The Economics of Welfare

Pigou, Arthur C.
Display paragraphs in this book containing:
First Pub. Date
London: Macmillan and Co.
Pub. Date
4th edition.
23 of 73

Part II, Chapter VIII


§ 1. WE have now to introduce a new conception, that of the relative variations of demand in different parts of the industrial field. If the amount of any sort of productive resource demanded at a given price at all points collectively is constant, and the amounts demanded at the several points individually are variable, the relative variation of demand between two periods, say, between two successive years, may be measured by the sum of the excesses of the amounts demanded in the second year, at those points where there are excesses, over the amounts demanded in the previous year at the same points. If the demand for the productive resource at all points collectively is not constant, this variation may be measured, either by the sum of the excesses of the amounts demanded in the second year over the amounts demanded in the previous year at the same points, or by the sum of the deficiencies of the amounts so demanded over the corresponding amounts of the previous year, according as the one or the other of these sums is the smaller.


§ 2. On the basis of this description, it can readily be shown that the influence of impediments to movement, in causing departures from equality of returns, is, in general, greater, the greater are the relative variations of demand in the sense just explained. For let attention be concentrated upon those impediments, which are not adequate to prevent returns from being equalised in "the long run," but suffice to prevent the movements required to equalise them immediately. If the various parts of the industrial field are fluctuating relatively to one another, impediments of this order will keep returns always unequal. Mill's illustration from wave movements on the ocean is wholly apposite. Under the influence of gravity, there is a constant tendency to equality of level in all parts; but, since, after any disturbance, this tendency takes time to assert itself, and since, before the necessary time has elapsed, some fresh disturbance is always introduced, equality of level does not in fact ever occur. It is evident that the average amount of inequality of level depends in part on the magnitude of these disturbances. It is, similarly, evident that the average degree of inequality of returns depends in part, in respect of any system of impediments to movement, upon the size of the relative variations which the demands for productive resources at different points in the industrial field undergo. It is the task of the present chapter to distinguish the principal influences upon which the magnitude of these relative variations in different circumstances depends.


§ 3. First, it is plain, in so far as the demands for the services of productive resources in different occupations and places are affected by independent causes, anything that promotes variations in any particular occupation or place is likely to enhance the relative variations of demand as a whole. Thus all the factors affecting particular industries, which we shall be studying for a different purpose in Chapter XX. of Part III., are also relevant here.


§ 4. Secondly, when the demands for a commodity that is being produced in several centres fluctuate as between the centres, anything that prevents variations of demand for the commodity from being reflected in variations of demand for productive resources lessens the relative variations of demand for these resources. The practice followed by some firms of giving out work on commission in times of over-pressure to other firms in the same industry that are temporarily slack has this effect. When the firms engaged in an industry combine into a single concern, this device can, of course, be carried further. Orders, at whatever point they originate, can be spread among the members of the combination in such wise that there are no relative variations in the several demands for resources to meet these orders.


§ 5. Thirdly, there are sometimes at work causes which bring about definite transfers of demand from one group of occupations to another. The most obvious of these are seasonal changes in climatic conditions; in the summer, for example, people want less gas for lighting but more petrol for motoring than they do in the winter. Shiftings of taste, under the influence of fashion, from one class of luxury article to another are on the same plane. Yet again, even when every individual's tastes remain unaltered, the transfer of income from people with one set of tastes to people with another set involves a transfer of demand from the products favoured by the first class to those favoured by the second. Thus an improvement in the incomes of the poor at the expense of the rich would cause the demand for poor men's goods to grow, and that for rich men's to contract: and this change would be reflected in the demands for productive resources to make the two sorts of goods.


§ 6. Fourthly, there are at work certain general causes, which affect the demand for productive resources in a large number of occupations in the same sense, but in different degrees. Thus the psychical, monetary and other factors that underlie what are commonly called cyclical fluctuations in industry involve, for reasons which I have endeavoured to explain elsewhere,*47 much more violent swings in the demand for instrumental goods than in that for the general run of consumption goods. This, of course, means that demands for productive resources to produce these two sorts of goods fluctuate relatively to one another. Any policy, therefore, that succeeded in mitigating the swing of cyclical industrial fluctuations, would, incidentally, also lessen the relative variations in the demands of different industries.


§ 7. If we suppose ourselves starting from a position of equilibrium and imagine any relative variation of demand to occur as between two occupations, until the appropriate transfer of productive resources has taken place the national dividend must fall below its maximum. But, nevertheless, it is for some purposes important to distinguish between relative variations in which the demands in both occupations have moved in the same direction and relative variations in which one demand has risen and the other fallen. Where wage earners maintain rigid rates of wages in the face of falling demand, it will be possible for those displaced from a depressed industry to find employment in an expanded one, but not possible for them to find employment in one which, though less depressed, is itself depressed in some degree. To eliminate obstacles to movement—provided that the expense of doing this were not too great—would, therefore, help the national dividend in the first case, but would have no effect upon it in the second.


§ 8. In conclusion it should be noted that relative variations in the demand for productive resources, whether between places or between occupations, may be expected to have different effects according as they take place rapidly or slowly. If one occupation or place declines slowly while another expands slowly, adjustment may be made to the new conditions without it being necessary for any actual transfer of resources to take place between them. All that need happen is that in the declining occupation or place, capital, as it wears out, and workpeople, as they retire or die, are not fully replaced; while newly created capital and young men and women coming to industrial age are turned into the expanding place or occupation to the extent required to fit the enlarged demand. In these conditions, no actual transfer of capital or labour being required, obstacles in the way of transfer do no harm to the national dividend. When, however, relative variations of demand take place rapidly, adjustment cannot be fully made in the way described, and, if it is to be made, actual transfers must take place. Here, therefore, obstacles to movement necessarily injure the dividend, and the elimination of them (if this could be accomplished without too great cost) would benefit it. Where exactly the line between "gradual" and "rapid" relative variations lies from the standpoint of this discussion, depends on the rates at which, in the occupations and places concerned, capital equipment normally decays and the labour force normally requires replacement. As will be observed presently in another connection, the proportion of the annual flow to total stock is, in general, much larger among women workers than among men workers.*48

Notes for this chapter

Cf. my Industrial Fluctuations, Part I. chapter ix.
Cf. post, Part III. Chapter IX. § 6.

Part II, Chapter IX

End of Notes

23 of 73

Return to top