SEARCH

SEARCH BY CITATION

Keywords:

  • Recruitment;
  • Beauty;
  • Likability;
  • Politicians;
  • Gender

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. MPs’ Extra-Parliamentary Activities
  5. 3. MPs’ Physical Appearance
  6. 4. Empirical Results
  7. 5. Conclusion
  8. References

Politicians have been shown to benefit electorally from an attractive physical appearance. Employing data on 614 German MPs, this note explores whether it also affects their success/failure in the market for extra-parliamentary activities. An attractive physical appearance is found to mainly benefit female MPs, especially for private-sector jobs. This is particularly driven by MPs’ perceived likability. While MP's perceived beauty is shown to have no direct effects for extra-parliamentary activities, our findings suggest important indirect effects.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. MPs’ Extra-Parliamentary Activities
  5. 3. MPs’ Physical Appearance
  6. 4. Empirical Results
  7. 5. Conclusion
  8. References

During the most recent UK elections of May 2010, significant controversy erupted when several Tory candidates – among which David Cameron (now Prime Minister) and Caroline Dinenage (now a Member of Parliament) – were alleged to employ digitally enhanced images of themselves on campaign posters. Part of the controversy derives from the fact that such alterations can manipulate voter preferences. For instance, Rosenberg and McCafferty (1987) – studying the effect of different photographs used by the same candidate on voter preferences in an actual election – and Rosenberg et al. (1986) – studying the effect of strategically manipulated photographs of 200 women in mock elections (making use of a make-up artist for the manipulation of women's image) – illustrate that ‘it is possible to shape a political candidate's image in a way which may affect electoral outcomes’ (Rosenberg et al., 1986, p. 345). Moreover, newspaper editors have long known – and exploited – the value of (un)favourable photographs to support (undermine) the electoral campaigns of candidates they endorse (oppose) (e.g., Barrett and Barrington, 2005). Still, besides the normative aspects involved, taking recourse to airbrushing clearly indicates that politicians believe in the electoral importance of physical attractiveness. Recent academic work suggests they are right to do so (e.g., Banducci et al., 2008; Berggren et al., 2010).

For politicians, however, the influence of physical appearance may extend beyond Election Day as MPs are often permitted to carry out jobs in addition to their political mandate (often referred to as moonlighting, Geys and Mause, 2013). Although such sideline jobs are welcomed in many traditions of representation as a source of additional knowledge and experience for MPs, they are often equated with personal greed and conflicts of interest by the broader public (and the media); a view prominently reflected in the passionate public discussion in Germany in fall 2012. As physical attractiveness can play an important role in employment settings (Hamermesh, 2011), one may wonder whether the influence of politicians’ physical (un)attractiveness persists into the market for extra-parliamentary jobs. We evaluate this proposition using a dataset of 197 female and 417 male German MPs’ physical appearance and extra-parliamentary activities over the period 10/2005–09/2007.

Besides evaluating whether physical attractiveness benefits politicians not just on, but also after Election Day, this question is interesting for two related reasons. First, while earlier research on the benefits of physical attractiveness in employment settings predominantly concentrates on private-sector jobs (Hamermesh, 2011), our sample allows extending the focus to also include public-sector activities as politicians’ extra-parliamentary activities cover both private- and public-sector jobs. Second, MPs’ extra-parliamentary jobs often concern the so-called ‘elite’ professions (e.g., board memberships, management consultants). It has been argued that beauty might in such a setting be detrimental for female individuals – a ‘beauty is beastly’ effect (Heilman and Saruwatari, 1979; Heilman and Stopeck, 1985; Ruffle and Shtudiner, 2012). The reason is that beauty stresses gender-related perceptions of individuals (e.g., Gillen, 1981; Heilman and Stopeck, 1985), which tends to work against women in masculinely sex-typed jobs with an institutionalized male bias (due to ‘Think Manager, Think Male’ stereotypes; Schein, 1973; Dodge et al., 1995). Previous empirical work related to this idea relies, however, largely on laboratory experiments with student subjects (see, however, Ruffle and Shtudiner, 2012).

Our main findings can be summarized as follows. First, we find that physical appearance shows a positive relation to MPs’ extra-parliamentary activities, especially for female MPs and particularly for private-sector jobs. Second, these results are mainly driven by MPs’ perceived likability, rather than perceived beauty. That is, the former has strong direct and, via its association with female MPs’ overall political success, indirect effects on politicians’ sideline activities, whereas MP's perceived beauty only has an indirect effect (via female politicians’ improved re-election odds and prolonged tenure) but no direct effect.

2. MPs’ Extra-Parliamentary Activities

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. MPs’ Extra-Parliamentary Activities
  5. 3. MPs’ Physical Appearance
  6. 4. Empirical Results
  7. 5. Conclusion
  8. References

Members of the German Bundestag are legally allowed to carry out professional activities in addition to their political mandate, but have to disclose activities falling into the following categories: (1) ‘paid activities in addition to the mandate’ (e.g., management consultant); (2) ‘member of the management, supervisory, administrative, advisory or other board in a private enterprise’; (3) similar activities ‘in local authorities or public corporations’; (4) similar activities ‘in clubs, associations and foundations not solely of local significance’; (5) ‘agreements on future activities or pecuniary advantages’; and (6) ‘investments in business companies’ (if MP has a voting share of more than 25%). Information on such activities for all 614 Bundestag members (197 women and 417 men) in the period 10/2005 to 09/2007 is extracted from the 2007 Official Handbook of the German Bundestag.1 Although self-reported, fines for misreporting equal to maximally half the MPs’ annual allowance imply that the data provided are likely to be complete and correct. Consequently, it has been employed in various studies (e.g., Becker et al., 2009; Mause, 2009; Niessen and Ruenzi, 2010). However, none of these links the data to (perceptions of) politicians’ physical appearance.

Table 1 presents the number of MPs with a given number of sideline jobs. The overall number of ancillary activities for each MP (‘all jobs’) is counted as the sum of activities in categories (1) through (4) and (6). Private-sector jobs are defined as activities in categories (2) and (6), plus category (1) for activities not performed in the political system (e.g., state secretary, minister). Sideline activities in the public-sector count ancillary activities in categories (1) and (3) – again excluding activities in the political system – plus category (4). We excluded political jobs as these mostly concern elected offices and cannot be interpreted as deriving from a labour market recruitment process. While Table 1 substantiates that most politicians do maintain extra-parliamentary activities (i.e., only 74 of the 614 German MPs in our sample have no such jobs), it also establishes that MPs are in general more likely to have public-sector (81% of MPs in our sample), rather than private-sector (51% in our sample), outside jobs. The median value lies at three outside jobs.

Table 1. Summary statistics politicians’ sideline jobs
 0 jobs1 job2 jobs3 jobs4 jobs5 jobs6 jobs7 jobs>7 jobs
Notes
  1. Entries represent the number of politicians reporting a given number of sideline activities. We report separate results for all sideline activities (‘all jobs’) and those in the public and private sectors.

All jobs7476101905856383487
Private sector304154663720126312
Public sector117105114816542301842

Note that we employ the number of outside activities as an indicator for MPs’ labour market success. While MPs’ remuneration might be considered as an additional measure and German MPs disclose some information on income received from outside activities (Geys and Mause, 2012; Peichl et al., 2013), we do not exploit this information here. The reason is that the income data are only made public in three income bands (€1000–3500, €3500–7000, and above €7000), and politicians themselves can indicate whether income is received annually or monthly. As this allows politicians to ‘game the system’ and hide substantial revenue streams, the income data lack reliability (Geys and Mause, 2012, p. 267; a similar view on analogous US data is also provided in Rosenson, 2007). Even so, it is important to highlight that the number of outside jobs employed here is positively correlated with a (admittedly crude) measure of earnings constructed from the available income information by assuming that the MP earns the lower end of the income band for each job reported (for further details, see Mause, 2009). Hence, the number of outside jobs appears to be a reasonable measure of success on the market for sideline jobs.

3. MPs’ Physical Appearance

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. MPs’ Extra-Parliamentary Activities
  5. 3. MPs’ Physical Appearance
  6. 4. Empirical Results
  7. 5. Conclusion
  8. References

To employ perceptions of politicians’ physical appearance as an explanatory factor, they should contain a common understanding of what it means to be, say, ‘beautiful’. Fortunately, various studies have shown this to be the case both within and across cultures (e.g., Berggren et al., 2010; Langlois et al., 2000). On this basis, we conducted a web-based survey asking individuals to evaluate photographs of our sample of 614 politicians (photos taken from the Official Handbook of the German Bundestag). These photographs have a standardized format (i.e. black-and-white headshots), which prevents variation in the characteristics of the pictures from affecting reported perceptions (Berggren et al., 2010). Consistent with previous work, we asked: ‘Based on the picture provided, what do you think of this person – compared to people living in your country – in terms of […]’, and repeated this question for five traits: that is ‘physical appearance or attractiveness’, ‘competence’, ‘likability (i.e. how nice, pleasant and agreeable you find this person)’, ‘trustworthiness’ (i.e. how ethical, honest and responsible you find this person) and ‘intelligence’. Respondents replied on a five-point scale from ‘very positive’ (5) to ‘very negative’ (1). No information beside the picture was provided. Although perceptions need not reflect politicians’ ‘true’ characteristics, they drive people's decisions and behaviour, which warrants their use here (Hamermesh, 2011).

In total, 4817 evaluations were obtained from 15 female and 13 male respondents between 20 and 60 years old (only 12 were students) with an average of 7.5 evaluations per picture. To minimize recognition bias, we used respondents in Belgium, France, Slovakia, Hungary, United States and United Kingdom (as well as one German respondent), and removed the observation when respondents recognized the person in the picture.2 The final number of evaluations per picture ranges from 3 to 24. This number lies close to the number of evaluations employed in Berggren et al. (2010) – who delete observations with fewer than three evaluations, and have on average nine evaluations per picture – and Hamermesh (2004) – who relies on four evaluations per picture. Clearly, one might worry that three evaluations is insufficient. To check this, we assess whether the coefficient of variation is higher for pictures with fewer evaluations (which would reflect more ‘noise’ surrounding the estimated evaluation). Comparing pictures obtaining less than 10 evaluations with those obtaining between 10 and 19 evaluations, we find no such effect. Hence, even relatively few evaluations appear to give a fairly precise estimate.

Overall, Spearman rank-order correlation tests show high interrater agreement. The test statistics range from ρ = 0.237 to 0.847 with associated p-values ranging from 0.12 to 0.00 (with one exception where ρ = 0.054; p < 0.10). Significance levels under the 90% confidence level thereby occur only when raters have less than 60 pictures in common. Even so, Berggren et al. (2010) show that their findings are robust to whether individuals evaluate over 500 pictures or only 4, which suggests that the properties of the rating distribution are not overly sensitive to the number of evaluations. Following the ‘truth of consensus’ method (Banducci et al., 2008), we calculate the average of the independent evaluations across raters for a given politician on each of the five characteristics.3 Summary statistics across all 614 politicians – as well as the subsamples of men and women – are given in Table 2. This illustrates that women are on average perceived to be more beautiful (2.88 vs. 2.54; p < 0.01), more likable (3.08 vs. 2.71; p < 0.01) and more trustworthy (3.08 vs. 2.64; p < 0.01) than men. No significant difference exists for perceived competence and intelligence, indicating no bias against female politicians in terms of intellectual and professional capacities.

Table 2. Summary statistics perceptions of politicians’ perceived traits
 BeautyCompetenceLikabilityTrustworthinessIntelligence
Notes
  1. Entries represent mean values, with standard deviation between brackets (5 is ‘best’ score and 1 is ‘worst’ score). ‘Men vs. Women’ gives t-value of difference-in-means t-test allowing for unequal variance across samples. ***significant at 1%; ** at 5% and * at 10%.

Overall

(N = 614)

2.646

(0.542)

2.944

(0.370)

2.825

(0.467)

2.779

(0.443)

3.025

(0.382)

Men

(= 417)

2.537

(0.496)

2.933

(0.364)

2.707

(0.436)

2.636

(0.398)

3.037

(0.373)

Women

(N = 197)

2.877

(0.565)

2.966

(0.383)

3.076

(0.432)

3.083

(0.375)

2.999

(0.398)

Men vs. Women7.226***1.0429.841***13.521***−1.139

4. Empirical Results

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. MPs’ Extra-Parliamentary Activities
  5. 3. MPs’ Physical Appearance
  6. 4. Empirical Results
  7. 5. Conclusion
  8. References

4.1. Physical appearance and extra-parliamentary jobs

To evaluate the relation between physical attractiveness and politicians’ extra-parliamentary jobs, we estimate the following regression model:

  • display math(1)

where Yi represents either MPi's total number of sideline activities (‘all jobs’) or his/her sideline activities in the private (‘Private-sector’) or public (‘Public-sector’) sector. Equation (1) is estimated separately for each of these three dependent variables. PhysAttrij is measured as the average of the independent evaluations for MPi across raters for three aspects of his/her (perceived) physical appearance (i.e. Beauty, Likability, Competence).4 To evaluate whether these effects of physical appearance differ across gender, we include Femalei (an indicator variable equal to 1 for female MPs, 0 for males) and an interaction of PhysAttrij with Femalei. To minimize missing-variable bias, Xi is a vector of control variables including MPi's party affiliation (dummies for SPD, FDP, GRÜNE and LINKE; CDU/CSU is reference category), home federal state (dummy for former East German Länder), type of mandate (i.e., party list vs. direct mandate), number of legislative periods attended, age, marital status (dummy: 1 if married), number of children, educational background (dummy: 1 if university education), profession (dummies for legal, economic and teacher background), religious affiliation and whether MPi holds an important political office (e.g. leader of parliamentary fraction or Bundestag (vice-)president). As the dependent variables only take non-negative integer values and have highly skewed distributions, we employ a negative binomial count model to estimate equation (1) (tests of overdispersion suggest that Poisson models are inappropriate).

The key findings are brought together in Table 3.5 Column (1) shows that, all else equal, MPs’ (perceived) beauty bears no significant relation to the total extent of their sideline activities. Indeed, for both men (inline image; p < 0.10) and women (inline image; p < 0.10),6 the coefficient estimates remain well below standard levels of statistical significance – thus, invalidating the presence of any ‘beauty is beastly’ effect in this setting. Separating private-sector (Column (2)) from public-sector (Column (3)) jobs provides similar insignificant results. The same also holds for MPs’ perceived competence, which returns statistically insignificant effects throughout Table 3 for both men and women. The measure of perceived likability, however, shows a significant positive relation to an MPs’ total number of outside jobs. The effect size is substantively meaningful: that is, evaluated at the mean of all variables included in the model, a 1 SD increase in a male politician's likability evaluation (0.436; see Table 2) increases the number of outside activities by approximately one tenth of a SD (or 0.388 jobs). For women, a 1 SD increase in perceived likability (0.432; see Table 2) induces an increase in extra-parliamentary jobs by almost one-third of a SD (or 1.036 jobs).

Table 3. Physical appearance and extra-parliamentary activities
VariableAll jobs (1)Private sector (2)Public sector (3)
Notes
  1. N = 614; t-values based on robust standard errors between brackets: ***significant at 1%; ** at 5% and * at 10%. Wald test indicates joint significance of all regressors (with R equal to the number of regressors minus one). Intercept and full set of controls are always included. Note that the variables for beauty, competence and likability are the mean of evaluations on a five-point scale (see text for details). Given the nature of the dependent variables, all estimations rely on negative binomial count models.

Beauty

(1 = plain; 5 = beautiful)

0.042

(0.39)

−0.021

(−0.10)

0.110

(0.94)

Competence

(1 = inept; 5 = competent)

−0.252*

(−1.74)

−0.268

(−1.25)

−0.262

(−1.59)

Likability

(1 = likable; 5 = likable)

0.239**

(2.05)

0.203

(1.11)

0.240*

(1.86)

Female × Beauty

−0.116

(−0.71)

0.258

(0.79)

−0.256

(−1.54)

Female × Competence

0.300

(1.32)

0.480

(1.18)

0.270

(1.11)

Female × Likability

0.405**

(2.07)

0.704*

(1.83)

0.347

(1.64)

Log pseudolikelihood−1443.79−826.28−1279.58
Wald χ² (R)163.46***189.77***128.13***

For public-sector jobs, this likability benefit holds for both women and men (although it is substantively stronger for the former), while for private-sector jobs, it applies only to women (inline image; χ² = 7.08; p < 0.01). Particularly, once again evaluated at the mean of all variables, a 1 SD increase in a politician's perceived likability increases the number of public-sector outside activities by almost one tenth of a SD (or 0.276 jobs) for men, and just under one quarter of a SD (or 0.670 jobs) for women. The same increase in likability gains women just over one-third of a SD or 0.334 private-sector jobs. One possible explanation of this difference between the public and private sector is that recruiting for public-sector jobs is usually not on a personal basis, but in many cases at least in part reflects partisan attachments (making that physical appearance matters less). It is important to note here that we also experimented with measures for a more complete set of perceived personality traits (i.e. beauty, competence, likability, intelligence and trustworthiness). This does not affect the main findings reported above. That is, in all cases, we find that i) physical appearance shows a positive relation to MPs’ extra-parliamentary activities and ii) likability is the most important factor among the different dimensions of physical appearance included here (details upon request).

The discovery of a significant likability effect supports earlier findings illustrating that ‘teachers, employees, defendants in court cases, salespeople, prospective employees, and even cocktail waitresses (…) reap the benefits of likability’ (Jayanti and Whipple, 2008; see also Casciaro and Sousa Lobo, 2005). In our setting, this effect might arise because likability may be particularly desirable in the type of jobs politicians entertain as extra-parliamentary activities, which are mostly of a representational nature. Nonetheless, one might wonder why perceived beauty – which has been frequently shown to play an important role in employment settings (Hamermesh, 2011) – is found to display no significant direct relation to MPs’ extra-parliamentary activities in Columns (1) through (3). From this perspective, it should be noted that although beauty has no significant direct effect, it is likely to have at least some indirect effect. The reason is that beauty correlates with likability, such that it may indirectly affect the dependent variable through its effect on likability.

More generally, it is at least conceivable that all attributes reflecting MPs’ perceived physical attractiveness employed here (e.g. beauty and likability) have additional indirect effects on politicians’ extra-parliamentary jobs as they increase the probability that politicians become (re-)elected (Banducci et al., 2008; Berggren et al., 2010), and may affect their political success more generally (e.g., by affecting the likelihood of obtaining leadership positions). Such political success is picked up by several control variables in vector Xi (e.g. their tenure in office or the indicator variables for important political offices), which often do display a significant association with sideline jobs (see, e.g. Becker et al., 2009; or Mause, 2009). Hence, one might argue that the analysis above gives only a partial view of physical appearance's relation to extra-parliamentary jobs because it does not explicitly illustrate the existence and/or strength of such indirect effects. We therefore turn to this more explicitly in the next section.

4.2. Physical appearance and political success7

To evaluate the indirect effects of MPs’ physical appearance on sideline jobs in more detail, we look at the relation between physical attractiveness and politicians’ political success using the following regression model (with subscript i referring to MPs):

  • display math(2)

where Successi represents either MPi's number of legislative periods (‘Tenure’, as a measure of re-election successes), his/her number of memberships in parliamentary committees (‘Committees’) or an indicator variable equal to 1 if the MP is minister, secretary of state, part of the leadership of the party's parliamentary group, (deputy) committee leader or Bundestag (vice-)president (‘Important office’). While PhysAttrij and Femalei are defined as above, the vector of control variables Ci now includes MPi's party affiliation, home federal state, type of mandate, number of legislative periods attended, age, marital status, number of children, educational background, profession and religious affiliation. The key findings are brought together in Table 4. Note that given the nature of the dependent variables, Columns (4) and (5) are estimated using a Poisson regression, whereas Column (6) is based on a logit model.

Table 4. Physical appearance and political success
VariableTenure (4)Committees (5)Important office (6)
Notes
  1. N = 614; t-values based on robust standard errors between brackets: ***significant at 1%; ** at 5% and * at 10%. Wald test indicates joint significance of all regressors (with R equal to the number of regressors minus one). Intercept and full set of sociodemographic background controls are always included. Note that the variables for beauty, competence and likability are the mean of evaluations on a five-point scale (see text for details). Given the nature of the dependent variables, Columns (4) and (5) are estimated using a Poisson regression, whereas Column (6) is based on a logit model.

Beauty

(1 = plain; 5 = beautiful)

0.018

(0.25)

0.041

(0.47)

−0.030

(−0.09)

Competence

(1 = inept; 5 = competent)

0.011

(0.12)

−0.049

(−0.47)

0.272

(0.65)

Likability

(1 = likable; 5 = likable)

0.013

(0.18)

0.042

(0.44)

0.418

(1.30)

Female × Beauty

0.143

(1.53)

−0.206*

(−1.66)

0.538

(1.06)

Female × Competence

−0.155

(−1.18)

−0.123

(−0.66)

0.477

(0.67)

Female × Likability

0.081

(0.68)

0.237

(1.54)

0.473

(0.86)

Log pseudolikelihood−1064.91−923.59−334.64
Wald χ² (R)418.10***36.44**39.66**

Table 4 first of all illustrates that, all else equal, perceived physical appearance shows no significant relation to male MPs’ tenure in office (Column (4)), their committee memberships (Column (5)) or the probability of obtaining a leadership position in German politics (Column (6)). To evaluate the results for female politicians, we – as before – have to add the baseline effects to those of the interaction terms, and recalculate the standard error around this point estimate taking into account the covariance of the coefficients of the baseline and interaction effects. This shows that, for female MPs’, perceived beauty is significantly positively related to higher tenure (inline image; p = 0.02), which indicates a higher re-election probability for physically attractive female MPs (consistent with, for instance, Banducci et al., 2008; Berggren et al., 2010). Perceived beauty displays a much weaker association with holding an important political office (inline image; p = 0.21), and even a negative connection to committee memberships (inline image; p = 0.09).

Interestingly, much of the reverse pattern is observed for perceived likability. This plays no significant role for female MPs’ tenure (inline image inline image; p = 0.31), but is significantly positively related to committee memberships (inline image; p = 0.02) and holding an important political office (inline image; p = 0.05). No significant effects are observed for perceived competence on all three measures of political success. Taken together, these results suggest that (perceived) physical beauty appears to provide a benefit to (particularly female) politicians in the electoral arena, but that (perceived) likability is a more valuable asset once the elections are over. This configuration may well reflect the varying preferences of the different decision makers at both time points of the democratic process: that is, voters in elections and other politicians after the elections.

As tenure and holding important political offices positively affect MPs’ outside jobs (β = 0.044; p < 0.05 for tenure, and β = 0.164; p < 0.10 for holding important political offices),8 the above results suggest an indirect effect from female MPs’ perceived beauty and likability on their extra-parliamentary activities. Still, calculating the strength of these indirect effects, we find that the gains associated with MPs’ physical appearance in terms of their political success translate into at best marginal (indirect) benefits in terms of additional outside jobs. For instance, evaluated at the mean of all variables, a 1 SD increase in a female politician's perceived beauty increases her tenure with 0.249 terms, which, in turn, translates into an increase in her outside activities by 0.044 jobs. Similarly, a 1 SD increase in perceived likability increases a female MPs’ probability of holding an important political office with 7.17 percent, which, in turn, translates into an increase in her outside activities by 0.039 jobs.

Before we conclude, it appears that in general the effect of likability is substantially stronger for female politicians, which, given women's higher average perceived likability, might put them at a competitive advantage. One potential explanation of the gendered nature of the likability effect is that personality characteristics generally associated with likability are mostly conceived of as feminine (Diekman and Eagly, 2000). This might make the male/likeable combination more conspicuous as it does not conform to expectations (Judge et al., 2012). As such gender-role incongruence (Nieva and Gutek, 1981) has been found to elicit penalties including social rejection, negative evaluations and lower income levels (Amanatullah and Tinsley, 2013; Costrich et al., 1975); it may explain the weaker likability premium among male politicians.

5. Conclusion

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. MPs’ Extra-Parliamentary Activities
  5. 3. MPs’ Physical Appearance
  6. 4. Empirical Results
  7. 5. Conclusion
  8. References

Recent studies indicate that physically attractive politicians generally obtain more votes (Banducci et al., 2008; Berggren et al., 2010). This research note extended this literature by arguing that physical attractiveness might well benefit politicians not just on, but also after Election Day. Using data on German MPs’ extra-parliamentary positions, we indeed illustrate that (perceived) likability benefits (especially female) MPs for both private- and public-sector jobs. The likability effects thus uncovered are both statistically and substantively meaningful. For MPs’ perceived physical attractiveness, we find that it may well have an indirect effect on politicians’ extra-parliamentary activities, but a direct effect of beauty cannot be substantiated.

These findings raise a number of important additional questions. First, do our results carry over to general employment settings? That is, is the fact that we analyse politicians – which are ‘preselected’ via elections in which their personality traits are an important factor for success – important for our results or would they likewise hold in traditional recruitment settings (where candidates are not similarly preselected)? Second, our findings suggest an unequal role for likability and competence, which immediately leads to the question whether we desire a ‘lovable fool’ or ‘competent jerk’ to work with (assuming a ‘lovable star’ is not available; labels taken from Casciaro and Sousa Lobo, 2005). This has, surprisingly, received little attention in either the labour economics or the occupational psychology literature thus far (see, however, Singh and Tor, 2008) and presents an interesting avenue for future (experimental) research.

Footnote
  1. 1

    The starting point of the data reflects a change in legislation, and implies we cannot observe whether jobs were held already prior to that date. We exclude category (5) below as we are interested in MPs’ current sideline activities. Note also that repeated activities (e.g., a series of public lectures) are treated as one activity.

  2. 2

    Given the international composition of our respondent sample, recognition of politicians was only a minor issue, and did not require the exclusion of any politician due a lack of valid (i.e. unbiased by personal recognition) evaluations.

  3. 3

    This implicitly assumes that all raters have the same perception of average beauty, competence, likability, and of the variation around that mean. Clearly, this need not be the case. Hence, we also calculated standardized scores by subtracting the mean rating of each rater from every individual rating (s)he makes, and normalizing the resulting score by dividing it with this rater's standard deviation (see also Belot et al., 2012; Johnson et al., 2010). The standardized and unstandardized measures are highly correlated (r = 0.97; p < 0.001) for all five personal traits – and using either provides similar results in the analysis below (details upon request).

  4. 4

    We did not include all five traits into the model due to the strong positive correlation between them, and the significant multicollinearity problems this induces.

  5. 5

    As results for the control variables follow those reported in the foregoing literature, we suppress these to preserve space (details upon request). See, for example, Becker et al. (2009) or Mause (2009).

  6. 6

    The estimated significance level for women is calculated taking into account that the standard error around the coefficient estimate also depends on the covariance of inline image and inline image (Brambor et al., 2006).

  7. 7

    We are grateful to an anonymous referee for pointing this out, and for excellent suggestions on how to approach this in more detail.

  8. 8

    Note that no significant relation is observed between committee memberships and MPs’ outside jobs.

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. MPs’ Extra-Parliamentary Activities
  5. 3. MPs’ Physical Appearance
  6. 4. Empirical Results
  7. 5. Conclusion
  8. References
  • Amanatullah, E. T. and C. H. Tinsley (2013), ‘Punishing Female Negotiators For Asserting Too Much…Or Not Enough: Exploring Why Advocacy Moderates Backlash Against Assertive Female Negotiators’, Organizational Behavior and Human Decision Processes 120, 110122.
  • Banducci, S. A., M. Thrasher, C. Rallings and J. A. Karp (2008), ‘Ballot Photographs as Cues in Low-Information Elections’, Political Psychology 29, 903917.
  • Barrett, A. W. and L. W. Barrington (2005), ‘Bias in Newspaper Photograph Selection’, Political Research Quarterly 58, 609618.
  • Becker, J., A. Peichl and J. Rincke (2009), ‘Politicians’ Outside Earnings and Electoral Competition’, Public Choice 140, 379394.
  • Belot, M., V. Bhaskar and J. van de Ven (2012), ‘Beauty and the Sources of Discrimination’, Journal of Human Resources 47, 851872.
  • Berggren, N., H. Jordahl and P. Poutvaara (2010), ‘The Looks of a Winner: Beauty and Electoral Success’, Journal of Public Economics 94, 815.
  • Brambor, T., W. R. Clark and M. Golder (2006), ‘Understanding Interaction Models: Improving Empirical Analyses’, Political Analysis 14, 6382.
  • Casciaro, T. and M. Sousa Lobo (2005), ‘Competent Jerks, Lovable Fools and the Formation of Social Networks’, Harvard Business Review 86, 9299.
  • Costrich, N., J. Feinstein, L. Kidder, J. Marecek and L. Pascale (1975), ‘When Stereotypes Hurt: Three Studies Of Penalties For Sex Role Reversals’, Journal of Experimental Social Psychology 11, 520530.
  • Diekman, A. B. and A. H. Eagly (2000), ‘Stereotypes as Dynamic Constructs: Women and Men of the Past, Present, and Future’, Personality and Social Psychology 26, 11711188.
  • Dodge, K. A., F. D. Gilroy and L. M. Fenzel (1995), ‘Requisite Management Characteristics Revisited: Two Decades Later’, Journal of Social Behavior and Personality 10, 253264.
  • Geys, B. and K. Mause (2012), ‘Delegation, Accountability & Legislator Moonlighting: Agency Problems in Germany’, German Politics 21, 255273.
  • Geys, B. and K. Mause (2013), ‘Moonlighting Politicians: A Survey and Research Agenda’, Journal of Legislative Studies 19, 7697.
  • Gillen, B. (1981), ‘Physical Attractiveness: A Determinant of Two Types of Goodness’, Personality and Social Psychology Bulletin 7, 277281.
  • Hamermesh, D. S. (2004), ‘Changing Looks and Changing ‘Discrimination’: The Beauty of Economists’, Economics Letters 93, 405412.
  • Hamermesh, D. S. (2011), Beauty Pays: Why Attractive People Are More Successful, Princeton University Press, Princeton.
  • Heilman, M. E. and L. R. Saruwatari (1979), ‘When Beauty is Beastly: The Effects of Appearance and Sex on Evaluations of Job Applicants for Managerial and Non-Managerial Jobs’, Organizational Behavior and Human Performance 23, 360372.
  • Heilman, M. E. and M. H. Stopeck (1985), ‘Being Attractive, Advantage or Disadvantage? Performance-Based Evaluations and Recommended Personnel Actions as a Function of Appearance, Sex and Job Type’, Organizational Behavior and Human Performance 35, 202215.
  • Jayanti, R. K. and T. W. Whipple (2008), ‘Like Me, Like Me Not: The Role of Physician Likability on Service Evaluations’, Journal of Marketing Theory and Practice 16, 7986.
  • Johnson, S. K., K. E. Podratz, R. L. Dibboye and E. Gibbons (2010), ‘Physical Attractiveness Biases in Ratings of Employment Suitability: Tracking Down the ‘Beauty is Beastly’ Effect’, Journal of Social Psychology 150, 301318.
  • Judge, T. A., B. A. Livingston and C. Hurst (2012), ‘Do Nice Guys – and Gals – Really Finish Last? The Joint Effects of Sex and Agreeableness on Income’, Journal of Personality and Social Psychology 102, 390407.
  • Langlois, J. H., L. Kalamakis, A. J. Rubinstein, A. Larson, M. Hallam and M. Smoot (2000), ‘Maxims or Myths of Beauty? A Meta-Analytic and Theoretical Review’, Psychological Bulletin 126, 390423.
  • Mause, K. (2009), ‘Nebentätige Bundestagsabgeordnete: Was offenbaren die Veröffentlichungspflichten?’, Perspektiven der Wirtschaftspolitik 10, 146174.
  • Niessen, A. and S. Ruenzi (2010), ‘Political Connectedness and Firm Performance: Evidence from Germany’, German Economic Review 11, 441464.
  • Nieva, V. and B. Gutek (1981), Women and Work, Praeger, New York.
  • Peichl, A., N. Pestel and S. Siegloch (2013), ‘The Politicians’ Wage Gap: Insights from German Members of Parliament’, Public Choice 156, 653676.
  • Rosenberg, S. W. and P. McCafferty (1987), ‘The Image and the Vote Manipulating Voters’ Preferences’, Public Opinion Quarterly 51, 3147.
  • Rosenberg, S. W., C. Bohan, P. McCafferty and K. Harris (1986), ‘The Image and the Vote: The Effect of Candidate Presentation on Voter Preference’, American Journal of Political Science 30, 108127.
  • Rosenson, B. A. (2007), ‘Explaining Legislators’ Positions on Outside Income Limits: Voting on Honoraria Ceilings in the U.S. Senate, 1981–83’, Public Choice 133, 111128.
  • Ruffle, B. J. and Z. Shtudiner (2012), Are Good-Looking People More Employable? Ben-Gurion University, Beer-Sheva.
  • Schein, V. E. (1973), ‘The Relationship between Sex Role Stereotypes and Requisite Management Characteristics’, Journal of Applied Psychology 57, 95100.
  • Singh, R. and X. L. Tor (2008), ‘The Relative Effects of Competence and Likability on Interpersonal Attraction’, Journal of Social Psychology 148, 253255.