Jean-Paul Faguet's latest book addresses one of the most substantive questions someone interested in politics must deal with: can democracy be simultaneously a responsive and efficient system of government? This superbly-written book dismisses any trade-off between the two elements, defending the position that genuine and extensive decentralisation, combined with favourable local conditions, can produce both the responsiveness and efficiency that citizens demand from democracy. In order to make the point, the book studies the decentralisation reform launched in Bolivia in 1994. Faguet properly places the reform in the immense literature on decentralisation and analyses it with an array of methods, from sophisticated econometric models to impressive qualitative research, that includes more than 300 interviews, conducted over more than a decade of research.

The book starts by highlighting the fact that the decentralisation process initiated in the 1994 Ley de Participación Popular (LPP) was a major disruption to a traditionally centralised state in its creation of more than 300 municipalities. Drawing on an impressive database covering the whole country, Faguet shows that the quality of local government improved immediately after the first election of mayors under the new LPP regime. Not only were huge resources directly allocated to new local governments but new authorities radically altered the way those resources were spent. Whereas the central state used to privilege spending on infrastructure and productive sectors, local governments overwhelmingly switched to human capital investments such as education, health and primary services, the sectors that citizens tend to rank as their highest priorities. That is, local government immediately became more responsive to the citizenry's needs. This is the most important empirical finding of the book which is statistically well-proven in Chapter 1 and confirmed with an econometric model in Chapter 4.

However, even if statistical and econometric analysis demonstrates that local governance generally improved in Bolivia after the LPP, this is not tantamount to saying that every local government's governance and resource management improved. As Faguet reminds us with lucid scepticism, the extensive literature on decentralisation shows that positive and negative cases coexist everywhere, with no conclusive evidence on what drives failure and success. In order to tackle this question the author proposes to study a positive and a failed case of good governance in Bolivia immediately after the LPP was adopted. Viacha, an important town in La Paz department near El Alto province, developed an inefficient, corrupt and irresponsive local government. The second case, Charagua, is a rural district in the eastern Chaco region that saw the emergence of a responsive, efficient and well-run government. This detailed qualitative contrasted comparison (almost one hundred pages) placed amid the decentralisation literature allows Faguet to infer a general theory of good decentralised local governance.

Faguet's central claim is that the effects of decentralisation actually depend less on the reform itself than on the social context in which it is implemented. Based on the premise that there are good and bad conditions for decentralisation, Faguet proposes a two-step, context-sensitive theory: 1) ‘Where local politics are nourished by a diverse, heterogeneous local economy and an active civil society rich in organised groups, political competition will tend to be open and substantive’; 2) ‘local government responsiveness and accountability are primarily the product of the openness and substantive competition of its politics’ (both quotations from p. 218). This model seems to explain quite well the contrasting outcomes of Charagua and Viacha in 1997.

Yet the proposed theory is more successful because of what it rejects than what it puts forward. Especially in the domain of public policy, but also for a long time in comparative politics, researchers abused arguments for institutional reforms as the main independent variables for political phenomena. Faguet rightly criticises those society-insensitive approaches and correctly highlights the fact that other variables, the economy, local politics, civil society organisation, might better explain political outcomes than a single institutional reform. In a nutshell, local context is crucial. However, once the institutional formal approach was defeated, the context-sensitive theory proposed in the book has its own flaws.

My first concern is that the idea that responsive and accountable governments are more likely to occur where the political competition is open and substantive does not seem particularly challenging. Actually, one could almost suggest some level of endogeneity among the factors. My second concern arises from the author's findings when he returns to Viacha and Charagua in 2009 to see how these cases developed since his account of 1997. While Charagua continued on its previous trajectory of responsible and efficient local government, Viacha's previously dismal situation had reversed and it had become a solid case of good local governance. It is unclear how the processes driving these two cases fit with the theory. Charagua follows the argument quite well: a dense and developed civil society brings open competition that fosters good governance. Yet, the Viacha case, ‘a remarkable transformation’ (p. 243), does not seem to properly follow the general explanation. Faguet's description suggests a responsible mayor is crucial for the governance improvement, but the general model does not provide room for voluntaristic individual-level accounts of political change. Moreover, one of the two large enterprises that dominated Viacha's economy in 1997 had left by 2009. Less economic heterogeneity coexisted with an improvement of governance, even though the theory predicted the opposite.

Actually, the main mechanism behind improved governance in Viacha seems to be a general learning process where every actor (politicians, electors, civil society organisations) gradually internalised and institutionalised the new decentralisation regime. Hence, the successful transformation of Viacha seems linked to practices launched by the institutional reform rather than by social conditions as Faguet persuasively had argued in the theoretical chapter. Viacha is a case where the LPP seems to have gradually produced good governance by transforming civil society, rather than a case where distinctive solid civil society drove the LPP into good governance.

My third concern results from the tension between a society/economy-driven theory and an empirical institutional-driven approach that reappears in the book's conclusions. Through the preceding chapters, the author has mainly argued that the success of decentralisation lies in contextual conditions rather than in the reform itself. Yet, suddenly, causality seems to reverse in the conclusions: ‘In Bolivia decentralization turned passive residents into engaged citizens …’ (p. 279); ‘Decentralization led to major changes …’ (p. 281); ‘The local dynamics that it [decentralization] set into motion proved virtuous, not only permitting good cases of local governance to improve but spurring distorted cases to heal themselves’ (p. 284). At this point the reader wonders where the explanation lies: is it in the peculiar Bolivian social conditions or in the virtuous formal reform that Sánchez de Lozada launched in 1994? The tension becomes clear when the author summarises the positive institutional aspects of the LPP that might be ‘replicated elsewhere’. As a result, readers closing the book wonder how the particularistic comparative politics argument turned into a universalistic public policy approach.