Political Parties: Constitutional Roles, Recognition, Rights and Regulations

Aim

This primer is a guide to the design of constitutional provisions on the roles, recognition, rights and regulation of political parties in a democratic political system. It discusses the prohibition of anti-democratic parties, internal party democracy, the funding and resourcing of parties, the registration of parties, the organization of party caucuses and the enforcement of party discipline through ‘anti-defection’ provisions.¹

Scope and content

Chapter 2 defines political parties, discusses their roles and functions in a democracy and introduces the fundamental question of how to balance the hybrid nature of parties as private associations that perform important public functions. This chapter also introduces the concept of ‘militant democracy’ as it applies to the constitutional recognition, rights, regulation and prohibition of political parties—themes that are developed in subsequent chapters.

Chapter 3 explores how constitutions can recognize and even encourage the role of parties in democratic politics. This might involve specific commitments to multiparty democracy, the prohibition of one-party states and the separation of the state from the party. It also considers recognition of the role of political parties in the legislature (party caucuses).

Chapter 4 examines how constitutions protect the right to join and support a political party, the right to take part in party activism and other rights that ensure parties can function effectively. It also discusses certain limitations on these rights as they apply to persons such as civil servants, judges, members and staff of fourth-branch institutions and members of the armed forces, who are supposed to remain impartial.

Chapter 5 considers constitutional rules on the regulation of political parties, as well as mechanisms for the registration and regulation of parties. It examines whether there should be a special body for these purposes.

Chapter 6 examines how parties organize themselves, the election of party leaders, internal decision-making processes and the importance—and limits—of democracy within parties.

Chapter 7 discusses political party financing, from transparency in donations to spending limits and auditing, the public funding of parties and the public provision of benefits in kind, such as guaranteed access to state broadcast media.

Chapter 8 focuses on a different way constitutions can interact with political parties—by banning them. It discusses the concept of ‘militant democracy’ and asks under what circumstances, if any, banning parties might be permissible. It also considers the process of banning parties and the implications of such bans.

Chapter 9 discusses how constitutions might seek to enforce party discipline, whether they should do so or whether informal means of enforcing party discipline are sufficient.

The primer also provides some decision-making questions for use in discussions or to help decision makers think through the coherence and consequences of their constitutional design choices (Chapter 10), as well as example tables of illustrative constitutional provisions (Chapter 11).

What is a political party?

The first question is definitional: What is a political party?

Definitions and understandings of what constitutes a political party vary, but for the purposes of this primer, we consider that, at their core, political parties are bodies of people, united by certain common principles or policies, who seek by common endeavours to hold elective office in order to pursue outcomes that give effect to those principles or policies.

The roles and functions of political parties

In the earliest days of modern constitutionalism—in the late 18th and early 19th centuries—the prevailing attitude was one of hostility towards political parties. They were dismissed by US ‘Founding Father’ James Madison as mere ‘factions’ fighting over the spoils of power, as opposed to pursuing the broader public interest (Drutman 2020).

By the late 19th century, however, it was increasingly recognized that political parties played a vital role in a democracy. According to one late-19th-century US commentator, political parties were ‘always and everywhere … the manifestation of political life, and the indispensable means of its growth’ (Morse 1896).

Today, the major roles and functions of political parties in a democracy are:

• Interest aggregation: Citizens have diverse interests that cannot all be met at the same time. Political parties aggregate the interests of party members, supporters and allied interest groups, through brokering, compromise and negotiation, into a manageable number of ‘packages’ on which voters can make informed and effective choices.
• Structuring the popular vote: Political parties structure election campaigns into debates about competing governing agendas, thereby facilitating effective choice and accountability. Election results are not just a sum of scattered individual preferences, but a reflection of collective public opinion. Voting for a party with a programme, in coordination with others who support the same party and programme, enables voters collectively to influence public policy.
• Formulating public policy: Political parties can translate broad principles into specific proposals for public policy. They may engage in policy research and sometimes have affiliated foundations in which policy innovations can be incubated. These policy goals and proposals, in turn, are the core elements of a party platform that sets out the party’s agenda for government.
• Mobilization and integration of citizens: Political parties are campaigning machines. They seek to convince voters to turn out and vote for their candidates. By engaging and enthusing voters, they help to ensure that election results reflect public opinion. This also integrates citizens into the political process and gives them a stake in the political system.
• Responsibility and accountability: Political parties commonly maintain a sustained identity over time, which means that even if individual candidates or leaders change, parties can be held collectively responsible by the voters for their policies, actions and conduct in office.
• Recruiting leaders for public office: Political parties identify future political leaders, and recruit, train and provide them with opportunities and the resources (human and financial) to secure public office.
• Enabling the enactment of legislation: Legislatures are large multi-member bodies, the members of which each have a single vote. Political parties enable the work of the legislature to be organized, by coordinating individual legislators to act collectively as party members to set the legislative agenda, propose legislation, offer their collective views and vote on it.
• Government selection: In parliamentary systems, political parties enable the selection of the government. The person who is the leader of the majority party or coalition in the Lower House is usually entitled to be appointed prime minister, and the prime minister can therefore be chosen and removed by internal leadership elections within the majority party. For example, between 2016 and 2022, the United Kingdom had four prime ministers who came to power, and lost power, by means of internal party decisions. In presidential systems, this role is less evident, since parties cannot usually remove a president, but parties are nonetheless instrumental in choosing presidential candidates either by decisions of a party convention or by means of primary elections.

In reality, not all parties are as effective as they should be at performing these functions. Parties in fragile or weakly consolidated democracies are often ‘poorly institutionalized, with limited membership, weak policy capacity and shifting bases of support; they are often based around narrow personal, regional or ethnic ties, rather than reflecting society as a whole; they are typically organizationally thin, coming to life only at election time; they may have little in the way of a coherent ideology; [and] they often fail to stand for any particular policy agenda. They are frequently unable to ensure disciplined collective action in parliament, with members shifting between parties’ (Reilly and Nordlund 2008). Corrupt, weak, unstable or ineffective political parties can distort or undermine the operation of a constitution.

Even when political parties perform these functions well, they are not an unmixed blessing. Nancy Rosenblum (2009) identifies a strain of ‘anti-partyism’ in some countries, which sees parties as a threat to democracy and good government. Critiques of political parties focus on three aspects. First, parties can undermine the ability of individual representatives to act in the best interests of their constituents or to act in accordance with their own personal judgement. Parties can also be criticized for encouraging a politics of artificial division and conflict, for ‘bickering’ and ‘intransigence’, or for institutionalizing dishonesty and insincerity by requiring politicians to stick to the party line even when they know it to be false. Second, parties are criticized for enabling a small professional political class to hold power, excluding ordinary citizens who do not line up behind partisan labels from public life. Third, parties may be viewed as vehicles of patronage, clientelism and corruption—a mechanism by which jobs and favours are distributed in return for political support. Rosenblum nonetheless makes a case for the continued value of parties in structuring democratic politics, enabling voters to make effective choices between coherent packages of personnel and policy offerings. She notes that ‘non-partisanship is not a synonym for independent thought: it is navigating without political orientation or organization’, while ‘partisans, on the other hand, take responsibility for telling a comprehensive public story about the economic, social, and moral changes of the time’ (Rosenblum 2009).

Over time, constitutional design has expanded to seek to accentuate the positive aspects of parties while limiting their negative tendencies. Today, partisan candidates are the norm in most democracies, while independent candidates, if they feature at all, are rare exceptions.

Constitutional status of political parties

There is a lot of variation among democratic constitutions in the provisions related to political parties: ‘A global comparison of the laws governing political parties reveals a significant degree of diversity—in the sources from which these laws derive, the specific functions they target, and the extent to which they regulate parties as political organizations and electoral actors’ (Gauja 2016: 4).

Unlike Communist constitutions or the constitutions of one-party regimes, liberal-democratic constitutions cannot—provided certain minimum democratic criteria are met (see Chapter 8)—specify which parties can exist, what their ideological orientation should be or which policy platforms they should endorse. Parties rise and fall according to public opinion. Even when highly regulated, political parties in liberal democracies are not state institutions.

Parties are therefore often described as intermediary institutions that sit between society and the state (Müller 2021). On the one hand, they are created by citizens to reflect their interests and to mobilize people politically to change policies and laws. In this respect, they are institutions of civil society, like trade unions or pressure groups. On the other hand, the distinctive purpose of a political party is to elect public officials to wield political power in public institutions under public law. In this respect, political parties have some public aspects to their roles and functions that are far too important to be left unregulated in the way that a tennis club, for example, would be unregulated.

A key question to be addressed in this primer, therefore, is how to balance the private and public nature of political parties. Should the constitution merely respect the rights of political parties as essentially private organizations, with freedom of action in the political system secured by constitutional provisions on freedom of expression and association? Or are parties such essential and powerful institutions that the public has a legitimate interest in their functioning, and which should therefore be subject to constitutional regulation? The trend over the past century has been for more constitutional regulation of political parties (Choudhry 2024).

Militant democracy

A focus on political parties having important public duties and therefore being the proper subject of constitutional protection and regulation could include what has been described as ‘militant democracy’, whereby the state seeks to prevent the rise of anti-democratic parties in order to save democracy from self-destruction.

Militant democracy originated as a set of constitutional techniques that seek to constrain the potential threat posed by anti-democratic political parties to constitutional democracy. The concept was developed by the German émigré Karl Lowenstein (1937) based on his diagnosis that the Nazis had come to power in Weimar Germany by initially complying with the rules on political competition in a democratic constitutional order. Before seizing power unconstitutionally, the Nazis formed a political party, contested elections, took their seats in the Reichstag, participated in the legislative process and formed a government through a constitutional mechanism at the invitation of the president. Militant democracy posits that the state and public authorities, while neutral between democratic political parties, cannot be neutral with regard to democracy itself. It must promote democracy and resist its enemies.

In response to its history, Germany became a leading example of militant democracy after World War II. This is expressed in Germany’s Basic Law of 1949, which (as amended in 2017) provides in article 21 that:

1. Political parties shall participate in the formation of the political will of the people. They may be freely established. Their internal organization must conform to democratic principles. They must publicly account for their assets and for the sources and use of their funds.
2. Parties that, by reason of their aims or the behaviour of their adherents, seek to undermine or abolish the free democratic basic order or to endanger the existence of the Federal Republic of Germany shall be unconstitutional.
3. Parties that, by reason of their aims or the behaviour of their adherents, are oriented towards an undermining or abolition of the free democratic basic order or an endangerment of the existence of the Federal Republic of Germany shall be excluded from state financing. If such exclusion is determined, any favourable fiscal treatment of these parties and of payments made to those parties shall cease.
4. The Federal Constitutional Court shall rule on the question of unconstitutionality within the meaning of paragraph (2) of this Article and on exclusion from state financing within the meaning of paragraph (3).
5. Details shall be regulated by federal laws.

These provisions, which are widely emulated, cover many of the issues that are addressed and explained in subsequent chapters of this primer. The first sentence of the first paragraph sets out the role of political parties and gives recognition to their place in a democratic political system: they ‘participate in the formation of the political will of the people’. The second sentence of the first paragraph sets out the rights of political parties, noting that ‘They may be freely established’. The third sentence of the first paragraph covers the regulation of political parties, stating that ‘Their internal organization must conform to democratic principles’. The final sentence of the first paragraph sets out a principle relating to the financing of political parties, explaining that ‘They must publicly account for their assets and for the sources and use of their funds’. The second paragraph deals with the prohibition of parties—forbidding parties that, ‘by reason of their aims or the behaviour of their adherents, seek to undermine or abolish the free democratic basic order or to endanger the existence of the Federal Republic’.

The third paragraph, which was added in 2017, is not directly related to the prohibition of parties, but rather to the public funding of parties, which may be denied to parties ‘oriented towards an undermining or abolition of the free democratic basic order or an endangerment of the existence of the Federal Republic of Germany’. This addition is significant because it sets a lower hurdle for a party to be denied public funding than to be banned as unconstitutional. The fourth paragraph is jurisdictional. It empowers the Federal Constitutional Court to rule on such questions of constitutionality and on the withdrawal of public funding. Finally, the federal legislature is empowered and required by the fifth paragraph to regulate political parties in accordance with these provisions.

In general terms, therefore, the structure of this primer—covering roles, recognition, rights, regulations, financing and prohibition—mirrors the structure of article 21 of Germany’s constitution. This primer returns at various points to this article not because it is the only example, or even perhaps the best example, but because it is it is an illustrative and influential example and provides a baseline against which other constitutional provisions on political parties can be compared.

International standards

Relatively few international instruments place obligations on countries in relation to the regulation of political parties (Gauja 2016). However, some authorities at least recognize the role and responsibilities of parties in a healthy democracy.

• The African Union African Charter on Democracy, Elections and Governance (2012) commits its signatory states to ‘Strengthening political pluralism and recognizing the role, rights and responsibilities of legally constituted political parties, including opposition political parties, which should be given a status under national law’.
• The Inter-American Democratic Charter (2001) recognizes that ‘a pluralistic system of political parties and organizations’ is an essential element of representative democracy (article 3), and that the ‘strengthening of these organizations is a priority’, as is a ‘balanced and transparent system for their financing’ (article 5).
• The Commonwealth Charter (2013) notes that political parties, alongside government and civil society, ‘are responsible for upholding and promoting democratic culture and practices and are accountable to the public in this regard’.
• The OSCE/ODIHR and the Venice Commission’s ‘Guidelines on Political Party Regulation’ (2020) contain a number of recommendations and principles on political parties, ordered around the principles set out in Table 4.1.

Declaring the roles of political parties

The most basic way in which a constitution can recognize political parties is to mention them. Today, the terms ‘political party’ or ‘political parties’ appear in 155 national constitutions currently in force. Some also recognize the roles and functions of parties in a democracy.

• Article 21 of the German Basic Law (1949) is an early and influential example of a constitutional provision that sets out the role and functions of parties. It states that ‘Political parties shall participate in the formation of the political will of the people’.
• A similar formulation referring to ‘political will’ can also be found in the Constitution of Romania (1991, article 8), which provides that ‘Political parties are established and carry out their activity under the conditions of the law. They contribute to the definition and expression of the citizens’ political will, respecting national sovereignty, territorial integrity, the rule of law, and the principles of democracy.’
• The Constitution of Ukraine (1996, article 36) similarly states that ‘Political parties in Ukraine promote the formation and expression of the political will of citizens, and participate in elections’. Note that this formulation includes a specific reference to the role of parties in participating in elections.

Another common expression recognizing the role of political parties derives from article 4 of the French Constitution (1958), which provides that ‘Political parties and groups shall contribute to the exercise of suffrage’. This formulation has been adopted across francophone Africa, such as in Benin (1990, article 5), Burundi (2018, article 79), Comoros (2018, article 35), the Democratic Republic of the Congo (2005, article 6), Côte d’Ivoire (2000, article 14), Niger (2010 article 9) and Senegal (2001, article 4).

Both expressions—the ‘formation of political will’ and ‘exercise of suffrage’—convey the idea that political parties perform a distinct institutional role by serving as mechanisms through which the political opinions of individual citizens are ascertained, consolidated and formed, and then expressed in political institutions through electoral processes. Both recognize that representative democracy does not take place mechanically or in a political vacuum, but requires initiative and action, the creation of narratives, ideas and identities, and the building and mobilization of political support—in all of which political parties participate.

These provisions also presuppose that political parties occupy a space between the people and the state, but distinct from both; that is, they are instruments for the people, on whose behalf they coordinate public control of elected institutions and state functions. Finally, these provisions presuppose that political parties are institutions that exist for democratic purposes and are implicitly constrained by these purposes.

Prohibition of one-party regimes

Some authoritarian regimes have expressly declared themselves to be one-party states. For example, the Constitution of the Soviet Union, adopted in 1936, declared that the Communist Party was ‘the vanguard of the working people’ and ‘the leading core of all organizations of the working people, both public and state’. Only the Communist Party and the various trade unions and organizations affiliated with the Communist Party had the right to nominate candidates for office. One-party states were not limited to the former Soviet bloc. The monopolization of power at a formal, constitutional level by one party was found in many other regimes across the ideological spectrum. For example, the 1973 Constitution of Syria (article 8) declared that ‘The leading party in society and the state is the Socialist Arab Baath Party’. Kenya was formally and officially a one-party state from 1982 until 1991, although it had been a one-party state in practice since the mid-1960s.

In addition to these formally recognized one-party regimes, other such regimes may be instituted on an informal basis, behind a facade of democracy. Other parties may be permitted and even officially encouraged so long as they work for the regime, either collaborating with the ruling party or by offering no more than nominal opposition that gives the superficial appearance of political pluralism and competition without its substance. Various forms of coercion and corruption have been used to reinforce the position of the ruling party and to keep opposition parties in a subordinate position.

Democratic constitutions have taken various steps to prevent one-party regimes:

• In Liberia, where the True Whig Party established an informal one-party state for more than 100 years, the 1986 Constitution (article 77) states that ‘Laws, regulations, decrees or measures which might have the effect of creating a one-party state shall be declared unconstitutional’.
• In the Democratic Republic of the Congo, in reaction to the one-party state of Mobutu Sese Seko, the Constitution (2005, article 7) prohibits the creation of a one-party state and declares it to be high treason.
• The Constitution of Bulgaria (1991, article 11) provides that ‘Political activity … shall be founded on the principle of political pluralism’ and that ‘No political party or ideology shall be proclaimed or affirmed as a party or ideology of the State’. In other words, the neutrality of the state in relation to political parties is affirmed.
• In Moldova, the Constitution (1994, article 2) attempts to preserve the non-partisanship of the state—the separation of state and party—by declaring that ‘Neither an individual person or a group of people, nor a social group, a political party or any other public organization may exercise State power on their own behalf’.

Guarantee of multiparty democracy

Beyond simply prohibiting one-party regimes, a constitution can make a positive commitment to multiparty democracy.

• In South Africa, the principle of ‘a multiparty system of democratic government’ is placed at the head of the Constitution (article 1). This principle is harder to amend than the other parts of the Constitution and can be regarded foundational to the meaning of the Constitution as a whole.
• In Kenya, the Constitution (article 4) declares that ‘The Republic of Kenya shall be a multi-party democratic State’.
• A similar approach is taken in Croatia, which declares that a ‘democratic multiparty system’ is among ‘the highest values of the constitutional order of the Republic of Croatia and the ground for interpretation of the Constitution’ (article 3).

Such provisions may be particularly helpful in situations where multiparty democracy is being re-established after a period of one-party rule. As mentioned above, Kenya was formerly (1982–1991) a one-party state in which an official monopoly on power by one party was recognized. It is rarely enough in such circumstances merely to repeal the provisions establishing a one-party state, although that is the approach taken by Bangladesh. Rather, it makes sense from a declaratory point of view to make an overt commitment to multipartyism.

Recognition of parties in the legislature

Once parties are sufficiently successful to win more than two or three seats in the legislature, they tend to form organizations. The most fundamental of these is a meeting of all legislators belonging to that party, generally known as the party’s ‘legislative caucus’, although other terms such as ‘faction’ or ‘parliamentary party’ are used in some contexts. The legislative caucus and its role in the internal structure of parties is discussed further in Chapter 6. There are usually also party whips—members of the legislative caucus who act as the hands, eyes and ears of the party leader, instruct members on how to vote, enforce party discipline and act as a means of communication between the legislative caucus and the party leader.

For the most part, these internal party organizations within the legislature were not recognized in the earliest constitutions and existed as informal—albeit often well-established—institutions. This has gradually changed. In keeping with the progressive expansion of the recognition, rights and regulation of parties, in terms of their national organizations and electoral functions, there has been a parallel expansion of the constitutional roles of parties within legislatures.

The recognition of parties within the legislature is particularly relevant in parliamentary systems, where they play a crucial role in government formation. Constitutions often refer to the appointment of the leader of the majority party, or coalition of parties, as prime minister—a provision that, incidentally, presupposes the existence not only of parties but also of recognized party leaders.²

• The Constitution of Antigua and Barbuda (section 69) states that, ‘Whenever there is occasion for the appointment of a Prime Minister, the Governor-General shall appoint as Prime Minister a member of the House who is the leader in the House of the political party that commands the support of the majority of members of the House’.
• The Constitution of Greece (article 37) provides that ‘the leader of the party having the absolute majority of seats in Parliament shall be appointed Prime Minister. If no party has the absolute majority, the President of the Republic shall give the leader of the party with a relative majority an exploratory mandate in order to ascertain the possibility of forming a Government enjoying the confidence of the Parliament.’
• The Constitution of Kosovo (article 84) requires the President to appoint the prime minister ‘after proposal by the political party or coalition holding the majority in the Assembly’.

Aside from these provisions, constitutions may also recognize the existence of parties in the organization of the work of the legislature. The 2015 Constitution of Armenia, for example, provides (article 105 and article 106) that ‘The factions [including Deputies of the same party] shall contribute to the formation of the political will of the National Assembly’, that ‘Seats within standing committees shall be allocated in proportion to the number of Deputies included in factions’ and that ‘The positions of chairpersons of standing committees shall be distributed among factions in proportion to the number of Deputies included in the faction’. This gives parties official recognition in the legislature and approximate proportionality of influence.³

Some countries provide for constitutional recognition of majority and minority leaders in the legislature, or a leader of the opposition, which is also a way of formalizing the roles of party caucuses and their leaders. The Constitution of Kenya (2010, article 108), for example, states that ‘There shall be a leader of the majority party and a leader of the minority party. The leader of the majority party shall be the person who is the leader in the National Assembly of the largest party or coalition of parties. The leader of the minority party shall be the person who is the leader in the National Assembly of the second largest party or coalition of parties.’

In addition to recognizing their role and functions, constitutions can give political parties rights to exist and to carry on their activities. Constitutions can also give citizens particular rights in relation to political parties: the right to join, the right to take part in partisan activities and the right not to be discriminated against on the basis of their party membership or activities.

General freedoms of expression, association and assembly

Democratic constitutions create a permissive legal environment for the creation of political parties: the freedoms of expression, association and assembly enable political parties to exist and to function. The rights to form and to join political parties, and for political parties to engage in political expression and assembly, flow by necessary implication from these constitutional rights. These rights are upheld and guaranteed by the courts. However, such provisions treat political parties no differently, in essence, from other forms of private association. The same limitations clauses that apply to the rights to expression, association and assembly generally also apply to political parties.

Specific rights to form and join parties

An alternative is to expressly entrench the rights to form and to join political parties. These rights exist alongside and in addition to the general freedoms of expression, association and assembly. This gives political parties some constitutional recognition and acknowledges them as a particular form of association that is sufficiently important to democratic political life to require specific constitutional protection.

This right can be expressed as an individual right to form or join political parties.

• The Constitution of The Bahamas (1973), for example, states that ‘no person shall be hindered in the enjoyment of his freedom of peaceful assembly and association, that is to say, his right to assemble freely and associate with other persons and in particular to form or belong to political parties, or to form or belong to trade unions or other association for the protection of his interests’ (section 24).
• The Constitution of the Solomon Islands (1978, section 13(1)), similarly provides that, ‘Except with his own consent, no person shall be hindered in the enjoyment of his freedom of assembly and association, that is to say, his right to assemble freely and associate with other persons and in particular to form or belong to political parties or to form or belong to trade unions or other associations for the protection of his interests’.
• The Constitution of Romania (1991, article 40(1)) provides that ‘Citizens may freely establish and adhere to political parties’.

This right can also be expressed as a right of parties to be freely formed.

• The German Basic Law (1949, article 21) states that political parties ‘may be freely established’.
• The Constitution of Poland (1997, article 11) makes establishing the free formation of political parties a duty of the republic, which ‘shall ensure freedom for the creation and functioning of political parties’.

Right to participate in partisan activities

The above formulations do not spell out which activities individuals can engage in once they have formed or joined a political party. Nor do they—beyond the general freedoms of expression, association and assembly—say what members of parties may actually do. To rectify this, some modern constitutions—mainly, incidentally, in Africa—protect political participation by party members and campaigners.

• The Constitution of South Africa (1996, article 19(1)) provides that ‘Every citizen is free to make political choices, which includes the right: to form a political party; to participate in the activities of, or recruit members for, a political party; and to campaign for a political party or cause’. The Constitution of Kenya (2010, article 38(1)) uses the exact same wording.
• The Constitution of Malawi (1994, article 40) is also similar: ‘Subject to this Constitution, every person shall have the right to form, to join, to participate in the activities of, and to recruit members for, a political party; to campaign for a political party or cause; to participate in peaceful political activity intended to influence the composition and policies of the Government; and freely to make political choices.’

This is an additional level of protection for the right of political parties not only to exist but also to freely carry out their activities and perform their essential functions in a democracy.

Alternatively, certain rights may be granted to political parties in their corporate capacity. For example:

• The Constitution of Ghana (1992, article 55) provides that ‘a political party is free to participate in shaping the political will of the people, to disseminate information on political ideas, social and economic programmes of a national character, and sponsor candidates for elections’.
• The Constitution of Zambia (1991, article 60) provides that ‘A political party has the right to: (a) disseminate information on social and economic programmes of a national character and its political ideology; (b) sponsor candidates for election or nomination to a State office in respect of which elections are required to be held; and (c) conduct primary elections for the selection of candidates’.

Non-discrimination on the basis of party affiliation

An important consideration alongside the freedom to join (or not to join) a political party is the right to be free from discrimination on the basis of party membership or affiliation. In many authoritarian or semi-authoritarian regimes, supporters of the ruling party are given preferential treatment by the government. Party membership might be a precondition for certain types of public sector employment, or those who are not members of the ruling party may be discriminated against in matters such as access to educational opportunities.

To prevent such discrimination, party membership or affiliation can be made a constitutionally protected characteristic for the purposes of anti-discrimination law. This is usually framed in terms of preventing discrimination on the grounds of ‘political opinions’.

• The Constitution of Belize (1981, section 16), for example, declares that ‘no person shall be treated in a discriminatory manner by any person or authority’ on the grounds, among other things, of their ‘political opinions’.
• The German Basic Law (1949, article 3) states that ‘No person shall be favoured or disfavoured because of … political opinions’.
• The Constitution of Nigeria (1999, section 42) provides that ‘A citizen of Nigeria of a particular … political opinion shall not, by reason only that he is such a person be subjected either expressly by, or in the practical application of, any law in force in Nigeria or any executive or administrative action of the government, to disabilities or restrictions to which citizens of Nigeria of other … political opinions are not made subject; or be accorded either expressly by, or in the practical application of, any law in force in Nigeria or any such executive or administrative action, any privilege or advantage that is not accorded to citizens of Nigeria of other … political opinions’.
• The Constitution of the Netherlands (1983, article 1) states that ‘Discrimination on the grounds of … political opinion … shall not be permitted’.
• The Constitution of Angola (2010, article 17) requires equal treatment of parties by the state and stipulates that ‘Political parties shall be entitled to equal treatment by entities exercising political power, impartial treatment by the state press and the right to exercise democratic opposition, under the terms of the Constitution and the law’.

Limits on partisan activities for certain office bearers

An important consideration is whether the rights to join political parties and to participate in party activities should be limited in the case of certain officials who are supposed to be impartial, notably judges, civil servants, military officers and the staff of independent institutions such as an electoral commission or boundaries commission. The integrity and credibility of such officials and institutions are essential to allow them to perform their proper functions in a democracy. This requires them to be non-partisan. Authoritarian regimes, moreover, typically try to pack these institutions with party loyalists, as an instrument of patronage and as a means of extending control.

Therefore, paradoxical though it might seem, the defence of democracy may require the rights of certain individuals to take part in partisan activities to be curtailed in the public interest. The extent of such curtailment, however, is a matter of proportion. In a famous case from Antigua and Barbuda (De Frietas v The Permanent Secretary, 1998), a civil servant was removed from office for political activity following their participation in a demonstration against government corruption, on the grounds that this action violated a prohibition on political activities under section 10 of the Civil Service Act. The Judicial Committee of the Privy Council (the apex court of Antigua and Barbuda) found that the removal was unlawful, however, because section 10 of the Civil Service Act was phrased too broadly, and because the blanket prohibition on all political activities by all civil servants was too much of a sacrifice of the freedom of expression.

The Constitution of Kenya, as noted above, protects the rights of all citizens to be members of political parties and to take part in partisan activities—and this includes civil servants. Civil servants are only prohibited from ‘holding office’ in a political party. Examples of such provisions in relation to judges include the following:

• Slovakia’s Constitution (article 137(1)) states that, ‘If an appointed judge of the Constitutional Court is a member of a political party or a political movement, he must surrender his membership prior to taking his oath’.
• Austria’s Constitution (article 147(4)) notes that ‘persons who are in the employ of or hold office in a political party cannot belong to the Constitutional Court’.
• In Ecuador, ‘Judges cannot hold any executive office in political parties and movements or participate as candidates in elections by universal suffrage or carry out activities of political or religious solicitation’ (Constitution of Ecuador, article 174).

Note the difference between these formulations. The ban in Slovakia applies to all members of a political party, whereas in Austria it applies only to employees and office bearers of a political party, and in Ecuador it applies to holders of executive office in parties.

In addition to the civil service and the judiciary, there are often specific rules on partisan membership or activities by members of the military or the security agencies. Partisan activities within the armed forces can give rise to military coups.

• The Constitution of Estonia (article 125) provides that ‘A person in active service [in the armed forces] shall not hold other elected or appointed office, or participate in the activities of any political party’.
• The Constitution of Paraguay (article 173) provides that ‘military [personnel] on active service will conform their actions to the laws and regulations, and they may not join any political party or movement, or realize any type of political activity’.

A constitution might include a broad provision limiting partisan activities that applies to several categories of public official. The Constitution of Romania (article 40(3)), for example, provides that ‘Judges of the Constitutional Court, people’s advocates, judges and prosecutors, active members of the Armed Forces, policemen and other categories of civil servants determined by an organic law shall not be members of political parties’. This provision is notable for two reasons. First, it prohibits membership and not just office bearing or other senior roles. Second, it allows for expansion of the categories to other civil service posts by organic law.

Alternatively, certain narrower prohibitions could be applied. The Constitution of Malta, for example, prohibits persons who are ‘officials’ or ‘candidates’ of any political party from being members of the Judicial Appointments Committee (article 96A), but members of the Electoral Commission or Public Service Commission are under no such prohibition. In the latter cases, membership of or standing for election to parliament are prohibited, but holding office in a political party is not.

In some countries, the president is barred from partisan activity. This is logical in parliamentary systems, where the cabinet is led by a prime minister and where the presidency is a mostly ceremonial office, often indirectly elected, and the incumbent is supposed to be a unifying, non-partisan figure. The Constitution of Estonia (article 84), for example, provides that, ‘Upon assuming office, the authority and duties of the President of the Republic in all elected and appointed offices shall terminate, and he or she shall suspend his or her membership of political parties for the duration of his or her term of office’.

The more executive power and policymaking authority the president has, the less logical a prohibition on the partisanship of the president becomes. To govern is to choose, and parties are useful in structuring people’s choices.

In addition to recognizing the role of political parties in a democratic system and protecting the rights of parties and their members, constitutions can also seek to regulate political parties.

This regulation can apply through the registration of political parties, the regulation of internal party democracy and party financing, and the auditing and disclosure of funds. The existence of such regulations means that parties are not entirely private bodies with internal organizations and functions that are largely beyond the scope of the legislator, but political bodies with legal responsibilities, and restrictions as well as rights.

Constitutional or statutory provisions?

An important general consideration is the extent to which such regulation can or should be directly provided for in the constitution, or set out in ordinary laws and other sub-constitutional rules. Since constitutions are usually harder to change than other laws, constitutional provisions are usually more resilient and provide greater resistance to manipulation by the party in power.

Rather than relying on ordinary laws, constitutionalization also protects regulatory principles from regressive jurisprudence that might otherwise seek to undermine them. In the United States, for example, campaign finance legislation was struck down on the grounds of unconstitutionality in Citizens United v Federal Election Commission, 558 US 310 (2010), which held that corporations have a constitutionally protected right to ‘free speech’ in the form of political donations. If the USA, like many other countries, had a constitutional rather than a statutory ban on corporate donations, this would not have occurred.

Another consideration is that constitutions have an expressive function: they say what matters and remind us of what we ought to care about. Putting certain principles and standards of party registration, organization or financing into the constitution is a way of giving them visibility, and thereby indirectly shaping the tone and direction of the public debate—and perhaps jurisprudence–on the issue.

On the other hand, detailed constitutional rules are less flexible than ordinary laws at responding to changing needs. In practical terms, it is not possible or even desirable to constitutionalize every detail. The legislature, the courts and bodies such as electoral commissions or similar must still have some discretion.

One solution is for the constitution to set out various basic standards or principles, such as fairness or the equal treatment of candidates, and for the details to be filled in by ordinary laws and other sub-constitutional rules, which must conform to those standards or principles.

• The Constitution of Ghana, for example, sets out various general rules on the organization, functioning and registration of political parties (article 55) but then provides that, subject to those provisions, ‘Parliament shall by law regulate the establishment and functioning of political parties’.
• The Constitution of Kenya similarly sets out some general principles (article 91) but then, within those principles and in order to give effect to them, gives legislative authority over most aspects of party regulation to parliament (article 92).
• The Constitution of Sierra Leone (2013, section 35(6)) makes the purposive intent of such legislation explicit, stating that, ‘Subject to the provisions of this Constitution, and in furtherance of the provisions of this section, Parliament may make laws regulating the registration, functions and operation of political parties’.

Regulatory powers in relation to parties

Regulatory power over political parties may be implicit. Broad grants of legislative power—such as the power to legislate for the ‘peace, order and good government’ of a particular territory—assume the ability of the legislature to make laws on political parties. However, if the rights of political parties to form and operate freely are explicitly guaranteed without a correspondingly explicit right on behalf of the legislature to regulate them, then legislation to regulate political parties might be struck down as unconstitutional. This would be an undesirable outcome, since it would deprive the state of a means of protecting itself against corruption. Many constitutions therefore explicitly declare the right—or even the duty—of the legislature to make laws on the regulation of political parties. As noted above, constitutions can go beyond a mere grant of power to specify the direction and purposes of such legislative rules.

The OSCE/ODIHR and the Venice Commission’s Guidelines on Political Party Regulation (2020) set out 11 principles for the exercise of this legislative or regulatory power over political parties. These principles start from an assumption of freedom of association but balance that with a recognition of the public roles and responsibilities of parties (see Table 5.1).

Registration of parties

Although the right of parties to be freely formed may be guaranteed, constitutions or laws enacted in accordance with constitutions may require the registration of political parties. Registration gives official recognition to a party. It may be necessary in order for potential candidates to get their name on the ballot, or access to public funds or other benefits, or even just to allow a party to operate.

• In Nigeria (Constitution of Nigeria 1999, article 222), ‘No association by whatever name called shall function as a party, unless the names and addresses of its national officers are registered with the Independent National Electoral Commission’.
• In Australia, a federal political party registration scheme was introduced by law to enable access to public funding, recognition on ballot papers and compliance with financial disclosure and reporting requirements (Kelly 2016).

In a survey of 37 democracies, Botelho and Garoupa (2023: 1662) found that 84 per cent had laws requiring the registration of political parties. It is less common for party registration rules to be included directly in constitutions, although there is a recent trend in that direction, which reflects a broader appreciation of the role that parties play in a democracy.

Granting or denying registration can be a means by which the responsible authorities are able to enforce constitutional regulations on parties:

• In Ghana, a political party, on seeking registration, must provide the Electoral Commission with a copy of its constitution (Constitution of Ghana 1996, article 55(7)), thereby providing a means of enforcing the requirement that ‘the internal organization of a political party shall conform to democratic principles’ (Constitution of Ghana 1996, article 55(5)).
• The Constitution of Liberia (article 79) similarly requires parties to provide the Electoral Commission with a copy of their constitution.

Registration requirements may be designed to weed out small parties or parties that have only limited or local appeal. This can be a way of consolidating the party system, encouraging catch-all and programmatic parties, and discouraging regionalist or ethnic parties, protest and single-issue parties, joke parties and anti-system parties.

• In Sierra Leone, a party must have a geographical organization, evidenced by having a registered office ‘in each of the Provincial Headquarter towns and the Western Area’ (Constitution of Sierra Leone, section 35(5)).
• In Ghana, a party cannot be registered unless ‘it has branches in all the regions of Ghana and is, in addition, organised in not less than two-thirds of the districts in each region’ (Constitution of Ghana, 1996, article 55(7)(b)).

Registration rules can be used to de facto ban a political party. If certain conditions are laid down for the registration of parties, and if registration is required in order for a candidate to appear on the ballot, then failure to comply with those conditions can effectively ban a party. Chapter 8 discusses the circumstances in which the banning of political parties is acceptable in a democracy. However, if registration requirements are made too onerous, the practical effect can be to thwart the exercise of the right to form and join political parties. Complex and expensive registration requirements—such as a need to gather a very large number of signatures in a short space of time or the need to pay large registration fees—can make the barriers to entry faced by new parties prohibitively high. Such rules can be used to undermine rather than strengthen democracy.

Party regulating bodies

Where parties are constitutionally regulated, responsibility for ensuring compliance with such regulations, and for overseeing the registration of parties and candidates, is usually entrusted to the electoral commission, electoral tribunal or similar supreme electoral management or electoral justice body. In Fiji, for example, article 75(2) of the Constitution gives the Electoral Commission responsibility for ‘monitoring and enforcing compliance with any written law governing elections and political parties’.

Sierra Leone, in contrast, restricts the Electoral Commission to matters concerning elections, and entrusts the regulation and registration of political parties to a separate Political Parties Registration Commission (Constitution of Sierra Leone 1991, article 34). However, the chief electoral commissioner, who is chair of the Electoral Commission, is an ex officio member of the Political Parties Registration Commission, so there is some overlap in membership. The Political Parties Registration Commission is responsible for ensuring that parties comply with the requirements set out in the Constitution, including the auditing of accounts (article 35(3)).

The Constitution of Kenya (article 88(4)) gives the Independent Electoral and Boundaries Commission responsibility for ‘regulation of the process by which parties nominate candidates for elections’, ‘regulation of the amount of money that may be spent by or on behalf of a candidate or party in respect of any election’, ‘the development of a code of conduct for candidates and parties contesting elections’, and ‘compliance with the legislation … relating to nomination of candidates by parties’. However, an independent Office of the Registrar of Political Parties has statutory authority to register and regulate political parties and to administer the public funding of parties.

Separating electoral management from party regulation has several advantages. First, it enables the Electoral Commission to avoid potential conflicts with political parties over issues of regulation or registration and so makes it easier for the Electoral Commission to maintain its neutrality. Second, dividing functions between two bodies provides greater resilience against attempts to ‘capture’ the political process.

Occasionally, other institutions may be involved in the regulation of political parties. For example, the Constitution of Papua New Guinea (section 129) requires parties to disclose their assets, income, sources of funds and election expenditure to ‘the Ombudsman Commission or some other authority prescribed by the law’.

Where registration of a party is denied, there is usually a means of appeal. In Liberia, for example, ‘A denial of registration or failure by the Elections Commission to register any applicant may be challenged by the applicant in the Supreme Court’ (Constitution of Liberia, article 79), while in Sierra Leone, ‘Any association aggrieved by a decision of the Political Parties Registration Commission … may appeal to the Supreme Court and the decision of the Court shall be final’ (Constitution of Sierra Leone, article 35(7)). In this way, the judiciary acts as a check and balance against the decisions of an independent fourth-branch body.

As discussed in Chapter 3, many of the democratic functions of political parties, such as educating voters through political participation and providing means for ordinary people to participate in policy formation, are most effectively performed by the mass party model. This is characterized by a large and active membership, which works through elected party officials and decision-making conferences. Active membership of a mass party can develop political skills in public speaking, campaigning and organization, and can maintain the loyalty and discipline needed to win elections and implement the party’s agreed manifesto. At their best, mass parties can be a form of democracy within democracy, as internal party democracy gives life, substance and resilience to representative government. Although mentioned in constitutions only in the most cursory terms, if at all, the institutions of internal party democracy of the major parties can be foundational to the health and quality of democracy.

Constitutional provisions on internal democracy

It is generally speaking extremely difficult for a constitution to guarantee the internal democracy of political parties. For practical reasons, and in deference to a residual respect for the traditional liberal-constitutional view that parties are essentially private bodies the internal autonomy of which should not be unnecessarily infringed by the state, many constitutions do no more than set broad boundaries for the democratic principles within which parties must organize themselves, or within which the legislature should regulate political parties.

Returning to article 21 of Germany’s 1949 Constitution, there is an attempt to constitutionally recognize such broad boundaries. It provides, in highly general terms, that the ‘internal organization’ of parties ‘must conform to democratic principles’. Other, similarly generic, examples of constitutional provisions on internal party democracy reference general ‘principles’ rather than establish specific procedural rules:

• ‘Internal organization of political parties shall be in accordance with fundamental constitutional principles’ (Constitution of Croatia 1991, article 6).
• ‘The internal structure and the functioning of political parties shall be based on democratic principles’ (Constitution of Panama 1972, article 138).
• ‘The internal structure and the operation of political parties shall be democratic’ (Constitution of Mozambique 2004, article 74).

What is meant by ‘conforming to democratic principles’ varies greatly between countries and between parties. There is no standard formula. If apex courts are called on to decide what meets the standard of ‘democratic principles’ without a clear legislative framework to rely on, there is a chance that the judiciary will be brought into situations of intense political controversy.

To avoid the judiciary having to operate in a legislative vacuum, some constitutions mandate legislatures to enact laws that establish rules on internal party democracy. For example, the Constitution of Nigeria (1999) states that ‘The National Assembly may by law provide guidelines and rules to ensure internal democracy within political parties, including making laws for the conduct of party primaries, party congresses and party conventions’ (article 228).

This approach has also been adopted in Nepal, where the Constitution requires political parties, on registering with the Election Commission, to submit a copy of the party’s constitution, which must provide for ‘the election of the office bearers at the federal and provincial levels, at least once in five years’ (article 269(4)(b)).

Some constitutions take a more specific approach. The Constitution of Colombia (1991, articles 107 to 109), for example, makes detailed rules on the internal democratic processes of parties, including provisions for the conduct of primary elections, and stipulates that, ‘In the case of popular ballots the rules relating to campaign financing and advertising and access to the State media which govern the ordinary elections shall apply. Those who participate in the ballot of a party or political movement or in inter-party elections may not register for another one in the same electoral process. The result of the ballots shall be binding.’

The Constitution of Ghana (1996, article 55(8)) requires that ‘A political party shall not have as a founding member, a leader or a member of its executive, a person who is not qualified to be elected as a member of Parliament or to hold any other public office’. It also demands that ‘The members of the national executive committee of a political party shall be chosen from all the regions of Ghana’ (article 55(9)).

Gender and minority inclusion

Political parties may be required by law to make provision for gender inclusion in the selection of their candidates for public office. According to International IDEA’s Gender Quota Database, as of 2024, 71 countries had legislatively mandated candidate quotas. Such legislation may be vulnerable to constitutional challenge if provision is not made for it in the constitution. Belgium and Portugal are examples of countries with constitutions that make general, broad provision for the promotion of gender equality and inclusion in public office while leaving the details to ordinary legislation:

• The Constitution of Belgium (article 11bis) states that ‘The law … guarantees that women and men may equally exercise their rights and freedoms, and in particular promotes their equal access to elective and public mandates’ (emphasis added).
• The Constitution of Portugal (article 109) declares that ‘The direct and active participation in politics by men and women is a fundamental instrument in the consolidation of the democratic system, and the law shall promote both equality in the exercise of civic and political rights and the absence of gender-based discrimination in access to political office’.

Since the nomination of candidates for public office is a key function of parties, these provisions, which potentially limit by gender who can be nominated, could be understood as enabling the legislature to influence an important matter of internal party decision making.

A constitution can extend this principle to internal elections within a party. For example, article 91 of the Constitution of Kenya requires political parties to ‘respect the right of all persons to participate in the political process, including minorities and marginalized groups’. This is reinforced by section 25(2) of Kenya’s Political Parties Act 2011, which requires governing bodies of political parties to reflect gender balance with no more than two-thirds of their members being of the same gender in order for the party to receive public funds.

The funding of political parties and election campaigns is part of a wider issue concerning the role of money in politics. This is a massive, complex area of legislation and practice, which cannot be captured in this primer. The aim here is not to provide a comprehensive overview of party and election financing, but merely to draw attention to certain constitutional rules that, in some countries, set a framework for laws and policies in this area. In addition, the financing of political parties is one area that requires constant review and revision. For example, digitalization has transformed how political parties raise and spend money online, but most countries have not kept pace. For further information about party financing rules, see International IDEA’s Funding of Political Parties and Election Campaigns: A Handbook on Political Finance (Falguera, Jones and Ohman 2014) and Combatting Corruption in Political Finance: Global Trends, Challenges, and Solutions (Hamada and Agrawal 2025).

Transparency

Among the most elementary ways to regulate party financing is to require transparency: to allow the financial transactions of parties—their sources of funds and how they are spent—to be open to inspection and to the gaze of the public and the press. This means banning secret or anonymous donations, or at least those above a certain relatively low threshold. Such bans are in place in Canada, Finland, France and the Netherlands, among other places (International IDEA n.d.).

Transparency does not limit how parties obtain or spend money, but it does force them to be publicly accountable. Voters and the media can at least see which donors support the party and who might therefore be influencing the party’s actions. Various constitutional provisions require financial transparency by political parties:

• ‘Political parties must make public the source and destination of their funds and their economic net worth’ (Constitution of Argentina 1994, article 38).
• ‘Private contributions to political parties will be submitted to the principle of publicity and will be regulated by law’ (Constitution of Costa Rica 1949, article 96).
• ‘The law shall lay down the rules governing the financing of political parties, particularly as regards … the requirements to publicise their assets and accounts’ (Constitution of Portugal 1976, article 51(6)).

Transparency requires not only an obligation on the parties to publish their accounts, but also a power by the relevant public authorities to audit those accounts. For example, in Liberia, the Constitution states that ‘The Elections Commission shall have the power to examine into and order certified audits of the financial transactions of political parties and independent candidates and their organizations. The Commission shall prescribe the kinds of records to be kept and the manner in which they shall be conducted by a certified chartered public accountant, [who is] not a member of any political party’ (article 82).

Donation rules

Beyond requirements on transparency in donations, constitutions may also prescribe criteria on how much money can be given to parties and the sources of donations.

Donation limits: The usual practice is for the constitution not to set a fixed cap on donations, but to grant the legislature the power to make provision by ordinary law. The Constitution of Colombia, for example, states that ‘the maximum amount of private contributions may also be limited in accordance with the applicable statute’ (Constitution of Colombia 1991, article 109).

Ban on corporate donations: Some countries ban corporate donations. This reflects the idea that political expression is a human right, not a right of corporate ‘legal persons’. Corporate donations give the very rich—who control corporations—disproportionate political influence compared to ordinary voters. They might also swing policies or contracting decisions in favour of that corporation. Banning corporate donations is therefore a way to help resist the oligarchic distortion of democracy. The Constitution of Liberia, for example, bans political donations from corporate and business organizations (article 82), but it also bans labour unions from contributing to political parties.

Ban on foreign donations: Some constitutions ban foreign donations or foreign sources of funds:

• ‘No political party shall hold or possess any funds or other assets outside Nigeria; or be entitled to retain any funds or assets remitted or sent to it from outside Nigeria’ (Constitution of Nigeria 1999, article 225).
• ‘Political parties that accept aid from foreign states, international institutions and persons and corporate bodies of non-Turkish nationality shall be dissolved permanently’ (Constitution of Turkey 2017, article 69).
• ‘No political party or organization may hold or possess any funds or other assets outside of Liberia’ (Constitution of Libera 1986, article 82).

Such bans on foreign funding can help to protect the political system from manipulation and subversion by malign foreign interests, in particular the agents of hostile powers. They can be an acceptable protective cordon for democracy in certain environments. On the other hand, excessive fear of foreign interference can be used by the authoritarians to limit access to support, such as capacity building and technical assistance, from friendly international organizations. Section 27 of Kenya’s Political Parties Act 2011 contains a balanced provision that prohibits foreign donations but does allow foreign agencies and foreign allied parties to provide technical assistance.

Spending rules

Constitutions may limit spending by political parties and electoral campaigns, or at least lay the foundations and general principles for such limits to be set by law:

• ‘The expenses which parties, movements, relevant citizen groups or candidates may incur in election campaigns, as well as the maximum amount of private contributions may also be limited in accordance with the applicable statute’ (Constitution of Colombia 1991, article 109).
• ‘A statute shall impose the maximum limit of electoral expenses, may prohibit certain forms of pre-electoral promotion and shall specify the conditions under which violation of the relevant provisions constitutes a ground for the forfeiture of parliamentary office on the initiative of the special body of the following section. The audit of the electoral expenses of political parties and parliamentary candidates is carried out by a special body which is constituted also with the participation of senior magistrates, as specified by law’ (Constitution of Greece 1974: article 29(2)).
• In Kenya, the Independent Electoral and Boundaries Commission is responsible for ‘regulation of the amount of money that may be spent by or on behalf of a candidate or party in respect of any election’ (Constitution of Kenya 2010, article 88(4)), while the Office of the Registrar of Political Parties is responsible for the disbursement of public funds to parties and the auditing of their accounts (Political Parties Act 2011).

Note that election spending or campaign spending is not necessarily the same as party spending. Many countries with a British institutional heritage have rules with origins in the 19th century that limit the spending of individual candidates in their constituencies, but this does not necessarily translate into limits on the expenditure of national campaigns. Similarly, parties spend money on more than just election campaigns. They also spend money on policy research, for which in some countries there are dedicated foundations affiliated to, but to some degree operationally autonomous from, the party, youth rallies and maintaining their party organization and bureaucracy. When drafting constitutional provisions on the funding of parties, it is therefore necessary to ensure that all types of spending are covered, and legislative loopholes are not created.

Prohibition of vote buying

‘Vote buying’ is a general term for a range of corrupt and illicit electoral practices, such as offering cash inducements, gifts or other payment in kind in return for votes. More subtle forms of bribery designed to circumvent rules, such as ‘loans’ that are not intended to be repaid, or the awarding of administrative favours, grants or public employment to political supporters, are also forms of vote buying.⁴ Vote buying undermines both democracy and good government. Many countries therefore have specific criminal laws against it. In Ghana, for example, section 33 of the Representation of the People Act 1992 forbids various vote-buying activities.

Detailed prohibitions on vote buying are rarely expressed in constitutions but covered under general rules enabling or requiring legislatures to enact laws to ensure free and fair elections or to prohibit ‘corrupt or illegal’ practices in relation to elections. The Constitution of Malta (article 56), for instance, states that ‘The election of members of the House of Representatives shall be free of illegal or corrupt practices and foreign interference’. It gives the Electoral Commission the authority to suspend an election if it ‘has reasonable ground to believe’ that the prevalence of such practices have ‘so extensively prevailed or have been of such nature that they may reasonably be expected to affect the result of the election’.

Public funding of parties

Some countries grant political parties the right to access public funds in order to perform their functions, in the form of regular funding to support their core activities, funding tied to specific electoral campaigns or both. According to the International IDEA Political Finance Database, more than 70 per cent of countries have provision for public funding of some sort, while 62 per cent of countries provide at least some degree of regular funding not only in the context of election campaigns.⁵ The International IDEA Handbook on Political Finance explains the arguments for and against public funding (Falguera, Jones and Ohman 2014). Rules on the public funding of political parties must consider two general principles: the Eligibility Rule and the Allocation Rule.

The Eligibility Rule determines who is eligible for public funding, what a legitimate ‘political party’ is for these purposes and the threshold in terms of size for a party to qualify for public funding. There is often a requirement that a party be registered and that it therefore meet the constitutional and/or statutory criteria for registration. Sometimes, to prevent the allocation of public funds to ghost parties with no real existence or very small personal parties, a party must also achieve a minimum base of support before being eligible for public funds.

• The Constitution of Colombia (articles 107 and 109) mandates the distribution of public funds only to parties with ‘legal personality’, which requires them to gain at least 3 per cent of the votes.
• In Malawi, the state must provide funds to ‘any political party which has secured more than one-tenth of the national vote’ in the most recent parliamentary elections (Constitution of Malawi, section 40).

The Allocation Rule determines how funds are distributed between eligible parties. Should all parties receive the same treatment or receive different amounts of funding in proportion to the size of their membership or the extent of their recent vote?

Where these principles are left to be decided by ordinary legislation, there is a risk that the incumbent majority will use its control of the legislature to manipulate the rules to its own advantage. To avoid this, constitutions may prescribe some basic principles that cannot be violated. For example, the Constitution of South Africa provides that, ‘To enhance multi-party democracy, national legislation must provide for the funding of political parties participating in national and provincial legislatures on an equitable and proportional basis’ (article 236). The question of whether the legislature’s rules are ‘equitable and proportional’ is not instinctively clear but is judicially reviewable. Further clarity and certainty can be achieved by including detailed allocation formulas in the constitution.

• In Costa Rica (Constitution of Costa Rica 1949, article 96), parties receive public funds if they receive 4 per cent of the votes or, at the provincial scale, if they elect at least one deputy. The Constitution of Costa Rica adds a further protection against majoritarian abuse of power, requiring that legislation ‘that establishes the procedures, means of control and other regulations for the application’ of the constitutional article on the public funding of parties must be approved by a two-thirds majority in the legislature (article 96).
• The Constitution of Mexico (1917, article 41) specifies that the total funding available to parties should be equal to the product of the number of voters and 65 per cent of the daily minimum wage; of this, 30 per cent is distributed equally among parties, and 70 per cent is distributed on the basis of vote share.

The interaction of the Eligibility Rule and the Allocation Rule determines who gets what. A system for the public funding of parties can distribute funds in various ways, but the rules must be seen to be fair and equitable if the system is to have broad public acceptability and cross-party support. It is important that the rules do not unfairly advantage any one party, and that they do not create a cartel of parties insulated from accountability or competition.

Broadcasting and other benefits in kind

Some countries give political parties various benefits in kind, the most important of which is access to public broadcasting. Despite the rise of the Internet, broadcast media outlets remain, in many countries, an essential mechanism for reaching the public at scale.

In some countries, political airtime is a commodity that can be bought and sold, which gives a louder voice to better-funded candidates. Elsewhere, constitutional provisions seek to level the playing field and to ensure that all parties have a chance to make their voices heard.

• The Constitution of Angola (2010, article 45) provides that ‘During general and local elections and referendums, candidates shall have the right to broadcasting time on state radio and television stations in accordance with the scope of the election or referendum, under the terms of the Constitution and the law’.
• In Brazil, ‘Political parties have the right to resources from party funds and to free radio and television time, as provided by law’ (Constitution of Brazil 1988, article 17).
• In Portugal, ‘Political parties … shall, in accordance with their size and representativity and with objective criteria that shall be defined by law, possess the right to broadcasting time on the public radio and television service’ (Constitution of Portugal 1976, article 40). This is ‘apportioned in accordance with each party’s proportional share of the seats in the Assembly’. The opposition parties also have a ‘right to reply’ to the government’s political statements, which must be given the ‘same duration and prominence’ as the government’s broadcasts.
• The Constitution of Ghana (1991, article 55) provides that ‘The State shall provide fair opportunity to all political parties to present their programmes to the public by ensuring equal access to the state-owned media. All presidential candidates shall be given the same amount of time and space on the state-owned media to present their programmes to the people.’
• The Constitution of Mexico (1917, article 41, III(A)) has detailed rules and formulas for the allocation of broadcasting rights to political parties. It also prohibits private individuals and legal entities from buying airtime on television or radio for partisan purposes. This takes the distribution of broadcast campaigning outside the sphere of the market and places it under equitable constitutional control. The scale of fundraising seen in US presidential elections, much of which is needed to buy broadcasting airtime, should be unnecessary under such rules.

Democratic constitutions must protect the rights of political parties and ensure free electoral competition between them. However, some constitutions also seek to forbid certain types of parties that are deemed to be dangerous to democracy itself. This is obviously a paradox, and one that gives rise to various theoretical arguments and counterarguments. In constitutional design, however, the question of whether to forbid certain parties, or types of parties, must be answered pragmatically, with reference to the effect on the quality and resilience of democracy.

Prohibition of anti-democratic parties

Article 21 of Germany's Constitution provides the paradigmatic example of constitutional rules banning anti-democratic parties. To recap, it states that ‘Parties that, by reason of their aims or the behaviour of their adherents, seek to undermine or abolish the free democratic basic order or to endanger the existence of the Federal Republic of Germany shall be unconstitutional’. It further provides that the Federal Constitutional Court must rule on the question of unconstitutionality (article 21(4)). This provision has several features:

• First, it applies to both the aims of political parties and to the behaviour of their adherents. In other words, a party might be unconstitutional because of its ideology and intent, or because of the way its leaders, members or supporters conduct themselves. The point is that to be a threat to the democratic constitutional order, a political party need not necessarily have an overly anti-democratic or anti-constitutional ideology. It can also be a threat if its behaviour—such as intimidating opponents or using street violence—is anti-democratic.
• Second, it protects not democracy as an abstract ideal but the more pragmatic and grounded ‘free democratic basic order’ established by the constitution. This has been interpreted to include the inviolability of human dignity (article 1), as well as democracy, federalism, the separation of powers and constitutional supremacy (article 20).
• Third, it provides a mechanism for the removal of public funding for a party if its ‘aims or the behaviour of their adherents, are oriented towards an undermining or abolition of the free democratic basic order or an endangerment of the existence of the Federal Republic of Germany’. Thus, the threshold for removal of funds is subtly lower than the threshold for declaring a party unconstitutional, which enables a stepped approach.
• Fourth, it vests the Federal Constitutional Court with the legal power to determine whether parties are unconstitutional, and also with the power to determine whether, even if a party is not unconstitutional, they should lose access to public funds.

Article 21 of the German Constitution is also interesting for what it does not say. It does not go into detail about the process of initiating a challenge to the constitutionality of a political party. This is provided by ordinary legislation, enacted in accordance with paragraph 5 of the article. The move to ban a political party can be initiated by the Federal Government, the federal lower house (Bundestag) or the federal upper house (Bundesrat). Where parties are active in only one state, the State Government can also initiate the process (Kehlbach 2024). The proceedings comprise two stages: preliminary proceedings to determine whether the application to ban a party is well-founded and not frivolous or vexatious, and a decision by the Constitutional Court on whether the party is unconstitutional. As an additional safeguard, a decision to declare a party unconstitutional must be backed by a two-thirds majority of the judges hearing the case (Bundesverfassungsgericht n.d.).

In other words, although the final decision on whether to ban a party in Germany is judicial, the decision on whether to institute proceedings against a party is always political, rather than legal, in nature. A party that is in government or that has a majority—or can prevent any other parties from having a majority—in either house of the Federal Parliament is in a position to stop any investigation into its constitutionality from reaching the Federal Constitutional Court. The process is also political in a broader sense because those initiating an attempt to ban a party must consider the political consequences of their actions. Banning a small fringe party on the grounds of unconstitutionality might be accepted without repercussions, but if a party has established a certain size and following, banning it could cause a crisis of legitimacy in the political system as a whole. On the other hand, in order to ban a party, the courts must be convinced there is a likelihood that the party will achieve its aims (Bundesverfassungsgericht n.d.). This acts as something of a catch-22, as a party that is too small to be taken seriously cannot be banned on legal grounds, while a party that is too big to be ignored is hard to ban for political reasons.

In consequence, only two parties have been banned under Germany’s article 21—a party of the far right in 1952 and the Communist Party in 1956. Two attempts were made to ban the far-right National Democratic Party (NDP), in 2001 and 2017, but these failed. In 2017, the Federal Constitutional Court ruled that the NDP was ‘aimed at abolishing the existing free democratic basic order’ but could not be declared unconstitutional because ‘there was no indication that it would succeed in achieving its anti-constitutional aims’ (Bundesverfassungsgericht 2017).

The political consequences of trying to ban a party can be seen in Türkiye. In 2002 and 2008, an attempt was made to ban the Justice and Development Party (AKP) on the grounds that it threatened the secularity of the Turkish republic as guaranteed by the constitution. Although these attempts were ultimately unsuccessful, they generated resentment and inadvertently helped the AKP to consolidate its power. Thus, banning an anti-democratic political party might be an effective means of preserving democracy if employed early, but such an attempt can be an escalatory measure if it is left too late.

Other examples of constitutions prohibiting anti-democratic parties include the following:

• The Constitution of Armenia (2015, article 46.4) states that ‘Parties that advocate the violent overthrow of the constitutional order or use violence for overthrowing the constitutional order shall be unconstitutional and are subject to prohibition by decision of the Constitutional Court’.
• The Constitution of Estonia (1992, article 48) states that ‘Organizations, unions, and political parties whose aims or activities are directed at changing the constitutional order of Estonia by force, or are otherwise in conflict with the law providing for criminal liability, are prohibited. Only a court may terminate or suspend the activities of, or fine, an organization, union or political party for a violation of the law.’
• The Constitution of Poland (1997, article 13) provides that ‘Political parties and other organizations whose programmes are based upon totalitarian methods and the modes of activity of Nazism, fascism and communism, as well as those whose programmes or activities sanction racial or national hatred, the application of violence for the purpose of obtaining power or to influence the State policy, or provide for the secrecy of their own structure or membership, shall be forbidden’.

Australia, in contrast, does not ban anti-democratic parties. In 1951, at the height of the Cold War and in the midst of the ‘Red Scare’—a fear of Communist subversion of western democracies—Australia held a referendum to amend the constitution to allow the prohibition of the Communist Party. This was defeated narrowly at the national level and failed to win approval in the required four out of six states. The terms of the proposed amendment were quite different from those of outlined in article 21 of Germany's Constitution. First, the German Constitution does not seek to ban specific parties. Rather, it sets out what a party must not be or do, and any party that breaks those rules can be banned. In Australia, in contrast, there was a targeted ban against a specific party (the Communist Party) and its associated ideological movement. Second, the proposed Australian ban was broader in scope in terms of its effects. It would have applied not only to a party as a collective entity but also to supposed members or supporters of the party, who could at the Government’s discretion be declared Communists and banned from public office and from trade union membership. The toll, in terms of the potential sacrifice of personal liberty, was therefore much higher than in Germany.

Even so, the rejection of the proposed anti-Communist amendment in Australia shows two contrasting approaches to or understandings of the relationship between democracy, individual liberty and the right of anti-democratic parties to exist. In Germany, coming out of totalitarian dictatorship, the constitution-makers in 1949 adopted a ‘militant democracy’ approach that required anti-democratic parties to be banned in the name of democracy. In Australia, which had not experienced an existential threat to democracy from totalitarian movements, democracy was understood as including the right of people to vote for anti-system parties, even Communists at the height of the Cold War, if they wished. In the German system, the bounds of what is politically acceptable are set, in accordance with the Constitution, by the Federal Constitutional Court. In Australia, they are set ultimately by the people at the ballot box. In practice, however, the difference is perhaps not so great. Even in the German model, as has been shown, a party that is sufficiently popular can be very difficult to ban, both procedurally and in terms of the political consequences. In both cases, if people persistently want to vote away their democracy by putting anti-democratic parties in power, they can. Democracy can be sustained only if the demos is democratic. Italy and Portugal provide other examples of bans targeted at a specific party or type of party—fascist parties (see Box 8.1).

Prohibition of paramilitary parties and activities

Constitutions may also expressly prohibit parties—or any association—of a paramilitary nature. This is an important limitation that separates political parties, which compete for power at the ballot box, from paramilitary forces, which compete for power through armed struggle. In an established and peaceful democracy, this should go without saying. However, it is possible for paramilitary movements to arise, unless prohibited, even in established democracies. In the UK, the Public Order Act of 1936, passed in response to the rise of violent fascist street marches, prohibits political parties from wearing uniforms and carrying out paramilitary activities. Constitutional examples of such prohibitions include the following:

• The Constitution of Paraguay (article 42) provides that ‘Secret associations and those of a paramilitary character are prohibited’.
• The Constitution of Mozambique (article 52(3)) provides that ‘Armed associations of a military or paramilitary nature, as well as associations that promote violence, racism, xenophobia or pursue aims that are against the law, shall be prohibited’.
• The Constitution of Portugal (article 46(4)) provides that ‘Armed associations, military, militarised or paramilitary-type associations … shall not be permitted’.

These prohibitions extend not only to parties but also to other groups or associations. They are, in effect, limitations on freedom of association, as well as aspects of party regulation.

Prohibition of parties on grounds of national security

Some countries are very concerned about challenges to their sovereignty. There is a fear of foreign incursions into or manipulation of their politics. Parties that are controlled or influenced by outside forces are treated as a potential threat to national security. This may give rise to a desire to ban such parties. In Pakistan, for example, parliament may subject political parties to ‘reasonable restrictions imposed by law in the interest of the sovereignty or integrity of Pakistan’. The Federal Government may seek to ban a party that is ‘operating in a manner prejudicial to the sovereignty or integrity of Pakistan’ and may ‘refer the matter to the Supreme Court whose decision … shall be final’ (Constitution of Pakistan, article 17(2)).

Prohibition of ethnic, religious or secessionist parties

In some countries, parties are based—whether formally or informally—on appeal to specific ethnic, linguistic, religious, territorial or tribal communities, and focus on allocating resources and patronage to their own groups. In such circumstances, elections are primarily expressions of communal identity and lose their effectiveness as mechanisms for deciding policy and holding governments to account. In fragile states, such parties may also weaken democratic institutions by stoking inter-communal tensions, increasing the risk of electoral violence or seeking extra-constitutional means to promote their particular interests. For this reason, some countries ban parties that organize or mobilize on the basis of race, region, religion, tribe, ethnicity and/or language. For example:

• The Constitution of Angola (2010, article 17) requires parties to have a ‘national character and scope’.
• The Constitution of Ghana (1992, section 55) provides that ‘Every political party shall have a national character, and membership shall not be based on ethnic, religious, regional or other sectional divisions’.
• The Constitution of Guyana (1980, article 160A) provides that ‘All persons, institutions and political parties are prohibited from taking any action or advancing, disseminating or communicating any idea which may result in racial or ethnic division among the people’ and requires parliament to make laws ‘preventing or disbarring any person or political party from contesting any election’ if they break these prohibitions.
• The Constitution of Sierra Leone (section 35(5)) prohibits the registration of a political party if ‘membership or leadership of the party is restricted to members of any particular tribal or ethnic group or religious faith’ or if it is ‘formed for the sole purpose of securing or advancing the interests and welfare of a particular tribal or ethnic group, community, geographical area or religious faith’. It also prohibits parties from using any tribal, ethnic or religious symbolism.
• The Constitution of Zambia (1991, article 60) requires parties to ‘have a national character’ and to ‘uphold national unity’.

At the same time, political parties representing a particular community can enable that community to participate directly in the political system, thereby conferring legitimacy on the political system by making it more responsive to diverse interests. For example, the existence of the Bloc Québécois in Canada has forced the Canadian political actors to pay particular attention to the specific needs and interests of Quebec and its French-speaking inhabitants. Cutting off that form of democratic expression by banning parties organized on demographic lines could harm the quality of democracy. The use of such provisions therefore requires careful consideration. In particular, care must be taken not to ban democratic parties advocating autonomy or peaceful secession if such a ban would lead to more violent alternatives.

Preventing the abuse of party banning

The ability to ban political parties is a very powerful tool in the hands of the state. Even if we accept the case for enabling the prohibition of anti-democratic parties, safeguards are necessary to prevent this power from being abused.

The Venice Commission’s ‘Guidelines on Party Political Regulation’ (OSCE/ODIHR and Venice Commission 2010) state that restrictions on parties ‘must have their basis in the law of the state constitution or parliamentary act’, must ‘conform to relevant international instruments’ and should be ‘clear, easy to understand, and uniformly applicable’. Furthermore, they must be ‘necessary in a democratic society’, ‘effective at achieving their specified purpose’ and ‘proportionate in nature’, and there should be a ‘presumption in favour of non-dissolution’. These Guidelines were developed based on European experiences and are not necessarily applicable in other contexts. However, they do recognize the need for balance and caution in exercising the power to ban parties.

One of the most common and important safeguards is that the decision to prohibit a party should be a judicial one taken by a senior court, not an administrative decision taken by the executive or a political decision taken by the legislature. However, this requires a neutral and independent judiciary. Semi-authoritarian regimes could use a ‘captured’ judiciary as a weapon.

The bottom line is that all these provisions are attempts to reconcile the paradox of democracy: that democracy can be destroyed if people vote for anti-democratic parties, and that restraints—including the banning of anti-democratic parties—might be necessary to save democracy from its own moments of weakness. However, such power, if granted and not well regulated, can also be used and abused by anti-democrats to shut down potential opposition.

This chapter discusses party discipline and whether constitutions should seek to reinforce it by means of ‘anti-defection clauses’. An anti-defection clause is a constitutional rule that causes members of a legislature to forfeit their seat if they ‘defect’ from their political party. Defection from a party is also known as ‘crossing the floor’ or ‘party switching’.

Where are anti-defection clauses found?

Anti-defection clauses are ‘more common in nascent democracies than in established democracies’ (Janda 2009: 5): 24 per cent of ‘new democracies’ have anti-defection clauses, whereas they are found in just 14 per cent of ‘older democracies’ (Janda 2009). ‘Semi-democratic’ regimes are most likely to have anti-defection clauses (33 per cent of them do so). This distribution appears logical, as anti-defection clauses might help to stabilize democratic party systems in new democracies, but their negative effects in terms of weakening legislatures and accountability mechanisms could trap these countries in a semi-democratic stage of development.

Anti-defection clauses are more often present in Commonwealth countries. Comparative studies have found that 58 per cent of Commonwealth countries had an anti-defection clause (Malhotra 2006), as opposed to just 14 per cent of countries globally (Janda 2009). However, in the oldest Commonwealth jurisdictions—Australia, Canada, New Zealand and the UK—loss of party membership, either by floor crossing or expulsion, does not result in legislators losing their seats.

Relevant principles: Party discipline and its limits

In most democracies, candidates who have been elected as a member of a party are usually expected to vote with that party and to support the party and its leader in public. Such ‘party discipline’ involves the subordination of a legislator’s individual preferences and priorities to those of the party, as determined by the party’s conference, party leadership or legislative caucus.

Some degree of party discipline is healthy and indeed essential. Without it, parties would be incapable of performing their democratic functions. It would be hard for voters to know what they had voted for, and there would be less connection between votes cast and policy outcomes.

On the other hand, excessive party discipline can be dangerous for democracy. If party discipline is too strict, members’ personal consciences, their political judgement and their ability to represent a specific constituency would all be undermined. Above all, members would lose the ability to hold the leadership of their party to account or to restrain it when it goes astray. Autocracy in the governing party can easily slide into autocracy in the state.

In a healthy democracy, there needs to be a balance between sufficient party discipline to ensure coherence, effective government and collective responsibility, and sufficient room for dissent when a member cannot, in good conscience, vote with their party. This might be because of irreconcilable differences of opinion or a particular constituency interest. Party loyalty should be the norm and the general expectation, but it should also be flexible and not require blind, unthinking obedience to the whips in all circumstances.

The need to strike a balance between these two principles is not normally addressed directly in the text of a constitution. Rather, a balance is normally worked out through informal practices, internal party rules and political pressures.

In many countries, a parliamentarian who regularly votes against his or her own party might be ostracized, harm their chances of promotion, be suspended from the whip or risk deselection by the party at the next election. In Westminster Model democracies, there is a convention that a minister who votes against the government must resign from ministerial office, although, in the absence of any anti-defection clause, they remain a member of parliament. Going against the party line can therefore be personally and politically costly, and individual members will have to weigh those costs before voting against the party.

Occasionally, members might even change parties—abandoning the party under whose banner they were elected, either to sit as an independent or to cross the floor to join another party. Historically, floor crossing has been an important mechanism for legislators to check the power of party leaders: the threat of ‘exit’ creates credible political pressure.

In other countries, however, party allegiances are more fluid. Opposition members might be seduced by the government in order to gain ministerial office or other personal benefits, or to secure benefits for their constituents, which in turn increases their chance of re-election. In some countries, moreover, having a member of parliament (MP) who is in the Cabinet is seen as beneficial to the voters in their constituency, since that MP will be in a position to channel public sector jobs, contracts, and development and infrastructure projects to that constituency (Vernon 2022).

Causes and types of defection

Table 9.1. identifies two types of defection. Defections of conscience are usually from the government to the official opposition, or from the majority to the minority. Typically, these will be caused by members of the governing party who disagree with the leadership on a matter of conscience or constituency interest, and decide that they can no longer support it. Defections of ambition are usually in the opposite direction—from the opposition to the government, or minority to majority. These are more often induced by corrupt motives such as material rewards or by a promise of public office.

Defections of either type can cause problems if they are too frequent. Defections of ambition are hard to defend, as they are essentially acts of corruption. While defections of conscience might be more noble, if every member were to vote on all issues according to their own personal opinions, party discipline would break down and coherent policymaking would become impossible.

Design of anti-defection clauses

An anti-defection clause has to determine: (a) what counts as defection; (b) who decides when a defection has taken place; (c) the consequences of the defection; and (d) whether there are any exceptions or mitigating circumstances that might lessen those consequences.

What is ‘defection’?

A change in a legislator’s party affiliation can arise in three ways: a legislator ‘crosses the floor’ to join another political party, a legislator voluntarily resigns from their party to sit as an independent or a member is expelled from the party.

• A narrow definition of defection is limited to the first two situations—a deliberate act of renouncing party membership and then either sitting as an independent or joining another party. This puts members in a relatively strong position, as it is up them whether they wish to resign from the party or remain within the fold, albeit as a somewhat rebellious member of the party.
• A broader definition would include cases where a member is thrown out of the party against their will. That places legislators in a more vulnerable and dependent position, reducing their ability to disagree with party leaders. The power to expel is the power to compel. This rule effectively shuts down any dissent within the party and makes it difficult for members of the legislature to question leaders or hold them accountable.
• At its most extreme, defection could be defined as including any act of apparent disloyalty to the party without a formal change in party affiliation. This could include voting against the party whip in the legislature or even abstaining. This is the case in Bangladesh (Constitution of Bangladesh, article 70), which has one of the most draconian and stifling anti-defection rules in existence. Similarly, in Fiji, a member is deemed to have vacated his or her seat in parliament if he or she ‘votes or abstains from voting in Parliament contrary to any direction issued by the political party for which he or she was a candidate at the time he or she was elected to Parliament, without obtaining the prior permission of the political party’ (Constitution of Fiji, article 63(1)(h)).

Who decides when a defection has taken place?

• It might be up to the individual member to declare that they have left their party or changed parties. This gives the individual member relatively greater protection. Even if they are obliged to report their change in party allegiance, it means they cannot just be deemed to have left the party by someone else, without their own participation.
• Alternatively, the party leadership might be responsible for declaring when a member has left the party. In extreme cases, this places a member totally at the mercy of the party leadership, liable to be dismissed from the legislature on the leader’s whim.
• Placing the decision in the hands of the speaker or other presiding officer might be a middle course, but this depends on the level of partisanship or otherwise of the office.

What are the consequences of a defection?

In countries with an anti-defection clause, the normal consequence of a defection, once it has been confirmed, is the removal of a member from their seat. However, this need not necessarily be immediate or automatic. In Belize, for example, a member has the right to appeal to the Supreme Court, and removal of a member takes place only once the opportunities for judicial review of the act of expulsion have been exhausted (Constitution of Belize, section 59A(5)).

Depending on the electoral system and the particular constitutional rules, it might be possible for a member who has lost their seat because of an anti-defection rule to be re-elected under a new party label. This is easy to arrange in systems where members are elected from single-member districts, since the expulsion of a member for defection necessarily triggers a by-election in his or her constituency. In such cases, a member who defects from their party could keep their seat if they are able to win a by-election, either as an independent or as a member of another party. By-election victories in such circumstances are rare but not unheard of. The principle embodied in such rules is that the people should ultimately decide.

A list-proportional electoral system makes it harder to hold by-elections, as the usual practice is to fill vacant seats from the same party as the member who has vacated their seat. This is the case, for example, in Portugal (Constitution of Portugal 1976, article 160(1)(c)) and South Africa (Constitution of South Africa 1996, article 47(c)). The modalities vary. It could be achieved by means of a pre-nominated substitute, by drawing the next name from the appropriate list based on the results of the previous general election or simply by allowing the party to nominate a replacement. In such cases, an anti-defection provision enforces party discipline but does not engage with the people. A member removed for defection is unlikely to be able to regain their seat, at least until the next general election. It is the party, therefore, and not the people that has the last word.

Arguments for anti-defection clauses

Consolidation of the party system: In some countries, the informal norms sustaining party loyalty and cohesion are insufficiently embedded. The party system is weak and fluid, and parties often fragment, split or merge, according to shifting factions and alliances. Members of the opposition can easily be bought off, typically ‘in pursuit of ministerial office, monetary benefits, and varied personal interests’ (Bari and Dey 2020: 479). In such circumstances, anti-defection provisions can help to consolidate the party system and encourage the formation of stronger, more disciplined and more programmatic parties that are better able to perform their functions as instruments of representation, government and accountability.

Democratic mandate: In most democracies, members of a legislature are elected primarily on the basis of their party label, and are therefore pledged to a party manifesto or platform that forms part of the political ‘contract’ between voters and their representatives. Voters expect their representatives to adhere to that party and manifesto, and to implement the policies of their party. According to this logic, legislators who break from their party have betrayed their voters. If they choose to do this, they should at least have to go back to their voters to seek a new mandate, either as an independent or as a member of another party.

Government stability: Anti-defection provisions have been proposed as a means of increasing government stability. This is especially relevant in systems where the executive holds office by virtue of enjoying the ‘confidence’ (political support) of a majority in the legislature. The anti-defection provision in the Constitution of Bangladesh, for example, was introduced in response to chronic government instability under the former regime of East Pakistan (Bari and Dey 2020). Reinforced by anti-defection rules, the government can count on the solidity of its majority and does not have to worry about its majority being whittled away between general elections. This is less of an issue in presidential systems, where the stability of the executive is assured—except in cases of impeachment—by the president’s fixed term of office.

Anti-corruption: Since defections are often corruptly induced, banning defections might be one way—alongside other measures—of limiting corruption.

Arguments against anti-defection clauses

Unsuitability of constitutional regulation: Simply banning defections may not be sufficient to consolidate the party system, or even necessarily beneficial. It treats a symptom rather than the cause. In India, for example, anti-defection rules do not ‘address the root causes of defection, such as lack of intra-party democracy, corruption, and electoral malpractices’ (Drishtiias 2023). Not every problem in democracy is a constitutional problem. Sometimes, constitutional fixes such as anti-defection provisions can obscure the need for changes in political culture, norms or sub-constitutional rules. Party loyalty is not so much a matter of ‘constitutional legislation’ as of ‘constitutional morality—the ethics of representative government’ (Mill 2010: 224–25).

Democratic mandate: While a democratic mandate does run through parties, members of a legislature are also individual representatives with personal mandates of their own. This is especially true in systems where candidates are elected individually rather than on party lists. They are expected to use their judgement to represent their constituents’ interests. In some cases, this might mean deviating from the party line. For instance, if a party in office were to change its principles or to renege on a major policy commitment it had made before an election, then a member might reasonably claim that disobedience to a leadership that promotes such changes is loyalty to the mandate of those by whom they were elected. To put it another way, they could say, ‘I did not abandon the party; the party abandoned me’.

Democratic deliberation and powers of parliament: Democracy involves more than just the fulfilment of a majoritarian mandate. It also requires discussion, debate and deliberation. Parliament is a place for parleying—that is, talking. Such talk has no real effect—it becomes merely theatrical or a form of ritual—if parliamentarians’ votes are pledged to their parties in advance, or if there is no possibility of an MP being undecided on how to vote going into a debate and convinced by the speeches of fellow members. It is this capacity to talk to one another and not past one another that gives a legislature a shared existence and a sense of purpose. In other words, strong anti-defection provisions weaken legislatures. According to Bari and Dey (2020: 470), the anti-defection clause in Bangladesh has prevented effective scrutiny and oversight of the executive and ‘impeded the institutionalisation of Parliament’.

Government stability: Anti-defection provisions may be unnecessary. There are other ways in a parliamentary system to secure the government’s majority and to reinforce executive stability without resorting to anti-defection provisions that could restrict the ability of MPs to scrutinize legislation and hold the government to account. One way is to empower the government to dissolve parliament and call new elections. MPs will be less likely to force a vote of no confidence or to dismiss a government if they could lose their own seats. Another is to hedge votes of no confidence with time limits, for example, by preventing a vote of no confidence within the first year or six months of a new prime minister being appointed (although this, of course, has the corresponding disadvantage that a prime minister who proves to be unfit for office may, during this period, be difficult to remove). A third option might be to require a ‘constructive vote of no confidence’, according to which a prime minister cannot be removed from office unless parliament designates a successor with majority support.⁶

Anti-corruption: Anti-defection clauses are neither the only way nor necessarily the best way to tackle corruption. Other approaches, such as creating an anti-corruption commission, strengthening the integrity of the civil service by limiting patronage in public appointments, adopting stricter party and campaign finance rules or requiring members of the legislature and others in public office to declare their interests and avoid conflicts of interest, might be more effective than anti-defection provisions without the associated disadvantages. Moreover, anti-defection clauses can have the opposite effect. By protecting the government from accountability to their own legislative caucus, strong anti-defection rules can help to facilitate or hide corrupt practices.

International law: There is some soft international guidance on anti-defection clauses. Since these clauses are most often found in Commonwealth countries, the Latimer House Principles apply. These state that, while ‘anti-defection measures may be necessary in some jurisdictions to deal with corrupt practices’, ‘the expulsion of members from parliament as a penalty for leaving their parties (floor crossing) should be viewed as a possible infringement of members’ independence’, and ‘the cessation of membership of a political party of itself should not lead to the loss of a member’s seat’ (Latimer House Principles, III(2)).

Moderate anti-defection clauses

Where informal political incentives to party loyalty are weak, anti-defection clauses might help to stabilize the executive (especially in parliamentary systems) while strengthening party cohesion and reducing certain specific types of political corruption. On the other hand, excessively strict anti-defection clauses can stifle political discussion, weaken the legislature, reduce political accountability and undermine democratic norms. In any circumstance where a member of the legislature stands to lose their seat simply for defying the leader of their party, any semblance of responsibility of the government to parliament, in parliamentary systems, is lost (Alam 2024).

Therefore, if anti-defection clauses are considered, care should be taken to draft such clauses in a moderate way that does not unduly limit the freedom of members of a legislature to perform their duties according to their conscience.

Moderating the definition of defection

There is an important distinction to be made between anti-defection clauses that penalize only a voluntary decision to leave a party and those that penalize being expelled from one’s party, voting against the party whip in the legislature or even abstaining. An anti-defection clause triggered by voluntary resignation may be appropriate and democratically acceptable. Broader clauses that cause members to be totally dependent on party leaders are risky and, on the evidence of the various countries where they have been tried, should be avoided.

However, even constitutional provisions that seem to limit removal from office to voluntary acts of resignation from a party can, in practice, be more broadly construed. In Belize, for example, the anti-defection rule is triggered only if a member ‘resigns from their party or crosses the floor’ (Constitution of Belize, section 59(2)(e)), which seems to imply that there has to be a deliberate, voluntary and unambiguous act by the member to leave their party. In practice, the rule has been interpreted more broadly. A member of the governing party who refused to support the government on the budget was said to have ‘crossed the floor’ even though he did not resign his party membership and was forced to resign his seat (Vernon 2024). In India, a provision in article 102 of the Constitution (as amended by the Constitution (Fifty-Second Amendment) Act 1985) requiring a member to forfeit their seat if they resign from their party has been interpreted as including ‘both formal resignation and giving up of membership by means of conduct, such as public expression of opposition to his party or support for a rival party’ (Bari and Dey 2020).

To prevent such situations arising, any anti-defection clause could specify that it does not apply to a member who is forced out of their party against their will, but only to those who voluntarily resign from their party or voluntarily join another party. Moreover, it should specifically and expressly protect the freedom of a member of the legislature to vote against their party, or to abstain from voting, without putting their seat in the legislature at risk. Even if these protections are implicit in the constitutional provisions on the privileges of members, they should be repeated in the anti-defection clause to give the courts and others clear guidance.

Moderating the effect of defection

It would also be possible to develop an anti-defection clause that would not immediately or automatically lead to the loss of a seat.

• Eligibility for recall: Some countries have rules that enable a member of the legislature to be removed from office following a recall vote or recall petition. Their constitutions or the law may specify certain grounds that can trigger a recall, such as a member having been found guilty of a criminal offence or having breached rules relating to parliamentary standards. Quitting one’s party or changing parties could be added as a ground that renders a member of parliament eligible for recall.
• Limitations on appointments: If the aim is to prevent defections of ambition without limiting defections of conscience, the consequence of defection could be to prohibit a member from being appointed to any ministerial office—or other lucrative public office—for the remainder of the term for which they were elected. That would discourage defections of ambition to the government in the hope of public office without penalizing defections of conscience from the government.

1. Are the roles and functions of political parties and their place within the democratic system recognized by the constitution?
2. Are the rights of political parties and the rights of persons to join (or not to join) such parties and take part in political activities properly guaranteed? How can the constitution balance the hybrid nature of parties as private organizations that need to be protected under freedom-of-association laws, on the one hand, and quasi-public bodies with a recognized role in the political system, on the other?
3. Who should be responsible for registering and regulating political parties? Should this function be entrusted to the electoral commission? Or should there be a specific institution, such as an office for the registrar of political parties, with this remit? What would be the pros and cons of separating party registration and regulation from electoral management?
4. Is there—and does there need to be—a constitutional provision on internal democracy within parties? How can this be enforced in ways that are not subjective? What is the right balance between democracy within parties and democracy between parties?
5. What about limits on party funding? Are there prohibitions on donations, secret donations or corporate donations? Are these prohibitions effectively and equally enforced? Is there a robust provision on transparency in party funding? In what ways could the constitutional provisions be strengthened?
6. Is there an existing arrangement for the public funding of parties? If not, what would be the pros and cons of introducing one? If there is an existing arrangement for public funding, is it working? Is it regarded as fair and reasonable by all parties? Does it disproportionately favour some parties over others? Is it vulnerable to manipulation by the government or ruling majority party? Is public funding affordable? How could the constitution improve this situation or protect it from future regressive change?
7. Should any political parties be prohibited? What about anti-democratic parties? (Is it possible to define ‘anti-democratic’?) Do the same criteria apply to parties that are democratic but seek to champion the interests of particular ethnic, religious, linguistic or territorial minorities? Who should decide on the question of prohibiting a political party? What rules and procedures are needed to prevent the abuse of that power?
8. Is party discipline too tight or too lax? If party discipline is too lax, would anti-defection provisions be appropriate? If so, what measures can be taken to ensure that they do not go too far, stifle the personal responsibility of legislators or weaken accountability? Is there a middle ground that can prevent ‘defections of ambition’ without altogether preventing ‘defections of conscience’?