Moderate Parliamentarism: Optimism and Classic Assumptions about Parliamentarism

José Antonio Cheibub

Texas A&M University

“Balancing Accountability and Effectiveness: A Case for Moderated Parliamentarism” by Tarunabh Khaitan is a rich and thoughtful defense and expansion of Ganghof’s model of semi-parliamentarism. Like in Ganghof’s formulation, the constitutional structure consists of an executive subject to parliamentary confidence and a dual-chamber legislature. The two legislative houses are symmetrical when it comes to legislation, but confidence is required in only one of them. Khaitan expands this structure by placing political parties at the center of the constitutional system. Moderate parliamentarism is not simply a way to structure executive-legislative relations; it is also defined by the nature and size of the party system in each of the two chambers. The system should be designed to optimize four constitutional principles related to political parties and the way to do this successfully is to simultaneously consider legislative-executive relations and the rules to compose the two legislative chambers.  

Khaitan’s analysis is profound and raises crucial issues for anyone interested in the design of democratic institutions. It is pitched as a way to resolve the trade-off between representation and accountability, which vexes all democratic regimes. Here, I will concentrate on two of the many issues it raises.   

Electoral Rules, Political Parties and Electoral Outcomes 

In moderate parliamentarism, electoral rules are supposed to shape the composition of each chamber. Regarding "the confidence chamber," they should aim at producing majorities that are sufficiently solid to support a government but without excluding significant interests. "The appointing chamber," in turn, would be based on proportional representation in districts of large magnitudes but moderated by thresholds of representation. Whereas accountability would be achieved through the operation of the confidence chamber, representation would come from the appointing chamber. 

How feasible, however, are these goals, given what we know about the effects of electoral systems? Take, to begin with, the confidence chamber. Although the existence of only two parties is not required, Khaitan has in mind a two-party system, in Sartori’s sense, that is, one in which two parties compete for a majority of seats and the party that wins it governs alone. Alternatively, consider the appointing chamber, where the most important interests would be represented, and the parties would operate within a competitive structure similar to Sartori’s moderate pluralism. How confident should we be that the rules proposed for the composition of each chamber would actually lead to what Khaitan has in mind? 

There is, of course, no doubt that electoral rules affect the nature and size of the party systems. But there is no way we can make specific predictions about how many parties will be represented, the way they will interact, and the nature of the interests these parties will represent on the basis of broad electoral principles.  We collectively know a lot about electoral systems, including the fact that the effect of electoral rules on all sorts of outcomes depends on contextual variables. 

Thus, generally speaking, electoral rules can be more or less permissive in the sense that they amplify or restrict the expression of existing interests at the district level. Whether they will actually do so, however, depends on voters’ primary goals with respect to the election, the availability and quality of information, and voters’ ability to process this information, among other things. In turn, competition at the district level can aggregate in multiple ways, leading to national chambers with a party system that is quite different from that prevailing at the district level. A country with SMDs where only two parties compete can lead to a national chamber containing a much larger number of parties, if the two parties competing at the district level are not the same throughout the national territory. Finally, even if rules were designed with firm knowledge of both the number of significant “voter types” to be represented and these voters’ distribution across the national space, that knowledge would likely become obsolete with time in light of economic, social and demographic changes. There is no guarantee that the electoral rules that today produce the desired outcomes will do the same tomorrow. 

Thus, while the simultaneous consideration of electoral rules and legislative-executive relations is a sensible proposition, we should not expect from it more than what is reasonable. 

Protecting Executives and Legislatures From Mutual Interference 

My second broad point has to do with the absence of any discussion of the executive proper and of parliamentary dissolution in Khaitan’s proposal. Executives are, in general, neglected in democratic theory; they are even more so in the context of parliamentary democracies, since the implicit assumption is that the connection between the government and the legislative majority based on the confidence principle is sufficient for the latter to discipline the former. Yet we know that the relationship between executives and parliaments can go awry and that the confidence principle alone is not sufficient to equilibrate parliamentarism as a form of government.  

In a recent paper, I show how European parliamentary systems were stabilized until after WWII. It is arguable that stabilization was made possible by constitutional developments in both new and old parliamentary democracies. In the post-war period, the principle of assembly responsibility was constitutionalized in almost all countries with parliamentarism and a large number of them introduced mechanisms that protect the government from fleeting and/or negative legislative majorities. Examples include the constructive vote of no confidence, first introduced in Germany, which requires the investiture of an alternative government if the existing one is to be removed by a legislative majority. They also include the introduction of provisions forbidding repeated no confidence motions, or their introduction at certain points in the cabinet’s or the assembly’s life cycle. Additionally, the decision rule for a successful no confidence vote was changed from simple to absolute majority, thus making it harder for the government to be removed. Finally, it was during this period that successful no confidence motions were dissociated from parliamentary dissolution, assuring the opposition that removing the government would not imply renewing the parliament.  

At the same time, governments’ ability to request a confidence vote became constitutionalized, with the clear and unambiguous stipulation that if such requests were not granted, the government had to resign. However, these same constitutions set a relatively low bar for approving such a request – simple as opposed to absolute majority – thus making it easier for the government to win. It is also after WWII that parliamentary constitutions adopted provisions that grant governments the kind of legislative powers which enabled them to more easily conform their preferences with those of a legislative majority. Examples include the vote bloqué (Art. 44.3) and guillotine (Art. 49.3) in the 1958 French Constitution, the ability to issue decree powers without the need to invoke a state of emergency (Italy 1948 Art. 77 and Spain 1978 Art. 86), and a series of provisions associated with the introduction, amendment, and approval of legislation in general and the budget law in particular. All of these changes can be interpreted as ways to protect executives in parliamentary democracies by making it harder for the opposition to obstruct legislative proceedings and, given the nature of the regime, to shake the stability of the government. To a large degree, they were responses to crises that most countries had, at one point or another, experienced. 

Parallel to these developments, parliamentary constitutions significantly removed the executive’s discretion to dissolve the assembly before the end of its constitutional term. This is important because it makes it harder, and sometimes even impossible, for prime ministers to manipulate the timing of legislative elections in order to expand their parliamentary base. In other words, it removed parliamentary dissolution as a way to advantage the incumbent. Thus, a large number of constitutions now require that certain actors be consulted about, and in some cases concur with, an early dissolution decision, including sometimes a legislative majority itself. In many recent constitutions, but also in some older but reformed ones, dissolution is conceived as a mechanism to resolve pre-specified extraordinary situations.  For example, it is allowed, and sometimes required, if government investiture is unsuccessful or the government does not form after multiple attempts; if parliament fails to vote on a bill to which the government has attached confidence; if the parliament has removed the government multiple times; if parliament fails to elect a head of state; or if parliament fails to adopt a budget by a certain date. Moreover, dissolution is not allowed within a certain time before the end of the term of the head of state; for a period of time after a parliament was first convened after an election; or within a certain time prior to the end of the parliament’s term. These provisions have in common the fact that they made dissolution less discretionary while allowing it to remain a mechanism for resolving conflicts in pre-specified situations. 

Thus, contemporary parliamentary constitutions acknowledge that the confidence provision alone is not sufficient to induce the kind of behavior expected of politicians and political parties by the classic view of parliamentarism. After so many regime collapses in the inter-war period, constitution makers in Europe recognized that the principle of assembly confidence needed to be regulated for it to function in a satisfactory way. In this sense, the absence of any problematization of the executive suggests that Khaitan’s proposal of moderate parliamentarism is still situated within the parameters of classic parliamentarism, of a system that is self-regulating. This is why constitutional attention must be directed at designing electoral rules that generate the necessary number of parties in each chamber, even if the connection between rules and outcomes is marred by uncertainty, particularly if formulated in terms of broad constitutional principles. These same rules, in combination with the confidence requirement, should be sufficient to induce the creation of political parties with specific normative attributes, even though parties are, for reasons Khaitan well understands and extensively elaborates, rather immune to constitutional nudges. In sum, for all its richness, Khaitan’s proposal for a moderate parliamentarism still rests on the usual assumptions about parliamentarism and is quite optimistic about electoral engineering. 

José Antonio Cheibub is Mary Thomas Marshall Professor in Liberal Arts in Texas A&M University 

Suggested Citation: José Antonio Cheibub, ‘Moderate Parliamentarism: Optimism and Classic Assumptions about Parliamentarism’ IACL-AIDC Blog (13 May 2021) https://blog-iacl-aidc.org/workshop-my-paper/2021/05/11moderate-parliamentarism-optimism-and-classic-assumptions-about-parliamentarism.