Abstract
This study focuses on regional networks of interprovincial agreements (IPAs) emerged at China’s Pan Pearl River Delta (PPRD). Coparticipation of eleven PPRD members in 192 IPAs is investigated. It examines the extent to which coparticipation in the seven policy areas by the PPRD members are correlated. It measures density, the degree to which PPRD members coparticipate in IPAs. Although regional economic development was the initial focus, more IPA coparticipation has been devoted to addressing the noneconomic policy concerns. The results of the study are relevant for understanding the complex patterns of regional collaboration in which subnational governments participate.
Introduction
Local governments around the world are increasingly facing cross-boundary issues such as air pollution. These issues are of regional significance if the decision by one government in one or more functional areas has significant influence on other governments and other functional areas (Feoick 2013). Instead of resorting to the central government for solutions, local governments may collaborate with each other in a number of different ways to solve the cross-boundary policy issues. The number of the regional collaborative initiatives has increased globally (Savitch, Gross, and Ye 2014).
This article examines a specific collaboration: the networks of interprovincial agreements (IPAs) emerged in China’s Pan Pearl River Delta (PPRD) area (Chen, Ma, and Suo 2015). IPAs include different types of agreements, such as joint planning, joint policy initiatives and declaration, or shared programs, and contracts with two or more participating. IPAs enable multiple provinces to pool their resources for addressing shared problems, promoting common goals, and coordinating policies. These IPAs can be considered a form of regional networked governance (Chen, Ma, and Suo 2015; Hu and Ma 2011; Thurmaier and Wood 2002).
Intergovernmental agreements have been used in the United States for many years (Chen and Thurmaier 2009; Blair and Janousek 2013). Intergovernmental collaboration has also become a worldwide issue (Tavares and Camões 2007; Park 2015; Carr 2015). The IPAs that are the focus of this research not only showcase how China’s provincial governments adapt to a globalized world but also provide important lessons to other countries seeking governance tools to address regional issues.
This article is organized as follows: the first section includes a brief review of the literature related to regional governance. The second section examines regional governance in China and the PPRD as a case study of IPA networks. The third section uses social network analyses to examine three aspects of the IPA networks. First, a network correlation analysis ascertains the extent to which different IPA networks correlate with each other. Second, a comparison of network density scores shows the extent to which PPRD members are collaborating in each IPA network. Third, among PPRD members, are some more connected or more influential than others? A core–periphery analysis is performed to address this question. The final section discusses the implications of these research findings for regional networked governance and its limitations and need for future research.
Horizontal Regional Governance
The literature suggests a shift to horizontal regional governance in the United States and China. There is extensive literature in the United States that studies vertical intergovernmental relations (IGRs; Oates 1999; Colan and Posner 2008). The literature on horizontal IGR, however, suggests three streams: centralist, polycentrist, and regionalist (Wallis 1996). All three are concerned about identifying regional governance structure that is instrumental to social and economic development (Nelson and Foster 1999). The centralists favor a top-down approach of consolidating small, multiple, and fragmented local governments. The polycentrists prefer a system of decentralized and overlapping governments (Leland and Thurmaier 2000). The interjurisdictional competition forces local governments to structure themselves in such a way that their bundle of services/taxes matches the preferences organizations and people they serve (Ostrom, Tiebout, and Warren 1961; Tiebout 1956). The primary concern of regionalists is the role of collaborative governance mechanism in overcoming the problems of collaboration (Feiock and Scholz 2010). They focus on identifying more flexible, voluntary, and horizontally linked arrangements to both make joint decisions and deliver services (Savitch and Vogel 2000).
There are many different forms of collaborative arrangements such as voluntary memorandums of understanding, mutual aid agreements, and joint planning agreements to more legally binding mutual contracts, from relatively simple joint service agreements to more complex, ongoing interactions involving multiple governments (Blair and Janousek 2013). Three major interstate collaborative mechanisms—interstate compacts, multistate legal actions, and uniform state—are embedded in a “Horizontal Federalism” (Bowman 2004).
One approach to understanding these many different forms of collaboration is called “complex networks” in which a large number of independent governments voluntarily cooperating through multiple, overlapping webs of interlocal agreements (Thurmaier and Wood 2002). Such range in the scope and depth of intergovernmental collaboration enables researchers to apply a network approach to studying regional governance (Leroux and Carr 2010). When individual jurisdictions engage in multiple policy and service relationships simultaneously, they develop a multiplexed network (Bae and Feiock 2012; Shrestha and Feoick 2009).
Regional Governance in China and the PPRD
The initial governance efforts of the People’s Republic of China for regional governance were hierarchical. This approach has limited success (Solinger 1977). During the postreform period from 1978, however, the focus changed to administrative and fiscal devolution to sustain the economic reforms (Shirk 1993), as a model of “Market-Preserving Federalism” (Qian and Weingast 1996). The central government encouraged local governments to undertake reforms, generate more revenues, and allocate resources to meet their own needs. At the horizontal level, the reform unleashed the entrepreneurial spirit of local governments, fueling the rapid growth of local economies, a market-oriented scenario that would be favored by polycentrists. Yet, just like centralists in the United States, opponents pointed out this produced “race-to-the-bottom” competition between local governments and encouraged local officials to protect their own markets by erecting administrative blockades (Chen 2004). While central authorities advocated for coordinating the national economy as a common market, local authorities in pursuit of revenue were dividing it up.
In response to administrative fragmentation and excessive competition, China’s local governments, like regionalists in the United States, sought more collaborative approaches to regional governance (Shi 2008). Ultimately, four major regional cooperative zones formed in China: the Beijing–Tianjin–Hebei metropolitan region, the Yangtze River Delta, the PPRD, and the mid-China region (Ye 2009). Accordingly, these four regions were created to solve the policy problems facing China that require regional collaboration—environmental protection, infrastructure, economic development, and public crises and emergencies (Ma 2010; Yan and Wang 2013).
The number of China’s intergovernmental agreements also grew rapidly. In the PPRD area, the number of IPAs participated by provinces started at four in 2003 but ultimately averaged eighteen new agreements annually (Ma, Suo, and Chen 2014). There are many forms of IPAs including vertical agreements signed either between central and local governments or between provincial and municipal governments, horizontal agreements between governments in the same administrative levels, and cross-jurisdictional agreements made between governments at the different administrative levels (Yang 2011). The IPAs in this study are horizontal agreements between provincial governments.
The PPRD includes nine mainland provinces of Fujian, Jiangxi, Hunan, Guangdong, Guangxi, Hainan, Sichuan, Guizhou, and Yunnan and two special administrative regions (SARs) of Hong Kong and Macau. The PPRD members span across China’s four geographic areas—eastern, southern, central, and southwestern parts. Most PPRD members are contiguous and located in the Pearl River Basin. They embraces one-fifth of China’s total territory and 55 percent of the mainland coastline. In 2012, the PPRD carried a third of China’s population (480 million) and contributed 36 percent to China’s total gross domestic product. It is a very vibrant regional economy with an average growth rate of 10.8 percent, 0.7 percentage higher than the national average (China Statistic Year Book 2013).
The data for this research were obtained from www.pprd.org.cn, an official website cosponsored by PPRD members. The provincial governments in the PPRD created a total of 192 IPAs from 2003 to 2013. These IPAs are involved in seven different policy areas: environmental protection, tourism promotion, transportation infrastructure, science/technological/cultural exchange, migrant labor, public health, and trade development. Table 1 summarizes IPAs by the number of participants across seven policy areas.
Distribution of IPAs by Number of Agreements and Participating Provinces Across Seven Policy Areas.
Network Analyses of the IPAs
Connections across Policy Arenas
The policy issues of regional significance are not isolated from each other but interconnected (Feoick 2013). The effects of one policy can influence another policy. It is important to study the extent of externality between different regional policy networks. Are participating jurisdictions of regional collaboration fully aware of such interconnection? For example, environmental degradation is usually a consequence of excessive economic development. Does participation of economic development IPA network also result in the increased participation of environmental protection IPA network? A quadratic assignment procedure regression in UCINET, a software for social network analysis (Borgatti et al. 2002), is used to estimate the correlation coefficients across seven IPA networks. The results are presented in Table 2.
Quadratic Assignment Procedure Correlations Across Seven Interprovincial Agreement Networks.
***p < .01.
**p < .05.
*p < .1.
Interprovincial collaborations on the regional economic development of different policy areas are found to reinforce each other. Three IPA networks related to economic development, transportation infrastructure, tourism promotion, and trade development, are highly correlated with each other (r = .65, .56, and .48, p < .01), indicating a positive policy functional externality between them. On the one hand, better transportation infrastructure as a result of coordinated regional development is instrumental to the more exchanges on trade and tourism across jurisdictions. On the other hand, growing interjurisdictional tourism and trade require more collaboration on infrastructure planning. In 2005, Guangdong signed collaborative agreements with four provinces to conduct interprovincial highway planning. Collaboration related to regional development and infrastructure planning also extended to railway, sea, and river transportation in 2011–2013. Seventeen IPAs were signed to promote regional trade. In 2003, mainland provinces signed closer economic partnership arrangements with Hong Kong and Macau on June 29 and October 18, respectively. The provinces within the PPRD region offer many different attractions for visitors to explore. A number of bilateral and multilateral agreements were signed to promote and coordinate tourism industry development.
The results indicate that the growing regional economy attracts a massive flow of labor across jurisdictions. The IPA network of migrant labor is found to be significantly correlated with three networks on economic development (r = .24, p < .1; r = .37, p < .05; and r = .32, p < .05). Well-developed transportation infrastructure facilitates the free flow of goods, material, and personnel. The PPRD has always witnessed huge waves of migrant workers. Millions of workers from moderately and underdeveloped provinces, say Sichuan, migrated to the developed province of Guangdong for job opportunities. The demand of coordinating human resource policies increases. The policies pertaining to social security and workers’ rights need to be coordinated across provinces. In 2006, an agreement on protecting migrant workers’ rights was signed among labor unions in different provinces.
The results do show that provincial governments at PPRD are increasingly paying attention to the environmental issues as a consequence of the booming economy. The IPA network of environmental protection is highly associated with those of transportation, tourism, and trade (r = .84, p < .001; r = .63, p < .001; r = .54, p < .001). IPAs in the policy area of environmental protection include those designed to address the issues of clean air and water pollution within PPRD. Many mainland PPRD members are loosely tied together through the tributaries of the Pearl River. Protecting water resources across the administrative boundary became their common concern. In 2012, all the mainland PPRD member provinces formulated a guideline on dealing with disputes on cross-boundary water pollution.
Science, technology, and cultural exchanges also follow the increasing economic and social interactions. It is the IPA network of science, technology and culture that is significantly related to all the other networks. The IPAs promoting collaboration on scientific and technological development and cultural exchanges are, for example, “PPRD Regional Cooperation Agreement in Science and Technology” and “PPRD Regional Cooperation Agreement of the Information Industry.”
However, there is little connection between regional collaborations on public health and many other policy areas. The public health IPA network only significantly correlates with two networks (S&T culture and trade development). A frequent movement of people makes it increasingly difficult to manage a public health crisis. Infectious diseases, such as outbreak of severe acute respiratory syndrome in 2002, do not recognize the administrative and geographic boundaries. As a reflection of their common concerns, a number of IPAs on public health were signed by Guangdong, Hong Kong, and Macau from 2004 to 2013.
Network Density: Collaboration among All Eleven Provinces
The next analyses examine the extent to which the eleven provinces are linked by coparticipating in IPAs. The number of linkages is calculated as the number of coparticipation, that is, the bilateral participation of one province to another province for a particular IPA (Chen, Ma, and Suo 2015). The number of province to province linkages for the environmental protection is summarized in Table 3. The Guangdong province, for example, participated in thirty-eight of the total of forty-two environmental IPAs. Guangdong coparticipated in twenty-five of the IPAs with Guangxi, twenty-one with Guizhou, and so on. The total coparticipation of Guangdong in all environmental protection IPAs with all the other provinces is 184. The total coparticipation for all provinces with each other for environmental participation is 816. A similar table can be constructed for each policy area.
Matrix of Coparticipation of Core and Peripheral Provinces in Interprovincial Agreement Network for Environment Protection.
A measure of density for each policy area provides a measure of the coparticipation. The measure of density is calculated using the total bilateral coparticipation for each policy area (i.e., 816 for environmental protection) divided by the total possible bilateral connections for the eleven provinces-55 (the figure 55 is calculated at the total combination of eleven items taken two at a time). A density score for environmental participation of 14.8 is obtained by dividing 816 by 55. The larger the density score, the greater the number of bilateral connections, more shared participation of the agreements, and the more extensive the interprovincial collaboration. Table 4 is a comparison of network density values and number of agreements across seven IPA networks.
Density Scores of Coparticipation for Environmental Protection.
A comparison of two columns suggests that the ranking of density scores is not necessarily the same as that of IPA numbers. The larger number of IPAs does not indicate a wider scope of collaboration. The environment IPA network ranks the highest with a density score of 14.8, indicating that the agreements of environmental protection involves the most connected networking ties among eleven PPRD members. There are, however, also forty-two agreements on tourism promotion. Yet, its network tourism density only ranks fifth with a score of 11.3, and the average number of tourism agreements that the PPRD members jointly participated is around eleven. In terms of density (coparticipation), S&T and culture IPA network moved up to the second place. The migrant labor IPA network moved from the fifth on the number of agreements up to the third on the ranking of density. The IPA network on transportation ranks fourth. The rankings for trade and public health IPA networks on density remain the same with those of number of agreements.
Network Density: Collaboration by Core and Peripheral Provinces
Not all provinces participate equally in the IPA network, and some provinces may be much better connected than others in PPRD regional networks. The implications of the core–periphery structure for regional governance have been well explored (Andrew and Feiock 2010). This analysis identifies a core group of PPRD members who have high density of ties among themselves, and periphery who have lower density of ties among themselves. The core members are able to coordinate their actions while those in the periphery are not. Provinces in the core enjoy a structural advantage over those in the periphery.
Table 3 also provides us with information for a core–periphery analysis. The coparticipation of eleven provinces in environmental protection IPA network is partitioned into two groups of core and periphery provinces. The core group includes six provinces of Guangdong, Guangxi, Guizhou, Hunan, Jiangxi, and Yunnan, with a total coparticipation of 297. The peripheral group has five provinces of Sichuan, Fujian, Hainan, Hong Kong, and Macau, and its total coparticipation is 113. The measure of group density for the core and periphery groups of provinces is calculated using the same formula used to calculate density for all eleven provinces. The total bilateral coparticipations for each group (i.e., 297 for the core group in IPAs on environmental protection) divided by the total possible bilateral connections (i.e., fifteen for the core group of six provinces, the figure 15 is computed at the total combination of six items taken two at a time). A density score for the core group of 19.8 is obtained by dividing 297 by 15. In the same vein, for the five peripheral provinces, the total bilateral links are 10 (the total combination of five items taken two at a time). A density score for the peripheral group of 11.3 is obtained by dividing 113 by 10.
A graphic presentation of the core–periphery structural configuration of interprovincial network for environmental protection IPAs is displayed in Figure 1. A circle refers to a core member and a square a peripheral one. The sizes of lines are in proportion to the number of coparticipation. The six provinces of Guangdong, Guangxi, Guizhou, Hunan, Jiangxi, and Yunnan (circles) are found to form a core substructure in the center of the PPRD with a density score of 19.8. The five squares (Sichuan, Fujian, Hainan, Hong Kong, and Macau) are in the periphery with a density score of 11.3. On average, the core members have 8.5 more coparticipation than the peripheral members.
Environmental protection.
Discussion and Conclusion
Since the late 1990s, efforts to form regional networks have been proliferated in many parts of the world. This has led to a shift of scholarly and practical interest toward horizontal and networked regional governance. Less research, however, has been conducted on the regional governance from a network perspective. Social network analysis can be a powerful analytical tool in researching regional collaboration. Studying the network structure will identify the gaps in regional collaboration. The network analysis also can help individual participating jurisdictions have a better understanding of their roles and positions in regional networks.
This article contributes to the literature conceptually and empirically by applying a network approach to unpacking the complex patterns of interaction in regional networked governance. It studied the coparticipation of eleven provinces in seven IPA policy networks emerged in China’s PPRD. The interpretation of our results may be local. But their insights could be useful to other countries, as they may face the same challenges of collaboration in regional governance. For subnational governments in other countries, three lessons can be drawn from our study.
First, regional collaboration in one area may generate collaboration in other areas. Ultimately, regional collaborations will become comprehensive as collaborations in different areas reinforce each another. In this study, many IPA networks are highly correlated with other, demonstrating complex interactions among regional policy arenas. For example, the increasing coparticipation of IPA network on environmental protection concurs with the growing shared participation of IPA networks on economic development (tourism, transportation, and trade). The correlation among some policy areas may not occur in other policy areas. For example, the public health IPA network is not significantly correlated to environmental protection and migrant labor.
Second, formation of regional networks should take into account both economic and noneconomic policy issues. For many local governments around the world, in response to economic conglomeration, the priority of regional governance is on the economic development. Similarly, China’s local governments have been always criticized for diversion of scarce public resources to economic development and diverted attention away from environmental and social concerns in regional collaboration (Xu and Yeh 2011). In the case of PPRD, although the starting point is regional economic integration, it has evolved into a comprehensive regional collaboration with many other policy issues. Our network analysis of density illuminates that the PPRD members have the highest coparticipation of IPAs on environmental protection, S&T and culture, and migrant labor.
Third, to overcome the collaborative challenges, one or a few leading network members should assume the responsibilities of managing and sustaining regional networks, as being suggested by a core–periphery analysis. The visualization of the core–periphery network structures provides us with some clues on why some PPRD members have more IPA coparticipation than the others. The PPRD is intended to serve as an informal partnership in which its members participate on a voluntary basis. The provinces of Guangdong, Guangxi, Guizhou, Hunan, Jiangxi, and Yunnan, located in the PPRD’s heartland, neighboring immediately with each other, are found to have a very extensive shared participation in environmental IPA network. This is because geographic proximity is a key factor in fostering interjurisdictional collaboration for addressing the issues of environmental protection.
The findings of this study should be interpreted with one important caveat. The IPA network structural characteristics revealed in this study are to a large extent due to the PPRD’s local conditions. Although a network approach this study utilized can be replicated in other countries, they have to be contextualized. The future studies of other regional networks in other countries should take into account each case’s specific context. This article is more descriptive in its nature. The extent to which members collaborate in regional networks could be attributed to complex interaction of geographic, economic, cultural, and political factors. The future studies should go beyond simple descriptive network measures. For example, a network regression can be used to better explore why some pairs of provinces have more extensive IPA coparticipations than the other pairs.
Footnotes
Acknowledgments
We thank four reviewers for their constructive comments on the manuscript. We are extremely grateful to Michael Scicchitano’s editing for making this article in a much better shape.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research is supported by the National Natural Science Foundation of China (Project Number: 71003013 and 71303032).