« НазадПродовжити »
§ 4. But already I seem to myself to hear a triple growl from the giant. (A) First, “ Have I not made it clear," I hear him ask, “ that for simplicity's sake I conceive of the resources applied to industry as a homogeneous flow (Economics of Welfare, p. 114)? And how am I touched then by that discontinuous £20 steel die, or by your distinction between fixed and floating resources, which I have decided to ignore?” And I can only reply, Yes, he seems to have made that clear. But I shall go on to urge three things. (1) There is no doubt that, in ordinary discussions of the phenomenon of decreasing cost, industries of my set (I) play a very large part. It seems to be of such cases that Prof. Edgeworth 1 speaks as “so important as often to obtain the title of Increasing Return par excellence,” and of the five sets of conditions which he distinguishes as “ attended with the attribute Increasing Return,” two indubitably turn on this principle of the discontinuity of investment and the economy of multiplication. (2) The case is covered by Professor Pigou's diagrams in so far that while, in the case of our medal-manufacturer, the “curve of marginal supply prices” (Economics of Welfare, p. 931) would become horizontal after the second medal, it would remain continuously below the supply-curve, and cut the demand-curve further to the right than the supply-curve does. (3) But I can appeal more directly to Professor Pigou. He states specifically (op. cit. pp. 275-6) that “ among railways there is ground for believing that, at all events until considerable development has been reached, this condition [strong action of the law of increasing returns) is generally satisfied. The reason is that the fixed plant of a railway cannot, in practice, be so made as to be capable of effecting less than a considerable minimum of transportation. The aggregate costs of arranging for rail transport for one ounce per week are very nearly as great as those of arranging for the transport of many thousand tons. ... This implies increasing returns acting strongly till a larger investment has been made, and afterwards-less strongly.” There can indeed be no question that Professor Pigou's analysis is intended to apply to railways; nor that, in Ripley's words, 3 “ from this fact [discontinuity of investment], therefore, rather than because of any marked economies of large-scale production,
1 “Contributions to the Theory of Railway Rates," ECONOMIC JOURNAL, 1911, p. 370.
: Ibid., pp. 553-5.
• Quoted by Edgeworth, “ Contributions,” etc., ECONOMIC JOURNAL, 1913, p. 217; who, while holding that other conditions generating " increasing return" have been too much ignored by writers on Railway Economics (ibid., 1911, p. 553), seems to concur in giving such considerations the primacy.
may it be affirmed that railroads offer a notable example of the law of increasing returns.” If, therefore, Professor Pigou tells me that his apparatus of a homogeneous flow of resources is not a suitable one for dealing with my Class (I) industries, I shall joyfully agree; but if he forbids me on that account to discuss them with him, I shall decline to obey.
$ 5. (B) But there is another door of escape for the wounded giant which I think it prudent to bar. We are dealing so far with a regime of competition; and he may tell me that my adopted medal-maker is not a true subject of that regime. If he were, and if he found that £10 58. were the ruling price for medals, he would produce not two, but ever such a large number. The fact that an output of two is associated with a price of £10 58. indicates that he is regulating output with an eye to the effect of his actions upon the price of his product, which is improper conduct in a denizen of the realm of competition, even though to act otherwise would, by preventing him covering his total costs of production, drive him out of business. Now poor David has no weapons with which to attack this mathematical conception of pure competition. But he knows, or thinks he knows, that his Class (I) industries exist, and that they do not all exercise monopoly powers, but that in their case, as in others, normal competitive price must in the long run cover supplementary as well as prime costs. He is even prepared to invent for his own use a meaning of the term “ competition,” which shall imply that producers are not in a position to make monopoly profits, but are free, and determined, in the long run to cover their standing charges. He suspects that printing, and in most countries in normal times railways, are conducted permanently on these lines, and he is tempted (but without falling) to hazard further attempts at filling this particular sub-box. He suspects too-a point to which he will return—that it is “ competitive” conditions of this kind that the State, if it takes over an industry, should (unless for good reason to the contrary) seek to emulate.
(C) There is one more loophole—will the giant make for it? Will he charge me with neglecting his warning (op. cit. p. 115 and p. 931, note) that whatever the scale of output, the flow of resources must be conceived of as organised appropriately to that scale? I shall not plead guilty; for whether my medalmaker is turning out two medals or fifty, he must be assumed to. be acting in the most sensible way open to him. The relatively high cost per unit of the smaller output is due, not to any lack
of judgment on his part in combining his factors, but to technical facts beyond his control.
§ 6. I shall assume, therefore, that I am entitled to discuss with Professor Pigou both my types of “ decreasing cost " industry, and to discuss them separately. Let us take Group (I) first.
Let us suppose that m units of fixed resources (typified by the steel die) have been sunk in the industry, and that n units of running resources (typified by metal and labour for stamping) are being employed in conjunction with them, and that p units of output are being produced. Professor Pigou's analysis (op. cit. p. 937) leads us to suppose that, in the social interest, p should be such that the additional units of output specifically attributable to the addition of the nth unit of running resources should sell for a price which adequately remunerates that nth unit of running resources. In our concrete case, if demand is such that fifty medals can be disposed of at 58. each, fifty medals should be produced, since the fiftieth medal is the “net product” of the marginal 58. worth of resources, and 5s. is the “ marginal supply price” for an output of fifty—the difference made to aggregate expenses of production by organising production so as to produce fifty instead of forty-nine (op. cit. p. 931).
Now I deny that this pth unit of output is in any significant sense the net product of the nth unit of running resources. It is the net product of [that unit + m units of fixed resources]. The fact is that these m units of fixed resources yield no product, unless at least one unit of running resources is applied in conjunction with them, and that in order to make sense at all we must credit them with a part of any additional output which appears at first sight to be specifically attributable to the addition of any nth unit of running resources. Professor Pigou's statement (op. cit. p. 190) that when any given aggregate of resources is devoted to an occupation, any one unit of those resources must be conceived of as yielding the same net product as any other, leads me to suppose that he would agree with this view; but, on the other hand, the rest of his analysis, and especially his treatment of the railway problem, leads me to suppose that he would not. For if the view is correct, the whole case for
1 In our concrete case, we can choose our units so that p always = n. If the special costs per medal for labour and materials either increase or diminish as output is expanded, the argument becomes slightly more complex in form, but is unaltered in substance.
carrying production in this group of “ decreasing cost ” industries beyond the competitive point seems to me to vanish. And consider how paradoxical that case is ! For if it is sound, the logical outcome seems to be that the State, if it takes over an industry of this type, is entitled to neglect altogether, in determining its price-and-output policy, the costs of the fixed capital embarked, paying for them presumably out of taxation, and pushing production to such a point that price covers only the special costs of the pth unit of output. Whereas, if my view is correct, the State (unless for clear cause shown) ought so to regulate output that aggregate receipts cover aggregate costs without yielding monopoly profits. Even if through want of capacity I have misunderstood the giant's analysis, let me try to pin him down to this supremely important point of practical policy. Suppose discrimination to be ruled out, as in the sale of Government publications (for I see an avenue of escape for him along the road of discrimination), how should such a nationalised enterprise (any broader social reasons for subsidisation being disregarded) fix the scale of its output and its charges ? Does his analysis or does it not lead directly to the conclusion that it should claim to be subsidised to the extent of the whole burden of the charges of the fixed original plant? And can this conclusion possibly be sound? And if not, what logical half-way house is there between this procedure and charging so as to cover total costs, i. e. simulating the conditions of free competition ?
§ 7. Now let us turn to my second group of " decreasing cost” industries—those in which an enlarged output is associated with a decreased cost per unit of output as a result of the introduction of improvements in technique and organisation. How far is it true that in these industries output is smaller than is socially desirable, and might be brought to a socially more desirable level by some form or other of State intervention ?
Now it is these industries that, at any rate in the prelude (op. cit. pp. 189–192) to his discussion of the whole matter, Professor Pigou seems to me to have uppermost in his mind. His explanation there given of the failure of competition to produce the best results must therefore be carefully noted. The employment of an additional unit of resources in any industry may, it appears, so modify the general organisation of the industry as to make each of the units of resources employed in it yield a different net product from what it otherwise would have done; but since, under pure competition, the individual who has made the extra
investment experiences only a very small part of the effects of this indirect impact upon general organisation, it is not to be expected that the probable nature and total magnitude of these effects should appreciably influence his actions. The relevance of this doctrine to “ increasing cost ” industries does not concern us here : as applied to “ decreasing cost ” industries, it can only mean that all the improvements in organisation from which “ decreasing cost" arises are of the nature of “ external economies ”—“ those dependent on the general development of the industry" , the familiar “ internal economies "_" those dependent on the resources of the individual houses of business engaged in it, on their organisation and the efficiency of their management "Lhaving vanished into thin air.2
Now once again I am conscious of being up against the mathematical theory of pure competition, and aware that that is so much the worse for me. But I cannot let it rest at that. I recall to mind the “representative firm,” which “ has its fair share of those internal and external economies, which appertain to the aggregate scale of production in the industry to which it belongs ” ; 3 whose size,“ while partly dependent on changes in technique and in the costs of transport, is governed, other things being equal, by the general expansion of the industry," and is therefore clearly to be regarded as capable of significant variation : 3 and which is nevertheless certainly a denizen of the realm of competition, and indeed plays a commanding part in the theory of normal competitive price. I even take the supreme risk of starting to fill the sub-box under consideration, by flinging into it at a venture electrical engineering and cinema-film manufacture, in order to confirm my impression that any likely occupant of the sub-box is almost sure to be carried on by firms who are capable of introducing and appropriating internal improvements in organisation and technique. (I cannot, therefore, bring myself to believe that, under any conception of competition which is appropriate to the matter in hand, the phenomenon of decreasing
1 Marshall, Principles, p. 266.
? So determined is the Professor to banish these old friends that, disturbed by the apparent theoretical incompatibility of pure competition with the prevalence of decreasing cost at all, he seems to hold (p. 192) that each firm is (? or would be if it were isolated) working under conditions of increasing cost while the industry as a whole is working under conditions of decreasing cost. I would prefer to offend the mathematical theory of competition than to follow him through this logical hole in his own logical net; therein agreeing with Professor Allyn Young, who “ cannot imagine 'external economies 'adequate to bring about this result”. (Quarterly Journal of Economics, August 1913, p. 678, note).
: Marshall, Principles, p. 459. My italics.