Principles of Economics

Marshall, Alfred
Display paragraphs in this book containing:
First Pub. Date
London: Macmillan and Co., Ltd.
Pub. Date
8th edition
20 of 71




§ 1. The law of or statement of tendency to Diminishing Return may be provisionally worded thus:


An increase in the capital and labour applied in the cultivation of land causes in general a less than proportionate increase in the amount of produce raised, unless it happens to coincide with an improvement in the arts of agriculture.


We learn from history and by observation that every agriculturist in every age and clime desires to have the use of a good deal of land; and that when he cannot get it freely, he will pay for it, if he has the means. If he thought that he would get as good results by applying all his capital and labour to a very small piece, he would not pay for any but a very small piece.


When land that requires no clearing is to be had for nothing, everyone uses just that quantity which he thinks will give his capital and labour the largest return. His cultivation is "extensive," not "intensive." He does not aim at getting many bushels of corn from any one acre, for then he would cultivate only a few acres. His purpose is to get as large a total crop as possible with a given expenditure of seed and labour; and therefore he sows as many acres as he can manage to bring under a light cultivation. Of course he may go too far: he may spread his work over so large an area that he would gain by concentrating his capital and labour on a smaller space; and under these circumstances if he could get command over more capital and labour so as to apply more to each acre, the land would give him an Increasing Return; that is, an extra return larger in proportion than it gives to his present expenditure. But if he has made his calculations rightly, he is using just so much ground as will give him the highest return; and he would lose by concentrating his capital and labour on a smaller area. If he had command over more capital and labour and were to apply more to his present land, he would gain less than he would by taking up more land; he would get a Diminishing Return, that is, an extra return smaller in proportion than he gets for the last applications of capital and labour that he now makes, provided of course that there is meanwhile no perceptible improvement in his agricultural skill. As his sons grow up they will have more capital and labour to apply to land; and in order to avoid obtaining a diminishing return, they will want to cultivate more land. But perhaps by this time all the neighbouring land is already taken up, and in order to get more they must buy it or pay a rent for the use of it, or migrate where they can get it for nothing*8.


This tendency to a diminishing return was the cause of Abraham's parting from Lot*9, and of most of the migrations of which history tells. And wherever the right to cultivate land is much in request, we may be sure that the tendency to a diminishing return is in full operation. Were it not for this tendency every farmer could save nearly the whole of his rent by giving up all but a small piece of his land, and bestowing all his capital and labour on that. If all the capital and labour which he would in that case apply to it, gave as good a return in proportion as that which he now applies to it, he would get from that plot as large a produce as he now gets from his whole farm; and he would make a net gain of all his rent save that of the little plot that he retained.


It may be conceded that the ambition of farmers often leads them to take more land than they can properly manage: and indeed almost every great authority on agriculture from Arthur Young downwards, has inveighed against this mistake. But when they tell a farmer that he would gain by applying his capital and labour to a smaller area, they do not necessarily mean that he would get a larger gross produce. It is sufficient for their argument that the saving in rent would more than counterbalance any probable diminution of the total returns that he got from the land. If a farmer pays a fourth of his produce as rent, he would gain by concentrating his capital and labour on less land, provided the extra capital and labour applied to each acre gave anything more than three-fourths as good a return in proportion, as he got from his earlier expenditure.


Again, it may be granted that much land, even in a country as advanced as England, is so unskilfully cultivated that it could be made to give more than double its present gross produce if twice the present capital and labour were applied to it skilfully. Very likely those are right who maintain that if all English farmers were as able, wise and energetic as the best are, they might profitably apply twice the capital and labour that is now applied. Assuming rent to be one-fourth of the present produce, they might get seven hundredweight of produce for every four that they now get: it is conceivable that with still more improved methods they might get eight hundredweight, or even more. But this does not prove that, as things are, further capital and labour could obtain from land an increasing return. The fact remains that, taking farmers as they are with the skill and energy which they actually have, we find as the result of universal observation that there is not open to them a short road to riches by giving up a great part of their land, by concentrating all their capital and labour on the remainder, and saving for their own pockets the rent of all but that remainder. The reason why they cannot do this is told in the law of diminishing return; that return being measured, as has already been said by its quantity, not its exchange value.


We may now state distinctly the limitations which were implied under the words "in general" in our provisional wording of the law. The law is a statement of a tendency which may indeed be held in check for a time by improvements in the arts of production and by the fitful course of the development of the full powers of the soil; but which must ultimately become irresistible if the demand for produce should increase without limit. Our final statement of the tendency may then be divided into two parts, thus:—


Although an improvement in the arts of agriculture may raise the rate of return which land generally affords to any given amount of capital and labour; and although the capital and labour already applied to any piece of land may have been so inadequate for the development of its full powers, that some further expenditure on it even with the existing arts of agriculture would give a more than proportionate return; yet these conditions are rare in an old country: and, except when they are present, the application of increased capital and labour to land will add a less than proportionate amount to the produce raised, unless there be meanwhile an increase in the skill of the individual cultivator. Secondly, whatever may be the future developments of the arts of agriculture, a continued increase in the application of capital and labour to land must ultimately result in a diminution of the extra produce which can be obtained by a given extra amount of capital and labour.


§ 2. Making use of a term suggested by James Mill, we may regard the capital and labour applied to land as consisting of equal successive doses*10. As we have seen, the return to the first few doses may perhaps be small and a greater number of doses may get a larger proportionate return; the return to successive doses may even in exceptional cases alternately rise and fall. But our law states that sooner or later (it being always supposed that there is meanwhile no change in the arts of cultivation) a point will be reached after which all further doses will obtain a less proportionate return than the preceding doses. The dose is always a combined dose of labour and capital, whether it is applied by a peasant owner working unaided on his own land, or at the charges of a capitalist farmer who does no manual labour himself. But in the latter case the main body of the outlay presents itself in the form of money; and when discussing the business economy of farming in relation to English conditions, it is often convenient to consider the labour converted at its market value into a money equivalent, and to speak of doses of capital simply, rather than of doses of labour and capital.


The dose which only just remunerates the cultivator may be said to be the marginal dose, and the return to it the marginal return. If there happens to be in the neighbourhood land that is cultivated but only just pays its expenses, and so gives no surplus for rent we may suppose this dose applied to it. We can then say that the dose applied to it is applied to land on the margin of cultivation, and this way of speaking has the advantage of simplicity. But it is not necessary for the argument to suppose that there is any such land: what we want to fix our minds on is the return to the marginal dose; whether it happens to be applied to poor land or to rich does not matter; all that is necessary is that it should be the last dose which can profitably be applied to that land*11.


When we speak of the marginal, or the "last" dose applied to the land, we do not mean the last in time, we mean that dose which is on the margin of profitable expenditure; that is, which is applied so as just to give the ordinary returns to the capital and labour of the cultivator, without affording any surplus. To take a concrete instance, we may suppose a farmer to be thinking of sending the hoers over a field once more; and after a little hesitation he decides that it is worth his while, but only just worth his while to do it. The dose of capital and labour spent on doing it, is then the last dose in our present sense, though there are many doses still to be applied in reaping the crop. Of course the return to this last dose cannot be separated from the others; but 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 hoeing*12.


Since the return to the dose on the margin of cultivation just remunerates the cultivator, it follows that he will be just remunerated for the whole of his capital and labour by as many times the marginal return as he has applied doses in all. Whatever he gets in excess of this is the surplus produce of the land. This surplus is retained by the cultivator if he owns the land himself*13.


It is important to note that this description of the nature of surplus produce is not a theory of rent: we shall not be ready for that till a much later stage. All that can be said here, is that this surplus produce may, under certain conditions, become the rent which the owner of the land can exact from the tenant for its use. But, as we shall see hereafter, the full rent of a farm in an old country is made up of three elements: the first being due to the value of the soil as it was made by nature; the second to improvements made in it by man; and the third, which is often the most important of all, to the growth of a dense and rich population, and to facilities of communication by public roads, railroads, etc.


It is to be noted also that in an old country it is impossible to discover what was the original state of the land before it was first cultivated. The results of some of man's work are for good and evil fixed in the land, and cannot be distinguished from those of nature's work: the line of division is blurred, and must be drawn more or less arbitrarily. But for most purposes it is best to regard the first difficulties of coping with nature as pretty well conquered before we begin to reckon the farmer's cultivation. Thus the returns that we count as due to the first doses of capital and labour are generally the largest of all, and the tendency of the return to diminish shows itself at once. Having English agriculture chiefly in view, we may fairly take, as Ricardo did, this as the typical case*14.


§ 3. Let us next inquire on what depends the rate of diminution or of increase of the returns to successive doses of capital and labour. We have seen that there are great variations in the share of the produce which man may claim as the additional result of his own work over what unaided nature would have produced; and that man's share is much larger with some crops and soils and methods of cultivation than with others. Thus broadly speaking it increases as we pass from forest to pasture land, from pasture to arable, and from plough land to spade land; and this is because the rate of diminution of the return is as a rule greatest in forests, rather less in pasture, still less in arable land, and least of all in spade land.


There is no absolute measure of the richness or fertility of land. Even if there be no change in the arts of production, a mere increase in the demand for produce may invert the order in which two adjacent pieces of land rank as regards fertility. The one which gives the smaller produce, when both are uncultivated, or when the cultivation of both is equally slight, may rise above the other and justly rank as the more fertile when both are cultivated with equal thoroughness. In other words, many of those lands which are the least fertile when cultivation is merely extensive, become among the most fertile when cultivation is intensive. For instance, self-drained pasture land may give a return large in proportion to a very slight expenditure of capital and labour, but a rapidly diminishing return to further expenditure: as population increases it may gradually become profitable to break up some of the pasture and introduce a mixed cultivation of roots and grains and grasses; and then the return to further doses of capital and labour may diminish less quickly.


Other land makes poor pasture, but will give more or less liberal returns to a great deal of capital and labour applied in tilling and in manuring it; its returns to the early doses are not very high, but they diminish slowly.


Again, other land is marshy. It may, as did the fens of east England, produce little but osiers and wild fowl. Or, as is the case in many tropical districts, it may be prolific of vegetation, but so shrouded with malaria that it is difficult for man to live there, and still more to work there. In such cases the returns to capital and labour are at first small, but as drainage progresses, they increase; afterwards perhaps they again fall off*15.


But when improvements of this kind have once been made, the capital invested in the soil cannot be removed; the early history of the cultivation is not repeated; and the produce due to further applications of capital and labour shows a tendency to diminishing return*16.


Similar though less conspicuous changes may occur on land already well cultivated. For instance, without being marshy, it may be in need of a little drainage to take off the stagnant water from it, and to enable fresh water and air to stream through it. Or the subsoil may happen to be naturally richer than the soil at the surface: or again, though not itself rich, it may have just those properties in which the surface soil is deficient, and then a thorough system of deep steam-ploughing may permanently change the character of the land.


Thus we need not suppose that when the return to extra capital and labour has begun to diminish, it will always continue to do so. Improvements in the arts of production may, it has always been understood, raise generally the return which can be got by any amount of capital and labour; but this is not what is meant here. The point is that, independently of any increase in his knowledge, and using only those methods with which he has long been familiar, a farmer finding extra capital and labour at his command, may sometimes obtain an increasing return even at a late stage in his cultivation*17.


It has been well said that as the strength of a chain is that of its weakest link, so fertility is limited by that element in which it is most deficient. Those who are in a hurry, will reject a chain which has one or two very weak links, however strong the rest may be: and prefer to it a much slighter chain that has no flaw. But if there is heavy work to be done, and they have time to make repairs, they will set the larger chain in order, and then its strength will exceed that of the other. In this we find the explanation of much that is apparently strange in agricultural history.


The first settlers in a new country generally avoid land which does not lend itself to immediate cultivation. They are often repelled by the very luxuriance of natural vegetation, if it happens to be of a kind that they do not want. They do not care to plough land that is at all heavy, however rich it might become if thoroughly worked. They will have nothing to do with water-logged land. They generally select light land which can easily be worked with a double plough, and then they sow their seed broadly, so that the plants when they grow up may have plenty of light and air, and may collect their food from a wide area.


When America was first settled, many farming operations that are now done by horse machinery were still done by hand; and though now the farmers have a strong preference for flat prairie land, free from stumps and stones, where their machines can work easily and without risk, they had then no great objection to a hill-side. Their crops were light in proportion to their acreage, but heavy in proportion to the capital and labour expended in raising them.


We cannot then call one piece of land more fertile than another until we know something about the skill and enterprise of its cultivators, and the amount of capital and labour at their disposal; and till we know whether the demand for produce is such as to make intensive cultivation profitable with the resources at their disposal. If it is, those lands will be the most fertile which give the highest average returns to a large expenditure of capital and labour; but if not, those will be the most fertile which give the best returns to the first few doses. The term fertility has no meaning except with reference to the special circumstances of a particular time and place.


But even when so limited there is some uncertainty as to the usage of the term. Sometimes attention is directed chiefly to the power which land has of giving adequate returns to intensive cultivation and so bearing a large total produce per acre; and sometimes to its power of yielding a large surplus produce or rent, even though its gross produce is not very large: thus in England now rich arable land is very fertile in the former sense, rich meadow in the latter. For many purposes it does not matter which of these senses of the term is understood: in the few cases in which it does matter, an interpretation clause must be supplied in the context*18.


§ 4. But further, the order of fertility of different soils is liable to be changed by changes in the methods of cultivation and in the relative values of different crops. Thus when at the end of last century Mr Coke showed how to grow wheat well on light soils by preparing the way with clover, they rose relatively to clay soils; and now though they are still sometimes called from old custom "poor," some of them have a higher value, and are really more fertile, than much of the land that used to be carefully cultivated while they were left in a state of nature.


Again, the increasing demand in central Europe for wood to be used as fuel and for building purposes, has raised the value of the pine-covered mountain slopes relatively to almost every other kind of land. But in England this rise has been prevented by the substitution of coal for wood as fuel, and of iron for wood as a material for ship-building, and lastly by England's special facilities for importing wood. Again, the cultivation of rice and jute often gives a very high value to lands that are too much covered with water to bear most other crops. And again, since the repeal of the Corn Laws the prices of meat and dairy produce have risen in England relatively to that of corn. Those arable soils that would grow rich forage crops in rotation with corn, rose relatively to the cold clay soils; and permanent pasture recovered part of that great fall in value relatively to arable land, which had resulted from the growth of population*19.


Independently of any change in the suitability of the prevailing crops and methods of cultivation for special soils, there is a constant tendency towards equality in the value of different soils. In the absence of any special cause to the contrary, the growth of population and wealth will make the poorer soils gain on the richer. Land that was at one time entirely neglected is made by much labour to raise rich crops; its annual income of light and heat and air, is probably as good as those of richer soils: while its faults can be much lessened by labour*20.


As there is no absolute standard for fertility, so there is none of good cultivation. The best cultivation in the richest parts of the Channel Islands, for instance, involves a lavish expenditure of capital and labour on each acre: for they are near good markets and have a monopoly of an equable and early climate. If left to nature the land would not be very fertile, for though it has many virtues, it has two weak links (being deficient in phosphoric acid and potash). But, partly by the aid of the abundant seaweed on its shores, these links can be strengthened, and the chain thus becomes exceptionally strong. Intense, or as it is ordinarily called in England "good" cultivation, will thus raise £100 worth of early potatoes from a single acre. But an equal expenditure per acre by the farmer in Western America would ruin him; relatively to his circumstances it would not be good, but bad cultivation.


§ 5. Ricardo's wording of the law of diminishing return was inexact. It is however probable that the inaccuracy was due not to careless thinking but only to careless writing. In any case he would have been justified in thinking that these conditions were not of great importance in the peculiar circumstances of England at the time at which he wrote, and for the special purposes of the particular practical problems he had in view. Of course he could not anticipate the great series of inventions which were about to open up new sources of supply, and, with the aid of free trade, to revolutionize English agriculture; but the agricultural history of England and other countries might have led him to lay greater stress on the probability of a change*21.


He stated that the first settlers in a new country invariably chose the richest lands, and that as population increased, poorer and poorer soils were gradually brought under cultivation, speaking carelessly as though there were an absolute standard of fertility. But as we have already seen, where land is free, everyone chooses that which is best adapted for his own purpose, and that which will give him, all things considered, the best return for his capital and labour. He looks out, therefore, for land that can be cultivated at once, and passes by land that has any weak links in the chain of its elements of fertility, however strong it may be in some other links. But besides having to avoid malaria, he must think of his communication with his markets and the base of his resources; and in some cases the need for security against the attacks of enemies and wild beasts outweighs all other considerations. It is therefore not to be expected that the lands which were first chosen, should turn out always to be those which ultimately come to be regarded as the most fertile. Ricardo did not consider this point, and thus laid himself open to attacks by Carey and others, which, though for the greater part based on a misinterpretation of his position, have yet some solid substance in them.


The fact that, in new countries, soils which an English farmer would regard as poor, are sometimes cultivated before neighbouring soils which he would regard as rich, is not inconsistent, as some foreign writers have supposed, with the general tenor of Ricardo's doctrines. Its practical importance is in relation to the conditions under which the growth of population tends to cause increased pressure on the means of subsistence: it shifts the centre of interest from the mere amount of the farmer's produce to its exchange value in terms of the things which the industrial population in his neighbourhood will offer for it*22.


§ 6. Ricardo, and the economists of his time generally were too hasty in deducing this inference from the law of diminishing return; and they did not allow enough for the increase of strength that comes from organization. But in fact every farmer is aided by the presence of neighbours whether agriculturists or townspeople*23. Even if most of them are engaged like himself in agriculture, they gradually supply him with good roads, and other means of communication: they give him a market in which he can buy at reasonable terms what he wants, necessaries, comforts and luxuries for himself and his family, and all the various requisites for his farm work: they surround him with knowledge: medical aid, instruction and amusement are brought to his door; his mind becomes wider, and his efficiency is in many ways increased. And if the neighbouring market town expands into a large industrial centre, his gain is much greater. All his produce is worth more; some things which he used to throw away fetch a good price. He finds new openings in dairy farming and market gardening, and with a larger range of produce he makes use of rotations that keep his land always active without denuding it of any one of the elements that are necessary for its fertility.


Further, as we shall see later on, an increase of population tends to develop the organization of trade and industry; and therefore the law of diminishing return does not apply to the total capital and labour spent in a district as sharply as to that on a single farm. Even when cultivation has reached a stage after which each successive dose applied to a field would get a less return than the preceding dose, it may be possible for an increase in the population to cause a more than proportional increase in the means of subsistence. It is true that the evil day is only deferred: but it is deferred. The growth of population, if not checked by other causes, must ultimately be checked by the difficulty of obtaining raw produce; but in spite of the law of diminishing return, the pressure of population on the means of subsistence may be restrained for a long time to come by the opening up of new fields of supply, by the cheapening of railway and steamship communication, and by the growth of organization and knowledge.


Against this must be set the growing difficulty of getting fresh air and light, and in some cases fresh water, in densely peopled places. The natural beauties of a place of fashionable resort have a direct money value which cannot be overlooked; but it requires some effort to realize the true value to men, women and children of being able to stroll amid beautiful and varied scenery.


§ 7. As has already been said the land in economic phrase includes rivers and the sea. In river-fisheries, the extra return to additional applications of capital and labour shows a rapid diminution. As to the sea, opinions differ. Its volume is vast, and fish are very prolific; and some think that a practically unlimited supply can be drawn from the sea by man without appreciably affecting the numbers that remain there; or in other words, that the law of diminishing return scarcely applies at all to sea-fisheries: while others think that experience shows a falling-off in the productiveness of those fisheries that have been vigorously worked, especially by steam trawlers. The question is important, for the future population of the world will be appreciably affected as regards both quantity and quality, by the available supply of fish.


The produce of mines again, among which may be reckoned quarries and brickfields, is said to conform to the law of diminishing return; but this statement is misleading. It is true that we find continually increasing difficulty in obtaining a further supply of minerals, except in so far as we obtain increased power over nature's stores through improvements in the arts of mining, and through better knowledge of the contents of the earth's crust; and there is no doubt that, other things being equal, the continued application of capital and labour to mines will result in a diminishing rate of yield. But this yield is not a net yield, like the return of which we speak in the law of diminishing return. That return is part of a constantly recurring income, while the produce of mines is merely a giving up of their stored-up treasures. The produce of the field is something other than the soil; for the field, properly cultivated, retains its fertility. But the produce of the mine is part of the mine itself.


To put the same thing in another way, the supply of agricultural produce and of fish is a perennial stream; mines are as it were nature's reservoir. The more nearly a reservoir is exhausted, the greater is the labour of pumping from it; but if one man could pump it out in ten days, ten men could pump it out in one day: and when once empty, it would yield no more. So the mines that are being opened this year might just as easily have been opened many years ago: if the plans had been properly laid in advance, and the requisite specialized capital and skill got ready for the work, ten years' supply of coal might have been raised in one year without any increased difficulty; and when a vein had once given up its treasure, it could produce no more. This difference is illustrated by the fact that the rent of a mine is calculated on a different principle from that of a farm. The farmer contracts to give back the land as rich as he found it: a mining company cannot do this; and while the farmer's rent is reckoned by the year, mining rent consists chiefly of "royalties" which are levied in proportion to the stores that are taken out of nature's storehouse*24.


On the other hand, services which land renders to man, in giving him space and light and air in which to live and work, do conform strictly to the law of diminishing return. It is advantageous to apply a constantly increasing capital to land that has any special advantages of situation, natural or acquired. Buildings tower up towards the sky; natural light and ventilation are supplemented by artificial means, and the steam lift reduces the disadvantages of the highest floors; and for this expenditure there is a return of extra convenience, but it is a diminishing return. However great the ground rent may be, a limit is at last reached after which it is better to pay more ground rent for a larger area than to go on piling up storey on storey any further; just as the farmer finds that at last a stage is reached at which more intensive cultivation will not pay its expenses, and it is better to pay more rent for extra land, than to face the diminution in the return which he would get by applying more capital and labour to his old land*25. From this it results that the theory of ground rents is substantially the same as that of farm rents. This and similar facts will presently enable us to simplify and extend the theory of value as given by Ricardo and Mill.


And what is true of building land is true of many other things. If a manufacturer has, say, three planing machines there is a certain amount of work which he can get out of them easily. If he wants to get more work from them he must laboriously economize every minute of their time during the ordinary hours, and perhaps work overtime. Thus after they are once well employed, every successive application of effort to them brings him a diminishing return. At last the net return is so small that he finds it cheaper to buy a fourth machine than to force so much work out of his old machines: just as a farmer who has already cultivated his land highly finds it cheaper to take in more land than to force more produce from his present land. Indeed there are points of view from which the income derived from machinery partakes of the nature of rent: as will be shown in Book V.

Note on the Law of Diminishing Return.


§ 8. The elasticity of the notion of diminishing return cannot be fully considered here; for it is but an important detail of that large general problem of the economic distribution of resources in the investment of capital, which is the pivot of the main argument of Book V. and indeed of a great part of the whole Volume. But a few words about it seem now to be called for in this place, because much stress has recently been laid on it under the able and suggestive leadership of Professor Carver*26.


If a manufacturer expends an inappropriately large amount of his resources on machinery, so that a considerable part of it is habitually idle; or on buildings, so that a considerable part of his space is not well filled; or on his office staff, so that he has to employ some of them on work that it is not worth what it costs; then his excessive expenditure in that particular direction will not be as remunerative as his previous expenditure had been: and it may be said to yield him a "diminishing return." But this use of the phrase, though strictly correct is apt to mislead unless used with caution. For when the tendency to a diminishing return from increased labour and capital applied to land is regarded as a special instance of the general tendency to diminishing return from any agent of production, applied in excessive proportion to the other agents, one is apt to take it for granted that the supply of the other factors can be increased. That is to say, one is apt to deny the existence of that condition—the fixedness of the whole stock of cultivable land in an old country—which was the main foundation of those great classical discussions of the law of diminishing return, which we have just been considering. Even the individual farmer may not always be able to get an additional ten or fifty acres adjoining his own farm, just when he wants them, save at a prohibitive price. And in that respect land differs from most other agents of production even from the individual point of view. This difference may indeed be regarded as of little account in regard to the individual farmer. But from the social point of view, from the point of view of the following chapters on population it is vital. Let us look into this.


In every phase of any branch of production there is some distribution of resources between various expenditures which yields a better result than any other. The abler the man in control of any business, the nearer he will approach to the ideally perfect distribution; just as the abler the primitive housewife in control of a family's stock of wool, the nearer she will approach to an ideal distribution of wool between the different needs of the family*27.


If his business extends he will extend his uses of each requisite of production in due proportion; but not, as has sometimes been said, proportionately; for instance the proportion of manual work to machine work, which would be appropriate in a small furniture factory would not be appropriate in a large one. If he makes the best possible apportionment of his resources, he gets the greatest (marginal) return from each appliance of production of which his business is capable. If he uses too much of any one he gets a diminishing return from it; because the others are not able to back it up properly. And this diminishing return is analogous to that which a farmer obtains, when he cultivates land so intensively that he obtains a diminishing return from it. If the farmer can get more land at the same rent as he has paid for the old, he will take more land, or else lie open to the imputation of being a bad business man: and this illustrates the fact that land from the point of view of the individual cultivator is simply one form of capital.


But when the older economists spoke of the Law of Diminishing Return they were looking at the problems of agriculture not only from the point of view of the individual cultivator but also from that of the nation as a whole. Now if the nation as a whole finds its stock of planing machines or ploughs inappropriately large or inappropriately small, it can redistribute its resources. It can obtain more of that in which it is deficient, while gradually lessening its stock of such things as are superabundant: but it cannot do that in regard to land: it can cultivate its land more intensively, but it cannot get any more. And for that reason the older economists rightly insisted that, from the social point of view, land is not on exactly the same footing as those implements of production which man can increase without limit.


No doubt in a new country where there is an abundance of rich land not yet brought under cultivation, this fixedness of the total stock of land is not operative. American economists often speak of the value, or rent, of land as varying with the land's distance from good markets, rather than with its fertility; because even now there is a great deal of rich land in their country which is not fully cultivated. And in like manner they lay but little stress on the fact that the diminishing return to labour and capital in general applied to the land by discreet farmers, in such a country as England, is not exactly on the same footing as the diminishing return to an inappropriate investment of their resources by indiscreet farmers or manufacturers in a disproportionately large number of ploughs or planing machines.


It is true that when the tendency to diminishing return is generalized, the return is apt to be expressed in terms of value, and not of quantity. It must however be conceded that the older method of measuring return in terms of quantity often jostled against the difficulty of rightly interpreting a dose of labour and capital without the aid of a money measure: and that, though helpful for a broad preliminary survey, it cannot be carried very far.


But even the recourse to money fails us, if we want to bring to a common standard the productiveness of lands in distant times or places; and we must then fall back on rough, and more or less arbitrary modes of measurement, which make no aim at numerical precision, but will yet suffice for the broader purposes of history. We have to take account of the facts that there are great variations in the relative amounts of labour and capital in a dose: and that interest on capital is generally a much less important item in backward than in advanced stages of agriculture, in spite of the fact that the rate of interest is generally much lower in the latter. For most purposes it is probably best to take as a common standard a day's unskilled labour of given efficiency: we thus regard the dose as made up of so much labour of different kinds, and such charges for the use and replacement of capital, as will together make up the value of, say, ten days' such labour; the relative proportions of these elements and their several values in terms of such labour being fixed according to the special circumstances of each problem*28.


A similar difficulty is found in comparing the returns obtained by labour and capital applied under different circumstances. So long as the crops are of the same kind, the quantity of one return can be measured off against that of another: but, when they are of different kinds, they cannot be compared till they are reduced to a common measure of value. When, for instance, it is said that land would give better returns to the capital and labour expended on it with one crop or rotation of crops than with another, the statement must be understood to hold only on the basis of the prices at the time. In such a case we must take the whole period of rotation together, assuming the land to be in the same condition at the beginning and the end of the rotation; and counting on the one hand all the labour and capital applied during the whole period, and on the other the aggregate returns of all the crops.


It must be remembered that the return due to a dose of labour and capital is not here taken to include the value of the capital itself. For instance, if part of the capital on a farm consists of two-year-old oxen, then the returns to a year's labour and capital will include not the full weight of these oxen at the end of the year, but only the addition that has been made to it during the year. Again, when a farmer is said to work with a capital of £10 to the acre, this includes the value of everything that he has on the farm; but the total volume of the doses of labour and capital applied to a farm during, say, a year, does not include the whole value of the fixed capital, such as machinery and horses, but only the value of their use after allowing for interest, depreciation and repairs; though it does include the whole value of the circulating capital, such as seed.


The above is the method of measuring capital generally adopted, and it is to be taken for granted if nothing is said to the contrary; but another method is more suitable occasionally. Sometimes it is convenient to speak as though all the capital applied were circulating capital applied at the beginning of the year or during it: and in that case everything that is on the farm at the end of the year is part of the produce. Thus, young cattle are regarded as a sort of raw material which is worked up in the course of time into fat cattle ready for the butcher. The farm implements may even be treated in the same way, their value at the beginning of the year being taken as so much circulating capital applied to the farm, and at the end of the year as so much produce. This plan enables us to avoid a good deal of repetition of conditioning clauses as to depreciation, etc., and to save the use of words in many ways. It is often the best plan for general reasonings of an abstract character, particularly if they are expressed in a mathematical form.


The law of diminishing return must have occupied thoughtful men in every densely peopled country. It was first stated clearly by Turgot (Œuvres, ed. Daire I. pp. 420, 1), as Prof. Cannan has shown; and its chief applications were developed by Ricardo.

Notes for this chapter

Increasing return in the earlier stages arises partly from economy of organization, similar to that which gives an advantage to manufacture on a large scale. But it is also partly due to the fact that where land is very slightly cultivated the farmer's crops are apt to be smothered by nature's crops of weeds. The relation between Diminishing and Increasing Return is discussed further in the last chapter of this Book.
"The land was not able to bear them, that they might dwell together: for their substance was great, so that they could not dwell together." Genesis xiii. 6.
As to this term see the Note at the end of the chapter.
Ricardo was well aware of this: though he did not emphasize it enough. Those opponents of his doctrine who have supposed that it has no application to places where all the land pays a rent, have mistaken the nature of his argument.
An illustration from recorded experiments may help to make clearer the notion of the return to a marginal dose of capital and labour. The Arkansas experimental station (see The Times, 18 Nov. 1889) reported that four plots of an acre each were treated exactly alike except in the matter of ploughing and harrowing, with the following result:—

Plot. Cultivation. Crop yields bushels per acre.
1 Ploughed once 16
2 Ploughed once and harrowed once 18½
3 Ploughed twice and harrowed once 21 2/3
4 Ploughed twice and harrowed twice 23¼

This would show that the dose of capital and labour applied in harrowing a second time an acre which had already been ploughed twice gave a return of 1 7/12 bushels. And if the value of these bushels, after allowing for expenses of harvesting, etc. just replaced that dose with profits, then that dose was a marginal one; even though it was not the last in point of time, since those spent on harvesting must needs come later.

Let us seek a graphical illustration. It is to be remembered that graphical illustrations are not proofs. They are merely pictures corresponding very roughly to the main conditions of certain real problems. They obtain clearness of outline, by leaving out of account many considerations which vary from one practical problem to another, and of which the farmer must take full account in his own special case. If on any given field there were expended a capital of £50, a certain amount of produce would be raised from it: a certain amount larger than the former would be raised if there were expended on it a capital of £51. The difference between these two amounts may be regarded as the produce due to the fifty-first pound; and if we suppose the capital to be applied in successive doses of £1 each we may speak of this difference as the produce due to the fifty-first dose. Figure 11.  Click to enlarge in new window.Let the doses be represented in order by successive equal divisions of the line OD. Let there now be drawn from the division of this line representing the fifty-first dose M, a line MP at right angles to OD, in thickness equal to the length of one of the divisions, and such that its length represents the amount of the produce due to the fifty-first dose. Suppose this done for each separate division up to that corresponding to the last dose which it is found profitable to put on the land. Let this last dose be the 110th at D, and DC the corresponding return that only just remunerates the farmer. The extremities of such lines will lie on a curve APC. The gross produce will be represented by the sum of these lines: i.e., since the thickness of each line is equal to the length of the division on which it stands, by the area ODCA. Let CGH be drawn parallel to DO, cutting PM in G; then MG is equal to CD; and since DC just remunerates the farmer for one dose, MG will just remunerate him for another: and so for all the portions of the thick vertical lines cut off between OD and HC. Therefore the sum of these, that is, the area ODCH, represents the share of the produce that is required to remunerate him; while the remainder, AHGCPA, is the surplus produce, which under certain conditions becomes the rent.
That is, we may substitute (fig. 11) the dotted line BA' for BA and regard A'BPC as the typical curve for the return to capital and labour applied in English agriculture. No doubt crops of wheat and some other annuals cannot be raised at all without some considerable labour. But natural grasses which sow themselves will yield a good return of rough cattle to scarcely any labour.

It has already been noticed (Book III. ch. III. § 1), the law of diminishing return bears a close analogy to the law of demand. The return which land gives to a dose of capital and labour may be regarded as the price which land offers for that dose. Land's return to capital and labour is, so to speak, her effective demand for them: her return to any dose is her demand price for that dose, and the list of returns that she will give to successive doses may thus be regarded as her demand schedule: but to avoid confusion we shall call it her "Return Schedule." Corresponding to the case of the land in the text is that of a man who may be willing to pay a larger proportionate price for a paper that would cover the whole of the walls of his room than for one that would go only half way; and then his demand schedule would at one stage show an increase and not a diminution of demand price for an increased quantity. But in the aggregate demand of many individuals these unevennesses destroy one another; so that the aggregate demand schedule of a group of people always shows the demand price as falling steadily with every increase in the amount offered. In the same way, by grouping together many pieces of land we might obtain a return schedule that would show a constant diminution for every increase of capital and labour applied. But it is more easy to ascertain, and in some ways more important to take note of, the variations of individual demand in the case of plots of land than in the case of people. And therefore our typical return schedule is not drawn out so as to show as even and uniform a diminution of return as our typical demand schedule does of demand price.

This case may be represented by diagrams. If the produce rises in real value in the ratio of OH' to OH (so that the amount required to remunerate the farmer for a dose of capital and labour has fallen from OH to OH'), the surplus produce rises only to AH'C', which is not very much greater than its old amount AHC, fig. 12, representing the first case.

Figures 12, 13, and 14.  Click to enlarge in new window.

The second case is represented in fig. 13, where a similar change in the price of produce makes the new surplus produce AH'C' about three times as large as the old surplus, AHC; and the third in fig. 14. The earliest doses of capital and labour applied to the land give so poor a return, that it would not be worth while to apply them unless it were intended to carry the cultivation further. But later doses give an increasing return which culminates at P, and afterwards diminishes. If the price to be got for produce is so low that an amount OH" is required to remunerate the cultivator for a dose of capital and labour, it will then be only just profitable to cultivate the land. For then cultivation will be carried as far as D"; there will be a deficit on the earlier doses represented by the area H"AE", and a surplus on the later doses represented by the area E"PC": and as these two are about equal, the cultivation of the land so far will only just pay its way. But if the price of produce rises till OH is sufficient to remunerate the cultivator for a dose of capital and labour, the deficit on the earlier doses will sink to HAE, and the surplus on the later doses will rise to EPC: the net surplus (the true rent in case the land is hired out) will be the excess of EPC over HAE. Should the price rise further till OH' is sufficient to remunerate the cultivator for a dose of capital and labour, this net surplus will rise to the very large amount represented by the excess of E'PC' over H'AE'.

In such a case as this the earlier doses are pretty sure to be sunk in the land; and the actual rent paid, if the land is hired out, will then include profits on them in addition to the surplus produce or true rent thus shown. Provision can easily be made in the diagrams for the returns due to the landlord's capital.
Figure 15.  Click to enlarge in new window.Of course his return may diminish and then increase and then diminish again; and yet again increase when he is in a position to carry out some further extensive change, as was represented by fig. 11. But more extreme instances, of the kind represented by fig. 15, are not very rare.
If the price of produce is such that an amount of it OH (figs. 12, 13, 14) is required to pay the cultivator for one dose of capital and labour, the cultivation will be carried as far as D; and the produce raised, AODC, will be greatest in fig. 12, next greatest in fig. 13, and least in fig. 14. But if the demand for agricultural produce so rises that OH' is enough to repay the cultivator for a dose, the cultivation will be carried as far as D', and the produce raised will be AOD'C', which is greatest in fig. 14, next in fig. 13, and least in fig. 12. The contrast would have been even stronger if we had considered the surplus produce which remains after deducting what is sufficient to repay the cultivator, and which becomes under some conditions the rent of the land. For this is AHCin figs. 12 and 13 in the first case and AH'C' in the second; while in fig. 14 it is in the first case the excess of AODCPA over ODCH, i.e. the excess of PEC over AHE; and in the second case the excess of PE'C' over AH'E'.
Rogers (Six Centuries of Work and Wages, p. 73) calculates that rich meadow had about the same value, estimated in grain, five or six centuries ago as it has now; but that the value of arable land, similarly estimated, has increased about fivefold in the same time. This is partly due to the great importance of hay at a time when roots and other modern kinds of winter food for cattle were unknown.
Thus we may compare two pieces of land represented in figs. 16 and 17, with regard to which the law of diminishing return acts in a similar way, so that their produce curves have similar shapes, but the former has a higher fertility than the other for all degrees of intensity of cultivation. Figures 16 and 17.  Click to enlarge in new window.The value of the land may generally be represented by its surplus produce or rent, which is in each case represented by AHC when OH is required to repay a dose of capital and labour; and by AH'C' when the growth of numbers and wealth have made OH' sufficient. It is clear that AH'C' in fig. 17 bears a more favourable comparison with AH'C' in fig. 16 than does AHC in fig. 17 with AHC in fig. 16. In the same way, though not to the same extent, the total produce AOD'C' in fig. 17 bears a more favourable comparison with AOD'C' in fig. 16, than does AODC in fig. 17 with AODC in fig. 16. (It is ingeniously argued in Wicksteed's Coordinates of Laws of Distribution, pp. 51, 2 that rent may be negative. Of course taxes may absorb rent: but land which will not reward the plough will grow trees or rough grass. See above, pp. 157, 8.)

Leroy Beaulieu (Répartition des Richesses, chap. II.) has collected several facts illustrating this tendency of poor lands to rise in value relatively to rich. He quotes the following figures, showing the rental in francs per hectare (2½ acres) of five classes of land in several communes of the Départements de l'Eure et de l'Oise in 1829 and 1852 respectively:—

Class I. Class II. Class III. Class IV. Class V.
A.D. 1829 58 48 34 20 8
A.D. 1852 80 78 60 50 40

As Roscher says (Political Economy, Sect. CLV.), "In judging Ricardo, it must not be forgotten that it was not his intention to write a text-book on the science of Political Economy, but only to communicate to those versed in it the result of his researches in as brief a manner as possible. Hence he writes so frequently making certain assumptions, and his words are to be extended to other cases only after due consideration, or rather re-written to suit the changed case."
Carey claims to have proved that "in every quarter of the world cultivation has commenced on the sides of the hills where the soil was poorest, and where the natural advantages of situation were the least. With the growth of wealth and population, men have been seen descending from the high lands bounding the valley on either side, and coming together at its feet." (Principles of Social Science, chap. IV. § 4.) He has even argued that whenever a thickly peopled country is laid waste, "whenever population, wealth, and the power of association decline, it is the rich soil that is abandoned by men who fly again to the poor ones" (Ib. ch. V. § 3); the rich soils being rendered difficult and dangerous by the rapid growth of jungles which harbour wild beasts and banditti, and perhaps by malaria. The experience of more recent settlers in South Africa and elsewhere does not however generally support his conclusions, which are indeed based largely on facts relating to warm countries. But much of the apparent attractiveness of tropical countries is delusive: they would give a very rich return to hard work: but hard work in them is impossible at present, though some change in this respect may be made by the progress of medical and especially bacteriological science. A cool refreshing breeze is as much a necessary of vigorous life as food itself. Land that offers plenty of food but whose climate destroys energy, is not more productive of the raw material of human wellbeing, than land that supplies less food but has an invigorating climate.

The late Duke of Argyll described the influence of insecurity and poverty in compelling the cultivation of the hills before that of the valleys of the Highlands was feasible, Scotland as it is and was, II. 74, 5.

In a new country an important form of this assistance is to enable him to venture on rich land that he would have otherwise shunned, through fear of enemies or of malaria.
As Ricardo says (Principles, chap. II.) "The compensation given (by the lessee) for the mine or quarry is paid for the value of the coal or stone which can be removed from them, and has no connection with the original or indestructible powers of the land." But both he and others seem sometimes to lose sight of these distinctions in discussing the law of diminishing return in its application to mines. Especially is this the case in Ricardo's criticism of Adam Smith's theory of rent (Principles, chap. XXIV.).
Of course the return to capital spent in building increases for the earlier doses. Even where land can be had for almost nothing, it is cheaper to build houses two stories high than one; and hitherto it has been thought cheapest to build factories about four stories high. But a belief is growing up in America, that where land is not very dear factories should be only two stories high, partly in order to avoid the evil effects of vibration, and of the expensive foundations and walls required to prevent it in a high building: that is, it is found that the return of accommodation diminishes perceptibly after the capital and labour required to raise two stories have been spent on the land.
See also the writings of Professors Bullock and Landry.
In this he will make large use of what is called below the "Substitution" of more for less appropriate means. Discussions bearing directly on this paragraph will be found in III. V. 1-3: IV. VII. 8; and XIII. 2: V. III. 3; IV. 1-4; V. 6-8; VIII. 1-5; X. 3: VI. I. 7; and II. 5.

The tendencies of diminishing utility and of diminishing return have their roots, the one in qualities of human nature, the other in the technical conditions of industry. But the distributions of resources, to which they point, are governed by exactly similar laws. In Mathematical phrase, the problems in maxima and minima to which they give rise are expressed by the same general equations; as may be seen by reference to Mathematical Note XIV.

The labour-part of the dose is of course current agricultural labour; the capital-part is itself also the product of labour in past times rendered by workers of many kinds and degrees, accompanied by "waiting."

End of Notes

20 of 71

Return to top