The four boundaries in the vertical dimension of Zheng et al.'s framework of boundaries are all identified in the case study. In addition, although the process boundary in the vertical dimension is not discovered in Zheng et al.'s framework of boundaries, it is identified in the case study, and the framework can be further extended. In the horizontal dimension, six boundaries are identified. The six boundaries are departmental boundary, process boundary, geographic

Fig. 6. A platform to facilitate information sharing across horizontal boundaries.

T.-M. Yang et al. / Government Information Quarterly 29 (2012) S51–S60

S59

social welfare information system. Five boundaries identified in the vertical dimension in the case study. Therefore, when the BLI needs to interact with twenty-five local government agencies, there might be one hundred and twenty-five boundaries in the vertical dimension that the BLI has to cross to fulfill its information-sharing need. In this example, crossing vertical boundaries is believed to be more complex than crossing horizontal boundaries. 7. Conclusion

Fig. 7. The five boundaries in the vertical dimension.

boundary, development level boundary, personal boundary, and sectoral boundary (see Fig. 8). All the boundaries in the horizontal dimension of Zheng et al.'s framework of boundaries are identified in the case study. The result of the boundaries identification in the case study could verify that Zheng et al.'s integrated framework of boundaries in crossboundary information sharing could be generalizable to different national contexts of e-Government. 6.2. Vertical boundaries are not always easier to cross than horizontal boundaries The case study does not assume that boundaries in the vertical dimension are easier to cross than boundaries in the horizontal dimension. Once outside the scope of the five centralized information systems in the vertical dimension, the story is different for a central government agency to acquire information from local government agencies when they have no connected business process. A central government agency will have to interact with each local government agency separately. For instance, in the case study, because of the established Household Registration Information System, the Bureau of Labor Insurance (BLI) can have information shared directly by the Department of Household Registration (DHR) to acquire household information of all twenty-five counties of Taiwan. The boundaries that the BLI needs to cross are just the horizontal boundaries between it and the DHR. However, when the BLI needs to have social welfare information sharing from the Department of Social Affairs (DSA) of each local government, the BLI has to deal with one local government agency by one local government agency because there is no centralized

In sum, a government agency commonly has to deal with different government agencies to have cross-boundary information sharing to run its operations or to make its operations more efficient in innovative ways. A government agency inevitably encounters different vertical and horizontal boundaries simultaneously, and the boundaries also interact with each other. The more boundaries a government agency has to cross in an information-sharing initiative, the more influential factors surrounding the boundaries the government agency has to encounter and resolve. By adopting Zheng et al.'s framework of boundaries, the discussion of the various boundaries in this paper can provide a more thorough lens to understand the preliminary complexity of cross-boundary information sharing and integration. Zheng et al.'s framework of boundaries is proved to be a useful analytical tool to perceive various vertical and horizontal boundaries in different national contexts of e-Government. In the case study, a new process boundary in the vertical dimension is identified. In addition, it is observed that centralized information systems in the case study can help to cross or remove various boundaries of information sharing among government agencies. Furthermore, it is verified in the case study that vertical boundaries are not always easier to cross than horizontal boundaries. For future studies, it will be interesting to further apply the framework of boundaries to other national contexts of e-Government, to explore whether there are other boundaries that remain unidentified, and to see how other contexts of e-Government interact with the boundaries in their information-sharing initiatives. Lastly, the framework of boundaries also provides a conceptual foundation for other related research to further perceive the complexity of information sharing arisen by influential factors surrounding the boundaries. Acknowledgment The authors want to thank Dr. Yu-Hsieh Sung, the Deputy Minister of the Research, Development and Evaluation Commission and other government officials of the Taiwan government for their important help during the interviewee recruitment of the research. The authors

Fig. 8. The six boundaries in the horizontal dimension.

S60

T.-M. Yang et al. / Government Information Quarterly 29 (2012) S51–S60

are also thankful to all the interviewees who participated in the research for their time, patience, and valuable information and suggestions. References Argote, L., Ingram, P., Levine, J. M., & Moreland, R. L. (2000). Knowledge transfer in organizations: Learning from the experience of others. Organizational Behavior and Human Decision Processes, 82(1), 1–8. Armstrong, D. J., & Cole, P. (Eds.). (2002). Managing distances and differences in geographically distributed work groups. Cambridge, MA: MIT Press. Artigas, F., Elefante, D., & Marti, A. (2009). Geographic information sharing: A regional approach in northern New Jersey, USA. Information Polity: The International Journal of Government & Democracy in the Information Age, 14(1/2), 127–139. doi:10.3233/ ip-2009-0165 (Article). Bekkers, V. (2007). The governance of back-office integration—Organizing co-operation between information domains. Public Management Review, 9(3), 377–400. Beyah, G., & Galivan, M. (2001). Knowledge management as a framework for understanding public sector outsourcing. Paper presented at the Hawaii international conference on system sciences (HICSS-34), Hawaii. Black, L. J., Carlile, P. R., & Repenning, N. P. (2004). A dynamic theory of expertise and occupational boundaries in new technology implementation: Building on Barley's study of CT scanning. Administrative Science Quarterly, 49(4), 572–607. Bouty, I. (2000). Interpersonal and interaction influences on informal resource exchanges between R&D researchers across organizational boundaries. Academy of Management Journal, 43(1), 50–65. Bryman, A. (2004). Social research methods (2nd ed.). Oxford, U.K.: Oxford University Press. Carlile, P. R. (2004). Transferring, translating, and transforming: An integrative framework for managing knowledge across boundaries. Organization Science, 15(5), 555–568. Chan, C. M. L., Lau, Y., & Pan, S. L. (2008). E-Government implementation: A macro analysis of Singapore's e-Government initiatives. Government Information Quarterly, 25(2), 149–346. Chen, Z., Gangopadhyay, A., Holden, S. H., Karabatis, G., & McGuire, M. P. (2007). Semantic integration of government data for water quality management. Government Information Quarterly, 24(4), 716–735. doi:10.1016/j.giq.2007.04.004. Dawes, S. S. (1996). Interagency information sharing: Expected benefits, manageable risks. Journal of Policy Analysis and Management, 15(3), 377–394. Dawes, S. S. (2008). The evolution and continuing challenges of e-Governance. Public Administration Review, 86–102 (Special issue). Dawes, S. S. (2009). Governance in the digital age: A research and action framework for an uncertain future. Government Information Quarterly, 26, 257–264. Dawes, S. S., Cresswell, A. M., & Pardo, T. A. (2009). From "need to know" to "need to share": Tangled problems, information boundaries, and the building of public sector knowledge networks. Public Administration Review, 69, 392–402. Espinosa, J. A., Cummings, J. N., Wilson, J. M., & Pearce, B. M. (2003). Team boundary issues across multiple global firms. Journal of Management Information Systems, 19(4), 157–190. Fedorowicz, J., Gogan, J. L., & Williams, C. B. (2007). A collaborative network for first responders: Lessons from the CapWIN case. Government Information Quarterly, 24(4), 785–807. doi:10.1016/j.giq.2007.06.001. Gil-Garcia, J. R., Chengalur-Smith, I., & Duchessi, P. (2007). Collaborative eGovernment: impediments and benefits of information-sharing projects in the public sector. European Journal of Information Systems, 16(2), 121–133. Gil-Garcia, J. R., & Martinez-Moyano, I. J. (2007). Understanding the evolution of eGovernment: The influence of systems of rules on public sector dynamics. Government Information Quarterly, 24(2), 266–290. doi:10.1016/j.giq.2006.04.005. Gil-Garcia, J. R., & Pardo, T. A. (2005). E-Government success factors: Mapping practical tools to theoretical foundations. Government Information Quarterly, 22(2), 187–216. Gil-Garcia, J. R., Pardo, T. A., & Burke, G. B. (2010). Conceptualizing information integration in government. In J. Scholl (Ed.), Electronic government: information, technology, and transformation (pp. 179–202). Armonk, NY: ME Sharpe. Gil-Garcia, J. R., Soon Ae, C., & Janssen, M. (2009). Government information sharing and integration: Combining the social and the technical. (Article). Information Polity: The International Journal of Government & Democracy in the Information Age, 14(1/2), 1–10. doi:10.3233/ip-2009-0176. Harrison, T. M., Pardo, T., Gil-Garcia, J. R., Thompson, F., & Juraga, D. (2007). Geographic information technologies, structuration theory, and the World Trade Center crisis. Journal of the American Society for Information Science and Technology, 58(14), 2240–2254. doi:10.1002/asi.20695. Kettl, D. F. (2006). Managing boundaries in American administration: The collaboration imperative. Public Administration Review, 10–19 (Special Issue, December). Klievink, B., & Janssen, M. (2008). Stage models for creating joined-up government: From local to nation-wide integration. The 9th annual international digital government research conference (Montreal, Canada). Klievink, B., & Janssen, M. (2009). Realizing joined-up government — Dynamic capabilities and stage models for transformation. Government Information Quarterly, 26(2), 275–284. Landsbergen, D. J., & Wolken, G. J. (2001). Realizing the promise: Government information systems and the fourth generation of information technology. Public Administration Review, 61(2), 206–220.

Layne, K., & Lee, J. (2001). Developing fully functional e-Government: A four stage model. Government Information Quarterly, 18(2), 122–136. Luna-Reyes, L. F., Gil-Garcia, J. R., & Cruz, C. B. (2007). Collaborative digital government in Mexico: Some lessons from federal Web-based interorganizational information integration initiatives. Government Information Quarterly, 24(4), 808–826. doi: 10.1016/j.giq.2007.04.003. Moon, M. J. (2002). The evolution of e-Government among municipalities: Rhetoric or reality? Public Administration Review, 62(4), 424–433. Navarrete, C., Gil-Garcia, J. R., Mellouli, S., Pardo, T. A., & Scholl, J. (2010). Multinational e-Government collaboration, information sharing, and interoperability: An integrative model. Paper presented at the the 43rd Hawaii international conference on system sciences (HICSS), Honolulu, HI, USA. Orlikowski, W. J. (2002). Knowing in practice: Enacting a collective capability in distributed organizing. Organization Science, 13(3), 249–273. Pardo, T. A., Cresswell, A. M., Dawes, S. S., & Burke, G. B. (2004). Modeling the social and technical processes of interorganizational information integration. Paper presented at the Hawaii international conference on system sciences (HICSS-37), Hawaii. Pardo, T. A., & Tayi, G. K. (2007). Interorganizational information integration: A key enabler for digital government. Government Information Quarterly, 24(4), 691–715. Schooley, B. L., & Horan, T. A. (2007). Towards end-to-end government performance management: Case study of interorganizational information integration in emergency medical services (EMS). Government Information Quarterly, 24(4), 755–784. Siau, K., & Long, Y. (2005). Synthesizing e-Government stage models — A metasynthesis based on meta-ethnography approach. Industrial Management & Data Systems, 105(3–4), 443–458. Strauss, A., & Corbin, J. (1998). Basics of qualitative research: Techniques and procedures for developing grounded theory (2nd ed.). Thousand Oaks, CA: Sage Publications. Tsai, W. (2002). Social structure of "coopetition" within a multiunit organization: Coordination, competition, and intraorganizational knowledge sharing. Organization Science, 13(2), 179–190. Wengraf, T. (2001). Qualitative research interviewing: Biographic narrative and semistructured methods. Thousand Oaks, CA: Sage Publication. Willem, A., & Buelens, M. (2007). Knowledge sharing in public sector organizations: The effect of organizational characteristics on interdepartmental knowledge sharing. Journal of Public Administration Research and Theory, 17(4), 581–606. Williams, C. B., Dias, M., Fedorowicz, J., Jacobson, D., Vilvovsky, S., Sawyer, S., & Tyworth, M. (2009). The formation of inter-organizational information sharing networks in public safety: Cartographic insights on rational choice and institutional explanations. (Article). Information Polity: The International Journal of Government & Democracy in the Information Age, 14(1/2), 13–29. doi:10.3233/ip-20090170. Yang, T. -M., & Maxwell, T. A. (2011). Information-sharing in public organizations — A literature review of interpersonal, intra-organizational and inter-organizational success factors. Government Information Quarterly, 28(2), 164–175. Zhang, J., & Dawes, S. S. (2006). Expectations and perceptions of benefits, barriers, and success in public sector knowledge networks. Public Performance & Management Review, 29(4), 433–466. Zheng, L., Jiang, Y., Yang, T. -M., & Pardo, T. A. (2008). Sharing information for product quality and food safety in China: Barriers and enablers. In T. Janowski, & T. A. Pardo (Eds.), Proceedings of the 2nd international conference on theory and practice of electronic governance, Vol. 351. (pp. 90–97)Cairo, Egypt: ACM International Conference Proceeding Series. Zheng, L., Yang, T. -M., Pardo, T. A., & Jiang, Y. (2009). Understanding the "boundary" in information sharing and integration. Big Island, Hawaii, USA: IEEE Computer Society. Proceedings of the 42nd Hawaii International Conference on System Sciences (HICSS 2009), 1–10 e-Government Track.

Tung-Mou Yang is an Assistant Professor in the Department of Library and Information Science at National Taiwan University, Taiwan. Tung-Mou Yang received his Ph.D. in Information Science from the University at Albany, SUNY, in 2011. His research interests include e-Government, cross-boundary information sharing, information management, and information systems. He is also interested in digital divide and other information-related sociotechnical systems.

Lei Zheng is an Assistant Professor in the School of International Relations and Public Affairs at Fudan University, China. Lei Zheng received his Ph.D. in Public Administration and Policy at the University at Albany, SUNY, in 2009. His research focuses on crossboundary information sharing, transnational information collaboration, government information disclosure, and comparative digital government research.

Theresa A. Pardo is Director of the Center for Technology in Government located at the University at Albany, SUNY. She is also a faculty member in Public Administration and Policy and Informatics at the University. Theresa has directed numerous applied research projects with government partners and written articles, research reports, book chapters and case studies focusing on IT innovation in government, crossboundary information sharing and integration, trust and knowledge sharing, preservation of government digital records, and XML.