The previous plate measured the surplus on one side of the market — the buyer's, the triangle of benefit he carries away beneath the curve of demand. But a market has two sides, and so does its surplus, for the seller no less than the buyer ordinarily comes away from a transaction ahead. The price he receives is the same on every unit he sells, the going price; but the units did not all cost him the same to supply, and on every one but the last he is paid more than the least he would have accepted for it. The accumulated excess of the price he gets over the price he would have taken is his surplus, the mirror image of the buyer's, and this plate is about it: what it is, where it comes from, and the surprising fact that, followed to its root, it turns out to be the oldest quantity in all of economics wearing a new name.
The construction runs exactly parallel to the one the last plate drew, with supply in the place of demand. Just as buyers differ in what they will pay, sellers differ in what they will accept, for they differ in what it costs them to supply. Some can produce cheaply and would part with their goods for very little; others can produce only at greater cost and must be paid more to make it worth their while. Arrange them in order of cost and one obtains the supply curve, sloping upward — the low-cost sellers at the bottom, willing to supply at a low price, and progressively higher-cost sellers entering only as the price climbs to meet them. At the market price, every seller whose cost lies below it supplies, and each is paid that single price regardless of his particular cost. The seller who could have produced for three and is paid eight has gained five; the sum of all such gains, across all the units supplied, is the producer's surplus — the area lying above the supply curve and below the line of the price, the precise reflection, flipped about the price line, of the consumer's surplus above it.
Set the two surpluses side by side and a satisfying whole emerges. The consumer's surplus is the region between the demand curve and the price; the producer's surplus is the region between the price and the supply curve; and together they fill the entire space between the two curves, from the vertical axis out to the quantity at which they cross. That total area — the sum of what buyers gain and what sellers gain — is the whole benefit the market confers, the gains from trade entire, and it is divided between the two parties along the price line that runs across the middle of it. The price, in this picture, is not only what changes hands but the boundary that apportions the spoils: raise it, and the buyers' share shrinks while the sellers' grows; lower it, and the reverse; and only at the price where supply meets demand is the total as large as it can possibly be — which is the deepest meaning of the claim that competitive markets are efficient. A tax, a monopoly, a quota — anything that pushes the quantity away from that crossing — leaves part of the area between the curves unclaimed by anyone, a benefit that simply fails to occur. But that loss is a later plate's subject.
Beneath the price sits the seller's gain and above it the buyer's, two surpluses pressed together along the line where they meet; and the whole region between the curve of demand and the curve of supply is the good the market does, parcelled out between the two parties to every trade.
It is tempting to suppose that the producer's surplus is simply the seller's profit, but it is not, and the difference is instructive. The surplus is the excess of the price over the cost of producing one more unit — over marginal, or variable, cost — summed across the units sold; it is, in other words, revenue minus the costs that vary with output. Profit is revenue minus all costs, the variable and the fixed alike. The producer's surplus therefore exceeds profit by exactly the fixed costs: it is the sum out of which the fixed costs must still be paid, with whatever then remains being profit proper. In the short run, when a firm's plant and equipment are given and their cost already sunk, the producer's surplus is the whole return to those fixed things — the money the operation throws off above its running costs, available to cover the investment in the factory and to reward its owner. And this is the clue that leads from the surplus to its source, for a return to a factor that is fixed in supply is precisely what an older part of the subject calls a rent.
Rent, in the economist's sense — as the second volume of this table set out at length — is the payment a factor of production receives over and above the least that would keep it in its present use, its earnings in excess of what it could get elsewhere. A factor that is fixed in supply, that cannot be added to however high its price may rise, earns the purest form of it: since it will be there whether it is paid much or little, everything it is paid above nothing is rent, a surplus wrung from its scarcity alone. Ricardo saw this first and most clearly in the land — the better acres, more fertile or better placed than the worst land in use, yield more for the same effort, and that excess yield becomes a rent to whoever owns them, a rent equal to their advantage over the marginal land that pays none at all. And here the two halves of the table fold together, for the producer's surplus in a market is nothing other than the sum of these rents: the excess returns earned by all the sellers more efficient than the marginal one, accruing in the end to whatever scarce thing made their efficiency possible. The seller's surplus, traced to its source, is the rent of the scarce factors that lie behind the supply.
Traced to its source, the seller's surplus is never quite the seller's: it is the rent of whatever made the selling possible and could not be multiplied — the land, the gift, the place, the patent — the payment a scarce thing commands beyond the cost of keeping it where it is.
But there is a complication of time, and it was Marshall who resolved it, with a coinage that has stuck. Land is fixed in supply permanently, and its rent is permanent too. Most other factors, though, are fixed only for a while. A factory, a blast furnace, a fleet of ships, a worker's hard-won skill — none of these can be conjured up overnight, so that in the short run their supply is as fixed as the land's, and they earn, in the short run, a surplus exactly like a rent: the excess of what they bring in over the mere cost of operating them. Marshall called this short-run surplus a quasi-rent, to mark that it behaves like a rent without being the rent of a permanently scarce thing. The difference lies wholly in the passage of time. In the long run, more factories can be built, more ships launched, more workers trained; the supply that was fixed becomes elastic; and the competition of the new entrants erodes the quasi-rent away, as it can never erode the rent of land. Marshall's point was a unifying one: rent is not a peculiarity of land but the general return to scarcity, and the only question is whether the scarcity is permanent, in which case we call the surplus a rent, or temporary, in which case we call it a quasi-rent and expect it, in time, to melt.
The symmetry of the two surpluses is elegant, and the dissolution of the seller's surplus into rent is one of the more satisfying unifications the subject affords; but a few honest qualifications belong here. The producer's surplus inherits all the measurement troubles of the consumer's — the supply curve must be known and is not simply handed to us, and the area is reckoned in money, with all the imperfection that entails. Its very identity shifts with the time horizon one adopts: large in the short run, when much is fixed and earning quasi-rent, it shrinks in the long run as entry competes the quasi-rents away, until, under perfect competition with free entry, the only producer's surplus that survives is the rent of whatever is genuinely and permanently scarce, and the pure profit of the marginal firm falls to nothing. This last point carries a quiet sting for the intuition that the seller's gain is his reward for effort or enterprise: in the long run, the framework insists, there is no such reward in a competitive market beyond the rent of scarce factors and the ordinary return on capital, and what looks like profit is most often a rent in disguise, or a quasi-rent on its way to disappearing. The seller's surplus is real enough, but it belongs, in the end, less to the seller than to the scarce things he commands.
With both surpluses now in hand — the buyer's beneath the demand curve, the seller's above the supply curve, and their sum the whole area between — the volume has quietly assembled the pieces of a larger claim it has so far only gestured at. If every voluntary exchange leaves both parties better off, each carrying away a surplus, then trade does not merely move goods from one hand to another but creates value in the moving, manufacturing benefit out of nothing more than a willing buyer and a willing seller finding each other at a price that lies between their valuations. That is the most quietly radical proposition in all of economics, and the most easily doubted, for it seems to promise something out of nothing. The next plate takes it up directly: the gains from trade entire, why a voluntary exchange must by its very nature enrich both sides, and what, if anything, is amiss with so convenient a conclusion. As the table says of every quantity, first to last: the argument remains open.