"$^{\text{a}}$" | $^{\text{b}}$" | $^{\text{c}}$" | $^{\text{d}}$" | $^{\text{e}}$" | $^{\text{f}}$" | $^{\text{g}}$" | $^{\text{h}}$" | $^{\text{i}}$" | $^{\text{j}}$" | $^{\text{k}}$" | $^{\text{l}}$" | $^{\text{m}}$" | $^{\text{n}}$" | $^{\text{o}}$" | $^{\text{p}}$" | $^{\text{q}}$" | $^{\text{r}}$" td colspan="4">
On the 29 hire land (and so on) |
---|
$$^\text{"a"}$$ Hobod. p. 120.
ATTEMPT AT RECONSTRUCTION
On the 32 hire land.
417
**$20$**
**$20$**
**$20$**
**$20$**
and that the price-determined differentials are $30-30-0$;
$a_{0}-a_{0}=a_{0}$; $a_{0}-a_{0}=a_{0}$, etc., increasing to a unit of price-
determined differential for the 32-hire land.
It must be fair to say that Hobson does not himself put his
case in the precise terms of the illustration offered; the details of
the scheme of statement are not his, but those of the present writer; but the
argument and the illustration are so closely connected with each other at
the end of assigning to the argument its clearest, shortest, and most
effective statement, that it would be unfair to Hobson's case to omit them.
It must, however, be at the same time admitted that this manner of
exposition is chosen in some measure with a view to bringing the issues
of criticism into clearer focus. It may be said that this exposition is too
elaborate, and it may be pleaded that it is not intentional; but there is perhaps
the more reason for all that whatever justification the facts afford be promptly submitted.
Though the worst granting land may pay no rent, the worst wheat
land might pay a positive rent, and though the best market-garden land in
this case can only be obtained for growing wheat by paying a little more
than its marginal rent, yet it will still pay a positive rent. The worst wheat
land will be a positive rent, and will enter into wheat prices; again, the worst market-garden land competing for a given market
may be obtained for growing wheat by paying a little more than its marginal
rent for wheat forms a marginal rent for market-garden land. So as we
see how much higher than its marginal rent the worst wheat land pays,
the differential rents must be measured, not from a no-rent margin, but
from a minimum specific rent of a higher and higher order, until we get
to the greatest possible rent which any land could obtain for the best agricultural use to
which it is adapted.
For the sake of simplicity I have assumed that the marginal rent is directly and exactly determined by the alternative use of the worst
land in question. This is not always true, but it is usually true in practically
the case. It is not necessary that the worst land should have an alter-
native use: it may be some better land, enjoying a differential rent as large as
marginal rent. In such cases it is clear that if the worst land could obtain
a marginal rent of $20$, per acre; superior qualities of wheat land might obtain a marginal rent of $25$, per acre; and inferior qualities of wheat
land might obtain a marginal rent of $30$, per acre. But if any one of these
of the land rented at $20$, had another use which would yield a rent of $25$,
it is evidently this land which fixes the marginal rent; it must receive
a rent of $25$. If any one of these lands had another use which would yield a rent of $30$,
and the use taken by the worst land measures its inferiority of wheat-
growing, then this land must receive a rent of $30$. If any one of these
soil land might continue to grow wheat, however little rent was paid;
its rent is directly determined by the cost of keeping in the supply of
wheat land which it occupies. It is evident that this cannot be done on
soil land and not the soil land which is the direct determinant of
418
VALUE AND DISTRIBUTION
Precisely where Hobson finds the ultimate forces of causation in the determination of the price of wheat is not readily made out; but it is clearly his view that, on the supply side, certain of these land hires, or, more accurately, certain portions of these hires, are to be regarded as causes of price rather than results, and that we must go to the market price of the land use, or of part of it, in order to explain the price of the product.
But (i) how are these price-determining land hires themselves determined, and (ii) through what bearing and in what manner do they bear on their price causes?
1. The determination of the value of the land use is held to be in principle precisely like that of the determination of the price of consumption goods, in the sense, that is, that both are market adjustments worked out under price for the supply side in the market for sale of wheat-growing power" (pp. 123-144).
"Now suppose that on the one hand, the worst wheat land in occupation... . It is quite legitimate to suppose that the owner of this land has been able to obtain a rent per acre of 16s., which might have been willing to contribute to supply even if it had been at 16s. instead of 20s. In such case it would be evident that the rent per acre determines the amount of rent per acre of the land at the margin of cultivation" (p. 125).
"The determining increment of supply is not necessarily identical with the increment of demand; for example, if wheat grows on the margin of cultivation. If the slackness of the demand for wheat causes a fall in its price below its cost of production, and if this fall is worked out of cultivation; it may be the case, if the latter has an alternative use and the former has not. The actual determination of rent by this method is not so simple as it appears at first sight. For although one part of the land supply, but many parts have alternative uses to which they would be worth paying a higher price for one to fall below a certain figure. It is reasonable to assume that all other parts of land use is always determined by the common position of one part of employment over another part. This assumption does not hold true when supply in preference to some others; the fact that at a different price per unit some other land would occupy this position need not concern us."
This is perhaps the opportune time for presenting a résumé of the different rent concepts and, so far as relevant to the present discussion, Hobson's system of terminology with regard to them:
Selecting out of our illustrative scheme the 30-bale tract or acre
ATTEMPT AT RECONSTRUCTION
419
entrepreneur bidding. True, the consumption good receives, on the cost side, its value from the value of the instrument; and from this point of view there is doubtless a distinction, in that the value of the cost good—in this case the value of the land—is not, through an appeal to cost fixation, as readily explained, either in whole or in part, as is the value of the other two factors. On the other hand, the land hire appears, as we have seen, to be in part derived from price and in part a determinant of price. But at any rate, as a question purely of the process of market adjustment, there is a complete parallel between the price of the land use and the price of the finished consumption good:
In so far as the price of uses of factors of production is reached by competition and bargaining (and this is our hypothesis throughout), the mode of determining rent, interest, and wages will be
of land, the rent concepts applicable to it would catalogue as follows:
1. The entire hire, 30 shillings.
2. The differential hire above the poorest land in actual use, 20 minus, say, 5 shillings.
3. The differential hire above the poorest land in use in the particular line of production, 30 minus 20 shillings.
4. The differential hire above the best alternative use, say 5 shillings.
(With the exception of the first item, all these differentials are evidently 1 : on the cranberry patch of our earlier discussions, this differential would include the entire hire.)
"Now it would seem that it is necessary to give to the term 'margin of occupation or employment' to describe the worst or least efficient part of supply, some special significance. It seems to me that this margin occupies the determinant place in any given market. I propose to speak of this portion as 'the determining portion of supply,' and of its owner as 'the determinant.' In this way it will be possible to describe what is happening in any market. The margin of supply may be described, in accordance with usage, as 'marginal land,' and its rent will be determined by rents obtained by hands of superior productivity contributing to this supply, and will be measured from the margin" (p. 129).
But upon the assumption, actually made by Hobson, that "the lands are funded," there is surely no occasion for ranking any of the lands as worse than others. For suppose that one were better than another; then both would be hired out at their full value; but if they were both hired out at their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neither could be hired out at anything less than their full value, then neither could be hired out at anything less than their full value; and if neithercouldbehiredoutatanythinglessthantheirfullvalue,
in which case there would not only have been no rent paid on any land but no rent paid on any hire.
A page from a book with text discussing economic theory.
420
VALUE AND DISTRIBUTION
essentially the same as that of determining the price of horses or wheat (p. 136).
But precisely because the consumption good is a good price-determined through its costs, it must be recognized that when goods have been exchanged for goods, . . . in order to understand more fully the nature of the bargain, we must regard any two commodities which have been exchanged as complexes of the various quantities of the factors of production that have entered into them in the various processes of production (p. 133).
This serves merely to emphasise that we understand prices, we must, on the supply side, refer to land costs, determine the actual quantity of the rent-cost elements in price; and thus we discover that a bargain for the sale or the exchange of finished commodities will depend, as far as supply forces are concerned, upon the conditions of a number of preceding, underlying bargains for the use of different kinds and qualities of land, capital, and labor power (p. 133). We regard the hiring of the factors of production as equivalent to the sale of their use,
and thus subject to all the possibilities of haggling,
oppression, and forced gains (p. 114).
Where shall we be able to isolate a rent which is price-
determining in its entirety, that is, without adulteration of any price-determined element? But in any case, whatever rent comes to be selected as price-determining, it must, perforce, be a rent fixed through the market process of competition between bidders for the use of land and uses.
And it is also to be understood that all these land uses are actually sold at the same ratio between their productivity which they offer and the price which they com-
mand; that is, they are all funded into units of
land-productivity service:
In just like manner, price-determined commodities by the relative eco-
nomies at which buyers and sellers bargaining for a given quantity of land use and not for a given sized piece of land, though the language of these proceedings has reference to the latter. The sub-
**ATTENTAT RECONSTRUCTION**
421
jective valuations [offer prices and refusal prices] of a single owner and a single tenant (the final pair) fix the limit for the price of a unit of this land power, the stronger of the two fixing the price point (page 30).
Adopting, then, the supposition that "what is really sold in the bargaining between land-owners and cultivators for the use of wheat land" is not merely wheat-growing power or wheat-growing land but wheat land "as units of wheat-growing power," we find that this supply is conceived as a fund of abstract productive units—we turn to examine the process by which these wheat-land hires are adjusted. It is significant that for this purpose Hobson appeals to the concrete and not to the abstract supply of...
But upon this definition (x), the criticism against the "whole mathematical treatment" was that it "rests upon the assumption of an infinite divisibility of supply." The fact that supply is not in any case divisible does not impair its utility as a basis for whole mathematical treatment. Thus "it is easy to see that there is a far greater variety of possible prices with a given quantity than on the horse market, a far greater variety of possible prices with a far narrower interval between them. This signifies a far closer and more effective competition than exists on the horse market. On the other, the result being that the limits between which ordinary competitors can deal are much wider. (p. 27.) But surely competition should then be perfect in the horse market? The assumption that it is really funded into precisely equal, interchangeable, and indivisible units of supply, as Hobson now ever, seems to be that Hobson regards land as funded for the purposes of cost theory, but as non-funded for the purposes of "forced-gain theory."
"Wealth or work will, in theory and usually in practice, rank as separate and distinct things—goods or bargains... goods which are held to be identical in size and quality. Now in the market for the sale of one of labour or land no formal reduction is made of their value into units of productive power. In place of sale is a quantity of productive power in land or labour, what is nominally sold is not what is actually bought. . . . How many labourers... But while the bargainers express themselves in terms of acres or labourers, the real object of their bargain is the use of land producing power. They do not wish to be reduced to reducing acres and labourers to units of productive power when they buy and sell.
"It is admitted (by whom?) that what is really sold ... is units of wheat-growing power. The fact that the nominal subject of bargain is land produces power does not affect this conclusion (p. 24). If so, then so much the worse for forced gains; but Hobson does not see this.
"It is the owner of the soil, land who, in fixing for the supply side the price per unit at $4., determines the amount of rent per acre of the land at the margin of cultivation—the $4. land (p. 14).
432
VALUE AND DISTRIBUTION
entrepreneurs and to their competitive bidding against one another; and it would thus seem that all the phenomena, forced gain or other, characterizing the price fixation of consumption goods must equally apply to land. Indeed this is indeed. Hence, if we regard the hiring of the factors of production as equivalent to the sale of their use, we are confronted with the investigation of the market for the sale of the use of various supplies of land, labor, and capital.
But, if so, how shall productivity be funded unless also the utility in every particular stock or series of commodities be also taken as funded? This view must find its argumentative support in the doctrine of natural rent as an organism doctrine. And what, then, becomes of forced gains?
We have seen that it is the xox-per-acre land which is believed to determine the price, that is, under the conditions as assumed, not the worst land, the xox land, but the best, the xox land, this latter being the land upon the alternative margin.
Suppose that some of the land rented at xox had another use which would yield a rent of xpo; it is evidently this land that fixes the marginal rent.... It is possible that the xox land might continue to grow wheat, however little rent was paid; its rent is simply determined by the cost of keeping in the supply of wheat land. The rent on this land will be xox. The rent on wheat will be the xox land and not the xox land which is the direct determinant of price for the supply side in the market for sale of wheat-growing power (p. 124). The argument... shows that it is not necessary that all rent potentialities do not enter into price, lets into price any rents which are paid for use of marginal land contributing to any supply. Land may be graded according to its economic uses; the different rents will be included in the marginal costs of production (p. 125).
The subjective valuations of a single owner and a single tenant (the final pair) fix the limits for the price of a unit of this land power, the utmost limit at two fixing the price per acre. This done, the rental acre is determined by its value as a field of land rent in each grade of land. If the bidding of the market fixes the price of a unit at xoz, the best land available for that supply may yield two units of power per acre, in which case the rent per acre is xox, the worst land only 36 a unit with a rent of one per acre (p. 126).
ATTEMPT AT RECONSTRUCTION 423
But if this is true, it must follow that all lands are equally cheap and equally dear; and if so, all talk of areas, or of acres, or of tracts of any sort becomes irrelevant. 1,000 acres of poorer land being both the price equivalent and the productivity equivalent of 100 acres of a better grade or of 10 acres of a still better grade, then the doctrine should, then, suffice to cancel all talk of marginal lands in any sense of marginal entrepreneur cost, or in any other sense than that of the nearness of the wheat land to the line of the corn land. Under the doctrine as presented, all lands are equally expensive for wheat purposes, and, so far as entrepreneur outlays or entrepreneur grounds of interest are concerned, neither the zot land nor the corn land can be distinguished in its rental. Here, indeed, it becomes clearly manifest that Hobson's analysis, unconsciously collectivist in standpoint, really involves the entire abandonment of the entrepreneur point of view.
For mark how this manner of computing entrepreneur costs by the displaced potentialities of the instrumental goods will affect the entrepreneur computation when car- ried over into the question of rent. It may be well to be recalled that Hobson protests vigorously—and rightly—against "that general tendency of economic science, especially in England," which "seems to treat all those of the sale of capital use and labor power to that of the sale of goods, but to mark off the sale of land use as subject to quite other economic laws" (p. 116).
I propose to bring the sale of the factors of production under the general law of exchange. For this purpose it is necessary (1) to co-ordinate the three factors with reference to the conditions which regulate their price; (2) to show that their sales are in essence identical, as economic processes, with the sale of com- modities (p. 117).
But if it is indeed true that the entrepreneur will reduce as his cost only what he pays out, it is somewhat open to him with regard to the application of his expenditure, but only the worth of what the laborers could respectively pro- duce in their next most productive lines of employment; and if, out of his own resources, he does not find sufficient means to be allowed to compute as cost only such part of this expense as represents, say, what his cotton machinery would have produced in a woolen mill, there is an end of all hope that
424
VALUE AND DISTRIBUTION
any entrepreneur anywhere will ever be able to determine his own or anyone's else cost of production of anything.
The full significance of the alternative-margin analysis, as presented by Hobson, will be best appreciated if put in the form of a summary:
1. Through the determinant power of the margin, we arrive at a price-appraisal of $z. per unit of land power, for the land powers in the marginal land—the zoz. land, a 6-unit tract.
2. By reflection from this price, margin-fixed, we get the same price for all other equal land powers, and thereby a fund of least-land-power productivity.
3. This marginal land, the zoz. land, also determines, on the land-cost side, the price of wheat, since this zoz. land was at the alternative-use margin, and held the position of marginal-instrument cost.
4. Having fixed the wheat price, and having fixed the rent accruing to those lands not themselves price-determining, this marginal land (or is it the owner of the marginal land?) apportions to these other lands their supra-cost differences—rents—and thus they receive their necessary (price-determining?) rentals, attributing thus to them—in our illustrative scheme,—price-determined rents of from $\frac{1}{4}$ up to $\frac{1}{2}$. These price-determined services from non-terminant lands are thus, as it seems, cheaper, for cost purposes, than the services of lands nearer to the alternative use, and yet the lands are funded into equal productivity units.
And thus it appears that the marginal-land instrument, at the margin, is determinant and strategic to a degree not before appreciated even by the most pronounced advocates of the margin-fixation doctrine.
But despite the fact that this manner of analysis makes impossible the entire entrepreneur-cost category, it still remains unclear whether the determinant margin is presented as an entrepreneur margin or as an instrument margin. The supply of wheat being a stock of similar items,
A diagram showing different levels of land productivity and corresponding prices.
ATTEMPT AT RECONSTRUCTION 425
its price was determined, on the demand side, not by the entire demand but by the marginal bidder, on the supply side, not by the entire supply and not by any marginal entrepreneur but by the cost of production upon the alternative-use margin. The rent of land power, seemingly a rent for unfunded land powers, functioning as one of the marginal and price-determining costs. The market value of units of productivity is therefore fixed by the law of diminishing returns to these limited units, in view of the market price for their products. But here again it is the owner of the land upon the alternative-use margin—the land without an appreciable wealth differentials—upon which the rent is fixed. This rent in turn, in fixing the price per unit of land use at 5c., determines the amount of rent per acre for the land at the margin of cultivation, that is, the per-acre rent of the poorest land, and so forth. The owners can thus obtain their differential of income as measured from this margin, according to their differentials of productiveness in the unit schedule.
And thus it is on the whole evident why, with reference to the problem of this margin, Hobson says :
Some other term is needed to mark that fact which occupies the determinant place in any given market. I prefer to speak of this portion as "the determining portion of the supply" and of its owner as "the determining owner" (p. 128).
Both, then, seem to be determining.
But there are other difficulties :
Whether the determinant portion of supply of land be the worst land or not makes no difference; the price of land power and so the rent of different qualities of land appear to be directly determined by the fact that some of the land has an alternative use, and that it may return a profit greater than its supply cost when a certain price is paid.
But though the alternative price . . . determines a lower limit of marginal rent, there is nothing to prevent the marginal rent rising higher than 5c., provided that there exists a possibility that use might yield only 25c., now, though the owner of that land would consent to take 25c. rent, he may be able to get 30c., because there is, for the time, an absolute scarcity of land available for this supply. In a word, he may be able, as the final seller, to take a forced gain of 5c., which corresponds precisely to the "forced
A page from a book with text discussing attempts at reconstruction.
426
VALUE AND DISTRIBUTION
gain" in the price of the horse in our analysis of a market for com-
modities. In such a case it might be best to distinguish the $x. from
the $25., and to class it as a third form of rent (p. 120).
But this is really not a new—a fifth-concept of land
rent; it is merely a landlord's quasi-resent subdivision of
concept.
And now note the implications: This 30x. rent upon
land having an alternative use of only $25. becomes the land-cost of the horse, and at the same time the cleverness of this particular landlord, as achieving in bargaining a forced gain of rent for his own benefit, has enabled him, as determinant owner of the mar-
ginal land, to force upon all other tenants a forced gain of
productivity, and to fix this price for all the different lands,
with the result that all other lands as well as his own are
achieving a forced gain of $x. above their "lower" limit of
marginality. It is not difficult to see why he does not do the
purpose; to deny it is to deny to the marginal land (or to
the marginal owner—which?) the determination of the
price of productivity; the general rent determination for
the different qualities of land amounts to nothing; it is to
deny the funding of land into productivity units.
But now as to the precise correspondence of these
forced gains with those gains in the horse market. If
there is the alleged correspondence, then so much the worse
for the horse case; for where are the other tenant bidders
that, with funded lands—or without—this 30x. or $25. is left?
In fact, if we assume that in the bargaining contest—the
bargaining contest of this marginal part—that is it really
assumed that, except to this one tenant candidate, the land
is not 30x. land in point of desirability, but only $25. land?
But this is not true, because it is not so. And if the land is really 30x. land to him and to other tenants generally, if only they would act upon their interests, are the $5., over and above the $25. lower limit, really forced in the horse market? Or is it a forced gain for the gainer's expense? Or is the full 30x. the fair market value of the value-producing power offered for sale, so that at any rent less than 30x., there would be a forced gain for the tenant at 30x.? If so, then all other lands which were worth only $25. to other cultivators, but 30x. to this one, what should be regarded as truly and justly its worth?
The validity of the argument from the intensive mar-
ATTEMPT AT RECONSTRUCTION 427
gin to prove the non-cost significance of land hires comes in for searching examination at Hobson's hands. His line of reasoning in criticism of Mill and of Marshall, while perhaps not of greater interest than the foregoing, is of considerably greater cogency.
Not overlong insisting upon the fact that the argument which at the intensive margin excludes rent from cost "can be similarly applied to show that interest and wages do not enter into price," Hobson proceeds to the discussion of what he regards as the fundamental error in the so-called "dosing" method of argument, whether one or another of the factors of production be taken as applied in the dosing fashion. His objection goes to the underlying supposition that any one of the three factors of production can be expe- diently and economically applied as the sole constituent of the expense dose. "The truth is that a certain har- mony of combination of factors exists for various produc- tive purposes ; ... if there is a short supply of one of them at the former quantity prices, then both of the others will be substituted, but at an increased cost per unit of product; this must indicate that the proportions between the factors were wrong; it would have been better to have had more of the first:
So when the final dose of capital and labor on a given piece of wheat land achieves a product which yields no rent, it means that with the same quantity of land use as sufficient for a smaller product, a larger amount of capital and labor use has been combined; that on more land use was employed, none was paid for (p. 138).
Or, better :
We may consider a piece of land as containing various land powers, some high, some low, some powers so low that they require so large a proportion of capital and labor to utilize them that they only just pay to work. These low natural powers yield no net economic powers of production (p. 138).
The cost is then the same for products raised upon non-marginal land powers as upon the intensive margin, and vice versa, since upon the intensive margin so much
428
VALUE AND DISTRIBUTION
more capital and labor cost [non-land cost, entrepreneur-capital cost] is incurred as the land-rent cost is less.¹¹
Or put it in still another way:
Suppose it [the extra product] is raised by a tenant farmer as part of rent, and an extra acre of hosing and ploughing on his land, [is roughly] worth one wage, but if, however, instead of this extra hosing and ploughing the farmer decided to hire one more acre of the same quality of land and spread the same amount of labor over both acres, the larger area would pay this last acre pays its rent but no wages. The labor of working the . . . last acre of land is certainly remunerated by wages, and at the same rate as . . . the other acres. Why, then, does it appear from this that it is true that the product of . . . the last acre pays no wage? (P. 140.)
¹¹No question can be raised as to the force and accuracy of this criticism for the purposes of the actual issue; but it is none the less true that the argument is not quite accurately made. There is, in fact, no such thing as a "rent" which is paid out of the income derived from the use of land. It is a fallacy to doubtless his own separate use, as depending upon his own personal equation, that he has received a reward for using his land for using credit. But for each entrepreneur his best combination is a different one from that of any other entrepreneur, precisely because as entrepreneurs have different resources at their command. This combination will be a new and different combination with every change in the resources available to him. In short, this argument, however, the dosing argument might avoid the entire force of this general line of attack by reformulating the dose application in terms of doses per unit of land rather than doses applied to the land.
But what is the size of the economic "dose"? Is there an intensive margin to the extensive margin of land?
If every other sort of land has its intensive margin, so must also the pasture land. But pasture land differs from all other sorts of land in that pastured cattle are in themselves capital downes. Absolutely non-rental land must therefore be distinguished from rental land. Types of expense: Unless (1) marginal land does not, for the earlier doses of expense, fall within the law of diminishing value returns; in this case and up to some point they may be considered as equivalent to con- tuting one dose; or (2) unless there is a limit of another sort to the withdrawal of any particular dose from any particular unit of land. The first unit is adequate in point of amount to the duration of the undertaking in hand, and to the nature of the undertaking in hand; the first unit dose represents a definite quantity of time; it represents also the taking of the business, its size and the time involved in obtaining remuneration for it.
But in any case the dose is not a capital-goods dose or a labor dose but an entrepreneur-capital dose, a matter of quantum of expense rather than of the nature or the detail of the technological means or other means relevant.
ATTEMPT AT RECONSTRUCTION 499
And why, also, when an extra dose of labor is applied to the land, assume that all the increase in product goes to remunerate the larger quantity of labor? If it be, indeed, true that more labor can now be applied to the land, why was it not before applied? The rent that was being paid before was paid on the basis of what the land was worth under proper utilization; and when it now comes to be so utilized and comes to produce what it ought all the while to have produced, it is still being paid at the same rate; it is not to be regarded as due in its entirety to the new labor but also in part to the opportunity now belatedly utilized:
If a tenant hires a piece of land and puts five doses of capital upon it when he ought to have put six, he pays a rent based upon the supposition that he will make a full economic use of the land, i. e., that he will get out of it six doses of product; but afterward adds the sixth dose, he only appears to pay no rent out of its produce, because he has all the while been paying a rent based upon the supposition that he was working his land with six doses (p. 147).
The land use is thus the basis of part of the price-determining cost; true, no more is now being paid for the land than was before paid, but the price now being paid for a utilized land service was before being paid for a non-utilized service; one does not have to pay more for the land when adequately supplied with labor and capital than before when it was inadequately supplied.
If I rent a piece of land for utility, in which all houses are three or four stories, and I shall not take into consideration the capacity of the property for building a three- or four-story house. If I choose to put a one-story house upon the ground, the rent I pay will be less than if I had built a three or four-story house. If I afterward add stories, it will seem that I pay no rent for this extra accommodation, but in reality I have been paying it all the time (p. 148).
Hobson's argument here appears to be unanswerable for the issues as they must present themselves, if "doses" are to be regarded as doses of the separate productive factors, rather than as doses of entrepreneur capital applied in every case under the direction of individual and peculiar entrepreneur initiative. The different entrepreneurs are
430
VALUE AND DISTRIBUTION
bidding for the use of entrepreneur capital—money, or goods, or credit, each in terms of money—or of instru-
mental or other goods as reduced to the money denomina-
tor, and so forth—is a bidding for the employment of capital. Each entrepreneur bid, whether for entrepreneur
capital, or for instrumental goods reduced to terms of entre-
preneur capital, is never accurately a bid to get more of
anything than he has already got. He bids for the use of his
employed labor, but rather to put with his entire situation
as a whole, with his productive complex as an aggregate,
in which he himself is a part. Thus, one entrepreneur will be directed by his bid to the purchase of a new piece of
machinery, another entrepreneur to the hiring of labor,
another to renting land or more land, another to the
purchase of fertilizers, hay barns, or work cattle or dairy
cattle, or other outlays for the "capital" nature etc.
And out of all the entrepreneur activities in the supply
of agents and instruments, and out of the entrepreneur
competitions for the control of these agents and instru-
ments, there is no less important activity than that of the
raw materials, hires for the instruments and agents, a rate
of time discount upon business capital, and a capitalization
for such of the productive facts as are susceptible of
capitalization.
Nothing, however, so far adduced denies that the entre-
preneur in making his bids proceeds upon what is essen-
tially to doing him good—out of the capital nature or
capital denominator of his own productive complex. But
even under this interpretation, Hobson is right in insisting that,
in accurate analysis, neither the entrepreneur's maxi-
mum bidding disposition nor any actual price output is
based upon any single agent apart and separate from its produc-
tivity of the dose fact under consideration. Any separate productivity of this sort the entrepreneur himself could not isolate. The whole complex must go together with his aggregate situation, his entrepreneur complex, and to become a constituent part of it; he can easily compute what he can afford to pay for the accruing advantages, due in part to his individual initiative and in part to those
—which the new fact bears in its own right, in part to its added productivity in its new setting, in part also to the added productivity which the old facts take on in their new association and relationship so far as the productivity is a matter of the interrelations of the different parts in
ATTEMPT RECONSTRUCTION 431
the entrepreneur complex—the entrepreneur being himself a part thereof—and a matter of the organization of the complex, there is a productivity which defies any attempt at distribution.
It remains true, however, that the entrepreneur can readily tell how much he would if necessary pay, and how much he must pay, and this is all that, in this aspect, is necessary for the validity of the dosing argument.
“Bun” Hobson says, “the ‘dosing’ illustration is vitiated by the assumption that he [the laborer] We may suppose that he [the laborer] is in full knowledge of the facts and has a full exercise of choice; as a consequence, he estimates that it just pays him to work five looms instead of four” (pp. 142, 143). Why say that the fifth loom pays him less or produces less or adds to his wage less than any other one loom out of the earlier four? “The fifth loom after it is added is found to be just as productive as any of the other four looms. The answer is plain. The fifth loom does not pay because its addition has injured his work with the other four looms.” That is, each of the later stock is producing less than each of the earlier stock produced before, but any one item of the present stock produces now as much as any other one item produces now.
That all this is true may be, and must be, admitted; but here again, there is no need for the argument as to the purpose of disturbing the dosing rule. Let it be assumed that, to the four looms, 500 in product was to be attributed, 125 for each loom; but that with the five looms only 600 in product was to be attributed, 120 for each. It is not true that in this second case the first four looms are to be credited with a return of 125 each and the fifth loom with a return of only 100. Under the new situation all are producing more than they did under the old one. But it is nevertheless true that the productivity of the fifth loom, 120, is achieved only on terms of reducing the productivity of each of the four looms from 125 to 120, with the result that their total productivity falls from 500 to 480. The loom with its 120 of productivity is only 100; this 100 is then, all that can at the outside be paid as the price of the value increase accruing to the whole situation by virtue of the hiring of the fifth loom.
**VALUE AND DISTRIBUTION**
The next objection, the third so called, is, in truth, not another objection, but merely another aspect of the first; as such, it must be accepted as tenable, but this only in view of the precise manner in which the issues in the discussion have defined themselves:
Professor Marshall, in treating the marginal dose of labor in agriculture (e.g., the last heaping applied to a field), admits that "the return of one man's labor cannot be separated from the others," but, he adds, "we ascribe to it all that part of the produce which we believe would not have been produced if the farmer had decided against the extra hoisting" (Book IV, chap. III, p. 144. par. 2).
Marshall's argument must be admitted to be fortu- nate; to state the best case for his theory, Mr. Hobson, believes, to state the thing affirmatively rather than nega- tively, the question not of what would be forfeited by the loss of one item out of the existing stock, but of what would be gained by the addition of another item. As the expense computation is forward-looking so must also the return computation be forward-looking.
But Hobson, in condemning, like Wieser before him, Marshall's method, and like Carver's backward- looking, negative method of value imputation to productive agents, fails, like Wieser, to see that no ultimate conclusion is thereby established. Wieser had somehow deduced the tenability of the forward-looking addition method; Hobson infers its impossibility from separate value imputation of any kind:
It is essential to productive agents that capital and labor shall all cooperate; it is impossible to assign to any one of them a product based upon the supposition of a separate productivity. Simi- larly, where there exists a necessary organic quantitative relation between the factors, no separate product can be put down to any single department.
But that this is true of the concrete commodity product does not necessitate the conclusion that it is true of the value product; and Hobson himself appears to assert that the increase of product at the margin or under the margin is due to the capital or to the labor application, since no valuables whatever is actually employed.
*As has already been argued, and as will later more fully appear,*
ATTEMPT AT RECONSTRUCTION
433
But that entrepreneurs do actually bid not only for additional supplies of competitive business capital, entre-
preneur capital, but also for different agents and instru-
ments of production, and that these are all to be valued
to terms of capital outlay, and so ranking under the com-
mon denominator computation of entrepreneur cost—must
still be admitted; and this leaves the dosing principle good
for all that, in the present connection, the rent-cost prob-
lem, as we have seen, is one which is not susceptible directly
with the dosing principle is in the attempt to apply it in
terms of the traditional categories of productive factors.
The tripartite division is altogether inadequate to the case.
But it is clear that the application of the entrepreneur
method of computation is not of necessity to concur
in the conclusion that all rents, or any particular class or
subclass of land rents, must be excluded from price-deter-
mining entrepreneur costs in their entirety. This is evident,
in entrepreneur computations servicable with reference to
expenses in general, or with reference to any particular
direction of expense. Thus we are constrained to deny
that "the law of rent is applicable" because "it is that the application
of the law of rent to the intensive curve of any single factor must be rejected as fallacious," but we none
the less agree that the argument is fallacious as used to show
that any form of land rent is irrelevant to the process of
price fixation.
THE VARIOUS RENTS IN THEIR RELATION TO VALUE
Hobson argues, as did earlier Say, and as does later
Fetter, that although land and labor are commonly regarded
as having concrete forms, we are prone to regard the pay-
ment for the use of capital as payment for the use of a
money value of a certain volume. But Hobson insists that
if any common law of price or of value is to be worked out,
both land and labor must be subjected to the abstract-value
Hobson's position is in fact correct with regard to the value product;
but his discussion under consideration fails a good way short of prov-
ing this. However, in a paper entitled "The Value of Land," in the Jan.
number of the Journal of Political Economy, Hobson has presented
a much more searching and in many respects a satisfactory analysis
of this question. See, later chap. xxi., p. 476, note.
434
VALUE AND DISTRIBUTION
measurement as cost values, or capital must be conceived as concrete forms of wealth serviceable in production:
The actuality of a science of industry as distinguished from a science of finance requires us to take the latter course, and to treat capital as consisting not in money but in concrete forms of wealth serviceable in production.
That one and the same treatment must be accorded to all productive agents and instruments ought to be accepted as past doubt, and it is to be counted to the especial honor of Hobson and Clark that they are first among modern economists in accepting this distinction. Their faults are still almost alone in this acceptance and emphasis. But in point of fact the uses of labor and the uses of land do attain a value which is distinct from their use as market-value items in the entrepreneur-cost computation. Hobson, indeed, has himself gone so far as to reduce these land uses to an abstract fund of productivity units. Other economists have attempted a similar form for labor, but labor, based on its being used up, does not require removal of all of its productive services to the common denomina- tor of market price.
But there can be no disappearance of these separate and concrete existences, through the mere fact that all arrive at the common denominator of exchange relations in market price. The fact that the customs and legal institu- tions of modern society make it possible for the laborer, as distinguished from the daily putting-forth of his labor power, need cause no perturbation; the labor use expresses itself in value precisely as do land uses and capital uses. The distinction between capital as a purely rental value but a capitalized value, remains despite that fact none the less concrete. So there is at this point no especial occasion for Hobson to insist upon the concrete ness of his theory. It is only when he attempts to carry this insistence so far as to call for the abandonment of the market-value expression of the capital goods and of their productive services. But this latter course is the one chosen, and it is here that Hobson fails. As it seems, he is compelled to abandon the common "business valuation of all capital as a valuation based upon the rate of interest" (p. 154). No relation is possible between this capital and our other factors of production. We must deal with the concrete forms which are thus valued!
ATTEMPT AT RECONSTRUCTION
435
And, as it seems, he makes both unintelligible and impos-
sible his earlier notion of land as an abstract fund of value-
productive units.
Land and capital, Hobson urges, require no rent or
interest for purposes of upkeep; there are indeed upkeep
charges, but all charges of this sort are met before the net
income of the land can be considered its inherent worth.
But it takes some wage to maintain labor in exist-
ence. "Thus it comes to pass that while the margin of
land is no-rent land, the margin of capital no-interest capi-
tal, the margin of labor is (say) 15% labor." But this 15%
wage does not in any sense correspond to interest or rent.
It is simply a wear-and-tear fund of labor, the expenditure
necessary to replace the labor power given out in a day's
work and to maintain the laboring population at their
present number. It is wages above 15% that correspond to positive rent and interest," and the fact that, unlike land, labor and capital have no sub-marginal representatives does not impair the setting; for there is a fund of capital safe to become actual
if the rate of interest calls for it [but, one infers, not safe to
relapse into non-existence if interest again fails], "while
any rise of payment to the marginal 15% labor will increase
the supply of labor power, either by raising the population
rate or by increasing the efficiency of labor, or by both"
(Fp. 135: 156).¹
¹Whether it be true or not that the prospect of obtaining interest
is a necessary condition to induce the creation of capital, it may dis-
tinctly be affirmed that interest is not necessary to secure the economic
maintenance of forms of capital that have been brought into existence"
(Jp. 135: 157). The maintenance of capital depends on its own inherent
deterioration of the land, or of the capital, or the slow consumption of
either these two elements. In other words, capital is a new accumulation.
Upkeep is a new capitalization; if rent or interest only cover this charge,
there is nothing left to overcome the difficulties present against
the perpetuation of capital. Interest is not necessary for the produc-
tional or intermediate goods. But it is fairly to be inferred that all
forms of capital are maintained by rent and interest, which Hobson
calls by Hobson as within cost. But here again it must be noted that since
the upkeep requirement applies to land equally with capital, there is
with land the same opportunity for abstinence and the same elasticity
436
VALUE AND DISTRIBUTION
And now to the question, What payments for use of land, capital, labor, enter as elements into market price of goods? it is replied that—
the same reasoning which shows that differential rents of land need not enter price shows also that differential payments for capital and labor do not enter price. For rent does not form an element of cost or price in agricultural produce, some of which is raised on no-rent land, so interest need not figure in the cost or price of manufactured goods, some of which are produced by no-interest capital. The cost of labor above the 1% depreciation fund need enter into the price of commodities partly produced by marginal labourers (p. 157).
It might, therefore, it is urged, be true than price should be determined by the cost upon no-rent land, cultivated by farmers obtaining no interest from their capital, and paying only a bare subsistence wage to their labourers:
But normally the last and most expensive portion of supply which reaches the market will be produced under conditions which exclude all rent and profit. . . . It will be more likely that the last portion of the supply will be produced partly on no-rent land, but paying an interest on capital and perhaps a wage far above 1%, partly on no-rent land, but paying a wage below its value on invested capital, partly by peasants paying rent or mortgage interest, but living on a bare subsistence wage. [And so] if the history of the market is to be traced, it would seem that if we closely traced, it might well be found that some quarters of it were raised on no-rent land, others on no-profit capital, others on subsistence wages; but that an average quarter of this most expensive portion consisted of a combination of these three elements—rental at all three. [And so we get as the determinant of price a cost] not necessarily the minimum of rent interest and wages, but the lowest average combination of the three. . . . Differential expenses of supply, in kind if not in degree, as with capital. And from the point of view of the individual calculation, either is immediately consummated in his own hands.
As to the 1% necessary upkeep for labor, it is to be said that whether it is paid out of wages or out of profits is entirely irrelevant as a business—a financial—consideration to any entre-preneur producer. One can afford to deteriorate his labourers where he could sustain them indefinitely without deterioration. But it is clear that the hirer of any agent has not the slightest interest to ask how much of his payment is for rent or interest and how much for upkeep; the pay-ment is in any case equally a hire and an expense.
ATTEMPT AT RECONSTRUCTION
437
production above this composite limit, whether they be rent, interest, or wages, will not enter into the market-price of the supply (pp. 158, 159).
Probably the fitting course for one who altogether fails to understand is to ask questions: What "late portion" is it which bears the cost of production divided into "parts" or "portions" raised under all possible forms of co-operation of productive agents, by "numbers of farmers working under widely different conditions"? Is it old, new, in new countries, some quarters raised on no-profit land, others on no-profit capital, others on subsistence wages? And even assuming that any portion of demand or of sup- ply, or any portion part of trade, determines the price, what was the sort of composite presentment if any such there could possibly be—to do with price fixation?
And if somewhere there were found a man working on good-for-nothing land, and paying his wages in the hands of laborers paid at precisely the cost of their keep, what warrant would there be for assuming that this man's cost would either fix or be the market price? Why, if his wages were paid at the cost of his keep, he might be grossly wronged under some forced-gain relationship or contract should not this exceptional farmer be making an especially and obnoxiously high rate of profits? Or, on the other hand, if his wages were paid at a rate below what he ought to lose, to prevent his making serious losses? And why should we, as the seeming beginning and occasion of all our troubles, have started with the assumption that good-for-nothing land is worth nothing? It may be worth nothing land at all, or that rickety, payless, valueless hay-rakes are capital at all merely by the fact of being in con- crete guise and in the similitude of farm machinery! And is it true that wages are paid at rates which correspond to interest or rent? For is there not at any rate this much of correspondence, that, under competitive condi- tions, each productive agent gets recompensed in some approximation to its efficiency in the process of value pro- duction?
Or if it be answered that this determining cost and this determining unit or portion of the supply are not the cost or unit or portion of the supply which are those whose then are they, and where are they? And if they need be of no one place, need they be of some one particular time, and why, or why not? And if they are some sort of
438
VALUE AND DISTRIBUTION
an average compounded of different marginal producers anywhere and everywhere, how compute supply as elastic through some composite-man's processes and choices?
And of what is the cost of production? That only 15s.-per-day labor would be computed as cost? That "the law of substitution has always to be taken into account" can rightly mean not that the actual rents and wages and interest are omitted from costs, but only that by so much as the cost exceeds in one direction in these directions is less, by so much must it be more in the others; excepting for differences in entrepreneurs all the different items of cost supply will enter into the price.
And as with land and rent, so with labor and wages; that the better lands are more highly paid than the poorer, or that the best laborers receive higher wages than the poorest, is not a matter of cost problem; for there is no reason to suppose that any grade of agents or instruments is better or worse paid than any other, in proportion to the efficiency rendered. The cheaper lands or the low-paid laborers may be at a disadvantage; but incon-
laborer rents, like land differentials, have no relevancy to the cost problem. If the 15s.-man's wages enter into price, so must the 20s., that a man $ is efficient gets, enter into price, and this difference between 15s. and 20s. is a cost.
And as with the better lands so with the better laborers or the better entrepreneurs; the best, as easily as the poorest, may be nearest to the margin of alternative occupation; but not the outlier. It is less, but not collisional reckoning of some alternative condition present, but the disparity as it is, has significance for the purposes of competitive cob.
CHAPTER XXII
DISTRIBUTION BY VALUE PRODUCTIVITY: CLARK
The central thesis of Professor Clark's *Distribution of Wealth* is that the different distributive shares are the correlatives of productive efficiency, and that under static conditions and with frictionless competition these shares would be accurately correlative. As corollaries from this
"John Bates Clark, The Distribution of Wealth, a Theory of Wages, Interest and Profit," New York, Macmillan Co., 1899, p. 30.
it is necessary to state that the distribution of the income of society is controlled by a natural law, and that this law, if it worked without friction, would give to every agent of production the same proportionate share of the product. However wages may be adjusted by bargains freely made between individual men, the price of labor itself is fixed by the law which is here claimed, to equal that part of the product of industry which is traceable to labor itself, and however interest may be adjusted by similarly free bargaining, the price of capital is that fractional product that is traceable to capital."—Ibid., Preface, p. i.
It is unnecessary to stop to discuss, or even to appraise, Clark's limitation of the subject-matter of the problem to concrete, material goods; e.g.,
"By wealth is meant those sources of human welfare which are material, and which are capable of being converted into material things that are appropriate and useful in this specific way; useful, are economic goods" (p. 45). "The great income of all society—that which is derived from real property—is distributed among all, of some use" (p. 13). It is unnecessary also to discuss his consistent utilitarianism; he does not claim that the law has any pronounced and unquestioning influence upon "natural law" as a causal, directive, and competitive agency; e.g.,
"If there is any law at all by which the income of society is divided into wages, interest, and profits; if so, what is this law? That is the question which I have been trying to answer. In short, a deep-seated natural law at work amid the confusing struggles of the labor market" (p. 6). "Where natural laws have their way, the object sought after will be attained; but how far this is gauged by the actual product of it [p. 2]. "If the law on which property is based operates in such a manner as it actually works," etc." (p. 9). "For the present, be it noted that exchanges divide and subdivide industry; they range its forces into groups and sub-groups; they make them more or less dependent by natural law" (p. 10). "Dynamic changes are in no sense and broader sense nature. Nature herself is continually disturbing the régime of
439
VALUE AND DISTRIBUTION
position or as steps in the argument by which it is reached, the following propositions become of especial significance natural justice but competition is trying to restore it" (p. 9), etc.
And it is unnecessary to discuss therewith the adoption of the asso-
ciated ethical view inclining to identify natural law with provi-
dential justice, since this view has been attacked both as a
scientific fact; it is fairly to be said that while all these are issues, they are not the main issue. The main issue is that Clark's theory of distribution must stand or fall. And while in this connection it is only just to say that in many of these formulations, such as, "A social order is right when it is one in which people work
without friction, etc., " Clark means in the main by "social"
and "natural" what he often has in mind something very appreciably different, though he himself might be unable to say precisely what---a mysterious, provi-
dential, immutable, and unchangeable order of things, which is indec-
ent, by which it comes about, among other things, that all these
natural relations are so arranged that man can live without friction,
to justify the ways of God to man, at the same time with furnishing a
deep scientific basis for an optimism which would otherwise and in
another way be impossible.
It is nevertheless to be said that this optimism, limitless in
quantity and unlimited in quality, is not only a very dangerous thing for the
writer to justify things which are least offensive when viewed apart
from their moral quality, has yet no necessary part or share in the
argument of this book.
And further: However much dissent or question may later come to
be expressed as to the fundamental thesis that the "income" that attaches to any productive function is gauged by the actual product of it; and that each person receives according to his contribution,
and to each a corresponding reward, such is the natural law of
distribution. It is true that this law does not always hold; it observes
that "more hangs upon the truth of it" than does really so hang. With this thesis once established he must still be indefinitely distant from Joubert's assertion that "the right of society to exist is not a certainty," and clearly untrue that "the right of society to exist in its present form, and the probable permanence of its present form, are certain." One
might conceiv in Clark's thesis and be yet the most radical of socialists. That rent paid as the precise correlate of the pro-
ductive efficiency of land is a matter of fact, and that it increases ownership in land. Carver has made this clear in his review of the
work under discussion.
The right of the present social order to exist depends upon the
law which governs not functional but personal distribution. Our only interest here lies in the question whether we have any reason to
see the vastly more important question of personal distribution. We need
to be shown how far we can go in saying that each individual shall receive from each individual producer the share which he individually creates,
and no more ---17---. In "Clark's Distribution of Wealth," Quarterly Journal of Economics, Vol. XXVII (1893).
But here again Clark's argument and conclusions are in no wise involved; the substance of the work is otherwise.
DISTRIBUTION BY VALUE PRODUCTIVITY 441
as indicating, for present purposes, the trend of the theoretical analysis.
1. Capital is regarded not as concrete capital goods but as an abstract, homogeneous value fund.
2. For static purposes, land is assimilated to capital goods and funded with them into abstract capital, socially viewed.
3. Labor is also funded into value-productivity units.
4. Value expresses marginal utility, and is fixed and determined by marginal utility.
5. All factor rents or hires are equally included in costs and made of price, and have relevancy to price; but these remunerations are rather value-derived than value-fixing—this last, however, not quite definitely beyond misinterpretation.
6. The possibility, under the funding principle and from the group point of view, is asserted of isolating the separate productivity of units of capital and of units of labor.
7. Distributive shares are determined by the value-productive power of the final unit of the funded productive factors.
8. Society is viewed as an organism, and derivative concepts of group pain, group pleasure, group utility, group cost, etc., are adopted and emphasized as necessary to the argument.
9. For dynamic purposes—purposes of retrospect or of prophecy—the tripartite division of productive factors, in the sense of mechanical and technological categories, is made important.*
*Propositions (3), (4), and (5) should in this connection call for no further discussion, excepting possibly to the extent that in the consideration of the remaining propositions these earlier doctrines may be deemed to be established. In any event, they need any proof by quotation or citation be adduced that the positions as above formulated are correct.
With regard to proposition (4) also, when once it is established as actually held, neither criticism nor discussion would appear to be necessary.
In the interests of space and as intimately associated with (4), proposition (8) will best come to here for such citation and quotation
**VALUE AND DISTRIBUTION**
Propositions (5), (6), and (7) present the issues with which the immediate discussion is chiefly concerned.
as may reasonably be called for, and for such discussion as the particular form of presentation may seem to demand:
"In every stage of economic evolution wealth consists of useful material things; but their utility is of the kind that we may call specific, i.e., they are useful only in so far as they belong to us. Outward material things that are appropriate and, in this specific way, useful, are economic goods.
"If one man has more of any good than another, he usually so to another, and is therefore in itself exchangeable" (p. 41). "If men do in fact use a commodity in such a way that its specific utility diminishes as they get more and more of them, then what they will give for any one of them will be gauged by the law of final utility. The propositions of the modern theory of value correspond with the facts of life, the theory of value being a theory of final utility."
"The primitive [Isolated] economy . . . cannot test final utilities in a market, for it has no exchanges. Can it not, then, test them at all? The answer is that it can. We must not hastily see that it does this, and that the purpose is exactly like that for which we have set up the theory of value. The law of final utility belongs in the first division of a theory of economics and has to be assumed in the second division (p. 43). The law of final utility fixes the point at which one producer stops creating a product and begins making another. A modern labourer, with money in his pocket, goes into a shop and buys a loaf of bread. He makes purchases and to spend each dime where, in view of the supply of different commodities, he would make his purchases. In these markets and prices are, therefore, modern phenomena, the study of which has no place in a division of the science devoted to universal truth; but when we come to consider how these laws operate we find that they are made in a modern market also directs the production of the isolated man away from what he would produce if he had no other life these laws direct the social demand for different goods offered in the shops; but in primitive life they control the manner in which a man hunts or fishes or grows food or makes tools; and they determine the most good. The law of final utility is common to both economies" (p. 45).
"The picture of an isolated man turning his own labor from making one thing, of which he now has a supply, to the making of a thing if he does not have a supply, is one that is common to all stages of modern life which is in danger of being overlooked. Through the laws of value we learn that when a man turns his labor to turning its collective energies from one direction to another, according to the law of final utility. Markets and values afford the mechanism for doing this. When a man turns his labor from making one thing to making collective energy for the making of some thing till it has enough of it, and then turns it back again on making something else. Through the mechanism of a falling price, society is warned to turn its energies to the making of something else; and its whole procedure is nothing more nor less than doing what an isolated man would do, if
DISTRIBUTION BY VALUE PRODUCTIVITY 443
It would not, perhaps, be fair to ask whether Clark regards costs as determining value, or rather value as he found his want of one commodity becoming satisfied. If we then individualize society, if we make it so to be in its entirety one isolated being, and if we consider that each of the members of this being, as well as the being itself, consists of independent beings as one organism, we find it doing what a solitary man would do, under the influence of the law of diminishing utility. Putting aside the question whether this is a correct organon in estimating the importance to itself of each of its own products, it is clear that when the product of any one member of the society is used up, the outcome of his activity, society acts as a solitary man would act under the influence of the law of diminishing utility" (p. 45, e.g.).
"Marginal utility", "the principle of final utility", by which values are fixed, is universal in its scope" (p. 47).
"By a law that Austrian studies have made familiar, the value of any article in this series of goods of one kind is based by the utility of the first unit consumed" (p. 48).
"If there are marginal laborers, in the sense that there are marginal quantities of labor employed in producing any particular good, these marginal men are likewise in a strategic position; for their product sets the standard of every one's wages" (p. 60).
"If there are no marginal laborers, then he gets from him an inducement is offered to other employees to take him at a better rate of pay, and thus to increase his earnings. The total quantity produced and the wages of social labor equal the product of a composite final unit of labor" (p. 61).
"How is this product to be measured? Take away one social unit of labor, and see what is lost by the withdrawal of it; or add one such unit to the existing stock. In either case it is possible to note the amount of product that is separately due to a unit of labor and to no other agent. Let us, then, withdraw what we have called "marginal labor". This will leave us with some portion of some labor from every industrial group that the community con- tains. We shall call this "composite labor". It includes all workers, weavers, etc., in carefully adjusted proportions, causing a final unit of labor to vanish from every specific industry (p. 156).
It will be especially in place to note here that Professor Seligman's interpretation of the law of supply and offer and to value is substantially identical with that of Professor Charles.
Under the "General Law of Value" it is said: "Value is at bottom determined by cost." But cost must always be considered as a variable; and therefore cost cannot be taken as constant. Cost must therefore be studied from the point of view of marginal utility; that is, how much more or less can be supplied to satisfy the marginal increments of demand. This is only another form of stating that the fundamental explanation of value is marginal efficiency or the capacity to satisfy wants. See also Professor Jevons' "Principles of Economics", p. 36.a). "The contribution or efficiency is the positive cause; the cost of production adjusts itself to this.
The cost is not necessarily constant.
"Cost of production is thus only a partial, and even then approxi-
444
VALUE AND DISTRIBUTION
determining costs; his analysis goes neither far nor deeply into the value problem. To be strictly relevant to his mate, explanation of value; marginal efficiency is the universal and ultimate explanation" (p. 65).
"There is no such thing as silver below the surface that is not mined because it will not pay; if the marginal efficiency or value of silver above the surface is increased by a tax on its production, it will be marketed, and the new marginal cost of production would adjust itself to the price. The price would not rise because the cost increased; but the increase in cost would be offset by the increase that would now be the new point of marginal efficiency" (p. 66a).
Professor Seligman's work is, however, of especial interest in this connection by virtue of the thorough, consistent, and systematic manner in which he has attacked the problem of utility. In view of its logical limits—whether of truth or of absurdity—as an interpreta-
tion of the facts, it may be said that it does not attack them, as is pertinent to this particular aspect of Professor Clark's doctrinal position may well be merged with a discussion having pri-
mary concern with the theory of distribution. The following is, for the most part, either reproduction or abridgment of the views expressed in "The Theory of Distribution" (see also my con-
nection. (See "Seligman, Social Values," Journal of Political Economy, March, 1906.)
It is true that, as also with Clark, much difficulty to deter-
mine the precise significance of such portions of the discussions as are devoted to the determination of individual utility, individual marginal utility, individual demand, entrepreneur supply computations, entrepreneur cost, individual sacrifice, individual income, etc., arises from their being aggregative, synthetical,
competitive, and differential, etc.; for in the main, all this individualistic discussion is concerned with aggregates as really and as prac-
tically beside the point; or, at least, as regards its relation to the point—not precisely side issues truly, and presumably not irrelevancies,
but rather those which are so closely related to the general subject matter of the doctrine, the realities of the objective business situation are sought elsewhere.
For this reason, it is taken as Seligman's intention to abandon any of his many formulations of the strictly individualistic sort; it still remains true that he does not deny that there is a measure of utility of different quantities of goods" (p. 12); that "when we speak of the value of a commodity, we think not of the total utility of the quantity consumed but only of its marginal utility"; that "the value
that of a definite quantity of other commodities" (p. 179).
But though this may be true in some cases, there must obviously be difficulty in finding someone to make the comparison: a pint of champagne sells for the same price as a sack of flour; are they therefore equally valuable? Or again: Is one unit more valuable than merely an equality of command over purchasing power? Equal utility to two persons does not necessarily mean equal value to both;
it is evidently not enough to assert that "value is not merely
the expression of utility in general, but only marginal utility,"
DISTRIBUTION BY VALUE PRODUCTIVITY 445
discussion, the question,—substantially the same question, nevertheless,—should be formulated to ask whether distribu-
(p. 108). If values are really to be resolved into a common denomi-
nator of utility something further is waiting to be done.
The solution as offered runs:
"It is evident that each sets a value (market value) upon things. . . . If an apple is worth twice as much as a nut, it is only that the apple has more utility than the nut. But those who express individual preferences, that the desire unsatisfied by the lack of an apple is twice as keen as that unsatisfied by the lack of the nut. When this is done, we have social utility (p. 107); it is the expression of social marginal utility" (p. 108).
Since the market price is determined by the comparison of diverse goods, and since this comparison is ordinarily made in society by their transfer from man to man, it is clear that the value with which exchange is carried on does not represent any social utility.
Value in exchange is nothing but the expression of its true value to the members of the social group, that is, of its marginal utility" (p. 83).
Exchange power is based on the comparative estimates of direct social utility and indirect social utility, that is, on the estimated utility that values have in society" (p. 108).
Thus, it appears that what we call "utility" that which underlies and explains value is not individual utility, but social utility. And we have, as we shall later see, in addition to this social-utility concept, another one which we may call "social cost," namely, social supply, social surplus, diminishing social utility, diminishing social return, a reduction in social utility due to increased cost; and finally, as the goal and summation of all this, social value, that is, market value. This will evidently bear looking into.
Some linguistic uses connected with collective nouns will offer a point of departure for our discussion. A noun like "a battle" or "an army" is used as a collective noun when it takes the plural verb; if regarded as a collection of units, it takes the plural verb. And so we say, "the committee was unable to agree"; "the committee was unable to agree" (with one another); "was unable to agree" (with the conference com-
mittee); "the army were marching" or "was marching"; "the crew was
or "were marching"; "the crew was unable to agree" (with one another); "the agreement has to be done by
more than one person."
Now, in many cases, though the act or the situation asserted is really one of each individual by himself, there is no occasion for instating it in singular form; thus we say "the battle was eating its dinner." We are involved in saying that "the battalion is eating its dinner;" it is a shorthand way of saying that each member of the battalion was able enough to think of a battalion as a unit, and the act of dining is a simple one in which all join, and in which all comport themselves in pretty much the same manner. In such cases as these where the interest proceeded upon, the purpose in hand, no importance attaches to the fundamental separation of the activities, and to their entire lack either of physical unity or of purposive co-operation; they are
**VALUE AND DISTRIBUTION**
tive shares are to be explained as derivative from value or as causes of value; is value, that is, the intermediate step simply similar—roughly simultaneous—and are thought of in block. True, one man eats rapidly and another slowly, some little and others much, but this difference is not due to value, but the expression serves, and implies its own limitations of accuracy.
And so of an army when we say that "it marches," no doubt is even false, because the men who march do not march by themselves but with their own muscles, use up his own tissues, and that presumably many are both, which is true enough. But the fact that such differences signifies values for the purposes of the thought in mind; each man is separately getting ahead, moving along, like all the rest; and so we say, "It marches." And yet it does not follow that if these processes inaccurate, were any pervertently bent on misinterpreting it. But when in concert they are all marching together, and one man is behind another's toe, or has a stomach ache, there is obvious difficulty. These things are not done jointly, co-operatively, by aggregates, and will not bear comparison with the aggregate of the whole. The collective idiom is obviously manifest.
And so also with public opinion, the preference, or habit, or custom, or convention, of society; and no harm need come of it. Despite the fact that none merely think nor choose in the manner implied, they are still members of society and have their ideas, and yet are members of society entitled to be included in any exact formulation of social laws. For it is only those who belong upon majorities of "most-everybodies" that is, no harm need come of it; if only there were not people to take the notion of a "social mind" seriously. It would be absurd to suppose that anyone could fail to accept as sober fact, a mere figure of speech, or at best a loose analogy drawn from the human body. In other words, if we are told that it is to be charged—or credited—that the society-as-an-organism formula has found its way into economic thought. And thus hereby a doctrine is made out of a mere metaphor. What is the use of reappearing; for have we not need of normals and averages? Else our doctrines would be mere fancies and fancies would be useless.
And so, by the aid of the sociologists, through the magic of the society-as-an-organism incantation, a resurrection miracle has lately been wrought in economics.
One hesitates to approach the invincible task of assigning primacy in this new school of thought; for that there is a new school, and that it has come to include a passably generous membership—somewhat heedless indeed—of those who have never had anything to do with any form of economic science; is the excuse, so far as there can be any for the present state of affairs. But I am afraid that Professor Seligman is the first writer who has seriously undertaken to carry the doctrine to its logical conclusions; and thus it necessarily comes to be true that he has succeeded in doing so. He has shown how practical bearing, impersonal, theoretical, and general in its reference, rather than primarily an examination of Professor Seligman's doctrine as such.
And yet one asks one's self why, if it is all thus easy--this magic word "social" making all things plain—if the heart of the mystery is
DISTRIBUTION BY VALUE PRODUCTIVITY 447
toward explaining distribution, or are the distributive shares to be taken as the intermediate step, and value the goal?
thus easily plucked out, why all this other talk of gross profits, neces-
sary and minimum cost, individual cost, individual utility, marginal
increases, etc., etc., is so much nonsense? What is the real problem here?
and the like? Why, indeed, any talk at all! Everything will explain as a
social resultant, if this passes as explanation.
In this doctrine of social cost, social sacrifice, and social value,
one must distinguish between the two kinds of considerations of all
allegiance to outlay cost, and a return to the distinctively pain-cost and
pleasure-balance terms of analysis.
"Cost" is a measure of value; but it is not, as
Ricardo thought, individual cost. Marginal utility determines value:
but it is not the same thing as marginal utility. The former is
"utility measure value, because . . . marginal cost is always equal to marginal utility." (p. 103)
"The sacrifice imposed upon society to secure anything is . . .
the exertion needed to replace it. . . . Thus, when we speak of
world costs, we mean that the price paid for something is its
value influenced by cost; we mean that value is influenced by cost
of production. We do not mean that the sacrifices made
upon any one individual, nor only that of the sacrificing party,
cost embodied in the commodity; or rather, the sacrifices, or cost, to
the individual does not result in a return to the community."
(p. 107)
"In reality, whatever the rate of exchange is only the
social utility and the social cost of which the marginal degrees are equal. If a knife exchanges for a book, it is because the demand in the community for knives exceeds their supply; and therefore the cost of parting with a book is equal to the marginal pleasure of society in getting a knife. But what is the reason for this equality? (Is it because
a knife exchanges for a book?) To put it more accurately, a knife will exchange for a book only because the sacrifice to society in making the knife, for which it receives a book in return, equals the pleasure received from making the book, for which it receives a book in return. This is true even though there may be less than the pleasure, but there will always be a marginal individual to whom pleasure and sacrifice are equal. The marginal pleasure in the aggregate will be greater than the marginal sacrifice in the aggregate. The balance or equilibrium is between the pains and pleasures of the
sum of individuals. The real cost to any member of society which influ-
ences value is not the subjective cost to him." (p. 107)
Now let us consider what happens when marginal pleasure or
the aggregate equals the marginal pain in the aggregate. What is
to be done? "There can be no such thing as equality of pain.
There is always a marginal person, it is said—one person "to whom
pleasure and sacrifice are equal." But this man can have little to do
with this matter. He has no interest in it. It would be impossible for
any member of society which influences value is not the subjective cost
to him; and, in point of fact, there is after all, no such man, for
"it is only the social utility and the social cost of which the marginal
448
VALUE AND DISTRIBUTION
Or, again, is the fundamental problem one of value or is it one of distribution? Or, finally, are the two problems not really two but one?
degrees are equal" we really think not "of the sacrifices imposed upon any one individual, but only of the social sacrifice or cost, embodied in the commodity." The sacrifice of the individual is the result and redress of the sacrifice to the community.
Nevertheless, it is clear to Professor Seligman—and to us—that there has been a great confusion in this matter. He says that "the estimate put by the individual on one commodity as compared with another is not necessarily the same as its true value to him. In exchange is nothing but the expression of its true value to the mem- bers of the social group, that is, of its marginal utility" (p. 163).
It is obvious that this statement is false. A commodity may come to have a greater indirect utility to its possessor than it has direct utility to others. It may be that he values it more than less than it has to him. And in this computation is he, or he is not, a part of society? "Its indirect utility to me is the result of its direct utility to others. But I do not know whether my own direct utility is the cause of the indirect utility to the possessor? And how did my own utility differ from other individuals?" The members of a social group, then, can "direct" marginal utility to some one may have existence!
It must be understood that with this latest school of value, as with its predecessors, the principle that each thing has a supposed fixed price is, for whatever it is worth, freely accepted; but all the while with this difference, that demand, as conceived by this latest school is not merely a measure of supply; it is also a measure of each with its own particular psychology and its peculiar explanation: "The demand for a thing is the aggregate demand depending on the social utility" (p. 20).
Likewise cost is essentially not a matter of individual outlay or sacrifice. It is rather a matter of preference and profit. Cost varies accordingly as individual preferences and profit may direct; cost is rather a matter of preference and profit.
"Since cost is a form of demilitarism, it follows that the real cost of importance affecting value is social cost and not individual cost." Value is "a measure of social sacrifice and not personal sense." . . . Evidently not of individual sacrifice. A street sweeper will pay for his services less than a skilled factory hand. Society is more willing to spare the former than the latter, for to replace the one society may give up more than to replace the other (p. 193).
Note here, precisely as when we were introductory discussing individual demand and individual cost, cost was made a derivative discussion.
DISTRIBUTION BY VALUE PRODUCTIVITY 449
It is evident that Clark regards distribution as in part a process under which an aggregate of value is, as product, apportioned to aggregates or groups of producers; in part of demand, so here, in the social computation, social marginal utility is presented as adequate and controlling for value; but meanwhile our real problem is how we can explain the fact that the social explanation obtains admirable and adequate recognition and expression.
"All value is the reflex of social marginal utility. We have now to study the nature of the social forces which operate to translate into actual possession the products of the efforts of individuals that "prize the group" (p. 22).
If this proposition is accepted, something will remain unasked; but the difficulty and the regrettable fact of it all is that the transition over from the individual psychology to an alleged social psychology is marked by a complete failure to understand what it means to take us for the purpose:
"Costs may be defined as those cost--not pains (or their money equivalents) taken, but pains saved. It is only because individual cost does tend adjust itself to the socially necessary cost that we can roughly speak of price of anything demanded as its value on the market" (p. 244).
"Socially, however, it is not the whole of what pro- chasers as a group are willing to give rather than make the article for themselves. If the individuals cannot reduce their cost, they will stop producing altogether. The law of supply and demand says that "prices will move" (p. 344). And so then, it appears that individual cost does affect the quantity produced. But this is not quite true. What is being given up, because what the producer can do, the rest of society can, if necessary, do. It is in this way that an equivalence is brought about between individual costs and prices. In other words, it is not an equivalence that cost of production may be said to influence value" (p. 344).
But in view of the fact that, by assumption, some individual pro- ducers will not be able "to reduce their costs and will stop producing," while some consumers will not be able "to pay more than they are able of the proposition that what some "producers can do, the rest of society can, if necessary, do"? Are all consumers to be accepted as like all the producers? Or are some consumers different from others? Are these differences alike?
Or perhaps the following should be cited upon this point:
"Cost of production does not mean individual cost. Value, as we know, is a social conception; the real cost of production which affects what is worth at any particular time."
"The law of exchange may be equally well stated as the law of comparison between two persons. I would rather have my furnace man than I could save much coal by tending the furnace myself; yet I prefer to look after my business, and let him tend the furnace because he knows how to do it better."
"The important point is not that a commodity costs the producer something, but that it saves the consumer something. It may save one consumer more than another, but its value depends on what it saves the
450
VALUE AND DISTRIBUTION
as a process under which the original group holding or imputation of value is subdivided among the smaller and subsidiary process groups representative of the progr-
social group as a whole. This saving of social cost is what is meant by socially necessary cost. . . . The cost to the individual pro-
ducer will be less than the amount which he would have to pay for the amount which the purchasers as a group are willing to give rather than make the sacrifice.
Evidently, however, this reduces individual cost to social demand; thereby cost comes, not to fix price, but to be fixed by price; to make the determination of price dependent upon the willingness of the social group to pay. This is the cost for the economic side of the analysis. And though we may find it difficult to apply this principle, we can further devise to explain that social price which lies behind individual cost.
Or the explanation for shutting out from consideration all indi-
vidual peculiarities may be that they are assumed, for theoretical pur-
poses, more or less negligible. But this is not so. As has already been suggested, we must return back to the reasoning by averages characteristic of an earlier, and possibly a better, time. But even then we cannot escape the conclusion that in our own day our author has himself spoken decisively in condemnation of the economic theory of averages. For it is clear that while in his "average-man" analysis once adopted, there is necessarily an end to all talk of margins, social or otherwise.
The reason is that these differences between individuals are fully recognised, but are regarded as somehow merged and lost to the social average. In other words, it is assumed that any use the reasoning repudiates the economic man, declines any overt and systematic acceptance of the method of averages, and adheres resolutely to the "average-man."
Just as upon p. 23 it was declared that "a dollar is of more worth to a poor man than to a rich man--its marginal utility is greater," and that "the price of an old master or a mediaeval missal will often be higher than that of a new one," so it was nevertheless contended that "in the ordinary transactions of life, where we deal with men who are not perfect economists, the worth of money may be neglected," so now, in the discussion of the theory of interest, similar differences are likewise merged and the method applied, and thus a more energetic analysis of the price-cost type.
"The only point in which most people differ from me is by saving, by waiting, by forbearing (p. 319). . . . The problem is one of marginal forbearance, that is, sacrifice at the margin where he must wait or save or forbear. Wherever he does not do so much is, the more remote is the margin where he will have to decide. The saving of money is something very different to a rich man and to a day labourer" (p. 320).
"When, therefore, we say that interest is the result of forbearance, we really mean that interest is the result of marginal forbearance, or forbearance at the margin. . . . This marginal point will indeed
DISTRIBUTION BY VALUE PRODUCTIVITY 451
ress of the commodity through its various stages from raw material to finished product; in part, also, as a process of partition of the subgroup holding of value between the individuals who have contributed to the production of it, so clear whether this threefold view of the total distribution process is purely logical and analytical—merely one way of looking at the case—or is rather intended as realistic-ly descriptive? The answer is that it is both, tracing the actual sequence and direction of the causal forces. But it is none the less important that an interpretation should be arrived at upon this point.
be a different one for the rich and the poor (but see different ratios-what for the rich and what for the poor). This is because the rate of interest as little as the relative wealth of the purchaser, but much more than his utility. The price of wheat is the expression of its marginal utility to the wheat-eating group; the interest on capital corresponds to the difference in the marginal utility of capital between the capitalist and his capital-using group. Value in the market is social value" (p. 308).
And all of this coherence logically with the doctrine of an earlier page:
Cost of production is the measure of value; but it is not, as Ricardo thought, a measure of individual utility; but it is not, as Jevons thought, individual utility. Both cost and utility measure something other than "marginal cost is always equal to marginal utility" (p. 198).
But it is more than possible that more than justifiable space and time have been devoted to this social theory of value. However, to query why, with all these social utilities, demands, sup-plies, costs, etc., we do not find them expressed in money values; however, to query why, with all these social utilities, demands, sup-plies, costs, etc., we do not find them expressed in money values; however, to query why, with all these social utilities, demands, sup-plies, costs, etc., we do not find them expressed in money values; however, to query why, with all these social utilities, demands, sup-plies, costs, etc., we do not find them expressed in money values; however, to query why, with all these social utilities, demands, sup-plies, costs, etc., we do not find them expressed in money values; however, to query why, with all these social utilities, demands, sup-plies, costs, etc., we do not find them expressed in money values; however, to query why, with all these social utilities, demands, sup-plies, costs, etc., we do not find them expressed in money values; however, to query why, with all these social utilities, demands, sup-plies, costs, etc., we do not find them expressed in money values; however, to query why, with all these social utilities, demands, sup-plies, costs, etc., we do not find them expressed in money values; however, to query why, with all these social utilities, demands, sup-plies, costs, etc., we do not find them expressed in money values; however, to query why, with all these social utilities, demands, sup-plies, costs, etc., we do not find them expressed in money values; however, to query why, with all these social utilities, demands, sup-plies, costs, etc., we do not find them expressed in money values; however,
But if wages rent and interest are admitted and accepted as categories of distribution they must follow that profit a surplus income over cost or other is also an individual category and that these incomes are distributed by means of shares to the recipients—are obviously costs to the entrepreneur,
and as such will affect their prices and therefore their market value so far at least as market value is affected by supply influences. In short whatever may be said about the law of diminishing returns in urging that no individual but only social cost is relevant to market value the school will be under obligations to work out as factors of social cost profits rents and interests profits and interests and,
and as surplus over social cost a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit a profit
In fact however it does not do this For as soon as we turn from the cloud land into separate clouds into separate shares we find ourselves to have descended from the cloud land mysteries of transcendental economics to the old and beaten paths of the traditional analysis.
452
VALUE AND DISTRIBUTION
If the wages and rent shares, as determined inside the subgroup, are fundamental and ultimate, the way is clear for an adequate and consistent entrepreneur-cost analysis. But if we start with the market value of these shares, we must forthwith set about to find what determined these rent and wage shares. Or if we can start with the market value as cause, each group share and sub-group imputation and inference will follow logically. In this case, however, the entrepreneur of course being taken to be present; but the difficulty will still remain of explaining that original value with which the explanation has improvidently set forth. For, even if we assume that subjective or imaginary or subjective value be invoked, there is still no help, inasmuch as each of these assumes as fundamental to it the existence of a supposed cause.
But how does Clark present the case and dispose of it? So far as, from discussions mainly introductory in character, a reply may be had to questions of this sort, it must be deduced from the following:
There is a kind of distribution that does not fix the rate of wages and rents, namely, a distribution how much one industry, as a whole, including its laborers, its capitalists and its entrepreneurs, shall get, as compared with other industries. It determines whether one whole shall get more than another whole or vice versa than another. This is an intermediate part of the general division of production, and it is accomplished by means of prices. When wheat, for example, is high in price, the farming industry is well paid, as com- pared with other industries; when wheat is low in price, it is ill paid. If what we have in mind is the so-called "market price" of an article,—the immediate price of any given supply of an article,—this kind of value governs what we may call group distribution. If steel for example is high in price, it is well paid by the group that produces it. This income distributes itself somewhere in the group; but how much of a laborers get, and how much capitalists and employers get, is a question that we do not now raise. The same thing happens with respect to all other things that go on within the groups. Group distribution is a preliminary division of this social income, and it deals with branches of industry in their entirety. The terms of this primary division of the social income depend upon the relative values of kinds of goods. Farmers want wheat to be dear, as miners want ore to be dear, etc. Prices then, fix the incomes of these groups.
*Clark, op. cit., p. 12.*
A page from a book.
DISTRIBUTION BY VALUE PRODUCTIVITY 453
The creation of such a general stock of commodities for use is a great synthesis, which goes on in a systematic way. One group of producers makes the article A, another group makes B, another C, etc. A is sold to consumers that is paid for by the supporters among the entire population that make up the market. The sale proceeds from this sale are divided in the same way, among all who have helped to make this article. The prices of completed articles thus fix the incomes of groups in their entirety. These groups are in an equally exact way related to subgroups within them, whose members thrive; and if the difference between the price of wool and the price of cloth is large, manufacturers thrive. It is market values that fix the income of each group.
Neither of these price-adjusting operations, however, directly fixes wages and interest. This is the final and critical part of distribution. It takes place within the subgroups, and it constitutes the third and last stage of distribution. In this stage, the income falls to farmers, manufacturers, etc., as such, have to be further subdivided; for a share must be paid to every laborer and to every capitalist.
This distribution goes on in three distinct stages. There are to be made a division, a subdivision, and a final subdivision of the social income. The first division fixes the income of industrial groups; the second fixes that of subgroups, and the final subdivision assigns wages and interest to individuals. The shares of the groups and those of the subgroups depend entirely on the prices of goods, and therefore the fixing of these values results in the adjustment of the terms of group distribution. Let us call $A$ a certain complete product, say bread; and let $A$ represent raw material, the standing wheat of which it is made. $A$ may then represent the wheat as threshed, or as flour, or as bread or as a loaf.
The difference between the price of $A$ and that of $A'$ determines the income of the flouring industry, etc. The income of each subgroup in the series, then, depends directly on prices.
A phrasing like this does not explain prices, however, brings us to what are called natural* or "normal" prices. These are the values, expressed in terms of money, to which, in the long run, market values tend to conform.
A certain degree of uniformity within the sphere of group distribution establishes the normal standards to which market values tend to conform. Market prices fix the incomes of the different groups, as such, and so control distribution in its early stages.
A deeper force, and one that also acts in distribution, controls normal price. Market prices are the cause of group distribution;
PHOTOGRAPHIC PRINT
454
VALUE AND DISTRIBUTION
normal prices are the effect of a certain phenomenon of distribu-
tion. . . . The movements that make prices "natural" are, in fact,
efforts on the part of different men to get their natural shares of income.
In view of the fact that it is said that "the social process of production includes exchange and distribution," and that "the theory of value and that of group [italics the present writer's] distribution are one and the same" (p. 24), the general trend of the doctrine of the foregoing might seem to be that the value field and the distribution field are one field, and that the two problems are but dif-
ferent aspects of one problem.
But on the whole, the position seems to be other than this, viz:
1. That market value controls group distribution;
2. That the market value distributed to the group con-
trols the values assigned to the subgroups;
3. That normal values control market values;
4. That (through fixing wage and interest incomes, and thereby fixing entrepreneur costs?) the distribution inside the group controls normal value.
Reading through the quoted passages in such fashion as to express the required sequence, and recognizing and accepting the attendant dangers of misinterpretation, the doctrine would run as follows:
Market prices fix the incomes of the separate groups. . . . The fixing of market values results in the terms of group distribution.
. . . Market values determine what each subgroup gets as such, and so control the distribution at its earlier stages. . . . The income of each subgroup in the series, then, depends directly on prices.
"Natural" or "normal" prices . . . are the values . . . to which, in the long run, market values tend to conform.
A certain force that operates within the sphere of group dis-
tribution establishes the natural order, so that market values tend to conform. A deeper force, and one that also acts in distribution, controls normal prices. Market prices are the cause of group distribution. . . . The adjustment of natural or normal
*Clark, op. cit., pp. 14-16.*
DISTRIBUTION BY VALUE PRODUCTIVITY 455
prices is a part of the distributive process. The movements that make prices "natural" are, in fact, efforts on the part of different men to get their natural shares of income.
Being again interpreted, this appears to say that normal value controls market value: that market value controls group and subgroup distribution; and that the distribution inside the group and subgroup is determined by normal value. As, then, the ultimate term in the causal sequence is this infra-subgroup distribution process; everything else is derivative herefrom.
If, however, there is any seeming of circuitry here, it is due rather to the sentence-juggling device than to the intrinsic thought; the difficulty is not that the explanation offered is inadequate but that in last analysis no explanation is offered; and perhaps none should, at this stage of the argument, be required.
Notice, at any rate, that it is said not only that the actual market values at any time is determined by the normal value, but only that they may remain true or be con- trolled by normal value; it may remain true, then—and it really must be true—that no matter what the trend of things or the necessary long-time adjustments of them may be, the market values will remain true at any time according to the situation, the forces, and the adjustments of that particular time. To subscribe to some long-time control over the short-time market values in accordance with this principle; the short-time value of any particular sort of commodity remains, for its time, the determinant of the group and subgroup distribution of that time. For purposes, then, of any short-time analysis, it would be necessary that something be offered as the way of explanation of these short-time market values.
But to the logical validity of Clark's position nothing need have been added in this regard; for his problem has been essentially changed by the problem of distribu- tion; all, then, that is necessary is that he explain static value. The position on this point is as follows: Static values, as regards group and subgroup distribution, are the result of the interplay between investment worked out in the infra-group distribution process.
But it nevertheless appears to be true that the magni- tude of the group distributions out of which, as a static problem, the specific wage and interest incomes are to be
456
VALUE AND DISTRIBUTION
apportioned, is controlled by static market values. The author's problem stands, then, if circuitry is to be avoided, as an alternative problem- either (1) to explain the static value of commodities in terms of distributive shares, or (2) to explain the distributive shares independently of the static values.
But it is forthwith to be added that something purporting to be an explanation of the distributive shares in this tertiary group, subgroup distribution is offered, which, in terms at least, makes no appeal to the support or aid of market values.
The influence that brings production to this natural state is the effort of men and capitalists to seize any special gain that may be offered to them, by moving to any group in which the price of the product is high. This is clearly an operation in group distribution. Thus an influence that originates in distribution brings about a state of affairs that is contrary to the law of normal (p. 18). Prices are at their natural level when labor and capital in one industry produce as much and get as much as they do in any other. Normal prices mean equalized wages and equalized interest. If the producer has more than he needs, he can sell his surplus. No laborer and no capitalist could acquire an enlarged producing power by leaving the industry that creates one of these commodities, andbetaking himself to one that creates another, the price of each of the commodities being equalized. This is a case of free choice . . . . . . a state of production; that ultimate influence that controls it is an action of the forces of distribution. . . . Market value falls within the science of distribution. On the surface, it is current market prices; but on closer examination it is a comparison among different groups or specific industries. These prices, however, are transient, and they fluctuate about certain more permanent standards. The tendency of group distribution to become normal, that is, to tend toward equality of wages and interest in different industries, draws prices toward the normal standard. . . . What then, is left to be treated under the title, exchange? Only the actual passage of goods from hand to hand.
That multitude determines value and of distribution may be of the general sort indicated is beyond question; all the facts must contain within themselves all the explanations, causally speaking, though perhaps not descriptively or interpretatively speaking. But, for this causal explanation to stand as complete, more must be offered than the mere disposition of men to seek the direction of
DISTRIBUTION BY VALUE PRODUCTIVITY 457
maximum pleasure or of minimum pain, or even the direc-
tion of maximum advantage or of minimum sacrifice; ac-
count must also be rendered of the needs and desires of men
for consumption goods, on the one side,—on the other side,
of human productive capacities and differentiations, and of
the obligations which they impose upon each other—compen-
sations, franchises, good-will, trade secrets, and purchasing
power—in all its differentiations and adaptations and
specializations and distributions.
But, however much equilibrium value or for any
temporary equilibrium value, an explanation of the sort
offered is the kind of explanation for which we are in
quest, and if it is of a character to suffice for the problem
of distribution, it will not suffice for that of value. It remains
the basis for the explanation of costs and market values,
and thus to establish distribution as fundamental to value,
why pursue this subject further or any subject farther?
Why need we go beyond the level at which we have the prob-
lem or the distribution problem? All the causes are surely
adequate to explain all the results. The difficulty, however,
is that on this level of explanation both value and distribu-
tive shares are left undisturbed and unexplained in some
order, and neither law—or can be shown to be the inter-
mediate cause of the other. And regarded from this point
of view, the explanation of distributive shares, whether
looked at as a whole or broken down into sub-groups or
a group nor a subgroup investigation; the individual choice
of personal activity and of investment application is a
choice wide as the whole field of capacity and opportunity.
But it is not so long ago in time being, as true
that the ultimate adjustment, when reached, is a dis-
tribution presenting equalized wages and equalized interest
(property rentals?), as the result of the free play of choice
between competing opportunities, and not at all stopping to wonder why these opportunities should become equal.
It is still to be objected that these competing opportunities are themselves mainly the expression of an existing value situation; each individual's choice is made within this situation as directed by it. That it is that entire situation of needs, capacities and posses-
sions, adjustments, and distributions, may be taken as fundamentally directive and causal, but not so the indi-
vidual choices made within the conditioning and directing
A page from a book with text about distribution by value productivity.
458
VALUE AND DISTRIBUTION
situation, nor, for that matter, all the choices, excepting with the tacit assumption of all the situation and in connection with it. The individual choice, or even the aggregate of choices, is not a cause but a result of the entire individual activity infinitesimal reacting as cause upon the entire situation.
And if any attempt is made, in the line of explanation, to go further than an appeal to this huge and vague situation aggregate, and if definite and detailed explanation for anything be offered in terms of intermediate causes and sequences, there will forthwith come the abandonment of any further appeal to the market value of corporate shares, group or other, or of distributive shares as the causes of value; the investigation will be compelled to refer itself to the social organization of production and methods and adjustments of a competitive society under entrepreneur organization and direction. Any investigation that approaches distribution, leaving value aside or treating it as a problem to be solved by some causal and taken-for-granted appeal to distributive shares as the basis of entrepreneur costs, is for practical purposes no solution at all.
But another solution may possibly be open, for whatever it is worth, a solution so readily deducible from the organic concept of society that it may well have seemed to dispense with any special labor of exposition:
Consider the case of a man like Crusoe who is over against Crusoe as a producer pain-burdened in his processes of production, and setting him upon some assumed island of definite possibilities and limitations, we have seen it to be possible for him to make himself happy within the isolated economy. If, then, society may be taken as having social needs and desires, and thereby a derivative social marginal utility and a social subjective worth and a social subjective value equation. In this sense, however, there are social pains and abstinences and sacrifices adequate to function as cost terms in the value equation, no further assumptions appear to be called for; a complete account has been provided of market value; for, fortunately, the environment does not also have to be obtained by this process of assumption.
And thus, with the market values all satisfactorily
DISTRIBUTION BY VALUE PRODUCTIVITY 459
explained, nothing remains to be done but to deduce the distributive shares.
Precisely so—but how deduce these? What problem of distribution is open? Distribution among whom? And under what conditions? The problem of value production, by assumption, bore the pains thereof. When, then, is this, or who are these now coming to demand that payment be had by some test of distinguishable, separable, and assignable content? The answer is that the whole of value fixation is a social fact—costs, pains, sacrifices, pleasures, marginal utilities, and effective utilities, all social facts attaching to the appetitive or emotional psychology of the social organism. There are three generic shares that are unlike in kind,¹ and that
the entire study of distribution is . . . a study of specific pro-
duction; is an analysis of the wealth-producing operation, and a tracing back to each of the three agencies that together bring wealth into existence, one part which is inseparably contributed to the joint result. To each agency corresponds a corresponding share, and each a corresponding reward—such is the natural law of distribution.*
That by Clark land and all other instrumental goods are reduced to one homogeneous fund of value units will at this point require neither illustration nor proof. Labor is similarly funded into a volume of homogeneous labor units. And, either as included within this labor fund or as basis of a distinct and separate fund, entrepreneur activity also must seemingly be subjected to the funding process and be reduced to a homogeneity of value units. Only so, in fact, is it possible to establish the central thesis of Clark’s argument—that the productive contribution consists in the remuneration of each and every unit in the fund to express, in precise equivalence, the productive contribution of that unit:
There is before us the picture of social labor co-operating with social capital. Both are governed by the law of diminishing returns, and their earnings are fixed by the productivity of their final units.
¹Clark, op. cit., pp. 3. &
460
VALUE AND DISTRIBUTION
Wages conform to the final unit of social labor and interest to the product of the final increment of social capital.
But "how may we measure labor, capital, and their products" so as to make certain whether this proportionality of remunerations to funded units everywhere holds? "We need, evidently, a universally applicable measure."
Proportionally, then, all units of capital are measured in terms of money; but it is necessary to know exactly what the money ultimately represents." This is ultimate underlying verity--this definitive fundamental fact in value--an objective fact which must be known before we can become homogeneous? "If it means either of these two things, it is still necessary to find some way in which to express a measurement of labor or of sacrifice. And to Clark "it is clear that the only way in which the quantity of such connections as these, be the basis of the measurement of the capital," for this would be circular in reasoning: "If we say that whatever produces a unit of consumers' wealth is productive, we must also say that whatever produces that, at any one time, all units of capital are equally productive." So it is only provisionally that the doses of capital are measured in terms of money; we must, it is said, go deeper than this.
A universal rule for measuring economic value is necessary, if the law of final productivity is to have scientific exactness. The entire study of wealth is, indeed, meaningless, unless there be a unit for measuring it. Ratios of exchange alone afford no answer to the economist's chief problem. For his problem is that com- munity consists in heterogeneous things. If they are ever added together, it must be because there is some one element found in all of them and this element is always the same. There is one element that has been added to all the diverse things that appear in inventory of social wealth. In every commodity there is a power of a certain kind which can be measured. . . . There resides in each of them a power which can be measured by a common denominator. Effective utility is the name by which this potency of goods will here be designated. . . . Effective utility is the basis of values.*
Effective utility is, then, the common denominator into which all values are to be resolved and by virtue of which
*Clark, op. cit., p. 373.
*Thid., p. 374.
*Thid., pp. 375-376.
DISTRIBUTION BY VALUE PRODUCTIVITY 461
each and all come to be rational, intelligible, and compa-
rable. And this is in line with the doctrine, already noted,
that "the principle of final utility by which values are fixed,
is universal in its scope."13
"Final utility universally gauges values."11
That is to say, effective utility and final utility are essentially the same. True, "amounts of wealth are usually stated in money;" but the coins are not really the measure; they merely express power over the things that afford service: "They will buy goods or set men to working.
There resides in each one of them a certain amount of influence, which is the only way to control effective utility." Effective utility is the name by which this potency of goods will here be designated.11
But admittedly some things have utility not because they give us pleasure but because they shield us from pain; and Clark insists that this pain-avoiding quality is the char-
acteristic and essential and ultimate fact in all effective utility; the marginal-utility analysis does not go deeply enough, or, rather, it does not bring out quite adequately the important aspect, the sacrifice aspect, of things of service:
Give to a man a barrel of flour and you make him by so much better off . . . If you had not given him the flour, he would have got it by some sacrifice; and what you have done is, in effect, to save him from sacrifice. This effect measures the value of the flour. Take away a barrel of flour from him who has, cognate the real detriment that he suffers, and you measure the effective utility in another way."
Is this subjective worth? It looks like it; but if it is assumed that we can measure subjective worth by something other than submitted to directly or, perhaps, shifted to something else, the doctrine diverges from the Austrian analysis, and—still more—diverges from the truth. And if it be admit-
ted that the loss has not, of necessity, to be made good, or that, if made good, it may be at the cost of some substituted
\begin{align*}
&^{13}\textit{Ibid., p. 47.} \\
&^{11}\textit{Ibid., p. 376.} \\
&^{10}\textit{Ibid., p. 377.} \\
&^{10}\textit{Ibid., p. 142.} \\
&^{10}\textit{Ibid., p. 378.}
\end{align*}
462
VALUE AND DISTRIBUTION
service, we have nothing but subjective worth, the cost aspect of utility, "the importance attached to a good as the indispensable condition, etc." a purely personal category and an unrelated feeling magnitude. Clark continues:
He must [7] have food, and will get it by sacrifice of some kind. He may choose to sacrifice his own life, or his children's life, or he may live on maize, and in that case the utility of the barrel of flour is gauged by the cost of the maize and the unsatisfied want of a better quality [or quantity?] of food.
But later upon the same page the labor-necessity view is fully adopted:
The bare measure in the case is one of pain; for the ultimate injury that is done to a man by depriving him of any one means of pleasure, resolves itself into putting him under the necessity of enduring a certain amount of personal sacrifice in the effort to secure some other means of pleasure.
But in any case, we have arrived at nothing more or other than the old difficulty with regard to either marginal utility or subjective worth; as a purely personal experience, unreliable to the experience of any other person, and, as absurdly few persons can be found who do not consider the marginal item of a series—related to any other experience of the same individual, it perhaps does not greatly matter whether it is considered as a mere means to an end of utility merely or to the labor pain of replacement. If mar- ket value is to be resolved into a homogeneous fund of utility units, there are some further steps to be taken, and these are of a very difficult sort. At these steps individual interest, ample of its nature, is discussed in earlier chapters have sufficiently emphasized. It remains then, to pre- sent, with the minimum of comment, Clark's proposed method:
It is this process . . . , of determining how important it is to have one thing by ascertaining how much it will cost to get very different thing, that reveals one special significance of a form of effective utility. Men pursue happiness in the generic, and the form in which they pursue it is that which gives them most general well-being, thus regarded in the abstract, is an occult but dominant fact in exchanges. A man may have a monopoly of one means of pro- moting happiness, yet he cannot set his own price for his wares. That is fixed by the cost entailed on the community by the effort
*Clark*, op. cit., p. 371.
DISTRIBUTION BY VALUE PRODUCTIVITY 463
to secure by any means whatever, an equal quantity of happiness.
Effective utility . . . . is measured by society as a whole; and in this lies the significance of the phrase, "measure of effective social utility" . . . . The price of a thing depends its importance, not to one man, but to all men, as organically related to each other. The efficient use of a thing is not to be measured by its value to different individual users, but to society as a whole it is constant. . . . Into the mysteries of distinctly social psychology, therefore, the measuring process is given place. It is not the nature of things that is the measure in this operation, simpler even than the act of the man who decides how important a horse is to himself by seeing how long he must work to get a boat and a tennis outfit.
It is also true that there is no word corresponding to the word social. There is such a thing as a unit of social improvement or detriment. It happens, however, that the detriment is more avail- able for measuring purposes than is the improvement; and so the final unit of measurement will be the negative. This is what we de- finitely social labor. Society, in short, sets value upon a thing by ascertaining how much work is necessary to replace it or to get an equivalent for it.
It would seem, then, that in order to measure a unit of social utility, we must first make precise the notion of a unit of "distinctly social labor." what, then, is this?
Doubletless utility exists for each man and sacrifice exists for each man; but because each man can make these estimates for himself, does it safely follow that society can do the same for itself? And if we know how the individual man does it, shall we also know how society does it? So it seems:
It may be assumed that while articles are made by individual workers. As such goods leave the makers' hands day after day, in a continuous supply, they seek purchasers. No one man will take many, but every man will take some. The fact that articles are made in great numbers by one man and consumed simply by many men, is the essential thing to be noted.
It is the users of an article that can best gauge the well-being that it gives them. They can estimate continually. Shall I buy this article? . . . Is this article or some other of equal cost the more desirable? . . . If each man could measure the useful- ness of an article by the effort that it costs him to get it, and if he could attain a fixed unit of effort, he could state the utility of a
* Ibid., p. 378*
464
VALUE AND DISTRIBUTION
number of articles in a sum total. Similarly, if all society acts in reality as one man, it makes such measurements of all commodities, and the trouble arising from the fact that there are many measures disappears. A market secures this result, for society acts as an individual unit—like an individual buyer.
And similarly for the sacrifice,—which we recall, has thus far been presented as a better measure of the effective utility than the mere market itself, although this is not to be taken as in any sense an abandonment of the notion that all value resolves into and is nothing but effective utility; sacrifice is somehow a better measure of utility than is utility itself; indeed, how can utility serve as its own measure? And it needs measure, while, it seems, sacrifice does not, or perhaps is more easily measured:
Work . . . consists of concrete acts of men; and these are as unlike in themselves as they can possibly be, so as to be measured by them. Can we make one sum of the labor involved in cutting wood, in playing violins, in setting type, etc.? Adding the unlike acts that constitute social labor is, it appears, as difficult as adding the like acts that constitute physical labor. There is a pervasive element in the actions, and one that can be measured. Such an element can be found; for, as utility is common to all com-
modities, so also is sacrifice common to all kinds of labor. There is service rendered to man, on the one hand, and there is bur-
den imposed upon him, on the other. Social self-service—the act of mankind ministering to its own needs—constitutes the whole eco-
nomic gain. The same thing may be found at which social costs of production offset and measure social gains. We can . . .
estimate pleasure in terms of pain.
An isolated worker is the user of his own products, and he naturally seeks to increase his output by increasing his effort. Additional product might be gained by prolonging the toil, but the advantage of having it could not compensate for the sacrifices of making it . . . "The man that we are studying is a society by himself," i.e., making things for himself and using them as a society regarded as a unit the same is true. It produces for itself, and the burden of its final labor measures the utility of its final products, which is the same as the effective utility of any of its products created by its own labour. Take away the articles that the society gains by the labor of a morning hour, the
\footnote{Clark, op. cit., p. 380.} \footnote{Ibid., p. 381.}
DISTRIBUTION BY VALUE PRODUCTIVITY 465
necessary food, clothing and shelter that it absolutely must have, and to make good the loss it will divert the work performed at the approach of evening, which would otherwise have produced the final luxuries. The laborer who has been employed by the employer, and by that final labor will be the ones really lost, and their utility is meas-
ured by the burden entailed in the creating of them. . . . Every-
thing that is produced is a social labor, whether that labor be performed early in the day or later, because an effective social utility that equals the absolute utility of the final complement of goods consumed; and this, again, is counterpoised and measured by the sacrifices made by those who perform that labor as its
final hour. Periods of labor are equal in effective disutility,
and this makes it possible to use the labor of any period of a given length as a unit for measuring values. . . . In the subjective valuation of commodities, one man's labor is always worth just as much as is the product of one.
Mere labor time is an accurate gauge of the values of different complements of goods."
But a group of goods to serve as a social unit of con-
sumption is one thing, and the separate items made by
different individual men may be another; i. e., mere labor
time does not measure equally the relative values of the
different articles that enter into the complement." Here
the answer appears to be that the utility of what I sell is
measured by the pain of society in producing what I get
in exchange for what I sell: "The pain that all other men
suffer in making products for him represents the cost to
them of what they get from him. . . . Price is then,
an indication of the social cost of acquisition of different
commodities." 18
Recalling now that each unit of labor commands under
normal conditions precisely the same remuneration as every
other unit, we become interested to know how much labor
of any given individual constitutes a labor unit. Evidently,
laborers are not paid the same per-diem wage, but only the
same wage per labor unit:
\begin{align*}
& \text{"Ibid., pp. 283-89, passim.} \\
& \text{"Ibid., p. 291.}
\end{align*}
466
VALUE AND DISTRIBUTION
A laborer of high grade embodies in himself more units of labor than does an inferior one.®
The fixed amount of labor is the sacrifice entailed by a quantity of distinctly social labor.™
A social unit of labor . . . is a composite unit consisting of some labor from every industrial group that the community consumes.
A minute would be a larger fraction of one man's day than of another's. It is accurate enough for our purpose, however, to say that the social labor is made up of a fixed fraction of a day's labor of every individual.
Evidently it does not matter what precise quantity of this social labor is taken as the unit of measurement; but take now the case of a man who makes and upon the market sells a commodity:
A commodity is actually measured for value on the basis of the social service that it renders. All society, in the end, incurs a marginal sacrifice that measures its value. The individual laborer measures his own value by the service rendered by the social labor that is induced by it and that measures its value.™
But how does a man actually go to work to exercise this inducing power over the social labor? Evidently he sells his product upon the market:
If money is used in the transactions, and if the price of $W$ and that of $X$ are equal, then $W$ is a commodity, and so is $X$, because they are both sold at the same price. But if we draw out equal quantities of social labor, it is made over to the miniature society for consumption, imparts to society as a whole a uniform addition to its enjoyment.™ Price is, then, an indication of the social cost of acquisition or division.
Thus are we able to know how much of each man's labor is necessary to constitute a unit of labor; and then we know that these amounts of labor will be equally paid—these funded equal units—and how do we know this? By the fact that commodities sell for the same money price. We draw out equal quantities of social labor. But this means that the products sell for the same money price, and that the two quantities of labor are equal quantities by the mere fact of their being valued at equal prices. And yet Clark has only a few pages back insisted that it was not so,
*Clark, op. cit., p. 63.*
*Ibid., p. 170.*
*Ibid., p. 297.*
*Ibid., p. 396.*
*Ibid., p. 391.*
• Clark, op. cit., p. 63. |
• Ibd., p. 170. |
• Ibd., p. 297. |
• Ibd., p. 396. |
• Ibd., p. 391. |
DISTRIBUTION BY VALUE PRODUCTIVITY 467
in good logic, to make "the product . . . the basis of measurement. If we say that whatever produces a unit of consumers' wealth is a unit of capital, we assert nothing by adding, at any one time, all units of capital are equally paid."
We have now to recall that in Clark's view this resolution of market value into some sort of underlying and controlling and determining homogeneity is absolutely essential to the funding of capital goods and of labor into value units, and likewise essential to the proof that in each fund the compensations tend to be—and normally are—the same for all of the units. That is to say, Clark rests the specific-productivity theory of distribution upon two bases, (1) that market values can be reduced to an ultimate homogeneity in terms of effective-utility units, it being for this purpose that the social-organism concept is invoked for services; (2) that all productive instruments, on the one hand, all productive instruments, on the other hand, are likewise subjected to the funding process, in such fashion that equal value productivity may be ascribed to the individual units of each fund, and this without appeal to the quantum of value productivity as determining or defining the unit.
It is, indeed, possible that Clark has admitted here more than he has actually argued; it is conceivable that neither of these interpretations is essential to the conclusion; it may be that the specific productivity of the productive factors may be worked out, and distributed among them according to their relatives and equivalents of specific productivity, and all this without appeal to any sort of homogeneity underlying and determining market values, and without any manner of recourse to the funding devices proposed. How, indeed, disprove it?
And it is also forthwith to be admitted that no proof has yet been offered—or can later be adulated—that society is an organism; but it is certain that no proof has yet been anywhere advanced that it is an organism. And it is further true that the sociologists themselves have long since mostly
*1944*, p. 374
468
VALUE AND DISTRIBUTION
abandoned the doctrine. But none the less may the doc-
trine be true, and fertile of more truth; this, however, is
sheer matter of faith; and, as all propositions purely of
faith are necessarily false, they cannot be made into facts
provable. And so likewise it may be true that the social
organism has methods of funding labor and of funding
capital that we wet not of, and—possibly enough—these
capital funds are not necessary to the production of goods not
derivative from the valuations organically placed upon the
products. We can deny no part of this in any sense of
thinking ourselves able to prove its falsity; to prove that
water babies do not exist would be to prove that water does
more than not see any water babies existing; we must see
a water baby not existing. We can, then, do no more than
to put the issue at stake in the evidence of our argument,
if any have been offered in support of the proposition.
Any criticism made or yet to be made must surely confine
itself to questioning the logical legitimacy of this estab-
lishing of fundamental propositions purely by faith, and to
a scrutiny of the logical tenability of the relations asserted
or assumed between such distinctively derivative proposi-
tions.
But if even faith-wise, these factor funds can be estab-
lished upon grounds other than the meditated basis that what-
ever produces a unit of consumers' wealth is a unit of capital," the explanation—if explanation it is—is most appar-
ently run substantially as follows:
"Use as 'effective utility,' i.e., measured by society
as a whole, as a measure of well-being, thus
regarded in the abstract, is an occult but dominant fact in
exchanges," and as "in this lies the significance of the
phrase 'measure of effective utility,' " it was upon
the working out of which it was said "must depend, by
some occult social-organism process, do we get an appraisal
and a funding of capital goods and of labor,—a funding
not dependent on the value of its product." The value pro-
duced and imputed to these productive factors is all this, also,
in such fashion that both a mystical, occult, and marvelous
parallelism and a precise proportionality are discoverable
between the value of the funded unit and the value of its
product.
But even if this social-organism method of getting
products and productive factors into abstract funds be
DISTRIBUTION BY VALUE PRODUCTIVITY 469
abandoned as, on the one hand, unproved, and as, on the other hand, unserviceable for its purpose unless logically abused, the productivity theory of distribution need not be, it repeats itself, rejected. It may be said that Clark has overestimated or wrongly located the difficulties of his problem; and it must be remembered that it is primarily this problem, and not Clark's solution of the problem, which is of interest to us in investigating how to be established. We may say that market values will not reduce to pain jelly or to utility jelly, whether by the social-organism concept or by any other method; the case of the productive expenditure being spent on some service cannot be declared hopeless, if only it be true either, (1) that the funding expedient is not essential, or (2) that any without appeal to the hypothetical psychology of the social organ-ism, the market values of products are actually homogeneous, or made homogeneous on some sort of workable basis; and it need not matter whether this other basis be something distinct from past cost in the getting or from well-being significance than the first.
It is, in any event, worthy of remark that the market values of products and the market values of factors are actually and patently homogeneous under the simple, every-day, and practical conditions of our economic life; homogeneity, that is to say, in terms of the money fact a conven-tional standard admirably adapted, as expressive of homogeneous, undifferentiated purchasing power, for meeting all requirements. The same thing holds true of homo-ogeneity. What, in fact, can capital as an abstract fund possibly be, if it be anything other than a market-value fund?
But, even so, we are immediately driven back upon the problem of how to make use of this homogeneity in such wise as to help forward the problem in hand, which is, in part, precisely this of how to explain these very market values of products and factors. This explanation must still remain true, as Clark has so well pointed out, that neither capital units nor labor units, no matter in what manner or how well funded, can logically be ascertained to possess equal value productivity and so to be equally rewarded in value compensation. And yet it is clear that the units are explained and established as such, only by the fact that they produce equal values or get equal compensations.
470
VALUE AND DISTRIBUTION
We seem compelled, then, to adventure the problem under its other statement, how to explain the remunera-
tions attributed by the market to the different, specific, con-
crete forms of labor, which are employed in different parts of
the productive process? Is it possible to regard the
remuneration as either normally or actually the precise
equivalent of the productive contribution? How deter-
mine the value-productivity relationship with the
remuneration? Is any method of isolation of factors
possible? Or is all that can be said merely that the
remuneration is the market value of the value-productivity contribution? And what is this value-productivity?
The value contribution? To this aspect of the problem, which,
after all, seems to be treated by Clark, we must now direct attention.**
But precisely what is our problem? It is clear that we are not now interested in any issue as to whether these
values are the cause or cost, or as to whether distributive shares are better regarded as primary or secondary;
no denial is suggested that all productive instruments and agents are hired in view of the value products to be derived from them; nor is it, at least, the value-
productivity of shares of distribution which is generally accepted; nor is there no question that the amount of value in the joint product of the factors is the equivalent, the
**The specific productivity of labor does wages—wages—that is thesis to be maintained—is not a matter of fact.**
"We have said that the specific productivity of labor does wages; and this means that pay conforms to the amount of product that is specified by the quantity of labor used in producing it. This implies that the products of the different units are equal. In this manner, we may say that each unit produces one. The earnings of a dollar are what the dollar creates; and this implies that in any one fund of capital, as it is described in terms of money, the productivity of labor is constant.
"A laborer's income may seem to come to him from another man; but in reality he receives his own nature makes to his own labor—it is his own virtual product" (p. 53).
"The income of each subgroup is not the value, not of a completed article, but only what it would cost if it implied a "sale" (p. 54).
"Paying interest is (somewhat) analogous to the process of capital as paying wages is buying the product of labor. The power of capital to create product is, then, the basis of interest" (p. 123).
"As value depends on final utility, so shares in distribution depend on final productivity" (p. sub.).
DISTRIBUTION BY VALUE PRODUCTIVITY 471
source, and the intermediate determinant of the aggregate compensations distributed to these factors, it mattering not at all for present purposes how this total value to be distributed may be determined. It would be easy to deny that each of these distributed compensations falls out in some general and approximate proportionality with the value increase contributed; but rather we are concerned to determine whether the general and specific productivity of the different co-operating productive factors can be so isolated and distinguished that, actually or logically, the compensation may be declared to be the precise equivalent of the product which they have produced. It is well to say, what is the meaning and what the warrant of the assertion that out of the distribution of a joint product the different productive agents receive the precise equivalent of the value which they have produced? Is it true that this precise productivity can be arrived at other than by sheer reference and appeal to what they get?
This is, in truth, our old problem of imputation, the problem which we shall discuss in another chapter, how to distribute between the different co-operating factors not only all the value product which, if not so co-operating, they might have produced, but, together with this, that increment of value which they have received as a very result, as it was the very purpose, of their co-operation.
Assume that in isolated uses, or in other combinations, each of four productive facts, e.g., land, machines, wage-earners, and entrepreneur, could command 3 of recompense, and that when the four are brought together, their aggregate product is not 12 but 13; what distributive principle should imply that each receives 3? If the first of Clark's principles of imputation be accepted, namely, that "the amount that is taken from the force when measures the effective power of that force," then this subtraction method will lead us to impute to each co-operating factor a specific productivity of $a$,-and this despite the obvious fact that the joint product is not 12 but 13.
Clark's second principle of imputation he does not himself recognize to be a second, but only the repetition of the
*Clark, op. cit., p. 161.*
**VALUE AND DISTRIBUTION**
first: "A similar test might have been made merely by adding a unit of labor, instead of taking one away." But if this principle be accepted we shall come into precisely the reverse difficulty. For if the proportionate increase in the original product is not in the same proportions—if, for the entrepreneur, the combination was the best combina- tion—no increase in any one factor alone could take place without its per-item productivity suffering. It follows that the market bid for a unit of the item, "taking one away" would distribute a product somewhat smaller than the actual product.
But when it is known what any particular entrepreneur will ask for his share and for the item in question to be used as part of his production complex, under his own direction, is it safe to attribute all of the increase in product to the new item? Is not the increase due rather to the mere "worn outness" of the old tools? And is it not certain that other entrepreneurs in different cir- cumstances and of different abilities must have different maximum bids? And which one out of all these different producers has the most productive per-item productivity of the item in question? And does the market hire neces- sarily or probably exhaust all the value significance of the item to the successful bidder? If so, then where were this items' proportional value confined to those produc- tive factors which are present in stocks?
Substantially the same difficulties exist for all attempts at the isolation of productive factors zones. In indiffer- ence, they are found. In fact, no such zones are to be found. If labor could somehow be found supplied with no tools, or with valueless tools, and working upon valueless land, this would be a perfectly possible case. The only reason of a limitation of product due to inadequacy of equipment. Where "the worn tool, the rickety engine, the unaworthy ship . . ." is at the point of abandonment, the labor that uses them is paid full wages. It is therefore not true that "the amount of this product corresponds with and expresses the rate of general wages," and would not be true even if all laborers were alike in their relations either to equipment or to entrepreneurship.
Nor—adapted to Clark's proposition—is there any zone
*Clark, op. cit., p. 97.*
DISTRIBUTION BY VALUE PRODUCTIVITY 473
of indifference, or place of isolation, for the last or mar-
ginal worker employed by any entrepreneur:
It is this most sterile of the fields, openings, or opportunities for labor that we describe graphically as an outermost zone within which men produce only their wages. This is the zone of indiffer-
ence from an employer's point of view, because, if he sets men working within this area, he must give them all that they produce
as wages.*
But, at the most, he needs do this only upon the assumption
that he is the marginal employer, and upon the added
assumption also, that the employer has no power to say that
there is nothing peculiar to them or him or in his land or other capital so that another employer may not push him to his highest possible bid. And even upon these assump-
tions, there is still a possibility that the marginal laborer is hired at precisely no gain; the employer may well have some uneXhauXed powers of supervision still left in him. And so, "if one employer offers to them less than by their own judgment they would offer [him] they are worth [to him]" there is no sufficient war-
rant for the belief that "another will offer more, pro-
vided competition is perfectly free and efficient."
It is true that the marginal laborer hired by the marginal employer is hired at no gain. Is there any good reason for supposing the extended super-
vision of the employer to be non-productive? Why, then,
does he not employ more laborers? The reason is that with
limited command of capital some of the supervising abili-
ties of the employer must, in any case, run to waste, it is
to be replied that this argument recurs to an assumed peculiarity of human nature. It does not follow from these
reasons the attempt to explain the wages as reflecting the isolated, objective, value-producing power of the laborer.
For note that the isolated and objective productivity in question is not necessarily identical with the subjective compre-
hension that the production takes place upon marginal land only, or at the marginal powers of land, and with the co-
operation of only marginal capital or the marginal powers of capital. It is not necessary to suppose that the production is
marginal in its relation to that marginal entrepreneur, and
to the supervisory productive powers of that entrepreneur,
*Clark, op. cit., p. 110.
474
VALUE AND DISTRIBUTION
who can pay least for the labor and who, hiring it, pays at the same time all that he can.
And now, assuming even this case of utmost payment—
a payment leaving no least residue of rent or interest or personal compensation for the comfort of the employer—where capital goods are measured by their laborers and goes as their distinct and unquestioned value productivity, it becomes so much the clearer that the wage outlays of the marginal employer will be less than they would have been had his, he has share in the results as fixed, except to the extent that as one item in the demand schedule, his demand has prevented that still lower wage which would have attended the threefold increase of his demand. The employment of some still less efficient and still weaker entrepreneurs. But as in such case the wages must admittedly have been lower under this otherwise excluded employer, it follows that to our marginal employer, or to the entire productive complex as a unit, there is no per se greater part gain than pro-
ductivity which Clark's analysis imparts to the laborer alone.*
*Clark himself recognizes the importance of the productive com-
plex—recognizes, that is—as perhaps in view of his abstract-capital con-
cept he may have been led to overlook. For example, a locomotive is not of a quantum of machinery, or of land, or of cars, or of rails, etc., but of a combination of these things. It is not a unit of capital expense,
the guise of an item or dose of productive outlay, a unit sum of capital expense; and this is admirable, only it is assumed that in concrete manufacture and distribution it is possible to measure out units of material capital goods, rather than be applied to to use or varying fractions thereof. In other words, we cannot measure out architecture, inclusive of labor, insur-
ance, advertising, taxes, and what not—
"For here, instance is a new locomotive. It has not been secured by any means other than by the application of a mechanical engine, made necessary by an enlarged traffic. Is it a final increment of capital? No; it is a combination of capital goods. A motor car or motor engine with an equipment of good cars, good rails, etc. This comple-
mentarity of producers' goods must always be considered. The same applies to railroads. They are not merely rails on the rails,
the rails, the bridges, the cars, etc., with which it is used" (p. 248).
But Clark does not recognize that each item has its specific productivity in the separate items of capital goods; and note also that this same argument applies in principle to all combinations of capital with labor—capital goods with other capital goods; or of labor with other labor.
This distinction between capital goods and labor is strange to get concrete things that are about to be. The capital of society has no existence till it is in the shape in which entrepreneurs use it. Till it is raw materials and tools for the manufacturer, merchandise for the
DISTRIBUTION BY VALUE PRODUCTIVITY 475
But we have now to notice a still more important and still more disastrous error in this method of analysis:
The product that can be attributed to this second increment of labor is, of course, not all that it creates by the aid of the capital that the earlier division of workers has surrendered to it; it is only what its presence adds to the product previously created.
This is crucial; in point of fact, the producer does not get all his increase; on the contrary, distribution is disturbed to the advantage of the capital; wages are suffering not merely from the fact of a diminishing production increment, but also from the fact that in every situation every unit of capital is deriving an increased increment which is to say, wages are suffering not only on the production side but on the distribution side.
The fundamental error in all analysis of this sort—and the later economics is full of it—is traceable to the assumption that the marginal-utility analysis for the individual man can be applied to society as a whole; and also that the method of computation supposedly valid for consumption goods can safely be applied to production goods disposed of under entrepreneur bidding:
There is a commercial principle which causes the final or marginal part of any commodity to be valued on the value of the whole group. The value of the whole crop . . . conforms to that of the marginal bushel. If there are marginal laborers, in the sense in which there are marginal quantities of wheat, cotton, iron, etc., then their wages will conform to their marginal position; for their products set the standard of everyone's wages.
. . . The last increment in the supply of any commodity fixes the general price of it.
[retailer, vehicles for the carrier, etc., capital has no existence at all. . . . Bidding for capital, then, is bidding for something which is not money—namely, a change of quality of working implements] (p. 29).
The investment of the capital of this railroad corporation is, in reality, a difference between two kinds of plants. One of these is the railroad as it stands. The other is the road built and equipped since 1860. It would seem if this process had had one degree less" (p. 25). All this presents the social concept of capital rather than the competitive.
[Clark, op. cit., p. 176.]
[Noble, p. go.]
476
VALUE AND DISTRIBUTION
The difficulties are thus several in the way of this method of solution of the distributive-imputation problem:
1. With complementary production goods, no separate and specific significance, like that attached to consumption goods, can be ascribed to any one item.
2. Changes in the relative supplies of co-operating goods work changes in the relative significance of all the different classes of the co-operating goods.
(Propositions (1) and (2) would hold either in a Cru-
soe or a Smithian economy.)
3. In the competitive economy all possibility of ascription of a single productive significance to any productive item disappears; the entrepreneur bidding different in equip-
ment and skill for the same skill will pay as much as many different productive potentialities as there are differ-
ent productive complexes. There can, therefore, be no one degree of productivity assigned to any particular item, and there is no warrant for supposing that the hire paid by the successfully bidding entrepreneur coincides with even his own appraisal of the prospective value of the productive complex.
The successful bidder pays at the minimum what someone else will pay; at the maximum, all that he can afford to pay; but that he is the successful bidder does not imply that he actually produces more than anyone else.
4. Even were productive factors always present in stocks rather than—as generally—in concretely differing items, an entire competitive society would have no way of adding up the total value of all its stocks to a fixed and stable volume of complementary stocks; even the individual does not commonly proceed in this way, but rather, as he enlarges his business unit, enlarges it in several directions simultaneously. The various kinds of stocks and classes of production goods do not take one after another each its separate turn in the process of increase.
*The precise relation of the marginal analysis to the specific value-
productivity issue will be brought out in clearer definition by an account of a recent paper by Professor J. R. Hicks, "Marginal Utility," J. Hobson: Hobson, "Marginal Units in the Theory of Distribution," *Journal of Political Economy*, September, 1915; Hobson: Hobson, "The Marginal Theory of Distribution," *Journal of Political Economy*, March,
1915; and Hobson, "The Marginal Theory of Distribution," *Journal of Political Economy*, September, 1915.
In the preceding chapter some account was given of Hobson's
DISTRIBUTION BY VALUE PRODUCTIVITY 477
reputation of the "dosing" method of analysis and of its purported bearing upon the relations of land rent to cost; and the opinion was there expressed that Hobsen's criticism, as formulated by his "Discri-
tion," was not well founded. In his later contributions to the discussion he adopts, as we shall see, a somewhat different view of the matter, but does not abandon the con-
cencyence notion that under any dosing analysis, properly applied, is to be found a certain proportion between the value of each factor and one of the different productive factors in the production process.
Meanwhile, and especially after the reply of Professor Carver, other interesting points have been brought forward which are directly germane to the original question. Clark's doctrine is necessarily involved in the assumption that all men are equally productive.
Hobsen denies that any entrepreneur, e.g., a farmer, deciding to employ a fifth laborer, can properly regard the resulting increase of product as representing an equivalent increase in productivity of this fifth man:
I. The work of this fifth man being by assumption the same as that of the other men, is equally productive, and is paid for at precisely the same rate, because of this precise equality in productivity; if actually employed, he would receive no more than he would get if he were because the employment of the fifth man has lowered the value effectively.
If the employer's managerial capacity leaves him yet able gain-
fully to supervise another—a fifth-man, it must follow that a share of the increased product will go to him. This is a mere fraction of managerial ability which hitherto had run to waste.
2. The payment made by the employer to his laborers does not make an increase in the labor solely, but an increase partly in labor, partly in capital, partly in land; and the dose is a land-labor-capital com-
pound, and not a pure labor-dose. The increase in value attributable to anything in the dose sort, must be imputed to this compound.
Hobsen also argues—perhaps unfortunately, at all events unneces-
sarily—that if the employer's payment were really the full equivalent of the product received by him from his laborers, then these laborers were actually ascertained—and if the productivity theorists were correct in their assumption that they are so—as regards both wages and expenses and measures the value productivity of the
wage-paid labor; it would therefore follow, inasmuch as the other men are supposed to be equally productive with them, that whatever they produce, that there could be no gain in the case for the entrepreneur who pays them wages. For if he pays them wages, he re-
gains a forced gain, something which morally, does not belong to him, that he gets a profit; and that Clark is justified in his view, that under such conditions there can be no equilibrium." all entrepreneur profit would disappear; but that there is, on account of some other factor (which I do not know about) about this equilibrium situation; and that even were it to be estab-
lished, the benefit of it would accrue not to wage-earners exclusively,
but to some other class.
Carver joins issue upon this question of profit; and taking profits
to indicate what is left over "after the other shares are paid," he rightly proceeds to infer that competition can never cancel all entre-
478
VALUE AND DISTRIBUTION
preneur remuneration; and he undertakes a detailed analysis to show that the entrepreneur, in employing more and more men, must finally come upon a man marginal for the purpose of his labor. He must thus find out what is the value of this man marginal, and he must thereby establish to be what this marginal man produces; that is, the product of his labor. But it is not sufficient to regard as producing no more, that the excess in product from the earlier man over what they receive really represents production at all, but it is enough (as Hobson says) "that a product of him who gets it. Seemingly also, as with Clarke, he gets it because he produces it, and is known to have produced it because he gets it."
Now one could know precisely the sense in which Hobson here uses the term "production" in this case alone, where it worth while for our purposes, toward an award of judgment upon the issue as joined, to consider whether the employer or the laborer pre-
pares profit. Certainly also Hobson is right in asserting that where, with free laborers, there is a product of good, and with five laborers a product of two, the employer will raise wages to 1.50 cents per day down profits, will raise the wages to 1.25 cents if we assume that [the] competition of employers is as full and free as that of the laborers.
On the other hand, Hobson is right in his view that a combination of laborers would be productive only if the employer took them all or if the employer must take the five laborers or none.
But, on the other hand, Hobson's argument is also right in insinuating that under the perfect competition of a static situation all profit would be canceled, since profit, as he uses the word, connotes merely that something has been added to the price by reason of imperfect competition, or to both.
But one can see that issue at present concerns us and such was not the issue which it was in Hobson's purpose to present. Nor, accurately speaking, does the question whether the work of the fifth man is independent or subdative concern us; nor does it concern us: for what does the question really mean? Does it mean that it is as important for the employer to employ five laborers as it is for four as it was to have a fourth man to go with three? The question as thus stated answers itself, and there is no other question. There is no question about whether the fifth man's labor is productive or not; it is only productivity relative to the entrepreneur and to his situations. Hobson's repetition of this question is not significant. It is either independent, or subdative; like all forms of utility or of service, it is a relation pure and simple. With a different entrepreneur it would be a different relation; but so long as we are considering any entrepreneur's situation or need, the significance of the instrument becomes a new question. The question whether its use is productive or non-productive can be such only in the sense of a contribution toward value creation under the particular terms and conditions of the situation as its owner finds. The question does not concern a non-productive contribution, but only the market value of this con-
tribution.
Hobson's substantial repetition of Clark's view that the dose is never one of labor alone, or of land, or of capital goods alone, but is
DISTRIBUTION BY VALUE PRODUCTIVITY 479
rather a land-labor-capital composite, a unit only in the sense of a unit complex, goes in one direction somewhat overfar, and in another direction not quite far enough. For surely with his last dollar or last thousand dollars he may buy all that he needs either of labor or capital goods; but as surely, he may buy nothing at all. The truth is that under any one situation or at any one time, one kind of enterprise may be more productive than another, and that the different entrepreneurs being different in degree and kind of ability and of equipment, there is, in truth, never any one situation, but only various situations, in which one kind of enterprise is more productive expense in one way, another in another, and these different ways may none of them be of the "composite" sort, or they may all be of this sort.
But, as we have seen, the truth is not far away; the dose with each enterprise is so small that it can be ignored without introducing a dose of competitive entrepreneur capital. Nowhere in economic theory more seriously perhaps than here has the traditional threefold classification of factors exerted its influence to peripate and complicate and vitiate.
It must be recognized here that, for the purposes of this problem, the distinction between the two classes is a mere abbrevi- ation of what is substantially the truth. Without all courting in the dictum that "the idea that different parts of a product can be created by different agents is a fallacy,"---this notion seeming to cut away from under Clark's feet the only thing that makes his argument valid---it is clear that every analysis---the central idea in the following must command entire approval: "To him it makes no difference whether he hires one agent or another." It is true that he gets the same reward for the same outlay in all cases" (p. 365).
CHAPTER XXIII
THE LAWS OF RETURN
To have established in economic theory, on the one hand, the distinction between long-time and short-time influences or, on the other hand, the distinction between the static and the dynamic in value problems is, irrespective of other and possibly greater grounds of obligation, to have placed the science under no small burden of debt. These are, however, really distinct services; for though, at first impression, they appear to rank as merely different formulations of the same principle of distinction, this is seemingly neither a sympathetic nor a fair interpretation of either.
Marshall's long-time reckoning points rather to such normal or static equilibria as are either reached or always in process of being reached within one general and established situation of fundamental conditions; the short-time reckoning allows for the minor perturbations and rearrangements which yet do not imply movements or tendencies in the direction of radical, permanent, or fundamental change. Clark's even more important recognition of dynamic forces in economics is concerned with those temporary and permanent modifications in the ultimate determinants of value—changes in those greater and ultimately directive forces which have been here somewhat awkwardly denominated situation facts, as the basis of situation costs.
It is, however, obvious, that the two lines of distinction are prone to coalesce, and that long-run influences and dynamic influences, while sufficiently distinct at the extreme, shade off into each other in intermediate cases.
The long-run price is distinguished from all dynamic equilibrium prices as distinguished from those actual and unstable prices likely to obtain at any particular moment. That there
480
THE LAWS OF RETURN
481
take place a merging of actual price into normal price requires the assumption both of a static society and of a long-run period. Nevertheless, the long-run computation is essentially one of an ideal, static, equilibrium price, the normal price, and is a concept far from new in the science.
But up to the time of Clark, the distinction between fundamental change and temporary flux had been only vaguely felt and not clearly defined.
But, as has already been urged, that the distinction between long-time and short-time influences is recognized as valid and illuminating, does not deny the cost relevancy of all those influences whose effect is believed to be transitory; the distinction rightly employed points merely to the temporary nature of these costs, to the probability or the certainty that they will later be greater or smaller. It is indeed, a strange paradox that we can speak with confidence upon the assumption that it is permanent. The distinction really points the way to an investigation of the influences lying behind costs, and to the level of costs which it is the trend of these influences to establish. Mere seasonal and climatic changes, droughts and crop failures, pesti-
lences, famines, whims and flurries of fashion, stand as static influences which render the long-time computation
"Clearly there is the necessity in economic reasoning of regard-
ing man as the subject and central point in economic science; his environment as a field within which he must operate; his en-
eration; his economic activity as his attempt to produce and distribute this product; his income as that which he receives for his work. Normal price is to be conceived as the line of least resistance not only for the buyers and sellers directly engaged, but also for the producers in other enterprises who are affected by their prices. This is why the highest remuneration. Market prices are found to fluctuate in either direction from their normal levels. They do so because of an effec-
tive adjustment of sacrifice, long or widely depart therefrom. In short, the normal price is that price at which no producer can, to his own satisfaction, alter his output without altering his produc-
tion. Prices generally would stand at their normal, if no producer or consumer could alter his output without altering his own manner of economic action. But like the ocean, market values have no rest. Prices ripple and wave above or below their ideal level, as dextrously as waves on an ocean's surface, slowly or rapidly change in force."-Davenport, op. cit., sec. 7.
A page from a book with text discussing economic principles.
482
VALUE AND DISTRIBUTION
safer for purposes of understanding the larger laws of price. According to Marshall, "the value of a thing in the long run tends to measure its cost of production"; not precisely that the value of anything tends to be fixed or governed by its cost of production; "we might as reasonably suppose that the value of an upper blade of a pair of scissors cuts a piece of paper;" but, "as a general rule, the shorter the period which we are considering, the greater must be the share of our attention which we give to the influence of demand on value; and the longer the period, the more important will be the influence of cost of production on value;" all of which must be recognized as valid, since, as we have seen, cost in one industry resolves itself mostly into the demands of other industries; only in the long run can these opportunity-cost influences make themselves adequately felt.
That, as men acquire larger knowledge, strength, and technical skill, they become more effective producers of
*Marshall, op. cit., pp. 426, 429.*
Perhaps this is as good a time as any other for working out some of the details according to which these short-time influences affect the "long-time" effects. The most striking fact is that they are not so plastic. What results must commonly follow some price rise in the product of one industry may not follow at all in another. It may be assumed, through a change in the habits of consumption—
There must first take place, in the short-time adjustment, an intensified utilization of all productive appliances in that industry; a recourse to all available sources of labor and materials; all the other productive agents therein employed, and at the same time an emphasis cast upon other industries for instruments, labor, and materials.
But no very great increase could be brought about in the value of those products whose prices had risen because their supply was mere fact of their being common to several different industries; the larger effect would be felt by those agents having, for the time being at least, an exclusive control over one or more such products.
But this "monopoly" advantage would not work itself out fully; there are many things which cannot be monopolized. And even though that these monopoly agents are competitively remitted by owners to producers; competition could hardly be so swift or so effective as, with rising prices, to bring down the prices of goods which are not affected by them. For example, if we consider the money prices of the intermediate goods, thereby assessing these goods as
THE LAWS OF RETURN
483
wealth; that, with larger and larger supplies of any con-
sumable good, there must go a smaller importance attach-
ing to each successive unit of supply; that, upon any given
area of land, successive increments of product are obtain-
able only on terms of increasing difficulty per unit of
product; that, in many lines of production, the greater
business has in point of economics the advantage over
the smaller business where propositions no one of which is markedly economic or technical in import,
or of a nature to present overserious difficulty of compre-
hension, or of a character to offer especial temptations to
controversy.
Not precisely so, however, for the same propositions as,
after subjection to the necessities of economic analysis,
reinterpretation for the purposes of economic investigation,
and reformulation into the language of economic doctrine,
they present themselves transformed and rearranged into
the well-known "economic laws of return".
Superficially, though not thereby incorrectly regarded,
the ordinary formulations of the law of diminishing return
may be distributed under three heads: (1) A law of falling
higher costs of production. To begin with, the machines could not be
easily moved or sold, and competing producers could not step into
possessions which they had not previously acquired by investing.
In fact, the advantages do not attach to any monopoly agent separately
and in isolation, but to appliances as members of a distinct producing
group, which group may consist either of two or more con-
stituent members. Thus the problem of insufficiency within the group
is at least as great as that of excessiveness. In other words, pre-
pratically speaking, the gains would appear as enhanced possibilities of manage-
ment ; but this is crude in theory.
And if we suppose himself the owner of the different
agents, the difficulty would be precisely the same difficulty; he could not
merely rent out his machines to others without losing control over them.
The monopoly goods might be—would-be—essential to the
situation, but so also, would the situation as a whole be essential; and
the enterprise would have to be carried on by itself. In such a
situation would be equally essential. He could not expeditiously rent or sell the machines to others. The group product would be unique. There-
fore no alternative application would be possible whereby to apportion their shares in the group product. The problem of complementarity is present in an aggravated form; the product is simply and irreducibly a group product.
1 |
A law of falling higher costs of production. |
2 |
In fact, the advantages do not attach to any monopoly agent separately and in isolation, but to appliances as members of a distinct producing group, which group may consist either of two or more constituent members. |
3 |
Thus the problem of insufficiency within the group is at least as great as that of excessiveness. In other words, practically speaking, the gains would appear as enhanced possibilities of management ; but this is crude in theory. |
4 |
If we suppose himself the owner of the different agents, the difficulty would be precisely the same difficulty; he could not merely rent out his machines to others without losing control over them. |
5 |
The monopoly goods might be—would-be—essential to the situation, but so also, would the situation as a whole be essential; and the enterprise would have to be carried on by itself. |
6 |
In such a situation would be equally essential. He could not expeditiously rent or sell the machines to others. |
7 |
The group product would be unique. |
8 |
Therefore no alternative application would be possible whereby to apportion their shares in the group product. |
9 |
The problem of complementarity is present in an aggravated form; the product is simply and irreducibly a group product. |
484
VALUE AND DISTRIBUTION
utility; (2) a law of falling product by weight and tale, a diminishing productivity in terms of concrete, objective, physical measures; (3) a law of falling value productivity.
1. The first formulation of the law with negative increase in the size of the productive complex, the return in utility falls short of proportional increase, obviously a direct corollary from the law of satiation, if not, indeed, a mere repetition of it; thus a law of unquestionable validity for the purposes of the individual reckoning, and indirectly of significance for problems of Crusoe valuation; but equally clearly, a law only vaguely and only average-wise applicable to group-utility computations; and, in any case, a law which can be made out only through the individual comparison of competing utilities.
It is, however, here to be noted that, were all the different agents and instruments of production keeping abreast in rapidity of increase, this fact of failing utility could have no necessary bearing upon exchange relations. But equally for one agent or for all agents taken collectively, the law of diminishing utility, of falling significance with relation to need, must hold, since it holds for all products and services subject by the fact of increasing productivity by weight and tale.
2. The law of falling volume of concrete product, relatively to the enlarging productive complex, has evidently little significance for purposes of any individual competitive reckoning, otherwise than as the weight-and-tale aspect of the case transforms itself into a value-return outcome. And here, also, if all agents were manifesting the same rapidity of increase, it must follow, as will later more fully appear, that none would claim nor as value need any law of diminishing return obtain.
3. The third rendering of the law, as one of diminishing value return, is the only formulation having direct significance for any purposes of the competitive reckoning; and in this regard, also, it will later appear that no matter how clearly manifest the utility fall may conceivably be, there is
THE LAWS OF RETURN
485
no possibility in industry of a generally falling value return.
So, upon the assumption of equality in the rates of increase among productive agents, there is possible neither a diminishing weight-and-tale productivity nor a diminishing value productivity.
Malthus and his successors long since made it clear enough that, looked at solely from the point of view of the land situation, the prospects of the human race are not encouraging. Increasing numbers of human beings must find the food problem progressively a more serious problem; overcrowded land is the same thing as poor land; a larger and larger share of human energies must, then, with expanding population be applied to the solution of the food problem. The law of diminishing return for land is a fundamental fact in human affairs, a fixed, opaque, and brutal fact, full of bad omens and sad prophesies.
And so long as there is any hope for this law of diminishing return there could be no possibility of land shortage, or of that inevitable derivative of land shortage, rent. Thus, as with increasing population, there falls out, per capita, a smaller product to divide, there must also go to the landlords a larger and larger proportion of the more and more tragically inadequate total. The social classes disinherited of land are doomed to a double and compounded pressure of adversity; this law of diminishing return unites them together even more closely than before.
But the optimists also have their innings. All this would be true, other things remaining the same. But other things are not to remain the same; for if there is a law of diminishing return, there is also, it is said, a law of increasing return. If, with relative land famine, a larger share of the productive energies at human disposal must be applied to the land, it will also be true that, with increasing methods and processes in manufactures, we can spare some land for the use of our productive energies. Who knows that progress in one
486
VALUE AND DISTRIBUTION
direction may not more than make good the deficit in the other direction?
And not this alone; progress is possible and is probable, not only in the technique of non-agricultural production, but also in agricultural production itself. Progress of this very sort has indeed been rapid even without the increasing pressure of need. For what has been the meaning of the redistribution of population? It has been, characterising the last two centuries. The urban population has far outstripped in rapidity of increase the agricultural population.
The growth of the small city as against the country, and of the great city as against the small city is one of the most obtrusive facts of modern life; the new and agricultural countries like America and Australia, equally with the older countries, manifest these population redistributions; and on the other hand, in point of the degree of the tendency to rapidly populate, all Europe fails at not at all behind the sparsely and newly settled countries. City growth is general in the modern world.
Why is it? It is fruitless to search for the fundamental explanation in the improvement of industrial processes. Only such men can work in manufacturing as can be spared from the processes of food production. As long as the food product from one man's labor sufficed for the food requirements of one man, so the entire population was contented to occupy itself agriculturally; when now one man's labor will feed three men, two-thirds of the population may be urban. So also, the development of transportation serves for the most part to explain, not why so large a proportion of the population is now agricultural, but only the distribution of the non-agricultural population. To the extent solely that transportation has opened up more land or lower grades of land to agricultural cultivation, it may be rather said that the processes of agricultural production is transportation responsible for the growth of non-agricultural employment. And precisely here it should be remarked that to the extent
THE LAWS OF RETURN
487
that, in the production of implements and appliances,
manufacturing is itself an agricultural process, to pre-
cisely this extent industrial improvement must have aided
the relative growth of the urban population.
Improving transportation, then, so far as it is not at the
same time to be regarded as improving agriculture, has
had its effect, not in emphasizing the growth of urban
as against agricultural population, but in fostering the
growth of the great city against the village and of the
great city as against the small city.
Looked at from a more distinctly technological point of
view, this truth would read that transportation has fostered
the giant industry as over against many small competing
units.
Malthus in his formulation of the law of diminishing
return for land was very plainly proceeding from a purely
social and general point of view, rather than from the
point of view of the distinct and independent and com-
petitive interest. That the law of increasing return, con-
ceived as summing up the optimistic effects in the social
outlook, is equally a non-competitive formulation is equally
clear.
But, after all, what part, if any, of all this raw material
of optimism is accurately speaking, embraced within the
economic law of increasing return? So far, all the
"returns" suggested have sounded in terms of social service
—of group or race utility, of quantum of productivity by
measure of concrete item product, a purely weight-and-
tale standard and basis of computation. And it is unques-
tionable that, for certain purposes and from certain points
of view, this interpretation of the laws of return is not
merely a possible one, but is the sole interpretation either
relevant or necessary. But it is equally beyond question that,
for certain other points of view and for certain other com-
putations, measures of utility return, at the one extreme
and of value return at the other extreme are much more to
488
VALUE AND DISTRIBUTION
the point. For most purposes in the competitive reckon-
ing only laws of value return can have significance.
But does the law of increasing return, accurately
formulated, have exclusive reference to such industrial
effects as are due to the development of the human fac-
tor in industry? Does it not also indicate that there is me-
dium power, in zeal and persistency of effort, in scientific
knowledge, in control of technological methods and appli-
ances, or, finally, in the advantages and methods of “team play” as exhibited in higher forms of organization? That
is to say, does the law merely affirm that the better the pro-
ducer the larger the aggregate social product, precisely as
in the mathematical reasoning we assert that the less
adequate in quantity or quality is the smaller the return
from any activity applied to the land?
But, so interpreted, the law of increasing return applies,
equally with the law of diminishing return, to agriculture;
and the law of diminishing return applies, equally with
increasing return, to manufacturing. Agriculture bene-
fits by good appliances, by good transportation, and by
zeal and care and intelligence in supervision and in organi-
zation; manufacturing suffers by every inadequacy of
equipment.
Or does the law of increasing return assert that some-
how, as manufacturing in the aggregate comes to employ
more men (or more capital?), it makes more than pro-
portionate increase in its weight-and-tale productivity?
So understood—and irrespective of the effect of concentra-
tion into larger and larger productive units—there is only
so much in the doctrine as may be implied through the
division of labor among different branches of manufac-
tures; and here again, the principle applies unequally to
different manufacturing industries, and while perhaps
applying more noticeably to manufactures as a whole than
to agriculture as a whole, applies to some branches of agri-
culture in higher degree than to some branches of manu-
facture.
-
THE LAWS OF RETURN
Or does the law in question assert that, with organiza-
tion into larger production units, there results an increase
in the weight-and-tale productivity of manufacturing
industries in the aggregate? Here again, the advocates of
large-scale work in England or among the banana farm-
ers of the North-western States insist that the law is also in
degree an agricultural law.*
Or is the law to the effect that, among competing units
of production, the relatively large competitor has the rela-
tively large weight-and-tale product?
Or does the law run that among competing producers
the relatively large units get better returns in value product
in proportion to the size of their production?
And in what character is the law interpreted in this com-
petitive and value sense, is it to be taken to compare the
average entrepreneur costs and average value produc-
tivity of different units of production, or rather only their marginal value costs and their marginal value productivity?
And this leads us to the question whether the law is framed as primarily of service in the determination of comparative profits and thereby as explaining the trend
*Carver makes the following especially illuminating observation:
"Confusion has sometimes resulted from a failure to distinguish the law of diminishing returns from the law of increasing returns and from the comparative economy of large- and small-scale production. It is,
for example, sometimes assumed that because a factory can run more economically and turn out product at a lower cost, than can a small one, but because a small one can run more economically than a large one on a given piece of land, or that it could do so under certain conditions in conjunction with large plants than with small one of the same kind.
Among the various questions on which the manager of such a unit has to determine his policy there are two which are particularly pertinent in which to combine the various factors: (1) What is the best size for which he should build? (2) What is the best combination in which he should operate? The first question relates to the varying productivity of an industrial unit when the factors affecting it are varied. The second question relates to Value productivity?. On the other hand the law which relates to the comparative productivity of large- and small-scale production has to do primarily with the size of the unit."—77 Distribution of Wealth, pp. 64, 65.
490
VALUE AND DISTRIBUTION
of industry toward the giant organization, or rather as explaining the bearing of giant organization upon market prices, and so forth, also the relation of these market prices to the productivity and the remunerations of the various productive agents.
At any rate, the law can hardly be one of increasing proportional value productivity with increasing size of the productive unit, unless the law is taken to apply not to industries taken as a group aggregate, but to the competing industries inside the group; for it may readily be true that the organization of any industry into the giant form should so reduce its cost per unit of product, by weight and tale, the aggregate value of the product should be a diminished one; and this might hold of manufactures as a whole as over against agriculture as a whole.
Not can the law rightly mean that greater value productivity goes, per unit of expense, with increasing size. This is not necessarily true; it is safe to assert only that to the greater industrial unit goes the relatively greater profit. For, where the elasticity of consumption is not great, and where the price of the product is not very high, lower prices may obtain to an extent to bring a lower value productivity for each, and a generally lower average of profits; and yet it may remain true that the greater units suffer least, that to the larger units there accrues a relative advantage.
Or does the law run only to the effect that, in industries of heavy investment and heavy fixed charges, the extra expense of such large items of product is less than proportional to the increase in production, a law which, as of necessity, says nothing as to the aggregate increase in value going with the increase of product, but leaves it possible to be assumed that the entrepreneur will limit his product at the point where the extra expense of production, together with the falling prices upon the original output, balances the extra value represented in the added items?
THE LAWS OF RETURN
If this last is the significance of the law, a danger signal is called for; monopoly production would, it is true, follow the policy outlined; but with competing producers really competing, there is, as trust promoters and trust apologists have correctly urged, and as the influences behind such competing production are clearly assured; competition may bring prices down nearly or quite to the level of the costs of the extra product, practically canceling the earning power of the fixed-charge portion of the investment.*
It appears, then, that to find what there really is in this law of increasing return it is necessary rigidly to exclude all influences of improving technique, developing human beings, and all influences ranking under increasing demand for products, and to confine ourselves to the sheer competitive advantages of combination and concentration,
(1) for increased weight-and-tale product per unit of expense, (2) for increased value product per unit of expense. Evidently (1) may be found without (2), though (2) is impossible in the absence of (1).
Note that this law does not explain why this law of increasing return might not characterize all industries. If it does not, or if it does so unequally, the reason must be sought in the peculiar nature of the industries in question. The law may fail to hold with certain industries, because by the nature of the instruments which they employ, or of the processes required, e.g., as with land, the business unit cannot greatly increase, the giant organization being impracticable; or the market may be of so limited powers of concentration that no great increase can be possible.
At any rate, the law is not one referring by necessity to the interdependence of factors or to the constitution of the business unit in respect to the factors included. The law might hold for one industry almost exclusively labor-employing, or land-employing, or machine-employing. For
* Cf. Marshall, op. cit., pp. 448-460.
491
403
VALUE AND DISTRIBUTION
the purposes of the law more labor may apply itself to labor as well as to land or capital; or the advantage of increasing size may be obtained by adding more capital goods to an existing capital undertaking, or by adding more acres of land to the acres already employed.
The law of diminishing returns is perhaps even more difficult in making precise and definite. There is the same tendency to oscillate between value formulations or implications and the weight-and-tale type of concept. Malthus, as we have seen, gave to the doctrine a distinctly social significance, and thereby of necessity a definitely weight-and-tale type of formulation.
But earlier than the time of Malthus' formulation of the law in terms of population and subsistence, the problem of agricultural returns had received thoughtful and authoritative consideration, whereof the two aspects of the law sought here could not largely discussion attained some measure of vague differentiation.
Canillon, for example, though regarding the problem as, on the outlay side, one of labor units of investment, had yet, in prophetic anticipation of Physiocratic doctrine, rendered over this labor into terms of value cost according to the quantum of subsistence material, land-productive power, embodied in these units. It is, however, true that the signification of this cost case is for the most part regarded as measurable in units of time.
But with Quesnay the talk is, on the cost side, wholly of labor and capital conceived as reduced to an entrepreneur common denominator, and as aggregated under the head of "depreciation", a competitive formulation and a value rendering; at the same time, the return, the product net, is conceived as a value surplus over the invested capital outlay.
Malthus' doctrine was evidently not directed to the elucidation of the law of rent, and was formulated in advance of any well-con sidered and widely accepted doctrine upon
THE LAWS OF RETURN
493
the land-rent problem. Mostly, perhaps, because the dis-
cussion was innocent of rental connotations, the formula-
tion was consistent and free from confusion. But in later
discussion, and especially since there has appeared to
exist a close relation between diminishing return and the
rent-cost issue, there has prevailed an almost uninterrupted
confusion.*
But all of this should become clearer after a catalogue of the different concepts of diminishing returns has been
attempted and an analysis of these concepts completed.
I. Based upon the law of satiety applicable to any stock of
consumption goods in the hands of any single indi-
*Proof of this assertion is in the nature of the case almost
impossible of giving; some random citations out of the latest of
economic literature are, however, offered:
"The law of diminishing returns is simply a part of the general
observation that the product (concrete product? value product?) of
any given amount of labor (time sum? pain sum? value sum?) and capital (how
are labor and capital aggregated?) used in producing it."—Carver,
"Natural Law."
"Though large applications of labor and capital may continue to
produce more than one man's share, yet each man will produce less with
labor and capital."—Id., p. 26. "[The weight-and-tale aspect of the
crop is here compared with labor somehow measured, plus capital of
some sort, and then multiplied by the number of men employed.]"
And on pages 28, 29, and 50 of the same work the tables given
make clear that "if one man had twice as much labor and twice as many
bushels of product, and state the results in bushels per day's labor."
"But on page 60 the value formulation of the law is presented:
"[Wherever] one man can get more than his share with his own labor,
and capital [how limited?] to the growing up one crop on more than
one great field, he will do so; but he will not try to get more than his share
to do so than to concentrate all his energies on his best land."
However, later on the same page the discussion lapses into the
other conception which is that if one man had only twenty days' labor to use, he could get more bushels by concentrating his efforts on his best land.
"The law of diminishing returns relates to the amount (?) which
can be produced on a given piece of land [area?] by varying amounts
of labor and capital. It does not relate to the amount which can be pro-
duced on any given piece of land but does not increase in proportion to its area.
Entirely justifiably Carver takes issue with Bullock (cf. "The Variation of Productive Forces," Quarterly Journal of Economics, August, 1904), in regarding the law of diminishing returns as applicable
494 VALUE AND DISTRIBUTION
vidual, and upon the derived concept of marginal utility, there has been, as we have already seen, somehow deduced not only to hand but to all forms of combinations of productive factors; Carver and others have shown that this is true. But it does not show every possible application of this extension of the law of diminishing returns which require a separate term for each and every kind of labor, land, and capital. . . . The following simple formula will have to suffice:
If X with Y will produce P |
Then X with aZ will produce more than aP (Increasing return) |
And X with aZ will produce more than aP (Increasing economy of large-scale production) |
less than aP (Diminishing economy of large-scale production) |
"It is assumed that a is a positive quantity greater than zero" (p. 60). But what? And I of what? We must here land as superfluous, plus labor and land somehow aggregated, and the whole set over against weight-and-scale product. So on pages go on and g.
"There is another factor - - - with which we must reckon, and to which we must give due consideration. This is the fact that establish- ment is a combination of various factors under one management, and the proportion between them is not fixed. Therefore, a question of the proportion between the factor called manage- ment, on the one hand, and all the other factors, on the other. [The] formula given above may be modified by adding a new factor of increasing or decreasing economy of large-scale production, may be modified as follows, to take account of this new factor:
If M with X with Y with Z will produce P |
Then M with aX with aY with aZ will produce more than aP (Increasing economy of large-scale production) |
less than aP (Diminishing economy of large-scale production.) |
The following quotations illustrate, within the limits of the para-
THE LAWS OF RETURN
A law of falling market price in society for any increasing supply of consumption goods. The haziness of reason- graph quoted, a shift from one point of view to the other: questions of "pay" as value questions, not "amount" questions.
An increase in the amount of land does not necessarily give an increase of land will never, in any normal case, increase the product as much as the labor is employed upon it. For example, suppose that a farmer pays to cultivate land beyond the point where diminishing returns begin, if he pays to cultivate it on the land which it never pays to cultivate it up to the point where it begins to pay, then an increase in the amount of land will increase the gross product. But since our little land is never profitably used in connection with labor, we may assume that an increase in the amount per unit of land, it follows that, in any normal case, an increase in the land with such given amount of labor will not increase the product as much as the labor.
In other cases the argument goes clearly over to the value-return points of view. Generally there can be no talk of substituting one factor for another, or of the various factors being combined in combination, excepting upon the value-return basis.
When we consider the question whether we pay to combine them in such proportions that if any one or two of them were increased it would increase the product, but not so much as a variable factor, or factors, we find that this is not always true. It depends upon proportion in which to combine the various factors of production?
As already pointed out, there are several ways to grow a hundred bushels of corn. One way is to use a great deal of labor and capital to pro- duce a heavy crop, but getting a small product per unit of labor. Another is to use little labor with much land and enabling the land to produce more than it could do without labor and capital. In this latter method, the former is the better method; but where land is cheap (this is not always true) the latter method is better. . . . The question which is the better method depends upon the relative cost of the two factors' (pp. 76, 78).
Seager's formal statement of the law reads as one of labor and capital act over time. He says: "The law that labor and capital has been paid for by the cultivation of an acre of land or the exploitation of a mine, increased applications of labor and capital yield less than proportional increases in output." This statement is based upon his belief that in the method of cultivation beyond the roughness scratching over the soil away and pressing down into the earth. As soon as this stage is reached when crops after a certain point has been passed, all experience confirms the law that further improvements afford less than propor- tionate increases in output. The same thing holds true with respect to rendering is elsewhere, but less formally presented: "The final 'dose' of labor and capital required to get a crop may be just as large as before its price because of the additional product they have produced thereon. It is to his interest to continue his cultivation so long as it is remunerative. But all earlier applications of labor and capital will be more than
496
VALUE AND DISTRIBUTION
ing by which has been achieved this affilation of demand price upon marginal utility, and of falling individual demand price upon falling personal marginal utility, and of market price upon some assumed social marginal utility, need not here again concern us. But the further step by which have been worked out a falling marginal significance and a falling marginal demand price for increasing supplies of productive instruments, requires especial attention at this point.
The law of falling price with increasing supplies of consumption goods holds in its usual formulation only because the demand schedule with any one line of consumption goods may be taken as a fixed fact; new supplies can be covered by the price received for what they added to the product" (p. 117).
Seligman: "In the case of agricultural land... additional doses of capital and labor will yield a relatively smaller produce."—Principes, p. 308. "Whenever double the amount of exercise yields more than double the product, we have the law of increasing returns or decreasing cost. When double the exercise just fails to do so, we have the law of constant returns or constant cost" (p. 450).
Gide (Principes d'Économie Politique, sixième éd., p. 132) states in one formulation both laws, that of return and that of value return, as follows: "We know that quantity of labor does not increase and that it does not particularly matter whether quantity of labor be compared with value of product or quantity of product be compared with value of labor; but when we compare quantity of labor with increas- dent agitateur ne peut, à la rigueur, accroître le rendement seulement, passé un certain seuil de travail qu'il croisse en sorte qu'il arrive un moment ou l'effort à exercer pour tirer le rendement serait hors de proportion avec le produit qu'il donne. C'est là une loi nouvelle que nous venons de blé... Supposons que ces 15 hect. de blé représentent 100 journaux de travail et que les frais de transport et de vente repre- sentent trois francs de frais; la proposition revient à dire que pour faire progresser le rendement deux fois plus de blé, soit 30 hect., les frais deviennent plus de trois francs de frais; pour doubler le produit, il faudra peut-être tripler, peut-être quadrupler les frais de transport et les frais. C'est là ce qu'on appelle la loi du rendement non-proportional (inproportional au travail)."
Pine's rendering of the law, while ambiguous for the purposes in hand, may, it must be admitted, easily be interpreted into accuracy;
THE LAWS OF RETURN
497
marketed only on terms of such price as will tap lower levels of price-paying disposition. If, however, the increase is one of a productive agent, there results a new and larger volume of value product and a rearrangement of the conditions of demand; the new level of remuneration is to be worked out only as the outcome of a new problem of distribution, upon the assumption of a new volume of value product and a new set of other factors, with an unadjusted set of productive agents. So, then, with population increasing relatively to the other factors, there may be expected a fall in the level of wages, but this only by virtue of two influences, (1) a less than proportional increase in the product to be distributed, (2) less favorable terms of distribution for labor relatively to the other agents concerned in the technological process. The rule and the reason:
"Increase of labor and capital devoted to the cultivation of a given piece of land will, at any rate after a certain degree of thoroughness of cultivation has been attained, yield more than proportionate, but that increase will be in a constantly decreasing proportion to the labor and capital to which it is due." -Economic Principles, p. 98.
In the main and in general purpose Peter's formulation sounds consistent with the statement that "the law of diminishing returns" is not absolute, e.g.: "Economic diminishing returns always has reference to value. If a particular kind and amount of a certain material is used in varying proportions in different processes, the added product will not always be in the same proportion to the value of the added agent as it was when it was first used. For example, whether the added material will add to strength, but what it will cost, and whether the result will justify the expense. So the economic problem of diminishing returns is not so much "the limitation of production" (Principles, p. 64). But it is none the less true that the discussion is always near to the point of dropping into utilitarian calculations: as for example, in discussing the question whether a particular industrial agent is the expression of the fact that there is an elastic limit to the utility of goods produced by that agent. And again upon page 71: "Diminishing returns of indirect agents is a special case of the universal law of the diminishing utility of goods. Diminishing returns are those cases where an additional unit of gratification has to do with direct or consumption goods. They are two species or forms of diminishing returns. The effect of an additional agent, added to a fixed number of other agents with which it is technically used, is credited with a diminished utility just as an additional supply of employable goods coming to meet a fixed demand, fails in value."
A page from an economics textbook.
498
VALUE AND DISTRIBUTION
soning valid for increasing supplies of consumption goods will not hold for production goods.
II. The Malthusian rendering of the law of diminishing returns has no concern with any principle of rising or falling utility or of rising or falling price, as supplies of consumption goods may either expand or contract. The assertion is merely that successive increments of labor applied to a fixed area of land must be remunerated by a less than proportional increase in objective, concrete, material output:
This law is purely a derivative from the technology of agriculture; if economic at all, it is only borrowed so; it conceives labor roughly in terms of time, or, possibly, in units of effort or stress or pain, but certainly not in terms of value measures or of wage outlays. Nor is the product thought of as in any sense a value aggregate; nor are profits distinguished from wages relevant to the objectively and exclusively social significance of the point of view. The tacit assumptions are (1) that land is present in unchanging quantities and qualities; (2) that labor, unmodified in quality but increasing in quantity, must apply itself to the fixed land situation. The law signifies merely that in agricultural production different productive factors are required; with sparse population some of these factors are gratuitously present and others must be paid for. With more people, more and more upon the non-gratuitous factors, this is, upon labor directly or upon stored-up wealth; all of which sums up to mean that for all social purposes crowded land is synonymous with poor land.
III. That successive increments of labor and capital, applied to a fixed area of land, must be rewarded by a less than proportionate increase in weight-and-tale product:
*As* will later be pointed out the prevailing classification of productive factors is a most dubious one, even for dynamic problems; land possesses many properties which cannot be reduced in the lump or aggregate, without distinction of grades or occupations. But for present purposes these considerations are, perhaps, not especially significant.
THE LAWS OF RETURN 499
This is a law midway in transition from a social and material computation to a competitive and value computation. Precisely how labor and capital are to be aggregated, whether upon a value basis, or upon the notion simply that capital is merely indirect labor, and how, if the value aspect of the labor and capital is accepted, they are conceived to be related to the land as mere area, and how labor value and capital value and land value are competitively relevant to each other, are problems incidentally past knowing. The law is one of meddled thinking.
- IV. That with an increasing money quantum of costs upon a fixed area of land goes a weight-and-tale product less than proportional to the increasing money costs:
This law is manifestly one of more nearly completed transition to a value basis. It assumes the land area as fixed and, together with this, a fixed entrepreneur activity of supervision; and it conceives the application of productive energies into the value of land, and at all other any distinction from its purpose between capital outlays for labor, and capital outlays for instrumental goods; but it is not at all clear that either the fixed land area or the fixed entrepreneur activity has been carried over into a value rendering; the product remains a non-value fact, though it would be an easy step to carry this over into the value denominator, by assuming that the same price level holds for the new product as for the old.*
V. Upon a fixed area of land, an increasing expense outlay with the entrepreneur activity of super-
*In this general field of analysis for which the best work known to the present writer is that of Professor Commons in his Distribution of Wealth (Macmillan, 1893). The following passages especially deserve citation:
"The land-owner does not produce goods for his own consumption, but for sale. Hence his land is valuable to him in proportion to the exchange value of the products which he can obtain from it. He cannot usually afford room for the production of so large a supply of goods as he might wish. The prices at which these products are sold depend on their relative values, and these values are determined by the general forces of society, operating throughout the world." . . . So far, then, as the given area is con- cerned, the price per unit of its product changes so little that we may
500
VALUE AND DISTRIBUTION
vision, is not remunerated with a proportional increase of value product:
Here the adoption of value outlay as over against value product is approximately complete; nevertheless, the land fact and the supervision fact are not fully assimilated to the value reckoning. Capital is conceived as invested indifferently in labor or in instrumental labor and hired; thus, the cost consists of the production complex is labor value + capital value + land area + entrepreneur activity; these, as combined in the productive process, give somehow a value product.
VI. That with a fixed investment in land, an entrepreneur cannot, with successive increments of capital outlay, obtain increases of product proportional in value to the increase in capital expense.
Here all parts of the production complex, with the exception of entrepreneur activity of supervision, are reduced to the capital-value denominator. But both land value and entrepreneur activity are conceived as constants.
VII. That with fixed and valued investment in land, and with fixed and valued entrepreneur activity in production, successive additions of expense for labor or for instrumental goods or for both are remunerated by less than proportional additions to the value product:
regard it as fixed and constant. The total value of its product varies, therefore, only with changes in its own price, and this is subject to the law of diminishing returns, so also must be its value (p. 38).
"The reason why average returns are not as high . . . because successive increments yield a less return than would the single increment. This is true even when the law of diminution of diminishing returns has reference to a possible set of circumstances which might have been realized had the investments been different. The investments might be increased or diminished . . . Where twenty laborers are employed with an aggregate product of 3,000 bushels, and an average product of 150 bushels per laborer; then if two more laborers be employed with an aggregate product of 2,000 bushels each, to the next two, a product of 140 bushels each; then if three more laborers be employed with an aggregate product of 1,800 bushels each, to the next two, a product of 130 bushels each; then if four more laborers be employed with an aggregate product of 1,600 bushels each, to the next two, a product of 120 bushels each; it is not the actual product of the different laborers, since they all possess equal efficiency; but from an analogy with what they would produce under the different conditions we are justified in dividing up the aggregate product in this manner" (p. 39).
THE LAWS OF RETURN
501
Subjected to reinterpretation, this appears to mean no more than if value-wise you double only a part of your outlays of production, you will fall more or less short of doubling the value output, a proposition not seriously questionable as doctrine and not especially fertile of new truth. There is also to be noted, in passing, the assumption made—an assumption common to practically all discussion in this connection—that the costs of land hire, entrepreneur capital hire, the productive process are actually and necessarily restricted to four directions of expenditure, land hire, capital hire, labor hire, and supervision charge.
VIII. A law of rising costs of production generally and of rising market prices for land products, with increasing population:
It is assumed that with increasing population there must go in increasing land scarcity; and it is assumed, in addition, that, because of the relatively inelastic quality of this increased consumption need, production must take place at a high level of entrepreneur cost, and that this production will cease only at the point where the necessary cost outweighs the possible price; and that in the process of adjusting and distributing the outlays, the increased selling-price of the product will mostly go for the services of these necessary agents which, relatively to the increasing demand are assumed to have become relatively scarce, e.g., the land.
That is to say, this formulation of the law conceives of all entrepreneur debits against production as equally costs of production under the value measure; in other words, it reduces land hires and all other hires, together with the entrepreneur's own necessary remuneration, to the value statement as supply-limiting resistances offered to entrepreneur activity as applied to this particular line of production.
But this formulation, while not open to attack for inaccuracy or for lack of business actuality, is evidently a formulation which assumes, as fundamental to the entrepreneur
502
VALUE AND DISTRIBUTION
computation, an existing system of value costs, and, among other things, assumes such action of distributive forces as have attributed a high rent, a value hire, to the better qualities of land, and this by very virtue of the fact that these better qualities of land excuse the producer from high alternative expenses for other productive agents.
That is to say, this law reports the results of an economic analysis of the distribution of values; but it is a law summarizing distributive results rather than a law of production making toward an ultimate solution of the value problem; in last analysis, it is a law of value determination only in so far as it points to the underlying and funda-
mental conditions determinative of entrepreneur activity, entrepreneur costs, and entrepreneur value adjustments.
The relative scarcity of those agents requisite for certain productive processes is an explanation of their cost; for the relatively high exchange valuation of these products. These high values are, in point of process, reached and adjusted as the outcome of the entrepreneur system of production; in the course of the process, as means to the result, and as motivated by the result, there come to be imputed to the scarce agents, under entrepreneur bidding, their high rental and sale values.
IX. That with increasing population there go in agriculture increasing rents and values by the entrepreneur, relatively to the value product:
But if this is true, it must be true only as based upon some distributive analysis not contained in the law and not in terms appealed to by the law. On the face of it, there appears to be no reason why, with increasing population, agriculture should become less profitable to the entrepre-
neur cultivator. If his costs, rent and other, become in the aggregate higher, so might also the value of his products be higher. But this does not mean that he must incur this again, if true, must deduce its warrant from some source outside the law,—by appeal, perhaps, to the same distribu-
tive analysis as that upon which the law itself is based.
THE LAWS OF RETURN 503
Why, indeed, must rent be higher, or, if higher, the other distributive shares also lower? The law—valid doubtless—is distributive in tenor; land receives its larger ratio of the value product, but this depends as much upon the peculiarities of the demand for the product as upon the technological conditions of its production. With some commodities, indeed, increasing difficulty in production cancels all the value of the agent. That there could be no land scarcity but for the limited productive powers of any limited area of land; that is, that high rents do not exist in the absence of such a limitation, because of the diminishing responsiveness of land, leaves it, in the absence of other conditions, still possible that rent should fall with the diminishing weight-and-tale return. This form of the law of diminishing return is, then, not so much a law illuminating other problems as deriving illumination from other solutions; and all laws either of diminishing or of increasing value return are necessarily of this sort.
X. That with increasing population there goes in agriculture an increasing entrepreneur value outlay relatively to weight-and-tale product:
A useless formulation excepting as working out to the purport of law VIII, viz., that the product being, relatively to costs, smaller in value, the price per unit must be higher.
XI. That with increasing population there goes a higher rental and exchange value for land.
The same assumption of progressive land scarcity is made here as in VIII; the law is, indeed, in some respects practically a restatement of VIII, but with some difference in emphasis. Underlying either law is the assumption that the food demand increases in approximate proportion to the increase in population, that despite the fact that labor and capital are contributory factors, their relation to land is rather complementary than substitutional, and that thereby the emphasis of demand falls upon land as productive instrument.
504
VALUE AND DISTRIBUTION
And it is to be noted that, together with increasing population, there is taken to go without saying an increased supply of labor effectiveness, and, almost without saying, an increased aggregate of instrumental goods—technological capital. But, in fact, were the average quality of labor so much deteriorated as to amount in the aggregate to no increase in labor power and as to permit no increase of instrumental goods, it would still be true, the food requirement being taken as increasing with the numerical increase in population, that the demand upon land would rest upon land. All lines of product would suffer in output, but non-food lines more markedly than food lines, and this primarily because of so high a value upon food as to displace, in some measure, other lines of production, to widen the land-value differentials above the marginal powers, and to give high rental values to the land.
But, if so, there must be behind all this the tacit conditio or assumption that the demand for agricultural products traces up to the specific economic quality of consumption. If with a doubled population, land powers being assumed to be fixed, there went no increase of labor power or of capital power, there could be no increase of land emphasis, if only it were as easy to cut down food consumption as other consumption.
But still more important, because more actual, is the possibility that with the increase in population should go a more marked progressive increase in the per-capita efficiency of men as wealth-producers: what then? The present rendering of the law appears to say that, with the factors of production other than land outstripping land in rate of increase, those products markedly of land origin will rise in value, and that therewith will also rise the rent and the value of the unit area of land.
But inasmuch as land is not the only productive factor in agriculture, it must be recognized that the demand for agricultural products is at the same time a demand for non-land instruments of production, that is to say, a demand
THE LAWS OF RETURN 505
for labor and capital. Therefore, were it true that labor and capital were in the main substitutionary rather than complementary in their technological relations to land in agricultural production, there could be no such redistribution of production-distribution advantages as is implied in the law as formulated. Thus, also, if it were true of land that it could take the place of labor and capital in any large share of their uses, rather than serving as co-operating complementary productive instrument with them, land could never have experienced the fall in rents and in market price characteristic of the past two centuries.
It follows that under normal conditions of increasing population, of only proportionately increasing food requirement, and of more than proportionately increasing labor effectiveness and capital expansion, there is no basis of foretelling the effect upon land rents and land values otherwise than accordingly as assumptions are made as to the direction of human development. Improvements in transportation or in the technique of production may be so great in some directions as to aggravate the demand for the best grades of land free land and thus to cancel the importance of rent in the computation of costs, and to leave the food supply to find its relative cost purely in the relative outlays for labor and for non-land instrumental goods.
But in the degree that the actual inelasticity of the consumption of agricultural products—not all are food products—is overstated in the assumption, and in the degree that the development of human productive powers may not take place so rapidly as to make this assumption obsolete, a solution undergo modification. On the other hand, a gain in land rent must follow upon an increase of non-substitutionary supplies of labor; but whether this would be accompanied with any marked rise in the prices of food products would depend greatly upon the character of the increase in the supply of labor, whether, for example, and in what degree, it were adapted to other than agricultural production.
506 VALUE AND DISTRIBUTION
If these solutions shall appear to be disappointingly vague and conjectural, the explanation must be sought in the nature of the problem and of the assumptions neces-
sarily attendant.
The foregoing analysis of the various different con-
cepts of diminishing return—and there are possibly still others waiting to be catalogued—should have sufficed to make it clear that such formulations of the law as promise significance or serviceability for economic purposes of any sort, competitive or collective, are limited in number.
I. A law of statistical generalization foretell-
ing a diminution in the per-capita command of consum-
able goods, by reason solely of the society coming to
contain more members, these being assumed to be substan-
tially unmodified in all relevant aspects.
a. A law in the dynamics of competitive economics; a forecast of changes in the relative distributive shares accruing to the different agents and instruments in produc-
tion terms, due to changes in the relative supply of these concrete factors; thereby,
changes in their relative value through the capitalization of
their income-earning power; and thereby, also, upon the
supply side (and for whatever the cost of production com-
putation may be worth), and solely by reason of these
distributive shares functioning as costs in competitive entre-
preneur production, a forecast of the relative changes in
the market price of the various sorts of commodities
technically dependent on variable factors, and the use
of these differently remunerated productive facts.
3. A static, competitive, entrepreneur law expressing
the disadvantages accruing to the entrepreneur from any
relative excess or defect in the quantities employed of any
productive agent or agents, in view of the existing levels
of compensation for these different agents—a law formu-
lating the disadvantage from the unskilful combination of
cost goods.
THE LAWS OF RETURN 507
It is to be noted with regard to Law (2), the population-food law, that the reasoning upon which it is formulated abstracts entirely from the possibility that human development may—at least in some measure—avail to enlarge the land supply, from the possibility also that agricultural technique and transportation effectiveness may appreciably modify the situation, and finally from the possibility that the sources of food supply may not remain essentially agricultural, and that the food requirement may not increase precisely in the ratio that the labor supply increases. That, upon the assumptions made, the social product, food included, will continue to be progressively affected seems to be a ready and necessary inference from the general principle that if some, but not all, of the productive factors are doubled, the weight-and-tale product will not fully double.
Law (3) carries over into the competitive field this principle that a shortage in any one factor of production affects the product unfavorably; those factors increasing least rapidly, i.e., those whose relative scarcity is less pronounced scarcity, and thereby acquire scarcity values as compared with the increasing factors, the degree of the scarcity depending upon the degree of technological monopoly held by the factor in question, that is, upon the substitutions and upon the terms of the substitutions possible in the case,—a changing problem with every change in the direction of human efficiency in the technique of production.
With every change in the per-capita productivity of society, human needs and desires being assumed as a constant, there goes, of course, a fall in the average significance of new increments of product, a law of diminishing utility return per item of weight-and-tale product. But while this law is consistent with an expanding weight-and-tale productivity—depends indeed, upon it—the law is not inconsistent with unchanging conditions of value productivity. For value productivity is distributive in reference—
A page from a book titled "The Laws of Return" on page 507.
508 VALUE AND DISTRIBUTION
has to do, that is, with the different factors considered as share-takers in the division of product, and with the relation of the different shares to one another. Being a question of value, of exchange relations, it becomes a question of the relative shares in the total output of value product. Thus if all factors in production were increasing with equal relative rapidity, two yards of cloth, two pairs of shoes, and two bushels of wheat being produced where before the product was only one of each, there is no necessary reason why there should be either increasing or decreasing returns to any factor in this particular situation. If, then, there is occasion to forecast diminishing returns for any factor, it seemingly must be by reason of some relative inelasticity of supply supposed to attach to some other factor or factors in production. It looks, then, as if the peculiar conditions surrounding the land supply must furnish the only serious basis and occasion for all diminishing-return discussion, so far as it is something more than an enmusement or disadvantage from a badly constructed problem complex.
But upon what assumption could there really take place
"It is of course unfortunate that the established terminology should speak of this principle as "diminishing returns to land," whereas there are diminishing returns of any sort, these not only for the land but for the factors co-operating with the land. This purity as a pro- duction principle is not necessarily true absolutely even for the other factors, but only for the factors in the aggregate, as a pro- duction principle. Inasmuch as diminishing returns may be either of the absolute or of the relative shares, the discussion has passed over from the field of technological production to the field of value distribution. The discussion has been made from the point of view of production, but at its results from the point of view of distribution. The result is that we have been dealing with products produced by any separate factor, but not with the value resulting attributed or imputed to any factor. And as a conclusive result, the fact may be thus, that despite diminishing returns to land, there may be an increasing return, both relative and absolute, to some among the factors, while at the same time there may be diminishing returns rather speak of diminishing returns to the non-land factors; while as a production principle it would seem that if any shares of values should arise; the talk should be of diminishing returns in agriculture, or of diminishing returns in landed production, but not of diminishing returns to the land."
14
THE LAWS OF RETURN 509
this equal proportional expansion in all productive factors, in such sort as to leave the value resultant one of unmodi-
fied exchange ratios and the distributive outcome one of unchanged proportionate shares? Only upon the assump-
tion that the increase in the supply of each factor of pro-
duction has been accompanied by a proportionate increase
in the demand for the various commodities especially
dependent technologically upon the respective factors. But
increase in labor supply, if it comes by increase in number
of human beings, rather than by expanding individual
efficiency, is certain to do at least one thing, viz., to bring
with it an approximately proportional increase in the food
requirement; and precisely the contrary will be the fact, if
the increased food supply accrues through higher per-
capita powers of production.
And precisely as the food requirement is the most inelastic of needs in face of shortage, so it is least expan-
sive in the event of expanding supplies. It follows that
what appears upon its face to be a prospect of constant
value return may turn out to be one either of falling or of
rising distributive advantage for land. And as we have
already seen, this is further complicated by considerations
relating to the direction in which technological develop-
ment may take place.
Our third law of return, the static law of the proper proportions for the productive factors in the entrepreneur
complex, is obviously the only one of the three laws having significance for the cost problem of any particular entre-
preneur at any particular time. Law (2) is a forecast of probable changes in the conditions under which the cost-
value problem must come to be worked out. That the hire
of any particular factor will be—or may be—greater or less than it now is, does not make greater or less the pres-
ent hire, and does not modify the present relation of the
present hire to present cost and present value. It is unneces-
A page from a book with text discussing economic principles.
510
VALUE AND DISTRIBUTION
sary to look ahead fifty years in order, by finding what the costs will be then, to know what the costs are now.
It must now be noted that, simply because this law of bad proportions between productive factors is an entrepreneur law, it must, in any accurate formulation, take into account the entrepreneur himself as a productive fact. As entre- preneur, he is a fixed quantity, and at some point or other the law of disadvantageous combination, of diminishing value return, must apply to any further increases in the magnitude of operations. It is only in this sense that the application of the law of increasing returns with the greater size of the production complex.
It remains once more to call attention to the danger-ously technological tenor of all this diminishing-return dis- cussion. For purposes of retrospect or of prophecy, and in the field of value distribution no less than in the field of social welfare, it is necessary to keep in mind the distinction to the broadly admitted threefold classification of productive factors. But it is equally clear that the classification is inaccurate and unwearable for close theoretical purposes; that the technological relations between the different pro- duction goods are in a state of constant flux; that the sub- stitutionary relation is everywhere found in a measure, and is almost as likely to obtain between classes of factors as within any one class, while the complementary relation is not applicable to any particular case; that the determination in any specific case of what share is land instrument and what share non-land is impracticable or impossible; and finally and chiefly, that in competitive production no one is concerned in any way either with the validity of the classification or with the accuracy of its application.
For, from the point of view of the individual, all ques- tions of cost are questions of value expense, and capital is merely the fund out of which the costs are advanced and
THE LAWS OF RETURN
511
into which the instrumental charges and all other production expenses are reduced. Thus while, if one likes, he may distribute this capital expense among its different technological and other items, including much that cannot be classified as either land, labor, or capital, it will remain true that the outlay for labor hire is a capital outlay, the outlay for capital goods likewise a capital outlay, the outlay for land also equally a capital outlay, and therefore that the law of diminishing return is, for competitive purposes, a law of diminishing return upon capital and not upon land or instruments. It is only when the falling return upon capital is or is not due to unfortunate combinations of technological items indefinite in variety and susceptible of indefinite classification and subclassification, tripartite or other, would leave it still true that, for all competitive purposes, the only falling return having significance or interest must be the falling return having value cost, expressed under the private-capital denominator. But all this will become clearer, as approached from a slightly different point of view, in the next chapter.
CHAPTER XXIV
THE DYNAMICS OF VALUE AND OF DISTRIBUTION
As has already been sufficiently shown, the long-time computation of costs, the dynamic aspect of economic theory, points not to a new value theory or even to a supplementary valuation theory since its tendencies and influences point for changes in the fundamental conditions under which the value problem must later be worked out, the costs fixed, and the production distribution reached.
These changes in fundamental conditions are evidently restricted to two sorts, (1) changes in the demands for consumption goods, and (2) changes in the supplies or in the ownership of the productive agents, instruments, rights, and opportunities.
That either of these two lines of influence may affect the remuneration of productive factors, and thereby their values, is readily seen; that changes in the demand for products should have the same result carries its own warrant in the very statement; so, also, that changes in the supply of any factor of production must affect the remuneration of commodities itself as obvious—looks, indeed, much simpler and clearer than it really is. But that changes in the supply of some factors have of necessity direct and important effects on the remunerations of other factors is a long distance away from self-approving.
It is nevertheless true:
A demand for means of production arises only when, on the one hand, we are obliged to employ them or else go without what they produce; and when, on the other hand, we can employ them, insomuch as they are distinguished from necessary goods. . . . It follows . . . that the effective demand for means of production must vary, not only when there is a variation in personal wants, but also when there is a variation in the quantity of complementary goods.
*Wiener*, *Natural Value*, p. 104.
518
DYNAMICS OF VALUE AND DISTRIBUTION 513
The demand for plows is truly derivative from the demand for foods, but not directly; in any event, it is equally to be said that the supply of land furnishes the demand for plows; lumber affords a demand for nails, horses for wagons, wagons for horses, plows for land, men for plows, plows for men, horses for stables, stables and horses for harnesses, harnesses for wagons, horses and wagons for harnesses and for drivers, etc.
It is true that not rarely the possibility of substitution exists between different orders or varieties of productive agents,—that one sort takes the place of another, supplies the demand for it, rather than furnishes a demand for it. But more commonly it is through the principle of complementarity rather than of substitution that most of the dynamic forces in economics exert their influence upon supplies of products upon entrepreneur bidding, upon costs and diminishing returns show up as increasing values of goods, as it is likewise under this same principle of complementarity that various perplexing theoretical problems both in static-value imputation and in dynamic-value imputation are presented. New combinations of productive factors entail what new value results and what new distributive outcomes?
The usual treatment of the problem, as one centering about questions of multiplying human beings as over against a geographically limited land supply, emphasizes what is, for economic theory, the first phase of the increasing phase of this investigation; for it is again to be emphasized that only in the expectation that some factor or factors of production will come to be relatively scarce is there anything to be discussed. The laws of increasing return are laws of change in one aspect of human productive ability and adaptability; the laws of diminishing return are laws reflecting the influence of change in the relative supplies of the different factors of production. It is, perhaps, true, practically speaking, that, but for the limitation upon the
514
VALUE AND DISTRIBUTION
land supply, there could be nowhere any question of diminishing returns for social or dynamic problems.
And it is again to be emphasized that, with changing supplies of productive factors, such changes of combination, of product, of value of product, and of values of agents as take place, must take place under the guidance of entrepreneurship. This is our problem, our enterprise management. Technological productive contributions are such only relatively to the entrepreneur need, the entrepreneur combination, and the entrepreneur bid. As distinguished, then, from the ordinary threefold classification of productive factors, land, labor, and capital, there must, at the least, be established a fourfold classification including the activity of entrepreneurship. And our most serious problem is therefore to decide whether, even as modified, the traditional distribution of productive categories is either workable in theory or serviceable in applications.
Precisely how the dynamic forces shall be classified is, from one point of view, perhaps not an important matter. Clark's fivefold division into changes, (1) in population, (2) in capital, (3) in industrial organization, (4) in human organization, (5) in human wants, is possibly more serviceable as any other. Making some effort, however, toward arriving at a classification more nearly approaching the ultimate, we shall, perhaps, settle upon something like the following: (1) modifications in humanity, (2) in environment. Under modifications in humanity are to be cata\-logued the following lines of change:
(a) in numbers
(b) in wants changes in aggregate wants,
changes in relative intensity,
changes in kind and direction.
(c) in capacity changes in indus\-triousness or strength,
changes in technique,
changes in organization.
A table showing changes in different aspects of society.
DYNAMICS OF VALUE AND DISTRIBUTION 515
Under modifications in environment:
(a) in land changes in the sources of food supply,
changes in the sources of raw materials of industry.
(b) non-land changes, capital goods.
(c) changes in loan fund and in property institutions
from the point of view of each enterprise—neur an objective, environmental fact;
more properly, perhaps, as modifications among men; perhaps properly to fall under "modifications in humanity."
But here again the question presents itself as to what purpose, other than schematic, this classification may be made to serve; but if for nothing further, it will at any rate afford a basis for a more general discussion on expansion.
Doubtless it is possible to make some broad generalizations with regard to the effects of increasing population upon land values and upon land rents in the aggregate, irrespective of whether all lands must equally share in these effects. Possibly also, though less securely, something might, in wide generalization, be said of the effects of increasing machinery upon rents or upon wages, all this, likewise, without attempt to differentiate the sorts of machinery which are used for distribution labor on different sorts and grades in its technological relation to machinery.
But the difficulty with all this is that all of it has its basis in the technological relation of different instruments and agents to one another, and that these technological relations will not classify in even a loose and general coincidence with the traditional threefold classification of productive factors.
Increase of literary ability is technologically irrelevant to the increase of inventive power in industry; increase of artisan skill may intensify the demand for some machinery, while the that it displaces other. Again, the multiplication of unskilled labor may in some directions displace skilled labor or the labor-saving appliances produced by it. It is, for example, practically impossible, because unprofitable,
516
516
VALUE AND DISTRIBUTION
to introduce labor-saving machinery into Mexico or India; crude labor in those countries is too cheap compared with artisan labor. At the same time, enlarged supplies of some grades of labor furnish a demand for other grades of labor, e.g., carpenters for architects and masons, masons for hood-carriers, spinners for weavers. In this respect, machinery creates demand not for land as complementary good, and not, in any appreciable degree, for labor, but for other machinery. In the history of English industry, the spinning jenny placed an immense premium upon the invention of the power loom, and both called shirly not for more men but for the introduction of power machinery.
And now increase in population, to the extent that it should be attended by the traditionally menaced poverty, would not greatly intensify the demand for champagne lands, or place high values on diamond fields; something would result of advantage for apple and prune orchards, a lower degree of advantage to orange and walnut groves, great stress upon fuel and iron mines, and famine rents for the sources of the coarser foods. New coal lands would injure the wood lots, and benefit the iron mines. New fisheries would produce lower the income from pasture lands, but perhaps intensify the demand for fish lots.
The truth is that all these relations, on the one hand, of complementarity, on the other, of substitution, depend upon the particular situation in point of technique, and have not even the remotest relation to the land-capital-labor classification. It is possible enough, it is indeed characteristic of modern life, that modifications in technique, that is to say, in the factors which determine the pressure upon land. This is especially noticeable in the effect of farm machinery toward the lowering of agricultural rents taken in the aggregate. Improvements in transportation work in the same direction, and at the same time
*Fully worked out, the doctrine of this paragraph has some important applications to immigration problems.*
DYNAMICS OF VALUE AND DISTRIBUTION 517
do another thing, they create accessibility; thus, practically speaking, they create land.
So, new knowledge is constantly sending much old machinery—and some new machinery—to the scrap pile, and this, not rarely, with the requirement of no new machinery.
So again, changes in the standards of living or changes merely in the direction of consumption may cancel the alleged effect of the multiplication of human beings upon rent and interest for food, vegetarianism would post- pose the land famine for a succession of centuries.
And changes in technique may conceivably be so far-reaching as to leave to humanity practically no land prob- lem at all, or, at least, only a mining-land and dwelling-land problem:
Chemistry may some time solve the problem of food production without recourse to agricultural methods. The secret once known, the nitrogen in the air of the back yard and the ton of coal in the bin may furnish food for an ordinary family for a year.
But none of this is to deny that there is a dynamics to the value problem, or rather to the distribution-and-cost aspect of the value problem; but only to deny that much more can be done than has been done. We know there, as well as may be, what are the probable changes in condi- tions, and then to deduce the probable and possible eco- nomic significance of some few of these changes.
Increasing population appears reasonably probable, though there is question enough as to the races and the
*Dayenport, op. cit., sec. 352.
It is a noteworthy as well as a perplexing fact that one of the economists who has written most extensively on the subject of cate- gories, and strongest in the conviction that interest and wages are cost facts while rent is not, should have given allegiance to doctrines which, sufficiently stated, are quite incompatible with his own conclusion. It must cancel the possibility of all discussion of the laws of returns as based upon any such assumption. The doctrine that rent is purely fixed, would strike the pen across most of the pages of his own admirable and suggestive work upon the distribution of wealth. (See Carver, Distribution of Wealth, p. 83; or quotation, note ante, p. 131.)
A page from a book with text discussing economics and population dynamics.
518
VALUE AND DISTRIBUTION
classes likely to furnish the increase. No attendant changes appear probable of a sort to prevent an increasing stress upon land, varying in degree, it is likely, according to the different qualities and capacities of the land, but probably affecting in some measure all or almost all lands. But the swamps of the Amazon or the African jungle may for a long time remain an undisturbed surplus. And in this connection also it would be possible for the growth of land rent to impose important changes in the distribution of purchasing power among the various classes, with more or less reliable modifications in the direction and volume of different consumption demands, with wide and conflicting circles of varying interaction.
From the land source, together with other influences bearing upon the consumable product at the disposal of the human race, more or less pronounced effects may be deduced as probable upon the standards of living of the different races and nations of men, these effects being different with the different races and nations, and depending upon that of each different race. And various interactions between standards of living and rates of multiplication may be more or less dubiously asserted. And possible bearings of standards of living upon wages may be worked out through the general relation of the density of population to the wage level.
But what is, in truth, other things remaining unchanged, the bearing of increasing population upon the wage level?
To make the question accurately intelligible it must be assumed that the different grades and kinds of labor increase proportionally. And even then will it do to assert that wages must fall? How comes it to be true, if it is true, that the volume of population influences the wage level? Is it, for example, possible to say, with Carver, that "the wages of labor are determined by an equilibrium of two forces,—the productivity of labor, on the one hand,
A page from a book with text discussing value and distribution.
DYNAMICS OF VALUE AND DISTRIBUTION 519
creating the demand for it, and the standard of living, on the other hand, limiting the supply of it."*
Not at all denying the bearing of these two forces as somehow influencing wages, each in its own way and time, it is yet to be objected that the ways and the times are separate, that the offered explanation of wages is really a mixture of long-time and short-time influences,—on the one side a static category, a situation, on the other side a dynamic variant making for possible changes,—and then a balance somehow struck between them. The analysis neither asserts nor denies that any of entrepreneur wages cannot confine the costs as they are with supposed causes of the costs, and with possible or probable variations of the costs. But, even so, the argument is open to further serious criticism; for in reality the standard of living is itself a derivative from the productivity of the labor; the standard of living, as supply term, set over against productivity as the demand term, will, then, hardly serve as a full explanation of wages. But however this may be, it is in any case impossible to equate the productivity of today with any supposed productivity of yesterday, or to equate the productivity of today cannot, for any purpose of present costs or present wages, or under any entrepreneur computation, be equated against the labor supply of some years hence. The wages of all the yesterday's and of today may possibly have something to do with the supply of labor twenty years hence; and the supply of labor of that time will doubtless equate against the demand of that time. The supply of today has precisely that same relation to the demand of today. Today's supply does not equate with the demand or supply of today with the demand or supply of any other time. Any alleged effect from wages, through standards of living, on the supply of labor,—whether, on the one hand, the position urged be that high wages and high standards of living stimulate the birth-rate and the percentage of matrimony, or whether, on the other hand,
*Carver,"The Marginal Theory of Distribution," Journal of Political Economy, March, 1903. p. 263.
519
520
VALUE AND DISTRIBUTION
effect be asserted to be precisely the reverse—may be equally well admitted or denied with equal irrelevance to all problems of the current adjustment of wages; productivity is as it is. Investigation of these lines of influence is, then, merely a more or less successful attempt at a historical explanation of the present labour supply, and, so far as the labour supply is concerned, this investigation is an attempt to explain some of the cause of the present conditions controlling or influencing the ruling level of wages. But the ruling level of wages will be the same whether or not the historical explanations offered be well supported. So the wages to rule twenty years hence may today be possible of vague conjecture; and in the making-up of the prophecy some bearing may be ascribed to the expected population totals of that time; and there is certainly, with regard to real justification, no attribution to the standards of living prevailing today. But all this is prophecy, and has nothing to say for the wages of today.
Nor—and this is the important fact for the present discussion—even after the twenty-years' term has expired, will such population changes as may have taken place have overmuch to say; it is vastly dangerous doctrine to assert, even on the supply side, the dependence of wages on the supply of labour. For example, suppose that increasing rights rightly runs that an increased supply diminishes price; but for production goods the doctrine, so far as it is applicable at all, applies in quite other significance and to quite different results. Whenever the very increase in supply itself implies and necessitates a change in the volume of demand, the demand-and-supply formula, entirely accurate for consumption goods, becomes, for production goods, entirely misleading unless used in very different senses.
If this is true, what can we hope? How can we know that the wages must fall? Is it certain that either the per-capita productivity by weight and tale or the per-capita value productivity must suffer? Not unless the other
DYNAMICS OF VALUE AND DISTRIBUTION 521
classes and qualities of agents have failed to make a corresponding increase. And suppose that they have not; with an increased labor supply the social dividend is increased; is it to be assumed that only the old total of wages can, under the new aggregate productivity of labor, be distributed among laborers? If labor has doubled and all kinds of it have doubled, but if, at the same time, the other productive factors have failed to increase or to increase with corresponding rapidity, it may be taken as true that not quite twice as much aggregated social product will be possible; for your social dividend will be limited by the product a larger relative share will go to the agents relatively scarce, and a somewhat smaller relative share to the laborers. And this is all there can possibly be of truth in the proposition that "the wages of labor are determined by an equilibrium of forces—the productivity of labor, on the one hand, creating the demand for it, and the standard of living of laborers, on the other hand, limiting the supply of it." (Carver.)
To put it another way: Since with the change in the supply of labor the value product to pay with is all the while changing, that is, the productivity demand is changing, the effect upon the wage level must sum up as the solution of two inquiries: (1) in what measure, relatively to the increase of labor, is there a resulting increase in the total product to be distributed? (2) in what measure does labor, in the distributive process, fail of receiving the whole of the increment of value returned from the labor increase? It is evident that an appeal to this law of supply formula does not promise great results for the purposes of this problem.
Inasmuch as changes in population can take place only with attendant changes in the demand for goods and in the production of goods and in the distribution of purchasing power, there is room here for all sorts of varying influence upon values and upon distributive shares. No especially
532
VALUE AND DISTRIBUTION
serious difficulties, however, present themselves in this regard.
Possible changes in the productivity of human effort, through increases in vigor, in earnestness, in trustworthy-ness, and in skill, require some attention. And, for the moment, let changes in technique be excluded from consideration, so that the case may stand simply as one of emphasized power in lines and methods and directions already established. Twice as efficient a man is, for production to double; twice as productive property to increase this amounts, then, to a population increase. Are land rents to be thereby increased, or is the value imputation in general to be otherwise affected disadvantageously to labor? Seemingly so; for while, with larger weight-and-take-productivity and a greatly augmented social product, the absolute share of labor may increase, its ratio share will suffer. But all this is subject to amendment accordingly as the increased demand for personal service upon higher productivity and higher wages, turns out to be directed. If, for example, all this new demand power were directed to the purchase of new sorts of personal service, or to lines of goods in which labor alone could be applied, the new labor power could furnish no new demand for land or for any other instrumental good, and no advantages in distribution could accrue to any of these non-human factors of production as against labor.
Better organization of production may be merely institutional, or it may point directly to an increase in entrepreneur efficiency; if, however, it be assumed that the increase of product is due solely to an improvement in entrepreneur ability, it may, prima facie, be expected that, out of the resulting higher social dividend, the larger share will be distributed to the non-increasing factors, employee labor being here included. But here also the breadth and the indiscriminating inclusiveness of the accepted classi-
A page from a book with text on it.
DYNAMICS OF VALUE AND DISTRIBUTION 523
fications destroy all promise of accuracy in the conclusions. Better organization does not apply equally to all grades and kinds of labor, or equally in all businesses and industries, or equally to all kinds and varieties of instruments. The threefold classification must be especially misleading here.
But even more misleading is the traditional classification as related to changes in technique. Hygiene may render pill-rolling machinery useless; inventions may largely displace both labor and instrumental goods, and may shift the supply of land from one region to another. But particular soil qualities of land. There is, in truth, no limit to the possible and the probable permutations here; here, indeed, it is always the unexpected that is the probable. How complicated these problems are, and how dependent for their solution upon assumptions tacitly made or unconsciously implied, may be seen in an analysis of the relations of improvements in transportation and in crop-raising technique to the rental values of land.
*Traditional discussion of the rent problem has assumed a practically inelastic consumption for the products of agriculture. Making no qualification about this assumption, it is evident that crops manifest a great elasticity of consumption. It is nevertheless to be asserted that this inelasticity has been, in almost all recent discussion, past or current, a very serious handicap to the development of theoretical deductions. It is at any rate clear that "the law of diminishing returns" fails to explain the theoretical equipment for the analysis of such movements. This law is only an extremely important fact in the supply aspect of the problem. But diminution returns do not appear to be the ultimate cause of rents; on their ultimate cause... No economic explanation in terms of supply alone is adequate to explain rents.... How rapidly, for example, prices may advance with an enlargement of demand, must depend upon the measure in which demand will be retired by rising prices. The same thing applies to the case of increasing up of new lands. Is incapable of estimate till something is known of the degree in which falling prices may be expected to attract a larger consumption than would have been attracted by rising prices. From the point of view of the facts peculiar to supply, but as well from the point of view of demand, rents are not explained by supply alone."
*Commodities vary greatly in the elasticity of consumption. With falling prices, the demand for books, for example, greatly expands, while higher prices lead to a contraction in demand for commodities. Where consumption is very elastic, small changes in price work large changes in consumption. It follows then, that small changes in supply
523
524 VALUE AND DISTRIBUTION
The relations of expanding capital supplies to land rentals and to wages can be, so far as they are profitable of must work small changes in price, while large changes in supply work only limited changes in price.
But the relation of supply is indefinite, the reverse of all this is true. Small increments in supply are marketable only at considerable decrease in price, while large increases in supply work enormous price reductions. The increase in supply is not always accompanied by the fact that the consumption by society of the products of the earth, and especially of food products, is increased in quantity. Consumption of food products cannot be very largely increased, nor is it possible without acute suffering greatly to reduce consumption. It is even more difficult to increase the consumption of goods useful for another by reason of minor changes in price, but the total volume of consumption may be increased by a great deal through relatively great price fluctuations. Were the fact otherwise, advances in rents following upon increased population would have been much less constant than they have been. The relative value of new supplies of land would be relatively unimportant.
Deafeningly rapid changes in prices of agricultural products are unremarkable unless at rapidly falling prices, it becomes evident that all causes tending to increase the per-acre productiveness of land will meet with a corresponding increase in the cost of producing it and the decrease of rental totals. Thus all progress in agricultural skill, like better methods of irrigation, and all improvements in the applica- tions of chemistry to the production of fertilizers, by which the per- acre output of land is increased, will tend toward the demise of the poorer qualities of land. The same will happen when modern facilities, by which new and more fertile lands are brought into use and the dissemination of power there within possible, less reduce the rent differential.
"And even with the new lands only equally fertile with the old, the introduction of these new lands will not increase the differential of advantage enjoyed by the lands alone," 14 30 7 1 but the difference in advantages would be lowered.
In the diagram the transportation charges of a. 4. 10 6 30 3 0 8 8 30 4 8 6 10 30 5 3 4
are shown as a line parallel to the axis representing the cost of transportation by one-half lowers the cost of transportation would cause some of the cheaper transportation would cause some of the more distant lands to be brought into cultivation.
The result would be that rents on these lands could be marketed without so great a fall in price as seriously affected them.
14 30 7 1 "Assuming, however, that the land opened up is of distinctly inferior quality, one might look forward to a greater differential between this wider differential of fertility can be sufficient to offset the diminished differential of transportation."
10 6 30 3 0
"Suppose that the land is 20, 26, 24, etc., in productiveness, and that the transportation charges are a. 4. 6. 8. 10."
|
|
|
|
14 |
30 |
7 |
1 |
10 |
6 |
30 |
3 |
8 |
8 |
30 |
4 |
6 |
10 |
30 |
5 |
4 |
12 |
30 |
7 |
2 |
14 |
30 |
7 |
16 |
30 |
8 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION | VALUE AND DISTRIBUTION523
discussion here, mostly deduced from the principles already established. But much depends upon the sense in which etc., as in the previous illustration. Each grade of land, from the margin, increases in rent by a for differential of fertility, and by a more for differential of freight.
"71. The law of differential falls to 1 for each grade, the rent payments will fall from $m$ = 10 + 10 + 8 + 4 + 15 + 15 + 15 + 6 + 3. Even when the price of land is lower without a material fall in prices, this would carry the rent payments only to $n$ = 10 + 10 + 8 + 4 + 15 + 15 + 15 + 6 + 3. However, prices would greatly fall upon the land that has become less fertile, and lands that have become lower without a material fall in prices, this would carry the rent payments only to $n$ = 10 + 10 + 8 + 4 + 15 + 15 + 15 + 6 + 3. However, prices would greatly fall upon the land that has become less fertile, and lands that have become lower without a material fall in prices, this would carry the rent payments only to $n$ = 10 + 10 + 8 + 4 + 15 + 15 + 15 + 6 + 3. Even when the price of land is lower without a material fall in prices, this would carry the rent payments only to $n$ = 10 + 10 + 8 + 4 + 15 + 15 + 15 + 6 + 3. However, prices would greatly fall upon the land that has become less fertile, and lands that have become lower without a material fall in prices, this would carry the rent payments only to $n$ = 10 + 10 + 8 + 4 + 15 + 15 + 15 + 6 + 3. Even when the price of land is lower without a material fall in prices, this would carry the rent payments only to $n$ = 10 + 10 + 8 + 4 + 15 + 15 + 15 + 6 + 3. However, prices would greatly fall upon the land that has become less fertile, and lands that have become lower without a material fall in prices, this would carry the rent payments only to $n$ = 10 + 10 + 8 + 4 + 15 + 15 + 15 + 6 + 3. Even when the price of land is lower without a material fall in prices, this would carry the rent payments only to $n$ = 10 + 10 + 8 + 4 + 15 + 15 + 15 + 6 + 3. However, prices would greatly fall upon the land that has become less fertile, and lands that have become lower without a material fall in prices, this would carry the rent payments only to $n$ = 10 + 10 + 8 + 4 + 15 + 15 + 15 + 6 + 3. Even when the price of land is lower without a material fall in prices, this would carry the rent payments only to $n$ = 10 + 10 + 8 + 4 + 15 + 15 + 15 + 6 + 3. However, prices would greatly fall upon the land that has become less fertile, and lands that have become lower without a material fall in prices, this would carry the rent payments only to $n$ = ...
| Land | Rent Payments |
|------|---------------|
| $x_+ x_-$ | $x_+ x_- - x_+ x_- \cdot (x_+ x_-)$ |
and some methods were devised of doubling the output of classes $x_+$ and $x_-$ it would still remain true that it involves some of the as-bushed land while the $x_+$ and $x_-$ lands would become go and $x_+$ lands with their increased output they would increase their rents by one-half.
"In no case, then, is it safe to assume that the mere fact of the extension of cultivation to inferior lands means of necessity an increase in rents. In order to secure such results as are observable tendencies in any case, there is required a Gregory King's law of obsolescence. This law requires that all products must be treated as elastic as is the demand for books, sewing-machines, or bicycles; improved arts of transportation would probably raise rents. If for example, there was an increase in wages due to improvements of transportation, and these expenses were reduced by one-half, it would be necessary to reduce rents by one-half also. If there was an increase in demand for products were such that the prices should sharply fall. In this case, rents would increase in the total.
"Unless we can determine what is the number of the demand as well as its elasticity at any given time we cannot determine how much rent will be received from any particular piece of land. The degree of rise or fall in rent can be guessed at only in view of the demand.
"Some questions we can hardly even guess at. It is com-
526 VALUE AND DISTRIBUTION
the term capital is to be understood. If taken as loan-fund or as competitive capital in the most inclusive sense, it would follow that cheaper capital, meaning merely lower rates of interest, might raise or might lower rents accord-
ingly as, under the prevailing conditions of technique, the borrowed purchasing power should be directed to comple-
mentary or to substitutionary goods.
But if the term capital is taken to mean non-land
productive factors, further assumptions become necessary.
How elastic is the consumption of agricultural products?
And how far is it taken as possible that non-land
instruments can be made to function successfully on lands which were before marginal? Costless capital goods might conceivably go so far in substituting lines as to destroy all differentials of servicability between different lands, or even as to render all lands equally valudef.
But all this detail grows tiresome, simply because there can never come any end to it; at the best, it is mostly
merely assumed that improvements in farm machinery work in line with improved fertilizers and improved methods of cultivation. This is correct for such machinery as increases the per-acre output. But for the main part, these labor-saving devices are not land-saving devices. They increase the output employed on a given amount of product; thereby they lower the margin of fertility, exactly reversely to what is desired. The reason for this lies in the fact that cultivation is carried so far upon inferior soils, as through a continuous extension of supply, to lower prices, and to do this so much as to extend the influence of those qualities in product which are overcome by the necessity of marketing the products at lower prices than their real value.
But a caution is needed here. Doubtless the lands especially affected by these improvements are those which have been rented. So likewise the opening up of America may have had more effect to advance rents here than to depress rents in Europe alone. This means that the rent-generating process has spread over the world-wide, so must also be the rent generalisations, if they are to be theoretically sound.
But neither theory nor fact does room for doubt remain as to the validity of the point at present urged. Improvements in trans-
portation, and improvements in the efficiency of human effort as applied to production, serve to reduce the pressure upon the land factor in production—function, that is to say,
as substitutionary rather than as complementary agents.
DYNAMICS OF VALUE AND DISTRIBUTION 527
a disciplinary gymnastic. But this much, at least, stands forth clearly: Every problem in the dynamics of value, in its distributive aspect, must seek its solution along two lines of inquiry: (1) how does the new development affect the social dividend; (2) does the new development bear, as complementary good or process, to make relatively greater the demand for the instrument or agent under examination, or rather is the relation one of substitutionary good or process, summing up, that is, in the emergence of a new competitor; or, practically, in an increase in the supply of the goods under examination?
The bearing of education upon the various distributive shares must be worked out as in parallel with other developmental influences making toward increase in the productive efficiency of human effort; but allowance must be made for the different effects upon different grades of human activity,—the probable effect, for example, to lessen the differentials in wages between skilled and unskilled laborers; or, vice versa. And other effects toward increase in the volume and in the variety of consumption requirements would come in here for discussion.
A seemingly more difficult question is that of the probable future tendency in the rate of time discount. So far as the trend of things is toward an increasing social productiveness and a corresponding increase in the value of the product, and thereby a diminished pressure of subsistence needs upon the individual income, it would seem safe to infer an increasing volume of deferred rights of consumption offering themselves upon the market. On the other hand, there appears probable a progressive development in the science and technique of industry. Temporarily, then, at any rate, the greater ease of saving may be offset, in effect upon the discount rate, by the increasing weight-and-tale productive-ness of indirect methods of production.
But here again much depends upon the degree and direc-
528
528
VALUE AND DISTRIBUTION
tion in which improvements in technique express them-
selves in the form of substitutionary devices and instru-
ments rather than of complementary instruments in relation
to the general stock of instrumental goods. It is con-
ceivable that knowledge of ways of getting on without
capital should greatly limit the capital-goods field of invest-
ment for the loan fund.
So the increase of savings and the growth of capital
goods might go so far, in directions substitutionary to
particular grades or qualities of labor, as to depress the
wages paid to laborers below their natural subsis-
tence minimum. But it is difficult to conceive how laborers
in the aggregate could seriously suffer through any pos-
sible increase of capital instruments produced in the main
by labor, unless, indeed, these instruments were of a sort to
function with reference to labor as substitutionary goods,
and at the same time, with reference to land, as comple-
mentary goods. In such case, manifestly, no difficulty
could arise by way of supplying consumption goods at
human disposal, but only with the distribution under which the distribution would take place; rent would tend
to absorb the entire social product. Parallel theoretical
possibilities present themselves in connection with such
developments in methods and devices as should tend to dis-
place either capital goods or labor or both.
And there is possibility in many industries of monopoly
organizations going on far and so profitably, through the
increase of sales and the production of goods by entrepreneurs,
and through the decrease of consumption loans on the part of the laborers, and through the displacement of labor
by substitutionary instruments, as to bring about either a progressive non-employment of labor, or an employment upon increasingly harsh conditions, and to
bring about at the same time a tendency toward an increas-
ing volume of consumption loans, represented either by
an increasing volume of indebtedness from the non-capital-
izing classes to the capitalizing classes, or, more probably,
A page from a book with text discussing value and distribution.
DYNAMICS OF VALUE AND DISTRIBUTION 529
an increasing resort to public debts both by the employee class seeking employment and by the capitalizing classes seeking avenues of investment.⁴
*But all of this leads up to questions falling within the problem of what has been termed the "fallacy of saving," upon which problem the present writer has written at length. The works of Ruskin, Robertson, Hobson, and Veblen seem to have done the best work here, not perhaps towards the solution of the problem, but to the development of its nature and character. In their discussions we find talk, with its assumption that capital must be both concrete goods and relatively labor-complementary goods, a very difficult task. And the whole subject is complicated by the fact that it is related with the phenomena of industrial depressions. (See page 202, note.)
The first point is that so long as industrial technique will permit the utilisation of new supplies of labour power, and so long as the products of these new supplies are dis-
planted products or the displaced consumption represented by the intermediate appliances, so long private saving may be a clear advantage to society. But it is not necessary to assume that the saving goes to the increase of productive equipment, rather than to private consumption loans, or to fiscal extravagance, or to the organs of war.
But on what terms can any agency making for increased volumes of production make use of private saving? It is not sufficient to say that the products are to be consumed, upon the condition, that is to say, that the rate of consumption keeps pace with the increase in productive efficiency.
But even so, something may depend upon what classes of society do the consuming. The more important classes are those who are able to purchase power shall further capitalize to the indefinite increase of investment.
Note that the question is not at all whether these disposable incomes were repaid earned, or were institutionally justifiable, or, on the contrary, whether they were distributed among all classes equally or unequally distributed, but only whether, once obtained, and however obtained, they are used for productive purposes or for private consumption forthwith by their recipients, or are better directed into the creation of new technological equipment.
We must now consider how far and in what degree tech-
nological capital may, in ordinary times, affect the distributive shares of other classes.
If it were true that, for whatever addition to product were due to the new instruments, an equivalent income accrued to the owners of these instruments, then there would be no change in the distribution of interest in the increase of capital. The problem is, then, in this aspect a purely technical one. If distribution were affected solely by the imputation of distributive shares, the other claimants will receive some share of that product mechanically attributable to the enlarged supply of capital. But if this is not so—that is to say—only does the employer's capital benefit the labourers or other claimants. Any solution which directly deduces from the fact of an increasing social dividend made possible by capital, or from the fact of
---
⁴ *But all of this leads up to questions falling within the problem of what has been termed the "fallacy of saving," upon which problem the present writer has written at length.* (See page 202, note.)
530
VALUE AND DISTRIBUTION
But let it be assumed that an advance in rentals is to accrue to certain classes of instrumental goods, e.g., land; will this advance in rentals express itself in a higher-per-
better tools in the hands of labor, a larger wage for the laborer, is a solution superficial in reasoning and inaccurate in conclusion.
Thus it is clear that such part of private capital as actually goes to the productive process does not necessarily increase the other claimsants. If and when the consuming disposition of society is keeping pace with the productive capacity of the country, it is evident that the technological equipment adds to product more than the distributive share imparted to the equipment.
It may be further asserted, true that the capitalist's role is here only one of postponed consumption on terms of receiving later an increased product. But this is not necessarily so. It may be an advantage to labor, so long as there is market both for its product and for the product of the extra equipment. But when and how far is there such a market? The answer depends upon whether by the rich generally a necessary fact, if the volume of consumption is to keep pace with the productive capacity of the country is preserved, what about the private capital which is unrepresented by technological equipment and the correlative incomes unrelated to increased product?
The question whether standards of consumption do commonly keep pace with productivity, and whether technological productive equipment comes to exist, and the question whether standards of consump-tion must commonly thus keep pace, are distinct and separate questions. In times of prosperity, when there is no lagging in times of post-crisis depression, the standards of consumption do move in step with productivity. This is not a theoretical considera-
tion, but that there is no theoretical necessity for this; and it appears equally clear that in post-crisis times there is a distinct and diametric restriction on consumption. (a) That there is a period during which there is temporarily a surplus, and (b) that in some measure there takes place in industry a reduction of consumption.
And it appears to be true that the very fact that, through developing techniques and increasing equipment, a high per-capita productivity obtains, has led to a reduction in consumption. This reduces individual need, explains how it may occur and does often occur against the will of those who have been accustomed to it. Under these restric-
tions of consumption, industry is subjected to periodic reverses and to the periodic wastes, insolvencies, and starvations which bad times connote.
We seem, then, to have come safely thus far; that, from the social point of view, saving should neither go to the extent of subtracting from present consumption more in utility than is added by the later increase in production; nor should it go beyond that extent to the extent that the later consuming disposition will not absorb it; the limit of rational saving are, then, set by the prospective elasticity of consumption.
But now, precisely where, if anywhere, does this leave us with regard to the problem of luxuriant consumption for those times when
DYNAMICS OF VALUE AND DISTRIBUTION 531
centage ratio of time discount, or in a higher capitalized value with an unchanged rate of discount?
Such reply as may be given is neither precisely to the general attitude is one of overemphasis—of overemphasize, that is, upon future consumption as against present consumption?
If in prosperous times the consumption of the rich displaces, in the main, the consumption of the poor, then it follows necessarily that any expansion of consumption in times of depression cannot be at the expense of the consumption of the rich. For it is only by the luxury of the rich employs productive energies that otherwise would not function, such harm, if any, so can result to others must be found in the direction of the consumption of the poor. In other words, excessive consumption, that is to say, not in the direction of any influence to restrict production, but rather in the direction of increasing the significance of those incomes. And if, in times like these, charity would be in any aspect justifiable, these luxurious expenditures have some obvious defects.
But what is much case are the economic effects of charity? People who can find no work to do live somehow out of the actual product of industry, whether by the using up of their own saved purchases or by borrowing money on interest. When they are helped through offered charity, their consumption is increased, and yet not at the expense of the consumption of others. If it be true that more goods have been caused to be produced, it would appear to be true that the charity has meant added consumption for the resident poor. But if this is so, since it is not possible to increase income, these others should be minded to increase their present rate of consumption, this liberal stimulus would be passed forward one degree.
The case would then, stand as follows: by means of a subscrip-
tions collection which was made for the purpose of aiding against the products of others has been collected in the present and canceled, instead of being used for purposes which might have been more profitable, and the collection has taken place at a time when society has been able to achieve the cancellation through the employment of productive energies.
This argument will, if valid—which is doubtful enough—mean much for the methods and the times of the carrying-forward of public work. But even without the support of this particular argument, it should be clear that all such collections are not made and taken in times only of slack employment, and ought to be paid for in times of plenty. For thus, if such practice, carried on in prosperous times and on terms of displaced production, and paid for in times of depression.
But what does the argument imply for the social advantage of such savings? It implies itself in two things: first, that productive equipment of society, but instead, flows into consumption loans or goes to finance fiscal deficits? Here nothing but condemnation is possible. Any private investment which for any considerable period of
532
VALUE AND DISTRIBUTION
one effect nor to the other. The value of each and every instrument will be a capitalization based upon the current discount rate; but this discount rate will be the point at which all the savings in loan-fund form, the general-purchasing-power form, find a market among the various demands of borrowers for consumer purposes, for socially productive purposes, and for privately acquired purposes. The technological demand is only a part of the entire demand.
time, takes toll from social product by other title than of equivalent addition to that product is a socially disastrous thing. No matter what perspective one may take on the question of the relation as between man and man may really be, the case is, in last analysis, nothing but seridism on the one side, and parasitism on the other.
CHAPTER XXV
THE ADJUSTMENT OF PRICE
Such examination of the psychology of utility and of valuation as falls within the sphere of economics to under-
take has already been attempted, with something over.
(See note, p. 307.) For present purposes it suffices that market prices may always safely—albeit superficially—be reduced to the sum total of the effects of demand and supply. But, even so, there remain some aspects both of demand and supply and of the process of adjustment still requiring attention.
For purposes of analysis and of exposition, the device of plotting demand and supply volumes into intersecting curves expressive, at their point of intersection, of the place and the method of market-price adjustment, has sufficiently demonstrated its claim to servicability.¹ It is,
¹Oddly enough, the general adoption of the plotting methods has not awaited to remove the old-times ambiguities connected with the demand and supply curves. The demand curve, which represents actual, and potential, or excluded, supply, are clearly brought out in the plotting diagram by means of a line representing the possible adjustments of price. Nevertheless, the bad logic of the terminology which employs the concepts of demand and supply to explain price movements is not removed. The demand for price, still abides. Ruskin's inspirational methods touched the heart of the case. "The quantity demanded is not what is bought," he says. "The quantity of a thing sold." I mean by it the force of a buyer's capable intention to buy. In good English, "a person who does not wish to get, but who wishes to have." (Manners and Manners, chap. 16.)
And so Pantaleoni:
"When price falls, a determined scale of wants being given, more consumers purchase than when, on the same principle, less wants are enumerated. Hence arises all that do with the extension or restriction of consumption in accordance with a given and determinate law of demand. This is a very common mistake. It is made in regard to an extension or restriction of the demand, which gives rise to endless ambiguities. By the use of the graphic method these ambiguities are avoided . . . The conclusion, to speak accurately, or
533
534 VALUE AND DISTRIBUTION
however, at the same time true that the uses actually made of curves of utility and curves of demand have been prolific of much loose thinking--this, for the most part, because of the lack of differentiation between curves of individual utility and curves of group or social utility, and between curves of individual demand price and curves of group demand price.
As referring to the individual, precisely what may demand or utilize curves be made to express? As of any one time, for any individual, it is undoubtedly possible to construct a curve of utility for all of the various items of a stock of precisely similar goods; but it is seemingly impossible, and it is certainly profitable, to attempt to include in the formulation more than one kind and one grade of goods. And with reference to any one kind and grade of goods, at any given point of time, a demand curve in terms of price is also easily possible of construction for the individual; and for any given point of time, a total-expenditure curve, expressive of the distribution of pur-
figuratively the demand, is extended or restricted; but it neither rises nor falls in accordance with the law of supply (op. cit., p. 440).
Plus also is accurate in essentials:
The state of demand may be really altered while the amount demanded remains constant in quantity. Price change, in fact, leads to extension or restriction of the amount demanded, or to its complete cessation. But this does not really describe the true nature of what is occurring (op. cit., p. 39).
The following, however, are examples of the more common but less definable usage:
Suppose that the general law of demand is that it varies directly with changes in the intensity of want, and inversely with changes in the prices that must be paid for goods. When demand increases proportionally to increase in price change, it is said to be "elastic" (op. cit., p. 66).
Having regard to this market, the supply of an article, in its technical sense, is the amount offered at a given price. It tends to increase as the price diminishes. . . . The demand for an article is the amount which would be purchased at a given price if only to increase as the price diminishes. . . . The market price for an article is "the price at which the demand is equal to the supply" (op. cit., p. 74).
Fetter: "In the case of any good . . . a change in its ratio to other goods will increase the demand" (op. cit., p. 80).
THE ADJUSTMENT OF PRICE 535
chasing power, in terms of some standard, among all the different grades and kinds of commodities is, theoretically, within our accomplishment.
But for any group of individuals, a utility curve is, as we have seen, a hopeless impossibility. A group demand-price curve for any one kind and grade of article is readily attainable; and a group demand-price curve for commodities in general is also a possibility; but this last only in the sense that, as such, no curve remains, but only an aggregate market-price adjustment expressive of the price relations of all the different exchanging goods. All this, however, is not true.
For any one kind and grade of commodity, the individual curve of falling utility per unit of commodity, as distinguished from a falling price-paying disposition, could have little or no significance as expressive of absolute utility magnitudes; not that some common denominator, in units of pleasure, or of satisfaction, or of desirability, or of choice, might not exist in the individual psychology; but, if existent, it could hardly be expressed, and, if expressible, could hardly be represented by a single curve. The curve is in the expression not of the absolute utility magnitude of the different items of the stock, but only of their relative significance. The marginal item also could become only vaguely quantitative in meaning, quantitative in the sense merely of asserting a smaller utility than that of any other item in the series.
There is, therefore, no measure function anywhere expressed by utility or marginal-utility analysis; the very fact that the series is a series, and that the law of satiation which it expresses implies that the items are of diminishing utility volume, makes each item incapable of serving accurately as the utility equivalent or measure of any other.
Nor, as we have seen, is it, with the individual demand-price schedule, possible to find a measure of utility in money. The limit of price offer expresses merely an
536 VALUE AND DISTRIBUTION
equivalence in utility between the thing in prospect and some foregone alternative. (See page 315.)
It follows that the total-expenditure schedule of the individual indicates not absolute utilities, but only what uses will be made of the individual's fund of purchasing power as against the competing claims of other desirable commodities. And no item of expenditure among all the items need be marginal in the sense of being at indifference between the actual direction and the alternative direction of purchase. The marginal unit of expenditure shall be merely a point on the curve, if at all, without any neces-
sary or probable implication as to the absolute size of it. Any attempted reduction for the individual of all the different commodities into one utility curve or schedule could, at best, be a mere repetition of the original expendi-
ture curve.
But it can hardly be too many times repeated that, so far as concerns utility schedules, we can never get beyond the individual. Society can have no utility curves or computa-
tions, unless upon some heroic assumption—nevertheless possibly inevitable in socialism—that all men are alike, or, at all events, that their differences may or must be over-
looked. As between different individuals, there can be no comparison of utilities either quantitatively or qualitatively.
There may, however, be constructed for any one kind and grade of commodity a social or market price-demand curve, a social or market price-supply curve, and a com-
modity marketable at the different prices set in the schedule. But neither for the group, nor for society as a whole, nor for any individual within the group is the price offer indicative of any absolute utility magnitude.4
As forcibly illustrating this misstep between the individual utility curve and the individual price-offer curve—between the possible indi-
vidual utility and the possible individual price-offer—the two utility curves, and between the individual demand-price curve and the group or social demand-price curve. Maxwells' "Theorie der Nutzen," p. 108; and the
marginal-utility theory. (Value and Distribution, p. 37) is espe-
cially worthy of citation. "The utility of successive increments of
commodity is represented by lines at right angles to O'M..... Ac"
Ac
THE ADJUSTMENT OF PRICE 537
The relations between the individual desire and the individual price-offer curve, and of both of these to the market-price-offer curve, require some further attention.
How much of today's fund of purchasing power, money or credit, shall a given individual turn toward the acquiring of wheat? Not merely the hunger of today, but the foreseen hunger of later days must be taken into the reckoning, as must also the expected future supplies of wheat and the expected future command of purchasing power, and over against all this must be examined the same
according to the marginal utility theory, the value of the whole commodity is determined by the utility of the last increment of supply." So "AU or MN represent severally the marginal utility or value per unit of the commodity.
A line graph showing a downward-sloping curve labeled "U" with points E, F, M, N.
Accurately, this curve might represent the utility curve of any one commodity for any one individual at any given time, or it would serve for his price-demand curve, or for the price-demand curve of society as heeded by him. But it is not possible that such a curve can serve for utility curve and price-demand curve together, whether for some commodities or for all commodities. It is impossible to depart from a utility curve where this latter is possible, in that the price-demand curve shows the effect of the desire for other things. Thus, even though we assume that all commodities have identical desires, there would be as many different demand curves as there were different commodities. The same thing would hold true with different days, even though the utility curve might conceivably do so. So, when Macfarlane is going to have AU or MN represent the demand for wheat on a particular day, he will be again confusing the possible interpretations of the diagram; he is really treating it as a social demand-price curve expressive of different volumes of purchasing disposition in view of the differing individual comparisons and decisions as to the respective applications of purchasing power.
538 VALUE AND DISTRIBUTION
total of considerations as bearing upon the competing claims of all other directions of expenditure. That is to say, a purely personal system of discounting future facts into bases of present activity must be applied over a wide commodity field, before the individual can decide, in any given case, whether he shall buy wheat or raiment, or rather hold for future occasions certain items of unspecialized purchasing power. Thus today's hunger-utility line, if it could be drawn and plotted on a graph, would be cut by two lines, one steeply descending, since it can be drawn by both lines of steep descent, since the appetite for food, and particularly for any one sort of food, is quickly satiated. Not so, however, when the long-time aspect is included in the computation. As the needs of days ahead, or even of weeks and months, make themselves felt in thought, the price-offer line descends not at all so sharply; possibilities of storage, of decay, of ravages by vermin, as well as possibilities or probabilities on the side of scarcity or abundance of the commodity in question power, all are data in the problem. But in view of each man's situation and prospects, the law of satiety holds, and a limit comes to the purchasing disposition as reaching out toward wheat. This curve does not, however, find its lowest offer item at the point of satiation but at the point where some stronger pull attaches upon the purchasing power in hand. The items of price offer in the individual's wheat-demand schedule will therefore proceed somewhat thenceward along at considerable intervals into the construction of the individual's general-expenditure schedule or curve, this last curve serving to express the same facts which, from another point of view, might stand as the individual's money-utility curve.
It is now to be noted that the individual's price-offer curve for any particular line of goods, and the individual expenditure schedule, are both worked out upon the assumption of a given price situation for each and every line and grade of commodity; very considerable modifications, therefore, in the amount of each commodity demanded, and in
THE ADJUSTMENT OF PRICE 539
the general distribution of purchasing power, might follow upon a change in the price of any single commodity. The action of each individual in the market is, as we have seen, to be regarded rather as the result of the market situation than as the cause, though each individual activity is in turn to be taken as part of the entire cause.
And out of all these individual offer dispositions, how construct the aggregate or social price-offer curve? This problem applies to only one commodity, but we refer only to the various commodities of food which will be, under given conditions, purchased at different levels of price. And here again an existing medium of exchange and an established general level of prices are assumed. Precisely how, otherwise, the aggregate demand bearing upon any one commodity could be expressed or formulated, or even, for present purposes, described, is, indeed, hard to conceive; to the present writer, at least, the problem would seem hopeless. The difficulty is that there seems to be clear enough but too precisely of a sort that will not combine with other individual solutions in a way to render possible construction an aggregate barter-demand curve.
For, as has been already pointed out, each individual desiring to barter away any part of his possessions, and failing to find an opportunity to exchange the particular item to be sold against the precise thing demanded, will barter for such third sort of commodity as seems to him most likely to serve both parties. It is easy to see why this is so; and it is present evident how these various lines of exchange and this multiplication of media could be made theoretically manageable in a market-value analysis. Seemingly each instance of barter would be a matter of separate bargain adjustment, modified only by the report of how similar trades were elsewhere being made, and also by each individual trader's opportunities and devisings as to some other possible roundabout method of achieving his ends.
The social demand-price curve presents nothing like similarity in direction to any of the individual demand-
540
VALUE AND DISTRIBUTION
price curves combining to make it. The individual price-offer curves of $A$, $B$, $C$, $D$, and $E$, say for bread, being assumed as respectively depicting price demands of 9, 8, 7, 6, 5; 8, 7, 6, 5; 4; 7; 6, 5; 4; 3; 2; 5; 4; 3; 2; 1, and as represented in plotting as lines of a 45-degree declension, would, as an aggregate demand volume of 9, 8; 7; 7; 7; 6, 6; 6; 5; 5; 5; 5; 4; 4; 4; 3; 3; 2; 2; 1, plot into a group demand curve represented as follows:
A graph showing a series of downward-sloping lines representing different demand levels for bread.
The value equation requires a supply as well as a demand term; but it does not require the assumption of a produced or of an elastic supply. To make, therefore, still clearer the concept of demand, and to prepare for the introduction of supply considerations, let there be assumed an existing supply, truly, but a supply fixed and limited in volume, in the sense that all sources of new supply are taken as, for the time being, closed.
Nor is this the only possible assumption: the economics of child-trading approximates this case; and the situation among the reservation Indians after a general issue of supplies is a still closer approximation. How would demand present itself under this gift-supply assumption? And how would values adjust themselves?
Trading actually goes on briskly in these cases, and doubtless would do so, if confined entirely to barrier processes ; but the barrier problem has already been sufficiently considered, and, even if capable of satisfactory analysis, would not afford an analysis serviceable for our existing
THE ADJUSTMENT OF PRICE 541
money economy. There is, therefore, also to be assumed an existing medium of exchange, a price standard.
Nor will it greatly advance our problem, the determination of the demand and of the price of any one article, to appeal to the proposition, obviously true for certain purposes, that the existing volume of commodities at any time is in one aspect demand, and in the other aspect supply, and that therefore any increase in the total supply is at the same time an increase in demand. The question remains, in hand in hand to determine what amounts of money or of equivalent purchasing power are, under all the conditions and influences bearing upon the situation, at present held to be expended for any given commodity at its various levels of price—what purchasing-power demands are now extended toward the commodity in question. All the different holders of different goods, the exchange prices of which are to be offered against the commodity under consideration, must be assumed to have made their respective commodity holdings into the homogeneous purchasing-power medium, before any one of these possessors of commodities or any one of these commodities can figure as data in the analysis of the fixation of price. Our question is, what money demands center upon any selected commodity; we have no concern with demand and supply as aggregates in relation to the entire market for the total of commodities.
Each and every individual in the aggregate group of five persons who may be supposed to purchase some share of the assumed fixed supply, if only the price turns out to be low enough to attract him; each, that is, represents potential demand. Individual A will take one item if the price fixes itself at 9; five items if the price is 5; and evidently this disposition not to give more than 5 for the fifth item, or more than 6 for the fourth item, etc., expresses a situation which must obtain some further expression, were we to attempt the construction of demand schedules for other consumers. We have, then, a schedule of different purchasing dispositions at different price levels,—our
542
VALUE AND DISTRIBUTION
earlier schedule under a different statement; one item pur-
chasable at price 9, three at 8, six at 7, ten at 6, fifteen at
5, nineteen at 4, twenty-two at 3, twenty-four at 2, twenty-
five at 1.
Does this complete our demand schedule? Suppose a
certain number of articles to be for sale; at what level will
the price be adjusted? Here we must obviously take
account of two different possible assumptions, (1) that the
holders of the supply will offer the price that they can get, (2) that these are refusal prices.
Upon the first assumption, that of a demand schedule or
curve expressing maximum price offers of $9, $8, $7, $7,
$6, $6, $5, $5, $5, $5, etc., ten items of unreserved supply can
be absorbed by the market only upon terms of a price as
low as $5.
No such result will, however, obtain under the second
assumption; here, obviously, the outcome must be a dif-
ferent one accordingly as different assumptions are made
with regard to the reservation prices set by the various
sellers. The truth is that a case falling within this second
class is a case where the supply schedule really contains
demand items; the seller plays two roles. If two men, one
with a price-offer limit of 30, and other of 10, want to buy
a certain horse, for which the owner will refuse anything
under 30, there are really three demand prices bearing
upon this horse. One is on the part of the seller,
of a willingness to sell at any price up to 30; another
we should reckon only two demands, but is rather like that
of an auction with an authorized bidder-in.
If now, together with our original price-offer schedule of
$9, $8, $7, $7, $7, $6, $6, $6, $5, $5, $5, $4, $4, $3,
$3, $2, $2, it be assumed that the ten supply items are
offered only as subject to a reservation schedule expressing
minimum prices of 10,$9,$8,$7,$5,$4,$3,$2,$1 and if also these
reservation prices are supposed to appear as demand prices
under the demand schedule, our problem will then present
itself as one with an unreserved supply of ten items as over-
THE ADJUSTMENT OF PRICE 543
against a demand schedule or curve of 10, 9, 8, 8, 7, 7, 7, 6, 6, 6, 6, 5, 5, 5, 5, 4, 4, 4, 4, 3, 3, 3, 2, 2, 2, 1. The price will then adjust itself at the point where the market demand will absorb ten items of supply, that is to say, at some price greater than 6 and not greater than 7.
The manner of market analysis especially characteristic of the Austrian school has, under the name of the marginal method, now established itself among practically all economists, although there remain different views enough as to the possible purposes which this analysis may serve.
"In isolated exchange, exchange between one buyer and one seller, the price is determined somewhere between the subjective valuation of the commodity by the buyer as upper limit and the subjective valuation of the seller as lower limit."
*Positive Theory of Capital*, p. 190.
No one can be disposed to dissent further than to remark that Boheim-Bawerk has here shifted his subjective-worth concept over to a somewhat different field. He has shown how it is possible to express a true subjective valuation expressed in terms of prices. B, the seller of the horse, and A, the buyer, work out the price-result by bidding. But it is not so much the result of their bidding that interests us more than it seems upon the face of it. What are these different subjective valuations? They are first of all those which B attaches to the horse; but also with the utility of what he must let go in order to get the horse. B likewise has really two things in mind—the horse to be sold and the money which he must give up in order to obtain it. In the problem a medium of exchange and tacitly and indirectly present a whole series of other conditions which have to be fulfilled before horses can be exchanged. For ideally simple conditions, the case should be assumed as one in which B has only horses for sale, or only sheep, and neither of them is worth anything except when they are exchanged together with property to be acquired. If, then, B's supply of horses is such that he cannot sell any more than one horse without selling stock with sheep, the terms of exchange may be anything—thirty horses for one sheep or thirty sheep for one horse; there are no limits but those imposed by nature on the number of animals which each finds in both horses and sheep a utility for himself; there comes about a valuation of horses and sheep in terms of each other. This is not so with horses or by A of his sheep in terms of some unrelated marginal usefulness; but a comparison by B of the marginal uselessness of his horses in terms of his sheep shows that he would rather have his sheep than his horses in terms of the horses offered. That is to say, upon a basis of one marginal utility for each exchange there can be no limit prices between which the exchange price must finally be found. Each trader
544 VALUE AND DISTRIBUTION
This explanation of value is based upon the assumption that, as the items of offer and demand become more numer-
ous, the margin interval within which the bidding process may be regarded as a whole, becomes so infinitesimally minute gradation of both offer and demand is assumed to be near an approach to infinitesimals—as to justify the treatment of the selling price as accurately a marginal price for both demand and supply. Added to this, the ordinary attendant conditions, namely, that all the commodities are of equal desirability, all the competitors in the market simultaneously, and "that the buyers and sellers make no mistakes" (which is not true), and that no taxes would prevent them from really pursuing their egoistic interest",—assuming, in short, a perfectly frictionless market, this may be admitted as an accurate account, descriptively, of the market adjustment. It must also be admitted that the point of adjustment expresses marginal utilities, or measures them, or is measured by them. As we have seen, two marginal utilities must be compared by each marginal trader—before he can sell one unit of commodity before a trader can become a marginal trader. It is still another matter to assert that these marginal traders are, as against the opposing in-pressing volunities of commodity and of money power, able to effectuate any ultimate price adjustment. It is yet even more questionable to assert that, while the market price coincides with the price limits of both marginal traders, the price is invariably determined by the action of only one trader. All these questions really resolve themselves into the one great question. What are the causative forces in the market adjustment?
must be concerned with two marginal utilities, and must have based his subjective valuation upon the outcome of this comparison. It is only upon this assumption that we can say that A will give up an apple for one horse, or B determine, at his limit, to accept five sheep for one horse. The prices are therefore fixed at ten to one as upper limit and five to one as lower limit.
The importance of the further assumption of an exchange medium and a monetary standard cannot be overestimated. To say that B will accept fifty dollars for his horse, and that A will, as limit, give him twenty-five dollars for a horse is gaining nothing; because of the horse things sixty dollars will purchase and that, even as against the utility of the things that sixty dollars will purchase, A prefers the utility of the horse.
*Positive Theory*, p. 204.
THE ADJUSTMENT OF PRICE 545
The illustration—quoted from *Positive Theory*—of isolated exchange has already received sufficient examination. No talk of determination of price by margins or at margins, but only between margins, can be made for this case.
Consider now the illustration of competition confined to the selling side. If $A$ is the only buyer, with 30 as his price limit, and if together with $B$, with a minimum price of 15, he buys 10 horses, then $A$ will buy $B_1$ at 15, $B_2$ at 20, and $B_3$ at 25, the price must be made at somewhere between 10 and 12 as the limit.*
This seems to give great support to the theory that the price adjustment either expresses a demand price or is limited in either direction by demand margins. Both the upper and lower limits are fixed by buyers' prices.
It may be understood why when a certain minimum of price is set by supply, the demand items outnumber the supply items, can a demand schedule furnish both price limits; but cases of this sort are presented only in the extreme cases of monopoly.
*Assume now that, in addition to $A$, and $A_2$, three other buyers, $A_3$, $A_4$, and $A_5$, compete for the horse, and their respective circumstances are such that they count the possession of the horse as worth 28, 24, and 20 respectively. Then $A_3$ will bid to the limit of 22, $A_4$ to 25, and $A_5$ to 28.* Thus at 26 $A_1$ and $A_3$ would close, so $A_1$ must pay a price somewhere between 26 and 30.
TWO-SIDED COMPETITION
Buyers |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
$A_1$ values a horse at 30, and would buy at any price under |
30 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
$A_2$ values a horse at 28, and would buy at any price under |
28 |
30 |
30 |
30 |
30 |
30 |
30 |
30 |
30 |
30 |
30 |
30 |
30 |
30 |
30 |
30 |
30 |
30 |
30 |
30 |
30 |
30 |
30 |
$A_3$ values a horse at 24, and would buy at any price under |
24 | A. | B. | C. | D. | E. | F. | G. | H. | I. | J. | K. | L. | M. | N. | O. | P. | Q. | R. | S. | T. | U. | V. | W. | X. | Y. | Z. |
---|
Values of Horse for Buyers A., B., C., D., E., F., G., H., I., J., K., L., M., N., O., P., Q., R., S., T., U., V., W., X., Y., Z. |
---|
Values of Horse for Buyers A., B., C., D., E., F., G., H., I., J., K., L., M., N., O., P., Q., R., S., T., U., V., W., X., Y., Z. |
---|
Values of Horse for Buyers A., B., C., D., E., F., G., H., I., J., K., L., M., N., O., P., Q., R., S., T., U., V., W., X., Y., Z. |
---|
Values of Horse for Buyers A., B., C., D., E., F., G., H., I., J., K., L., M., N., O., P., Q., R., S., T., U., V., W., X., Y., Z. |
---|
Values of Horse for Buyers A.
546 VALUE AND DISTRIBUTION
Sellers |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
$B_1$ rates a horse at $10$, and would sell at any price over $10$ |
$B_2$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$B_3$ rates a horse at $17$, and would sell at any price over $17$ |
$B_4$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
$\ldots$ |
At any price over $20$ only six horses are demanded and five offered. The solution becomes essentially different when the rising demand reaches the point where at that price $A_4$ is compelled to cease bidding, and there are now only five sellers against five buyers. The bargain is concluded at $20$. But at the price $21 \frac{1}{2}$ there will be a sixth possible seller in the form of $B_6$. The limits within which the price must necessarily be determined are narrowed to $21 \text{ and } 21 \frac{1}{2}$. But meanwhile observe that, though in an isolated exchange, $A_4$ and $B_4$ get at the terms of sale by haggling, and though, in the case of the one seller $B_4$, in face of $A_4$, and $A_5$, and $A_6$, he gets at the terms of sale by haggling, it is hazardous step to conclude that any similar pairing-off can obtain under the two-sided competition of the ordi- |
1Positive Theory, pp. 203-206.
Reasoning the problem so as to make reservation prices appear in the demand schedule, we find that they stand as eight items for sale without reservation as against offers of $10$, $11$, $15$, $17$, $18$, $20$, $20 \frac{1}{2}$, $21$, $21 \frac{1}{2}$, etc., or twenty-six items. The price outcome is, of course, the same as before.
It is worthy of note that this view of supply as having also a demand side is not new. It was first suggested by Mr. Hume, who saw to be furnished from the demand schedule affords scant comfort to the demander. In his view, therefore, the number of items in the supply schedule must determine between which limit pair of demands the price shall finally settle. Otherwise than by somehow making allowance for this supply schedule, we are led back to demand forces and explained by them, the demand point of view will fail us.
But, in whichever manner the problems are analyzed, it is evident that no warrant has yet been given for asserting the paramount importance of any single factor in determining prices. We have seen, noting that for cases where individuals have excluded haggling, "we need see no more than that there is a certain law" (Positive Theory, p. .). The existence and origin of any one determining force must still be held in doubt. To justify the Austrian interpretation, supply has yet to be received into demand.
THE ADJUSTMENT OF PRICE 547
nary market, no matter how idealized may be assumed to be the conditions. It can hardly be true that, in order to reach the price adjustment, any particular individuals must give up their entire stock of goods. The reason has it that the least anxious actual buyer who would pay 22, if necessary, and the least anxious actual dealer with his limit of 20—the marginal bargainers—arrange the price adjustment through the process of haggling, which is a slow but gaining process. They certainly need not; all that the perfect market assumes is that such a price be reached as shall leave no one having the willingness to sell below the price to which he is willing to buy, and no one willing to leave unsupplied no purchaser who would yet take the commodity at any slightest fraction above the price established. For this price which will fulfil this condition may be established only by those who have no other chance; it is sufficient that it will not be disturbed. The chances are extremely countless to one that the marginal traders will not get together to haggle, and it is no mean clear that these are the traders of expected markets responsive to high prices. They are the most indifferent of all, in point of the volume of quasi-rents at stake, may not indeed fairly imply that they are the least interested in the particular penny or two, so be contented with a penny or two, whether number or per cent as stake, but the kind of people playing for these pennies, will mostly determine who will do the haggling and how much haggling will be done. The kind of shopping and bar-gain-coin mania deserve especial attention in this connection. There is no sufficient reason for supposing them to be purchasers at or near the margin of indifference.
And if we suppose that these people alone fix the margin chiefly do the haggling, their activity could be effective in setting the last touch to the price adjustment only so far as they were assumed not to be marginal. It is of little consequence whether or not the market haggling is not a force to modify the outcome; and, even upon the assumption of an inter-marginal area, it could be only within this narrowest of limits and as putting, so to speak, a stop to further change. But it is not necessary to fix the terms of the exchange. Certainly, in the broad view, these marginal or quasi-marginal bargainers are the results of the price limit, and not the cause of it. The marginals limit whether of demand or supply, differs from any other item only that through it as marginal increment
548
VALUE AND DISTRIBUTION
a determination may schematically be made of just what effect it, or any other single item, has upon the price adjustment, measurement being made from the point at which all the buyers who have left the market can have left the price. Not to the soldier who fires the last gun is the victory to be accounted, nor is the smallest boy who touches a fire-cracker to be held responsible for the Four-of-July explosion. But there is no doubt that a marginal buyer, the marginal price must coincide with his valuation; but neither the point of adjustment nor the buyer at this point is necessarily the lowest price. The least price will be forced out among all the buyers. True it is that, if he were not in the case, the price would have been other; but so is this true of all other buyers. The marginal demand is one among many demands, and each of these demands has its part in the resulting adjustment; but it is the entire demand in equilibrium with the entire supply which gives this market adjustment. Almost as well talk of the child who chases the wave up and down the shingle as fixing the wave-front.
For most purposes the marginal traders are observers. It is true that their added weight in the market may move the price on either side of it; but they are not those on which they build or to which they add is made by thousands of other demands in face of thousands of offers.
*That* this need saying is evident not merely from numerous cases of careless statement—some of the present writer's among them—but from cases where the marginal doctrine is made the basis of really bad decisions.
"At first sight it may appear strange that so few persons, and those so few only in certain classes, should be affected; but on closer examination this will be found quite natural. If all are to exchange at one market price, the price must be such as to suit all exchanges. Now, if we consider how far any one party can least capable contracting party suits in a higher degree all the more capable parties, we shall find that none of them can suit better than one who cannot suit at all. In short, no party whom he cannot suit, afford the standard for the height of price."
Positive Economics.
"We may go a little farther, and affirm that, so far from the money or commodity prices being regulated by some arrangement of ratios between the precious metals and other commodities that money demand can hardly ever be the regulator.... [I]t can hardly ever be supposed that any great change in the amount of supply is adjusted, and by which the ratio of exchange between the precious metals and other articles will be finally settled."—Griffen,
The Case against Bimetallism, pp. 94, 95.
THE ADJUSTMENT OF PRICE 549
The fact appears to be that the marginal method of analysis is of very limited application as an account of the concrete facts of industry, and is of even less value as a statement of causal sequences. As a thoroughly rationalized statement of facts, it may not be denied that it approaches the rationale as a formulation of the logic implicit in the market—it has, in some directions, an important function in economic investigation; but it says merely that, with the various occasions for which it is applicable, the forces at work in the market would work out in conformity with the illustrative scheme. It has nothing to say as to the nature or causal interplay of these forces.
This is in no sense to deny the important service of the marginal method, but rather to define and limit its purpose. Only by such close analysis of what is characteristic in marginal phenomena can we hope to understand the importance of value to market influences become intelligible, or a rational and detailed account of the ultimate relations of demand and supply to each other, and of both to market prices, become possible.
Precisely as demand at any given time must include all the purchasing dispositions in possession of money or of equivalent purchasing power translated into terms of money—credits, deposit rights, goods appraised in terms of money—so the supply schedule, in whatever manner it is formulated, must allow for all the commodities for sale on any money-price terms. As an intermediate step in the elucidation of this problem, price adjustments have been found to be formulated as a derivative from the price adjustment.
And here we may stop to question whether anything is really gained by distributing into the demand schedule the demand elements hidden in supply. Is the price adjustment thereby made either more intelligible or easier of manipulation?
Probably not; no preference is urged; it is only when, presently, other cost-of-production influences, the supply volume of any particular commodity is to be explained, or when an analysis is attempted of the influences by which
549
550 VALUE AND DISTRIBUTION
the volume of supply will, with passing time, be modified, that theoretical significance attaches to the proposed recast- ing of the demand and supply schedules.
But it remains true that, speaking generally of modern entrepreneur production, goods when once produced are sold for what they will bring; which really amounts to say- ing that in the main the practical significance of all reservation prices must be sought in this field.
For it is here best to endeavour to attempt some account of the bearing of cost-of-production influences upon supply. All influences tending to restrict the relative output of any commodity express themselves in the entrepreneur compu- tation as cost influences of the most unquestionable sort; and chief among these are the value-productive opportuni- ties open in other lines of production.
But, in point of fact, maximum cost of production is purely a matter of individual entrepreneur, there is hardly any limit to the influences that may bear to cancel or limit his production; but each such influence, by the very fact that it is supply-restricting, is the basis or the expression of a cost; whether it applies to one or to many or to all lines of product is irrelevant to the present pur- pose. No individual entrepreneur knows or cares as to the effect upon the relative values of different supplies. Possibly enough, the hazard or risk, the medium, or the ill repute, or the hope of profit, may make some costs appear more equally as serious burdens as in the line of production pursued; but there is, in any event, the alternative of non- production. The only cost-of-production question is, then, what remuneration must be received to maintain the out- put. This remuneration must be sufficient to keep the pro- ducer from deserting his line of production for another line, or from retiring, entirely or in part, to leisure. This the cost margin may be one of change of product or of restriction of product or of retirement. But, in any case, cost is the moosey statement of the necessary competi- sation.
A page from a book about economics.
THE ADJUSTMENT OF PRICE 551
Superficially viewed, there is, however, apparent force in the doctrine that all such burdens or expenses—if any such there are—as are common to all industries, could be omitted from cost computations, not as excluded from cost, but as irrelevant to relative costs; for, after all, it is, in the broad view, true that cost as bearing upon value must be cost in the relative sense. And it must be admitted that it would be possible to regard entrepreneur qualities, capacities, conditions, and preferences as the sole variant influences upon cost, taken in this relative sense, were it not for the fact that the different industries vary greatly in their techniques of response to these influences. The same entrepreneur agent or class of agents as against another. So were all industries alike in this respect, or, so far as the technology of production were concerned, were the different agents and instruments practically interchangeable at their established price levels, so that no agent or instrument could be in relatively short supply, their cost would remain no basis for variations in the relative costs of commodities, other than through such differences as depended strictly upon differences in entrepreneurs.
As the outcome of this discussion—more or less repetitious—we arrive at a clearer view of the relation of costs to value and of the ultimate determinants of value as expressing themselves through costs. Taking entrepreneurs as they are, with all their differences, and as competitively operating under given and common conditions of technology and resources. Among these entrepreneurs, with their different abilities and adaptations over against the conditions of human needs and wants on the one side, and over against the existing supply of instruments and the actual limitations and conditions upon the supply of these instruments, on the other side, we have a full explanation for the relative volumes of products and for their respective market prices. Cost of production is merely the entrepreneur computation under which these
A page from a book discussing the adjustment of price.
554 VALUE AND DISTRIBUTION
underlying forces and conditions exert their determinative influence upon the relative volumes of products.
Cost, then, is to the entrepreneur something more than a return for the outlays of production, often something more also than a personal justification for activity as against idleness. The producer must in most cases also justify his occupation against any other line of gainful activity open to him. As bearing upon price cost is, we repeat, nothing less than the sum total of expenditure in money, necessary to the bringing-forth of product. The search for the maximum price remuneration for the productive energies and opportunities at one's disposal discloses what is commonly the most important element in cost in any particular line of production.
Or, again, cost may be stated as the refusal price below which, as a margin, the advantages of some alternative of production or of recreation will tip the scales of choice. In this case, however, while the most important constituent, the relative advantages of some other line of employment, the compensations held forth for the ministry to other demands, supply takes on a distinctly dynamic aspect: it is no longer an inert or passive fact as the recipient of demand, and as having the capacity to affect value only as it satiates demand. It has rather the aspect of a resistance, since it is, in large part and in the relative sense, almost solely the expression of demands in other directions.
If it be granted that the supply schedule is equally a demand schedule, the futility of the contest between the utility school and the cost school becomes apparent. Each of the contestants will have to make room for the other, with both elements recognized as dynamic in precisely the same sort. Relatively speaking, however, the demand schedule is doubtless, a fixed one. However fully, then, the principle and its nature as demand can be recognized, there is much to say for the view that, given man as he is, with his equipment of desires and tastes and habits and
THE ADJUSTMENT OF PRICE 553
customs, modifications in price are most profitably studied from the point of view of variations in the supply term. Diminishing relative cost, how little now here, are the characteristics of a progressive economy. The fundamental causes of modification in value are, therefore, to be found in those changes of productive processes which, by diminishing the relative cost of particular commodities, expand their supply. A better process implies simply that per unit of product it is now necessary to divert a smaller total of productive forces from other lines of production.
Prices fall until the rising curve of product cuts the falling curve of price. In this new condition, the derived remunerations distribute themselves in view of the marginal urgency of the different demands. But the more clearly, then, is it an inadequate formulation to say that the market, having become overstocked at the old price, the increased supply makes it necessary to make sale connections with wants of lower intensity—that the larger supply has, so to speak, uncovered lower levels of demand, so that the market price is still too high for all but the most favored purchasers. Doubtless so it is, but it is equally the marginal supply price. Neither the relative-marginal-utility-demand items nor the marginal-utility-supply items can be, either alone or in conjunction, taken as fixing value, but only as commensurate with value.
CHAPTER XXVI
DISTRIBUTION
Fundamental to all problems of costs, and, indeed, to all aspects of distribution, is the problem of the fund or quantity of goods to be distributed. What is the distributive power of a given fund?
Suppose Farmer A to have succeeded during the year in covering his living expenses and to have added $500 to the value of his farm; what is his income, $500 or $1,000?
Mr. Cannan stands for the view that within income must be included (1) the things enjoyed, (2) the increase made in personal capital, the $1,000 solution. Professor Fisher, on the contrary, insisting that income is strictly to be distinguished from "present income," which consists purely of psychic services, the flow, while capital stands solely for the fund of possessions from which this flow is derived—holds that the $500 of improvement added to the farm is really not present income but rather the postpone-ment of present income, and that the increase in the value of the farm is merely the present worth, the capitalized value, of a prospective increase in income.
For the purposes of this present discussion, Professor Fisher's distinction may be disregarded; one; human affairs may, if one likes, be regarded in two aspects, on the one side, the appetitive side, an analysis of human desires for valuable services, on the other side, an investigation of the productive processes and the distributive forces under which these desires reach their more or less of satisfaction. Consumption is the final goal of production, psychic income (utility) the ultimate significance of the production-distribu- tion process. If by drought or fire or murrain and before any
*Economic Journal*, Vol. VII. p. 364.
* Ibid., pp. 334-335.
554
DISTRIBUTION 555
enjoyment had accrued, our capitalizing farmer had seen his improvements canceled, they could not be held to have furnished income, in any sense with which ultimately the distribution problem is concerned.
But how about the products of the farmer's kitchen garden, making part of his $300 of "living" supplies, or about the eggs and butter produced and consumed upon the farm, and so on generally about all those commodities and services which might have been, but were not, marketed?
Or how about the fresh air breathed and the fine views commanded and the neighborhood privileges enjoyed upon the farm—all facts controlled and appropriated under the right and title of the farm ownership, and all in their share contributing to its command of those rental incomes, of the long series of which the value of the farm stands as the capitalized sum? Or what about that annually marketable fact pass through the market crucible, before this fact can stand as an item in the social *distributendum*?
And what shall be said of the housewife's activities of the women folk at home, their errand-goings and slipper-bringings and nuzzlelike ministrations? Are these to be accounted unproductive merely by the fact that they are not, in any usual sense, paid for? And if productive, what is the relation of the product to the national dividend? What shall be done with such a question as this? And how exclude this, without applying the same rule to the efforts of the actors, teachers, and preachers? Or is the line of distinction still this one of appearance or non-appearance in market exchanges?
And, theoretically, these are far from being the most serious difficulties. If it be agreed that, even at the margin of withdrawal, work may be paid for, and that we abandon at the point where the pleasure of the activity, plus the pleasure from the product, is outweighed by the advantages attaching to other pleasant work or to leisure, where, then, shall be drawn the line between work and
A page from a book with text discussing agricultural economics.
556 VALUE AND DISTRIBUTION
play? Shall it be said that anything is work which is done with any slightest regard to the resulting product, and that only that is play which is done purely and solely for the very joy of the activity? This is probably the more com-
mon formulation; but what is *product* for the purposes of the case in hand? If one hires someone to play the violin for him, the activity of the player is clearly productive; that is why it is paid for. And if gratuitously rendered—a girl or boy playing on a street corner—what is it worth as the less a service and the less productive? And why is not equally productive the act of playing for oneself? Is the distinction purely in the fact that in this last case there is no external and marketable somewhat? Or is the dis-
tinction merely one of the degree of roundaboutness of the path by which the service arrives?
Surely if one spend some days in making a violin, later to be used for one's enjoyment, the process of making must be regarded as productive. Is the distinction then between the productive and the unproductive one of exter-
nality of result? or of roundaboutness? If one apply him-
self to grievous study in preparation for the making of
violins, this must be regarded as productive effort, unless,
indeed, the denial rest upon the fact that as yet there are no external results, the study being regarded rather as a preparation for producing than as actual production, and the skill acquired being used to produce, and its remun-
eration rather as wages than as interest.
But would it at all matter for the purpose, were the study a preparation not to make violins for oneself, but to play
violins to oneself? In economic usage, it is difficult to call this sort of preparation *work*, but in any lay sense of
the term, difficult to call it anything else. But, clearly enough, no product has yet manifested itself, of a sort to
rank as a psychic income, or to function as rent or hire or wage without first having aggregate product to be dis-
tributed. It seems, then, that not all desirable results fall within the dividend concept.
DISTRIBUTION 557
Possibly another line of approach to the problem will better serve. Leisure and recreation cannot altogether be excluded from the field of economic reasoning, since they rank as among the costs setting a limit to productive activity; not merely the outlays of production, but the pains of production, and likewise the pleasure displacements of production, are facts affecting the money recompense required to induce work. That is to say, recreation wants value-affecting influences to be taken into account in the computation of costs.
All this, however, does not involve the inference that, *from the market-value point of view*, recreation is productive. Recreation never looks toward marketable product, and is therefore irrelevant to market value, otherwise than in this aspect of cost. In the *market-value sense*, at any rate, it is not produced.
But how can there be individual point of view, say, in the Crusoe reckoning? Shall basking in the sun be, from this point of view, accorded productivity, even though it be a productivity without activity? Plainly, the result is a satisfaction, a utility, a psychic income. What does it signify in Crusoe's economic life--his *wirthschaftlichen Leben*--whether or not somewhere in the process an external fact presents itself? The truth seems to be that basking in the sun and eating fruit is a play which fails to be economic activity not by any test of externality, but only by the fact that free goods are not economic goods; they need no economizing. In the economic sense, the production of free utilities is not production at all. So if, aside from any market aspect of the case, violin-making were an enjoyable activity--a costless process, it could not become economically productive by the mere fact that a desirable thing resulted from it, any more than the playing of the violin for self- amusement becomes productive by the fact that it gives amusement. No matter how greatly prized mud pies may possibly be when once they are made, they are not thereby economic goods--and this, simply because they are
558
VALUE AND DISTRIBUTION
free-costless-goods; one may have them to the limit of his desires, exempt from all conditions of burden.
But again, what if the case be one where the recreation is really appreciated as a costly thing, as displacing some valuable product which, but for the overbalancing claims of recreation, would have been produced? In this case, the recreation must, it seems, be accepted as a productive fact; the resulting gain being greater than free goods of equal value obtained on terms of conscious sacrifice. Many a man foregoes an outing, not solely as a question of the expense of the trip, but in part because of the attendant suspension of earnings.
And surely the pleasure that, in any way, one pays for has a value; it is, indeed, the essential characteristic of all valuable psychic income that, in some sort, it costs to get it.
We conclude, then, that much that is called play, and much that is ignored, either as unproductive or as irrelevant to market-value computations, must, in the individual psychology, be held to be productive, and that only such activities are, in the individual reckoning, unproductive as are, in the actual thought of the individual, held to be costless.
This conclusion receives corroboration from the fact that where another plays the violin for your pleasure, and yet does it without charge, the naive common-sense would rank the case as neither more nor less productive than is one's own activity for one's own amusement.
But to declare an activity productive, whether individually or socially, is not precisely to establish its product as an item in the social *distribuendum*. Not all product is implicated in the distributive process; there is much productive activity which has no conflicting and conflicting claims on behalf of co-owners. Productive factors can attach, or, at all events, do attach, and which has no other relevancy to the distributive problem than is implied
DISTRIBUTION 559
in the fact that the production of it may have had some bearing to restrict the quantum of distributive goods.*
Regarding, then, the social dividend as made up solely of products ripe for consumption, and as including not all of these, but only such as, either by the conditions of their production or by the manner of their consumption, come to be involved within the distributive process, we are now concerned with what all this means in respect to service within the economic process, as distinguished from crime, or warlike foray, or other non-market predation, fall to their recipients under the guise of economic income. But not all incomes are received by title either of independent production or of co-operative contribution to production. If we are to render any adequate account of the forces determining the apportionment of the social distribution, it will be necessary to explain the actual distribution of purely productive and purely non-productive distributive process is adequate to do only in part. And productivity of income is something other and more inclusive than mere technological productivity, present or past. Goods for consumption are acquired through the possession and offer of current purchasing power; and this in turn may have been acquired by inheritance, by gift, by speculation, by gambling, by stealing, as well as by title of
*It would be hard to decide whether the net result of this discussion is concurrence in the prevailing doctrine or divergence from it. So far as this book is concerned, however, no effort has been directed to making the social-dividend concept precise. The discussions on the nature of productive and non-productive elements of income are both pertinent and illuminating for the purposes of the problem; and yet—as it seems to this writer—the social dividend is something else than merely a sum representing individual incomes. Smart (The Distribution of Income, chap. x) has discussed with great care the question whether the work of the housewife is productive; but he does so only in order to show that it is not productive. It is true that his conclusion is that she is not productive; but it is also true that he does not extend his analysis beyond this point. The prevailing opinion in extending the notion of productiveness, at the same time verging toward extreme restriction in deciding what share of the aggregate product is to be included in the distribution.
559
560
VALUE AND DISTRIBUTION
having produced, or of having helped produce, a valuable commodity. For, in truth, not merely the distribution of the landed and other instrumental, income-commanding wealth in society, but also the distribution of general purchasing power and of rights to wield and direct the application of general purchasing power, are, at any moment in society, to be explained only by appeal to a long and complex history, a distribution resting, no doubt, in part upon technological progress, in part upon the growth of capital, in part also tracing back to bad institutions of property rights and inheritance, to bad taxation, to class privileges, to stock-exchange manipulation, political favor, legislative and administrative corruption, pensions, tolls, royalties, perquisites, patents, interest on public loans, interest on consumption loans, and, as well, to every sort of vested right in iniquity.
And some of these mere rights of tribute come to be included in the production process, and to rank there as valuable market advantage or opportunity to such individuals as control these rights, e. g., business blackmail, royalties on patents and processes, tolls, transportation impositions, and the like. But there being no apparent method of bringing this class of facts within the orderly sequences of economic law, we shall—perhaps—do well to dismisse them from our discussion, merely stopping, however, to remark that they are facts which indicate that these incomes are so far vested as to promise future revenues—are capitalized under the discount principle, are salable like other acquiretive goods, are wealth for all individual ends of gain or of social prestige, and carry with them the right to participate in the enjoyment of the social product.
But none the less is there a distribution by right of productive contribution. And under this title must, among other things, be discussed the compensations allotted to human labor and human supervisory activity, as wages,
A page from a book with text discussing value and distribution.
DISTRIBUTION 561
salaries, and profits,—to the owners of instrumental goods, as rents, and to capital-owners, as time discount upon wealth in its time aspect.
But in order that the difficulty of the distribution problem be not exaggerated or its importance disparaged, recourse must be had to the principle that production precedes and conditions limits consumption, and that therefore the production-distribution process logically precedes and determines the latter. The various economic influences as have no basis in productivity rendered, and as modify—even profoundly—the ultimate apportionment of consumption rights. Interest received upon public war loans is of this secondary-distribution sort; so pensions, sinecure salaries, subsidies, profits upon corrupt contracts, and not a few of the secondary effects of taxation.
The chief theoretical difficulty in the subject is, indeed, to draw the line between this primary and this secondary distribution. Taxes are used for many other inter-actions; for example, consumption loans, by affecting the supply of funds for loan in productive directions, exercise an influence upon the discount costs of production. Such taxation, also, as can be appreciated by the contributor as falling upon his productive process rather than upon his consumption, are treated by him as production costs. Taxes also which burden a distinct line of raw materials function as cost charges on productive processes; they thereby burden distinct lines of consumption, and thus disturb the relative volumes of consumption goods to be marketed, may superficially appear to have no further effect than to redistribute the productive energies of society, but, nevertheless, by modifying the relative hires of productive agencies, do appreciably disturb the distribution of purchasing power in society. Privately achieved or publicly granted monopolies of productive patents, process royalties, trade secrets of method, and any other form of control over some or all methods of supply—all command rents, and thereby affect the distribution of purchasing power; and, on the other hand,
A page from a book with text about distribution.
562 VALUE AND DISTRIBUTION
the opportunities and advantages paid for under the form of these rents hold the same relation to cost of production as are held by land and other instrument differentials of advantage; these costs, in turn, are mostly passed along under the guise of enhanced market prices, and are ultimately mainly a burden upon the consumers of the goods, whereby, again, redistributions of purchasing power are initiated.
Monopolies or privileges of sale—as distinguished from those of production—function in this regard like taxes upon consumption. Transportation charges, whether justifiable or predatory, also operate like taxes, and are production costs or mere consumption tribute accordingly as the original incidence is upon production rather than upon consumption; but in either case the final burden rests in most cases and for the most part upon consumers.*
The broad principle for all problems of cost of production is, however, that any outlay or sacrifice for a differential opportunity, whether this be a mere permit or a license, or be attached to the possession of some agent or instrument of production, is a cost. For, as we have seen, any production cost is merely another way of looking at what is, from another point of view, a distributive share in the product. But that all agent or instrument hires are costs is far from saying that they include all costs. The technological point of view, which sums up costs as a total
*From "The Theory of Production," p. 307. It may have very considerable significance may be made to important problems in the theory of taxation and to the general principles upon which import duties should be based. The question whether the materialisation of society should therfore receive some illumination. But all this is left for further discussion on the basis of application and theoretical analysis. It must here suffice to note that such imperfections as, from the present point of view, are incident to the competitive process in its various stages—the results of the primary—the production—distribution, as to the political and property rights—taxation—social insurance—public works—taxes due place, and to the modifications of the primary distribution due to reactions upon it from the secondary distribution. (See in this connection note, p. 349.)
DISTRIBUTION 563
of wages, interest, rent, and profits, is in its general acceptance little short of astounding; for even if taxes, insurance, advertising, and like outlays may finally be traced to labor or capital-goods bases—-which, by the way, is not a simple matter with, say, taxes to pay interest on war debts—it is, at all events, clear that these are not outlays for labor or capital as technological factors in production. The tripartite, or any other technological classification of productive factors, must be especially misleading for purposes of the entrepreneur-cost computation.
And it is worth it to be repeated that the mere fact that cost-distribution shares are received through the entrepreneur as intermediary, does not imply either that no part of the entrepreneur remuneration, profit, is cost, or that all of it is cost. For it is precisely at this point of entrepreneur remunerations that costs and distributive shares fail of coincidence. All of the entrepreneur remuneration is a distributive share, but only for the marginal entrepreneur, or only for the marginal item of each entrepreneur's product; it is all other distributional shares—the result of entrepreneurship, all unnecessary or supra-minimum profits, are distributive shares falling outside of costs.
But this does not mean that all occupation or instrument-employment differentials above the next most attractive opportunity are non-cost facts wherever found. Cost is an entrepreneur reckoning; the entrepreneur knows what his own occupation-employment requires; he forces him to pay; but he cannot know and he need not care, what hire in some other employment the agent or instrument might command. For competitive purposes, occupation differentials are non-cost facts only for those individuals who receive the hire of them and to whom it, at the same time, falls to compute costs. It is the entrepreneur alone whose occupation differentials fulfill both these requirements. The self-employed laborer—entrepreneurship at its simplest—computes his costs as the money state-
A page from a book about economics.
564 VALUE AND DISTRIBUTION
ment of his best alternative line of conduct, whether this be one of independent production, or of wage-earning, or of leisure. The employee might—but without any bearing upon market cost of production—compute for himself in what degree his compensation was greater than an equivalent for his pain, or greater than his wage under another employer, or in another industry. And so, with equal irrelevancy to any market-value problem, might the labour computer compute that his rent return was greater (1) than that which he himself could make out of the land, or (2) than another in the same line of production would pay, or (3) than some tenant in some other line of production would pay. Or a collectivist society could properly compute as its land cost of any given product only the displaced alternative products. But the entrepreneur must compute as his cost not what he would pay, if he paid less, but what he does pay, as compensated thereto by all the facts of the situation.
Land worth 100 as wheat rent but only go in its next best use would permit, for collectivist computations, only go of cost; in a competitive society, this land will pay its owner in rent to more than it could command in any other line of production, and may, under the actual tenant, pay the owner 1 of rent more than any other tenant could or would pay. This is not the cost the cultivating tenant 100 of rent, it is a 100 cost for him.
This does not carry the cost computation to its closest approximation to accuracy, though even at its closest, something, as we have seen, must commonly be lacking to the entire accuracy of the productive imputation. If the actual renter at 100 is conscious that he could, in another line of production, make the land count him for 102 of return, the while that it is actually paying him 103 in which case he must compute against its actual capacity of 103 a cost not of 100, the rent outlay, but of 102, the foregone opportunity. His cost, so far as it is a land cost, is in his best foregone alternative; in the case supposed,
A page from a book with text on it.
DISTRIBUTION 565
this best alternative was not to keep his money in his pocket. The necessary price to induce the production of the wheat was not, in point of land cost, 100 but 102.*
Were all entrepreneurs, albeit of unequal abilities, yet equal in equipment of wealth, credit, and instrumental goods, and alike in adaptation to the equipment in hand, alike also in relative adaptation to alternate lines of employment, all costs would be equal in each respective line of production, and no entrepreneur more marginal than any other, or marginal as a different output of common product.
But, even under these conditions, wages for explaining all profits to fall to the general wages level—if such a level there were—unless they were also assumed that *all* men were equally able and equally well equipped and equally disposed to undertake entrepreneurship. With fluid and perfect competition among unlike entrepreneurs, instrument rents and time discounts would be forced so high a level that the last dose of expense, and each instrumental good employed therewith, would be employed at a rate of remuneration so high as barely to leave the entrepreneur inducible.
*The law of costs, correctly formulated, is applicable to all things competitively produced. In this case, the scarcity goods may enter into the production process. In the accurate sense, the term monopoly is used only when there is complete absence of restricted competition. But, in any case, the law has no reference to the underlying influences explaining the actual cost situation; it takes the situation as given.
And even in monopoly conditions the cost law may, without undue violence, be made to cover the computation under which a restriction on output becomes operative. For example, suppose that a manufacturer cancels some sales of his product because he cannot sell all he can manufacture; or that he cancels some sales of his product because he cannot sell all he can produce; or that he cancels some sales of his product because he cannot sell all he can supply; or that he cancels some sales of his product because he cannot sell all he can make; or that he cancels some sales of his product because he cannot sell all he can offer; or that he cancels some sales of his product because he cannot sell all he can charge for; or that he cancels some sales of his product because he cannot sell all he can charge for; or that he cancels some sales of his product because he cannot sell all he can charge for; or that he cancels some sales of his product because he cannot sell all he can charge for; or that he cancels some sales of his product because he cannot sell all he can charge for; or that he cancels some sales of his product because he cannot sell all he can charge for; or that he cancels some sales of his product because he cannot sell all he can charge for; or that he cancels some sales of his product because he cannot sell all he can charge for; or that he cancels some sales of his product because he cannot sell all he can charge for; or that he cancels some sales of his product because he cannot sell all he can charge for; or that he cancels some sales of his product because he cannot sell all he can charge for; or that he cancels some sales of his product because he cannot sell all he can charge for; or that he cancels some sales of his product because he cannot sell all he can charge for; or that he cancels some sales of his product because he cannot sell all he can charge for; or that he cancels some sales of his product because he cannot sell all he can charge for; or that he cancels some sales of his product because he cannot sell all he can charge for; or that he cancels some sales of his product because he cannot sell all he can charge for; or that he cancels some sales of his product because he cannot sell all he can charge for; or that he cancels some sales of his product because he cannot sell all he can charge for; or that他
The monopoly computation applies this principle: on the credit side of the account is placed (a) the extra outlays of production; (b) the charges against which are (i) the extra outlays of production, and (ii) the loss in price suffered by those who have been induced to reduce their purchases of the new items. The point of equation between the two sides of the account is the limit upon production.
For cost purposes, in truth, small occasion exists for any extended discussion either of monopoly problems, or of the relation of corporate
566
VALUE AND DISTRIBUTION
ment further to burden himself with supervision and further to increase the volume of his product. But still there would exist no warrant for asserting the equality of costs with the aggregate compensations of the productive factors, in any other sense than that competition could carry these compensations no higher. The distributive shares out of the product would be the higher for all entrepreneurs, the entire class alone being thus subjectively to instrumental goods and to employee labor. Instrumental goods of different sorts would be better paid relatively to labor or to entrepreneur ability accordingly as they were respectively in limited supply.
For it is to be remembered that, the demand for con-
sumption good being assumed, the demand for any instru-
mental good or agent is conditioned by the quantum of
instrumental goods or agents adapted to co-operate with
it and under its service as well as by the price of it just
as it is the limitation upon the supply of productive
instruments and agents that makes value possible, so
organisation to the traditional theory and terminology of the science.
However important and indeed, overshadowing in modern business the phenomena of the later methods of business organisation may be, not much interest attaches to them in this connection. For distributive theory does great significance attach to the corporate aspects of business organisation, excepting those are, in practical affairs, purely formal.
As to corporations purely and simply as such, there is little greater interest in their theoretical aspects than in those of the state. The theoretical aspects of the new problems presented have already been sufficiently summed up in an earlier examination of the bearing of corporate organisation on distributive theory and terminology of the cost-of-production problem. (See note, p. 88.)
Monopolies also offer few difficulties of analysis so far as the effects are confined exclusively either within the field of primary distribution or within that of secondary distribution. In vari-
ary distribution; very often, however, this simplicity of effects is not present.
Particularly are the activities of the operator in the field of "high finance" difficult to distribute between the two categories of market-value and market-price. This is because he acts as market
predator or parasitism—secondary distribution.
The activities of the two-waist-bonnet-trimmer, the
diamond-polisher, the parasitic-medicine manufacturer, the clown, and
DISTRIBUTION 567
it is the relatively limited supply of one class of productive factors that attaches to it a high remuneration relative to other increasing factors. The ultimate explanation for the value of any commodity, by it repeated, is found, on the one side, in the demand for the commodity, on the other, in the fact that the supply of productive means is limited, whether by the absolute scarcity of these productive facts, or by a scarcity due to the diverting influences exerted by the demand for other commodities. And the actual level of remuneration is everywhere reached through the bidding of entrepreneurs for increased productive means, which they obtain either by their own efficiency already in hand; and the actual payment is thus commonly in some rude approximation to the amount which the successful bidder is able to pay for the purpose of enlarging his production complex.
With goods present in stocks of similar items, this approximation is theoretically close accordingly as entre-preneur competition is close. Where each item of goods
the prostitute, are, in the economic sense of the term, clearly enough productive; on the other hand, the three-card monte man, the shell-game maker, and the lottery player, are not only less valuable intelligence, ranked as agencies of secondary distribution, under methods more or less analogous to those employed in primary trade; but even here there is a line between these two extremes as to be ranked the methods most distinctly characteristic of high finance, at all events, most notorious in prostitution.
The promoter or underwriter is in the business of producing stocks and bonds for sale to individuals who do not produce the products are commonly of considerably greater worth than those of the patent-medicine vendor, and indeed, are often of the very highest title no matter how they may be produced. The capital and personal wealth, much good-will or franchise or monopoly, more of pros- pectors and less of producers, and yet more than any combination of ingenuity skill of organization, combined into a marketable commodity most profit-giving to the producer, commonly, truly, of moderate advantage to him alone and not generally available to the general public. All this is hard to rule out of the category of market productivity.
The later processes by which the market is rigged through bear stories and through artificially low dividends, or by declaration of unearned profits and other devices designed to make possible predation, be safely classified as entirely within the field of secondary distribution. (Cf. Veblen, *Theory of Business Enterprise*, passim.)
A page from a book with text about economics.
568
VALUE AND DISTRIBUTION
is sui generis, the room for haggling is appreciably more ample and the point of price adjustment is quite possibly found at considerably below what the successful bidder would, if necessary, have consented to pay.
With the recognition that entrepreneurs are different must come the abandonment not only of the notion that profits can arrive, in any conceivable state of equilibrium, at equality, but also of the notion that, at no matter what point of development in technique, there can ever be any one best combination of productive factors among various different factors or classes of productive agents. There is indeed, no such best combination for any one entrepreneur, excepting upon the assumption of an established level and ratio of prices and of hires upon the different productive agents; with each change in these relative hires, that which was best becomes not best, and the production complex undergoes a reconstruction. And finally, with varying financial resources, the best combination is again a different combination. The entrepreneur has to make his own effective combinations to the conditions set by their capital and their credit; what additions or subtractions of different factors are expedient depends so much upon what would be advantageous if the entrepreneur could command the necessary resources, as upon what he can with reasonable caution attempt.
CHAPTER XXVII
SUMMARY OF DOCTRINE
In the interests of economy of space, and to avoid the further detailed repetition of what has already been sufficiently set forth in earlier pages, the following propositions are presented:
*Value* is a ratio of exchange between two goods, quantitatively specified. The concept of a general market value depends upon the assumption of an established medium of exchange, and finds its only expression as *price*.
The primary fact in the explanation of value is the existence of human needs and desires. *Utility* as expressed in the existence of *goods* is merely the relation of adaptedness of the thing or fact to the human need or desire. Limitation upon the supply of goods relatively to the need gives value. Thus value in productive goods is ultimately explained by human desires over against a limitation of supply due either to want of instrumental goods or to the irksomeness of effort or to both.
With great esteem for good singing and with the rarity of good singers, the high gains of prima donnas find sufficient explanation.
With scarce iron mines and a relatively high need for iron, a high value is given to iron readily exploited.
With relatively scarce supplies of land, and a high need for wheat, the high value of wheat land would be explained, irrespective of the fact that various other uses for land further greatly restrict the supply of wheat land.
Human needs and their relative intensity being assumed, the value-causal sequence runs from relative scarcity of agents to relative scarcity of products; from relative scarcity of products to high exchange power of products—
50
570
VALUE AND DISTRIBUTION
high value, relatively high price; from relatively high price of products to relatively high remuneration of agents; from relatively high remuneration of agents to relatively high present worth of agents.
Under the competitive activity of various and differing entrepreneurs, each seeking his most advantageous line of activity in view of his particular situation in point of capital, credit, ability, and preference, market supplies of products are worked out in adjustment with the price demand; and under the competitive bidding of entrepreneurs for productive auxiliaries, the market values of instruments and agents are worked out, and the cost situation confronting each individual entrepreneur determined.
That underlying the competitions and costs of entrepreneurs is a situation, a controlling complex of fundamental facts, under the influence and direction and determination of which the details of market production and of value adjustment take place, and with changes in which most common commodities play their part. This situation is a situation in which, furnishing us with the principle from the point of view of which must be examined the dynamics of value and of distribution, a group of problems having to do with the manner and degree of change in market adjustments attendant upon different probable or possible changes in the underlying situation.
Neither in utility on the demand side nor in pain cost on the supply side can there be found a common denominator or standard or determinant of market value, or of price as its money expression. The only common denominator of value is found in the selection of a conventional standard for the purpose, a price commodity.
Neither in terms of market-value equivalence nor in terms of pain or cost, but only in terms of
SUMMARY OF DOCTRINE 571
utility equivalence is to be sought the standard of deferred payments. Equivalence in terms of unspecialized purchasing power, expressed under some conventional price standard, is the only resource for the case.
The equation of demand with supply is an explanation for value only in the sense that the entrepreneur-cost situation and the condition of price demand reflect and express the effects of the underlying and determining situation. Therefore both demand and supply themselves require analysis and explanation.
DEMAND
Utility, marginal utility, and subjective worth are primarily categories leading up to the explanation of the demand side of the value equation, as expressed in terms of purchasing power, and as bearing upon the price adjustment of any particular commodity.
Marginal utility—a purely individual category and an absolute magnitude—is a step toward explaining subjective worth—a purely subjective and individual fact and an absolute feeling magnitude, the cost aspect of marginal utility. Two subjective worths in comparison explain price offer, or refusal price; this latter being merely a demand fact in another aspect.
To different men, utilities, marginal utilities, and subjective worths are, as such, incapable of comparison; nor is it possible to give to utility, marginal utility, or subjective worth expression or measurement in terms of money. A maximum demand price expresses merely the equivalence, in point of subjective worth, of the thing bid for and the thing otherwise to be obtained through the purchasing power.
cost
The emergence of value is not dependent upon cost-of-production influences as a prerequisite, but only upon there
572
VALUE AND DISTRIBUTION
being a supply limited relatively to human desires. But so far as the cost-of-production investigation bears to explain the relative volumes of supply of different commodities, it bears to explain the values of these commodities.
For competitive purposes, cost of production is purely a computation of the individual entrepreneur; for any item or quantum of product, it is the price statement of the compensation necessary to the forthcoming of that product.
Outlay costs to the entrepreneur are distributive shares to the recipients; the distributive share of the entrepreneur also--his profit--is cost, to the extent that it is necessary profit.
But the distribution that takes place under the production process and at last and current is not the only distribution process in society. Such incomes as are not otherwise explained by title of separate or co-operative productivity, find their explanation in those other social facts and forces which distribute purchasing power in society.
All productive-distributive compensations come by the same and equal title of contribution to value productivity; but they are the market value of the value contribution rather than the accurate equivalent of the value productive-ness; this last varies for each instrument with each entre-preneur, and is nowhere precisely ascertainable by any.
Only relative costs of production have to do with the exchange relations of goods.
All influences making to increase the indemnity-price total which a commodity must afford to its producer if it is to be produced, rank, under the price denominator, as costs; chief among these influences is commonly opportunity cost--demand in another direction functioning as resistance in the given direction.
SUMMARY OF DOCTRINE 573
The resistance attractiveness of recreation or of rest may also be included within the opportunity-cost concept if interpreted broadly. Instrument and agent costs are often accurately to be reckoned as costs only in this opportunity aspect.
As a cost concept capital is neither technological nor social in significance; it stands for the total invested fund of value inclusive of all instrument values, and of all general purchasing power devoted to the gain-seeking enterprise; it is an acquitative category.
For competitive purposes, the capital concept should be formulated in the individual, private, and competitive sense. It should include all things, facts, or rights having value so that to them abstinence—the postponement of consumption—is applied. Capital in this sense is a private fund of wealth possessed under the time denominator and viewed in the time aspect. The market value of any basis of income is the present worth of its entire series of putative incomes.
Market time—discount—interest in the accurate and ultimate sense—is the premium, expressed as a rate per cent, per unit of time, which any fact, as present purchasing power expressed in terms of the conventional standard, commands over future purchasing power likewise expressed.
The surplus in any instrument hire over upkeep or depreciation is a market time-discount fact; expressed as a ratio between the value of the instrument and the hire, it is interest in the accurate sense.
All costs are merely sacrifices of production reduced to terms of the price denominator. Costs, then, include, among other items, all necessary indemnities for capital outlays in production, and a time-discount charge upon the capital fund invested.
THE PRICE AND DISTRIBUTION
price is a less limited measure to human desires. But so far as the production-distributive mechanism bears to explain the relation between price of different commodities, it must be called the cause of these inequalities.
If a productive person's cost of production is purely a compensation of the minimum expenditure for any item or element of product, it is the price statement of the entrepreneur's measure to the functioning of that product.
Thus only in the entrepreneurs are distributive shares to the remuneration distributive share of the entrepreneur himself profits equal to the extent that it is necessary profit.
But the distribution that takes place under the production process and as part and parcel of it, is not the only distributive process in society. Such incomes are received otherwise than by use of separate or cooperative productivity, find their explanation in some other social facts and forces which distribute purchasing power in society.
All productive-distributive compensations come by the same and equal title of contribution to value productivity; but they are the market value of the value contribution rather than the accurate equivalent of the value productive-ness; this last varies for each instrument with each entrepreneur, and is nowhere precisely ascertainable by any.
Only relative costs of production have to do with the exchange relations of goods.
All influences making to increase the indemnity-price total which a commodity must afford to its producer if it is to be produced, rank, under the price denominator, as costs; chief among these influences is commonly opportunity cost—demand—supply—cost—functioning—and—
A page from a book discussing economic principles.
SUMMARY OF DOCTRINE 573
The resistance attractiveness of recreation or of rest may also be included within the opportunity-cost concept if interpreted broadly. Instrument and agent costs are often accurately to be reckoned as costs only in this opportunity aspect.
As a cost concept capital is neither technological nor social in significance; it stands for the total invested fund of value, inclusive of all instrument values, and of all general purchasing power devoted to the gain-seeking enterprise; it is an acquisition category.
For competitive purposes, the capital concept should be formulated in the individual, private, and competitive sense. It should include all things, facts, or rights having value so that to them abstinence—the postponement of consumption—applies. Capital in this sense is a private fund of wealth expressed under the price denominator and viewed in this aspect. The market value of any basis of income is the present worth of its entire series of putative incomes.
Market time—discount—interest in the accurate and ultimate sense—is the premium, expressed as a rate per cent. per unit of time, which any fact, as present purchasing power expressed in terms of the conventional standard, commands over future purchasing power likewise expressed.
The surplus in any instrument hire over upkeep or depreciation is a market time-discount fact; expressed as a ratio between the value of the instrument and the hire, it is interest in the accurate sense.
All costs are merely sacrifices of production reduced to terms of their denominator. Costs, then, include among other necessary indemnities for capital and a time-discount charge upon the
A page from a book with text on it.
574
VALUE AND DISTRIBUTION
But the cost to the individual entrepreneur is not a funda-
mental explanation of anything; it assumes values upon instrumental
facts as a step toward value explanation. Nor does the aggregate
activity of all individuals explain the value of any particular goods,
unless and until the great underlying facts of human wants and
capacities, and of instrumental equipment and opportunity are in-
cluded in the survey.
Static analysis takes as definitive and ultimate the actually existing total situation, inclusive of human needs and productive powers, and with all the existing supplies and existing limitations of equipment and opportunity, and all this irrespective of how far the situation has been modified by past activity, by past acquisi-
tion, and irrespective of whether human activity has in the past added or subtracted relevant elements, aspects, or facts. Not the outlays that produce goods or their prices but the scarcity of
incomes, but the scarcity of these productive factors relatively to the
human need, is responsible for the emergence of scarcity of prod-
ucts anywhere and for the relative scarcity of products which under-
lie and determine prices.
But the inadequacy of the general equipment does not explain
the market values of any particular line of products, that is to say,
the exchange relations between different classes of goods. Inside the
general situation, each particular line of production will reveal out
of the relative inadequacy of productive equipments for the various
lines of commodities, in view of the relative strength of the pur-
chasing power disposition in these various commodity directions.
Here enter the influences of various different lines of production to
restrict the supplies of productive factors in each particular line of industry.
All rent outlays, whether for land or for other instru-
mental goods, and all wage outlays and all discount charges
upon the capital fund employed in production are equally to
be included within costs of production as an intermediate
explanation of the supply side of the value equation.
Nowhere is the distinction between price-determining and price-
determined costs valid. In the main, the value of each productive
fact is variable, because it is subject to change by new experi-
ments, each, through its products, in its small measure, a value-
affecting influence. So also each individual activity bearing upon price or related to price, whether, on one hand, of production or
of sale or do so, on the other hands of price offer or of price
refusal, is, in the main price-determined, because chosen in view
SUMMARY OF DOCTRINE 575
of the actual price situation and in adaptation to this situation; but each such activity, as affecting in its own small measure the aggregate of supply and demand, must thereby and pro tanto act as a price-determining influence.
The only one of the several rent concepts important to the cost analysis is that of the actual hire; but as opportunity cost, the land or any other productive fact may figure as cost at something vaguely more than the actual hire paid.
Costs to the entrepreneur are mostly but not entirely traceable (1) to value servicability to entrepreneurs in other lines of production, or (2) to value servicability to entrepreneurs in the same line of production, or (3) to alternative value possibilities of the productive facts, inclusive of the entrepreneur's own productive power, when under his own employment. But pain and weariness and displaced recreation have also their place in fixing the total remuneration necessary to the forthcoming of product.
All margins are ultimately personal and not instru- ment margins. Instruments are marginal only with refer- ence to the entrepreneur and relatively to him and to his situation; marginality is a psychological attitude with reference to productive activity or to the productive employ- ment of instruments.
Marginal instruments are variously understood to indicate (1) valueless instruments, a market-value standing, not inconsistent with the rendering of services for which the user would, if necessary, pay an appropriate price; (2) instruments whose usefulness has no real value nor personal-value significance; (3) instruments which at the actual market charge are just barely worth employing by the actual employer.
Number one is a concept derivative from the relationships of instrumental goods to entrepreneur activity, but not necessarily giv- ing a precise expression to any one of these relationships. Numbers two and three are merely different expressions of relative to entrepreneur servicability. Marginality is, in last analysis, an entre- preneur attitude with relation to one's own productive activity or to the productive efficiency of agents and instruments.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
INDEX
A
ABSENCE: Carver on, 355, n. See Cost of Production; Interest; Cap-
ital.
ABSTRACT CAPITAL, 170-74. See Capital.
Abstract: Scholastic terminology criticized, 308-11. See Boehm-
Bauerk, Wieser; Utility; Cost and Utility; Subjective Value.
AVERAGE MAN, 466, n.
As bank fund: chap. xi; Cairns's view, 1865; Ricardo's confusion,
41; related to banking, 159-69.
Various concepts: Boehm-Ba-
uerk, 149-50; Fisher, 149-50;
154-55; Fisher, 149, n.; 157-61;
Smith, 149; Ricardo, 151; J. S.
Mill, 152-53; Veblen, 153; Clark,
170-74.
Abstract capital: 170-74.
Competitive capital: chap. xi.
Capital: N. B. abstract capital,
172, n.; 173, n.; absence and
presence of capital: distinction
distinguished, 133, n.; on op-
portunity cost, 383; on price re-
lations with interest rate, 150;
wage determination, 218; measure of
capital value, 228; on profit,
493, n.; criticism of Clark, 440, n.;
competition and cost of production.
Charity: expenditure for, 239, n.
Clark, J. B.: chap. xiiil: abstract
capital, 228; N. B. abstract capital,
184, 279, n.; on interest—contro-
versy with boehm-Bauerk, 202-23;
on money costs of production,
of output, 434-454; labor as fund,
279-80; on profit, 279-80; on
income forms, 480, 514.
Commons, J. R.: laws of return,
165.
COMPETITIVE CAPITAL. See Capital.
CONSUMPTION: productive and un-
productive consumption.
COST OF PRODUCTION: various cost
concepts: chap. xi; Ricardian doc-
trine: chap. xi; labor-purchase
value concept: labor-cost theory:
labour-value concept: labor-price cost,
383-84; labor-cost theory: labor-
price cost: labor-cost theory: labor-pro-
ductivity and interest and profit:
chap. xiv. See Cost of Production;
Loan Fund.
577
578
VALUE AND DISTRIBUTION
costa—Marshall, 731; cost in Crusoe economy, 84;
Cypriot, 60; cost, 5; Adam Smith's doctrine, 10-12; Ricardo's doctrine, 48-49; Austrian doctrine, 335; multiplied utility at, 85, 347; 366; any duplication of cost, 335; useful to cost, 290-5; as related to profit, 290-5; as related to profit opportunity—Fetter, 69; Seager, 99; n.—Flint, 99; n.; Carver, 100; n.; Kneale, 101; collectivistic cost, 291; 336.
Abraham Lincoln—Senor, 50; capital-use cost, 5.
Investors—J. S. Mill's explanation, 59.
Wage-assistance cost, 34.
Profit and cost—Seyyed Ali, 65; relation of profit to cost, 88; Hadley's doctrine—Wakker, p. 91; Fetter's, 104; Senor, 106; Frazier, n.; cody: Seligman's, 100.
Cost merely relative, 85; 367; com-
petitive costs—Mill, 60; is opportunity cost, 60.
Theories of cost by J. S. Mill: adopted by J. S. Mill, chs. 5-6.
Econ and cost—chap. xii.
Mill's doctrine—chap. xii.
Ricardo's controversy with Say,
1821-22—Senor, chs. 88-89.
Senor's chs. 88-89: Adam Smith's,
ch. vii., p. 77; Ricardo's controve-
rsy—ch. vii., p. 77; Ricardo's controve-
rsy—ch. vii., p. 77; Ricardo's controve-
rsy—ch. vii., p. 77; Ricardo's controve-
rsy—ch. vii., p. 77.
Marginal cost—chap. xvii., p. 162.
Clashes between theory and fact: extensive and intensive margins, n.; past and present determinations determined, n.; n., n.; n., n.; marginal determination—Senor's determina-
tion, chap. xxii.
Land as opportunity cost—J. S.
Mill: chs. xiii.-xiv.; Jevons: chs.
xv.; A. S. Johnson: Macfie-
son: chap. xvi.; Hume: chap. xvi.; spo-
ts: Marshall: chs. iii.-iv.; calculations of different factors to cost,
135-37.
Risk cost, chs.
Utility vs income, chs.
Ultimate determinants-situation
costs: chs. xxx., xxx., xxx.;
Skill and cost, chs.
D
DAVENPORT, H. J.: Outlines of Eco-
nomic Theory, cited, 186 n., 187;
departmental price index system:
price, xxxp.; normative value, xxxp.;
xxxp.; xxxp.
DIFFERED PAYMENTS: See Standard of Deferred Payments.
Difficult problems, chap. xxxv.: de-
leted, xxxv.
DIETZEL: Heinsen criticized by
Boehm-Bawerk, 135-37.
Distribution: chap. xxxvi.: by
value productivity, chap. xxxii.;
by value of production, chap. xxxiv.;
by degree of utility (utility), xxxii.;
size of xxxp.; views of Hobbes and Car-
bonel on distribution
DYNAMICS: of value and distribu-
tion, chap. xxxvii.
Effective UTILITY: See Marginal Utility.
Entrepreneur CAPITAL: See Capital.
EXPENSES OF PRODUCTION: See Cost of Production.
FACTORS OF PRODUCTION: a techno-
logical classification, xxxvii.-xxxix.;
Adam Smith's view on labor and capi-
tal: his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii., n.; his view on labor and capital,
xxxvii.,
FETTEr: F.A.: definition of de-
mand, xxxiii.: interest theory, xxxiii.;
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income taxation: income taxation:
income
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHER-WARNER ON HEDONISM,
FISCHERA
INDEX
579
Flaxx, A. W.: definition of demand, 533 n., profits related to cost, 90, n., 106, n.; expense cost and opportunity cost compared with return, 496, n.
FINDING of productive factors, 463.
Foucault, J. B., 428, 446; Holmes on, 495, n.
G
Gallant, F., 107: 211, n., 213.
GIDE, CHARLES: on laws of return, 496, n.
GILBERT, ROBERT: determination of value, 548, n.
GONNER, E. C. K., eo.
GREENE, L. P., 93.
H
HADELEY, A. T.: definition of de-
mand, 533 n., n.; profits related to cost,
496.
HEDONISM, 48, n.; n.; 50.
HOBSON, J. A.: chap. vii: forced gain,
gain; cost of production, 414;
displaced rent and cost, 490;
comparative advantage and savings,
530 n.; utility and value,
326 n.
HOLLANDER, J.: rent and cost, 490;
intensive margin, 266.
HYND, A. M.: rent and cost, 490.
HINMAN, H. S.: labor measure of value,
777 n., n.
INCREASING RETURNS. See Returns.
INCOME: vs. capital, 556.
INTEREST: chap. xlv; iv; defined,
188; problem stated, 290-291; pro-
ductivity of labor increased by cost;
cost as cont of production; 756;
rewards to labor and capital; Bawer
on, 849; Fisher, 843; n.;
Ricardo's difficulty; 384-385.
LITERATURE: as to relative rent credit,
258 n.
J
JEVONS, W. S.: relation to eco-
nomics and politics; on profit and loss,
319, n.; definition of wealth, 314,
n.; displaced rent and cost, 490;
surplus and deficit in production,
JOHNSON, A. S.: displaced rent as
cost, 290.
JOHNSON, J. F.: on loan capital,
174.
L
LABOR: as cost: see Cost of Produc-
tion; as fund: 953, 459; produc-
tive power of labor: see Pro-
ductivity; as cost denominator, 15,
25, 118; as cost determinant; 15;
as comparative advantage; see
Comparative Advantage; Ricardo,
chap. iii; Cairnes, chap.
vi. See also Production.
LANKTON: see P. B. A., eo.
LAND: as capital: see Capital; as
industries: 663, 474, 479; its op-
kern: see Rent; tax on land:
taxes on land: see Taxes on Land.
LATOURELLE: J., measurement of
value, 777 n., n.
LOAN FUND: Ricardo on, 160. See
Capital.
LENTARY: 539 n., n.
M
MACPHERSON, C. W.: use of land
and manure curves, 636 a.; forced gains
and losses: see Losses; effect of
profit: rent; post rent: land costs;
labor as fund: 463 a.; interest,
280 a.
MACKENZIE: J., on Hedonism, 207,
n.
McCUlough: J. R.: interest as cost,
280 a.
MALMSTRÖM: T. R.: laws of return,
chap. xiii: doctrine of cost; 53j;
definition of profit and loss; labor
as cost measure and determi-
nation of value; Ricardo's difficulty,
n.; various kinds of profit and loss:
n.; etc.; cost doctrine criticized by
Ricardo.
MARIONI: extensive and intensive:
see Cost of Production; personal
use of land: see Use of Land;
in price fixation; 405; nature of
return in gain; cause or result;
various kinds of profit and loss:
BIBLIOGRAPHY [360 ff.] (Coxen)
The value produced in any country is the sum total of all the values produced in that country
VALUE AND DISTRIBUTION |
MARGINAL UTILITY: chap. xvii; as value determinant, chap. xxi. |
im, 310 n., 311 n., on Gallian, 236 n. |
|
MASSENET: chap. xx; measurement of utility by price, 375; on Levant, 237 n.; displaced rent as cost, 287 n. |
on consumption, 287 n.; on cost, 290 n. |
PYSCHOGRA: cost theory, 107; See PRINCIPLES OF ECONOMICS. |
COMMODITY: chap. xx; on cost, 290 n.; long-time reckoning, 480; cost of production, 480 n. |
on cost, 290 n.; on labor, 290 n. |
PRESBURY, N.: G. on cost, 288 n. |
MAYK, KALE: on money, 293; labor measure of value, 177 n.; on labor (fund), 290 n. |
|
PLEASURE: See Utility. |
MENGER, ANTON: relation to eco- nomic doctrine, 334; definition of subjective value, 334; on subjective valuation, 334; on imputation, 360; capitalization, 360; |
|
POPULATION: redistribution of, 466 |
MONETARY SYSTEMS: chap. xxviii; on cost, chap. xvii; marginal cost and relative costs, chap. xvi; on cost, chap. xix; displaced rent as cost, xxxi. See Capitalist. Productivity. |
|
on cost, xxxi; normal, xxxi; rent and price, xxxi. See Rent. |
MERCANTILISTS: value doctrine, 107; notion of productivity; see Pro- ductivity. |
|
FACTORS OF PRODUCTION: |
MILL, JAMES: interest as cost, 37; |
|
Productivity: defined, 191-24; |
on interest rate as cost with Mathias, xxxiv n., xxxv n., xxxvi n. |
on consumption, xxxiv n.; Mer- cialism: See Mercantilism. |
on consumption, xxxiv n.; Mer- cialism: See Mercantilism. |
See Capitalist. Fixed and circulating. |
on circulation of money, xxxiv n.; ex- ecutive action, xxxiv n.; funded, |
on circulation of money, xxxiv n.; ex- ecutive action, xxxiv n.; funded, |
MILLER: JAMES: interest as cost, |
460 n.; on cost of production according to chapter xix. See Fac- tors of Production. |
460 n.; on cost of production according to chapter xix. See Factors of Production. |
cost, chap. xvii; marginal cost and relative costs, chap. xvi; on cost, |
|
|
cost, xxxi; displaced rent as cost, |
|
|
xxxi. See Capitalist. Productivity. |
|
|
MONEY: function in deferred pay- ment, xxxi; See Price. |
|
|
MONEY: cost in $55 n., | NON-COMPETITIVE GROUPS: 71-80. | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
---|
|
---|
|