Two Bakers Walk Into A Cartel

We test a simple bakery price-fixing pact and watch it unravel, then use that failure to explain how transaction costs shape institutions, from unenforceable contracts to the rise of the corporate trust. We connect the common law doctrine of restraint of trade to the Sherman Act, modern merger fights, and a real-world egg pricing controversy where evidence is hard to untangle from supply shocks.
• a two-baker cartel that collapses as soon as cheating pays
• why courts historically refuse to enforce restraint of trade agreements
• how the trust device unifies control and removes the need to “trust” rivals
• how the Sherman Antitrust Act targets trusts and criminalizes collusion
• why merger discretion creates power and invites rent seeking
• why high prices and parallel pricing are not proof of conspiracy
• how price indexes can reduce transaction costs and also enable coordination
• two antitrust “twedges” that poke at efficiency arguments
• a listener letter on lateness norms and why uncertainty matters
If any listeners have some ideas about it, please do write in.

Links:

• Egg producers settle
• Bartholomae, T.C. (1923). “The Common Law on Restraint of Trade.” University Journal of Business, 1(4), 451–466.
•  Munger, M.C. (2021). “Giants among us: do we need a new antitrust paradigm?” Constitutional Political Economy. https://doi.org/10.1007/s10602-021-09350-w
• Sherman Antitrust Act (1890)
• William Baxter Joke 
• Robert Bork, Antitrust Paradox

If you have questions or comments, or want to suggest a future topic, email the show at taitc.email@gmail.com !

You can follow Mike Munger on Twitter at @mungowitz 

Transcript

Two Bakeries Try Price Fixing

Michael Munger 0:00

This is Mike Munger, the knower of important things. Here's the scenario. You and I run the only two bakeries in a small town. We make basically the same bread in the same way. So we compete mostly on price. We meet for coffee. Hey, it's a small town. People meet up. And we agree. Bread goes up to five dollars a loaf starting Monday. We draw up a simple contract, both sign it, shake hands, deal. How does that deal work out for us? Well, the answer depends on transaction cost, actually, as you might expect. A new twedge, letters, and book it a week. And more. Straight out of Creedmoor, this is Tidy C.

SPEAKER_02 0:48

I thought they talk about a system where there were no transaction costs. It's an imaginary system. There always are transaction costs.

SPEAKER_01 0:58

When it is costly to transact, institutions matter. And it is costly to transact.

Michael Munger 1:09

So we have a deal. Monday morning I drive past your shop, your sign says bread, five dollars. That's all good. Thursday afternoon, about 4 p.m. I drive past your shop, your sign says bread, $4.50. I storm into your shop. You say, well, a guy told me I could sell more bread, a lot more, if I cut my price. And you know what? It's true. I'm gonna sell out today. Business is great. Back at my shop, I see we've hardly sold any bread. Everyone's going to the shop with cheaper bread. You cheated. Well, what can I do about that? In almost any other line of business, the answer is obvious. I sue you. We had an agreement. You breached it. It caused me damage. A court will make you pay. That's what contract law is for. It's the machinery that makes promises credible even between people who don't trust each other, which honestly is most people most of the time. But if our agreement was to fix the price of bread, if it was in the old language a contract in restraint of trade, for the first century of American legal history, I could walk into that courthouse and the judge would throw me out, wouldn't find me, wouldn't lecture me, just refuse to hear it. Dismissed. Not justiciable. Go home. So I cut my price to $425. You're furious. Hey, I thought we had a deal. I say, ha, you're one to talk. We end up fussing at each other. We realize that a contract won't work because of transaction cost. We can't have trust because the contract is not enforceable. We can have triangulation and transfer. We can have the agreement that it's going to be $5. We can both sign it. We can find each other. And we can say that the agreement itself requires that there's $5 prices. But when I look, it turns out that you cheated. Well, if we go back centuries in English and then American common law, courts developed a doctrine that agreements in restraint of trade, that is, deals whose whole purpose was to suppress competition, were void. They're not exactly illegal, they're not criminal, they're just unenforceable. You could sign the paper, shake the hand, swear the oath on your grandmother's Bible. Didn't matter. One party reneged, the other had no legal recourse. There's a legal historian, TC Bartholomew, who wrote a wonderful plain spoken summary of this doctrine back in 1923 in a piece called The Common Law on Restraint of Trade. Here's how he put it. I want to read it carefully because it's pretty cool. It's the whole ballgame. Competitors sometimes enter into agreements between themselves by which they covenant not to compete with each other in order to secure larger profits. The agreement may be a mere informal understanding or it may be written. The restraint may take the form of limiting output, controlling supplies, fixing prices, dividing territory, etc. The courts have generally held such agreements invalid on the ground that by concentrating the control of a market in the hands of certain parties, to the exclusion of others, prices may be arbitrarily advanced, quality deteriorated, etc., to the detriment of the public. End of quote. So notice what the courts are not doing. They're not sending sheriffs out to break up a meeting where you and I plotted our bread cartel. They're not finding us. They're just declining to help either of us overcome the transaction costs that are involved in a deal when there's incentives to cheat. It's not perfect, but it's not bad. And before the Industrial Revolution, where large firms and large areas or markets became standard, this was a pretty satisfactory solution generally. So if we go back to our two bakers with a small market, the not enforceable solution's a good one. We're going to have to work pretty hard to find a workable solution. If we both cut price, we don't make much money. But we can't have an agreement. Having high transaction cost is actually the point. The law is creating a setting where I have high transactions costs that are difficult to get rid of.

Restraint Of Trade Means No Lawsuit

Michael Munger 5:44

So we go see a lawyer. The lawyer suggests we merge into a kind of corporate partnership called a trust. The trust device solves the cheating problem not by making the agreement enforceable, but by eliminating the need for any agreement at all. The ineffective way to was to have company A and Company B contract with each other about price. And then each of them has to decide whether to abide by that agreement and enforce the agreement on the other side, even though each could make more money by shading their price lower. The modern way, especially in industries with substantial economies of scale, railroads, iron and steel or oil, is for the shareholders of both companies to transfer their shares to a small board of trustees, receiving, in quotes, trust certificates. In return, those certificates entitled the original owners to their proportional share of the combined earnings, but stripped them of voting control. And by the modern solution of trusts, I mean something around 1870, 1880. Now, it might seem like you want voting control, but it's precisely the control of price that leads to cheating and loss of cartel profits. The trust device puts pricing decisions beyond your control, but it does the same for your competitor, the other Baker, in our example. This is the Hobbes Coase solution. You give up the liberties you want others to have to give up also. You give up your ability to cut price, your competitor does too, and everybody's happy. Of course, the original price fixing contract did that too, but it's not enforceable. Trusts are enforceable. Trustees now held the voting stock, maybe just five to nine people in one room, and I have to think they were probably smoking cigars. And with it, the power to elect the directors and officers of every formally independent company in the combination. And there's no conflict, there's no benefit to cutting price, at least not below the profit maximizing price for the combined entity. Now, the Sherman Antitrust Act of 1890 was written specifically to close this loophole. It bans, quoting, every contract, combination in the form of trust or otherwise, or conspiracy and restraint of trade. So that phrase, in the form of trust or otherwise, was Congress explicitly saying the ownership consolidation workaround won't get you out of this. This isn't a mechanism that would let two businesses raise prices together today. It's the specific historical trick that was a response to the arms race. We have that the common law will not allow us to enforce price agreements. Well, we'll use trusts to get around that. And then the law caught up. Antitrust law was built to shut down that loophole, which is why trust became the generic word for the whole body of law aimed at it, even though it was only true for a relatively short historical period, 15 or 20 years, that trusts were the problem. I should

Trusts Solve Cheating By Ownership

Michael Munger 9:00

note that this problem of coordination and uh agreements about price is quite an old problem. Adam Smith wrote insightfully about it, quoting from Book 1, Chapter 10 of Wealth of Nations, people of the same trade seldom meet together even for merriment and diversion, but the conversation ends in a conspiracy against the public or in some contrivance to raise prices. It is impossible, indeed, to prevent such meetings by any law which either could be executed or would be consistent with liberty and justice. But though the law cannot hinder people of the same trade from sometimes assembling together, it ought to do nothing to facilitate such assemblies, much less to render them necessary. A regulation which obliges all of those of the same trade in a particular town to enter their names and places of abode in a public register facilitates such assemblies. A regulation which enables those of the same trade to tax themselves in order to provide for their poor, their sick, their widows, and orphans by giving them a common interest to manage, renders such assemblies necessary. An incorporation not only renders them necessary, but makes the active majority binding upon the whole. End of quote. And again, that's Adam Smith, Wealth of Nations, Book One, Chapter 10. That's totally a transaction cost story. It may be hard to keep people from meeting and conspiring, but you shouldn't require it. You shouldn't reduce the transaction cost of coming up with contracts and restraint of trade. So Smith wasn't worried about the meeting. He knew that states couldn't very well ban conversation. He's saying the state should not encourage such assemblies. The state should make it more expensive, not less expensive, to negotiate deals on price and output. Of course, the not enforceable aspect is a big part of the transaction cost argument. And Smith just took that for granted, frankly. Everybody in the cartel wants everybody else to hold the line while reserving the right to chisel a little on price themselves. That's not hypothetical. It's the single most robust finding in the empirical literature on cartels, going back to the railroad pools of the 1870s and all the way up to OPEC quota busting today. Cartels collapse from the inside constantly and consistently, because the incentive to cheat is baked into the arithmetic. That was a bakery joke. If you're a 19th century industrialist, you figured out the courts won't enforce your cartel agreement, you don't need to throw up your hands. You look for a different institutional form, one that doesn't need a course court to enforce the agreement because there's no longer a promise between the firms within a trust. The railroad pricing pools of the 1870s kept collapsing in exactly the way we would predict since the price agreements could not be enforced. Companies stopped pooling and started trusting. That bad pun is transaction cost truth. I can't trust my competitors, but if we form a trust, that single entity no longer has any incentive to cut price. It's just the standard Coast Theorem logic. Resources will be bought and sold to produce the highest profit form of ownership structure. So remember, for me, trust is one of the three fundamental types of transaction cost, triangulation, transfer, and trust. But the noun trust, as opposed to the verb, means an actual person, a trustee, holding property for the benefit of someone else. It is first seen in the early 1400s, meaning a confidence placed in a person who holds property entrusted to him by its legal owner. By the mid-1400s, this had solidified into a specific legal instrument sense, the condition of being legally entrusted. And a trust is that which is committed to another for safekeeping. That underlies Will's estates, fiduciary law today. The legal trust is a device for combining competing firms under unified control, and it's quite recent. Coming near the end of the classic industrial revolution. To be fair, there was no need for the legal fiction in many nations, because it was quite common for those nations' governments to sell monopolies outright. And then the state was selling what it could do best, provide violence. And the firm or cartel would just act with government sanction. It's unsurprising that trust originated in the United States, but not for the reason that many cap anti-capitalist writers allege. Most countries never pretended to want competition, but the U.S. did, so there was a conflict. The use of the word trust in that specific sense appears to date to 1877, according to Etim Online, roughly five years before Standard Oil's famous 1882 trust agreement. Standard Oil didn't coin the word so much as to give it its defining high-profile instance. The company's attorney, Samuel C. T. Dodd, adopted the existing legal trust instrument, shareholders conveying their stock to a small board of trustees in exchange for a trust certificate as a way to unify control over Standard Oil's many state chartered subsidiaries. That 1882 arrangement became the template everyone had in mind, which is why trust then generalized to mean any monopolistic competition, giving us antitrust, and of course, by 1903, Theodore Roosevelt, the trust buster.

Sherman Act Closes The Trust Loophole

Michael Munger 14:52

So in 1890, when Congress passed the Sherman Act, it does two things at once. We should spell out the competition title, uh the title of the act, because it tells you a lot about it. An act to protect trade and commerce against unlawful restraints and monopolies. Well, it turns out that now, in looking back at the Sherman Act, the left has discovered originalism because antitrust law in the United States since 1978-1979 has focused on consumer welfare. But if you look at the original debates in the passage of the Sherman Act, it's perfectly true that there were many other considerations. Well, what does the Sherman Act do? First, it makes restraint of trade agreements illegal per se, not just unenforceable, but affirmatively punishable with fines and criminal penalties. The text of the Act says, in part, this every contract, combination in the form of trust or otherwise, or conspiracy in restraint of trade or commerce among the several states or with foreign nations is hereby declared to be illegal. End of quote. And so technically that has to be across state borders because this is a federal law. The idea of interstate commerce has just been eviscerated where it no longer, that that part of it has to be among the several states is no longer important. Antitrust can operate anywhere. Section two. I edited it for length. Notice that there's two parts. The first is about structure, and the second is about performance or activity. So the this kind of anticipates the structure, conduct, performance, industrial organization paradigm. The Sherman Act goes after monopoly structure itself, meaning that sheer consolidated size, the trust device that had routed around the old contract unenforceability rule, could now be a violation of all on its own. Size alone can be an offense, regardless of whether there was ever a contract to refuse to enforce. So price fixing could happen even if a single firm had grown just by excellence, not by trust or price fixing. But then, if size alone is an offense, even if it was achieved by excellence, it raises the question of industry. If you have a really high market share, of what market? Criminal liability under the Sherman Act is active. It doesn't wait for one cartel member to sue another. The state itself has the discretion to act by investigating and prosecuting. Sherman Act applies whether or not anybody inside the arrangement or someone outside the arrangement ever complains. So that philosophy's gone quite far. And now much of the neo-Brandeissian counter-revolution in antitrust is just obsessed with maximizing the number of choices. They have a not very good economics 101 microeconomic textbook definition of competition. And they would like airlines to be like wheat, thousands and thousands of small producers, none of whom can decide what the price is. This is not hypothetical. Spirit Airlines actually faced the problem of trying to merge to avoid bankruptcy. They wanted to merge, be acquired by Jet Blue. This was turned down by the Justice Department. Now the reasons are actually pretty complicated, and Spirit's bankruptcy is a little more uh complex than just saying that the antitrust decision to prevent the merger was the cause of the bankruptcy. But it certainly didn't help. The common law's refusal to enforce restraint of trade contracts didn't require a regulator to guess the right market structure, didn't require anybody to prove intent, and didn't require an army of prosecutors who make their living by bringing cases. After all, saying, well, nope, we're going to let this go through means that you don't get any litigation experience if you're an attorney for the Justice Department. So there's a really if the government has discretion and attorneys make uh get their experience for increasing their future value as litigators by litigating, there's an obvious problem in terms of the asymmetry of incentives. But the Sherman X criminal layer was necessary. That trust loophole was real. Non-enforcement alone then couldn't reach a merger. But it's worth noticing that we need both pieces: the non-enforceability of contracts under the common law, if they're in contracts and restraint of trade, and some kind of regulation of market structure. The problem is the discretion of market structure is a great opportunity for someone who wants to exercise government control.

Merger Review And Government Discretion

Michael Munger 20:36

So dirigists of all parties, the left and the right, have recognized that the ability to withhold permission for a merger can be used to hold up the creation of value. They can, in effect, engage in rent seeking themselves, because they have an excessive and frankly unnatural power over commerce. There was a recent example of this, and uh it's a complicated enough case that I'm not sure what the right answer is, but let me tell you about it.

Egg Price Index Allegations And Evidence

Michael Munger 21:07

The U.S. Justice Department and 17 states reached a settlement agreement with three major egg producers and that resolved allegations that the companies had illegally colluded for years to raise price, including when the cost soared to record highs in 2025. State and federal government accused CalMaine Foods, Versova, and Hickman's Egg Ranch of a behind-the-scenes arrangement to artificially inflate the daily price quotations for eggs between June of 2022 and March of 2025. In particular, the government claimed that the companies coordinated on what bids they would submit to Ernerberry Publications, a company that runs an index. Which is a key to determining how much grocery stores, restaurants, and others pay for billions of eggs each year. Now, the AP story said that the investigation found. That investigation didn't find anything. This is an allegation. Famously, prosecutors could probably get a grand jury to indict a ham sandwich. And this wasn't even an indictment. This was just an announcement by the government that these are the things that they were accusing the egg industry of. It's a charge brought by a prosecutor as a way of cultivating public opinion. No trial, no evidence presenting, and certainly no finding of fact in a courtroom. So we kind of have to be careful. Now, price fixing is hard to prove in a market like eggs because there's a real, there was a real, massive, verifiable supply shock that was happening at the same time as the alleged collusion. Avian flu killed millions of egg-laying hens starting in 2022, and any economist would predict that that supply shock alone should send prices sharply higher. Supply collapsed while demand for that grocery staple stayed inelastic. So an honest, fully competitive market and a market with some illegal coordination layered on top would both produce observationally the same symptom. Exit 623 a dozen in March of 2025. High prices by themselves are not evidence of a conspiracy. They are equally consistent with the legitimate scarcity story. And all the companies charging the same price is not evidence of a conspiracy because it's a highly competitive market, and that's actually what the neo-Brandeisan tyros want from their model of perfect competition. No one is a price giver, everyone's a price taker. So high prices where everybody's charging the high price could be a very competitive market. So the specific theory here isn't that three companies met in a room and set a price. The complaint alleges that Calmain, Versova, and Hickman's coordinated on what bids they submitted to Earnerberry Publications, a private reporting service whose daily index is used as the reference price on which wholesale egg supply contracts across the industry are based. Now, it's really a transaction cost reduction device in the first place. It's a way of choosing price quickly and cheaply rather than holding auctions. Can you imagine holding auctions for eggs? I mean, you could have them for large lots, but if you have some mechanism for deciding what the overall price is, a kind of index, then everyone can use that because their contracts will just point to that variable overall signal. But it could be the same structural idea as the trust device we discussed earlier. Rather than a contract to fix price, which is unenforceable and easy to defect from, you engineer a mechanism where everyone's price moves together through a shared input that goes into an index. No one needs to trust anyone to hold up their end. You just charge the price that is implied. Now, the strongest piece of evidence that was reported is behavioral rather than actual documentary. There's no smoking gun, there's no evidence that there was actual communication. However, price quotations dropped significantly once Calmain, Versova, and Hickmans learned of the Department of Justice investigation. They learned of this, of course, because they are instructed to preserve documents in March of 2025. That timing is a little bit sketchy. If the elevated prices were purely a function of ongoing bird flu outbreak and reduced flock sizes, there's no obvious reason that receiving a subpoena and a legal hold notice would cause the price index to fall, and it fell pretty dramatically. Still, the breathless news stories have phrases that are doing a lot of heavy lifting. Behind the scenes arrangements, secretly communicated, illegally coordinated. Those are all things that the Attorney General press statement said. There is no finding of fact. There is not one bit of evidence that that happened. Companies settle civil antitrust actions all the time, regardless of the strength of the underlying case, because litigation is expensive and reputational exposure is costly. So it's worth noting that the settlement here, $3.3 million total, plus they had to donate some eggs worth not quite $10 million at retail, is really small. Calmain alone had reported profits of $1.22 billion, that's with a B, for fiscal 2025. And so the a number of people just pointed out that these companies might treat that sort of little flea bite of a fine as one of the costs of doing business. It's certainly not any kind of proof of guilt. We also have no evidence of their innocence, which tells you something about how complicated price fixing litigation can be.

Antitrust Jokes About Efficiency

Michael Munger 27:24

Whoa, that sound means it's time for the twedge. I have two, um, and I should remind you, twedges are never funny, so it's not surprising that these aren't funny. First, a story about William Baxter, former professor at Stanford Law School. He served as 1981 to 1983 as the assistant attorney general in charge of antitrust at the Department of Justice. And here's the story. While strolling one evening on a quiet street near his Washington residence, Bill and his wife Carol were confronted by an armed robber. They emerged minus some personal property, but without injury. At the next antitrust division holiday party, the incident became a skit in which the robber demanded Bill's watch. Bill resisted, trying to say no. The robber points out it would only be a wealth transfer, a mere redistribution that would not affect total economic efficiency or social welfare. Reassured that the robber shares his standards of economic analysis and acknowledging that the point is analytically correct, Bill immediately surrenders his watch. End of quote. Well, the the the point is that one of the alleged costs of monopoly is called the Harburger triangle, which is the dead weight losses that arise from mutually beneficial transactions that fail to take place. That means that the rectangle in that diagram, which is what is being referred to here, is simply a transfer from consumers to producers. And that's the reason why I think it is more defensible to use consumer welfare. Um, because the concern for consumer welfare means that we put a thumb on the scale against using the rectangle as being purely neutral. So if you understood none of that, the point of the joke is that most people understand none of that. And the fact that the assistant attorney general in charge of antitrust would credit a robber who understood it is the thing that's supposed to be funny. The second wedge: a Democratic senator, federal judge, and an antitrust specialist at Department of Justice walk into the Spirit Airlines check-in area. Okay, no, they don't, because Spirit Airlines doesn't exist, and none of those three rich people would be writing in it if it did. The point is that most of the people who didn't want Spirit Airlines to merge never used Spirit Airlines in the first place, and it was no harm to them when it disappeared. They could keep their little textbook definition of competition without any cost to themselves.

Why Some Places Tolerate Lateness

Michael Munger 30:13

Some letters. First, the episode on lateness. The issue is not knowing the expected delay. I was working as a photographer on a concert Thursday night. The event page said 8 p.m. Then in Tiny Print it said doors 8 p.m. concert 9 p.m. So I emailed the producers and asked when the concert was actually starting. They replied 10 p.m. The concert actually started at 10.50 p.m. Welp. I wonder why these societies, that is, the ones where punctuality is scorned, work this way. That to me is the most interesting question. Do they value serendipity more than other societies? I have to say there's something nice about unpredictability in that it allows me to do what I want instead of what I planned. But then again, I could have just not made any plans in the first place and not left all those other people waiting. All the best, PG. Well, thanks, PG. And it it is an interesting question because I whine and make moral claims about people who are late, but in fact, there's at least another side to this, and there are plenty of societies where it probably would be possible to have a different norm, and yet they don't. And uh rock concerts in particular. Uh it's not so much true for Broadway shows, it's not true for a lot of other kinds of shows, but rock concerts in particular, there seems to be something about the experience in waiting for two hours. People are willing to pay for that, and I don't understand why. If any listeners have some ideas about it, please do write in.

Book Recommendation And Goodbye

Michael Munger 31:54

Book of the week this week is Robert Bork's Antitrust Paradox. There's a new edition in 2021 by Bork Publishing. It's still worth reading, perhaps now more than ever. I really just reread it myself and found it terrific. Well, that's it for this week. We'll talk again on next Tuesday on Tidy C.