To briefly revisit the argument, the state and state-centric factors, or vari­ ables, remain primary. When identifying these state-centric factors use was made of various ideas and concepts from Chinese political economy literature of the past. The most useful of these was the Fragmented Authoritarianism model developed in the mid-1980s by Lieberthal and Oksenberg to explain the policy process in China's energy sector. It provided a means of examin­ ing state authority more precisely by looking at the roles of central and local state leaders, the bureaucracy, and how they interact in determining out­ comes through the policy process. Specifically, the model showed how those top leaders at the apex of the state system can exert a high degree of author­ itarian and decisive control over the policy process. However, these authori­ tarian features are counterbalanced by an institutionalized form of structural bureaucratic fragmentation that takes the form of functional vertical (t/oo) bureaucratic systems, or xitong(s), linking the central government to grass­ roots productive units, and territorial horizontal (kuai) systems linking units to local state government bodies. These different systems overlap and often work poorly with each other, and can generally form a major obstacle to the

realisation central decision makers' policy objectives. This tends to blunt the authoritarian elements of the state system and gives rise to a more protract­ ed policy process in which outcomes often differ from policy intentions.