That said, the AUH still does represent a notable break from clientelism. Most evidence suggests that the AUH is not very selectively administered, additional benefits are not targeted at key voters, and there is little to no political coercion involved (Zarazaga 2012).20Therefore, while understanding clientelism does not involve sweeping, black-and-white comparisons, we can still draw a line between incremental differences in the specifics and implementation of policies. Clientelism is secretive, non-universal, and politicized, while non-clientelism is the opposite.21As such, the swing inherent in the AUH is, in any conception of clientelism, a significant event. 1.2 Literature on Clientelism and Social Policy In recent decades, clientelism has been a much-studied phenomenon in political science. However, surprisingly limited literature addresses cases like 20All of this said, the fact that the AUH applies essentially only to the lower and lower-middle classes is still an important detail later in the thesis. See Chapter 5. 21A relatively simple test for the division between clientelism and non-clientelism is provided by Stokes et al.: “Distributive strategies can be divided into two categories: those with public and binding rules about who gets what, and those in which the rules are absent or hidden… precisely because their patent unfairness would hurt the prospects of the office seekers who deploy them” (2013, 245-246). It is this underlying differentiation between clientelistic and non-clientelistic programs that I rely on for the majority of the thesis.
Chapter 1 — 21 the AUH in which there is a sudden programmatic transition away from clientelism. Yet while it may not directly address the questions at hand, the existent literature does help frame the basic foundation of the thesis. Most importantly, scholars such as Kitschelt and Wilkinson have been able to construct a relatively clear model of clientelist versus programmatic voter linkages, as seen in Figure 1 (2007). This dichotomy helps elaborate on Stokes et al.’s model of clientelism, forming a basic foundation of this thesis (2013). Clientelism can be defined as contingent resource exchanges that are mediated through brokers or party members (Kitschelt and Wilkinson 2007). Moreover, clientelism tends to involve private or small group goods and vote monitoring of some form. On the other hand, programmatic exchanges are the opposite: non-contingent exchanges of public, large group, or universally available private goods without broker involvement (Kitschelt and Wilkinson 2007). Additionally, under programmatic schemes, votes are not monitored and parties tend to be aligned on ideological or policy grounds.
Chapter 1 — 22Figure 1 — Clientelistic v. programmatic voter-linkages22From this literature, we can also attempt to explain the AUH through various existing hypotheses. These theories primarily explain why parties over time tend towards or against clientelism in the long run. However,