Gender and Dynastic Political Selection

Abstract

Throughout history and across countries, women appear more likely than men to enter politics on the heels of a close family relative or spouse. To explain this dynastic bias in women’s representation, we introduce a theory that integrates political selection decisions with informational inequalities across social groups. Candidates with dynastic ties benefit from the established reputations of their predecessors, but these signals of quality are more important to political newcomers such as women. Legislator-level data from twelve democracies and candidate-level data from Ireland and Sweden support the idea that dynastic ties are differentially more helpful to women, and that the quality of predecessors may be more relevant for the entry and evaluation of female successors than their male counterparts. The role of informational inequalities is also reflected in the declining dynastic bias over time (as more women enter politics), and in the differential effect of a gender quota across Swedish municipalities.

Keywords

dynasties gender representation gender quota Ireland Sweden

Introduction

The proportion of women in legislatures around the world has nearly doubled over the last two decades—from roughly 13 percent in 1999 to 24 percent in 2019—and this steady inflow of women into power has coincided with a heightened research interest in the topic of gender representation.¹ Many studies have pointed to institutional factors to explain variation in women’s descriptive representation, showing that it tends to be higher under proportional representation (PR) electoral systems, and when parties or countries adopt gender quotas with placement mandates (for a review, see Krook, 2018). Other studies have investigated the social backgrounds, network resources, and ambition of women who seek and win office (e.g., Lawless and Fox, 2005; Norris and Lovenduski, 1995; Sanbonmatsu, 2006; Schwindt-Bayer, 2011; Thomsen and Swers, 2017).

Despite huge strides in our understanding of when and how women enter politics, however, one of the most striking pathways to power for women remains poorly understood: political dynasties. In an emerging literature on dynastic political selection—that is, the entry of new politicians who are related by blood or marriage to current or former politicians—it has been observed across a variety of countries that women are more likely to be dynastic than their male counterparts.² This dynastic bias in women’s representation seems to apply to national-level executive and legislative offices (Baturo and Gray, 2018; Jalalzai and Rincker, 2018; Smith, 2018), as well as local-level offices (Geys, 2017; Labonne et al., 2019), but also appears to decline over time.

To illustrate these general empirical patterns, Figure 1 uses panel data covering all national-level legislators serving between 1945 and 2016 in twelve democracies: Australia, Canada, Finland, Iceland, Ireland, Israel, Italy, Japan, New Zealand, Norway, Switzerland, and the United States.³ The dynastic bias in women’s representation can be calculated by taking the ratio of dynastic legislators among the women and men in each post-election legislative session, and then taking the difference across sexes:

(\frac{Number of dynastic women}{Number of all women}) - (\frac{Number of dynastic men}{Number of all men})

(1)

Positive values thus indicate that a higher proportion of women are dynastic than their male counterparts.

Figure 1.

The dynastic bias in women’s representation in twelve democracies, 1945–2016.

Panel (a) of Figure 1 plots the average dynastic bias across all sessions for each country, and shows an aggregate bias in nine out of the twelve. Panel (b) illustrates the broad shift over time by dividing the data into two equal time periods: 1945 to 1980 and 1981 to 2016. The average dynastic bias decreased between these two periods in all but two countries, Israel and Norway, where the dynastic bias was already comparatively small (or negative) in the aggregate. Panel (c) pools the data and illustrates a negative correlation between the share of women in each legislature over time and the share of dynastic legislators among them.⁴ Finally, panel (d) directly plots the dynastic bias against the proportion of women in each legislature, showing a negative correlation between women’s representation and the difference in dynastic ties across sexes. In sum, the comparative data reveal two motivating empirical patterns: first, across many democracies, female legislators are more likely to be dynastic than their male counterparts; second, this dynastic bias tends to shrink over time (coinciding with an increase of women entering into politics).

To explain these patterns, we introduce a theory whereby dynastic ties function as a signal of “quality” that shapes political recruitment and selection decisions in the context of imperfect information about the qualifications of men and women. The first component of the theory is an assumption that the selectorate (which includes either the party elites or voters making decisions about political selection) has less accurate—and in some cases downward-biased—information about the true qualifications of women. The second component of the theory is that this informational inequality can be partly overcome by dynastic ties: the qualifications of would-be junior members of dynasties can be inferred from the observed qualifications of the senior member(s). Although this inherited signal of quality helps both male and female juniors, the effect is differentially larger for women due to their general informational disadvantages in a male-dominated political marketplace. This leads to a larger proportion of dynastic politicians among women than men. A clear implication from the theory is that the competence or quality of a (would-be) senior should strongly predict the entry or evaluation of a female junior, but less so a male junior.

The informational disadvantage for women also helps us understand why the dynastic bias would shrink over time as more women enter politics. The inflow of women into political offices should result in improved information about women as politicians among the selectorate and, accordingly, reduce the reliance on dynastic ties in recruiting and evaluating women. The increased entry of women into politics, and in particular into the ranks of the party elite who influence recruitment, also reduces the network and homosocial advantages of men in the communication of skills. As the selectorate’s perception of women’s competence becomes more accurate—or, alternatively, less negatively biased—the gender-differential role of dynastic ties should be mitigated.

We test our core prediction of inferred competence from senior dynasts using candidate-level data from two country cases: Ireland and Sweden. These cases have different histories, cultures, and institutions, but share the benefit of candidate-level data with accurate information on gender and dynastic ties. In the candidate-centered context of national elections in Ireland under the single-transferable vote (STV) system, we find that the difference in electoral strength between dynastic women and non-dynastic women is larger than the difference for their male counterparts, but that this difference has declined over time, such that dynastic status plays less of a role in the election of women in recent decades. In addition, although the quality of male politicians helps to predict the future entry of both male and female juniors, voters appear to impute quality from the dynastic seniors to dynastic juniors more so in the case of women than for men, as inferred from more similar vote shares between female dynastic juniors and their predecessors. In the party-centered context of local PR elections in Sweden, the nomination of dynastic women in electable positions can be tied to four measures of the quality of the dynastic senior; in contrast, these relationships are absent for male dynastic juniors. Thus, the differential utilization of a dynastic senior’s quality to evaluate women versus men appears to occur both among voters (though not necessarily party elites) in Ireland’s candidate-centered system, and among party elites in Sweden’s party-centered system.

We also explore the implications of our theory for the effect of gender quotas on the dynastic bias.⁵ The introduction of a quota delivers a shock to the selectorate, which is forced to increase the proportion of nominated (or elected) women. Our theory predicts that this will lead to an increase in the recruitment of dynastic women, and that this effect should vary positively with the magnitude of the increase in women required by the quota. We test this auxiliary prediction with the Swedish case, where the Central Party Board of the Social Democratic Party (SAP) introduced a zipper quota in 1994, obligating all municipality-level party organizations to alternate gender on the ballot for local elections. Our analysis indicates that the quota led to a quantitatively small, but positive, increase in dynastic women. However, as predicted, this effect was only temporary, with patterns reversing to the pre-quota level within two elections.

Our theory and extensive set of empirical tests make novel contributions to several research literatures. Our theoretical framework introduces the mechanism of informational differences at the selectorate level as a cause of lagging representation of women—a mechanism which should also be applicable to other political minorities. More importantly, our argument links research on women’s disadvantages in political recruitment to research on the advantages of dynastic ties to produce insights into the gender differences in dynastic recruitment observed in various contexts around the world. Our theory of dynastic ties as a signaling mechanism for candidate quality not only helps to explain the dynastic bias in women’s representation, but also the persistence of dynastic recruitment in modern democracies more generally. As such, it offers a point of departure for future studies of the political entry of other underrepresented groups, such as young people and LGBTQ⁺ individuals, who are underrepresented in politics but whose relatives do not face similar barriers to entry.

A Theory of Gender and Dynastic Selection

Our theory to explain the dynastic bias in women’s representation combines elements from political science, labor economics, and sociology—in particular, empirical and theoretical insights about the role of gender and information in recruitment and selection. We propose that dynastic ties mediate political selection in a context of imperfect information about the quality of individual men and women. Dynastic ties are more helpful for women than men because they help women overcome an informational disadvantage in the evaluation of their competence on the part of parties and voters.

The Selectorate and Screening Discrimination

The process of selection into politics involves two key sets of actors: the candidates themselves (who must want to run) and the selectorate, which is a sub-group of citizens who evaluate the qualifications of candidates. For our purposes, we will refer to the selectorate in a general sense to mean both party elites (or primary voters) who determine nominations, as well as general-election voters who determine which candidates are ultimately elected (i.e., the electorate).⁶ For both party elites and voters, the selectorate can be assumed to prefer candidates with higher “valence” (Groseclose, 2001; Stokes, 1963), a composite characteristic of skills, quality, and integrity (Dal Bó et al., 2017).

Research on labor markets outside the world of politics shows that selectorates are generally less accurate in assessing women’s job qualifications relative to those of men (e.g., Bjerk, 2008; Cornell and Welch, 1996). Job recruitment decisions are susceptible to “screening discrimination,” a type of statistical discrimination whereby recruiters are better at distinguishing between good and bad candidates among some groups, such as men, more than others, such as women. This lower accuracy in evaluating women’s qualifications may be because the selectorate itself is composed mostly of men (Lipman-Blumen, 1976; McPherson et al., 2001)⁷; because women are relative newcomers to the labor market (e.g., Goldin, 2014); or because women simply have fewer resources to signal their qualifications. Regardless of the reason, lower accuracy leads to a smaller reliance on individual valence in the perception of female job candidates, and a greater reliance on information about the average valence among all women. Naturally, this creates a disadvantage for women with above-average valence, who are perceived as less qualified than above-average men with the same true valence.

Political selection is a process where screening discrimination should also be prevalent, since the underlying quality of candidates is difficult to observe. A successful career in politics is not dependent on any specific qualification or educational degree; selectors are instead forced to evaluate qualifications that are gained over the course of diverse life paths. Moreover, individual contributions to political outcomes are often hard to disentangle and assess. Other traits, such as party loyalty or integrity, are observable over time and through repeated interactions, but not easily reportable on a CV.

Consistent with the idea of screening discrimination, research on political parties shows that selectors are indeed less informed about women than men as candidates. For example, Sanbonmatsu (2006) finds that political elites in the U.S. believe there to be more uncertainty about women than men as candidates. Crowder-Meyer (2013) studies political recruitment at the local level in the U.S., and finds that women have the largest disadvantage when party recruiters are men and rely on internal party networks to get information on viable candidates. In Swedish local parties, recruiters also report more uncertainty and perceived risks to nominating women, in particular if they have children (Widenstjerna, 2019). Women also tend to have lower access to political resources, including time, money, organizational affiliations, and connections (e.g., Niklasson, 2005; Norris and Inglehart, 2006), the latter of which are important channels for elites to obtain information about potential candidates.

As noted previously, the screening discrimination that results from these inequalities in information across sex may result in the underestimation of valence for above-average women. This effect is particularly relevant for the political sector given that there is positive selection into politics from the population—that is, the average candidate has better qualifications than the average citizen (e.g., Dal Bó et al., 2017). Indeed, most of the candidates who are reviewed by the selectorate will have true qualifications that exceed the average qualifications of their respective gender among the population.

In addition to this kind of screening discrimination, the underestimation of women’s qualifications could also stem from taste-based discrimination or negative stereotypes. A considerable body of research shows that many voters hold a negative view of women as candidates (e.g., Huddy and Terkildsen, 1993; McDermott, 1997), although recent research based on survey experiments suggests that voters in many contexts might no longer hold such discriminatory views (e.g., Clayton et al., 2020; Teele et al., 2018).⁸ In the presence of such gender discrimination, women may need to be more qualified and work harder at constituency work than their male counterparts in order to reach the same level of support from voters and parties, at least until they have a chance to prove their true quality (e.g., Anzia and Berry, 2011; Barnes et al., 2017; Folke and Rickne, 2016; Fulton, 2012; Horiuchi et al., 2020).⁹ This type of gender discrimination is not a necessary precondition for the theoretical role of dynastic ties that we suggest. To the extent that a negative bias against women might matter in a given context, it will only augment the disadvantage of women and, hence, add to the importance of the role that dynastic ties can play in bridging this disadvantage.¹⁰

In sum, screening discrimination will create a situation where male candidates will receive a better evaluation than female candidates with the same level of qualifications. This happens because the evaluation of women’s qualifications will be pulled down more by the group average, while men will be judged more according to their own valence. In addition, the more the selectorate relies on the perceived group average as an informational shortcut for evaluating women, the greater the disadvantage for women who are above-average in valence. To the extent that information about women is less accurate because of their relative newness to politics or the dominance of men in the selectorate, a reduction in either is expected to reduce this bias.

Dynastic Ties and the Signaling of Political Qualifications

A dynastic relationship can improve the information the selectorate possesses about a new candidate through two potential mechanisms. First, the selectorate can draw on information already gained about the qualifications of an experienced senior politician to infer the individual qualifications of the potential dynastic junior—that is, the new candidate who is linked to the senior politician via family ties. Second, the dynastic senior can facilitate the junior’s own communication of his or her individual quality to the selectorate. This can occur through raising the junior’s level of political ambition or through sharing networks or financial resources that facilitate the signaling of quality to the selectorate.

The idea of inferred quality from seniors to juniors has recently been formalized in the theoretical model of Besley and Reynal-Querol (2017). An assumption is that the selectorate is able to observe, and recall reasonably well, the qualifications of those who have held office within a given family. When the selectorate has imperfect information about a new candidate’s true qualifications, it can make use of the fact that—in the case of dynastic juniors—it nevertheless possesses some amount of information from the evaluation of his or her dynastic senior(s). It is reasonable for the selectorate to assume that qualifications for political office, like education and human capital accumulation, might be correlated across generations and within couples (due to assortative mating), and so an informational deficit about new candidates can be mitigated by information about the candidate’s family members who are (or were) themselves politicians. Importantly, although this inherited signal of quality should help both male and female juniors, the effect will be differentially larger for women due to screening discrimination bringing down the evaluations of non-dynastic women more so than those of non-dynastic men.

The second possible mechanism is that the dynastic senior can facilitate the junior’s own appreciation or communication of his or her individual quality, either by boosting political ambition or by sharing networks or financial resources. It is well documented that longer tenures in office increase the probability that a dynastic junior will follow into politics (e.g., Dal Bó et al., 2009; Querubin, 2016; Smith, 2018), with political socialization and inherited resources—such as accumulated networks and financial war chests—theorized as a key mechanism behind the ambition of juniors. Dynastic seniors, especially those of high valence, might therefore raise the probability of both male and female relatives seeking to enter electoral politics. However, because women tend to have lower levels of political ambition and access to resources than men (e.g., Galais et al., 2016; Lawless and Fox, 2005), any effect of dynastic ties should be differentially larger for women since (non-dynastic) men’s ambition and access to resources already tend to be at a higher level. Another implication of this logic is that junior women might tend to follow only exceptionally high-valence seniors, whereas men might follow seniors of any level of quality.¹¹

Ultimately, these two potential mechanisms—operating on what might be considered the “demand” and “supply” sides of political selection—are challenging to parse with observational data, and are both likely to play some role in generating the observed dynastic bias in women’s representation. Fortunately, the logic of each leads to a common testable hypothesis about the differential role of dynastic ties for men and women to take to our empirical investigations. If the selectorate relies more on the dynastic signal—the qualifications of the dynastic senior—when judging the qualifications of new female candidates compared to judging the qualifications of new male candidates, or if high-valence seniors are more instrumental in bridging gender gaps in ambition or resources for women versus men, then:

Hypothesis 1: the qualifications of a dynastic senior will have a stronger correlation with the selection or electoral success of a female dynastic junior than a male dynastic junior.

The larger the informational inequality between men and women, the larger the dynastic bias in women’s representation will be. Put differently, the less the selectorate knows about women, the more weight it will place on the valence of junior female candidates’ dynastic seniors and, accordingly, the larger the advantage of dynastic women relative to non-dynastic women. Hence, we should see a declining informational inequality over time as more women enter politics, as we have already documented descriptively in Figure 1. This theoretical expectation can be expressed as a second hypothesis:

Hypothesis 2: the dynastic bias in women’s representation will be larger when female politicians are less common.

This second hypothesis leads to an auxiliary prediction about the effect of gender quotas, which we can test empirically. The presence of women can vary across time, but also across space. Our general argument is that the fewer women there are in politics, the more ignorant the selectorate is about the quality of women as candidates. Thus, the use of dynastic informational shortcuts to evaluate potential female recruits should be starkest where there are currently fewer women. If parties in this situation are forced to select more women due to the introduction of a gender quota, the selectorate will rely on dynastic ties to a larger extent. The introduction of a gender quota should thus lead to a larger increase in the share of dynastic women in parties that are forced to increase the overall share of women by larger amounts.

Testing the Theory in Two Country Cases

We test our main hypothesis on candidate-level data from two countries—one with a candidate-centered electoral system (Ireland) and one with a party-centered system (Sweden). In Ireland, candidates are elected using the STV system in multimember districts that range in magnitude ( $M$ ) from three to five seats.¹² The electoral quota (also known as the Droop quota, and not to be confused with the gender quota we will discuss later for the case of Sweden) is defined as $\frac{Total votes cast}{M + 1} + 1$ . Voters rank candidates in order of preference, and those who receive enough votes to surpass the quota are elected, with their surplus votes redistributed to the next-preference candidate on each of the surplus ballots. If no candidate’s preference votes reach the quota, then the candidate with the lowest number of votes is eliminated and his or her votes are redistributed. This process continues until all seats are filled.

In Sweden, candidates are elected from multimember districts using closed or semi-open list PR. Voters are permitted to cast a preference vote for a single candidate on their chosen party ballot, but in practice these votes rarely alter the rank order determined by parties.¹³ In what follows, we present the tests for each country separately.

Ireland

Our dataset for Ireland includes all candidates in national Dáil elections and by-elections (i.e., off-cycle elections to fill vacancies) held between 1918 and 2016. The coding of dynastic ties is based on verified information from yearly political almanacs, biographical dictionaries, census records, and newspaper reports.¹⁴ We use the same definition of “dynastic” as in Figure 1 (i.e., a relation to a national-level politician—in this case, members of either of Ireland’s legislative chambers, the Dáil and Seanad). In our empirical analysis, we restrict the sample to candidates running between 1944 and 2016 to allow sufficient time for dynasties to emerge.

The analysis proceeds in two steps. First, we test simply whether being a dynastic junior is associated with a larger relative amount of electoral support among women than among men, and whether this difference decreases over time. This part of the analysis serves to establish the candidate-level dynamics of the dynastic bias in women’s representation (already documented for elected legislators in Figure 1), since we can directly compare electoral results for dynastic and non-dynastic candidates of each gender over time. For our purposes, we operationalize electoral support as the candidate’s share of the Droop quota obtained with first-preference votes (henceforth, we shall refer to this normalized measure of a candidate’s electoral support as the candidate’s “vote share” to avoid confusion with our discussion of gender quotas). We argue that this vote share measure reflects voters’ appreciation of aspects of the candidate, such as intelligence and charisma, that are related to the candidate’s true quality. We focus on first-time candidates because this is the period when candidates’ relationships to previous politicians should be most relevant to voters’ evaluations.¹⁵

Figure 2 plots, for each gender, the average vote share for dynastic and non-dynastic first-time candidates over time (grouped into decades due to sparsity of first-time female candidates in some individual elections). The descriptive patterns in Figure 2 illustrate that, indeed, (1) members of dynasties perform better in elections, (2) that the difference is larger for women than for men, and that (3) the differential impact of dynastic status for women decreases with time as the share of candidates who are women increases—as theoretically expected.¹⁶ In the early decades of the sample, dynastic women performed much better on average than non-dynastic women, and often even better than dynastic men. In the most recent decades, in contrast, the gap in electoral strength (as measured through the vote share) of dynastic versus non-dynastic women has narrowed.

Figure 2.

Average vote share of first-time dynastic and non-dynastic candidates in Ireland.

In the second step of the analysis, we test our main hypothesis that the evaluation of dynastic women is more closely associated with the evaluations of their dynastic senior(s) than it is for their male counterparts. For this part of the analysis, we consider two relevant selectorates. The party selectorate first chooses whom to nominate, and then each voter (the electorate) chooses which candidate to support among those who are nominated.¹⁷ We can thus test the extent to which each of these selectorates uses the dynastic signal—that is, the quality of (would-be) male seniors—in evaluating new (would-be) juniors of each sex.

To capture the parties’ use of the signal, we measure whether a previously elected male politician is followed into candidacy by a (male or female) junior sometime after his first election. Our key explanatory variable, which serves as a proxy for the “quality” of the would-be senior, is the politician’s average vote share obtained over the course of his entire career. Formally, we estimate the following baseline model:

P r e c e d e J u n i o r_{i} = α_{t} + α_{p} + β_{1} A v e r a g e V o t e S h a r e_{i} + ε_{i t},

(2)

where $P r e c e d e J u n i o r_{i}$ is a binary indicator of whether a (male or female) relative or spouse of the politician runs for election at any point in time after his own first election. The vectors of intercepts $α_{t}$ and $α_{p}$ are fixed effects for time (grouped into decades) and party. The estimate of interest, $β_{1}$ , captures the relationship between the electoral qualifications of the senior politician, $A v e r a g e V o t e S h a r e_{i}$ , and the event of having a junior of each specific gender nominated in a future election.

If our theoretical expectations are correct, then higher vote share averages should be more strongly correlated with the recruitment of female juniors than male juniors. Most especially, dynastic women should tend to follow high-quality male politicians, as the selectorate is less likely to favor a female relative of a weaker male politician if it uses the male politician’s past record to infer the would-be junior’s quality; in contrast, a male junior should be better able to signal his own quality, regardless of the quality of his predecessor(s). Nevertheless, we also have to keep in mind that the average probability of having a female junior is much lower than that of having a male junior due to the simple fact that women overall have a much lower probability of being nominated (11.28% of male politicians precede a female junior, compared to 26.23% who precede a male junior).

To examine voters’ use of the dynastic signal of quality, we relate the vote share of a (male or female) dynastic junior in his or her first election attempt to the average vote share obtained by his or her senior predecessor(s) in prior elections. In other words, the baseline regression model for voters’ evaluations is simply:

\begin{array}{l} J u n i o r V o t e S h a r e_{i} = α_{t} + α_{p} + β_{1} A v e r a g e V o t e \\ S h a r e o f S e n i o r {(s)}_{i} + ε_{i}, \end{array}

(3)

where $J u n i o r V o t e S h a r e_{i}$ is the proportion of the electoral quota earned by the first-time dynastic candidate with first-preference votes, $α_{t}$ and $α_{p}$ are again fixed effects for decade and party, and $A v e r a g e V o t e S h a r e o f S e n i o r {(s)}_{i}$ is the mean of the vote shares earned across all prior elections by the junior’s predecessor(s). If our theoretical expectations are correct, then the vote share of a first-time dynastic woman should be more strongly correlated with the past vote shares of her senior predecessor(s) than is the vote share of a male dynastic counterpart and his predecessor(s).

Table 1 shows the regression results. The top panel displays the results for Equation 2; the bottom panel displays the results for Equation 3. Specifications 1 and 3 estimate the baseline models with no other controls. Specifications 2 and 4 add individual-level control variables.¹⁸ The results for the top panel of regressions indicate that higher quality predicts both male and female successors among male politicians. All estimates for both men and women juniors are positive and statistically significant. However, contrary to the prediction of our theory, the dynastic signal does not appear to be more important for women than for men—the marginal estimated effect is less than half as large—although we note that the variance in predecessor quality is much greater for male juniors than female juniors, with many male juniors succeeding predecessors with both higher and lower average quality than the female juniors.¹⁹ Since the average probability of having a female follower is also about half the size of having a male follower (owing to the low number of women in politics in Ireland) it means that the relative importance is about the same for men and women. In other words, although the results confirm the central assumption of our theory—that selectors use the quality of the seniors to evaluate juniors—this signal does not appear to be more important for women when they are being evaluated by the party selectorate in Ireland.²⁰

Table 1.

Relationship Between a (would-be) Senior’s Quality and Dynastic Outcomes for Men and Women in Ireland.

DV=	Precede male junior		Precede female junior
	Overall = 26.23%		Overall = 11.28%
	(1)	(2)	(1)	(2)
Average vote share	0.554*** (0.083)	0.486*** (0.086)	0.200** (0.078)	0.222*** (0.083)
Observations	709	709	709	709
$R^{2}$	0.092	0.140	0.035	0.077
Party and time FE	Yes	Yes	Yes	Yes
Individual controls	No	Yes	No	Yes
DV =	Male junior vote share		Female junior vote share
	Overall Mean = 0.587		Overall Mean = 0.577
	(3)	(4)	(3)	(4)
Average vote share of senior(s)	0.118 (0.099)	0.107 (0.100)	0.656*** (0.216)	0.517*** (0.161)
Observations	190	190	51	51
$R^{2}$	0.287	0.298	0.547	0.621
Party and time FE	Yes	Yes	Yes	Yes
Individual controls	No	Yes	No	Yes

All regressions estimated with OLS. The sample for the top panel includes all elected male politicians (n = 709) who stopped running between 1944 and 2011 (i.e., allowing for a junior to appear in 2016, the last election in the data). The dependent variable is whether the male politician (a potential senior) preceded a male or female dynastic candidate (junior). Average vote share (measured over the politician’s entire career) serves as a proxy for quality. Specification 1 includes only party and decade fixed effects. Specification 2 adds individual-level controls for age (five groups) and fixed effects for the first year (Dáil session) the candidate ran. Appendix Table B.2 shows the results of alternative specifications. The sample for the bottom panel is male juniors (n = 190) and female juniors (n = 51) whose predecessor(s) ran in a previous Dáil election. The dependent variable is a (male or female) dynastic candidate’s vote share in his or her first election attempt. Average vote share of senior(s) is the average vote share obtained over the course of all preceding family members’ electoral careers (including any members, beyond the elected seniors, who may have failed to win election). Specification 3 includes only party and decade fixed effects. Specification 4 adds the individual-level control variable for age (five groups). Robust standard errors are shown in parentheses.

p < .1. **p < .05. ***p < .01.

The results in the lower panel of Table 1, in contrast, paint a picture more in line with our theoretical predictions. The vote share of juniors is much more strongly correlated with the vote share of the senior(s) among female juniors compared to male juniors. The estimate for men in both specifications is small and statistically insignificant; for women, the estimate is about 5 times larger, and statistically significant, even though the number of women is fewer. This gap is entirely consistent with the prediction of our theory, and confirms that dynastic women’s evaluations (in this case, by voters) are more closely related to the quality of their predecessors. In other words, women appear more dependent on the dynastic signal than men when they are evaluated by voters.²¹

Sweden

Our Swedish dataset contains the whole universe of candidates for 290 municipal councils, 17 county councils, and the national parliament (Riksdag), covering nine elections over a 30-year period (1982–2010). We focus our analysis on the municipal councils, which hold substantial political power and offer a large sample size for testing our model. Each council has between 31-101 elected members. The municipal party organization is the main selectorate, with each implementing its own procedure for ranking candidates on the ordered ballot, usually involving internal nominations by party clubs (in parties on the left of the ideological spectrum), or by internal primaries (in the center-right parties).²² A complete record of every politician’s personal ID code allows us to anonymously match each person to his or her 1) siblings, 2) parents, and 3) spouse, using highly accurate register data.²³ We can thus define a dynastic politician as someone with a close family member who previously held political office at the local, county, or national level.

While our measurement of family ties is extremely accurate, and covers a wider range of political offices than in the Irish case, it does not capture more distant family relationships, and is temporally truncated because we can only verify politicians as dynastic if they had a relative in office from 1982 onwards. The most accurate measurement of dynastic politicians will thus exist for recent election(s) in the sample, but the truncation will be less important when we compare dynastic men and women at any given point in time. The Swedish data also contain detailed background variables from administrative records—most importantly, the person’s level of education and, for men, evaluations of IQ and leadership abilities from the military enlistment process (further discussed later).

Since Sweden was not among the twelve democracies in the comparative data presented earlier (owing to the time truncation), we begin by establishing that there is a smaller proportion of women than men in office, but that dynastic politicians account for a larger share of the women than the men. Both of these conditions are shown in Figure 3, which uses data from all municipal councilors from 1998 to 2010. We exclude the five earliest elections from the empirical tests to sidestep the influence of measurement error in dynastic status due to time truncation.

Figure 3.

Proportion of men and women among local councilors (left) and the proportion of dynastic politicians among them (right) in Sweden.

We use two approaches to measure the quality of the dynastic senior in the Swedish case. These two approaches are complementary in the sense that each addresses a relative weakness in the other. First, we use the highest List Rank on the ballot achieved by each politician during his career. This measure captures the party’s evaluation of the person’s suitability for top posts, since rank on the ballot approximates the internal power structure within the party (Folke and Rickne, 2016). The higher (lower numerically in terms of rank) that a candidate has reached, the more favorably he has been viewed by the party, which implies that he possesses characteristics that the party values, such as intelligence, loyalty, and so on. However, using the position in the party hierarchy to measure the evaluation of a politician comes with the drawback that power and influence could have an independent effect on the evaluation of a politician’s dynastic junior. If there is a gender difference in this potential impact, we risk confounding our measurement of the selectorate’s evaluation with the politician’s power.

The second set of measures captures the dynastic senior’s individual qualifications in terms of education and personal traits, each of which has previously been shown to predict internal success within a party (see Dal Bó et al., 2017). When using these measures, we can hold list rank constant, controlling for the political power of the senior. The first measure is Years of Education, which is a common proxy for quality, and broadly captures enhanced practical skills, signaling ability, and civic engagement (e.g., Besley and Reynal-Querol, 2017; Franceschet and Piscopo, 2008). The second and third measures come from Sweden’s military enlistment register and capture scores of the recruits’ Cognitive Ability and Leadership Skills. Operationally, both scores are measured on a discrete 1 to 9 scale that is approximately normally distributed with a mean of 5.²⁴

Our general empirical strategy for the Swedish case is similar to our strategy for analyzing nominations in Irish case. We relate the quality of a (potential) dynastic senior to the probability of having a male junior or—in a separate regression—a female junior, elected in the future. To make the analysis comparable to the Irish case, and because we only have the last two qualification measures for men, we restrict our sample to male (would-be) seniors. As with the Irish test, the intuition is that, if women face greater informational disadvantages in recruitment and selection than their male counterparts, then it may mean that only the highest quality male politicians will be followed by female juniors; in contrast, male juniors might follow predecessors of varying quality.

A key difference with the Irish analysis is that our outcome variable is whether the (would-be senior) male politician precedes a future junior who is elected, rather than simply nominated. This restriction reflects the fact that list positions play the largest part in determining electoral outcomes for individual candidates, and parties will focus most on the qualifications of the candidates who are most likely to represent them in office (owing to a list position that comes with a high probability of election), especially if the concern is evaluating potential future behavior in office (Widenstjerna, 2019).²⁵ The formal regression equation used for the analysis is:

P r e c e d e J u n i o r_{i} = α_{m} + α_{t} + α_{p} + β_{1} Q u a l i t y_{i} + β_{2} F i r s t Y e a r_{i} + ε_{i t},

(4)

where $P r e c e d e J u n i o r_{i}$ is a binary indicator of whether a relative or spouse of the politician (the dynastic junior) enters into a local council at any point in time after the senior’s own first term in office. Politicians who married their dynastic senior at any point after being nominated are not included in this definition.²⁶ The estimate of interest, $β_{1}$ , captures the relationship between the qualifications of the senior politician, $Q u a l i t y_{i}$ , and the event of having a junior of each specific gender. The vectors of intercepts $α_{m}$ , $α_{t}$ , and $α_{p}$ are fixed effects for municipality ( $m$ ), election period ( $t$ ), and party ( $p$ ), and $F i r s t Y e a r_{i}$ represents fixed effects for the first year in which the candidate was ever elected.²⁷

We estimate three specifications of Equation 4 and report the results in Table 2. The first specification for each gender (columns 1 and 4) includes only the senior’s qualifications, $Q u a l i t y_{i}$ , and the fixed effects. In columns 2 and 5, we add controls for a set of individual control variables for the dynastic senior.²⁸ In columns 3 and 6, we add fixed effects for the senior’s highest list rank to control for the political power of the senior. Note that we could also expect a positive correlation between quality and power, so the estimate on quality will likely have a downward bias after including this control.

Table 2.

Relationship Between the Qualifications of the Dynastic Senior and the Event of Having a Male Dynastic Follower (Left) or a Female Dynastic Follower (Right) in Sweden.

	Precede male junior			Precede female junior
	Overall = 4.2%			Overall = 6.4%
	(1)	(2)	(3)	(4)	(5)	(6)
List rank	0 (0.04)	0 (0.04)		−0.10** (0.04)	−0.10** (0.04)
Observations	23,667	23,667		23,667	23,667
Years of education	−0.01(0.05)	−0.07(0.05)	−0.08* (0.05)	0.18*** (0.05)	0.14*** (0.05)	0.12** (0.05)
Observations	23,327	23,327	23,325	23,327	23,327	23,325
Cognitive ability	−0.06(0.20)	−0.11(0.20)	−0.13(0.20)	0.47* (0.24)	0.44* (0.24)	0.43* (0.24)
Observations	5,927	5,927	5,926	5,927	5,927	5,926
Leadership skills	−0.05(0.17)	−0.04(0.17)	−0.08(0.17)	0.37* (0.21)	0.35* (0.21)	0.34(0.21)
Observations	4,954	4,954	4,953	4,954	4,954	4,953
Municipality, party, time, First Year FE	Yes	Yes	Yes	Yes	Yes	Yes
Individual controls	No	Yes	Yes	No	Yes	Yes
List rank FE	No	No	Yes	No	No	Yes

All regressions estimated with OLS as linear probability models, and include fixed effects for election municipality, year, the first year the politician was elected, and party. Control variables include five age categories (30–49 = reference), and binary indicators for being foreign born and for having at least one foreign-born parent. The number of observations varies across models due to data availability. Robust standard errors are reported in parentheses.

p < .1. **p < .05. ***p < .01.

The results for each of the four quality measures are presented in separate panels in Table 2. Each provides support for our hypothesis. First, there is a strong negative correlation between List Rank (the lower the value, the higher the rank) and the probability of having a female junior. Moving up one position on the ballot is associated with a one percentage point (1.5 percent in relative terms) higher probability of having a dynastic female junior. There is no relationship between list rank and having a male junior. For each of the three other qualification measures, we find a positive and statistically significant relationship with the event of having a female junior, but not with having a male junior. One additional year of education is associated with an almost 3 percent higher probability of having a female junior. For the two draft-based measures, moving one point on the 9-point scale increases the probability by between 6 and 8 percent. Thus, the relationship between our quality measures and having a female dynastic junior are not only statistically significant, but also substantial.

Importantly, none of the estimates is substantively affected by adding the individual control variables, which suggests that the estimates are not simply picking up a spurious relationship. Furthermore, the estimates on the three direct qualification measures are barely affected when we control for the senior’s list rank. This is important for the interpretation of these estimates, as it shows that the three qualification measures are not simply capturing the political power of the senior. In sum, the results for the case of Sweden also provide empirical support for our core hypothesis that the selectorate’s evaluation of the qualifications of a dynastic senior has a stronger correlation with its evaluation of a female dynastic junior than with its evaluation of a male dynastic junior. In this case, it is the party selectorate that appears to make greater use of the dynastic signal to evaluate women than men.

Discussion

The findings in both of our case studies teach us something new about how a political selectorate evaluates male and female dynastic candidates. For candidate groups who are newcomers to politics—in this case women—the quality of a dynastic predecessor appears to be used as an informational shortcut. For groups that the selectorate is accustomed to evaluating, and therefore has better information about—in this case men—this informational shortcut is less important (in the case of Irish voters), or not important at all (in the case for Swedish party elites). Our analysis thus suggests that the differential role of the dynastic signal for men and women might apply to some extent across different types of selectorates—the Irish results illustrate the process for voter evaluations, after women have already selected into running as candidates (which does not appear to depend on predecessor quality any more than it does for men), while the Swedish results illustrate the process for evaluations made by party elites in composing lists of candidates.

Although the results for both country cases generally point in the expected direction of our hypotheses, data limitations prevent us from claiming that these results are completely conclusive. The comparative patterns in Figure 1, for example, suggest that institutional contexts might condition the relationship between gender and dynastic recruitment. Smith (2018) argues that the incentives to recruit dynastic candidates—in general, regardless of gender—should be higher under candidate-centered contexts (such as Ireland) than in party-centered contexts (such as Sweden), owing to differences in the electoral value of the “personal vote” (Carey and Shugart, 1995). Although we find evidence of the dynastic bias in women’s representation in Sweden, many of the other party-centered PR cases in the comparative data, such as Norway and Switzerland, appear to feature a lower dynastic bias, while candidate-centered cases, such as Ireland, New Zealand, the United States, and Canada, are among the cases with the highest dynastic bias. Future research should further explore how such contextual differences might mediate the relationships we have documented in our analysis.

Gender Quotas and Dynastic Selection

In this section, we examine our auxiliary prediction, which relates to the implications of our theory for the impact of a gender quota on the dynastic selection of women. The decrease in the dynastic bias over time, already evident in the comparative data in Figure 1, suggests that a reduction in informational inequalities (due to more women entering politics) can reduce the extent to which the selectorate must rely on the dynastic signal in evaluating women. The reverse naturally also applies. The more ignorant the selectorate is about the quality of women, the larger the relative reliance on dynastic ties in their recruitment and selection. The effect will be most dramatic when a party with relatively few women is forced to rapidly increase the number of women in its ranks.

We can explore this auxiliary prediction by evaluating the relative effect of the introduction of a gender quota across municipalities with varying existing levels of gender representation. Quotas are introduced in organizations where women’s initial presence is relatively low, that is, where there is a perceived need to increase women’s descriptive representation. From the intuition of our theory, we can deduce that fewer women means a greater relative ignorance about women’s quality, and hence a larger reliance on the dynastic signal to fill the reserved seats that open up through the quota. We expect that, in the short run, a gender quota will lead to an increase in the probability that a recruited woman is dynastic. Notably, the quota should be unrelated to the probability that a recruited man is dynastic because the selectorate was already capable of evaluating men’s true qualifications prior to the introduction of the quota and this capability would not change with the reform. Because women’s entry through the quota results in them having a presence in the organization—thus building information about their true qualifications—we expect the dynastic bias in women’s recruitment to decrease in the long run following the imposition of the quota.

We test this prediction using the same identification strategy as O’Brien and Rickne (2016), who exploit the fact that Sweden’s largest political party, SAP, introduced a zipper quota in the 1994 local elections.²⁹ This quota mandated local parties to alternate male and female names on their lists, leading to near-equity because seats are counted from the top of the ballot. The smaller the proportion of elected women prior to the quota, the larger the forced increase.³⁰ The quota was imposed by the central party organization, whose hand was forced by an outside group that threatened to start a feminist party, thus creating an exogenous shock to the local parties’ recruitment practices.

We use a difference-in-difference specification that isolates the exogenous quota shock, restricting the data sample to the new (first-term) politicians in each election (1988–2010). The regression specification is estimated at the level of the individual, and takes the form:

D y n a s t i c_{i} = α + β_{t} Δ w_{1994 - 1991, m} * E l e c t i o n_{t} + γ_{t} E l e c t i o n_{t} + θ_{m} M u n_{m},

(5)

where the outcome variable $D y n a s t i c_{i}$ is a binary indicator that takes the value 1 if the newly recruited politician was dynastic, and 0 if the person was not dynastic. The treatment is denoted $Δ w_{1994 - 1991, m}$ and measures the change in the proportion of elected women in the local SAP in municipality $m$ between the years 1991 and 1994. To estimate the effect of the quota, the impact measure is interacted with a dummy variable for each election between 1988 and 2010: $E l e c t i o n_{t}$ . The 1991 election is used as the reference year. This means that the key vector of estimates $β_{t}$ captures the difference in the probability that a newly elected politician is dynastic in the election year in question, for example in 1994 (the first year with the quota) compared to the reference year, 1991.

Figure 4 graphically displays the estimates of $β_{t}$ from the regression analysis.³¹ To make the estimates more easily interpretable, we scale the quota impact variable so that the estimate is interpreted as the change in the probability that a politician is dynastic if the party were forced to increase the proportion of elected women by ten percentage points, an amount which corresponds to the average of the variable across all municipalities (at the time, there were 285 municipalities).

Figure 4.

Effect of the quota-induced change in the proportion of elected women in Sweden on the recruitment of dynastic men and women.

The plotted estimates and 95% confidence intervals in Figure 4 suggest that the quota temporarily increased the recruitment of dynastic women. The solid dots for the elections in 1994 and 1998 indicate that a 10-percentage-point increase in the proportion of women, brought on by the quota, increased the probability that each recruited woman was dynastic by about 3 percentage points. Due to the large standard errors, neither estimate is statistically significant (the 1994 estimate falls short with the closest possible margin), which means that the estimates should be interpreted with some caution. The estimate is close to zero in all of the last three elections. For men (point estimates and 95% confidence intervals represented with hollow dots), there is no corresponding change in dynastic recruitment. We also find that the effect for women decreases over time—the estimates for the 2002, 2006, and 2010 elections are all close to zero. These patterns are consistent with the theory, in that the quota should ultimately decrease screening discrimination of female candidates by bringing more women into politics.

An alternative interpretation of our quota results is that men forced out by the quota might orchestrate the nomination of their female relatives to take their place in the councils. We examined if this is the case. In the year of the quota’s introduction (1994), 1,652 men left the ranks of Social Democratic municipal councilors and 1,492 women joined (out of 6,103 total seats). Among these men, 103 were dynastic seniors and 16 had a dynastic female junior enter directly in the first election with the quota. But among the women entrants, these 16 dynastic juniors were just 9 percent (16/184) of all dynastic entrants. Moreover, this ratio was higher in other years (roughly 18 percent in both 1988 and 1991). It thus seems unlikely that the effect that we see in Figure 4 is driven by the recruitment of dynastic women as “placeholders” for male relatives.³²

Conclusion

Across many democracies, a dynastic relationship to a previous politician is a dramatically more prominent channel of recruitment into office for women than for men. In this study, we have provided comparative evidence across thirteen countries that women’s entry into politics is often characterized by this dynastic bias. Our theory to explain this pattern offers the novel hypothesis that socioeconomic groups who are newcomers in the political arena—such as women—are more reliant on signaling their qualifications via dynastic seniors who are already insiders. One way of understanding this claim is that selectorates, both voters and parties, want to avoid the risk associated with an unknown candidate, and so use shortcuts to impute missing information. Our theory provides a new perspective on dynastic recruitment, offers an explanation for the gender difference in this recruitment channel, and generates valuable predictions for the relationship between dynastic status and candidate quality, as well as for the impact of gender quotas.

We find clear empirical support for the signaling effect of dynastic ties in our empirical case studies of Ireland, a candidate-centered system, and Sweden, a party-centered system. In Ireland, women’s overall lower electoral strength can be bridged by having dynastic seniors, and the extent of electoral success enjoyed by such dynastic women is closely associated with that of their seniors. Illustrating the gender difference for the signaling channel, the same relationship between the quality of the dynastic senior and the electoral success of the junior is not as prominent among men. Similar patterns in the differential relationship between the quality of seniors and dynastic men and women prevail in Sweden.

We also provide evidence that the dynastic bias in women’s representation decreases over time in most democracies, a pattern which our theory implies is driven by a decreasing level of screening discrimination against women as more women enter into politics. These findings suggest room for optimism with regard to the pathways for improving gender representation around the world. Political dynasties have generally been more common in developing democracies (Smith, 2018), where at the same time, gender representation has often lagged behind—at least until recent shifts owing in part to the adoption of gender quotas (Matland, 1998; Tripp and Kang, 2008).³³ Although women may initially rely more heavily on dynastic ties to get elected when women in politics are scarce, these early pioneers should improve the selectorate’s information about women’s quality, thus reducing barriers for future women.

Future studies of this process could be expanded to investigate recruitment patterns for other political minorities and newcomers, such as young people (among whom there also appears to be a large dynastic bias), and openly-LGBTQ⁺ politicians.³⁴ For these groups, a dynastic senior could help with inferred quality assessments or direct encouragement to seek office. Interestingly, and in contrast, the same cannot be said for historically underrepresented social-class minorities, ethnic or racial minorities, or people with an immigrant background. These groups are systematically less likely to elect a (potential) dynastic senior, meaning that our theory can help explain the absence of these groups from candidacy and election, rather than their disproportionate entry via dynastic ties.

In addition, our findings suggest an avenue for future research on the quality of male and female dynastic juniors. Previous research conjectures that dynastic women might be relatively unqualified and controllable “proxies” or “placeholders” standing in for male relatives (e.g., Jalalzai, 2013), although the evidence for this quality difference is mixed (Labonne et al., 2019). It is possible that dynastic ties help parties select more qualified women when high-quality seniors provide a signal of high-quality daughters, sisters, or wives. Our theory suggests that using dynastic ties to guide the selection of politicians is not always bad for ensuring meritocratic selection. A negative overall effect might, however, exist if dynastic recruits from some political minorities are advantaged over potentially more qualified candidates of the same group. Finally, future research should separately theorize about the different dynastic recruitment paths for different types of women, such as wives or widows versus daughters.³⁵

Our findings regarding gender quotas also have important implications for both academics and advocates of women’s political representation. The estimated effects of the quota are imprecise, but indicate that a gender quota might lead to a disproportionate inflow of dynastic women in the short term. Dynastic ties appear to play a role in the recruitment of women when parties are forced to raise women’s descriptive representation by affirmative action. This may be somewhat surprising in the Swedish case, given that the local SAP parties had an average proportion of women that already exceeded 30 percent before the quota, with strong women’s branches acting as pipelines into candidacy. An inflow of dynastic women under these circumstances of a healthy supply of women available to run suggests that similar consequences would likely be observed in other contexts.

Another important way to extend our understanding of gender and dynastic recruitment could be through in-depth studies of countries with different political institutions, party practices of recruitment, and norms and traditions. In addition, although we have focused on how dynastic ties operate as a signaling device in the recruitment and election processes in advanced democracies, the processes and patterns may differ in new or developing democracies. Another promising avenue for future research would be to explore the introduction of institutional reforms, such as quotas, new candidate selection procedures, or new electoral rules, that force parties to shift their recruitment behavior, or—with an eye to the theoretical narrative presented here—reforms that change the ability of candidates to showcase their true qualifications to voters.

Finally, the normative implications of our theory and empirical results merit further exploration. Our results paint a picture of a political recruitment process where women need compensating resources to help pave the way to political power. Even with these resources, the pathways to politics are still unequal. We should therefore view gender differences in the recruitment and promotion of dynastic politicians as an important symptom of a much larger problem in the political system.

Supplemental Material

FRS-CPS-APPENDIX – Supplemental material for Gender and Dynastic Political Selection

Supplemental material, FRS-CPS-APPENDIX for Gender and Dynastic Political Selection by Olle Folke, Johanna Rickne and Daniel M. Smith in Comparative Political Studies

Footnotes

Acknowledgements

We thank Mona-Lena Krook, Jon Fiva, Thomas Däubler, Simon Chauchard, anonymous referees, and attendees at panels or workshops at the American Political Science Association, International Political Science Association, European Conference on Politics and Gender, Gothenburg University, Harvard University, and Université Laval for helpful comments, and Jonas Allerup, Johan Arntyr, Mark Daley, Sirus Dehdari, Max Goplerud, Roza Koban, Elin Molin, Darragh Nolan, and Aaron Roper for valuable research assistance. Replication materials are available at Dataverse: .

Authors’ Note

In addition to their primary institutional affiliations, Olle Folke is also an affiliated researcher at the Swedish Institute for Social Research at Stockholm University, Sweden, and Johanna Rickne is also a part-time Professor of Economics at Nottingham University, United Kingdom.

Declaration of Conflicting Interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Folke and Rickne thank the Riksbankens Jubileumsfond (Grant Number P16-0786:1) and the Knut and Alice Wallenberg Foundation for financial support. Smith thanks the Weatherhead Center for International Affairs at Harvard University for financial support.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Daniel M. Smith

Supplemental Material

Supplemental material for this article is available online at the CPS website

Notes

Author Biographies

Olle Folke is an associate professor at the Department of Government at Uppsala University and affiliated researcher at the Swedish Institute for Social Research, Stockholm University. His primary research fields are applied political economics and gender and politics.

Johanna Rickne is a professor of Economics at the Swedish Institute for Social Research, Stockholm University, and part-time Professor of Economics at Nottingham University. Her research interests include gender and politics and labor economics.

Daniel M. Smith is an associate professor of comparative politics in the Department of Government at Harvard University. His research focuses on political institutions and representation in democracies.

References

Anzia

S. F.

Berry

C. R.

(2011). The Jackie (and Jill) Robinson effect: Why do congresswomen outperform congressmen? American Journal of Political Science, 55(3), 478–493.

Barnes

T. D.

Branton

R. P.

Cassese

E. C.

(2017). A reexamination of women’s electoral success in open seat elections: The conditioning effect of electoral competition. Journal of Women, Politics and Policy, 38(3), 298–317.

Basu

(2016). Women, dynasties, and democracy in India. In Chandra

(Ed.), Democratic dynasties: State, party, and family in contemporary Indian politics (pp. 127–160). Cambridge University Press.

Batto

N. F.

(2018). Legacy candidates in Taiwan elections, 2001–2016: Just a bunch of bullies. Asian Survey, 58(3), 486–510.

Baturo

Gray

(2018). When do family ties matter? the duration of female suffrage and women’s path to high political office. Political Research Quarterly, 71(3), 695–709.

Besley

Reynal-Querol

(2017). The logic of hereditary rule: Theory and evidence. Journal of Economic Growth, 22(2), 123–144.

Bjerk

(2008). Glass ceilings or sticky floors? Statistical discrimination in a dynamic model of hiring and promotion. The Economic Journal, 118(530), 961–982.

Carey

J. M.

Shugart

M. S.

(1995). Incentives to cultivate a personal vote: A rank ordering of electoral formulas. Electoral Studies, 14(4), 417–440.

Casas-Arce

Saiz

(2015). Women and power: Unpopular, unwilling, or held back?” Journal of Political Economy, 123(3), 641–669.

10.

Clayton

Robinson

A. L.

Johnson

M. C.

Muriaas

(2020). (How) do voters discriminate against women candidates? Experimental and qualitative evidence from Malawi. Comparative Political Studies, 53(3–4), 601–630.

11.

Cornell

Welch

(1996). Culture, information, and screening discrimination. Journal of Political Economy, 104(3), 542–571.

12.

Crowder-Meyer

(2013). Gendered recruitment without trying: How local party recruiters affect women’s representation. Politics & Gender, 9(4), 390–413.

13.

Dahlerup

(Ed.) (2006). Women, quotas and politics. Routledge.

14.

Dal Bó

Finan

Folke

Persson

Rickne

. (2017). Who becomes a politician? The Quarterly Journal of Economics, 132(4), 1877–1914.

15.

Dal Bó

Snyder

. (2009). Political dynasties. Review of Economic Studies, 76(1), 115–142.

16.

Esteve-Volart

Bagues

(2012). Are women pawns in the political game? Evidence from elections to the Spanish senate. Journal of Public Economics, 96(3–4), 387–399.

17.

Folke

Rickne

(2016). The glass ceiling in politics: Formalization and empirical tests. Comparative Political Studies, 49(5), 567–599.

18.

Folke

Persson

Rickne

(2016). The primary effect: Preference votes and political promotions. American Political Science Review, 110(3), 559–578.

19.

Franceschet

Piscopo

J. M.

(2008). Gender quotas and women’s substantive representation: Lessons from Argentina. Politics & Gender, 4(3), 393–425.

20.

Fulton

S. A.

(2012). Running backwards and in high heels: The gendered quality gap and incumbent electoral success. Political Research Quarterly, 65(2), 303–314.

21.

Gagliarducci

Paserman

M. D.

(2012). Gender interactions within hierarchies: Evidence from the political arena. The Review of Economic Studies, 79(3), 1021–1052.

22.

Galais

Öhberg

Coller

(2016). Endurance at the top: Gender and political ambition of Spanish and Swedish MPs. Politics & Gender, 12(3), 596–621.

23.

Gallagher

(1980). Candidate selection in Ireland: The impact of localism and the electoral system. British Journal of Political Science, 10(4), 489–503.

24.

Gertzog

I. N.

(1980). The matrimonial connection: The nomination of congressmen’s widows for the house of representatives. The Journal of Politics, 42(3), 820–833.

25.

Geys

(2017). Political dynasties, electoral institutions and politicians’ human capital. The Economic Journal, 127(605), F474–F494.

26.

Goldin

(2014). A pollution theory of discrimination: Male and female differences in occupations and earnings. In Boustan

L. P.

Frydman

Margo

R. A.

(Eds.), Human capital in history: The American record (pp. 313–353). University of Chicago Press.

27.

Groseclose

(2001). A model of candidate location when one candidate has a valence advantage. American Journal of Political Science, 45(4), 862–886.

28.

Horiuchi

Smith

D. M.

Yamamoto

(2020). Identifying voter preferences for politicians’ personal attributes: A conjoint experiment in Japan. Political Science Research and Methods, 8(1), 75–91.

29.

Huddy

Terkildsen

(1993). Gender stereotypes and the perception of male and female candidates. American Journal of Political Science, 37(1), 119–1147.

30.

Ibarra

(1992). Homophily and differential returns: Sex differences in network structure and access in an advertising firm. Administrative Science Quarterly, 37(3), 422–447.

31.

Jalalzai

(2013). Shattered, cracked, or firmly intact? Women and the executive glass ceiling worldwide. Oxford University Press.

32.

Jalalzai

Rincker

(2018). Blood is thicker than water: Family ties to political power worldwide. Historical Social Research, 43(4), 54–72.

33.

Kincaid

D. D.

(1978). Over his dead body: A positive perspective on widows in the U.S. congress. The Western Political Quarterly, 31(1), 96–104.

34.

Krook

M. L.

(2018). Electoral systems and women’s representation. In Herron

E. S.

Pekkanen

R. J.

Shugart

M. S.

(Eds.), The Oxford handbook of electoral systems (pp. 175–192). Oxford University Press.

35.

Labonne

Parsa

Querubin

(2019). Political dynasties, term limits and female political empowerment: Evidence from the Philippines (NBER Working Paper No. 26431). National Bureau of Economic Research, Inc.

36.

Lawless

J. L.

Fox

R. L.

(2005). It takes a candidate: Why women don’t run for office. Cambridge University Press.

37.

Lipman-Blumen

(1976). Toward a homosocial theory of sex roles: An explanation of the sex segregation of social institutions. Signs, 1(3), 15–31.

38.

Matland

R. E.

(1998). Women’s representation in national legislatures: Developed and developing countries. Legislative Studies Quarterly, 23(1), 109–125.

39.

McDermott

M. L.

(1997). Voting cues in low-information elections: Candidate gender as a social information variable in contemporary United States elections. American Journal of Political Science, 41(1), 270–283.

40.

McDonagh

(2002). Political citizenship and democratization: The gender paradox. American Political Science Review, 96(3), 535–552.

41.

McPherson

Smith-Lovin

Cook

J. M.

(2001). Birds of a feather: Homophily in social networks. Annual Review of Sociology, 27(1), 415–444.

42.

Niklasson

(2005). Contact capital in political careers: Gender and recruitment of parliamentarians and political appointees (PhD thesis). University of Gothenburg.

43.

Nishizaki

(2018). New wine in an old bottle: Female politicians, family rule, and democratization in Thailand. The Journal of Asian Studies, 77(2), 375–403.

44.

Niven

(1998). Party elites and women candidates: The shape of bias. Women & Politics, 19(2), 57–80.

45.

Norris

Lovenduski

(1995). Political recruitment: Gender, race and class in the British parliament. Cambridge University Press.

46.

Norris

Inglehart

(2006). Gendering social capital: Bowling in women’s leagues? In O’Neill

Gidengil

(Eds.), Gender and social capital (pp. 73–98). Routledge

47.

O’Brien

Rickne

(2016). Gender quotas and women’s political leadership. American Political Science Review, 110(1), 112–126.

48.

Pinkston

J. C.

(2003). Screening discrimination and the determinants of wages. Labour Economics, 10(6), 643–658.

49.

Querubin

(2016). Family and politics: Dynastic persistence in the Philippines. Quarterly Journal of Political Science, 11(2), 151–181.

50.

Sanbonmatsu

(2006). Where women run: Gender and party in the American States. University of Michigan Press.

51.

Schwindt-Bayer

L. A.

(2011). Women who win: Social backgrounds, paths to power, and political ambition in Latin American legislatures. Politics & Gender, 7(1), 1–33.

52.

Smith

D. M.

(2018). Dynasties and democracy: The inherited incumbency advantage in Japan. Stanford University Press.

53.

Smith

D. M.

Martin

(2017). Political dynasties and the selection of cabinet ministers. Legislative Studies Quarterly, 42(1), 131–165.

54.

Stokes

D. E.

(1963). Spatial models of party competition. American Political Science Review, 57(2), 368–377.

55.

Stuart

T. E.

Hoang

Hybels

R. C.

(1999). Interorganizational endorsements and the performance of entrepreneurial ventures. Administrative Science Quarterly, 44(2), 315–349.

56.

Teele

D. L

Kalla

Rosenbluth

(2018). The ties that double bind: Social roles and women’s underrepresentation in politics. American Political Science Review, 112(3), 525–541.

57.

Thomsen

D. M.

Swers

M. L.

(2017). Which women can run? Gender, partisanship, and candidate donor networks. Political Research Quarterly, 70(2), 449–463.

58.

Tripp

A. M.

Kang

(2008). The global impact of quotas: On the fast track to increased female legislative representation. Comparative Political Studies, 41(3), 338–361.

59.

Widenstjerna

(2019). Vem väljar vem—och varför? [Who recruits whom to politics, and why?] (PhD thesis). Mid Sweden University.

60.

Zetterberg

(2008). The downside of gender quotas? Institutional constraints on women in Mexican state legislatures. Parliamentary Affairs, 61(3), 442–460.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.61 MB