The volume has shown the wage to be a contest — bounded above by what labour produces, and pressed below that mark by the buyer's power. This plate takes up the most direct of all the attempts to intervene in that contest: the floor set, by law, beneath the wage, below which the bargain may not be driven. And it begins where both its title and the history of the idea begin — in the competitive model, the textbook world in which a floor, however kindly meant, must cost the very workers it was raised to help their jobs. That this confident verdict has, in the event, so often miscarried is the story the plate has to tell; but the miscarriage cannot be understood without first taking the model in earnest, at its full and genuine strength.
The argument is clean, and it is formidable. In a competitive labour market the wage comes to rest where the labour offered equals the labour wanted — that resting point being the price at which no willing worker is left unhired and no willing employer unstaffed. Set a floor above it, and the meeting fails: at the higher wage employers want fewer hands, and more people want to work, and the gap between the two is, by the plainest definition, unemployment. The minimum wage, on this telling, draws a line through the low-paid and sorts them into two camps — those who keep their places, now better paid, and those who lose them, or never find them, and are paid nothing at all — and the second camp is drawn from the youngest, the greenest, the least skilled, the very people in whose name the wage was raised. George Stigler set the case down, with his usual unsentimental rigour, in 1946: a minimum wage relieves some of the poor by pricing others wholly out of work, and if the end in view is the income of the poor, far better to hand them the money outright than to outlaw the low bargain that is, for some, the only one on offer. For the better part of the century this was the profession's settled teaching, and it drew its great authority from following with something near geometric necessity from a model that nearly everyone held.
But the model says rather more than the slogan drawn from it, and the surplus is everything. How much harm the floor does depends entirely upon how far above the clearing wage it is set — upon what the trade calls the bite, the ratio the minimum bears to the typical, or median, wage. A floor laid below the point where the market would have cleared binds nothing and costs nothing. A floor set modestly above it grazes employment. A floor set far above it cuts to the bone. The competitive model nowhere holds that a minimum wage as such destroys work; it holds that a sufficiently high one does, and it leaves the height at which the damage begins to be found out by other means. Fifteen dollars is a feather in a high-wage city and a millstone in a poor rural county, for the bite is always local; and a great part of the heat in the minimum-wage quarrel is, in truth, a disagreement about the bite, carried on as though it were a disagreement about the principle.
The competitive model nowhere holds that a minimum wage destroys work; it holds that a sufficiently high one does — and leaves the height to be found out.
And when the question was at length put to the world instead of to the blackboard, the model's confident forecast stumbled. The natural experiment of David Card and Alan Krueger — New Jersey raising its floor while neighbouring Pennsylvania stood pat — found, just where the competitive model demanded a fall in employment, none worth the name; and the result has, on the whole, outlasted the long and bitter fight waged against it. Why it should be so is the proper business of the plate that immediately follows this one, upon monopsony; for where the employer holds power over the wage — and the eighth plate makes the case that he very often does — a floor pitched within the right band can lift the wage and the level of employment together, by stripping the monopsonist of the very motive that made him hire few. Here it is enough to mark the finding and draw its lesson: the competitive model is the right instrument only for a market that is in fact competitive, and the low-wage labour market, precisely where the minimum bites, frequently is not. The floor that the textbook swore must cost jobs has, at modest heights, again and again declined to.
To weigh the minimum wage by its effect on employment alone, in any case, is to read but one column of a long account. A floor that scarcely moves the number employed may still do a great deal of work: it raises the pay of those at the very bottom, and, by a ripple that travels some way up the ladder, of those ranged just above them; it shifts income out of profit, or onto the prices in the shop, or into lower turnover and steadier effort, in proportions that differ from one trade to the next; and where it bites hard it may hasten the coming of the machine that does the task instead — a matter the eleventh plate takes up in its turn. The question is never merely how many jobs, but whose incomes rise and whose fall, by how much, and at whose charge — the distributional question this volume has held, from its opening page, to lie beneath the smooth surface of every wage.
And here the orthodox critique recovers a point that the monopsony rebuttal leaves standing. The minimum wage is a blunt weapon against poverty, for the roll of those who earn it and the roll of those who are poor overlap only loosely: many a minimum-wage earner is a teenager, or the second earner in a comfortable house, and many of the poor hold no job at all, and so lie beyond the reach of any wage floor whatever. An instrument aimed more truly at poor families is the earnings subsidy — the earned-income credit — which tops up the pay of the low-waged in poor households without forbidding a single bargain. Yet the two are less rivals than partners. An earnings subsidy with no floor beneath it can be quietly pocketed by the employer, who trims the wage in the knowledge that the public purse will close the gap, so that the help meant for the worker leaks away into profit; the floor stops the leak. The wage floor and the wage subsidy do their best work in harness — the one curbing the employer's power over the price, the other reaching the poor whom the floor, by itself, cannot touch.
Where, then, does the floor begin to bite in good earnest? On this the evidence is honestly divided, and a scrupulous plate will not pretend it otherwise. One school, marshalling the modern natural experiments — the telltale bunching of jobs at the new minimum, anatomised by Arindrajit Dube and Doruk Cengiz and their colleagues — finds that the minimum wages actually placed on the books have lifted pay with little loss of employment, even at heights once thought rash. Another, gathered about David Neumark and his collaborators, finds real disemployment, falling hardest upon the youngest and least skilled, and cautions that the want of an effect within the range so far studied is no warrant for its absence beyond. Both have hold of something true. The modest floors of recent memory, imposed upon labour markets that were often monopsonistic to begin with, have cost few jobs and raised much pay; but the competitive model's warning has not been struck out, only fenced in — drive the floor high enough, past any slack the market can absorb, and even the most sympathetic evidence looks for the jobs to vanish. The honest position is that the minimum wage is neither the slayer of jobs the old orthodoxy dreaded nor the costless gift its warmer friends sometimes advertise: it is a tool with a range of safe working, whose upper edge we are still, one labour market at a time, feeling for.
Neither the slayer of jobs the old orthodoxy dreaded nor the costless gift its friends advertise: a tool with a range of safe working, whose upper edge we are still feeling for.
Return, last, to the question the volume puts to every wage. The minimum wage is the law's own hand reaching directly into the contest the first plate described — fixing, by statute, a level below which the bargain shall not be forced, however lopsided the power gathered at the table. The competitive model judged that reaching-in a costly folly, upon the supposition that the wage it overrode was already the fair clearing price of a free market. The whole burden of this volume is that the supposition is often false — that the wage beneath the floor was frequently not the verdict of competition at all, but the register of the employer's power, the monopsony of the eighth plate — and that a floor, rightly set, does not trample a fair price so much as correct an unfair one. The floor is worth, asserted by the law against power; and like every wage in these pages it cannot be judged from the serene diagram alone, but only by asking what contest it interrupts, and in whose favour the interruption falls. Pitched too high, it will indeed destroy what it touches; kept within its range, it is among the very few instruments that can raise the wage of the powerless without first waiting for their power to grow. Where that range gives out is a question, as the table is fond of saying, still open.