Seki Asano

Estimation of the Brazilian Consumer Demand System

In this study we estimate the Brazilian consumer demand system through family expenditure data, which cover all consumption categories. The model is estimated from family-level expenditures on seven consumption categories, and a new set of regional cost-of-living indexes. The sources for expenditures are the national expenditure surveys conducted in 1986/87 and 1995/96, which collected data from 11 metropolitan areas. Corresponding price indexes were constructed from detailed commodity prices, also from each metropolitan area. The salient features of our study are: a) price variations come from both time and regional differences, which allows us to estimate price elasticities with high precision; b) we have large variations in income (total expenditures) which is rarely available in aggregated data; and c) we can control for time specific factors by exploiting the panel structure of the data set. Contrary to the rule of thumb in empirical studies of demand systems, the estimation results display consistency with demand theory, and elasticity estimates are close to economic common sense. The estimated system will serve as a microeconomic basis for evaluating various policy-related issues. Neste estudo, estimamos o sistema de demanda de consumo brasileiro atraves de dados de dispendios das familias, que cobrem todas as categorias de consumo, e de um novo conjunto de indices regionais de custo de vida. As fontes de dados de dispendios sao as duas Pesquisas de Orcamentos Familiares (POF) do IBGE, realizadas em 1986/87 e 1995/96, que coletaram dados das nove areas metropolitanas, Distrito Federal e municipio de Goiânia, e que serviram de base para as atualizacoes do Sistema Nacional de Indices de Precos ao Consumidor. Os indices de precos correspondentes foram construidos a partir de precos detalhados de bens e servicos, tornados publicos pelo IBGE para cada uma das regioes pesquisadas. Ate onde sabemos, este e o primeiro estudo desse genero com base nas POFs. O trabalho destaca-se da literatura existente por: a) contar com variacoes de precos tanto ao longo do tempo como entre regioes, o que nos permite estimar elasticidades de precos com alta precisao; b) observar grandes variacoes na renda (dispendio total), o que e raramente disponivel em dados agregados; e c) poder usar controles para fatores especificos de tempo, explorando a estrutura de painel do banco de dados. Ao contrario do que costuma acontecer em estudos empiricos de sistemas de demanda, os resultados da estimacao mostram consistencia com a teoria da demanda, e as elasticidades estimadas sao proximas ao esperado pelo senso economico comum. O sistema estimado servira como base microeconomica para avaliar varias questoes relacionadas a politica economica.

Green revolution and labour demand in rice farming: The case of central Luzon, 1966–90

While the adoption of modern rice varieties (MVs) increased labour requirements for crop care and other activities, it might have induced the adoption of labour‐saving technologies. Thus, whether the net impact of MV adoption on labour demand is significantly positive is still debated in the green revolution literature. Using farm‐level survey data collected in Central Luzon from 1966 to 1990, this study estimated simultaneously the labour‐saving technology adoption and the labour demand functions. We found, among other things, that MVs increased labour demand significantly without inducing the widespread adoption of labour‐saving technologies.

Joint Allocation of Leisure and Consumption Commodities: A Japanese Extended Consumer Demand System 1979—90

In this study we estimate the parameters of a household expenditure function which includes joint choice of leisure and consumption commodities in scope without a separability assumption. We have used Japanese prices, wage rate, labour supply, and expenditure data on ten commodity groups, collected from 47 cities over 12 years. This data set has the advantage that separate observations are available for each data point for all the variables. We employed the AI demand system, for estimation. Controlling for time-specific effects, the result implied a definite rejection of the weak separability of labour supply and commodity choice, and non-rejection of the homogeneity and symmetry restrictions on the demand system. All the own-price elasticities are significantly negative, and both substitutes and complements are observed across commodity groups. As for the negativity, all but one of the eigenvalues of the substitution matrix are negative. The result as a whole showed consistency with demand theory. The estimated compensated labour supply elasticity is 0.39, which is in reasonable agreement with the previous studies.

Testing the constancy of relative risk aversion: An analysis of japanese household financial asset data

Abstract The paper examined the constancy of relative risk aversion (RRA) with respect to total asset from Japanese individual asset data. The paper paid attention to the fact that a substantial portion of households do not hold any risky assets. Since the OLS yields a biased estimate, we modified the estimation procedure that derives the relationship between RRA and a discrete choice of holding risky assets. The result suggests that, contrary to our speculation, constant RRA is not consistent with the data, and that there is a moderate decreasing tendency linked to total asset but that its rate of change tapers off as total asset increases.

Optimal commodity taxes for Brazil based on AIDS preferences

O presente estudo tem como objetivo calcular a estrutura otima da tributacao sobre o consumo no Brasil. As simulacoes baseiam-se no sistema de demanda Almost Ideal Demand System estimados com dados da POF/IBGE. O trade off entre equidade e eficiencia e incorporado no modelo com a introducao de um coeficiente de aversao a desigualdade social na funcao de bem-estar social, sujeita a restricao de receita do governo. Alem da hipotese de que o unico instrumento de politica tributaria disponivel seja a tributacao sobre o consumo, nosso modelo tambem admite a concessao de uma transferencia uniforme lump sum de renda per capita do governo para todos os agentes economicos. Os resultados mostram que a estrutura tributaria sobre o consumo caracteriza-se pela seletividade das aliquotas, que se torna mais significativa para valores altos do parâmetro de aversao a desigualdade. Este resultado e revertido para uma estrutura tributaria regressiva com a introducao de uma transferencia uniforme lump sum de renda do governo para todos os agentes no modelo. Quando tetos bastante restritivos sao impostos aos valores das transferencias, o primeiro resultado e restabelecido. Com relacao a forma escolhida da preferencia do consumidor, nossos resultados tambem mostram relativa robustez com os obtidos em Siqueira (1998).

Distributional effects of optimal commodity taxes combined with minimum income programs in Brazil

Commodity taxes play an important role in Brazil and raise around 60% of the total tax revenue. This heavy reliance renders commodity taxation one of the main tools available to the government for collecting revenue and securing redistribution. In fact, Brazilian income inequity is one of the highest in the world: the wealthiest 1% of population, equivalent to 1.6 million people, earn together as much as the 50% poorest, around 80 million. The purpose of this paper is a partial equilibrium numerical micro-simulation of the distributional effects of optimal commodity taxation combined with minimum income transfers made by the government to households. The approach used to measure households welfare is a money metric indirect utility or equivalent income [King (1983)], obtained from an Almost Ideal Demand System set of parameter estimates. We plug it into the equivalent variation formula to evaluate the equity effects specified in terms of the equivalent income. The data source is a 1995-1996 national household expenditure survey, though estimated parameters come from a sample comprising a 1987-1988 wave as well. We find that our proposed minimum income programs combined with selectiveness in commodity tax structure would be useful as redistribution income instrument among households in Brazil. These results can provide some valuable contribution in the context of the increasing discussion about minimum income programs in Brazil associated with demographic characteristics such as education and family size. Impostos sobre o consumo tem um importante papel no Brasil e arrecadam cerca de 60% da receita tributaria total. Essa forte dependencia faz com que a tributacao sobre o consumo seja um dos principais instrumentos distributivos e de arrecadacao do governo na receita tributaria. De fato, a desigualdade de renda brasileira e uma das mais altas do mundo: o 1% mais rico da populacao (o equivalente a 1,6 milhao de pessoas) ganha uma renda igual a dos 50% mais pobres; estes ultimos representam cerca de 80 milhoes de pessoas. O objetivo deste estudo e realizar uma simulacao dos efeitos distributivos da tributacao otima sobre o consumo associada a programas de transferencia de renda minima feitos pelo governo. O arcabouco utilizado para medir o bem-estar dos agentes economicos e a money metric indirect utility ou renda equivalente [King (1983)], com base nos parâmetros do sistema de demanda quase ideal (Almost Ideal Demand System). Nos adotamos a medida de variacao equivalente, especificada em termos de renda equivalente, para avaliar os efeitos de equidade. A fonte de dados deste estudo e obtida na Pesquisa de Orcamentos Familiares (POF) do periodo de 1995-1996, embora a estimacao dos parâmetros de demanda tambem tenha como base a POF de 1987-1988. Os resultados mostram que os programas de renda minima, combinados com seletividade na estrutura de tributacao sobre o consumo, podem ser muito uteis como instrumentos de redistribuicao de renda no Brasil. Os resultados apresentados podem ser de valiosa contribuicao, principalmente no contexto da crescente discussao sobre programas de renda minima no Brasil associados com caracteristicas demograficas, tais como educacao e estrutura familiar.

Efficient and Equitable Commodity Taxation: Micro-Simulations Based on an Estimated Brazilian Consumer Demand System

In this study we calculate the optimal commodity tax structure for Brazil. The micro-simulations are based on a complete demand system estimated with a flexible functional form [Almost Ideal Demand System, by Deaton and Muellbauer (1980)]. The data source is a 1995/96 national household budget survey. Preference parameters’ estimates are consistent with microeconomic demand theory and allow for a highly accurate optimal commodity tax simulation. The model features the maximization of a social welfare function, subject to a balanced government budget requirement. It is assumed that the only tax policy instrument available to the government is consumption goods’ and services’ taxation. The trade-off between equity and efficiency is taken into account by introducing the government’s aversion to inequality into the social welfare function. We also extend our analysis by allowing for a uniform per capita lumpsum payment to be made by the government to all households. Our results show that the commodity tax structure is characterized by selective tax rates. More specifically, we found that the commodities in which the lower household’s expenditure classes spend most, such as food and housing, should be subsidized. As expected, the degree of selectivity is more significant for higher inequality aversion parameters. Moreover, as the tax revenue goal increases, so do all commodity tax rates. By introducing a uniform lump-sum transfer, however, this result is reversed when the central planner’s degree of aversion to inequality is high enough. On the other hand, poll transfer levels are unreasonably high; when we cap them with a binding ceiling, the former pattern is restored. We believe that our empirical findings provide a valuable contribution for the current tax policy debate in Brazil, where distributive goals have a great importance in the agenda. O presente estudo tem como objetivo calcular a estrutura otima da taxacao de bens e servicos no Brasil. As simulacoes baseiam-se em um sistema de demanda de consumo brasileiro estimado com uma forma funcional flexivel [Almost Ideal Demand System, de Deaton e Muellbauer (1980)]. As fontes de dados sao as duas Pesquisas de Orcamento Familiar (POF) do IBGE, realizadas nos periodos 1986/87 e 1995/96, que coletaram dados das nove areas metropolitanas, do Distrito Federal e do municipio de Goiânia. O modelo caracteriza-se pela maximizacao de uma funcao de bem-estar social, sujeita a restricao de receita do governo. O trade-off entre equidade e eficiencia e considerado pela introducao de um parâmetro que indica o grau de aversao a desigualdade na funcao de bem-estar social. Alem da hipotese de que o unico instrumento de politica tributaria do governo seja a taxacao de bens e servicos, o modelo tambem admite a concessao de uma transferencia uniforme lump sum de renda per capita do governo para todos os agentes economicos. Os resultados mostram que a estrutura de taxacao de bens e servicos caracterizase pela seletividade das aliquotas e, em particular, que os bens que as classes de renda inferiores gastam mais, tais como alimentacao e habitacao, deveriam ser subsidiados. Como esperado, o grau de seletividade das aliquotas e mais significativo para valores altos do parâmetro de aversao a desigualdade. Esse resultado e revertido para uma estrutura de Ramsey tax quando e introduzida uma transferencia uniforme lump sum. Por outro lado, quando impomos tetos bastante restritivos aos valores das transferencias lump sum (que foram encontrados extremamente altos na estrutura otima irrestrita), o primeiro resultado “anti- Ramsey” e restabelecido. O calculo do imposto otimo pode ser de contribuicao valiosa para o atual debate da politica tributaria no Brasil onde, entre as prioridades, se encontram os objetivos redistributivos.

Equity in the Income Tax Rate Structure: Measurement on the Income Elasticity of Marginal Utility and Its Application

What is considered equitable when it comes to increase in tax payments as income increases? In spite of the importance of “vertical equity in taxation”, economic theory has not been successful in proving its theoretical foundation regarding how the exact amount of increase of tax payment can be assessed in practice while maintaining the vertical equity among different income groups.