In his famous statement about the perils of presidentialism, Juan Linz argued that newly emerging democracies ought to avoid adopting a presidential form of government. One of Linz’s reasons had to do with the winner-take-all-nature of presidential elections. By definition, such elections are zero-sum games where the losing candidates have little to no prospect of sharing in executive power. By having a single indivisible and powerful executive office, presidential elections amplify the gap between winning and losing, and can contribute to creating and deepening political divisions, which is precisely what a new democracy ought to minimize.
At the same time, having the people elect the head of their state directly can strengthen the legitimacy of the democratic foundations of the new constitutional order. When conducted in a fair, free, and transparent manner observing the highest standards of electoral integrity, direct presidential elections can play an important role in aiding the development of civic and political values such as electoral participation, competitiveness, and accountability. If the population is not imbued by such values, the new democratic system may soon hollow out and become a procedural mechanism with no substantive values informing and guiding it.
An intermediate constitutional solution is the adoption of a semi-presidential system of government, which, according to scholars like Maurice Duverger and Robert Elgie, is characterized by a presidency that is elected directly by the people but that is also considerably weaker in power and authority than the chief executive of a presidential system of government. The relative weakness of the semi-presidential head of state is underscored by the fact that the office shares executive power with the prime minister who, as the head of government, is responsible to the legislature. Semi-presidentialism often becomes an attractive constitutional choice in new democracies precisely because it has the advantage of encouraging popular participation in the political system without concentrating too much power in a single executive office.
The experience of the ten post-communist democracies in Eastern and Central Europe is an excellent case in point. At the time of their transition to democracy in the early 1990s, only half of the ten states had a semi-presidential executive: Bulgaria, Lithuania, Poland, Slovenia, and Romania. The remaining five states (the Czech Republic, Estonia, Hungary, Latvia, and Slovakia) remained parliamentary systems with both the prime minister and the president elected by the legislature. By 2015 Hungary, Estonia, and Latvia have remained the only three cases with indirectly elected heads of state. At first Slovakia, in 1998, and then the Czech Republic, in 2012, enacted constitutional changes to move from indirect to direct presidential elections. In both of these cases, the adoption of direct presidential elections was the result of repeated failures by parliament to ensure a smooth and efficient process. In Slovakia, the National Council was unable to elect a president after several unsuccessful rounds of balloting in 1998, whereas in the Czech Republic both the 2003 and the 2008 elections were characterized by legislative tumult leading to renewed calls for delegating the choice of president to the people.
What can constitutional designers and engineers learn from the history of presidential elections in the post-communist region? Insofar as political stability and legitimacy are concerned, there are three important lessons:
- The semi-presidential model is a reliable and good constitutional choice as long as the formal powers of the head of state are kept modest. Some of the most serious constitutional battles between a directly elected president and the legislature took place in the three states where the head of state had the greatest formal powers (Lithuania, Poland, and Romania). However, while subsequent constitutional changes in these states modified the term or the powers of the president, none of them did away with the directly elective nature of the office.
- Indirect presidential elections can be a major source of political instability, and loss of public trust in the legislature, unless the rules of the game are kept efficient. The examples of Slovakia and the Czech Republic showed that inclusive and consensus-oriented rules do not work in the long run. In both cases, the constitution required that the winning candidate obtain a highly qualified majority of votes, which proved to be extremely difficult, or even impossible, in an already fragmented multi-party parliament.
- Efficient rules for indirect presidential elections ought to combine a simple majority threshold for winning with a fixed number of rounds in which the election must be completed. As the cases of Hungary and Estonia show, efficient rules will typically favor the candidate of the incumbent governing coalition and as such will further concentrate executive power in its hands. However, this may be a small price to pay relative to the instability that can be caused by the failure of consensus-oriented inclusive rules.
In short, the post-communist cases suggest that the ideal form of choosing the president of a new democracy may be either direct election by the people or indirect election by parliament using efficient, result-oriented rules and procedures. While presidentialism as a constitutional system may be perilous for the stability of a new democracy, as Juan Linz argued, there is also a great danger in adopting parliamentary processes of presidential election which can themselves become the source of political instability.