Scott Mainwaring and Mathew Shugart, Presidentialism and Democracy in Latin America (1997), Chapters 1, 11.

Place in Literature
Generally, a reiteration of the arguments in Shugart and Carey (1992), with responses to objections made by Linz (1994) and Stepan and Skach (1994) in “The Failure of Presidential Democracy.”
Main Argument: Mainwaring and Schugart address Linz’s arguments about the perils of presidentialism. The authors disagree that presidentialism is particularly oriented towards winner-takes-all results. They argue that the superior record of parliamentary systems has rested partly on where parliamentary government has been implemented and claim that presidentialism has some advantages that partially offset its drawbacks. The authors aim to call more attention to institutional combinations and constitutional designs and to suggest that the advantages of parliamentarism may not be as pronounced as Linz argued.
Method:
Categorises presidential and parliamentary systems: looks at three other issues that may play a role in a society’s likelihood of sustaining democracy: income level, population size, and British colonial heritage.
Key themes:
1. Presidentialism has important institutional variations, so it should not be compared dichotomously against parliamentarism. It is not a homogenous regime type. The most important variations among presidential regimes are (a) how closely they approximate “pure” presidentialism, (b) the president’s legislative powers, (c) party system fragmentation, and (d) party discipline.
2. These institutional variations have consequences for how well presidents can implement their agenda and how well the regime functions.
Chapter 11: Presidentialism and the Party System
Main Argument: The nature of the party system (namely the number of parties, and party discipline) makes a fundamental difference in how presidential systems function [394-5]. Presidentialism should be studied in light of the number of parties and degree of party system fragmentation [434]
– Advantage of presidential systems: they offer more opportunity to check executive power than parliamentary systems
– Presidentialism is likely to function better if party system fragmentation is moderate (around 3.5 parties) [398]
Party Discipline:
– Party discipline facilitates the building of a well organised party system and a “institutionalised party system”
– Linz (1994) & Sartori (1994) argue that Presidential systems function better with disciplined parties
– Less disciplined parties may hinder accountability because ideology, decision-making is more opaque [420]
– There is a symbiotic relationship between presidents’ constitutional and partisan powers  elective affinity [429]
Three hypothesis about Presidentialism: [430-1]
1. potentially dominant presidencies will occur with lower levels of partisan powers and only in constitutions with Presidentialist origins
2. Proactive presidencies will occur with lower levels of partisan powers and only in constitutions drafted and ratified primarily by elected representatives
3. Potentially marginal presidencies will tend to occur at higher levels of partisan power