This thesis adopts three stepwise perspectives to look at earnings inequality. It applies Czech data from two surveys, Microcensus and Living Conditions, covering the period 1988-2008, and European dataset EU-SILC 2008 and 2009 for international comparisons. The first essay "Personal Earnings Inequality" analyzes personal earnings distribution in the Czech Republic since the early transition from communism, using relative distribution method. The trend of "hollowing of the middle" was confirmed in the early transition, but this phenomenon later subsided. Earnings polarization was apparent for all sex and education subgroups between 1988 and 1996. In international comparison, earnings of men and highly educated are more homogenous than earnings of their counterparts in most analyzed countries. The second essay "Gender Wage Gap" quantifies the structure of gender wage gaps in four Central-East European countries (CEE), using the Heckman regression model and Oaxaca- Blinder decomposition. The observed gender wage gap is substantially higher in the Czech Republic and Slovakia than in Hungary and Poland. A relatively small but positive part of the observed gender wage gap can be explained by gender differences in characteristics in the Czech Republic and Slovakia, with a high contribution of job...

In this three-part essay, we explore the classical issue of the interplay between inequality and growth and the role of government in the underlying dynamics using modern macroeconomic analytical tools and and econometric methods. In the first chapter, we frame the issues within an environment of endogenous labor supply where the government serves as both a facilitator of production and a source of redistribution. Herein we model and numerically simulate the effects on inequality and growth of an expansionary fiscal policy in the harnessing of externalities emanating from productive government capital which is subject to relative congestion. The results from this assessment indicate that congestion accelerates the time-path to steady-state convergence while moderating the distributional consequences of fiscal expansions and strengthening the potential for a tradeoff between instantaneous and long-run policy outcomes. Through numerical simulations we further demonstrate the inability of the capital income tax to ensure redistribution in the long run for significantly high levels of congestion such that the sole possibility for the joint realization of economic growth and decreasing inequality resides in the deployment of a hybrid tax scheme which disproportionately strengthens the return to labor. In the second chapter, we explore the distributional properties of the Barro (1990) model of productive government spending in the presence of endogenous labor, distortionary taxes and congestion externalities. We derive an optimal tax combination and demonstrate the effects on growth and inequality which arise from its enablement in circumstances where the government share is both optimally and sub-optimally determined for varying levels of congestion. Utilizing the endogenous response of labor to capital ownership, we show that depending on the tax regime adopted, a conflict between equity and efficiency exists regardless of whether inequality is evaluated in terms of income or welfare. In the third chapter, we utilize an extensive database to establish the strength of response of poverty to changes in economic growth as being positively influenced by improving institutional indices where poverty is evaluated at the $2.00/day margin. Accordingly, we establish the possibility that the war against poverty can be fought as much by policies that promote growth as by the effectuation of structural reforms which advance healthy economic development.

This dissertation examines the macroeconomic effects of poverty and inequality. The second chapter considers the effect of poverty and subsistence consumption constraints on economic growth in a two-sector occupational choice model. I find that in the presence of risk taking, subsistence consumption constraints result in a dramatic slow down in terms of economic growth. The third chapter (joint with Shankha Chakraborty) proposes a model in which agents face endogenous mortality and direct preferences over inequality. I find that the greater the scale of relative deprivation the worse the mortality outcomes are for individuals. The fourth chapter looks at the relationship between inequality and the demand for redistribution when individuals have social status concerns. I show that under social status concerns an increase in consumption inequality results in higher taxation and lower growth. This dissertation includes unpublished coauthored material.

DINIZ , Marcelo Bentes; ARRAES, Ronaldo Albuquerque. Contribuições ao estudo da desigualdade de renda entre os Estados Brasileiros. 2005. 209f. Tese (doutorado) – Universidade Federal do Ceará, Programa de Pós-Graduação em Economia - CAEN, Fortaleza-CE, 2005 / Submitted by Mônica Correia Aquino (monicacorreiaaquino@gmail.com) on 2011-09-05T12:32:04Z No. of bitstreams: 1 Tese de Marcelo Bentes Diniz seguro 2011.pdf: 1255471 bytes, checksum: 5489c8efa5a5c84c060efe7138dbf02f (MD5) / Approved for entry into archive by Mônica Correia Aquino(monicacorreiaaquino@gmail.com) on 2011-09-05T12:32:49Z (GMT) No. of bitstreams: 1 Tese de Marcelo Bentes Diniz seguro 2011.pdf: 1255471 bytes, checksum: 5489c8efa5a5c84c060efe7138dbf02f (MD5) / Made available in DSpace on 2011-09-05T12:32:49Z (GMT). No. of bitstreams: 1 Tese de Marcelo Bentes Diniz seguro 2011.pdf: 1255471 bytes, checksum: 5489c8efa5a5c84c060efe7138dbf02f (MD5) Previous issue date: 2005 / Makes an analysis considering the household income concept aplied for brasilian states. Besides a large preliminary discussion concerning partial factors who explain inequality, including those ones assign to national literature, this thesis presents two relevant parts. The characterization of probability distribution that describe income distribution for brasilian states and a different methodology to calculate inequality and poverty indicators. Concerning this goal are present two information groups: Kernel Functions and “best adjustment” probability distributions for every state. By the way, are presents two theoretical structure, from result different econometrics framework, on the way to explain income inequality in Brazil. Thus, to catch the relationship between household income inequality and economic growth, was used first a framework based upon a simultaneous equation models applied in cross-sectional data for brazilian states with some extensions to different revenue percentiles. Otherwise, considering a theoretical model with just one equation and following panel data methodology to Random Effects and Instrumental Variables estimation. At the and was made an exercise using Quantiles Regressions to catch the selected variables effects on different groups of states. / Faz uma análise da desigualdade de renda considerando o conceito de renda domiciliar e tendo como unidade espacial os Estados brasileiros. Além de ampla discussão preliminar sobre os fatores que parcialmente explicam desigualdade, inclusive, os apontados pela literatura nacional, esta tese apresenta duas partes empíricas relevantes. A caracterização das distribuições de probabilidade que descrevem as distribuições de renda para os estados brasileiros, inclusive, apresentando diferente metodologia para cálculo de indicadores de desigualdade e pobreza. Dentro desse objetivo são apresentados dois conjuntos de informação: funções de Kernel e as distribuições de probabilidade “de melhor ajuste” as distribuições de renda de cada estado. Por seu turno, também são apresentadas duas estruturas teóricas, do qual decorrem diferentes contextos econométricos, com a finalidade de explicar as causas da desigualdade de renda no Brasil. Assim, com objetivo de captar a relação entre desigualdade e crescimento utiliza-se um modelo econométrico na estrutura de equações simultâneas para dados cross-section, com extensões para diferentes percentis da renda. Por seu turno, considerando um modelo teórico com apenas uma equação e, na estrutura de dados em painel, são realizadas estimações pelo método de Efeito Aleatório e Variáveis Instrumentais. Por fim, é elaborado um exercício com Regressões Quantílicas para captar como as variáveis selecionadas atingem diferentes grupos de Estados.

No description available.

Malaysian politics has long been dominated by tensions arising out of inter-ethnic inequality. However, economic policy in the earlier years of independence took a laissez-faire approach. Following the racial riots in 1969, there was a re-think on policy and the New Economic Policy (NEP) was promulgated in 1970. The underlying objective of the policy was to achieve national unity, which entailed improving the economic and social status of the Malay (Bumiputera) community visa-vis the non-Malays (non-Bumiputeras), especially the Chinese. The policy also attempted to transfer ownership of industries to the Malay (Bumiputera) community to develop a capitalist economy under the control of the Malay (Bumiputera) ethnic group. This study argues that, whilst the policy was successful at the outset in generating economic growth and reducing poverty, especially in the rural areas, it became obsolete even for that narrow purpose. The exclusive focus of the policy on inter-ethnic inequality made it insensitive to the problem of intra-Malay inequality. Due to the very success of NEP in the earlier years, the Malay community has become less homogeneous and cross-cutting cleavages have begun to emerge. Also, Malay tolerance of intra-Malay inequality has begun to erode. Many of Malaysia's current economic and political problems can be explained by the failure of income redistribution policy to reflect these changes within the Malay community. The argument is presented here using an analysis of the trends in income distribution. The required data are obtained from the literature and also from the Malaysian Family Life Survey (MFLS).

Using the IIASA/VID dataset of populations by age, sex and level of education, I calculate education Gini coeffients and decompose the overall degree of educational inequality into age, sex and within-group components. I analyze the relative relevance of these components for inequality reduction and investigate the distributional outcomes of education expansion. I find that, on average, equalization between males and females, younger and older cohorts as well as within these subgroups of the population has significantly contributed to declining educational inequality over the observed sample period around the globe. But the relative role of these components fluctuates in the process of education expansion. First, as improvements are initiated by enhancing the educational opportunities of the youth, the gap between cohorts widens in transition phases but vanishes thereafter. Second, gaps between sexes have been reduced but are predicted to widen again if either males or females are the first to enter higher education levels. To a lesser extent, this is also true for gaps within population subgroups which can be due to the ethnic background or the social and economic status of people. / Series: INEQ Working Paper Series

Faz uma anÃlise da desigualdade de renda considerando o conceito de renda domiciliar e tendo como unidade espacial os Estados brasileiros. AlÃm de ampla discussÃo preliminar sobre os fatores que parcialmente explicam desigualdade, inclusive, os apontados pela literatura nacional, esta tese apresenta duas partes empÃricas relevantes. A caracterizaÃÃo das distribuiÃÃes de probabilidade que descrevem as distribuiÃÃes de renda para os estados brasileiros, inclusive, apresentando diferente metodologia para cÃlculo de indicadores de desigualdade e pobreza. Dentro desse objetivo sÃo apresentados dois conjuntos de informaÃÃo: funÃÃes de Kernel e as distribuiÃÃes de probabilidade âde melhor ajusteâ as distribuiÃÃes de renda de cada estado. Por seu turno, tambÃm sÃo apresentadas duas estruturas teÃricas, do qual decorrem diferentes contextos economÃtricos, com a finalidade de explicar as causas da desigualdade de renda no Brasil. Assim, com objetivo de captar a relaÃÃo entre desigualdade e crescimento utiliza-se um modelo economÃtrico na estrutura de equaÃÃes simultÃneas para dados cross-section, com extensÃes para diferentes percentis da renda. Por seu turno, considerando um modelo teÃrico com apenas uma equaÃÃo e, na estrutura de dados em painel, sÃo realizadas estimaÃÃes pelo mÃtodo de Efeito AleatÃrio e VariÃveis Instrumentais. Por fim, Ã elaborado um exercÃcio com RegressÃes QuantÃlicas para captar como as variÃveis selecionadas atingem diferentes grupos de Estados. / Makes an analysis considering the household income concept aplied for brasilian states. Besides a large preliminary discussion concerning partial factors who explain inequality, including those ones assign to national literature, this thesis presents two relevant parts. The characterization of probability distribution that describe income distribution for brasilian states and a different methodology to calculate inequality and poverty indicators. Concerning this goal are present two information groups: Kernel Functions and âbest adjustmentâ probability distributions for every state. By the way, are presents two theoretical structure, from result different econometrics framework, on the way to explain income inequality in Brazil. Thus, to catch the relationship between household income inequality and economic growth, was used first a framework based upon a simultaneous equation models applied in cross-sectional data for brazilian states with some extensions to different revenue percentiles. Otherwise, considering a theoretical model with just one equation and following panel data methodology to Random Effects and Instrumental Variables estimation. At the and was made an exercise using Quantiles Regressions to catch the selected variables effects on different groups of states.

CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / O Brasil à notadamente marcado por elevados Ãndices de desigualdade na distribuiÃÃo de renda e altos nÃveis de pobreza, levando-nos a questionar se esses fatores possuem efeitos sobre o crescimento econÃmico. O presente trabalho tem como objetivo, verificar os efeitos da desigualdade e pobreza sobre o crescimento econÃmico no Brasil no perÃodo entre 1987 e 2002. O modelo empÃrico apresentado à estimado a partir de dados em painel, tomando como unidade de anÃlise os estados brasileiros, utilizando MÃnimos Quadrados Generalizados - MQG. O modelo proposto à estimado de forma que se possa obter a mÃdia da taxa de crescimento em cada perÃodo, e o valor inicial das variÃveis explicativas dentro desse perÃodo. Os resultados obtidos nos levaram a conclusÃo de que a desigualdade de renda, em geral, possui um efeito positivo e significativo sobre o crescimento econÃmico no Brasil, com o efeito de variÃveis como a pobreza e indigÃncia variando de acordo com as regiÃes e os perÃodos a que se considera. Concluiu-se tambÃm uma tendÃncia à convergÃncia condicional em que a taxa de crescimento do PIB per capita à relacionada inversamente a variÃvel PIB per capita inicial quando se considera 2 perÃodos de 8 anos. PorÃm as variÃveis educacionais se mostram positivos e significativos, estando de acordo com evidÃncias na literatura. / Brazil is well-known by elevated indexes of inequality distribution and high levels of poverty, which makes us to ask if those factors have effects on economic growth. This work aims to observe the effects of inequality and poverty on economic growth in Brazil during the period 1987 - 2002. The empirical model presented is estimated from a panel dataset from Brazilian States, using Generalized Least Squares - GLS. The proposed model is estimated considering different periods, in a way that is possible to obtain the mean of the growth rate from each period, and the initial value of the independent variables in each period. The results allow us to conclude that the income inequality, in general, has a positive and significant effect on economic growth in Brazil, and the effect of variables as poverty and indigence varies according to the regions and periods that are analyzed. A trend to the conditional convergence was also concluded where the tax of growth of the per capita income is related inversely the changeable initial per capita income when it considers 2 periods of 8 years. However the educational variable if show significant and positives, being in accordance with evidences in literature.

In this paper, we present evidence on the relationship between inequality and long-term development using data on different Brazilian regions. A new framework of analysis is provided in the sense that our empirical approach is developed within a constant de jure institutional environment - Brazil - accounting for possible differences in the de facto institutional environments (Brazilian regions) rooted in distinct colonial experiences within the same national territory. New inequality indicators are constructed from scratch for Brazilian municipalities in 1920 (using the Census of 1920, which, surprisingly, had thus far been ignored for such purposes). We find no significant relationship between economic (land) inequality (proxied by the Land Gini) and political concentration (proxied by the percentage of eligible voters) for Brazilian municipalities in the early twentieth century. And although our econometric analysis indicates a positive robust relationship between economic inequality and long-term development indicators for Southeastern states (São Paulo, the center of coffee production in the nineteenth and twentieth centuries and a state with a large influx of European immigrants, which became the most dynamic Brazilian region; and Minas Gerais, the gold cycle region, shaped also by cattle-farming and coffee production), we find no relationship for Pernambuco, a state in the Northeast region representative of the old agrarian structure of colonial sugar plantations; and a positive and robust relationship for Rio Grande do Sul, a Southern state with a colonial experience more similar to that of the United States and Canada. We found no evidence of a robust relationship between the percentage of eligible voters and long-term development, a surprising result in light of the results provided in development literature, but likely consistent with a politically captured system with very low levels of enfranchisement. These results are shown to hold even when controlling for proxies for structural changes that happened in this time span, namely: urbanization, industrialization, and immigration. Moreover, land inequality in 1920 is at most weakly related to contemporaneous income inequality for Minas Gerais and São Paulo, but significant for Pernambuco and Rio Grande do Sul. In other words, evidence suggests that the positive effects of inequality are associated to a particular structural organization at a specific time, in contrast to a more structural inequality, which, as exemplified by the cases of Pernambuco and Rio Grande do Sul, would have negative or no significant effects on long-term development. Finally, we find no robust relationship between the overall land Gini and long-term economic development. These results highlight the importance of the study of historical and social elements in their respective context, as the results are consistent with the picture of a rural Brazil dominated by agrarian elites within a complex institutional environment. / Este estudo apresenta novas evidências sobre a relação entre desigualdade e desenvolvimento de longo prazo a partir de dados de diferentes regiões brasileiras. A análise é realizada a partir de uma original estratégia de identificação: trabalhamos dentro de um ambiente institucional de jure constante - o Brasil - permitindo possíveis efeitos heterogêneos a partir de ambientes institucionais de facto (estados brasileiros) diferentes, resultantes de diferentes experiências coloniais. Novos indicadores de desigualdade são construídos a partir de dados primários para os municípios brasileiros em 1920 (usamos o Censo de 1920, que não foi sistematicamente utilizado para tais propósitos): o índice de Gini da distribuição de terras (entre donos de terras e considerando toda a população) e a porcentagem de potenciais eleitores. Primeiro, não encontramos uma relação significativa entre a desigualdade da distribuição de terras e a relativa concentração política para os municípios considerados no início do século XX. Segundo, e, de certo modo, surpreendente, encontramos, através de exercícios econométricos, relações entre desigualdade e desenvolvimento no longo-prazo particulares para cada conjunto de observações: (i) uma relação positiva entre desigualdade da distribuição de terras e desenvolvimento para os estados da região Sudeste, São Paulo (o centro da produção cafeeira nos séculos XIX e XX que recebeu um forte fluxo de imigrantes e que se tornou o estado brasileiro mais dinâmico) e Minas Gerais (estado particularmente influenciado pelo ciclo do ouro, moldado também pela atividade de criação e produção de café); (ii) uma ausência de relação significativa entre os indicadores de desigualdade no início do século XX e desenvolvimento contemporâneo para o estado de Pernambuco (estado da região Nordeste, representativo da antiga estrutura colonial de produção de açúcar); e (iii) uma relação negativa entre desigualdade e desenvolvimento para o Rio Grande do Sul (estado da região Sul, com colonização mais associada a países da América do Norte). Terceiro, não encontramos uma relação estatisticamente robusta entre nosso indicador de concentração política e desenvolvimento no longo-prazo. O que seria um resultado possivelmente contra-intuitivo à luz da literatura internacional, é provavelmente consistente com um sistema político capturado e níveis bastante baixos de acesso ao voto. Os resultados acima são mantidos mesmo após controlarmos para proxies de mudanças estruturais ocorridas no período, entre elas: urbanização, industrialização e imigração. Além desses resultados, apresentamos evidência de que a desigualdade da distribuição da terra em 1920 é, no máximo, fracamente relacionada à desigualdade contemporânea para Minas Gerais e São Paulo, enquanto é significante para o Rio Grande do Sul e Pernambuco. Em outras palavras, nossas evidências sugerem que os efeitos positivos da desigualdade no início do século estão associados a uma particular organização em um período histórico específico, em contraste com uma desigualdade mais estrutural, exemplificada pelos casos de Pernambuco e Rio Grande do Sul, no qual os efeitos da desigualdade são negativos ou insignificantes no desenvolvimento de longo prazo. Finalmente, não encontramos uma relação estatisticamente robusta entre o índice de Gini da distribuição de terras considerando toda a população e o desenvolvimento das regiões consideradas. Estes resultados ressaltam a importância do estudo de elementos históricos no seu respectivo contexto, uma vez que são consistentes com um Brasil rural dominado por elites agrárias em um complexo ambiente institucional.