Which of the following is more likely to be a problem in a presidential system than in a parliamentary system?

Get help with access

Institutional access

Access to content on Oxford Academic is often provided through institutional subscriptions and purchases. If you are a member of an institution with an active account, you may be able to access content in one of the following ways:

IP based access

Typically, access is provided across an institutional network to a range of IP addresses. This authentication occurs automatically, and it is not possible to sign out of an IP authenticated account.

Sign in through your institution

Choose this option to get remote access when outside your institution. Shibboleth / Open Athens technology is used to provide single sign-on between your institution’s website and Oxford Academic.

1. Click Sign in through your institution.
2. Select your institution from the list provided, which will take you to your institution's website to sign in.
3. When on the institution site, please use the credentials provided by your institution. Do not use an Oxford Academic personal account.
4. Following successful sign in, you will be returned to Oxford Academic.

If your institution is not listed or you cannot sign in to your institution’s website, please contact your librarian or administrator.

Sign in with a library card

Enter your library card number to sign in. If you cannot sign in, please contact your librarian.

Society Members

Society member access to a journal is achieved in one of the following ways:

Sign in through society site

Many societies offer single sign-on between the society website and Oxford Academic. If you see ‘Sign in through society site’ in the sign in pane within a journal:

1. Click Sign in through society site.
2. When on the society site, please use the credentials provided by that society. Do not use an Oxford Academic personal account.
3. Following successful sign in, you will be returned to Oxford Academic.

If you do not have a society account or have forgotten your username or password, please contact your society.

Sign in using a personal account

Some societies use Oxford Academic personal accounts to provide access to their members. See below.

Personal account

A personal account can be used to get email alerts, save searches, purchase content, and activate subscriptions.

Some societies use Oxford Academic personal accounts to provide access to their members.

Viewing your signed in accounts

Click the account icon in the top right to:

• View your signed in personal account and access account management features.
• View the institutional accounts that are providing access.

Signed in but can't access content

Oxford Academic is home to a wide variety of products. The institutional subscription may not cover the content that you are trying to access. If you believe you should have access to that content, please contact your librarian.

Institutional account management

For librarians and administrators, your personal account also provides access to institutional account management. Here you will find options to view and activate subscriptions, manage institutional settings and access options, access usage statistics, and more.

How do we evaluate forms of government or any other set of formal political institutions? One prominent idea is to evaluate them in terms of their causal consequences. Good institutions are those that lead to good results or outcomes. Another prominent idea is that some institutions are inherently more valuable or democratic than others. This idea is more controversial in political theory, and it is not obvious how the two types of evaluation relate to one another. One goal of this chapter is to clarify the approach to the normative evaluation of democratic institutions taken in this book.

A second, more specific goal is to reject the widespread idea that the direct election of the chief executive—most notably under presidentialism—makes a form of government inherently more democratic (Arato 2000: 321; Calabresi 2001: 67; Lijphart 1992a: 13). This rejection is an important part of my overall argument against presidentialism and in favor of semi-parliamentarism. I also reject the suggestion that semi-parliamentary government is inferior to pure parliamentarism on purely procedural grounds (Meinel 2019, 2021).

The third goal is to clarify three more general desiderata in the egalitarian evaluation of democratic institutions: (a) to distinguish the democratic equality embodied in formal procedures (procedural equality) from that realized in the overall political processes (process equality); (b) to specify what a particular institutional scheme is compared to; and (c) to consider the two dimensions of political equality in a representative democracy, horizontal and vertical, in conjunction.

The first three sections develop the conceptual framework: they distinguish three ways to value formal democratic procedures, explain how one institutional scheme can be more democratic than another, and highlight the distinction between vertical and horizontal inequality. I then use this framework to show that presidentialism is not democratically superior and semi-parliamentarism not democratically inferior.

Three ways to value formal democratic procedures

Authors that postulate the inherent democratic superiority of presidentialism have never clarified what it means to say that one institutional scheme is inherently more democratic than another. To do so, it is helpful to distinguish three ways to value formal procedures. Figure 4.1 illustrates this distinction with a concrete example, which we will come back to later. The figure depicts a stylized (incomplete) model of some of the causal effects of the mechanical proportionality of electoral systems—a purely procedural feature that has been considered as inherently more democratic in the literature (Christiano 1996; McGann 2006).

Fig. 4.1

Formal procedures, processes, and outcomes (an illustration)

Source: author’s own composition.

Mechanical proportionality requires that x% of the votes of any party—real and hypothetical—is translated into x% of seats. The degree to which it is realized depends, among other things, on how many seats are to be won in a given district (district magnitude). This procedural feature influences important aspects of the political process, three of which are singled out for illustrative purposes. First, high proportionality is likely to increase citizens’ subjective feeling of being represented by a party (Blais et al. 2014; Rodden 2020). It facilitates the emergence of multiple parties with distinct multidimensional platforms, so that voters are more likely to find a party that they feel close to ideologically. Second, multiple parties in parliament and government tend to reduce the so-called “clarity of responsibility” in a political system. This clarity is generally considered to be maximized when a single party dominates the entire political process (Powell 2000; Schwindt-Bayer and Tavits 2016). Third, mechanical proportionality is likely to influence turnout, partly through the two aforementioned variables. The feeling of being represented is likely to increase turnout (Blais et al. 2014), while lacking clarity of responsibility might reduce it (Park et al. 2019). Mechanical proportionality is also likely to influence certain outcomes of the political process, such as the degree of income redistribution. This influence may partly run through turnout (Kenworthy and Pontusson 2005).

While we will return to some of these causal hypotheses in Chapters 5 and 6, here the model is merely used for illustration. It helps us to distinguish three ways of evaluating a purely procedural feature, such as mechanical proportionality: in terms of

1.

its causal effects on the outcomes of politics, such as income redistribution;

2.

its causal effects on features of the democratic process, such as voter feelings, clarity of responsibility, and turnout;

3.

its potential non-instrumental value , that is, the value that is independent of any causal consequences it might have.

I assume that when authors claim presidentialism to be inherently more democratic than parliamentarism, they have the third, non-instrumental value of political institutions in mind. While they may also have separate concerns about the causal consequences of presidentialism (e.g. Lijphart 1992a), these are discussed in Chapter 9. Here, our goal is to specify a purely proceduralist interpretation of “more democratic.” To check whether such an interpretation is meaningful, we first have to consider objections to the idea that formal procedures can have non-instrumental value at all.

Two main groups deny it. One argues that we should only care about outcomes. Its members are often called “instrumentalists” because they see democratic procedures and processes merely as instruments for achieving desirable outcomes, such as a fair income distribution or, more abstractly, justice and truth. They deny that the kind of political equality we associate with representative democracy has any non-instrumental value (Wall 2007; but see Viehoff 2017). Instrumentalists of this type fall into two broad camps: those who believe that representative democracy as we know it is the best instrument (e.g. Bagg 2018; Landemore 2017) and those who doubt this (e.g. Brennan 2016).

The second group that denies the non-instrumental value of formal procedures has a more complex position. It rejects the kind of instrumentalism we have just discussed and embraces political equality as having non-instrumental value. Members of this group have a more robust commitment to democracy because they believe that the substantive outcomes that we ought to pursue in a political system must, in some adequately egalitarian way, be determined by the citizens themselves. For example, James L. Wilson (2019: 111, n. 26) argues that political equality requires the “appropriate consideration” of citizens’ political judgment. This concern about equality in the process of political decision-making allows this group to potentially justify adopting democratic arrangements over non-democratic ones, even when the latter lead to better substantive outcomes. It is the achievement or approximation of some equality standard, which I call process equality, that is seen as having some kind of non-instrumental value.

When it comes to the evaluation of formal institutions or procedures, any process equality standard is also a kind of outcome standard (see also Estlund 2009: 248–251). So, the second group, too, values formal procedures instrumentally: in terms of their causal consequences for process equality (as well as substantive outcomes). For example, they do not accept procedural features such as mechanical proportionality as a “requirement” for political equality (Wilson 2019: 194). They insist that disproportional electoral systems, such as the so-called first-past-the-post system, are not, in themselves, “undemocratic” (Kolodny 2014: 288; Beitz 1989). Instead, formal institutions such as those of the electoral system must be evaluated in terms of their consequences for an egalitarian political process, all things considered. This group needs a theory of how—through features such as clarity of responsibility or turnout—formal procedures affect overall process equality.

What about those that do ascribe some non-instrumental value to formal institutions, or certain aspects of these institutions, such as mechanical proportionality? They do not deny the importance of overall process equality but suggest that the kind of equality embodied in formal procedures, what I call procedural equality, has some kind of priority and establishes some kind of baseline. This baseline deserves special attention and departures from it deserve a special justification. A proportional representation (PR) system with maximal mechanical proportionality is seen as one example of such a baseline institution (Christiano 1996, 2008; McGann 2006).

Before we move on, let me emphasize that I use important terms differently than much of the literature on the justification of democracy. This literature contrasts “instrumental” and “procedural” reasons for democracy but does not typically distinguish between procedural and process equality. It therefore often fails to acknowledge that concerns about process equality are also concerns about the causal effects of formal institutions. I highlight this distinction between procedural and process equality and use the term “procedural” with a narrow focus on the evaluation of formal institutions. It is this evaluation that we are concerned with when we try to understand whether one institutional scheme is inherently “more democratic” than another.

How one institutional scheme can be inherently more democratic than another

What kind of priority does the value of procedural equality have, and what kind of baseline does it establish? I believe that process equality and substantive outcomes must have moral priority over procedural equality. For example, if it were true that high mechanical proportionality consistently undermined process equality (e.g. by leading to less and more unequal turnout, etc.), we would have reason to avoid it. However, it does not follow that procedural equality has no distinct value at all or that an analytical focus on it is misplaced. It merely means that the non-instrumental value of procedural equality is conditional and that an analytical focus on it cannot be justified on purely moral grounds. I elaborate on both points in turn.

To say that procedural equality has conditional value means that its value can be undercut by considerations about causal effects (Christiano 2008). It can be valued for its own sake but only under certain background conditions. More specifically, procedural equality may be valued non-instrumentally as a particularly visible component of a fair democratic process but only if its causal consequences do not run counter to overall process equality. If they do, procedural equality may lose its non-instrumental value (see also Viehoff 2019).

But if procedural equality is subordinate in this way, why focus on it in the first place? Why not pick some conception of process equality (and desirable substantive outcomes) as our target variable and reason backwards from the available causal knowledge to the desirable set of institutions (Beitz 1989; Kolodny 2014; Wilson 2019)? My answer to these questions highlights the importance and difficulty of publicly justifying procedural inequalities in the real world. We know as a matter of social-scientific fact that instrumentalist justifications of procedural inequalities—including those about process equality—are often insincere or reflect well-known cognitive biases (confirmation bias, status quo bias, etc.). They are often made in a self-interested manner, especially by powerful actors who benefit from some institutional scheme or can predict to do so in the future (e.g. Colomer 2005; Klarman 2016). Against this background, the point of focusing on procedural equality is to shift the onus of justification onto those that argue for procedural inequality.

I suggest that this shift is grounded in an explanatory presumption, not a moral one, the underlying logic being that of Ockham’s Razor (Sober 2015). Since the simplest and thus prima facie most likely explanation for any procedural inequality is that it benefits powerful groups or actors, a crucial task in real-world deliberation about justifiable political institutions is to distinguish genuine instrumentalist justifications from pseudo-justifications driven by self-interest and cognitive bias. An important task for political theory is to inform this real-world deliberation. Only if a genuine justification for highly visible procedural inequalities exists can it be publicly available in a way that reassures free and equal citizens that they are not treated unjustly and/or as social inferiors (Christiano 2008; Gaus 1996; Kolodny 2014; Viehoff 2019).

Finally, let us note a problem with the notion of a baseline. It might be thought to imply that there is some ideal set of procedures, which uniquely embodies the value of political equality and thus ought to be approximated (Christiano 1996; McGann 2006). One problem with this thought is that the requirements for procedural equality cannot be uniquely specified without any (implicit) instrumentalist assumptions. For example, a concern about procedural equality in making decisions might lead us to use majority rule, but it may also lead us to flip a coin (Estlund 2008, 2009). Moreover, we will see below that when we consider a representative democracy, the desiderata of formal procedural equality along the horizontal and vertical dimensions of the political process may conflict. Hence, it is impossible to determine some ideal design of representative democracy on purely procedural grounds.

A solution to this problem is to conceive of the justification of procedural inequalities in a strictly comparative manner (Wiens 2012). A comparative approach avoids evaluating certain institutional schemes tout court. When we compare institutional schemes only along one specific dimension, while keeping others constant, we are often able to say that one institutional scheme is (conditionally) preferable to another in virtue of its greater procedural equality. I will call such a scheme procedurally preferable, regardless of whether it is the scheme we ought to adopt, all things considered.

Take, for instance, the comparison between two parliamentary systems that use proportional representation with closed lists in a single district, but different legal thresholds of representation. These thresholds imply that parties have to surpass a certain vote share to win parliamentary seats at all. Suppose that the thresholds are at 5% and 10%, respectively. It is meaningful to say that the former institutional scheme is procedurally preferable and, in this sense, more democratic because it nullifies the votes of fewer (sincerely voting) citizens. Hence, if the scheme with the 5% threshold achieved process equality and good substantive outcomes equally well, it is the one we ought to choose.1 This is different from saying that this scheme approximates some ideal set of democratic institutions, but it directs our attention where it belongs. Whenever someone proposes to establish or maintain procedural inequality, we must ask whether some alternative scheme could not achieve the relevant goals equally well but with less procedural inequality.

Taking vertical inequality seriously

In a representative democracy, procedural equality has two analytical dimensions, horizontal and vertical (Dworkin 2000), but many normative discussions focus only on the former (Ganghof 2015b). As Abizadeh observes:

Having equal opportunity to wield power in selecting representatives and to influence representatives once selected may be a way of instantiating horizontal equality between non-representatives. But horizontal equality fails to address the formal vertical inequality intrinsic to representative democracy: between representatives empowered to decide legislation and policy and non-representatives who are not. The tendency to parachute a notion of political equality forged with direct democracy tacitly in mind—as equal say in majoritarian decision-making—into a theory of representative democracy (Waldron 1999) fails to take seriously this vertical inequality and the fact that elections select office-holders rather than decide laws. Being treated as an equal qua selector (and having equal opportunity to influence representatives) is therefore insufficient for political equality [emphasis in the original].

Vertical inequality in the formal procedures of democracy may be justified in terms of process equality and/or substantive outcomes. Normative theorists have focused on two types of comparisons. One is between direct and representative democracy. Proponents of representation justify procedural vertical inequality in terms of greater overall process equality and/or better outcomes. For example, Christiano (2008) focuses on process equality and argues that the intellectual division of labor achieved through representation increases everyone’s control over society so much that it overcompensates for procedural vertical inequality. That is, “even the power of the least powerful is likely to be greater than under direct democracy” (Christiano 2015: 102). Landa and Pevnick (2020b) focus more on good outcomes and justify representation as a sort of compromise between full democracy and “epistocracy,” the rule of the knowers (see also Brennan 2016).

The second comparison is between electoral and lottery-based representation. Abizadeh (2020) argues that the only way to make the unavoidable vertical inequality of representation compatible with the value of political equality is to treat citizens equally not qua selectors but qua candidates for office (see also Guerrero 2014; Landemore 2020). Office-holding is seen as a good that consists in extra opportunities to wield power over political decisions and that cannot be distributed equally. Hence the superior institutional scheme is to give everyone “an equal chance or opportunity to hold office” (Abizadeh 2020: 7). However, while this solution establishes a certain form of procedural equality, whether it is better overall can be questioned. Landa and Pevnick (2020a) defend electoral representation in terms of both process equality and better substantive outcomes.

There has been less interest in a third type of comparison, that between different degrees of procedural vertical inequality under different forms of representative government. The normative literature is virtually silent on this matter, even at an abstract level. For example, an influential article by Niko Kolodny (2014: 317–318) makes rather detailed claims about horizontal equality and institutional design, but it ignores differences between forms of government and specifies only the most rudimentary procedural requirements for acceptable vertical inequality. Most notably, he requires that the principal controls the selection of the agent and that the agent can be replaced after a short and limited term.

It is here that the claims in political science and constitutional theory about the alleged democratic superiority of presidentialism enter the picture. As I understand them, these are claims about reduced procedural inequality along the vertical dimension. They are misleading, however, because they fail to adequately distinguish and specify the relevant comparisons and disregard conflicts between vertical and horizontal equality.

Is presidentialism inherently more democratic?

It has often been suggested that presidentialism is inherently more democratic than parliamentarism. Arend Lijphart (1992a: 13) states that a “major advantage of presidential government is that its popular election of the chief executive can be regarded as more democratic than the indirect ‘election’—formal or informal—of the executive in parliamentary systems.” Similarly, Andrew Arato (2000: 321) suggests that the critique of presidentialism “was rarely based on normative considerations for the simple reason that, under a democracy, direct elections are always preferable to indirect elections that can always deny office to the candidate the voters actually prefer.” Calabresi (2001: 67) concurs and suspects that “[f]or many readers this advantage of presidentialism over parliamentarianism may be dispositive just by itself.” Similar claims can be found in other works (Moe and Caldwell 1994: 172; von Mettenheim 1997).

The underlying argument is not spelled out, though. One intuition is that direct election conserves some of the putative normative appeal of direct democracy (von Mettenheim 1997). Another idea is that the need for direct authorization varies with the power of the office: “Democracy does not require the popular election of all public officials, of course, but the argument that heads of government, who are the most important and powerful office-holders in democracies, should be directly elected has great validity” (Lijphart 1992a: 13; see also Calabresi 2001: 67).

As I interpret these ideas, they suggest that because presidential government gives citizens as the principal more direct control over the selection of the chief executive as a particularly powerful agent, it is procedurally preferable to parliamentary government. This suggestion is mistaken for two reasons: (a) it conflates two distinct comparisons; and (b) it disregards the possibility of conflict between procedural considerations along the vertical and horizontal dimensions.

Consider first the comparison between a presidential system with the direct election of the chief executive and an otherwise identical presidential system, in which presidential selection is processed through intermediate agents such as the members of the Electoral College of the United States. In this comparison, all the other elements of the compared systems remain fixed, so that direct election is indeed procedurally preferable. Concerns about horizontal and vertical equality do not conflict, but point in the same direction.2 An institution like the Electoral College violates horizontal procedural equality because the votes of some citizens, those in more populous states, do not have the same weight as those of others. The Electoral College echoes the unequal representation of citizens in the United States Senate because states are accorded College votes according to the number of representatives in Congress. It also threatens to undermine vertical equality because its members might choose to violate their mandate and not elect the candidate that won in the respective state. The intermediate agents may thus have more control over the selection of the president than other citizens. This possibility has been subject to legal controversy. In this particular comparison, it is meaningful to say that the direct election of the president is procedurally preferable and, in this sense, more democratic.3

The comparison between a presidential and a parliamentary system of government is different. We have already seen that the arguments about this comparison are not well developed, but a charitable interpretation might go as follows. It is procedurally preferable, along the vertical dimension of procedural equality, that each and every agent in a representative democracy be directly elected by the entire electorate: each member of parliament, each member of the cabinet, each member of the Supreme or Constitutional Court, and so on. That representative democracy is not ultimately set up in this way is because of weighty instrumental reasons: Members of parliament ought to be accountable to their parties or local constituencies; there must be some hierarchy in the cabinet to create clarity of responsibility; judges ought to be shielded from electoral competition; and so on. Hence, the idea might be that the general procedural preference for direct election does not survive instrumentalist scrutiny for most individual agents, but it does for the heads of government, as the most important and powerful office-holders.

The problem is that this kind of argument about vertical procedural equality may conflict with reasonable concerns about horizontal equality. After all, it is not enough to directly elect a set of individual agents. These agents must also interact with each other under specific rules in order to produce collectively binding decisions. Our concerns about procedural equality must also include these horizontal rules of interaction, which might well point towards making the chief executive an agent of the assembly.

Consider, for instance, McGann’s (2006) justification for a parliamentary system of government. He argues that political equality requires that decision-making power be concentrated in a legislative assembly that makes internal decisions by simple majority rule and whose members are elected under rules that are mechanically proportional. In this way, citizens have formally equal opportunity to influence binding decisions via groups of representatives with similar views (see also Christiano 1996). Moreover, McGann (2006: 85) notes that one way to extend this equality in legislative voting to the process of agenda-setting is to let the assembly majority select and deselect the chief executive and cabinet. In this account, it is procedurally preferable along the horizontal dimension of equality to select the chief executive indirectly.

A proponent of presidentialism might respond to McGann by insisting that a presidential system could be designed to match the horizontal procedural equality achieved by proportional-representation (PR) parliamentarism (Colomer and Negretto 2005). The mechanical proportionality in assembly elections could match that of a parliamentary system, presidents could be denied absolute veto power over legislation, and whatever specific powers (e.g. in legislative agenda-setting) presidents might have could be fairly authorized by an absolute majority of voters in a separate presidential election.4 When we compare this kind of system to the one favored by McGann, purely procedural considerations are insufficient to rank them. While the proponent of presidentialism can point to the procedural value of electing chief executives directly, McGann can point to the value of having them selected by a proportionally elected majority coalition.

We can certainly try to weigh the conflicting considerations against one another but not without bringing in instrumentalist assumptions about process equality and the requirements of adequate representation. For example, the proponent of parliamentarism might deny that a single human being can adequately represent a heterogeneous citizenry and emphasize that a prime minister must continuously accommodate the preferences of the majority in a proportionally elected assembly to stay in office. One might also argue from a social choice perspective that a PR parliamentary system is more reliable in empowering the median voter (if it exists) as the Condorcet winner or, at least, in preventing the victory of the Condorcet loser (see Colomer and Negretto 2005; McGann et al. 2002).5 By contrast, the proponent of presidentialism might point to the fact that the endogenous selection of the chief executive in a pure parliamentary system can lead to a bias against whichever side on the general left–right dimension is fragmented into a greater number of parties (Döring and Manow 2015). This is because the party that leads the cabinet-formation process is often the largest party but not necessarily the one preferred by a majority of voters. Both sides have to make assumptions about human psychology and the causal effects of institutions; they cannot make purely procedural arguments. Hence, the claim that presidentialism is inherently more democratic than parliamentarism is false.

The neglect of direct recall

It is also worth noting that the proponents of this claim do not apply their concerns about vertical procedural equality consistently. After all, the vertical equality between principals and agents not only depends on how the agents are selected but also—and more fundamentally—on how their authority can be revoked.6 The possibility of recalling all directly elected agents is arguably procedurally preferable to its absence. Under presidentialism, the possibility of directly recalling a directly elected president would reduce procedural inequality along the vertical dimension without affecting the horizontal dimension.7 This possibility is preferable in the same way in which it is preferable, on egalitarian grounds, that we as individuals can fire our doctors or lawyers. It is striking that proceduralist arguments for presidentialism neglect this procedural superiority of direct recall.

To be sure, one can claim that the recall of directly elected representatives would have undesirable causal effects on process equality and substantive outcomes. But such a claim has to be part of a more general instrumentalist evaluation. In Chapter 9, I argue that presidentialism cannot be defended on instrumental grounds. The argument essentially reverses the logic of Lijphart (1992a) and Calabresi (2001). The direct power that the holders of the office of the chief executive exert over their citizens does not give us procedural reasons for their direct election (as we have seen), but it does give us instrumental reasons to make their authority politically revocable by some collective and representative entity.

Is semi-parliamentarism less democratic?

Let us finally consider the comparison of parliamentarism and semi-parliamentarism from the perspective of procedural equality. Some authors worry that there is something democratically defective about semi-parliamentary government (Meinel 2019, 2021; Weale 2018). Here I want to respond to Meinel; Weale’s concerns are discussed in Chapter 6.

Meinel (2019, 2021) sees semi-parliamentarism as an interesting response to the challenges faced by pure parliamentary government, such as the increased partisan fragmentation of parliaments. However, he considers semi-parliamentary government to violate a “principle of egalitarian representation” (Meinel 2019: 212; see also Meinel 2021: 135-136). The idea seems to be that under parliamentarism, the equality of the members of parliament symbolizes the equality of citizens. Since semi-parliamentarism creates a privileged group of assembly members (those that can participate in the no-confidence procedure), it gives up on this symbolic representation of citizens.

I want to make two main points in response. First, I do not view Meinel’s concern as one about the procedural equality of citizens. I have argued that this equality conditionally requires that citizens have equal institutional entitlements, not that their status is symbolized in a particular manner. When we think about semi-parliamentary government in terms of these entitlements and specify the relevant comparisons systematically, we can see that it can be procedurally preferable to parliamentarism.

Meinel’s comparison of parliamentarism and semi-parliamentarism lacks precision because he does not specify the relevant levels of mechanical (dis)proportionality in the electoral system. Elsewhere in his book he considers the 5% legal threshold of representation in the German electoral system to be instrumentally justified because it makes an “indispensable contribution” (2019: 121) to the formation of stable governing majorities in a parliamentary system. Yet, such a threshold means that the actual or potential voters of below-threshold parties are procedurally discriminated against. By denying these voters representation in parliament, they are denied all the opportunities associated with it, including the opportunity to participate, via their chosen parties, in the vote of no confidence procedure. As a result, whether or not parliamentarism treats citizens more or less equally than semi-parliamentarism in purely procedural terms depends on the level of the respective thresholds of exclusion.

To see this, let us consider the simple version of semi-parliamentary government introduced in Chapter 3: a legal threshold of confidence authority (see Ganghof 2018a and Chapter 8). Parties whose vote share is above the threshold of representation but below the threshold of confidence authority gain representation in parliament and the rights associated with it, but not the right to participate in the no-confidence procedure. The bicameral versions of semi-parliamentarism that we find in Australia create essentially the same result. The implicit electoral threshold created by the majoritarian electoral systems of the first chamber becomes the threshold of confidence authority, while the lower implicit threshold of the proportional systems of the second chamber becomes the threshold of representation.

Semi-parliamentary government can be procedurally preferable to parliamentary government because a threshold of confidence authority denies certain citizens fewer rights than a threshold of representation. Whether this is the case depends on the respective thresholds. When we compare a parliamentary system with a 5% threshold of representation to a semi-parliamentary system without such a threshold but a 5% threshold of confidence authority, the latter is procedurally preferable, everything else being equal. It denies the actual or potential voters of below-threshold parties fewer opportunities. They are merely denied the opportunity to influence the formation and dismissal of the government, not the opportunity to participate fairly in legislative deliberation and voting. It is precisely in this sense that semi-parliamentarism can be considered more democratic than parliamentarism, everything else being equal.8

When we vary the thresholds in the comparison, the evaluation becomes more complicated. Consider, for instance, a parliamentary system with a 5% threshold of representation and a semi-parliamentary system with no such threshold but a 10% threshold of confidence authority. The reduced procedural inequality in legislation and deliberation under semi-parliamentarism must now be weighed against the increased procedural inequality in choosing the government. Here instrumental concerns, for example about the importance of the government’s control of the legislative agenda, must enter the picture and a purely procedural comparison becomes inconclusive.

My second response to Meinel is to grant that the symbolism of semi-parliamentarism might raise valid instrumentalist concerns, as it makes the procedural discrimination of certain voters more visible. When parliamentary systems deny some voters fair representation in parliament, this discrimination becomes hidden to some extent once parliament is formed. It is almost as if these voters did not exist. By contrast, if these voters are represented in a semi-parliamentary system but denied participation—via their representatives—in the formation and dismissal of the government, the procedural discrimination becomes highly visible. This visibility might have negative causal effects, for example, on the overall support for the democratic system. This is an empirical hypothesis worth considering, although it does not seem to find a lot of initial support in the bicameral cases of semi-parliamentarism in Australia (Stone 2008).

Conclusion

Presidentialism is not inherently more democratic than parliamentarism—a purely procedural comparison of these two systems is inconclusive. Parliamentarism is also not preferable to semi-parliamentarism on purely procedural grounds, but semi-parliamentarism is preferable if the relevant thresholds of exclusion are held constant in the comparison. Denying the voters of below-threshold parties power over the cabinet creates less procedural inequality than denying them any representation in the assembly. There are no conclusive procedural reasons for presidential government or against semi-parliamentary government.

Notes

1

I am not concerned here with the relative importance of process equality and substantive outcomes because my goal is not to present a justification of democracy. Whether the approach outlined here implies a justification of representative of democracy as we know it depends on our causal knowledge.

2

For a more detailed discussion of the Electoral College from the perspective of political equality, see Wilson (2019).

3

Of course, defenders of the Electoral College might accept this but insist that its instrumental benefits undercut the procedural value of direct election. This question need not concern us here, although it is worth noting that there are also weighty instrumental reasons against the Electoral College, including those associated with process equality. One reason is that presidential campaigns tend to focus on a few swing states, thus potentially not giving adequate consideration to the interests and judgments of citizens in other states.

4

Perhaps the most egalitarian way to directly elect a president is an alternative vote (or “ranked choice”) system. All voters can rank as many candidates as they like, and the candidates with the lowest vote shares are sequentially eliminated and their votes reallocated to determine the candidate with an absolute majority (more than 50% of all votes). For further discussion, see Chapter 8.

5

The Condorcet winner (loser) is the alternative that would win (lose) every pairwise majority contest.

6

Abstract discussions about the equality between principal and agent also ignore this crucial aspect of their relationship (e.g. Kolodny 2014). This is surprising, given that all of the archetypical agents Kolodny and others use to motivate the argument for delegation (doctors, lawyers, accountants, and financial planners) can usually have their authority revoked at any time and for whatever reason. The call for the possibility of recalling public officers also has a long pedigree in political thought (Qvortrup 2020; Whitehead 2018).

7

Of course, the democratic superiority of direct recall applies more generally. In particular, it is procedurally preferable under any form of government that citizens can recall individual members of parliament (when these are directly elected in geographically defined districts) and/or the parliament as a whole.

8

A fuller evaluation must also take into account the procedures regulating inter-branch relations; see Chapter 8.

Which of the following ways does a parliamentary system differ from a presidential system?

What are the potential disadvantages of a presidential system quizlet?

What is presidential and parliamentary system of government?

What makes a parliamentary government different from the United States?