The Distribution of Public Expenditure: The Case of Health Care

Abstract

In contrast to the amount of attention paid recently to the distribution of income and wealth, very little work has been done in Britain on the distribution of public expenditure. The gap is particularly wide in the case of the distribution of expenditure conferring such as that on the public provision of education or health care. There are a number of North American studies, particularly on education (see, for example, Gillespie, 1965; Hansen and Weisbrod, 1969; Pechman, 1970; and Judy, 1970). However the only recent British work is the annual estimates of tax and services produced by the Central Statistical Office (see Harris, 1977); and these do not pretend to be an elaborate study of the distribution of expenditure on benefits-in-kind, the distribution being simply assumed to be equal for individuals of the same age and sex. This omission is a serious one. Since expenditure on benefits-in-kind is an important part of the so-called social wage, and since the latter has become a focus of political attention (being used, for example, as a partial justification for the holding down of private wages), it seems important to establish exactly how such expenditure is distributed between different groups in the population. This paper is an attempt to do so for one element of the wage: the distribution of public expenditure on the National Health Service (NHS). The paper relates public expenditure on the Health Service to the incidence of illness by socioeconomic group, in order to see whether the position in society where an individual is placed affects the amount of medical treatment he receives when he is ill. It is divided into five parts. In Section I the distribution of morbidity (or ill health) among different groups in the population is discussed. In Section II an attempt is made to estimate the costs of the health service facilities used by different groups, and to relate this to the morbidity of the groups. The calculations are very tentative, and should not be regarded as in any way definitive. In Section III some possible explanations for the results are discussed. In Section IV some of the implications of the results are explored, and the conclusions are briefly summarized in Section V. The paper is a substantially revised version of evidence submitted to the Royal Commission on the Health Service (Le Grand, 1976). Access to more data has permitted the use of a method of calculation superior to that used in the Royal Commission evidence. Reassuringly, the change in the methods of calculation has not greatly changed the results. For reasons described below the basic source for the data used in the calculations is the health section of the General Household Survey undertaken by the Office of Population Censuses and Surveys (henceforth OPCS),