Estimating food demand

Past food demand equals use by definition and is determined from historical measures and correlations. To estimate long-term future food demand, populations at country level are segmented by income group and by urban versus rural (Bijl et al. 2018). Food quality is implemented by assessing food categories on the basis of end-use functions (energy, protein, luxury-taste, vitamins-minerals, dietary fibres and frying). Major food categories are animal, fruit and vegetables, luxuries, oils and oil crops, pulses and staples (cereals, roots and tubers). Historical data are used to construct correlations between (changes in) income and dietary patterns. Additional assumptions are made on food habits and trends such as lactose intolerance, veg(etari)an food and food waste.

An important economic variable is the relation between (changes in) food demand and income (growth): the income elasticity, defined as the percentage increase in expenditures on a food item F per percent increase in income I:                                                                                                                      [INSERT FORMULA]     (eqn. 1)

 is the change during the period of observation, usually one year. If the ratio of food expenditure and income in year t equals Rt and the income growth rate is α, then it can be shown that:

[INSERT FORMULA]     (eqn. 2)

The part of income spent on food will decline if ε< 1. For staple crops, one can expect ε< 1 when incomes rise from low to medium. For luxury items, ε> 1 on such a path. The price-elasticity p is defined in similar fashion:

(eqn. 3)

with p the price of the food commodity. Because it is assumed that amounts decrease for increasing price, it is defined with a minus sign. Normally, the fraction of income spent on a food item will tend to decline with increasing price and  will be in the order of 1.

If the elasticity is less than one, the product for which this happens is considered a necessity. Examples are potatoes and other starchy staple food items. If the elasticity exceeds one, it is called a luxury good. Income elasticities and similarly defined price-elasticities are estimated from statistical data.

Statistical data show that food consumption measured as the fraction of income spent on food tends to decline with a rise in income, in other words: ε< 1. Already in 1895, the Belgian economist Engel analysed the cost of living of worker families and observed that food expenditures as a proportion of income decreased with increasing household income. In Belgium, for instance, the share of food expenditures in total household expenditures was around 60-65% between 1860 and 1920, then declined slowly to 50% by 1950 and dropped precipitously to less than 18% by 1980 (Swinnen et al. 2001). Economists speak about Engel’s law, which is confirmed by more recent trends in emerging economies. Besides time-series, it can be observed in cross-household data: higher income declines of the population have higher expenditures on food but as a fraction of income these are much lower than for lower income deciles  (Sabates et al. 2001; Figure 1).

[INSERT FIGURE 1]

Figure 1. Fraction of income spent on food and monthly expenditures on food for three South-American countries for 10 income deciles. Data from Sabates et al. (2001).

Literature

Sabates, R, B. Gould and H. Villarreal (2001). Food policy 26(2001)571-586

Swinnen , J., A. Banerjee and H. de Gorter (2001). Economic development, institutional change, and the political economy of agricultural protection – An econometric study of Belgium since the 19th century. Agricultural Economics 26(2001)25-43