A thesis submitted in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Political Science

Chapter 5: teh reform of the state within the re-democratiosation context. 5.1. Introduction 5.2. Gaining terrain within the political transition 5.3. Policy making within the context of the unexécted declide of SEDAP-PR 5.4. The coup de grace: Collor de Mello's downsizing of the federal executive 5.5. Conclusion Chapter 6. Public management policy change in Brazil: 1995 - 1998. 6.1. Introduction 6.2. A sprint at the start: climbing the agenda and formulating a proposal in time record 6.4. Implementation during negotiation 6.5. Heading towards an anticlimax 6.6. Conclusion Part Three. Chapter 7: Comparing brazilian public management constitutional reforms. 7.1. Overview 7.2. The public management issue dynamics 7.3. Delaing with attention, ambiguity, and chance 7.4. Steering in turbulent waters: comparing the unfolding of the political stream in the four cases 7.5. Comparing the way issue definition was processed 7.6. Comparing the alternative specification processes 7.7. Comparing how enterpreneurs performed their functions 7.8. Comparing the decision-making process in the four episodes 7.9. Comparing the policy out comes 7.10. Conclusion Chapter 8: conclusion. 8.1. Introduction 8.2. The cumulative consequences of institutional layering for policy change 8.3. The complex nuances of constitutional change 8.4. The decisive but limited role of entrepreneurs 8.5. Final remarks

The Brazilian state apparatus is a patchwork built upon a cascade of reform attemts over a span of sixty years. Layers of legislation have produced a complex and inconsistent institutional landscape that has made policy change a hard task. Four major public management policy reforms resultedin significant concstitucional changes. Those reforms produced outcomes in the form of Constitutional provisions and they reshaped the modus operandi of the federal administration. The approach adopted is comparative because it is oriented towards identifying similat]rities and differences among the four cases selected. The pattern of the Brazilian cases is intriguinding because the frequency, variety, and speed of the reforms followed a zigzag course. They do not fit in with the long term, gradual, and cumulative transformations the look olace in many developed countris. Each case study corresponds to a historical episode. The material is organized and analycatically narrated around inconnected events within the episode. Those events were significant un themselves and the clearly hoghlighted aspects os the means-ends chains. The framework adopted for the analysis was developed by Kingdon to explain pre-decisional and decision-making stages of the policy-making process. The reason in the Brazilian context, characterized by fluidity, unclear technologiesm chance, temporal sorting, and attention allocation. Policy change took place when events in the political, problem and policy streams coverged to generate windows of oppotunity, during wich public entrepreneurs puched theis policy colutions through top decision-making arenas. The investigation is centred on the dynamics of the policy processes. The main conclusion of the thesis refers to the essential role performed by entrepreneurs in getting the reforms approved. The influence of the reformers in establiching the default implementation model varied in the four cases because of problems of political feasibility, reformers', rotation, policity discontinuties, transitional legislation inconsistencies, and lack of top down political support.