“…The empirical literature, which is reviewed in detail in the next section, has adopted a range of different approaches when testing for the presence of welfare gains under decentralization and, in almost all instances, the empirical evidence is favorable to decentralized institutional arrangements. Some papers have focused on whether the outcomes of certain public policies (or indeed general subjective well-being) improve when provision is decentralized (Hoxby, 2000;Treisman, 2002;Cantarero and Pascual, 2008;Bjørnskov, Dreher and Fischer, 2008;Kyriacou and Roca-Sagalés, 2011;Díaz-Serrano and Rodríguez-Pose, 2012;Falch and Fischer, 2012;or Caldeira, Foucault, Rota-Graziosi, 2012 and2015); others on whether decentralized public policies are more responsive to expenditure needs (Faguet, 2004;Esteller-Moré and Solé-Ollé, 2005;Borge, Brueckner and Rattsø, 2014); others have examined whether politicians have better knowledge about voter preferences in small jurisdictions (Dahlberg, Mörk and Ågren, 2005); while a final strand of the empirical literature has turned its attention to the question of whether public expenditure composition varies as a consequence of decentralization (González-Alegre, 2010; Arze del Granado, Martínez-Vazquez and McNab, 2012;and again, Borge, Brueckner and Rattsø, 2014). The approach followed by most of these papers is general enough to accommodate welfare gains due to greater levels of allocative efficiency and of productive efficiency.…”