Demand Analysis- Segment 2

1. Types of data required for estimating elasticity

The empirical demand analysis is based on the data obtained from two main sources of statistical observations:

1. Family Budget Data. Family budget data are collected through sample surveys covering households which are representatives of different classes of people with respect to their income, family size, social class, etc., and their expenditure on different items of consumption like food, clothing, housing, power and fuel, etc., during a time period of 12 months or over is collected.

In order to study the influence of income level on the expenditure habits of the people, we carry out an experiment consisting of the following main steps:

1.                                                               i.      The first step is to select a group of households which are as homogeneous as possible w.r.t. regional environments, social and economic characteristics, family size and other factors that affect the demand, without making any reference to their family income.
2.                                                             ii.      The next step consists in regulating the family income by allotting the households to different income levels at random, randomisation being resorted to neutralise the effect of factors other than income.
3.                                                           iii.      Finally, a detailed account of the expenditure of each household during a period of 12 months on various budget items is compiled.

In demand analysis, by family budget data we mean the data collected through an experiment with the above listed three steps. Here expenditure (rather than income) is interpreted in terms of demand function.

1. Time Series Data. The statistical market data are usually the time series data relating to the prices of the commodities and their quantities bought (or sold) at that price at different points of time.

The treatment of such data is quite analogous to that of family budget data except that demand is now primarily regarded as a function of price and not of income. We have already seen that the market price of any commodity settles at a level known as the ‘equilibrium price’ (say), which is the intersection of the supply and demand curves, viz., d=f(p) and s=Ø(p). A variation in the price of a commodity over time means a shift in either or both the demand and supply curves if both the curves d=f(p), s=Ø(p) remain fixed, the market data remains more or less static and does not provide enough number of points according to their estimation. If both the demand and supply curves shift their positions then the market statistics would give us a picture of variation of demand (and supply) curves consequent upon the variations in the equilibrium price ‘p₁’ and in this case it is unlikely to trace either the supply or the demand function closely. However, if one of the two curves remain fixed and the other changes its position, then the family budget data provides a number of points on the fixed curve and hence the curve, is determined. Thus, for example, in order to determine the demand curve, we assume that it is relatively fixed while the supply curve shifts its position over the time period under consideration. This assumption is more or less valid for staple consumer goods, especially food articles.

Again the demand for any commodity does not depend only on its price but also on a number of factors such as income, the price of the substitution (i.e., price of related commodities), etc. Hence, for sound statistical analysis of demand, we should either take into account those factors explicitly or eliminate their effect on the demand and the price.

Click here for government certification in Accounting, Banking & Finance