A tax is a price, and a price is an incentive. Tax something and you make it more expensive, so people do less of it. Ronald Reagan once mocked how readily governments reach for that lever:

Government's view of the economy could be summed up in a few short phrases: If it moves, tax it. If it keeps moving, regulate it. And if it stops moving, subsidize it.— Ronald Reagan, remarks to the White House Conference on Small Business, 1986 (delivered, to laughter, as a jab at over-government)

It was a partisan applause line, not a neutral description — but the mechanics underneath are real and apply to any tax. A tax discourages whatever it touches. Usually that discouragement is an unwanted side effect: we tax cigarettes mostly for revenue and get less smoking, we tax income and get a little less work. Occasionally the discouragement is the whole point — taxing things we'd like less of — an idea we'll sharpen later. Either way the quantity shrinks, which is exactly the wedge we're about to draw. And the flip side of Reagan's line — "subsidize it" — is the mirror image, a tax with the sign reversed, waiting later in this chapter.

So, the wedge. Picture any market at its equilibrium, the one we measured in Chapter 6. Now the government places a tax of, say, $15 on each unit. The immediate effect is a tax wedge: the price the buyer pays and the price the seller keeps can no longer be the same number, because $15 of every transaction now goes to the government. The buyer pays one price; the seller pockets that price minus $15. The market settles at a smaller quantity, because some trades that barely made sense before don't survive the wedge.

Here's the first surprise, and it's the heart of the chapter. It does not matter whether the law taxes the seller or the buyer — the outcome is identical. Tax the seller $15 per unit and they try to charge more; tax the buyer $15 and they offer sellers less. Either way the same wedge opens, the same quantity trades, and the same split of the burden emerges. The legal incidence— whose name is on the check — has no effect on the economic incidence, who actually ends up worse off. As economists put it: who pays the tax is decided not by the laws of Congress but by the laws of supply and demand.

If the law doesn't decide who bears a tax, what does? The answer is the most useful idea in this chapter: the burden falls on whichever side can least escape it. "Escape" is just elasticity in plain clothes. An elastic side has alternatives — substitutes to buy instead, other markets to sell into — so when a tax pushes the price the wrong way for them, they simply do less of it and dodge most of the burden. An inelastic side is stuck: no good substitutes, no easy exit, so they swallow the tax to keep trading.

What makes one side elastic and the other stuck? A handful of things, and you can feel every one of them without touching a number. Demand is more elastic when buyers have somewhere to go: close substitutes exist (one cola with another right beside it on the shelf), the purchase is a discretionary luxury rather than a necessity, it eats a big share of the budget so a price jump really stings, it's defined narrowly ("a Honda Civic" has substitutes; "a car" has almost none), or buyers have time to adjust. Flip each of those and demand turns inelastic — no substitutes, a necessity, a trivial share of spending, broadly defined, no time to react. The supply side runs on the same logic: sellers are elastic when they can redeploy easily (spare capacity, time, transferable skills) and inelastic when they're stuck (a built shipyard, land that can't be moved).

Train the reflex. For each good below, decide elastic or inelastic— and why — before you tap.

This is centuries-old wisdom. The line attributed to Jean-Baptiste Colbert, Louis XIV's finance minister, is that "the art of taxation consists in so plucking the goose as to obtain the largest amount of feathers with the least possible amount of hissing." Strip away the cynicism and it's an elasticity rule in disguise: pluck where there's least hissing — tax the side that can't fly off. The goose that can escape (the elastic side) hisses loudly and loses few feathers; the goose that's stuck (the inelastic side) is plucked bare. The reason a luxury-yacht tax fails and a gasoline tax succeeds in raising revenue is the same reason: one goose can fly, the other can't.

That's the whole yacht story. The wealthy buyers were elastic — boats abroad, used boats, no boat at all — so they escaped the tax by not buying. The boat builders were inelastic — their shipyards and skills couldn't pivot to making something else — so the burden rolled downhill onto them as lost work. The rule, worth memorizing: the more elastic side pays a smaller share of the tax; the more inelastic side pays a larger share. Think of relative elasticity as relative ability to flee. Whoever is more stuck, pays more.

You can even read it off the graph without the rule. In our anchor market, demand is steeper (slope −2) than supply is shallow (slope +1) — the steeper curve is the less elastic one, so buyers are more stuck, and indeed buyers bore $10 of the $15 tax while sellers bore only $5. The numbers and the intuition agree: the inelastic side carries more. (We're working from the intuition of elasticity here — who's stuck. The number that makes it precise gets its own chapter later.)

A tax does two things at once, and it's vital to keep them separate. First, it raises revenue: the government collects the tax on every unit still sold. That's a transfer — surplus moving from buyers and sellers to the public purse, to fund roads or schools or whatever the state does. On the graph it's a rectangle: the tax per unit (the height of the wedge) times the quantity still traded (the width). In our market, $15 × 15 units = $225 of revenue.

Second, and more subtly, a tax destroys value. Some trades that used to happen — where the buyer valued the good above the seller's cost — no longer clear once the wedge is driven in. Those gains from trade simply vanish, captured by no one, not even the government. This is deadweight loss, the same triangle from Chapter 6, now caused by a tax. It sits between supply and demand, from the new quantity out to the old one.

Here the taxes-as-incentive idea bites the government back. Point the incentive at the tax rate itself: if a higher rate collects more per unit, why not keep raising it forever? Because the rate is also an incentive to shrink the very thing being taxed. At a 0% rate the government collects nothing — obviously. But at a 100% rate it also collects almost nothing: if the state takes every dollar, almost no one bothers to earn, sell, or report the dollar in the first place, so the base all but vanishes. Zero at one end, near-zero at the other — so somewhere between, revenue must peak. Plot revenue against the rate and you get a hump: the Laffer curve.

Everything so far has answered who does pay a tax and what it costs — pure positive economics, the same answer for any analyst who runs the numbers. Now the question that actually dominates every real tax fight: who should pay? That word — should— has crossed a line. It is a normative question, and no supply-and-demand diagram answers it.

Start with what economics can say, because it's more than you'd guess. We can classify any tax by how its burden tracks income. A progressive tax takes a bigger share of income the more you earn — the federal income tax, with its rising brackets. A regressive tax takes a bigger share the less you earn — a flat sales tax is the classic case, because lower-income households spend a larger fraction of their income, so a tax on spending bites them harder. A proportional (or flat) tax takes the same share at every level. Which bucket a tax falls into is a factual, checkable question — positive economics; two experts will agree.

What economics can't settle is which of those is fair. Here two reasonable principles collide. The benefit principle says you should pay for what you use — a gas tax that funds the roads you drive on feels fair on those grounds. The ability-to-pay principle says the burden should fall where it's easiest to bear, which points toward progressive taxation, taking more from those who have more. Both are coherent; they often recommend opposite taxes; and nothing in economics crowns a winner, because the choice rests on values — fairness, desert, what we owe each other — that the model simply does not contain.

This is the discipline's honest boundary, worth stating plainly: economics is unusually powerful at the positive half — tracing a tax's consequences with real precision — and has no special authority over the normative half, whether those consequences are good. It's the same efficiency-versus-equity split we first drew in Chapter 5. A good citizen uses the economics to get the consequences right, then argues fairness openly, as the genuine values question it is. And this course returns to the case for taxes and redistribution, at full strength, later on. 

Everything you learned about taxes applies to subsidies, just in reverse. In Concept 1, a tax drove a wedge that made buyers pay more than sellers received, and shrank the quantity. A subsidy does the exact opposite: it drives a wedge that lets sellers receive more than buyers pay, with the government covering the gap, and the quantity grows. Where the tax wedge pushed the two prices apart with a gap the government collected, the subsidy wedge pushes them apart with a gap the government pays.

We'll use it on the same market we've used all unit — demand P = 100 − 2Q, supply P = 40 + Q, which cleared at $60 and 20 units. A $15-per-unit subsidy means sellers now get $15 more than buyers hand over, so they're willing to supply at any buyer-price $15 below their old cost. The effective supply shifts down by $15 — the mirror of the tax, which shifted it up. The market grows to a larger quantity, buyers pay less, sellers receive more, and the government pays the difference on every unit.

Concept 2 showed that it doesn't matter who signs the check on a tax; tax incidence is determined by elasticity. The same is true for subsidies: the more inelastic side captures the larger share of a subsidy — exactly the side that would have borne the larger share of a tax. E stands for "escape" as well as "elasticity." In this case, the party that cannot escape is stuck. Stuck means you can't walk away, and the side that can't walk away both eats the tax and pockets the subsidy. So the entire question "will this subsidy actually help the people it's aimed at?" reduces to one prior question: which side here is stuck?

We're going to use the example of student loans in the higher ed market. Buyers in a college market have options: skip college, choose a trade, and lots of universities. The elasticity of demand is far more elastic than suppliers. Supply is inelastic because universities that can't or won't add seats quickly. In part, this is because their prestige is bound up in scarcity: you will often find that some suppliers don't want to increase supply because it runs contrary to their incentives. Buyers can spend the aid; sellers can't easily produce more in response. So the stuck side is the sellers, and the sellers capture the subsidy: when you pass subsidies to the students, the extra demand bids up the price, tuition rises, and much of the aid lands in the schools' budgets rather than lowering the real cost to students. Subsidize buyers — students, with loans — in a market where supply is inelastic—and watch as they get very little out of the deal. The intended beneficiary is not the actual beneficiary. Watch the same wedge, now in a steep-supply market:

Higher education is the sharpest test of everything in this chapter, because it is one of the clearest cases of a subsidy that ends up benefiting suppliers. When a good is expensive and important, the reflex is to intervene by subsidizing demand. But if the reason it's expensive is that supply is constrained, that intervention predictably backfires — it inflates the price and enriches incumbents, defeating its own purpose. The deeper economics points to a better solution: if the problem is that supply can't grow, the answer is to free supply to grow — to let the market do what markets do.

Note that the case we are making is not that subsidies are bad. Rather, the case is that there are predictable effects flowing from the imposition of a subsidy. It may well be worth the distortions on net if we help people get to college, but a good economist will spot the potential problems: increased price, and increased (distorted) quantity.

On the demand side, decades of expanding aid against a slow-moving supply of seats is exactly the Bennett setup. Those decades have made demanders grateful, but haven't helped them much: prices have gone up, and quantity has expanded dramatically.

But look at the supply side and you find the market suppressed in at least two ways. Harvard receives enough money to vastly expand its offerings, but chooses not to. Why? Because the prestige of Harvard is in the limit to its number of spots. The Ivy Leagues command high prices because they are prestigious. This tendency to suppress output is what monopolists do in our last unit. The "free market" response to high prices is to let markets expand, not to subsidize demand.

But there is another sense in which education suffers from a monopoly: for most families, which school a child attends is decided by their home address — a local monopoly, assigned, with no competition, no entry, no price signal, no rival down the street fighting for the student. Economics has a sharp prediction about any monopoly insulated from competition: it will be less responsive, less efficient, and slower to improve than it would be under rivalry. The stagnation many feel in assigned schooling is what the market model predicts when you remove competition.

School choice — vouchers, charters, open enrollment — is the proposal to let that market function: let families choose, let schools compete for them, let good ones grow and weak ones feel pressure to improve. Milton Friedman proposed it in 1955 on exactly this logic. The market prediction is not mainly that any one student's test scores will jump; it's that competition disciplines producers — that schools forced to compete for families will get better and use resources more carefully, just as competition does in other markets. That is the genuine, market-based alternative to subsidizing a suppressed system: don't pour more demand into a market you've prevented from working — let it work.

Step back and look at the whole chapter at once. A tax aimed at the rich landed on boat builders. Aid aimed at students landed, in part, in college budgets. Neither policy failed because anyone meant harm. Both failed the same test: they asked who deserved the burden or the help, and never asked who could escape it. Good intentions are not a defense against elasticity. Before you cheer any tax or any subsidy, ask the question the yacht tax never asked: who is stuck?