Incumbency advantage is not restricted to established majoritarian systems

To date, most scholarly works have focused on incumbency advantage in the US and consider how it operates in majoritarian contexts. In a recent paper, Mert Moral, H. Ege Ozen and Efe Tokdemir drew on the case of Turkey to explore whether the incumbency operates in multi member district systems. They found that although it is not as marked as in the US context, considerable incumbency advantage persisted in the more proportional system.

Credit: Aris Gionis CC BY-NC 2.0

Incumbency advantage is an empirical regularity, which can briefly be described as the frequent re-election of incumbent members of legislatures. The literature on incumbency advantage, which originates from observations of incumbent members of the Congress and mostly built on the dynamics of American Politics, suggests a number of theoretical explanations for such regularity. Incumbents are found to discourage high quality challengers from running in elections, and have higher name recognition and larger resources due to their stronger local ties. Incumbency also works as a heuristic cue to voters, which largely decreases information costs associated with voting.

To date, most scholarly works have focused on incumbency advantage in the US and explore how the single member district (SMD) electoral system creates incentives for individual candidates to develop close personal relationships with their constituencies, which eventually lead to high levels of name recognition and therefore to safer electoral races for candidates running in the same district from one election to the next.

Recently, we published an article in Electoral Studies where we asked whether the incumbency advantage is a phenomenon unique to SMD systems. If this is not the case, then under which conditions or institutional arrangements would political parties and individual candidates enjoy the benefits of incumbency? Does a similar pattern exist in multi member district (MMD) systems, particularly in closed-list PR systems where party control on the nomination of candidates is arguably higher? We used Turkey as an MMD example revealing the characteristics of both stable and unstable party systems to assess the extent of incumbency advantage conditional on varying levels of district magnitude and party system instability.

Incumbency advantage is not unique to majoritarian systems

In the article, we argue that incumbency advantage is not peculiar to SMD systems. Rather, it is a phenomenon which is applicable to other electoral systems, even those where voters base their evaluations on party lists. In other words, we suggest that political parties can gain additional electoral benefits by strategically fielding incumbent candidates.

One of the circumstances that motivates political parties to field incumbent candidates is district magnitude, which is often considered as the primary factor leading to high levels of incumbency advantage in SMD. The main source of scepticism against an incumbency advantage in MMD systems is the higher cost of information that voters need to possess due to higher numbers of individual candidates running in distinct party lists in each electoral district. That said, in most MMD systems, the number of candidates is proportional to the size (i.e., demography) of electoral districts. An incumbent candidate, in the conventional definition of incumbency, in smaller district is more likely to enjoy an incumbency advantage. Due to the benefits of higher name recognition, we also hypothesise that reputable incumbents transferred from other districts or other parties are likely to be fielded by political parties in larger districts especially in turbulent times where the cost of information to voters is higher.

The second condition that we underline in our study relates to party system instability. In unstable party systems, it is not political parties but political elites who have a more prominent role in politics, which makes parties field individuals more familiar to voters. In our article we therefore hypothesise that political parties receive electoral gains from fielding incumbents transferred from other parties or districts when the level of party-system instability is high.

The case of Turkey

In order to test our theoretical expectations we employed an original dataset on incumbent and non-incumbent parliamentarians in Turkey, where the closed-list PR makes incumbency advantages even less likely to be observed and hence provides more conservative tests for our hypotheses. Moreover, varying district magnitudes across electoral districts in Turkey enable us to look at how individual and party incentives differ across smaller and larger districts. We limit our analysis to the period after the military coup in 1980, to the seven parliamentary elections between 1987 and 2011.

Turkey is also a good case for studying the effects of party-system stability, as the elections in 1987 and 2002 indicated a distinct pattern compared to the other five elections we examined. Clientelistic relationships between candidates/parties and their constituencies, as well as the electoral rules in Turkey, have otherwise followed a quite stable trajectory between 1987 and 2011. However, political shocks, namely the military coup and the 2001 economic crisis, affected the electoral outcomes in the 1987 and 2002 elections. We therefore divided our sample into two, and analysed the 1987 and 2002 general elections separately from other elections.

Strategies adopted in the context of party-system instability

Our empirical analyses reveal that party system instability in the Turkish context made political parties adopt different strategies compared to parties in more established and institutionalised party systems. To start with the interactive models in the article, incumbency advantage, in its conventional definition, decreases when district size increases. However, our findings also confirm an inverse relationship between the incumbency advantage and district magnitude in MMD systems when alternative party strategies are taken into account.

Figure 1 below shows the marginal effect of two different measures of incumbency that we account for. While incumbents running from same parties in same districts create an advantage that varies between 2.8 and 1.3 percentage points in smaller districts, political parties also benefit from transferring well-recognised incumbents and make them run in larger districts to decrease the cost of information to voters and benefit from their name recognition. In larger districts, the electoral gains of parties from fielding incumbent candidates vary between 1.5 and 11.6 percentage points conditional on district magnitude.

While we found a small incumbency advantage in non-interactive models, our most important findings is that inter- and intraparty transfers were mainly used to enable parties to enjoy a relatively high level of incumbency advantage in 1987 and 2002 elections, which were stigmatised with high levels of party system instability. Accordingly, we can confirm that political parties receive electoral gains from fielding incumbent candidates that they transfer from other parties or from other districts when the party-system instability is high.

Conclusion

Our study shows a considerable incumbency advantage in Turkey, albeit lower than estimated levels in studies focusing on the US. Incumbency advantage is higher in districts with smaller magnitudes that resemble the SMD systems, but such effect is also conditional on other factors affecting voter behavior such as party entry and exit. Moreover, intra and inter-party transfers of incumbent candidates benefit political parties in turbulent times. In these respects, we believe that further research on incumbency advantage in PR should take into account the conditionality on party system stability and electoral institutions.

This post represents the views of the authors and not those of Democratic Audit UK or the LSE. Please read our comments policy before posting.

Mert Moral is PhD Candidate in Political Science at Binghamton University (SUNY). His research interests include political polarization, voting behavior, voter turnout, electoral institutions, representation, political methodology, and Turkish politics. His research has been published or is forthcoming in International Political Science Review, Electoral Studies, and International Journal of Electronic Governance.

Ege Ozen is from Turkey, where he received his BA degree in political science and sociology and his MA degree in history and political science. Currently, he is a doctoral candidate in political science at Binghamton University (SUNY) specialized in comparative politics. His research interests are on voting behavior, domestic politics of the Middle Eastern and North African countries, elections and electoral processes, political parties, and representation. His Ph.D. dissertation is on voting behavior in Egypt, Tunisia, and Turkey.

Efe Tokdemir is a PhD candidate in Political Science at Binghamton University (SUNY), New York. He is currently a visiting scholar at Mershon Center for International Security Studies at the Ohio State University. Tokdemir is specializing in International Relations with a focus on foreign policy and foreign public opinion, and terrorism. He has forthcoming and published articles in Journal of Peace Research, Conflict Management and Peace Science, International Political Science Review, and Electoral Studies. To learn more about his research please visit www.efetokdemir.com