Agenda de Seminários Acadêmicos

Agenda Qualificações

Agenda Pré-Defesas

Agenda Defesas

  • Marcos Hitoshi Endo Open or Close

    Área: Economia - Área: Economia Aplicada

    Curso: Doutorado
    Data: 11/12/2020, às 08h30 (horário de Brasília)
    Local: Devido à contingência da COVID-19, a banca ocorrerá totalmente online e será transmitida publicamente.

    Link da transmissão ao vivo: https://stream.meet.google.com/stream/0b25e435-546e-49c4-a08c-38a76963f02f

    Obs: Apenas usuários com contas "@usp" podem acessar a transmissão. Caso usuários que não possuam estas contas queiram assistir a sessão, deverão nos enviar (com antecedência) e-mail para "posgrad@fearp.usp.br" e enviaremos as instruções.

    Título: Essays in consumption predictability
    Autor: Marcos Hitoshi Endo

    Banca: Prof(a). Dr(a). Fábio Augusto Reis Gomes (Presidente)

    Prof(a). Dr(a). Gian Paulo Soave (Universidade Federal da Bahia - UFBA)

    Prof(a). Dr(a). Eurilton Alves Araújo Júnior (Banco Central (BACEN))

    Prof(a). Dr(a). Paulo Rogério Faustino de Matos (Universidade Federal do Ceará - UFCE)

    Prof(a). Dr(a). Luciano Nakabashi (FEA-RP)

    Prof(a). Dr(a). Alex Luiz Ferreira (FEA-RP)

    Resumo: 

    This doctoral dissertation is composed of three self-contained essays that study the relationship between the aggregate consumption growth and the main macroeconomic variables investigated by the consumption literature. The first essay examines the predictability of the Brazilian aggregate consumption growth taking into account four factors: intertemporal substitution, rule-of-thumb behavior, liquidity constraints and habit formation. The results suggest that estimations are plagued by weak instruments, thus we use valid (robust) confidence intervals that suggest that income growth rate and/or credit growth rate are the relevant predictors. The second essay studies whether consumption responds symmetrically or asymmetrically to predictable income. Instead of finding evidence for myopia or liquidity constraints, the findings support the “perverse asymmetry” hypothesis raised by Shea (1995a). The third essay studies the consumption predictability in the US. Using structural break tests and robust confidence intervals to deal with the parameter stability and weak instruments, respectively, the results show evidence that the relevant predictability factors have changed over time.