When you get a great deal at the store, say, you buy a nice backpack for $5 on clearance that you would have paid $50 for, you receive $45 in consumer surplus. When you get a great college job that pays way above the market rate, say, $25 per hour when you would have insisted on at least $15, you get $10 an hour of producer surplus. Last chapter we learned about the gains from trade, which we could only describe. Now, we see that they can be measured — in dollars, right off a supply-and-demand graph. This is the power of markets, quantified.

Most of economics asks what will happen — where the price lands, how much gets traded, etc. This is a different kind: it asks how good an outcome is for the people living in it, and whether some other outcome might be better. Welcome to a branch of the discipline called welfare economics. A warning about the word, because it trips up almost everyone. In everyday American speech, "welfare" means public assistance — a government check. Here it means something older and broader: well-being, how well a person's life is going. A welfare analysis isn't a study of welfare programs. It's an accounting of how much better off a market leaves the people in it — and how much worse off when the market is blocked.

In the last chapter, "gains from trade" came from specialization — Crusoe fishing, Friday gathering, each doing what they sacrifice least to make. Ch 5But value is also created whenever a buyer and seller simply exchange, with no specializing at all: if you'd happily pay $90 for a thing and I'd let it go for $45, the trade manufactures value out of the gap between us. These are two engines — gains from specialization and gains from exchange — and they are not rivals; they compound. You specialize to make a thing at low cost, then exchange it with someone who values it more than you do, and value is created at both steps. The tool we build now measures both, and we can use a simple supply and demand curve to measure it. What we're looking for is the wedge between what a unit is worth to a buyer and what it costs a seller.

Remember back to chapter 2: each point on a demand curve is somebody's willingness to pay — the marginal value of that unit to that buyer. The curve slopes down because, one buyer to the next and one unit to the next, that value falls. So the demand curve is a ranked list of what each unit is worth to the people who want it.

Now suppose the market price is $60 and you'd have happily paid $90. You buy, hand over $60, and walk away with $30 of value you didn't have to pay for. That $30 is your consumer surplus— the value you captured beyond what you spent. Add up that gap for every buyer in the market and you get total consumer surplus: on the graph, it's the area below the demand curve and above the price.

Sellers have a mirror-image story. The supply curve is each seller's willingness to accept— the least they'd take, which is their cost of making that unit. A seller who'd have sold for $45 but receives $60 keeps $15 of producer surplus. Summed across sellers, that's the area above the supply curve and below the price. Put the two together — consumer surplus plus producer surplus — and you have total surplus: the entire value the market created by letting these trades happen.

Those triangles are no classroom abstractions. They are real measures of how much better off you are, in economic terms, from specialization and trade; and some of them are simply enormous. Try this: what does Google cost you? Nothing — the price is zero. On the graph, that means your consumer surplus is the entire area under your demand curve. In a 2019 study, the economists Erik Brynjolfsson, Avinash Collis, and Felix Eggers set out to measure it by flipping the question around. Instead of asking what people would pay for search, they asked how much people would have to be paid to give up every search engine for a year. The median answer was $17,530. Email came in around $8,400; digital maps around $3,600. None of that shows up in GDP, which counts what people spend, not what they get. It shows up only as surplus — a quiet, enormous gap between what these services cost and what they're worth.

The equilibrium quantity maximizes total surplus — exactly, not approximately. Every unit a buyer values above its cost (demand above supply) creates surplus, and at the market price it trades. Every unit beyond that would cost more to make than anyone will pay, and at the market price it doesn't. Produce less and you leave good trades on the table; produce more and you force bad ones.

Economists love (like love love) efficiency. We have a bunch of different kinds. They're all lovely. Three are worth keeping straight:

That a competitive equilibrium passes the allocative and Pareto tests at once is important enough to be called a theorem: the First Fundamental Theorem of Welfare Economics, or just the First Welfare Theorem. A competitive market equilibrium is Pareto efficient. Nobody plans it; each person watches only the price and their own situation. It's the gains from trade of Ch 5and the spontaneous order of Ch 3with a dollar figure attached: Adam Smith's invisible hand, weighed and counted.

If that sounds too good to be true, Vernon Smith agreed. As a Harvard graduate student he had watched a classroom market run by his professor, Edward Chamberlin, in which prices wandered all over the place. So in 1956, on the first day of his own class at Purdue, Smith tried again. He gave each student a secret card — buyers a maximum they could pay, sellers a minimum they could accept — and let them call out bids and offers across the room, round after round. No one knew the curves but Smith. He expected a mess. Instead, prices closed in on the equilibrium he had drawn, and the class captured nearly all the possible gains from trade. The result survived decades of variations, and in 2002 it helped earn him a Nobel Prize.

The theorem does have fine print. It holds only when:

When a condition fails, the efficiency of the market collapses. Much of the second half of this course is a tour of those failures. The best-known critiques of markets — Arthur Pigou's on spillovers, Joseph Stiglitz's on hidden information — are claims that the fine print fails, not that the theorem is wrong.

Note that the economist's language fails us a bit here. When a layperson hears an economist blathering on about "efficiency," they naturally think: Do they even care about regular human well-being, or just cold corporate profits? But notice the pivot: in welfare economics, efficiency is human flourishing. When an economy achieves allocative and productive efficiency, it isn't just optimizing a spreadsheet; it is liberating people from the crushing friction of poverty. Look at the alternative—price controls and distorted markets that create empty shelves, endless rationing lines, and wasted human potential. By contrast, the systems that reliably generate economic efficiency are the same ones that give people the security, health, and freedom to pursue meaningful work, raise families, and actually flourish.

Now, let us return to the equity and efficiency debate. If efficiency is so important, is it always at odds with equitable distribution? Yes, allocatively efficient markets will also be Pareto efficient, and that's pretty cool. But you could still live in a world of haves and have-nots, rich trillionaires and those eking out a life under crushing poverty.

The Second Fundamental Theorem of Welfare Economics gives us a possible bridge between efficiency and equity. Any Pareto-efficient outcome — including one far more equal than the market happens to produce — can be reached through a competitive market; but rather than playing games with prices that lead to devastating distortions, the best path is redistribution at the start. Consider a one-time tax on rich to poor: you can still achieve efficiency under such a scheme. You don't have to shrink the pie to slice it differently: change who owns what, and leave the prices alone.

The second theorem also has assumptions, however. The theorem assumes behavior-agnostic lump-sum transfers, but redistribution tends to change behavior. Real taxes and benefits always change something — how  much people work, how much they earn — so real redistribution carries some deadweight loss of its own. We can't quite say that "fairness is free." But we might be able to say "there is usually a more effective way to pursue fairness than freezing a price." The second welfare theorem suggests that maybe the best way to solve poverty is to simply give people money, which we'll take up as a potential policy at the end of the chapter.

If the equilibrium quantity maximizes total surplus, then anything that pushes the market away from it must shrink the pie. The lost surplus has a name: deadweight loss— value that disappears, captured by no one. This is the crucial difference from a simple transfer. When a tax moves $10 from your pocket to the government's, that's a transfer: you lose, they gain, the surplus still exists somewhere. Deadweight loss is worse — it's surplus that stops existing at all, because the trade that would have created it never happens.

Picture our anchor market with something holding the quantity at 10 instead of the efficient 20. Every unit between 10 and 20 was a trade where the buyer valued the good more than it cost to make — a trade that should have happened and created surplus. Block them, and that surplus is simply gone. It shows up on the graph as a triangle between the demand and supply curves, from the restricted quantity out to the efficient one. (In the next concept we'll see that "something holding the quantity at 10" is, concretely, a price control.)

Think back to the beginning of the chapter: that clumsy Christmas gift that someone wants to return? It is deadweight loss in a Santa hat: a "trade" where the giver guesses wrong about the recipient's willingness to pay, so value is destroyed. The fix is the one Chapter 2 implied — let people choose for themselves, because no one knows your marginal utility better than you. Waldfogel's surveys bear this out. Gifts from partners and parents, who know your tastes, lost roughly a tenth of their value; gifts from aunts, uncles, and grandparents lost about a third. And the relatives who know you least turn out to be the likeliest to hand over cash or a gift card. Grandparents, it seems, are intuitive economists. Cash isn't unromantic; it's efficient.

Begin with a question the last three concepts now let us answer in a sharper way: what happens when the price is simply wrong? Suppose it sits above the equilibrium — too high. Sellers, tempted by the high price, want to sell more than buyers want to buy; unsold units pile up as excess supply (often just called a surplus — a word we'll untangle below). Or suppose it sits below equilibrium — too low. Buyers want more than sellers will provide, and a shortage (excess demand) opens. We met the mechanics of both back in Chapter 3. Ch 3What we can add now is the welfare reading: a price away from equilibrium isn't just out of balance — it's inefficient. Value-creating trades are going unmade, total surplus sits below its maximum, and a deadweight-loss triangle has opened.

In a free market, none of this lasts. Excess supply has sellers shading their prices down to move stock; a shortage has buyers bidding the price up. The price slides back to the crossing, the triangle closes, and efficiency is restored — that was Chapter 3's lesson, now with a welfare price tag. Disequilibrium is a force, not a resting place. So the inefficiency of a "wrong" price is, in a free market, temporary: the market corrects itself.

Unless it can't. A price controlis a law that forbids the price from moving — and it is precisely the case where the inefficiency does not self-correct. It is held open, by force of law, indefinitely. There are two kinds, mirror images of each other:

Either way, the controlled price stops the two quantities from meeting, and one side ends up rationed. The rule for how much actually trades is simple and worth memorizing: the short side wins. Whichever quantity is smaller at the controlled price — what sellers will supply under a ceiling, or what buyers will buy under a floor — is the amount that actually changes hands. Because that amount is always below the efficient quantity, a binding control always destroys surplus. The unmade trades between the controlled quantity and the efficient one are a deadweight-loss triangle, exactly the one from the last concept — only now we know its cause has a name.

Anyone who drove in America in the 1970s has seen this diagram come to life. Price controls on oil and gasoline — an outgrowth of President Nixon's 1971 freeze on wages and prices — held the price at the pump below what the market would have set. When oil supplies tightened in 1973 and again in 1979, the price wasn't allowed to rise and ration the gas, so something else did the rationing: the line. Drivers waited for hours, some arriving before dawn. Several states let you buy only on odd or even days, depending on the last digit of your license plate. The remaining controls were lifted in early 1981, and the nationwide gas lines of that decade never returned.

If the gas line is what a binding ceiling looks like, a cheese cave is what a binding floor looks like. For decades the U.S. government propped up milk prices with a floor, and it enforced the floor the only way a floor can really be enforced: by promising to buy whatever farmers couldn't sell at the support price. Farmers responded exactly as the supply curve predicts — they produced more — and the government bought the difference, turning it into cheese, butter, and powdered milk that could be stored. By 1981 it owned more than 560 million pounds of cheese, over two pounds for every American, much of it stacked in underground caves in Missouri and Kansas. Storing all that dairy cost around a million dollars a day, and some of it was going moldy. That December, President Reagan began giving it away: the five-pound blocks of "government cheese" that a generation of low-income families still remembers.

Notice what the cheese does to our triangle. The diagram assumes only the 15 units buyers want get produced. But when the government stands ready to buy the excess, farmers produce all 30 — so society pays the full cost of 15 extra units nobody wanted at that price, and then pays again to store them. The $37.50 triangle is a floor's minimum cost. A floor backed by government purchases costs far more.

Should we simply give cash to the poor? Everything in this chapter bears on the answer. Surplus measures what help is actually worth to the person receiving it. The Second Welfare Theorem says you can redistribute without freezing prices. Price controls showed what happens when you try to help by forcing a price to stand still. The economist's instinct leans toward cash. But the case for in-kind help is stronger than it first looks.

Many readers of this book know one in-kind system from the inside. The welfare program of The Church of Jesus Christ of Latter-day Saints is built on the in-kind side of this table by design. Its handbook asks bishops to avoid giving cash when possible, to provide groceries and services through bishops' orders, to pay bills directly to the provider rather than to the member, and to invite those who receive help to work or serve as they are able. The logic is targeting and relationship: a bishop knows the family, the help is shaped to what they actually need, and the goal is self-reliance rather than dependence. It's a thoughtful answer to the same question economists argue about — and a reminder that where you land depends partly on evidence and partly on what you think help is for.

If you side with cash, what would a cash-first system look like? In 1962 Milton Friedman — the century's most famous champion of free markets — proposed an answer in Capitalism and Freedom: a negative income tax. Pick an income threshold. Earn above it and you pay tax, as now. Earn below it and the tax runs in reverse: the government pays you a fraction of the gap. No storehouses, no vouchers, no price held still by law — just cash, scaled to need, delivered through the tax system.

Friedman's pitch was that it could replace much of the patchwork of programs that had grown up one at a time. Fig 6.5 is one attempt to draw that patchwork.

Two features make Friedman's design clever. Because you keep part of every extra dollar you earn, it never fully punishes work, the way programs that cut benefits off at a hard line can. No more welfare cliff. And because it runs through the tax system, it needs little new bureaucracy. The idea came remarkably close to law: in 1969 President Nixon proposed a version, the Family Assistance Plan, which passed the House and died in the Senate. Its closest descendant is the Earned Income Tax Credit, created in 1975 and now one of the largest antipoverty programs in the country. Federal experiments with the full idea in the 1970s found what the in-kind side would predict: recipients worked somewhat less — modestly for primary earners, more for second earners and teenagers. Friedman explains the idea himself in a short clip linked in Learn More.

So: the sweater, or the cash? For your cousin, the economics is easy — cash, or at least a gift card. For a society, it's harder, and this chapter has given you the tools to see why. Economics can tell you that cash usually delivers more well-being per dollar, that price controls destroy surplus, and roughly what each design costs. It can't tell you how much help a society owes, or whether help should come with conditions and a relationship attached. Those are questions for citizens.