By Martin Lundin, Oskar Nordström-Skans, Pär Zetterberg
Who runs for office? Standing as a candidate in public elections could be characterized as the ultimate act of political participation, and successful candidates make decisions that have major impacts on people’s lives. Yet, we know little about the exact determinants of why some citizens choose to run for office whereas others do not. An interesting line of research on recruitment has documented the role of political ambition and the personal traits that are related to this ambition, but we know next to nothing about what the key personal experiences or life events are that trigger individuals to enter into representative politics.
In a new issue (vol. 38, no. 2) of Political Behavior, we investigate a specific personal experience, documenting its potential role of generating political candidates: the experience of holding office within a civil organization. This particular experience is interesting since the potential causal link between organizational activities and individual political participation has been debated for more than a century. Already in the 19th century, Alexis de Tocqueville pointed at the democratic potential of voluntary associations: they could operate as “schools of democracy” for individuals participating in their activities. Following in that tradition, political scientists have commonly assumed that these associations serve as recruitment channels, or training grounds, for new political leaders. But the causal effect of acquiring and holding a leadership position within a civil organization on political career patterns has rarely been tested empirically.
As always, there are good reasons for this lack of research. One such reason is data limitations: it is relatively easy to find information on already elected representatives but much harder to find data on suitable “controls” –those who could have, but never did, become politicians. In addition, much of the research on the role of voluntary associations has focused on broader outcomes, such as party membership, voting and protests. Partly, this is also likely to be due to lack of proper data since most surveys contain too few cases where subjects actually entered into representative politics. Finally, studies on this topic have rarely been able to fully take problems of self-selection bias into account. Individuals who participate in civil organizations are likely to be more politically motivated and skilled than others. It is therefore inherently difficult to empirically separate the causal impact of organizational activities from the selectivity of the individuals who participate in these activities.
To fill in the gaps, our study examines whether gaining office within one specific kind of organization – student organizations – is causally related to political candidacy using unique Swedish data. The case is particularly well-suited for a causal analysis of this topic since offices are allocated in well-documented elections. Both the students who succeed and the students who fail in acquiring leadership positions in the student organizations can be followed over time. The organizations also have a fairly well-recognized status, and a relatively large fraction of subjects do enter into regular politics later in life. In order to derive evidence with an unambiguous causal interpretation, our empirical approach relies on the logic of Regression Discontinuity (RD) designs. We compare students who were elected into the SU councils with a small margin to those who just missed to be elected into the same councils. To facilitate the analysis we collected unique archive data on a large number of candidates to student union (SU) councils at Swedish universities during a period spanning several decades (1982–2005). This data is linked to register data from Statistics Sweden covering candidate lists in all public elections at all governmental levels (national, regional and local) in Sweden from 1991 to 2010.
We find a significant and substantial positive causal effect of acquiring office in a student organization on political candidacy later in life. Students who just barely managed to get elected into the SU councils (for a one-year term) had a six percentage points larger probability of becoming a candidate in a public election later in life than students who just missed out on their seat in the same council. These six percentage points should be compared to a baseline probability of around 18 percent to become a candidate among those who participated in the SU-elections but failed to acquire a seat. Thus, we find that participating in a SU council raises the candidacy rate by 34 percent in relative terms. In our view, this is a sizable effect. The effect persists at least a decade after the SU elections.
The results thus provide strong evidence to an increasingly contested issue within the field of political participation by showing that experiences within voluntary associations matter for individuals’ political involvement. More precisely, they show that acquiring a leadership position within a civil organization triggers (or enables) them to take the key step from political interest to participation in electoral politics. We interpret the results as strong evidence for a notion, rarely highlighted in political recruitment literature, that there are arenas between the family and representative institutions on which politically engaged citizens acquire crucial experiences that make them enter into public elections.