Part Three


 THE PROJECTION OF DEMAND




 Chapter 1

 DEMAND AND INCOME

 Given the expected growth in population and in income per capital, the estimation of fu-
ture demand depends primarily on the relation assumed to exist between demand and in-
come. This is normally expressed in terms of the income elasticity of demand for the dif-
ferent products and product-groups under consideration. The statistical measurement of
these income elasticities may be carried out by one, or both, of two different methods.
Time-series data relating to income and consumption per capital and to changes in prices
over a suitable period may yield serviceable estimates both of income elasticities and of
price elasticities. Alternatively, survey data relating to consumption or expenditure by
different groups of households in a given period may be analysed to show the relation be-
tween total household income or expenditure and the demand for individual items. A single
survey however will not permit the estimation of price elasticities, since the price situa-
tion must normally be assumed to be the same for all the households taking part in it.

 In the present study, reliance has had to be placed mainly on the second or cross-sec-
tional method, since satisfactory time-series data for a sufficient number of years are
not available. Whenever possible however, the changes which have taken place in recent
years in average consumption levels (some of which are of substantial magnitude) have
been taken into account in determining the elasticity to be used for purposes of projection.

 Expenditure surveys for cost-of-living index purposes have been made in several West
Indian territories, and some estimates of income elasticities based on this material have
already been published (Ref. D. 1.). For the purpose of the present study, a number of
special analyses were made of the results of the survey carried out in Jamaica in 1958.
Use has also been made of some of the results of other surveys, but it has not been possi-
ble to analyse these in the same datail.

 The Jamaica survey was carried out by the Government Department of Statistics, and
a report based on it was issued in 1960 (Ref. D. 2). It covered 1,160 households divided
roughly according to the distribution of total population between the main urban area,
Kingston, the other main towns on the island, and the rural areas. The survey records
were made available by courtesy of the Department, but it bears no responsibility for the
use which has been made of them for this study.

 An itemised food budget, including both purchased foods and those obtained as gifts or
from domestic production, was requested from each participating household for two weeks
at different periods of the year. Information was also asked for on other principal catego-
ries of expenditure during the year, and on the earnings of members of the household.