|tors have effective ideas (and the ability to implement them) that have been stymied by lack of control over the requisite policy levers.|
|•||We find some evidence that states use increased programmatic flexibility to develop pedagogical innovations in areas with high returns, such as preschool. The cases show that sub-national administrators are crafty at eking out more power and control than they are officially given. Through scrutinizing legal codes or simply gaining confidence to defy central norms, sub-national authorities seemed to take "decentralization" reforms, however limited de jure, as signals that they should do more, de facto. This is particularly true for regulation and policy development. The state-level SEP delegations (SCEPs) previously had little incentive to pursue such innovation, since they would have been violating norms established by their superiors in their own bureaucracy.|
|•||Well-targeted central compensatory resources and policies are needed to alleviate inequity. Sub-national administrators immediately lobbied central compensatory programs to effectively support their own innovations.|
|•||It is also clear that non-decentralized federal compensatory programs (such as CONAFE in this case) continue to have a key impact and supporting role in programs developed by the state. We find some evidence that states have used the decentralization legislation and their increased flexibility in program design to lobby and involve federal compensatory programs effectively. Part of this cooperation stems from the willingness of states to commit resources under their control and combine them with federal funds.|
Given these preliminary findings, we can tentatively assert that decentralizing the traditionally highly centralized educational bureaucracies in Latin America can free up sub-national officials to pursue reform strategies that can positively impact the problems of dropout and retention rates. Naturally, the ultimate success in this area is a long-term process and will depend on the precise nature of the decentralization reform and the specific powers actually transferred from the center. Nevertheless, the Mexican results allow for guarded optimism.
This chapter grew out of the research project for the Inter-American Development Bank (IDB) entitled "Decentralization and Recentralization: Lessons from the Social Sectors in Mexico and Central America." The research assistance of Norma Mogrovejo and Melissa O'Brien was invaluable. Comments from Michael Jacobs, Steve Doherty, and Bill Savedoff are gratefully acknowledged. The opinions expressed do not reflect those of the IDB or its board of directors. The author is solely responsible for all statements, interpretations, and errors.