Analytical Framework

Throughout the reviewed literature, it seems that relatively little of it attempts to assess the role of impact science in relation to the decision-making surrounding the Lancang cascade development. As described by Magee (2005), the “existence of scientific research does not ensure that the results will have any bearing on hydropower development, planning, or implementation…”. That said, however, the increasing number of third-party groups (e.g. academic institutions and social organizations) and

government watchdogs (e.g. the Ministry of Environmental Protection) attempting to influence hydropower decision making in China (see Magee 2005, 2006, 2011; Mertha 2008) suggests that the role of science is, at the very least, to legitimize, delegitimize, critique, support, or contextualize development.

Darrin Magee (2005, 2006a, 2006b, 2011) has touched upon the role of science in hydropower development in China, primarily through an examination of the hydropower development on the Nu and Upper Mekong (Lancang) rivers in Yunnan and the lens of development priorities within China. In Magee (2005), he contends that in debates over development of major rivers, typically socioeconomic and political concerns often weigh more heavily than that of the traditional “hard” water sciences of hydrology, morphology, aquatic biology, and so on. Magee goes on to argue that both the “hard” sciences

  5  

mentioned above as well as social science should both be examined together in

comprehensive, interdisciplinary, cooperative studies. Magee (2006a, b), examined the role of energy companies in hydropower project implementation and motivation. While expounding on the complexity of the decision-making apparatus, Magee found that the energy needs of southeast Asia and China (e.g. Guangzhou) coupled with the competition between the Huaneng and Huadian power companies to get approval for mega-projects in an atmosphere of state and quasi-state actors, creates a general push for hydropower development. In addition, Magee (2006a,b) uses geographical scale as analytical device, exploring the notion of the politics of scale justifying Yunnan’s part of the “the west” in China used to justify both regional integration and justifying government policies and needs.

“Just as studying erosion in the Mekong delta requires attention to hydrologic and socioeconomic processes throughout the entire Lancang-Mekong watershed, so does understanding economic development in Guangdong necessitate consideration of one of the primary driving forces behind that development, namely electric provision” (Magee 2006b, 41).

Magee (2011) also examines the role of other motivating forces behind dam development – domestic energy needs, regional integration, poverty alleviation, low-carbon

development, etc. – and China’s interaction with the region. However, it only briefly touches upon the use of science within hydropower development. All of Magee’s abovementioned studies offer significant insight into the frameworks surrounding hydropower development in the Upper Mekong, and are essential to understanding the context within which downstream impact science exists. This will be explored more in-depth in chapter 3.

Brown et al (2009), Tullos 2009, and Tullos et al (2010) all examine perception of impacts from large-scale dam construction among stakeholders. Brown et al (2009) and Tullos et al (2010) found that among government and hydropower industry

representatives (specifically in Yunnan in Tullos et al’s study), large-scale dam benefits were seen to outweigh the impacts of downstream impacts while also perceiving impacts

  6  

as of lesser magnitude. This is opposed to academics and NGO representatives who decidedly viewed large-scale impacts as negative, with the negatives outweighing the positives, and viewing impacts as having greater influence than others. Tullos (2009) examined the use of environmental impact assessments (EIAs), and its success (or lackthereof) of mitigating negative impacts of large projects like the Three Gorges dam.

The uncertainty of impacts in scientific assessments can greatly impact mititgation measures and the extent to which the EIA is included in the development and

management process. Generally uncertainty is riven by different time and spatial scales, a failure to consider interdisciplinary links (Brown et al 2009), and the cumulative impacts of dams, espcially if they’re in a cascade (Tullos 2009).

Of the literature reviewed, I found very little English-language research, outside the abovementioned work, touching upon the use and role of science in the development of the Upper Mekong. More specifically, while the perception of impacts through an examination of scientifically backed downstream impacts was examined in Brown et al (2009) and Tullos et al (2010), I did not find any literature examining the specific scientific evidence that government and hydropower representatives were using when describing downstream impacts, nor the potential reasons for its use. This research seeks to contribute to the limited English-language literature on the role of science in decision-making on China’s development of the Upper Mekong, and bridge the perceived gap between the literature addressing downstream dam impacts (typically scientific papers), the use of science in decision-making, and potentially competing scientific frameworks that are available to decision-makers in China. To address this, Kingdon’s “multiple streams” (MS) framework will provide an overarching analytical framework for the information gathered in this research, with important contributions to the framework from the concepts of Haas’ “epistemic communities” and Mertha’s contribution to the well-known “fragmented authoritarianism” framework.

Kingdon’s MS model seeks to explain why certain policies are ultimately adopted or fail. The framework identifies two sets of factors – participants and processes – that impact agenda setting and alternative specifications to a given issue under governmental consideration. Participant factors consider actors inside or outside the government, their level of visibility, and their influence. Influence of participants is dependent upon

  7  

whether they’re “visible”, i.e. those who receive a lot of press and public attention, which tend to have more control on setting the agenda and less control over alternatives.

“Hidden” participants (e.g. interest groups, researchers, civil servants, epistemic communities, etc.), on the other hand, are more effective in specifying alternatives.

Kingdon also raised the concept of “policy entrepreneurs”, also used under Mertha’s model (discussed below). Policy entrepreneurs are advocates for policy proposals or the prominence of an idea. John Kingdon describes them as follows:

"These entrepreneurs …could be in and out of government, in elected or appointed positions, in interest groups or research organizations. But their defying characteristic…is their willingness to invest their resources - time, energy, reputation, and sometimes money - in the hope of a future return…[including] in the form of policies of which they approve." (Kingdon 2002, 122)

These policy entrepreneurs often bide their time until chance opportunities arise,

indicating they may not simply be providing a solution in response to an issue but rather have already formulated a solution and have been waiting for a particularly salient issue to arise (Kingdon 2002; Mertha 2008). This framework was originally conceived to address the United State’s federal system, and the two most important actors affecting agenda setting were also the most visible: the administration and Congress.

This raises the concern of adapting an analytical framework from its original democratic context to the authoritarian regime in China. Kingdon’s multiple streams framework originally sought to understand the changes in policies in the US congress from the efforts of policy entrepreneurs. In a democratic context, these actors are decidedly derived heavily from both governmental and non-governmental actors, with non-governmental forces having much more say in the policy process than in an authoritarian regime. In China, however, civil society and non-governmental

organizations (NGOs) work more in a supplementary role, than a typically adversarial role in western nations. The key adaptation to the multiple streams framework for authoritarian regimes, as this research intends to show, is to better understand those entities with which decision-makers in China consult with. Therefore, while in recent

  8  

times civil society has become much more ingrained in the policy process in China, in an authoritarian regime their role remains relative to the support of bureaucratic bodies within China’s government, and increasingly public opinion (if only to a certain extent).

As such, within an authoritarian regime the use and understanding of science seems more dependent on those bodies with which the government consults, in this research described as epistemic communities. Kingdon’s framework differs in the context in China with the use of focusing events. Hydropower has been an accepted, dominant framework within China as a means of continued developed. As such, many of the driving policy streams behind hydropower development on the Upper Mekong have not faced many opportunity windows that changed the course of viewing hydropower as a key development tool.

Rather, it seems that many of the events have not shifted the course of streams pushing for hydropower development, but have only further solidified hydropower’s position in the government’s framework surrounding continued development. This will be discussed more in-depth in chapters three and four.

In addition, in Kingdon’s MS model, policy entrepreneurs typically follow two basic principles when proposing a new policy idea. Their ideas are selected only when they are “technically feasible” and “value acceptable” (Zhu 2008). “Technically feasible”

means that actors supporting a proposal must delve into details and technicalities, and gradually eliminate inconsistencies and specify mechanisms (Kingdon 1995; Zhu 2008).

“Value acceptable” means that the proposal is both acceptable to the policy entrepreneur themselves, and the intended target (Zhu 2008). These two concepts, however, are admittedly ambiguous and difficult to determine. “Technical feasibility” is “a

multidimensional notion consisting of legal, administrative, financial and technological feasibilities, among others” (Zhu 2008, 317), which remains difficult to prove by the sheer number of factors needing to be tested. “Value acceptability” could consist of political views, ideology, political or national culture, and so on, the characteristics of which are not elaborated upon in Kingdon’s work. In the context of an authoritarian Chinese government, policy entrepreneurs must function very differently to their Western counterparts. First, the doctrine of the Chinese Communist Party (CCP) is the most important and dominant element in China’s political system. Singular party rule equates to no technical distribution of power among the highest state agencies as there would be

  9  

in say the United States. Rather, leaders from the government, legislative, and other consultative agencies discuss and decided upon courses of action for various future policy endeavors, while also sometimes making direct policies to guide governmental affairs. As such, the decisions of the CCP affect the entire bureaucracy and the decision-making systems within it. This singular party means that regular periodical transfer of political power does not occur as in would in Western politics. Rather, change over occurs in shifts in senior leadership positions like in the State Council, thus making the “politics stream” (discussed below) relatively stable (Zhu 2008). Second, under the cadre personnel system, government representatives, both in the party and government, are selected and appointed by their supervisors and predecessors (Lieberthal 1995; Zhu 2008). Therefore, government officials are often relatively insulated from policy actors outside the government; they typically not held accountable to social policy actors or the public, but rather their superiors within the government (Zhu 2008).

As this research will show, the trusted actors within the government and the trusted groups with which the government consults with under situations of uncertainty are vital to their understanding of given policy decisions. In addition, it is not necessarily problematic if officials fail to initiate policy changes and analysis when ordinary

problems occur, even if these situations open a “policy window” (Zhu 2008). However, officials are more likely to pay attention in to problems or change their perspective if in a given case they must deal with powerful or persuasive external forces powerful enough to call into question officials’ positions (i.e. higher level officials or bodies in the

government hierarchy, an increasingly commercialized media, etc.) (Zhu 2008). This is an important factor, as policy entrepreneurs and officials are notably averse to taking any amount of risk lest they bring about reprisal from higher officials. As such, actors and bureaucracies rarely take on the role of a “policy entrepreneur” as Kingdon defines it in China. Rather, these groups arise either in times of certainty of legal backing and institutional mandates, like with SEPA in mid-2000s (discussed in chapter three and four), or in NGOs and other actors outside the government itself. Zhu (2008) argues that for policy entrepreneurs outside the government employ a “technical infeasibility model”

which contends that “third sector” policy entrepreneurs promote change by making politically acceptable, but technically infeasible policy proposals for change. By doing

  10  

this, these actors may shift the conversation and spur a change from decision-makers that can be done given what is technically feasible. In Kingdon’s MS model, technical

feasibility is a requirement for the success of a given policy entrepreneur’s efforts. In China, technical feasibility is no necessarily required.

“…in China today the policy entrepreneurs’ “technically infeasible strategy” serves to maximize the likelihood that the policy proposal will gain leadership attention, through the attention of media and public – where a “technically feasible strategy” may leave the entrepreneurs’ proposal lost in the lacunae of bureaucracy” (Zhu 2008, 331).

As touched upon above, participants in the MS framework interact under the context of the three process factors, consisting of the streams themselves: the problem stream, policy stream, and politics stream. The problem stream examines how and why certain problems gain attention from the government. Actors compete to define problems within this stream according to their goals, values, and information (Stone 2002; Turin 2012). Systemic indicators such as “focusing events” (e.g. disasters or crises) and feedback contribute to shaping how problems are framed and brought to the attention of the government (Kingdon 2002). The policies stream, consists of various alternative policy solutions, proposals, strategies, and initiatives to a given problem, which must meet a “criteria for survival” within the “primeval soup” of policy discourse (Kingdon 2002; Zhu 2008). These criteria – technical feasibility, value acceptability, applicability to a prominent political campaigns, and anticipation of future constraints, among others – determine its likelihood for diffusing among decision makers. The result is what Kingdon describes as a “short list” of ideas upon which general consensus is founded (Turin 2012). The politics stream consists of macro-political conditions, elements that create an environment either conducive or hostile to certain types of ideas: a change over in the State Council, a prominent new political campaign, interest group pressure, energy demand, local resistance, and so on. It is within this steam that visible actors (e.g. the central and provincial government, the NDRC, etc.) may have strong impacts over policy agendas, while “organized forces” (e.g. interest groups, hydropower companies, etc.) maintain important influences over the alternatives considered.

  11  

Kingdon theorized that major changes in policy are possible after a convergence of all three streams to create a “policy window” or “window of opportunity”. Policy windows “exist when some consensus has been reached that government should address a certain problem, a suitable alternative is available and matched with the problem, and the political climate is conducive to change” (Turin 2012). Meaning that policy entrepreneurs and related actors then have the opportunity to get a solution, pet policy, or get attention for an issue into the mainstream decision-making network. According to Kingdon,

typically these policy entrepreneurs facilitate large policy changes by taking advantage of policy windows. Policy entrepreneurs prioritize the solution or issue based on its

likelihood to gain traction (Kingdon 2002). Typically these openings occur during a change in the political stream (e.g. a change over in the State Council, a prominent new political campaign, etc.) or a new problem captures the attention of government officials.

These opportunities “close” when the problems are deemed “solved”, acted upon, or a new issue arises (Kingdon 2002). While both “problem” and “policy” streams often already exist on the governmental agenda, a convergence all three streams offers a better guarantee of a given issue to exist within the “decision agenda” of a government

(Kingdon 2002). This, in turn, allows for and creates policy changes on a given issue or existing policy.

Under Kingdon’s framework, these streams typically exist separately from one another, needing to come together under certain situations to create a “policy window”.

Within the context of China, however, a number of scholars have found that they are, in fact, very much intertwined (Bi 2007; Zhu 2008). As is also found in this research, Bi (2007) and Zhu (2008) found that the problem stream and politics streams build off one another as the factors conducive to or hostile to new ideas are directly involved with the its relation to the CCP, the stability of the government, and prominent initiatives within the government. Therefore, the problem stream – how and why problems are deemed relevant – is tied directly to the politics stream, as they exist in reference to decision-makers positions. The policy stream then follows from what is both feasible and acceptable among decision-makers and officials generally unwilling to engage in risky policy changes. As mentioned above, due to the structure of the government, decision-makers are also generally isolated from differing policy streams because of their

  12  

answering to superiors within the government rather than outside actors (Bi 2007; Zhu 2008). As such, as this research will argue, those trusted groups with which they consult with on policy streams with certain levels of uncertainty (i.e. downstream impact science) could potentially have major impacts on how the policy stream changes.

To summarize, I believe the Kingdon MS framework can effectively be adapted to the authoritarian China context through an understanding that the three streams are very much intertwined. In addition, given the bureaucratic structure of China’s government restricting the courses of actions of policy entrepreneurs both within and outside the government, their relationships with policy change dynamics are significantly different than their Western counterparts, as described above. However, given these modifications, the MS framework seems to offer good insight into both policy change and their related actors within China.

The next step is to establish the broader political system within which these policy streams exist. To do this, this study uses Lieberthal and Oskenberg’s fragmented

authoritarianism (FA) framework with Mertha’s (2008) additions as a basic construct of understanding of China’s governmental system. Briefly, the FA framework describes Chinese politics as a system in which the policies made at the center become increasingly malleable to the parochial organizational and political goals of various agencies and regions charged with enforcing the policy. Therefore the policy outcomes of a given policy are the result of the incorporation of the interests of the implementation agencies into the policy itself. The model originally illustrated the segmentation of authority among the functions of administrative organs in order to explain economic policy change, exploring the fundamental change to China’s bureaucracy brought from economic

decentralization. As such, the negotiations among bureaucratic organs factored significantly in policy-making. Lieberthal and Oksenberg’s FA model asserts that authorities are divided into a number of “clusters” based on policy interests and bureaucratic mandates. Therefore, the achievement and/or retention of their policy interests through consensus seeking becomes the core rationale in decision-making in China (Lieberthal and Oksenberg 1988).

The governmental organs are organized by “vertical” lines (條, tiao) and

“horizontal” lines (塊, kuai) of hierarchy. Vertical lines gather bureaucratic organs along

  13  

with similar functions and jobs; horizontal lines refer to bureaus at the same hierarchical level, but with different functions in a sort of non-binding professional relationship (Lieberthal 1995). Economic reform and decentralization stressed the “kuai” hierarchy, thus making decisions be achieved through “an enormous amount of discussion and bargaining among officials to bring the right people on board” (Lieberthal 1995, 173). As such, FA would provide opportunities for actors to halt or delay important policies from

with similar functions and jobs; horizontal lines refer to bureaus at the same hierarchical level, but with different functions in a sort of non-binding professional relationship (Lieberthal 1995). Economic reform and decentralization stressed the “kuai” hierarchy, thus making decisions be achieved through “an enormous amount of discussion and bargaining among officials to bring the right people on board” (Lieberthal 1995, 173). As such, FA would provide opportunities for actors to halt or delay important policies from