from the effects of the 1943 hurricane was still incomplete. It seems necessary to suppose
that consumption levels were abnormally low at this period and that the short-fall in sup-
plies was not made good by imports. In predicting future levels of demand, therefore, the
experience of the 1950's may perhaps be disregarded; at all events it seems impossible to
ignore the survey evidence as to income elasticity so far as future years are concerned.
(It will however be observed from Table 3.4.i. that Jamaican consumption of oils and fats
was still much lower in 1958 than that of other territories.)

 The consumption of cereal products appears (both from the deflated expenditure estimates
and from the tonnage statistics of total supplies) to have increased considerably more than
would have been expected on the basis of the survey evidence as to income elasticity. To
give a close fit to the deflated expenditure estimates the income elasticity in 1958 would
have to be raised to rather more than unity; the supply statistics indicate an increase of
average consumption per capital between 1950 and 1958 of about 45 percent, which if wholly
due to the rise in income levels would suggest an income elasticity in 1958 of about 0.75,
two-thirds greater than the estimate derived from the survey of about 0.45. An elasticity
significantly exceeding the latter figure however would be impossible to reconcile with the
general evidence relating to income elasticities for cereals in other underdeveloped areas,
which do not seem to exceed 0.5 even in the low-income countries of Asia and Africa where
standards of living and of food consumption are undoubtedly lower in many cases than in the
West Indies. 1/ Mainly for this reason, therefore, it has seemed desirable, in spite of this
conflict of evidence, to adhere to the survey-based estimate of 0.45 as the average income
elasticity in terms of quantity for the cereals group as a whole in Jamaica.

 By contrast with cereals, the average consumption of roots and starchy vegetables,
which in countries such as Jamaica represent an important alternative to cereals as sour-
ces of carbohydrates, seems to have been falling. However, the supply statistics for this
group, which includes many products grown by rural households for their own use, are in
the nature of things somewhat conjectural. Substitution in response to price changes may
account for part of the change in the relative consumption of root vegetables and of cereals,
since the average prices of the former appear to have risen somewhat during the 1950's in
relation to those of cereals.

 l:ven in the absence of price changes, however, a rise in the consumption of purchased
cereals such as bread and rice at the expense of the subsistence carbohydrate foods would
be one of the consequences to be expected from a shift in consumption habits towards a
diet of the urban type and away from that characteristic of the traditional rural mode of
life. It is evident from the survey analysis that the consumption of roots and starchy ve-
getables is considerably lower in Kingston households than in rural households. This lower
consumption in Kingston is important even though it conflicts with the difference in consump-
tion levels which would have been predicted from the estimates of elasticity within both
the rural and urban groups of households (the latter being in every case positive, as they
are for all other product groups). In interpreting the survey estimates of elasticity how-
ever, it must be remembered that they relate to groups of food that include a number of


i/ See "Agricultural Products, Projections for 1970", Food and Agriculture Organization,
 1962; Annex, Table M4.