Numéraire
Volume the Seventeenth · of the table of twenty-six
The Institutions · Group IV · № 17 · Tx

The Tax

who really pays

The tax is the fourth of the institutions, and in one respect the most revealing: it is the policy in which the gap between what is intended and what results is the widest in all of economics. A tax is the compulsory claim by which the state takes, without a direct return, the resources to fund the foundation the market rests on — the courts that enforce the contract, the power that backs the money, the public goods the market will not provide for itself. But the volume's subject is not chiefly what the tax funds; it is who, in the end, pays for it. For the law that levies a tax names a payer — the person who must hand it over — and we assume that the named payer is the one who bears it. He is not. The burden of a tax, once imposed, does not stay where the law puts it; it slides through prices, the buyer's rising and the seller's falling, until it settles on whoever cannot escape the taxed trade — the inelastic, the stuck — regardless of whom the law named. The legal incidence, who pays, and the economic incidence, who bears, are two different things, and only the first is the law's to choose. The twelve plates follow the burden of a tax from the wedge it drives and the surplus it destroys, through the elasticities that divide it and the optimum that would tame it, to a discovery that joins this volume to the question of the state itself: that the market, not the law, decides who bears a tax, so that the state can choose whom to tax, but not whom to burden.

in print · № XVII · 12 of 12 plates
← The complete table of volumes

The argument of the volume

Begin with what a tax does to a market. A tax drives a wedge between the price the buyer pays and the price the seller receives: the buyer pays more, the seller gets less, and the gap goes to the state. This wedge does more than transfer money; it shrinks the quantity traded, for at the higher price some buyers withdraw and at the lower price some sellers do, and the trades that no longer happen — mutually beneficial exchanges the wedge has made unprofitable — are a pure loss, the deadweight loss, collected by no one. Dupuit was first to see it: the cost of a tax is not the revenue it raises but the revenue plus this lost surplus, the trades destroyed as well as the trades transferred. And the loss grows with the square of the rate, so that broad, light taxes do less damage than narrow, heavy ones. A tax always costs the economy more than it brings the treasury.

But the wedge, having a size, must be divided — and how it divides is the volume's central question. The buyer pays somewhat more, the seller receives somewhat less; who bears the larger share? Not whom the law names. It makes no difference, to who bears a tax, on which side the law lays it: the price adjusts to the same division either way, because the burden flows to whoever cannot escape the taxed trade. The side that can readily retreat — the elastic side — sheds the burden by buying or selling less; the side that is stuck — the inelastic side — absorbs it. So the burden settles on the inelastic, on the necessity no one can forgo, the labor that must be sold, the land that cannot move; and elasticity is a fact about the market, not the law. The legal incidence, the payer the law names, does not determine the economic incidence, the bearer the burden finds. The payroll tax falls on the worker, not the employer; the corporate tax on people, never on the corporation, which is no one and can bear nothing. The law names who pays; the market decides who bears.

From this divide the volume's findings follow. The design of a good tax is a conflict between efficiency, which would tax the inelastic, who distort least, and equity, which would tax the able, who can best bear it — and the two collide, for the inelastic are often the poor, and the able are those whose effort taxation discourages; Mirrlees gave the conflict its deepest form, in the optimal income tax that redistributes without killing the effort it taxes. One tax escapes the conflict entirely: the corrective tax on a harm, which mends the market rather than distorting it, taxing the pollution or the congestion the market overproduces, so that its deadweight loss is negative and it pays a double dividend. And beneath all of it lies the deepest turn, which the flagship draws out: that the tax, the most direct-seeming exercise of the state's power — a bare command to pay — is the policy whose true effect most escapes the state's control, for the state can compel the payment but not the bearing, can tax but not burden. Yet whoever bears it, the tax funds what the market cannot fund itself — the public goods defeated by free-riding, the institutions on which exchange rests — and it must be compulsory precisely because those goods are public, fundable only by removing the choice not to pay. The tax is the price of the market, exacted by a power the market cannot question, and borne by whoever, in the end, cannot escape.

The argument remains open. What follows is an attempt to fix its exact edges.

The twelve plates of this volume

  1. i The Wedgein print Dupuit; the wedge a tax drives between buyer and seller, and the trades it destroys — the deadweight loss no one collects. Read the plate →
  2. ii The Tariffin print Metzler; the tariff, the terms of trade, and the conditions under which a duty can undo its own protection. Read the plate →
  3. iii The Shiftin print Jenkin; how the burden divides — not as the law dictates but by elasticity, settling on whoever cannot escape the taxed trade. Read the plate →
  4. iv The Optimumin print Mirrlees; the conflict of efficiency and equity, and the optimal tax that redistributes without killing the effort it taxes. Read the plate →
  5. v The Correctivein print Pigou; the tax that mends rather than wounds — taxing a harm to make its author reckon the whole of what it costs. Read the plate →
  6. vi The Landin print Henry George; the land value tax, borne wholly by the owner and distorting nothing, because land cannot move. Read the plate →
  7. vii The Bracketin print progressive taxation and the income tax; the larger share, and what the larger share costs in effort foregone. Read the plate →
  8. viii The Burdenin print the deepest turn: the law names who pays, but the market decides who bears — the state can tax, but it cannot choose whom to burden. Read the plate →
  9. ix The Restraintin print public choice; the state as a revenue-maximizer, and the constitutional limits on the power to tax. Read the plate →
  10. x The Consentin print consent and the social contract; by what right the state takes, and what makes the taking just rather than theft. Read the plate →
  11. xi The Loopholein print avoidance and evasion; the shadow economy, and the gap between the statutory rate and the rate that is paid. Read the plate →
  12. xii The Deficitin print the deficit; spending beyond taxes, the budget constraint, and the deferred taxation of borrowing. Read the plate →

In closing

This volume took up the fourth of the institutions: the tax, the compulsory claim by which the state takes the resources to fund the foundation the market rests on — and the policy in which the gap between intention and effect is the widest in all of economics. Its subject was not chiefly what the tax funds but who pays for it; and its answer was that the question has two meanings, the legal and the economic, and that only the first is the law's to settle.

The volume traced the burden of a tax from the wedge to the bearer. Dupuit showed what a tax does — drives a wedge between buyer and seller, destroys the trades the wedge makes unprofitable, and so costs the economy the revenue plus a deadweight loss that grows with the square of the rate. Jenkin showed how the wedge divides — not as the law dictates but by elasticity, the burden settling on whoever cannot escape the taxed trade, so that the legal incidence and the economic are two different things. Mirrlees showed how a tax might best be laid, balancing the efficiency that would tax the inelastic against the equity that would tax the able, in an optimal schedule that redistributes without killing the effort it taxes. And Pigou showed the one tax that mends rather than wounds — the corrective tax on a harm, whose deadweight loss is negative, taxing the bad to spare the taxing of goods.

And then the volume found, beneath the question of who pays, the limit of a power. The tax seems the most direct exercise of the state's authority — a bare command, without exchange or consent, that this person shall pay. Yet Harberger showed that the burden, once imposed, does not stay where the law puts it but slides through the whole economy, by the cold logic of elasticity, to settle on whoever cannot move — so that a corporation, being no one, bears nothing, and its tax finds its way to people the legislator never named. The state can choose whom to tax; it cannot choose whom to burden. And yet, whoever bears it, the tax funds the foundation the market rests on — the public goods the commons could not provide, the courts that enforce the contract, the power that backs the money — and it must be compulsory, because those goods are public and free-riding would defeat any voluntary funding of them. The tax is the price of the market, exacted by compulsion and borne by the inelastic. With the tax, the state has its revenue; but the tax has exposed the question of the state's power itself — what it may rightly command, and what restrains it. That is the question of the last institution, the rule, to which the family now turns.

The law names who pays a tax; the market, by the cold logic of elasticity, decides who bears it — and the two are rarely the same. Strip away the name on the statute, and ask who cannot escape the taxed trade, for there the burden settles, whatever the law intended. Who really pays?