The costs of acquiring parcels by condemnation are usually significantly higher than those for property acquired by negotiation, suggesting that Right-of-Way (R/W) acquisition costs may best be described by two different regression equations. This paper develops a switching regression model of acquisition cost to simultaneously predict the probability of whether a parcel will go to condemnation rather than be acquired via negotiation and the corresponding acquisition costs under these two regimes. The error terms of the selection equation and the two cost equations follow a trivariate normal distribution to reflect correlations across unobserved factors (such as a land owner's tenacity or a site's view value). When this model is calibrated using data on properties acquired across the state of Texas for transportation projects between 2008 and 2011, results suggest that R/W appraisers and staff should pay special attention to commercially used parcels in urban areas involving a partial taking with a relatively small remainder. Comparison of cost estimates between the two regimes (condemnation vs. negotiation) suggests that condemned parcels will have, on average, 78% higher acquisition costs across the 1,710 acquired properties and 51% greater price variation. These results suggest that it is much more costly to acquire a property and more difficult to accurately predict its costs if it cannot be acquired via negotiation. The application of model estimates to an example corridor highlights the value of simulation to capture all modeling uncertainties. This two-regime model is further extended to a three-category multinomial endogenous switching, allowing for differential cost estimates across negotiation-deed, administrative settlement, and condemnation contexts. A model of acquisition time -- from the agency's initial-offer date to its final possession date is also developed, to examine the effects of condemnation on acquisition duration. The results suggest that condemnation proceedings add approximately 7 to 8 months, on average, to parcel acquisitions by the Texas Department of Transportation. Taken together, such switching models for condemnation versus negotiated settlement highlight the benefits of avoiding condemnation proceedings in R/W acquisition. Estimation results illuminate the relative importance of various parcel and owner attributes, impacting the nature and cost of acquisition, and enhancing opportunities for R/W staff to identify more contentious properties and establish more reliable budget estimates. / text

No description available.

Submitted by Ana Luiza Neves de Holanda Barbosa (ana.barbosa@ipea.gov.br) on 2012-01-17T17:48:38Z No. of bitstreams: 2 tese_final_analuizaholanda.pdf: 1401652 bytes, checksum: fd6723241862cf776f8dcdc02a400af2 (MD5) tese_final_analuizaholanda.pdf: 1401652 bytes, checksum: fd6723241862cf776f8dcdc02a400af2 (MD5) / Approved for entry into archive by Andrea Virginio Machado (andrea.machado@fgv.br) on 2012-03-05T19:50:53Z (GMT) No. of bitstreams: 2 tese_final_analuizaholanda.pdf: 1401652 bytes, checksum: fd6723241862cf776f8dcdc02a400af2 (MD5) tese_final_analuizaholanda.pdf: 1401652 bytes, checksum: fd6723241862cf776f8dcdc02a400af2 (MD5) / Made available in DSpace on 2012-03-07T19:29:24Z (GMT). No. of bitstreams: 2 tese_final_analuizaholanda.pdf: 1401652 bytes, checksum: fd6723241862cf776f8dcdc02a400af2 (MD5) tese_final_analuizaholanda.pdf: 1401652 bytes, checksum: fd6723241862cf776f8dcdc02a400af2 (MD5) Previous issue date: 2011-09-12 / This study investigates the wage gap between the public and private sectors in Brazil. The analysis is carried out with 2009 microdata from the Pesquisa Nacional por Amostra de Domicílios (PNAD) and it takes account its complex sample design. The main contribution of this study is the estimation of a switching endogenous regression model that corrects the selection bias in the choice of employment sector. This model allows for the identification of some factors that determine the entrance of the individual in the public sector labor market. The public-private wage gap is calculated by gender as labor supply varies significantly between women and men. The results show that public sector wages are higher than those in the private sector. In particular, the public-private wage gap for women is higher than for men. / A presente tese engloba três artigos sobre diferencial de salários e estimação de demanda no Brasil. O primeiro artigo investiga o diferencial de salários entre os trabalhadores dos setores público e privado. A principal contribuição deste estudo é a estimação de um modelo de regressão com mudança endógena (endogenous switching regression model), que corrige o viés de seleção no processo de escolha setorial realizada pelos trabalhadores e permite a identificação de fatores determinantes na entrada do trabalhador no mercado de trabalho do setor público. O objetivo do segundo trabalho é calcular a elasticidade-preço e a elasticidade-despesa de 25 produtos alimentares das famílias residentes nas áreas rurais e urbanas do Brasil. Para tanto, foram estimados dois sistemas de equações de demanda por alimentos, um referente às famílias residentes nas áreas rurais do país e o outro sistema associado às famílias residentes nas áreas urbanas. O terceiro artigo busca testar a validade do modelo unitário para solteiros(as) e a validade do modelo de racionalidade coletiva de Browning e Chiappori (1998) para casais no Brasil. Para tanto, foi estimado um sistema de demanda do consumo brasileiro com base no modelo QUAIDS, que apresenta uma estrutura de preferências flexível o suficiente para permitir curvas de Engel quadráticas.