The Vertical View

The formal description of the education system's bureaucratic actors portrays the focus of policy activity as being at the SEdC, albeit with active collaboration with or coordination of other ministries and commissions. But the educational policy process also involves much vertical movement. The State Council needs to approve major policy and, at times, mediate disagreements between participants. Provinces, counties, municipalities, and individual schools also are centrally involved. And there is substantial autonomy at the lower levels.

What is the relative power of these different levels within the educational system? Most generally, the central-government level (SEdC) is seen as having the greatest power, though provincial and local levels also have responsibilities that give them power over schools. It is the SEdC that determines broad policies, provides the outline for curriculum in precollegiate education, determines texts, and, to a large extent, runs higher education (through its control of the university entrance-examination system and the determination of academic majors). Provincial education authorities run the secondary education system (through their control of the secondary school entrance exams), provide some financing of precollegiate education, and control some aspects of the nonkey sector of higher education (its financing, student recruitment, and, in part, job assignments). Local governments provide the majority of funds for local elementary and secondary schools.

It should be noted also that the power of different levels of the education bureaucracy is complex because of overlap in the function of central, provincial, and local levels; regional variation; range in types of policy and kinds of resources that connote power; and the distinction between state-run and community-run (minban ) schools. Power over funding, for example, depends on the level and type of school (e.g., key or nonkey). Power over the curriculum is also somewhat influenced by more than one level and varies by place: though in the 1980s the SEdC controlled a nationally unified curriculum, provinces and localities had the authority to supplement that curriculum (and varied greatly in the degree to which they did so). In the late 1980s Shanghai used its own experimental curriculum in place of the national curriculum. Finally, depending on the type of education policy (e.g., curriculum, funding, administrative structure), even individual schools can exert a kind of veto power in their ability to obstruct, delay, or reinterpret policy.

In short, the pattern of power within the education system is complex. The pattern has also changed. Over the 1980s there occurred nationally an increase in the relative strength, autonomy, and influence of the local bureaucratic level and individual school units as greater responsibility for finance devolved to local levels and schools were encouraged to rely on entrepreneurial solutions to many operational problems. At the same time, in policies regarding the admissions systems, bureaucratic structures, rules, and standards of evaluation, educational reforms have tied schools more closely to unified plans or central authority. The result of this two-sided change is the increased autonomy of lower levels of the educational system within a more circumscribed boundary of action.

With respect to teacher policy, the patterns of fragmented power parallel those described above for education generally. Formally, the departments responsible for teachers at the central level (in the SEdC) have provincial counterparts in the Bureau of Education (BOE).^[5] These provincial offices have power over teacher policy through provincial administration of teacher education,^[6] through direct administration of provincially run secondary schools, and indirectly through the oversight of county and district education offices, where the bulk of elementary and secondary schooling is administered. Provinces, like the counties and

districts below them, play a noticeable role in interpreting and in effect shaping educational policy.

In this structure there is much delegation of discretionary authority. The Center decides that teacher competence must be assured, but municipalities choose the subjects in which to test their elementary teachers (ZGJYB , 10 November 1987, 1). The central government announces a 10 percent wage increase for teachers, but provinces and localities are authorized to find the money and make decisions about the actual allocations (ZGJYB , 3 December 1987, 1). And the former Ministry of Education announced that teacher education would be upgraded, but local institutions had to decide what that meant in curricular terms (Paine 1986).

Nonetheless, stopping with the observation that there is delegation of discretionary authority assigns too much rationality, foresight, and decisive clarity to the central education bureaucracy. Instead, closer examination of the process of educational policy formation suggests less rationality, more interaction, and iteration. Horizontal and vertical interactions produce a process and a set of policies that are at one time more responsive, vaguer, more heterogeneous, and slower than the delegation or discretion model suggests.

Two cases of policy formation—one regarding a regulatory issue, the other distributive in character, are analyzed below. While each illustrates the profoundly interactive nature of the process, the two differ in the locus of and approaches to authority and influence. The first concerns regulations for teacher competence. Through this case we see the grouping process, the powerful role of subordinates, and the variety of coping strategies invoked. The second case concerns efforts to improve the social, economic, and political situation of teachers. Here we observe the limits of the education sector and the subsequent need for cross-sector alliances and cooperation.