« НазадПродовжити »
not decisive. A large part of the dividends were paid out of reserves; while the profits actually made in the two years would represent the earnings of differential advantages over marginal firms. Employment depends on the ability of the marginal firm to pay its way; there may, therefore, be great unemployment and yet profits be made by firms that are not marginal.
There is not much justification in these facts for the common assumption that the pre-war rate of wages can be taken as a starting-point in wage discussions. Even if an improvement in the world economic situation makes possible a restoration of the pre-war level of real wages on the average, the changes we have reviewed render it unlikely that all occupations will be able to secure full employment at that level. Some trades will have gained and some lost by the war.
We have to look elsewhere for an indication of the general level that post-war conditions will allow. The problem is to find some sort of a base line, by reference to which allowance can be made for the various changes altering the relation of wages in one trade to wages in others. It arises because occupations vary widely in the speed with which they respond to changed economic conditions. At the one extreme are the iron and steel trades, that have accepted the full force of the commercial depression by sticking to their sliding scales, and, since 1921, the coal industry; at the other extreme are the employees of public authorities, effectively insulated from any immediate shock from changes in the economic situation. The cost-of-living sliding scale has done something to relate wages in the latter occupations to commercial conditions, but only by accident, so that they certainly cannot be taken as our base line. The former group, on the other hand, exaggerates the fluctuations in wages which the change in the economic situation requires; nevertheless they form a better indication than the other group of the national economic fortunes.
The indication we are looking for is probably to be found in the important export industries. Just because they work for foreign markets, they are compelled to adjust their costs to a competitive level, and wages in them reflect the changed conditions of the post-war world. Moreover, a country dependent for maintaining a reasonable standard of life on foreign trade cannot for long allow the workers in industries working mainly or exclusively for the home market to enjoy better conditions than workers in export industries, without running the risk of starving the latter and so depriving the former of the source of their good conditions. Before this indication can be used, however, it needs correction in two or three respects. In the first place some allowance has to be made for the inclusion in the average of industries like engineering, in which overcrowding induced by the war has forced wages lower than they would have fallen if influences on the demand side alone had been operative. In the second place, allowance must be made for the effect of transient influences (at least one hopes that they are transient), like the occupation of the Ruhr, that tend to depress wages in the more sensitive industries (or raise them, as in the case of coal) below the level which represents a fair allowance for the more permanent effects of the war. On the other hand, it is possible that wage rates are still at a level at which full employment is unlikely in some industries, in which after great reductions unemployment is still four and five times as great as in a bad pre-war year. It may be undesirable to reduce them further in these cases, but the alternative is to check the influx of new labour and assist the efflux of labour at present dependent on these industries.
To conclude; the post-war wages problem is the problem of restoring to the wages system the stability that it possessed before the war. This can be done only by modifying the pre-war system to allow for the accumulated changes of five years of war. In making these modifications we are forced to seek some sort of a guide as to the general relation which post-war wages will bear to pre-war wages, and I have suggested that this guide is to be found, with certain corrections, in the average level of wages in export industries rather than in the pre-war level adjusted to allow for the increased cost of living. I have not tried to do more than define the problem, and to catalogue some of the factors in it that the union and association officials, arbitrators and others who fix wages will have to take into account, congratulating myself that I do not share their responsibility. But even this may be of use in view of the urgency and difficulty of the problem. Before the war a mistake in adjusting wages to commercial conditions was not serious; it might cause a slight increase in unemployment, but the rising tide of demand, as society got richer, would correct the mistake. To-day, with perhaps a fifth of the labour force of the country standing idle, there is not the same certainty that society is getting richer. The problem of fixing wages is therefore more delicate and the stakes involved in every decision are heavier.
THOSE EMPTY BOXES § 1. A YEAR ago the pages of this Journal were enlivened by a battle of giants. The contest raged over the practical utility of certain refinements of analytical economics : in particular it was debated whether the theoretical sorting of industries into boxes labelled “ diminishing return” and “ increasing return” could be turned to practical account. Into that mêlée it is not my purpose to enter directly, though I think that my suggestions, if correct, may have some bearing upon its issue. My purpose is an even more presumptuous one-to engage, albeit armed with an apparatus as defective as David's, one of the giants upon his own ground, and to cast a pebble which, if it glances innocuous from that august brow, may perhaps at least elicit an explanatory roar.
My aim then is to echo, and I hope to expand, certain criticisms which have been wafted across the Atlantic of the analytical mechanism elaborated by Professor Pigou in his Economics of Welfare for dealing with these conceptions of diminishing and increasing return. I desire to suggest that the character of that mechanism (1) renders the “ filling of the boxes " unnecessarily difficult; (2) prompts to a use of the boxes, if filled, which under certain conditions of political and economic development might become positively misleading and dangerous. And I desire to suggest that this unfortunate result has its roots-I will not say in an unhappy choice of terminology (there are limits even to David's presumption), but in a tendency to discover a simplicity of parellelism where none exists, and to submit disparate materials to an identical logical process. The boxes, if I may make free with the metaphor, are not in my view properly to be loaded upon the same cab. It is almost as though one were a hat-box, and the other a monstrous compound of a box at the opera and a box growing alongside a garden path.
§ 2. Now first, while far from insisting that this question of terminology is fundamental, I must be allowed my private grumble about it. By industries which obey the laws respectively of diminishing and increasing return I take to be meant those in which the average full expenses per unit of product respectively increase and diminish as the scale of production is expanded.1 In spite of Dr. Marshall's arguments to the contrary,2 I wish heartily that English-speaking writers could agree to speak of such industries as obeying respectively the laws of increasing and decreasing cost. What superfluous confusions about the emergence of rent in manufacturing industries, what regrettable misunderstandings of the teaching of economic science about the relation of agricultural progress to the problem of population, could have been avoided if the phrase “ the law of diminishing return” could have been tied down to denoting the results of either (a) applying successive doses of one factor to a fixed quantity of all the others, or (6) applying successive doses of all the factors but one to a fixed quantity of that one! And even that much duplicity of function would seem to entitle it to extra pay! I look forward to the time when this hard-worked phrase, relieved of all other duties, shall thus be confined to expressing the great unifying principle which runs through the whole theory of distribution-surely a sufficient honour to content the most ambitious company of words.
I propose, therefore, to speak of “ decreasing cost” industries and “ increasing cost” industries; but I repeat that I do not intend that this choice of phrasing shall alter the substance of my argument.
§ 3. And now to business. Professor Pigou's most momentous conclusion is that under conditions of free competition production in “increasing cost” industries is carried further, and in“ decreasing cost "industries less far than the true interests of society require. I shall not attempt to summarise his argument, but shall give such references as will enable the reader to refresh his memory of it.
Let us confine ourselves at present to “ long-period ” conditions, in the sense that we are not concerned with the effects of cyclical or other temporary fluctuations of demand. And let us deal first with “ decreasing cost " industries in the sense defined. Now the first point is that, so far as I can see, this apparently harmless phrase covers two sets of analytically quite distinct phenomena. As the scale of output of an industry is increased, the average cost 1 per unit may fall for either of the two following reasons. (I) In some branches of production the process of investment of fixed capital is, from the nature of the case, lumpy and discontinuous; and once such a process is completed, the larger the number of units of output produced (or, to put the same thing in a different way, the larger the number of units of floating resources of all kinds employed) the smaller the share of fixed capital charges which each unit has to bear, and the less therefore, in general, its full cost. “We may take the production of a medal as a type of such a supply. For we may suppose that it costs a manufacturer of medals £20 to produce a steel die, and after that has been made it costs him 58. for the metal and stamping of each medal. If then he produced only one medal the cost would be £20 5s. If he produced two medals, £10 58. each. If he produced fifty medals, the cost would be 138. each; and so on.” 2 Note first that such a statement may well represent, not an evanescent state of affairs, but the permanent conditions under which an industry (such as newspaper enterprise ?) is operated. Note secondly that the element of time, so often the source of all our troubles, is here as nearly as possible innocent : a raising of the demand schedule for the medals (occasioned, say, by the establishment of a permanent state of war) will instantaneously lower their average cost.
? Professor Pigou does not himself use this definition; but it appears consistent with, and is indeed derived from, the definition and properties of his “supply-curve ” (Economics of Welfare, p. 933).
2 Principles, p. 319 n.
3 See Professor Bullock's most valuable article, “The Variation of Productive Forces,” Quarterly Journal of Economics, August 1902.
• A distinguished Colonial professor still, I believe, boasts publicly of his intention of disproving the existence of the law of diminishing return from land.
No. 133.- VOL. XXXIV.
But (II) decreased average cost may accompany expanded output for another reason, namely, because, given time, methods of technique and of organisation are capable of improvement in any one of a myriad different ways, so that ultimately a larger output can be produced at a lower cost per unit than that at which a smaller output, was previously produced. I do not pretend that this class of cases is entirely unconnected with the first class; for the “ economies of large-scale production,” and among them the installation of large and specialised pieces of plant, are among the causes of the prevalence of decreasing cost in my sense (II). But the difference between the classes can be seen by reflecting that nothing but a raising of the demand schedule can be relied upon to establish a lowered cost in Class (I), while the progress of time, the enterprise of producers and the occurrence of invention 3 are expected, without necessarily any alteration of normal demand, to produce this result in Class (II).
1 I use this word henceforth as an abbreviation for "full expenses of production.”
2 Cunynghame, A Geometrical Political Economy, p. 57.
3 Dr. Clapham's query whether invention is to be classed among the causes of “decreasing cost" must clearly, on my definition, be answered in the affirmative (ECONOMIC JOURNAL, 1922, pp. 314, 562).