Studies of policy diffusion are grounded on the perspective that domestic policy decisions are affected by foreign choices. This influence, in turn, produces policy convergence (Braun 2006; Drori, Meyer, and Hwang 2006; Dobbin, Simmons, and Garrett 2007). Despite this common view, the mechanisms by which these approaches explain diffusion are quite different. Some theorists argue that since countries compete at a global scale to export their products and attract foreign investment, governments engage in similar policies in order to reduce the barriers for trade and exchange (Green 1999; Twining 2005). Other theorists employ the notion of coercion to describe …show more content…
Yet, any of these mechanisms seems to be producing the expected policy convergence. As Table 1 shows, private enrolment ratio on secondary education have risen in the regional average, but this increment varies substantially across countries. In addition, the policies employed to promote privatization are quite different. For instance, Chile encourages competition between private and public schools developing a quasi-market of education (Valenzuela, Bellei, and De los Rios 2014). In contrast, Argentina provides just marginal subsidies to some private schools, but the increment of private enrolment is rather driven by families opting out from public schools given their poor performance (Narodowski …show more content…
First, these approaches provide an unclear definition of what is diffused (Howlett and Rayner 2008). Therefore, we are left with a blurred idea about how exactly are policies converging. Sometimes, studies focus on the dissemination of particular policy instruments such as vouchers or public-private partnerships (PPP) (Robertson, Mundy, and Verger 2012; Verger 2012). Some others, scholars engage with the adoption of policy objectives or concepts such as school choice (Forsey, Davies, and Walford 2008). And still some others show convergence in particular policy specifications such as the reduction of government education expenditure (Jakobi and Teltemann 2011). These diverse units of analysis may show convergence in some elements while obscures variance in other aspects. A second interconnected problem is the predominance of large-n studies as the preferred methodology to show diffusion and convergence (Howlett and Rayner 2008; Marsh and Sharman 2009). Generally, these studies aim at testing correlation between a proxy of the hypothesized mechanisms (e.g. foreign aid or signed international conventions), and the dependent variable of interest (Schofer and Meyer 2005; Suárez, Ramirez, and Koo 2009). But they tell us little about the patterns in which policy change actually occurs. Finally, although a handful of studies employed qualitative methods to study diffusion, they