Paper

Choosing an Electoral System

| April 29, 2014

Iraq's Three Electoral Experiments, their Results, and their Political Implications

Executive Summary

As the drama of the Middle East’s democratic upheaval unfolds, the design of electoral systems is a crucial but underreported part of the story. Our original analysis of Iraqi elections in 2005 and 2010 demonstrate that small changes in how votes become seats can have a major impact on who governs. As such, they offer critical lessons for those shaping the contours of the democracies struggling to emerge in the Middle East today.

Iraq held three general parliamentary elections between 2005 and 2010. Each employed a different electoral system and occurred under very different political environments, making comparisons difficult.

This paper seeks to enable comparisons and probe deeply the implications of Iraq’s different electoral systems over time. It does so through a combination of political analysis and the reconstruction of results in each election had the system of the other elections been employed. This effort – based on the actual polling data from the polling centers – produces election results for nine “elections” – three actual and six counterfactuals. In addition to this technical analysis, we then consider whether – had everyone voted the same but a different electoral system been used – the political results would have been different.

This detailed study allows us to make some striking, practical conclusions.

First and perhaps most important, it confirms that the electoral system chosen in any particular instance can have profound effects on the outcome of that election, not only in the sense of seats gained, but, in parliamentary systems like Iraq, in the process and dynamics of governmental formation. For example, had the election system used in the December 2005 election been employed for the March 2010 parliamentary election, Prime Minister Maliki’s party, the State of Law, would have emerged as the winner by two seats, in contrast to the actual elections, where his party won two seats fewer than the Iraqiya party. Such an outcome, though tiny in the shifting of seats, would have unequivocally given Prime Minister Maliki the right to form the next government, eliminating the nine months of political wrangling and stalemate that occurred after the actual elections.

Second, this study also suggests some recommendations for those crafting electoral strategies, once they have determined their goals. For instance, the debate over whether a single district system or a multi-district system based on the provinces would better advantage small, secular parties now seems misguided. This study reveals that the actual mechanisms for allocating seats within a district are far more influential in determining the scope of representation than the shift from a single district to multiple ones. As discussed in detail above, the district-based system of 2010 was much more favorable to larger parties – and harmful to smaller ones – than the single district system of January 2005 on account of the mechanisms used to distribute seats within the provinces.

Third, this study provides some valuable insights to parties looking to advocate for an electoral system that best meets their needs – one that increases their representation in parliament. In this case, the first issue to consider is whether the turnout of the electorate of the party in question is likely to be higher that that of the party’s competitors. If so, then the single-district system, which equates turnout with representation, is likely to be the most advantageous. The Kurds, who traditionally have higher turnout rates than other groups in Iraq, consistently fared best in the single-district system. Conversely, if the party has reason to believe its constituency will be more modest in its turnout – either due to boycotts or factors such as difficult weather or terrain – than the single-district system should be avoided; the link between turnout and representation becomes less important the smaller the electoral unit and the allocation of seats to it.

Fourth, this study has important implications for parties, not only by pointing to what electoral system they should lobby for, but by suggesting strategies they should employ to win elections once the electoral system is finalized. All three elections, and the six counterfactuals constructed around them, clearly show the importance of building coalitions with other parties to increase chances of representation. This recommendation is, of course, more applicable the smaller the party.

The electoral system chosen for elections, particularly in nascent, fragile democracies like Iraq, must be treated as a strategic matter. Yet the formulation of electoral systems is often done without adequate transparency or appreciation for their importance.

As many countries in the Middle East move to build the institutions of their futures, the question of what election law should be used will loom large in countries embracing democracy for the first time. The Iraq experience underscores the importance of choosing an electoral system with care, and understanding how the most-seemingly technical parameters – such as the mechanism for how seats are allocated within provinces – can have a significant impact on the nature of the parliament that is produced. In countries like Egypt and Libya, where elections may be used to select people to write or modify country constitutions, it is essential that new election laws are consistent with the principles espoused by the new leaders and the newly-empowered people – and are clearly understood by those competing in elections for the first time.

Click here to read the full paper:

For more information on this publication: Please contact the Belfer Communications Office
For Academic Citation: O'Sullivan, Meghan L. and Razzaq al-Saiedi. “Choosing an Electoral System.” Paper, April 29, 2014.

The Authors

Meghan L. O'Sullivan