How to use this page

This recap covers the first 31 minutes of the September 17 class, which finished the gains-from-exchange topic (the September 10 and 15 sessions have their own Part 1 and Part 2 pages in this module; the rest of the September 17 period opened Markets and Coordination and is in the Lecture 5 module). Class delivered the fee ceiling and the rival dealer, the licensing sequel with a long tour of real-world entry restrictions, a TopHat item, and the used-lamp exercise on transaction costs. Read the three-step diagnostic first. Then use the lecture section to reconnect each idea to an example from class. The box near the end pulls the whole three-session topic together, because the deck’s closing study cards went by quickly. Finish with the application checks. The separate transcript has the full explanation, including the certificate-of-need, liquor-license, and taxi-medallion stories.

What you should be able to do

1. Compute the most a lone middleman can charge from the buyer’s worth and the seller’s minimum, and show how a rival’s offer changes each party’s holdings and the middleman’s spread.
2. Given a proposed licensing or entry rule, name who is excluded, who gains the protected margin, why the incumbent will spend resources to win it, and what evidence would distinguish protection from genuine screening.
3. Decide whether a trade with a stated arranging cost can leave both parties better off, and say what a lower cost of trading does to the number of trades.

Movement 03 · Blocked trades and middlemen (concluded)

Where we left off: costless trade stops where the worths meet, and costs of trading cut it short

The session opened with a retrieval. Sam and Joe trade until their marginal worths are the same; at that equilibrium no gains from trade remain, Joe having sold water to Sam and Sam, in effect, granola bars to Joe. Add a cost of trading and there is less trade; lower the cost and there is more. Uber is the real-world case: before it, matching a person who wanted a ride with a person willing to give one was very costly, and Uber found a cheap way to bring the two sides together. The newcomer in the camp did the same thing. He noticed that Sam and Joe, who receive identical care packages but have different preferences, could both gain, arranged the trade, and kept part of the gain as the return for being alert to the opportunity.

A lone middleman can take the whole gain. Rivals trim it.

If the newcomer is the only trader in camp, the most he can charge for arranging the ten-bottle deal is 4 bars, the gap between the most Sam would pay (8) and the least Joe would accept (4); beyond that one of them refuses because the deal would make him worse off. In the original deal Sam gave up 7 bars for the ten bottles and Joe received 5, so the newcomer kept 2. Now a rival appears (the instructor drew on the class’s earlier green-Scantron exchange to cast a student as the rival). The rival offers Sam the same ten bottles for 6.5 bars instead of 7, and offers Joe 5.5 bars for his ten bottles instead of 5. Both take the better deal. Sam ends with 30 bottles and 13.5 bars instead of 13; Joe with 10 bottles and 25.5 bars instead of 25; the spread falls from 2 bars per ten bottles to 1. If entrepreneurs can enter freely, they keep entering while it is still profitable, and competition drives the camp to the equilibrium Sam and Joe would have reached themselves.

License only “approved” traders. Who gains?

The original newcomer is not happy about the upstart: to keep his customers he must now keep fewer than 2 bars. So what does he lobby the guards for? Not a better offer but restrictions on trade. He is an upstanding member of the community; the rival is a fly-by-night who waters down the water and puts poison in the granola bars; Sam and Joe are not smart enough to know they are being hustled; the camp needs licensed, trained, ethical middlemen. And who will decide what licensed, trained, and ethical mean? The guards are not experts, so they ask the newcomer. Once the rule is in place the newcomer gains and Sam and Joe lose, because the rival was willing to give them a better deal.

The instructor then toured real cases of the same move. Certificate-of-need laws: in some communities a new hospital must first show the existing hospitals that another hospital is needed, and the existing hospitals are rarely persuaded, even though more hospitals would lower the price of care. Liquor licenses in the town where he tended bar for six years: a new license needed the unanimous approval of the existing bars, who were not keen on more competition, though consumers would have gained. Doctors, dentists, barbers, and lawyers: the American Medical Association and the American Bar Association limit how many students can be admitted to and graduate from medical and law schools, which lowers the supply of doctors and lawyers and raises their wages; the bar’s campaign began when paralegals were drawing up deeds, trusts, and wills at prices lawyers could not match. His own view, offered as a provocation students need not share: most of these arguments boil down to “we do not want competition.” A mechanic who under-tightens the lug nuts on his car could kill him, and nobody licenses that job; he trusts private certification such as AAA garages and public doctor ratings more than a license exam passed thirty years ago; a restaurant that poisons customers or a surgeon who loses patients loses business; and economists who have looked find no meaningful quality improvement from licensing dentists or physicians, which suggests the point is restricting supply, not safety. A student raised medical tourism; the reply was that lower professional standards abroad can exist without legal licensing, and that a reputation for bad outcomes drives down what people will pay there.

Two consequences of a successful entry restriction close the frame. First, the incumbent excludes capable rivals and can charge a higher price, so customers lose the alternatives; hence the instructor’s refusal, in the exam-day Scantron market, to sell any one student an exclusive license (“I want free entry in the classroom”). Second, because the protection raises the incumbent’s earnings in perpetuity, he is willing to spend a great deal to obtain it. Quality might rise, but consumers may then be paying for more quality than they value; they might have preferred a cheaper, lower-quality service.

Why reform is so hard: the transitional gains trap

While the poll ran, a further question: what happens to the value of a medical license if licensing is repealed? Or of a New York taxi medallion when Uber innovates around the requirement? It falls, and the holders fight hard; New York’s taxis eventually got Uber brought into the fold and regulated like themselves, and much of Uber’s advantage there went away. In the taxi drivers’ defense: the medallion rule is about a century old, and the people who won it and enjoyed the restricted entry are long gone. Later drivers paid a market price for the medallion that already reflected the restriction, so they earn only a competitive return; deregulation would inflict a loss on people who never got the windfall. The same holds for cosmetology licenses. Economists call this the transitional gains trap. Even when everyone agrees the rule was a bad idea, the people who would be harmed by repeal have a case, so the practical advice is not to create the privilege in the first place.

TopHat check: which finding would show the rule protects incumbents?

The item asked which finding would most strongly indicate that the proposed licensing rule primarily protects incumbent dealers rather than improving buyer outcomes. That a license commands a price because it promises future net earnings is true of any durable right and does not settle the question; that screening produces more consistent quality would show buyers gaining; that licensed traders meet training and ethical standards before entering is the rule’s own claim. The answer is that capable rival traders are excluded, leaving customers without alternative offers, which is exactly what the rival had been providing.

The next trade must cover its real cost

A buyer values a used lamp at $18; its owner would accept $10; arranging the trade uses resources worth $9. Can the trade leave both better off? No. The most the buyer will pay is $18 and the least the seller will take is $10, so the total gain available is $8, and nobody spends $9 to arrange an $8 trade. Let a technology cut the arranging cost to $3 and the trade happens, with $5 left to share. That is the Uber story again, and eBay’s: before eBay, the person with a lamp to sell and the person who wanted one could not find each other cheaply; eBay brings the two sides together, captures a small part of the value, and both parties gain. It also shows the limit on eBay’s fee. Charge more than the whole gain and no trade happens and eBay earns nothing; charge a high but feasible share and competitors appear offering to take less, and people switch. This is why a principles textbook’s “infinite number of buyers and sellers” is not required for competition: a market with a single seller can behave as if it were competitive, because a price even a little above the competitive level invites entry. The rival who has not entered yet is a check on the firm. Uber made money; Lyft noticed; fares came down as the two bid against each other, just as the rival newcomer bid down the spread in camp. The deck’s closing study cards were passed over (“we basically covered that”); their content is in the box below.

The close

For the first time this term the feedback slide was reached at the end of a lecture rather than at the top of the next one. The instructor explained that the lecture decks are revised with the responses from the Google Form, so feedback changes what the class sees; the form link is also in the module. He then set the gains-from-exchange TopHat page to review, opened the new lecture’s page, and began Markets and Coordination (Lecture 5 module).

The whole topic in one place

Gains from exchange ran across three sessions: September 10 (the cold open, the camp tables, the jacket, the two-panel figure), September 15 (the crossing, Pareto-optimality, the forbidden swap, the middleman, the transaction-cost wedge), and this one. The deck’s closing cards, which class did not dwell on, say this.

Worth and gains. Exchange can improve both positions. Differing personal worths create room for gain; payments divide it; the quantity of goods need not grow. Trade stops where the two worths meet (Sam 30 bottles, Joe 10, price 0.6 bars, two bars of gain each), and that allocation is Pareto-optimal, which says nothing about fairness.

The stopping rule. The next gain must cover the next real cost. With zero transaction cost the worths meet; with a real cost per trade, fewer trades happen (0.8 bars a bottle blocks all of them, 0.4 allows five, 0.2 seven and a half, zero allows ten). Middlemen can lower those costs, and entry and reputation discipline their service.

Institutions. Bans can redirect or dissipate scarcity rents (the parking permit, first-come first-served). Licensing can screen quality and restrict entry; evidence must distinguish the two effects.

Exam 1 covers this topic from the posts and what was delivered in class across the three parts; the instructor said a study guide and a practice exam will be posted.

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

Example from class	Economic lesson
Sam would give 8 bars, Joe would take 4, and the newcomer kept 2	The middleman’s fee is capped by the whole gain (4); anything less leaves both sides a strict gain.
The rival offers Sam 6.5 and Joe 5.5	Competition among middlemen returns part of the gain to the traders; each is half a bar better off and the spread falls from 2 to 1.
The green-Scantron arbitrage on exam day, and the refusal to sell one student an exclusive license	Open entry keeps a middleman’s cut small; an exclusive right would recreate the protected margin.
“He waters down the water; the camp needs licensed, trained, ethical middlemen”	An incumbent lobbies for entry restrictions dressed as consumer protection, and asks to write the standard.
Certificate-of-need laws; the liquor license that needed every other bar’s approval	Incumbents given a veto over entry use it; the excluded rivals and the consumers bear the cost.
The AMA and the ABA limiting graduates; paralegals and wills	Restricting the supply of a profession raises its members’ wages; the rule appeared when cheaper rivals did.
Nobody licenses the person who tightens your lug nuts	The harm-prevention rationale is applied selectively, which is evidence that the motive is often restricting supply.
AAA-certified garages; doctor ratings; the restaurant that poisons customers	Reputation and private certification can do the screening a license claims to do, and they update.
Economists find no quality improvement from licensing dentists or physicians	Evidence, not the label, tells screening from protection; here it points to protection.
What happens to a taxi medallion’s value when Uber arrives	A protected right capitalizes future earnings into a price today; competition destroys that value, so holders fight.
The driver who paid full price for a medallion the original lobbyists no longer hold	The transitional gains trap: later buyers earn only a normal return, which makes repeal painful and reform hard.
The $18 lamp, the $10 owner, and the $9 arranging cost	A trade happens only if its gain ($8) covers its real cost; at $3 the trade occurs with $5 to share.
eBay’s pricing problem, and the competitor who offers to take less	A middleman’s fee is limited by the gain and by rivals, including the one who has not entered yet.
Uber, then Lyft	A single seller can behave competitively because entry is open; a second entrant bids fares down, like the rival in camp.

Question 1 — One camper would give up to 9 cans of tuna for a blanket; another would part with his blanket for as few as 5. A lone middleman arranges the trade. What is the most tuna he can keep, and is that his profit?

4 cans, the gap between 9 and 5. Charge more and one side refuses. It is a ceiling on his fee, not his profit; whatever the arranging actually costs him comes out of the 4.

Question 2 — The rival in camp offered Sam 6.5 bars and Joe 5.5. Suppose a third dealer enters. What happens to the spread, and where does the process stop?

The spread falls again as the third dealer undercuts, and it keeps falling while arranging trades is still profitable. With costless entry it stops when no gain is left to capture, at the price where Sam’s and Joe’s worths meet: the equilibrium they would reach trading directly.

Question 3 — A state proposes that only licensed interior designers may sell design services, with the licensing board drawn from practicing designers. Using class, what would you check before believing the consumer-safety rationale?

Who is excluded, who sets the standard, and what the evidence shows. If capable rivals are shut out and prices rise with no observable quality gain, it is protection; the incumbents’ seat on the board is the tell from the camp story. Real screening is possible, but it has to be demonstrated, and reputation or private certification may already be doing it.

Question 4 — A friend says: “Repealing the taxi medallion rule would be fair; the drivers got a monopoly for a century.” What did class say complicates this?

Today’s drivers mostly bought their medallions at a price that already reflected the restriction, so they earn a competitive return, not a windfall; the people who gained from the rule are gone. Repeal imposes a loss on people who never got the gain. That is the transitional gains trap, and it is why the advice is to avoid creating such privileges rather than to count on undoing them.

Question 5 — A buyer values a bike at $250 and its owner would sell for $190. Listing, meeting, and vetting the buyer would use $70 of resources. Does the trade happen? What if an app cuts that to $25?

The gross gain is $60. At a $70 arranging cost nobody does it. At $25 it happens and $35 remains to share among the buyer, the seller, and the app. The app’s fee is a transfer; the resources it uses are the real cost.

Question 6 — “eBay is the only real marketplace for used lamps, so it can charge whatever it likes.” Give two limits from class.

First, the gain itself: a fee larger than the gap between buyer worth and seller minimum kills the trade and eBay’s revenue with it. Second, entry: a fee that leaves a fat margin invites a competitor to offer sellers a better split, and people switch. A single seller facing open entry acts as if it were in competition.

Question 7 — Why did the instructor refuse to sell one student the exclusive right to sell green Scantrons on exam day, even though that student would have paid for it?

Because the license would convert an open market into a closed one: the student would exclude capable rivals and charge classmates more, and much of the value of the privilege would end up with the seller of the license rather than with anyone providing Scantrons. Free entry keeps the arbitrage cheap for the buyers.

Bottom line

A middleman can keep at most the whole gain from the trade he arranges, and a single rival trims his cut toward zero. When he responds by asking for a rule that only approved traders may trade, ask who is excluded and who writes the standard; screening can be real, but the evidence usually shows a protected margin, and once the privilege is capitalized into a license price it is very hard to repeal. Underneath all of it is one test: a trade happens only when its gain covers its real cost, so anything that lowers the cost of trading, from Uber to eBay to a camp newcomer, creates trades that were not happening before. Next: how prices coordinate millions of such trades among strangers.