Eight plates have now been spent building the margin into an apparatus of some intricacy — the falling gradient of utility, the rule that levels the margins of every use, the scissors of supply and demand, the marginal reckoning of cost and of the product of factors, the rates of substitution along an indifference curve. It is an impressive machine, and a student may be forgiven for feeling, at this point in the volume, that the margin has become something only an economist could love or operate. The purpose of this plate is to undo that impression, by showing that the whole of the apparatus, for the person who actually has a decision to make, collapses into a single instruction of three words. Think at the margin. Everything else in this volume is the elaboration, the proof, and the qualification of those three words; but the words themselves are the usable residue, the one thing that a person who learns no other economics should carry away, and they are very nearly enough on their own.
The rule is this. When you are deciding how much of something to do — and almost every real decision is a decision about how much, not whether — do not ask whether the thing as a whole is worthwhile, and do not consult the average return it has yielded so far. Ask instead what one more unit of it would add, and what one more unit would cost: the marginal benefit and the marginal cost of the next increment. If the marginal benefit exceeds the marginal cost, do a little more; if it falls short, do a little less; and you have found the right amount when the two have come into balance, when the benefit of one more unit has fallen to just equal its cost. The best quantity of anything, in a phrase, is the quantity at which marginal benefit equals marginal cost. That single condition is what all the machinery of this volume has been describing from its various angles — the equimarginal rule, the scissors, the hiring of factors up to the point where their marginal product meets their price are all special cases of it — and it is the whole of what the margin asks of anyone who would use it to decide.
What gives the rule its standing as the most useful single idea in economics is the sheer breadth of what it governs. A firm deciding how much to produce applies it, expanding output until the cost of the last unit made rises to meet the price it fetches. A consumer deciding how to spend applies it, as the fourth plate showed, buying of each good up to the point where the last penny spent on it yields the same satisfaction as the last penny spent on any other. A worker deciding how many hours to labour applies it, working until the value of one more hour's leisure surpasses the wage that hour would earn. A student deciding how long to study, a government deciding how much of a river to clean, a factory deciding how much of its pollution to abate — each faces a question of how much, and each is answered by the same comparison of the next increment's benefit against its cost. The rule does not care what the activity is. Wherever there is a quantity to be chosen, and a benefit and a cost that both move with it, the margin is where the answer lies, and the three words apply unchanged.
Almost every question that opens with the words how much — how much to make, to buy, to work, to spend, to risk, to abate — has at bottom the same three-word answer; and the answer is not a number but an instruction: think at the margin.
To see why the rule must be taught, and is not simply obvious, it helps to notice how differently the untrained mind tends to reason. Left to itself, ordinary thinking gravitates to wholes, and to averages, and to absolutes. It asks whether an activity is good or bad, as though the choice were all-or-nothing, when the real question is how much of it to do. It reaches for the average — the typical cost, the usual return — when only the marginal figure bears on the decision, as the plate on cost laboured to show. And confronted with something it values highly, it slides into the language of the absolute: safety is priceless, no expense should be spared, we must do everything possible. The marginal habit of mind is the steady refusal of all three temptations. It declines the all-or-nothing framing and asks instead for the right quantity; it sets the average aside and attends only to the increment; and it meets every is priceless with the quiet, unwelcome question of how much. This is the distinctive cast of thought that economic training instils, and it is why economists so often seem to be answering a different question from the one that was asked — colder, narrower, more arithmetical — when in truth they are only asking how much of a good thing, where others had asked whether it was good.
Nowhere does this habit unsettle people more than when it is turned upon the things they hold sacred, and the rule's hardest application — the one that has earned economists their reputation for cold-bloodedness — is the pricing of human life. To the marginal mind the question is unavoidable: a society has only so much to spend, and every pound spent reducing one risk is a pound not spent reducing another, so that the question of how much to spend to save a life is not a question it can decline to answer, but only a question it can decline to answer honestly. Economists answer it by observing what people themselves will pay to reduce small risks to their own lives — the extra wage that dangerous work must offer before it is taken, for instance — and inferring from those choices an implied value of a statistical life. The procedure strikes many as monstrous, a reduction of the infinite worth of a person to a sum of money. But the alternative is not to leave life unpriced; it is to price it blindly, and worse. For a public budget that refuses on principle to count the cost of saving a life will pour resources into one hazard far past the point of usefulness while another, cheaper to remedy, goes unaddressed — and will thereby save fewer lives, with the same money, than a frank reckoning at the margin would have saved. The refusal to do the arithmetic does not keep our hands clean; it only hides, in the lives we failed to save, the price we were paying all along.
To ask what a human life is worth seems monstrous until one sees that a budget which will not ask it must answer it anyway — in the lives it quietly fails to save with the money it spent, uncounted, somewhere else.
This is the deep vindication of the marginal rule, and it is worth stating plainly, because the rule is so often mistaken for an apology for callousness when it is in fact a discipline of conscience. Given that resources are finite, and that any pound spent here cannot also be spent there, the way to do the most good — to save the most lives, relieve the most suffering, produce the most welfare from what one has — is precisely to push each use to the margin where its last unit of benefit equals what that unit costs, and to move resources from the uses where the marginal benefit is low to those where it is high. To think at the margin is to refuse to waste; and waste, where the stakes are lives, is not a venial sin. The hospital that spares no expense on one patient has, by that very generosity, less to spend on the next; the safety rule that ignores its own cost buys less safety, in total, than a cost-conscious rule would buy. The marginal habit, rightly understood, is not the enemy of our higher values but their faithful servant, the only method that takes seriously both the worth of the end and the scarcity of the means. Its coldness is the coldness of arithmetic done honestly in the service of something warm.
None of which is to pretend that the rule is unconditional, for it rests on assumptions that the remaining plates of this volume are devoted to questioning. It assumes, first, that one can move smoothly, taking a little more or a little less as the comparison directs — that the margin is a place one can actually occupy; yet a great deal of the world comes in lumps that cannot be halved, and the next plate is given over to that difficulty. It assumes that the marginal benefit and the marginal cost can be known, or at least estimated, when often they are shrouded in an uncertainty the tidy condition conceals. It assumes, more controversially still, that decisions really are made this way — that firms and people actually do equate margins, rather than merely behaving, in the end, somewhat as if they did — a claim that provoked one of the sharpest controversies in the subject's history, taken up two plates hence. And it assumes, at the root, that the benefits and costs being weighed can be laid on a common scale at all, which is exactly what is denied by those who hold that some values are incommensurable, not to be traded against money or against one another at any rate whatever. The rule is powerful precisely where these assumptions hold, and no further; its empire is wide, but it has borders.
Within those borders, though, the rule is the nearest thing the discipline has to a gift it can hand to anyone. The elaborate theory of the margin is for specialists; the three words are for everyone, and a person who genuinely absorbs them — who learns to ask, of every decision about quantity, what the next unit adds and what it costs, and to ignore the totals and averages and absolutes that clamour for attention — has acquired the larger part of what economics can do for a practical life. It is the revolution's residue, the whole vast apparatus boiled down to a maxim a person can carry in a sentence and apply in an afternoon. And yet the very first of the assumptions on which it rests is one the world delights in violating. The rule supposes that one can always take a little more or a little less; but bridges and factories and human beings do not come in fractions, and there are decisions in which the margin offers no foothold at all, because the smallest step available is a whole and indivisible lump. To the things that will not be halved, and to the trouble they make for the smooth calculus of the margin, the next plate now turns. As the table says of every quantity, first to last: the argument remains open.