IN HONG KONG
E. R. LIM *
Introduction The objectives of consumer demand analysis are first, to isolate a few major variables from the many and complex factors influencing consumer behaviour, and second, to verify empirically that this relatively small number of parameters gives a reasonable explanation of a wide range of observations on consumer behaviour. The quantitative knowledge thus obtained is necessary for the study of economic development and for the formulation of economic policy. In addition they are a necessary input to the evolution of consumer theory. The data for the present study come from the household expenditure survey conducted by the Hong Kong Department of Commerce and Industry from 1963 to 1964. The survey covered almost three thousand households in all census districts of Hong Kong and included information on monthly
household expenditure on more than 90 items of commodities and services as well as information on household particulars such as size and composition of household, occupation, employment status and industrial group of head of household, etc. Description, methodology and summary results of the survey are published in The Household Expenditure Survey, 1963/64 and the Consumer Price Index.2 The 90-odd items of expenditure in the survey are classified into eight groups for this study: foodstuffs (including meals away from home but excluding alcoholic drink); housing; fuel and light; clothing and footwear; durable goods; miscellaneous goods (including alcoholic drink and tobacco); transport * E.R. Lim is Lecturer in Economics, University of Hong Kong.
1 This paper presents the preliminary results of the author’s work on the household sector in Hong Kong. Transcription of data and computations on the IBM-360 were financed by a research grant from the University of Hong Kong. The co-operation of the Commerce and Industry Department, Hong Kong, in making the data available is also gratefully acknowledged. 2 Statistics Branch, Commerce and Industry Department, Hong Kong, The Household Expenditure Survey, 1963/64 and the Consumer Price Index, Government Printer, Hong Kong, 1965.
and vehicles ; and services. No grouping of observations was attempted, and individual data of 2,762 households were used. Choice of Variables
The major concern of this study is to isolate and quantify the effects of household income1 on the eight expenditure items. Specifically, we are interested in estimating the form and parameters of the income-consumption equations (commonly known as Engel curves) for these items and to analyse the effects of other household characteristics such as occupation and employment status of head of household on the parameters of these equations. The direct product is a set of income elasticities for each category of expenditures and, if necessary, for each type of household. The study of this income-consumption relationship from cross sectional The assumption is that data such as we have involves a crucial assumption. differences in the consumption behaviour of the many households observed at a point of time represent also a “representative” household’s differences in behaviour, as its determining parameters change accordingly. In other words, in analysing the income-consumption behaviour, we are assuming that the household with $200 income will behave as another household with a $400 income, if and when its own income doubles. This assumption is not as arbitrary as it may seem at first sight. To be able to formulate any laws of consumer behaviour at all, we must necessarily It assume that there are certain predictable elements in consumer behaviour. is these predictable elements which we are interested in quantifying, and if they exist at all, they must be evident in the measurable differences of behaviour now between the household receiving $200 and the household receiving $400...