This article was downloaded by: [Washington State University Libraries ] On: 05 November 2014, At: 23:52 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T3JH, UK German Politics Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/1oi/ fgrp20 Magic Mayors: Predicting Electoral Success from Candidates' Physical Attractiveness under the Conditions of a Presidential Electoral System Ulrich Rosar , Markus Klein & Tilo Beckers Published online: 20 Nov 2012. To cite thisarticle: Ulrich Rosar , Markus Klein & Tilo Beckers (2012) Magic Mayors: Predicting Electoral Success from Candidates' Physical Attractiveness under the Conditions of a Presidential Electoral System, German Politics, 21:4, 372-391, DOI: 10.1080/ 09644008.2012.739609 To link to thisarticle: http://dx.doi.org/ 10.1080/09644008.2012.739609 PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the "Content") contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sublicensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions Magic Mayors: Predicting Electoral Success from Candidates' Physical Attractiveness under the Conditions of a Presidential Electoral System ULRICH ROSAR, MARKUS KLEIN and TILO BECKERS A range of empirical studies has shown that candidates' physical attractiveness can substantially influence the outcome of political elections. This applies to different countries, different electoral systems, and different levels of political systems, and equally affects simple direct or list candidates and front-runners. However, no previous investigation using actual election results has been made into whether candidates' attractiveness also has an effect under the conditions of a presidential electoral system. Theoretical reasons can be formulated that suggest attractiveness is ineffective under these circumstances. In order to clarify this point empirically, we analysed the 2009 North Rhine-Westphalia mayoral elections. Yet the results of the analyses clearly show that candidates' attractiveness has a substantial influence. Taking into account earlier findings, the influence of physical attractiveness in political elections appears to be resistant, to a large degree, to varying constraints. INTRODUCTION: PHYSICAL ATTRACTIVENESS, ELECTORAL SUCCESS, AND THE INFLUENCE OF A PRESIDENTIAL ELECTORAL SYSTEM The general thesis of the personalisation of politics and its specification for the field of voting behaviour has been a prominent topic of empirical voting research.1 Proponents of the thesis claim that the relevance of politicians in politic media coverage as well as in individual decision-making rationales is on the rise. Furthermore, non-political aspects of politicians are said to become more important. The claimed changes, that is, tendencies towards more personalisation over time, could not be validated in empirical studies. Nevertheless, in the course of this line of research some aspects of voting behaviour have received attention that were unnoticed beforehand. The investigation of non-political characteristics of political candidates brought about a rising number of empirical studies focusing on physical attractiveness. The results of these studies may be summarised in a simple way: Physical attractiveness matters - even in political elections. This was not only verified by multiple experimental studies, there is also a number of investigations that can prove on the basis of actual election results that a candidate's physical attractiveness may substantially influence electoral success. In addition to studies for Germany, Finland, France, Ireland, Switzerland and the UK4 there are also studies for Australia and Canada.5 Furthermore, the effect of candidate attractiveness has been studied in relation to different levels of political systems. Besides German Politics, Vol.21, No.4, December 2012, pp.372-391 ISSN 0964-4008 print/1743-8993 online http://dx.doi.org/10.1080/09644008.2012.739609 © 2012 Association for the Study of German Politics MAGIC MAYORS 373 f\ 7 X national parliamentary elections studies have also been carried out on local, regional and supranational elections.9 The conditions of individual electoral systems have thus varied accordingly. The spectrum of systems covered ranges from simple majority voting through the single transferable vote and panachage to outright proportional representation. If one compares the empirical findings of these various studies, it appears that the influence of physical attractiveness is astonishingly robust, regardless of the institutional and politico-cultural context. Moreover one gets the impression that it hardly makes a difference whether the analysis focuses on constituency candidates, candidates from lists or front-runners.10 The overall impression gained is that candidates' attractiveness has a ubiquitous impact on electoral success. If this impression were to prove correct, it would also mean that there is relatively little opportunity to eliminate the influence of physical attractiveness which is doubtless undesirable from the point of view of normative democratic theory. Nevertheless, it must be considered that - despite the complexity of previous studies - a conclusive case has not yet been empirically investigated: no study has been conducted that uses actual election results while also investigating the influence of physical attractiveness under the conditions of mayoral elections. This type of election is an object of inquiry in its own right, that is, it is a highly relevant field for political election research: it is both quantitatively widespread as well as important from the substantive perspective of political participation due to the direct local and social proximity of the candidates and the electorate. Apart from this, the mayoral elections we study are a presidential electoral race and thus a system that is also important because of the possible transfer to national elections. This study is path-breaking because no other study has analysed under real-life conditions whether physical attractiveness still has an effect when a central post in the political executive needs to be filled in accordance with the principle of majority voting. This deficiency is important once the role of non-political qualities in election campaigns such as mayoral races becomes a research topic (see the discourse on personalisation in politics and signalling theory, e.g. on labour markets).11 It is a significant research objective to study the aforementioned deficiency because the possibility exists, at least theoretically, that the candidates, under such conditions, are so prominent that they compete intensively on the basis of their suitability for the desired office. At the same time this would mean, however, that candidate traits which are apolitical, such as gender, ethnicity and also physical attractiveness, come less into play. If it were proved that the physical attractiveness of a candidate does not influence the election outcome under the conditions of a presidential electoral system, this would have two important consequences. First, it would allow the conclusion that, at least when it comes to voting for one individual who will receive a great deal of political power, voters choose more carefully than it has previously appeared. Second, it would identify an institutional adjusting mechanism that would enable - if so desired - the effect of candidate attractiveness to be suppressed, albeit selectively, by altering the electoral system. The present investigation seeks to make progress towards bridging this gap in research. This project is supported by data obtained in connection with the North Rhine-Westphalia mayoral elections which took place in 2009. The mayors in North-Rhine Westphalia have very strong political positions in their respective cities 374 GERMAN POLITICS or counties because they are both head of the administration and the local parliament. Since 1999, their position is further consolidated because they are directly elected by the people and thus democratically legitimised and empowered to act as strong representatives. Besides this, in 2009, their term of office has been raised from five to six years while the legislature of the local or regional parliament is only five years. Due to these changes, in 2009, the election of the mayor and the parliamentary election took place together for the last time, and now the mayoral election is becoming politically important in its own right.15 In the following we will outline the mechanisms by which the effect of the physical attractiveness of candidates in elections unfolds and explain how this effect could be modified in a presidential electoral system. This will be followed by a brief explanation of the decision to carry out our investigation using the example of the North Rhine-Westphalia mayoral elections and a description of these base data. Next, we will present the results of our empirical analyses. The paper will finish with a summary of the central findings and several brief conclusions. THEORETICAL BACKGROUND: HOW PHYSICAL ATTRACTIVENESS WORKS, AND WHY A PRESIDENTIAL ELECTORAL SYSTEM COULD SUPPRESS THE INFLUENCE OF PHYSICAL ATTRACTIVENESS The effects of people's physical attractiveness on their chances of success in various social contexts have already been extensively described elsewhere.16 The authors of the present paper have already carried out a basic adaptation of the aspects of attrac- 1 7 tiveness research that are relevant for empirical election research. Therefore, by way of summary, we will only outline here the major elements of the chain of effect that also enable the physical attractiveness of candidates to have an impact in elections. The main elements in this chain of effect are Attractiveness Consensus, Attractiveness Stereotype, Attractiveness Attention Boost, Attractiveness Glamour Effect and Attractiveness Treatment Advantage. Attractiveness Consensus forms the basis for the effectiveness of physical attractiveness. This term describes the view that human beauty is not so much in the eye of the beholder, but lies primarily in the features of the person being observed. A person's physical attractiveness is thus a personal feature that can be determined rela- 18 tively unambiguously. Therefore, there is a widespread interpersonal and intercul-tural consensus in how different observers rate the attractiveness of a given person.19 Attractiveness, similar to gender or ethnicity, can be considered an ascriptive human feature that is open to objective measurement. Building on the Attractiveness Consensus, a person's attractiveness can give rise to a range of well-documented effects: the Attractiveness Stereotype is in line with the 20 phrase 'What is beautiful is good', other people attribute attractive persons a priori with socially highly desirable personality traits. Attractive persons are therefore seen, for example, as being more assertive, determined, efficient, industrious, intelligent and competent; and also more honest, reliable, creative, empathetic and sympathetic. Furthermore, attractive persons attract more notice. They gain more attention, are noticed more frequently, observed more intensively and their statements and actions 22 are better memorised. This is the Attractiveness Attention Boost. MAGIC MAYORS 375 In actual interactions the Attractiveness Treatment Advantage means that attractive persons may be better treated by others than is the case for less good-looking contemporaries. They encounter more trust and respect and experience more help and 23 support. Even a public misdemeanour does not necessarily undo these advantages since the Attractiveness Glamour Effect means that the wrongdoing of attractive 24 persons is relativised in the perception of third parties. On the one hand, this can happen through third parties attributing the causes to factors for which the perpetrator is not responsible. On the other hand, it may be that the misdemeanour is simply tri-vialised. Consequently, the above-mentioned mechanisms lead to attractive persons 25 having a clear Attractiveness Competition Advantage over their unattractive rivals. This also applies to the world of politics. In conjunction with political elections, this results in the following simple causal mechanism: due to the Attractiveness Consensus, voters have very similar evaluations of the attractiveness of candidates standing for election. The Attractiveness Attention Boost entails greater, more frequent and more intense attention for attractive candidates and means that their standpoints and actions are better remembered. When forming a voting preference this would likely in turn lead to these candidates and their party simply being more 'present' in the considerations of the voter. This should give a significant initial advantage. Aside from this, the Attractiveness Stereotype results in attractive candidates being ascribed with superior political motivation and capability. This should give a second significant benefit. Due to the Attractiveness Treatment Advantage it may be that those voters who have not fully made up their minds on election day, tend (in the polling booth) to cast their vote ultimately for attractive candidates or their parties. This should give a third meaningful benefit. In terms of voting, it does not necessarily matter if an attractive candidate holds a different view on political issues: any contradictions can be resolved by the Attractiveness Glamour Effect in that their importance is underplayed or excuses are cited - for example that the candidate is forced to toe the party line. This should give a fourth significant advantage. As a result, the likelihood increases of attractive candidates or their parties getting more votes on polling day than less attractive rivals. Nevertheless, these mechanisms assume that voters - consciously or subconsciously - tend to use (also) irrelevant external characteristics such as physical attractiveness as heuristics when making their voting decision. One may accept that this is essentially the case for all voters or at least for a large proportion of the electorate. However, one can also suppose that the extent to which such heuristics are applied depends on the political function of the contested positions, the form of electoral law and the quantity and quality of the information available on the candidates. It could therefore make a difference whether the position being filled is just a member of parliament or a key leadership position at the head of the political executive; or whether it is a leadership position in which political decisions can be taken to a large extent independently of parliamentary influence and where accountability lies primarily with the electorate. It could also matter whether voters can only indirectly decide who undertakes the role - for example where votes only affect the composition of the parliament which itself in turn determines the leader of the executive - or whether they directly make the decision through a very simple election process. Of course, it ought also to be the case that voters find it easier to make their decision 376 GERMAN POLITICS based on candidates' relevant qualities and political viewpoints if extensive information is easily accessible. It can doubtless be seen as given that elections in the context of presidential systems involve deciding on whom to fill the key position in the political executive. It is unimportant whether a nation's president, a federal state governor or a municipal mayor is being elected. The candidates themselves are always at the centre of the voting decision. It ought also to be indisputable that under these conditions the parties tailor their election campaigns to the candidates and the media focus reporting on the candidates, their personalities and their political agendas. Voters would therefore have easy access to a maximum of substantial information about the rival candidates. However, clear differences would probably arise with regard to the actual form of electoral procedure. Empirically, this ranges from simple majority voting through French two-stage run-off elections to the US elector system. The more transparent, simple and direct the mode of voting, the clearer the connection that is created between voters' ballots and the election outcome and therefore the greater the incentive to engage with the candidates who are standing before the election. Ideally, therefore, elections conducted on the basis of a simple majority voting system should be used to investigate whether or not candidate attractiveness has a proven effect in a presidential electoral system. DATABASE: FORTY-NINE MAYORAL ELECTIONS WITH 264 CANDIDATES AT THE SAME POINT IN TIME, AND UNDER THE SAME CONDITIONS OF ELECTORAL LAW Admittedly it is not entirely simple to compile a sufficient number of elections with an adequate number of candidates without encountering serious problems elsewhere that can diminish the validity of the findings. An obvious strategy could be to investigate internationally comparable presidential elections at a national level within a particular time period that is as recent as possible. However, one then faces the problem of either accepting the disadvantage of widely varying electoral rules or dramatically reducing the number of elections that can be analysed. In addition, concentrating on presidential elections in one country over a prolonged time period does not represent a viable alternative. If the USA were used, for example, a nation with the longest continual presidential tradition with fifty-six presidential elections and more than double the number of relevant candidates, the analysis would have to encompass a period of around 220 years. This would involve statistically hard-to-control periodic effects resulting from territorial expansion, civil war and successive broadening of the right to vote. And a problem would be caused not least by the fact that, because of the Electoral College system, the winner of popular voting is not necessarily elected president. However, if one switches to the subnational level, the data rapidly improve. This particularly applies when one looks at the local level and the election of mayors. Here it is generally the case that a number of elections are held in parallel within a country at a certain time and under the uniform conditions of a certain electoral system. Methodological problems such as those that can arise when comparing international cross-sectional data or national longitudinal data are ruled out here. This is why we decided to carry out our investigation at the level of local elections. There was also another reason for choosing the 2009 North Rhine-Westphalia mayoral MAGIC MAYORS 377 elections: electoral law was changed in this German state following the mayoral elections in 2004. Up until 2004, mayors were elected directly by the electorate in two rounds of voting, following the model of French majority voting. In 2009, only one 26 voting stage was used, in other words simple majority voting. Voters were advised of the change in the run-up to the 2009 election through state information campaigns and the benefits and disadvantages of the new system were widely discussed in the media. Therefore, we may assume that voters at the 2009 North Rhine-Westphalia mayoral elections were aware of the fact, that they vote for a candidate for the central position of the local political executive under an extremely simplified majority voting system. On 30 August 2009, mayoral elections took place in forty-nine metropolises and counties in North Rhine-Westphalia. In total, 264 candidates participated in the elections. The number of candidates in the local area varied between two and eleven with a median of five. In the initial step in retrieving data, candidates' personal and political background was researched on the internet in advance of the elections and portrait photographs were sought out that could form an appropriate basis for later attractive- 27 ness measurements. This was successful for all 264 mayoral candidates, which was largely due to the candidates' election campaign homepages. Only in eighteen cases (6.8 per cent) was it necessary to resort to other online sources. These sources were mainly online information provided by the local press or regional television on the mayoral elections. Details of candidates' personal and political background were coded in a joint dataset as they were to be later entered into the investigation as control variables. For the same reason, for every mayoral candidate, the logarithmised number of his or her rivals and their supporting party or parties was coded in the data set. Candidates' portrait photographs were scaled to the same height and retouched where necessary so that no clue as to their party affiliation was recognisable. The processed photographs were then incorporated into an online questionnaire for gauging attractiveness, as described in detail later. In a second step, after election day the official result was recorded for each of the forty-nine mayoral elections. For this, we relied wholly on the relevant websites of the regional returning officers for North Rhine-Westphalia. The percentages of the votes for the 264 candidates were reflected in the data set that already contained information on personal and political background, the number of rival candidates and the supporting parties. In a third and final data-collecting step, it remained for the physical attractiveness of mayoral candidates to be determined. The Truth of Consensus Method employed in attractiveness research was used as a basis. This method of determining attractiveness involves a group of raters, with each one independently rating a person's attractiveness. Averaging the individual ratings then gave a so-called 'attractiveness score' for each person assessed. The foundation for this process was the aforementioned Attractiveness Consensus, the observation that a person's attractiveness is due to a 30 characteristic which various observers perceive in a very similar way. Any variations in attractiveness rating by different observers were therefore essentially attributable to secondary differences in taste between the observers. At the same time this also meant that even with a very small group of assessors a comparable, valid and reliable attractiveness measurement could be achieved. In the literature it is generally seen as 378 GERMAN POLITICS sufficient if the attractiveness measurement is based on the ratings of two dozen assessors. The attractiveness scores are then so stable that one is unlikely to achieve a differ- 31 ent result even with 10,000 assessors. Regardless of this conclusion, there are studies that used half a dozen or fewer assessors or even a single assessor and which reveal significant and substantial effects of physical attractiveness.32 The attractiveness measurements of mayoral candidates used in this study are based on assessments given by participants via the online access panel Sozioland. Although Respondi AG supplied the assessors free of charge, participation in the survey was incentivised by Respondi AG. Because the assessment was carried out by means of an online questionnaire, all assessors were able to evaluate the candidates' photographs on their home computer. The questionnaire was structured in such a way that each assessor was randomly presented only with the candidates who stood against each other in one of the forty-nine mayoral elections. The order in which the randomly selected candidate photographs appeared was randomised anew for each assessor. Each candidate photograph was presented on its own page and the assessors did not receive any information about the persons being evaluated. The assessment was then carried out via a seven-point scale displayed underneath the photograph. The scale featured the extremes unattractive (coded as 0 in the data set) and attractive (coded as 6 in the data set). In all, 3,500 participants from the Sozioland panel were invited to participate in the survey. A total of 2,450 people took up the invitation and evaluated the portrait photographs with which they were presented. They also answered some appended demographic questions. After the survey was complete, 136 people were excluded because either they had not assessed all the displayed photographs or had not answered essential personal questions in full. A further forty people were retrospectively disqua- 34 lifted because they are not eligible to vote in Germany. Finally, thirteen people were disqualified because they were presented with the mayoral candidates who stood in the municipality in which they are entitled to vote. Due to the exclusion of these groups, the number of eligible assessors for the investigation fell to 2,261 assessors. Women, representing 56 per cent of assessors in the remaining sample, were slightly overrepre-sented. The age ranged from 18 to 80 years with an arithmetic mean of 35 years and a standard deviation of 12 years. As each assessor was only presented with a proportion of the candidate photographs for evaluation, each candidate photograph was assessed on average by only 36 forty-six assessors. Nevertheless, this value is significantly above the minimum stan-dard of twenty-four assessors suggested by Henss. Furthermore, a reliability analysis of the individual assessments confirmed the expected clear consensus in assessors' attractiveness evaluations. According to usual practice in attractiveness research, the assessors were considered as variables and the photographs as cases. The average Cronbach's a was 0.80. ° The ratings of the assessors were calculated according to the Truth of Consensus Method for every mayoral candidate by averaging the attractiveness score and reflecting the result in the previously created data set as another variable. The average attractiveness score for the mayoral candidates investigated was 2.10. The lowest value was 0.53 and the highest 4.29. These modest figures are connected to the fact that in the specific population of our investigation, women are significantly underrepresented and older people are significantly overrepresented. This can also be clearly seen in the number of cases in Table 1. At the same time, the MAGIC MAYORS 379 TABLE 1 THE ATTRACTIVENESS OF THE CANDIDATES AT THE NORTH RHINE-WESTPHALIA MAYOR ELECTIONS 2009 BY SEX, BY AGE AND BY SUPPORTING PARTIES Mean Std dev. N All 2.10 .69 264 By sex of candidate Female 2.67 .71 39 35*** Male 2.00 .64 225 By age of candidate 40 years and younger 2.50 .76 29 41 up to 50 years 2.15 .79 76 51 up to 60 years 2.07 .60 110 61 years and older 1.86 .55 49 By supporting party/parties Conservative Party 2.21 .58 44 .24* Liberal Party 2.25 .42 4 Conservative Party and Liberal Party 2.16 .59 29 Social Democratic Party 2.12 .66 46 Green Party 2.35 .59 3 Social Democratic Party and Green Party 2.39 .74 35 Socialist Party 1.94 .64 24 Others 1.92 .76 79 Note: **p < .050; **p < .010; ***p < .001 (p-values relate to F-test in ANOVA). statistical data shown in Table 1 also confirm the finding outlined in the literature that attractiveness varies with age and gender. In addition, the photographs we used - to the extent allowed by the investigation's sample - give a clear impression of the variables that favour a positive attractiveness rating, uncoupled from gender and age.39 Before we now turn to the findings of the empirical analyses, a final, methodological comment must be made about the database: 264 candidates stood in the 2009 North Rhine-Westphalia mayoral elections and we succeeded in collecting all relevant information for all 264 candidates. The analyses reported in the following section are therefore based on a census. As a result it was not necessary to apply inference statistics. As there are diverging opinions within the research community on this subject,40 we will report the significances in relation to the information, but we will not interpret them. EMPIRICAL ANALYSIS: PHYSICAL ATTRACTIVENESS WORKS - EVEN UNDER THE CONDITIONS OF A PRESIDENTIAL ELECTORAL SYSTEM All the following reported empirical analyses were performed as OLS regressions. The percentage share of the vote that individual candidates were able to gain in the election in which they participated represents the dependent variable. In our initial approach to the subject of the investigation, we calculated a model which only incorporated candidates' attractiveness score as a predictor (see Model 1A in Table 2). If one follows the findings of this model, a candidate's share of the vote rises on average by 2.95 percentage points if his or her physical attractiveness rises by 1 scale point. If one considers that there is a difference of 3.76 scale points between the most unattractive candidate (attractiveness score = 0.53) and the most attractive candidate 380 GERMAN POLITICS TABLE 2 OLS-REGRESSION MODELS OF THE DETERMINANTS OF THE ELECTORAL SUCCESS AT THE NORTH RHINE-WESTPHALIA MAYOR ELECTIONS 2009 Winner and 1st All candidates follow-up Model Model 1A IB Model 1C Model ID Model 2 Intercept 12 37*** —89 93*** -39.97* 22.41* -57.65 Candidates' attractiveness 2.95+ 5.66" 4 gg*** 1.30* 4.35** Candidate is female -11.02*** -7.80** -2.64* -7.27* Candidates' age 3 72*** 2.38** .70* 3.68** Candidates' age (squared) -.02** -.01+ -.04** Candidate holds an academic -1.66 -2.26 1.53 6.11 title Candidate holds a title of -6.20 -1.49 1.20 -.- nobility Candidate has an immigration -8.21 -6.19 .31 -.- background Candidate has come out as 9.69 10.88 .93 2.18 homosexual Candidate is married 5.58 1.69 -.75 1.15 Candidate is at least one time -4.84 -2.40 -.91 -2.70 divorced Candidate is incumbent 33.43*** 16.92*** 14.34*** Log. number of candidates' -9.86*** -3.26*** -5.23** opponents Supporting parties Conservative Party (reference category) Liberal Party -30.08*** Conservative Party and .55 .24 Liberal Party Social Democratic Party -6.02*** -5.93** Green Party -28.26*** Social Democratic Party and -3.02 -1.27 Green Party Socialist Party -32.78*** -.- Others -33.46*** -9.70 R2 (per cent) 1.1 16.4 57.0 91.9 60.3 N 264 264 264 264 98 Note: Entries are unstandardised coefficients; +p < .100; *p < .050; **p < .010; ***p < .001; (p-values relate to f-test). -.- predictor has no variance in this subpopulation. (attractiveness score = 4.29), attractiveness can theoretically effect a change in vote share of up to eleven percentage points. Given the expectation that candidates' physical attractiveness should not have an influence under the conditions of a presidential electoral system, this is a more than considerable effect. Nevertheless, this initial modelling did not take into account a range of candidate variables that can also affect the election result and which could - at least partially -be correlated with the physical attractiveness of a candidate. For this reason, we considered the following candidate variables in a second analytical stage: gender,41 age and squared age,42 immigration background,43 admission of own homosexuality,44 academic title,45 title of nobility46 and marital status.47 If one includes these control MAGIC MAYORS 381 variables in the analysis this does not lead to the effect of physical attractiveness being lowered. On the contrary, it increases considerably (see Model IB in Table 2). 48 This also does not change when the logarithmic number of rival candidates and a possible incumbency49 are considered (see Model 1C in Table 2). A clear reduction in the effect of physical attractiveness is only seen when an addition control is made for the strategic potential vote of the candidates.50 According to the results reported in Model ID in Table 2, the vote share increases on average by only 1.30 percentage points when physical attractiveness alters by 1 scale point. Just on the basis of the figures, the maximum change in vote share for the given attractiveness range of 3.76 scale points would be almost five percentage points according to the model's findings. Yet even this effect must still be ranked as substantially important since it could be decisive in a neck-and-neck race. Furthermore, it should be noted that the variables by which the strategic potential vote was operationalised demonstrate a systematic relationship to candidates' physical attractiveness (see again Table 1). In particular, candidates who are supported by the Socialist Party or others have, on average, only a rather modest level of attractiveness. In a multivariate analysis that would probably lead to a part of the attractiveness effect being absorbed by dummy variables on strategic potential vote and the influence of attractiveness therefore being underestimated. Thus far it appears as though candidate attractiveness can also influence electoral success in mayoral elections. Our original expectation that a presidential electoral system could cancel out the effect of physical attractiveness would then be disproved. Nevertheless, there is a point of criticism with regard to our analyses not previously discussed that could relativise our findings. According to the arguments of Banducci et a/.,51 political elections at a local level are low-salience, low-information elections which leave plenty of scope for candidate attractiveness to influence the election result. They base this assumption on a number of studies that achieved similar results in relation to candidate gender or ethnicity. CO However, we would question this view; nonetheless the implications that could impact on our investigation should be empirically tested. If one assumes that in mayoral elections usually a maximum of two candidates - the subsequent winner and runner-up - have a realistic chance of victory, then rational, interested and well-informed voters should focus on these two candidates if they do not want to waste their vote.54 However, these voters have to then see the election as having at least a degree of significance in order for them to inform themselves at all about who are the two most promising candidates, and they would have to acquire a certain degree of additional information in order to decide between the two most promising candidates. For this sub-electorate, a mayoral election would then be to a lesser extent a low-salience, low-information election and it would indeed be plausible that these voters make their decision less on irrelevant candidate traits and more on relevant personal characteristics and issues. For our investigation this would then also mean that the effect of physical attractiveness diminishes or disappears if the empirical analyses are performed only for this subpopulation of election winners and runners-up. Model 2 in Table 2 shows the result of the relevant calculations. Indeed, the effect of physical attractiveness changes dramatically. It is now three times as strong as in the comparable model for all mayoral candidates (Model ID in Table 2) and has an unstandardised regression coefficient of 4.35. 382 GERMAN POLITICS CONCLUSION: PHYSICAL ATTRACTIVENESS MATTERS - AND OUTWEIGHS ELECTORAL SYSTEM, LEVEL OF ELECTION, AND POLITICAL CULTURE The starting point for the present investigation was the establishment that a range of empirical studies had essentially proven the influence of candidates' physical attractiveness on the results of political elections, demonstrating not only the effect of attractiveness in different political contexts, but also at different levels of the political system and in varying electoral law conditions for simple constituency and list candidates as well as front-runners. Despite the complexity of existing bodies of work, among those studies using real elections results as a database, no single study could be found that investigated the influence of candidate attractiveness on electoral outcome under the conditions of a presidential electoral system. In other words, there is no such study that has tested the extent to which candidate attractiveness still has an effect when a central position in the political executive is being filled on the basis of majority voting. This gap in research is therefore primarily of importance because a good theoretical argument can be made as to why a candidate's physical attractiveness should not be relevant under these conditions. On the other hand, this is relevant because - in the case of supporting empirical findings - it would provide starting points for eliminating the influence of physical attractiveness which is doubtless undesirable from the point of view of normative democratic theory. In order to bridge this gap, we tested whether or not the influences of candidates' physical attractiveness on their electoral success can be proved using the 2009 North Rhine-Westphalia mayoral elections. The empirical findings of the analytical models produced were overwhelmingly positive. Even under the conditions of a presidential electoral system, candidate attractiveness still has an impact. When seen together with studies conducted on the effect of physical attractiveness in political elections in other contexts, these findings support the impression that a candidate's outer appearance affects the election result, on the whole regardless of the conditions of a specific election: it appears that a country's political tradition does not play a role, the system under which the election is held appears to be of secondary importance, and the levels of political system at which the election takes place do not appear to have a substantial impact on the strength of the attractiveness effect. It remains to be seen whether this effect can be fully confirmed were these aspects to be systematically varied quasi-experimentally within a study, or whether adjusting mechanisms could be identified that influence the effectiveness of candidate attractiveness. But this would be an ambitious and attractive assignment for further research projects. ACKNOWLEDGEMENTS We would like to thank the anonymous reviewers for the constructive criticism and many helpful comments. We would also like to thank Svenja Döbler and Laura Mly-narek for proof reading. ABOUT THE AUTHORS Ulrich Rosar is Professor of Sociology at the Heinrich-Heine-University Düsseldorf. He holds a PhD from the University of Bamberg and a Habilitation from the University MAGIC MAYORS 383 of Cologne. His research covers the areas of political sociology, attractiveness research, voting and methods of social research. Core publication with the coauthors: 'The Frog Pond Beauty Contest: Physical Attractiveness and Electoral Success of the Constituency Candidates at the North Rhine-Westphalia State Election 2005', European Journal of Political Research 47/1 (2008), pp.64-79; 'Fabulous Front-Runners: Eine Analyse zur Bedeutung der physischen Attraktivität von Spitzenkandidaten für den Wahlerfolg ihrer Parteien', Politische Vierteljahresschrift 50/4 (2009), pp.754-73; together with Hermann Dülmer, Markus Klein, Dieter Ohr und Markus Quandt, 'Response Sets in the Measurement of Values: A Comparison of Rating and Ranking Procedures', International Journal of Public Opinion Research 16/4 (2004), pp.474-83. Markus Klein is Professor of Political Sociology at the Leibniz University Hannover. He holds a PhD from the University of Mainz and a Habilitation from the University of Cologne. His research areas are voting and value research, attractiveness research and methods of social research. Publications: see above and editor of (together with Tim Spier et al), Parteimitglieder in Deutschland (Wiesbaden: Verlag für Sozialwissenschaften, 2011); (together with Ulrich Rosar), 'Wirkungen des TV-Duells im Vorfeld der Bundestagswahl 2005 auf die Wahlentscheidung: Eine empirische Analyse unter besonderer Berücksichtigung von Medieneinflüssen auf die Siegerwahrnehmung und subjektiven Erwartungshaltungen an die Debattenperformance der Kandidaten', Kölner Zeitschrift für Soziologie und Sozialpsychologie 59/1 (2007), pp.81-104; forthcoming: 'Postmaterialism', in Bertrand Badie, Dirk Berg-Schlosser and Leonardo Morlino (eds), International Encyclopedia of Political Science (Thousand Oaks, CA: Sage, 2011). Tilo Beckers is Akademischer Rat (equiv. of assistant professor) for Sociology and Methods of Social Research at the Heinrich-Heine-University Düsseldorf. He holds a PhD from the University of Cologne. His research focus is in the areas of (cross-national) survey research, research in values, attitudes and morality. Publications: see above and: 'Comparing Comparisons as a Cross-Validation Strategy (CC-CV): A Proposal for Applied Survey Research Using the Example of Attitudes towards Economic Liberalism', in Tilo Beckers, Klaus Birkelbach, Jörg Hagenah and Ulrich Rosar (eds), Komparative empirische Sozialforschung (Wiesbaden: VS-Verlag, 2010); forthcoming, together with Pascal Siegers and Anabel Kuntz, 'Congruence and Performance of Value Concepts in Social Research', Survey Research Methods 6/1 (2012), pp.13-24. NOTES 1. See, among many other works, M.P. Wattenberg, The Rise of Candidate-Centered Politics: Presidential Elections of the 1980s (Cambridge, MA: Harvard University Press, 1991); J. Lass, Vorstellungsbilder über Kanzlerkandidaten: Zur Diskussion um die Personalisierung von Politik (Wiesbaden: Deutscher Universitätsverlag, 1995); F.U. Pappi and S. Shikano, 'Personalisierung der Politik in Mehrparteiensystemen am Beispiel deutscher Bundestagswahlen seit 1980', Politische Vierteljahresschrift 42/3 (2001), pp.355-87; F. Brettschneider, Spitzenkandidaten und Wahlerfolg (Wiesbaden: Westdeutscher Verlag 2002). 2. U. Rosar, 'Fabulous Front-Runners: Eine Analyse zur Bedeutung der physischen Attraktivität von Spitzenkandidaten für den Wahlerfolg ihrer Parteien und zu den Möglichkeiten der Gegensteuerung durch das Wahlrecht', Politische Vierteljahresschrift 50/4 (2009), pp.754-73. GERMAN POLITICS See A.W. Barrett and L.W. Barrington, Ts a Picture Worth a Thousand Words? Newspaper Photograph and Voter Evaluations of Political Candidates', Harvard International Journal of Press/Politics 10/4 (2005), pp.98-113; T.L. Budesheim and SJ. DePaula, 'Beauty or the Beast? The Effects of Appearance, Personality, and Issue Information on Evaluations of Political Candidates', Personality and Social Psychology Bulletin 20/4 (1994), pp.339-48; M. Klein and U. Rosar, 'Ist Deutschland reif für eine Kanzlerin? Eine experimentelle Untersuchung aus Anlass der Bundestagswahl 2005', in F. Brettschneider, O. Niedermayer, B. Pfetsch and B. Wessels (eds), Die Bundestagswahl 2005: Analysen aus Sicht der Wahlforschung, der Kommunikationswissenschaft und der Parteienforschung (Wiesbaden: VS Verlag für Sozialwissenschaften, 2007), pp.271-91; M. Klein and U. Rosar, 'Sie, Sie, Sie oder Er? Angela Merkel im Spiegel der Daten einer experimentellen Befragung', in O.W. Gabriel, J.W. Falter and B. Weßels (eds), Wahlen und Wähler: Analysen aus Anlass der Bundestagswahl 2005 (Wiesbaden: VS Verlag für Sozialwissenschaften, 2009), pp.346-57; K.E. Lewis and M. Bierly, 'Toward a Profile of the Female Voter: Sex Differences in Perceived Physical Attractiveness and Competence of Political Candidates', Sex Roles 22/1-2 (1990), pp.1-12; E.D. Riggle, V.C. Ottati, R.S. Wyer, J. Kuklinski and N. Schwarz, 'Basis of Political Judgements: The Role of Stereotypic and Nonstereotypic Information', Political Behavior 14/1 (1992), pp.67-87; K. Rohrbach and U. Rosar, Merkel Reloaded: Eine experimentelle Untersuchung zur Relevanz von Geschlecht und physischer Attraktivität aus Anlass der Bundestagswahl 2009 (Hannover and Köln: Working Paper, Leibniz Universität Hannover und Universität zu Köln, 2010); S.W. Rosenberg, L. Bohan, P. McCafferty and K. Harris, 'The Image and the Vote: The Effect of Candidate Presentation on Voter Preference', American Journal of Political Science 30/1 (1986), pp.108-27; S.W. Rosenberg, S. Kahn and T. Tran, 'Creating a Political Image: Shaping Appearance and Manipulating the Vote', Political Behavior 13/4 (1991), pp.345-67; J.N. Schubert and MA. Curran, 'Stereotyping Effects in Candidate Evaluation: The Interaction of Gender and Attractiveness Bias', Paper Prepared for Presentation at the 2001 Annual Meeting of the Midwestern Political Science Association, Northern Illinois University, DeKalb, 2001; C.K. Sigelman, D.B. Thomas, L. Sigelman and F.D. Robich, 'Gender, Physical Attractiveness, and Electability: An Experimental Investigation of Voter Biases', Journal of Applied Social Psychology 16/3 (1986), pp.229-48; L. Sigelman, C.K. Sigelman and C. Fowler, 'A Bird of a Different Feather? An Experimental Investigation of Physical Attractiveness and the Electability of Female Candidates', Social Psychology Quarterly 50/1 (1987), pp.32-43. S.A. Banducci, JA. Karp, M. Thrasher and C. Railings, 'Ballot Photographs as Cues in Low-Information Elections', Political Psychology 29/6 (2008), pp.903-17; N. Berggren, H. Jordahl and P. Poutvaara, 'The Looks of a Winner: Beauty, Gender and Electoral Success', Journal of Public Economics 94/1-2 (2010), pp.8-15; F. Buckley, N. Collins and T. Reidy, 'Ballot Paper Photographs and Low-Information Elections in Ireland', Politics 27/3 (2007), pp. 174-81; M. Klein and U. Rosar, 'Physische Attraktivität und Wahlerfolg: Eine empirische Analyse am Beispiel der Wahlkreiskandidaten bei der Bundestagswahl 2002', Politische Vierteljahresschrift 46/2 (2005), pp.263-87; G. Lutz, The Electoral Success of Beauties and Beasts: FORS Working Papers 2009-2 (Lausanne: Swiss Foundation for Research in Social Sciences, 2009); Rosar, 'Fabulous Front-Runners'; U. Rosar and M. Klein, 'And the Winner Is ... - Ein Drei-Länder-Vergleich zum Einfluss der physischen Attraktivität von Direktkandidaten auf den Wahlerfolg bei nationalen Parlamentswahlen', in T. Beckers, K. Birkelbach, J. Hagenah and U. Rosar (eds), Komparative empirische Sozialforschung (Wiesbaden: VS Verlag für Sozialwissenschaften, 2010), pp.307-35; U. Rosar and M. Klein, Physical Attractiveness and Electoral Success: An Empirical Micro-Macro-Analysis at the Example of the Constituency-Candidates at the German Federal Election 2005 (Köln and Hannover: Working Paper, Universität zu Köln und Leibniz Universität Hannover, 2010); U. Rosar and M. Klein, Pretty Politicians: Die physische Attraktivität von Spitzenkandidaten, ihr Einfluss bei Wahlen und die These der Personalisierung des Wahlverhaltens (Köln and Hannover: Working Paper, Universität zu Köln und Leibniz Universität Hannover, 2010); U. Rosar, M. Klein and T. Beckers, 'The Frog Pond Beauty Contest: Physical Attractiveness and Electoral Success of the Constituency Candidates at the North Rhine-Westphalia State Election of 2005', European Journal of Political Research 47/1 (2008), pp.64-79. Another study looking at the context of the United States can also be found in the investigation by A. Todorov, A.N. Mandisodza, A. Goren and C.C. Hall, 'Inferences of Competence from Faces Predict Election Outcomes', Science 308/5728 (2005), pp.1623-26. The following clarifications may be adapted for C.C. Bailew, II and A. Todorov, 'Predicting Political Elections from Rapid and Unreflective Face Judgments', Proceedings of the National Academy of Sciences of the United States of America 104 (2007), pp.17948-53. In this study Todorov et al. investigated the inferences of competence based on portrait photographs. They were able to show that at the American Senate elections of 2000, 2002 and 2004 there was a 71.6 per cent probability of correctly predicting the winner in each state based on these inferences of competence. In the elections for the House of Representatives in 2002 and 2004, this figure was still 66.8 per cent. However, the authors claimed that they 'ruled out MAGIC MAYORS 385 the possibility that the age, attractiveness, and/or familiarity with the faces of the candidates could account for the relationship between inferences of competence and election outcomes': Todorov et ai, 'Inferences of Competence', p. 1625. In a commentary on this study Zebrowitz and Montepare pointed out that the decisive variable behind the inferences of competence could be the 'mature-' or 'babyfacedness' of the candidates. See L.A. Zebrowitz and J.M. Montepare, 'Appearance DOES Matter', Science 308/5728 (2005), pp.1565-6. They went on to say that from the point of view of sociopsychological attractiveness research these are classic features of male attractiveness/unattrac-tiveness. Poutvaara et ai, who carried out an investigation for Finland, came to the conclusion, however, that the 'babyfacedness' of a candidate is indeed linked to the assessment of his or her political competence, but at the same time tends to be either uncorrected with electoral success or tends to be positively correlated. See P. Poutvaara, H. Jordahl and N. Berggren, 'Faces of Politicians: Babyfacedness Predicts Inferred Competence But Not Electoral Success', Journal of Experimental Social Psychology 45/5 (2009), pp.1132-5. See also M.G. Efran and E.W.J. Patterson, 'Voters Vote Beautiful: The Effect of Physical Appearance on a National Election', Behavioral Science 6/4 (1974), pp.352-6; A. King and A. Leigh, Beautiful Politician: Working Paper (Adelaide: University of South Australia, 2007); A. Leigh and T. Susilo, 'Is Voting Skin-Deep? Estimating the Effect of Candidate Ballot Photographs on Election Outcomes', Journal of Economic Psychology 30/1 (2009), pp.61-70. 6. Berggren et ai, 'The Looks of a Winner'; Efran and Patterson, 'Voters Vote Beautiful'; King and Leigh, Beautiful Politician; Klein and Rosar, 'Physische Attraktivität und Wahlerfolg'; Lutz, The Electoral Success of Beauties and Beasts; Klein, 'And the Winner Is ...'; Rosar and Klein, Physical Attractiveness and Electoral Success. 7. Banducci et ai, 'Ballot Photographs as Cues in Low-Information Elections'; Berggren et ai, 'The Looks of a Winner'; Buckley et ai, 'Ballot Paper Photographs and Low-Information Elections in Ireland'; Leigh and Susilo, 'Is Voting Skin-Deep?'. 8. U. Rosar and M. Klein, 'Mein(schöner)Prof.de. Die physische Attraktivität des akademischen Lehrpersonals und ihr Einfluss auf die Ergebnisse studentischer Lehrevaluationen', Kölner Zeitschrift für Soziologie und Sozialpsychologie 61/4 (2009), pp.621-45; Rosar and Klein, Pretty Politicians; Rosar et ai, 'The Frog Pond Beauty Contest'. 9. M. Klein and U. Rosar, Front-Runner's Attractiveness and Electoral Success at the European Elections 2004: An Empirical Analysis. Working Paper (Hannover und Köln: Leibniz Universität Hannover und Universität zu Köln, 2010). 10. See the empirical findings of Banducci et ai, 'Ballot Photographs as Cues in Low-Information Elections'; Berggren et ai, 'The Looks of a Winner'; Efran and Patterson, 'Voters Vote Beautiful'; King and Leigh, Beautiful Politician; Klein and Rosar, 'Physische Attraktivität und Wahlerfolg'; Klein and Rosar, Front-Runner's Attractiveness and Electoral Success at the European Elections 2004; Rosar, 'Fabulous Front-Runners'; Rosar and Klein, 'And the Winner Is ...'; Rosar and Klein, Physical Attractiveness and Electoral Success; Rosar et ai, 'The Frog Pond Beauty Contest'. 11. See Spence's signalling theory. He describes the decision-making process under uncertainty from the perspective of employers. See M. Spence, 'Job Market Signaling', Quarterly Journal of Economics 3/87 (1973), pp.355-74. Spence differentiates between indices and signals and he underlines the importance of the possibility to use signals actively: 'There is not much that the applicant can do about indices. Signals, on the other hand, are alterable and therefore potentially subject to manipulation by the job applicant' : Spence, 'Job Market Signaling', p.357. See also K.J. Arrow, 'Higher Education as a Filter', Journal of Public Economics 2/3 (1973), pp.193-216; A. Weiss, 'Human Capital vs. Signalling Explanations of Wage', Journal of Economic Perspectives 9/4 (1995), pp.133-54; M. Spence, 'Signaling in Retrospect and the Informational Structure of Markets', American Economic Review 92/3 (2002), pp.434-59. In our context, physical attractiveness is an index but may also be turned into a signal because beauty is subject to possible manipulations which may become important in mayoral races. See the application of Hönekopp et al. relating body fitness and physical attractiveness ratings: J. Hönekopp, U. Rudolph, L. Beier, A. Liebert and C. Müller, 'Physical Attractiveness of Face and Body as Indicators of Physical Fitness in Men', Evolution and Human Behavior 28/2 (2007), pp.106-11. 12. North Rhine-Westphalia is the biggest of sixteen German federal states with a population of roughly 18,000,000 inhabitants. It consists of fifty-four subunits, twenty-three metropolises (so-called Kreisfreie Städte) and thirty-one counties (so-called Landkreise). On 30 August 2009 there were mayoral elections in forty-nine of the fifty-four subunits (twenty-seven counties and twenty-two metropolises with a total of roughly 14,400,000 citizens aged over 16 who are eligible to vote). Because of death or demission of the incumbent, five subunits held snap elections before 2009. Therefore they were excluded from our analysis. 13. The German title for the mayor of a metropolis is Oberbürgermeister (literal translation: chief mayor). The title for the mayor of a county is Landrat (literal translation: county administrator). To simplify matters, we use the title of mayor for both. GERMAN POLITICS Council (Rat) in the cities, County Council (Kreistag) in the counties. A. Kost, 'Kommunalpolitik in Nordrhein-Westfalen', in A. Kost and H.-G Wehling (eds), Kommunalpolitik in den deutschen Ländern (Wiesbaden: Verlag für Sozialwissenschaften, 2010), pp.231-54. N. Etcoff, Survival of the Prettiest (New York: Doubleday & Company, 2000); E. Hatfield and S. Sprecher, Mirror, Mirror: The Importance of Looks in Everyday Life (Albany, NY: State University of New York Press, 1986); G.L. Patzer, The Physical Attractiveness Phenomena (New York: Plenum, 1985); see also the meta-analyses by A.H. Eagly, R.D. Ashmore, M.G Makhijani and L.C. Longo, 'What Is Beautiful Is Good, But...: A Meta-Analytic Review of Research on the Physical Attractiveness Stereotype', Psychological Bulletin 110/1 (1991), pp.109-28; A. Feingold, 'Matching for Attractiveness in Romantic Partners and Same-Sex Friends: A Meta-Analysis and Theoretical Critique', Psychological Bulletin 104/2 (1988), pp.226-35; A. Feingold, 'Gender Differences in Effects of Physical Attractiveness on Romantic Attraction: A Comparison across Five Research Paradigms', Journal of Personality and Social Psychology 59/5 (1990), pp.981-93; A. Feingold, 'Good-Looking People Are Not What We Think', Psychological Bulletin 111/2 (1992), pp.304-41; M. Hosoda, E.F. Stone-Romero and G. Coats, 'The Effects of Physical Attractiveness on Job-Related Outcomes: A Meta-Analysis of Experimental Studies', Personnel Psychology 56/2 (2003), pp.431-62 ; LA. Jackson, J.E. Hunter and C.N. Hodge, 'Physical Attractiveness and Intellectual Competence: A Meta-Analytic Review', Social Psychology Quarterly 58/2 (1995), pp.108-22; J.H. Langlois, L. Kalakanis, A.J. Rubenstein, A. Larson, M. Hallam and M. Smoot, 'Maxims or Myths of Beauty? A Meta-Analytic and Theoretical Review', Psychological Bulletin 126/3 (2000), pp.390-423; R. Mazzella and A. Feingold, 'The Effects of Physical Attractiveness, Race, Socioeconomic Status, and Gender of Defendants and Victims on Judgments of Mock Jurors: A Meta-Analysis', Journal of Applied Social Psychology 24/15 (1994), pp.1315-44. Rosar et al., 'The Frog Pond Beauty Contest'. B. Köhler, 'Physische Attraktivität und Persönlichkeitsmerkmale', in M. Amelang and H.-J. Ahrens (eds), Brennpunkte der Persönlichkeitsforschung. Band 1 (Göttingen: Hogrefe, 1984), pp.l40ff. J.F. Cross and J. Cross, 'Age, Sex, Race, and the Perception of Facial Beauty', Developmental Psychology 5/3 (1971), pp.433-9; M.R. Cunningham, 'Measuring the Physical in Physical Attractiveness: Quasi-Experiments on the Sociobiology of Female Beauty', Journal of Personality and Social Psychology 50/5 (1986), pp.925-35; M.R. Cunningham, A.P. Barbee and C.L. Pike, 'What Do Women Want? Facialmetric Assessment of Multiple Motives in the Perception of Male Physical Attractiveness', Journal of Personality and Social Psychology 59/1 (1990), pp.61-72; M.R. Cunningham, A.R. Roberts, A.P. Barbee, P.B. Druen and C.-H. Wu, '"Their Ideas of Beauty Are, on the Whole, the Same As Ours": Consistency and Variability in the Cross-Cultural Perception of Female Physical Attractiveness', Journal of Personality and Social Psychology 68/2 (1995), pp.261-79; M.R. Cunningham, P.B. Druen and A.P. Barbee, 'Angels, Mentors, and Friends: Trade-offs among Evolutionary, Social, and Individual Variables in Physical Appearance', in J.A. Simpson and D.T. Kenrick (eds), Evolutionary Social Psychology (Mahwah, NJ: Lawrence Erlbaum, 1997), pp. 109-40; R. Henss, 'Zur Beurteilerübereinstimmung bei der Einschätzung der physischen Attraktivität junger und alter Menschen', Zeitschrift für Sozialpsychologie 18/2 (1987), pp.118-30; R. Henss, "Spieglein, Spieglein an der Wand ...": Geschlecht, Alter und physische Attraktivität (Weinheim: Psychologie Verlags Union, 1992); R. Henss, Gesicht und Persönlichkeitseindruck (Göttingen: Hogrefe, 1998); A.H. Iliffe, 'A Study of Preferences in Feminine Beauty', British Journal of Psychology 51/1 (1960), pp.267-73; D. Jones, 'Sexual Selection, Physical Attractiveness and Facial Neoteny: Cross-Cultural Evidence and Implications', Current Anthropology 36/5 (1995), pp.723-48; D. Jones and K. Hill, 'Criteria of Facial Attractiveness in Five Populations', Human Nature 4/3 (1993), pp.271-96; R. Kowner and T. Ogawa, 'The Role of Raters' Sex, Personality, and Appearance in Judgments of Facial Beauty', Perceptual and Motor Skills 81/1 (1995), pp.339-49; G. Rhodes, K. Harwood, S. Yoshikawa, M. Nishitani and I. McLean, 'The Attractiveness of Average Faces: Cross-Cultural Evidence and Possible Biological Bias', in G. Rhodes and L.A. Zebrowitz (eds), Facial Attractiveness: Evolutionary, Cognitive, and Social Perspectives (Westport, CT and London: Ablex Publishing, 2002), pp.35-58; G. Rhodes, K. Lee, R. Palermo, M. Weiss, S. Yoshikawa, P. Clissa et al., 'Attractiveness of Own-Race, Other-Race, and Mixed-Race Faces', Perception 34/3 (2005), pp.319-40. K.K. Dion, E. Berscheid and E. Walster, 'What Is Beautiful Is Good', Journal of Personality and Society Psychology 24/3 (1972), pp.285-90. See S. Chaiken, 'Communicator Physical Attractiveness and Persuasion', Journal of Personality and Social Psychology 37/8 (1979), pp.1387-97; M. Dermer and D.L. Thiel, 'When Beauty May Fail', Journal of Personality and Social Psychology 31/6 (1975), pp. 1168-76; Dion et al., 'What Is Beautiful Is Good'; Eagly et al., 'What Is Beautiful Is Good, But...'; Feingold, 'Good-Looking People Are Not What We Think'; Henss, Gesicht und Persönlichkeitseindruck; A.G. Miller, 'Role of Physical Attractiveness in Impression Formation', Psychonomic Science 19/4 (1970), pp.241-3; R.K. Unger, MAGIC MAYORS 387 M. Hilderbrand and T. Madar, 'Physical Attractiveness and Assumptions about Social Deviance: Some Sex by Sex Comparisons', Personality and Social Psychology Bulletin 8/2 (1982), pp.293-301. 22. J.K. Maner, M.T. Gailliot, D.A. Rouby and S.L. Miller, 'Can't Take My Eyes Off You: Attentional Adhesion to Mates and Rivals', Journal of Personality and Social Psychology 93/3 (2007), pp.389-401; J.K. Maner, D.T. Kenrick, V.D. Becker, A.W. Delton, B. Hofer, C J. Wilbur et al., 'Sexually Selective Cognition: Beauty Captures the Mind of the Beholder', Journal of Personality and Social Psychology 85/6 (2003), pp. 1107-20; M. Mulford, J. Orbell, C. Shatto and J. Stockard, 'Physical Attractiveness, Opportunity and Success in Everyday Exchange', American Journal of Sociology 103/6 (1998), pp.1565-93. 23. P.L. Benson, S.A. Karabenic and R.M. Lerner, 'Pretty Pleases: The Effects of Physical Attractiveness on Race, Sex and Receiving Help', Journal of Experimental Social Psychology 12/5 (1976), pp.409-15; F. Bian, The Effects of Attractiveness on Helping Behavior (Claremont, CA: Harvey Mudd College, 1997); FJ.M. Dabbs and N.A. Stokes, III, 'Beauty Is Power: The Use of Space on the Sidewalk', Socio-metry 38/4 (1975), pp.551-7; K.K. Dion and E. Berscheid, 'Physical Attractiveness and Peer Perception among Children', Sociometry 37/1 (1974), pp.1-12; JJ. Hartnett, K.O. Bailey and C.S. Hartley, 'Body Height, Position and Sex as Determinants of Personal Space', Journal of Psychology 87/1 (1974), pp.129-36; Hatfield and Sprecher, Mirror, Mirror, Langlois et al., 'Maxims or Myths of Beauty?'; A. Marwick, Beauty in History: Society, Politics and Personal Appearance c. 1500 to the Present (London: Thames and Hudson, 1988); V. McCabe, 'Facial Proportions, Perceived Age, and Caregiving', in T.R. Alley (ed.), Social and Applied Aspects of Perceiving Faces (Hillsdale, NJ: Earlbaum, 1988), pp. 89-95; Mulford et al., 'Physical Attractiveness, Opportunity and Success in Everyday Exchange'; J.M. Ritter, RJ. Casey and J.H. Langlois, 'Adults' Responses to Infants Varying in Appearance of Age and Attractiveness', Child Development 62/1 (1991), pp.68-82; C. White Stephan and J.H. Langlois, 'Baby Beautiful: Adult Attributions of Infant Competence as a Function of Infant Attractiveness', Child Development 55/2 (1984), pp.576-85; M. Wilson and J.F. Dovidio, 'Effects of Perceived Attractiveness and Feminist Orientation on Helping Behavior', Journal of Social Psychology 125/4 (1985), pp.415-20; R.K. Wilson and C.C. Eckel, 'Judging a Book by Its Cover: Beauty and Expectations in the Trust Game', Political Research Quarterly 59/2 (2006), pp.189-202. 24. J.N. Bassili, 'The Attractiveness Stereotype: Goodness or Glamour?', Basic and Applied Social Psychology 2/4 (1981), pp.235-52; Dion et al., 'What Is Beautiful Is Good'; K. Grammer, Signale der Liebe: Die biologischen Gesetze der Partnerschaft (Frankfurt a.M.: dtv, 2002), p.169. 25. Rosar et al, 'The Frog Pond Beauty Contest'. 26. The run-off election was abolished because in 1999 and 2004 the turnout in the second round of the mayoral elections was much lower than in the first round. This was considered to be problematic from a legitimation perspective. Vice versa, in 2009, some mayors were elected into office with less than 40 per cent of the votes. Therefore, the run-off was reintroduced in the communal election law in 2011. See the extensive website of the Ministry of internal and communal affairs (Ministeriums für Inneres und Kommunales des Landes Nordrhein-Westfalen): www.wahlen.nrw.de. 27. In order to counter a possible objection to our findings in this regard: it could be said that a counterargument to the use of photographs as the basis for determining attractiveness is that the attractiveness of the persons depicted can be artificially increased beyond their natural appearance through cosmetic measures and digital or manual retouching, thus falsifying the attractiveness measurement. See A. Hergovich, S. Hasenegger and K. Koller, 'Eine empirische Studie zum Einfluss von Make-up auf die Beurteilung der Attraktivität', in A. Hergovich (ed.), Psychologie der Schönheit: Physische Attraktivität aus wissenschaftlicher Perspektive (Wien: WUV-Universitätsverlag, 2002), pp.129-35; S.W. Rosenberg and P. McCafferty, 'The Image and the Vote: Manipulating Voters' Preferences', Public Opinion Quarterly 51/1 (1987), pp.31 -47; S.W. Rosenberg, S. Kahn and T. Tran, 'Creating a Political Image: Shaping Appearance and Manipulating the Vote', Political Behavior 13/4 (1991), pp.345-67. Furthermore, the staging of the photograph and the fashion, clothing style, ornamentation, gesture and facial expression of the depicted person or the technical quality of the photograph can influence the attractiveness rating. See A.J. Elliot and D. Niesta, 'Romantic Red: Red Enhances Men's Attraction to Women', Journal of Personality and Social Psychology 95/5 (2008), pp.1150-64; D.R. Osborn, 'Beauty Is as Beauty Does? Make Up and Posture Effects on Physical Attractiveness Judgments', Journal of Applied Social Psychology 26/1 (1996), pp.31-51; H.T. Reis, I. McDougal Wilson, C. Monestere, S. Berstein, K. Clark, E. Seidl et al., 'What Is Smiling Is Beautiful and Good', European Journal of Social Psychology 20/3 (1990), pp.259-67. This is all correct but is of low importance in respect of this investigation: each of these implications works against the basic hypothesis that physical attractiveness has a provable influence on the election result. In this respect there is merely a tightening of the test conditions. Another argument for not using portrait photographs as the basis for attractiveness assessment is that portrait shots do not adequately give an overall impression of the person. This GERMAN POLITICS reservation is, however, unfounded. That is to say, it can be shown that the assessment of a person's attractiveness hardly alters when a full-length photograph is used instead of a portrait shot. Moreover, the assessment of facial attractiveness matches to a high degree the assessment of the entire person. See E. Brunswik, Perception and the Representative Design of Psychological Experiments (Berkeley, CA: University of California Press, 1956); K. Grammer, B. Fink, A. Juette, G. Ronzal and R. Thornhill, 'Female Faces and Bodies: N-Dimensional Feature Space and Attractiveness', in Rhodes and Zebrowitz (eds), Facial Attractiveness, pp.91-126; M. Snyder, E. Berscheid and P. Glick, 'Focusing on the Exterior and the Interior: Two Investigations of Personal Relationships', Journal of Personality and Social Psychology 48/6 (1985), pp. 1427-39. And even when video sequences or observations in natural surroundings form the basis for assessment, the attractiveness rating does not alter compared to that of a portrait photograph. See T.A. Brown, T.F. Cash and S.W. Noles, 'Perception of Physical Attractiveness among College Students: Selected Determinants and Methodological Matters', Journal of Social Psychology 126/3 (1999), pp.305-16. To be precise, the logarithmised number of rival candidates is not an individual variable, but a contextual variable that has the same expression for all candidates in a particular mayoral election. Accordingly, it would actually be appropriate to build a hierarchically structured data set and conduct upcoming analyses as hierarchical-linear regression analyses. In half of all mayoral elections involved in the investigation, however, no more than five candidates were standing. In addition, detailed analyses were to be presented with only two candidates from each election. According to Joop Hox, these numbers of level-1 units per level-2 unit are far too low to allow the derivation of reliable and stable multi-level models: see J.J. Hox, Multilevel Analysis: Techniques and Applications (New Jersey/ London: Lawrence Erlbaum Associates, 2002). It was therefore decided to create a simple, non-hier-archically structured data set and to conduct all analyses by means of simple OLS regressions. Patzer, The Physical Attractiveness Phenomena, p.17. See K. Grammer, B. Fink, A.P. M0ller and R. Thornhill, 'Darwinian Aesthetics: Sexual Selection and the Biology of Beauty', Biological Review 78/3 (2003), pp.385-407; Henss, 'Zur Beurteilerübereinstimmung bei der Einschätzung der physischen Attraktivität junger und alter Menschen'; Henss, "Spieglein, Spieglein an der Wand .Iliffe, 'A Study of Preferences in Feminine Beauty'. Henss, "Spieglein, Spieglein an der Wand ...", p.308. See J.E. Biddle and D.S. Hamermesh, 'Beauty, Productivity, and Discrimination: Lawyers' Looks and Lucre', Journal of Labor Economics 16/1 (1998), pp. 172-201; C. Davis, G. Claridge and J. Fox, 'Not Just a Pretty Face: Physical Attractiveness and Perfectionism in the Risk for Eating Disorders', International Journal of Eating Disorders 27/1 (2000), pp.67-73; D.S. Hamermesh and J.E. Biddle, 'Beauty and the Labour Market', American Economic Review 84/5 (1994), pp.1179-81; D.S. Hamermesh and A.M. Parker, Beauty in the Classroom: Professors' Pulchritude and Putative Pedagogical Productivity. NBER Working Paper 9853 (Cambridge, MA: National Bureau of Economic Research, 2006); H.N. Mocan and E. Tekin, Ugly Criminals. NBER Working Paper 12019 (Cambridge, MA: National Bureau of Economic Research, 2005), pp.8-13; P. Roszell, D. Kennedy and E. Grabb, 'Physical Attractiveness and Income Attainment among Canadians', Journal of Psychology 123/6 (1989), pp.550f.; C. Stelzer, S.M. Desmond and J.H. Price, 'Physical Attractiveness and Sexual Activity of College Students', Psychological Report 60/2 (1987), pp.567-73; D. Umberson and M. Hughes, 'The Impact of Physical Attractiveness on Achievement and Psychological Well-Being', Social Psychology Quarterly 50/3 (1987), pp.231f.; W.R. Zakahi, R.L. Duran and M. Adkins, 'Social Anxiety, Only Skin Deep? The Relationship between Ratings of Physical Attractiveness and Social Anxiety', Communication Research Reports 11/1 (1994), pp.23-31. Here, we would like to express our great thanks to Respondi AG. Without the generous support of Respondi AG, the investigation as it is presented here would not have been possible. This exclusion turned out to be excessive as there were no systematic differences between the scoring of this group of people and the other assessors. However, in doing so, the intention was to counter a potential objection that this assessor sample has a systematic bias compared to the eligible voting public. When an assessor is eligible to vote in the municipality in which the depicted mayoral candidate is running, it cannot be ruled out that the assessor already has a preconceived opinion (be it positive or negative) of that candidate. This personal appraisal could then, of course, colour the assessment of attractiveness which might lead to a distorted attractiveness score. Besides this, we may not exclude the possibility that raters being eligible to vote in another city or county of North Rhine-Westphalia have similarly distorted scores. These raters may be confronted with a picture of a candidate from a neighbouring community whom they know or they perceive as this candidate. In fact, this risk is a minor one and not relevant: we computed the attractiveness scores of all candidates excluding all raters who were eligible to vote in North Rhine-Westphalia; next, we computed the correlations with the attractiveness scores as used in our analyses and found an almost perfect correlation (Pearson's r = .99 with p < .0001, n = 264). MAGIC MAYORS 389 36. The minimum is thirty-one and the maximum is fifty-nine. The standard deviation is 6.50. 37. Henss, "Spieglein, Spieglein an der Wand ...", p.308. 38. The minimum is 0.68, the maximum 0.90, the median 0.82 and the standard deviation is 0.06. 39. See Etcoff, Survival of the Prettiest; Grammer, Signale der Liebe; M. Gründl, 'Was ist Schönheit?', in A. Hauner and E. Reichart (eds), Body talk: Der riskante Kult um Körper und Schönheit (Frankfurt a.M.: dtv 2004), pp.9-33; M. Gründl, 'Attraktivitätsforschung: Auf der Suche nach der Formel für Schönheit', in C. Gutwald and R. Zons (eds), Die Macht der Schönheit (München: Wilhelm Fink Verlag, 2007), pp.49-70; G. Rhodes and L.A. Zebrowitz (eds), Facial Attractiveness: Evolutionary, Cognitive, and Social Perspectives (Westport, CT and London: Ablex Publishing, 2002). 40. See, for an overview, J. Behnke, 'Lassen sich Signifikanztests auf Vollerhebungen anwenden? Einige Essayistische Anmerkungen', Politische Vierteljahresschrift 46/1 (2005), pp.O-l-O-15; R.A. Berk, B. Western and R.E. Weiss, 'Statistical Inference for Apparent Populations', Sociological Methodology 25 (1995), pp.421-58. 41. As politics might be considered an arena with masculine connotations, we assume that female candidates have a systematic competitive disadvantage. This theoretically links to an interaction effect between the gender of direct candidates and their physical attractiveness. According to the so-called Beauty Is Beastly Effect it can be assumed that as attractiveness increases, not only does the attribution of socially desirable personality traits rise, but so too does stereotyping according to gender-role cliches. See first and foremost M.E. Heilman and L.R. Saruwatari, 'When Beauty Is Beastly: The Effects of Appearance and Sex on Evaluation of Job Applicants for Managerial and Nonmanagerial Jobs', Organizational Behavior and Human Performance 23/3 (1979), pp.360-72; but also the following work by H. Friedman and L.A. Zebrowitz, 'The Contribution of Typical Sex Differences in Facial Maturity to Sex Role Stereotypes', Personality and Social Psychology Bulletin 18/4 (1992), pp.430-8; Barry Gillen, 'Physical Attractiveness: A Determinant of Two Types of Goodness', Personality and Social Psychology Bulletin 7/2 (1981), pp.277-81; M.E. Heilman and M.H. Stopeck, 'Being Attractive, Advantage or Disadvantage?', Organizational Behavior and Human Performance 35 (1985), pp.202-15; Rosar and Klein, 'Mein(schöner)Prof.de. Die physische Attraktivität des akademischen Lehrpersonals und ihr Einfluss auf die Ergebnisse studentischer Lehrevaluationen'; Schubert and Curran, 'Stereotyping Effects in Candidate Evaluation'; S. Sczesny, 'A Closer Look Beneath the Surface: Various Facets of the Think-Manager-Think-Male Stereotype', Sex Roles 49/7-8 (2003), pp.353-63; S. Sczesny and U. Kühnen, 'Meta-Cognitions about Biological Sex and Gender-Stereotypic Physical Appearance: Consequences for the Assessment of Leadership Competence', Personality and Social Psychology Bulletin 30/1 (2004), pp.13-21; and the critical remarks in K.E. Podratz and R.L. Dipboye, 'In Search of the "Beauty Is Beastly" Effect', Paper Presented at the 17th Annual Conference of the Society for Industrial and Organizational Psychology, Toronto, Canada, 12-14 April 2002. In the political sphere this should mean that with increasing attractiveness women are also ascribed with characteristics that are seen as counterproductive by voters in this specific context. As a consequence, this should result in outer appearance having a negative effect on female candidates' share of the vote. We have empirically tested this possibility but could not find any strong effects that would support the existence of a Beauty Is Beastly Effect. So as not to increase the complexity of the discourse unnecessarily, we have decided to refrain from discussing the Beauty Is Beastly Effect and explaining the associated empirical analyses. 42. We assume that the age of mayoral candidates does not linearly affect election chances as the voters expect a mayor to have a certain minimum degree of life experience on the one hand, but they also take into account the physical and cognitive ability of a candidate when making their decision on the other. While experience should tend to increase as age progresses, the physical and cognitive ability can be seen to decrease as age increases. In order to be able to model these counter-running effects of age appropriately, we included age in the analysis model as both a linear and a squared value. 43. We assume that within the German electorate there is resentment against people from a migrant background. Accordingly, mayoral candidates who are themselves immigrants or whose parents came from abroad should have a systematic competitive disadvantage. 44. We assume that within the German electorate there is resentment against homosexuals. Mayoral candidates who live openly as homosexuals should therefore have a systematic competitive disadvantage. 45. The academic title of doctor can signify intellectual competence. We therefore make the assumption that mayoral candidates with a doctorate have a systematic competitive advantage. 46. Noble heritage is still considered prestigious in Germany. Hence we assume that mayoral candidates who bear a title of nobility have a clear competitive advantage. 47. According to the commonly held belief of many voters, a political leader should have a suitable wife -be it because she usually undertakes representative duties, or because it signifies that the political leader leads an orderly private life. The same should also be true of a female political leader and her husband. Hence we assume that mayoral candidates who are married at the time of the election enjoy a GERMAN POLITICS competitive advantage. However, the fact that a candidate has a spouse at the time of the election is not necessarily a guarantee of an orderly private life. There are politicians who were married every time they stood for election - albeit to a different wife each time. For this reason we also take into account in the empirical analyses whether a mayoral candidate has been previously divorced at the time of the election. If this is the case, the candidate would be supposed to have a systematic competitive disadvantage. If the number of rival candidates in an election increases, the vote share of each individual standing mayoral candidate should tend to decrease. Because we assume that the potential to absorb votes is distributed very differently between candidates - and depending on their individual strategic potential vote - we incorporated the number of rival candidates logarithmically in the empirical analyses. It is a truism of political science research that incumbents have a strategic advantage over their challengers. We have therefore avoided providing a detailed explanation here of the suppositions behind it. For simplicity, a candidate's strategic potential vote was coded by us using dummy variables that indicate which party or parties support a candidate. Here, we made a distinction between the Conservative Party (CDU) as a reference category, the Liberal Party (FDP), the Conservative Party and Liberal Party (CDU and FDP), the Social Democratic Party (SPD), the Green Party (Bündnis '90/Die Grünen), the Social Democratic Party and Green Party (SPD and Bündnis '90/Die Grünen), the Socialist Party (Die Linke) and Others (extreme left wing parties, extreme right wing parties, other splinter parties, local voters' associations and independent candidates). These dummy variables summarise from the total population of our investigation the potential voters to whom candidates could successfully appeal in principle. For the objective of this study it is insignificant that these potential voters comprise both loyal party supporters and people who support the candidate's party or parties purely for reasons of their alignment on various political issues. Banducci et ai, 'Ballot Photographs as Cues in Low-Information Elections'. L. Huddy and N. Terkildsen, 'Gender Stereotypes and the Perception of Male and Female Candidates', American Journal of Political Science 37/1 (1993), pp. 119-47; L. Huddy and N. Terkildsen, 'The Consequences of Gender Stereotypes for Women Candidates at Different Levels and Types of Office', Political Research Quarterly 46/3 (1993), pp.503-25; C.E. Jones and M.L. demons, 'A Model of Racial Crossover Voting: An Assessment of the Wilder Victory', in G.A. Persons (ed.), Dilemmas of Black Politics: Issues of Leadership and Strategy (New York: Harper Collins, 1993), pp.128-46; M.S. Leeper, 'The Impact of Prejudice on Female Candidates: An Experimental Look at Voter Inference', American Politics Quarterly 19/2 (1991), pp.248-61; M.L. McDermott, '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 (1997), pp.270-83; M.L. McDermott, 'Race and Gender Cues in Low-Information Elections', Political Research Quarterly 51/'4 (1998), pp.895-918; K. Reeves, Voting Hopes or Fears? White Voters, Black Candidates and Racial Politics in America (New York: Oxford University Press, 1997); E.D. Riggle, P.M. Miller, T.G. Sheilds and M.M.S. Johnson, 'Gender Stereotypes and Decision Context in the Evaluation of Political Candidates', Women & Politics 17/3 (1997), pp.69-88; Riggle et ai, 'Basis of Political Judgements'; S. Rosenwasser and N.G. Dean, 'Gender Role and Political Office: Effects of Perceived Masculinity/Femininity of Candidate and Political Office', Psychology of Women Quarterly 13/1 (1989), pp.77-85; S. Rosenwasser and J. Seale, 'Attitudes toward a Hypothetical Male or Female Presidential Candidate - a Research Note', Political Psychology 9/4 (1988), pp.591-8; C.K. Sigelman, L. Sigelman, B.J. Walkosz and M. Nitz, 'Black Candidates, White Voters: Understanding Racial Bias in Political Perceptions', American Journal of Political Science 39/1 (1995), pp.243-65; E.R.A.N. Smith and R.L. Fox, 'The Electoral Fortunes of Women Candidates for Congress', Political Research Quarterly 54/1 (2001), pp.205-21; C. Spohn and D. Gillespie, 'Adolescents' Willingness to Vote for a Woman for President: The Effect of Gender and Race', Women & Politics 7 (1987), pp.31-49; N. Terkildsen, 'When White Voters Evaluate Black Candidates: The Processing Implications of Candidate Skin Color, Prejudice, and Self Monitoring', American Journal of Political Science 37/4 (1993), pp.1032-53. We would primarily question this view because Niclas Berggren, Henrik Jordahl and Panu Poutvaara have already proven, using Finland as an example, that the effect of candidate attractiveness is only slightly weaker for local elections as opposed to national elections: see Berggren et ai, 'The Looks of a Winner'. In addition, we also have basic doubts about the low-salience, low-information assumption in relation to the 2009 North Rhine-Westphalia mayoral elections. First, the level of information -as outlined above - was very good in the run-up to the election because of extensive media reporting, party election campaigns and state information campaigns. Voters could obtain information easily and readily compare the mayoral candidates standing in their municipality. Second, the election did not aim merely to fill low-ranking positions on the local board but to appoint a leading post in the local executive. Although this does not make mayoral elections in any way presidential elections, their significance MAGIC MAYORS 391 in our opinion can appear much greater than is the case in the minds of voters when it comes to other local elections. 54. We are, of course, aware that in such an arrangement a vote for an unpromising candidate can be rational when it is an expressive voting act. See G. Brennan and A. Hamlin, Democratic Devices and Desires (Cambridge: Cambridge University Press, 2000); G. Brennan and L. Lomasky, Democracy and Decision: The Pure Theory of Electoral Preference (Cambridge: Cambridge University Press, 1993). Nonetheless, the motives for such an act are probably, as a general rule, ideological or programmatic in nature and this would restrict the scope for the effect of physical attractiveness per se.