The rural expenditure scenario

Rural people, be it in Bangladesh or elsewhere, try to transform their given resources in various ways to shape up their living pattern.  Both income and expenditure issues become relevant as outcomes of their livelihood strategy. Households spend earned incomes not only to buy daily necessities (consumption expenditures), but also to create capital which is likely to generate income in future. This kind of expenditures is called investment. However, in tandem with increased income, levels of food intake tend to increase. But an increase in income leads to different expenditure patterns for various commodities.
There is a famous, and almost universal, law in economics, called Engel's Law, which states that demand for food increases at a relatively lower rate than the rise in income. This is the reason why the role of agriculture tends to be diminished in the process of economic development. Of various food items, again, the demand for staple grains has the lowest rate of increase followed by potato and vegetables.
If we want to have modest idea about consumption-induced linkages in the economy, we should have at least some knowledge about the daily expenditure pattern of households. With that end in view, the first task would be to know about linkages and then provide an explanation for the linkages based on field-level data.
It could be evident from available empirical works that linkages generally pass through three channels: production, consumption and labour market. Supposing that the income of a household has increased, the income effect might push household towards more food consumption. But if substitution effect gets stronger as an offshoot of consumption diversification, the household might become interested to consume more non-food crops and services. The main point is that demand for food decreases while that of non-food items increases, paripassu an increase in per capita income.
With the help of income-induced multiplier, some researchers have concluded that in rural areas of Asia, for each dollar increase in agricultural value added, there is an additional $0.50 to $1.0 value added in the rural non-farm economy (which also includes local towns). It is assumed that most of the multiplier - from two-thirds to four-fifths - originates from the consumption-induced multiplier. In the context of India, some researchers estimated that each Rs 100 increase in agricultural income generates an additional Rs 37 to Rs 54 of rural non-farm income.
Of course, the multiplier from agricultural growth is found to be stronger in regions with better infrastructure, higher population density and higher per capita agricultural income. In another exercise in Indian context, it was found that each dollar of additional agricultural income leads to an additional $0.80 of income in the region's non-farm economy as a result of the impacts of income. Researchers have adduced this to growth linkages. In another study on Bangladesh,  Mahabub Hossain has shown that the demand for non-food and service items in rural areas increases along with a rise in agricultural income.
The most recent data (2014) show that on average, roughly two-thirds of the total expenditure find their ways to food items - rice and wheat claiming about one-fourths. Non-cereal foods such as daal, vegetable oil and spices cost 20 per cent and non-crop items such as milk, meat and fish cost 18 per cent of total expenditure. The relatively more nationally representative Household Income and Expenditure Survey (HIES) dataset shows an almost similar expenditure pattern. Also, housing is a part of the major non-food expenditure sector accounting for around one-tenths of the total expenditure. Again, rural households, on an average, spend  about 5 per cent each on health, education and clothing.
When disaggregated by income strata, the bottom 40 per cent in income groups seem to spend about two-thirds of  total expenditure on food items. This is pitted against around 54 per cent by the top 10 per cent, thus, implying that the poor segment in general tend to spend more on food than the rich. Sometimes, they do it even at the cost of education, health etc. These findings are quite in tune with the existing economic theory. Second, at higher levels of income, the share of non-crop food items, such as meat, milk, egg, fish etc. outweighs that at lower levels of income. In other words, the higher is the level of income, the more is the consumption diversification with less importance attached to crop-specific food items.
However, an opposite scenario prevails in case of non-food expenditure. The bottom 40 per cent of the household appears to spend a little over one-thirds on the non-food items; the top 10 per cent spend about half of the budget on these items. Again, the bottom 40 per cent spends only half of that by top 10 per cent on education (4 per cent versus 8 per cent). By and large, the expenditure allocation seemingly shows that the poor segments in rural areas spend more on current consumption and the rich on future consumption via capital formation and investment.
The expenditure elasticity for food reflects that the elasticity of demand for food is low (relatively inelastic demand). It means a Tk 1 rise in budget leads to Tk 0.78 rise in food expenditure. However, the elasticity of expenditure with respect to education is higher. This implies that as expenditure rises by 1 per cent and educational expenses by 3.4 per cent. Elasticity of the health expenditure also indicates high sensitivity for health care facilities when income increases. Both education and health elasticity shows that rural  households are willing to spend more on human capital.
One important issue is how elasticity changes depending on distribution of income. It can be observed that food elasticity decreases consistently with income quartile; the bottom quartile has the highest elasticity while the top quartile has the lowest elasticity. This supports the previous position that as household income increases, expenditure share in food items decreases and hence, the elasticity of food also decreases. However, we find the exact opposite scenario for education sector where the elasticity is high for the top quartile group and lowest in the bottom quartile. Similar is the trend in health.
The writer is a former Professor of
 Economics at Jahangirnagar University.
abdul.bayes@brac.net