Income and bean consumption patterns in Zambia

Abstract

The literature shows that increases in incomes lead to changes in the allocation of income or expenditure shares to different food products. The purpose of this thesis is to identify the effect of income on expenditure share allocations among different food groups. The study was particularly interested in beans and how changes in incomes affect the share of bean expenditures. We used data from the 2010 Zambia Living Conditions Monitoring Survey (LCMS). The LCMS covers the whole country and provides segmentation of the respondents, across the region and rural versus urban. It also provides detailed information on the income and expenditure distributions of respondent households. This allowed for the achievement of the overall objective of this thesis: understanding how beans and other food products responded to income changes as well as other demographic and socio-economic variables. The food share is the proportion of total household income that was allocated to food. The results show that food averages about 40% of income but varied significantly across the four income groups. It was 92% for those earning less than ZMW300 per month and 37% for those earning between ZMW300 and ZMW750 per month. It was down to 22.6% for those earning between ZMW750 and ZMW2.1 million per month had a food share of total income of only 10.8%, similar to the average U.S. consumer. These averages were found to be statistically different across the income groups. We found that Zambians allocated about 40% of their food expenditure to cereals compared to 5% to pulses and 3.5% to beans. They allocated a higher proportion of their food expenditure to fruits and vegetables than to beans and/or to pulses. This shows that legumes are very low on the food hierarchy in Zambia. However, across income categories, it was found that consumers in the second income group (ZMW300 and ZMW750 per month) allocated the most of their food expenditure to beans, about 3.9%, while those in the highest income group (ZMW750 and ZMW2.1 million per month ) allocated the least, about 3%. The biggest influencing demographic factor for pulses and beans’ shares of food expenditure was locale, with urban consumers having about 1.1 and 0.8 percentage points higher share of food expenditures allocated to beans than rural consumers. The respective t-values were 15.58 and 16.96. All the demographic and socio-economic variables were statistically significant at or below the 5% level. There was no difference between the allocation of people in the highest income group and those in the lowest income group. The results suggest that if the long-term objective is to reap the nutritional benefits of beans, there may be value in focusing on two principal policy variables: education and income enhancement. However, because education is correlated with income, the benefits of undertaking this policy initiative would more than benefit the bean consumption. It should unleash across the economy a more productive workforce that understands the health benefits of its food choices.