No abstract

No abstract

Single-equation studies of the demand for specific commoditygroups represent the oldest and most extensive branch of applied economics. Among commodities, tobacco is of particular interest on both economic and econometric grounds : estimates of the elasticities of demand with respect to income and price are of practical significance to the government, given the U.K. tax structure, and to firms engaged in the industry, in view of its highly concentrated character; and the imposition of a specific tax a t a high and frequently varied rate reduces the identification problem.1 I t is not surprising, therefore, that this good has attracted the attention of previous investigators.* Although the latter have left unsettled some important theoretical questions, which will be considered below, the principal objective of this paper is to examine the impact on demand of a noneconomic factor, viz. the widespread publicity given in recent years to the contribution of smoking to disease. While the medical profession has long been aware of the dangers of smoking, the concern of the general public does not appear to have been aroused until the publication in 1962 of a report on the subject by the Royal College of Physicians. Evidence of the deleterious effects of tobacco has continued to accumulate sincc then, and oficial measures have been taken to discourage smoking, notably by a ban on television advertising of cigarettes. Accordingly, the first quarter of 1962 will lie treated as tlie dividing line in the tests for a change in tastes. 7'he Foriii o j I k i i i m d . Pwzctioit the gcnc!ral formAs in prwious studies, the demand function is postulated to be of(1) *i a m indebted to D. J. Coppock, J. Johnston, 1). B. \V. 1,aidlcr and J . hl.Parkin for their invaluable comments on an earlier draft of this papw. The usual disclninicr applies.'In tlie c e t o i s paribus CBSC of competitive theory, the imposition of a specific tax would generally causr the price paid by consumers and the price received by producers t o movc in opposite tlircctioris; thus thcre could br 110 question of the retail price acting as a proxy f(Jr the price relevant t o the supply function. In this case the argument is niuch less conclusive: rough estimates indicate that on an annual basis thcrc is an almost perfect linrar correlation bctwccn the cwv-and ex-lax prices of cigarettes; and even the impact effect of six out of thc cight postwar changcs in the tobacco duty was to raise both prices. (Both the exceptions vcciirred in 1968). Nevertheless, the argument that a t least some price increases were exogenous to the demand function appears tenable, and represents an important addition t o the ease for identification.

Corporate retentions have a well-determined effect on consumers' expenditure, which cannot be explained by an impact of retentions on capital gains and thence on household wealth; on the contrary, increases in retentions are associated with subsequent capital losses. The timing of the relationship between retained profits and expenditure is consistent with direct corporate purchases of on-the-job consumption. Copyright � 2008 The Author.

Corporate retained profits have well-determined effects on both total and non-durable consumers' expenditure, with coefficients which do not differ significantly from those on disposable income in a standard consumption function; the effects also appear in a life-cycle model which excludes disposable income. Retentions convey no useful information about future values of the other arguments of either consumption function. The implications of an alternative explanatory hypothesis, of a discrepancy between the principles and practice of national accounting, are not rejected.