Long before there was an economy to study, there was the problem of the fair exchange. Aristotle set it down in the fifth book of the Ethics, where he divided justice in two: the justice that apportions a city’s common goods, and the justice that governs what passes between two hands. An exchange is just, he held, when it holds an equality — not a flat equality, for a house is not a shoe, but a proportionate one, each thing answering to the other according to its worth. Money was devised, he thought, precisely to make unlike things commensurable, so that the builder and the shoemaker might trade as equals.

From this came two convictions that would hold for two thousand years. The first was that a price could be set against a standard of justice, and found wanting. The second, its shadow, was that money which bred money — interest taken on a loan — was a thing against nature, since coin was made to measure exchange, not to grow.¹1Aristotle, Nicomachean Ethics V — justice in exchange, and the proportionate reciprocity that money exists to make possible. Usury as ‘contrary to nature’ — money breeding money — belongs to the Politics I. The fair exchange and the forbidden usury were the two poles of the moral economy, and the wage lived between them.

The Schoolmen of the Middle Ages took Aristotle’s equality and made it doctrine. Thomas Aquinas asked, plainly, whether a man may sell a thing for more than it is worth, and answered that he may not: to sell above the just price — the iustum pretium — is to take what is not freely given, and is against justice.²2Thomas Aquinas, Summa Theologiae II–II, q. 77, a. 1 — whether one may sell a thing for more than its worth; the just price as the common estimation, and the wrong of pricing to a buyer’s need. The wage was the same question turned around. A just wage was one that sustained the worker in the condition his labor and his station merited — enough to live, and to live as what he was.

But the doctrine held a quiet surprise, and everything turns on it. What was the just price? Not, for most of the Schoolmen, a figure handed down by a confessor. It was the communis aestimatio — the common estimation, the price a thing commonly fetched in that place at that time, among honest dealers. The just price was, in the ordinary case, the going price. What justice forbade was not the market but its abuses: the fraud, the cornering of a scarce good, the man who raised his price because he had seen that his buyer was desperate. Take those away, and the price the market reached was the price justice asked for.

It fell to the last of the Schoolmen to see how far that idea reached. In the universities of sixteenth-century Spain — at Salamanca above all — a remarkable company of theologians, Vitoria and de Soto, Azpilcueta and Molina, set out to work the just price through to its end, and arrived somewhere their successors would recognize better than their forebears could have.

The just price, they reasoned, could not come from the cost of making a thing, nor from the labor bound up in it, nor from any worth it carried in itself — for the same thing was justly dearer in a year of scarcity and justly cheaper in a year of plenty, though nothing within it had changed. It came instead from common estimation: from how badly the thing was wanted, and how little of it there was to be had. Price was the work of human need meeting human supply, and the just price was simply the price a free market reached when no one cheated and no one was cornered.³3The School of Salamanca — Vitoria, Domingo de Soto, Martín de Azpilcueta (Dr. Navarrus), Luis de Molina. That they reached a subjective, supply-and-demand account of value is the celebrated — and much-debated — thesis of Marjorie Grice-Hutchinson, The School of Salamanca (1952), later pressed by Hayek and Rothbard. They had, in the course of defending a medieval doctrine, very nearly arrived at the modern one — utility, scarcity, supply and demand — three centuries before the men who would be given the credit.

The just price, they concluded, was only the market price — honestly arrived at.

After Salamanca the moral price had little left to do. If the just price was the market price, then the market price could simply be called the price, and the adjective fell away. When Adam Smith came to write of a ‘natural price’ toward which the market gravitates, the machinery was the Schoolmen’s, but the justice had gone out of it; what remained was mechanism. For a century and more the question of whether a wage was just was not so much answered as retired — folded into the larger confidence that a free exchange, by being free, was already fair. The moral economy did not lose the argument. It was quietly informed that the argument was over.

It was not over. In 1891, into a Europe of factory towns and the first great combinations of labor, Leo XIII set down the encyclical Rerum Novarum, and with it returned the just wage to the world as a reproach to the market rather than a gloss upon it. Employer and worker, the Pope allowed, were free to agree on wages. But beneath that freedom lay something older and more binding: a dictate of nature, he wrote, more imperious and more ancient than any bargain between man and man — that a wage be enough to keep a frugal and upright worker alive.⁴4Leo XIII, Rerum Novarum (1891), § 45 — beneath the freedom to bargain lies ‘a dictate of natural justice more imperious and ancient than any bargain between man and man.’ If the free price fell below that line, the freedom did not make it just; it made the injustice consensual.

The idea, once restored, would not stay still. Pius XI widened it to the family wage — enough to keep not a man but a household. In America a priest-economist, John Ryan, turned the encyclical into a programme and a figure, argued the living wage as a legal right, and carried it close enough to power to take a nickname from the New Deal he helped to shape.⁵5Pius XI, Quadragesimo Anno (1931), on the family wage; John A. Ryan, A Living Wage (1906), who carried the argument into the New Deal and was known, in the end, as the ‘Right Reverend New Dealer.’ When the Fair Labor Standards Act set out, in 1938, to secure ‘the minimum standard of living necessary for health, efficiency, and general well-being,’ it was speaking the just wage in the flat prose of a statute.⁶6Fair Labor Standards Act of 1938, § 2(a) — the congressional finding against conditions below ‘the minimum standard of living necessary for health, efficiency, and general well-being of workers.’

Freedom did not make the wage just. It only made the injustice consensual.

So the oldest framework leaves two bequests, and they do not agree. From the just price as common estimation descends the whole market tradition — the conviction that a price freely reached is a fair one, that the wage is worth whatever it fetches, that to interfere is to deface a natural thing. From the just price as a moral floor descends its opposite — the conviction that a price can be unjust however freely it was struck, and that the law may say so. The defense of the market and the indictment of it were born in the same cloister, of the same parents, a few pages apart.

That is why the argument never closes, and why it is not, at bottom, an economic argument at all. Every contest over a minimum wage, every living-wage campaign, every claim that a market ‘clears’ and so settles the matter, is a late round of a dispute the Schoolmen were already having: whether the price the market reaches is the just price, or only a price. The wage was the first place the moral and the market met. It is still the place they have not finished meeting.

Before the clock there were no hours to sell. A medieval ploughman did not work an eight-hour day or a twelve-hour one; he worked until the field was sown or the light was gone, and the length of his labor was set by the task and the season, not by a number on a dial. Time, where it was reckoned at all, was reckoned against the things of life — an egg boiled for the space of a Pater Noster, a journey ran from cockcrow to noon. What mattered was whether the work was done, not how long it had taken, and the two were not yet the same thing.

This was not idleness; it was a wholly different relation to time — what one historian named task-orientation, in which the shape of the day follows the work to be done rather than the clock.¹1E. P. Thompson, ‘Time, Work-Discipline, and Industrial Capitalism,’ Past & Present 38 (1967) — the classic account of the passage from task-orientation to labour measured by the clock. Its rhythm was uneven by our lights, hard bouts of work broken by long idle ones, and it had a feature we have wholly lost: when the task was finished, the day was finished. No one was owed a surplus hour, because no one was buying hours. A man was paid, if he was paid in coin at all, for the sheaves he bound or the cloth he wove — by the piece, by the thing, by the job. The wage by time did not exist, because the thing it would one day sell did not exist.

That thing had to be invented, and it was invented, of all places, in the monastery. The Benedictine houses of early medieval Europe had bound themselves to a rule, and the rule required prayer at fixed points of the day and night — the canonical hours, from matins before dawn to compline at dusk. To keep them, a community could not depend on a single monk watching the sky; so the houses built mechanisms to ring a bell at the appointed times, and timekeeping in the West became, for the first time, regular, communal, and announced aloud.²2Lewis Mumford, Technics and Civilization (1934), on the Benedictine origin of regular timekeeping and the clock as ‘the key-machine of the modern industrial age.’ On the medieval work-bell, or Werkglocke, see Jacques Le Goff, Time, Work, and Culture in the Middle Ages (1980).

From the cloister the bell came down into the town. Across the fourteenth and fifteenth centuries mechanical clocks rose in the towers of European cities, striking the hours over the marketplace as they had struck them over the choir — and what they summoned was no longer only prayer. The same instrument that had called monks to matins came to call laborers to work and to send them home: the Werkglocke, the work-bell, that fixed when a weaver’s day began and when it might end. The clock, Lewis Mumford would argue — not the steam-engine — was the key machine of the industrial age: the first whose product was neither cloth nor grain but time, counted out in identical units. Eternity had been the monk’s measure; the hour became the merchant’s. The hour, like the just price before it, was born in the cloister.

But a bell that strikes the hours is not yet a market in them. For time to be bought and sold it had first to become a quantity — uniform, divisible, the same everywhere and always. Here the clock did something subtler than ring: it made the hours equal. The hours people had kept before were seasonal hours, the daylight cut into twelve parts whatever the length of the day, so that an hour of June was longer than an hour of December, and an hour of daylight longer than an hour of night. Such hours could be lived by; they could not be priced, for you cannot set a fixed rate on a unit that swells and shrinks with the calendar. The mechanical clock swept the seasonal hour away and put the equal hour in its place — and only then was there a measure of time to which a wage could be fixed.³3On the seasonal or ‘temporal’ hours — daylight divided into twelve parts that lengthened and shrank across the year — giving way to the uniform equal hour of the mechanical clock: Gerhard Dohrn-van Rossum, History of the Hour (1996).

With the equal hour, labor could at last be sold as cloth was sold: so much measured quantity, at a price. The man once paid for the sheaves he bound became the man paid for the hours he gave, and the question of the wage shifted under everyone’s feet. It was no longer what is this work worth, but what is an hour of this man worth — and an hour, unlike a task, has no natural end. Benjamin Franklin drew the moral with a tradesman’s bluntness in 1748: remember, he wrote, that time is money. He that could earn ten shillings in a day and sat idle half of it had not merely lost his ease; he had thrown away five shillings he might have had.⁴4Benjamin Franklin, ‘Advice to a Young Tradesman’ (1748): ‘Remember that time is money.’ Idle time is reckoned, for the first time, as wages thrown away. Once time was money, to rest was to spend.

Work had been a thing you finished. Now it was a thing you spent.

Once the wage was paid by the hour, everything turned on how many hours there were. The task had set its own limit; the clock set none. In the first factories the working day stretched to fill whatever the engine and the gaslight allowed — ten hours, twelve, fourteen, sixteen, six days in the week, with children at the looms beside their mothers. The wall that dusk and exhaustion had once placed around labor was gone, and in its absence the length of the day became the bitterest question of the age.

The answer, when it came, came against the market rather than from it. As early as 1817 the millowner Robert Owen, sickened by what the factories had made of the people inside them, set the demand in three equal parts — eight hours’ labor, eight hours’ recreation, eight hours’ rest — and the figure became a banner carried for a hundred years. The law followed by inches: the Ten Hours Act of 1847, the long agitation of the eight-hour leagues, and at last, in 1938, the Fair Labor Standards Act, which fixed the American week at forty hours and ruled that an hour beyond them must be paid at one and a half — a century and more after Owen had named the number.⁵5Robert Owen’s demand of 1817 — ‘eight hours’ labour, eight hours’ recreation, eight hours’ rest’; the Factory (Ten Hours) Act of 1847; the United States’ eight-hour movement; and the forty-hour week and overtime premium of the Fair Labor Standards Act of 1938. On the working day itself, Karl Marx, Capital I, ch. 10. The day was given a wall again, and this time the wall was written into statute: past a certain hour a worker’s time became his own once more, and an employer who wanted it had to pay a penalty to reach across the line.

The clock-wage was never only one thing; from the worker’s side it had always worn two faces. It was a cage. It was the bell that opened the day and shut it, the timesheet, the foreman docking a quarter-hour for a late arrival; and in time it was the stopwatch of Frederick Taylor, who turned the clock inward to measure not merely the hours a man was bought for but the motions within them, so that even the seconds inside the hour might be made to give up their yield.⁶6Frederick W. Taylor, The Principles of Scientific Management (1911) — the stopwatch, which turned the clock from a measure of the hours bought into a measure of the labour packed inside them. To be paid by time is to have one’s time owned, minute by counted minute. It is the discipline every worker has resented.

And yet the very thing that made the hour a cage made it also a wall. A quantity that can be counted can be bounded; a day with a measured length has, by that fact, an end. The clock that told a man his time belonged to his master after the morning bell told him, with the same precision, that it belonged to himself after the evening one. The overtime rate went further still: it put a price on the boundary itself, made the employer pay more to cross it, and so turned the edge of the working day into something the law would stand and defend. The hour disciplined; but the hour also protected. What the worker resented as a cage was, seen from the other side, the wall that kept the work on its own side of the door.

The hour was a cage. But the cage had a wall — and the wall faced both ways.

That wall is the thing we are now losing, and we are losing it from both ends at once. At the top of the labor market the salaried professional carries the office in his pocket: the message arrives at eleven at night and an answer is expected by morning, and because he is paid by the year and not by the hour, no clock marks where his time stops being his employer’s. At the bottom the gig worker is paid exactly as the ploughman was — by the task, by the ride, by the delivery, by the piece — and the protections the long fight had bolted onto the hourly wage do not reach him, because he has no hours to protect. At both ends the world is sliding back toward task-time: work measured again by what is done, not by how long it takes.

But it is task-time stripped of the one mercy the old task-time carried — the natural wall, the fallen dark and the finished field, that told a man his day was over. The medieval laborer owned no hours, yet when the task was done he was free; the modern worker owns no hours either, and the task is never done, because the device in his pocket can always furnish another. We spent five centuries learning to sell our time, and one hard century learning to fence it, and now, the fence dissolved, we find we have kept the selling and lost the bound. It is no accident that France, in 2017, judged it necessary to write into law a right to disconnect — a right, in plain words, to be permitted to stop.⁷7France’s droit à la déconnexion — the ‘right to disconnect’ — enacted in the El Khomri labour law (Law No. 2016-1088, art. 55) and in force from 1 January 2017: the bounded day re-erected, against the always-on device, by statute. The wall the eight-hour movement won with marches, a later age must try to win back with a law against the telephone.

To earn, we feel, is to deserve, and to deserve is very nearly to possess: the wage we have earned seems already ours, withheld only by the accident of payday. The word will not support the feeling. Wage entered English around 1300 from the Norman French wage, a northern variant of gage — a pledge, a security, a stake thrown down. It is the very same word as wager, and a near cousin of wed, which once meant to bind oneself by a pledge or a gift of earnest money, and of mortgage, the ‘dead pledge’ of the lawyers. A wage, in the grain of the word, is not a payment but a pledge — something staked against a performance still to come.

Older still, and plainer, is the word the language threw away. Before English borrowed the French term, its own word for a workman’s hire was lean — the same word, and at root the same thing, as loan.¹1Wage < Anglo-Norman and Old North French wage (Old French gage) ‘pledge, pay,’ from Frankish *wadi- ‘pledge’; a doublet of gage, cognate with wed (Old English weddian, to pledge or betroth) and with mortgage, ‘dead pledge.’ The older native English word for wages was lean, related to loan. (OED; Online Etymology Dictionary.) The two senses had not yet parted: to be paid for one’s labour and to make a loan were once spoken in a single breath, because at bottom they were a single act. The wage was a loan the worker made, and hoped to have returned. We kept the word for the lending, and forgot it had ever meant a wage — but the thing it named has not changed in the least.

The pledge in the word was once a pledge in the hand. When a servant was hired in medieval England the bargain was sealed not by a signature but by a coin: the earnest penny, the arles, the ‘God’s penny’ — a token pressed into the new servant’s palm to bind him to the work. Once he had taken the earnest he was bound to appear on the appointed day; the coin was not his wages but the pledge of them, the little stake that made the larger promise real.²2The earnest penny — also arles, or ‘God’s penny’ (Latin Argentum Dei) — a coin or token given to bind a bargain, ‘notably for the hiring of a servant’; the same gesture as the soldier’s ‘King’s shilling.’ At the hiring (or ‘mop’) fairs, held from the fourteenth century, servants were engaged by the year and their wages paid at its end. The same gesture enlisted a soldier, who ‘took the King’s shilling’ and was from that moment the King’s man.

And the wages themselves came after — long after. At the hiring fairs that ran from the fourteenth century into the nineteenth, labourers stood in the market wearing the badges of their trade — a shepherd’s twist of wool, a milkmaid’s pail — and were hired by the year. Board and lodging they had as they went; but their wages were paid at the end of the year’s service, in a lump, when the work was already done. A man gave a year of his life on the strength of a shilling and a promise. The pledge bound him at the start; the payment, if it came, came at the close; and between the two lay the long unpaid middle, the part the worker advanced on trust.

This is the strange asymmetry at the heart of the wage, and it shows most clearly by contrast. When you buy a loaf, you and the baker exchange in the same instant: the bread crosses the counter as the coin does, and neither of you need trust the other longer than a moment. The wage bargain is nothing like this. The worker delivers first — an hour, a day, a week, a year of labour — and is paid afterward, in arrears, once the period he has already worked is over. For the whole of that period he has given value and received nothing but a promise. He is, in the plainest sense, a lender: he advances his labour to his employer and waits to be repaid.³3In an ordinary sale, payment and delivery are simultaneous. The wage contract is almost unique in requiring one party — the worker — to perform in full before the other pays. Wages paid after the work are said, tellingly, to be paid ‘in arrears.’

No one calls it lending, because the loan is invisible, uncounted, and bears no interest, and because the worker has no choice but to make it. But a loan it is. Every Friday’s pay clears a debt the worker ran up from Monday; and the following Monday he begins to run it up again. The balance is never in his favour and never settled for long: he is permanently his employer’s creditor, by an amount equal to all the work he has done since he was last paid. The wage does not precede the labour, the way a price precedes a purchase. It follows the labour, always a step behind it.

He does not earn his wage and then collect it. He lends his labour, and waits.

To be a creditor is to bear a risk, and the worker bears his with almost no protection. For most of industrial history the first risk was that he would not be paid in money at all. Under the truck system the millowner and the mineowner paid in tokens or scrip, redeemable only at the company’s own shop — the ‘tommy shop’ — where the prices were set by the very hand that paid the wage; the worker’s labour came back to him as overpriced goods, or as credit in a ledger he could not check. It took a century of legislation — in Britain the Truck Acts, the first of them in 1831, which required that wages be paid in ‘the current coin of the realm’ — merely to establish that a wage must be paid in money the employer did not himself control.⁴4The truck system — payment in scrip or goods redeemable at the company store (the ‘tommy shop’; in Mexico, the tienda de raya). The Truck Act 1831 required payment in ‘the current coin of the realm’; its protections were extended to nearly all manual workers in 1887.

The deeper risk was, and remains, that the employer cannot pay, or will not. When a firm fails, the workers who have already done the week’s or the month’s work do not stand first in the queue; they line up as creditors among the others, claimants upon an estate that may have nothing left for them — having lent their labour to a debtor now insolvent. The merchant who sells a company its coal is owed a debt and knows it, and secures himself accordingly. The worker who sells it his labour is owed a debt too, but is rarely told so, and almost never secured. He has extended credit without being told he was a lender, and learns it only when the credit goes unpaid.

Because the worker performs first, the simplest way to cheat him is the oldest: take the work and withhold the pay. We have a flat little phrase for it now — wage theft — and measured in money it is among the largest thefts in the country. When employers shave hours, deny overtime, pocket tips, or simply never send the final cheque, they are taking from people who have already done the work and cannot take it back. The Economic Policy Institute reckons that the wages recovered for victims in a single recent year came to roughly three times everything taken in all the robberies reported that year; the true total, most of it never recovered, runs into the tens of billions.⁵5Economic Policy Institute estimates of wage theft — unpaid overtime, off-the-clock work, stolen tips, misclassification, withheld final pay. Sums recovered in 2012 (about $933 million) ran near three times all reported U.S. robberies; fuller estimates reach the tens of billions a year.

And where the law has made the plain theft harder, the advance has instead been turned into a market. A young industry now offers to lend the worker his own earned-but-unpaid wages, a few days early, for a fee — ‘earned wage access,’ it is called, as though reaching what one has already earned were a service one might be sold. The worker’s standing loan to his employer, the invisible credit at the centre of the wage, has at last been noticed by finance — not in order to abolish it, but in order to charge him for it. He pays, now, to borrow back his own labour.

The man with a gun takes your wallet. The employer takes the week you cannot un-work.

So we come back to the word, which was right all along. A wage is a wager. The worker stakes his labour — hours of a life he cannot have again — on the promise of a payment he can collect only once the labour is spent; and, like any bettor, he may find when he comes to collect that the house is empty or the game was rigged. The much-praised freedom of free labour, the freedom to sell one’s work to whom one will, holds inside it a freedom less often praised: the freedom to be cheated by the one you worked for, and to stand in the queue of creditors when he fails. The medieval serf was bound, and protected; the modern worker is free, and exposed. He is paid, when he is paid, for a trust he was never told he had extended.

Plate i found in the wage a measure of value, and asked whether it could be just. Plate ii found in it a measure of time, and watched the clock that made the hour saleable. This plate finds a third thing in it, plainer and more human than either: a measure of trust — and the trust, unlike the value and the time, is furnished wholly by the weaker party. A wage is the price of the work, reckoned in money and paid in time; but before it is any of that, it is a promise — made by the strong to the weak — that the work already done will be paid for. Every payday is that promise kept. The long argument over the wage is, at the last, an argument about a debt: who owes it, who carries its risk, and what is owed to the one who, the work being finished, must wait.

The wage is so nearly universal now that a working life without one is hard to picture; to be without a wage is to be in trouble, and we reach at once for the word unemployed. But unemployment is a young idea. For most of the human past, and for most people in it, a livelihood did not run through an employer at all. A family lived from the land — a strip in the open field, a beast on the common pasture, pigs turned into the lord’s wood, fuel cut from the turf, the right to glean what the harvest left behind.¹1On the open-field system and the rights of common — pasture, estovers (wood), turbary (turf), gleaning — by which households drew a subsistence from land they did not individually own: J. M. Neeson, Commoners (1993); E. P. Thompson, Customs in Common (1991). These were not charity and not wages; they were rights — customary, often unwritten, but real, attaching to the place and the person — and on them a household could subsist without ever selling its labour to anyone.

Such a life was neither easy nor equal. It was hedged by its own customs and its own lords; it could be bitterly poor, and it bound a person to a place. But it had one feature the wage does not: it required no buyer. A commoner who hired himself out did so to add to his living, not to have one at all; if no one hired him this week, he did not starve, because the land was still there and his right to it still held. The wage sat at the edge of his life, not at its centre. To move it to the centre — to make the wage the single thing standing between a person and hunger — something first had to happen to the land.

What happened to the land was enclosure. Over centuries, and then in a great rush, the open fields and commons of England were fenced, hedged, and parcelled into private property, and the rights of common that had hung upon them were extinguished. The first wave, in the Tudor century, was driven by wool: it paid to turn arable strips and shared pasture into private sheep-runs, and so, as Thomas More wrote in 1516, the once-meek sheep had grown so great a devourer that it would ‘eat up and swallow down the very men themselves’ — the husbandmen thrust out, whole villages pulled down, to make room for flocks.²2Thomas More, Utopia (1516), Bk. I, on enclosure for sheep. The later, larger wave was statutory: between 1604 and 1914 Parliament passed over 5,200 Inclosure Acts, enclosing roughly 6.8 million acres — better than a fifth of England.

The second wave was slower, vaster, and entirely lawful. Between the seventeenth century and the nineteenth, Parliament passed more than five thousand Inclosure Acts, enclosing some six and three-quarter million acres — better than a fifth of England — each act a stroke of the pen that converted common land into private title and dissolved the rights by which the poor had lived upon it. There is a real case for enclosure: consolidated private fields could be worked far more efficiently than scattered strips, and the harvests that fed a fast-growing nation came in part from it. But for the commoner the improvement was also an eviction. Stripped of the land and the rights that had been his living, he had left to sell only one thing — his labour — and from that day he had to find a buyer for it, or go under.

The commons did not require a buyer. The wage does.

Enclosure pushed people off the land; the law then saw to it that they could not simply roam free of work. The dispossessed who took to the roads — seeking harvest work, or alms, or anywhere to be — were met not with relief but with the whip. A long line of Tudor statutes against ‘vagabonds and sturdy beggars’ made being poor and masterless a punishable condition: the able-bodied man without a master could be whipped, branded, set in the stocks. The most savage of them, an Act of 1547, decreed that a stubborn vagabond might be made a slave for two years and driven to work by beating or chaining — a law so brutal it proved unenforceable and was repealed within three years, but whose intent was unmistakable.³3The Tudor ‘bloody legislation against the expropriated’ (Marx’s phrase, Capital I, ch. 28): the vagrancy statutes of 1494–1597, of which the Vagabonds Act 1547 (1 Edw. 6 c. 3), prescribing two years’ bondage for the able-bodied vagrant, was the harshest; it was repealed in 1550. To be without work was not a misfortune to be cushioned; it was a fault to be corrected.

The point of the whip was not cruelty for its own sake but instruction: there was to be no living outside the labour market, no honourable idleness for the poor, no third thing between a master and a punishment. What enclosure had made physically necessary — that a man must sell his labour — the vagrancy laws made legally compulsory. Between them they did not so much offer the wage as ensure there was nothing else.

There was one celebrated attempt to keep a door open, and its fate has shadowed the argument ever since. In 1795, with bread prices soaring, the magistrates of Berkshire, meeting at Speenhamland, resolved to top up the wages of the poor out of the parish rates on a sliding scale tied to the price of the loaf and the size of the family, so that no labourer’s income — whatever his master paid him — would fall below what bread cost to live.⁴4The Speenhamland scale (1795) keyed relief to the price of the gallon loaf and the number of children. Condemned by the Royal Commission of 1834 (and later by Polanyi) for depressing wages and pauperising labour, its ill effects were judged greatly exaggerated by later historians (Mark Blaug, 1963; George Boyer). It was abolished by the Poor Law Amendment Act 1834. It was, in effect, a guaranteed minimum income — a wage subsidy two centuries before the phrase — and it spread across much of southern England.

What it did is still disputed, and the dispute is the whole modern quarrel in miniature. To its critics — the reformers of 1834, and a century later Karl Polanyi — it was a disaster: by making up the difference, it freed employers to pay beneath subsistence and heaped the rest onto the rates, so that wages sank, the labourer was pauperised, and his independence drained away. To later historians — Mark Blaug, George Boyer — the charge was overdrawn: the system mostly kept people from starving, steadied a countryside in crisis, and fitted the broken, seasonal rhythm of farm work. But the official verdict of the day was damning, and in reaction it produced the New Poor Law of 1834, which swept the subsidy away, raised the workhouse, and laid down the iron rule of ‘less eligibility’ — that relief must be made so much worse than the meanest paid work that no one would ever choose it. The open door was not merely closed; it was walled up and made a place of dread, precisely so that the wage, however poor, would always look the better bargain.

It took another century to say plainly what had happened, and it was Karl Polanyi who said it best. Land and labour, he argued in 1944, are not really commodities at all: a commodity is something produced in order to be sold, and labour is only another name for human beings, as land is only another name for nature. Neither was made for the market; both were driven into it. To run a whole society as one great self-regulating market, these ‘fictitious commodities’ had to be treated as though they were real ones — human life priced and sold by the hour like so much cloth — and a market that did that to people, left to itself, would grind them to pieces. And so, Polanyi held, society always struck back: the very century that freed the labour market also filled with factory acts and unions and poor relief — a ‘double movement,’ the market’s advance met by society’s defence.⁵5Karl Polanyi, The Great Transformation (1944), on land, labour, and money as ‘fictitious commodities’ and on the ‘double movement.’ The origin story is Karl Marx’s ‘so-called primitive accumulation,’ Capital I, part VIII.

Marx had told the origin story earlier, and harder. The wage-labouring class, he insisted, was not born of thrift — a few industrious souls saving while the rest squandered, as the comfortable legend ran. It was made, by force, through what he called primitive accumulation: the prising of the people from the land, enclosure and eviction and the bloody statutes, until there existed a great mass of men who were, in his phrase, free in a double sense — free to sell their labour as they pleased, and free of any property by which they might otherwise have lived. The liberty of the wage worker and his dispossession, he saw, were a single event told twice.

He was free to sell his labour, and free of anything else to live on. The two freedoms were the same.

Set the commons and the wage side by side, and the bargain is plain. The commons gave security without freedom: the commoner could not easily be starved, for the land was there — but neither could he easily leave, bound as he was to his place, his lord, the slow grip of custom. The wage gives the opposite: freedom without security. The wage worker may go anywhere, leave anyone, sell his labour to the highest bidder — a real and precious liberty the serf never knew — but he must find a buyer, week upon week, or fall, for there is no longer any land beneath him to break the fall. We did not simply lose the commons, nor simply gain the wage. We traded a floor for a door.

Nearly everything built since has been an effort to have both at once — to keep the freedom of the wage while restoring, by other means, the security of the commons. Poor relief, factory acts, social insurance, unemployment pay, the minimum wage: each is, in its way, an attempt to slide a floor back under the worker without binding him again to a master or a place. The boldest version of the idea is also among the earliest. In 1797, watching these same enclosures, Thomas Paine argued that because the earth had once been ‘the common property of the human race,’ everyone dispossessed by private property was owed compensation — not as charity but as a right — and proposed a national fund to pay every person a sum on coming of age and a pension ever after: a floor of money to stand in for the floor of land.⁶6Thomas Paine, Agrarian Justice (1797): the earth as ‘the common property of the human race,’ and a national fund — raised from a ground-rent on inherited land — to pay each person a lump sum at twenty-one and a pension from fifty, ‘as a right, and not as a charity.’ An ancestor of the basic-income idea. Two centuries on the same proposal returns under a new name, the universal basic income, and draws the same two replies it always has: that a floor a person can live on without working is the truest freedom a modern society can offer — and that it is merely Speenhamland again, a subsidy that lets wages fall and calls it kindness. Plate i asked whether the wage was just; plate ii, what it sells; plate iii, the trust it is staked upon. This plate asks the question beneath the others — whether it must be taken at all — and finds the answer was settled long ago, by the closing of every other door. The measure of a society may be how freely it can let that door open again.

We talk about wages as though they were facts of the same hard kind as weights and distances. There is, we feel, a rate for the job — the going rate — and the only real questions are whether a given employer pays it, beats it, or falls short. The wage, on this picture, is a reading taken off the work itself: the labour has a worth, and the wage reports it, faithfully or not. Nearly everything in this plate is an argument that the picture is false — that there is no single number written on the back of the work, waiting to be read, and that the search for ‘what the work is really worth’ is a search for a thing that was never there.

This is not to say that wages are arbitrary, or that anything at all can be paid. They are penned within limits, and the limits are real. But the limits are wide, and what happens between them is not measurement; it is a contest. To see this, one need only find the two walls of the pen — the lowest wage that is possible and the highest — and notice how much room lies between them.¹1That the terms of a bargain between two parties are not fixed by the goods alone but left indeterminate within a range, to be settled by bargaining, is an old result — F. Y. Edgeworth’s ‘indeterminacy of contract’ (Mathematical Psychics, 1881).

The lower wall is subsistence. There is a wage beneath which a worker simply cannot be had — not as a matter of fairness but of fact: paid too little to live, or to live near the work, or to keep returning to it, he is not hired at that price, because at that price he does not exist. The classical economists hardened this floor into a law. Ricardo named subsistence the ‘natural price’ of labour; Lassalle called it the ‘iron law of wages,’ the grim doctrine that pay must always sink back to the bare cost of keeping a worker alive and breeding his replacement. They were wrong that it was a law — across the last two centuries real wages climbed far above subsistence and stayed there — but they were right that it was a floor.²2The subsistence floor: David Ricardo’s ‘natural price’ of labour (Principles, 1817) and Ferdinand Lassalle’s ‘iron law of wages.’ The doctrine that wages must stay at subsistence was falsified by the long rise of real wages; the floor, however, is real enough. Below it, no labour.

The upper wall is the worker’s product. There is a wage above which an employer will not go — again as fact, not generosity: pay a man more than the value his labour adds to what is sold, and he is kept at a loss, and not kept long. The most that can be paid is, at the very limit, the whole of what the worker’s labour brings in; past that the bargain ruins the one who makes it. Between this ceiling and that floor lies the entire range of possible wages — and it is not a narrow space. The value a worker’s labour adds may be several times what it costs to keep him alive. Somewhere in that wide gap the actual wage sits; and nothing in the work itself says where.

There is no number on the back of the work. There is a floor, a ceiling, and a contest between.

What decides the place within the gap was named, once and for all, by Adam Smith. The wage, he saw, is settled by a ‘dispute’ between two parties whose strength is wildly unequal. The masters are fewer, and combine the more easily; the law, in his day, allowed their combinations and forbade the workmen’s; and, above all, the masters can wait. ‘In all such disputes,’ he wrote in 1776, ‘the masters can hold out much longer’ — a landlord or manufacturer might live a year or two upon the stock he had already laid by, though he hired no one, while few workmen could subsist a month, and scarce any a year, without their wages.³3Adam Smith, The Wealth of Nations (1776), Bk I, ch. 8: the masters fewer and combining more easily, the law on their side, and able to ‘hold out much longer’; and their ‘tacit, but constant and uniform combination, not to raise the wages of labour above their actual rate.’ In the long run the two may need each other equally; but the worker’s need is immediate, and the master’s is not. Hunger keeps poor time, and the side that can afford to wait names the price.

This is the quiet engine beneath the ‘market wage.’ We seldom hear of the masters combining, Smith observed, only because their combination is the silent and settled state of things — ‘a sort of tacit, but constant and uniform combination, not to raise the wages of labour above their actual rate’ — whereas the workers’ combinations, being loud and desperate, are the ones the world notices and the law strikes down. The wage that results is called the rate of the market, as though no one had set it. But it has been set — by the steadier nerve and the fuller larder of the stronger side.

If the wage is fixed by who can hold out, then the worker’s one hope of moving it is to win the power to hold out too — and that is what the combination is for. A union is, at bottom, a machine for waiting: a strike fund lets workers refuse the offered wage for a week or a month, as the master can; and a pledge among them not to undercut one another denies him the cheaper hand he would otherwise hire in their place. Combination is the attempt to make the two larders equal. The law grasped this at once, and for a long age forbade it: the Combination Acts of 1799 and 1800 made it a crime for workmen to band together to raise their wages or shorten their hours — while the parallel ban on the masters’ combinations, exactly as Smith would have foretold, was in no recorded case ever enforced.⁴4The Combination Acts 1799–1800 criminalised workers’ combinations (the matching prohibition on masters’ combinations went unenforced); repealed in 1824 through the campaign of Francis Place and Joseph Hume, re-restricted in 1825 after a wave of strikes. Unions remained in contested legal territory until the Trade Union Act 1871. Repealed in 1824, re-restricted the very next year after a wave of strikes, the right to combine spent the whole nineteenth century in contested legal ground before the union was at last wholly lawful.

What the union does — where it works — is not to discover the true worth of the labour, which has no single worth to discover, but to shift the wage within the range, toward the ceiling and away from the floor, by lending the workers’ side the one thing it lacked: the power to wait. Its critics reply that it can shift other things too — that by raising its own members’ pay it may raise prices, or shut out the workers it does not speak for, the young and the outside, who would gladly have taken the lower wage. The objection is real, and unsettled. But it too is an argument about who shall win the contest, and on what terms — not about a number inscribed in the nature of the work.

Economics has never quite agreed with itself about any of this, and the disagreement is old. The theory that has ruled the last century says the gap is an illusion — that in a properly competitive market the wage is driven, precisely, to the worker’s marginal product, the value added by the last worker hired, so that the indeterminacy closes and the wage becomes a determinate, measurable thing after all. Its founder, John Bates Clark, went further and called the outcome just: the distribution of income, he wrote in 1899, is governed by ‘a natural law’ that gives to every agent of production ‘the amount of wealth which that agent creates.’ If true, it is a kind of consolation — you are paid exactly what you make — and it is the picture in most of the textbooks.⁵5John Bates Clark, The Distribution of Wealth (1899): in competition the wage equals labour’s marginal product, which Clark cast as a ‘natural law’ rewarding each with what it creates. Later economists (e.g. George Stigler) faulted this ‘productivity ethics’: the theory explains the demand for labour, not what anyone deserves.

But it holds only if the market is the frictionless, many-buyered thing the model assumes, and real labour markets are not. Where employers are few, or workers tied down, or knowledge poor, the employer commands a power the model denies him — the power of the only, or the largest, buyer, which Joan Robinson named monopsony in 1933 — and can pay beneath what the worker adds without losing him, because the worker has nowhere better to go. The plainest sign came in 1994, when David Card and Alan Krueger found that a rise in New Jersey’s minimum wage had not destroyed the fast-food jobs the competitive theory said it must — impossible, if wages already sat at the worker’s product; entirely expected, if employers had been paying less.⁶6Joan Robinson coined ‘monopsony’ — a single dominant buyer — in The Economics of Imperfect Competition (1933). Card and Krueger’s New Jersey minimum-wage study (1994) found no resulting job loss, consistent with employer wage-setting power; the result was, and remains, contested (Neumark and Wascher, 2000). See also Alan Manning, Monopsony in Motion (2003). The finding was fought over, and is still; but it reopened the older view, in which the wage is not read off the work but won or lost in the gap. Which theory holds is not settled, and likely differs from one market to the next. What is hardly in doubt is that the determinate wage — the one with a single right value — is the special case, not the rule.

You are paid what you make, says one theory; you are paid what you can get, says the other. Only the second has ever explained why the same work pays two prices.

If this is right, the wage is not a fact but a settlement: the point, within a wide and indeterminate range, at which the contest between the two parties has for the moment come to rest — and it moves the instant their relative power moves. The same work, of identical worth, pays more in a tight labour market than a slack one, more where the union is strong than where it is broken, more where a minimum wage shores up the floor than where none does, more where a worker can leave than where one company owns the only mill in the valley. Not one of these alters what the labour is ‘worth,’ because worth was never the variable. They alter who can hold out — and so they alter the wage.

Here the folio’s first question returns, transformed. Plate i asked whether a wage could be just, and followed the long effort to say what justice in a price might mean. The answer this plate gives is harder than the schoolmen hoped: there may be no single just wage to find, because there is no single wage the work requires — only a range, within which ‘just’ is one of the things being fought over, and within which power, left to itself, will keep the wage near the floor. The minimum wage, the union, the law of combination are not discoveries of the true rate; they are thumbs laid on the scale, attempts to settle the contest a little above where power alone would settle it. A wage measures the value of work, we began by saying; it measures its hours; it is the trust the work is staked upon; it is the thing we are compelled to seek. And it is, this plate adds, the prize in a contest — so that the truest measure of a labour market is not whether it pays the worth of the work, for no such worth exists, but how nearly equal it has made the two who bargain over the wage.

Ask whether someone works, and you are asking, without quite knowing it, whether they are paid. The man at the desk works; the woman raising three children at home ‘does not work,’ though her day is longer and harder. The retiree ‘no longer works’; the volunteer ‘isn’t working’; the parent at home is ‘off work’ — each phrase quietly equating work with the wage, and treating everything done without one as some other kind of thing. The wage, then, is not only a price attached to work; it is the badge that turns activity into work in the first place. What it touches becomes labour, counted and seen; what it does not touch becomes love, or duty, or rest — anything but work.¹1How the household’s labour came to be reclassed as economically ‘unproductive’ over the nineteenth century: Nancy Folbre, ‘The Unproductive Housewife: Her Evolution in Nineteenth-Century Economic Thought’ (Signs, 1991), and the ‘cult of domesticity’ that placed the home outside the realm of economic interest.

It was not always so plainly the case, and it did not happen by accident. Across the nineteenth century, as paid work moved out of the household and into the mill and the office, the household’s own labour — which had always been work, and hard — was slowly redescribed as something outside the economy. The housewife, once a productive member of a producing household, became in the new economic language ‘unproductive’: a dependent, a consumer of the wage rather than a maker of anything. The line between work and not-work was drawn — and it was drawn around the door of the home.

The line had a wage of its own, and a name: the family wage. Through the nineteenth and twentieth centuries, workers and their unions fought for a wage high enough that a single earner could keep a whole household — and the earner they pictured was the man. A ‘family wage’ was a man’s wage sufficient to support a wife and children, so that the wife need not, and ideally could not, take paid work of her own. Seen one way, this was a genuine and decent demand: a wage a family can actually live on is something many would still wish for, and its slow vanishing — the arrival of the two-earner household merely to manage what one wage once managed — is widely and rightly mourned.²2The family or ‘breadwinner’ wage carried a genuine demand for a living income alongside a gendered division of labour: the man as waged worker, the woman as unpaid homemaker dependent on his wage. The male-breadwinner model, though much eroded, remains culturally engrained.

But the same demand drew the line this plate is about. It made the man the worker, whose labour was work and was paid; and it made the woman the homemaker, whose labour was not work, was not paid, and was recast as the natural outflowing of her love and her duty. Her cooking and cleaning and child-rearing — without which his wage-earning could not have happened at all — disappeared, in the official view, into the private warmth of the home, where by definition nothing economic was held to occur. The father’s wage, as one writer has put it, hid the mother’s work.

And it was work the whole edifice depended on. No worker arrives at the workplace ready-made, out of nowhere, fed and clothed and rested and grown. He has to be made — daily, by meals cooked and clothes washed and a home kept warm; and across generations, by children borne, raised, taught, and sent out in their turn to be workers too. Economists since Marx have called this the reproduction of labour-power: the labour that produces and maintains the labourer, the one input without which no other input exists.³3The ‘reproduction of labour-power’ is Marx’s (Capital, vol. I); the worker’s capacity to labour must be restored daily and renewed across generations. Feminist theorists — Mariarosa Dalla Costa, Silvia Federici, Leopoldina Fortunati, the Bielefeld school — argued that this reproduction is performed largely by women’s unpaid domestic and care work, a blind spot in Marx, who imagined the worker’s renewal mainly as commodities bought. It is real, skilled, exhausting, unceasing work — and in the world the family wage built, it was done overwhelmingly by women, for nothing. The factory could not open in the morning if the home had not done its unpaid work the night before.

Here is the heart of the matter, and its strangest feature. The unwaged labour of the home is not outside the economy — a soft private margin to the hard public business of earning. It is the foundation the earning stands on: vast in quantity, and prior in logic, since the waged worker is himself its product. Yet because no wage touches it, it does not appear — not in the wage, not in the ledgers, not in the great figure we call the size of the economy. The thing the system most depends upon is the thing it is built not to see.

They say it is love. We say it is unwaged work.

In 1972 a circle of feminists gave the hidden labour a demand, and the demand was deliberately startling: wages for housework. Meeting in Padua, Mariarosa Dalla Costa, Selma James, Silvia Federici, and Brigitte Galtier founded a campaign to insist that the work of the home be recognised and paid — by the state, by capital, by whoever had been profiting from its being free.⁴4The International Wages for Housework Campaign grew from the International Feminist Collective, founded in Padua in 1972; its early text is Dalla Costa and James, The Power of Women and the Subversion of the Community (1972). The demand was largely a ‘lever’ — a way to make the labour visible — as Federici’s title Wages Against Housework (1975) suggests. The point was only partly the money. Chiefly it was to compel a recognition: by demanding a wage for housework, they made it impossible to keep pretending housework was not work. The wage, which had drawn the line, was turned about and used to rub it out — for if the labour deserved a wage, then a wage-less labour was exactly what it had always been, and the word for that is unpaid, not free. Their slogan put it with great economy: what the world calls love, they answered, is unwaged work.

The demand was contested from the start, and not only by its enemies. Among feminists it was argued that paying women to keep house might bind them only more tightly to it — gild the cage rather than open it — and that the deeper aim was not a wage for the role but freedom from it. Federici half-conceded as much in the very title of her best-known essay, Wages Against Housework: the wage was a lever to make the labour visible and refusable, not a wish to be paid forever for scrubbing the floor. Yet whether one took the demand at its word or as a provocation, it had done its work the instant it was spoken — by making the invisible labour impossible to un-see.

What the wage does to our idea of work, a second instrument does to our idea of the economy, and far more officially. Gross domestic product — the single number by which nations measure their size and their success — counts only what crosses a particular line the statisticians call the production boundary, and that line runs almost exactly where the wage’s does. A meal cooked in a restaurant counts; the same meal cooked at home does not. A child minded in a paid nursery counts; the same child minded by its parent does not. Nursing in a hospital counts; nursing a dying parent through the night does not.⁵5The production boundary of the System of National Accounts excludes unpaid household and care work. A. C. Pigou’s paradox (The Economics of Welfare, 1920): a man who marries his housekeeper lowers the national income. The ILO (2018) estimates that unpaid care work, valued at the minimum wage, would equal some 9% of global GDP — about US$11 trillion — with women performing over three-quarters of it. The economist Arthur Pigou marked the paradox a century ago: a man who employs a housekeeper adds her wage to the national income, but should he marry her, and her work go on unchanged and unpaid, the national income falls. The labour did not vanish. Only its wage did — and with the wage went its place in the count.

The omission is not the simple scandal it can be made to sound. The accounts were built, in the mid-twentieth century, to measure market activity for the pressing purposes of taxation, money, and war; their architects, Simon Kuznets among them, knew and said that the figure was never a measure of welfare or of all useful work, and that putting a value on unpaid labour was genuinely hard — at what wage, the worker’s foregone earnings or a servant’s, and reckoned how? These are real difficulties, not mere excuses, and statisticians now build ‘satellite accounts’ to estimate the missing labour rather than ignore it. But estimate it they must, for it is not small. The International Labour Organization reckons the world’s unpaid care work, valued only at the minimum wage, at some nine per cent of global output — around eleven trillion dollars a year — with women doing more than three-quarters of it. A tenth of the economy, by the economy’s own arithmetic, is work it has chosen not to count.

The labour did not vanish. Only its wage did.

Step back, and the wage shows a second face. It is a price, yes — for the value of work, as the first of these plates had it; for hours; for trust; paid under compulsion; sized by power. But it is also a boundary: a line that sorts the world into work and not-work, economy and not-economy, the counted and the uncounted — and the sorting is a decision, not a discovery. Nothing in the labour of feeding one’s own child is less real, less skilled, or less needful than the labour of serving a stranger the same food for money; the only difference is the wage, and the wage is just what has been used to tell them apart. What falls outside the line is not less work. It is only less visible, less paid, and less counted — and, as it happens, more often women’s.⁶6Marilyn Waring, If Women Counted: A New Feminist Economics (1988), showed how the global accounts render women’s unpaid work — and much of nature — ‘of no account.’ She later cautioned that simply folding such work into a GDP that also counts armaments and the felling of forests as gains might be the wrong remedy — that the measure itself, not only its boundary, is in question.

Where to draw the line is, then, a real question, with serious answers on more than one side — whether to pay for care or to provide it, whether to fold it into the headline figure or to keep a truer one beside it, whether naming love as labour honours it or cheapens it. The folio has watched the wage measure value, sell time, stake trust, compel obedience, and ride the seesaw of power; this last plate finds it doing something quieter and larger than any of these — deciding, by the mere fact of what it does and does not touch, what shall count as work at all, and so what shall count as the economy, and so what shall be seen. A society shows itself not only in how it pays for work, but in how much real work it has arranged not to call work. The line is no law of nature. It was drawn, and it can be drawn again.

Economics begins, in its simplest form, with a promise: that any market will clear if only the price is free to move. Too much wheat, and the price of wheat falls until the surplus is bought; too few buyers, and the seller lowers his asking until one appears. Apply the same logic to labour, as the classical economists did, and unemployment ought to cure itself — for when workers go begging, their wage should fall, cheap labour should tempt employers to hire, and the slack should be taken up, the labour market clearing like any other. It was a tidy promise, and the twentieth century broke it in plain sight: in the depths of slumps, with millions wanting work, wages did not fall nearly enough to clear the market, and the unemployment did not cure itself.¹1Downward nominal wage rigidity. Plot year-to-year pay changes and a spike appears at zero — a fifth or more of workers held at last year’s figure. Keynes (1936) called downward rigidity ‘the normal case.’ It is real but partial and historically variable: in the interwar slump wages were cut about as often as raised, modern wages are stickier, and its exact extent is still argued.

The reason is a stubborn fact: nominal wages almost never fall. Plot the year-on-year changes in workers’ pay and a strange pile-up appears exactly at zero — a great many workers whose wage is frozen, held at last year’s figure, and very few whose pay is openly cut. Wages rise; wages stall; but wages seldom retreat. The number behaves not like a price free to move either way but like a thing that can climb and cannot descend. Economists name this downward rigidity, which states the fact without explaining it. The explanation is stranger than the fact.

There is a small machine that does exactly what the wage does, and lends it its truest name: the ratchet. A toothed wheel or bar and a spring-loaded pawl resting against the teeth — the thing may move freely one way, clicking forward tooth by tooth, but the instant it tries to move back, the pawl drops into a tooth and locks it. Motion in one direction is easy; in the other, all but impossible. The wage is a ratchet of just this kind. It climbs, tooth by tooth, through the good years, and when the bad years come and the market leans on it to descend, it holds.

And a price that cannot fall has a consequence, for the pressure to lower it does not vanish when the wage refuses to move; it is diverted. If the cost of labour will not come down, then in a slump the employer reduces not the price he pays but the number he pays it to: the adjustment other markets make by lowering a price, the labour market makes by shedding people. The rigidity of the wage is paid for in jobs. But that is the consequence; the cause is what we must look at first — why the wage will not bend.

Every other price bends to the market. The wage digs in.

Why will it not bend? The deepest answer was found not by theorising but by asking. In the early 1990s the economist Truman Bewley, dissatisfied with the tidy models, went and interviewed more than three hundred employers, managers, union officers, and counsellors of the unemployed, and put to them the plain question: in a downturn, why not cut pay rather than cut jobs? The answer that came back, again and again, was not about productivity curves or the supply of labour. It was about morale.²2Truman Bewley, Why Wages Don’t Fall During a Recession (1999), built on some three hundred interviews. The dominant reason managers gave for not cutting pay was morale: the saving is outweighed by the damage to effort and loyalty, so firms prefer layoffs — which remove the aggrieved worker — to cuts, which aggrieve everyone who stays. A pay cut, the managers told him, devastates a workforce — not in proportion to the money lost, but out of all proportion to it — because employees receive it not as an adjustment to conditions but as an insult: a judgment that they are worth less, valued less, owed less than before. And a workforce that feels cheated does not work as it did: effort slackens, loyalty curdles, the best ones leave, and the small saving on the payroll is swallowed many times over by the damage to the spirit of the place.

This is why employers so often prefer the seemingly crueller course of laying people off. A layoff, brutal as it is, removes the aggrieved worker from the premises and leaves the morale of those who remain intact — even, at times, relieved. A pay cut keeps everyone and aggrieves them all. The logic is cold, but it discloses something warm at its centre: that the wage is bound up with a worker’s sense of worth so tightly that to lower it is to wound, and that the employer, needing willing rather than sullen hands, cannot afford the wound. The wage is not only what the worker costs. It is what the worker is told he is worth, and no number is read more personally.

There is a piece of evidence that shows just how far the wound lies in the telling rather than the sum. Suppose inflation is running at twelve per cent and an employer raises wages by five: the worker is, in real terms, seven per cent poorer than before, his money buying noticeably less. Now suppose instead that prices are flat and the employer cuts the wage by seven: in real terms, the very same loss. Yet people judge the two cases utterly differently. The raise-amid-inflation is felt as a raise, and accepted; the outright cut is felt as a robbery, and condemned as plainly unfair.³3Surveys find the asymmetry stark: Kahneman, Knetsch and Thaler (1986) report that a 7% pay cut when inflation is zero is judged far less fair than a 5% raise when inflation is 12% — though the two are equivalent in real terms. See also Shafir, Diamond and Tversky, ‘Money Illusion’ (1997), and Tobin’s (1972) argument that mild inflation ‘greases the wheels’ of the labour market. Economists call this the money illusion — the worker’s eye fixed on the face value of the wage rather than its purchasing power — but it is less an illusion than a moral instinct. A cut is something an employer does to you, an act with an author and a victim; inflation is weather, an impersonal drift with no one to blame. The first is an injury; the second, only a misfortune.

This single quirk has a vast and largely benign consequence: it is the reason a little inflation eases the workings of the labour market. Because wages can be lowered in real terms by stealth — held flat while prices drift upward — but not in nominal terms by decree, an economy with gently rising prices can quietly bring real wages down where a slump requires it, and spare itself the unemployment that an attempt at open cuts would bring. The wage’s refusal to fall is got around, humanely, by never asking it to fall in the only terms it watches.

It was John Maynard Keynes who saw that this one rigidity was enough to overturn the classical promise. If the wage will not fall, he argued in 1936, then the labour market has no mechanism to clear: nothing drives the price of labour down to the level at which all who want work are hired, and so a slump can settle into a grim equilibrium of idle hands and willing workers — unemployment that persists not because the jobless are too choosy or too dear, but because the wage, across the whole economy, will not descend to meet them.⁴4J. M. Keynes, The General Theory of Employment, Interest and Money (1936): with money wages rigid downward, the labour market need not clear, and involuntary unemployment can persist. Keynes’s own remedy was not to force wages down — he thought it futile and destructive — but to raise demand to meet the rigid wage. His own remedy, notably, was not to force wages down — he thought the attempt both futile and destructive — but to lift demand until it met the rigid wage, and the idle hands were wanted once more.

The theorists who came after filled in why the wage holds, and their answers, though dressed in the cool language of models, all circle the same human core. The efficiency-wage school showed that an employer may pay above the market on purpose and never cut, because a well-paid worker who fears to lose a good thing works harder and shirks less — so that a little unemployment becomes, perversely, the discipline that keeps the employed honest; others modelled the wage as a gift exchanged for loyalty, above-market pay repaid in above-ordinary effort.⁵5Efficiency-wage theory: Shapiro and Stiglitz (1984) modelled unemployment as a ‘worker-discipline device’ — firms pay above market so that the threat of job loss deters shirking; Akerlof (1982) modelled the wage as a ‘partial gift exchange,’ above-market pay reciprocated with above-minimum effort. Each is a fragment of the same picture Bewley’s managers drew in plainer words: that the wage buys not only hours but willingness, and that willingness is wrecked by a cut. None of this, it should be said, makes the rigidity total. Wages do fall sometimes — for those who change jobs, for the newly hired, and in the harshest depressions even for those who stay; in the slumps of the nineteenth century pay was cut about as readily as it was raised. The wage is sticky, not frozen, and how sticky is argued still. But stickier than other prices it plainly is, and that difference is the whole of the puzzle.

It is, of all prices, the one with a conscience.

Set all the explanations side by side and they converge on a single sentence: the wage will not fall because it means too much. Other prices are only prices — the meeting of supply and demand, indifferent to the dignity of the thing priced. The wage alone is also a verdict, a relationship, a daily statement to a person of how much they are valued; and a statement of that kind cannot be revised downward without insult, however the market may be moving. The ratchet is the shape this meaning takes when it is drawn: the wage rises with good fortune and locks against bad, because to unsay what it has said — to tell a worker he is worth less than last year — is a thing that workers and employers alike will go to great lengths to avoid.

But a conscience can have its victims, and this one does. The same rigidity that spares the employed the insult of a cut loads the whole weight of a downturn onto another group — those who lose their jobs, and those who cannot find one. If wages fell a little for everyone, more might keep working; because they do not fall, fewer do, and the cost is borne in full by the unemployed, the young, the outsiders who would gladly have worked at the lower wage the insiders refused.⁶6The insider–outsider theory (Lindbeck and Snower, 1988): incumbent ‘insiders,’ protected by the costs of replacing them, can keep their wages high, so the burden of adjustment in a downturn falls on ‘outsiders’ — the unemployed, who cannot bid their way in. The wage’s dignity is real, and so is its price, and the price is paid by those it has shut out — who receive no verdict at all, because they receive no wage. The folio has watched this one number measure value, sell time, stake trust, compel obedience, settle by power, and draw the bounds of work; here it does the strangest thing of all. It declines to fall, out of respect for the worker — and lets the unemployment it thereby causes fall, instead, upon the workers it does not have.

Adam Smith drew the distinction at the very start of economics, and it has been half-forgotten ever since. Every thing, he wrote, has two prices. There is its nominal price — the sum of money it fetches — and there is its real price, which is something else entirely: “the toil and trouble of acquiring it.” Money is only the veil; behind it lies the true cost, paid in effort and time. “Labour,” he concluded, “is the real measure of the exchangeable value of all commodities” — and, he added, the first one: it was not by gold or silver but by labour that all the wealth of the world was originally purchased.¹1Adam Smith, Wealth of Nations (1776), Bk I ch. 5, ‘Of the Real and Nominal Price of Commodities.’ Smith’s claim that labour determines value — the classical labour theory — was later displaced by the marginalist view that value is subjective. What survives, and is used here, is the milder and sturdier claim: that labour-time is a measure of value, a yardstick, whatever sets the value being measured.

A caution is owed at once, for Smith meant more than this and overreached. He thought labour not only measured value but created it — the ‘labour theory of value’ — and on that point the economists who followed corrected him: a thing’s worth, they showed, is set by desire and scarcity at the margin, not by the sweat in it, or a diamond would be cheaper than water. So let us keep only the sober half of Smith’s insight, which no correction has touched: that whatever fixes the value of a thing, the cost of it to you, the buyer, is measurable in your labour — in the portion of your working life you must surrender to have it. Labour need not be the source of value to be its truest ruler.

Here is what that ruler means for a person who works. Your wage is a rate of exchange — so much money for so much of your life — and, like any exchange rate, it runs both ways. It turns your hours into money; and it turns money, all money, back into hours. Read the second direction and every price in the world quietly changes its dress. The coat does not cost two hundred pounds; it costs, at your wage, a day and a half of your life. The car does not cost a year’s salary; it costs a year. The house costs a decade. This is the time-price of a thing — the hours of work it takes to buy it — and it is a truer figure than the money price, because money is a shifting yardstick (inflation stretches it, currencies shrink it) while an hour of a human life is, near enough, a constant thing, the same for the Babylonian and for us.²2The ‘time-price’ of a good — the hours of labour required to earn its purchase — strips away the drift of money and currency. It is, in Smith’s sense, money turned back into the toil it stands for: as he put it, what we buy with money is ‘purchased by labour’ as surely as what we make with our own hands; the money merely ‘saves us this toil.’

Once seen, it cannot be unseen, and it changes the question one asks of a price. The question is no longer “can I afford this?” — a question about the balance in an account — but “is this worth the hours of my life it will cost?” — a question about the life itself. Every purchase is the spending of a portion of one’s allotted time, converted into money for convenience and then converted back into goods. The wage is merely the rate at the window where the two currencies, money and life, are exchanged.

The real price of a thing is the part of your life it costs.

Hold this ruler against history and a remarkable thing appears — the great, under-told fact of the modern age. Measured in money, prices only ever seem to rise; measured in life, in the hours of work a thing costs, the material world has grown astonishingly, almost unbelievably cheaper. The economist William Nordhaus showed it with the oldest human want there is: light. He traced the labour-price of illumination down the centuries — the hours of work it took to buy a fixed quantity of light — and found a collapse beyond anything the money figures suggest.³3William Nordhaus, ‘Do Real-Output and Real-Wage Measures Capture Reality? The History of Lighting Suggests Not’ (1996). The labour to buy 1,000 lumen-hours fell from roughly 41 hours (a Babylonian oil lamp) to about 5 hours (1800), to 13 minutes (1900), to under a second today. His larger point: ordinary price and real-wage indices badly undercount such gains, because they cannot see a new good arrive. To light an evening by a Babylonian oil lamp cost something like forty hours of labour; by a tallow candle around 1800, some five hours; by a gas or kerosene lamp in 1900, about thirteen minutes; and today, by an electric bulb, a fraction of a second. An hour of work now buys roughly three hundred and fifty thousand times the light it bought in Babylon. The labour that once purchased a single hour of light will now purchase a lifetime of it.

And it is not only light. The same collapse, if less extreme, has overtaken nearly every made thing — food, cloth, books, sugar, glass, a mile of travel, a word sent across the world. Each costs a small fraction of the life it cost a century or two ago. This is the deep reason the modern person is rich in a way no king once was: not because the numbers in the account are larger, but because the hours behind each thing are so few. Read in the currency of life, the made world has been, quietly and continuously, marked down.

But the same ruler tells a second story, harder and stranger, and an honest measure must report both. Not everything has cheapened. Some of the things that matter most — a house, a year of schooling, a course of medical care, the minding of a child or an aged parent — cost, in many places, more hours of life than they did a generation ago, not fewer. The reason was named by the economist William Baumol and is half a paradox: these are the things whose making cannot be sped up.⁴4Baumol’s ‘cost disease’ (Baumol and Bowen, 1966; The Cost Disease: Why Computers Get Cheaper and Health Care Doesn’t, 2012): in labour-intensive sectors — care, teaching, the arts — productivity cannot rise much, yet wages must keep pace with the rest of the economy, so their cost climbs. It is a major cause, though not the only one; affluence and administrative growth also drive service prices up. A string quartet takes as many player-hours to perform today as it did in 1800; a nurse can tend only so many patients, a teacher teach only so many children, however clever the age becomes. Where the loom and the chip multiply what an hour of work can make, the bedside and the classroom cannot — and so, as wages rise across the whole economy, the cost of these unspeedable things rises with them, climbing in hours even as the made things fall.

This is the quiet arithmetic behind a feeling common in our time: that of being materially flooded and yet somehow pressed — surrounded by cheap abundance while the few things that anchor a life, a home and health and the raising of children, drift further out of reach. The made things grew cheap; the cared-for things, and the things there is only so much of, grew dear. An age can be, at once, the richest there has ever been in goods and the most anxious in the things that goods cannot be turned into.

The made things grew cheap. The cared-for things grew dear.

For all its power, the ruler measures only what it can reach, and an honest hand must mark where it stops. The time-price tells, better than any money figure, how dear the buyable world is in hours of life — but much of a life is not buyable. It says nothing of the hours themselves once they are free: of leisure, of safety, of clean air and quiet, of love and meaning, none of which has a price in labour because none is for sale.⁵5The time-price reaches only what is bought and sold; it is mute on leisure, safety, environment, and meaning, and reports averages that can hide wide differences by income. A measure may be illuminating without being complete — the caution Kuznets pressed on national income applies no less here. And it speaks always in averages, which can lie by soothing: a thing whose time-price has fallen for the comfortable may have risen for the poor, who earn fewer pounds an hour and so pay more of their life for the same loaf. Nordhaus reached for light precisely because he distrusted the official measures; the lesson holds against his own. Every measure, the time-price included, is a lamp that lights one room and leaves the rest of the house dark.

So the measure is to be trusted as far as it sees and no further — which is far enough to overturn the gloomy folk-conviction that everything only ever gets worse, and not so far as to license the cheerful one that everything only ever gets better. It says, soberly, that the world of made things has grown astonishingly cheap in the one currency that counts, the hours of a life; that some of the dearest things have moved the other way; and that the truest material verdict on an age is got not by counting its money but by asking how much of a life its goods cost — a question money was invented, in part, to help us forget.

There is a word for the thing in which all other values are reckoned. Economists call it the numéraire — the unit of account, the one good against which every price is expressed, its own value taken as the fixed point so that all the rest can be measured from it.⁶6In the pure theory of exchange (Léon Walras and after), the numéraire is the commodity chosen as the unit of account, its own price set to one, in terms of which all other prices are stated. Money is the numéraire of daily life; the deeper one, Smith’s ‘first price,’ is labour — the toil for which money is only a receipt. In daily life money plays this part, and plays it well enough that we forget it is itself only a stand-in. For money’s value floats; it is measured against other things as surely as they are measured against it. The deeper numéraire, the one that does not float because it is the substance and not the symbol, is the one Smith pointed to at the beginning: human labour, human time. It is the measure beneath the measure — the unit in which money itself is finally denominated, for a pound is worth nothing except the minutes of living it can command or save.

And this is the name the folio has carried, plate upon plate, unexplained, at its head. For the wage is exactly where the two currencies meet — the rate at which a portion of a human life is turned into the price of everything else, and everything else into a portion of a human life. The wage is the numéraire made flesh. These pages have watched it measure value and sell time, stake trust and compel obedience, settle by power and bound the world of work and refuse, out of dignity, to fall; and under every one of those offices lies this last and first one — that the wage is the place where money is at length translated back into what it always stood for, which is a stretch of someone’s irreplaceable life. Everything you have ever bought, you bought with that.

When you buy a loaf from a baker, you acquire a loaf, and nothing more. You do not acquire the right to stand in the bakery directing his hands, telling him to knead longer or bake hotter; the transaction is complete in the exchange of bread for coin, and the two of you are, before it and after, equals and strangers. This is a sale, and it is how the market coordinates very nearly everything — by a multitude of such exchanges, each one finished the moment its price is paid. But an employment is not like this, and the whole of the matter lies in the difference. When you engage someone for a wage, you do not buy a finished thing from them. You buy the right to tell them, within bounds and as your needs unfold, what things to make and how to make them.¹1The law’s line between an employee and an independent contractor turns on exactly this: the ‘control test.’ What makes one an employee is not what they produce but whether the hirer has the right to control how the work is done. A contract for services buys a result; a contract of service buys obedience. The contractor sells you a result and keeps command of his method; the employee sells you the command itself.

This is why employment has always felt categorically unlike trade, and why a wage is not simply a price among prices. A price clears a single exchange and dissolves the tie. A wage opens a tie and keeps it open — establishing, for as long as it is paid, a standing relation in which one party proposes and the other disposes, one directs and the other is directed. To be paid for a thing is to remain free of the payer. To be paid a wage is to enter, for the working day, into his service.

For most of the history of the wage, this was said with a bluntness that now gives one pause. The bond between the one who paid and the one who worked was known, in law, as master and servant — and the law meant both of its words. The master had the right to command; the servant, the duty to obey; and the obedience was secured not, as a broken sale would be, by a suit for damages between equals, but by the criminal law itself. Under the Master and Servant Acts — a line of English statutes running back to the Statute of Artificers of 1562 — a servant who left his place, or worked negligently, or was merely disobedient, could be haled before a magistrate (very often an employer himself) and sentenced to gaol, to hard labour, even to the whip.²2Master and Servant law made a worker’s breach of contract a crime, punishable by imprisonment, while a master’s breach was a civil matter only. Criminal prosecutions were heavily used in nineteenth-century industry; abolished in 1875 (the parties renamed ‘employer and workman’), after which, studies find, wages rose and moved more freely with demand. The asymmetry was the very heart of it: a master who broke the bargain owed, at most, money; a servant who broke it had committed an offence against the peace. Only in 1875 were the criminal sanctions struck off and the parties renamed, in the cool modern words, employer and workman — and where the sanctions fell away, wages promptly rose, and moved more freely, as though a hand had been lifted from them.

We are right to feel the distance between that world and our own; the servant who could be jailed for walking off is not the employee who resigns by email. But the distance is one of force and degree, not of kind. The criminal whip is gone; the structure it once enforced — that the wage buys a right to direct, and the worker an obligation to comply — remains the quiet armature of every job. We have made the obedience voluntary and revocable. We have not made it cease to be obedience.

You do not sell your work. You sell your agreement to be told what work to do.

Why should such a relation exist at all? It is a fair question, and a deep one, because the market — that vast machinery of prices and willing exchange — is supposed to coordinate everything without anyone giving an order to anyone. If a carmaker needs a thousand bolts, why does he not simply buy them, and the labour to fit them, by contract in the open market, as he buys his steel? Why instead gather hundreds of people under one roof, set them on a wage, and govern them by command — raising, in the very midst of the free market, a little zone of planning where the prices fall silent and the order of the day descends from above? Ronald Coase put exactly this question in 1937, and the answer founded a science.³3R. H. Coase, ‘The Nature of the Firm’ (1937). Using the market is itself costly — one must find prices, negotiate, and contract for every transaction — and these costs can be spared by bringing the activity inside a firm and coordinating by authority. The firm’s size is set at the margin, where the cost of organising one more task by command equals the cost of buying it in the market. Using the market, he saw, is not free: to coordinate by price one must seek out suppliers, discover what they charge, bargain, and draw a contract for every single transaction — and these costs, the costs of working the price mechanism, may be escaped by drawing the activity inside a firm and coordinating it by authority instead. A firm is a stretch of the economy where the market has been switched off on purpose. It is, in a phrase Coase borrowed, one of the ‘islands of conscious power in this ocean of unconscious co-operation’ — a lump of deliberate planning in the buttermilk of the market.

And the size of the island is no accident; it is fixed, like all things in economics, at a margin. A firm draws one more task inside its command whenever that is cheaper than buying the task in the market, and stops — leaves the task outside, to be bought — when it is not. The shoreline of the island is drawn precisely where the cost of organising by command comes level with the cost of organising by price. Inside, all is run by the wage and the order; outside, by the contract and the price.

What, then, does the worker actually agree to, when he steps onto the island and takes the wage? Not to perform some one, specified task — that would be a contractor’s bargain, the sale of a defined result. He agrees to a far more open thing: to do, within limits, whatever he is told. Herbert Simon, who modelled the employment relation with great exactness, saw that this open-endedness is its whole purpose.⁴4Herbert Simon, ‘A Formal Theory of the Employment Relationship’ (1951), distinguished a sales contract (a defined thing for a price) from an employment contract, in which the worker agrees, within limits, to do what the employer later directs — a ‘zone of acceptance.’ It is the rational choice precisely when the employer cannot foresee what he will need. A sale settles in advance exactly what shall change hands; but an employer often cannot know, at the hour of hiring, what he will need a month hence — and so he buys, in place of a fixed task, a range of possible tasks, a zone within which he may afterwards direct the worker as the day requires. Simon called it the worker’s ‘zone of acceptance’: the set of orders the worker agrees, beforehand and for the wage, to treat as legitimate and to obey. The wage is, in part, the price of the breadth of that zone — of how much of a person’s own discretion is signed over to another’s choosing.

This is why the employment relation is so useful — far more useful than a sheaf of contracts could be — for it is adaptable. The world changes faster than contracts can be rewritten; the firm that can simply re-aim its people, point them at the new problem, the new product, the sudden emergency, without renegotiating terms at every turn, moves at the speed of a command and not the speed of a haggle. The wage buys pliancy: a general, standing readiness to be useful in ways not yet named. One is paid not for what one will do, which is unknown, but for the promise to find out, and to do it.

And yet — is any of this truly authority? A formidable line of economists says it is not, and the challenge is bracing. Armen Alchian and Harold Demsetz argued in 1972 that the ‘command’ of the employer is a fiction dressed as power.⁵5Alchian and Demsetz, ‘Production, Information Costs, and Economic Organization’ (1972): the firm ‘has no power of fiat, no authority’; an employer can only dismiss or sue, ‘just as I can fire my grocer.’ The firm is on this view a ‘nexus of contracts.’ Later theorists (Williamson; Hart) recovered a real role for authority, lodged in the residual control rights of ownership. What, they asked, can an employer actually do to a disobedient worker? He cannot fine him at will, nor jail him — those days ended in 1875 — nor compel his body by any means. He can do just two things: cease to employ him, or sue him for breaking their agreement. But these are exactly the powers that any party to any exchange already holds. One can fire a grocer, by taking one’s custom elsewhere, just as an employer can fire a worker; the employer, they insisted, has no fiat, no authority, beyond what one trading partner holds over another. On this reading the firm is no hierarchy at all but a ‘nexus of contracts’ — a knot in which many ordinary, voluntary agreements happen to be tied — and the language of masters and commands is a misdescription of what is, at bottom, only more exchange among equals who stay free to walk away.

There is real force in this, and yet it proves too much, for it cannot say why working inside a firm feels, and functions, so unlike shopping. The economists who came after answered that the difference is genuine: that ownership carries a residual right of control — the right to decide all the things the contract never foresaw — and that this is a real fiat the market does not confer, the reason the law treats a firm as, in the quarrels of its own members, a kind of court unto itself. The truth lies between the grocer and the sovereign. The worker is not owned and may leave; the employer’s last sanction is only dismissal, which is the market’s sanction and not a gaoler’s. But for as long as the bond endures, the worker does, day upon day, simply obey within the zone — does not renegotiate the filing of each document or the typing of each letter, but does as he is bid — and a thousand such obediences, directed from a centre, are a different thing from a thousand market bargains, whatever the ultimate freedom beneath them. The wage is the price of an authority consented to and revocable: more than a purchase, less than a dominion, and exactly the bond that makes a firm a firm.

More than a purchase. Less than a dominion.

The line between command and price, once drawn, does not hold still — and in our own time it is moving, plainly, beneath our feet. The long arc has run toward freedom: from the servant who could be jailed for leaving, to the employee ‘at will’ who may quit on a whim and be dismissed on one, the worker’s submission has grown ever more conditional, ever more revocable. And now the island itself is, in places, washing back into the sea. The very logic Coase named is running in reverse: as the digital networks make it cheaper to find, price, rate, and contract with a stranger for a single task — a ride, a delivery, a design, an hour of labour summoned through an app — work that the firm had once swallowed is being cast back out into the market, recoded from employment into contracting, from the wage into the fee, from command into a price per piece.⁶6As the cost of contracting falls — search and coordination made cheap by digital platforms — Coase’s trade-off runs in reverse: work once internalised under employment (the wage, command) flows back out into the market (the fee, the contract). The blurring of the line between employee and independent contractor is the live legal edge of the shift. See A. Sundararajan, The Sharing Economy (2016). Whether this is liberation (each worker his own master at last, a free agent contracting as an equal) or its cunning opposite (the direction retained through the soft command of the algorithm and the rating, while the wage’s old protections are quietly shed) is the fierce question of the present labour economy — and it is, at bottom, this plate’s question: where the market should end, and command begin.

For that line — the shoreline of the island — is what the wage has marked all along. The folio has watched this single price measure value and sell time, stand as a pledge and fall under compulsion, settle by power and bound the world of work, refuse from dignity to drop, and serve at the last as the measure of all things. Here it does the most defining thing of all, the thing that parts a job from a sale: it buys not a person’s labour but a person’s agreement to be directed in it — a will placed, for a price and a time, at another’s command. That is what it is to be employed and not merely paid; and the whole vast architecture of firms — of offices and factories and the ordered working day — is built of nothing else but that agreement, multiplied.

Begin with a single, under-noticed asymmetry between the two ways a business pays out. The wage is a fixed sum: agreed in advance, owed in full, indifferent to how the venture turns out. The worker promised a hundred pounds a week is owed that hundred whether the firm triples its sales or loses half of them; his claim does not swell in the boom or shrink in the slump; and it is senior — the law and the queue alike put it ahead of the owners’, so that wages are paid before profits, in good times and in the bankruptcy court the same. Profit is the mirror-image. It is not a fixed sum but a residual — what is left over, if anything, once the wages and the rent and the suppliers and the interest have all been met. Frank Knight, who thought about this harder than anyone, put it exactly: profit is the residual, what remains after every contract has settled, and the one who takes it bears the brunt of all that no one foresaw.¹1Frank Knight, Risk, Uncertainty and Profit (1921). Knight separated risk (odds that can be measured, and so insured) from true uncertainty (which cannot), and argued that profit is the reward for bearing the latter: the residual left once the fixed, contractual claims — wages foremost — are paid. When the firm does well the residual swells and the owner grows rich; when it does badly the residual thins to nothing; and when the firm loses money the residual turns negative — the owner does not merely fail to gain but actually loses, while the wage is still paid out in full.

Hold those two shapes in mind, for they are the whole of this plate: a claim that is flat — the same in every weather — and a claim that is the gap between the firm’s fortune and that flat line, and so takes every rise and every fall. The worker holds the first. The owner holds the second. Everything that follows is a consequence of which claim is whose.

Set the two claims side by side and what they describe is an insurance policy. The worker, holding the fixed claim, has been insulated from the firm’s fortunes: his income does not shake when the firm’s does. The owner, holding the residual, has absorbed all the shaking — the variation that might have fallen on wages gathered, instead, onto profit. This is not a metaphor but a mechanism, and the economists who formalised it, Martin Baily and Costas Azariadis in the 1970s, said so outright: a firm is, among other things, an insurer — a party able to bear risk, shielding its more risk-averse workers from the shocks of the market by promising a steady wage and taking the fluctuations onto itself.²2The ‘implicit contract’ theory of Baily (1974) and Azariadis (1975): a risk-neutral firm insures risk-averse workers against shocks by paying a steady wage and absorbing the variation, in exchange for a lower average wage — the premium. Workers at firms that insure them best are found to accept pay a few per cent lower for the security. Knight had seen it half a century before, and called it the heart of the entire system: the bold, he wrote, ‘assume the risk and insure the doubtful and timid’ by guaranteeing them a settled income in return for the uncertain remainder — and that, he said, simply is the enterprise and wage system of industry. And, like any insurance, it is not free. The worker pays a premium — not as a separate fee, but folded into a lower wage: the steady hundred he accepts is, on average, somewhat less than the swinging sum his work would fetch if he shouldered its risk himself, and that discount, the price of safety, is what he pays the owner for carrying the risk in his stead.

You sold the gamble, and kept the floor.

This bargain answers a question the earlier plates left standing. We have seen why a person must take some wage — the alternatives were closed — and what they submit to in taking it: direction, the order, command. But neither tells us why so few of us strike out as owners of our own enterprises, claiming the residual instead of the wage; why the world is so overwhelmingly made of employees. The answer is risk. Most people are risk-averse: offered a certain fifty thousand, or a venture worth fifty-five in expectation but with a real chance of nothing, they take the fifty thousand — and are right, by their own lights, to do it. The wage is the institutional shape of that preference. It sorts us.³3Kihlstrom and Laffont (1979) modelled exactly this sorting: those least averse to risk become entrepreneurs, taking the uncertain residual; those more averse become employees, taking the certain wage. The wage and the firm are, on this view, the institutions a risk-averse population builds to shed risk onto the few willing to hold it. Those most willing to bear uncertainty become the entrepreneurs and the owners, and claim the residual with its risk; those who would rather not become the wage-earners, ceding the residual for a sum they can count on. To take a wage is, in part, to have declined a gamble — to have chosen the certain floor over the uncertain everything, and to have paid for the choice with the upside it forecloses. It is the prudent course, and the common one. But it is a choice, and it has a price.

Seen from one side, this is among the kindlier arrangements in economic life, and its defenders are not wrong to insist on it. The worker is genuinely protected. The firm may have a ruinous year — a failed product, a market that evaporates — and the worker, unlike the owner, is spared the ruin: his wage is paid, his household holds steady, the blow falls elsewhere. He has been shielded from the very thing that breaks the entrepreneur, the lurch and plunge of an income lashed to fortune; and the profit the owner draws in the fat years is, on this reading, no theft but a fair return for having carried, the whole time, a burden the worker was relieved to set down. Nor is it a swindle in which one side’s gain is the other’s loss: the worker, by his own preference, is better off with the steady wage than with the gamble, and the firm is better off bearing a risk it is large and diversified enough to absorb — both are improved against a world in which the bargain did not exist.⁴4In the implicit-contract models both parties gain relative to a bare spot market for labour: the worker buys a smoother income, the firm earns a higher average profit for supplying the insurance. On this account the wage bargain is a mutual improvement, not a zero-sum extraction. No one chose the safety with eyes shut. The premium buys something real and dear — a life that can be planned — and most of us, most of the time, are glad to pay it.

Seen from the other side, the identical structure is the quiet engine of the divide between labour and capital — and its critics are not wrong either. For if the wage is fixed and the profit residual, then every gain beyond the agreed wage, however vast, flows to the owners and none to the workers who helped make it. When an enterprise does not merely survive but soars — when a firm comes to be worth a hundred times what was put in — the workers who built it draw the same fixed wage they were promised, while the residual, now enormous, falls entirely to capital. Marx gave this reading its hardest edge: the residual, to him, was surplus — value made by labour and taken by the owner under the cover of a fair wage.⁵5For Marx the residual was surplus value, created by labour and appropriated by capital. One need not adopt his framework to note the asymmetry of risk: a shareholder diversifies across many firms and bears only what cannot be diversified away, while a worker cannot diversify a single job and the human capital sunk in it — so labour often carries the more concentrated risk, yet capital takes the residual. One need not follow him the whole way to feel the asymmetry, which is sharpened by a fact the insurance story tends to pass over: that the owner’s risk is often far smaller than it looks. A shareholder spreads his stake across many firms, so that any single failure costs him little; he bears only the risk that cannot be diversified away. The worker cannot diversify. He has one job, one employer, one stream of income, and his human capital — the years and the skill he has sunk into this work — is staked, undivided, on this one firm. It is not at all clear that the party bearing the larger and more concentrated risk is the one who draws the residual; often it is the reverse. And the insurance, besides, is imperfect: when the bad year comes the firm may simply break its unwritten promise and lay the worker off — so that the cover lapses precisely when it is most needed, and the worker proves to have borne the gravest risk of all, the loss of the whole income, after everything.

All the gain falls on one side of the line.

In truth the fixed wage and the residual are not two boxes but the two ends of one line, and a working life may sit anywhere along it. Every device that hands the worker a piece of the residual — a bonus tied to profit, a commission, a grant of shares, a partnership, the membership of a cooperative — slides him toward the owner’s end, trading a little of his safety for a little of the upside; and done well, the share of fortune is added on top of the steady wage, so that the worker keeps his floor and gains a window on the gain.⁶6Profit-sharing, stock grants, employee-ownership plans, and worker cooperatives give workers a share of the residual — moving them toward ownership. The gain is greatest, and the danger least, when the residual share is added atop a secure wage rather than substituted for it; staking one’s whole livelihood on a single firm concentrates risk severely. But the line runs the other way too. The gig contract that pays only by the task and guarantees nothing has stripped the wage of its insurance — handed the worker back all the variation a wage was built to absorb — without giving him, in return, any share of the residual his platform earns: the risk of the owner with the claim of the labourer, the worst of both ends. Where a worker sits on this line — how much of his income is the safe, fixed, capped thing, and how much the risky, residual, unbounded thing — is among the most consequential facts about his working life, and among the quietest. Most people, asked, would not be sure where they stand.

So this single price, which we have watched measure the worth of all things, sell the hours of a life, stand as a pledge and fall under compulsion, settle by power and bound the day’s work, refuse from dignity to drop, and buy the obedience that makes a firm — does one thing more, beneath all of those. It shelters. It is the roof the worker takes in exchange for the harvest: he gives up his claim on the bumper year, and in return the failed year cannot ruin him. The wage is the storm-insurance of a life, and its premium is the fortune one agrees never to have.

When a machine can do what you are paid to do, it does not merely compete with you, as another worker would, for a share of the wage; it threatens to abolish the wage altogether, for a task done by a machine is a task for which no one need be paid. This is the oldest fear in the history of industrial work, and it arrives, like a tide, in waves. In 1811 the skilled framework-knitters and weavers of the English Midlands and North, watching the new powered frames and looms turn out cheaper goods and put them out of their trades, did the direct thing: they broke the machines.¹1The Luddites (1811–13) — skilled textile workers — smashed the powered frames and looms that were destroying their trades. The breaking of machines was made a capital offence, and fourteen were hanged. David Ricardo, in the 1821 third edition of his Principles (ch. 31, ‘On Machinery’), reversed his earlier optimism and conceded that machinery could be gravely injurious to the labouring class. They were not fools raging at progress; they were craftsmen who understood, precisely, that their hard-won skills were about to be made worthless — and they were right, for their wages did vanish. The state hanged fourteen of them. A decade later the fear reached the calmest mind in the science: David Ricardo, who had taught in his first edition that machinery must benefit every class, added to his third, in 1821, a chapter recanting. He had become convinced, he wrote, that the substitution of machinery for human labour was ‘often very injurious to the interests of the class of labourers.’ And a century after that, Keynes gave the fear its lasting name, warning in 1930 of a ‘new disease’ — technological unemployment, the unemployment that comes when our skill at economising labour outruns our skill at finding new uses for it. The names change; the dread is one. And in our own day it has taken its most unsettling form, for the newest machines threaten not the hand but the mind.

And yet the dread has been, in its central prophecy, wrong — wrong for two hundred years, wrong with a consistency that ought to humble it. For the machine that can do your work has a second face. Where it does not replace the worker it complements him: the weaver at the power-loom made more cloth in a day than the handloom weaver could in a week, and the productivity that poured from the machine, spread through the economy, lifted wages it never touched. And by making goods cheap, the machine freed the spending of everyone to flow toward wants that had gone unmet — wants that became new trades, new industries, new kinds of work the displaced, or their children, could turn to. The proof is written in the largest single fact of the modern economy. In 1870 almost half of all Americans worked the land; in our own century, scarcely one in a hundred does²2In the United States, agriculture employed about 46% of the workforce in 1870 and roughly 1% today — a displacement of labour more total than any machine has yet wrought elsewhere, yet it ended not in mass idleness but in reabsorption into new work, as real wages rose many-fold across the two centuries of mechanisation. — a displacement of the workforce more total than any robot has yet wrought — and it produced not mass idleness but the richest society in history, the displaced absorbed, over generations, into work their great-grandparents had no names for. The lamplighter and the elevator operator vanished; the programmer and the radiographer appeared. Employment did not collapse. Across the whole long sweep, through every panic, real wages rose.

The machine is the rival and the benefactor both.

The error in the recurring fear has a name, and naming it is the beginning of the reassurance. It is the ‘lump of labour’ fallacy — sometimes, in honour of the machine-breakers, the Luddite fallacy — the belief that there is a fixed quantity of work to be done in the world, a lump, so that every task handed to a machine is a task taken from a human being for good. But the lump does not exist. The quantity of work to be done is not fixed; it grows, because cheaper production and higher incomes call forth new demands without end, and meeting them takes new labour. The clearest way to see this is to stop thinking of whole jobs and think instead of tasks, as the economists who study automation now do.³3In the task framework of Acemoglu and Restrepo (‘Automation and New Tasks,’ 2019), technology has two opposed effects: a displacement effect, as machines take over tasks once done by people, and a reinstatement effect, as new tasks are created for which human labour is needed. Whether employment falls or grows turns on whether reinstatement keeps pace with displacement. A machine does not usually swallow an occupation whole; it takes over particular tasks within it and leaves others — while the wider change calls into being entirely new tasks, and new occupations, that did not exist before. Automation thus works at two frontiers at once: it displaces labour from the tasks it absorbs, and it reinstates labour at the new tasks it creates. Whether work, on the whole, shrinks or grows is a race between those two frontiers — and for two centuries, the second has kept ahead of the first.

But the reassurance, however well-earned across centuries, conceals two hard things, and honesty requires holding them up to the light. The first is the matter of when. ‘In the long run,’ the optimist says, the machine has always been a benefactor — and it is true. But, as Keynes also wrote, in the long run we are all dead;⁴4‘In the long run we are all dead’ — Keynes, A Tract on Monetary Reform (1923), rebuking economists who console the present with the equilibrium of a distant future. His 1930 guess that the rich countries would be four to eight times better off within a century has largely come true in the aggregate — but the gains pooled unevenly, much of the rise accruing above the median rather than to the worker in the middle. and the short run, the only run a person actually inhabits, can be merciless. The handloom weaver displaced at forty did not live to see his grandchildren prosper in the mills; he lived to watch his wage fall to nothing and his skill turn to dust. The twenty thousand Nottingham knitters thrown out of work in a single year had no retraining scheme and no safety net; they were, in the plain phrase of the day, cast onto the scrap heap. The rising line the centuries draw is, seen close, a sequence of cliffs — and it is never the same people who fall and who climb. The gains are general and the losses particular, and a statistic that nets the one against the other describes no one who ever lived. Even the great enrichment Keynes foresaw arrived crookedly, much of it gathering at the top, so that the average climbed far faster than the wage of the man in the middle.

The second hard thing is whether this time is like the others at all. Every machine until now has had a ceiling to what it could take. The early engines took muscle, and labour withdrew into the work of the hand and the eye and the mind. The computer took routine — the predictable, rule-bound tasks, clerical and manual — and the labour market split, or ‘polarised’: the middle, where the routine work clustered, was hollowed out, while the top (the cognitive, the creative, the work of judgment) and the bottom (the non-routine manual work of care and service) grew, the displaced of the middle pressed downward more often than up.⁵5Job polarisation (Autor and Dorn, 2013; Goos and Manning, 2007): computerisation hollowed routine middle-skill work while high- and low-skill work grew, often pushing the displaced into lower-paid jobs. Artificial intelligence may break the pattern: studies of exposure (e.g. Frey and Osborne, 2017, and later work on large language models) find its reach greatest among higher-skill cognitive tasks — and expert forecasts diverge more widely than on almost any question in economics. But always there was higher ground — a domain of human judgment, imagination, and understanding into which the wage could retreat, secure in the faith that no machine could follow it there. The artificial intelligence of the present age is the first machine that climbs toward that ground. It writes, it draws, it diagnoses, it argues, it codes; the very faculties that were labour’s last redoubt — analysis, language, the weighing of judgments — are the ones it now contests. No one knows what this means. The forecasts diverge more wildly than on almost any question in the science: some see the same false alarm that has sounded for two centuries, the lump-of-labour fallacy in newer dress; others see the first machine for which the old comfort fails, because the high ground, this time, has no higher ground above it. It may be that new work blooms as it always has. It may be that this is the time the tide does not turn. The honest answer is that we are inside the experiment, and cannot yet read its result.

Always, before, there was higher ground.

There is one thing more the machine does to the wage, and it does it even to those it never replaces. It sets a ceiling. No worker is paid, for long, much more than it would cost to do without him — and the cost of the machine that could do his work is precisely that limit, the silent ceiling above his wage.⁶6A worker is not, for long, employed at a wage above the cost of the machine that could replace the job. The machine need not be installed to bind the wage; its availability, and its ever-falling price, cap what the worker can demand. As the price of machines falls — its one reliable tendency — the ceiling descends. He need not be replaced for the threat to bind: the mere existence of the machine, and its falling price, caps what he may ask, for he cannot demand more than his replacement would cost. And the price of machines does, over time, only one thing — it falls. So the ceiling descends, year upon year, pressing down toward the wage from above, whether or not the machine is ever installed. To earn a wage at all is, in this sense, to be paid in the shadow of one’s own replacement: the worker keeps his place only so long, and only so far up, as the machine that could take it remains dearer than he is.

This, then, is the last of the forces the folio has traced through this single price. We have watched the wage measure the worth of things and sell the hours of a life; stand as a pledge and fall under compulsion; settle by power and bound the world of work; refuse from dignity to drop; serve as the measure of all value; buy the obedience that makes a firm; and shelter the worker from the firm’s own risk. Here, at the last, we find the rival that shadows all the rest — the machine that can do the work, and so threatens, in the end, to make the wage itself unneeded. The wage has always lived in that shadow, and lives there still. What is new is only how high up the light the shadow has begun to climb.

Ask people why they do their work, and only some of the answer is the wage. They speak of the use of it, the pride in it, the people it serves, the self it lets them become; of a thing worth doing for its own sake. We have a word for work like that — a calling, a vocation, from the Latin for a voice, as though the work summoned the worker to it.¹1The idea of worldly work as a ‘calling’ (German Beruf) — a vocation valued beyond its pay — received its classic statement from Max Weber, who traced it to the Reformation: ordinary labour reimagined as a summons, dignified and chosen in itself, not merely endured for a wage. The word is old and half-religious, but the thing it names is ordinary: most people want their work to mean something, and not only to pay. This meaning is a kind of second wage, paid not in coin but in significance, and it appears on no slip. Yet it is not free of the first wage, the one that does. The two are entangled, and the manner of their entanglement is among the stranger facts in the economics of work — for they do not add together as one would expect. They trade against each other; and worse, the one can destroy the other.

Begin with the trade. Adam Smith saw it in 1776, in the very chapter where he set out what makes one wage differ from another: of all the forces that lift a wage or lower it, the first he named was the agreeableness or disagreeableness of the work itself.²2Adam Smith, Wealth of Nations (1776), Book I, ch. 10. Wages adjust for the non-monetary character of work — its ease or hardship, honour or dishonour, safety or danger — so that it is not the wage but ‘the whole of the advantages and disadvantages’ of employments that tends, under free choice, to equality. Agreeable and honourable work can therefore command less pay. A pleasant, honourable, meaningful job, he reasoned, draws people to it for reasons other than the pay, and so need not pay as much; a dirty, dangerous, or shameful one must bribe people in, with a wage high enough to make up for what the work itself costs them. It is not the wage that equalises across employments, Smith concluded, but the whole of the advantages and disadvantages — the wage plus everything the work is and gives. The meaning, then, is part of the pay; and the more of it a job offers, the less of the other pay it need offer. This is the vocation penalty, and it falls most heavily where the meaning runs deepest and is most taken for granted. The care of the sick and the old, the teaching of the young, the making of art, the tending of souls and of communities — the work most steeped in purpose — is, with a regularity that ought to trouble us, among the worst paid, in part because those who do it will do it for less, for the meaning, and everyone knows that they will.

Work that pays in meaning pays less in money.

That meaning lowers the wage is surprising enough. The deeper strangeness is that the wage can lower the meaning. Money and significance are not merely substitutes traded at a stable rate; the act of paying can reach into the meaning and spoil it. The clearest demonstration is also the most famous. In 2000 two economists studied a set of daycare centres in Haifa troubled by parents who came late to collect their children. The centres began to fine the latecomers — and lateness went up, not down, and stayed up.³3Gneezy and Rustichini, ‘A Fine Is a Price’ (Journal of Legal Studies, 2000): fining late parents increased lateness, and removing the fine did not undo it. Cf. the psychological evidence that extrinsic rewards can undermine intrinsic motivation (Deci, 1971; Deci and Ryan), and Titmuss’s celebrated — though empirically contested — claim that paying for blood would reduce donation (The Gift Relationship, 1970). The fine had not deterred the parents; it had released them. Where before a late parent felt the tug of an obligation — they were imposing on the teachers, keeping them past their evenings — now they had a price, and a low one, and could simply pay it and feel nothing. A moral relation had been converted into a market transaction, and the conversion did not run backward: when the centres dropped the fine, the lateness did not fall back to where it had begun. The guilt, once priced, did not return for free. The economists titled their paper, precisely, ‘A Fine Is a Price’ — and the lesson reaches far past daycares. Pay people to give blood and some give less, not more, for a gift has been turned into a sale; reward a child for reading and the reading can curdle into a chore done for the reward, dropped the moment the reward stops. Psychologists call it the crowding-out of intrinsic motivation: under the wrong conditions, a payment offered for what was freely done drives out the freedom, and leaves less behind than there was before money entered the room.

A price can drive out what no price could buy.

Two readings contend over these facts, and the folio owes both their due. The first is generous, and not wrong. The vocation penalty, on this reading, is no injustice but a free and open bargain: no one conscripts the nurse, the teacher, the artist; they choose the work knowing it pays less, because it pays in the other currency, and the lower wage is simply the price they willingly give for the meaning they take.⁴4On the compensating-differential view this is no wrong but an exchange: those who prize a job’s meaning sort into it and accept lower pay, the meaning being part of what they are paid. And the deeper good is real — to have work one would do for love is among the genuine fortunes of a life, not a delusion to be dispelled. The meaning is part of their compensation, as real to them as money; to insist they are simply underpaid is to ignore what they themselves are buying with the wage they forgo. And the deeper good is real beyond dispute: to have work one would do for love — work that is a calling and not merely a job — is among the great fortunes of a life, the opposite of the far commoner bargain of selling one’s hours at a task one is glad to leave at the door. That some work can be loved is not a trap to be exposed; it is a thing to wish for everyone, and the counsel to seek it is, at heart, kind.

The second reading is darker, and also not wrong. For the very same fact — that people will do beloved work for less — is, from the other side, a lever, and one that gets pulled. If the carer will care for the meaning, then the meaning can be used to underpay the care; if the devotion is reliable, the wage can be set below what the devotion is worth and the difference kept by whoever pays, secure that the work will be done regardless, for love. A whole creed has grown up to oil this lever — the cheerful modern command to ‘do what you love’ — and its critics have shown how neatly it serves: if work is love, then to ask fair pay for it seems crass, a betrayal of the passion, and so the love becomes, in one writer’s phrase, the most elegant of anti-worker arguments, dressing labour as devotion precisely so that it need not be paid or bounded as labour.⁵5Miya Tokumitsu, ‘In the Name of Love’ (2014): the ‘do what you love’ ethic can disguise work as passion and so license underpayment and overwork — ‘emotionally satisfying work is still work.’ The penalty falls hardest on care work long assigned to women and assumed to flow from natural love (cf. plate vi, the unwaged). The penalty falls hardest, and least by accident, on the work long assigned to women — the nursing, the teaching, the minding, the feeding — which is presumed to flow from a natural love, and so held, by a very old habit, scarcely worth paying for at all. Emotionally satisfying work is still work; and to refuse to call it work, however tender the refusal sounds, is to hold the door open for the plainest exploitation.

Both readings are true at once, which is what makes the matter hard, and their shared truth fits in a sentence: the meaning of work is real, and so is its abuse. The wage cannot settle between them, because the wage cannot see the meaning at all. It measures the price of the work and is blind to its worth, and the two are not the same number. This is the limit the folio has been circling since it named the wage the measure of all value: the wage measures everything except the one thing that makes work most worth doing — and when it tries to, when a price is laid upon a calling, it is as apt to corrode the worth as to capture it.⁶6Crowding-out is not a law but a tendency, and a conditional one — rewards can also ‘crowd in’ motivation — sharpest where the work is genuinely a calling. The remedy the critics urge is not to stop paying but to pay fairly and honour the meaning: the two currencies together, not one offered in place of the other. The decent course is therefore neither to deny the meaning (and so license the underpayment) nor to trust the meaning to pay the rent (for it cannot). It is to pay fairly in the first currency and to honour the second — to treat beloved work as work, fully paid, and as a calling, fully respected — and to refuse the bargain, always offered in love’s name, that asks the worker to take the meaning instead of the money rather than alongside it.

So we reach the edge of what the wage can do. It can measure the worth of a thing and sell the hours of a life; stand as a pledge and fall under compulsion; settle by power and bound the world of work; refuse to drop; serve as the measure of value; buy obedience; shelter against risk; and stand in the shadow of the machine. The one thing it cannot do is say why the work was worth doing. That part of the reward was never money, and was never for sale — and it is the part, when the work is a calling, that the worker would keep though all the rest were taken away.

To the employer there is nothing mysterious about the wage: it is a cost, the price of an input, a figure on the same ledger as rent and materials and power — and, like all those figures, to be kept as low as the work allows. This is not cruelty but arithmetic. Every pound paid in wages is a pound not kept as profit, and a firm that pays more than its rivals for the same labour will, all else equal, be undersold by them and driven under; so the discipline of competition presses, always, toward the lowest wage the firm can pay and still get the work done. And the pressure is not slight, for labour is, in most enterprises and in the economy entire, the largest cost there is — well over half of all the income a modern economy generates is paid out as wages and salaries.¹1Compensation of employees is the majority of national income — on the order of sixty per cent — and for the typical firm the wage bill is its largest controllable expense. Hence the standing competitive pressure to hold it down: to the one who pays it, the wage is the great cost. The wage is the great expense, and to the one who pays it, the case for holding it down is as plain as a column of figures.

Now step back from the single ledger to the whole economy, and a second nature of the wage appears, invisible from up close. The wages that every firm pays out do not vanish; they become the incomes of households — and households, having earned, spend. That spending is not a sideshow but the main event: consumer spending is roughly two-thirds of all the activity an economy like America’s produces, by far the largest source of the demand that firms exist to satisfy.²2Personal consumption is about sixty-eight per cent of US gross domestic product — the largest single component of demand — and it is funded, overwhelmingly, by wages. The worker the firm sees as a cost is, through the till of some other firm, the economy’s principal customer. And what funds that spending is, overwhelmingly, wages. So the worker, whom the firm sees only as a cost, is also — at one remove, through the till of some other firm — the customer. The wage paid at one factory buys the goods made at another; the cost on one ledger is the revenue on a second. Seen whole, the wage is not money leaving the economy but money circulating within it: out of firms as cost, into households as income, back to firms as demand, around and around.

One firm’s wages are another firm’s revenue.

Here the two natures of the wage collide, and the collision is one of the most consequential errors in economics. Each firm, reasoning soundly from its own ledger, concludes that cutting its wages will help it: lower costs, the same sales — for its handful of workers are a negligible part of the market for its goods — and so more profit. And each firm is right, about itself. But what is true of one firm is not true of all firms together; for if every firm cuts its wages, then every household’s income falls, and the spending that is every firm’s revenue falls with it. The saving each firm counted on is swallowed by the demand they all lose. This is the fallacy of composition, the mistake of assuming that what holds for a part must hold for the whole³3The fallacy of composition: what is true of a part need not be true of the whole. Keynes charged the classical economists with exactly this — taking the demand curve for a single industry, where a wage cut leaves sales untouched, and carrying it ‘without substantial modification to industry as a whole’ (General Theory, ch. 19). The paradox of thrift is its twin. — and it is the very error Keynes laid at the door of the economics of his day: taking the demand for a single industry, where a wage cut does no harm to sales, and carrying it over, unaltered, to industry as a whole, where it does. The same trap catches the thrifty. It is prudent for one household to save more; but if all households save more at once and no one spends, the demand for everyone’s labour collapses, and all are made poorer by their common prudence. Virtue in the part can be ruin in the whole.

There is a name for the moment this fallacy stops being academic: a depression. When demand weakens and firms, each acting prudently, cut wages and shed workers to defend their margins, the cut incomes mean cut spending, the cut spending means weaker demand, and the weaker demand invites the next round of cuts — a spiral turning downward, every turn of it justified, the whole of it catastrophic. This was the argument with which Keynes, in 1936, broke the orthodoxy of his age.⁴4Keynes, General Theory (1936), ch. 19, ‘Changes in Money-Wages.’ Against the ‘Treasury View’ that wage cuts would cure the mass unemployment of the Depression, he argued that the labour market does not clear like a market for goods, because the wage is also the source of demand — so cutting it can deepen a slump rather than end it. The classical doctors had prescribed, for the mass unemployment of the Depression, exactly what each firm’s ledger advised: cut wages, and the labour market will clear, as any market does when its price falls. Keynes answered that the labour market is not like the market for fish, because the wage is not only the price of labour but the wellspring of the demand for everything — so that cutting it does not merely lower a price, it lowers the spending of the very people whose spending the recovery requires. The cure, applied to all at once, could deepen the disease. To lower the wage in a slump might be, for the economy entire, to dismiss its customers in the name of sparing its costs.

The worker is also the customer.

And yet the demand-side argument, left to itself, curdles into its own kind of fallacy, and honesty requires the other hand. For the wage is a cost — really, not merely in appearance — and a cost cannot be wished into a pure blessing. The classical tradition Keynes attacked was not foolish; it held a truth he himself granted: that a wage cut can, in certain circumstances, restore employment, by making it profitable to hire again.⁵5Keynes conceded a wage cut is ‘quite capable in certain circumstances of affording a stimulus to output.’ And the famous proof of the demand argument — that Ford doubled his workers’ pay in 1914 so they could buy his cars — is a myth: he raised wages to stem a turnover that had reached 370 per cent, an efficiency wage; even had every worker bought a Model T, the higher wages would have cost more than the cars returned. Push the demand logic too far and it yields an absurdity: that an economy might enrich itself merely by decreeing higher wages — paying everyone more so that everyone may buy more, without end. But this is the fallacy of composition stood on its head. A single firm cannot conjure customers by overpaying its own workers, and an economy cannot conjure wealth by paying wages its production has not earned; it can conjure only inflation, or unemployment, as the wages it mandates outrun the value its work creates. Even the cherished proof of the demand argument — that Henry Ford doubled his men’s pay in 1914 so they could afford his cars — dissolves on inspection into a myth: Ford raised wages to stanch a ruinous flight of labour, and had every worker bought a Model T, the wages would still have cost him more than the cars returned. Wages are a real cost, paid for real reasons, and no economy can pay itself rich.

So the wage wears two faces, and both are true, and the difficulty is that no one who pays a wage can see both at once. From where the single firm stands, the wage is a cost, and the case for cutting it is sound. From where the whole economy stands, that same wage is a customer’s income, and the case for cutting it — all of them, together — can be ruinous. The firm is not wrong about the part; it is merely blind to the whole, because the whole is visible from no place inside it. This is why the question that divides the two faces — whether, in a given slump, wage flexibility will cure unemployment or deepen it — has stayed unsettled for the better part of a century,⁶6The quarrel endures between the heirs of Keynes and of the classics. Which face governs depends on the case — how deep the slump, how idle the capacity, whether falling wages and prices revive spending by other routes or merely drain it. Both faces are real; neither alone is the whole. fought between those who trust the market to clear and those who fear the spiral, the truth lying, as it usually does, in the changing particulars: how deep the slump, how idle the capacity, how frightened the saver.

And so the folio’s long anatomy widens, at the last, from the single wage to the sum of them all. We have followed the wage as a thing agreed between two parties — its justice, its hours, its trust, its compulsion, its amount, its boundary, its rigidity, its measure; the authority it buys, the risk it sheds, the machine that threatens it, the meaning it cannot hold. Here we find that the wage is never only a private bargain. It is also a public flow — the great circulating river of demand from which every seller drinks, including the very employers who, one by one, would dam it. The wage each firm pays is a cost to that firm and a livelihood to the economy; and the whole stands or falls on a sum that no single party to it was ever placed to see.

Strip the wage of everything the earlier plates have found in it — its justice, its measure, its meaning — and a hard residue remains: the wage is paid because, for most people and at the margin, work is something one would rather, just then, not do. Not that work is hateful; much of it is bearable and some of it is loved. But the marginal hour — the extra one, at the end of a long day, when the body is tired and the evening beckons — that hour has a cost to the one who works it, a cost the economists named without flinching: the disutility of labour.¹1The marginalist account (W. S. Jevons, The Theory of Political Economy, 1871): work carries a disutility that mounts with each hour, and a person supplies labour up to the point where the pain of one more hour just equals the worth of its pay. Cf. Smith’s ‘toil and trouble’ (plate viii). The least wage that will draw one from leisure is the reservation wage. Against it stands the pull of one’s own time: rest, family, idleness, the pursuit of whatever else the hour might hold. The wage is the bribe that overcomes that pull — the sum that must be offered before a person will hand an hour of their life to another’s purpose rather than keep it for their own. Below some figure, no one comes; each worker has, in effect, a price beneath which their time is simply not for sale. The wage is what clears it.

How much of their life a person sells, then, is a balance — struck hour by hour — between the wage an hour would earn and the worth of the hour itself. And here a subtlety enters that overturns the easy assumption that higher pay must always mean more work. A rise in the wage pulls in two opposite directions at once. On one side, it makes each hour of leisure more expensive: an hour not worked now forgoes more money than before, so the worker is tempted to surrender some leisure and work more — the substitution effect, trading rest for the suddenly dearer wage.²2A wage rise raises the opportunity cost of leisure — the earnings given up by not working — which pulls toward more work (the substitution effect); and it raises real income, and leisure is a normal good, wanted more as one grows richer — which pulls toward less work (the income effect). Which prevails is not fixed in advance. But on the other side, the same rise makes the worker richer — able to reach any given standard of living in fewer hours — and a richer person, like anyone with more to spend, tends to want more of a good thing, and leisure is a good thing. So the higher wage also tempts the worker to take some of their new wealth as free time, and work less — the income effect. The two pulls contend. Which prevails is not settled in advance; it turns on the worker, the wage, and how much leisure they already hold.

The wage is the price of an hour of a life.

For most workers, at low and middling wages, the first pull wins: when pay rises, hours rise with it, for the money is needed and the leisure given up is cheap. This is the labour-supply curve we expect — more pay, more work. But it does not hold all the way up. As the wage climbs higher, the second pull strengthens, for two reasons at once: each additional pound now buys less that the worker lacks — the gap between earning thirty thousand and thirty-five is keenly felt, between a hundred and fifty thousand and a hundred and fifty-five, scarcely at all — while each hour of life surrendered grows correspondingly dearer. Past some point the income effect overtakes the substitution effect, and the curve does something strange: it bends backward. Higher pay now buys fewer hours, not more. The worker, grown rich enough, begins to spend a rising wage on the one thing no wage can manufacture — time. Drawn out in full, the supply of labour is not a rising line but a leaning one: up at first, then folding back upon itself, so that the very rich, were they to follow it, would work less than the merely comfortable.³3The backward-bending labour supply curve: the substitution effect dominates at low wages (the curve slopes up), the income effect at high (it bends back), with a vertical ‘turn’ between. The diminishing marginal utility of income sharpens the bend. It is observed in practice — for instance in the Canadian labour market, and most clearly among some groups of workers.

Past a point, the wage buys time, not work.

This backward bend, written small in one worker’s choices, was once read large as the destiny of all. If a rising wage lets the individual buy back their hours, then surely rising wealth would let humanity buy back its leisure — and the richer the world grew, the less it would need to work. No one made the promise more famously than Keynes. In 1930, in the very pit of the Depression, he looked a hundred years ahead and foresaw not ruin but abundance: by the time his contemporaries’ grandchildren had grown — by about now — the world would be four to eight times richer, the age-old struggle for subsistence all but solved, and people would work, he ventured, perhaps fifteen hours a week.⁴4J. M. Keynes, ‘Economic Possibilities for our Grandchildren’ (1930). He predicted that by about 2030 living standards would be four to eight times higher and the ‘economic problem’ nearly solved, leaving a ‘fifteen-hour week’ — and a new difficulty: how to use the leisure won, how ‘to live wisely and agreeably and well.’ The machines would do the labouring; the people would be freed. And freedom, Keynes saw, would bring its own difficulty — for the first time, he wrote, man would face his real and permanent problem: how to fill the leisure that science had won him, how to live wisely and agreeably and well. The economic problem solved, the human one would only begin.

He was half right, which is the most interesting way to be wrong. The wealth arrived — by some measures the world outran even his guess. The leisure did not.⁵5The verdict on Keynes: the wealth came, the fifteen-hour week did not. Hours fell for a century, then roughly stalled, far above it. He overrated satiation — wants prove insatiable and relative, driving a ‘rat race’ — and Veblen’s leisure class (The Theory of the Leisure Class, 1899), in which leisure displayed status, has inverted: today the highest earners work the longest, and busyness signals rank. Cf. plate xii — work is also where meaning is found. Hours of work fell steeply for a century and then, in the richest countries, roughly ceased to fall, settling far above fifteen a week; and then came the strangest turn of all. For most of history, leisure had been the very badge of rank — the mark of the gentleman was that he did not work, and Veblen gave the idle rich their name, the leisure class, whose conspicuous uselessness advertised their wealth. That has inverted. Today it is the highest earners who keep the longest hours, and busyness, not idleness, that signals importance; the modern eminence boasts not of their leisure but of how little of it they have. The backward bend, real enough in the textbook, was overwhelmed in the world — partly because Keynes mistook our wants, supposing they would be sated once our needs were met, when in truth wants keep no ceiling and climb with every rung, so that we run to hold our place in a race with no finishing line; and partly because work was never only the sacrifice this plate began by calling it. It is also, for many, where meaning and standing and selfhood are found — the calling of an earlier leaf — so that even those rich enough to stop do not, for to stop would be to lose not income only but a self. Keynes half-saw it: everybody, he allowed, will need to do some work to be content. The old Adam is strong in us. We were offered the leisure, and we have, so far, declined it.

So the wage is, at the last, the price of a portion of a life — the rate at which a person turns their finite hours into the means of living, and the instrument by which they settle, mostly without ever deciding, how much of themselves to sell and how much to keep. The labour-supply curve, with its hopeful backward bend, holds out a real possibility: that wealth might be spent on time, that a life grown richer might grow freer.⁶6The bend is genuine, and some do claim it — trading a higher wage for shorter hours, buying back their time. That most do not owes partly to wants that never rest and partly to the goods of work itself (plate xii). The curve only offers the choice; what a person makes of it the wage cannot say. Some take it. But the same curve, folding back, measures how little of the promise we have claimed — how, offered the chance to work less, most of us have chosen, or been driven, to want more instead. The wage can buy back an hour. Whether anyone spends it so is a question the wage cannot answer; it belongs to the worker, and to whatever they believe their hours are for.

And so the folio nears the end of its anatomy. We have weighed the wage’s justice and its hours, its trust and its compulsion, its amount and its boundary, its rigidity and its measure; the authority it buys, the risk it sheds, the machine that rivals it, the meaning it cannot hold, the demand it sustains. Here, beneath all of these, lies the plainest fact of any: that the wage is paid for time, and time is the one thing a wage can never make more of. It can only set the rate at which we hand it over — an hour of the only life we have, for the coin to go on living it.

That schooling pays is not in doubt. The high-school graduate out-earns the dropout, the college graduate out-earns the high-school graduate, and the gap is wide — by one careful count the typical college graduate earns on the order of a quarter to a half more than the worker who stopped before, a premium that holds, remarkably, across almost every occupation there is.¹1The college wage premium is large — one tabulation finds a median of about 28 per cent across occupations, and more in raw full-time terms — and it appears in some 90 per cent of occupations, including many with no academic content at all. The fact is not in dispute; its cause is. It holds even in trades where nothing taught in a lecture hall could plausibly matter — the credentialed bartender, the degree-holding cashier, out-earning their uncredentialed fellows at the very same tasks. So the fact stands firm: education and the wage rise together, and the diploma marks the rise. The dispute is not whether the educated earn more. It is why.

The first and oldest answer is the natural one: school makes people better at work. This is the theory of human capital, given its modern form by Gary Becker — that knowledge and skill are themselves a kind of capital, built up by schooling and training as surely as a machine is built up by investment, and that the educated earn more because they can produce more.²2Human-capital theory (G. Becker, Human Capital, 1964; J. Mincer; T. Schultz): education and training build productive capacity, so the educated earn more because they are more productive. A degree is an investment — its costs (tuition, and earnings forgone while studying) paid for a higher wage across a working life — and the measured return is substantial. On this view a degree is an investment, undertaken like any other: the student pays its costs — the tuition, and the years of earnings forgone while studying — in exchange for a higher wage across a working life. The surgeon earns her premium because medical school genuinely taught her medicine; the engineer earns his because he can do what the untrained cannot. There is nothing mysterious here, and a great deal of truth in it: much of what schooling teaches is real and useful, the professions plainly require it, and the workers who hold it can, in fact, do things the rest of us cannot. Were this the whole story, the wage would simply be paying, exactly as it ought, for a skill that did not exist before and now does.

But there is a second answer, stranger and more unsettling, first set out by Michael Spence in 1973. Suppose, Spence said, that school taught nothing useful whatever — and watch how the premium could survive even so. People differ in ability, and that ability is hard for an employer to see across a hiring desk. But it is easier for the able to get through school than for the unable: they find the work less costly, less punishing, less likely to defeat them. So the able go further in school — not because it improves them, but because they can; and employers, knowing this, pay for the diploma as a proxy for the ability that reaching it reveals. The schooling sorts rather than builds. On this account the premium is just as real, but its cause is not what it appears: the educated earn more not because school made them productive, but because arriving at the diploma proved them able to begin with.³3Signaling theory (M. Spence, ‘Job Market Signaling,’ 1973; Nobel 2001; cf. K. Arrow’s ‘filter,’ 1973, and J. Stiglitz’s ‘screening,’ 1975). Since ability is costly to observe and the able find schooling less costly, education sorts on pre-existing ability and the credential signals it — so the premium can survive even if school builds no skill at all. Here is the whole quarrel in a single line: in the first theory, school creates the gap between workers; in the second, it merely discloses a gap that was already there.

Does school create the difference, or only reveal it?

How could one ever tell the two apart, when both predict the very same thing — that the educated earn more? The deciding evidence, such as it is, lies in a quirk the human-capital story strains to explain. If each year of school simply added skill, then each year ought to add roughly equally to the wage. It does not. The year that grants the credential pays out of all proportion to the rest: by careful estimate the twelfth grade — the one that yields the high-school diploma — lifts earnings more than the ninth, tenth, and eleventh combined, and the final year of college more than the first three together.⁴4The sheepskin effect (diplomas were once written on sheepskin): the credential year pays disproportionately — twelfth grade more than grades nine through eleven combined, the last college year more than the first three — and the gap persists after controlling for ability. It strains pure human capital, since the learning is spread evenly while the wage leaps at the credential. Its reading is itself debated — a last semester is no costly signal in Spence’s strict sense — but the paper plainly earns beyond the years. The leap is not in the learning, which is spread evenly across the years; it is in the paper, which comes only at the end. Economists call it the sheepskin effect, after the hide on which diplomas were once written, and it is the signal theory’s strongest exhibit: the identical coursework, lacking the credential, pays a fraction of what it pays with it — as though the market valued not what was learned but the proof of having finished. Offered a Princeton education with no diploma, or a Princeton diploma with no education, the worker who wanted the higher wage should take the paper.

If much of what the diploma earns is signal rather than skill, an uncomfortable consequence follows, and Bryan Caplan has pressed it hardest: that a large share of our vast outlay on education buys no new productivity at all, only a reshuffling of who stands where in the queue. By Caplan’s reckoning — contested, but seriously argued — as much as four-fifths of the wage premium may be signaling.⁵5B. Caplan, The Case Against Education (2018), estimates roughly 80 per cent of the premium is signaling, making schooling largely a wasteful ‘arms race’ of credential inflation. The figure is much disputed: critics put pre-existing ability alone at a quarter to a half of the premium, and the mainstream view is a hybrid in which human capital is substantial. How much is skill and how much signal is genuinely unsettled. If so, education becomes in part an arms race: each person rationally seeks more of it to stand out, but as all do, the credential demanded for any given job ratchets upward — yesterday’s diploma becomes today’s bare minimum — while the work itself grows no more skilled. What is privately essential turns out to be socially close to wasteful, effort spent only to hold one’s place. And yet the case must not be overstated, for the other hand is heavy too. Education plainly does build real skills, nowhere more than in the technical and professional fields that keep the modern world running; the able are not merely sorted but genuinely trained; and the careful attempts to weigh how much of the premium is signal arrive nowhere near agreement — some put the pre-existing-ability share at a quarter to a half, leaving real room for skill that schooling truly makes. The honest verdict is a mixture whose proportions no one has fixed: the diploma is part investment and part ornament, part skill and part sorting, and which part rules remains among the most consequential open questions in the economics of work.

Part skill, part sorting — and the wage pays for both.

So the piece of paper does at least three things at once, and the wage pays for all of them without troubling to say which. It certifies a skill that schooling truly built — the human capital that lets the surgeon operate. It signals an ability the holder brought to the gate and merely proved by passing through — the sorting that would lift her wage even had school taught her nothing. And it marks, quietly, an advantage she may have carried all along: the talent or fortune or family that made the diploma reachable for her and not for others, so that some of what looks like a return to education is really the return to whatever made the education possible.⁶6The premium braids three strands — human capital (skill built), signaling (ability revealed), and the bias of pre-existing advantage (those who would have earned more in any case being likeliest to get the schooling). They are tangled so tightly in the data that no one has cleanly separated them, which is why the value of education stays among economics’ most consequential unsettled questions. These three strands are braided so tightly through every educated wage that no one has cleanly pulled them apart, and perhaps no one can. The diploma arrives as a single thing, and is paid for as one.

And so the folio’s anatomy comes nearly to its close. We have weighed the wage’s justice and its hours, its trust and its compulsion, its amount and its boundary, its rigidity and its measure; the authority it buys, the risk it sheds, the machine that rivals it, the meaning it cannot hold, the demand it sustains, the leisure it might buy back. Here we have asked why one wage towers over another, and found the answer folded inside a credential whose worth we cannot fully account for — part the skill it builds, part the rank it confers. The wage rewards the educated richly, and certainly. What exactly it is rewarding, it has never had to know.

We asked, at the outset, the simplest question — what is a wage? — and have spent fifteen leaves discovering that it has no simple answer, or rather too many. It is a moral fact, the first leaf found, older and more imperious than any bargain; a price paid for time, said the second, once time had been made into a thing that could be sold; a wager, a pledge advanced on trust, the worker performing first and paid in arrears; the toll exacted once the commons were closed and there was no living but the wage; a figure floating in a range with no fixed point, settled by power; the boundary that lets some labour count as work and the rest go uncounted; the one price that will not fall; the measure beneath all other prices, the numéraire in which the cost of everything is secretly told.¹1The fifteen readings, i–xv, each argued in its own leaf. The question they answer — what a wage is — has been put by moral philosophy, by history, by law, by economics, and by sociology, and each discipline has answered in its own coin; the folio’s fifteen answers run across them all. And it is the purchase of obedience, the right to direct a person within bounds; the worker’s insurance, a steady claim bought with a lower average wage; the machine’s oldest rival; the silence in which a calling’s meaning goes unpriced; the demand that turns out to buy back the world’s own product; the price of an hour torn from a finite life; and the premium on a paper whose worth no one can fully account. Fifteen answers. Each was argued; each, in its leaf, was true. And no two of them are the same.

Set the fifteen answers side by side and the temptation is to ask which is the real one — to suppose that beneath the moral fact and the market price and all the rest there must lie a single true nature, and that the others are appearances, overlays, distortions laid upon it. Every partial student of the wage yields to this temptation in his own direction. The economist sees the price and takes the morality for sentiment laid over it; the moralist sees the justice and takes the market for a distortion of it; each mistakes the one face turned toward him for the whole.²2Each lens flatters its own object. But the disagreement between the economist and the moralist is not, at bottom, about the facts; it is about which face of a many-faced thing to call its true one — a question the thing itself does not answer, because it wears them all at once. But the leaves will not permit the reduction. The wage is not really a price with a moral gloss, nor really a moral fact with a market mechanism bolted on; it is, irreducibly and at the same time, both — and thirteen other things as well, each as real as the next. A thing with fifteen true natures is not a thing whose real nature we have missed. It is a thing whose nature is to be many. That is the first lesson of the folio and the last: the wage cannot be reduced, and the wish to reduce it is the common error of everyone who has only ever seen one of its faces.

A thing with many true natures has no one nature.

Why should one thing be made to carry so much? Because of where it sits. The wage is the precise point at which a human life — with all that the earlier leaves loaded into it: its hours, its trust, its meaning, its moral claim — is exchanged for a share of the world’s product, with all that this in turn carries: its prices, its markets, its mathematics. It is the hinge between the human and the economic, the one place where a person and an economy meet and are forced to agree on a number.³3The fourth leaf, after Polanyi, called labour a ‘fictitious commodity’ — a thing sold on markets that was never made to be sold, because it is only the life of a person looked at as a thing to hire. The wage is the price at that impossible junction, which is why it must speak both languages — the moral and the mechanical — at once. Nothing else sits quite there. A price for wheat is only economic; a duty of justice is only moral; but the wage is the price of a person’s time, which is to say the price of a portion of a person, and so it is condemned to be both at once — to carry the ethics because it is about a life, and the arithmetic because it is struck in a market. Everything the folio found in the wage flows from that single fact of position: that it is the place where a life is turned into a number, and the number turned back into a life.

From this single fact of position, two consequences follow, and between them they explain much about the world. The first is that the argument over the wage can never be won. When people contend over what a wage should be — over a minimum, over a gap, over what is fair — they are very often not disagreeing about the facts at all but pointing at different faces of the same thing: one sees the justice owed, another the price that clears a market, a third the jobs that might be lost, a fourth the demand that might be gained, and each is looking at something that is really there.⁴4The minimum-wage quarrel is the folio’s faces colliding: the justice of the first leaf against the indeterminate range and disputed employment effects of the fifth, the floor that will not fall of the seventh, the household demand of the thirteenth. Each is true; none is the whole; so the dispute has no clean resolution, only a balance struck and re-struck. No side can simply prevail, because no single face is the whole truth, and the wage genuinely is every one of the things the contending parties take it to be. The permanence of the quarrel is not a sign that the question is muddled or the disputants foolish. It is a sign that the thing in dispute really does possess all those natures at once, so that each party holds a real piece of it, and none holds all.

Each party holds a piece of it; none holds all.

The second consequence is that the wage cannot be cleanly reformed. Because its faces are not separate but joined — because the same number is at once a cost and an income, a measure and an incentive, a moral claim and a market price — any hand laid on one of them moves all the others, commonly in ways the reformer never meant.⁵5Raise the floor and you have touched not justice alone but the ratchet that will not let it fall (vii), the machine that may be called to replace the dearer worker (xi), the household income that is someone’s demand (xiii), and the unpaid labour just past the wage’s edge (vi). Subsidize the schooling and you feed the signal and the arms race (xv) as surely as the skill. No lever moves one face alone. Raise the floor, and you have touched not only justice but the ratchet that will not let it fall again, the machine that may be summoned to replace the dearer worker, the household income that becomes another’s demand, and the unpaid labour just beyond the wage’s edge. Subsidize the schooling, and you feed the signal and the arms race as surely as the skill. There is no lever that moves a single face by itself. This is why wage policy is the most treacherous kind, its every measure trailing consequences through all the others, and its results so endlessly contested: one is never merely adjusting a price, but always, in the same motion, adjusting fifteen things that happen to share one number.

So we end where the word itself ends: at a sum. A wage is, in the plainest sense, a sum of money — and it is also the sum of everything the folio has found, the total into which a life’s hours and the world’s machinery are added together and rendered as a single figure. The fifteen leaves have not reduced that figure to any one of its meanings; they have shown why it cannot be reduced — why the wage is condemned, by its station at the meeting of the human and the economic, to be moral and mechanical, free and forced, personal and systemic, all at once and without remainder.⁶6The eighth leaf named the wage the numéraire — the measure beneath all prices. The folio’s final claim widens that one: the wage is the measure beneath all prices and a moral fact and a bond and a boundary and the rest, held together in a single number — and that this is not a confusion to be cleared but the truth to be kept. To understand the wage is not to find the one reading that dissolves the others. It is to hold all fifteen at once, and to let none of them be the last word.

The folio is finished; its subject is not. The first leaf closed by saying that the argument over the wage remains open, and the fifteen that followed did not close it, and could not — for the argument is not a defect in our understanding but a property of the thing, the permanent and structural openness of a number in which a life is at stake. We have not settled what the wage is. We have drawn instead the map of why it can never quite be settled: fifteen readings, one number, and a human life folded inside it — waiting, as it always has, to be priced again tomorrow morning.