How to use this page

This recap covers the whole September 15 class, the second session on gains from exchange (the first 34 minutes of the topic, from September 10, have their own Part 1 recap and transcript). Class re-ran reservation values and the two-panel figure, then delivered the crossing, Pareto-optimality, the forbidden swap, the middleman, and the transaction-cost figure, and stopped there. The fee ceiling, the rival dealer, and licensing were not reached; they are summarized below from the slides and the post, clearly marked, because the instructor said the wrap-up comes Thursday. Read the four-step diagnostic first. Then use the lecture sections to reconnect each idea to an example from class. Finish with the application checks. The separate complete transcript is available when you want the full explanation, including the car-dealership story, the interest-rate aside, and the exchange about phones and Uber.

What you should be able to do

1. Read two personal worth curves on one axis to find the price and quantity where costless trading stops, state each side’s gain, and explain why neither side trades past that point.
2. Define a Pareto-optimal allocation, explain why voluntary exchange tends to reach one, and distinguish that test from fairness and from productive efficiency.
3. Say where the value of a good goes when trading it is forbidden; list the work a middleman does; and predict how the quantity traded changes when the real cost per unit traded falls.

Movement 02 · Personal worth and the stopping point

Both expect to gain when the price lies between their worths (recap)

Marginal personal worth is a reservation value: the most a buyer would give to acquire a good, or the least a seller would accept to give it up. The instructor’s weekend car purchase made the point. He was prepared to pay the asking price; the dealer’s minimum turned out to be lower; neither side knew the other’s number, and bargaining is the process of finding out where the price can land. Every dollar the buyer bargains off the price is a dollar of surplus that moves from the seller to the buyer; the gains from exchange stay positive, and the haggling only divides them. One reason a dealer’s minimum falls when interest rates rise: a car on the lot earns nothing, and the money from selling it could earn the market rate, so holding inventory costs more (the same logic says Costco-style bulk buying makes less sense when rates are high). The jacket again: a buyer worth $40, a seller worth $25, a price of $32 gives $8 to the buyer and $7 to the seller, and any price between $25 and $40 leaves both better off. No seller takes $24 and no buyer pays $41.

Different worths at the same start make the trade

Both men hold 20 bottles. Sam values his 20th at a full bar, Joe values his at two-tenths, so Joe becomes the seller of water without producing any, and they trade until the marginal worths are the same: Joe’s 10th bottle and Sam’s 30th are each worth six-tenths of a bar. Offer Joe half a bar for one more and he refuses, because parting with another bottle now needs at least 0.6. Why do the worths move? The law of demand: Sam consumes more water and values the next bottle less; Joe consumes less and values it more. Flip the goods and the same story runs in bars, since bars are what Sam pays with. The efficient-market aside applies the idea to stocks: if a stock were systematically overvalued people would sell it, so you should not expect to find mispriced stocks lying around any more than $20 bills on the campus sidewalk. The two triangles on the figure are the gains: Sam’s buyer surplus above the price line and Joe’s seller surplus below it (which is also Joe’s consumer surplus for granola bars). Sam pays Joe six bars for the ten bottles, each gains two, and no dealer takes a cut because here the two men can see the gain themselves; in the real world, seeing gains that others miss and bringing the two sides together is what entrepreneurs are paid for. The first TopHat check asked which interpretation best explains the exchange of 10 bottles at 0.6 bars a bottle: the trade lets marginal worth reach 0.6 for both parties, creating a four-bar gross gain. They do not end with equal bottles, and the trade would not happen if it benefited Joe alone.

With costless exchange, trade stops where the worths meet

Take Joe’s diagram, flip it, and lay it over Sam’s: that is how the course goes from demand and demand to supply and demand. Reading left to right from the 20-bottle start, Sam values the next bottle more at every point until the lines cross. At the crossing the marginal worths are equal; economists call that equilibrium. Beyond it Sam’s worth is below Joe’s, so no offer Sam makes is accepted; and if for some reason they overshot, Sam would become the seller and they would trade back toward the crossing. Trading from 20 and 20 to 30 and 10 leaves Sam two bars better off and Joe two bars better off, and those two triangles measure the gains from trade. Make the trade illegal and both lose exactly that. Change the endowments (Sam 35, Joe 5) and the roles flip, but they still trade until the worths are equal. Two asides carried the equilibrium idea outward: at the gas pump everyone buys until the marginal worth of the last gallon equals the price, so in equilibrium every buyer’s marginal worth of gasoline is the same even though they buy different amounts; and in equilibrium goods of different quality must carry different prices, which is why an Omega and a Casio, or Monster cable and Walmart cable, cannot be perfect substitutes at different prices (the instructor’s conjecture: the same average quality with a lower variance is what the higher price buys). Equilibrium is the marble at rest in the bowl: nothing moves it except a disturbance, such as new care packages or changed tastes.

Pareto-optimal means no gain left. It does not mean fair.

The allocation at the crossing is Pareto-optimal: it is not possible to make one person better off without making another worse off. Take a bottle from Joe and hand it to Sam and Sam gains, Joe loses, and Joe would never agree. Economics supplies no values, so the label says nothing about whether the result is fair, moral, or just; some ethical criteria call it fair and others do not, and economics cannot choose among them. “Is” statements are scientific, “ought” statements are ethical, and “no mutual gain is left” is an “is.” Nor does it say anything about productive efficiency, since nothing is produced. Two student questions extended the idea. Is the classroom Pareto-optimal right now? Probably: taking a phone from one student and giving it to another makes one better off and one worse off, and if the second student wanted the phone badly enough he could offer a dollar for a minute of it. The catch is transaction costs: neither may know the other is willing, so some improving trades do not happen. That is the middleman’s opening, and it is what Uber did for people with cars and people who wanted rides, gains that were too expensive to capture with a phone book. The second TopHat check: after costless voluntary exchange stops at 30 and 10, the allocation is Pareto-optimal but its fairness remains a separate question. It is not fair because they gained equally, not productively efficient because they traded voluntarily, and it cannot still be improved without hurting someone. A last aside drew the limit: hiring a hitman is mutually beneficial to the two parties and still not something the law should allow, and strictly it is not Pareto-optimal either, because the third party wants a say.

Movement 03 · Blocked trades and middlemen

The swap is forbidden. Does the value disappear?

You hold a parking permit and would rather have your friend’s library desk; he would rather have your permit; the university forbids the swap. Why? First, banning transfers keeps the power to assign spaces with the university, and whoever assigns them may be collecting something for it; costless trading among permit holders would make that go away. Second, the assigner can capture the benefits in permitted forms. Harvard charges below its clearing price, and you can tell because good grades are necessary but not sufficient: political and donor connections, extracurriculars, and entrance essays do the rationing, and you cannot pay cash in lieu of the essay. Charging less than the market price lets the decider choose on criteria the decider likes, nefarious or not. Third, nonprice competition burns real resources: allocate parking first-come, first-served and everyone arrives early and circles the lot. Blocking a trade kills the gain between the two traders, not the value; it is one way of reallocating power.

“Cut out the middleman.” And do whose work?

Buy straight from the farmer and save money, says a friend. Then you must find a farmer who has what you want and wants your dollars; judge the barrel of corn yourself; and carry it home. Publix does all three: it finds the farmer, it judges quality, and it keeps milk cold so that it is there when you want it. It also brings the discipline of repeated dealings. A farmer who sells to you once has an incentive to misrepresent; one who sells to Publix for fifty years does not, because Publix stops buying, sues, or finds another farmer, and both sides come to rely on reputation. A student asked how Aldi sells for less. Fewer employees, goods left in their shipping boxes, a pear bin you sort yourself: Aldi charges you in time rather than dollars, so it serves people whose time is cheap (students, retirees), while Publix, Trader Joe’s, and Whole Foods serve people with more dollars and stronger preferences. In equilibrium the pears cost the same at both stores; only the composition of the cost differs, just as the cost of living in Auburn, Alabama includes the plane ticket to a Broadway show. The third TopHat check: buying directly from a farmer requires the shopper to arrange matching, assess quality and delivery, and manage travel, storage, and timing. The tasks do not disappear; they land on the shopper (or the seller).

Lower transaction costs make more trades worthwhile

Now let each bottle traded use up real resources. From the 20-and-20 start the gap between the worths is 1.0 minus 0.2, or 0.8 bars. At a cost of 0.8 a bottle the cost eats the whole gain and nobody trades. At 0.4 some trades happen, but they stop where the remaining gap just covers the cost: five bottles, not the equilibrium ten. At 0.2 they trade more, seven and a half bottles, and still stop short; as long as the cost is positive they never reach the crossing. Only with no transaction cost do the worths meet and ten bottles move. The camp assumed costless trade; the real world does not, and what middlemen do is arrange the trades that the cost of finding and judging would otherwise block. The fourth TopHat check: when the real cost per bottle falls from 0.4 to 0.2, trade volume rises from 5 to 7.5 bottles, not to 10, because ten needs a zero cost. Cheaper trading means more trading, never less.

Not reached in class: the fee ceiling, a rival dealer, and licensing

Class stopped after the transaction-cost figure. The last three slides of the topic were not shown; the instructor said the wrap-up comes Thursday. Their content is in the post, and it is summarized here so the page is complete.

How much can a lone middleman charge? Go back to the camp. Sam would give up to 8 bars for ten bottles and Joe would take as few as 4. A newcomer with no rival can charge at most the whole gap, 4 bars, paying Joe barely enough to release the water and charging Sam just short of what it is worth to him. At that limit both are indifferent; a smaller fee leaves each a strict gain. That is a ceiling on his fee, not his profit: his own costs still come out of it.

Rivals trim it. The newcomer actually kept 2 bars (7 from Sam, 5 to Joe). Let a second dealer enter and offer Sam the ten bottles for 6.5 bars and Joe 5.5 bars for them: each man gains another half bar, and the spread falls from 2 bars per ten bottles to 1. Rival offers, and the option to trade directly, limit the fee.

License only “approved” traders? The dealers propose that only licensed, trained, ethical middlemen may trade in camp, and that the existing dealers decide who qualifies. Two things can be true at once. Screening can help buyers when it credibly signals quality. And restricted entry can protect the incumbents’ margin by removing the rival offers that trimmed it. Which effect matters more is an empirical question, and the label “consumer protection” does not settle it; ask whether capable rivals can still enter, what comparable services cost, and whether quality is observably better. One more consequence: a credible, durable, transferable right to a protected stream of future earnings can be sold today for a price that capitalizes those earnings, so early license holders may gain a windfall while later buyers who pay full value expect only an ordinary return.

What continues Thursday

The instructor closed with “we will wrap up this lecture and start market prices on Thursday.” Expect a short wrap-up of gains from exchange (the material in the box above, and the topic’s recap) before the next topic, Markets and Coordination, begins. Thursday’s class gets its own recap and transcript page. Also from the housekeeping at the top of class: Exam 1 is Tuesday, September 29 (green Scantron, pencil, calculator allowed, open note; a study guide and a practice exam will be posted; the TA runs the review on Thursday, September 24), the course schedule is being updated, and TopHat grades are now in Canvas with the four lowest dropped.

Examples are memory cues; the right column is the principle each example should help you recover.

Example from class	Economic lesson
The car dealer says no, then $67,000, then $65,000 and floor mats	Each side has a reservation value the other cannot see; bargaining searches for a price between them and only divides the gain.
A car sitting on the lot while interest rates rise	Holding inventory has an opportunity cost, what the money could earn; a higher rate lowers a seller’s reservation value.
Sam offers half a bar for one more bottle at the crossing	Once the worths are equal at 0.6, no price below the seller’s worth is accepted; trading has stopped.
No $20 bills on the campus sidewalk; no systematically mispriced stocks	Where gains from trade exist, people exploit them until they are gone; that is what equilibrium means.
Joe’s diagram flipped and laid over Sam’s	Supply is demand read from the other side; trade stops where the two worth curves cross, at 30 and 10 and a price of 0.6.
Two triangles, two bars each; make the trade illegal	The triangles measure the gains from trade; a ban costs each man exactly his triangle.
Everyone at the pump; the 10th gallon for one driver, the 8th for another	Each buyer stops where marginal worth equals the price, so in equilibrium all buyers share the same marginal worth at different quantities.
The Omega and the Casio; Monster cable and Walmart cable	Different prices in equilibrium imply different quality on some dimension; perfect substitutes would sell at one price.
The marble in the bowl	An equilibrium persists until something disturbs it: new endowments or changed worths.
Taking a phone at gunpoint and handing it to a classmate	Pareto-optimal: nobody can gain without someone losing; a voluntary offer of a dollar is how an improvement would happen.
“I think this is fair, but my ethics might not be yours”	Pareto-optimality is an “is”; fairness is an “ought” that economics does not decide.
Uber, and the phone book you would have used instead	Transaction costs block improving trades; a middleman who lowers them releases gains that were too expensive to capture.
Hiring a hitman	Mutually beneficial to two parties is not the same as permitted, and with a harmed third party it is not Pareto-optimal.
Harvard’s entrance essay you cannot pay cash to skip	A price below the clearing level shifts allocation to the decider, who collects the value in permitted forms.
First-come, first-served parking	Nonprice competition dissipates part of the value in real resources (arriving early, circling).
The barrel of corn at the farm; the farmer who sells to Publix for fifty years	A middleman finds, judges, and carries; repeated dealings and reputation discipline honesty.
Aldi’s pear bin versus Publix; the cost of living in Auburn	The full cost is the same in equilibrium; stores differ in whether you pay in dollars or in time.
A cost of 0.8, 0.4, 0.2, then 0 bars per bottle	Trades happen while the next gain covers the next real cost: 0, 5, 7.5, then 10 bottles.

Question 1 — Partway through the trading, Sam’s worth of the next bottle is 0.8 bars and Joe’s is 0.4. Will another bottle change hands, and at what prices?

Yes, at any price between 0.4 and 0.8 bars. Sam gains whenever he pays less than 0.8 and Joe gains whenever he receives more than 0.4, so the gap is a gain waiting to be split. They keep going until the worths meet at 0.6.

Question 2 — At the crossing (Sam 30, Joe 10) a classmate says Sam should keep buying because he still likes water. What is wrong with that?

Liking water is not the test; the price is. Sam’s worth of a 31st bottle is below 0.6 and Joe would not part with a 10th bottle for less than 0.6, so no price satisfies both. The half-a-bar offer from class is refused for exactly this reason. Any further swap makes at least one of them worse off, and in fact reversing it would help both.

Question 3 — A camp rule bans all trading. Sam and Joe each keep their 20 bottles and 20 bars. How much worse off is each, and what happened to the total number of goods?

Each loses the two bars of gain his triangle measures, four bars in all, while the totals stay at 40 bottles and 40 bars. Nothing was destroyed; value was left uncreated. Whoever enforces the rule now decides who holds what, which is the point the parking-permit slide makes.

Question 4 — “The economist said the outcome is Pareto-optimal, so she is saying it is fair.” Correct the speaker.

Pareto-optimal is an “is” statement: no reallocation can help one person without hurting another. Fair is an “ought” judgment that depends on an ethical criterion economics does not supply. The label also says nothing about productive efficiency. Sam ending with 30 bottles to Joe’s 10 might strike you as fair or unfair; either view is consistent with the allocation being Pareto-optimal.

Question 5 — A concert venue sells every ticket for $40 and they are gone in a minute; the rest of the day people camp outside for returns and post pleas on social media. What does that tell you about the price, and where did the value go?

The price is below the clearing level, the same signal as Harvard’s entrance essay: something other than willingness to pay is doing the rationing. The value did not vanish. Part of it goes to whoever controls the tickets, collected in permitted forms (favors, connections, first pick for friends), and part is burned in nonprice competition, the hours spent camping and refreshing. Blocking the money price reallocates power; it does not erase the value.

Question 6 — A friend says an app that lets people rent out their parking driveways “creates nothing, it just takes a cut.” Answer with the phone-and-Uber discussion.

Before the app, the driveway owner and the driver could not find each other cheaply, so a trade that would have helped both did not happen: an improvement blocked by transaction costs. The app does what the newcomer did in camp and what Uber did for rides. It finds the two sides and brings them together, and its cut is payment for the gain it makes possible. The gain is real even though no driveway was built.

Question 7 — The real cost of arranging a trade is 0.2 bars a bottle and 7.5 bottles move. A new messenger cuts the cost to zero. How many more bottles move, and why does the answer stop at 10?

2.5 more bottles, from 7.5 to 10. With no cost the trades continue until the worths meet at the crossing, and beyond the crossing there is no gain left to cover any cost, even a zero one. Trades happen while the next gain covers the next real cost; at zero cost that is every trade up to the crossing and none past it.

Question 8 (from the box above) — Sam would give up to 8 bars for ten bottles; Joe would take as few as 4. A lone dealer arranges the trade. What is the most he can keep, and what happens when a second dealer offers Sam 6.5 and Joe 5.5?

At most 4 bars, the whole gap between 8 and 4; take more and one side walks away. That is a ceiling on his fee, not his profit. The rival’s offer gives Sam and Joe each another half bar and cuts the spread from 2 bars to 1: Sam ends with 13.5 bars instead of 13, Joe with 25.5 instead of 25. Rival offers and the option to trade directly are what keep a middleman’s cut below the ceiling.

Bottom line

Costless trading stops where the two worths meet, and the two triangles there measure the gains: two bars each for Sam and Joe. That allocation is Pareto-optimal, which means no mutual gain is left and nothing more; whether it is fair is a separate question economics does not answer. Forbid the trade and the gain dies while the value of deciding who gets what moves to the decider, collected in permitted forms or burned in queues. Let trading cost real resources and fewer trades happen; a middleman earns his fee by finding, judging, and carrying more cheaply than you can, and rivals keep that fee below the whole gain. Thursday wraps this up and turns to the prices that coordinate millions of such trades.