To-day there exist large areas of land which may be had for nothing. Of this land some is cultivated which pays no rent; some is not cultivated at all. Why, then, is it that some land will bear rent under such circumstances? Obviously, because that land is more desirable than the land which may be had for nothing. And how much rent will it bear ? It is equally obvious that assuming the cultivators to be of equal degrees of efficiency, the rent will be measured by the difference in desirability. Let us illustrate this by a diagram.

Suppose the above to represent all land, arranged in seven groups according to desirability, each small parallelogram representing four bushels of product per acre in excess of the amount of the product necessary to pay for marketing the crop. Then the first group, deducting enough of the product to pay for transportation, will put in the market 28 bushels of wheat per acre; the second, 24; the others, 20, 16, 12, 8, and 4, respectively. Now if the people are few and need a small part of the land, they may cultivate only A, or the most desirable land. As long as there is enough of this land, if it is of equal desirability, there will be no rent, for no man will pay rent for what he can get for nothing. But the time may come, with increasing population, when more land is needed, and cultivation is driven to B. Land is still free there, but all of group A has now been appropriated. If, then, any man insists upon cultivating land which belongs to an owner in group A, he must pay for the privilege. How much must he pay ? Evidently it is 4 bushels per acre, since in group A he can produce 28 bushels per acre, while in group B he can produce only 24 bushels. The land in B, which is free land, is now the extensive margin of cultivation; that is, the grade of land which will just pay for cultivation and no more. The normal reward to labor in agriculture is the total return to cultivation on this margin of cultivation, after deducting a return for the capital invested in the work. The surplus product from the superior land in other words, the advantage which owners of land in A have over the tillers of the free land is rent. And it is rent whether the owners of the land in A work the land themselves or lease it to others. The money expression of the rent in this case is the market price of the 4 bushels of wheat that represent the differential advantage in production possessed by workers on the more desirable land.

If population increases still further, without any improvement in the arts of production or of consumption, the margin of cultivation will in time descend to land in group C, where the number of bushels produced by a given amount of labor and capital is less than before. Land in B will now return a rent of 4 bushels per acre, while land in group A will return a rent of 8 bushels per acre. If the margin of cultivation is later forced down to E, then rents on land in B will equal one-half, and on land in A will equal four-sevenths, of the entire product of such land.

Intensive Cultivation. With the figure in mind, let us place ourselves again at the point where all the A land is taken and men are beginning to seek new means of production. We have assumed that they will take up new land in group B. This is not the only possibility, however. It is probable that land in A may be made to produce more than it has, if the amount of labor and capital expended upon it is increased. In other words, it will be possible to cultivate the old land more intensively at a profit. Suppose that ten men formerly cultivated 100 acres of A land, raising 2800 bushels of wheat, and that now eleven men put their labor upon the 100 acres. It may be that the 100 acres will now produce 3060 bushels, in which case it is evident that the labor of the eleventh man has made a difference of 260 bushels. The 2800 bushels raised by the ten men meant 280 bushels of crop per man. In accordance with the law of diminishing returns, the eleventh man does not increase the output proportionately, but he is still producing two bushels more than he would if he were to work on the B land, where by our assumption ten men could produce only 2400 bushels on each 100 acres. The owner will give such a laborer only what he could get elsewhere, on the B land, which would be 240. The difference between the 240 bushels and the 260 bushels, the owner of the superior land takes for himself. Encouraged by this, the owner thinks of hiring a twelfth man, but concludes that he would thus secure a crop of only 3280 bushels. Hence the twelfth man would increase the output by only 220 bushels, while he would have to be paid 240 bushels, the amount that he could earn by working free land in group B. All new laborers, therefore, in excess of one for every ten of the earlier laborers, would find it more profitable to put their labor upon the free land. Hence, as the demand for agricultural produce increases relatively to the supply, new labor and capital are expended upon land already under cultivation as well as upon land not before used. The rent of such land is increased by the surplus yielded by every addition of labor and capital. In other words, there is a change in both the intensive and extensive margins of cultivation. With every increase in the price of produce, and with every fall in the extensive margin of cultivation, more labor may be employed profitably on land already cultivated. Thus the landowner, who in the case last supposed could not afford to employ a twelfth laborer, may be able to employ thirteen or even more when the extensive margin of cultivation has fallen to group 0 or D.

From the foregoing it is clear that the theory of rent is based upon the law of diminishing returns, which has already been explained in a previous chapter. It is evident that, barring improvements in the arts of production or consumption, each addition to the number of mouths which must be filled, at least beyond a limited number, makes harder the task of drawing sustenance from the earth. But we know that improvements have hitherto more than kept pace with increasing population.