Sustaining Democracy and Promoting Development in Presidential System*

by Rikha


As an emerging democracy and developing country, Indonesia is now struggling to sustain its democracy and at the same time promote its development. These two agendas are under constitutional framework of presidential system. As argued by both proponents and opponents of the presidential system, it has advantages and disadvantages (Lijphart, 1992; Linz, 1990, Mainwaring 1993)[1]. One of the disadvantages is less effective executive-legislative relationship due to lack of incentives for cooperation. Separation of powers that aims to provide check and balance between government branches has the serious unintended consequences of policy gridlock. As a presidential system, Indonesia also suffers from this disadvantage that could destabilize its democracy and potentially hinder its development in the future.

This policy paper aims to propose a reform of the relationship between the two branches of government. It is strongly recommended to limit presidential legislative power to constitute a more balanced relationship and, thus, to sustain democracy. However, a more balanced relationship would not necessarily increase the effectiveness of policy-making for development. Therefore, it is also suggested to initiate joint development agendas agreed on by both branches in their first year of governing to secure national development agendas from deadlock. These two reforms would provide incentives toward more cooperative behavior in the executive and legislative relationship.

Arguments and analysis where the proposed reform is based on will be structured as follow. Firstly, this paper will assess the president power compare to the House. Basically, this assessment will answer the question of how much power is enough for a president. It is important to note that the presidential power here refers to the formal power as written in Indonesian Constitution, not informal power that a president might have. Secondly, comparative analysis will be carried out with two purposes. The first is to identify ‘international standard’ of presidential power required for sustaining democracy. The second is to explore international experiences in dealing with gridlock situations. Lastly, this paper will evaluate the plausible impact of the proposed reform on the institutional behaviour in the context of Indonesia’s multiparty system.

Assessing Indonesian presidential power

It is the logic of veto players (Tsebelis, 1995) underlying the principle of separation of power in presidential system that aims to provide checks and balances between government branches. However, keeping the balance between the two government branches (executive and legislative) is not as easy as it may sound. On the one hand, if a president is too weak, he or she tends to become a lame duck. As most presidents play a role as agenda setters, this situation would hinder development agendas (Mainwaring, 1993). On the other hand, a too powerful president tends to be authoritarian and lead to democratic instability. This is confirmed by Mainwaring and Shugart (1997) by arguing that presidents with more legislative power are likely to be more vulnerable from democratic breakdown than others.

One way of balancing the constitutional power between the two branches of government is by making adjustments to strengthen or weaken the president’s power. Thus, it is important to measure the power of Indonesian president prior to the adjustment. In measuring the constitutional power of Indonesian president, a tool developed by Shugart and Carey (1992) is relevant.

Shugart and Carey (1992) identify ten important powers of a president[2], divided into two categories: legislative and non-legislative power. Six legislative powers include package veto, partial veto, law-decree, exclusive introduction of legislation, budgetary power, and proposal for referenda. Non-legislative power includes cabinet formation, cabinet dismissal, censure and dissolution of assembly. Each dimension of power is scaled from the least powerful (weighted 0) to the most powerful (weighted 4). So, the total score for legislative power will be in between 0 to 24 and for non-legislative power will be in between 0 to 16.

Regarding legislative power, it is explicitly written in the Constitution that the president has veto power both for the whole package and a certain part of law proposed by the House. Article 20 of the Constitution stipulates that every proposed law should be jointly agreed by both sides. If agreement cannot be reached, the proposed law cannot be discussed anymore in that assembly session. Referring to that provision, the president apparently has strong veto power for both the whole package and partial veto over a law. For this power, the president scores 8 out of 8. Moreover, constitutionally the president has authority to make a law, known as decree-law, particularly in crisis situation. Although a decree-law is subject to be reviewed by the House in the soonest assembly meeting, the president has full discretionary power to define the term of crisis by his or her own interest. With these few restrictions and temporary decree authority, the president is considered to score 3 out of 4 for holding this power. In terms of exclusive rights to initiate legislation, the president can proactively propose legislation to the House (article 15). However, legislation proposed by the president can be amended unrestrictedly by the House. Thus, the president scores 1 out of 4 for having this power. The remaining two legislative powers are scored 0 since the president only have power to propose budget, without any control to restrict the House to amend it. For proposing a referendum, it is neither the president nor the House have the authority since it is not regulated in the Constitution. Summing up the total score of the president’s legislative power, it is accounted for 12 out of 24.

Looking at non-legislative power, it is stated in the Constitution that Indonesian president has full authority in cabinet formation and dismissal without requiring any confirmation from the House (article 17). Therefore, president gets full score (8 out of 8) for holding these two powers. Moreover, in the absence of the House’s authority to censure or remove a minister bring the president additional score of 4. However, the president cannot dissolve or dismiss the House (article 7C) and the president gets score 0 for this strict limitation. The total score of presidential non-legislative power is 12 out of 16.

The measurement both legislative and non-legislative power of a president above does not provide how much power should a president has. The following analysis will provide a comparative answer for that question.

Comparative analysis of presidential power and gridlock mitigation

As discussed above, it is not an easy task to determine how much strength is enough for a president. Simply determine this by median point in the sense that legislative and executive should have the same score would be misleading since in Montesquieu’s original idea of separation of power, legislative power belongs to the House, not to the president. Therefore, comparative analysis could provide best practice analysis as a benchmark for dealing with ‘how strong is enough’ controversy.

This comparative analysis is based on work from Shugart and Carey (1992), Carey (1997), Mainwaring (1993) and Nino (1992). Focus will be put on five countries which are considered to be successful and stable presidential democracy: the U.S., Costa Rica, Venezuela, Dominican Republic and Colombia. Lessons from other presidential countries will also be analysed specially to learn how those countries smooth the policy making process and deal with gridlock situation. 

Shugart and Carey (1992) measured 32 presidential regimes and they found long and stable presidential system at low presidential legislative power. In their scale, those countries only have score in the range of 0 to 2 for presidential legislative power, including the United State of America (score 2), Dominican Republic (score 2), Costa Rica (score 1), and Venezuela (score 0). The Colombian presidential system (score 5) is debatable, as it is not clear whether it has had continuous democracy or not (Mainwaring 1993, p. 204). In terms of non-legislative power, those countries have score 12, except for the U.S. which score 11.

From the comparative perspective, whereas Indonesian president’s non-legislative power is considered in the acceptable range to maintain democracy in the sense that it has the same score with those of successful democracies; Indonesian president’s legislative power is not. The result of power measurement suggests that the Indonesian president hold too much legislative power and this could potentially harm its democracy. We can see from other countries experiences, particularly from the most successful democracy in the world, the U.S., that the president holds very limited legislative power. There is a tendency that the weaker the legislative power of a president, the higher the probability of a country to be a stable democracy.

Reducing presidential legislative power, however, would raise two challenges. First, to what extent the power should be reduced and what kind of power should be kept and what kinds should be eliminated. Second, reduced presidential power might lead to the lame duck phenomenon as a result of gridlock situation which eventually would impede development agendas.

Addressing those two challenges, international experiences might provide some answers. First, looking deeper at successful presidential democracies, all presidents in those countries hold legislative power only in the form of veto power over proposed law. In the case of the U.S. and Dominican Republic, a president has a veto power over a law proposed by the congress, but they must reject it as a whole package, not partial veto. The House can override the president’s veto over a law; however, it requires two-third of the legislative support (Shugart and Carey, 1992). In Costa Rica, it even requires less than two-thirds of legislative support (Carey 1997, p. 202) and no veto power is held by the Venezuelan president. Therefore, referring these successful democracies, it is strongly recommended to equip Indonesian president with only the whole package of veto power subject to override by two-third of members of HoR while keeping the executive power as it is. This would promote more balanced power and check between the president and the House.  

Limited presidential legislative power, however, might discourage the president from taking initiative in initiating development policies. The worst of all is that when gridlock situation starts to happen. Instead of check and balance, the two branches might be destructive and block any developmental initiatives which mostly come from the presidents. As argued by Mainwaring (1993), the gridlock situation is more likely to happen in a multiparty system. In a multiparty system, like Indonesia, a president hardly gets support from the House unless the president’s party won the majority of the seats. If president’s party only got small number of seats in the House, just like Indonesian president in 2004-2009, he or she would run a minority government. A minority government with a less powerful president is unlikely to be effective in implementing his or her development agendas. In short, democracy would be sustained but at the expense of development.

Dealing with this issue, there should be a mechanism that can secure development agendas from the possibilities of policy-making gridlock. This would be a binding commitment or joint major development agendas between the president and the House. The joint agendas should be made in the beginning or at least in the first year of new elected government. This does not mean that either the president or the House could not propose other new agendas apart from it. Both are still encouraged to propose new agendas or even to change the agreed agendas. However, it would be more difficult to do the later since it is already agreed and both have veto power to override changes. What makes it different from normal gridlock situation is that when joint agendas are in place, any gridlock situation would have less effect on the development since the agendas is already agreed. Without binding co-agendas, a lasting gridlock situation would jeopardize most of development agendas.

This attempt actually had been tried in Venezuela where ‘the president presented in the first year of his term of office a program of social and economic development to be approved by both Houses in joint session’ (Nino 1992, p. 129). Although this reform was adopted as part of attenuation of the presidential system in Venezuela, the idea can also be applied to mitigate the gridlock effects in presidential system as explained above. Moreover, Indonesia actually experienced this kind of mechanism for 32 years under president Soeharto’s regime. The fundamental problem with the mechanism at that time is that the development agendas that are mandated to president by the National Assemblies were actually dictated by the president to appear as if he acted constitutionally. The difference with the proposed reform is that the binding joint agendas here is discussed and agreed together by the both branches of government.

Plausible impacts of proposed reform on political institution

It is argued by Armijo, Faucher & Dembinska (2006) that the relationship between the executive and the legislative are influenced by many factors such as electoral rules, the political party system and the rules of legislative procedure. As Indonesia is adopting proportional representation that leads to multiparty system, the examination will only focus on how would the proposed reform influence institutional behaviour in the context of Indonesia multiparty system.

The presidential constitution offers presidents two strategies to achieve their policy goals: through passing a law and/or cabinet appointing (Neto, 2006). Related to the former, veto is the most ubiquitous tool by which president can shape legislation (Carey 1997, p. 202). However, as demonstrated by Neto (2006) in the case of America Latin, with extensive veto power and decree-law presidents tend to appoint their cabinet mostly with technocrats and cronies. Although the appointment might be optimal for the president in the short run, he argues, it might be also dangerous to alienate parties and legislatures and can lead to political instability in the long-run. Conversely, a president with limited legislative power would generate incentive to cooperate and negotiate with legislative especially in cabinet appointment.

On the other hand, establishing joint development agendas would provide another incentive for president to seek support form legislatures. He or she can accommodate as many as possible parties’ agendas in the exchange of their support. Even in minority president case where the president’s party has smaller seats in legislative and in a multiparty system where the president should get support not only from its own party but also from many parties in the House, this tool can be used to consolidate legislature’s support. In short, this joint agenda would act like a written coalition treaty among parties in the House and the president in the form of development program which is put into a binding law.


This proposal recommends two reforms. First, limit the presidential legislative power. This would not only stabilize democracy through more balanced power but also generate incentive for corporation between the two government branches. Second, joint development agenda would mitigate gridlock situation that might happen in the presidential system. Beside this would provide more incentive for the president to cooperate with other parties in the House, joint development agendas would also secure national development.


Armijo, LE, Faucher, P & Dembinska, M 2006, ‘Compared to what?: assessing Brazil’s political institutions’, Comparative Political Studies, vol. 39, No. 6, pp. 759-786.

Carey, JM 1997, ‘Strong candidates for a limited office: presidentialism and political parties in Costa Rica’ in Mainwaring, S & Shugart, MS (eds.), Presidentialism and Democracy in Latin America, Cambridge University Press, New York.

Lijphart, A 1992, ‘Introduction’ in Lijphart, A (ed.), Parliamentary Versus Presidential Government, Oxford University Press, New York.

Linz, JJ 1990, ‘The perils of presidentialism’, Journal of Democracy, vol. 1, no.1, pp. 51-69.

Mainwaring, S 1993, ‘Presidentialism, multipartism, and democracy: the difficult combination, Comparative Political Studies, vol. 26, no. 2, pp. 198-228.

Mainwaring, S & Shugart, MS 1997‚ ’Conclusion: presidentialism and the party system’ in Mainwaring, S & Shugart, MS (eds.), Presidentialism and Democracy in Latin America, Cambridge University Press, New York.

Neto, OA 2006, ‘The presidential calculus: executive policy making and cabinet formulation in the Americas’, Comparative Political Studies, vol. 39, No. 4, pp. 415-440.

Nino, CS 1992, ‘Ideas and attempts at reforming the presidentialist system of government in Latin America’ in Lijphart, A (ed.), Parliamentary Versus Presidential Government, Oxford University Press, New York.

Republic of Indonesia 2002, The Forth Amended Constitution, Republic of Indonesia, Jakarta, viewed October 31, 2009, <>

Shugart, MS & Carey, JM 1992, Presidents and Assemblies: Constitutional Design and Electoral Dynamics, Cambridge University Press, New York.

Tsebelis, G 1995, ‘Decision making in political systems: veto players in presidentialism, parliamentarism, multicameralism and multipartyism’, British Journal of Political Science, vol. 25, No. 3, pp. 289-325.

[1] Presidential system has three important advantages namely executive stability, more limited government, and greater democracy. Presidential, however, suffers three disadvantages of executive-legislative deadlock, temporal rigidity, and ‘winner-take-all’ government.

[2] A more comprehensive measurement is proposed by Frye (1997) covering 27 types of power including appointing several high official positions like chief of central bank, commander-in-chief of armed force, etc. In relation to the president and the House relationship, however, Shugart and Carey’s (1992) measurement is simpler and highly accurate since it uses a scale of 0 to 4 for each dimension of power.

*This piece of paper was written back in 2010 as part of assignment for the class Comparative Governments and Politics at the Australian National University.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s