Rethinking Democratization in the Philippines: Elections, Political Families, and Parties 57 In presidential systems, there is an inherent tension between national and particularistic representation, with the executive generally addressing broader national policy matters and the legislature usually tied down to parochial district interests (Shugart and Carey 1992). One institutional response to this tension has been the practice of endowing the executive with signicant legislative and related powers vis-à-vis the legislature akin to what O'Donnell has referred to as “delegative democracy”. In the absence of a strong and stable party system, this is a convenient mechanism to address the competing demands of governmental efciency and representation but is also fraught with a lot of problems. The key problem lies in how to check the overly strong powers of the presidency once the executive is allowed to act as the surrogate legislator to solve the problems of collective action embedded in the naturally particularistic legislature. It is in this context that the role of a well-institutionalized and stable party system can best be appreciated as an important mediating and aggregating mechanism for the competing demands of district interests and national policy concerns. Without this important institutional mechanism of a party mediating between the president and the legislature and local ofcials and also attending to intra-party affairs, presidential powers become heightened and prone to abuse. One major rule that has had a direct bearing on the party system since 1987 concerns the one-term limit imposed on the presidency. A one-term rule can weaken the president's incentive to strengthen parties because the sanctioning or disciplining effect of prospective elections is lost. In short, if the incumbent president is not qualied for reelection, the same ofcial faces no strong pressure to strengthen the party to win prospective elections.23 Moreover, a oneterm rule for the presidency lowers the entry barrier for prospective candidates for the next round of elections resulting in the proliferation of candidates and their ephemeral parties (Kasuya 2009). This has been validated by the big number of presidential candidates in every presidential election since 1992: seven candidates in 1992; ten in 1998; ve in 2004; nine in 2010; and ve in 2016 or an average of seven candidates during the last ve presidential elections. Each presidential candidate not afliated with any of the established parties normally creates a new party just for the election contest and which disappears with the defeat of the candidate. However, there is a counter-argument in support of term limits for the presidency. First, the incumbent normally cannot use the powers of ofce to commit fraud on his behalf (although it can be done to favor his party). Second, with reelection not an option, the incumbent's fear of being prosecuted for illegal activities once out of ofce may keep the executive more honest (Fearon 1999:62). It appears that the Philippine experience does not strongly support these arguments. The present study has shown that political families have routinely bypassed the term limits imposed on incumbents by relying on family