Quantifying the structure of food demand in China: An econometric approach

Abstract

This article examines food demand structure and its dynamics for 11 commodities in urban China. The analysis is based on household-level expenditure survey data for two cross-sectional surveys of Chinese households pertaining to food expenditure patterns during 1995 and 2003. Pre-committed components of commodity demands, that are insensitive to economic variables, are explored.We use the generalized quadratic almost ideal demand system (GQAIDS) for its empirical superiority to the generalized almost ideal demand system (GAIDS), and estimate the associated parameters via full information maximum likelihood procedure (FIML) accounting for endogeneity of total expenditures on food for home consumption (FAH). We also use quality-adjusted commodity unit values to control for quality differences resulting from commodity aggregation and food choice. Furthermore, we derive GQAIDS elasticity formulas, and estimate income elasticities without restrictions.The results partially support the hypothesis that an average Chinese household has incorporated elements of Western diet (fine grains) into traditional Chinese food diet over time. Moreover, the outcome of a simple test developed here points to possible preference changes for a majority of food staples under study.JEL classifications: Q11, Q13, Q17

Energy spending and household characteristics of floating population: Evidence from Shanghai

“…modified versions (Banks et al, 1997;Hovhannisyan and Gould, 2010). In this family of models prices play an important role.…”

Food consumption of outgoing rural migrant workers in urban area of China: a QUAIDS approach

“…Then, the chi-square (χ 2 ) value of the DWH statistic shows whether endogeneity exists in the model. Under the situation of endogeneity, the full-information maximum likelihood (FIML) (Al-Zand and Hassan, 1977;Henneberry, 2009, 2011;Hovhannisyan and Gould, 2011) and the GMM estimator (Banks et al, 1997;Abdulai and Aubert, 2004) are applied to estimate the QUAIDS model. If the error terms are not normally distributed, however, the FIML method may produce poor results (Zheng, 2008).…”

“…However, the use of the unit value may ignore the difference in the quality of commodities, thus inducing a potential measurement error (Gao, 1997;Zheng, 2008;Hovhannisyan and Gould, 2011). Therefore, a procedure for adjusted-quality prices following Cox and Wohlgenant (1986) is applied.…”

“…Among the factors affecting the food consumption demand, household income and food price are somewhat significant, while the effect of the family structure is also noticeable (Bai et al, 2010;Hovhannisyan and Gould, 2011;Zheng and Henneberry, 2011). Factors such as household ownership of a refrigerator, urban population size, and the characteristics of the geographic region also significantly affect food consumption (Huang and Rozelle, 1998;Yen et al, 2004;Gould and Villarreal, 2006;Bai et al, 2008;Wu et al, 2014).…”

A Influência Dos Preços Sobre a Demanda Domiciliar Por Frutas E Hortaliças: Uma Análise Por Classes De Renda

“…Além de considerar os preços e dispêndio, o sistema de demanda estimado também inclui outras variáveis que captam a heterogeneidade dos consumidores (Pollak;Wales, 1981;Hovhannisyan;Gould, 2011). Dessa forma, o modelo estimado é o seguinte: (Mutuc et al, 2007).…”