The opening plate showed rent to be a differential — the surplus of the better land over the worse — and it rested that whole account upon a single strange figure: the worst land in use, the marginal land, which pays no rent of its own and yet, the present plate will show, sets the price for every richer acre above it. How can the poorest and least productive plot under cultivation be the one that governs the price of corn, and through it the rent of all the land besides? The answer is the marginal principle, the most powerful single idea in all of economics — the idea that value is determined not by averages or by totals but at the margin. It made its first appearance, quietly, in Ricardo's theory of rent; and two generations later it would be generalized into the very foundation of the modern science. This plate follows the marginal land from that humble beginning to its imperial reach, and shows, at the end, that the wage of the volume before and the rent of this one are a single idea, seen twice.
Begin with the paradox itself, and its resolution. The marginal land sets the price precisely because it pays no rent. Its produce only just repays the labour and the capital laid out upon it, leaving nothing over; and the price of corn must be high enough to make even this poor plot worth the working — for if it were not worth working, its corn would be absent from the market, and the price would rise until it was. So the price is pinned to the cost of production on the marginal land, the one plot into whose costs no rent enters. And every better plot, meeting that same price with lower costs — a greater yield for the same outlay — throws off a surplus, which is rent. The worst land governs the price not in spite of paying no rent but on account of it: it is the plot at which price exactly equals cost, the zero-mark from which all the rents above are measured. It rules because it is the margin, and the margin is where the price is made.
Ricardo's margin, however, had two faces, and the second is the more far-reaching. The first is the extensive margin already described: the worst land in use, out at the ragged edge of cultivation. But there is also the intensive margin — the last dose of labour and capital applied to a plot already in use. For a farmer may wring more corn from the same field by working it harder, with more labour and more manure and more ploughing; but he meets diminishing returns, each successive dose yielding a little less than the one before. He goes on adding doses until the last one barely repays its own cost, earning no surplus — while the earlier, more productive doses each earn one. And so rent arises even upon a single plot of perfectly uniform quality, out of nothing but the diminishing returns to successive doses upon it. This was the deeper discovery: that rent is not, at bottom, a matter of the difference between good land and bad, but of the difference between the inframarginal application and the marginal — a difference that is wholly general, arising wherever a fixed factor is worked together with a variable one under diminishing returns. The intensive margin cut rent free from its dependence on differences of soil, and made it a general law of the surplus that accrues to whatever is fixed.
Rent arises not, at bottom, from the difference between good land and bad, but between the inframarginal dose and the marginal.
And here is the plate's largest theme. Ricardo's marginal land was the first appearance in all of economics of an idea that would, half a century on, remake the entire subject: that value is settled at the margin. For decades the idea lay half-buried in the particular soil of rent. Then, in the eighteen-seventies, three men working in ignorance of one another — Jevons in England, Menger in Austria, Walras in Switzerland — dug it out and generalized it, in the revolution that founded economics as it has been practised ever since. They saw that what Ricardo had found for land held for everything: that the worth of a good is fixed not by its total usefulness nor its average cost but by its marginal utility, the value of the last unit; that the price of a factor is fixed by its marginal product, the output of the last unit employed — which is the marginal-productivity theory that the volume before found installed as the orthodox account of the wage; that the firm produces up to the point where marginal cost meets marginal revenue, and no further. The marginal land was the seed of all of it. And the kinship with the wage is exact, not merely analogical: the marginal worker, like the marginal land, earns no rent — his product only just covers his wage — while the inframarginal workers, the more productive or the more scarce, earn a surplus above their wage, which is their rent. The wage of The Margin and the rent of The Differential are one principle, applied once to labour and once to land.
Once the marginal principle is grasped, rent escapes the cornfield altogether and is seen to be everywhere. For the logic is perfectly general: wherever a factor of any kind is supplied in a fixed amount, the margin sets its price, and every unit more productive or more scarce than the marginal one earns a surplus above its cost — a rent. The uniquely gifted surgeon earns, above the salary that would just keep her in medicine, a rent upon her scarce talent; the firm with the better process earns a rent above the costs of the marginal firm; the worker with the rare skill earns, as the volume before allowed, a rent that the orthodox prefer to call a return on human capital. Economic rent, in this general sense, is the surplus that any factor earns over and above its transfer earnings — over what would be needed to keep it where it is — and it is found in every market where anything scarce commands a price. Land's rent is only the purest specimen: the case where the factor's supply is the most nearly fixed, and its entire return is therefore rent. The marginal land taught the whole of economics to see rent in everything.
But the principle, for all its reach, settles only half of the question — and the half it leaves untouched is the very half this volume cares about most. The margin determines where the price is set, and therefore how large the rent is. It says nothing at all about who gets the rent. The inframarginal surplus, the rent upon the better land, exists because of the margin; but it falls to the landlord because of property — because he holds the title to the better land, and may charge others for leave to use it. The margin is neutral; ownership is not. That very same surplus which the marginal principle locates so exactly could, under a different settlement of property, fall instead to the cultivator, or to the public, or be taxed clean away — and the margin, the price, the efficient use of the land would none of them alter by a hair. This is the deep reason behind the land-value tax that a later plate will reach: because the rent is a surplus fixed by the margin and not by the owner's exertion, it can be taxed without disturbing the price or the use of the land at all, the margin standing wholly undisturbed. The marginal principle hands us the price, and the size of the rent; but whether that rent is earned or extracted — who ought to have the surplus the margin creates — is a question of property and of justice that the margin is constitutionally unable to answer.
The margin is neutral; ownership is not. The margin creates the surplus; property alone decides who shall keep it.
The principle has its limits, too, which an honest plate must mark. The marginal land is, in sober truth, an abstraction not always easy to point at: it is defined by the going price, which is itself fixed at the margin — a circle that is logically sound, the two being settled together in equilibrium, but practically slippery, for the "worst land in use" shifts with every turn of demand and technique and is seldom a plot one may stand upon and name. And more tellingly, the marginal principle, which serves so cleanly for land, grows troubled when it is carried over to capital — for, as the volume before recorded, the Cambridge capital controversy showed that the "marginal product of capital" cannot be cleanly defined when capital is a heterogeneous heap of different machines, so that the marginal theory of the wage stands on firmer ground than the marginal theory of profit. For land, the principle is at its most secure: land's marginal product is plain, its supply genuinely fixed, its rent the very textbook case. The marginal land is the principle's first triumph, and its cleanest; it is only in the harder matters — capital, and the distribution entire — that the principle begins to strain.
Return, then, to the question of the volume. The marginal land is where the price is made and the rent is measured — the poorest plot, paying nothing, governing all. And the principle it seeded, that value is settled at the margin, is the unifying idea of the modern science, binding the rent of this volume to the wage of the last as two applications of one law: the margin sets the price, and the surplus above it is rent. But the principle is, in the end, serenely neutral upon the question this volume exists to put. It tells us that the rent is real, and how large it is, and that it is a surplus and not a cost. It does not tell us whether the owner of the scarce land has earned it. That the margin creates the surplus is a matter of price; that the landlord captures it is a matter of property; and whether that capture is earned or merely seized is the matter of the rent — the question the margin frames with perfect precision, and leaves, with perfect indifference, open. The worst land sets the price for all. Who shall have the surplus it leaves is the argument of the volume; and it remains, as the table says of every price within it, open.