The Answer Is Transaction Costs

Parking Lots, Transaction Costs of the Price Mechanism, and the Pork Pie Fedora Rogue

Michael Munger

Send us a text

Two seemingly similar parking lots at Wrightsville Beach, North Carolina, couldn't be more different in the emergent behaviors they foster. From the orderly lines of the 85-space lot to the chaotic dynamics of the smaller 19-space circular lot, discover how price rationing, queue formation, and transaction costs play critical roles in these everyday systems.

Things take s a quirky turn with the arrival of a man in a pork pie fedora who disrupts these parking norms, buying spaces directly from beachgoers. This unconventional behavior prompts a deeper discussion on the breakdown of social rules and the challenges of maintaining order when outsiders intervene. 

Book o'da'week:  Jonathan Haidt, The Anxious Generation: How the Great Rewiring of Childhood is Causing an Epidemic of Mental Illness. 2024, Penguin Press. 

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


Speaker 1:

This is Mike Munger, the knower of important things from Duke University Beach parking and the problem of using price rationing. A new twedge, plus this week's letters and more Straight out of Creedmoor. This is Tidy C. I thought they'd talk about a system where there were no transaction costs. It's an imaginary system. There always are transaction costs. When it is costly to transact, institutions matter, and it is costly to transact.

Speaker 1:

The city of Wrightsville Beach on the North Carolina coast operates municipal parking lots. They charge for the privilege of parking. Now what happens? There's two large lots near the beach condo where we spend a month every summer. One of them is large, about 85 parking spaces, the other is small. It only has 19 slots. The current charge is $6 an hour or $30 for a day. The rate applies from 9 am to 8 pm. Now on some days, cloudy weekdays for example, the lots aren't full. But on high demand days, on sunny weekends and almost all the time in July and August, which are the beach going times, there are more people willing to pay $6 per hour to park than there are spaces for them to park. So we need a way of rationing access to the spaces that price doesn't fully take care of.

Speaker 1:

Now, if you pull into a space that's unoccupied and there are other unoccupied spaces, you've replicated the old Lockean process of acquiring ownership. That is, you can take this thing by combining your car with it and there's as much and as good for others who are also looking for parking because there's other empty spaces. Now, if you fail to pay, you're going to be ticketed. Ultimately, you'll be towed. So let's ignore the price. That's fixed externally. That's true for everyone, regardless of whether there are or there are not other empty parking spaces, because that's really the point.

Speaker 1:

The question is, what if parking spaces are scarce? What if there are more people who want to pay $6 an hour to park than there are parking spaces that allow them to pay $6 per hour to park? Because you don't just need $6, you need a parking space. So once you pull in, you have an exclusive right to use that space. You can park there and I can't, but you have no right to transfer the space to someone else. It's not yours, you have just acquired the use rights by pulling your car into it first. So it's not clear what happens when a car pulls out of a space to leave the parking lot because there's more than one other person when the lot is full who wants that now vacated parking space? But that's where the allocation problem emerges, the one that I'm interested in, at least, since many people are willing to pay the money price. Whose space is this now vacated space, now that the previous owner has left? Is this now vacated space now that the previous owner has left? Well, perhaps surprisingly, there's a clear pattern of behavior in these two parking lots the large one with 85 spaces, the small one with 19,. But the patterns of behavior are different. I've watched it for a long time and noticed that the patterns of behavior are different.

Speaker 1:

When the large lot fills up, a line forms. There's only one entrance. The first car in line waits at the turnoff from the main road and the other cars temporarily park in order of arrival off on the right shoulder, with two wheels on the grass, so that they don't block through traffic on the big road. As I've noted before, this constitutes a kind of surge pricing, because the cost of a parking spot is the sum of the fixed money price $6 per hour and the congestion tax of waiting in a longer, shorter line. If the line's too long, you're not willing to pay the time costs of obtaining a space and so you pull out of line. Remember, the cost of anything is the sum of the money price and the cost and inconvenience of acquiring it. So I've argued that, for example, starbucks has surge pricing. The latte always costs the same $11, but the line could be longer or shorter and you might decide yes, I will stand in that short line and pay my $11. Or no, I'm not going to stand in that long line and pay my $11 because it's too expensive. Total cost is the sum of the money price and the cost and inconvenience of acquiring the thing by paying that price.

Speaker 1:

When the smaller lot fills up, cars waiting for a spot disperse around the lot with four cars each parked in an area that commands four or five parking spaces. If someone comes back from the beach and pulls out of a spot, the car that is staked out the relevant segment gets to pull in. So I've watched these two parallel systems play out many times over the course of weeks Over the past decade. The large lot system is more rational and predictable. It seems more fair. Cars get parking in the order of their arrival. Further, because the big lot is nearly five times as big, the rate at which cars are leaving spots is far greater, and so the line moves faster.

Speaker 1:

But there's a downside. For just those reasons, the line at the big lot is really really long. The line at the small lot is usually one or two cars at most, plus the four cars that have staked out territories within the lot itself. Now you might get lucky in the small lot and only have to wait 10 or 20 minutes. You'll have to wait for an hour or more at the large lot. On the other hand, you also might get unlucky at the small lot because you only have four or five spots in the territory that you've staked out. None of those people might leave for a long time. You might have to wait for two hours or more.

Speaker 1:

There are two differences in the lots. The obvious one is size One's large and one's small. One's more than four times as big as the other, but the other is access. The smaller lot is a circle, meaning that there are in effect two entrances. Now it's supposed to be one way. There's a big sign that says one way, do not enter. But a car near the entrance could hurriedly pull around and take the short part of the circle, pulling into the vacated spot before, the car that was waiting in line could drive the long part of the circle. Now the large lot only has the one entrance. There's no way to pull around the cars ahead of you in line to jump the queue, making the more orderly line there self-enforcing.

Speaker 1:

Now, as far as I can tell, the same people behave differently in the two settings, based on what economists call the comparative statics differences in size and access. But these are the same people. They're choosing between one lot or another based on the expected waiting time and their preferences, and so comparative statics means we're changing something. Usually what we do is we have an equilibrium, we change something and then we examine the properties of the new equilibrium. This is a natural experiment. Since the two lots are only about 300 yards apart, you could switch between one or the other, but there's two different little subcultures, which means that, since assignment is either random or by choice, I can go pick the other one if there's a shorter line. It must be true that it is the difference in the two observable features I'm claiming size of the lot and access, where one has only one entrance and the other has two entrances that together explain the differences in the allocation mechanisms that have emerged. So if you're willing to take a risk that you have to wait a long time and you're willing to argue about the boundaries between the made-up territories, you'll choose the smaller lot. The average wait time there is smaller, it's true, although the variation in wait time for the smaller lot is much larger.

Speaker 1:

Over the July 4th extended weekend this year, I saw a remarkable thing. I saw a man try to break the emergent conventions, or at least to try to Now. It was in the large lot and his car was fourth in line. I happened to be standing in the lot watching, thinking about the way that the line worked, because of course I was. That's what I do.

Speaker 1:

By chance, a young woman was standing beside me and she was talking to herself oh, I can't believe it. He's doing it again. Why does he do this? She was referring to, and, as it turns out, I asked, and she told me, her father. It was obvious who she meant, though a round gentleman in a worn straw pork pie fedora. Just think about that image for a minute.

Speaker 1:

A round gentleman with a pork pie fedora, running around in a parking lot under the hot sun, accosting people who were obviously heading to their car. So people come up from the beach. They're obviously heading to their car. They've got their cart, their chair and then the umbrella, but he was accosting them back by the exit from the sandy beach. He asked them if he could buy their parking spot and he offered to pay. He found someone who would accept $10 for their spot. Then he ran to his car and I watched this, pulled around the line to go, wait by the spot that he had bought. And you can pull around the line if you pull around the road on the grass, but who does that? You have to go into the exit lane and there's a lot of people there. He waited then by the soon-to-be-emptied spot with his turn blinker on, which is the standard I-claim-this-signal-in-parking-lot language, and it worked. The paid guy pulled out Pork pie boy pulled in, jumping the queue. The daughter at this point was hiding behind the bathroom building. She was mortified with embarrassment because apparently this kind of behavior was common. But that behavior is so out of bounds that I think the people ahead of the buyer assumed something else must be going on. Nobody gets to buy a space. Maybe they're family members or something.

Speaker 1:

What's interesting is that humans are pretty good at discovering and then following rules for making allocations of scarce resources in an orderly fashion. The institution of the line is deeply ingrained in most of us to the point where we're angry if someone cuts in line, or maybe even we're ahead of us, or if someone cuts in line behind us. Even that makes us angry because they're violating the rule. It's also interesting that humans are adaptable. In the smaller lot, which is a different institution, which has territories rather than queuing, a different arrangement has emerged to solve the problem of different size and double entrance. Problem is that an otherwise stable Ostromian system can be invaded by outsiders who don't know or don't accept the norms of the system. If enough people tried to use that bidding system that Pork Pie Boy used, allocation norms now observed in equilibrium could quickly be shattered. And so when I say Ostromian, I'm talking about Lynn Ostrom and her recognition that if you have the same group of people over time solving the same problem, you'll find that norms emerge that allow them to solve property rights and common pool resource problems. But if you have new people, if they're constantly turning over and some people don't know or just don't accept the norms that until now have governed and organized the rationing of the scarce resource, then things can break down. So in this case, the two different norms you learn from the person ahead of you and it gives us a focal point to coordinate on. But if you have as is often true, I think, in a vacation region constant turnover, new people coming in almost by definition, particularly on the busy days when the lots are most likely to be full, then you may get a shattering of existing institutions. Whoa, that sound means it's time for the twedge.

Speaker 1:

This week's economics joke. Now I've heard this joke told about many places. I was at the University of Texas. I was actually told that it was true, but since then I've heard that about every university I've ever visited. So it seems unlikely that it happened at the University of Texas. But anyway, here's the joke.

Speaker 1:

It was the final examination for the required intro economics class. There are 420 students in the class and the class is required for the already overcrowded econ major, so they're trying to dissuade students by making it really hard. It's a multiple choice exam. The room only holds 500, so 420 people in the room meant a lot of people were sitting close together.

Speaker 1:

During the exam, the professor and the TAs noticed that one student was just blatantly copying from the student beside him. He also took a piece of paper from his backpack and used it to answer some of the question, referring to the piece of paper. You can't use notes. This is a blatant violation. But the professor and the TAs waited to avoid disrupting hundreds of other students with a commotion. Now the cheating student finished, comes up to the front. The professor hissed we saw you cheating. There is no reason for you to turn in that exam. You're going to fail. The student looks incredulous and angry, drew himself up and said in a staged, a stage whisper do you know who I am? Um, no, we don't and we don't care. You cheated Seriously, none of you know who I am. The student asked again, looking at the TAs who all shrugged. Now, in this class discussion section was voluntary, so no role was taken and none of the TAs knew him. The professor shook his head no and we don't care. Fair enough, said the student, darted over, stuck his exam deep in the middle of the pile of other exams, vaulted the railing and ran away singing Na na, na, na, na, na, na na, hey, hey, hey, goodbye. The point is that once the exam was buried in all the other exams, since the student was careful to establish that no one actually knew his name. There was no way of telling who it was that had cheated and since it was in the middle of the exam, you couldn't look to see if it was a lot like the students on either side, because the transaction cost of figuring out the identity of this person were just too high.

Speaker 1:

In this week's letter, jh writes Hi, mike, your podcast is really thought-provoking, many thanks. My question concerns the observation to consumers, all costs are transaction costs. Here's an excerpt from your AIER article of the same name. Reading quoting from my little article. The point is that to the buyer, all costs are transaction costs, including the purchase price. You can raise the purchase price if you cut the transaction costs by even more and the consumer will be happy. They'll buy more Because to the consumer, all costs are transaction costs. So my question should two people sharing almost everything in common income, housing, cost of living, demographics, culture and so on still value many of their transaction costs differently? This relates to how people value their non-purchase triple T transaction costs and triple T transaction costs. Jh is kindly referring to my taxonomy of triangulation, transfer and trust. It's at that point that JA gets to his real question. Back to the letter.

Speaker 1:

For example, my wife loves going to Costco to get a deal. The purchase price she pays for household items are less at Costco than at other closer stores or online options. For Costco this includes driving 10 miles, finding parking, fighting the crowds, suffering through long checkout lines, accepting the risk that a purchased item may not work or that there's too much of it and it goes bad. Why is it? I often say to myself that most of us have a desire to eat more than we need. Whether we're having peanut butter on toast or leftover rice, we eat more than we need. To me, the cost of time, gas, car maintenance, environmental damage, annoyance and risk are quite high, far beyond the cost of using Amazon to buy the same stuff and just have it delivered. On the cost of using Amazon to buy the same stuff and just have it delivered. For my wife, well, she thinks this isn't so bad and she values those non-purchase triple T items much lower.

Speaker 1:

First reaction to the question about different individual transaction cost valuation might be well, exactly, you both value the same inputs to the non-purchase triple T cost differently. So that makes perfect sense. But does it? Why would two identical people value these transaction costs so differently. This feels like the domain of behavioral economists to help us understand why two similarly situated individuals would value transaction costs so differently. Thanks, jh. End of letter. Well, thanks to you, jh, that's a terrific letter.

Speaker 1:

I'm a subjectivist and so I think that some people like chocolate and some people like vanilla ice cream, and that's pretty much all you can say. We can look how, at the margin, those people might react differently to changes in prices or changes in income, but the basic things that people like are idiosyncratic and subjective. The same thing is true about the things that people dislike. So you could imagine someone who likes going into a crowd. It's sort of a social opportunity. Some people like to go to church, I think. So they feel they get out of the house and they participate in some large social gathering. And I've never been to Costco because I, like you, would rather sit at home and order it from Amazon, even if it's slightly more expensive, and just have it delivered. But it could be a little bit like going up for communion, could be a little bit like going up for communion. You go through the line and going through that ceremony. You could imagine, at least, that someone either likes it or doesn't really dislike it, and so, as a result, we're going to find that people are going to differ in how they treat the triple T costs triangulation, transfer and trust. Some people are going to want to reduce those. Some people don't really mind them. Now my claim is that both of them are costs, so if we could reduce them, that is, the line is slightly shorter. People don't actively like standing in line, but some people will dislike it less. What I like about this question is that once we think about our reaction to transaction cost as being idiosyncratic, that might differ more than people's different reaction to price, although even then the price elasticity, which means my responsiveness to changes in price, might differ across people because my preferences might differ. And so it is an interesting question Should we add the triangulation, transfer and trust to the consumption problem when we talk about preference, indifference curves and people's response to price? So again, thanks JH for that question.

Speaker 1:

It's time for Book of the Week. This week's book is Jonathan Haidt's new book, the Anxious Generation how the Great Rewiring of Childhood is Causing an Epidemic of Mental Illness. Professor Haidt, who I know personally, has written a series of books on this question, but this may be the most provocative. It was published by Penguin Press in March of 2024, so it's a new book. It's pretty easygoing, it's not exactly a beach book, but it is worth reading and I recommend it. The next episode will be released on Tuesday, july 16th. We'll have a new topic, some letters and, of course, a hilarious new twedge. All that and more next week on Tidy C.