The Bottleneck

Three plates have been spent on land, and it is time to confess that land was only ever the example. Everything the law of rent describes — a factor fixed in supply, a surplus that demand alone determines, a price determined by rent rather than determining it — depends not at all on the factor's being soil. It depends only on the factor's being scarce, and fixed, and pressed upon by demand. Land is the purest case, because its supply is fixed forever and by nature; but it is a case, not the category. The category is larger, and once seen it is seen everywhere.

The economist who saw it most clearly, and gave the seeing its lasting vocabulary, was Alfred Marshall, whose Principles of Economics (1890) did for the generation around 1900 what Ricardo's Principles had done eighty years before: it set the terms of the discipline. Marshall's contribution here was a single fertile idea — that the surplus Ricardo had found in land is a special case of a surplus that attaches, for a time, to any factor whose supply cannot quickly adjust. He called it quasi-rent, and with it he turned a theory of ground rent into a theory of the whole economy.¹

Begin with a machine. A manufacturer builds one — a press, a loom, a turbine — at considerable cost. The moment it is finished, something important changes about it: its cost of construction is sunk, gone, beyond recovery, and therefore irrelevant to every decision that follows. What governs the machine's use from now on is not what it cost to build but what it costs to run; and so long as the machine earns more than that running cost, it will be kept working, for the only alternative is to leave it idle and earn nothing at all.²

This means that the machine's earnings, in the period before more machines like it can be built, are set not by its cost but by demand. If the product it makes is keenly wanted, the machine earns far above its running cost, and the surplus is large; if the product is little wanted, the surplus is small; but in either case the surplus is whatever demand leaves after the running cost is met. The machine, in the short run, behaves exactly as land does: its supply is fixed, demand alone fixes its return, and the return is a residual — price-determined, not price-determining. The fixed factor need not be a gift of nature. It need only be fixed.³

Marshall gave this surplus its name. The income earned by a factor whose supply is fixed in the short run he called quasi-rent — rent-like, because while the fixity lasts it obeys every law of rent, but only quasi, because the fixity will not last. In the short run the distinction between quasi-rent and true rent is invisible. Both are surpluses thrown off by a factor that cannot multiply; both are governed by demand; both could be taxed away without changing anyone's behaviour, since the factor sits there earning its surplus regardless. Stand close to a machine earning a fat return in a boom and you cannot tell its income from the rent of land. The two look identical, behave identically, and are, for the moment, the same thing.

The word quasi carries the whole of the difference, and the difference is not of kind but of duration. True rent is the surplus of a factor fixed forever; quasi-rent is the surplus of a factor fixed for now. To see what for now does to the picture, one has only to let time pass.⁴

Let the boom continue, and the machines earning their fat returns will not go unnoticed. Other manufacturers, seeing the surplus, build machines of their own; the supply that was fixed in the short run begins, in the long run, to grow; and as it grows, the product becomes more plentiful, its price falls, and the surplus on every machine shrinks. The process halts only when the return on a machine has fallen to the level that just suffices to call forth new ones — the cost of building them, plus the ordinary return on the capital they absorb. At that point the quasi-rent has been competed away entirely, and what remains is merely the normal return to capital, the long-run supply price of the machine. The surplus was real, but it was temporary; it lasted exactly as long as the supply stayed fixed, and no longer.⁵

Here is the deep result, the one that reorganises the theory of distribution. The difference between rent and the ordinary returns to capital and labour is not a difference of kind; it is a difference of time horizon. In the very short run, every factor is fixed and earns a quasi-rent. Over the long run, the reproducible factors — capital, ordinary labour — watch their quasi-rents competed away as supply adjusts, and are left earning only their cost of reproduction. Only the factors that cannot be reproduced — land, and the handful of things as fixed as land — go on earning a true and permanent rent. Rent is not a separate species of income sitting beside wages and profit; it is what every income would be if its factor could never be increased, and wages and profit are merely the rents of things that can. Time, Marshall held, lies near the centre of almost every difficulty in economics, and nowhere more than here, where it alone divides the rent that lasts from the rent that fades.⁶

Once rent is unchained from land, it appears wherever the same conditions are met — wherever supply is fixed, or barred from adjusting, and demand presses against the limit. The condition is general, and so the phenomenon is general. A drug under patent earns a rent, because the law forbids anyone else to make it. A surgeon of rare gifts, or a singer, earns a rent, because the supply of such gifts cannot be expanded to meet the want of them. A taxi medallion, a liquor licence, a banking charter earns a rent, because the number of them is capped. A platform that everyone must use to reach everyone else earns a rent, because its position cannot be duplicated. A brand a century in the building earns a rent, because a century is how long a rival would need to match it. Each is a bottleneck — a narrow place that others must pass through and one party controls — and at each, the controller collects a surplus above what was needed to bring the thing into being.⁷

The bottlenecks differ in their nature and in their permanence, and the plates to come take them one by one: the rent of scarce talent in the next, the rent of legal restriction in the one after, the rents of patent and platform in the modern plates near the end. But they share a skeleton, and it is Ricardo's — a fixed supply, a pressing demand, a surplus to whoever holds the fixed thing. Land was the first bottleneck economics understood. It was not the last, nor, in a modern economy, the largest.⁸

There is a distinction here that the volume has been circling since its first plate, and quasi-rent forces it into the open. Land rent, the earlier plates insisted, is unearned: no one made the land, and its rent rewards no act of creation. But quasi-rent is different, in a way that matters morally and analytically both. Someone built the machine. Someone risked the capital, and might have lost it; someone spent the decade acquiring the rare skill. The quasi-rent that the machine or the skill earns is, taken over its whole life, the reward for that building and risking and acquiring — and it must be, or the building would never be done. A manufacturer who could never earn a surplus over his running costs would never construct the machine in the first place; the quasi-rents of its fat years are precisely what repay the cost and the hazard of making it. Seen over the full span, quasi-rent is not a windfall but a return on investment, the indispensable lure that calls forth the capital and the effort a society needs.⁹

And yet. The same income, viewed at a single instant, is indistinguishable from a pure rent; and the difference between a quasi-rent that is rewarding genuine creation and a rent that is merely being extracted can be desperately hard to read from the outside — the more so when the surplus is large, and lasting, and defended. A quasi-rent that springs from luck rather than effort; one inflated by an artificial barrier the holder lobbied into being; one that persists, decade upon decade, long after the investment that justified it has been repaid many times over, because something keeps the supply from ever adjusting — such a surplus has drifted across the line, from the reward of creation toward the extraction of a toll. The whole moral argument of this volume lives in that drift, and in the difficulty of seeing it. The plates ahead will watch a series of quasi-rents and ask, of each, on which side of the line it has come to rest.¹⁰

It is best, in the end, to stop thinking of rent as a box and start thinking of it as a pole. At one extreme stand land and the things as permanently scarce as land — a unique site, a finite ore, a band of the spectrum — whose supply can never grow, and whose rent is therefore pure and perpetual. At the other extreme stand the perfectly reproducible things, made freely and at once to meet any demand, which earn no rent at all because their supply forever answers their price. And between the poles lies almost everything: the vast middle country of quasi-rents, where supply adjusts but slowly, and the surplus lasts a while — a year, a generation, the term of a patent, the life of a brand — before it fades, if it fades. Every asset in an economy can be placed somewhere on that line by asking a single question: how fast can its supply respond? The nearer the fixed pole, the more rent-like and lasting its return; the nearer the reproducible pole, the more fleeting.¹¹

This is the map the rest of the volume will walk. The land plates explored the fixed pole; the plates to come move through the middle country, where the modern economy increasingly lives — among talents that cannot be copied, licences that cannot be had, platforms and patents and networks whose fixity is engineered and whose quasi-rents show, year upon year, a stubborn refusal to erode.¹² The concept's reach runs wider still: Marshall's own extension of it — the composite quasi-rent shared between two factors worth far more together than apart — became, generations later, a seed of the modern theory of the firm, in which the scramble to capture a jointly created surplus helps explain why some exchanges are folded inside a company rather than trusted to the open market.¹³ Whether the modern quasi-rents' refusal to fade is the honest reward of creation or the quiet entrenchment of a toll is the question the volume now carries into its second half. Ricardo found rent in the ground; Marshall found it everywhere the ground's condition was met; and the everywhere has only grown.