Purpose – The purpose of this paper is to fill a knowledge gap by examining the changes in fruit and
vegetable consumption of Chinese households.
Design/methodology/approach – Using 1993 and 2001 household survey data from three selected
provinces in China, the authors estimated a quantile regression (QR) model to demonstrate how
changes of fresh fruit and vegetable consumption over time may differ across regions, and
additionally, how these changes may differ over the entire distribution.
Findings – Results show significant increases in fresh fruit consumption for all provinces; in
addition, the pattern of changes over time differs across the entire distribution. In contrast, significant
decreases of fresh vegetable consumption are evident, and results are robust across regions; however,
the disparities of fresh vegetable consumption across regions are not significant.
Research limitations/implications – The results may shed some light on the national food policy.
First, any food policy that may affect prices of fresh fruits and vegetables will likely affect households
in lower percentiles more than those in upper percentiles. In addition, based on the findings,
households in Guangdong may have a higher risk of inadequate fruit consumption. Lower level
consumption of fruits in Guangdong may be caused by its relatively high prices of fruits and perhaps
the shifting consumption pattern to a more meat-based diet as income increases.
Originality/value – There has been considerable interest in estimating food demand structure in
China due to its huge market for food products. However, little is known about the fruits and
vegetables products. In addition, most of the previous studies used the linear regression-type model for
analysis, which fails to capture the effects of the exogenous factors on the entire distribution. To fill the
knowledge gap, this paper uses a QR model with the different-in-difference method to examine the
changes in fruit and vegetable consumption of Chinese households