All three essays in this thesis are concerned with the interrelation of family, gender and labour market outcomes. The first paper investigates family earnings mobility between parents and sons, and parents and daughters, highlighting the role of assortative mating. The results suggest that daughters are more mobile than sons. I also find that Sweden has a higher degree of mobility compared to the U.S., and that assortative mating is an important underlying channel for earnings transmission. The difference in mobility between the two countries does not inherently depend on factors affecting the marriage match. Moreover, adult economic outcomes are more dependent on family background for those at the lower end of the earnings distribution. The second study analyses the long-run effects of an increase in family size on the 1980-2005 labour market outcomes of Swedish men and women. The decision to have (more) children is dependent on current and future labour market prospects. I use the exogenous variations in the sex composition of the first two children to overcome this endogeneity problem. My findings suggest that having an additional child has a stronger negative impact on earnings than on participation. However, mothers experience a substantial but not complete long-term recovery in earnings. The third paper illustrates the difficulty in disentangling the underlying channels of intergenerational earnings persistence using a path analysis model. On closer examination, such a model has a potential shortcoming since the covariates are correlated to other unobserved factors. The results suggest that education is the most influential mechanism in the earnings transmission process, while IQ, mental ability and BMI are of secondary importance. However, education is sensitive to the inclusion of other covariates and the order in which these are entered into the equation. / At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 2: Manuscript. Paper 3: Manuscript.

No description available.

Submitted by Murilo Esteves de Santi (murilo_es@hotmail.com) on 2016-07-12T17:47:22Z No. of bitstreams: 1 biblioteca.pdf: 575242 bytes, checksum: a061c3686ed763a0e9468ec7a8c130a0 (MD5) / Approved for entry into archive by GILSON ROCHA MIRANDA (gilson.miranda@fgv.br) on 2016-11-29T11:40:40Z (GMT) No. of bitstreams: 1 biblioteca.pdf: 575242 bytes, checksum: a061c3686ed763a0e9468ec7a8c130a0 (MD5) / Approved for entry into archive by Maria Almeida (maria.socorro@fgv.br) on 2016-12-06T13:22:08Z (GMT) No. of bitstreams: 1 biblioteca.pdf: 575242 bytes, checksum: a061c3686ed763a0e9468ec7a8c130a0 (MD5) / Made available in DSpace on 2016-12-06T13:23:59Z (GMT). No. of bitstreams: 1 biblioteca.pdf: 575242 bytes, checksum: a061c3686ed763a0e9468ec7a8c130a0 (MD5) Previous issue date: 2016 / I study the impact of the changes in the U.S. labor market that took place in the last few decades - such as the increase in the college wage premium and the reduction in the gender wage gap - on the intergenerational persistence of income, with a particular emphasis on the marriage market channel. To motivate my analysis, I document a positive cross-country correlation between intergenerational persistence of income (and education) and educational assortative mating. I then develop an overlapping generations model in which parents invest in their children's education and individuals choose whom they are going to marry, and estimate the model to fit the postwar U.S. data. My results suggest that both of these changes have affected the intergenerational earnings persistence, but that the marriage decision plays only a very small role in these results.

Submitted by Lorena Hakak Marçal (lhakak@gmail.com) on 2016-07-15T01:32:34Z No. of bitstreams: 1 main_tese.pdf: 1165233 bytes, checksum: 3c715e88460a5637db424c8ffb2ac64b (MD5) / Rejected by Letícia Monteiro de Souza (leticia.dsouza@fgv.br), reason: Prezada Lorena, Favor alterar seu trabalho de acordo com as normas ABNT: 1: O nome Getulio não tem acento. Favor alterar de todas as páginas que constam o nome incorreto. 2: Capa, contra-capa e demais sessões devem estar em português, língua oficial da Fundação. 3: A numeração das páginas não deve aparecer até a Introdução, porém devem contabilizar desde a primeira página. Além disso, a paginação deve estar a direita da página. 4: O título das sessões Agradecimentos, Abstract e Resumo devem ser em fonte de tamanho 12, em caixa alta, negritado e centralizado. Atenciosamente, Letícia 3799-3631 on 2016-07-15T11:53:44Z (GMT) / Submitted by Lorena Hakak Marçal (lhakak@gmail.com) on 2016-07-15T17:39:58Z No. of bitstreams: 1 main_tese.pdf: 1164308 bytes, checksum: c13c0208ff92eb7996f56a8420fa7c53 (MD5) / Rejected by Letícia Monteiro de Souza (leticia.dsouza@fgv.br), reason: Prezada Lorena, Favor fazer a ultima alteração em seu trabalho para que possa ser aprovado: No cabeçalho, onde se encontra 'FUNDAÇÃO GETULIO VARGAS' Favor colocar "ESCOLA DE ECONOMIA DE SÃO PAULO", pois deve permanecer em português. Atenciosamente, Letícia Monteiro 3799-3631 on 2016-07-15T18:33:29Z (GMT) / Submitted by Lorena Hakak Marçal (lhakak@gmail.com) on 2016-07-15T20:34:11Z No. of bitstreams: 1 main_tese.pdf: 1164303 bytes, checksum: 064ad0fc52a52bcb5d49ae307cd0359b (MD5) / Approved for entry into archive by Letícia Monteiro de Souza (leticia.dsouza@fgv.br) on 2016-07-15T20:35:37Z (GMT) No. of bitstreams: 1 main_tese.pdf: 1164303 bytes, checksum: 064ad0fc52a52bcb5d49ae307cd0359b (MD5) / Made available in DSpace on 2016-07-15T20:43:19Z (GMT). No. of bitstreams: 1 main_tese.pdf: 1164303 bytes, checksum: 064ad0fc52a52bcb5d49ae307cd0359b (MD5) Previous issue date: 2016-06-20 / Society has changed in the past decades raising questions to be asked by social scientists and their impacts on family units. In this thesis we aim to analyze how agents’ decisions on marriage and education can be interconnected assuming that men and women have preferences for intra-group marriage. In our framework we find that preferences for intra-group marriage can increase the proportion of men and women who decide to get married and study. We also show that empirically for Brazilian data there is a positive assortative mating between people with same traits, such as, education, religion or race. In addition, married couples that share the same religion tend to have the same level of schooling. We investigate how changes in marital sorting, educational composition and returns to education that occurred in Brazil in the last years can impact in household income inequality. We calculate counterfactual scenarios for Gini Coefficient keeping one of these three variables fixed in one year and comparing the counterfactual values with the actual one. If marriage were formed randomly, the Gini Coefficient would be lower than the actual one. Keeping the returns to education fixed in year 2014 we also show that the counterfactual Gini would be lower than the actual one. / A sociedade mudou nas últimas décadas abrindo a possibilidade para cientistas sociais estudarem essas mudanças e analisar os seus impactos na unidade familiar. Nesta tese pretendemos analisar como as decisões dos agentes com relação a decisão de casar e estudar pode estar conectado considerando que homens e mulheres têm preferências pelo casamento intragrupo. No modelo estudado encontramos que as preferências para o casamento intragrupo podem aumentar a proporção de homens e mulheres que decidem se casar e estudar. Mostramos também que empiricamente há um positive assortative mating entre pessoas com as mesmas características, tais como, educação, religião ou raça. Além disso, a probabilidade de casais casados na mesma religião aumenta a probabilidade dos casais estarem casados dentro do mesmo nível de escolaridade. Considerando as mudanças em como os casais se formam, a composição educacional e os retornos da educação que aconteceram no Brasil nos últimos anos, investiga-se os impactos dessas mudanças na desigualdade de renda dos casais. Calculamos cenários contrafactuais para o Coeficiente de Gini mantendo uma dessas três variáveis fixas em um determinado ano, comparando o contrafactual estimado com o Gini real. Se o casamento for formado aleatoriamente com relação à educação, o Coeficiente de Gini seria menor do que o real. Mantendo os retornos da educação fixos no ano de 2014 encontramos um Gini contrafactual menor do que o real.

A critical mechanism underpinning current biological diversity is the extent to which one species mates with, or avoids mating with, another. However, little is known about the factors that mediate hybridisation, especially during the initial and rarely observed stages of secondary contact when interspecific interactions have not responded to selection. In particular, whilst hybridisation is ultimately a behavioural phenomenon, the role of behaviour in mediating hybridisation and how it is influenced by environmental and circumstantial factors is rarely investigated. Recently introduced species provide us with unequalled opportunities to study these factors. In this thesis I examine the role of behavioural mechanisms, in particular male-male competition and mate choice, in mediating mating patterns between two genetically and phenotypically distinct lineages of European wall lizard (Podarcis muralis) that have come into recent secondary contact through human introductions. In Chapter Two, I investigated how sexual selection during allopatry is responsible for creating stark differences in phenotypic traits such as body size and weapon performance evident in the two lineages today, ultimately explaining the strong biases in dominance during territorial disputes between males. However, I also show that even given this asymmetry in male competitive ability, the extent to which it extrapolates into greater access to females in naturalistic, outdoor enclosures depends strongly on the spatial clustering of basking sites, a critically important resource for many ectotherms. In contrast to initial predictions suggested by asymmetries in male competition outlined in the previous chapter, in Chapter Three I show that both paternity and courtship behaviour was strongly assortative in the outdoor enclosures. Further investigation through staged experiments on olfactory mate choice, mating trials and analyses on specific behavioural data obtained in an enclosure experiment, I show that lineage based dominance actually contributes to assortative mating patterns in conjunction with weak conspecific male choice. In contrast, female choice seems to play no role in mediating the mating patterns observed between the two lineages. In Chapter Four I had the rare opportunity to examine the morphological and behavioural factors that predict why animals should hybridise in the first place, using the data obtained in the enclosure experiment above. I found that hybridisation was particularly common between small individuals of the larger lineage and large individuals of the smaller lineage; a result that corroborates the mechanisms determining the assortative patterns uncovered in Chapter Three. Additionally, hybridisation rates were particularly high in less dominant individuals, which I suggest is due to subordinate males having reduced opportunities for courting conspecific females due to male-male competition, requiring them to become less ‘choosy’ and therefore more likely to mate with heterospecifics. Finally, secondary contact cannot occur without at least one lineage coming into a new environment, and yet relatively little attention is paid to how this environmental change can affect the signals involved in intraspecific communication and mate choice. In Chapter Five I show that a change in the amount of time male lizards spend thermoregulating (a likely consequence of arriving in a new environment) significantly changes the chemical composition of their scent marks. However, whilst female lizards were able to detect these effects, they did not seem to base their mating decisions on them. Nevertheless, this result raises interesting questions about the potential function and consequences of this plasticity, and highlights the importance of considering plasticity in chemical communication in heterogeneous environments. Overall, this thesis shows the critically important role of behaviour in mediating intra- and interspecific mating patterns during recent secondary contact. In particular, it highlights how the direction and extent of hybridisation and competition are influenced by the degree to which differing morphological and behavioural phenotypes interact over a heterogeneous environment, particularly during the initial stage of secondary contact when mate choice has not had the chance to respond to the selective pressures of hybridisation.

Educational assortative mating and earnings inequality has both increased in both Europe and the United States in the last decades. As a result, educational assortative mating, or educational homogamy, has been suggested as a potential explanation for the increase in earnings inequality. According to this hypothesis increased sorting on education will lead to polarization between lower and higher-educated couples where the advantages of the latter will compound on one another and lead to increased economic inequality.   The majority of the studies to date report a non-relationship between educational assortative mating and earnings inequality, one of the exceptions being a study of Denmark. This exception has led sociologists to theorize that the impact of educational assortative mating could be especially strong in the Nordic countries. In this study I test this hypothesis by employing a novel decomposition method, the Theil-index, to answer if increases in educational assortative mating are associated with increases in earnings inequality in Sweden between 2000 and 2010, using data from the Standard of Living Survey (LNU).   The result is a non-relationship between homogamy and earnings inequality and an overall decrease in earnings inequality in the sample. The result is another null result for the hypothesis that educational homogamy leads to inequality, and points to a larger discrepancy between singles and couples than between couples. If corroborated, this decrease in earnings inequality would mean a divergence, in earnings inequality, between partnered individuals and the general population. Future studies should focus on the extent of this divergence.

Assortative Mating, the tendency for mate selection to occur on the basis of similar traits, plays an essential role in understanding the genetic variation on academic achievements and intelligence (IQ). It is an important mechanism explaining spousal concordance. We used principal component analysis (PCA) for spousal correlation. There is a significant positive correlation between spouses by the new variable PC1 (correlation coefficient=0.515, p<0.0001). We further research the genetic factor that affects IQ by using the same data. We performed a low density genome-wide association (GWA) analysis with a family-based association test to identify genetic variants that associated with intelligence as measured by WAIS full-score IQ (FSIQ). NTM at 11q25 (rs411280, p=0.000764) and NR3C2 at 4q31.23 (rs3846329, p=0.000675) were 2 novel genes that haven't been associated with IQ from other studies. This study may serve as a resource for replication in other populations and a foundation for future investigations.

En raison des forts coûts en temps et en énergie associés à chaque reproduction, les femelles ne sont généralement pas aussi disponibles que les mâles pour se reproduire. Les mâles entrent donc souvent en compétition pour accéder aux femelles disponibles. Ceci conduit à une forte sélection sexuelle chez les mâles. Un des exemples les plus frappants de compétition entre mâles peut être observé chez certaines espèces de crustacés chez qui les femelles ne sont sexuellement réceptives que pour un temps très limité. Les mâles ont donc évolué une stratégie de gardiennage précopulatoire grâce à laquelle ils monopolisent une femelle plusieurs jours avant qu’elle ne devienne réceptive. Ce comportement mâle est lui-même coûteux en temps et en énergie. En conséquence, il a été suggéré que les mâles devraient devenir sélectifs envers les femelles du fait du fort investissement que chaque reproduction représente pour eux. A l’aide d’un modèle mathématique, nous prédisons que les mâles effectuant de longs gardiennages précopulatoires devraient préférer s’apparier avec les grandes femelles plus fécondes. Toutefois, cette sélectivité devrait rester faible du fait de la forte compétition pour accéder aux femelles libres. Nous suggérons plutôt que les mâles devraient chercher à s’apparier avec des femelles de bonne qualité après s’être initialement apparié avec une femelle. Quand les mâles en couple rencontrent une femelle libre de meilleure qualité que leur propre femelle, ils devraient quitter leur femelle pour s’accoupler avec la nouvelle femelle. Contrairement à cette prédiction, nos expériences ont montré que les mâles en couple d’un crustacé amphipode Gammarus pulex ne changeaient pas systématiquement de femelle quand nous leurs proposions une femelle de meilleure qualité que leur propre femelle. Ils décidaient de changer de partenaire uniquement quand leur femelle était de mauvaise qualité, indépendamment de la qualité de la nouvelle femelle libre. D’autres expériences sont nécessaires pour comprendre le caractère adaptatif de ce comportement de changement de partenaire, seulement basé sur une partie de l’information disponible. Ces deux études soulignent la difficulté d’inférer des patrons de reproduction uniquement à partir des préférences individuelles. Dans la première étude, les mâles étaient contraints par la compétition pour accéder aux femelles libres. Dans la seconde, le processus de prise de décision des mâles conduisait à un comportement de choix apparemment sous-optimal. Ces contraintes n’ont que rarement été prises en compte malgré leur grande importance lorsqu’il s’agit de comprendre les causes comportementales d’un patron de reproduction très répandus chez les crustacés à gardiennage précopulatoire : l’homogamie pour la taille. Il a principalement été suggéré que ce patron de reproduction était issu d’une préférence mâle pour les grandes femelles associée à un avantage des grands mâles pour accéder aux femelles. Cette hypothèse n’a malgré tout reçu que peu de support empirique. A l’aide d’un modèle par simulation individu centrée, nous avons donc testé l’hypothèse selon laquelle une préférence mâle pour la distance à la mue des femelles serait à l’origine de l’homogamie pour la taille chez les crustacés à gardiennage précopulatoire. Quand les mâles préfèrent s’apparier avec des femelles qui sont strictement plus proches de la mue qu’eux, les couples formaient un patron d’homogamie pour la taille. Puisque plusieurs préférences différentes peuvent conduire à un même patron de reproduction, ce résultat souligne l’importance de considérer le processus complet de mise en couple pour étudier le lien entre les préférences individuelles et les patrons de reproduction. Les stratégies de femelles peuvent aussi jouer un rôle important dans les processus de mise en couple. Contrairement aux mâles, les femelles ont été décrites comme préférant les gardiennages courts du fait des coûts associés à la mise en couple. / Because of strong costs associated with each mating event, females are usually not as available for reproduction as males at any given time. Males are therefore in competition with each other for access to receptive females, hence leading to strong sexual selection. One textbook case of such a mating system occurs in moulting crustaceans where females can only be fertilized during a short period following their moult. This has favoured the evolution male strategies to monopolize females before their period of receptivity. Such a precopulatory mate guarding is widespread among many taxa and represents one of the most striking example of males’ competitive traits favoured by sexual selection. However, recent investigations have suggested that because males’ sexually selected traits often involve opportunity or mortality costs, males should become choosy towards females. Using a theoretical approach, we showed that males performing long lasting mate guarding should choose larger, more fecund females. However, under sequential encounter of potential mates, competition for female access decreases male choosiness before entering in precopula. We rather suggest that males should become choosy after initial pairing with a female. When encountering an unpaired female of better quality than their current female, paired males should switch partners. Contrary to our expectations, even under simultaneous encounters of two females, males did not seem to assess their relative quality. Instead they decided to change partner when their own female was of low absolute quality. This led to several cases where males forewent the possibility of increasing their fitness. Further investigations are needed to understand the adaptive significance of using only a subset of information in decision making. These two cases highlight the difficulty of inferring mating patterns from mating preferences only. In the first case, male preference was constrained by competition for access to females while in the second one, sampling processes led to apparent suboptimal mate choices. These potential constraints on decision making have rarely been acknowledge in precopulatory mate guarding crustaceans in spite of their major importance when inferring the causes of a well-known pairing pattern occurring in these species: size-assortative pairing. Size assortment among pairs has mainly been considered to come from a male directional preference for larger females associated with a large male advantage in getting access to preferred females. However, this hypothesis has received contrasted empirical support and little is known about the underlying pairing process causing size-assortative pairing. We investigated theoretically the possibility that a state-dependent male mating preference could account for size-assortative pairing. When males chose females which were exclusively closer to moult than them, assortative pairing by size arose under strong male-male competition. Because several preferences can account for a given pattern, this result emphasises the importance of considering the whole pairing process when studying the link between preferences and mate choice. Female strategies may also be of great importance during the pairing process. Contrary to males, females have been suggested to prefer short precopulatory mate guarding due to costs associated with pairing. Such a sexual conflict over guarding duration may have major effects on co-evolutionary dynamics between males and females traits. Proving its occurrence is yet challenging because empirical studies often lack a full economical survey of costs and benefits for females associated with male traits. Females benefits associated with long lasting precopulatory mate guarding have particularly been overlooked in previous studies. Here, we proposed several potential benefits for females and discuss their influence on sexual conflict over guarding duration.

Orientador: Marcus Aloizio Martinez de Aguiar / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Fisica Gleb Wataghin / Made available in DSpace on 2018-08-15T21:06:55Z (GMT). No. of bitstreams: 1 Andrade_ElizabethMachadoBaptestini_D.pdf: 4491574 bytes, checksum: 117d970a1c273ecd6ef9533aa742bb0f (MD5) Previous issue date: 2010 / Resumo: Nesse trabalho, estudamos doismecanismos de formação de espécies. No primeiro deles, consideramos um modelo espacial de especiação neutra totalmente probabilístico, sem barreiras geográficas ou interações ecológicas. A população evolui devido a influência de reprodução sexuada, mutações e recombinação. O modelo é baseado em acasalamento seletivo dependente de duas distâncias críticas, uma no espaço físico e outra no espaço dos genomas. Os vínculos introduzidos por essas duas distâncias permitem que a população se divida em grupos reprodutivamente isolados. Nossos resultados mostram que essa dinâmica gera padrões de diversidade consistentes com padrões observados na natureza, como distribuição de abundâncias do tipo log-normal, lei de potência para curvas espécie-área, taxas de especiação e extinção constantes e maior número de espécies para baixas dimensões. No segundo, nós generalizamos um modelo de especiação simpátrica baseado em competição intraespecífica, proposto por Dieckmann e Doebeli. Nesse modelo, uma população assexuada, inicialmente idêntica, evolui por seleção direcional para um fenótipo ótimo, onde a competição intraespecífica induz à seleção disruptiva. Nós mostramos que a forma das funções de competição e distribuição de recursos afetam a probabilidade de dois fenótipos coexistirem. Nós desenvolvemos um modelo analítico e simulações computacionais e comparamos os resultados de ambas abordagens / Abstract: In this work, we have studied two different mechanisms of species formation. In the first one, we considered a probabilistic spatial neutral model of speciation, without physical barriers or any kind of ecological interaction. The population evolves under the combined influences of sexual reproduction, mutation and recombination. The model is based on assortative mating and it depends on two critical distances, one in the genetic space and one in the physical space. The constraints imposed by these two distances allow the population to split in reproductively separated groups. Our results show that this kind of dynamics creates patterns of biodiversity in agreement with observed data, like lognormal distributions of species abundance, power law species-area relationships, steady speciation and extinctions rates and more species in low dimensions. In the second model, we generalized a sympatric speciation model based on intraspecific competition, proposed by Dieckmann and Doebeli. In that model, an assexual population, initially identical, evolves by directional selection to an optimal phenotype, where intraspecific competition induces disruptive selection. We show that the shape of the competition and carrying capacity kernels affects the likelihood of emergence of two coexisting phenotypes. We developed an analytical and a computational model and we compared the results of both approaches / Doutorado / Física da Matéria Condensada / Doutora em Ciências

Human mate choice is far from random. Assortative mating can be either positive (homogamy), when people prefer and choose partners with self-similar characteristics, or negative (heterogamy, complementarity), when people prefer self-dissimilar partners. Over one hundred years of research, it has been shown that people generally couple based on the principle of homogamy. This thesis seeks to address the following two goals. First, it critically reviews the current state of knowledge in positive assortative mating (in particular, empirical support, factors affecting homogamy, mechanisms of homogamy, relationship and genetic impact of homogamy, and methodological pitfalls of research). This section includes theoretical papers deal with further mechanisms of assortative mating (homogamy, imprinting-like effect, heterogamy, complementarity). Second, the thesis provides further test of assortative mating in 'ideal partners' (preferences) and actual partners, in the context of sex, sexual orientation (heterosexual and non-heterosexual), and population (Brazil and Czech Republic). Results of these studies show that the principle of homogamy is valid irrespective of sex and population. However, they find a stronger tendency for homogamy in actual partners among heterosexuals than in homosexuals, although...